Token | r |
Token ID | 81 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
DEFENSE | +1.288 |
Vert | +1.149 |
HQ | +0.966 |
Administrator | +0.958 |
Crest | +0.955 |
Neg logit contributions | |
isa | -1.226 |
awa | -1.201 |
favour | -1.196 |
ophon | -1.182 |
ethical | -1.112 |
Token | ga |
Token ID | 4908 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
DEFENSE | +0.761 |
Vert | +0.690 |
HQ | +0.597 |
Crest | +0.595 |
Administrator | +0.595 |
Neg logit contributions | |
isa | -0.505 |
awa | -0.487 |
favour | -0.484 |
ophon | -0.480 |
ethical | -0.443 |
Token |
|
Token ID | 198 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
isa | +0.401 |
awa | +0.387 |
favour | +0.384 |
ophon | +0.382 |
swat | +0.357 |
Neg logit contributions | |
DEFENSE | -0.466 |
Vert | -0.419 |
HQ | -0.354 |
Crest | -0.353 |
Administrator | -0.352 |
Token |
|
Token ID | 198 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
DEFENSE | +0.790 |
Vert | +0.708 |
HQ | +0.600 |
Administrator | +0.597 |
Crest | +0.597 |
Neg logit contributions | |
isa | -0.675 |
awa | -0.656 |
favour | -0.652 |
ophon | -0.647 |
ethical | -0.605 |
Token | Read |
Token ID | 5569 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
DEFENSE | +0.561 |
Vert | +0.500 |
HQ | +0.421 |
Crest | +0.421 |
Administrator | +0.421 |
Neg logit contributions | |
isa | -0.509 |
awa | -0.491 |
favour | -0.488 |
ophon | -0.487 |
ethical | -0.456 |
Token | Next |
Token ID | 7406 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
DEFENSE | +0.157 |
Vert | +0.140 |
HQ | +0.117 |
Administrator | +0.117 |
Crest | +0.117 |
Neg logit contributions | |
isa | -0.153 |
awa | -0.149 |
favour | -0.148 |
ophon | -0.147 |
ethical | -0.139 |
Token |
|
Token ID | 198 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
isa | +3.611 |
favour | +3.333 |
ophon | +3.301 |
awa | +3.175 |
swat | +2.972 |
Neg logit contributions | |
DEFENSE | -7.144 |
Vert | -6.589 |
HQ | -5.870 |
PDATE | -5.803 |
7601 | -5.782 |
Token |
|
Token ID | 198 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
isa | +1.260 |
favour | +1.193 |
ophon | +1.171 |
awa | +1.109 |
abet | +1.099 |
Neg logit contributions | |
DEFENSE | -1.697 |
Vert | -1.511 |
PDATE | -1.361 |
Crest | -1.295 |
HQ | -1.291 |
Token | L |
Token ID | 43 |
Feature activation | -7.50e-02 |
Pos logit contributions | |
isa | +1.539 |
ophon | +1.465 |
awa | +1.438 |
favour | +1.432 |
abet | +1.387 |
Neg logit contributions | |
DEFENSE | -1.553 |
Vert | -1.357 |
Crest | -1.160 |
Ft | -1.150 |
Administrator | -1.147 |
Token | AT |
Token ID | 1404 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
DEFENSE | +0.305 |
Vert | +0.275 |
HQ | +0.237 |
Administrator | +0.235 |
Crest | +0.235 |
Neg logit contributions | |
isa | -0.223 |
awa | -0.218 |
favour | -0.217 |
ophon | -0.214 |
ethical | -0.199 |
Token | EST |
Token ID | 6465 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
DEFENSE | +0.088 |
Vert | +0.079 |
HQ | +0.067 |
Crest | +0.067 |
Administrator | +0.066 |
Neg logit contributions | |
isa | -0.081 |
awa | -0.078 |
favour | -0.078 |
ophon | -0.077 |
abet | -0.073 |
Token | s |
Token ID | 82 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
isa | +0.486 |
favour | +0.467 |
ophon | +0.461 |
awa | +0.460 |
swat | +0.430 |
Neg logit contributions | |
DEFENSE | -0.635 |
Vert | -0.570 |
Crest | -0.486 |
Administrator | -0.486 |
HQ | -0.486 |
Token | )} |
Token ID | 38165 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
isa | +0.096 |
awa | +0.093 |
favour | +0.093 |
ophon | +0.092 |
ethical | +0.086 |
Neg logit contributions | |
DEFENSE | -0.122 |
Vert | -0.109 |
HQ | -0.093 |
Administrator | -0.093 |
Crest | -0.093 |
Token | 12 |
Token ID | 1105 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
isa | +0.894 |
awa | +0.874 |
favour | +0.865 |
ophon | +0.863 |
ethical | +0.812 |
Neg logit contributions | |
DEFENSE | -0.889 |
Vert | -0.786 |
HQ | -0.656 |
Administrator | -0.655 |
Crest | -0.655 |
Token | . |
Token ID | 13 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
DEFENSE | +2.733 |
Vert | +2.454 |
HQ | +2.065 |
Crest | +2.049 |
Administrator | +2.042 |
Neg logit contributions | |
awa | -2.650 |
isa | -2.599 |
favour | -2.544 |
ophon | -2.534 |
ethical | -2.395 |
Token | Na |
Token ID | 11013 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
isa | +0.872 |
favour | +0.847 |
awa | +0.845 |
ophon | +0.830 |
swat | +0.777 |
Neg logit contributions | |
DEFENSE | -1.059 |
Vert | -0.949 |
HQ | -0.806 |
Administrator | -0.802 |
Crest | -0.799 |
Token | 3 |
Token ID | 18 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
DEFENSE | +0.222 |
Vert | +0.195 |
HQ | +0.160 |
Administrator | +0.159 |
Crest | +0.159 |
Neg logit contributions | |
isa | -0.253 |
awa | -0.246 |
favour | -0.245 |
ophon | -0.244 |
ethical | -0.230 |
Token | { |
Token ID | 1391 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
isa | +3.833 |
favour | +3.703 |
awa | +3.584 |
ophon | +3.544 |
swat | +3.305 |
Neg logit contributions | |
DEFENSE | -6.730 |
Vert | -6.123 |
Administrator | -5.340 |
onel | -5.337 |
HQ | -5.328 |
Token | ( |
Token ID | 7 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
DEFENSE | +0.769 |
Vert | +0.689 |
HQ | +0.585 |
Administrator | +0.582 |
Crest | +0.582 |
Neg logit contributions | |
isa | -0.649 |
awa | -0.629 |
favour | -0.626 |
ophon | -0.621 |
ethical | -0.580 |
Token | 33 |
Token ID | 2091 |
Feature activation | -8.20e-02 |
Pos logit contributions | |
DEFENSE | +1.462 |
Vert | +1.279 |
HQ | +1.037 |
Administrator | +1.030 |
Crest | +1.030 |
Neg logit contributions | |
isa | -1.820 |
awa | -1.778 |
favour | -1.768 |
ophon | -1.758 |
ethical | -1.663 |
Token | s |
Token ID | 82 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
DEFENSE | +0.318 |
Vert | +0.286 |
HQ | +0.243 |
Administrator | +0.242 |
Crest | +0.242 |
Neg logit contributions | |
isa | -0.261 |
awa | -0.254 |
favour | -0.251 |
ophon | -0.250 |
ethical | -0.233 |
Token | )} |
Token ID | 38165 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
DEFENSE | +0.214 |
Vert | +0.192 |
HQ | +0.164 |
Administrator | +0.163 |
Crest | +0.163 |
Neg logit contributions | |
isa | -0.171 |
awa | -0.165 |
favour | -0.165 |
ophon | -0.163 |
ethical | -0.152 |
Token | )} |
Token ID | 38165 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
isa | +0.412 |
favour | +0.396 |
awa | +0.393 |
ophon | +0.393 |
ethical | +0.365 |
Neg logit contributions | |
DEFENSE | -0.528 |
Vert | -0.475 |
HQ | -0.406 |
Administrator | -0.405 |
Crest | -0.405 |
Token | 7 |
Token ID | 767 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
isa | +1.390 |
awa | +1.352 |
favour | +1.347 |
ophon | +1.338 |
ethical | +1.254 |
Neg logit contributions | |
DEFENSE | -1.427 |
Vert | -1.271 |
HQ | -1.063 |
Administrator | -1.057 |
Crest | -1.056 |
Token | . |
Token ID | 13 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
DEFENSE | +1.289 |
Vert | +1.156 |
HQ | +0.978 |
Crest | +0.970 |
Administrator | +0.970 |
Neg logit contributions | |
isa | -1.158 |
awa | -1.153 |
favour | -1.125 |
ophon | -1.119 |
ethical | -1.051 |
Token | N |
Token ID | 399 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
isa | +1.334 |
favour | +1.303 |
awa | +1.292 |
ophon | +1.268 |
swat | +1.192 |
Neg logit contributions | |
DEFENSE | -1.584 |
Vert | -1.420 |
HQ | -1.201 |
Administrator | -1.194 |
onel | -1.188 |
Token | b |
Token ID | 65 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
isa | +0.434 |
awa | +0.416 |
favour | +0.415 |
ophon | +0.414 |
ethical | +0.386 |
Neg logit contributions | |
DEFENSE | -0.523 |
Vert | -0.467 |
Administrator | -0.397 |
Crest | -0.397 |
HQ | -0.396 |
Token | 3 |
Token ID | 18 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
isa | +2.697 |
favour | +2.586 |
ophon | +2.547 |
awa | +2.509 |
ethical | +2.416 |
Neg logit contributions | |
DEFENSE | -2.580 |
Vert | -2.261 |
PDATE | -1.878 |
onel | -1.868 |
Administrator | -1.861 |
Token | { |
Token ID | 1391 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
isa | +3.771 |
favour | +3.650 |
awa | +3.535 |
ophon | +3.495 |
swat | +3.264 |
Neg logit contributions | |
DEFENSE | -6.621 |
Vert | -6.069 |
PDATE | -5.286 |
onel | -5.280 |
Administrator | -5.273 |
Token | ( |
Token ID | 7 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
DEFENSE | +0.167 |
Vert | +0.150 |
Administrator | +0.127 |
Crest | +0.127 |
HQ | +0.127 |
Neg logit contributions | |
isa | -0.137 |
favour | -0.132 |
awa | -0.131 |
ophon | -0.131 |
ethical | -0.122 |
Token | 244 |
Token ID | 25707 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
DEFENSE | +1.402 |
Vert | +1.230 |
HQ | +1.004 |
Administrator | +0.998 |
Crest | +0.998 |
Neg logit contributions | |
isa | -1.669 |
awa | -1.629 |
favour | -1.620 |
ophon | -1.610 |
ethical | -1.521 |
Token | s |
Token ID | 82 |
Feature activation | -8.46e-03 |
Pos logit contributions | |
isa | +0.354 |
favour | +0.340 |
ophon | +0.336 |
awa | +0.335 |
swat | +0.313 |
Neg logit contributions | |
DEFENSE | -0.459 |
Vert | -0.413 |
HQ | -0.352 |
Crest | -0.352 |
Administrator | -0.352 |
Token | )} |
Token ID | 38165 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
DEFENSE | +0.033 |
Vert | +0.030 |
HQ | +0.026 |
Administrator | +0.025 |
Crest | +0.025 |
Neg logit contributions | |
isa | -0.026 |
awa | -0.026 |
favour | -0.025 |
ophon | -0.025 |
ethical | -0.024 |
Token | )} |
Token ID | 38165 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
isa | +0.082 |
awa | +0.079 |
favour | +0.079 |
ophon | +0.078 |
ethical | +0.073 |
Neg logit contributions | |
DEFENSE | -0.103 |
Vert | -0.093 |
HQ | -0.079 |
Administrator | -0.079 |
Crest | -0.079 |
Token | 51 |
Token ID | 6885 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
isa | +0.949 |
awa | +0.922 |
favour | +0.922 |
ophon | +0.905 |
swat | +0.852 |
Neg logit contributions | |
DEFENSE | -1.165 |
Vert | -1.051 |
HQ | -0.891 |
Administrator | -0.887 |
Crest | -0.884 |
Token | . |
Token ID | 13 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
DEFENSE | +2.050 |
Vert | +1.832 |
HQ | +1.551 |
Crest | +1.546 |
Administrator | +1.538 |
Neg logit contributions | |
isa | -1.814 |
awa | -1.765 |
favour | -1.748 |
ophon | -1.745 |
ethical | -1.632 |
Token | N |
Token ID | 399 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
isa | +0.216 |
favour | +0.209 |
awa | +0.209 |
ophon | +0.206 |
swat | +0.192 |
Neg logit contributions | |
DEFENSE | -0.264 |
Vert | -0.237 |
Administrator | -0.202 |
HQ | -0.201 |
Crest | -0.201 |
Token | d |
Token ID | 67 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
isa | +0.197 |
awa | +0.190 |
favour | +0.189 |
ophon | +0.189 |
ethical | +0.177 |
Neg logit contributions | |
DEFENSE | -0.217 |
Vert | -0.194 |
Administrator | -0.163 |
Crest | -0.163 |
HQ | -0.162 |
Token | 4 |
Token ID | 19 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
isa | +0.864 |
favour | +0.833 |
awa | +0.831 |
ophon | +0.826 |
ethical | +0.780 |
Neg logit contributions | |
DEFENSE | -0.768 |
Vert | -0.675 |
Crest | -0.555 |
Administrator | -0.554 |
HQ | -0.552 |
Token | + |
Token ID | 10 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
isa | +4.386 |
favour | +4.353 |
awa | +4.213 |
ophon | +4.081 |
swat | +3.977 |
Neg logit contributions | |
DEFENSE | -5.914 |
Vert | -5.409 |
PDATE | -4.688 |
onel | -4.607 |
Ground | -4.555 |
Token | { |
Token ID | 1391 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
isa | +3.590 |
favour | +3.526 |
awa | +3.408 |
ophon | +3.372 |
swat | +3.156 |
Neg logit contributions | |
DEFENSE | -5.080 |
Vert | -4.607 |
PDATE | -3.944 |
Crest | -3.924 |
onel | -3.920 |
Token | ( |
Token ID | 7 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
DEFENSE | +0.871 |
Vert | +0.781 |
HQ | +0.660 |
Administrator | +0.655 |
Crest | +0.655 |
Neg logit contributions | |
isa | -0.775 |
awa | -0.756 |
favour | -0.750 |
ophon | -0.745 |
swat | -0.696 |
Token | 10 |
Token ID | 940 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
DEFENSE | +2.611 |
Vert | +2.345 |
HQ | +1.937 |
7601 | +1.897 |
Administrator | +1.890 |
Neg logit contributions | |
isa | -2.953 |
awa | -2.916 |
favour | -2.887 |
ophon | -2.860 |
swat | -2.705 |
Token | s |
Token ID | 82 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
DEFENSE | +1.332 |
Vert | +1.194 |
HQ | +1.017 |
Administrator | +1.008 |
Crest | +1.007 |
Neg logit contributions | |
isa | -1.124 |
awa | -1.091 |
ophon | -1.083 |
favour | -1.078 |
ethical | -1.008 |
Token | s |
Token ID | 82 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
DEFENSE | +0.543 |
Vert | +0.488 |
HQ | +0.416 |
Administrator | +0.415 |
Crest | +0.414 |
Neg logit contributions | |
isa | -0.432 |
awa | -0.417 |
favour | -0.416 |
ophon | -0.412 |
ethical | -0.384 |
Token | )} |
Token ID | 38165 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
isa | +0.092 |
awa | +0.089 |
favour | +0.089 |
ophon | +0.088 |
ethical | +0.082 |
Neg logit contributions | |
DEFENSE | -0.117 |
Vert | -0.105 |
HQ | -0.090 |
Administrator | -0.089 |
Crest | -0.089 |
Token | 58 |
Token ID | 7618 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
isa | +0.588 |
awa | +0.572 |
favour | +0.569 |
ophon | +0.566 |
ethical | +0.531 |
Neg logit contributions | |
DEFENSE | -0.601 |
Vert | -0.534 |
HQ | -0.447 |
Administrator | -0.445 |
Crest | -0.445 |
Token | . |
Token ID | 13 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
DEFENSE | +2.169 |
Vert | +1.947 |
HQ | +1.642 |
Crest | +1.627 |
Administrator | +1.623 |
Neg logit contributions | |
awa | -2.108 |
isa | -2.059 |
favour | -2.018 |
ophon | -2.007 |
ethical | -1.899 |
Token | Re |
Token ID | 797 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
isa | +0.624 |
favour | +0.606 |
awa | +0.605 |
ophon | +0.595 |
swat | +0.556 |
Neg logit contributions | |
DEFENSE | -0.779 |
Vert | -0.699 |
HQ | -0.595 |
Administrator | -0.593 |
Crest | -0.591 |
Token | 1 |
Token ID | 16 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
isa | +1.056 |
favour | +1.017 |
awa | +1.012 |
ophon | +1.010 |
ethical | +0.951 |
Neg logit contributions | |
DEFENSE | -1.005 |
Vert | -0.882 |
Administrator | -0.732 |
Crest | -0.732 |
HQ | -0.730 |
Token | { |
Token ID | 1391 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
isa | +3.803 |
favour | +3.660 |
awa | +3.563 |
ophon | +3.539 |
swat | +3.275 |
Neg logit contributions | |
DEFENSE | -6.572 |
Vert | -5.982 |
Administrator | -5.206 |
PDATE | -5.202 |
onel | -5.184 |
Token | ( |
Token ID | 7 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
DEFENSE | +0.764 |
Vert | +0.686 |
HQ | +0.582 |
Administrator | +0.579 |
Crest | +0.579 |
Neg logit contributions | |
isa | -0.649 |
awa | -0.633 |
favour | -0.627 |
ophon | -0.623 |
ethical | -0.582 |
Token | 25 |
Token ID | 1495 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
DEFENSE | +1.561 |
Vert | +1.369 |
HQ | +1.114 |
Crest | +1.105 |
Administrator | +1.105 |
Neg logit contributions | |
isa | -1.913 |
awa | -1.872 |
favour | -1.860 |
ophon | -1.849 |
ethical | -1.748 |
Token | s |
Token ID | 82 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
isa | +0.241 |
favour | +0.232 |
awa | +0.232 |
ophon | +0.230 |
ethical | +0.214 |
Neg logit contributions | |
DEFENSE | -0.307 |
Vert | -0.276 |
HQ | -0.235 |
Administrator | -0.235 |
Crest | -0.234 |
Token | )} |
Token ID | 38165 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
isa | +0.349 |
favour | +0.335 |
awa | +0.334 |
ophon | +0.332 |
ethical | +0.309 |
Neg logit contributions | |
DEFENSE | -0.443 |
Vert | -0.398 |
HQ | -0.340 |
Administrator | -0.339 |
Crest | -0.339 |
Token | : |
Token ID | 25 |
Feature activation | +7.22e-01 |
Pos logit contributions | |
isa | +1.975 |
favour | +1.841 |
ophon | +1.783 |
swat | +1.666 |
awa | +1.654 |
Neg logit contributions | |
DEFENSE | -3.669 |
Vert | -3.310 |
PDATE | -3.063 |
Administrator | -2.885 |
chwitz | -2.883 |
Token | 47 |
Token ID | 2857 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
isa | +2.482 |
favour | +2.364 |
ophon | +2.330 |
awa | +2.286 |
swat | +2.208 |
Neg logit contributions | |
DEFENSE | -2.606 |
Vert | -2.273 |
PDATE | -2.006 |
Administrator | -1.934 |
HQ | -1.928 |
Token | ] |
Token ID | 60 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
