Token | 2010 |
Token ID | 3050 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
Balanced | +0.504 |
Dominican | +0.450 |
Nicarag | +0.449 |
Augusta | +0.431 |
KB | +0.423 |
Neg logit contributions | |
breaching | -0.768 |
etheus | -0.709 |
Eid | -0.701 |
plets | -0.696 |
hump | -0.679 |
Token | United |
Token ID | 1578 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
Balanced | +0.389 |
Dominican | +0.345 |
Nicarag | +0.343 |
Augusta | +0.328 |
KB | +0.321 |
Neg logit contributions | |
breaching | -0.696 |
etheus | -0.646 |
Eid | -0.639 |
plets | -0.635 |
hump | -0.620 |
Token | States |
Token ID | 1829 |
Feature activation | +7.32e-03 |
Pos logit contributions | |
Balanced | +0.080 |
Nicarag | +0.070 |
Dominican | +0.070 |
Augusta | +0.067 |
KB | +0.066 |
Neg logit contributions | |
breaching | -0.140 |
etheus | -0.130 |
Eid | -0.130 |
plets | -0.128 |
hump | -0.125 |
Token | 10 |
Token ID | 838 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
Balanced | +0.011 |
Dominican | +0.010 |
Nicarag | +0.010 |
Augusta | +0.009 |
KB | +0.009 |
Neg logit contributions | |
breaching | -0.018 |
etheus | -0.017 |
Eid | -0.016 |
plets | -0.016 |
hump | -0.016 |
Token | 53 |
Token ID | 4310 |
Feature activation | -5.16e-03 |
Pos logit contributions | |
Balanced | +0.329 |
Dominican | +0.297 |
Nicarag | +0.296 |
Augusta | +0.285 |
KB | +0.282 |
Neg logit contributions | |
breaching | -0.426 |
etheus | -0.391 |
Eid | -0.386 |
plets | -0.383 |
hump | -0.373 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
breaching | +0.013 |
Eid | +0.012 |
etheus | +0.012 |
plets | +0.012 |
hump | +0.012 |
Neg logit contributions | |
Balanced | -0.007 |
Nicarag | -0.006 |
Dominican | -0.006 |
Augusta | -0.006 |
KB | -0.006 |
Token | # |
Token ID | 1303 |
Feature activation | +9.24e-01 |
Pos logit contributions | |
Balanced | +1.938 |
Dominican | +1.690 |
Nicarag | +1.677 |
Augusta | +1.604 |
KB | +1.561 |
Neg logit contributions | |
breaching | -4.084 |
etheus | -3.804 |
Eid | -3.770 |
plets | -3.753 |
hump | -3.656 |
Token | 15 |
Token ID | 1314 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
Balanced | +1.575 |
KB | +1.403 |
Dominican | +1.383 |
Nicarag | +1.350 |
Augusta | +1.347 |
Neg logit contributions | |
breaching | -2.085 |
etheus | -1.986 |
pestic | -1.896 |
plets | -1.876 |
Eid | -1.876 |
Token | honestly |
Token ID | 12698 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
Balanced | +0.235 |
Dominican | +0.209 |
Nicarag | +0.209 |
Augusta | +0.201 |
KB | +0.198 |
Neg logit contributions | |
breaching | -0.356 |
etheus | -0.329 |
Eid | -0.325 |
plets | -0.323 |
hump | -0.314 |
Token | if |
Token ID | 611 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
Balanced | +0.327 |
Dominican | +0.296 |
Nicarag | +0.295 |
Augusta | +0.285 |
KB | +0.282 |
Neg logit contributions | |
breaching | -0.398 |
etheus | -0.367 |
Eid | -0.360 |
plets | -0.359 |
hump | -0.347 |
Token | we |
Token ID | 356 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
Balanced | +0.239 |
Dominican | +0.218 |
Nicarag | +0.217 |
Augusta | +0.211 |
KB | +0.208 |
Neg logit contributions | |
breaching | -0.259 |
etheus | -0.237 |
Eid | -0.233 |
plets | -0.231 |
hump | -0.224 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
Balanced | +0.755 |
Dominican | +0.674 |
Nicarag | +0.671 |
Augusta | +0.645 |
KB | +0.631 |
Neg logit contributions | |
breaching | -1.156 |
etheus | -1.068 |
Eid | -1.053 |
plets | -1.050 |
hump | -1.022 |
Token | United |
Token ID | 1578 |
Feature activation | +8.11e-02 |
Pos logit contributions | |
Balanced | +0.430 |
Dominican | +0.382 |
Nicarag | +0.380 |
Augusta | +0.363 |
KB | +0.355 |
Neg logit contributions | |
breaching | -0.766 |
etheus | -0.711 |
Eid | -0.702 |
plets | -0.699 |
hump | -0.681 |
Token | States |
Token ID | 1829 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
Balanced | +0.111 |
Dominican | +0.098 |
Nicarag | +0.098 |
Augusta | +0.093 |
KB | +0.091 |
Neg logit contributions | |
breaching | -0.209 |
etheus | -0.195 |
Eid | -0.192 |
plets | -0.191 |
hump | -0.187 |
Token | 3 |
Token ID | 513 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
Balanced | +0.069 |
Dominican | +0.062 |
Nicarag | +0.061 |
Augusta | +0.059 |
KB | +0.058 |
Neg logit contributions | |
breaching | -0.110 |
etheus | -0.101 |
Eid | -0.100 |
plets | -0.100 |
hump | -0.097 |
Token | 240 |
Token ID | 16102 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
Balanced | +0.142 |
Dominican | +0.126 |
Nicarag | +0.125 |
Augusta | +0.119 |
KB | +0.118 |
Neg logit contributions | |
breaching | -0.256 |
etheus | -0.237 |
Eid | -0.235 |
plets | -0.234 |
hump | -0.228 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
Balanced | +0.147 |
Nicarag | +0.128 |
Dominican | +0.128 |
Augusta | +0.121 |
KB | +0.119 |
Neg logit contributions | |
breaching | -0.302 |
etheus | -0.281 |
Eid | -0.279 |
plets | -0.276 |
hump | -0.271 |
Token | # |
Token ID | 1303 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Balanced | +1.782 |
Dominican | +1.558 |
Nicarag | +1.529 |
Augusta | +1.490 |
Container | +1.448 |
Neg logit contributions | |
breaching | -3.997 |
etheus | -3.738 |
plets | -3.718 |
Eid | -3.702 |
hump | -3.580 |
Token | 10 |
Token ID | 940 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
Balanced | +1.784 |
Dominican | +1.603 |
Nicarag | +1.602 |
KB | +1.577 |
Augusta | +1.550 |
Neg logit contributions | |
breaching | -2.213 |
etheus | -2.022 |
plets | -1.979 |
Eid | -1.970 |
hump | -1.923 |
Token | Happy |
Token ID | 14628 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
Balanced | +0.527 |
Dominican | +0.471 |
Nicarag | +0.469 |
Augusta | +0.454 |
KB | +0.446 |
Neg logit contributions | |
breaching | -0.758 |
etheus | -0.702 |
Eid | -0.691 |
plets | -0.688 |
hump | -0.667 |
Token | it |
Token ID | 340 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Balanced | +0.116 |
Dominican | +0.106 |
Nicarag | +0.106 |
Augusta | +0.102 |
Container | +0.100 |
Neg logit contributions | |
breaching | -0.130 |
etheus | -0.120 |
plets | -0.117 |
Eid | -0.117 |
horns | -0.112 |
Token | worked |
Token ID | 3111 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
Balanced | +0.175 |
Dominican | +0.157 |
Nicarag | +0.157 |
Augusta | +0.151 |
KB | +0.149 |
Neg logit contributions | |
breaching | -0.236 |
etheus | -0.217 |
Eid | -0.215 |
plets | -0.213 |
hump | -0.207 |
Token | 2002 |
Token ID | 6244 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
Balanced | +1.326 |
Dominican | +1.192 |
Nicarag | +1.185 |
Augusta | +1.142 |
KB | +1.117 |
Neg logit contributions | |
breaching | -1.931 |
etheus | -1.783 |
Eid | -1.762 |
plets | -1.752 |
hump | -1.704 |
Token | United |
Token ID | 1578 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
Balanced | +0.059 |
Dominican | +0.052 |
Nicarag | +0.052 |
Augusta | +0.049 |
KB | +0.049 |
Neg logit contributions | |
breaching | -0.104 |
etheus | -0.096 |
Eid | -0.096 |
plets | -0.095 |
hump | -0.093 |
Token | States |
Token ID | 1829 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
Balanced | +0.175 |
Nicarag | +0.154 |
Dominican | +0.153 |
Augusta | +0.146 |
KB | +0.145 |
Neg logit contributions | |
breaching | -0.314 |
etheus | -0.292 |
Eid | -0.290 |
plets | -0.286 |
hump | -0.281 |
Token | 3 |
Token ID | 513 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
Balanced | +0.187 |
Dominican | +0.166 |
Nicarag | +0.166 |
Augusta | +0.159 |
KB | +0.155 |
Neg logit contributions | |
breaching | -0.307 |
etheus | -0.285 |
Eid | -0.281 |
plets | -0.280 |
hump | -0.272 |
Token | 472 |
Token ID | 37856 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
Balanced | +0.077 |
Dominican | +0.069 |
Nicarag | +0.069 |
Augusta | +0.066 |
KB | +0.065 |
Neg logit contributions | |
breaching | -0.106 |
etheus | -0.097 |
Eid | -0.096 |
plets | -0.096 |
hump | -0.093 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Balanced | +0.133 |
Nicarag | +0.117 |
Dominican | +0.116 |
Augusta | +0.110 |
KB | +0.108 |
Neg logit contributions | |
breaching | -0.265 |
Eid | -0.247 |
etheus | -0.246 |
plets | -0.242 |
hump | -0.238 |
Token | # |
Token ID | 1303 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
Balanced | +1.802 |
Dominican | +1.575 |
Nicarag | +1.552 |
Augusta | +1.497 |
Container | +1.452 |
Neg logit contributions | |
breaching | -3.909 |
etheus | -3.651 |
Eid | -3.614 |
plets | -3.611 |
hump | -3.504 |
Token | 11 |
Token ID | 1157 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Balanced | +1.035 |
Dominican | +0.889 |
Nicarag | +0.878 |
KB | +0.871 |
Augusta | +0.849 |
Neg logit contributions | |
breaching | -2.101 |
etheus | -1.958 |
Eid | -1.940 |
plets | -1.929 |
pestic | -1.888 |
Token | On |
Token ID | 1550 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
Balanced | +0.231 |
Dominican | +0.207 |
Nicarag | +0.207 |
Augusta | +0.199 |
KB | +0.196 |
Neg logit contributions | |
breaching | -0.317 |
etheus | -0.293 |
Eid | -0.289 |
plets | -0.287 |
hump | -0.278 |
Token | September |
Token ID | 2693 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
Balanced | +0.625 |
Dominican | +0.554 |
Nicarag | +0.550 |
Augusta | +0.538 |
KB | +0.521 |
Neg logit contributions | |
breaching | -1.045 |
etheus | -0.979 |
plets | -0.959 |
Eid | -0.958 |
hump | -0.930 |
Token | 20 |
Token ID | 1160 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Balanced | +0.555 |
Dominican | +0.495 |
Nicarag | +0.494 |
Augusta | +0.475 |
KB | +0.466 |
Neg logit contributions | |
breaching | -0.848 |
etheus | -0.788 |
plets | -0.771 |
Eid | -0.770 |
hump | -0.750 |
Token | June |
Token ID | 2795 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
Balanced | +0.675 |
Dominican | +0.607 |
Nicarag | +0.604 |
Augusta | +0.582 |
KB | +0.569 |
Neg logit contributions | |
breaching | -0.996 |
etheus | -0.923 |
plets | -0.906 |
Eid | -0.905 |
hump | -0.877 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
Balanced | +0.559 |
Dominican | +0.499 |
Nicarag | +0.498 |
Augusta | +0.478 |
KB | +0.470 |
Neg logit contributions | |
breaching | -0.864 |
etheus | -0.799 |
Eid | -0.790 |
plets | -0.784 |
hump | -0.765 |
Token | Canada |
Token ID | 3340 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
Balanced | +0.397 |
Dominican | +0.352 |
Nicarag | +0.350 |
Augusta | +0.334 |
KB | +0.327 |
Neg logit contributions | |
breaching | -0.712 |
etheus | -0.661 |
Eid | -0.653 |
plets | -0.650 |
hump | -0.634 |
Token | 16 |
Token ID | 1467 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
Balanced | +0.469 |
Dominican | +0.419 |
Nicarag | +0.417 |
Augusta | +0.400 |
KB | +0.392 |
Neg logit contributions | |
breaching | -0.754 |
etheus | -0.699 |
Eid | -0.690 |
plets | -0.686 |
hump | -0.668 |
Token | 10 |
Token ID | 940 |
Feature activation | +3.30e-02 |
Pos logit contributions | |
breaching | +0.043 |
etheus | +0.039 |
Eid | +0.039 |
plets | +0.039 |
hump | +0.038 |
Neg logit contributions | |
Balanced | -0.034 |
Dominican | -0.031 |
Nicarag | -0.031 |
Augusta | -0.030 |
KB | -0.030 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Balanced | +0.044 |
Nicarag | +0.038 |
Dominican | +0.038 |
Augusta | +0.036 |
KB | +0.036 |
Neg logit contributions | |
breaching | -0.085 |
Eid | -0.080 |
etheus | -0.079 |
plets | -0.078 |
hump | -0.077 |
Token | # |
Token ID | 1303 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
Balanced | +1.842 |
Dominican | +1.610 |
Nicarag | +1.599 |
Augusta | +1.525 |
KB | +1.484 |
Neg logit contributions | |
breaching | -3.872 |
etheus | -3.604 |
Eid | -3.574 |
plets | -3.558 |
hump | -3.466 |
Token | 14 |
Token ID | 1415 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Balanced | +1.369 |
Dominican | +1.250 |
KB | +1.227 |
Augusta | +1.199 |
Nicarag | +1.194 |
Neg logit contributions | |
breaching | -2.092 |
etheus | -1.986 |
Eid | -1.907 |
pestic | -1.894 |
plets | -1.881 |
Token | On |
Token ID | 1550 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
Balanced | +0.263 |
Dominican | +0.235 |
Nicarag | +0.235 |
Augusta | +0.226 |
KB | +0.222 |
Neg logit contributions | |
breaching | -0.397 |
etheus | -0.367 |
Eid | -0.362 |
plets | -0.360 |
hump | -0.350 |
Token | September |
Token ID | 2693 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
Balanced | +0.559 |
Nicarag | +0.507 |
Dominican | +0.506 |
Augusta | +0.487 |
KB | +0.477 |
Neg logit contributions | |
breaching | -0.743 |
etheus | -0.697 |
plets | -0.686 |
Eid | -0.675 |
hump | -0.652 |
Token | 20 |
Token ID | 1160 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
Balanced | +0.481 |
Dominican | +0.440 |
Nicarag | +0.440 |
Augusta | +0.425 |
KB | +0.420 |
Neg logit contributions | |
breaching | -0.538 |
etheus | -0.488 |
Eid | -0.488 |
plets | -0.478 |
horns | -0.464 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
Balanced | +0.713 |
Dominican | +0.640 |
Nicarag | +0.639 |
Augusta | +0.614 |
KB | +0.605 |
Neg logit contributions | |
breaching | -1.011 |
etheus | -0.931 |
Eid | -0.922 |
plets | -0.913 |
hump | -0.891 |
Token | United |
Token ID | 1578 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Balanced | +0.560 |
Dominican | +0.496 |
Nicarag | +0.493 |
Augusta | +0.471 |
KB | +0.461 |
Neg logit contributions | |
breaching | -1.011 |
etheus | -0.938 |
Eid | -0.928 |
plets | -0.922 |
hump | -0.900 |
Token | States |
Token ID | 1829 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Balanced | +0.221 |
Nicarag | +0.194 |
Dominican | +0.194 |
Augusta | +0.185 |
KB | +0.183 |
Neg logit contributions | |
breaching | -0.392 |
etheus | -0.364 |
Eid | -0.362 |
plets | -0.357 |
hump | -0.350 |
Token | 21 |
Token ID | 2310 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
Balanced | +0.180 |
Dominican | +0.161 |
Nicarag | +0.160 |
Augusta | +0.154 |
KB | +0.150 |
Neg logit contributions | |
breaching | -0.288 |
etheus | -0.267 |
Eid | -0.263 |
plets | -0.263 |
hump | -0.255 |
Token | 50 |
Token ID | 1120 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
breaching | +0.128 |
etheus | +0.117 |
Eid | +0.116 |
plets | +0.114 |
hump | +0.111 |
Neg logit contributions | |
Balanced | -0.105 |
Dominican | -0.095 |
Nicarag | -0.095 |
Augusta | -0.092 |
KB | -0.091 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
breaching | +0.259 |
Eid | +0.242 |
etheus | +0.240 |
plets | +0.236 |
hump | +0.233 |
Neg logit contributions | |
Balanced | -0.132 |
Nicarag | -0.116 |
Dominican | -0.116 |
Augusta | -0.110 |
KB | -0.108 |
Token | # |
Token ID | 1303 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
Balanced | +1.775 |
Dominican | +1.549 |
Nicarag | +1.538 |
Augusta | +1.469 |
KB | +1.432 |
Neg logit contributions | |
breaching | -3.704 |
etheus | -3.451 |
Eid | -3.420 |
plets | -3.401 |
hump | -3.316 |
Token | 16 |
Token ID | 1433 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
Balanced | +1.506 |
Dominican | +1.360 |
Nicarag | +1.340 |
KB | +1.333 |
Augusta | +1.332 |
Neg logit contributions | |
breaching | -1.761 |
etheus | -1.634 |
Eid | -1.592 |
plets | -1.583 |
hump | -1.538 |
Token | Glad |
Token ID | 28442 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Balanced | +0.268 |
Dominican | +0.239 |
Nicarag | +0.239 |
Augusta | +0.230 |
KB | +0.226 |
Neg logit contributions | |
breaching | -0.409 |
etheus | -0.379 |
Eid | -0.374 |
plets | -0.372 |
hump | -0.361 |
Token | to |
Token ID | 284 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
Balanced | +0.213 |
Dominican | +0.194 |
Nicarag | +0.193 |
Augusta | +0.187 |
KB | +0.187 |
Neg logit contributions | |
breaching | -0.229 |
etheus | -0.210 |
Eid | -0.205 |
plets | -0.204 |
hump | -0.197 |
Token | hear |
Token ID | 3285 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
Balanced | +0.172 |
Dominican | +0.156 |
Nicarag | +0.156 |
Augusta | +0.150 |
KB | +0.148 |
Neg logit contributions | |
breaching | -0.212 |
etheus | -0.195 |
Eid | -0.192 |
plets | -0.190 |
hump | -0.184 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
Balanced | +0.489 |
Dominican | +0.438 |
Nicarag | +0.436 |
Augusta | +0.419 |
KB | +0.411 |
Neg logit contributions | |
breaching | -0.734 |
etheus | -0.678 |
Eid | -0.670 |
plets | -0.665 |
hump | -0.648 |
Token | United |
Token ID | 1578 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Balanced | +0.613 |
Dominican | +0.544 |
Nicarag | +0.540 |
Augusta | +0.515 |
KB | +0.503 |
Neg logit contributions | |
breaching | -1.107 |
etheus | -1.028 |
Eid | -1.014 |
plets | -1.010 |
hump | -0.984 |
Token | States |
Token ID | 1829 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
Balanced | +0.202 |
Nicarag | +0.178 |
Dominican | +0.177 |
Augusta | +0.169 |
KB | +0.168 |
Neg logit contributions | |
breaching | -0.356 |
etheus | -0.330 |
Eid | -0.328 |
plets | -0.324 |
hump | -0.317 |
Token | 228 |
Token ID | 29041 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
Balanced | +0.075 |
Dominican | +0.067 |
Nicarag | +0.067 |
Augusta | +0.064 |
KB | +0.063 |
Neg logit contributions | |
breaching | -0.121 |
etheus | -0.112 |
Eid | -0.111 |
plets | -0.110 |
hump | -0.107 |
Token | 65 |
Token ID | 2996 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
breaching | +0.172 |
etheus | +0.157 |
Eid | +0.155 |
plets | +0.154 |
hump | +0.150 |
Neg logit contributions | |
Balanced | -0.135 |
Dominican | -0.122 |
Nicarag | -0.122 |
Augusta | -0.117 |
KB | -0.116 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
Balanced | +0.243 |
Nicarag | +0.214 |
Dominican | +0.213 |
Augusta | +0.202 |
KB | +0.198 |
Neg logit contributions | |
breaching | -0.480 |
Eid | -0.448 |
etheus | -0.446 |
plets | -0.439 |
hump | -0.432 |
Token | # |
Token ID | 1303 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
Balanced | +1.766 |
Dominican | +1.545 |
Nicarag | +1.532 |
Augusta | +1.465 |
KB | +1.425 |
Neg logit contributions | |
breaching | -3.702 |
etheus | -3.446 |
Eid | -3.420 |
plets | -3.401 |
hump | -3.315 |
Token | 13 |
Token ID | 1485 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
Balanced | +1.281 |
Dominican | +1.171 |
Nicarag | +1.147 |
KB | +1.120 |
Augusta | +1.109 |
Neg logit contributions | |
breaching | -1.676 |
etheus | -1.558 |
Eid | -1.526 |
plets | -1.505 |
pestic | -1.472 |
Token | On |
Token ID | 1550 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
Balanced | +0.022 |
Dominican | +0.020 |
Nicarag | +0.020 |
Augusta | +0.019 |
KB | +0.019 |
Neg logit contributions | |
breaching | -0.032 |
etheus | -0.030 |
Eid | -0.029 |
plets | -0.029 |
hump | -0.028 |
Token | September |
Token ID | 2693 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
Balanced | +0.423 |
Nicarag | +0.384 |
Dominican | +0.383 |
Augusta | +0.369 |
KB | +0.360 |
Neg logit contributions | |
breaching | -0.568 |
etheus | -0.534 |
plets | -0.526 |
Eid | -0.517 |
hump | -0.499 |
Token | 20 |
Token ID | 1160 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
Balanced | +0.559 |
Dominican | +0.506 |
Nicarag | +0.505 |
Augusta | +0.487 |
KB | +0.481 |
Neg logit contributions | |
breaching | -0.677 |
etheus | -0.620 |
Eid | -0.613 |
plets | -0.606 |
hump | -0.591 |
Token | 2012 |
Token ID | 2321 |
Feature activation | +9.62e-02 |
Pos logit contributions | |
Balanced | +0.835 |
Dominican | +0.747 |
Nicarag | +0.743 |
Augusta | +0.714 |
KB | +0.698 |
Neg logit contributions | |
breaching | -1.289 |
etheus | -1.193 |
Eid | -1.176 |
plets | -1.172 |
hump | -1.141 |
Token | United |
Token ID | 1578 |
Feature activation | +8.23e-02 |
Pos logit contributions | |
Balanced | +0.136 |
Dominican | +0.120 |
Nicarag | +0.120 |
Augusta | +0.115 |
KB | +0.112 |
Neg logit contributions | |
breaching | -0.244 |
etheus | -0.226 |
Eid | -0.224 |
plets | -0.222 |
hump | -0.217 |
Token | States |
Token ID | 1829 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
Balanced | +0.117 |
Nicarag | +0.103 |
Dominican | +0.102 |
Augusta | +0.098 |
KB | +0.097 |
Neg logit contributions | |
breaching | -0.207 |
etheus | -0.192 |
Eid | -0.192 |
plets | -0.189 |
hump | -0.185 |
Token | 5 |
Token ID | 642 |
Feature activation | +9.41e-02 |
Pos logit contributions | |
Balanced | +0.008 |
Dominican | +0.007 |
Nicarag | +0.007 |
Augusta | +0.007 |
KB | +0.007 |
Neg logit contributions | |
breaching | -0.012 |
etheus | -0.012 |
Eid | -0.011 |
plets | -0.011 |
hump | -0.011 |
Token | 667 |
Token ID | 28933 |
Feature activation | -6.26e-03 |
Pos logit contributions | |
Balanced | +0.163 |
Dominican | +0.148 |
Nicarag | +0.147 |
Augusta | +0.142 |
KB | +0.141 |
Neg logit contributions | |
breaching | -0.209 |
etheus | -0.192 |
Eid | -0.189 |
plets | -0.188 |
hump | -0.183 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
breaching | +0.016 |
Eid | +0.015 |
etheus | +0.015 |
plets | +0.015 |
hump | +0.015 |
Neg logit contributions | |
Balanced | -0.008 |
Nicarag | -0.007 |
Dominican | -0.007 |
Augusta | -0.007 |
KB | -0.007 |
Token | # |
Token ID | 1303 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
Balanced | +1.746 |
Dominican | +1.527 |
Nicarag | +1.514 |
Augusta | +1.451 |
KB | +1.410 |
Neg logit contributions | |
breaching | -3.685 |
etheus | -3.429 |
Eid | -3.403 |
plets | -3.388 |
hump | -3.299 |
Token | 12 |
Token ID | 1065 |
Feature activation | +3.35e-02 |
Pos logit contributions | |
Balanced | +1.247 |
Dominican | +1.132 |
Nicarag | +1.117 |
KB | +1.092 |
Augusta | +1.088 |
Neg logit contributions | |
breaching | -1.640 |
etheus | -1.526 |
Eid | -1.495 |
plets | -1.476 |
pestic | -1.441 |
Token | I |
Token ID | 314 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
Balanced | +0.054 |
Dominican | +0.048 |
Nicarag | +0.048 |
Augusta | +0.046 |
KB | +0.045 |
Neg logit contributions | |
breaching | -0.079 |
etheus | -0.073 |
Eid | -0.072 |
plets | -0.071 |
hump | -0.069 |
Token | 'm |
Token ID | 1101 |
Feature activation | +6.79e-04 |
Pos logit contributions | |
Balanced | +0.050 |
Dominican | +0.045 |
Nicarag | +0.045 |
Augusta | +0.044 |
KB | +0.043 |
Neg logit contributions | |
breaching | -0.064 |
etheus | -0.058 |
Eid | -0.058 |
plets | -0.057 |
hump | -0.055 |
Token | glad |
Token ID | 9675 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Balanced | +0.001 |
Dominican | +0.001 |
Nicarag | +0.001 |
Augusta | +0.001 |
KB | +0.001 |
Neg logit contributions | |
breaching | -0.001 |
etheus | -0.001 |
Eid | -0.001 |
plets | -0.001 |
hump | -0.001 |
Token | TL |
Token ID | 24811 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
Balanced | +0.606 |
Dominican | +0.536 |
Nicarag | +0.533 |
Augusta | +0.508 |
KB | +0.498 |
Neg logit contributions | |
breaching | -1.107 |
etheus | -1.027 |
Eid | -1.016 |
plets | -1.010 |
hump | -0.986 |
Token | AD |
Token ID | 2885 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
Balanced | +0.882 |
Nicarag | +0.787 |
Dominican | +0.784 |
Augusta | +0.754 |
KB | +0.745 |
Neg logit contributions | |
breaching | -1.352 |
etheus | -1.245 |
Eid | -1.242 |
plets | -1.221 |
hump | -1.198 |
Token | T |
Token ID | 51 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
breaching | +0.112 |
etheus | +0.103 |
Eid | +0.102 |
plets | +0.101 |
hump | +0.098 |
Neg logit contributions | |
Balanced | -0.088 |
Dominican | -0.079 |
Nicarag | -0.079 |
Augusta | -0.076 |
KB | -0.075 |
Token | 235 |
Token ID | 28878 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
Balanced | +0.191 |
Dominican | +0.169 |
Nicarag | +0.168 |
Augusta | +0.161 |
KB | +0.158 |
Neg logit contributions | |
breaching | -0.351 |
etheus | -0.326 |
Eid | -0.323 |
plets | -0.320 |
hump | -0.313 |
Token | 21 |
Token ID | 2481 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
Balanced | +0.071 |
Dominican | +0.064 |
Nicarag | +0.064 |
Augusta | +0.061 |
KB | +0.061 |
Neg logit contributions | |
breaching | -0.088 |
etheus | -0.081 |
Eid | -0.080 |
plets | -0.079 |
hump | -0.077 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
breaching | +0.455 |
Eid | +0.427 |
etheus | +0.420 |
plets | +0.413 |
hump | +0.410 |
Neg logit contributions | |
Balanced | -0.240 |
Nicarag | -0.213 |
Dominican | -0.212 |
Augusta | -0.201 |
KB | -0.196 |
Token | # |