isa | +1.627 |
favour | +1.535 |
ophon | +1.513 |
awa | +1.426 |
abet | +1.422 |
Neg logit contributions | |
DEFENSE | -2.222 |
Vert | -1.972 |
PDATE | -1.777 |
Crest | -1.690 |
Administrator | -1.685 |
Token | < |
Token ID | 1279 |
Feature activation | +7.22e-04 |
Pos logit contributions | |
isa | +0.840 |
awa | +0.814 |
favour | +0.808 |
ophon | +0.798 |
swat | +0.739 |
Neg logit contributions | |
DEFENSE | -1.261 |
Vert | -1.147 |
HQ | -0.992 |
Crest | -0.987 |
Administrator | -0.985 |
Token | sh |
Token ID | 1477 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
isa | +0.002 |
awa | +0.002 |
favour | +0.002 |
ophon | +0.002 |
ethical | +0.002 |
Neg logit contributions | |
DEFENSE | -0.003 |
Vert | -0.003 |
HQ | -0.002 |
Administrator | -0.002 |
Crest | -0.002 |
Token | o |
Token ID | 78 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
isa | +2.488 |
ophon | +2.345 |
awa | +2.336 |
favour | +2.333 |
abet | +2.213 |
Neg logit contributions | |
DEFENSE | -3.272 |
Vert | -2.920 |
Crest | -2.522 |
HQ | -2.517 |
Administrator | -2.515 |
Token | ja |
Token ID | 6592 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
isa | +4.336 |
ophon | +4.105 |
favour | +3.978 |
awa | +3.952 |
ethical | +3.800 |
Neg logit contributions | |
DEFENSE | -5.930 |
Vert | -5.235 |
Crest | -4.606 |
Administrator | -4.573 |
PDATE | -4.540 |
Token | > |
Token ID | 29 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
isa | +1.409 |
awa | +1.352 |
favour | +1.348 |
ophon | +1.345 |
ethical | +1.257 |
Neg logit contributions | |
DEFENSE | -1.624 |
Vert | -1.445 |
Administrator | -1.226 |
Crest | -1.226 |
HQ | -1.226 |
Token | Frank |
Token ID | 5278 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
DEFENSE | +0.227 |
Vert | +0.204 |
HQ | +0.173 |
Crest | +0.172 |
Administrator | +0.172 |
Neg logit contributions | |
isa | -0.196 |
awa | -0.190 |
favour | -0.189 |
ophon | -0.188 |
ethical | -0.175 |
Token | er |
Token ID | 263 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
DEFENSE | +0.298 |
Vert | +0.271 |
HQ | +0.235 |
Administrator | +0.235 |
Crest | +0.234 |
Neg logit contributions | |
isa | -0.186 |
awa | -0.179 |
favour | -0.178 |
ophon | -0.176 |
ethical | -0.162 |
Token | Z |
Token ID | 57 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
isa | +1.999 |
awa | +1.931 |
favour | +1.914 |
ophon | +1.899 |
abet | +1.776 |
Neg logit contributions | |
DEFENSE | -2.291 |
Vert | -2.049 |
HQ | -1.740 |
Crest | -1.735 |
Administrator | -1.727 |
Token | 5 |
Token ID | 642 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
DEFENSE | +0.868 |
Vert | +0.765 |
HQ | +0.629 |
Administrator | +0.627 |
Crest | +0.627 |
Neg logit contributions | |
isa | -0.974 |
awa | -0.947 |
favour | -0.944 |
ophon | -0.935 |
ethical | -0.883 |
Token | ); |
Token ID | 1776 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
DEFENSE | +2.454 |
Vert | +2.196 |
HQ | +1.846 |
Administrator | +1.830 |
Crest | +1.828 |
Neg logit contributions | |
isa | -2.259 |
awa | -2.225 |
favour | -2.197 |
ophon | -2.193 |
ethical | -2.055 |
Token | ++ |
Token ID | 19969 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
isa | +1.905 |
awa | +1.876 |
favour | +1.871 |
ophon | +1.815 |
swat | +1.751 |
Neg logit contributions | |
DEFENSE | -1.532 |
Vert | -1.360 |
HQ | -1.102 |
Crest | -1.094 |
Administrator | -1.082 |
Token | count |
Token ID | 9127 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
isa | +1.612 |
favour | +1.520 |
awa | +1.520 |
ophon | +1.491 |
ethical | +1.412 |
Neg logit contributions | |
DEFENSE | -2.139 |
Vert | -1.916 |
Crest | -1.671 |
Administrator | -1.660 |
Ft | -1.640 |
Token | ; |
Token ID | 26 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
isa | +0.476 |
awa | +0.461 |
favour | +0.459 |
ophon | +0.456 |
ethical | +0.427 |
Neg logit contributions | |
DEFENSE | -0.519 |
Vert | -0.462 |
HQ | -0.390 |
Crest | -0.389 |
Administrator | -0.388 |
Token | } |
Token ID | 1782 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
isa | +1.691 |
awa | +1.668 |
favour | +1.661 |
ophon | +1.596 |
swat | +1.538 |
Neg logit contributions | |
DEFENSE | -1.714 |
Vert | -1.543 |
HQ | -1.289 |
Crest | -1.286 |
Administrator | -1.273 |
Token | RAW |
Token ID | 33782 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
isa | +4.666 |
awa | +4.618 |
favour | +4.587 |
ophon | +4.395 |
swat | +4.277 |
Neg logit contributions | |
DEFENSE | -4.692 |
Vert | -4.204 |
HQ | -3.541 |
Administrator | -3.525 |
chwitz | -3.502 |
Token | _ |
Token ID | 62 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
isa | +1.906 |
favour | +1.799 |
ophon | +1.787 |
awa | +1.715 |
abet | +1.680 |
Neg logit contributions | |
DEFENSE | -2.561 |
Vert | -2.274 |
Crest | -1.987 |
PDATE | -1.984 |
Administrator | -1.972 |
Token | ASS |
Token ID | 10705 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
DEFENSE | +0.460 |
Vert | +0.406 |
HQ | +0.334 |
Administrator | +0.332 |
Crest | +0.332 |
Neg logit contributions | |
isa | -0.514 |
awa | -0.501 |
favour | -0.499 |
ophon | -0.496 |
ethical | -0.467 |
Token | ERT |
Token ID | 17395 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
isa | +2.402 |
favour | +2.313 |
ophon | +2.310 |
awa | +2.262 |
abet | +2.227 |
Neg logit contributions | |
DEFENSE | -0.920 |
Vert | -0.724 |
Crest | -0.516 |
Administrator | -0.503 |
Ft | -0.489 |
Token | ( |
Token ID | 7 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
isa | +2.534 |
favour | +2.437 |
awa | +2.420 |
ophon | +2.360 |
ethical | +2.228 |
Neg logit contributions | |
DEFENSE | -3.250 |
Vert | -2.939 |
Crest | -2.529 |
Administrator | -2.521 |
HQ | -2.521 |
Token | s |
Token ID | 82 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
DEFENSE | +0.991 |
Vert | +0.878 |
HQ | +0.730 |
Administrator | +0.727 |
Crest | +0.726 |
Neg logit contributions | |
isa | -1.021 |
awa | -0.994 |
favour | -0.988 |
ophon | -0.983 |
ethical | -0.924 |
Token | )} |
Token ID | 38165 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
isa | +0.104 |
favour | +0.100 |
awa | +0.100 |
ophon | +0.099 |
ethical | +0.092 |
Neg logit contributions | |
DEFENSE | -0.133 |
Vert | -0.119 |
HQ | -0.102 |
Administrator | -0.102 |
Crest | -0.102 |
Token | 11 |
Token ID | 1367 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
isa | +1.002 |
favour | +0.974 |
awa | +0.971 |
ophon | +0.964 |
swat | +0.904 |
Neg logit contributions | |
DEFENSE | -0.962 |
Vert | -0.853 |
HQ | -0.709 |
Administrator | -0.704 |
Crest | -0.702 |
Token | . |
Token ID | 13 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
DEFENSE | +1.638 |
Vert | +1.470 |
HQ | +1.242 |
Crest | +1.232 |
Administrator | +1.231 |
Neg logit contributions | |
isa | -1.493 |
awa | -1.492 |
favour | -1.451 |
ophon | -1.445 |
ethical | -1.359 |
Token | h |
Token ID | 289 |
Feature activation | -8.03e-02 |
Pos logit contributions | |
isa | +1.187 |
favour | +1.155 |
awa | +1.151 |
ophon | +1.130 |
swat | +1.060 |
Neg logit contributions | |
DEFENSE | -1.436 |
Vert | -1.288 |
HQ | -1.093 |
Administrator | -1.086 |
Crest | -1.083 |
Token | 3 |
Token ID | 18 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
DEFENSE | +0.269 |
Vert | +0.237 |
HQ | +0.195 |
Administrator | +0.195 |
Crest | +0.194 |
Neg logit contributions | |
isa | -0.299 |
awa | -0.290 |
favour | -0.289 |
ophon | -0.288 |
ethical | -0.271 |
Token | { |
Token ID | 1391 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
isa | +3.542 |
favour | +3.432 |
awa | +3.319 |
ophon | +3.291 |
swat | +3.075 |
Neg logit contributions | |
DEFENSE | -6.011 |
Vert | -5.489 |
PDATE | -4.791 |
Administrator | -4.770 |
onel | -4.766 |
Token | ( |
Token ID | 7 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
DEFENSE | +0.636 |
Vert | +0.570 |
HQ | +0.484 |
Administrator | +0.482 |
Crest | +0.482 |
Neg logit contributions | |
isa | -0.535 |
awa | -0.518 |
favour | -0.516 |
ophon | -0.512 |
ethical | -0.478 |
Token | 165 |
Token ID | 20986 |
Feature activation | +3.02e-03 |
Pos logit contributions | |
DEFENSE | +1.491 |
Vert | +1.308 |
HQ | +1.067 |
Administrator | +1.061 |
Crest | +1.061 |
Neg logit contributions | |
isa | -1.780 |
awa | -1.737 |
favour | -1.728 |
ophon | -1.718 |
ethical | -1.623 |
Token | s |
Token ID | 82 |
Feature activation | +2.88e-03 |
Pos logit contributions | |
isa | +0.009 |
favour | +0.009 |
awa | +0.009 |
ophon | +0.009 |
swat | +0.008 |
Neg logit contributions | |
DEFENSE | -0.012 |
Vert | -0.011 |
HQ | -0.009 |
Crest | -0.009 |
Administrator | -0.009 |
Token | )} |
Token ID | 38165 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
isa | +0.009 |
awa | +0.009 |
favour | +0.009 |
ophon | +0.009 |
ethical | +0.008 |
Neg logit contributions | |
DEFENSE | -0.011 |
Vert | -0.010 |
HQ | -0.009 |
Administrator | -0.009 |
Crest | -0.009 |
Token | return |
Token ID | 1441 |
Feature activation | -9.77e-02 |
Pos logit contributions | |
isa | +1.721 |
favour | +1.694 |
awa | +1.689 |
ophon | +1.635 |
swat | +1.559 |
Neg logit contributions | |
DEFENSE | -1.994 |
Vert | -1.775 |
HQ | -1.530 |
Crest | -1.497 |
7601 | -1.497 |
Token | normal |
Token ID | 3487 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
DEFENSE | +0.363 |
Vert | +0.324 |
HQ | +0.274 |
Crest | +0.272 |
Administrator | +0.271 |
Neg logit contributions | |
isa | -0.327 |
awa | -0.319 |
favour | -0.317 |
ophon | -0.313 |
swat | -0.294 |
Token | ; |
Token ID | 2162 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
isa | +0.417 |
awa | +0.403 |
favour | +0.401 |
ophon | +0.398 |
ethical | +0.374 |
Neg logit contributions | |
DEFENSE | -0.402 |
Vert | -0.354 |
Crest | -0.296 |
HQ | -0.296 |
Administrator | -0.294 |
Token | # |
Token ID | 1303 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
isa | +1.579 |
awa | +1.550 |
favour | +1.545 |
ophon | +1.504 |
swat | +1.432 |
Neg logit contributions | |
DEFENSE | -1.584 |
Vert | -1.405 |
HQ | -1.182 |
Crest | -1.166 |
7601 | -1.162 |
Token | endif |
Token ID | 32088 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
isa | +0.413 |
awa | +0.400 |
favour | +0.399 |
ophon | +0.395 |
abet | +0.371 |
Neg logit contributions | |
DEFENSE | -0.450 |
Vert | -0.400 |
HQ | -0.339 |
Administrator | -0.337 |
Crest | -0.337 |
Token | } |
Token ID | 1782 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
isa | +1.416 |
awa | +1.392 |
favour | +1.384 |
ophon | +1.357 |
swat | +1.283 |
Neg logit contributions | |
DEFENSE | -1.524 |
Vert | -1.357 |
HQ | -1.156 |
Administrator | -1.136 |
Crest | -1.135 |
Token | If |
Token ID | 1002 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
isa | +4.458 |
favour | +4.440 |
awa | +4.435 |
ophon | +4.303 |
swat | +4.116 |
Neg logit contributions | |
DEFENSE | -4.901 |
Vert | -4.324 |
HQ | -3.740 |
Administrator | -3.668 |
7601 | -3.654 |
Token | you |
Token ID | 345 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
isa | +2.336 |
awa | +2.282 |
favour | +2.281 |
ophon | +2.223 |
swat | +2.109 |
Neg logit contributions | |
DEFENSE | -2.532 |
Vert | -2.252 |
HQ | -1.922 |
Crest | -1.895 |
7601 | -1.885 |
Token | are |
Token ID | 389 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
isa | +1.030 |
favour | +1.019 |
awa | +1.001 |
ophon | +0.977 |
swat | +0.934 |
Neg logit contributions | |
DEFENSE | -1.362 |
Vert | -1.242 |
HQ | -1.069 |
Crest | -1.050 |
Administrator | -1.044 |
Token | curious |
Token ID | 11040 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
isa | +2.179 |
favour | +2.156 |
awa | +2.139 |
ophon | +2.070 |
swat | +2.000 |
Neg logit contributions | |
DEFENSE | -2.253 |
Vert | -2.016 |
HQ | -1.712 |
Crest | -1.663 |
onel | -1.660 |
Token | to |
Token ID | 284 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
isa | +0.958 |
awa | +0.926 |
favour | +0.925 |
ophon | +0.912 |
swat | +0.866 |
Neg logit contributions | |
DEFENSE | -0.959 |
Vert | -0.855 |
HQ | -0.713 |
Crest | -0.705 |
Administrator | -0.703 |
Token | 10 |
Token ID | 940 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
DEFENSE | +2.611 |
Vert | +2.345 |
HQ | +1.937 |
7601 | +1.897 |
Administrator | +1.890 |
Neg logit contributions | |
isa | -2.953 |
awa | -2.916 |
favour | -2.887 |
ophon | -2.860 |
swat | -2.705 |
Token | s |
Token ID | 82 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
DEFENSE | +1.332 |
Vert | +1.194 |
HQ | +1.017 |
Administrator | +1.008 |
Crest | +1.007 |
Neg logit contributions | |
isa | -1.124 |
awa | -1.091 |
ophon | -1.083 |
favour | -1.078 |
ethical | -1.008 |
Token | )} |
Token ID | 38165 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
isa | +0.127 |
awa | +0.123 |
favour | +0.122 |
ophon | +0.121 |
ethical | +0.113 |
Neg logit contributions | |
DEFENSE | -0.157 |
Vert | -0.141 |
HQ | -0.120 |
Administrator | -0.119 |
Crest | -0.119 |
Token | K |
Token ID | 509 |
Feature activation | -4.78e-03 |
Pos logit contributions | |
isa | +0.820 |
favour | +0.797 |
awa | +0.795 |
ophon | +0.782 |
swat | +0.731 |
Neg logit contributions | |
DEFENSE | -1.041 |
Vert | -0.940 |
HQ | -0.798 |
Administrator | -0.795 |
Crest | -0.793 |
Token | b |
Token ID | 65 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
DEFENSE | +0.017 |
Vert | +0.015 |
Crest | +0.013 |
Administrator | +0.013 |
HQ | +0.013 |
Neg logit contributions | |
isa | -0.017 |
favour | -0.016 |
awa | -0.016 |
ophon | -0.016 |
ethical | -0.015 |
Token | 7 |
Token ID | 22 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
isa | +2.338 |
favour | +2.222 |
ophon | +2.193 |
awa | +2.190 |
ethical | +2.069 |
Neg logit contributions | |
DEFENSE | -2.357 |
Vert | -2.081 |
Administrator | -1.727 |
PDATE | -1.725 |
Crest | -1.723 |
Token | { |
Token ID | 1391 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
isa | +4.101 |
favour | +4.038 |
awa | +3.900 |
ophon | +3.878 |
swat | +3.658 |
Neg logit contributions | |
DEFENSE | -5.058 |
Vert | -4.696 |
PDATE | -3.914 |
HQ | -3.912 |
Crest | -3.893 |
Token | ( |
Token ID | 7 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
DEFENSE | +0.921 |
Vert | +0.826 |
HQ | +0.700 |
Crest | +0.694 |
Administrator | +0.694 |
Neg logit contributions | |
isa | -0.806 |
awa | -0.784 |
favour | -0.778 |
ophon | -0.773 |
ethical | -0.722 |
Token | 3 |
Token ID | 18 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
DEFENSE | +2.176 |
Vert | +1.933 |
HQ | +1.586 |
Crest | +1.562 |
Administrator | +1.560 |
Neg logit contributions | |
isa | -2.541 |
awa | -2.496 |
favour | -2.477 |
ophon | -2.458 |
swat | -2.319 |
Token | s |
Token ID | 82 |
Feature activation | +5.90e-03 |
Pos logit contributions | |
DEFENSE | +0.869 |
Vert | +0.781 |
HQ | +0.666 |
Administrator | +0.661 |
Crest | +0.661 |
Neg logit contributions | |
isa | -0.717 |
awa | -0.696 |
favour | -0.690 |
ophon | -0.687 |
ethical | -0.641 |
Token | )} |
Token ID | 38165 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
isa | +0.019 |
awa | +0.018 |
favour | +0.018 |
ophon | +0.018 |
ethical | +0.017 |
Neg logit contributions | |
DEFENSE | -0.023 |
Vert | -0.021 |
HQ | -0.018 |
Administrator | -0.017 |
Crest | -0.017 |
Token | s |
Token ID | 82 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
DEFENSE | +0.073 |
Vert | +0.066 |
HQ | +0.056 |
Crest | +0.056 |
Administrator | +0.056 |
Neg logit contributions | |
isa | -0.058 |
favour | -0.056 |
awa | -0.056 |
ophon | -0.055 |
ethical | -0.051 |
Token | )} |
Token ID | 38165 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
isa | +0.094 |
awa | +0.091 |
favour | +0.090 |
ophon | +0.090 |
ethical | +0.083 |
Neg logit contributions | |
DEFENSE | -0.117 |
Vert | -0.105 |
HQ | -0.090 |
Administrator | -0.089 |
Crest | -0.089 |
Token | 55 |
Token ID | 5996 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
isa | +0.557 |
favour | +0.543 |
awa | +0.542 |
ophon | +0.537 |
swat | +0.505 |
Neg logit contributions | |
DEFENSE | -0.536 |
Vert | -0.478 |
HQ | -0.395 |
Administrator | -0.393 |
Crest | -0.391 |
Token | . |
Token ID | 13 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
DEFENSE | +1.799 |
Vert | +1.610 |
HQ | +1.359 |
Crest | +1.351 |
Administrator | +1.349 |
Neg logit contributions | |
isa | -1.633 |
awa | -1.616 |
favour | -1.580 |
ophon | -1.578 |
ethical | -1.482 |
Token | Ra |
Token ID | 7567 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
isa | +0.697 |
favour | +0.677 |
awa | +0.675 |
ophon | +0.664 |
swat | +0.622 |
Neg logit contributions | |
DEFENSE | -0.845 |
Vert | -0.759 |
HQ | -0.643 |
Administrator | -0.640 |
Crest | -0.639 |
Token | 1 |
Token ID | 16 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
isa | +0.744 |
favour | +0.714 |
awa | +0.711 |
ophon | +0.709 |
ethical | +0.670 |
Neg logit contributions | |
DEFENSE | -0.696 |
Vert | -0.610 |
Crest | -0.506 |
Administrator | -0.506 |
HQ | -0.502 |
Token | { |
Token ID | 1391 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
isa | +3.587 |
favour | +3.475 |
awa | +3.395 |
ophon | +3.333 |
swat | +3.135 |
Neg logit contributions | |
DEFENSE | -5.655 |
Vert | -5.148 |
PDATE | -4.442 |
onel | -4.435 |
Administrator | -4.428 |
Token | ( |
Token ID | 7 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
DEFENSE | +0.691 |
Vert | +0.619 |
HQ | +0.525 |
Administrator | +0.523 |
Crest | +0.523 |
Neg logit contributions | |
isa | -0.583 |
awa | -0.565 |
favour | -0.562 |
ophon | -0.558 |
ethical | -0.521 |
Token | 55 |
Token ID | 2816 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
DEFENSE | +1.936 |
Vert | +1.709 |
HQ | +1.394 |
Crest | +1.375 |
Administrator | +1.372 |
Neg logit contributions | |
isa | -2.351 |
awa | -2.314 |
favour | -2.293 |
ophon | -2.279 |
ethical | -2.151 |
Token | s |
Token ID | 82 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
DEFENSE | +0.457 |
Vert | +0.412 |
HQ | +0.350 |
Crest | +0.348 |
Administrator | +0.348 |
Neg logit contributions | |
isa | -0.382 |
awa | -0.372 |
favour | -0.368 |
ophon | -0.367 |
ethical | -0.342 |
Token | )} |
Token ID | 38165 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
isa | +0.100 |
favour | +0.096 |
awa | +0.096 |
ophon | +0.095 |
ethical | +0.089 |
Neg logit contributions | |
DEFENSE | -0.127 |
Vert | -0.114 |
HQ | -0.097 |
Administrator | -0.097 |
Crest | -0.097 |
Token | s |
Token ID | 82 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
isa | +0.443 |
favour | +0.424 |
ophon | +0.418 |
awa | +0.416 |
abet | +0.390 |
Neg logit contributions | |
DEFENSE | -0.586 |
Vert | -0.525 |
Crest | -0.449 |
HQ | -0.448 |
Administrator | -0.448 |
Token | )} |
Token ID | 38165 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
isa | +0.287 |
favour | +0.275 |
awa | +0.274 |
ophon | +0.273 |
ethical | +0.254 |
Neg logit contributions | |
DEFENSE | -0.367 |
Vert | -0.330 |
HQ | -0.282 |
Administrator | -0.281 |
Crest | -0.281 |
Token | 14 |
Token ID | 1478 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
isa | +0.931 |
awa | +0.902 |
favour | +0.898 |
ophon | +0.888 |
ethical | +0.840 |
Neg logit contributions | |
DEFENSE | -0.895 |
Vert | -0.796 |
HQ | -0.660 |
Crest | -0.655 |
Administrator | -0.654 |
Token | . |
Token ID | 13 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
DEFENSE | +2.124 |
Vert | +1.905 |
HQ | +1.607 |
Crest | +1.593 |
Administrator | +1.589 |
Neg logit contributions | |
awa | -1.990 |
isa | -1.975 |
favour | -1.925 |
ophon | -1.917 |
ethical | -1.806 |
Token | e |
Token ID | 304 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
isa | +1.132 |
favour | +1.102 |
awa | +1.098 |
ophon | +1.078 |
swat | +1.013 |
Neg logit contributions | |
DEFENSE | -1.301 |
Vert | -1.163 |
HQ | -0.982 |
Administrator | -0.976 |
Crest | -0.971 |
Token | 3 |
Token ID | 18 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
DEFENSE | +0.226 |
Vert | +0.198 |
HQ | +0.161 |
Administrator | +0.160 |
Crest | +0.160 |
Neg logit contributions | |
isa | -0.270 |
awa | -0.263 |
favour | -0.262 |
ophon | -0.260 |
ethical | -0.246 |
Token | { |
Token ID | 1391 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
isa | +3.470 |
favour | +3.376 |
awa | +3.282 |
ophon | +3.226 |
swat | +3.020 |
Neg logit contributions | |
DEFENSE | -5.737 |
Vert | -5.242 |
PDATE | -4.545 |
HQ | -4.537 |
Administrator | -4.534 |
Token | ( |
Token ID | 7 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
DEFENSE | +0.922 |
Vert | +0.828 |
HQ | +0.702 |
Administrator | +0.698 |
Crest | +0.698 |
Neg logit contributions | |
isa | -0.786 |
awa | -0.767 |
favour | -0.760 |
ophon | -0.754 |
ethical | -0.704 |
Token | 8 |
Token ID | 23 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
DEFENSE | +1.479 |
Vert | +1.296 |
HQ | +1.056 |
Administrator | +1.050 |
Crest | +1.050 |
Neg logit contributions | |
isa | -1.781 |
awa | -1.738 |
favour | -1.729 |
ophon | -1.719 |
ethical | -1.624 |
Token | 26 |
Token ID | 2075 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
DEFENSE | +0.549 |
Vert | +0.482 |
HQ | +0.393 |
Administrator | +0.391 |
Crest | +0.390 |
Neg logit contributions | |
isa | -0.660 |
awa | -0.643 |
favour | -0.641 |
ophon | -0.636 |
ethical | -0.601 |
Token | s |
Token ID | 82 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
DEFENSE | +0.009 |
Vert | +0.008 |
HQ | +0.007 |
Administrator | +0.007 |
Crest | +0.007 |
Neg logit contributions | |
isa | -0.007 |
favour | -0.006 |
awa | -0.006 |
ophon | -0.006 |
ethical | -0.006 |
Token | ! |
Token ID | 0 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
isa | +0.862 |
awa | +0.836 |
favour | +0.834 |
ophon | +0.816 |
swat | +0.773 |
Neg logit contributions | |
DEFENSE | -1.072 |
Vert | -0.968 |
HQ | -0.818 |
Crest | -0.815 |
Administrator | -0.814 |
Token |
|
Token ID | 198 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
isa | +0.484 |
awa | +0.468 |
favour | +0.465 |
ophon | +0.459 |
swat | +0.429 |
Neg logit contributions | |
DEFENSE | -0.615 |
Vert | -0.554 |
HQ | -0.473 |
Crest | -0.469 |
Administrator | -0.469 |
Token |
|
Token ID | 198 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
isa | +0.503 |
favour | +0.480 |
ophon | +0.473 |
awa | +0.462 |
abet | +0.443 |
Neg logit contributions | |
DEFENSE | -0.645 |
Vert | -0.576 |
PDATE | -0.497 |
Crest | -0.490 |
HQ | -0.490 |
Token | In |
Token ID | 818 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
isa | +0.305 |
awa | +0.292 |
favour | +0.291 |
ophon | +0.290 |
abet | +0.274 |
Neg logit contributions | |
DEFENSE | -0.330 |
Vert | -0.291 |
Crest | -0.246 |
Administrator | -0.246 |
HQ | -0.244 |
Token | the |
Token ID | 262 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
isa | +0.479 |
favour | +0.466 |
awa | +0.465 |
ophon | +0.456 |
abet | +0.432 |
Neg logit contributions | |
DEFENSE | -0.480 |
Vert | -0.428 |
HQ | -0.359 |
Administrator | -0.355 |
Crest | -0.353 |
Token | meantime |
Token ID | 14324 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
isa | +0.680 |
awa | +0.668 |
favour | +0.667 |
ophon | +0.651 |
swat | +0.622 |
Neg logit contributions | |
DEFENSE | -0.558 |
Vert | -0.493 |
HQ | -0.399 |
Crest | -0.395 |
7601 | -0.395 |
Token | , |
Token ID | 11 |
Feature activation | -4.20e-03 |
Pos logit contributions | |
isa | +3.693 |
awa | +3.605 |
favour | +3.531 |
ophon | +3.439 |
swat | +3.304 |
Neg logit contributions | |
DEFENSE | -5.369 |
Vert | -4.900 |
7601 | -4.229 |
HQ | -4.167 |
onel | -4.137 |
Token | happy |
Token ID | 3772 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
DEFENSE | +0.016 |
Vert | +0.014 |
HQ | +0.012 |
Administrator | +0.012 |
Crest | +0.012 |
Neg logit contributions | |
isa | -0.014 |
awa | -0.014 |
favour | -0.014 |
ophon | -0.013 |
swat | -0.013 |
Token | Monday |
Token ID | 3321 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
isa | +1.747 |
awa | +1.716 |
favour | +1.702 |
ophon | +1.654 |
swat | +1.606 |
Neg logit contributions | |
DEFENSE | -1.926 |
Vert | -1.731 |
7601 | -1.450 |
Administrator | -1.429 |
onel | -1.423 |
Token | to |
Token ID | 284 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
isa | +0.411 |
awa | +0.401 |
favour | +0.395 |
ophon | +0.388 |
swat | +0.368 |
Neg logit contributions | |
DEFENSE | -0.473 |
Vert | -0.425 |
Administrator | -0.355 |
HQ | -0.355 |
7601 | -0.355 |
Token | you |
Token ID | 345 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
DEFENSE | +0.415 |
Vert | +0.369 |
HQ | +0.309 |
Administrator | +0.307 |
Crest | +0.307 |
Neg logit contributions | |
isa | -0.406 |
awa | -0.395 |
favour | -0.393 |
ophon | -0.390 |
ethical | -0.366 |
Token | s |
Token ID | 82 |
Feature activation | +2.09e-03 |
Pos logit contributions | |
DEFENSE | +0.227 |
Vert | +0.204 |
HQ | +0.174 |
Administrator | +0.173 |
Crest | +0.173 |
Neg logit contributions | |
isa | -0.185 |
awa | -0.180 |
favour | -0.178 |
ophon | -0.177 |
ethical | -0.165 |
Token | )} |
Token ID | 38165 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
isa | +0.007 |
awa | +0.006 |
favour | +0.006 |
ophon | +0.006 |
ethical | +0.006 |
Neg logit contributions | |
DEFENSE | -0.008 |
Vert | -0.007 |
HQ | -0.006 |
Administrator | -0.006 |
Crest | -0.006 |
Token | 53 |
Token ID | 7192 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
isa | +0.965 |
favour | +0.939 |
awa | +0.937 |
ophon | +0.924 |
swat | +0.862 |
Neg logit contributions | |
DEFENSE | -1.199 |
Vert | -1.084 |
HQ | -0.917 |
Administrator | -0.916 |
7601 | -0.912 |
Token | . |
Token ID | 13 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
DEFENSE | +1.835 |
Vert | +1.641 |
HQ | +1.387 |
Crest | +1.379 |
Administrator | +1.375 |
Neg logit contributions | |
isa | -1.683 |
awa | -1.675 |
favour | -1.630 |
ophon | -1.627 |
ethical | -1.531 |
Token | e |
Token ID | 304 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
isa | +0.746 |
favour | +0.725 |
awa | +0.723 |
ophon | +0.710 |
swat | +0.665 |
Neg logit contributions | |
DEFENSE | -0.937 |
Vert | -0.843 |
HQ | -0.716 |
Administrator | -0.712 |
Crest | -0.711 |
Token | 6 |
Token ID | 21 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
isa | +0.423 |
awa | +0.409 |
favour | +0.409 |
ophon | +0.406 |
ethical | +0.384 |
Neg logit contributions | |
DEFENSE | -0.371 |
Vert | -0.325 |
Administrator | -0.266 |
HQ | -0.266 |
Crest | -0.265 |
Token | { |
Token ID | 1391 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
isa | +3.694 |
favour | +3.618 |
awa | +3.508 |
ophon | +3.428 |
swat | +3.270 |
Neg logit contributions | |
DEFENSE | -5.369 |
Vert | -4.899 |
PDATE | -4.240 |
onel | -4.195 |
Administrator | -4.170 |
Token | ( |
Token ID | 7 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
DEFENSE | +0.521 |
Vert | +0.467 |
HQ | +0.396 |
Crest | +0.393 |
Administrator | +0.393 |
Neg logit contributions | |
isa | -0.453 |
awa | -0.442 |
favour | -0.438 |
ophon | -0.435 |
swat | -0.406 |
Token | 8 |
Token ID | 23 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
DEFENSE | +1.606 |
Vert | +1.417 |
HQ | +1.161 |
Crest | +1.149 |
Administrator | +1.148 |
Neg logit contributions | |
isa | -1.884 |
awa | -1.844 |
favour | -1.832 |
ophon | -1.820 |
ethical | -1.717 |
Token | s |
Token ID | 82 |
Feature activation | +1.23e-03 |
Pos logit contributions | |
DEFENSE | +0.948 |
Vert | +0.850 |
HQ | +0.723 |
Administrator | +0.719 |
Crest | +0.717 |
Neg logit contributions | |
isa | -0.798 |
awa | -0.777 |
favour | -0.768 |
ophon | -0.768 |
ethical | -0.716 |
Token | )} |
Token ID | 38165 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
isa | +0.004 |
awa | +0.004 |
favour | +0.004 |
ophon | +0.004 |
ethical | +0.004 |
Neg logit contributions | |
DEFENSE | -0.005 |
Vert | -0.004 |
HQ | -0.004 |
Administrator | -0.004 |
Crest | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +6.90e-01 |
Pos logit contributions | |
isa | +2.274 |
favour | +2.145 |
ophon | +2.096 |
swat | +1.961 |
abet | +1.943 |
Neg logit contributions | |
DEFENSE | -3.310 |
Vert | -2.932 |
PDATE | -2.747 |
chwitz | -2.543 |
HQ | -2.530 |
Token | Park |
Token ID | 25478 |
Feature activation | +8.59e-01 |
Pos logit contributions | |
isa | +2.264 |
ophon | +2.160 |
favour | +2.102 |
awa | +2.082 |
ethical | +2.001 |
Neg logit contributions | |
DEFENSE | -2.668 |
Vert | -2.269 |
Crest | -2.014 |
Administrator | -1.994 |
Ft | -1.972 |
Token | er |
Token ID | 263 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
isa | +2.588 |
ophon | +2.452 |
awa | +2.437 |
favour | +2.431 |
abet | +2.282 |
Neg logit contributions | |
DEFENSE | -3.493 |
Vert | -3.164 |
HQ | -2.693 |
7601 | -2.678 |
Administrator | -2.662 |
Token | , |
Token ID | 11 |
Feature activation | +9.65e-01 |
Pos logit contributions | |
isa | +2.400 |
ophon | +2.279 |
awa | +2.276 |
favour | +2.265 |
abet | +2.111 |
Neg logit contributions | |
DEFENSE | -3.137 |
Vert | -2.819 |
HQ | -2.419 |
Administrator | -2.398 |
7601 | -2.392 |
Token | A |
Token ID | 317 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
isa | +2.862 |
favour | +2.727 |
ophon | +2.638 |
awa | +2.600 |
swat | +2.519 |
Neg logit contributions | |
DEFENSE | -3.799 |
Vert | -3.516 |
PDATE | -3.119 |
chwitz | -2.991 |
HQ | -2.964 |
Token | . |
Token ID | 13 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
isa | +2.431 |
favour | +2.324 |
awa | +2.319 |
ophon | +2.312 |
abet | +2.169 |
Neg logit contributions | |
DEFENSE | -2.828 |
Vert | -2.533 |
HQ | -2.148 |
Crest | -2.133 |
Administrator | -2.129 |
Token | K |
Token ID | 509 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
isa | +3.868 |
favour | +3.652 |
ophon | +3.533 |
awa | +3.489 |
swat | +3.372 |
Neg logit contributions | |
DEFENSE | -4.980 |
Vert | -4.644 |
PDATE | -4.074 |
chwitz | -3.946 |
7601 | -3.904 |
Token | . |
Token ID | 13 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
isa | +0.986 |
awa | +0.953 |
favour | +0.950 |
ophon | +0.945 |
ethical | +0.886 |
Neg logit contributions | |
DEFENSE | -1.053 |
Vert | -0.937 |
HQ | -0.787 |
Crest | -0.784 |
Administrator | -0.783 |
Token | ( |
Token ID | 357 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
isa | +3.082 |
favour | +2.946 |
ophon | +2.891 |
awa | +2.877 |
swat | +2.734 |
Neg logit contributions | |
DEFENSE | -3.444 |
Vert | -3.123 |
7601 | -2.616 |
HQ | -2.605 |
Administrator | -2.586 |
Token | 2010 |
Token ID | 3050 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
isa | +0.764 |
favour | +0.738 |
awa | +0.736 |
ophon | +0.730 |
abet | +0.690 |
Neg logit contributions | |
DEFENSE | -0.677 |
Vert | -0.596 |
HQ | -0.489 |
Administrator | -0.484 |
Crest | -0.481 |
Token | ). |
Token ID | 6739 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
isa | +0.718 |
awa | +0.696 |
favour | +0.692 |
ophon | +0.686 |
ethical | +0.639 |
Neg logit contributions | |
DEFENSE | -0.923 |
Vert | -0.831 |
HQ | -0.710 |
Administrator | -0.707 |
Crest | -0.707 |
Token | class |
Token ID | 1398 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
isa | +0.050 |
awa | +0.048 |
favour | +0.048 |
ophon | +0.048 |
ethical | +0.045 |
Neg logit contributions | |
DEFENSE | -0.062 |
Vert | -0.056 |
HQ | -0.048 |
Administrator | -0.047 |
Crest | -0.047 |
Token | parameter |
Token ID | 11507 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
DEFENSE | +1.180 |
Vert | +1.059 |
Administrator | +0.911 |
Crest | +0.910 |
HQ | +0.906 |
Neg logit contributions | |
isa | -0.925 |
awa | -0.890 |
ophon | -0.889 |
favour | -0.880 |
ethical | -0.832 |
Token | and |
Token ID | 290 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
isa | +2.301 |
favour | +2.230 |
awa | +2.225 |
ophon | +2.178 |
swat | +2.043 |
Neg logit contributions | |
DEFENSE | -2.471 |
Vert | -2.215 |
HQ | -1.849 |
Crest | -1.837 |
7601 | -1.834 |
Token | can |
Token ID | 460 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
isa | +0.512 |
awa | +0.498 |
favour | +0.496 |
ophon | +0.486 |
swat | +0.458 |
Neg logit contributions | |
DEFENSE | -0.608 |
Vert | -0.547 |
HQ | -0.465 |
Crest | -0.459 |
7601 | -0.458 |
Token | you |
Token ID | 345 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
isa | +0.388 |
awa | +0.378 |
favour | +0.375 |
ophon | +0.370 |
swat | +0.348 |
Neg logit contributions | |
DEFENSE | -0.417 |
Vert | -0.373 |
HQ | -0.314 |
Crest | -0.310 |
Administrator | -0.309 |
Token | pass |
Token ID | 1208 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
isa | +0.498 |
favour | +0.489 |
awa | +0.487 |
ophon | +0.474 |
swat | +0.448 |
Neg logit contributions | |
DEFENSE | -0.702 |
Vert | -0.636 |
HQ | -0.552 |
Crest | -0.544 |
Administrator | -0.542 |
Token | it |
Token ID | 340 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
isa | +3.992 |
awa | +3.952 |
favour | +3.880 |
ophon | +3.741 |
swat | +3.582 |
Neg logit contributions | |