Token ID | 1303 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
Balanced | +1.699 |
Dominican | +1.482 |
Nicarag | +1.473 |
Augusta | +1.407 |
KB | +1.369 |
Neg logit contributions | |
breaching | -3.542 |
etheus | -3.299 |
Eid | -3.270 |
plets | -3.251 |
hump | -3.171 |
Token | 18 |
Token ID | 1507 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
Balanced | +1.368 |
KB | +1.252 |
Dominican | +1.243 |
Augusta | +1.235 |
Nicarag | +1.223 |
Neg logit contributions | |
breaching | -1.541 |
etheus | -1.472 |
Eid | -1.398 |
pestic | -1.389 |
plets | -1.384 |
Token | Good |
Token ID | 4599 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
Balanced | +0.311 |
Dominican | +0.278 |
Nicarag | +0.278 |
Augusta | +0.268 |
KB | +0.263 |
Neg logit contributions | |
breaching | -0.450 |
etheus | -0.416 |
Eid | -0.410 |
plets | -0.408 |
hump | -0.396 |
Token | stuff |
Token ID | 3404 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
Balanced | +0.563 |
Dominican | +0.510 |
Nicarag | +0.510 |
Augusta | +0.498 |
KB | +0.495 |
Neg logit contributions | |
breaching | -0.570 |
etheus | -0.520 |
plets | -0.508 |
Eid | -0.508 |
hump | -0.487 |
Token | ! |
Token ID | 0 |
Feature activation | +6.11e-02 |
Pos logit contributions | |
Balanced | +0.655 |
Nicarag | +0.590 |
Dominican | +0.588 |
Augusta | +0.573 |
KB | +0.566 |
Neg logit contributions | |
breaching | -0.806 |
etheus | -0.741 |
plets | -0.726 |
Eid | -0.724 |
hump | -0.703 |
Token | national |
Token ID | 2260 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
Balanced | +0.039 |
Nicarag | +0.036 |
Dominican | +0.036 |
Augusta | +0.035 |
KB | +0.034 |
Neg logit contributions | |
breaching | -0.043 |
etheus | -0.040 |
Eid | -0.039 |
plets | -0.039 |
hump | -0.037 |
Token | economics |
Token ID | 12446 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
breaching | +0.820 |
etheus | +0.751 |
Eid | +0.741 |
plets | +0.738 |
hump | +0.713 |
Neg logit contributions | |
Balanced | -0.740 |
Nicarag | -0.676 |
Dominican | -0.676 |
Augusta | -0.651 |
KB | -0.640 |
Token | correspondent |
Token ID | 21076 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
breaching | +0.095 |
etheus | +0.088 |
plets | +0.086 |
Eid | +0.086 |
hump | +0.083 |
Neg logit contributions | |
Balanced | -0.080 |
Nicarag | -0.073 |
Dominican | -0.073 |
Augusta | -0.070 |
KB | -0.069 |
Token | for |
Token ID | 329 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Balanced | +0.079 |
Dominican | +0.071 |
Nicarag | +0.071 |
Augusta | +0.069 |
KB | +0.067 |
Neg logit contributions | |
breaching | -0.112 |
etheus | -0.104 |
plets | -0.102 |
Eid | -0.102 |
hump | -0.099 |
Token | McCl |
Token ID | 23780 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
Balanced | +0.214 |
Dominican | +0.194 |
Nicarag | +0.193 |
Augusta | +0.189 |
KB | +0.185 |
Neg logit contributions | |
breaching | -0.241 |
etheus | -0.221 |
Eid | -0.217 |
plets | -0.217 |
hump | -0.210 |
Token | atch |
Token ID | 963 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Balanced | +0.108 |
Dominican | +0.089 |
Nicarag | +0.089 |
KB | +0.083 |
Augusta | +0.083 |
Neg logit contributions | |
breaching | -0.331 |
Eid | -0.308 |
etheus | -0.308 |
plets | -0.303 |
hump | -0.300 |
Token | y |
Token ID | 88 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
Balanced | +2.009 |
Dominican | +1.756 |
Nicarag | +1.750 |
Augusta | +1.700 |
KB | +1.688 |
Neg logit contributions | |
breaching | -3.131 |
etheus | -2.909 |
Eid | -2.812 |
plets | -2.810 |
hump | -2.748 |
Token | Newsp |
Token ID | 36765 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
Balanced | +0.590 |
Dominican | +0.534 |
Nicarag | +0.532 |
Augusta | +0.514 |
KB | +0.506 |
Neg logit contributions | |
breaching | -0.733 |
etheus | -0.678 |
Eid | -0.669 |
plets | -0.663 |
hump | -0.646 |
Token | apers |
Token ID | 5656 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
breaching | +0.133 |
etheus | +0.120 |
Eid | +0.120 |
plets | +0.117 |
hump | +0.115 |
Neg logit contributions | |
Balanced | -0.127 |
Dominican | -0.117 |
Nicarag | -0.116 |
Augusta | -0.113 |
KB | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
Balanced | +0.890 |
Dominican | +0.806 |
Nicarag | +0.800 |
Augusta | +0.771 |
KB | +0.765 |
Neg logit contributions | |
breaching | -1.141 |
etheus | -1.056 |
Eid | -1.040 |
plets | -1.035 |
hump | -1.009 |
Token | Jay |
Token ID | 9180 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
Balanced | +0.111 |
Dominican | +0.102 |
Nicarag | +0.102 |
Augusta | +0.099 |
KB | +0.098 |
Neg logit contributions | |
breaching | -0.098 |
etheus | -0.089 |
plets | -0.086 |
Eid | -0.086 |
hump | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
breaching | +0.173 |
etheus | +0.159 |
Eid | +0.158 |
plets | +0.155 |
hump | +0.152 |
Neg logit contributions | |
Balanced | -0.132 |
Nicarag | -0.119 |
Dominican | -0.119 |
Augusta | -0.114 |
KB | -0.113 |
Token | impl |
Token ID | 23928 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
breaching | +0.078 |
etheus | +0.071 |
Eid | +0.071 |
plets | +0.070 |
hump | +0.068 |
Neg logit contributions | |
Balanced | -0.056 |
Dominican | -0.051 |
Nicarag | -0.051 |
Augusta | -0.049 |
KB | -0.048 |
Token | Board |
Token ID | 5926 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
Balanced | +0.304 |
Dominican | +0.274 |
Nicarag | +0.274 |
Augusta | +0.264 |
KB | +0.260 |
Neg logit contributions | |
breaching | -0.394 |
etheus | -0.362 |
Eid | -0.358 |
plets | -0.354 |
hump | -0.345 |
Token | { |
Token ID | 1391 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
breaching | +0.750 |
Eid | +0.690 |
etheus | +0.689 |
plets | +0.677 |
hump | +0.669 |
Neg logit contributions | |
Balanced | -0.408 |
Nicarag | -0.367 |
Dominican | -0.364 |
Augusta | -0.343 |
KB | -0.336 |
Token | pub |
Token ID | 2240 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
breaching | +0.342 |
Eid | +0.314 |
etheus | +0.312 |
plets | +0.307 |
hump | +0.303 |
Neg logit contributions | |
Balanced | -0.227 |
Nicarag | -0.206 |
Dominican | -0.205 |
Augusta | -0.195 |
KB | -0.194 |
Token | fn |
Token ID | 24714 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Balanced | +0.803 |
Dominican | +0.711 |
Nicarag | +0.705 |
Augusta | +0.683 |
KB | +0.670 |
Neg logit contributions | |
breaching | -1.302 |
etheus | -1.206 |
Eid | -1.191 |
plets | -1.189 |
hump | -1.152 |
Token | copy |
Token ID | 4866 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
Balanced | +2.218 |
Dominican | +1.976 |
Nicarag | +1.970 |
Augusta | +1.949 |
Container | +1.900 |
Neg logit contributions | |
breaching | -2.819 |
etheus | -2.640 |
plets | -2.580 |
Eid | -2.557 |
horns | -2.466 |
Token | _ |
Token ID | 62 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
breaching | +0.812 |
etheus | +0.739 |
Eid | +0.734 |
plets | +0.723 |
hump | +0.704 |
Neg logit contributions | |
Balanced | -0.764 |
Nicarag | -0.701 |
Dominican | -0.700 |
Augusta | -0.671 |
KB | -0.664 |
Token | board |
Token ID | 3526 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
Balanced | +0.080 |
Dominican | +0.073 |
Nicarag | +0.072 |
Augusta | +0.070 |
KB | +0.069 |
Neg logit contributions | |
breaching | -0.100 |
etheus | -0.092 |
Eid | -0.091 |
plets | -0.090 |
hump | -0.088 |
Token | ( |
Token ID | 357 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
breaching | +0.296 |
etheus | +0.269 |
Eid | +0.267 |
plets | +0.263 |
hump | +0.257 |
Neg logit contributions | |
Balanced | -0.271 |
Dominican | -0.247 |
Nicarag | -0.247 |
Augusta | -0.238 |
KB | -0.235 |
Token | & |
Token ID | 1222 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
breaching | +0.120 |
etheus | +0.111 |
Eid | +0.110 |
plets | +0.109 |
hump | +0.106 |
Neg logit contributions | |
Balanced | -0.071 |
Dominican | -0.063 |
Nicarag | -0.063 |
Augusta | -0.060 |
KB | -0.060 |
Token | 157 |
Token ID | 18458 |
Feature activation | +6.80e-02 |
Pos logit contributions | |
Balanced | +0.532 |
Dominican | +0.479 |
Nicarag | +0.478 |
Augusta | +0.460 |
KB | +0.456 |
Neg logit contributions | |
breaching | -0.708 |
etheus | -0.652 |
Eid | -0.646 |
plets | -0.639 |
hump | -0.624 |
Token | ] |
Token ID | 60 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Balanced | +0.105 |
Nicarag | +0.093 |
Dominican | +0.093 |
Augusta | +0.089 |
KB | +0.088 |
Neg logit contributions | |
breaching | -0.163 |
etheus | -0.150 |
Eid | -0.150 |
plets | -0.148 |
hump | -0.145 |
Token |
|
Token ID | 198 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
Balanced | +0.197 |
Dominican | +0.177 |
Nicarag | +0.177 |
Augusta | +0.171 |
KB | +0.168 |
Neg logit contributions | |
breaching | -0.260 |
etheus | -0.240 |
Eid | -0.236 |
plets | -0.235 |
hump | -0.229 |
Token |
|
Token ID | 198 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
breaching | +0.131 |
etheus | +0.120 |
Eid | +0.120 |
plets | +0.118 |
hump | +0.115 |
Neg logit contributions | |
Balanced | -0.100 |
Nicarag | -0.090 |
Dominican | -0.090 |
Augusta | -0.087 |
KB | -0.086 |
Token | Th |
Token ID | 817 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
Balanced | +0.045 |
Nicarag | +0.040 |
Dominican | +0.040 |
KB | +0.038 |
Augusta | +0.038 |
Neg logit contributions | |
breaching | -0.065 |
etheus | -0.060 |
Eid | -0.059 |
plets | -0.058 |
hump | -0.057 |
Token | rill |
Token ID | 20190 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Balanced | +1.006 |
Nicarag | +0.922 |
Dominican | +0.912 |
KB | +0.889 |
Augusta | +0.882 |
Neg logit contributions | |
breaching | -1.166 |
Eid | -1.051 |
etheus | -1.049 |
plets | -1.031 |
hump | -1.009 |
Token | ride |
Token ID | 6594 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
Balanced | +2.095 |
Dominican | +1.859 |
Nicarag | +1.833 |
Augusta | +1.803 |
KB | +1.779 |
Neg logit contributions | |
breaching | -2.855 |
etheus | -2.704 |
plets | -2.629 |
Eid | -2.604 |
hump | -2.542 |
Token | [ |
Token ID | 685 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
breaching | +0.150 |
etheus | +0.138 |
Eid | +0.136 |
plets | +0.135 |
hump | +0.130 |
Neg logit contributions | |
Balanced | -0.129 |
Nicarag | -0.117 |
Dominican | -0.117 |
Augusta | -0.113 |
KB | -0.111 |
Token | edit |
Token ID | 4370 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
breaching | +0.143 |
etheus | +0.131 |
Eid | +0.130 |
plets | +0.128 |
hump | +0.125 |
Neg logit contributions | |
Balanced | -0.113 |
Dominican | -0.103 |
Nicarag | -0.102 |
Augusta | -0.099 |
KB | -0.098 |
Token | ] |
Token ID | 2361 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
Balanced | +0.136 |
Dominican | +0.123 |
Nicarag | +0.123 |
Augusta | +0.118 |
KB | +0.116 |
Neg logit contributions | |
breaching | -0.187 |
etheus | -0.172 |
Eid | -0.170 |
plets | -0.168 |
hump | -0.164 |
Token |
|
Token ID | 198 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
Balanced | +0.237 |
Dominican | +0.213 |
Nicarag | +0.213 |
Augusta | +0.204 |
KB | +0.202 |
Neg logit contributions | |
breaching | -0.328 |
etheus | -0.302 |
Eid | -0.298 |
plets | -0.296 |
hump | -0.288 |
Token | June |
Token ID | 2795 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Balanced | +0.463 |
Dominican | +0.416 |
Nicarag | +0.414 |
Augusta | +0.399 |
KB | +0.391 |
Neg logit contributions | |
breaching | -0.676 |
etheus | -0.625 |
Eid | -0.615 |
plets | -0.613 |
hump | -0.595 |
Token | 2009 |
Token ID | 3717 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
Balanced | +0.711 |
Dominican | +0.635 |
Nicarag | +0.633 |
Augusta | +0.607 |
KB | +0.596 |
Neg logit contributions | |
breaching | -1.108 |
etheus | -1.026 |
Eid | -1.014 |
plets | -1.007 |
hump | -0.982 |
Token | Germany |
Token ID | 4486 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
Balanced | +0.336 |
Dominican | +0.296 |
Nicarag | +0.295 |
Augusta | +0.281 |
KB | +0.276 |
Neg logit contributions | |
breaching | -0.619 |
etheus | -0.575 |
Eid | -0.569 |
plets | -0.565 |
hump | -0.552 |
Token | 247 |
Token ID | 30179 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
Balanced | +0.588 |
Dominican | +0.525 |
Nicarag | +0.522 |
Augusta | +0.502 |
KB | +0.490 |
Neg logit contributions | |
breaching | -0.934 |
etheus | -0.867 |
plets | -0.852 |
Eid | -0.850 |
hump | -0.826 |
Token | 34 |
Token ID | 2682 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
Balanced | +0.502 |
Dominican | +0.448 |
Nicarag | +0.446 |
Augusta | +0.429 |
KB | +0.425 |
Neg logit contributions | |
breaching | -0.757 |
etheus | -0.699 |
Eid | -0.691 |
plets | -0.686 |
hump | -0.669 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Balanced | +0.147 |
Nicarag | +0.130 |
Dominican | +0.129 |
Augusta | +0.122 |
KB | +0.120 |
Neg logit contributions | |
breaching | -0.281 |
Eid | -0.264 |
etheus | -0.261 |
plets | -0.257 |
hump | -0.253 |
Token | # |
Token ID | 1303 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
Balanced | +1.622 |
Dominican | +1.415 |
Nicarag | +1.409 |
Augusta | +1.343 |
KB | +1.311 |
Neg logit contributions | |
breaching | -3.358 |
etheus | -3.128 |
Eid | -3.101 |
plets | -3.081 |
hump | -3.007 |
Token | 20 |
Token ID | 1238 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Balanced | +1.425 |
KB | +1.287 |
Nicarag | +1.280 |
Dominican | +1.279 |
Augusta | +1.251 |
Neg logit contributions | |
breaching | -1.629 |
etheus | -1.527 |
Eid | -1.460 |
plets | -1.456 |
pestic | -1.441 |
Token | Nice |
Token ID | 18460 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
Balanced | +0.259 |
Dominican | +0.233 |
Nicarag | +0.232 |
Augusta | +0.224 |
KB | +0.221 |
Neg logit contributions | |
breaching | -0.352 |
etheus | -0.325 |
Eid | -0.320 |
plets | -0.318 |
hump | -0.309 |
Token | , |
Token ID | 11 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
Balanced | +0.290 |
Nicarag | +0.263 |
Dominican | +0.262 |
Augusta | +0.255 |
KB | +0.253 |
Neg logit contributions | |
breaching | -0.307 |
etheus | -0.282 |
Eid | -0.277 |
plets | -0.276 |
hump | -0.264 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.82e-03 |
Pos logit contributions | |
Balanced | +0.024 |
Nicarag | +0.022 |
Dominican | +0.022 |
Augusta | +0.021 |
KB | +0.021 |
Neg logit contributions | |
breaching | -0.027 |
etheus | -0.025 |
Eid | -0.025 |
plets | -0.025 |
hump | -0.024 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
Balanced | +0.637 |
Dominican | +0.571 |
Nicarag | +0.569 |
Augusta | +0.546 |
KB | +0.536 |
Neg logit contributions | |
breaching | -0.950 |
etheus | -0.877 |
Eid | -0.868 |
plets | -0.861 |
hump | -0.839 |
Token | United |
Token ID | 1578 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
Balanced | +0.598 |
Dominican | +0.531 |
Nicarag | +0.527 |
Augusta | +0.503 |
KB | +0.492 |
Neg logit contributions | |
breaching | -1.086 |
etheus | -1.009 |
Eid | -0.996 |
plets | -0.992 |
hump | -0.966 |
Token | States |
Token ID | 1829 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
Balanced | +0.110 |
Nicarag | +0.097 |
Dominican | +0.096 |
Augusta | +0.092 |
KB | +0.091 |
Neg logit contributions | |
breaching | -0.190 |
etheus | -0.176 |
Eid | -0.175 |
plets | -0.173 |
hump | -0.169 |
Token | 6 |
Token ID | 718 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
Balanced | +0.088 |
Dominican | +0.078 |
Nicarag | +0.078 |
Augusta | +0.075 |
KB | +0.073 |
Neg logit contributions | |
breaching | -0.141 |
etheus | -0.131 |
Eid | -0.129 |
plets | -0.129 |
hump | -0.125 |
Token | 12 |
Token ID | 1065 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Balanced | +0.344 |
Dominican | +0.311 |
Nicarag | +0.309 |
Augusta | +0.299 |
KB | +0.296 |
Neg logit contributions | |
breaching | -0.439 |
etheus | -0.403 |
Eid | -0.396 |
plets | -0.395 |
hump | -0.384 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
breaching | +0.331 |
etheus | +0.307 |
Eid | +0.307 |
plets | +0.301 |
hump | +0.296 |
Neg logit contributions | |
Balanced | -0.172 |
Nicarag | -0.151 |
Dominican | -0.151 |
Augusta | -0.143 |
KB | -0.140 |
Token | # |
Token ID | 1303 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
Balanced | +1.612 |
Dominican | +1.407 |
Nicarag | +1.400 |
Augusta | +1.334 |
KB | +1.302 |
Neg logit contributions | |
breaching | -3.334 |
etheus | -3.103 |
Eid | -3.078 |
plets | -3.056 |
hump | -2.984 |
Token | 17 |
Token ID | 1558 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Balanced | +1.322 |
Dominican | +1.194 |
Nicarag | +1.180 |
KB | +1.170 |
Augusta | +1.152 |
Neg logit contributions | |
breaching | -1.635 |
etheus | -1.538 |
plets | -1.488 |
Eid | -1.482 |
pestic | -1.450 |
Token | Good |
Token ID | 4599 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
Balanced | +0.169 |
Dominican | +0.151 |
Nicarag | +0.151 |
Augusta | +0.145 |
KB | +0.143 |
Neg logit contributions | |
breaching | -0.247 |
etheus | -0.228 |
Eid | -0.225 |
plets | -0.224 |
hump | -0.218 |
Token | to |
Token ID | 284 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
Balanced | +0.605 |
Dominican | +0.546 |
Nicarag | +0.546 |
Augusta | +0.534 |
KB | +0.530 |
Neg logit contributions | |
breaching | -0.627 |
etheus | -0.572 |
plets | -0.561 |
Eid | -0.560 |
hump | -0.538 |
Token | hear |
Token ID | 3285 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
Balanced | +0.235 |
Dominican | +0.213 |
Nicarag | +0.212 |
Augusta | +0.206 |
KB | +0.204 |
Neg logit contributions | |
breaching | -0.253 |
etheus | -0.232 |
Eid | -0.228 |
plets | -0.225 |
hump | -0.218 |
Token | to |
Token ID | 284 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
Balanced | +0.585 |
Dominican | +0.539 |
Nicarag | +0.536 |
Augusta | +0.520 |
KB | +0.513 |
Neg logit contributions | |
breaching | -0.580 |
etheus | -0.530 |
Eid | -0.524 |
plets | -0.518 |
hump | -0.502 |
Token | five |
Token ID | 1936 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
breaching | +0.146 |
etheus | +0.138 |
Eid | +0.135 |
plets | +0.134 |
horns | +0.129 |
Neg logit contributions | |
Balanced | -0.152 |
Nicarag | -0.141 |
Dominican | -0.141 |
Augusta | -0.136 |
Container | -0.134 |
Token | war |
Token ID | 1175 |
Feature activation | -7.52e-02 |
Pos logit contributions | |
Balanced | +0.250 |
Nicarag | +0.228 |
Dominican | +0.228 |
Augusta | +0.218 |
KB | +0.215 |
Neg logit contributions | |
breaching | -0.291 |
etheus | -0.271 |
Eid | -0.268 |
plets | -0.265 |
hump | -0.256 |
Token | - |
Token ID | 12 |
Feature activation | +4.24e-04 |
Pos logit contributions | |
breaching | +0.147 |
etheus | +0.135 |
Eid | +0.133 |
plets | +0.131 |
hump | +0.127 |
Neg logit contributions | |
Balanced | -0.148 |
Dominican | -0.136 |
Nicarag | -0.136 |
Augusta | -0.131 |
KB | -0.130 |
Token | cr |
Token ID | 6098 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Balanced | +0.001 |
Dominican | +0.001 |
Nicarag | +0.001 |
Augusta | +0.001 |
KB | +0.001 |
Neg logit contributions | |
breaching | -0.001 |
etheus | -0.001 |
Eid | -0.001 |
plets | -0.001 |
hump | -0.001 |
Token | imes |
Token ID | 999 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Balanced | +0.269 |
Nicarag | +0.251 |
Dominican | +0.251 |
Augusta | +0.244 |
KB | +0.243 |
Neg logit contributions | |
breaching | -0.148 |
etheus | -0.127 |
Eid | -0.126 |
plets | -0.121 |
hump | -0.118 |
Token | charges |
Token ID | 4530 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
Balanced | +2.124 |
Dominican | +1.948 |
Nicarag | +1.946 |
Augusta | +1.862 |
Container | +1.836 |
Neg logit contributions | |
breaching | -2.550 |
etheus | -2.481 |
Eid | -2.420 |
plets | -2.393 |
unicorn | -2.309 |
Token | , |
Token ID | 11 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
Balanced | +0.725 |
Dominican | +0.662 |
Nicarag | +0.658 |
Augusta | +0.636 |
KB | +0.624 |
Neg logit contributions | |
breaching | -0.925 |
etheus | -0.870 |
Eid | -0.852 |
plets | -0.845 |
horns | -0.821 |
Token | including |
Token ID | 1390 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
Balanced | +0.021 |
Dominican | +0.019 |
Nicarag | +0.019 |
Augusta | +0.018 |
KB | +0.018 |
Neg logit contributions | |
breaching | -0.027 |
etheus | -0.025 |
Eid | -0.025 |
plets | -0.024 |
hump | -0.024 |
Token | murder |
Token ID | 5123 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
Balanced | +0.583 |
Nicarag | +0.548 |
Dominican | +0.547 |
Augusta | +0.529 |
Container | +0.519 |
Neg logit contributions | |
breaching | -0.492 |
etheus | -0.472 |
plets | -0.461 |
Eid | -0.457 |
hump | -0.433 |
Token | , |
Token ID | 11 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
breaching | +0.151 |
etheus | +0.140 |
Eid | +0.138 |
plets | +0.137 |
hump | +0.133 |
Neg logit contributions | |
Balanced | -0.110 |
Dominican | -0.100 |
Nicarag | -0.100 |
Augusta | -0.096 |
KB | -0.094 |
Token | / |
Token ID | 14 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
Balanced | +0.520 |
Dominican | +0.472 |
Nicarag | +0.472 |
Augusta | +0.455 |
KB | +0.450 |
Neg logit contributions | |
breaching | -0.608 |
etheus | -0.556 |
Eid | -0.551 |
plets | -0.543 |
hump | -0.530 |
Token | Getty |
Token ID | 6633 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
Balanced | +0.125 |
Dominican | +0.115 |
Nicarag | +0.115 |
Augusta | +0.111 |
KB | +0.110 |
Neg logit contributions | |
breaching | -0.109 |
etheus | -0.098 |
Eid | -0.098 |
plets | -0.096 |
hump | -0.093 |
Token | Images |
Token ID | 5382 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
Balanced | +0.226 |
Dominican | +0.198 |
Nicarag | +0.198 |
Augusta | +0.188 |
KB | +0.184 |
Neg logit contributions | |
breaching | -0.445 |
etheus | -0.414 |
Eid | -0.411 |
plets | -0.406 |
hump | -0.398 |
Token | ) |
Token ID | 8 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
breaching | +0.141 |
etheus | +0.127 |
Eid | +0.126 |
plets | +0.124 |
horns | +0.121 |
Neg logit contributions | |
Balanced | -0.139 |
Nicarag | -0.126 |
Dominican | -0.126 |
Augusta | -0.123 |
KB | -0.121 |
Token | Story |
Token ID | 8362 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Balanced | +0.494 |
Dominican | +0.444 |
Nicarag | +0.442 |
Augusta | +0.424 |
KB | +0.418 |
Neg logit contributions | |
breaching | -0.706 |
etheus | -0.653 |
Eid | -0.643 |
plets | -0.641 |
hump | -0.621 |
Token | Highlights |
Token ID | 30776 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Balanced | +0.169 |
Dominican | +0.142 |
Nicarag | +0.139 |
Augusta | +0.134 |
KB | +0.130 |
Neg logit contributions | |
breaching | -0.433 |
etheus | -0.406 |
Eid | -0.402 |
plets | -0.400 |
hump | -0.392 |
Token | The |
Token ID | 383 |
Feature activation | +2.79e-02 |
Pos logit contributions | |
Balanced | +2.063 |
Dominican | +1.835 |
Nicarag | +1.828 |
Augusta | +1.765 |
KB | +1.725 |
Neg logit contributions | |
breaching | -2.677 |
etheus | -2.492 |
plets | -2.458 |
Eid | -2.390 |
hump | -2.350 |
Token | United |
Token ID | 1578 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
Balanced | +0.051 |
Dominican | +0.046 |
Nicarag | +0.046 |
Augusta | +0.044 |
KB | +0.043 |
Neg logit contributions | |
breaching | -0.060 |
etheus | -0.055 |
Eid | -0.054 |
plets | -0.054 |
hump | -0.052 |
Token | States |
Token ID | 1829 |
Feature activation | -9.81e-02 |
Pos logit contributions | |
Balanced | +0.235 |
Dominican | +0.206 |
Nicarag | +0.204 |
Augusta | +0.194 |
KB | +0.190 |
Neg logit contributions | |
breaching | -0.456 |
etheus | -0.425 |
Eid | -0.420 |
plets | -0.418 |
hump | -0.408 |
Token | will |
Token ID | 481 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
breaching | +0.209 |
etheus | +0.191 |
Eid | +0.190 |
plets | +0.187 |
hump | +0.182 |
Neg logit contributions | |
Balanced | -0.178 |
Dominican | -0.161 |
Nicarag | -0.161 |
Augusta | -0.155 |
KB | -0.154 |
Token | run |
Token ID | 1057 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
Balanced | +0.118 |
Dominican | +0.107 |
Nicarag | +0.107 |
Augusta | +0.103 |
KB | +0.102 |
Neg logit contributions | |
breaching | -0.138 |
etheus | -0.127 |
Eid | -0.125 |
plets | -0.124 |
hump | -0.120 |
Token | rigorous |
Token ID | 22888 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
breaching | +0.062 |
etheus | +0.057 |
Eid | +0.056 |
plets | +0.056 |
hump | +0.054 |
Neg logit contributions | |
Balanced | -0.063 |
Nicarag | -0.057 |
Dominican | -0.057 |
Augusta | -0.055 |
KB | -0.055 |
Token | testing |
Token ID | 4856 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
Balanced | +0.721 |
Nicarag | +0.656 |
Dominican | +0.655 |
Augusta | +0.635 |
KB | +0.624 |
Neg logit contributions | |
breaching | -0.681 |
etheus | -0.624 |
plets | -0.614 |
Eid | -0.609 |
hump | -0.585 |
Token | to |
Token ID | 284 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
Balanced | +0.259 |
Dominican | +0.233 |
Nicarag | +0.233 |
Augusta | +0.224 |
KB | +0.221 |
Neg logit contributions | |
breaching | -0.329 |
etheus | -0.303 |
Eid | -0.300 |
plets | -0.297 |
hump | -0.289 |
Token | meet |
Token ID | 1826 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
Balanced | +0.136 |
Dominican | +0.123 |
Nicarag | +0.123 |
Augusta | +0.119 |
KB | +0.118 |
Neg logit contributions | |
breaching | -0.163 |
etheus | -0.149 |
Eid | -0.147 |
plets | -0.145 |
hump | -0.142 |
Token | MIL |
Token ID | 31515 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
Balanced | +0.226 |
Dominican | +0.205 |
Nicarag | +0.205 |
Augusta | +0.198 |
KB | +0.196 |
Neg logit contributions | |
breaching | -0.243 |
etheus | -0.222 |
Eid | -0.219 |
plets | -0.217 |
hump | -0.211 |
Token | STD |
Token ID | 32147 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Balanced | +0.671 |
Nicarag | +0.597 |
Dominican | +0.593 |
Augusta | +0.569 |
KB | +0.565 |
Neg logit contributions | |
breaching | -1.062 |
etheus | -0.990 |
Eid | -0.977 |
plets | -0.969 |
hump | -0.943 |
Token | 810 |
Token ID | 40215 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
Balanced | +1.776 |
Nicarag | +1.531 |
Dominican | +1.520 |
Augusta | +1.467 |
KB | +1.444 |
Neg logit contributions | |
breaching | -2.908 |
etheus | -2.713 |
plets | -2.694 |
Eid | -2.689 |
hump | -2.575 |
Token | G |
Token ID | 38 |
Feature activation | -9.56e-02 |
Pos logit contributions | |
Balanced | +0.290 |
Dominican | +0.261 |
Nicarag | +0.260 |
Augusta | +0.250 |
KB | +0.248 |
Neg logit contributions | |
breaching | -0.383 |
etheus | -0.353 |
Eid | -0.349 |
plets | -0.347 |
hump | -0.337 |
Token | standards |
Token ID | 5423 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
breaching | +0.206 |
etheus | +0.189 |
Eid | +0.187 |
plets | +0.185 |
hump | +0.180 |
Neg logit contributions | |
Balanced | -0.172 |
Nicarag | -0.155 |
Dominican | -0.155 |
Augusta | -0.149 |
KB | -0.148 |
Token | , |
Token ID | 11 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
Balanced | +0.456 |
Dominican | +0.410 |
Nicarag | +0.409 |
Augusta | +0.393 |
KB | +0.389 |
Neg logit contributions | |
breaching | -0.592 |
etheus | -0.546 |
Eid | -0.540 |
plets | -0.535 |
hump | -0.520 |
Token | guaranteeing |
Token ID | 50156 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
breaching | +0.037 |
etheus | +0.034 |
Eid | +0.034 |
plets | +0.033 |
hump | +0.032 |
Neg logit contributions | |
Balanced | -0.032 |
Dominican | -0.029 |
Nicarag | -0.029 |
Augusta | -0.028 |
KB | -0.028 |
Token | Guard |
Token ID | 4932 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
Balanced | +0.021 |
Dominican | +0.019 |
Nicarag | +0.019 |
Augusta | +0.018 |
KB | +0.018 |
Neg logit contributions | |
breaching | -0.032 |
etheus | -0.030 |
Eid | -0.029 |
plets | -0.029 |
hump | -0.029 |
Token | { |
Token ID | 1391 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
Balanced | +0.143 |
Dominican | +0.127 |
Nicarag | +0.127 |
Augusta | +0.121 |
KB | +0.119 |
Neg logit contributions | |
breaching | -0.247 |
etheus | -0.229 |
Eid | -0.227 |
plets | -0.224 |
hump | -0.219 |
Token | ... |
Token ID | 2644 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Balanced | +0.046 |
Nicarag | +0.041 |
Dominican | +0.041 |
Augusta | +0.039 |
KB | +0.039 |
Neg logit contributions | |
breaching | -0.065 |
etheus | -0.060 |
Eid | -0.059 |
plets | -0.058 |
horns | -0.057 |
Token | } |
Token ID | 1782 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
breaching | +0.353 |
etheus | +0.326 |
Eid | +0.324 |
plets | +0.319 |
horns | +0.313 |
Neg logit contributions | |
Balanced | -0.218 |
Nicarag | -0.196 |
Dominican | -0.195 |
Augusta | -0.186 |
KB | -0.183 |
Token | pub |
Token ID | 2240 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
breaching | +0.669 |
etheus | +0.611 |
Eid | +0.608 |
plets | +0.594 |
horns | +0.590 |
Neg logit contributions | |
Balanced | -0.511 |
Nicarag | -0.466 |
Dominican | -0.463 |
Augusta | -0.445 |
KB | -0.441 |
Token | fn |
Token ID | 24714 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Balanced | +0.795 |
Dominican | +0.717 |
Nicarag | +0.717 |
Augusta | +0.690 |
KB | +0.682 |
Neg logit contributions | |
breaching | -1.024 |
etheus | -0.940 |
Eid | -0.932 |
plets | -0.920 |
hump | -0.897 |
Token | pin |
Token ID | 6757 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
Balanced | +2.126 |
Dominican | +1.903 |
Nicarag | +1.901 |
Augusta | +1.871 |
Container | +1.841 |
Neg logit contributions | |
breaching | -2.492 |
etheus | -2.296 |
plets | -2.287 |
Eid | -2.256 |
hump | -2.171 |
Token | () |
Token ID | 7499 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
breaching | +0.045 |
etheus | +0.041 |
Eid | +0.040 |
plets | +0.040 |
hump | +0.039 |
Neg logit contributions | |
Balanced | -0.041 |
Dominican | -0.037 |
Nicarag | -0.037 |
Augusta | -0.036 |
KB | -0.036 |
Token | -> |
Token ID | 4613 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
Balanced | +0.705 |
Nicarag | +0.632 |
Dominican | +0.628 |
Augusta | +0.596 |
KB | +0.589 |
Neg logit contributions | |
breaching | -1.160 |
etheus | -1.067 |
Eid | -1.063 |
plets | -1.052 |
hump | -1.033 |
Token | Guard |
Token ID | 4932 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
Balanced | +0.984 |
Dominican | +0.868 |
Nicarag | +0.865 |
Augusta | +0.840 |
Container | +0.826 |
Neg logit contributions | |
breaching | -1.493 |
etheus | -1.384 |
plets | -1.358 |
Eid | -1.356 |
hump | -1.316 |
Token | ; |
Token ID | 2162 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
breaching | +0.078 |
etheus | +0.073 |
Eid | +0.072 |
plets | +0.071 |
hump | +0.070 |
Neg logit contributions | |
Balanced | -0.044 |
Dominican | -0.039 |
Nicarag | -0.039 |
Augusta | -0.037 |
KB | -0.037 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
Balanced | +0.267 |
Nicarag | +0.239 |
Dominican | +0.239 |
Augusta | +0.229 |
KB | +0.227 |
Neg logit contributions | |
breaching | -0.392 |
etheus | -0.361 |
Eid | -0.359 |
plets | -0.354 |
hump | -0.346 |
Token | United |
Token ID | 1578 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
Balanced | +0.440 |
Dominican | +0.389 |
Nicarag | +0.388 |
Augusta | +0.370 |
KB | +0.363 |
Neg logit contributions | |
breaching | -0.791 |
etheus | -0.734 |
Eid | -0.727 |
plets | -0.721 |
hump | -0.704 |
Token | States |
Token ID | 1829 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Balanced | +0.108 |
Dominican | +0.096 |
Nicarag | +0.095 |
Augusta | +0.091 |
KB | +0.089 |
Neg logit contributions | |
breaching | -0.201 |
etheus | -0.186 |
Eid | -0.184 |
plets | -0.183 |
hump | -0.179 |
Token | 6 |
Token ID | 718 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
Balanced | +0.155 |
Dominican | +0.138 |
Nicarag | +0.138 |
Augusta | +0.132 |
KB | +0.129 |
Neg logit contributions | |
breaching | -0.257 |
etheus | -0.239 |
Eid | -0.236 |
plets | -0.235 |
hump | -0.228 |
Token | 83 |
Token ID | 5999 |
Feature activation | -9.90e-02 |
Pos logit contributions | |
Balanced | +0.241 |
Dominican | +0.218 |
Nicarag | +0.217 |
Augusta | +0.210 |
KB | +0.208 |
Neg logit contributions | |
breaching | -0.304 |
etheus | -0.279 |
Eid | -0.274 |
plets | -0.273 |
hump | -0.265 |
Token | Posts |
Token ID | 12043 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
breaching | +0.255 |
Eid | +0.239 |
etheus | +0.237 |
plets | +0.233 |
hump | +0.230 |
Neg logit contributions | |
Balanced | -0.131 |
Nicarag | -0.117 |
Dominican | -0.116 |
Augusta | -0.110 |
KB | -0.107 |
Token | Last |
Token ID | 4586 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
Balanced | +1.514 |
Dominican | +1.322 |
Nicarag | +1.316 |
Augusta | +1.254 |
KB | +1.226 |
Neg logit contributions | |
breaching | -3.103 |
etheus | -2.887 |
Eid | -2.864 |
plets | -2.843 |
hump | -2.778 |
Token | Edited |
Token ID | 34212 |
Feature activation | +7.41e-02 |
Pos logit contributions | |
Balanced | +0.345 |
Nicarag | +0.292 |
Dominican | +0.289 |
Augusta | +0.265 |
KB | +0.255 |
Neg logit contributions | |
breaching | -1.067 |
Eid | -1.003 |
etheus | -0.996 |
plets | -0.977 |
hump | -0.972 |
Token | : |
Token ID | 25 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
Balanced | +0.131 |
Nicarag | +0.118 |
Dominican | +0.117 |
Augusta | +0.113 |
KB | +0.111 |
Neg logit contributions | |
breaching | -0.162 |
Eid | -0.147 |
etheus | -0.147 |
plets | -0.144 |
hump | -0.142 |
Token | 2013 |
Token ID | 2211 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
Balanced | +0.304 |
Dominican | +0.272 |
Nicarag | +0.272 |
Augusta | +0.261 |
KB | +0.257 |
Neg logit contributions | |
breaching | -0.459 |
etheus | -0.424 |
Eid | -0.420 |
plets | -0.416 |
hump | -0.406 |
Token | - |
Token ID | 12 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Balanced | +0.344 |
Nicarag | +0.309 |
Dominican | +0.308 |
Augusta | +0.294 |
KB | +0.291 |
Neg logit contributions | |
breaching | -0.510 |
Eid | -0.469 |
etheus | -0.468 |
plets | -0.458 |
hump | -0.451 |
Token | Josh |
Token ID | 8518 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
breaching | +0.568 |
etheus | +0.517 |
Eid | +0.503 |
plets | +0.502 |
hump | +0.483 |
Neg logit contributions | |
Balanced | -0.663 |
Dominican | -0.619 |
Nicarag | -0.614 |
Augusta | -0.597 |
KB | -0.588 |
Token | a |
Token ID | 257 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
breaching | +0.456 |
etheus | +0.417 |
Eid | +0.408 |
plets | +0.407 |
hump | +0.393 |
Neg logit contributions | |
Balanced | -0.454 |
Dominican | -0.421 |
Nicarag | -0.418 |
Augusta | -0.404 |
KB | -0.398 |
Token | � |
Token ID | 564 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
breaching | +0.137 |
etheus | +0.126 |
Eid | +0.123 |
plets | +0.122 |
hump | +0.117 |
Neg logit contributions | |
Balanced | -0.141 |
Dominican | -0.130 |
Nicarag | -0.129 |
Augusta | -0.125 |
KB | -0.123 |
Token | � |
Token ID | 250 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
Balanced | +0.098 |
Dominican | +0.089 |
Nicarag | +0.088 |
Augusta | +0.086 |
KB | +0.085 |
Neg logit contributions | |
breaching | -0.105 |
etheus | -0.096 |
Eid | -0.095 |
plets | -0.093 |
hump | -0.091 |
Token | st |
Token ID | 301 |
Feature activation | -1.21e-03 |
Pos logit contributions | |
breaching | +0.060 |
etheus | +0.055 |
Eid | +0.055 |
plets | +0.054 |
hump | +0.053 |
Neg logit contributions | |
Balanced | -0.053 |
Dominican | -0.048 |
Nicarag | -0.048 |
Augusta | -0.046 |
KB | -0.046 |
Token | ern |
Token ID | 1142 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
breaching | +0.002 |
etheus | +0.002 |
Eid | +0.002 |
plets | +0.002 |
hump | +0.002 |
Neg logit contributions | |
Balanced | -0.002 |
Dominican | -0.002 |
Nicarag | -0.002 |
Augusta | -0.002 |
KB | -0.002 |
Token | lecture |
Token ID | 19143 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
Balanced | +2.292 |
Dominican | +2.163 |
Nicarag | +2.118 |
Augusta | +2.049 |
Container | +2.042 |
Neg logit contributions | |
breaching | -2.232 |
etheus | -2.048 |
plets | -2.022 |
Eid | -2.000 |
hump | -1.911 |
Token | , |
Token ID | 11 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
Balanced | +0.296 |
Dominican | +0.269 |
Nicarag | +0.268 |
Augusta | +0.256 |
KB | +0.255 |
Neg logit contributions | |
breaching | -0.380 |
etheus | -0.352 |
Eid | -0.346 |
plets | -0.343 |
hump | -0.334 |
Token | � |
Token ID | 447 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
breaching | +0.073 |
etheus | +0.068 |
Eid | +0.066 |
plets | +0.066 |
hump | +0.064 |
Neg logit contributions | |
Balanced | -0.063 |
Dominican | -0.058 |
Nicarag | -0.058 |
Augusta | -0.056 |
KB | -0.055 |
Token | � |
Token ID | 251 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
Balanced | +0.035 |
Dominican | +0.032 |
Nicarag | +0.032 |
Augusta | +0.031 |
KB | +0.030 |
Neg logit contributions | |
breaching | -0.038 |
etheus | -0.035 |
Eid | -0.034 |
plets | -0.034 |
hump | -0.033 |
Token | but |
Token ID | 475 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
breaching | +0.247 |
etheus | +0.228 |
Eid | +0.224 |
plets | +0.222 |
hump | +0.216 |
Neg logit contributions | |
Balanced | -0.210 |
Dominican | -0.191 |
Nicarag | -0.190 |
Augusta | -0.184 |
KB | -0.181 |
Token | < |
Token ID | 1279 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
Balanced | +0.515 |
Dominican | +0.470 |
Nicarag | +0.470 |
Augusta | +0.453 |
KB | +0.448 |
Neg logit contributions | |
breaching | -0.566 |
etheus | -0.515 |
Eid | -0.511 |
plets | -0.504 |
hump | -0.490 |
Token | T |
Token ID | 309 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
breaching | +0.440 |
etheus | +0.397 |
Eid | +0.396 |
plets | +0.388 |
horns | +0.378 |
Neg logit contributions | |
Balanced | -0.444 |
Nicarag | -0.410 |
Dominican | -0.409 |
Augusta | -0.395 |
KB | -0.392 |
Token | > |
Token ID | 1875 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
Balanced | +0.215 |
Dominican | +0.192 |
Nicarag | +0.191 |
Augusta | +0.184 |
KB | +0.182 |
Neg logit contributions | |
breaching | -0.321 |
etheus | -0.296 |
Eid | -0.293 |
plets | -0.290 |
hump | -0.283 |
Token | { |
Token ID | 1391 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
breaching | +0.377 |
etheus | +0.345 |
Eid | +0.344 |
plets | +0.339 |
horns | +0.334 |
Neg logit contributions | |
Balanced | -0.237 |
Dominican | -0.216 |
Nicarag | -0.216 |
Augusta | -0.205 |
KB | -0.202 |
Token | pub |
Token ID | 2240 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
breaching | +0.591 |
Eid | +0.543 |
etheus | +0.540 |
plets | +0.531 |
horns | +0.529 |
Neg logit contributions | |
Balanced | -0.355 |
Nicarag | -0.326 |
Dominican | -0.326 |
Augusta | -0.308 |
KB | -0.306 |
Token | fn |
Token ID | 24714 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Balanced | +0.839 |
Dominican | +0.747 |
Nicarag | +0.743 |
Augusta | +0.717 |
KB | +0.703 |
Neg logit contributions | |
breaching | -1.257 |
etheus | -1.161 |
Eid | -1.150 |
plets | -1.142 |
hump | -1.110 |
Token | new |
Token ID | 649 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
Balanced | +2.089 |
Nicarag | +1.875 |
Dominican | +1.874 |
Augusta | +1.828 |
Container | +1.798 |
Neg logit contributions | |
breaching | -2.474 |
etheus | -2.289 |
Eid | -2.250 |
plets | -2.249 |
hump | -2.152 |
Token | ( |
Token ID | 357 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
Balanced | +0.505 |
Dominican | +0.459 |
Nicarag | +0.458 |
Augusta | +0.445 |
KB | +0.439 |
Neg logit contributions | |
breaching | -0.537 |
etheus | -0.489 |
Eid | -0.484 |
plets | -0.477 |
hump | -0.464 |
Token | t |
Token ID | 256 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
breaching | +0.151 |
etheus | +0.140 |
Eid | +0.139 |
plets | +0.137 |
hump | +0.134 |
Neg logit contributions | |
Balanced | -0.091 |
Dominican | -0.080 |
Nicarag | -0.080 |
Augusta | -0.077 |
KB | -0.076 |
Token | : |
Token ID | 1058 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
breaching | +0.298 |
etheus | +0.269 |
Eid | +0.264 |
plets | +0.260 |
hump | +0.253 |
Neg logit contributions | |
Balanced | -0.311 |
Dominican | -0.286 |
Nicarag | -0.284 |
Augusta | -0.277 |
Container | -0.274 |
Token | T |
Token ID | 309 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
Balanced | +0.850 |
Dominican | +0.764 |
Nicarag | +0.761 |
Augusta | +0.738 |
Container | +0.731 |
Neg logit contributions | |
breaching | -1.027 |
etheus | -0.947 |
Eid | -0.929 |
plets | -0.925 |
hump | -0.896 |
Token | ) |
Token ID | 8 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
Balanced | +0.078 |
Dominican | +0.071 |
Nicarag | +0.071 |
Augusta | +0.069 |
KB | +0.068 |
Neg logit contributions | |
breaching | -0.082 |
etheus | -0.074 |
Eid | -0.073 |
plets | -0.072 |
hump | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
Balanced | +0.052 |
Dominican | +0.047 |
Nicarag | +0.047 |
Augusta | +0.045 |
KB | +0.045 |
Neg logit contributions | |
breaching | -0.069 |
etheus | -0.063 |
Eid | -0.062 |
plets | -0.062 |
hump | -0.060 |
Token |
|
Token ID | 198 |
Feature activation | -9.10e-03 |
Pos logit contributions | |
breaching | +0.119 |
etheus | +0.109 |
Eid | +0.109 |
plets | +0.107 |
hump | +0.104 |
Neg logit contributions | |
Balanced | -0.091 |
Nicarag | -0.082 |
Dominican | -0.082 |
Augusta | -0.079 |
KB | -0.078 |
Token | Re |
Token ID | 3041 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
breaching | +0.022 |
etheus | +0.021 |
Eid | +0.020 |
plets | +0.020 |
hump | +0.020 |
Neg logit contributions | |
Balanced | -0.014 |
Nicarag | -0.012 |
Dominican | -0.012 |
KB | -0.012 |
Augusta | -0.012 |
Token | un |
Token ID | 403 |
Feature activation | +6.13e-03 |
Pos logit contributions | |
Balanced | +0.555 |
Dominican | +0.508 |
Nicarag | +0.505 |
Augusta | +0.492 |
KB | +0.487 |
Neg logit contributions | |
breaching | -0.565 |
etheus | -0.513 |
Eid | -0.509 |
plets | -0.497 |
hump | -0.485 |
Token | ite |
Token ID | 578 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
Balanced | +0.011 |
Dominican | +0.010 |
Nicarag | +0.010 |
Augusta | +0.010 |
KB | +0.010 |
Neg logit contributions | |
breaching | -0.013 |
etheus | -0.012 |
Eid | -0.012 |
plets | -0.012 |
hump | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
breaching | +3.274 |
Eid | +2.972 |
etheus | +2.958 |
plets | +2.895 |
horns | +2.842 |
Neg logit contributions | |
Balanced | -2.924 |
Dominican | -2.681 |
Nicarag | -2.667 |
KB | -2.563 |
Augusta | -2.561 |
Token | GL |
Token ID | 10188 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
Balanced | +0.078 |
Nicarag | +0.071 |
Dominican | +0.071 |
Augusta | +0.069 |
Container | +0.067 |
Neg logit contributions | |
breaching | -0.088 |
etheus | -0.082 |
Eid | -0.079 |
plets | -0.079 |
hump | -0.077 |
Token | a |
Token ID | 64 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
Balanced | +0.561 |
Dominican | +0.498 |
Nicarag | +0.498 |
Augusta | +0.481 |
KB | +0.474 |
Neg logit contributions | |
breaching | -0.916 |
etheus | -0.843 |
Eid | -0.839 |
plets | -0.830 |
hump | -0.809 |
Token | DOS |
Token ID | 35178 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
breaching | +0.242 |
etheus | +0.222 |
Eid | +0.220 |
plets | +0.217 |
hump | +0.211 |
Neg logit contributions | |
Balanced | -0.188 |
Nicarag | -0.170 |
Dominican | -0.170 |
Augusta | -0.164 |
KB | -0.162 |
Token | ( |
Token ID | 357 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
breaching | +0.890 |
etheus | +0.816 |
Eid | +0.808 |
plets | +0.798 |
hump | +0.777 |
Neg logit contributions | |
Balanced | -0.738 |
Dominican | -0.668 |
Nicarag | -0.668 |
Augusta | -0.645 |
KB | -0.636 |
Token | . |
Token ID | 13 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
breaching | +1.190 |
etheus | +1.091 |
Eid | +1.081 |
plets | +1.068 |
hump | +1.041 |
Neg logit contributions | |
Balanced | -0.940 |
Dominican | -0.850 |
Nicarag | -0.849 |
Augusta | -0.818 |
KB | -0.808 |
Token |
|
Token ID | 198 |
Feature activation | +6.10e-02 |
Pos logit contributions | |
breaching | +0.029 |
etheus | +0.027 |
Eid | +0.027 |
plets | +0.026 |
hump | +0.026 |
Neg logit contributions | |
Balanced | -0.022 |
Dominican | -0.020 |
Nicarag | -0.020 |
Augusta | -0.019 |
KB | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +8.04e-02 |
Pos logit contributions | |
Balanced | +0.101 |
Dominican | +0.091 |
Nicarag | +0.091 |
Augusta | +0.087 |
KB | +0.086 |
Neg logit contributions | |
breaching | -0.140 |
etheus | -0.129 |
Eid | -0.127 |
plets | -0.126 |
hump | -0.123 |
Token | Israel |
Token ID | 14040 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
Balanced | +0.132 |
Nicarag | +0.118 |
Dominican | +0.118 |
KB | +0.115 |
Augusta | +0.113 |
Neg logit contributions | |
breaching | -0.186 |
etheus | -0.171 |
Eid | -0.169 |
plets | -0.168 |
hump | -0.164 |
Token | se |
Token ID | 384 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
Balanced | +0.074 |
Dominican | +0.067 |
Nicarag | +0.067 |
Augusta | +0.064 |
KB | +0.063 |
Neg logit contributions | |
breaching | -0.095 |
etheus | -0.087 |
Eid | -0.087 |
plets | -0.086 |
hump | -0.084 |
Token | izes |
Token ID | 4340 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
breaching | +1.195 |
etheus | +1.090 |
Eid | +1.081 |
plets | +1.069 |
hump | +1.046 |
Neg logit contributions | |
Balanced | -0.971 |
Nicarag | -0.886 |
Dominican | -0.884 |
Augusta | -0.850 |
KB | -0.841 |
Token | every |
Token ID | 790 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
breaching | +2.940 |
Eid | +2.661 |
etheus | +2.641 |
plets | +2.586 |
hump | +2.554 |
Neg logit contributions | |
Balanced | -2.675 |
Dominican | -2.431 |
Nicarag | -2.429 |
Augusta | -2.351 |
KB | -2.337 |
Token | opportunity |
Token ID | 3663 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
Balanced | +0.485 |
Dominican | +0.448 |
Nicarag | +0.447 |
Augusta | +0.430 |
Container | +0.425 |
Neg logit contributions | |
breaching | -0.524 |
etheus | -0.490 |
plets | -0.475 |
Eid | -0.469 |
horns | -0.452 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
Balanced | +0.524 |
Dominican | +0.482 |
Nicarag | +0.481 |
Augusta | +0.466 |
KB | +0.459 |
Neg logit contributions | |
breaching | -0.532 |
etheus | -0.487 |
Eid | -0.477 |
plets | -0.475 |
hump | -0.460 |
Token | reiterate |
Token ID | 49852 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Balanced | +0.020 |
Dominican | +0.018 |
Nicarag | +0.018 |
Augusta | +0.018 |
KB | +0.018 |
Neg logit contributions | |
breaching | -0.020 |
etheus | -0.018 |
Eid | -0.018 |
plets | -0.018 |
hump | -0.017 |
Token | its |
Token ID | 663 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
breaching | +0.495 |
etheus | +0.454 |
Eid | +0.446 |
plets | +0.443 |
hump | +0.429 |
Neg logit contributions | |
Balanced | -0.451 |
Dominican | -0.412 |
Nicarag | -0.412 |
Augusta | -0.398 |
KB | -0.392 |
Token | K |
Token ID | 509 |
Feature activation | +5.30e-02 |
Pos logit contributions | |
breaching | +0.114 |
etheus | +0.105 |
Eid | +0.104 |
plets | +0.103 |
hump | +0.100 |
Neg logit contributions | |
Balanced | -0.084 |
Dominican | -0.075 |
Nicarag | -0.075 |
Augusta | -0.073 |
KB | -0.071 |
Token | ild |
Token ID | 688 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
Balanced | +0.094 |
Dominican | +0.085 |
Nicarag | +0.085 |
KB | +0.082 |
Augusta | +0.082 |
Neg logit contributions | |
breaching | -0.116 |
etheus | -0.106 |
Eid | -0.105 |
plets | -0.103 |
hump | -0.101 |
Token | on |
Token ID | 261 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
Balanced | +0.061 |
Dominican | +0.054 |
Nicarag | +0.054 |
Augusta | +0.051 |
KB | +0.050 |
Neg logit contributions | |
breaching | -0.115 |
etheus | -0.107 |
Eid | -0.106 |
plets | -0.105 |
hump | -0.103 |
Token | an |
Token ID | 272 |
Feature activation | -2.37e-03 |
Pos logit contributions | |
Balanced | +0.063 |
Dominican | +0.057 |
Nicarag | +0.057 |
Augusta | +0.054 |
KB | +0.054 |
Neg logit contributions | |
breaching | -0.083 |
etheus | -0.077 |
Eid | -0.076 |
plets | -0.075 |
hump | -0.073 |
Token | Un |
Token ID | 791 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
breaching | +0.005 |
etheus | +0.005 |
Eid | +0.005 |
plets | +0.005 |
hump | +0.005 |
Neg logit contributions | |
Balanced | -0.004 |
Dominican | -0.004 |
Nicarag | -0.004 |
Augusta | -0.003 |
KB | -0.003 |
Token | iting |
Token ID | 1780 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Balanced | +0.035 |
Dominican | +0.032 |
Nicarag | +0.032 |
Augusta | +0.031 |
KB | +0.031 |
Neg logit contributions | |
breaching | -0.032 |
etheus | -0.029 |
Eid | -0.029 |
plets | -0.028 |
hump | -0.027 |
Token | Care |