DEFENSE | -4.950 |
Vert | -4.476 |
HQ | -3.859 |
7601 | -3.826 |
PDATE | -3.800 |
Token | around |
Token ID | 1088 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
DEFENSE | +0.389 |
Vert | +0.348 |
HQ | +0.294 |
Administrator | +0.291 |
Crest | +0.291 |
Neg logit contributions | |
isa | -0.346 |
awa | -0.337 |
favour | -0.335 |
ophon | -0.331 |
swat | -0.310 |
Token | easily |
Token ID | 3538 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
isa | +1.642 |
favour | +1.607 |
awa | +1.590 |
ophon | +1.552 |
swat | +1.471 |
Neg logit contributions | |
DEFENSE | -1.873 |
Vert | -1.666 |
HQ | -1.406 |
7601 | -1.402 |
onel | -1.391 |
Token | as |
Token ID | 355 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
isa | +0.152 |
favour | +0.147 |
awa | +0.147 |
ophon | +0.144 |
swat | +0.136 |
Neg logit contributions | |
DEFENSE | -0.168 |
Vert | -0.150 |
HQ | -0.127 |
Crest | -0.126 |
Administrator | -0.126 |
Token | opposed |
Token ID | 6886 |
Feature activation | +3.42e-02 |
Pos logit contributions | |
isa | +0.695 |
favour | +0.676 |
awa | +0.676 |
ophon | +0.659 |
swat | +0.620 |
Neg logit contributions | |
DEFENSE | -0.874 |
Vert | -0.787 |
HQ | -0.674 |
7601 | -0.668 |
Crest | -0.661 |
Token | s |
Token ID | 82 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
DEFENSE | +1.269 |
Vert | +1.141 |
HQ | +0.973 |
Administrator | +0.966 |
Crest | +0.965 |
Neg logit contributions | |
isa | -1.038 |
awa | -1.019 |
favour | -1.004 |
ophon | -0.998 |
ethical | -0.930 |
Token | )} |
Token ID | 38165 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
DEFENSE | +1.114 |
Vert | +1.002 |
HQ | +0.850 |
Administrator | +0.845 |
Crest | +0.844 |
Neg logit contributions | |
isa | -0.934 |
awa | -0.927 |
favour | -0.906 |
ophon | -0.899 |
ethical | -0.841 |
Token | 3 |
Token ID | 513 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
isa | +1.559 |
favour | +1.519 |
awa | +1.512 |
ophon | +1.491 |
swat | +1.408 |
Neg logit contributions | |
DEFENSE | -1.586 |
Vert | -1.414 |
HQ | -1.178 |
Administrator | -1.175 |
7601 | -1.168 |
Token | . |
Token ID | 13 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
DEFENSE | +2.305 |
Vert | +2.069 |
HQ | +1.746 |
Crest | +1.730 |
Administrator | +1.727 |
Neg logit contributions | |
awa | -2.158 |
isa | -2.144 |
favour | -2.088 |
ophon | -2.078 |
ethical | -1.958 |
Token | g |
Token ID | 308 |
Feature activation | -9.03e-02 |
Pos logit contributions | |
isa | +1.225 |
favour | +1.194 |
awa | +1.183 |
ophon | +1.163 |
swat | +1.093 |
Neg logit contributions | |
DEFENSE | -1.449 |
Vert | -1.300 |
HQ | -1.099 |
Administrator | -1.094 |
Crest | -1.089 |
Token | 3 |
Token ID | 18 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
DEFENSE | +0.299 |
Vert | +0.261 |
Administrator | +0.216 |
Crest | +0.214 |
HQ | +0.213 |
Neg logit contributions | |
isa | -0.343 |
favour | -0.330 |
ophon | -0.329 |
awa | -0.329 |
ethical | -0.311 |
Token | { |
Token ID | 1391 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
isa | +3.390 |
favour | +3.287 |
awa | +3.200 |
ophon | +3.132 |
swat | +2.977 |
Neg logit contributions | |
DEFENSE | -5.496 |
Vert | -5.038 |
PDATE | -4.391 |
onel | -4.370 |
Administrator | -4.345 |
Token | ( |
Token ID | 7 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
DEFENSE | +0.879 |
Vert | +0.789 |
HQ | +0.669 |
Administrator | +0.666 |
Crest | +0.665 |
Neg logit contributions | |
isa | -0.747 |
awa | -0.727 |
favour | -0.721 |
ophon | -0.716 |
ethical | -0.669 |
Token | 0 |
Token ID | 15 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
DEFENSE | +1.764 |
Vert | +1.550 |
HQ | +1.263 |
Administrator | +1.247 |
Crest | +1.245 |
Neg logit contributions | |
isa | -2.168 |
awa | -2.126 |
favour | -2.108 |
ophon | -2.099 |
abet | -1.979 |
Token | s |
Token ID | 82 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
DEFENSE | +1.207 |
Vert | +1.086 |
HQ | +0.926 |
Administrator | +0.921 |
Crest | +0.920 |
Neg logit contributions | |
isa | -0.979 |
awa | -0.957 |
favour | -0.946 |
ophon | -0.940 |
ethical | -0.876 |
Token | )} |
Token ID | 38165 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
DEFENSE | +0.891 |
Vert | +0.801 |
HQ | +0.680 |
Administrator | +0.676 |
Crest | +0.676 |
Neg logit contributions | |
isa | -0.742 |
awa | -0.733 |
favour | -0.719 |
ophon | -0.713 |
ethical | -0.667 |
Token | leg |
Token ID | 1455 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
isa | +0.571 |
awa | +0.550 |
favour | +0.547 |
ophon | +0.545 |
ethical | +0.507 |
Neg logit contributions | |
DEFENSE | -0.707 |
Vert | -0.635 |
HQ | -0.541 |
Administrator | -0.539 |
Crest | -0.538 |
Token | r |
Token ID | 81 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
isa | +0.724 |
awa | +0.701 |
favour | +0.699 |
ophon | +0.696 |
abet | +0.656 |
Neg logit contributions | |
DEFENSE | -0.624 |
Vert | -0.547 |
HQ | -0.449 |
Administrator | -0.445 |
Crest | -0.445 |
Token | ini |
Token ID | 5362 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
DEFENSE | +0.222 |
Vert | +0.191 |
HQ | +0.150 |
Administrator | +0.148 |
Crest | +0.148 |
Neg logit contributions | |
isa | -0.338 |
awa | -0.329 |
favour | -0.328 |
ophon | -0.326 |
ethical | -0.310 |
Token | , |
Token ID | 11 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
isa | +1.864 |
awa | +1.779 |
favour | +1.773 |
ophon | +1.769 |
abet | +1.643 |
Neg logit contributions | |
DEFENSE | -2.398 |
Vert | -2.161 |
HQ | -1.849 |
Administrator | -1.832 |
Crest | -1.829 |
Token | A |
Token ID | 317 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
isa | +2.237 |
favour | +2.141 |
ophon | +2.074 |
awa | +2.050 |
swat | +1.970 |
Neg logit contributions | |
DEFENSE | -2.855 |
Vert | -2.620 |
PDATE | -2.285 |
HQ | -2.218 |
chwitz | -2.198 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
isa | +2.205 |
favour | +2.110 |
awa | +2.107 |
ophon | +2.099 |
abet | +1.970 |
Neg logit contributions | |
DEFENSE | -2.485 |
Vert | -2.221 |
HQ | -1.882 |
Crest | -1.865 |
Administrator | -1.864 |
Token | D |
Token ID | 360 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
isa | +3.879 |
favour | +3.667 |
ophon | +3.536 |
awa | +3.515 |
swat | +3.375 |
Neg logit contributions | |
DEFENSE | -4.829 |
Vert | -4.500 |
PDATE | -3.898 |
7601 | -3.810 |
HQ | -3.787 |
Token | ., |
Token ID | 1539 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
isa | +0.422 |
awa | +0.411 |
favour | +0.409 |
ophon | +0.406 |
ethical | +0.381 |
Neg logit contributions | |
DEFENSE | -0.435 |
Vert | -0.387 |
HQ | -0.324 |
Administrator | -0.323 |
Crest | -0.322 |
Token | Bart |
Token ID | 13167 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
isa | +2.851 |
favour | +2.736 |
awa | +2.705 |
ophon | +2.691 |
abet | +2.548 |
Neg logit contributions | |
DEFENSE | -2.804 |
Vert | -2.474 |
HQ | -2.076 |
7601 | -2.053 |
PDATE | -2.031 |
Token | ini |
Token ID | 5362 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
isa | +1.927 |
ophon | +1.837 |
favour | +1.811 |
awa | +1.802 |
abet | +1.713 |
Neg logit contributions | |
DEFENSE | -2.087 |
Vert | -1.832 |
HQ | -1.552 |
Crest | -1.546 |
Administrator | -1.541 |
Token | , |
Token ID | 11 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
isa | +1.427 |
awa | +1.370 |
favour | +1.365 |
ophon | +1.358 |
ethical | +1.262 |
Neg logit contributions | |
DEFENSE | -1.808 |
Vert | -1.626 |
HQ | -1.388 |
Administrator | -1.380 |
Crest | -1.379 |
Token | )} |
Token ID | 38165 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
isa | +0.316 |
favour | +0.303 |
ophon | +0.301 |
awa | +0.300 |
ethical | +0.279 |
Neg logit contributions | |
DEFENSE | -0.407 |
Vert | -0.366 |
HQ | -0.313 |
Administrator | -0.312 |
Crest | -0.312 |
Token | 22 |
Token ID | 2534 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
isa | +0.852 |
awa | +0.827 |
favour | +0.823 |
ophon | +0.816 |
ethical | +0.764 |
Neg logit contributions | |
DEFENSE | -0.959 |
Vert | -0.858 |
HQ | -0.724 |
Administrator | -0.721 |
Crest | -0.721 |
Token | . |
Token ID | 13 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
DEFENSE | +2.026 |
Vert | +1.817 |
HQ | +1.535 |
Crest | +1.526 |
Administrator | +1.522 |
Neg logit contributions | |
awa | -1.860 |
isa | -1.853 |
favour | -1.804 |
ophon | -1.795 |
ethical | -1.690 |
Token | R |
Token ID | 371 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
isa | +1.072 |
favour | +1.042 |
awa | +1.038 |
ophon | +1.020 |
swat | +0.957 |
Neg logit contributions | |
DEFENSE | -1.278 |
Vert | -1.144 |
HQ | -0.969 |
Administrator | -0.963 |
Crest | -0.960 |
Token | c |
Token ID | 66 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
isa | +0.250 |
awa | +0.241 |
favour | +0.239 |
ophon | +0.237 |
ethical | +0.218 |
Neg logit contributions | |
DEFENSE | -0.406 |
Vert | -0.369 |
HQ | -0.321 |
Administrator | -0.320 |
Crest | -0.320 |
Token | 3 |
Token ID | 18 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
isa | +2.860 |
ophon | +2.718 |
favour | +2.716 |
awa | +2.655 |
ethical | +2.553 |
Neg logit contributions | |
DEFENSE | -2.927 |
Vert | -2.529 |
PDATE | -2.156 |
Administrator | -2.152 |
Crest | -2.132 |
Token | { |
Token ID | 1391 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
isa | +3.286 |
favour | +3.164 |
awa | +3.068 |
ophon | +3.063 |
swat | +2.823 |
Neg logit contributions | |
DEFENSE | -5.601 |
Vert | -5.106 |
Administrator | -4.451 |
HQ | -4.447 |
PDATE | -4.429 |
Token | ( |
Token ID | 7 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
DEFENSE | +0.137 |
Vert | +0.123 |
Administrator | +0.104 |
Crest | +0.104 |
HQ | +0.104 |
Neg logit contributions | |
isa | -0.113 |
favour | -0.108 |
ophon | -0.107 |
awa | -0.107 |
ethical | -0.100 |
Token | 270 |
Token ID | 20233 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
DEFENSE | +0.776 |
Vert | +0.679 |
Administrator | +0.552 |
HQ | +0.552 |
Crest | +0.551 |
Neg logit contributions | |
isa | -0.933 |
awa | -0.905 |
favour | -0.903 |
ophon | -0.897 |
ethical | -0.847 |
Token | s |
Token ID | 82 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
isa | +0.941 |
favour | +0.899 |
ophon | +0.882 |
awa | +0.867 |
swat | +0.824 |
Neg logit contributions | |
DEFENSE | -1.266 |
Vert | -1.136 |
Crest | -0.976 |
Administrator | -0.974 |
HQ | -0.968 |
Token | )} |
Token ID | 38165 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
isa | +0.656 |
favour | +0.627 |
ophon | +0.623 |
awa | +0.616 |
abet | +0.577 |
Neg logit contributions | |
DEFENSE | -0.861 |
Vert | -0.773 |
HQ | -0.663 |
Administrator | -0.662 |
Crest | -0.662 |
Token | $ |
Token ID | 720 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
isa | +0.464 |
awa | +0.448 |
favour | +0.446 |
ophon | +0.443 |
swat | +0.416 |
Neg logit contributions | |
DEFENSE | -0.530 |
Vert | -0.472 |
HQ | -0.400 |
Administrator | -0.399 |
Crest | -0.399 |
Token | s |
Token ID | 82 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
DEFENSE | +0.272 |
Vert | +0.237 |
Administrator | +0.199 |
Crest | +0.198 |
HQ | +0.196 |
Neg logit contributions | |
isa | -0.285 |
ophon | -0.272 |
favour | -0.272 |
awa | -0.270 |
ethical | -0.256 |
Token | Format |
Token ID | 26227 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
isa | +0.591 |
ophon | +0.564 |
favour | +0.562 |
awa | +0.560 |
abet | +0.522 |
Neg logit contributions | |
DEFENSE | -0.781 |
Vert | -0.697 |
Crest | -0.601 |
Administrator | -0.600 |
HQ | -0.598 |
Token | ) |
Token ID | 8 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
isa | +0.106 |
awa | +0.101 |
favour | +0.101 |
ophon | +0.101 |
swat | +0.094 |
Neg logit contributions | |
DEFENSE | -0.135 |
Vert | -0.121 |
Crest | -0.104 |
HQ | -0.103 |
Administrator | -0.103 |
Token | } |
Token ID | 1782 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
isa | +1.658 |
favour | +1.615 |
awa | +1.614 |
ophon | +1.602 |
swat | +1.515 |
Neg logit contributions | |
DEFENSE | -1.801 |
Vert | -1.586 |
HQ | -1.354 |
Crest | -1.328 |
Administrator | -1.322 |
Token | else |
Token ID | 2073 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
isa | +3.032 |
awa | +2.977 |
favour | +2.963 |
ophon | +2.904 |
swat | +2.793 |
Neg logit contributions | |
DEFENSE | -3.354 |
Vert | -2.943 |
HQ | -2.537 |
Ground | -2.471 |
7601 | -2.458 |
Token | { |
Token ID | 1391 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
isa | +3.316 |
ophon | +3.140 |
favour | +3.138 |
awa | +3.114 |
swat | +2.886 |
Neg logit contributions | |
DEFENSE | -5.562 |
Vert | -4.996 |
HQ | -4.418 |
Crest | -4.388 |
Administrator | -4.377 |
Token | $ |
Token ID | 720 |
Feature activation | -5.84e-03 |
Pos logit contributions | |
DEFENSE | +0.239 |
Vert | +0.215 |
HQ | +0.184 |
Crest | +0.183 |
Administrator | +0.183 |
Neg logit contributions | |
isa | -0.185 |
awa | -0.179 |
favour | -0.178 |
ophon | -0.177 |
swat | -0.165 |
Token | font |
Token ID | 10331 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
DEFENSE | +0.019 |
Vert | +0.016 |
Crest | +0.014 |
Administrator | +0.014 |
HQ | +0.013 |
Neg logit contributions | |
isa | -0.022 |
ophon | -0.022 |
favour | -0.022 |
awa | -0.021 |
ethical | -0.020 |
Token | 2 |
Token ID | 17 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
isa | +0.260 |
awa | +0.249 |
favour | +0.249 |
ophon | +0.249 |
abet | +0.233 |
Neg logit contributions | |
DEFENSE | -0.269 |
Vert | -0.236 |
Crest | -0.200 |
HQ | -0.199 |
Administrator | -0.199 |
Token | = |
Token ID | 796 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
isa | +0.637 |
ophon | +0.594 |
favour | +0.589 |
awa | +0.575 |
abet | +0.537 |
Neg logit contributions | |
DEFENSE | -1.260 |
Vert | -1.138 |
Crest | -1.006 |
Administrator | -1.006 |
HQ | -1.003 |
Token | Civil |
Token ID | 32610 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
DEFENSE | +1.677 |
Vert | +1.503 |
HQ | +1.179 |
Ground | +1.167 |
Administrator | +1.145 |
Neg logit contributions | |
awa | -2.498 |
isa | -2.468 |
favour | -2.450 |
ophon | -2.405 |
swat | -2.305 |
Token | Beat |
Token ID | 12568 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
DEFENSE | +0.319 |
Vert | +0.286 |
HQ | +0.242 |
Administrator | +0.240 |
Crest | +0.239 |
Neg logit contributions | |
isa | -0.282 |
favour | -0.270 |
ophon | -0.269 |
awa | -0.267 |
swat | -0.252 |
Token |
|
Token ID | 198 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
DEFENSE | +1.732 |
Vert | +1.531 |
Administrator | +1.302 |
HQ | +1.299 |
Crest | +1.293 |
Neg logit contributions | |
isa | -1.586 |
awa | -1.550 |
favour | -1.548 |
ophon | -1.519 |
ethical | -1.437 |
Token |
|
Token ID | 198 |
Feature activation | -2.15e-03 |
Pos logit contributions | |
isa | +0.163 |
favour | +0.156 |
ophon | +0.154 |
awa | +0.152 |
abet | +0.144 |
Neg logit contributions | |
DEFENSE | -0.204 |
Vert | -0.182 |
HQ | -0.155 |
Crest | -0.155 |
Administrator | -0.155 |
Token | C |
Token ID | 34 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
DEFENSE | +0.008 |
Vert | +0.007 |
Crest | +0.006 |
Administrator | +0.006 |
Ft | +0.006 |
Neg logit contributions | |
isa | -0.008 |
awa | -0.007 |
ophon | -0.007 |
favour | -0.007 |
ethical | -0.007 |
Token | ald |
Token ID | 1940 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
DEFENSE | +1.265 |
Vert | +1.121 |
HQ | +0.931 |
Administrator | +0.914 |
Crest | +0.909 |
Neg logit contributions | |
isa | -1.470 |
awa | -1.460 |
favour | -1.443 |
ophon | -1.427 |
swat | -1.351 |
Token | well |
Token ID | 4053 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
DEFENSE | +5.464 |
Vert | +4.896 |
Administrator | +4.299 |
onel | +4.236 |
ows | +4.223 |
Neg logit contributions | |
awa | -5.046 |
favour | -4.824 |
isa | -4.724 |
ophon | -4.569 |
swat | -4.453 |
Token | also |
Token ID | 635 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
isa | +0.909 |
favour | +0.877 |
awa | +0.877 |
ophon | +0.869 |
swat | +0.820 |
Neg logit contributions | |
DEFENSE | -0.874 |
Vert | -0.780 |
HQ | -0.646 |
7601 | -0.636 |
Crest | -0.636 |
Token | wants |
Token ID | 3382 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
DEFENSE | +0.699 |
Vert | +0.621 |
HQ | +0.518 |
Crest | +0.515 |
Administrator | +0.515 |
Neg logit contributions | |
isa | -0.695 |
awa | -0.676 |
favour | -0.673 |
ophon | -0.668 |
ethical | -0.627 |
Token | to |
Token ID | 284 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