Token ID | 7276 |
Feature activation | -7.59e-02 |
Pos logit contributions | |
breaching | +2.909 |
Eid | +2.652 |
etheus | +2.615 |
plets | +2.578 |
horns | +2.534 |
Neg logit contributions | |
Balanced | -2.270 |
Nicarag | -2.086 |
Dominican | -2.070 |
Augusta | -1.992 |
KB | -1.985 |
Token | told |
Token ID | 1297 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
breaching | +0.171 |
etheus | +0.157 |
Eid | +0.156 |
plets | +0.154 |
hump | +0.150 |
Neg logit contributions | |
Balanced | -0.128 |
Dominican | -0.116 |
Nicarag | -0.115 |
Augusta | -0.111 |
KB | -0.110 |
Token | the |
Token ID | 262 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
breaching | +0.185 |
etheus | +0.172 |
Eid | +0.169 |
plets | +0.168 |
hump | +0.163 |
Neg logit contributions | |
Balanced | -0.152 |
Dominican | -0.136 |
Nicarag | -0.136 |
Augusta | -0.133 |
KB | -0.129 |
Token | ABC |
Token ID | 9738 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
Balanced | +0.027 |
Dominican | +0.024 |
Nicarag | +0.024 |
Augusta | +0.023 |
KB | +0.023 |
Neg logit contributions | |
breaching | -0.043 |
etheus | -0.040 |
Eid | -0.040 |
plets | -0.039 |
hump | -0.039 |
Token | demand |
Token ID | 3512 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Balanced | +0.668 |
Dominican | +0.604 |
Nicarag | +0.603 |
Augusta | +0.589 |
KB | +0.578 |
Neg logit contributions | |
breaching | -0.746 |
etheus | -0.689 |
Eid | -0.678 |
plets | -0.673 |
hump | -0.650 |
Token | ( |
Token ID | 357 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
breaching | +0.120 |
etheus | +0.110 |
Eid | +0.109 |
plets | +0.108 |
hump | +0.105 |
Neg logit contributions | |
Balanced | -0.099 |
Dominican | -0.089 |
Nicarag | -0.088 |
Augusta | -0.085 |
KB | -0.084 |
Token | old |
Token ID | 727 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Balanced | +0.455 |
Dominican | +0.412 |
Nicarag | +0.410 |
KB | +0.396 |
Augusta | +0.394 |
Neg logit contributions | |
breaching | -0.571 |
etheus | -0.526 |
Eid | -0.519 |
plets | -0.515 |
hump | -0.499 |
Token | ) |
Token ID | 8 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
breaching | +0.271 |
etheus | +0.247 |
Eid | +0.245 |
plets | +0.242 |
hump | +0.235 |
Neg logit contributions | |
Balanced | -0.249 |
Dominican | -0.227 |
Nicarag | -0.226 |
Augusta | -0.219 |
KB | -0.217 |
Token | website |
Token ID | 3052 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
breaching | +0.089 |
etheus | +0.082 |
Eid | +0.081 |
plets | +0.081 |
hump | +0.078 |
Neg logit contributions | |
Balanced | -0.070 |
Dominican | -0.063 |
Nicarag | -0.063 |
Augusta | -0.061 |
KB | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +7.60e-02 |
Pos logit contributions | |
Balanced | +0.272 |
Dominican | +0.245 |
Nicarag | +0.244 |
Augusta | +0.236 |
KB | +0.233 |
Neg logit contributions | |
breaching | -0.339 |
etheus | -0.311 |
Eid | -0.308 |
plets | -0.305 |
hump | -0.296 |
Token | Nevertheless |
Token ID | 15933 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
Balanced | +0.128 |
Dominican | +0.115 |
Nicarag | +0.115 |
Augusta | +0.111 |
KB | +0.109 |
Neg logit contributions | |
breaching | -0.173 |
etheus | -0.159 |
Eid | -0.156 |
plets | -0.156 |
hump | -0.152 |
Token | we |
Token ID | 356 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
breaching | +2.635 |
etheus | +2.390 |
Eid | +2.389 |
plets | +2.328 |
hump | +2.287 |
Neg logit contributions | |
Balanced | -2.309 |
Nicarag | -2.118 |
Dominican | -2.116 |
Augusta | -2.044 |
KB | -2.008 |
Token | worked |
Token ID | 3111 |
Feature activation | -5.32e-02 |
Pos logit contributions | |
breaching | +0.383 |
etheus | +0.351 |
Eid | +0.350 |
plets | +0.345 |
hump | +0.337 |
Neg logit contributions | |
Balanced | -0.275 |
Dominican | -0.248 |
Nicarag | -0.248 |
Augusta | -0.238 |
KB | -0.235 |
Token | hard |
Token ID | 1327 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
breaching | +0.107 |
etheus | +0.097 |
Eid | +0.096 |
plets | +0.095 |
hump | +0.092 |
Neg logit contributions | |
Balanced | -0.103 |
Dominican | -0.094 |
Nicarag | -0.094 |
Augusta | -0.091 |
KB | -0.090 |
Token | to |
Token ID | 284 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
Balanced | +0.418 |
Dominican | +0.381 |
Nicarag | +0.379 |
Augusta | +0.367 |
KB | +0.362 |
Neg logit contributions | |
breaching | -0.452 |
etheus | -0.414 |
Eid | -0.408 |
plets | -0.404 |
hump | -0.392 |
Token | cover |
Token ID | 3002 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
Balanced | +0.030 |
Dominican | +0.027 |
Nicarag | +0.027 |
Augusta | +0.026 |
KB | +0.026 |
Neg logit contributions | |
breaching | -0.038 |
etheus | -0.035 |
Eid | -0.034 |
plets | -0.034 |
hump | -0.033 |
Token | ac |
Token ID | 330 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
breaching | +0.741 |
etheus | +0.673 |
Eid | +0.668 |
plets | +0.658 |
hump | +0.643 |
Neg logit contributions | |
Balanced | -0.658 |
Dominican | -0.599 |
Nicarag | -0.597 |
Augusta | -0.578 |
KB | -0.571 |
Token | illus |
Token ID | 44342 |
Feature activation | +8.94e-02 |
Pos logit contributions | |
breaching | +0.262 |
etheus | +0.238 |
Eid | +0.236 |
plets | +0.230 |
hump | +0.225 |
Neg logit contributions | |
Balanced | -0.254 |
Dominican | -0.234 |
Nicarag | -0.233 |
KB | -0.226 |
Augusta | -0.225 |
Token | produces |
Token ID | 11073 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
Balanced | +0.154 |
Dominican | +0.140 |
Nicarag | +0.139 |
Augusta | +0.134 |
KB | +0.132 |
Neg logit contributions | |
breaching | -0.198 |
etheus | -0.183 |
Eid | -0.180 |
plets | -0.178 |
hump | -0.173 |
Token | l |
Token ID | 300 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
breaching | +0.070 |
etheus | +0.064 |
Eid | +0.062 |
plets | +0.062 |
hump | +0.059 |
Neg logit contributions | |
Balanced | -0.076 |
Dominican | -0.070 |
Nicarag | -0.069 |
Augusta | -0.067 |
KB | -0.067 |
Token | actic |
Token ID | 12009 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
breaching | +0.335 |
etheus | +0.302 |
Eid | +0.299 |
plets | +0.291 |
hump | +0.286 |
Neg logit contributions | |
Balanced | -0.359 |
Dominican | -0.329 |
Nicarag | -0.329 |
KB | -0.319 |
Augusta | -0.317 |
Token | acid |
Token ID | 7408 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
breaching | +0.446 |
etheus | +0.389 |
Eid | +0.383 |
plets | +0.375 |
hump | +0.359 |
Neg logit contributions | |
Balanced | -0.802 |
Dominican | -0.749 |
Nicarag | -0.748 |
Augusta | -0.730 |
KB | -0.725 |
Token | that |
Token ID | 326 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
breaching | +2.653 |
Eid | +2.396 |
etheus | +2.394 |
plets | +2.359 |
hump | +2.310 |
Neg logit contributions | |
Balanced | -2.292 |
Nicarag | -2.101 |
Dominican | -2.098 |
Augusta | -2.024 |
KB | -1.989 |
Token | reduces |
Token ID | 12850 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
breaching | +0.072 |
etheus | +0.067 |
Eid | +0.066 |
plets | +0.065 |
hump | +0.063 |
Neg logit contributions | |
Balanced | -0.057 |
Dominican | -0.051 |
Nicarag | -0.051 |
Augusta | -0.050 |
KB | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
breaching | +0.437 |
etheus | +0.397 |
Eid | +0.389 |
plets | +0.386 |
hump | +0.370 |
Neg logit contributions | |
Balanced | -0.501 |
Dominican | -0.462 |
Nicarag | -0.461 |
Augusta | -0.447 |
KB | -0.442 |
Token | pH |
Token ID | 22918 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
breaching | +0.235 |
etheus | +0.213 |
Eid | +0.209 |
plets | +0.208 |
hump | +0.198 |
Neg logit contributions | |
Balanced | -0.271 |
Dominican | -0.250 |
Nicarag | -0.249 |
Augusta | -0.242 |
KB | -0.240 |
Token | of |
Token ID | 286 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
breaching | +0.782 |
etheus | +0.729 |
Eid | +0.720 |
plets | +0.710 |
horns | +0.684 |
Neg logit contributions | |
Balanced | -0.601 |
Dominican | -0.534 |
Nicarag | -0.532 |
Augusta | -0.516 |
KB | -0.508 |
Token | of |
Token ID | 286 |
Feature activation | -8.99e-02 |
Pos logit contributions | |
breaching | +0.188 |
etheus | +0.173 |
Eid | +0.171 |
plets | +0.169 |
hump | +0.164 |
Neg logit contributions | |
Balanced | -0.149 |
Dominican | -0.135 |
Nicarag | -0.134 |
Augusta | -0.129 |
KB | -0.127 |
Token | revolt |
Token ID | 27360 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
breaching | +0.179 |
etheus | +0.163 |
Eid | +0.160 |
plets | +0.159 |
hump | +0.154 |
Neg logit contributions | |
Balanced | -0.176 |
Nicarag | -0.161 |
Dominican | -0.161 |
Augusta | -0.156 |
KB | -0.154 |
Token | than |
Token ID | 621 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
Balanced | +0.109 |
Dominican | +0.099 |
Nicarag | +0.098 |
Augusta | +0.095 |
KB | +0.093 |
Neg logit contributions | |
breaching | -0.142 |
etheus | -0.131 |
Eid | -0.130 |
plets | -0.128 |
hump | -0.125 |
Token | the |
Token ID | 262 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
Balanced | +0.499 |
Nicarag | +0.455 |
Dominican | +0.454 |
Augusta | +0.440 |
KB | +0.435 |
Neg logit contributions | |
breaching | -0.523 |
etheus | -0.480 |
Eid | -0.471 |
plets | -0.466 |
hump | -0.454 |
Token | un |
Token ID | 555 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
breaching | +0.078 |
etheus | +0.071 |
Eid | +0.070 |
plets | +0.069 |
hump | +0.067 |
Neg logit contributions | |
Balanced | -0.082 |
Nicarag | -0.075 |
Dominican | -0.075 |
Augusta | -0.073 |
KB | -0.072 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
Balanced | +0.249 |
Nicarag | +0.230 |
Dominican | +0.229 |
Augusta | +0.224 |
KB | +0.222 |
Neg logit contributions | |
breaching | -0.163 |
etheus | -0.142 |
Eid | -0.141 |
plets | -0.136 |
hump | -0.136 |
Token | tut |
Token ID | 9732 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
breaching | +2.441 |
etheus | +2.200 |
Eid | +2.194 |
plets | +2.153 |
hump | +2.103 |
Neg logit contributions | |
Balanced | -2.407 |
Dominican | -2.197 |
Nicarag | -2.194 |
Augusta | -2.130 |
KB | -2.109 |
Token | el |
Token ID | 417 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
Balanced | +0.216 |
Dominican | +0.193 |
Nicarag | +0.193 |
Augusta | +0.185 |
KB | +0.182 |
Neg logit contributions | |
breaching | -0.366 |
etheus | -0.339 |
Eid | -0.335 |
plets | -0.331 |
hump | -0.325 |
Token | age |
Token ID | 496 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
breaching | +0.078 |
etheus | +0.071 |
Eid | +0.070 |
plets | +0.069 |
hump | +0.068 |
Neg logit contributions | |
Balanced | -0.060 |
Dominican | -0.055 |
Nicarag | -0.055 |
Augusta | -0.053 |
KB | -0.052 |
Token | of |
Token ID | 286 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
breaching | +0.621 |
etheus | +0.572 |
Eid | +0.565 |
plets | +0.558 |
hump | +0.544 |
Neg logit contributions | |
Balanced | -0.486 |
Dominican | -0.439 |
Nicarag | -0.439 |
Augusta | -0.422 |
KB | -0.417 |
Token | a |
Token ID | 257 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
breaching | +0.198 |
etheus | +0.181 |
Eid | +0.178 |
plets | +0.176 |
hump | +0.171 |
Neg logit contributions | |
Balanced | -0.198 |
Nicarag | -0.181 |
Dominican | -0.180 |
Augusta | -0.175 |
KB | -0.173 |
Token | = |
Token ID | 796 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
breaching | +0.658 |
etheus | +0.603 |
Eid | +0.599 |
plets | +0.591 |
hump | +0.577 |
Neg logit contributions | |
Balanced | -0.518 |
Dominican | -0.470 |
Nicarag | -0.469 |
Augusta | -0.451 |
KB | -0.443 |
Token | im |
Token ID | 545 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
breaching | +0.038 |
etheus | +0.034 |
Eid | +0.034 |
plets | +0.033 |
hump | +0.032 |
Neg logit contributions | |
Balanced | -0.035 |
Nicarag | -0.032 |
Dominican | -0.032 |
Augusta | -0.031 |
KB | -0.031 |
Token | utils |
Token ID | 26791 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
breaching | +0.247 |
etheus | +0.223 |
Eid | +0.222 |
plets | +0.218 |
hump | +0.214 |
Neg logit contributions | |
Balanced | -0.225 |
Dominican | -0.205 |
Nicarag | -0.204 |
Augusta | -0.198 |
KB | -0.197 |
Token | . |
Token ID | 13 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
breaching | +1.401 |
etheus | +1.276 |
Eid | +1.276 |
plets | +1.252 |
hump | +1.220 |
Neg logit contributions | |
Balanced | -1.212 |
Dominican | -1.105 |
Nicarag | -1.103 |
Augusta | -1.065 |
KB | -1.042 |
Token | res |
Token ID | 411 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
Balanced | +0.314 |
Nicarag | +0.282 |
Dominican | +0.282 |
Augusta | +0.271 |
KB | +0.270 |
Neg logit contributions | |
breaching | -0.422 |
etheus | -0.387 |
Eid | -0.383 |
plets | -0.378 |
hump | -0.369 |
Token | ize |
Token ID | 1096 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
breaching | +0.103 |
etheus | +0.095 |
Eid | +0.094 |
plets | +0.093 |
hump | +0.091 |
Neg logit contributions | |
Balanced | -0.058 |
Nicarag | -0.051 |
Dominican | -0.051 |
Augusta | -0.049 |
KB | -0.049 |
Token | ( |
Token ID | 7 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
breaching | +2.497 |
etheus | +2.281 |
Eid | +2.262 |
plets | +2.236 |
hump | +2.178 |
Neg logit contributions | |
Balanced | -2.296 |
Dominican | -2.119 |
Nicarag | -2.106 |
Augusta | -2.031 |
KB | -1.978 |
Token | image |
Token ID | 9060 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
breaching | +1.142 |
etheus | +1.043 |
Eid | +1.035 |
plets | +1.020 |
hump | +0.994 |
Neg logit contributions | |
Balanced | -0.971 |
Nicarag | -0.881 |
Dominican | -0.881 |
Augusta | -0.850 |
KB | -0.841 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
breaching | +0.760 |
etheus | +0.695 |
Eid | +0.689 |
plets | +0.680 |
hump | +0.663 |
Neg logit contributions | |
Balanced | -0.632 |
Dominican | -0.573 |
Nicarag | -0.573 |
Augusta | -0.552 |
KB | -0.544 |
Token | height |
Token ID | 6001 |
Feature activation | +6.82e-02 |
Pos logit contributions | |
breaching | +0.278 |
etheus | +0.256 |
Eid | +0.253 |
plets | +0.251 |
hump | +0.244 |
Neg logit contributions | |
Balanced | -0.210 |
Dominican | -0.189 |
Nicarag | -0.189 |
Augusta | -0.182 |
KB | -0.179 |
Token | = |
Token ID | 28 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
Balanced | +0.111 |
Dominican | +0.099 |
Nicarag | +0.099 |
Augusta | +0.095 |
KB | +0.094 |
Neg logit contributions | |
breaching | -0.159 |
etheus | -0.146 |
Eid | -0.145 |
plets | -0.143 |
hump | -0.140 |
Token | on |
Token ID | 319 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
Balanced | +0.345 |
Dominican | +0.312 |
Nicarag | +0.311 |
Augusta | +0.298 |
KB | +0.296 |
Neg logit contributions | |
breaching | -0.460 |
etheus | -0.426 |
Eid | -0.420 |
plets | -0.417 |
hump | -0.404 |
Token | any |
Token ID | 597 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
breaching | +0.129 |
etheus | +0.118 |
Eid | +0.116 |
plets | +0.115 |
hump | +0.112 |
Neg logit contributions | |
Balanced | -0.119 |
Dominican | -0.109 |
Nicarag | -0.109 |
Augusta | -0.105 |
KB | -0.104 |
Token | of |
Token ID | 286 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
Balanced | +0.143 |
Dominican | +0.130 |
Nicarag | +0.130 |
Augusta | +0.125 |
KB | +0.123 |
Neg logit contributions | |
breaching | -0.169 |
etheus | -0.155 |
Eid | -0.153 |
plets | -0.152 |
hump | -0.146 |
Token | his |
Token ID | 465 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
breaching | +0.046 |
etheus | +0.041 |
Eid | +0.041 |
plets | +0.040 |
hump | +0.039 |
Neg logit contributions | |
Balanced | -0.050 |
Dominican | -0.046 |
Nicarag | -0.046 |
Augusta | -0.045 |
KB | -0.044 |
Token | dis |
Token ID | 595 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
breaching | +0.167 |
etheus | +0.154 |
Eid | +0.151 |
plets | +0.150 |
hump | +0.145 |
Neg logit contributions | |
Balanced | -0.159 |
Dominican | -0.146 |
Nicarag | -0.146 |
Augusta | -0.141 |
KB | -0.139 |
Token | semb |
Token ID | 4428 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Balanced | +0.212 |
Dominican | +0.193 |
Nicarag | +0.193 |
Augusta | +0.186 |
KB | +0.185 |
Neg logit contributions | |
breaching | -0.233 |
etheus | -0.212 |
Eid | -0.210 |
plets | -0.207 |
hump | -0.202 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
breaching | +2.710 |
etheus | +2.496 |
Eid | +2.464 |
plets | +2.449 |
hump | +2.383 |
Neg logit contributions | |
Balanced | -2.051 |
Nicarag | -1.845 |
Dominican | -1.844 |
Augusta | -1.779 |
KB | -1.746 |
Token | . |
Token ID | 13 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
breaching | +0.337 |
etheus | +0.310 |
Eid | +0.303 |
plets | +0.302 |
hump | +0.291 |
Neg logit contributions | |
Balanced | -0.317 |
Dominican | -0.291 |
Nicarag | -0.290 |
Augusta | -0.279 |
KB | -0.275 |
Token | � |
Token ID | 564 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
breaching | +0.076 |
etheus | +0.070 |
Eid | +0.069 |
plets | +0.068 |
hump | +0.067 |
Neg logit contributions | |
Balanced | -0.060 |
Dominican | -0.054 |
Nicarag | -0.054 |
Augusta | -0.052 |
KB | -0.052 |
Token | � |
Token ID | 250 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
breaching | +0.137 |
etheus | +0.125 |
Eid | +0.124 |
plets | +0.122 |
hump | +0.118 |
Neg logit contributions | |
Balanced | -0.136 |
Dominican | -0.124 |
Nicarag | -0.124 |
Augusta | -0.120 |
KB | -0.119 |
Token | Six |
Token ID | 21447 |
Feature activation | +7.51e-02 |
Pos logit contributions | |
breaching | +0.131 |
etheus | +0.121 |
Eid | +0.120 |
plets | +0.118 |
hump | +0.115 |
Neg logit contributions | |
Balanced | -0.093 |
Dominican | -0.084 |
Nicarag | -0.083 |
Augusta | -0.080 |
KB | -0.079 |
Token | gas |
Token ID | 3623 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
breaching | +0.458 |
etheus | +0.418 |
Eid | +0.412 |
plets | +0.408 |
horns | +0.392 |
Neg logit contributions | |
Balanced | -0.468 |
Dominican | -0.430 |
Nicarag | -0.430 |
Augusta | -0.415 |
Container | -0.408 |
Token | pipelines |
Token ID | 31108 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
breaching | +0.561 |
etheus | +0.513 |
Eid | +0.508 |
plets | +0.500 |
hump | +0.485 |
Neg logit contributions | |
Balanced | -0.508 |
Dominican | -0.462 |
Nicarag | -0.461 |
Augusta | -0.446 |
Container | -0.441 |
Token | and |
Token ID | 290 |
Feature activation | -9.63e-02 |
Pos logit contributions | |
Balanced | +0.245 |
Dominican | +0.222 |
Nicarag | +0.221 |
Augusta | +0.213 |
KB | +0.210 |
Neg logit contributions | |
breaching | -0.320 |
etheus | -0.299 |
Eid | -0.294 |
plets | -0.293 |
hump | -0.282 |
Token | wells |
Token ID | 22571 |
Feature activation | -4.96e-02 |
Pos logit contributions | |
breaching | +0.184 |
etheus | +0.168 |
Eid | +0.165 |
plets | +0.164 |
hump | +0.157 |
Neg logit contributions | |
Balanced | -0.196 |
Dominican | -0.180 |
Nicarag | -0.180 |
Augusta | -0.174 |
KB | -0.171 |
Token | s |
Token ID | 264 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
breaching | +0.108 |
etheus | +0.100 |
Eid | +0.099 |
plets | +0.098 |
hump | +0.095 |
Neg logit contributions | |
Balanced | -0.086 |
Dominican | -0.078 |
Nicarag | -0.078 |
Augusta | -0.075 |
KB | -0.074 |
Token | ited |
Token ID | 863 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Balanced | +0.020 |
Nicarag | +0.018 |
Dominican | +0.018 |
Augusta | +0.018 |
KB | +0.017 |
Neg logit contributions | |
breaching | -0.023 |
etheus | -0.021 |
Eid | -0.021 |
plets | -0.021 |
hump | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
breaching | +2.564 |
etheus | +2.315 |
Eid | +2.301 |
plets | +2.258 |
hump | +2.224 |
Neg logit contributions | |
Balanced | -2.175 |
Nicarag | -1.965 |
Dominican | -1.954 |
KB | -1.890 |
Augusta | -1.888 |
Token | mainland |
Token ID | 22779 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
breaching | +0.031 |
etheus | +0.029 |
Eid | +0.028 |
plets | +0.028 |
horns | +0.027 |
Neg logit contributions | |
Balanced | -0.028 |
Dominican | -0.026 |
Nicarag | -0.026 |
Augusta | -0.025 |
KB | -0.024 |
Token | communities |
Token ID | 5348 |
Feature activation | -5.19e-03 |
Pos logit contributions | |
Balanced | +0.128 |
Dominican | +0.107 |
Nicarag | +0.106 |
Augusta | +0.099 |
KB | +0.096 |
Neg logit contributions | |
breaching | -0.385 |
etheus | -0.363 |
Eid | -0.359 |
plets | -0.357 |
hump | -0.349 |
Token | that |
Token ID | 326 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
breaching | +0.011 |
etheus | +0.011 |
Eid | +0.010 |
plets | +0.010 |
hump | +0.010 |
Neg logit contributions | |
Balanced | -0.009 |
Dominican | -0.008 |
Nicarag | -0.008 |
Augusta | -0.008 |
KB | -0.008 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
Balanced | +0.214 |
Dominican | +0.195 |
Nicarag | +0.194 |
Augusta | +0.188 |
KB | +0.184 |
Neg logit contributions | |
breaching | -0.253 |
etheus | -0.233 |
Eid | -0.229 |
plets | -0.228 |
hump | -0.220 |
Token | right |
Token ID | 826 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
breaching | +0.804 |
etheus | +0.737 |
Eid | +0.728 |
plets | +0.721 |
hump | +0.700 |
Neg logit contributions | |
Balanced | -0.681 |
Dominican | -0.617 |
Nicarag | -0.616 |
Augusta | -0.595 |
KB | -0.587 |
Token | ? |
Token ID | 30 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
Balanced | +0.315 |
Dominican | +0.282 |
Nicarag | +0.282 |
Augusta | +0.272 |
KB | +0.269 |
Neg logit contributions | |
breaching | -0.443 |
etheus | -0.409 |
Eid | -0.404 |
plets | -0.402 |
hump | -0.389 |
Token |
|
Token ID | 198 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
breaching | +0.055 |
etheus | +0.051 |
Eid | +0.050 |
plets | +0.050 |
hump | +0.049 |
Neg logit contributions | |
Balanced | -0.041 |
Dominican | -0.037 |
Nicarag | -0.037 |
Augusta | -0.035 |
KB | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
breaching | +0.365 |
etheus | +0.335 |
Eid | +0.334 |
plets | +0.328 |
hump | +0.321 |
Neg logit contributions | |
Balanced | -0.278 |
Nicarag | -0.251 |
Dominican | -0.251 |
Augusta | -0.241 |
KB | -0.238 |
Token | Any |
Token ID | 7149 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
breaching | +0.188 |
etheus | +0.172 |
Eid | +0.171 |
plets | +0.168 |
hump | +0.165 |
Neg logit contributions | |
Balanced | -0.139 |
Dominican | -0.124 |
Nicarag | -0.124 |
KB | -0.121 |
Augusta | -0.119 |
Token | ways |
Token ID | 1322 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
breaching | +0.043 |
etheus | +0.039 |
Eid | +0.039 |
plets | +0.038 |
hump | +0.037 |
Neg logit contributions | |
Balanced | -0.047 |
Dominican | -0.043 |
Nicarag | -0.043 |
Augusta | -0.042 |
KB | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
breaching | +2.505 |
etheus | +2.283 |
Eid | +2.277 |
plets | +2.232 |
hump | +2.185 |
Neg logit contributions | |
Balanced | -2.118 |
Dominican | -1.925 |
Nicarag | -1.925 |
Augusta | -1.853 |
KB | -1.830 |
Token | at |
Token ID | 379 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
breaching | +0.491 |
etheus | +0.450 |
Eid | +0.444 |
plets | +0.441 |
hump | +0.428 |
Neg logit contributions | |
Balanced | -0.413 |
Dominican | -0.375 |
Nicarag | -0.374 |
Augusta | -0.361 |
KB | -0.357 |
Token | le |
Token ID | 293 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
breaching | +0.299 |
etheus | +0.274 |
Eid | +0.270 |
plets | +0.269 |
hump | +0.260 |
Neg logit contributions | |
Balanced | -0.248 |
Dominican | -0.225 |
Nicarag | -0.224 |
Augusta | -0.217 |
KB | -0.214 |
Token | ast |
Token ID | 459 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Balanced | +0.346 |
Nicarag | +0.311 |
Dominican | +0.310 |
Augusta | +0.295 |
KB | +0.292 |
Neg logit contributions | |
breaching | -0.566 |
etheus | -0.519 |
Eid | -0.514 |
plets | -0.510 |
hump | -0.499 |
Token | things |
Token ID | 1243 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
Balanced | +0.281 |
Dominican | +0.255 |
Nicarag | +0.255 |
Augusta | +0.247 |
KB | +0.244 |
Neg logit contributions | |
breaching | -0.317 |
etheus | -0.293 |
Eid | -0.286 |
plets | -0.285 |
hump | -0.275 |
Token | iao |
Token ID | 13481 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
Balanced | +0.487 |
Dominican | +0.437 |
Nicarag | +0.435 |
Augusta | +0.423 |
KB | +0.422 |
Neg logit contributions | |
breaching | -0.659 |
etheus | -0.607 |
Eid | -0.598 |
plets | -0.593 |
hump | -0.575 |
Token | 8 |
Token ID | 23 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
Balanced | +0.391 |
Nicarag | +0.357 |
Dominican | +0.357 |
Augusta | +0.345 |
KB | +0.341 |
Neg logit contributions | |
breaching | -0.452 |
etheus | -0.413 |
Eid | -0.411 |
plets | -0.405 |
hump | -0.391 |
Token | and |
Token ID | 290 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
breaching | +0.423 |
etheus | +0.390 |
Eid | +0.386 |
plets | +0.381 |
hump | +0.372 |
Neg logit contributions | |
Balanced | -0.314 |
Dominican | -0.283 |
Nicarag | -0.282 |
Augusta | -0.271 |
KB | -0.269 |
Token | D |
Token ID | 360 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
breaching | +0.621 |
etheus | +0.571 |
Eid | +0.563 |
plets | +0.558 |
hump | +0.540 |
Neg logit contributions | |
Balanced | -0.520 |
Nicarag | -0.472 |
Dominican | -0.472 |
Augusta | -0.455 |
KB | -0.448 |
Token | endi |
Token ID | 43109 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
breaching | +0.493 |
etheus | +0.449 |
Eid | +0.446 |
plets | +0.438 |
hump | +0.428 |
Neg logit contributions | |
Balanced | -0.434 |
Dominican | -0.395 |
Nicarag | -0.394 |
KB | -0.381 |
Augusta | -0.381 |
Token | … |
Token ID | 1399 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
breaching | +0.314 |
etheus | +0.289 |
Eid | +0.286 |
plets | +0.281 |
hump | +0.275 |
Neg logit contributions | |
Balanced | -0.250 |
Dominican | -0.226 |
Nicarag | -0.225 |
Augusta | -0.218 |
KB | -0.216 |
Token | Then |
Token ID | 3244 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
breaching | +2.572 |
etheus | +2.351 |
Eid | +2.341 |
plets | +2.302 |
hump | +2.250 |
Neg logit contributions | |
Balanced | -2.043 |
Nicarag | -1.850 |
Dominican | -1.849 |
Augusta | -1.778 |
KB | -1.754 |
Token | after |
Token ID | 706 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
breaching | +0.027 |
etheus | +0.025 |
Eid | +0.025 |
plets | +0.025 |
hump | +0.024 |
Neg logit contributions | |
Balanced | -0.024 |
Dominican | -0.022 |
Nicarag | -0.022 |
Augusta | -0.022 |
KB | -0.021 |
Token | breakfast |
Token ID | 12607 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
breaching | +0.370 |
etheus | +0.341 |
Eid | +0.332 |
plets | +0.332 |
hump | +0.320 |
Neg logit contributions | |
Balanced | -0.360 |
Dominican | -0.330 |
Nicarag | -0.329 |
Augusta | -0.320 |
KB | -0.313 |
Token | , |
Token ID | 11 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
breaching | +0.676 |
etheus | +0.623 |
Eid | +0.608 |
plets | +0.607 |
hump | +0.586 |
Neg logit contributions | |
Balanced | -0.560 |
Dominican | -0.508 |
Nicarag | -0.507 |
Augusta | -0.490 |
KB | -0.481 |
Token | they |
Token ID | 484 |
Feature activation | -9.43e-02 |
Pos logit contributions | |
breaching | +0.661 |
etheus | +0.608 |
Eid | +0.593 |
plets | +0.592 |
hump | +0.572 |
Neg logit contributions | |
Balanced | -0.571 |
Dominican | -0.520 |
Nicarag | -0.519 |
Augusta | -0.502 |
KB | -0.492 |
Token | the |
Token ID | 262 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
breaching | +0.256 |
etheus | +0.235 |
Eid | +0.231 |
plets | +0.229 |
hump | +0.223 |
Neg logit contributions | |
Balanced | -0.217 |
Nicarag | -0.196 |
Dominican | -0.196 |
Augusta | -0.190 |
KB | -0.187 |
Token | quarterback |
Token ID | 7880 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
Balanced | +0.039 |
Nicarag | +0.036 |
Dominican | +0.036 |
Augusta | +0.034 |
KB | +0.034 |
Neg logit contributions | |
breaching | -0.044 |
etheus | -0.040 |
plets | -0.039 |
Eid | -0.039 |
hump | -0.038 |
Token | competition |
Token ID | 5449 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
breaching | +0.457 |
etheus | +0.421 |
Eid | +0.415 |
plets | +0.411 |
hump | +0.400 |
Neg logit contributions | |
Balanced | -0.337 |
Dominican | -0.303 |
Nicarag | -0.302 |
Augusta | -0.291 |
KB | -0.287 |
Token | between |
Token ID | 1022 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
breaching | +1.166 |
etheus | +1.072 |
Eid | +1.062 |
plets | +1.050 |
hump | +1.024 |
Neg logit contributions | |
Balanced | -0.867 |
Dominican | -0.781 |
Nicarag | -0.780 |
Augusta | -0.750 |
KB | -0.740 |
Token | Trevor |
Token ID | 25389 |
Feature activation | +9.90e-02 |
Pos logit contributions | |
breaching | +0.205 |
etheus | +0.188 |
plets | +0.184 |
Eid | +0.184 |
hump | +0.179 |
Neg logit contributions | |
Balanced | -0.180 |
Nicarag | -0.163 |
Dominican | -0.163 |
Augusta | -0.157 |
KB | -0.155 |
Token | Siem |
Token ID | 45196 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Balanced | +0.203 |
Dominican | +0.186 |
Nicarag | +0.186 |
Augusta | +0.180 |
KB | +0.178 |
Neg logit contributions | |
breaching | -0.188 |
etheus | -0.170 |
Eid | -0.168 |
plets | -0.166 |
hump | -0.161 |
Token | ian |
Token ID | 666 |
Feature activation | -3.73e-03 |
Pos logit contributions | |
breaching | +2.414 |
Eid | +2.201 |
etheus | +2.194 |
plets | +2.154 |
hump | +2.092 |
Neg logit contributions | |
Balanced | -2.175 |
Nicarag | -2.014 |
Dominican | -1.994 |
Augusta | -1.935 |
KB | -1.888 |
Token | and |
Token ID | 290 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
breaching | +0.009 |
etheus | +0.008 |
Eid | +0.008 |
plets | +0.008 |
hump | +0.008 |
Neg logit contributions | |
Balanced | -0.006 |
Dominican | -0.005 |
Nicarag | -0.005 |
Augusta | -0.005 |
KB | -0.005 |
Token | Pa |
Token ID | 11243 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
breaching | +0.287 |
etheus | +0.264 |
plets | +0.259 |
Eid | +0.255 |
hump | +0.251 |
Neg logit contributions | |
Balanced | -0.256 |
Nicarag | -0.233 |
Dominican | -0.232 |
Augusta | -0.225 |
KB | -0.219 |
Token | xton |
Token ID | 22874 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
breaching | +0.290 |
etheus | +0.249 |
Eid | +0.246 |
plets | +0.240 |
hump | +0.229 |
Neg logit contributions | |
Balanced | -0.576 |
Nicarag | -0.540 |
Dominican | -0.539 |
Augusta | -0.526 |
KB | -0.522 |
Token | Lynch |
Token ID | 16093 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
breaching | +1.055 |
etheus | +0.971 |
Eid | +0.960 |
plets | +0.950 |
hump | +0.926 |
Neg logit contributions | |
Balanced | -0.789 |
Dominican | -0.710 |
Nicarag | -0.709 |
Augusta | -0.684 |
KB | -0.673 |
Token | something |
Token ID | 1223 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
breaching | +0.167 |
etheus | +0.154 |
Eid | +0.152 |
plets | +0.150 |
hump | +0.146 |
Neg logit contributions | |
Balanced | -0.145 |
Dominican | -0.131 |
Nicarag | -0.131 |
Augusta | -0.127 |
KB | -0.125 |
Token | else |
Token ID | 2073 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
Balanced | +0.142 |
Nicarag | +0.129 |
Dominican | +0.129 |
Augusta | +0.124 |
KB | +0.122 |
Neg logit contributions | |
breaching | -0.162 |
etheus | -0.150 |
Eid | -0.147 |
plets | -0.146 |
hump | -0.141 |
Token | , |
Token ID | 11 |
Feature activation | +6.37e-02 |
Pos logit contributions | |
Balanced | +0.053 |
Dominican | +0.048 |
Nicarag | +0.048 |
Augusta | +0.046 |
KB | +0.046 |
Neg logit contributions | |
breaching | -0.064 |
etheus | -0.059 |
Eid | -0.058 |
plets | -0.057 |
hump | -0.056 |
Token | like |
Token ID | 588 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
Balanced | +0.131 |
Dominican | +0.120 |
Nicarag | +0.120 |
Augusta | +0.117 |
KB | +0.115 |
Neg logit contributions | |
breaching | -0.120 |
etheus | -0.109 |
Eid | -0.107 |
plets | -0.107 |
hump | -0.102 |
Token | general |
Token ID | 2276 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
breaching | +0.419 |
etheus | +0.383 |
Eid | +0.376 |
plets | +0.372 |
hump | +0.360 |
Neg logit contributions | |
Balanced | -0.415 |
Dominican | -0.380 |
Nicarag | -0.379 |
Augusta | -0.367 |
KB | -0.362 |
Token | ize |
Token ID | 1096 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
Balanced | +0.505 |
Dominican | +0.456 |
Nicarag | +0.455 |
Augusta | +0.438 |
KB | +0.432 |
Neg logit contributions | |
breaching | -0.658 |
etheus | -0.606 |
Eid | -0.599 |
plets | -0.593 |
hump | -0.576 |
Token | the |
Token ID | 262 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
breaching | +2.363 |
etheus | +2.139 |
Eid | +2.135 |
plets | +2.094 |
hump | +2.049 |
Neg logit contributions | |
Balanced | -2.215 |
Dominican | -2.020 |
Nicarag | -2.020 |
Augusta | -1.951 |
KB | -1.936 |
Token | data |
Token ID | 1366 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
breaching | +0.095 |
etheus | +0.087 |
Eid | +0.085 |
plets | +0.085 |
hump | +0.082 |
Neg logit contributions | |
Balanced | -0.087 |
Dominican | -0.079 |
Nicarag | -0.079 |
Augusta | -0.076 |
KB | -0.075 |
Token | by |
Token ID | 416 |
Feature activation | -8.00e-02 |
Pos logit contributions | |
breaching | +0.583 |
etheus | +0.535 |
Eid | +0.528 |
plets | +0.522 |
hump | +0.507 |
Neg logit contributions | |
Balanced | -0.503 |
Dominican | -0.457 |
Nicarag | -0.456 |
Augusta | -0.441 |
KB | -0.435 |
Token | population |
Token ID | 3265 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
breaching | +0.151 |
etheus | +0.137 |
Eid | +0.135 |
plets | +0.134 |
hump | +0.129 |
Neg logit contributions | |
Balanced | -0.164 |
Dominican | -0.151 |
Nicarag | -0.151 |
Augusta | -0.146 |
KB | -0.144 |
Token | or |
Token ID | 393 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
breaching | +0.361 |
etheus | +0.330 |
Eid | +0.327 |
plets | +0.322 |
hump | +0.314 |
Neg logit contributions | |
Balanced | -0.306 |
Dominican | -0.278 |
Nicarag | -0.278 |
Augusta | -0.268 |
KB | -0.265 |
Token | K |
Token ID | 509 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
breaching | +0.036 |
etheus | +0.033 |
Eid | +0.032 |
plets | +0.032 |
hump | +0.031 |
Neg logit contributions | |
Balanced | -0.027 |
Dominican | -0.025 |
Nicarag | -0.025 |
Augusta | -0.024 |
KB | -0.023 |
Token | ild |
Token ID | 688 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
Balanced | +0.130 |
Dominican | +0.116 |
Nicarag | +0.116 |
Augusta | +0.112 |
KB | +0.110 |
Neg logit contributions | |
breaching | -0.186 |
etheus | -0.172 |
Eid | -0.170 |
plets | -0.168 |
hump | -0.164 |
Token | on |
Token ID | 261 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
Balanced | +0.316 |
Dominican | +0.278 |
Nicarag | +0.278 |
Augusta | +0.266 |
KB | +0.262 |
Neg logit contributions | |
breaching | -0.553 |
etheus | -0.512 |
Eid | -0.508 |
plets | -0.502 |
hump | -0.493 |
Token | an |
Token ID | 272 |
Feature activation | -1.68e-02 |
Pos logit contributions | |
Balanced | +0.072 |
Dominican | +0.065 |
Nicarag | +0.065 |
Augusta | +0.063 |
KB | +0.062 |
Neg logit contributions | |
breaching | -0.094 |
etheus | -0.086 |
Eid | -0.085 |
plets | -0.084 |
hump | -0.082 |
Token | Un |
Token ID | 791 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
breaching | +0.039 |
etheus | +0.036 |
Eid | +0.036 |
plets | +0.036 |
hump | +0.035 |
Neg logit contributions | |
Balanced | -0.027 |
Nicarag | -0.024 |
Dominican | -0.024 |
Augusta | -0.023 |
KB | -0.023 |
Token | iting |
Token ID | 1780 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
Balanced | +0.109 |
Dominican | +0.099 |
Nicarag | +0.099 |
Augusta | +0.096 |
KB | +0.095 |
Neg logit contributions | |
breaching | -0.110 |
etheus | -0.100 |
Eid | -0.099 |
plets | -0.098 |
hump | -0.095 |
Token | Care |
Token ID | 7276 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
breaching | +2.570 |
Eid | +2.334 |
etheus | +2.295 |
plets | +2.252 |
hump | +2.244 |
Neg logit contributions | |
Balanced | -1.935 |
Nicarag | -1.782 |
Dominican | -1.779 |
KB | -1.706 |
Augusta | -1.705 |
Token | 's |
Token ID | 338 |
Feature activation | +7.30e-03 |
Pos logit contributions | |
breaching | +0.085 |
etheus | +0.078 |
Eid | +0.077 |
plets | +0.076 |
hump | +0.074 |
Neg logit contributions | |
Balanced | -0.077 |
Dominican | -0.070 |
Nicarag | -0.069 |
Augusta | -0.067 |
KB | -0.066 |
Token | financial |
Token ID | 3176 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
Balanced | +0.013 |
Dominican | +0.011 |
Nicarag | +0.011 |
Augusta | +0.011 |
KB | +0.011 |
Neg logit contributions | |
breaching | -0.016 |
etheus | -0.015 |
Eid | -0.015 |
plets | -0.015 |
hump | -0.014 |
Token | counselling |
Token ID | 47586 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
breaching | +0.077 |
etheus | +0.070 |
Eid | +0.069 |
plets | +0.069 |
hump | +0.066 |
Neg logit contributions | |
Balanced | -0.086 |
Dominican | -0.079 |
Nicarag | -0.079 |
Augusta | -0.076 |
KB | -0.076 |
Token | service |
Token ID | 2139 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
Balanced | +0.240 |
Dominican | +0.218 |
Nicarag | +0.218 |
Augusta | +0.210 |
KB | +0.208 |
Neg logit contributions | |
breaching | -0.269 |
etheus | -0.248 |
Eid | -0.244 |
plets | -0.242 |
hump | -0.234 |
Token | ue |
Token ID | 518 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
breaching | +0.146 |
etheus | +0.131 |
Eid | +0.130 |
plets | +0.127 |
hump | +0.123 |
Neg logit contributions | |
Balanced | -0.182 |
Dominican | -0.168 |
Nicarag | -0.168 |
Augusta | -0.163 |
KB | -0.162 |
Token | gas |
Token ID | 3623 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
breaching | +0.412 |
etheus | +0.372 |
Eid | +0.368 |
plets | +0.361 |
hump | +0.351 |
Neg logit contributions | |
Balanced | -0.482 |
Dominican | -0.444 |
Nicarag | -0.444 |
Augusta | -0.431 |
KB | -0.427 |
Token | des |
Token ID | 748 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
breaching | +0.478 |
etheus | +0.439 |
Eid | +0.433 |
plets | +0.427 |
hump | +0.416 |
Neg logit contributions | |
Balanced | -0.430 |
Dominican | -0.391 |
Nicarag | -0.390 |
Augusta | -0.377 |
KB | -0.373 |
Token | ulf |
Token ID | 4754 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
Balanced | +0.110 |
Dominican | +0.099 |
Nicarag | +0.099 |
Augusta | +0.095 |
KB | +0.094 |
Neg logit contributions | |
breaching | -0.144 |
etheus | -0.131 |
Eid | -0.131 |
plets | -0.127 |
hump | -0.126 |
Token | ur |
Token ID | 333 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
breaching | +0.487 |
etheus | +0.450 |
Eid | +0.447 |
plets | +0.440 |
hump | +0.433 |
Neg logit contributions | |
Balanced | -0.284 |
Dominican | -0.251 |
Nicarag | -0.251 |
Augusta | -0.239 |
KB | -0.236 |
Token | ization |
Token ID | 1634 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
Balanced | +0.237 |
Dominican | +0.215 |
Nicarag | +0.214 |
Augusta | +0.208 |
KB | +0.208 |
Neg logit contributions | |
breaching | -0.224 |
etheus | -0.206 |
Eid | -0.203 |
plets | -0.198 |
hump | -0.194 |
Token | systems |
Token ID | 3341 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
breaching | +2.439 |
etheus | +2.208 |
Eid | +2.201 |
plets | +2.165 |
hump | +2.117 |
Neg logit contributions | |
Balanced | -2.091 |
Nicarag | -1.900 |
Dominican | -1.896 |
Augusta | -1.827 |
KB | -1.808 |
Token | have |
Token ID | 423 |
Feature activation | -4.59e-02 |
Pos logit contributions | |
breaching | +0.711 |
etheus | +0.654 |
Eid | +0.648 |
plets | +0.639 |
hump | +0.623 |
Neg logit contributions | |
Balanced | -0.553 |
Dominican | -0.499 |
Nicarag | -0.498 |
Augusta | -0.480 |
KB | -0.474 |
Token | been |
Token ID | 587 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
breaching | +0.089 |
etheus | +0.081 |
Eid | +0.080 |
plets | +0.079 |
hump | +0.077 |
Neg logit contributions | |
Balanced | -0.092 |
Dominican | -0.085 |
Nicarag | -0.084 |
Augusta | -0.082 |
KB | -0.080 |
Token | installed |
Token ID | 6589 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
Balanced | +0.228 |
Dominican | +0.205 |
Nicarag | +0.205 |
Augusta | +0.197 |
KB | +0.193 |
Neg logit contributions | |
breaching | -0.316 |
etheus | -0.293 |
Eid | -0.290 |
plets | -0.287 |
hump | -0.279 |
Token | to |
Token ID | 284 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
breaching | +0.396 |
etheus | +0.361 |
Eid | +0.357 |
plets | +0.353 |
hump | +0.342 |
Neg logit contributions | |
Balanced | -0.386 |
Dominican | -0.353 |
Nicarag | -0.352 |
Augusta | -0.341 |
KB | -0.337 |
Token | collagen |
Token ID | 45452 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
breaching | +0.219 |
etheus | +0.198 |
Eid | +0.194 |
plets | +0.193 |
hump | +0.185 |
Neg logit contributions | |
Balanced | -0.254 |
Nicarag | -0.234 |
Dominican | -0.234 |
Augusta | -0.226 |
KB | -0.223 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
breaching | +1.244 |
etheus | +1.143 |
Eid | +1.129 |
plets | +1.115 |
hump | +1.086 |
Neg logit contributions | |
Balanced | -1.054 |
Dominican | -0.956 |
Nicarag | -0.954 |
Augusta | -0.920 |
KB | -0.909 |
Token | bio |
Token ID | 13401 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
breaching | +0.199 |
etheus | +0.182 |
Eid | +0.177 |
plets | +0.177 |
hump | +0.169 |
Neg logit contributions | |
Balanced | -0.224 |
Dominican | -0.206 |
Nicarag | -0.205 |
Augusta | -0.199 |
KB | -0.197 |
Token | ap |
Token ID | 499 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
Balanced | +0.241 |
Nicarag | +0.217 |
Dominican | +0.217 |
Augusta | +0.209 |
KB | +0.208 |
Neg logit contributions | |
breaching | -0.307 |
etheus | -0.282 |
Eid | -0.280 |
plets | -0.275 |
hump | -0.270 |
Token | at |
Token ID | 265 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
breaching | +0.930 |
etheus | +0.852 |
Eid | +0.845 |
plets | +0.834 |
hump | +0.814 |
Neg logit contributions | |
Balanced | -0.739 |
Nicarag | -0.668 |
Dominican | -0.668 |
Augusta | -0.643 |
KB | -0.636 |
Token | ite |
Token ID | 578 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
breaching | +0.507 |
etheus | +0.460 |
Eid | +0.459 |
plets | +0.445 |
horns | +0.443 |
Neg logit contributions | |
Balanced | -0.414 |
Dominican | -0.371 |
Nicarag | -0.371 |
KB | -0.362 |
Augusta | -0.362 |
Token | ( |
Token ID | 357 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
breaching | +2.410 |
etheus | +2.195 |
Eid | +2.178 |
plets | +2.151 |
hump | +2.093 |
Neg logit contributions | |
Balanced | -2.090 |
Dominican | -1.900 |
Nicarag | -1.898 |
Augusta | -1.832 |
KB | -1.810 |
Token | hard |
Token ID | 10424 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
breaching | +0.045 |
etheus | +0.042 |
Eid | +0.041 |
plets | +0.040 |
hump | +0.039 |
Neg logit contributions | |
Balanced | -0.037 |
Nicarag | -0.034 |
Dominican | -0.034 |
KB | -0.033 |
Augusta | -0.032 |
Token | carbon |
Token ID | 6588 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
Balanced | +0.517 |
Nicarag | +0.474 |
Dominican | +0.473 |
Augusta | +0.455 |
KB | +0.454 |
Neg logit contributions | |
breaching | -0.499 |
etheus | -0.456 |
Eid | -0.451 |
plets | -0.443 |
hump | -0.427 |
Token | ate |
Token ID | 378 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
breaching | +0.733 |
etheus | +0.671 |
Eid | +0.661 |
plets | +0.652 |
hump | +0.633 |
Neg logit contributions | |
Balanced | -0.693 |
Dominican | -0.632 |
Nicarag | -0.630 |
Augusta | -0.612 |
KB | -0.605 |
Token | bone |
Token ID | 9970 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
breaching | +1.323 |
etheus | +1.212 |
Eid | +1.196 |
plets | +1.178 |
hump | +1.146 |
Neg logit contributions | |
Balanced | -1.223 |
Dominican | -1.114 |
Nicarag | -1.112 |
Augusta | -1.077 |
KB | -1.066 |
Token | presidential |
Token ID | 4787 |
Feature activation | +8.32e-03 |
Pos logit contributions | |
breaching | +0.174 |
etheus | +0.158 |
Eid | +0.156 |
plets | +0.154 |
hump | +0.149 |
Neg logit contributions | |
Balanced | -0.186 |
Dominican | -0.171 |
Nicarag | -0.171 |
Augusta | -0.166 |
KB | -0.164 |
Token | contender |
Token ID | 24190 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
Balanced | +0.019 |
Nicarag | +0.018 |
Dominican | +0.018 |
Augusta | +0.018 |
KB | +0.017 |
Neg logit contributions | |
breaching | -0.013 |
etheus | -0.012 |
Eid | -0.012 |
plets | -0.012 |
hump | -0.011 |
Token | in |
Token ID | 287 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
breaching | +1.342 |
etheus | +1.226 |
Eid | +1.218 |
plets | +1.200 |
hump | +1.172 |
Neg logit contributions | |
Balanced | -1.096 |
Dominican | -0.991 |
Nicarag | -0.991 |
Augusta | -0.954 |
KB | -0.945 |
Token | a |
Token ID | 257 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
breaching | +0.053 |
etheus | +0.048 |
Eid | +0.047 |
plets | +0.047 |
hump | +0.046 |
Neg logit contributions | |
Balanced | -0.054 |
Dominican | -0.050 |
Nicarag | -0.050 |
Augusta | -0.048 |
KB | -0.047 |
Token | Rolling |
Token ID | 21567 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
breaching | +0.130 |
etheus | +0.119 |
Eid | +0.117 |
plets | +0.116 |
hump | +0.112 |
Neg logit contributions | |
Balanced | -0.129 |
Dominican | -0.118 |
Nicarag | -0.118 |
Augusta | -0.114 |
KB | -0.112 |
Token | Stone |
Token ID | 8026 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Balanced | +0.449 |
Dominican | +0.420 |
Nicarag | +0.419 |
Augusta | +0.410 |
KB | +0.405 |
Neg logit contributions | |
breaching | -0.262 |
etheus | -0.228 |
Eid | -0.224 |
plets | -0.221 |
hump | -0.211 |
Token | profile |
Token ID | 7034 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
breaching | +2.263 |
etheus | +2.043 |
Eid | +2.039 |
plets | +1.966 |
hump | +1.934 |
Neg logit contributions | |
Balanced | -2.202 |
Nicarag | -2.013 |
Dominican | -1.995 |
Augusta | -1.947 |
KB | -1.938 |
Token | published |
Token ID | 3199 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
Balanced | +0.134 |
Dominican | +0.121 |
Nicarag | +0.121 |
Augusta | +0.117 |
KB | +0.115 |
Neg logit contributions | |
breaching | -0.170 |
etheus | -0.156 |
Eid | -0.154 |
plets | -0.153 |
hump | -0.149 |
Token | Wednesday |
Token ID | 3583 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
breaching | +0.612 |
etheus | +0.559 |
Eid | +0.556 |
plets | +0.546 |
hump | +0.535 |
Neg logit contributions | |
Balanced | -0.482 |
Nicarag | -0.435 |
Dominican | -0.434 |
Augusta | -0.418 |
KB | -0.414 |
Token |
|
Token ID | 198 |
Feature activation | -5.57e-03 |
Pos logit contributions | |
breaching | +0.251 |
etheus | +0.230 |
Eid | +0.228 |
plets | +0.225 |
hump | +0.219 |
Neg logit contributions | |
Balanced | -0.218 |
Dominican | -0.198 |
Nicarag | -0.197 |
Augusta | -0.191 |
KB | -0.188 |
Token |
|
Token ID | 198 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
breaching | +0.013 |
etheus | +0.012 |
Eid | +0.011 |
plets | +0.011 |
hump | +0.011 |
Neg logit contributions | |
Balanced | -0.009 |
Dominican | -0.008 |
Nicarag | -0.008 |
Augusta | -0.008 |
KB | -0.008 |
Token | 1940 |
Token ID | 16236 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
breaching | +0.016 |
etheus | +0.015 |
Eid | +0.015 |
plets | +0.014 |
hump | +0.014 |
Neg logit contributions | |
Balanced | -0.014 |
Dominican | -0.012 |
Nicarag | -0.012 |
Augusta | -0.012 |
KB | -0.012 |
Token | s |
Token ID | 82 |
Feature activation | -6.97e-02 |
Pos logit contributions | |
breaching | +0.663 |
etheus | +0.609 |
Eid | +0.603 |
plets | +0.596 |
hump | +0.581 |
Neg logit contributions | |
Balanced | -0.519 |
Dominican | -0.469 |
Nicarag | -0.468 |
Augusta | -0.451 |
KB | -0.445 |
Token | . |
Token ID | 13 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
breaching | +0.153 |
etheus | +0.139 |
Eid | +0.138 |
plets | +0.137 |
hump | +0.133 |
Neg logit contributions | |
Balanced | -0.123 |
Nicarag | -0.111 |
Dominican | -0.111 |
Augusta | -0.107 |
KB | -0.106 |
Token | The |
Token ID | 383 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
breaching | +0.047 |
Eid | +0.044 |
etheus | +0.043 |
plets | +0.043 |
hump | +0.042 |
Neg logit contributions | |
Balanced | -0.034 |
Nicarag | -0.031 |
Dominican | -0.031 |
Augusta | -0.029 |
KB | -0.029 |
Token | most |
Token ID | 749 |
Feature activation | -5.37e-03 |
Pos logit contributions | |
Balanced | +0.075 |
Dominican | +0.069 |
Nicarag | +0.068 |
Augusta | +0.066 |
KB | +0.065 |
Neg logit contributions | |
breaching | -0.088 |
etheus | -0.080 |
Eid | -0.079 |
plets | -0.078 |
hump | -0.076 |
Token | famous |
Token ID | 5863 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
breaching | +0.010 |
etheus | +0.009 |
Eid | +0.009 |
plets | +0.009 |
hump | +0.009 |
Neg logit contributions | |