DEFENSE | +1.100 |
Vert | +0.983 |
HQ | +0.830 |
Administrator | +0.828 |
Crest | +0.827 |
Neg logit contributions | |
isa | -0.980 |
awa | -0.951 |
favour | -0.946 |
ophon | -0.940 |
ethical | -0.880 |
Token | enact |
Token ID | 10865 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
DEFENSE | +0.712 |
Vert | +0.640 |
HQ | +0.545 |
Administrator | +0.542 |
Crest | +0.542 |
Neg logit contributions | |
isa | -0.576 |
awa | -0.558 |
favour | -0.556 |
ophon | -0.551 |
swat | -0.513 |
Token | of |
Token ID | 286 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
isa | +0.887 |
awa | +0.854 |
favour | +0.848 |
ophon | +0.840 |
ethical | +0.772 |
Neg logit contributions | |
DEFENSE | -1.471 |
Vert | -1.339 |
HQ | -1.165 |
Crest | -1.160 |
Administrator | -1.159 |
Token | � |
Token ID | 564 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
isa | +1.923 |
awa | +1.872 |
favour | +1.863 |
ophon | +1.827 |
swat | +1.727 |
Neg logit contributions | |
DEFENSE | -2.111 |
Vert | -1.881 |
HQ | -1.603 |
7601 | -1.579 |
Administrator | -1.575 |
Token | � |
Token ID | 250 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
isa | +0.418 |
favour | +0.397 |
ophon | +0.393 |
awa | +0.393 |
abet | +0.368 |
Neg logit contributions | |
DEFENSE | -0.670 |
Vert | -0.607 |
Crest | -0.530 |
HQ | -0.529 |
Administrator | -0.526 |
Token | Right |
Token ID | 11028 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
DEFENSE | +0.414 |
Vert | +0.365 |
HQ | +0.303 |
Administrator | +0.303 |
Crest | +0.303 |
Neg logit contributions | |
isa | -0.435 |
awa | -0.422 |
favour | -0.420 |
ophon | -0.418 |
ethical | -0.394 |
Token | eous |
Token ID | 15303 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
DEFENSE | +0.721 |
Vert | +0.649 |
HQ | +0.554 |
Administrator | +0.552 |
Crest | +0.552 |
Neg logit contributions | |
isa | -0.563 |
awa | -0.545 |
favour | -0.542 |
ophon | -0.538 |
ethical | -0.501 |
Token | : |
Token ID | 25 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
isa | +0.168 |
favour | +0.162 |
awa | +0.162 |
ophon | +0.160 |
swat | +0.151 |
Neg logit contributions | |
DEFENSE | -0.165 |
Vert | -0.146 |
HQ | -0.123 |
7601 | -0.121 |
Administrator | -0.121 |
Token | Dis |
Token ID | 3167 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
DEFENSE | +4.925 |
Vert | +4.381 |
Crest | +3.730 |
Administrator | +3.679 |
Ft | +3.658 |
Neg logit contributions | |
isa | -4.887 |
awa | -4.816 |
favour | -4.786 |
ophon | -4.770 |
ethical | -4.442 |
Token | pat |
Token ID | 8071 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
DEFENSE | +1.960 |
Vert | +1.790 |
HQ | +1.550 |
Crest | +1.540 |
onel | +1.538 |
Neg logit contributions | |
isa | -1.234 |
awa | -1.231 |
favour | -1.211 |
ophon | -1.190 |
swat | -1.110 |
Token | ches |
Token ID | 2052 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
isa | +0.683 |
awa | +0.649 |
favour | +0.645 |
ophon | +0.638 |
ethical | +0.575 |
Neg logit contributions | |
DEFENSE | -1.565 |
Vert | -1.436 |
HQ | -1.272 |
Administrator | -1.268 |
Crest | -1.267 |
Token | from |
Token ID | 422 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
DEFENSE | +1.667 |
Vert | +1.475 |
Crest | +1.227 |
Administrator | +1.225 |
HQ | +1.221 |
Neg logit contributions | |
isa | -1.760 |
awa | -1.715 |
favour | -1.704 |
ophon | -1.699 |
ethical | -1.594 |
Token | the |
Token ID | 262 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
DEFENSE | +1.804 |
Vert | +1.604 |
Crest | +1.344 |
Administrator | +1.342 |
HQ | +1.338 |
Neg logit contributions | |
isa | -1.765 |
awa | -1.720 |
favour | -1.710 |
ophon | -1.701 |
ethical | -1.594 |
Token | th |
Token ID | 400 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
isa | +0.215 |
awa | +0.210 |
favour | +0.208 |
ophon | +0.207 |
ethical | +0.194 |
Neg logit contributions | |
DEFENSE | -0.236 |
Vert | -0.211 |
HQ | -0.178 |
Administrator | -0.177 |
Crest | -0.177 |
Token | U |
Token ID | 471 |
Feature activation | +5.05e-02 |
Pos logit contributions | |
isa | +0.653 |
awa | +0.633 |
favour | +0.633 |
ophon | +0.624 |
swat | +0.594 |
Neg logit contributions | |
DEFENSE | -0.546 |
Vert | -0.482 |
HQ | -0.392 |
Crest | -0.389 |
Administrator | -0.387 |
Token | . |
Token ID | 13 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
isa | +0.163 |
awa | +0.158 |
favour | +0.158 |
ophon | +0.156 |
ethical | +0.146 |
Neg logit contributions | |
DEFENSE | -0.193 |
Vert | -0.173 |
HQ | -0.147 |
Crest | -0.146 |
Administrator | -0.146 |
Token | S |
Token ID | 50 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
DEFENSE | +1.701 |
Vert | +1.495 |
HQ | +1.197 |
Administrator | +1.184 |
Crest | +1.175 |
Neg logit contributions | |
awa | -2.357 |
isa | -2.345 |
favour | -2.300 |
ophon | -2.285 |
ethical | -2.181 |
Token | . |
Token ID | 13 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
DEFENSE | +2.214 |
Vert | +1.975 |
HQ | +1.667 |
Administrator | +1.666 |
Crest | +1.666 |
Neg logit contributions | |
isa | -1.974 |
awa | -1.949 |
favour | -1.913 |
ophon | -1.902 |
ethical | -1.788 |
Token | Circuit |
Token ID | 13588 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
DEFENSE | +1.016 |
Vert | +0.863 |
HQ | +0.662 |
Administrator | +0.658 |
Crest | +0.657 |
Neg logit contributions | |
isa | -1.715 |
awa | -1.679 |
favour | -1.671 |
ophon | -1.663 |
ethical | -1.583 |
Token | Court |
Token ID | 3078 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
DEFENSE | +3.677 |
Vert | +3.125 |
Administrator | +2.525 |
Crest | +2.449 |
Ft | +2.406 |
Neg logit contributions | |
isa | -5.626 |
awa | -5.609 |
ophon | -5.538 |
favour | -5.508 |
ethical | -5.241 |
Token | of |
Token ID | 286 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
DEFENSE | +1.443 |
Vert | +1.310 |
Administrator | +1.143 |
Crest | +1.136 |
HQ | +1.134 |
Neg logit contributions | |
isa | -0.908 |
awa | -0.883 |
ophon | -0.870 |
favour | -0.869 |
ethical | -0.800 |
Token | Appeal |
Token ID | 31876 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
isa | +2.156 |
favour | +2.099 |
ophon | +2.042 |
awa | +1.998 |
abet | +1.945 |
Neg logit contributions | |
DEFENSE | -2.052 |
Vert | -1.869 |
HQ | -1.587 |
PDATE | -1.554 |
onel | -1.533 |
Token | 's |
Token ID | 338 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
DEFENSE | +2.300 |
Vert | +2.047 |
HQ | +1.715 |
Administrator | +1.707 |
Crest | +1.707 |
Neg logit contributions | |
isa | -2.208 |
awa | -2.147 |
favour | -2.136 |
ophon | -2.121 |
ethical | -1.990 |
Token | 3 |
Token ID | 513 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
isa | +0.634 |
favour | +0.617 |
awa | +0.606 |
ophon | +0.600 |
swat | +0.567 |
Neg logit contributions | |
DEFENSE | -0.667 |
Vert | -0.596 |
HQ | -0.506 |
7601 | -0.495 |
onel | -0.492 |
Token | ive |
Token ID | 425 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
isa | +0.377 |
favour | +0.367 |
awa | +0.359 |
ophon | +0.356 |
abet | +0.336 |
Neg logit contributions | |
DEFENSE | -0.385 |
Vert | -0.345 |
HQ | -0.289 |
7601 | -0.288 |
Crest | -0.287 |
Token | of |
Token ID | 286 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
isa | +0.561 |
favour | +0.542 |
awa | +0.536 |
ophon | +0.524 |
swat | +0.492 |
Neg logit contributions | |
DEFENSE | -0.727 |
Vert | -0.659 |
HQ | -0.563 |
Crest | -0.557 |
7601 | -0.555 |
Token | Peoples |
Token ID | 41770 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
isa | +0.103 |
favour | +0.100 |
awa | +0.099 |
ophon | +0.098 |
swat | +0.092 |
Neg logit contributions | |
DEFENSE | -0.102 |
Vert | -0.091 |
HQ | -0.076 |
Crest | -0.075 |
7601 | -0.075 |
Token | ' |
Token ID | 6 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
DEFENSE | +1.739 |
Vert | +1.564 |
Administrator | +1.340 |
HQ | +1.339 |
Crest | +1.337 |
Neg logit contributions | |
isa | -1.313 |
awa | -1.277 |
favour | -1.259 |
ophon | -1.258 |
ethical | -1.168 |
Token | Comm |
Token ID | 1520 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
DEFENSE | +0.901 |
Vert | +0.804 |
HQ | +0.667 |
7601 | +0.661 |
Crest | +0.661 |
Neg logit contributions | |
isa | -0.917 |
favour | -0.894 |
awa | -0.887 |
ophon | -0.877 |
abet | -0.826 |
Token | iss |
Token ID | 747 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
DEFENSE | +2.527 |
Vert | +2.259 |
HQ | +1.920 |
Administrator | +1.900 |
Crest | +1.891 |
Neg logit contributions | |
awa | -2.320 |
isa | -2.307 |
favour | -2.274 |
ophon | -2.253 |
swat | -2.126 |
Token | ariat |
Token ID | 21621 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
DEFENSE | +4.582 |
Vert | +4.056 |
HQ | +3.391 |
Administrator | +3.366 |
Ft | +3.359 |
Neg logit contributions | |
awa | -4.645 |
isa | -4.641 |
favour | -4.512 |
ophon | -4.479 |
swat | -4.258 |
Token | of |
Token ID | 286 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
DEFENSE | +2.016 |
Vert | +1.813 |
Administrator | +1.553 |
HQ | +1.553 |
Crest | +1.547 |
Neg logit contributions | |
isa | -1.547 |
awa | -1.507 |
favour | -1.487 |
ophon | -1.484 |
ethical | -1.382 |
Token | Defence |
Token ID | 16721 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
isa | +0.157 |
favour | +0.154 |
awa | +0.151 |
ophon | +0.151 |
swat | +0.143 |
Neg logit contributions | |
DEFENSE | -0.117 |
Vert | -0.104 |
HQ | -0.084 |
7601 | -0.082 |
Crest | -0.082 |
Token | No |
Token ID | 1400 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
DEFENSE | +1.699 |
Vert | +1.528 |
HQ | +1.307 |
Administrator | +1.304 |
Crest | +1.301 |
Neg logit contributions | |
isa | -1.321 |
awa | -1.282 |
favour | -1.270 |
ophon | -1.264 |
ethical | -1.176 |
Token | . |
Token ID | 13 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
DEFENSE | +1.256 |
Vert | +1.123 |
HQ | +0.946 |
Administrator | +0.944 |
Crest | +0.943 |
Neg logit contributions | |
isa | -1.128 |
awa | -1.103 |
favour | -1.093 |
ophon | -1.086 |
ethical | -1.016 |
Token | , |
Token ID | 11 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
DEFENSE | +0.068 |
Vert | +0.060 |
HQ | +0.051 |
Crest | +0.050 |
Administrator | +0.050 |
Neg logit contributions | |
isa | -0.063 |
favour | -0.061 |
awa | -0.061 |
ophon | -0.060 |
swat | -0.057 |
Token | especially |
Token ID | 2592 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
DEFENSE | +0.022 |
Vert | +0.020 |
HQ | +0.016 |
Crest | +0.016 |
Administrator | +0.016 |
Neg logit contributions | |
isa | -0.022 |
awa | -0.021 |
favour | -0.021 |
ophon | -0.021 |
swat | -0.019 |
Token | in |
Token ID | 287 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
DEFENSE | +2.618 |
Vert | +2.320 |
Administrator | +1.947 |
HQ | +1.943 |
Crest | +1.938 |
Neg logit contributions | |
isa | -2.614 |
awa | -2.548 |
ophon | -2.510 |
favour | -2.508 |
ethical | -2.367 |
Token | B |
Token ID | 347 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
DEFENSE | +0.591 |
Vert | +0.522 |
HQ | +0.431 |
Administrator | +0.428 |
Crest | +0.427 |
Neg logit contributions | |
isa | -0.637 |
favour | -0.621 |
awa | -0.619 |
ophon | -0.613 |
swat | -0.578 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
DEFENSE | +0.063 |
Vert | +0.057 |
HQ | +0.049 |
Administrator | +0.048 |
Crest | +0.048 |
Neg logit contributions | |
isa | -0.057 |
awa | -0.056 |
favour | -0.055 |
ophon | -0.055 |
ethical | -0.052 |
Token | C |
Token ID | 34 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
DEFENSE | +0.607 |
Vert | +0.537 |
Administrator | +0.448 |
HQ | +0.447 |
Crest | +0.447 |
Neg logit contributions | |
isa | -0.577 |
favour | -0.557 |
awa | -0.553 |
ophon | -0.550 |
swat | -0.521 |
Token | ., |
Token ID | 1539 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
DEFENSE | +4.151 |
Vert | +3.668 |
HQ | +3.052 |
Administrator | +3.037 |
Crest | +3.036 |
Neg logit contributions | |
isa | -4.904 |
awa | -4.882 |
favour | -4.770 |
ophon | -4.750 |
ethical | -4.488 |
Token | some |
Token ID | 617 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
isa | +0.430 |
favour | +0.419 |
awa | +0.416 |
ophon | +0.412 |
swat | +0.389 |
Neg logit contributions | |
DEFENSE | -0.444 |
Vert | -0.396 |
HQ | -0.329 |
Crest | -0.328 |
Administrator | -0.326 |
Token | of |
Token ID | 286 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
isa | +0.698 |
favour | +0.679 |
awa | +0.677 |
ophon | +0.668 |
swat | +0.625 |
Neg logit contributions | |
DEFENSE | -0.886 |
Vert | -0.798 |
HQ | -0.679 |
Administrator | -0.675 |
Crest | -0.675 |
Token | the |
Token ID | 262 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
isa | +0.675 |
favour | +0.656 |
awa | +0.651 |
ophon | +0.647 |
swat | +0.605 |
Neg logit contributions | |
DEFENSE | -0.766 |
Vert | -0.684 |
HQ | -0.575 |
Administrator | -0.571 |
7601 | -0.571 |
Token | more |
Token ID | 517 |
Feature activation | -5.72e-02 |
Pos logit contributions | |
isa | +0.306 |
favour | +0.300 |
awa | +0.297 |
ophon | +0.293 |
swat | +0.277 |
Neg logit contributions | |
DEFENSE | -0.333 |
Vert | -0.298 |
HQ | -0.250 |
7601 | -0.248 |
Administrator | -0.247 |
Token | , |
Token ID | 11 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
isa | +0.483 |
favour | +0.475 |
awa | +0.471 |
ophon | +0.469 |
swat | +0.447 |
Neg logit contributions | |
DEFENSE | -0.279 |
Vert | -0.236 |
Crest | -0.180 |
HQ | -0.178 |
Administrator | -0.177 |
Token | Soviet |
Token ID | 7570 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
isa | +1.301 |
favour | +1.292 |
ophon | +1.256 |
awa | +1.249 |
swat | +1.184 |
Neg logit contributions | |
DEFENSE | -1.373 |
Vert | -1.247 |
Crest | -1.030 |
HQ | -1.028 |
7601 | -1.022 |
Token | People |
Token ID | 4380 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
DEFENSE | +1.320 |
Vert | +1.165 |
HQ | +0.930 |
Crest | +0.924 |
7601 | +0.922 |
Neg logit contributions | |
isa | -1.692 |
favour | -1.660 |
awa | -1.642 |
ophon | -1.625 |
swat | -1.549 |
Token | 's |
Token ID | 338 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
DEFENSE | +2.488 |
Vert | +2.179 |
Administrator | +1.817 |
HQ | +1.808 |
Crest | +1.794 |
Neg logit contributions | |
isa | -2.798 |
awa | -2.748 |
favour | -2.703 |
ophon | -2.690 |
ethical | -2.552 |
Token | Comm |
Token ID | 1520 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
isa | +1.020 |
favour | +0.998 |
ophon | +0.986 |
awa | +0.982 |
swat | +0.935 |
Neg logit contributions | |
DEFENSE | -0.585 |
Vert | -0.503 |
7601 | -0.382 |
HQ | -0.372 |
Crest | -0.371 |
Token | iss |
Token ID | 747 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
DEFENSE | +2.092 |
Vert | +1.864 |
HQ | +1.601 |
Administrator | +1.578 |
Ft | +1.574 |
Neg logit contributions | |
isa | -1.869 |
awa | -1.854 |
favour | -1.824 |
ophon | -1.815 |
ethical | -1.706 |
Token | ar |
Token ID | 283 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
DEFENSE | +4.193 |
Vert | +3.663 |
HQ | +3.044 |
Administrator | +3.001 |
Ft | +2.995 |
Neg logit contributions | |
isa | -4.900 |
awa | -4.884 |
favour | -4.758 |
ophon | -4.736 |
ethical | -4.512 |
Token | of |
Token ID | 286 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
DEFENSE | +2.684 |
Vert | +2.391 |
Administrator | +2.017 |
HQ | +2.014 |
Crest | +2.007 |
Neg logit contributions | |
isa | -2.519 |
awa | -2.467 |
favour | -2.438 |
ophon | -2.428 |
ethical | -2.281 |
Token | Defense |
Token ID | 5947 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
isa | +0.424 |
favour | +0.413 |
ophon | +0.407 |
awa | +0.404 |
swat | +0.385 |
Neg logit contributions | |
DEFENSE | -0.345 |
Vert | -0.308 |
HQ | -0.250 |
7601 | -0.248 |
Crest | -0.246 |
Token | S |
Token ID | 311 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
DEFENSE | +1.041 |
Vert | +0.931 |
HQ | +0.774 |
Crest | +0.770 |
7601 | +0.767 |
Neg logit contributions | |
isa | -0.999 |
favour | -0.968 |
awa | -0.964 |
ophon | -0.956 |
swat | -0.897 |
Token | emy |
Token ID | 3065 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
DEFENSE | +0.614 |
Vert | +0.551 |
HQ | +0.469 |
Administrator | +0.467 |
Crest | +0.467 |
Neg logit contributions | |
isa | -0.504 |
awa | -0.491 |
favour | -0.487 |
ophon | -0.483 |
ethical | -0.451 |
Token | attempt |
Token ID | 2230 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
isa | +1.235 |
favour | +1.206 |
awa | +1.200 |
ophon | +1.167 |
swat | +1.128 |
Neg logit contributions | |
DEFENSE | -1.460 |
Vert | -1.321 |
HQ | -1.112 |
Administrator | -1.108 |
7601 | -1.102 |
Token | at |
Token ID | 379 |
Feature activation | -7.58e-02 |
Pos logit contributions | |
isa | +0.683 |
favour | +0.662 |
awa | +0.661 |
ophon | +0.651 |
swat | +0.610 |
Neg logit contributions | |
DEFENSE | -0.828 |
Vert | -0.744 |
HQ | -0.632 |
Administrator | -0.629 |
Crest | -0.628 |
Token | the |
Token ID | 262 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
DEFENSE | +0.281 |
Vert | +0.251 |
HQ | +0.211 |
Administrator | +0.210 |
Crest | +0.210 |
Neg logit contributions | |
isa | -0.255 |
awa | -0.247 |
favour | -0.247 |
ophon | -0.244 |
swat | -0.230 |
Token | B |
Token ID | 347 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
isa | +0.802 |
awa | +0.781 |
favour | +0.780 |
ophon | +0.769 |
swat | +0.731 |
Neg logit contributions | |
DEFENSE | -0.726 |
Vert | -0.641 |
HQ | -0.528 |
Administrator | -0.524 |
Crest | -0.521 |
Token | . |
Token ID | 13 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
DEFENSE | +0.057 |
Vert | +0.051 |
HQ | +0.043 |
Administrator | +0.043 |
Crest | +0.043 |
Neg logit contributions | |
isa | -0.049 |
awa | -0.048 |
favour | -0.047 |
ophon | -0.047 |
ethical | -0.044 |
Token | C |
Token ID | 34 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
isa | +0.156 |
favour | +0.151 |
awa | +0.151 |
ophon | +0.149 |
swat | +0.141 |
Neg logit contributions | |
DEFENSE | -0.160 |
Vert | -0.142 |
HQ | -0.119 |
Administrator | -0.118 |
Crest | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
DEFENSE | +4.389 |
Vert | +3.823 |
Crest | +3.284 |
Ft | +3.240 |
HQ | +3.195 |
Neg logit contributions | |
isa | -4.484 |
awa | -4.363 |
ophon | -4.329 |
favour | -4.301 |
abet | -4.062 |
Token | Legislature |
Token ID | 20815 |
Feature activation | -7.22e-02 |
Pos logit contributions | |
DEFENSE | +0.196 |
Vert | +0.174 |
HQ | +0.144 |
Administrator | +0.143 |
Crest | +0.143 |
Neg logit contributions | |
isa | -0.203 |
awa | -0.198 |
favour | -0.197 |
ophon | -0.195 |
swat | -0.184 |
Token | ." |
Token ID | 526 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
DEFENSE | +0.263 |
Vert | +0.234 |
HQ | +0.196 |
Crest | +0.195 |
Administrator | +0.195 |
Neg logit contributions | |
isa | -0.247 |
awa | -0.240 |
favour | -0.240 |
ophon | -0.237 |
swat | -0.223 |
Token |
|
Token ID | 198 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
isa | +0.373 |
awa | +0.362 |
favour | +0.360 |
ophon | +0.358 |
ethical | +0.334 |
Neg logit contributions | |
DEFENSE | -0.443 |
Vert | -0.397 |
HQ | -0.337 |
Administrator | -0.335 |
Crest | -0.335 |
Token |
|
Token ID | 198 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
DEFENSE | +1.407 |
Vert | +1.262 |
HQ | +1.068 |
Administrator | +1.063 |
Crest | +1.062 |
Neg logit contributions | |
isa | -1.233 |
awa | -1.214 |
favour | -1.193 |
ophon | -1.187 |
ethical | -1.111 |
Token | ahl |
Token ID | 15668 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
DEFENSE | +1.846 |
Vert | +1.665 |
HQ | +1.422 |
Crest | +1.419 |
Administrator | +1.419 |
Neg logit contributions | |
isa | -1.469 |
awa | -1.447 |
favour | -1.427 |
ophon | -1.410 |
ethical | -1.326 |
Token | from |
Token ID | 422 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
DEFENSE | +0.673 |
Vert | +0.600 |
HQ | +0.505 |
Crest | +0.502 |
Administrator | +0.502 |
Neg logit contributions | |
isa | -0.622 |
awa | -0.602 |
favour | -0.600 |
ophon | -0.596 |
ethical | -0.558 |
Token | the |
Token ID | 262 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
DEFENSE | +0.205 |
Vert | +0.183 |
HQ | +0.152 |
Administrator | +0.151 |
7601 | +0.150 |
Neg logit contributions | |
isa | -0.210 |
awa | -0.203 |
favour | -0.203 |
ophon | -0.201 |
swat | -0.189 |
Token | Max |
Token ID | 5436 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
isa | +0.288 |
favour | +0.281 |
awa | +0.280 |
ophon | +0.277 |
ethical | +0.265 |
Neg logit contributions | |
DEFENSE | -0.177 |
Vert | -0.152 |
HQ | -0.117 |
Administrator | -0.115 |
7601 | -0.115 |
Token | Plan |
Token ID | 5224 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
DEFENSE | +1.110 |
Vert | +0.980 |
HQ | +0.811 |
Administrator | +0.808 |
Crest | +0.807 |
Neg logit contributions | |
isa | -1.179 |
awa | -1.144 |
favour | -1.139 |
ophon | -1.135 |
ethical | -1.067 |
Token | ck |
Token ID | 694 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
DEFENSE | +0.926 |
Vert | +0.816 |
HQ | +0.672 |
Administrator | +0.668 |
Crest | +0.668 |
Neg logit contributions | |
isa | -1.034 |
awa | -1.006 |
favour | -1.002 |
ophon | -0.996 |
ethical | -0.939 |
Token | Institute |
Token ID | 5136 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
DEFENSE | +3.006 |
Vert | +2.594 |
HQ | +1.947 |
Administrator | +1.914 |
Ft | +1.903 |
Neg logit contributions | |
isa | -5.770 |
awa | -5.729 |
favour | -5.639 |
ophon | -5.582 |
swat | -5.354 |
Token | for |
Token ID | 329 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
isa | +0.551 |
favour | +0.540 |
ophon | +0.532 |
awa | +0.531 |
ethical | +0.498 |
Neg logit contributions | |
DEFENSE | -0.640 |
Vert | -0.569 |
HQ | -0.486 |
Crest | -0.482 |
7601 | -0.480 |
Token | Dem |
Token ID | 1897 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
isa | +1.213 |
favour | +1.182 |
awa | +1.169 |
ophon | +1.160 |
ethical | +1.098 |
Neg logit contributions | |
DEFENSE | -1.226 |
Vert | -1.091 |
HQ | -0.916 |
7601 | -0.899 |
Administrator | -0.895 |
Token | ographic |
Token ID | 6826 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
isa | +1.082 |
awa | +1.045 |
ophon | +1.041 |
favour | +1.041 |
ethical | +0.982 |
Neg logit contributions | |
DEFENSE | -0.953 |
Vert | -0.838 |
Crest | -0.687 |
HQ | -0.685 |
Administrator | -0.683 |
Token | Research |
Token ID | 4992 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
DEFENSE | +0.393 |
Vert | +0.344 |
HQ | +0.279 |
Crest | +0.277 |
Administrator | +0.276 |
Neg logit contributions | |
isa | -0.478 |
favour | -0.464 |
awa | -0.464 |
ophon | -0.460 |
ethical | -0.436 |
Token | Scientists |
Token ID | 23782 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
isa | +0.751 |
favour | +0.713 |
ophon | +0.707 |
awa | +0.705 |
abet | +0.662 |
Neg logit contributions | |
DEFENSE | -0.941 |
Vert | -0.849 |
HQ | -0.722 |
onel | -0.718 |
PDATE | -0.716 |
Token | at |
Token ID | 379 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
isa | +0.465 |
favour | +0.452 |
awa | +0.449 |
ophon | +0.441 |
swat | +0.414 |
Neg logit contributions | |
DEFENSE | -0.590 |
Vert | -0.536 |
HQ | -0.455 |
Crest | -0.452 |
Administrator | -0.450 |
Token | the |
Token ID | 262 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
DEFENSE | +1.481 |
Vert | +1.302 |
Crest | +1.088 |
HQ | +1.080 |
Administrator | +1.079 |
Neg logit contributions | |
isa | -1.580 |
awa | -1.545 |
ophon | -1.532 |
favour | -1.531 |
ethical | -1.441 |
Token | Max |
Token ID | 5436 |
Feature activation | +6.22e-02 |
Pos logit contributions | |
isa | +0.661 |
awa | +0.644 |
favour | +0.643 |
ophon | +0.636 |
ethical | +0.607 |
Neg logit contributions | |
DEFENSE | -0.414 |
Vert | -0.356 |
HQ | -0.276 |
Administrator | -0.272 |
Crest | -0.270 |
Token | Plan |
Token ID | 5224 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
isa | +0.221 |
ophon | +0.213 |
awa | +0.211 |
favour | +0.211 |
ethical | +0.199 |
Neg logit contributions | |
DEFENSE | -0.220 |
Vert | -0.195 |
HQ | -0.161 |
Administrator | -0.161 |
Crest | -0.160 |
Token | ck |
Token ID | 694 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
DEFENSE | +0.663 |
Vert | +0.584 |
HQ | +0.480 |
Administrator | +0.478 |
Crest | +0.478 |
Neg logit contributions | |
isa | -0.742 |
awa | -0.723 |
favour | -0.719 |
ophon | -0.715 |
ethical | -0.674 |
Token | Institute |
Token ID | 5136 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
DEFENSE | +2.991 |
Vert | +2.571 |
HQ | +1.916 |
Administrator | +1.876 |
Crest | +1.864 |
Neg logit contributions | |
isa | -5.804 |
awa | -5.768 |
favour | -5.671 |
ophon | -5.617 |
swat | -5.393 |
Token | for |
Token ID | 329 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
isa | +1.118 |
favour | +1.101 |
ophon | +1.080 |
awa | +1.078 |
ethical | +1.007 |
Neg logit contributions | |
DEFENSE | -1.308 |
Vert | -1.165 |
HQ | -0.998 |
Crest | -0.985 |
onel | -0.984 |
Token | Heart |
Token ID | 8894 |
Feature activation | +5.37e-03 |
Pos logit contributions | |
isa | +1.415 |
favour | +1.369 |
ophon | +1.360 |
awa | +1.353 |
abet | +1.284 |
Neg logit contributions | |
DEFENSE | -1.479 |
Vert | -1.327 |
HQ | -1.113 |
7601 | -1.094 |
onel | -1.092 |
Token | and |
Token ID | 290 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
isa | +0.020 |
awa | +0.020 |
favour | +0.019 |
ophon | +0.019 |
ethical | +0.018 |
Neg logit contributions | |
DEFENSE | -0.018 |
Vert | -0.016 |
HQ | -0.013 |
Administrator | -0.013 |
Crest | -0.013 |
Token | Lung |
Token ID | 39337 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
isa | +0.673 |
awa | +0.646 |
favour | +0.643 |
ophon | +0.637 |
ethical | +0.582 |
Neg logit contributions | |
DEFENSE | -1.209 |
Vert | -1.104 |
HQ | -0.965 |
Administrator | -0.962 |
Crest | -0.961 |
Token | III |
Token ID | 6711 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
isa | +0.741 |
awa | +0.714 |
favour | +0.711 |
ophon | +0.708 |
swat | +0.663 |
Neg logit contributions | |
DEFENSE | -0.788 |
Vert | -0.704 |
HQ | -0.591 |
Crest | -0.588 |
Administrator | -0.588 |
Token | / |
Token ID | 14 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
DEFENSE | +1.015 |
Vert | +0.900 |
HQ | +0.749 |
Administrator | +0.745 |
Crest | +0.745 |
Neg logit contributions | |
isa | -1.037 |
awa | -1.008 |
favour | -1.003 |
ophon | -0.997 |
ethical | -0.937 |
Token | Dem |
Token ID | 11522 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
DEFENSE | +2.509 |
Vert | +2.221 |
HQ | +1.835 |
Administrator | +1.804 |
Crest | +1.804 |
Neg logit contributions | |
isa | -2.774 |
awa | -2.718 |
favour | -2.710 |
ophon | -2.678 |
abet | -2.538 |
Token | ocrat |
Token ID | 35128 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
isa | +0.072 |
awa | +0.071 |
favour | +0.070 |
ophon | +0.069 |
ethical | +0.065 |
Neg logit contributions | |
DEFENSE | -0.074 |
Vert | -0.067 |
HQ | -0.056 |
Administrator | -0.055 |
Crest | -0.055 |
Token | ) |
Token ID | 8 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
DEFENSE | +0.253 |
Vert | +0.228 |
HQ | +0.198 |
Administrator | +0.196 |
Crest | +0.195 |
Neg logit contributions | |
isa | -0.186 |
awa | -0.182 |
favour | -0.181 |
ophon | -0.179 |
ethical | -0.167 |
Token | Buy |
Token ID | 14518 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
DEFENSE | +1.523 |
Vert | +1.358 |
HQ | +1.150 |
Administrator | +1.149 |
Crest | +1.144 |
Neg logit contributions | |
isa | -1.320 |
awa | -1.291 |
favour | -1.278 |
ophon | -1.273 |
ethical | -1.192 |
Token | Photo |
Token ID | 5555 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
DEFENSE | +3.118 |
Vert | +2.609 |
Crest | +2.013 |
Administrator | +1.993 |
HQ | +1.982 |
Neg logit contributions | |
awa | -5.712 |
isa | -5.690 |
ophon | -5.561 |
favour | -5.548 |
ethical | -5.374 |
Token |
|
Token ID | 198 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
DEFENSE | +2.364 |
Vert | +2.106 |
Administrator | +1.784 |
Crest | +1.772 |
HQ | +1.769 |
Neg logit contributions | |
isa | -2.150 |
awa | -2.105 |
favour | -2.076 |
ophon | -2.072 |
ethical | -1.953 |
Token |
|
Token ID | 198 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
isa | +0.858 |
favour | +0.805 |
ophon | +0.794 |
awa | +0.759 |
abet | +0.743 |
Neg logit contributions | |
DEFENSE | -1.155 |
Vert | -1.029 |
PDATE | -0.924 |
Crest | -0.885 |
onel | -0.882 |
Token | Sign |
Token ID | 11712 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
isa | +0.806 |
ophon | +0.768 |
favour | +0.756 |
awa | +0.755 |
abet | +0.726 |
Neg logit contributions | |
DEFENSE | -0.744 |
Vert | -0.642 |
Crest | -0.554 |
Administrator | -0.549 |
Ft | -0.542 |
Token | ing |
Token ID | 278 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
DEFENSE | +1.202 |
Vert | +1.072 |
HQ | +0.904 |
Administrator | +0.903 |
Crest | +0.901 |
Neg logit contributions | |
isa | -1.087 |
awa | -1.056 |
favour | -1.050 |
ophon | -1.041 |
ethical | -0.976 |
Token | According |
Token ID | 4784 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
DEFENSE | +0.092 |
Vert | +0.083 |
HQ | +0.071 |
Crest | +0.070 |
Administrator | +0.070 |
Neg logit contributions | |
isa | -0.073 |
awa | -0.070 |
favour | -0.070 |
ophon | -0.070 |
abet | -0.065 |
Token | to |
Token ID | 284 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
DEFENSE | +0.856 |
Vert | +0.767 |
HQ | +0.648 |
Administrator | +0.647 |
Crest | +0.646 |
Neg logit contributions | |
isa | -0.753 |
awa | -0.735 |
favour | -0.726 |
ophon | -0.722 |
ethical | -0.676 |
Token | Mur |
Token ID | 5921 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
DEFENSE | +0.626 |
Vert | +0.559 |
HQ | +0.471 |
Administrator | +0.469 |
Crest | +0.469 |
Neg logit contributions | |
isa | -0.568 |
awa | -0.552 |
favour | -0.549 |
ophon | -0.545 |
ethical | -0.510 |
Token | ad |
Token ID | 324 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
DEFENSE | +1.429 |
Vert | +1.295 |
HQ | +1.118 |
Administrator | +1.117 |
Crest | +1.116 |
Neg logit contributions | |
isa | -0.981 |
awa | -0.960 |
favour | -0.951 |
ophon | -0.941 |
ethical | -0.871 |
Token | Do |
Token ID | 2141 |
Feature activation | +9.89e-02 |
Pos logit contributions | |
isa | +1.614 |
awa | +1.547 |
favour | +1.544 |
ophon | +1.534 |
abet | +1.442 |
Neg logit contributions | |
DEFENSE | -1.778 |
Vert | -1.588 |
HQ | -1.342 |
Administrator | -1.324 |
7601 | -1.323 |
Token | ğ |
Token ID | 33133 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
isa | +0.231 |
awa | +0.221 |
favour | +0.219 |
ophon | +0.217 |
ethical | +0.197 |
Neg logit contributions | |
DEFENSE | -0.468 |
Vert | -0.429 |
HQ | -0.378 |
Administrator | -0.377 |
Crest | -0.376 |
Token | uk |
Token ID | 2724 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
DEFENSE | +4.946 |
Vert | +4.496 |
HQ | +3.873 |
onel | +3.857 |
Administrator | +3.855 |
Neg logit contributions | |
awa | -3.644 |
isa | -3.594 |
favour | -3.539 |
ophon | -3.400 |
ethical | -3.275 |
Token | an |
Token ID | 272 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
DEFENSE | +1.233 |
Vert | +1.106 |
HQ | +0.942 |
Crest | +0.942 |
Administrator | +0.941 |
Neg logit contributions | |
isa | -1.007 |
awa | -0.981 |
favour | -0.974 |
ophon | -0.964 |
ethical | -0.900 |
Token | li |
Token ID | 4528 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
DEFENSE | +1.258 |
Vert | +1.128 |
HQ | +0.959 |
Administrator | +0.958 |
Crest | +0.958 |
Neg logit contributions | |
isa | -1.053 |
awa | -1.030 |
favour | -1.023 |
ophon | -1.011 |
ethical | -0.947 |
Token | , |
Token ID | 11 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
isa | +0.258 |
awa | +0.250 |
favour | +0.249 |
ophon | +0.247 |
ethical | +0.230 |
Neg logit contributions | |
DEFENSE | -0.325 |
Vert | -0.293 |
HQ | -0.250 |
Administrator | -0.248 |
Crest | -0.248 |
Token | writing |
Token ID | 3597 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
isa | +1.063 |
favour | +1.028 |
awa | +1.025 |
ophon | +1.018 |
swat | +0.954 |
Neg logit contributions | |
DEFENSE | -1.278 |
Vert | -1.162 |
HQ | -0.985 |
Crest | -0.973 |
Administrator | -0.966 |
Token | in |
Token ID | 287 |
Feature activation | +8.67e-02 |
Pos logit contributions | |
DEFENSE | +0.567 |
Vert | +0.508 |
HQ | +0.427 |
Crest | +0.424 |
Administrator | +0.424 |
Neg logit contributions | |
isa | -0.514 |
favour | -0.499 |
awa | -0.499 |
ophon | -0.492 |
swat | -0.462 |
Token | his |
Token ID | 465 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
isa | +0.357 |
favour | +0.357 |
awa | +0.347 |
ophon | +0.343 |
swat | +0.333 |
Neg logit contributions | |
DEFENSE | -0.255 |
Vert | -0.222 |
HQ | -0.177 |
7601 | -0.173 |
Administrator | -0.172 |
Token | northern |
Token ID | 7840 |
Feature activation | -8.95e-02 |
Pos logit contributions | |
DEFENSE | +0.229 |
Vert | +0.203 |
HQ | +0.170 |
Administrator | +0.168 |
7601 | +0.168 |