Balanced | -0.011 |
Nicarag | -0.010 |
Dominican | -0.010 |
Augusta | -0.010 |
KB | -0.010 |
Token | case |
Token ID | 1339 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
breaching | +2.448 |
etheus | +2.225 |
Eid | +2.222 |
plets | +2.177 |
hump | +2.143 |
Neg logit contributions | |
Balanced | -1.963 |
Nicarag | -1.762 |
Dominican | -1.755 |
Augusta | -1.694 |
KB | -1.688 |
Token | was |
Token ID | 373 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
breaching | +0.269 |
etheus | +0.248 |
Eid | +0.244 |
plets | +0.243 |
hump | +0.235 |
Neg logit contributions | |
Balanced | -0.209 |
Dominican | -0.191 |
Nicarag | -0.190 |
Augusta | -0.183 |
KB | -0.180 |
Token | a |
Token ID | 257 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
breaching | +0.259 |
etheus | +0.238 |
Eid | +0.235 |
plets | +0.232 |
hump | +0.225 |
Neg logit contributions | |
Balanced | -0.217 |
Dominican | -0.198 |
Nicarag | -0.197 |
Augusta | -0.190 |
KB | -0.186 |
Token | Norwegian |
Token ID | 22158 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
Balanced | +0.047 |
Dominican | +0.043 |
Nicarag | +0.043 |
Augusta | +0.041 |
KB | +0.040 |
Neg logit contributions | |
breaching | -0.059 |
etheus | -0.055 |
Eid | -0.054 |
plets | -0.053 |
horns | -0.051 |
Token | woman |
Token ID | 2415 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Balanced | +0.137 |
Dominican | +0.126 |
Nicarag | +0.126 |
Augusta | +0.122 |
KB | +0.119 |
Neg logit contributions | |
breaching | -0.133 |
etheus | -0.122 |
Eid | -0.119 |
plets | -0.119 |
horns | -0.114 |
Token | brothers |
Token ID | 9397 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
breaching | +0.063 |
etheus | +0.056 |
Eid | +0.056 |
plets | +0.055 |
hump | +0.053 |
Neg logit contributions | |
Balanced | -0.087 |
Dominican | -0.080 |
Nicarag | -0.080 |
Augusta | -0.078 |
KB | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | -9.45e-02 |
Pos logit contributions | |
Balanced | +0.349 |
Dominican | +0.316 |
Nicarag | +0.315 |
Augusta | +0.303 |
KB | +0.300 |
Neg logit contributions | |
breaching | -0.434 |
etheus | -0.400 |
Eid | -0.394 |
plets | -0.390 |
hump | -0.380 |
Token | famous |
Token ID | 5863 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
breaching | +0.193 |
etheus | +0.178 |
Eid | +0.174 |
plets | +0.174 |
hump | +0.168 |
Neg logit contributions | |
Balanced | -0.180 |
Nicarag | -0.164 |
Dominican | -0.164 |
Augusta | -0.158 |
KB | -0.156 |
Token | ( |
Token ID | 357 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
breaching | +2.090 |
etheus | +1.904 |
Eid | +1.896 |
plets | +1.865 |
hump | +1.832 |
Neg logit contributions | |
Balanced | -1.710 |
Dominican | -1.548 |
Nicarag | -1.544 |
Augusta | -1.491 |
KB | -1.484 |
Token | or |
Token ID | 273 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
Balanced | +0.197 |
Dominican | +0.179 |
Nicarag | +0.179 |
KB | +0.173 |
Augusta | +0.171 |
Neg logit contributions | |
breaching | -0.260 |
etheus | -0.241 |
Eid | -0.237 |
plets | -0.235 |
horns | -0.228 |
Token | infamous |
Token ID | 16526 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
breaching | +0.330 |
etheus | +0.302 |
Eid | +0.298 |
plets | +0.294 |
hump | +0.284 |
Neg logit contributions | |
Balanced | -0.324 |
Dominican | -0.297 |
Nicarag | -0.297 |
Augusta | -0.287 |
KB | -0.283 |
Token | , |
Token ID | 11 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
breaching | +2.217 |
etheus | +2.000 |
Eid | +1.989 |
plets | +1.960 |
hump | +1.911 |
Neg logit contributions | |
Balanced | -2.183 |
Dominican | -1.993 |
Nicarag | -1.991 |
Augusta | -1.927 |
KB | -1.904 |
Token | depending |
Token ID | 6906 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
Balanced | +0.065 |
Nicarag | +0.060 |
Dominican | +0.060 |
Augusta | +0.058 |
KB | +0.057 |
Neg logit contributions | |
breaching | -0.057 |
etheus | -0.052 |
Eid | -0.051 |
plets | -0.050 |
hump | -0.048 |
Token | on |
Token ID | 319 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
breaching | +0.785 |
etheus | +0.724 |
Eid | +0.719 |
plets | +0.711 |
hump | +0.696 |
Neg logit contributions | |
Balanced | -0.503 |
Nicarag | -0.448 |
Dominican | -0.447 |
Augusta | -0.428 |
KB | -0.423 |
Token | one |
Token ID | 530 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
breaching | +0.315 |
etheus | +0.288 |
Eid | +0.283 |
plets | +0.281 |
hump | +0.272 |
Neg logit contributions | |
Balanced | -0.290 |
Dominican | -0.265 |
Nicarag | -0.265 |
Augusta | -0.255 |
KB | -0.251 |
Token | � |
Token ID | 447 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
Balanced | +0.026 |
Dominican | +0.024 |
Nicarag | +0.024 |
Augusta | +0.023 |
KB | +0.023 |
Neg logit contributions | |
breaching | -0.034 |
etheus | -0.031 |
Eid | -0.031 |
plets | -0.030 |
hump | -0.030 |
Token | regimes |
Token ID | 25879 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Balanced | +0.495 |
Nicarag | +0.450 |
Dominican | +0.449 |
Augusta | +0.429 |
Container | +0.420 |
Neg logit contributions | |
breaching | -0.608 |
etheus | -0.562 |
Eid | -0.548 |
plets | -0.548 |
hump | -0.529 |
Token | , |
Token ID | 11 |
Feature activation | +5.93e-02 |
Pos logit contributions | |
Balanced | +0.285 |
Dominican | +0.259 |
Nicarag | +0.258 |
Augusta | +0.248 |
KB | +0.243 |
Neg logit contributions | |
breaching | -0.370 |
etheus | -0.345 |
Eid | -0.339 |
plets | -0.339 |
hump | -0.326 |
Token | one |
Token ID | 530 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
Balanced | +0.110 |
Nicarag | +0.100 |
Dominican | +0.100 |
Augusta | +0.096 |
KB | +0.094 |
Neg logit contributions | |
breaching | -0.124 |
etheus | -0.114 |
plets | -0.112 |
Eid | -0.112 |
hump | -0.108 |
Token | that |
Token ID | 326 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
breaching | +0.216 |
etheus | +0.200 |
Eid | +0.197 |
plets | +0.196 |
hump | +0.190 |
Neg logit contributions | |
Balanced | -0.176 |
Dominican | -0.160 |
Nicarag | -0.160 |
Augusta | -0.154 |
KB | -0.151 |
Token | un |
Token ID | 555 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
Balanced | +0.230 |
Dominican | +0.209 |
Nicarag | +0.209 |
Augusta | +0.200 |
KB | +0.196 |
Neg logit contributions | |
breaching | -0.301 |
etheus | -0.279 |
plets | -0.273 |
Eid | -0.273 |
hump | -0.265 |
Token | ites |
Token ID | 2737 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Balanced | +0.515 |
Nicarag | +0.471 |
Dominican | +0.470 |
Augusta | +0.455 |
KB | +0.452 |
Neg logit contributions | |
breaching | -0.492 |
etheus | -0.443 |
Eid | -0.437 |
plets | -0.429 |
hump | -0.423 |
Token | them |
Token ID | 606 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
breaching | +2.199 |
etheus | +1.981 |
Eid | +1.979 |
plets | +1.936 |
hump | +1.891 |
Neg logit contributions | |
Balanced | -2.199 |
Dominican | -2.007 |
Nicarag | -2.006 |
Augusta | -1.947 |
KB | -1.933 |
Token | with |
Token ID | 351 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
Balanced | +0.258 |
Dominican | +0.234 |
Nicarag | +0.233 |
Augusta | +0.224 |
KB | +0.221 |
Neg logit contributions | |
breaching | -0.328 |
etheus | -0.303 |
Eid | -0.299 |
plets | -0.296 |
hump | -0.288 |
Token | their |
Token ID | 511 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
Balanced | +0.029 |
Nicarag | +0.027 |
Dominican | +0.027 |
Augusta | +0.026 |
KB | +0.025 |
Neg logit contributions | |
breaching | -0.030 |
etheus | -0.028 |
Eid | -0.027 |
plets | -0.027 |
hump | -0.026 |
Token | Israeli |
Token ID | 6085 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Balanced | +0.081 |
Dominican | +0.074 |
Nicarag | +0.074 |
Augusta | +0.071 |
KB | +0.069 |
Neg logit contributions | |
breaching | -0.093 |
etheus | -0.086 |
Eid | -0.084 |
plets | -0.084 |
hump | -0.081 |
Token | counterparts |
Token ID | 16054 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
Balanced | +0.243 |
Nicarag | +0.225 |
Dominican | +0.225 |
Augusta | +0.217 |
KB | +0.214 |
Neg logit contributions | |
breaching | -0.215 |
etheus | -0.196 |
Eid | -0.192 |
plets | -0.191 |
hump | -0.182 |
Token |
|
Token ID | 198 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
Balanced | +0.551 |
Dominican | +0.495 |
Nicarag | +0.493 |
Augusta | +0.475 |
KB | +0.470 |
Neg logit contributions | |
breaching | -0.780 |
etheus | -0.722 |
Eid | -0.708 |
plets | -0.705 |
hump | -0.686 |
Token | 55 |
Token ID | 2816 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
Balanced | +0.636 |
Dominican | +0.575 |
Nicarag | +0.571 |
KB | +0.558 |
Augusta | +0.553 |
Neg logit contributions | |
breaching | -0.767 |
etheus | -0.700 |
Eid | -0.687 |
plets | -0.686 |
hump | -0.666 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
Balanced | +0.218 |
Dominican | +0.184 |
Nicarag | +0.184 |
Augusta | +0.173 |
KB | +0.169 |
Neg logit contributions | |
breaching | -0.571 |
etheus | -0.534 |
Eid | -0.531 |
plets | -0.526 |
hump | -0.516 |
Token |
|
Token ID | 198 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
Balanced | +1.533 |
Dominican | +1.376 |
Nicarag | +1.367 |
Augusta | +1.321 |
KB | +1.304 |
Neg logit contributions | |
breaching | -2.214 |
etheus | -2.050 |
plets | -2.011 |
Eid | -2.006 |
hump | -1.948 |
Token |
|
Token ID | 198 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
Balanced | +0.590 |
Dominican | +0.530 |
Nicarag | +0.528 |
Augusta | +0.509 |
KB | +0.503 |
Neg logit contributions | |
breaching | -0.836 |
etheus | -0.773 |
Eid | -0.758 |
plets | -0.756 |
hump | -0.735 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
Balanced | +0.499 |
Dominican | +0.438 |
Nicarag | +0.434 |
KB | +0.416 |
Augusta | +0.415 |
Neg logit contributions | |
breaching | -0.964 |
etheus | -0.893 |
Eid | -0.885 |
plets | -0.877 |
hump | -0.859 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
Balanced | +1.190 |
Dominican | +1.051 |
Nicarag | +1.046 |
Augusta | +1.004 |
KB | +0.987 |
Neg logit contributions | |
breaching | -2.030 |
etheus | -1.880 |
Eid | -1.864 |
plets | -1.849 |
hump | -1.805 |
Token | this |
Token ID | 428 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
Balanced | +0.278 |
Nicarag | +0.251 |
Dominican | +0.251 |
Augusta | +0.241 |
KB | +0.239 |
Neg logit contributions | |
breaching | -0.360 |
etheus | -0.329 |
Eid | -0.328 |
plets | -0.322 |
hump | -0.315 |
Token | comment |
Token ID | 2912 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
Balanced | +0.086 |
Dominican | +0.080 |
Nicarag | +0.080 |
Augusta | +0.077 |
KB | +0.076 |
Neg logit contributions | |
breaching | -0.072 |
etheus | -0.065 |
Eid | -0.064 |
plets | -0.064 |
hump | -0.061 |
Token | on |
Token ID | 319 |
Feature activation | +6.89e-02 |
Pos logit contributions | |
Balanced | +0.079 |
Nicarag | +0.071 |
Dominican | +0.070 |
Augusta | +0.067 |
KB | +0.066 |
Neg logit contributions | |
breaching | -0.125 |
etheus | -0.115 |
Eid | -0.115 |
plets | -0.112 |
hump | -0.110 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
Balanced | +0.113 |
Dominican | +0.101 |
Nicarag | +0.101 |
Augusta | +0.097 |
KB | +0.096 |
Neg logit contributions | |
breaching | -0.159 |
etheus | -0.146 |
Eid | -0.145 |
plets | -0.143 |
hump | -0.140 |
Token | the |
Token ID | 262 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
breaching | +0.334 |
etheus | +0.306 |
Eid | +0.298 |
plets | +0.296 |
hump | +0.285 |
Neg logit contributions | |
Balanced | -0.378 |
Nicarag | -0.347 |
Dominican | -0.347 |
Augusta | -0.337 |
KB | -0.331 |
Token | art |
Token ID | 1242 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
breaching | +0.121 |
etheus | +0.111 |
Eid | +0.108 |
plets | +0.108 |
hump | +0.103 |
Neg logit contributions | |
Balanced | -0.136 |
Dominican | -0.126 |
Nicarag | -0.126 |
Augusta | -0.122 |
Container | -0.119 |
Token | of |
Token ID | 286 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
breaching | +0.364 |
etheus | +0.334 |
Eid | +0.331 |
plets | +0.327 |
hump | +0.319 |
Neg logit contributions | |
Balanced | -0.289 |
Dominican | -0.261 |
Nicarag | -0.261 |
Augusta | -0.252 |
KB | -0.249 |
Token | nin |
Token ID | 13462 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
breaching | +0.349 |
etheus | +0.320 |
Eid | +0.311 |
plets | +0.310 |
hump | +0.297 |
Neg logit contributions | |
Balanced | -0.423 |
Dominican | -0.390 |
Nicarag | -0.390 |
Augusta | -0.379 |
KB | -0.372 |
Token | j |
Token ID | 73 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
breaching | +0.066 |
etheus | +0.059 |
Eid | +0.059 |
plets | +0.058 |
hump | +0.056 |
Neg logit contributions | |
Balanced | -0.080 |
Dominican | -0.074 |
Nicarag | -0.073 |
Augusta | -0.072 |
KB | -0.071 |
Token | utsu |
Token ID | 36567 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
Balanced | +0.131 |
Dominican | +0.115 |
Nicarag | +0.114 |
Augusta | +0.108 |
KB | +0.107 |
Neg logit contributions | |
breaching | -0.274 |
etheus | -0.255 |
Eid | -0.254 |
plets | -0.251 |
hump | -0.246 |
Token | that |
Token ID | 326 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
Balanced | +1.282 |
Dominican | +1.164 |
Nicarag | +1.157 |
Augusta | +1.121 |
KB | +1.089 |
Neg logit contributions | |
breaching | -1.708 |
etheus | -1.600 |
Eid | -1.561 |
plets | -1.560 |
hump | -1.517 |
Token | he |
Token ID | 339 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
Balanced | +0.106 |
Dominican | +0.096 |
Nicarag | +0.096 |
Augusta | +0.092 |
KB | +0.091 |
Neg logit contributions | |
breaching | -0.124 |
etheus | -0.114 |
Eid | -0.112 |
plets | -0.112 |
hump | -0.108 |
Token | had |
Token ID | 550 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
breaching | +0.034 |
etheus | +0.031 |
Eid | +0.031 |
plets | +0.030 |
hump | +0.030 |
Neg logit contributions | |
Balanced | -0.026 |
Dominican | -0.024 |
Nicarag | -0.024 |
Augusta | -0.023 |
KB | -0.022 |
Token | learned |
Token ID | 4499 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
Balanced | +0.409 |
Dominican | +0.378 |
Nicarag | +0.378 |
Augusta | +0.366 |
KB | +0.356 |
Neg logit contributions | |
breaching | -0.411 |
etheus | -0.380 |
Eid | -0.370 |
plets | -0.367 |
hump | -0.353 |
Token | from |
Token ID | 422 |
Feature activation | -3.77e-03 |
Pos logit contributions | |
breaching | +1.275 |
etheus | +1.168 |
Eid | +1.155 |
plets | +1.141 |
hump | +1.111 |
Neg logit contributions | |
Balanced | -1.080 |
Dominican | -0.981 |
Nicarag | -0.980 |
Augusta | -0.945 |
KB | -0.933 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Balanced | +0.082 |
Nicarag | +0.074 |
Dominican | +0.074 |
Augusta | +0.072 |
KB | +0.070 |
Neg logit contributions | |
breaching | -0.105 |
etheus | -0.097 |
plets | -0.096 |
Eid | -0.095 |
hump | -0.092 |
Token | 't |
Token ID | 470 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
Balanced | +0.211 |
Dominican | +0.191 |
Nicarag | +0.190 |
Augusta | +0.184 |
KB | +0.182 |
Neg logit contributions | |
breaching | -0.270 |
etheus | -0.249 |
Eid | -0.245 |
plets | -0.243 |
hump | -0.237 |
Token | seem |
Token ID | 1283 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
breaching | +0.295 |
etheus | +0.271 |
Eid | +0.267 |
plets | +0.265 |
horns | +0.256 |
Neg logit contributions | |
Balanced | -0.248 |
Dominican | -0.226 |
Nicarag | -0.225 |
Augusta | -0.218 |
KB | -0.215 |
Token | in |
Token ID | 287 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
Balanced | +0.407 |
Dominican | +0.376 |
Nicarag | +0.376 |
Augusta | +0.363 |
KB | +0.359 |
Neg logit contributions | |
breaching | -0.406 |
etheus | -0.377 |
Eid | -0.368 |
plets | -0.365 |
hump | -0.351 |
Token | the |
Token ID | 262 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
Balanced | +0.034 |
Dominican | +0.031 |
Nicarag | +0.031 |
Augusta | +0.031 |
KB | +0.030 |
Neg logit contributions | |
breaching | -0.035 |
etheus | -0.032 |
Eid | -0.031 |
plets | -0.031 |
hump | -0.030 |
Token | mood |
Token ID | 10038 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
Balanced | +0.083 |
Dominican | +0.076 |
Nicarag | +0.075 |
Augusta | +0.073 |
KB | +0.072 |
Neg logit contributions | |
breaching | -0.090 |
etheus | -0.083 |
Eid | -0.082 |
plets | -0.081 |
hump | -0.078 |
Token | to |
Token ID | 284 |
Feature activation | +4.93e-02 |
Pos logit contributions | |
Balanced | +1.674 |
Dominican | +1.530 |
Nicarag | +1.523 |
Augusta | +1.480 |
Container | +1.450 |
Neg logit contributions | |
breaching | -1.901 |
etheus | -1.761 |
Eid | -1.727 |
plets | -1.719 |
hump | -1.645 |
Token | talk |
Token ID | 1561 |
Feature activation | -7.35e-02 |
Pos logit contributions | |
Balanced | +0.091 |
Nicarag | +0.083 |
Dominican | +0.083 |
Augusta | +0.080 |
KB | +0.079 |
Neg logit contributions | |
breaching | -0.102 |
etheus | -0.094 |
plets | -0.092 |
Eid | -0.092 |
horns | -0.088 |
Token | , |
Token ID | 11 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
breaching | +0.155 |
etheus | +0.143 |
plets | +0.140 |
Eid | +0.139 |
hump | +0.135 |
Neg logit contributions | |
Balanced | -0.134 |
Dominican | -0.123 |
Nicarag | -0.123 |
Augusta | -0.118 |
KB | -0.116 |
Token | and |
Token ID | 290 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
Balanced | +0.080 |
Dominican | +0.073 |
Nicarag | +0.073 |
Augusta | +0.070 |
KB | +0.069 |
Neg logit contributions | |
breaching | -0.094 |
etheus | -0.087 |
Eid | -0.085 |
plets | -0.085 |
hump | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
Balanced | +0.034 |
Dominican | +0.030 |
Nicarag | +0.030 |
Augusta | +0.029 |
KB | +0.029 |
Neg logit contributions | |
breaching | -0.041 |
etheus | -0.038 |
plets | -0.037 |
Eid | -0.037 |
hump | -0.036 |
Token | 120 |
Token ID | 7982 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
Balanced | +0.781 |
Dominican | +0.695 |
Nicarag | +0.693 |
Augusta | +0.668 |
KB | +0.663 |
Neg logit contributions | |
breaching | -1.031 |
etheus | -0.951 |
plets | -0.929 |
Eid | -0.926 |
horns | -0.899 |
Token | Hz |
Token ID | 7399 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Balanced | +0.051 |
Dominican | +0.045 |
Nicarag | +0.045 |
Augusta | +0.042 |
KB | +0.041 |
Neg logit contributions | |
breaching | -0.113 |
etheus | -0.106 |
Eid | -0.105 |
plets | -0.104 |
hump | -0.102 |
Token | We |
Token ID | 775 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
Balanced | +0.356 |
Dominican | +0.318 |
Nicarag | +0.317 |
Augusta | +0.307 |
KB | +0.302 |
Neg logit contributions | |
breaching | -0.453 |
etheus | -0.419 |
Eid | -0.410 |
plets | -0.408 |
horns | -0.396 |
Token | 've |
Token ID | 1053 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
Balanced | +0.274 |
Dominican | +0.247 |
Nicarag | +0.247 |
Augusta | +0.238 |
KB | +0.235 |
Neg logit contributions | |
breaching | -0.350 |
etheus | -0.321 |
Eid | -0.318 |
plets | -0.314 |
hump | -0.306 |
Token | also |
Token ID | 635 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Balanced | +0.388 |
Dominican | +0.354 |
Nicarag | +0.353 |
Augusta | +0.342 |
KB | +0.337 |
Neg logit contributions | |
breaching | -0.399 |
etheus | -0.365 |
Eid | -0.358 |
plets | -0.356 |
hump | -0.342 |
Token | included |
Token ID | 3017 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
Balanced | +0.394 |
Dominican | +0.357 |
Nicarag | +0.356 |
Augusta | +0.344 |
KB | +0.339 |
Neg logit contributions | |
breaching | -0.456 |
etheus | -0.419 |
Eid | -0.412 |
plets | -0.409 |
hump | -0.395 |
Token | a |
Token ID | 257 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
Balanced | +1.634 |
Dominican | +1.467 |
Nicarag | +1.458 |
Augusta | +1.427 |
KB | +1.410 |
Neg logit contributions | |
breaching | -1.705 |
etheus | -1.565 |
plets | -1.540 |
Eid | -1.531 |
hump | -1.471 |
Token | comparison |
Token ID | 7208 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
Balanced | +0.476 |
Dominican | +0.429 |
Nicarag | +0.429 |
Augusta | +0.415 |
KB | +0.412 |
Neg logit contributions | |
breaching | -0.510 |
etheus | -0.467 |
Eid | -0.461 |
plets | -0.459 |
horns | -0.442 |
Token | above |
Token ID | 2029 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
breaching | +0.625 |
etheus | +0.571 |
Eid | +0.566 |
plets | +0.558 |
hump | +0.545 |
Neg logit contributions | |
Balanced | -0.526 |
Dominican | -0.477 |
Nicarag | -0.477 |
Augusta | -0.460 |
KB | -0.454 |
Token | against |
Token ID | 1028 |
Feature activation | +8.88e-02 |
Pos logit contributions | |
Balanced | +0.071 |
Dominican | +0.064 |
Nicarag | +0.064 |
Augusta | +0.061 |
KB | +0.060 |
Neg logit contributions | |
breaching | -0.088 |
etheus | -0.081 |
Eid | -0.080 |
plets | -0.079 |
hump | -0.077 |
Token | 3 |
Token ID | 513 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
Balanced | +0.175 |
Dominican | +0.159 |
Nicarag | +0.158 |
Augusta | +0.154 |
KB | +0.152 |
Neg logit contributions | |
breaching | -0.177 |
etheus | -0.162 |
plets | -0.158 |
Eid | -0.158 |
horns | -0.152 |
Token | think |
Token ID | 892 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
breaching | +0.237 |
etheus | +0.218 |
Eid | +0.216 |
plets | +0.213 |
hump | +0.207 |
Neg logit contributions | |
Balanced | -0.186 |
Dominican | -0.168 |
Nicarag | -0.167 |
Augusta | -0.161 |
KB | -0.160 |
Token | around |
Token ID | 1088 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
breaching | +0.026 |
etheus | +0.024 |
Eid | +0.023 |
plets | +0.023 |
hump | +0.023 |
Neg logit contributions | |
Balanced | -0.023 |
Dominican | -0.021 |
Nicarag | -0.021 |
Augusta | -0.020 |
KB | -0.020 |
Token | 15 |
Token ID | 1315 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
breaching | +0.081 |
etheus | +0.075 |
Eid | +0.074 |
plets | +0.073 |
hump | +0.071 |
Neg logit contributions | |
Balanced | -0.068 |
Dominican | -0.062 |
Nicarag | -0.062 |
Augusta | -0.060 |
KB | -0.059 |
Token | - |
Token ID | 12 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
breaching | +0.169 |
etheus | +0.157 |
Eid | +0.155 |
plets | +0.154 |
hump | +0.150 |
Neg logit contributions | |
Balanced | -0.092 |
Dominican | -0.080 |
Nicarag | -0.080 |
Augusta | -0.077 |
KB | -0.076 |
Token | 20 |
Token ID | 1238 |
Feature activation | +2.76e-03 |
Pos logit contributions | |
breaching | +0.154 |
etheus | +0.141 |
Eid | +0.139 |
plets | +0.136 |
hump | +0.133 |
Neg logit contributions | |
Balanced | -0.146 |
Dominican | -0.133 |
Nicarag | -0.132 |
KB | -0.130 |
Augusta | -0.129 |
Token | minutes |
Token ID | 2431 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
Balanced | +0.004 |
Dominican | +0.003 |
Nicarag | +0.003 |
Augusta | +0.003 |
KB | +0.003 |
Neg logit contributions | |
breaching | -0.007 |
etheus | -0.007 |
Eid | -0.007 |
plets | -0.006 |
hump | -0.006 |
Token | for |
Token ID | 329 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
Balanced | +1.302 |
Dominican | +1.174 |
Nicarag | +1.166 |
Augusta | +1.135 |
KB | +1.114 |
Neg logit contributions | |
breaching | -1.744 |
etheus | -1.633 |
plets | -1.598 |
Eid | -1.593 |
hump | -1.532 |
Token | longest |
Token ID | 14069 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
Balanced | +0.315 |
Nicarag | +0.281 |
Dominican | +0.281 |
Augusta | +0.269 |
KB | +0.265 |
Neg logit contributions | |
breaching | -0.470 |
etheus | -0.438 |
Eid | -0.430 |
plets | -0.429 |
hump | -0.416 |
Token | . |
Token ID | 13 |
Feature activation | -7.57e-02 |
Pos logit contributions | |
breaching | +0.475 |
etheus | +0.436 |
Eid | +0.430 |
plets | +0.426 |
hump | +0.413 |
Neg logit contributions | |
Balanced | -0.400 |
Dominican | -0.362 |
Nicarag | -0.362 |
Augusta | -0.349 |
KB | -0.346 |
Token | This |
Token ID | 770 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
breaching | +0.172 |
etheus | +0.158 |
Eid | +0.156 |
plets | +0.155 |
hump | +0.151 |
Neg logit contributions | |
Balanced | -0.127 |
Dominican | -0.115 |
Nicarag | -0.115 |
Augusta | -0.110 |
KB | -0.109 |
Token | will |
Token ID | 481 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
Balanced | +0.126 |
Dominican | +0.114 |
Nicarag | +0.114 |
Augusta | +0.110 |
KB | +0.108 |
Neg logit contributions | |
breaching | -0.157 |
etheus | -0.144 |
Eid | -0.142 |
plets | -0.142 |
hump | -0.136 |
Token | how |
Token ID | 703 |
Feature activation | +8.78e-02 |