Neg logit contributions | |
isa | -0.220 |
favour | -0.219 |
awa | -0.214 |
ophon | -0.212 |
swat | -0.202 |
Token | B |
Token ID | 347 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
DEFENSE | +0.260 |
Vert | +0.225 |
HQ | +0.178 |
Administrator | +0.176 |
Crest | +0.175 |
Neg logit contributions | |
isa | -0.372 |
favour | -0.364 |
awa | -0.363 |
ophon | -0.360 |
swat | -0.343 |
Token | . |
Token ID | 13 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
DEFENSE | +0.174 |
Vert | +0.156 |
HQ | +0.133 |
Administrator | +0.131 |
Crest | +0.131 |
Neg logit contributions | |
isa | -0.157 |
awa | -0.155 |
favour | -0.152 |
ophon | -0.151 |
ethical | -0.142 |
Token | C |
Token ID | 34 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
isa | +0.137 |
favour | +0.132 |
ophon | +0.130 |
awa | +0.129 |
swat | +0.123 |
Neg logit contributions | |
DEFENSE | -0.145 |
Vert | -0.128 |
Crest | -0.107 |
HQ | -0.107 |
Administrator | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
DEFENSE | +4.102 |
Vert | +3.629 |
Crest | +3.049 |
HQ | +3.021 |
Administrator | +3.006 |
Neg logit contributions | |
isa | -4.430 |
awa | -4.378 |
ophon | -4.259 |
favour | -4.193 |
abet | -4.029 |
Token | riding |
Token ID | 10311 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
DEFENSE | +0.180 |
Vert | +0.161 |
HQ | +0.134 |
Administrator | +0.133 |
Crest | +0.133 |
Neg logit contributions | |
isa | -0.172 |
favour | -0.168 |
awa | -0.168 |
ophon | -0.166 |
swat | -0.156 |
Token | who |
Token ID | 508 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
DEFENSE | +1.573 |
Vert | +1.398 |
Crest | +1.187 |
Administrator | +1.182 |
HQ | +1.182 |
Neg logit contributions | |
isa | -1.458 |
awa | -1.415 |
ophon | -1.401 |
favour | -1.388 |
abet | -1.311 |
Token | worry |
Token ID | 5490 |
Feature activation | -7.19e-02 |
Pos logit contributions | |
isa | +0.940 |
favour | +0.927 |
awa | +0.905 |
ophon | +0.893 |
swat | +0.846 |
Neg logit contributions | |
DEFENSE | -1.100 |
Vert | -1.002 |
HQ | -0.847 |
Crest | -0.835 |
Administrator | -0.830 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
DEFENSE | +0.249 |
Vert | +0.221 |
HQ | +0.183 |
Administrator | +0.180 |
Crest | +0.180 |
Neg logit contributions | |
isa | -0.259 |
favour | -0.253 |
awa | -0.252 |
ophon | -0.249 |
swat | -0.236 |
Token | Her |
Token ID | 9360 |
Feature activation | -6.03e-01 |
Pos logit contributions | |
DEFENSE | +1.355 |
Vert | +1.203 |
HQ | +1.001 |
Administrator | +0.996 |
Crest | +0.995 |
Neg logit contributions | |
isa | -1.376 |
awa | -1.341 |
favour | -1.333 |
ophon | -1.323 |
ethical | -1.244 |
Token | od |
Token ID | 375 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
DEFENSE | +2.442 |
Vert | +2.219 |
HQ | +1.886 |
Crest | +1.882 |
Administrator | +1.868 |
Neg logit contributions | |
awa | -1.791 |
isa | -1.772 |
favour | -1.732 |
ophon | -1.668 |
swat | -1.594 |
Token | ot |
Token ID | 313 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
isa | +0.608 |
ophon | +0.589 |
favour | +0.583 |
awa | +0.573 |
abet | +0.546 |
Neg logit contributions | |
DEFENSE | -0.644 |
Vert | -0.571 |
HQ | -0.480 |
Administrator | -0.477 |
Crest | -0.476 |
Token | us |
Token ID | 385 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
isa | +0.092 |
favour | +0.089 |
ophon | +0.089 |
awa | +0.088 |
abet | +0.083 |
Neg logit contributions | |
DEFENSE | -0.083 |
Vert | -0.073 |
HQ | -0.060 |
Administrator | -0.060 |
Crest | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
isa | +0.265 |
favour | +0.257 |
awa | +0.256 |
ophon | +0.255 |
swat | +0.238 |
Neg logit contributions | |
DEFENSE | -0.289 |
Vert | -0.259 |
HQ | -0.218 |
Administrator | -0.216 |
Crest | -0.215 |
Token | though |
Token ID | 996 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
isa | +0.371 |
favour | +0.362 |
awa | +0.360 |
ophon | +0.358 |
swat | +0.335 |
Neg logit contributions | |
DEFENSE | -0.401 |
Vert | -0.359 |
HQ | -0.303 |
Crest | -0.298 |
Administrator | -0.298 |
Token | , |
Token ID | 11 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
DEFENSE | +4.478 |
Vert | +3.985 |
Crest | +3.413 |
Administrator | +3.406 |
Ft | +3.390 |
Neg logit contributions | |
isa | -4.127 |
awa | -3.988 |
ophon | -3.919 |
favour | -3.839 |
ethical | -3.716 |
Token | got |
Token ID | 1392 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
isa | +0.370 |
favour | +0.363 |
awa | +0.358 |
ophon | +0.354 |
swat | +0.335 |
Neg logit contributions | |
DEFENSE | -0.414 |
Vert | -0.372 |
HQ | -0.315 |
Crest | -0.310 |
Administrator | -0.309 |
Token | at |
Token ID | 379 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
isa | +0.407 |
awa | +0.398 |
favour | +0.397 |
ophon | +0.390 |
swat | +0.369 |
Neg logit contributions | |
DEFENSE | -0.447 |
Vert | -0.401 |
HQ | -0.336 |
Crest | -0.333 |
Administrator | -0.332 |
Token | least |
Token ID | 1551 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
DEFENSE | +0.207 |
Vert | +0.185 |
HQ | +0.156 |
Crest | +0.154 |
Administrator | +0.154 |
Neg logit contributions | |
isa | -0.192 |
favour | -0.186 |
awa | -0.186 |
ophon | -0.184 |
swat | -0.173 |
Token | one |
Token ID | 530 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
isa | +0.833 |
favour | +0.813 |
awa | +0.811 |
ophon | +0.797 |
swat | +0.750 |
Neg logit contributions | |
DEFENSE | -1.001 |
Vert | -0.902 |
HQ | -0.768 |
Administrator | -0.760 |
Crest | -0.757 |
Token | � |
Token ID | 247 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
isa | +0.057 |
awa | +0.055 |
favour | +0.054 |
ophon | +0.054 |
ethical | +0.050 |
Neg logit contributions | |
DEFENSE | -0.082 |
Vert | -0.074 |
HQ | -0.064 |
Crest | -0.064 |
Administrator | -0.064 |
Token | s |
Token ID | 82 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
DEFENSE | +0.586 |
Vert | +0.527 |
HQ | +0.450 |
Administrator | +0.448 |
Crest | +0.448 |
Neg logit contributions | |
isa | -0.468 |
awa | -0.454 |
favour | -0.452 |
ophon | -0.448 |
ethical | -0.418 |
Token | why |
Token ID | 1521 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
DEFENSE | +0.305 |
Vert | +0.267 |
HQ | +0.215 |
Administrator | +0.214 |
Crest | +0.214 |
Neg logit contributions | |
isa | -0.386 |
awa | -0.377 |
favour | -0.376 |
ophon | -0.372 |
swat | -0.353 |
Token | D |
Token ID | 360 |
Feature activation | +8.41e-02 |
Pos logit contributions | |
DEFENSE | +0.118 |
Vert | +0.106 |
HQ | +0.091 |
Administrator | +0.090 |
Crest | +0.090 |
Neg logit contributions | |
isa | -0.093 |
awa | -0.090 |
favour | -0.090 |
ophon | -0.089 |
ethical | -0.083 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
isa | +0.253 |
awa | +0.245 |
favour | +0.244 |
ophon | +0.242 |
ethical | +0.224 |
Neg logit contributions | |
DEFENSE | -0.341 |
Vert | -0.308 |
HQ | -0.264 |
Administrator | -0.263 |
Crest | -0.263 |
Token | C |
Token ID | 34 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
DEFENSE | +0.682 |
Vert | +0.605 |
HQ | +0.504 |
Crest | +0.501 |
Administrator | +0.501 |
Neg logit contributions | |
isa | -0.688 |
awa | -0.668 |
favour | -0.665 |
ophon | -0.660 |
ethical | -0.620 |
Token | . |
Token ID | 13 |
Feature activation | -9.10e-02 |
Pos logit contributions | |
DEFENSE | +4.297 |
Vert | +3.777 |
HQ | +3.236 |
Crest | +3.202 |
Administrator | +3.175 |
Neg logit contributions | |
isa | -4.246 |
awa | -4.222 |
favour | -4.161 |
ophon | -4.085 |
ethical | -3.865 |
Token | Republicans |
Token ID | 4734 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
DEFENSE | +0.336 |
Vert | +0.300 |
HQ | +0.251 |
Administrator | +0.250 |
Crest | +0.250 |
Neg logit contributions | |
isa | -0.307 |
awa | -0.297 |
favour | -0.296 |
ophon | -0.295 |
ethical | -0.276 |
Token | are |
Token ID | 389 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
DEFENSE | +0.414 |
Vert | +0.372 |
HQ | +0.317 |
Administrator | +0.315 |
Crest | +0.315 |
Neg logit contributions | |
isa | -0.335 |
awa | -0.325 |
favour | -0.323 |
ophon | -0.320 |
ethical | -0.299 |
Token | scrambling |
Token ID | 36502 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
DEFENSE | +0.105 |
Vert | +0.094 |
HQ | +0.079 |
Crest | +0.078 |
Administrator | +0.078 |
Neg logit contributions | |
isa | -0.095 |
awa | -0.093 |
favour | -0.093 |
ophon | -0.091 |
swat | -0.086 |
Token | to |
Token ID | 284 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
isa | +0.268 |
awa | +0.260 |
favour | +0.259 |
ophon | +0.257 |
swat | +0.240 |
Neg logit contributions | |
DEFENSE | -0.312 |
Vert | -0.280 |
HQ | -0.236 |
Crest | -0.235 |
Administrator | -0.235 |
Token | your |
Token ID | 534 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
DEFENSE | +2.293 |
Vert | +2.035 |
HQ | +1.699 |
Administrator | +1.693 |
Crest | +1.691 |
Neg logit contributions | |
isa | -2.298 |
awa | -2.236 |
favour | -2.220 |
ophon | -2.213 |
ethical | -2.077 |
Token | first |
Token ID | 717 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
DEFENSE | +2.097 |
Vert | +1.866 |
HQ | +1.577 |
Administrator | +1.576 |
Crest | +1.575 |
Neg logit contributions | |
isa | -1.892 |
awa | -1.834 |
ophon | -1.824 |
favour | -1.811 |
abet | -1.703 |
Token | time |
Token ID | 640 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
isa | +0.090 |
favour | +0.087 |
awa | +0.087 |
ophon | +0.085 |
swat | +0.081 |
Neg logit contributions | |
DEFENSE | -0.102 |
Vert | -0.091 |
HQ | -0.077 |
Administrator | -0.076 |
Crest | -0.076 |
Token | back |
Token ID | 736 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
DEFENSE | +1.270 |
Vert | +1.132 |
HQ | +0.945 |
Crest | +0.938 |
Administrator | +0.938 |
Neg logit contributions | |
isa | -1.240 |
awa | -1.207 |
favour | -1.205 |
ophon | -1.187 |
swat | -1.121 |
Token | , |
Token ID | 11 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
DEFENSE | +0.437 |
Vert | +0.389 |
HQ | +0.325 |
Administrator | +0.323 |
Crest | +0.322 |
Neg logit contributions | |
isa | -0.423 |
awa | -0.411 |
favour | -0.411 |
ophon | -0.404 |
swat | -0.382 |
Token | especially |
Token ID | 2592 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
DEFENSE | +0.402 |
Vert | +0.357 |
HQ | +0.297 |
Crest | +0.294 |
7601 | +0.293 |
Neg logit contributions | |
isa | -0.402 |
favour | -0.392 |
awa | -0.390 |
ophon | -0.381 |
swat | -0.367 |
Token | if |
Token ID | 611 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
DEFENSE | +4.237 |
Vert | +3.731 |
Administrator | +3.131 |
Ft | +3.122 |
Crest | +3.112 |
Neg logit contributions | |
isa | -4.389 |
awa | -4.276 |
ophon | -4.275 |
favour | -4.198 |
ethical | -4.013 |
Token | you |
Token ID | 345 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
DEFENSE | +2.180 |
Vert | +1.929 |
HQ | +1.601 |
Administrator | +1.595 |
Crest | +1.594 |
Neg logit contributions | |
isa | -2.274 |
awa | -2.214 |
favour | -2.201 |
ophon | -2.189 |
ethical | -2.059 |
Token | 've |
Token ID | 1053 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
DEFENSE | +2.032 |
Vert | +1.833 |
HQ | +1.574 |
Administrator | +1.569 |
Crest | +1.569 |
Neg logit contributions | |
isa | -1.490 |
awa | -1.443 |
favour | -1.432 |
ophon | -1.424 |
ethical | -1.322 |
Token | been |
Token ID | 587 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
DEFENSE | +1.135 |
Vert | +1.020 |
HQ | +0.862 |
Administrator | +0.853 |
Crest | +0.852 |
Neg logit contributions | |
isa | -0.998 |
favour | -0.969 |
awa | -0.968 |
ophon | -0.952 |
swat | -0.897 |
Token | there |
Token ID | 612 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
DEFENSE | +0.404 |
Vert | +0.364 |
HQ | +0.306 |
Crest | +0.302 |
Administrator | +0.302 |
Neg logit contributions | |
isa | -0.360 |
favour | -0.351 |
awa | -0.350 |
ophon | -0.343 |
swat | -0.325 |
Token | founder |
Token ID | 9119 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
isa | +1.193 |
awa | +1.156 |
favour | +1.154 |
ophon | +1.140 |
swat | +1.067 |
Neg logit contributions | |
DEFENSE | -1.414 |
Vert | -1.274 |
HQ | -1.082 |
Crest | -1.069 |
Administrator | -1.065 |
Token | of |
Token ID | 286 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
isa | +0.193 |
awa | +0.188 |
favour | +0.186 |
ophon | +0.185 |
ethical | +0.172 |
Neg logit contributions | |
DEFENSE | -0.257 |
Vert | -0.232 |
Administrator | -0.199 |
HQ | -0.199 |
Crest | -0.198 |
Token | Cot |
Token ID | 42266 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
isa | +1.436 |
awa | +1.395 |
favour | +1.394 |
ophon | +1.367 |
ethical | +1.295 |
Neg logit contributions | |
DEFENSE | -1.552 |
Vert | -1.386 |
HQ | -1.172 |
Administrator | -1.152 |
7601 | -1.152 |
Token | é |
Token ID | 2634 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
isa | +0.125 |
awa | +0.120 |
favour | +0.119 |
ophon | +0.117 |
swat | +0.106 |
Neg logit contributions | |
DEFENSE | -0.268 |
Vert | -0.246 |
HQ | -0.217 |
Crest | -0.216 |
Administrator | -0.216 |
Token | Or |
Token ID | 1471 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
DEFENSE | +1.600 |
Vert | +1.430 |
HQ | +1.195 |
Crest | +1.190 |
Administrator | +1.187 |
Neg logit contributions | |
isa | -1.568 |
awa | -1.537 |
favour | -1.525 |
ophon | -1.509 |
swat | -1.426 |
Token | phan |
Token ID | 19080 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
isa | +0.822 |
awa | +0.795 |
favour | +0.791 |
ophon | +0.786 |
ethical | +0.734 |
Neg logit contributions | |
DEFENSE | -0.984 |
Vert | -0.882 |
HQ | -0.749 |
Administrator | -0.746 |
Crest | -0.745 |
Token | Consulting |
Token ID | 41005 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
DEFENSE | +4.759 |
Vert | +4.309 |
Administrator | +3.719 |
Crest | +3.694 |
HQ | +3.679 |
Neg logit contributions | |
awa | -3.759 |
isa | -3.655 |
favour | -3.627 |
ophon | -3.588 |
ethical | -3.338 |
Token | in |
Token ID | 287 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
DEFENSE | +1.416 |
Vert | +1.265 |
HQ | +1.069 |
Administrator | +1.065 |
Crest | +1.064 |
Neg logit contributions | |
isa | -1.259 |
awa | -1.224 |
favour | -1.216 |
ophon | -1.207 |
ethical | -1.130 |
Token | Silver |
Token ID | 7698 |
Feature activation | -6.69e-03 |
Pos logit contributions | |
isa | +0.147 |
favour | +0.144 |
awa | +0.142 |
ophon | +0.141 |
swat | +0.135 |
Neg logit contributions | |
DEFENSE | -0.109 |
Vert | -0.095 |
HQ | -0.076 |
Administrator | -0.074 |
7601 | -0.074 |
Token | Spring |
Token ID | 8225 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
DEFENSE | +0.023 |
Vert | +0.021 |
HQ | +0.017 |
Administrator | +0.017 |
Crest | +0.017 |
Neg logit contributions | |
isa | -0.024 |
awa | -0.023 |
favour | -0.023 |
ophon | -0.023 |
ethical | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
isa | +0.872 |
favour | +0.825 |
ophon | +0.824 |
awa | +0.815 |
swat | +0.766 |
Neg logit contributions | |
DEFENSE | -1.213 |
Vert | -1.106 |
7601 | -0.945 |
HQ | -0.935 |
Administrator | -0.931 |
Token | photo |
Token ID | 4590 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
DEFENSE | +0.221 |
Vert | +0.195 |
HQ | +0.162 |
Administrator | +0.162 |
Crest | +0.162 |
Neg logit contributions | |
isa | -0.230 |
awa | -0.224 |
favour | -0.223 |
ophon | -0.222 |
ethical | -0.209 |
Token | of |
Token ID | 286 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
DEFENSE | +0.686 |
Vert | +0.622 |
HQ | +0.537 |
Administrator | +0.535 |
Crest | +0.535 |
Neg logit contributions | |
isa | -0.464 |
awa | -0.448 |
favour | -0.445 |
ophon | -0.442 |
ethical | -0.408 |
Token | the |
Token ID | 262 |
Feature activation | -7.46e-02 |
Pos logit contributions | |
DEFENSE | +0.694 |
Vert | +0.621 |
HQ | +0.524 |
Crest | +0.520 |
Administrator | +0.520 |
Neg logit contributions | |
isa | -0.616 |
awa | -0.597 |
favour | -0.596 |
ophon | -0.590 |
swat | -0.553 |
Token | " |
Token ID | 366 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
DEFENSE | +0.311 |
Vert | +0.282 |
HQ | +0.243 |
Administrator | +0.242 |
Crest | +0.242 |
Neg logit contributions | |
isa | -0.216 |
awa | -0.209 |
favour | -0.208 |
ophon | -0.205 |
ethical | -0.190 |
Token | O |
Token ID | 46 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
isa | +0.766 |
favour | +0.730 |
awa | +0.730 |
ophon | +0.726 |
ethical | +0.691 |
Neg logit contributions | |
DEFENSE | -0.896 |
Vert | -0.791 |
Administrator | -0.677 |
Crest | -0.677 |
Ft | -0.669 |
Token | lymp |