Pos logit contributions | |
Balanced | +0.148 |
Nicarag | +0.136 |
Dominican | +0.135 |
Augusta | +0.131 |
KB | +0.128 |
Neg logit contributions | |
breaching | -0.159 |
etheus | -0.147 |
plets | -0.144 |
Eid | -0.144 |
hump | -0.138 |
Token | U |
Token ID | 471 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
Balanced | +0.169 |
Nicarag | +0.156 |
Dominican | +0.155 |
Augusta | +0.150 |
KB | +0.147 |
Neg logit contributions | |
breaching | -0.176 |
etheus | -0.163 |
plets | -0.159 |
Eid | -0.158 |
hump | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
Balanced | +0.177 |
Nicarag | +0.160 |
Dominican | +0.160 |
Augusta | +0.154 |
KB | +0.154 |
Neg logit contributions | |
breaching | -0.228 |
etheus | -0.209 |
Eid | -0.206 |
plets | -0.204 |
hump | -0.199 |
Token | S |
Token ID | 50 |
Feature activation | +6.21e-03 |
Pos logit contributions | |
breaching | +0.322 |
etheus | +0.293 |
Eid | +0.289 |
plets | +0.286 |
hump | +0.277 |
Neg logit contributions | |
Balanced | -0.319 |
Dominican | -0.291 |
Nicarag | -0.291 |
Augusta | -0.282 |
KB | -0.280 |
Token | . |
Token ID | 13 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
Balanced | +0.011 |
Dominican | +0.010 |
Nicarag | +0.010 |
Augusta | +0.009 |
KB | +0.009 |
Neg logit contributions | |
breaching | -0.014 |
etheus | -0.012 |
Eid | -0.012 |
plets | -0.012 |
hump | -0.012 |
Token | generals |
Token ID | 26770 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
breaching | +0.176 |
etheus | +0.160 |
Eid | +0.159 |
plets | +0.157 |
hump | +0.152 |
Neg logit contributions | |
Balanced | -0.160 |
Nicarag | -0.146 |
Dominican | -0.146 |
Augusta | -0.141 |
KB | -0.139 |
Token | are |
Token ID | 389 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
Balanced | +0.057 |
Dominican | +0.052 |
Nicarag | +0.052 |
Augusta | +0.050 |
KB | +0.049 |
Neg logit contributions | |
breaching | -0.070 |
etheus | -0.065 |
Eid | -0.064 |
plets | -0.063 |
hump | -0.061 |
Token | appointed |
Token ID | 9899 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
breaching | +0.161 |
etheus | +0.147 |
Eid | +0.145 |
plets | +0.144 |
horns | +0.138 |
Neg logit contributions | |
Balanced | -0.177 |
Dominican | -0.163 |
Nicarag | -0.163 |
Augusta | -0.158 |
KB | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
Balanced | +0.228 |
Dominican | +0.206 |
Nicarag | +0.206 |
Augusta | +0.198 |
KB | +0.196 |
Neg logit contributions | |
breaching | -0.288 |
etheus | -0.265 |
Eid | -0.261 |
plets | -0.260 |
hump | -0.252 |
Token | He |
Token ID | 679 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
breaching | +0.090 |
etheus | +0.083 |
Eid | +0.082 |
plets | +0.081 |
hump | +0.079 |
Neg logit contributions | |
Balanced | -0.069 |
Dominican | -0.062 |
Nicarag | -0.062 |
Augusta | -0.060 |
KB | -0.059 |
Token | explained |
Token ID | 4893 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
Balanced | +0.021 |
Dominican | +0.019 |
Nicarag | +0.019 |
Augusta | +0.018 |
KB | +0.018 |
Neg logit contributions | |
breaching | -0.026 |
etheus | -0.024 |
Eid | -0.024 |
plets | -0.023 |
hump | -0.023 |
Token | IR |
Token ID | 4663 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
breaching | +0.143 |
etheus | +0.133 |
Eid | +0.132 |
plets | +0.130 |
hump | +0.127 |
Neg logit contributions | |
Balanced | -0.095 |
Dominican | -0.085 |
Nicarag | -0.085 |
Augusta | -0.081 |
KB | -0.081 |
Token | ST |
Token ID | 2257 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
breaching | +0.054 |
etheus | +0.050 |
Eid | +0.049 |
plets | +0.049 |
hump | +0.047 |
Neg logit contributions | |
Balanced | -0.042 |
Dominican | -0.038 |
Nicarag | -0.038 |
Augusta | -0.036 |
KB | -0.036 |
Token | Y |
Token ID | 56 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
breaching | +0.334 |
etheus | +0.307 |
Eid | +0.303 |
plets | +0.300 |
hump | +0.292 |
Neg logit contributions | |
Balanced | -0.264 |
Dominican | -0.239 |
Nicarag | -0.239 |
Augusta | -0.229 |
KB | -0.227 |
Token | LE |
Token ID | 2538 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
breaching | +0.035 |
etheus | +0.032 |
Eid | +0.032 |
plets | +0.031 |
hump | +0.030 |
Neg logit contributions | |
Balanced | -0.027 |
Dominican | -0.025 |
Nicarag | -0.025 |
KB | -0.024 |
Augusta | -0.024 |
Token | would |
Token ID | 561 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
Balanced | +0.378 |
Dominican | +0.337 |
Nicarag | +0.336 |
Augusta | +0.324 |
KB | +0.324 |
Neg logit contributions | |
breaching | -0.496 |
etheus | -0.458 |
Eid | -0.449 |
plets | -0.447 |
hump | -0.433 |
Token | fit |
Token ID | 4197 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
Balanced | +0.332 |
Dominican | +0.303 |
Nicarag | +0.301 |
Augusta | +0.292 |
KB | +0.289 |
Neg logit contributions | |
breaching | -0.365 |
etheus | -0.335 |
Eid | -0.329 |
plets | -0.326 |
hump | -0.315 |
Token | . |
Token ID | 13 |
Feature activation | +8.99e-02 |
Pos logit contributions | |
Balanced | +0.752 |
Dominican | +0.678 |
Nicarag | +0.677 |
Augusta | +0.656 |
KB | +0.648 |
Neg logit contributions | |
breaching | -0.853 |
etheus | -0.791 |
plets | -0.775 |
Eid | -0.768 |
hump | -0.742 |
Token | Scottish |
Token ID | 11905 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
Balanced | +0.158 |
Dominican | +0.141 |
Nicarag | +0.141 |
Augusta | +0.137 |
Container | +0.134 |
Neg logit contributions | |
breaching | -0.199 |
etheus | -0.184 |
plets | -0.180 |
Eid | -0.179 |
horns | -0.173 |
Token | team |
Token ID | 1074 |
Feature activation | -8.30e-02 |
Pos logit contributions | |
Balanced | +0.681 |
Dominican | +0.621 |
Nicarag | +0.615 |
Augusta | +0.593 |
KB | +0.591 |
Neg logit contributions | |
breaching | -0.700 |
etheus | -0.652 |
plets | -0.634 |
Eid | -0.632 |
hump | -0.603 |
Token | ? |
Token ID | 30 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
breaching | +0.177 |
etheus | +0.165 |
Eid | +0.161 |
plets | +0.160 |
hump | +0.155 |
Neg logit contributions | |
Balanced | -0.149 |
Dominican | -0.136 |
Nicarag | -0.135 |
Augusta | -0.130 |
KB | -0.129 |
Token | Personnel |
Token ID | 41623 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
breaching | +0.046 |
etheus | +0.043 |
plets | +0.042 |
Eid | +0.042 |
horns | +0.040 |
Neg logit contributions | |
Balanced | -0.039 |
Dominican | -0.035 |
Nicarag | -0.035 |
Augusta | -0.034 |
KB | -0.033 |
Token | title |
Token ID | 3670 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
Balanced | +0.016 |
Dominican | +0.015 |
Nicarag | +0.015 |
Augusta | +0.014 |
KB | +0.014 |
Neg logit contributions | |
breaching | -0.014 |
etheus | -0.013 |
plets | -0.013 |
Eid | -0.013 |
hump | -0.012 |
Token | win |
Token ID | 1592 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
breaching | +0.448 |
etheus | +0.414 |
Eid | +0.408 |
plets | +0.407 |
hump | +0.393 |
Neg logit contributions | |
Balanced | -0.342 |
Dominican | -0.311 |
Nicarag | -0.309 |
Augusta | -0.298 |
KB | -0.295 |
Token | , |
Token ID | 11 |
Feature activation | -1.61e-03 |
Pos logit contributions | |
Balanced | +0.050 |
Dominican | +0.045 |
Nicarag | +0.045 |
Augusta | +0.044 |
KB | +0.043 |
Neg logit contributions | |
breaching | -0.064 |
etheus | -0.059 |
plets | -0.058 |
Eid | -0.057 |
hump | -0.056 |
Token | however |
Token ID | 2158 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
breaching | +0.003 |
etheus | +0.003 |
plets | +0.003 |
Eid | +0.003 |
hump | +0.003 |
Neg logit contributions | |
Balanced | -0.003 |
Dominican | -0.003 |
Nicarag | -0.003 |
Augusta | -0.003 |
KB | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +6.92e-02 |
Pos logit contributions | |
breaching | +0.594 |
etheus | +0.543 |
Eid | +0.539 |
plets | +0.532 |
hump | +0.519 |
Neg logit contributions | |
Balanced | -0.490 |
Dominican | -0.444 |
Nicarag | -0.444 |
Augusta | -0.428 |
KB | -0.423 |
Token | would |
Token ID | 561 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
Balanced | +0.117 |
Dominican | +0.106 |
Nicarag | +0.106 |
Augusta | +0.102 |
KB | +0.100 |
Neg logit contributions | |
breaching | -0.154 |
etheus | -0.143 |
Eid | -0.140 |
plets | -0.140 |
hump | -0.136 |
Token | not |
Token ID | 407 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
Balanced | +0.465 |
Dominican | +0.426 |
Nicarag | +0.424 |
Augusta | +0.411 |
KB | +0.403 |
Neg logit contributions | |
breaching | -0.548 |
etheus | -0.505 |
Eid | -0.496 |
plets | -0.491 |
horns | -0.474 |
Token | . |
Token ID | 13 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
breaching | +0.140 |
etheus | +0.129 |
Eid | +0.127 |
plets | +0.125 |
hump | +0.121 |
Neg logit contributions | |
Balanced | -0.124 |
Dominican | -0.113 |
Nicarag | -0.113 |
Augusta | -0.109 |
KB | -0.108 |
Token |
|
Token ID | 198 |
Feature activation | -8.82e-02 |
Pos logit contributions | |
Balanced | +0.085 |
Dominican | +0.076 |
Nicarag | +0.076 |
Augusta | +0.073 |
KB | +0.072 |
Neg logit contributions | |
breaching | -0.115 |
etheus | -0.106 |
Eid | -0.104 |
plets | -0.104 |
hump | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | +3.05e-03 |
Pos logit contributions | |
breaching | +0.199 |
etheus | +0.183 |
Eid | +0.181 |
plets | +0.179 |
hump | +0.175 |
Neg logit contributions | |
Balanced | -0.149 |
Dominican | -0.134 |
Nicarag | -0.134 |
Augusta | -0.129 |
KB | -0.127 |
Token | Yes |
Token ID | 5297 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
Balanced | +0.005 |
Dominican | +0.004 |
Nicarag | +0.004 |
KB | +0.004 |
Augusta | +0.004 |
Neg logit contributions | |
breaching | -0.007 |
etheus | -0.007 |
Eid | -0.006 |
plets | -0.006 |
hump | -0.006 |
Token | Your |
Token ID | 7120 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Balanced | +0.001 |
Dominican | +0.001 |
Nicarag | +0.001 |
KB | +0.001 |
Augusta | +0.001 |
Neg logit contributions | |
breaching | -0.002 |
etheus | -0.002 |
Eid | -0.002 |
plets | -0.002 |
hump | -0.001 |
Token | Inner |
Token ID | 24877 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
Balanced | +0.196 |
Dominican | +0.173 |
Nicarag | +0.173 |
Augusta | +0.167 |
KB | +0.164 |
Neg logit contributions | |
breaching | -0.269 |
etheus | -0.252 |
plets | -0.246 |
Eid | -0.246 |
hump | -0.237 |
Token | Ass |
Token ID | 2195 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
Balanced | +0.802 |
Dominican | +0.734 |
Nicarag | +0.730 |
Augusta | +0.710 |
KB | +0.700 |
Neg logit contributions | |
breaching | -0.783 |
etheus | -0.711 |
Eid | -0.701 |
plets | -0.693 |
hump | -0.671 |
Token | hole |
Token ID | 13207 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
Balanced | +0.729 |
Dominican | +0.679 |
Nicarag | +0.670 |
Augusta | +0.660 |
KB | +0.653 |
Neg logit contributions | |
breaching | -0.488 |
etheus | -0.428 |
Eid | -0.425 |
plets | -0.416 |
horns | -0.401 |
Token | is |
Token ID | 318 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Balanced | +0.528 |
Dominican | +0.475 |
Nicarag | +0.472 |
Augusta | +0.460 |
KB | +0.449 |
Neg logit contributions | |
breaching | -0.654 |
etheus | -0.605 |
Eid | -0.593 |
plets | -0.593 |
hump | -0.571 |
Token | never |
Token ID | 1239 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
Balanced | +0.262 |
Dominican | +0.239 |
Nicarag | +0.238 |
Augusta | +0.231 |
KB | +0.226 |
Neg logit contributions | |
breaching | -0.265 |
etheus | -0.244 |
Eid | -0.239 |
plets | -0.238 |
hump | -0.229 |
Token | off |
Token ID | 572 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
Balanced | +0.489 |
Dominican | +0.449 |
Nicarag | +0.445 |
Augusta | +0.435 |
KB | +0.425 |
Neg logit contributions | |
breaching | -0.457 |
etheus | -0.424 |
Eid | -0.414 |
plets | -0.410 |
hump | -0.395 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
breaching | +0.050 |
etheus | +0.046 |
Eid | +0.045 |
plets | +0.045 |
horns | +0.043 |
Neg logit contributions | |
Balanced | -0.048 |
Dominican | -0.043 |
Nicarag | -0.043 |
Augusta | -0.042 |
KB | -0.041 |
Token | alert |
Token ID | 7995 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
Balanced | +0.316 |
Dominican | +0.286 |
Nicarag | +0.283 |
Augusta | +0.278 |
KB | +0.271 |
Neg logit contributions | |
breaching | -0.368 |
etheus | -0.344 |
Eid | -0.336 |
plets | -0.336 |
horns | -0.320 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
breaching | +0.061 |
etheus | +0.057 |
Eid | +0.056 |
plets | +0.055 |
hump | +0.053 |
Neg logit contributions | |
Balanced | -0.055 |
Dominican | -0.049 |
Nicarag | -0.049 |
Augusta | -0.048 |
KB | -0.047 |
Token | He |
Token ID | 679 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
Balanced | +0.212 |
Dominican | +0.191 |
Nicarag | +0.191 |
Augusta | +0.185 |
KB | +0.182 |
Neg logit contributions | |
breaching | -0.253 |
etheus | -0.232 |
Eid | -0.227 |
plets | -0.226 |
hump | -0.220 |
Token | - |
Token ID | 12 |
Feature activation | +7.01e-02 |
Pos logit contributions | |
Balanced | +0.066 |
Dominican | +0.059 |
Nicarag | +0.059 |
Augusta | +0.057 |
KB | +0.057 |
Neg logit contributions | |
breaching | -0.081 |
etheus | -0.075 |
Eid | -0.074 |
plets | -0.073 |
hump | -0.071 |
Token | point |
Token ID | 4122 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
Balanced | +0.125 |
Dominican | +0.113 |
Nicarag | +0.113 |
KB | +0.109 |
Augusta | +0.109 |
Neg logit contributions | |
breaching | -0.151 |
etheus | -0.139 |
Eid | -0.137 |
plets | -0.135 |
hump | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
breaching | +0.098 |
etheus | +0.090 |
Eid | +0.089 |
plets | +0.088 |
hump | +0.086 |
Neg logit contributions | |
Balanced | -0.078 |
Dominican | -0.071 |
Nicarag | -0.071 |
Augusta | -0.068 |
KB | -0.067 |
Token | To |
Token ID | 1675 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Balanced | +0.059 |
Dominican | +0.053 |
Nicarag | +0.053 |
Augusta | +0.051 |
KB | +0.051 |
Neg logit contributions | |
breaching | -0.079 |
etheus | -0.073 |
plets | -0.071 |
Eid | -0.071 |
hump | -0.069 |
Token | get |
Token ID | 651 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
Balanced | +0.255 |
Dominican | +0.232 |
Nicarag | +0.231 |
Augusta | +0.224 |
KB | +0.221 |
Neg logit contributions | |
breaching | -0.283 |
etheus | -0.259 |
Eid | -0.255 |
plets | -0.252 |
horns | -0.244 |
Token | a |
Token ID | 257 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
Balanced | +0.269 |
Dominican | +0.247 |
Nicarag | +0.245 |
Augusta | +0.238 |
KB | +0.233 |
Neg logit contributions | |
breaching | -0.284 |
etheus | -0.261 |
Eid | -0.256 |
plets | -0.255 |
hump | -0.245 |
Token | better |
Token ID | 1365 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
Balanced | +0.142 |
Dominican | +0.130 |
Nicarag | +0.130 |
Augusta | +0.125 |
KB | +0.123 |
Neg logit contributions | |
breaching | -0.146 |
etheus | -0.134 |
Eid | -0.132 |
plets | -0.130 |
horns | -0.126 |
Token | sense |
Token ID | 2565 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
breaching | +0.263 |
etheus | +0.239 |
Eid | +0.236 |
plets | +0.235 |
horns | +0.225 |
Neg logit contributions | |
Balanced | -0.270 |
Dominican | -0.247 |
Nicarag | -0.247 |
Augusta | -0.239 |
KB | -0.234 |
Token | of |
Token ID | 286 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
Balanced | +0.574 |
Dominican | +0.519 |
Nicarag | +0.518 |
Augusta | +0.499 |
KB | +0.493 |
Neg logit contributions | |
breaching | -0.728 |
etheus | -0.668 |
Eid | -0.662 |
plets | -0.654 |
hump | -0.637 |
Token | who |
Token ID | 508 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
Balanced | +0.122 |
Dominican | +0.111 |
Nicarag | +0.111 |
Augusta | +0.108 |
KB | +0.105 |
Neg logit contributions | |
breaching | -0.127 |
etheus | -0.116 |
Eid | -0.114 |
plets | -0.113 |
hump | -0.110 |
Token | keeps |
Token ID | 7622 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
Balanced | +0.063 |
Dominican | +0.057 |
Nicarag | +0.057 |
Augusta | +0.055 |
KB | +0.054 |
Neg logit contributions | |
breaching | -0.076 |
etheus | -0.070 |
Eid | -0.069 |
plets | -0.069 |
hump | -0.066 |
Token | Mech |
Token ID | 18202 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
breaching | +0.663 |
etheus | +0.609 |
Eid | +0.601 |
plets | +0.596 |
hump | +0.581 |
Neg logit contributions | |
Balanced | -0.505 |
Dominican | -0.455 |
Nicarag | -0.454 |
Augusta | -0.438 |
KB | -0.431 |
Token | who |
Token ID | 508 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
Balanced | +0.206 |
Dominican | +0.186 |
Nicarag | +0.186 |
Augusta | +0.180 |
KB | +0.180 |
Neg logit contributions | |
breaching | -0.238 |
etheus | -0.217 |
Eid | -0.215 |
plets | -0.212 |
horns | -0.207 |
Token | is |
Token ID | 318 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Balanced | +0.130 |
Dominican | +0.117 |
Nicarag | +0.117 |
Augusta | +0.112 |
KB | +0.110 |
Neg logit contributions | |
breaching | -0.177 |
etheus | -0.163 |
Eid | -0.161 |
plets | -0.160 |
hump | -0.155 |
Token | still |
Token ID | 991 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
breaching | +0.236 |
etheus | +0.218 |
plets | +0.216 |
Eid | +0.214 |
hump | +0.205 |
Neg logit contributions | |
Balanced | -0.230 |
Dominican | -0.211 |
Nicarag | -0.211 |
Augusta | -0.204 |
KB | -0.198 |
Token | conducting |
Token ID | 14523 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Balanced | +0.268 |
Dominican | +0.245 |
Nicarag | +0.245 |
Augusta | +0.237 |
KB | +0.231 |
Neg logit contributions | |
breaching | -0.294 |
etheus | -0.273 |
plets | -0.269 |
Eid | -0.267 |
horns | -0.255 |
Token | the |
Token ID | 262 |
Feature activation | -6.48e-02 |
Pos logit contributions | |
Balanced | +0.544 |
Nicarag | +0.504 |
Dominican | +0.503 |
Augusta | +0.489 |
KB | +0.479 |
Neg logit contributions | |
breaching | -0.423 |
etheus | -0.383 |
plets | -0.378 |
Eid | -0.374 |
horns | -0.353 |
Token | most |
Token ID | 749 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
breaching | +0.125 |
etheus | +0.114 |
plets | +0.113 |
Eid | +0.112 |
hump | +0.107 |
Neg logit contributions | |
Balanced | -0.131 |
Nicarag | -0.120 |
Dominican | -0.120 |
Augusta | -0.116 |
KB | -0.113 |
Token | detailed |
Token ID | 6496 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
breaching | +0.154 |
etheus | +0.138 |
Eid | +0.136 |
plets | +0.135 |
hump | +0.130 |
Neg logit contributions | |
Balanced | -0.189 |
Nicarag | -0.174 |
Dominican | -0.174 |
Augusta | -0.169 |
KB | -0.167 |
Token | and |
Token ID | 290 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
breaching | +0.343 |
etheus | +0.313 |
Eid | +0.308 |
plets | +0.308 |
hump | +0.292 |
Neg logit contributions | |
Balanced | -0.373 |
Nicarag | -0.343 |
Dominican | -0.343 |
Augusta | -0.332 |
KB | -0.326 |
Token | longest |
Token ID | 14069 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
breaching | +0.107 |
etheus | +0.097 |
Eid | +0.096 |
plets | +0.095 |
hump | +0.091 |
Neg logit contributions | |
Balanced | -0.122 |
Dominican | -0.113 |
Nicarag | -0.112 |
Augusta | -0.109 |
KB | -0.108 |
Token | term |
Token ID | 3381 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
breaching | +0.996 |
etheus | +0.909 |
Eid | +0.901 |
plets | +0.890 |
hump | +0.865 |
Neg logit contributions | |
Balanced | -0.868 |
Dominican | -0.788 |
Nicarag | -0.788 |
Augusta | -0.760 |
KB | -0.751 |
Token | video |
Token ID | 2008 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
Balanced | +0.491 |
Dominican | +0.442 |
Nicarag | +0.440 |
Augusta | +0.429 |
KB | +0.420 |
Neg logit contributions | |
breaching | -0.624 |
etheus | -0.581 |
plets | -0.565 |
Eid | -0.564 |
hump | -0.546 |
Token | , |
Token ID | 11 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
breaching | +0.110 |
etheus | +0.101 |
Eid | +0.099 |
plets | +0.099 |
hump | +0.095 |
Neg logit contributions | |
Balanced | -0.089 |
Dominican | -0.081 |
Nicarag | -0.081 |
Augusta | -0.079 |
KB | -0.077 |
Token | you |
Token ID | 345 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
Balanced | +0.091 |
Dominican | +0.082 |
Nicarag | +0.082 |
Augusta | +0.080 |
KB | +0.078 |
Neg logit contributions | |
breaching | -0.102 |
etheus | -0.094 |
Eid | -0.092 |
plets | -0.092 |
hump | -0.089 |
Token | just |
Token ID | 655 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
breaching | +0.428 |
etheus | +0.392 |
Eid | +0.387 |
plets | +0.382 |
hump | +0.371 |
Neg logit contributions | |
Balanced | -0.382 |
Dominican | -0.348 |
Nicarag | -0.347 |
Augusta | -0.336 |
KB | -0.332 |
Token | might |
Token ID | 1244 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
breaching | +0.791 |
etheus | +0.717 |
Eid | +0.713 |
plets | +0.702 |
hump | +0.685 |
Neg logit contributions | |
Balanced | -0.745 |
Nicarag | -0.680 |
Dominican | -0.679 |
Augusta | -0.657 |
KB | -0.649 |
Token | be |
Token ID | 307 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
breaching | +1.090 |
etheus | +0.997 |
Eid | +0.989 |
plets | +0.976 |
hump | +0.952 |
Neg logit contributions | |
Balanced | -0.897 |
Dominican | -0.812 |
Nicarag | -0.811 |
Augusta | -0.782 |
KB | -0.773 |
Token | featured |
Token ID | 8096 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
breaching | +0.439 |
etheus | +0.403 |
Eid | +0.394 |
plets | +0.389 |
hump | +0.374 |
Neg logit contributions | |
Balanced | -0.506 |
Dominican | -0.470 |
Nicarag | -0.469 |
Augusta | -0.455 |
KB | -0.447 |
Token | on |
Token ID | 319 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
Balanced | +0.206 |
Dominican | +0.185 |
Nicarag | +0.185 |
Augusta | +0.177 |
KB | +0.175 |
Neg logit contributions | |
breaching | -0.310 |
etheus | -0.287 |
Eid | -0.284 |
plets | -0.281 |
hump | -0.274 |
Token | the |
Token ID | 262 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
Balanced | +0.031 |
Dominican | +0.028 |
Nicarag | +0.028 |
Augusta | +0.027 |
KB | +0.027 |
Neg logit contributions | |
breaching | -0.034 |
etheus | -0.031 |
Eid | -0.030 |
plets | -0.030 |
hump | -0.029 |
Token | Warriors |
Token ID | 12090 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Balanced | +0.060 |
Dominican | +0.055 |
Nicarag | +0.055 |
Augusta | +0.053 |
KB | +0.052 |
Neg logit contributions | |
breaching | -0.069 |
etheus | -0.063 |
plets | -0.062 |
Eid | -0.062 |
hump | -0.060 |
Token | scoreboard |
Token ID | 50198 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
breaching | +0.349 |
etheus | +0.320 |
Eid | +0.316 |
plets | +0.312 |
hump | +0.304 |
Neg logit contributions | |
Balanced | -0.303 |
Dominican | -0.275 |
Nicarag | -0.275 |
Augusta | -0.265 |
KB | -0.263 |
Token | the |
Token ID | 262 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
Balanced | +0.260 |
Dominican | +0.237 |
Nicarag | +0.236 |
Augusta | +0.229 |
KB | +0.226 |
Neg logit contributions | |
breaching | -0.273 |
etheus | -0.251 |
Eid | -0.246 |
plets | -0.246 |
hump | -0.236 |
Token | bill |
Token ID | 2855 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
Balanced | +0.046 |
Dominican | +0.042 |
Nicarag | +0.042 |
Augusta | +0.040 |
KB | +0.040 |
Neg logit contributions | |
breaching | -0.051 |
etheus | -0.047 |
Eid | -0.046 |
plets | -0.046 |
hump | -0.044 |
Token | as |
Token ID | 355 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
breaching | +0.467 |
etheus | +0.429 |
Eid | +0.425 |
plets | +0.420 |
hump | +0.409 |
Neg logit contributions | |
Balanced | -0.360 |
Dominican | -0.325 |
Nicarag | -0.325 |
Augusta | -0.313 |
KB | -0.309 |
Token | an |
Token ID | 281 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
breaching | +0.030 |
etheus | +0.028 |
plets | +0.027 |
Eid | +0.027 |
hump | +0.026 |
Neg logit contributions | |
Balanced | -0.033 |
Dominican | -0.030 |
Nicarag | -0.030 |
Augusta | -0.029 |
KB | -0.029 |
Token | anti |
Token ID | 3098 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
breaching | +0.254 |
etheus | +0.229 |
Eid | +0.224 |
plets | +0.222 |
hump | +0.215 |
Neg logit contributions | |
Balanced | -0.321 |
Dominican | -0.297 |
Nicarag | -0.297 |
Augusta | -0.289 |
KB | -0.285 |
Token | - |
Token ID | 12 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
Balanced | +0.505 |
Dominican | +0.449 |
Nicarag | +0.448 |
Augusta | +0.430 |
KB | +0.427 |
Neg logit contributions | |
breaching | -0.845 |
etheus | -0.783 |
Eid | -0.775 |
plets | -0.768 |
hump | -0.749 |
Token | trans |
Token ID | 7645 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
breaching | +0.398 |
etheus | +0.364 |
Eid | +0.359 |
plets | +0.353 |
hump | +0.345 |
Neg logit contributions | |
Balanced | -0.342 |
Dominican | -0.312 |
Nicarag | -0.310 |
KB | -0.302 |
Augusta | -0.300 |
Token | measure |
Token ID | 3953 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
Balanced | +0.647 |
Nicarag | +0.587 |
Dominican | +0.586 |
Augusta | +0.564 |
KB | +0.559 |
Neg logit contributions | |
breaching | -0.779 |
etheus | -0.712 |
Eid | -0.708 |
plets | -0.705 |
hump | -0.676 |
Token | intended |
Token ID | 5292 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
breaching | +0.084 |
etheus | +0.077 |
Eid | +0.076 |
plets | +0.076 |
hump | +0.073 |
Neg logit contributions | |
Balanced | -0.065 |
Dominican | -0.059 |
Nicarag | -0.059 |
Augusta | -0.057 |
KB | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | +7.24e-02 |
Pos logit contributions | |
Balanced | +0.196 |
Nicarag | +0.180 |
Dominican | +0.180 |
Augusta | +0.174 |
KB | +0.173 |
Neg logit contributions | |
breaching | -0.175 |
etheus | -0.157 |
Eid | -0.155 |
plets | -0.153 |
hump | -0.149 |
Token | discriminate |
Token ID | 28433 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Balanced | +0.141 |
Dominican | +0.129 |
Nicarag | +0.129 |
Augusta | +0.125 |
KB | +0.123 |
Neg logit contributions | |
breaching | -0.144 |
etheus | -0.132 |
Eid | -0.130 |
plets | -0.128 |
hump | -0.124 |
Token | what |
Token ID | 644 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
breaching | +0.021 |
etheus | +0.019 |
Eid | +0.019 |
plets | +0.019 |
hump | +0.018 |
Neg logit contributions | |
Balanced | -0.019 |
Dominican | -0.017 |
Nicarag | -0.017 |
Augusta | -0.017 |
KB | -0.016 |
Token | I |
Token ID | 314 |
Feature activation | +7.65e-02 |
Pos logit contributions | |
breaching | +0.159 |
etheus | +0.146 |
Eid | +0.144 |
plets | +0.143 |
hump | +0.139 |
Neg logit contributions | |
Balanced | -0.121 |
Dominican | -0.109 |
Nicarag | -0.109 |
Augusta | -0.105 |
KB | -0.103 |
Token | want |
Token ID | 765 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
Balanced | +0.133 |
Dominican | +0.121 |
Nicarag | +0.120 |
Augusta | +0.116 |
KB | +0.114 |
Neg logit contributions | |
breaching | -0.169 |
etheus | -0.155 |
Eid | -0.153 |
plets | -0.151 |
hump | -0.147 |
Token | to |
Token ID | 284 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
breaching | +0.248 |
etheus | +0.226 |
Eid | +0.223 |
plets | +0.221 |
hump | +0.215 |
Neg logit contributions | |
Balanced | -0.216 |
Dominican | -0.197 |
Nicarag | -0.196 |
Augusta | -0.189 |
KB | -0.187 |
Token | be |
Token ID | 307 |
Feature activation | -7.91e-03 |
Pos logit contributions | |
Balanced | +0.196 |
Dominican | +0.177 |
Nicarag | +0.177 |
Augusta | +0.170 |
KB | +0.167 |
Neg logit contributions | |
breaching | -0.268 |
etheus | -0.247 |
Eid | -0.242 |
plets | -0.240 |
horns | -0.233 |
Token | known |
Token ID | 1900 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
breaching | +0.017 |
etheus | +0.016 |
Eid | +0.016 |
plets | +0.016 |
hump | +0.015 |
Neg logit contributions | |
Balanced | -0.014 |
Dominican | -0.013 |
Nicarag | -0.013 |
Augusta | -0.012 |
KB | -0.012 |
Token | for |
Token ID | 329 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
breaching | +0.610 |
etheus | +0.561 |
Eid | +0.557 |
plets | +0.550 |
hump | +0.537 |
Neg logit contributions | |
Balanced | -0.438 |
Dominican | -0.392 |
Nicarag | -0.392 |
Augusta | -0.376 |
KB | -0.372 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Balanced | +0.351 |
Dominican | +0.318 |
Nicarag | +0.317 |
Augusta | +0.306 |
KB | +0.301 |
Neg logit contributions | |
breaching | -0.451 |
etheus | -0.416 |
Eid | -0.409 |
plets | -0.406 |
hump | -0.395 |
Token | � |
Token ID | 447 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
breaching | +0.033 |
etheus | +0.030 |
Eid | +0.030 |
plets | +0.029 |
hump | +0.029 |
Neg logit contributions | |
Balanced | -0.024 |
Dominican | -0.022 |
Nicarag | -0.022 |
Augusta | -0.021 |
KB | -0.020 |
Token | � |
Token ID | 251 |
Feature activation | -8.82e-02 |
Pos logit contributions | |
breaching | +0.109 |
etheus | +0.100 |
Eid | +0.098 |
plets | +0.097 |
hump | +0.095 |
Neg logit contributions | |
Balanced | -0.091 |
Dominican | -0.082 |
Nicarag | -0.082 |
Augusta | -0.079 |
KB | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
breaching | +0.196 |
etheus | +0.180 |
Eid | +0.177 |
plets | +0.176 |
hump | +0.171 |
Neg logit contributions | |
Balanced | -0.154 |
Dominican | -0.138 |
Nicarag | -0.138 |
Augusta | -0.133 |
KB | -0.131 |
Token | is |
Token ID | 318 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
breaching | +0.174 |
etheus | +0.159 |
Eid | +0.157 |
plets | +0.155 |
hump | +0.151 |
Neg logit contributions | |
Balanced | -0.166 |
Dominican | -0.151 |
Nicarag | -0.151 |
Augusta | -0.146 |
KB | -0.145 |
Token | best |
Token ID | 1266 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
breaching | +0.454 |
etheus | +0.415 |
Eid | +0.408 |
plets | +0.403 |
hump | +0.391 |
Neg logit contributions | |
Balanced | -0.451 |
Dominican | -0.412 |
Nicarag | -0.412 |
Augusta | -0.399 |
KB | -0.393 |
Token | to |
Token ID | 284 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
breaching | +0.412 |
etheus | +0.373 |
Eid | +0.369 |
plets | +0.362 |
hump | +0.352 |
Neg logit contributions | |
Balanced | -0.428 |
Dominican | -0.392 |
Nicarag | -0.392 |
Augusta | -0.379 |
KB | -0.375 |
Token | wash |
Token ID | 13502 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
Balanced | +0.244 |
Dominican | +0.222 |
Nicarag | +0.221 |
Augusta | +0.213 |
KB | +0.211 |
Neg logit contributions | |
breaching | -0.271 |
etheus | -0.248 |
Eid | -0.244 |
plets | -0.240 |
horns | -0.233 |
Token | the |
Token ID | 262 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
breaching | +0.071 |
etheus | +0.065 |
Eid | +0.064 |
plets | +0.063 |
hump | +0.061 |
Neg logit contributions | |
Balanced | -0.064 |
Nicarag | -0.058 |
Dominican | -0.058 |
Augusta | -0.056 |
KB | -0.056 |
Token | Tu |
Token ID | 16749 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
Balanced | +0.174 |
Nicarag | +0.158 |
Dominican | +0.158 |
Augusta | +0.152 |
KB | +0.150 |
Neg logit contributions | |
breaching | -0.201 |
etheus | -0.185 |
Eid | -0.182 |
plets | -0.180 |
hump | -0.175 |
Token | pper |
Token ID | 2848 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
breaching | +0.512 |
etheus | +0.470 |
Eid | +0.466 |
plets | +0.458 |
hump | +0.447 |
Neg logit contributions | |
Balanced | -0.395 |
Nicarag | -0.358 |
Dominican | -0.357 |
Augusta | -0.344 |
KB | -0.342 |
Token | ware |
Token ID | 1574 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
breaching | +1.684 |
etheus | +1.543 |
Eid | +1.529 |
plets | +1.511 |
hump | +1.475 |
Neg logit contributions | |
Balanced | -1.321 |
Dominican | -1.195 |
Nicarag | -1.193 |
Augusta | -1.149 |
KB | -1.134 |
Token | container |
Token ID | 9290 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
breaching | +0.067 |
etheus | +0.062 |
Eid | +0.061 |
plets | +0.060 |
hump | +0.059 |
Neg logit contributions | |
Balanced | -0.051 |
Nicarag | -0.046 |
Dominican | -0.046 |
Augusta | -0.044 |
KB | -0.044 |
Token | completely |
Token ID | 3190 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
Balanced | +0.546 |
Nicarag | +0.495 |
Dominican | +0.493 |
Augusta | +0.474 |
Container | +0.468 |
Neg logit contributions | |
breaching | -0.693 |
etheus | -0.643 |
Eid | -0.632 |
plets | -0.624 |
hump | -0.606 |
Token | . |
Token ID | 13 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
breaching | +0.035 |
etheus | +0.033 |
Eid | +0.032 |
plets | +0.032 |
hump | +0.031 |
Neg logit contributions | |
Balanced | -0.027 |
Dominican | -0.025 |
Nicarag | -0.025 |
Augusta | -0.024 |
KB | -0.023 |
Token | / |
Token ID | 14 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
Balanced | +0.354 |
Dominican | +0.321 |
Nicarag | +0.321 |
Augusta | +0.310 |
KB | +0.307 |
Neg logit contributions | |
breaching | -0.418 |
etheus | -0.383 |
Eid | -0.379 |
plets | -0.374 |
hump | -0.364 |
Token | kg |
Token ID | 10025 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
Balanced | +0.056 |
Dominican | +0.051 |
Nicarag | +0.051 |
Augusta | +0.049 |
KB | +0.048 |
Neg logit contributions | |
breaching | -0.072 |
etheus | -0.066 |
Eid | -0.065 |
plets | -0.064 |
hump | -0.063 |
Token | ), |
Token ID | 828 |
Feature activation | +2.00e-02 |
Pos logit contributions | |
Balanced | +0.120 |
Dominican | +0.109 |
Nicarag | +0.109 |
Augusta | +0.106 |
KB | +0.105 |
Neg logit contributions | |
breaching | -0.118 |
etheus | -0.107 |
Eid | -0.106 |
plets | -0.105 |
hump | -0.102 |
Token | a |
Token ID | 257 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Balanced | +0.036 |
Dominican | +0.033 |
Nicarag | +0.033 |
Augusta | +0.032 |
KB | +0.031 |
Neg logit contributions | |
breaching | -0.043 |
etheus | -0.039 |
Eid | -0.039 |
plets | -0.038 |
hump | -0.037 |
Token | selective |
Token ID | 21792 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
breaching | +0.204 |
etheus | +0.187 |
Eid | +0.184 |
plets | +0.182 |
horns | +0.175 |
Neg logit contributions | |
Balanced | -0.213 |
Dominican | -0.196 |
Nicarag | -0.196 |
Augusta | -0.189 |
KB | -0.187 |
Token | serotonin |
Token ID | 32236 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
breaching | +0.059 |
etheus | +0.055 |
plets | +0.054 |
Eid | +0.053 |
hump | +0.051 |
Neg logit contributions | |
Balanced | -0.057 |
Dominican | -0.052 |
Nicarag | -0.052 |
Augusta | -0.050 |
KB | -0.050 |
Token | re |
Token ID | 302 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
breaching | +1.674 |
etheus | +1.498 |
Eid | +1.478 |
plets | +1.455 |
hump | +1.404 |
Neg logit contributions | |
Balanced | -2.214 |
Dominican | -2.048 |
Nicarag | -2.046 |
Augusta | -1.990 |
KB | -1.969 |
Token | upt |
Token ID | 37623 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
breaching | +0.107 |
etheus | +0.096 |
Eid | +0.095 |
plets | +0.093 |
hump | +0.091 |
Neg logit contributions | |
Balanced | -0.135 |
Nicarag | -0.125 |
Dominican | -0.125 |
Augusta | -0.121 |
KB | -0.121 |
Token | ake |
Token ID | 539 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Balanced | +0.153 |
Dominican | +0.132 |
Nicarag | +0.132 |
Augusta | +0.125 |
KB | +0.123 |
Neg logit contributions | |
breaching | -0.321 |
etheus | -0.299 |
Eid | -0.297 |
plets | -0.293 |
hump | -0.288 |
Token | inhibitor |
Token ID | 41704 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
Balanced | +0.301 |
Dominican | +0.274 |
Nicarag | +0.273 |
Augusta | +0.264 |
KB | +0.262 |
Neg logit contributions | |
breaching | -0.346 |
etheus | -0.320 |
Eid | -0.313 |
plets | -0.312 |
hump | -0.301 |
Token | ( |
Token ID | 357 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
Balanced | +0.047 |
Dominican | +0.042 |
Nicarag | +0.042 |
Augusta | +0.041 |
KB | +0.040 |
Neg logit contributions | |
breaching | -0.055 |
etheus | -0.051 |
Eid | -0.050 |
plets | -0.050 |
hump | -0.048 |
Token | toward |
Token ID | 3812 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
breaching | +0.509 |
etheus | +0.467 |
Eid | +0.457 |
plets | +0.456 |
hump | +0.436 |
Neg logit contributions | |
Balanced | -0.487 |
Dominican | -0.447 |
Nicarag | -0.446 |
Augusta | -0.432 |
KB | -0.422 |
Token | extinction |
Token ID | 21935 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
breaching | +0.542 |
etheus | +0.496 |
plets | +0.482 |
Eid | +0.478 |
hump | +0.451 |
Neg logit contributions | |
Balanced | -0.694 |
Dominican | -0.645 |
Nicarag | -0.641 |
Augusta | -0.626 |
KB | -0.611 |
Token |
|
Token ID | 198 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
Balanced | +0.255 |
Dominican | +0.230 |
Nicarag | +0.230 |
Augusta | +0.221 |
KB | +0.217 |
Neg logit contributions | |
breaching | -0.342 |
etheus | -0.315 |
Eid | -0.311 |
plets | -0.310 |
hump | -0.299 |
Token |
|
Token ID | 198 |
Feature activation | +4.46e-02 |
Pos logit contributions | |
breaching | +0.090 |
etheus | +0.083 |
Eid | +0.082 |
plets | +0.081 |
hump | +0.079 |
Neg logit contributions | |
Balanced | -0.068 |
Nicarag | -0.061 |
Dominican | -0.061 |
Augusta | -0.059 |
KB | -0.058 |
Token | How |
Token ID | 2437 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Balanced | +0.072 |
Dominican | +0.065 |
Nicarag | +0.065 |
KB | +0.063 |
Augusta | +0.062 |
Neg logit contributions | |
breaching | -0.105 |
etheus | -0.096 |
Eid | -0.095 |
plets | -0.094 |
hump | -0.091 |
Token | Vera |
Token ID | 44413 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
Balanced | +0.279 |
Dominican | +0.258 |
Nicarag | +0.258 |
Augusta | +0.251 |
KB | +0.243 |
Neg logit contributions | |
breaching | -0.255 |
etheus | -0.235 |
plets | -0.229 |
Eid | -0.227 |
horns | -0.215 |
Token | Rubin |
Token ID | 34599 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
breaching | +1.998 |
etheus | +1.842 |
Eid | +1.826 |
plets | +1.804 |
hump | +1.762 |
Neg logit contributions | |
Balanced | -1.376 |
Dominican | -1.234 |
Nicarag | -1.232 |
Augusta | -1.183 |
KB | -1.166 |
Token | changed |
Token ID | 3421 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
breaching | +0.498 |
etheus | +0.456 |
Eid | +0.452 |
plets | +0.447 |
hump | +0.436 |
Neg logit contributions | |
Balanced | -0.403 |
Dominican | -0.364 |
Nicarag | -0.364 |
Augusta | -0.351 |
KB | -0.347 |
Token | science |
Token ID | 3783 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
breaching | +0.511 |
etheus | +0.465 |
Eid | +0.457 |
plets | +0.454 |
hump | +0.436 |
Neg logit contributions | |
Balanced | -0.508 |
Dominican | -0.467 |
Nicarag | -0.466 |
Augusta | -0.452 |
KB | -0.442 |
Token |
|
Token ID | 198 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
breaching | +0.070 |
etheus | +0.064 |
Eid | +0.063 |
plets | +0.063 |
hump | +0.061 |
Neg logit contributions | |
Balanced | -0.053 |
Dominican | -0.048 |
Nicarag | -0.048 |
Augusta | -0.046 |
KB | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | +8.97e-02 |
Pos logit contributions | |
Balanced | +0.053 |
Dominican | +0.048 |
Nicarag | +0.048 |
Augusta | +0.046 |
KB | +0.045 |
Neg logit contributions | |
breaching | -0.071 |
etheus | -0.065 |
Eid | -0.065 |
plets | -0.064 |
hump | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
breaching | +1.296 |
Eid | +1.195 |
etheus | +1.183 |
plets | +1.159 |
hump | +1.141 |
Neg logit contributions | |
Balanced | -1.013 |
Nicarag | -0.916 |
Dominican | -0.915 |
Augusta | -0.875 |
KB | -0.871 |
Token |
|
Token ID | 198 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
breaching | +0.327 |
Eid | +0.299 |
etheus | +0.299 |
plets | +0.294 |
hump | +0.287 |
Neg logit contributions | |
Balanced | -0.253 |
Nicarag | -0.228 |
Dominican | -0.228 |
Augusta | -0.219 |
KB | -0.216 |
Token | Hon |
Token ID | 29478 |
Feature activation | +4.18e-03 |
Pos logit contributions | |
breaching | +0.162 |
etheus | +0.148 |
Eid | +0.146 |
plets | +0.145 |
hump | +0.141 |
Neg logit contributions | |
Balanced | -0.130 |
Dominican | -0.118 |
Nicarag | -0.118 |
KB | -0.114 |
Augusta | -0.114 |
Token | orable |
Token ID | 10475 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Balanced | +0.007 |
Dominican | +0.007 |
Nicarag | +0.006 |
Augusta | +0.006 |
KB | +0.006 |
Neg logit contributions | |
breaching | -0.009 |
etheus | -0.009 |
Eid | -0.008 |
plets | -0.008 |
hump | -0.008 |
Token | M |
Token ID | 337 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
breaching | +0.205 |
etheus | +0.188 |
plets | +0.182 |
Eid | +0.181 |
hump | +0.176 |
Neg logit contributions | |
Balanced | -0.212 |
Dominican | -0.195 |
Nicarag | -0.194 |
Augusta | -0.189 |
KB | -0.184 |
Token | ention |
Token ID | 1463 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
Balanced | +0.051 |
Dominican | +0.047 |
Nicarag | +0.046 |
Augusta | +0.045 |
KB | +0.044 |
Neg logit contributions | |
breaching | -0.052 |
etheus | -0.047 |
Eid | -0.046 |
plets | -0.045 |
hump | -0.045 |
Token | : |
Token ID | 25 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
breaching | +2.015 |
etheus | +1.849 |
Eid | +1.837 |
plets | +1.811 |
hump | +1.768 |
Neg logit contributions | |
Balanced | -1.553 |
Dominican | -1.401 |
Nicarag | -1.401 |
Augusta | -1.347 |
KB | -1.332 |
Token |
|
Token ID | 198 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
breaching | +0.252 |
etheus | +0.231 |
Eid | +0.227 |
plets | +0.226 |
hump | +0.220 |
Neg logit contributions | |
Balanced | -0.196 |
Dominican | -0.177 |
Nicarag | -0.177 |
Augusta | -0.171 |
KB | -0.168 |
Token |
|
Token ID | 198 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
Balanced | +0.067 |
Nicarag | +0.060 |
Dominican | +0.060 |
Augusta | +0.058 |
KB | +0.057 |
Neg logit contributions | |
breaching | -0.089 |
etheus | -0.081 |
Eid | -0.081 |
plets | -0.080 |
hump | -0.078 |
Token | 706 |
Token ID | 35402 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
Balanced | +0.169 |
Dominican | +0.153 |
Nicarag | +0.152 |
KB | +0.148 |
Augusta | +0.147 |
Neg logit contributions | |
breaching | -0.229 |
etheus | -0.209 |
Eid | -0.207 |
plets | -0.205 |
hump | -0.200 |
Token | Main |
Token ID | 8774 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
breaching | +0.139 |
etheus | +0.128 |
Eid | +0.126 |
plets | +0.125 |
hump | +0.122 |
Neg logit contributions | |
Balanced | -0.100 |
Dominican | -0.090 |
Nicarag | -0.090 |
Augusta | -0.087 |
KB | -0.085 |
Token | its |
Token ID | 663 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
Balanced | +0.158 |
Dominican | +0.144 |
Nicarag | +0.144 |
Augusta | +0.139 |
Container | +0.137 |
Neg logit contributions | |
breaching | -0.166 |
etheus | -0.153 |
plets | -0.149 |
Eid | -0.149 |
hump | -0.143 |
Token | food |
Token ID | 2057 |
Feature activation | -8.49e-02 |
Pos logit contributions | |
breaching | +0.537 |
etheus | +0.492 |
Eid | +0.487 |
plets | +0.482 |
hump | +0.469 |
Neg logit contributions | |
Balanced | -0.430 |
Dominican | -0.389 |
Nicarag | -0.389 |
Augusta | -0.374 |
KB | -0.370 |
Token | - |
Token ID | 12 |
Feature activation | -5.30e-03 |
Pos logit contributions | |
breaching | +0.162 |
etheus | +0.146 |
Eid | +0.144 |
plets | +0.143 |
hump | +0.138 |
Neg logit contributions | |
Balanced | -0.173 |
Dominican | -0.159 |
Nicarag | -0.158 |
Augusta | -0.154 |
KB | -0.152 |
Token | ordering |
Token ID | 34555 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
breaching | +0.011 |
etheus | +0.010 |
Eid | +0.010 |
plets | +0.009 |
hump | +0.009 |
Neg logit contributions | |
Balanced | -0.010 |
Dominican | -0.009 |
Nicarag | -0.009 |
Augusta | -0.009 |
KB | -0.009 |
Token | mobile |
Token ID | 5175 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
breaching | +0.633 |
etheus | +0.580 |
Eid | +0.574 |
plets | +0.568 |
hump | +0.553 |
Neg logit contributions | |
Balanced | -0.517 |
Dominican | -0.468 |
Nicarag | -0.468 |
Augusta | -0.451 |
KB | -0.445 |
Token | application |
Token ID | 3586 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
Balanced | +0.146 |
Dominican | +0.130 |
Nicarag | +0.129 |
Augusta | +0.124 |
KB | +0.122 |
Neg logit contributions | |
breaching | -0.238 |
etheus | -0.221 |
Eid | -0.218 |
plets | -0.217 |
hump | -0.210 |
Token | . |
Token ID | 13 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
breaching | +1.809 |
etheus | +1.651 |
Eid | +1.647 |
plets | +1.618 |
hump | +1.579 |
Neg logit contributions | |
Balanced | -1.472 |
Nicarag | -1.336 |
Dominican | -1.335 |
Augusta | -1.287 |
KB | -1.272 |
Token | A |
Token ID | 317 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
breaching | +0.183 |
etheus | +0.168 |
Eid | +0.166 |
plets | +0.164 |
hump | +0.160 |
Neg logit contributions | |
Balanced | -0.143 |
Dominican | -0.129 |
Nicarag | -0.129 |
Augusta | -0.124 |
KB | -0.123 |
Token | meal |
Token ID | 9799 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
breaching | +0.110 |
etheus | +0.101 |
Eid | +0.099 |
plets | +0.099 |
horns | +0.095 |
Neg logit contributions | |
Balanced | -0.103 |
Dominican | -0.094 |
Nicarag | -0.094 |
Augusta | -0.090 |
KB | -0.090 |
Token | for |
Token ID | 329 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
Balanced | +0.248 |
Dominican | +0.225 |
Nicarag | +0.223 |
Augusta | +0.215 |
KB | +0.213 |
Neg logit contributions | |
breaching | -0.302 |
etheus | -0.279 |
Eid | -0.274 |
plets | -0.273 |
hump | -0.264 |
Token | two |
Token ID | 734 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
Balanced | +0.065 |
Dominican | +0.059 |
Nicarag | +0.059 |
Augusta | +0.057 |
KB | +0.056 |
Neg logit contributions | |
breaching | -0.070 |
etheus | -0.065 |
plets | -0.063 |
Eid | -0.063 |
horns | -0.061 |
Token | d |
Token ID | 67 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
breaching | +0.442 |
etheus | +0.404 |
Eid | +0.400 |
plets | +0.394 |
hump | +0.384 |
Neg logit contributions | |
Balanced | -0.385 |
Dominican | -0.349 |
Nicarag | -0.349 |
Augusta | -0.337 |
KB | -0.333 |
Token | aughters |
Token ID | 13441 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
breaching | +0.219 |
etheus | +0.194 |
Eid | +0.192 |
plets | +0.187 |
hump | +0.181 |
Neg logit contributions | |
Balanced | -0.322 |
Dominican | -0.300 |
Nicarag | -0.299 |
Augusta | -0.291 |
KB | -0.289 |
Token | died |
Token ID | 3724 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
Balanced | +0.605 |
Dominican | +0.547 |
Nicarag | +0.544 |
Augusta | +0.525 |
KB | +0.515 |
Neg logit contributions | |
breaching | -0.807 |
etheus | -0.745 |
Eid | -0.732 |
plets | -0.725 |
hump | -0.705 |
Token | at |
Token ID | 379 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
breaching | +0.750 |
etheus | +0.689 |
Eid | +0.682 |
plets | +0.674 |
hump | +0.657 |
Neg logit contributions | |
Balanced | -0.584 |
Dominican | -0.529 |
Nicarag | -0.528 |
Augusta | -0.509 |
KB | -0.501 |
Token | the |
Token ID | 262 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
breaching | +0.106 |
etheus | +0.097 |
plets | +0.095 |
Eid | +0.095 |
hump | +0.092 |
Neg logit contributions | |
Balanced | -0.099 |
Dominican | -0.091 |
Nicarag | -0.090 |
Augusta | -0.089 |
Container | -0.086 |
Token | ages |
Token ID | 9337 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
breaching | +0.336 |
etheus | +0.308 |
Eid | +0.302 |
plets | +0.300 |
hump | +0.290 |
Neg logit contributions | |
Balanced | -0.316 |
Dominican | -0.290 |
Nicarag | -0.289 |
Augusta | -0.280 |
KB | -0.274 |
Token | of |
Token ID | 286 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
breaching | +1.756 |
etheus | +1.603 |
Eid | +1.596 |
plets | +1.571 |
hump | +1.536 |
Neg logit contributions | |
Balanced | -1.421 |
Nicarag | -1.283 |
Dominican | -1.282 |
Augusta | -1.236 |
KB | -1.224 |
Token | 12 |
Token ID | 1105 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
breaching | +0.244 |
etheus | +0.226 |
Eid | +0.223 |
plets | +0.221 |
hump | +0.215 |
Neg logit contributions | |
Balanced | -0.180 |
Dominican | -0.162 |
Nicarag | -0.162 |
Augusta | -0.156 |
KB | -0.153 |
Token | and |
Token ID | 290 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
breaching | +0.236 |
etheus | +0.217 |
Eid | +0.215 |
plets | +0.212 |
hump | +0.207 |
Neg logit contributions | |
Balanced | -0.169 |
Dominican | -0.152 |
Nicarag | -0.151 |
Augusta | -0.145 |
KB | -0.143 |
Token | 18 |
Token ID | 1248 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
breaching | +0.296 |
etheus | +0.272 |
Eid | +0.269 |
plets | +0.266 |
hump | +0.259 |
Neg logit contributions | |
Balanced | -0.223 |
Dominican | -0.201 |
Nicarag | -0.201 |
Augusta | -0.193 |
KB | -0.190 |
Token | , |
Token ID | 11 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
breaching | +0.095 |
etheus | +0.088 |
Eid | +0.087 |
plets | +0.086 |
hump | +0.084 |
Neg logit contributions | |
Balanced | -0.076 |
Dominican | -0.069 |
Nicarag | -0.069 |
Augusta | -0.067 |
KB | -0.066 |