Token ID | 6760 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
DEFENSE | +0.501 |
Vert | +0.452 |
HQ | +0.388 |
Administrator | +0.386 |
Crest | +0.386 |
Neg logit contributions | |
isa | -0.374 |
awa | -0.363 |
favour | -0.360 |
ophon | -0.357 |
ethical | -0.332 |
Token | ics |
Token ID | 873 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
DEFENSE | +4.275 |
Vert | +3.826 |
HQ | +3.304 |
Crest | +3.267 |
Administrator | +3.241 |
Neg logit contributions | |
awa | -3.931 |
isa | -3.919 |
favour | -3.872 |
ophon | -3.805 |
abet | -3.624 |
Token | Wi |
Token ID | 11759 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
DEFENSE | +0.126 |
Vert | +0.112 |
HQ | +0.095 |
Administrator | +0.094 |
Crest | +0.094 |
Neg logit contributions | |
isa | -0.113 |
awa | -0.110 |
favour | -0.109 |
ophon | -0.109 |
ethical | -0.102 |
Token | - |
Token ID | 12 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
isa | +0.755 |
awa | +0.731 |
favour | +0.726 |
ophon | +0.721 |
ethical | +0.671 |
Neg logit contributions | |
DEFENSE | -0.980 |
Vert | -0.883 |
HQ | -0.754 |
Administrator | -0.752 |
Crest | -0.752 |
Token | Fi |
Token ID | 10547 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
isa | +0.195 |
awa | +0.187 |
favour | +0.187 |
ophon | +0.185 |
ethical | +0.173 |
Neg logit contributions | |
DEFENSE | -0.257 |
Vert | -0.231 |
Administrator | -0.198 |
HQ | -0.198 |
Crest | -0.198 |
Token | police |
Token ID | 1644 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
DEFENSE | +0.700 |
Vert | +0.628 |
HQ | +0.535 |
Administrator | +0.533 |
Crest | +0.533 |
Neg logit contributions | |
isa | -0.575 |
awa | -0.558 |
favour | -0.554 |
ophon | -0.551 |
ethical | -0.514 |
Token | list |
Token ID | 4868 |
Feature activation | -9.09e-02 |
Pos logit contributions | |
DEFENSE | +2.904 |
Vert | +2.695 |
onel | +2.324 |
HQ | +2.310 |
7601 | +2.305 |
Neg logit contributions | |
awa | -1.977 |
isa | -1.880 |
favour | -1.837 |
ophon | -1.798 |
swat | -1.698 |
Token | _ |
Token ID | 62 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
DEFENSE | +0.345 |
Vert | +0.309 |
HQ | +0.262 |
Administrator | +0.261 |
Crest | +0.261 |
Neg logit contributions | |
isa | -0.297 |
awa | -0.287 |
favour | -0.286 |
ophon | -0.284 |
ethical | -0.266 |
Token | images |
Token ID | 17566 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
DEFENSE | +1.586 |
Vert | +1.389 |
HQ | +1.107 |
Crest | +1.087 |
Administrator | +1.087 |
Neg logit contributions | |
isa | -2.204 |
awa | -2.188 |
favour | -2.161 |
ophon | -2.146 |
swat | -2.033 |
Token | ( |
Token ID | 7 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
isa | +0.923 |
favour | +0.884 |
ophon | +0.870 |
awa | +0.850 |
abet | +0.812 |
Neg logit contributions | |
DEFENSE | -1.187 |
Vert | -1.059 |
PDATE | -0.907 |
HQ | -0.907 |
Administrator | -0.907 |
Token | args |
Token ID | 22046 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
DEFENSE | +2.174 |
Vert | +1.978 |
HQ | +1.727 |
Crest | +1.714 |
Administrator | +1.711 |
Neg logit contributions | |
isa | -1.288 |
awa | -1.276 |
favour | -1.243 |
ophon | -1.232 |
swat | -1.131 |
Token | [" |
Token ID | 14692 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
isa | +1.342 |
favour | +1.279 |
ophon | +1.275 |
awa | +1.250 |
abet | +1.189 |
Neg logit contributions | |
DEFENSE | -1.576 |
Vert | -1.406 |
Administrator | -1.192 |
Crest | -1.185 |
Ft | -1.185 |
Token | images |
Token ID | 17566 |
Feature activation | +3.53e-03 |
Pos logit contributions | |
DEFENSE | +3.120 |
Vert | +2.832 |
HQ | +2.194 |
7601 | +2.140 |
ows | +2.140 |
Neg logit contributions | |
awa | -5.285 |
favour | -5.045 |
isa | -5.030 |
ophon | -5.007 |
swat | -4.791 |
Token | "] |
Token ID | 8973 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
isa | +0.014 |
favour | +0.013 |
ophon | +0.013 |
awa | +0.013 |
abet | +0.012 |
Neg logit contributions | |
DEFENSE | -0.011 |
Vert | -0.010 |
Administrator | -0.008 |
HQ | -0.008 |
Crest | -0.008 |
Token | ): |
Token ID | 2599 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
DEFENSE | +0.876 |
Vert | +0.799 |
HQ | +0.696 |
Administrator | +0.693 |
Crest | +0.692 |
Neg logit contributions | |
isa | -0.528 |
awa | -0.516 |
favour | -0.507 |
ophon | -0.502 |
ethical | -0.463 |
Token | # |
Token ID | 1303 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
isa | +0.656 |
favour | +0.643 |
awa | +0.629 |
ophon | +0.625 |
swat | +0.601 |
Neg logit contributions | |
DEFENSE | -0.540 |
Vert | -0.473 |
HQ | -0.388 |
Administrator | -0.382 |
Crest | -0.380 |
Token | load |
Token ID | 3440 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
DEFENSE | +1.511 |
Vert | +1.361 |
HQ | +1.171 |
Administrator | +1.166 |
Crest | +1.163 |
Neg logit contributions | |
isa | -1.158 |
awa | -1.132 |
ophon | -1.115 |
favour | -1.109 |
ethical | -1.034 |
Token | . |
Token ID | 13 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
isa | +1.561 |
favour | +1.507 |
awa | +1.494 |
ophon | +1.482 |
swat | +1.381 |
Neg logit contributions | |
DEFENSE | -1.944 |
Vert | -1.756 |
Crest | -1.488 |
HQ | -1.486 |
Administrator | -1.476 |
Token | The |
Token ID | 383 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
isa | +0.585 |
favour | +0.558 |
ophon | +0.556 |
awa | +0.556 |
swat | +0.518 |
Neg logit contributions | |
DEFENSE | -0.738 |
Vert | -0.666 |
HQ | -0.570 |
onel | -0.566 |
PDATE | -0.558 |
Token | poll |
Token ID | 3278 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
DEFENSE | +0.227 |
Vert | +0.204 |
HQ | +0.170 |
Crest | +0.168 |
Administrator | +0.167 |
Neg logit contributions | |
isa | -0.220 |
favour | -0.215 |
awa | -0.213 |
ophon | -0.211 |
swat | -0.200 |
Token | below |
Token ID | 2174 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
DEFENSE | +2.161 |
Vert | +1.917 |
HQ | +1.598 |
Crest | +1.589 |
Administrator | +1.588 |
Neg logit contributions | |
isa | -2.196 |
awa | -2.144 |
favour | -2.120 |
ophon | -2.114 |
ethical | -1.988 |
Token | , |
Token ID | 11 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
isa | +0.440 |
favour | +0.431 |
awa | +0.428 |
ophon | +0.424 |
swat | +0.404 |
Neg logit contributions | |
DEFENSE | -0.347 |
Vert | -0.303 |
HQ | -0.245 |
Administrator | -0.243 |
Crest | -0.242 |
Token | commissioned |
Token ID | 20462 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
DEFENSE | +0.641 |
Vert | +0.569 |
HQ | +0.473 |
Administrator | +0.470 |
Crest | +0.469 |
Neg logit contributions | |
isa | -0.659 |
favour | -0.642 |
awa | -0.642 |
ophon | -0.634 |
swat | -0.597 |
Token | by |
Token ID | 416 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
DEFENSE | +4.149 |
Vert | +3.648 |
Crest | +3.086 |
Administrator | +3.084 |
HQ | +3.076 |
Neg logit contributions | |
isa | -4.226 |
awa | -4.141 |
ophon | -4.067 |
favour | -3.999 |
abet | -3.851 |
Token | Bloomberg |
Token ID | 15689 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
isa | +0.187 |
favour | +0.183 |
awa | +0.182 |
ophon | +0.180 |
swat | +0.169 |
Neg logit contributions | |
DEFENSE | -0.194 |
Vert | -0.173 |
HQ | -0.145 |
Administrator | -0.143 |
7601 | -0.143 |
Token | Politics |
Token ID | 17554 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
DEFENSE | +1.838 |
Vert | +1.638 |
HQ | +1.381 |
Administrator | +1.374 |
Crest | +1.372 |
Neg logit contributions | |
isa | -1.709 |
awa | -1.663 |
favour | -1.650 |
ophon | -1.641 |
ethical | -1.538 |
Token | and |
Token ID | 290 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
isa | +0.237 |
favour | +0.230 |
awa | +0.227 |
ophon | +0.227 |
ethical | +0.211 |
Neg logit contributions | |
DEFENSE | -0.289 |
Vert | -0.260 |
HQ | -0.220 |
Crest | -0.219 |
Administrator | -0.219 |
Token | Saint |
Token ID | 9281 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
DEFENSE | +0.088 |
Vert | +0.078 |
HQ | +0.064 |
Administrator | +0.064 |
Crest | +0.064 |
Neg logit contributions | |
isa | -0.093 |
favour | -0.090 |
awa | -0.090 |
ophon | -0.089 |
swat | -0.084 |
Token | Sh |
Token ID | 911 |
Feature activation | +3.96e-04 |
Pos logit contributions | |
isa | +0.146 |
awa | +0.142 |
favour | +0.142 |
ophon | +0.140 |
swat | +0.133 |
Neg logit contributions | |
DEFENSE | -0.151 |
Vert | -0.134 |
HQ | -0.113 |
Administrator | -0.111 |
Crest | -0.111 |
Token | ag |
Token ID | 363 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
isa | +0.001 |
awa | +0.001 |
favour | +0.001 |
ophon | +0.001 |
ethical | +0.001 |
Neg logit contributions | |
DEFENSE | -0.002 |
Vert | -0.002 |
HQ | -0.002 |
Administrator | -0.002 |
Crest | -0.002 |
Token | gy |
Token ID | 1360 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
isa | +1.066 |
favour | +0.993 |
ophon | +0.988 |
awa | +0.978 |
swat | +0.898 |
Neg logit contributions | |
DEFENSE | -2.139 |
Vert | -1.958 |
Crest | -1.711 |
Administrator | -1.710 |
HQ | -1.702 |
Token | just |
Token ID | 655 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
DEFENSE | +0.144 |
Vert | +0.128 |
HQ | +0.108 |
Administrator | +0.107 |
Crest | +0.107 |
Neg logit contributions | |
isa | -0.132 |
awa | -0.128 |
favour | -0.127 |
ophon | -0.126 |
swat | -0.119 |
Token | stand |
Token ID | 1302 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
DEFENSE | +1.444 |
Vert | +1.283 |
HQ | +1.077 |
Crest | +1.072 |
Administrator | +1.072 |
Neg logit contributions | |
isa | -1.386 |
awa | -1.348 |
favour | -1.339 |
ophon | -1.333 |
ethical | -1.252 |
Token | back |
Token ID | 736 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
DEFENSE | +1.255 |
Vert | +1.120 |
HQ | +0.940 |
Administrator | +0.935 |
Crest | +0.933 |
Neg logit contributions | |
isa | -1.182 |
awa | -1.147 |
favour | -1.140 |
ophon | -1.132 |
ethical | -1.061 |
Token | and |
Token ID | 290 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
DEFENSE | +3.912 |
Vert | +3.395 |
Crest | +2.875 |
Administrator | +2.859 |
HQ | +2.842 |
Neg logit contributions | |
isa | -4.350 |
favour | -4.185 |
ophon | -4.162 |
awa | -4.149 |
ethical | -3.933 |
Token | watch |
Token ID | 2342 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
DEFENSE | +0.974 |
Vert | +0.883 |
HQ | +0.759 |
Crest | +0.755 |
Administrator | +0.755 |
Neg logit contributions | |
isa | -0.690 |
awa | -0.668 |
favour | -0.666 |
ophon | -0.658 |
swat | -0.611 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
DEFENSE | +0.562 |
Vert | +0.503 |
HQ | +0.423 |
Administrator | +0.418 |
Crest | +0.417 |
Neg logit contributions | |
isa | -0.521 |
awa | -0.507 |
favour | -0.505 |
ophon | -0.500 |
swat | -0.472 |
Token | � |
Token ID | 564 |
Feature activation | -6.57e-03 |
Pos logit contributions | |
isa | +0.373 |
awa | +0.358 |
favour | +0.357 |
ophon | +0.354 |
swat | +0.334 |
Neg logit contributions | |
DEFENSE | -0.447 |
Vert | -0.402 |
HQ | -0.342 |
Crest | -0.336 |
Administrator | -0.336 |
Token | � |
Token ID | 250 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
DEFENSE | +0.028 |
Vert | +0.026 |
HQ | +0.022 |
Crest | +0.022 |
Administrator | +0.022 |
Neg logit contributions | |
isa | -0.018 |
awa | -0.017 |
favour | -0.017 |
ophon | -0.017 |
abet | -0.016 |
Token | Ireland |
Token ID | 49752 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
DEFENSE | +0.197 |
Vert | +0.174 |
Crest | +0.147 |
Administrator | +0.147 |
HQ | +0.146 |
Neg logit contributions | |
isa | -0.187 |
awa | -0.179 |
favour | -0.179 |
ophon | -0.178 |
ethical | -0.168 |
Token | will |
Token ID | 481 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
DEFENSE | +0.164 |
Vert | +0.147 |
HQ | +0.124 |
Administrator | +0.123 |
Crest | +0.123 |
Neg logit contributions | |
isa | -0.149 |
favour | -0.145 |
awa | -0.144 |
ophon | -0.142 |
swat | -0.134 |
Token | travel |
Token ID | 3067 |
Feature activation | +7.20e-03 |
Pos logit contributions | |
DEFENSE | +1.503 |
Vert | +1.347 |
Crest | +1.161 |
Administrator | +1.161 |
Ft | +1.160 |
Neg logit contributions | |
isa | -1.095 |
awa | -1.061 |
ophon | -1.052 |
favour | -1.035 |
ethical | -0.980 |
Token | to |
Token ID | 284 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
isa | +0.023 |
favour | +0.023 |
awa | +0.023 |
ophon | +0.022 |
swat | +0.021 |
Neg logit contributions | |
DEFENSE | -0.028 |
Vert | -0.025 |
HQ | -0.021 |
Administrator | -0.021 |
7601 | -0.021 |
Token | Armenia |
Token ID | 32369 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
isa | +0.367 |
favour | +0.358 |
awa | +0.354 |
ophon | +0.350 |
swat | +0.332 |
Neg logit contributions | |
DEFENSE | -0.335 |
Vert | -0.296 |
HQ | -0.244 |
7601 | -0.241 |
Administrator | -0.240 |
Token | next |
Token ID | 1306 |
Feature activation | +8.68e-01 |
Pos logit contributions | |
isa | +0.462 |
favour | +0.447 |
awa | +0.442 |
ophon | +0.436 |
swat | +0.411 |
Neg logit contributions | |
DEFENSE | -0.537 |
Vert | -0.482 |
HQ | -0.404 |
Administrator | -0.403 |
7601 | -0.402 |
Token | month |
Token ID | 1227 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
favour | +2.938 |
isa | +2.934 |
awa | +2.814 |
ophon | +2.749 |
swat | +2.680 |
Neg logit contributions | |
DEFENSE | -3.157 |
Vert | -2.832 |
HQ | -2.361 |
7601 | -2.352 |
PDATE | -2.335 |
Token | where |
Token ID | 810 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
DEFENSE | +1.258 |
Vert | +1.124 |
HQ | +0.945 |
Administrator | +0.941 |
Crest | +0.940 |
Neg logit contributions | |
isa | -1.151 |
awa | -1.117 |
favour | -1.113 |
ophon | -1.103 |
ethical | -1.033 |
Token | they |
Token ID | 484 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
isa | +0.867 |
awa | +0.842 |
favour | +0.842 |
ophon | +0.830 |
ethical | +0.783 |
Neg logit contributions | |
DEFENSE | -0.811 |
Vert | -0.716 |
HQ | -0.594 |
Administrator | -0.588 |
Crest | -0.587 |
Token | will |
Token ID | 481 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
isa | +0.682 |
favour | +0.660 |
awa | +0.658 |
ophon | +0.647 |
swat | +0.604 |
Neg logit contributions | |
DEFENSE | -0.930 |
Vert | -0.845 |
HQ | -0.725 |
Administrator | -0.719 |
Crest | -0.719 |
Token | meet |
Token ID | 1826 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
DEFENSE | +1.200 |
Vert | +1.085 |
Crest | +0.945 |
Administrator | +0.944 |
Ft | +0.942 |
Neg logit contributions | |
isa | -0.740 |
awa | -0.715 |
ophon | -0.706 |
favour | -0.700 |
ethical | -0.652 |
Token | ker |
Token ID | 41927 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
isa | +0.161 |
awa | +0.156 |
favour | +0.156 |
ophon | +0.154 |
swat | +0.144 |
Neg logit contributions | |
DEFENSE | -0.205 |
Vert | -0.185 |
HQ | -0.158 |
Administrator | -0.157 |
Crest | -0.157 |
Token | at |
Token ID | 265 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
DEFENSE | +0.045 |
Vert | +0.042 |
Crest | +0.036 |
HQ | +0.036 |
Administrator | +0.036 |
Neg logit contributions | |
isa | -0.028 |
awa | -0.027 |
favour | -0.027 |
ophon | -0.026 |
swat | -0.024 |
Token | otic |
Token ID | 6210 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
isa | +0.853 |
awa | +0.820 |
ophon | +0.819 |
favour | +0.818 |
ethical | +0.774 |
Neg logit contributions | |
DEFENSE | -0.775 |
Vert | -0.678 |
HQ | -0.564 |
Administrator | -0.559 |
Crest | -0.559 |
Token | plugs |
Token ID | 37008 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
DEFENSE | +1.759 |
Vert | +1.583 |
Crest | +1.341 |
HQ | +1.338 |
Administrator | +1.337 |
Neg logit contributions | |
isa | -1.479 |
awa | -1.439 |
favour | -1.425 |
ophon | -1.422 |
ethical | -1.330 |
Token | , |
Token ID | 11 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
isa | +0.091 |
awa | +0.088 |
favour | +0.088 |
ophon | +0.087 |
ethical | +0.081 |
Neg logit contributions | |
DEFENSE | -0.104 |
Vert | -0.093 |
HQ | -0.079 |
Administrator | -0.078 |
Crest | -0.078 |
Token | mim |
Token ID | 17007 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
DEFENSE | +0.165 |
Vert | +0.146 |
HQ | +0.124 |
Administrator | +0.123 |
Crest | +0.122 |
Neg logit contributions | |
isa | -0.151 |
awa | -0.147 |
favour | -0.147 |
ophon | -0.145 |
swat | -0.136 |
Token | icking |
Token ID | 7958 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
isa | +2.251 |
ophon | +2.156 |
favour | +2.042 |
awa | +2.039 |
ethical | +1.904 |
Neg logit contributions | |
DEFENSE | -4.311 |
Vert | -3.921 |
HQ | -3.480 |
PDATE | -3.446 |
Administrator | -3.436 |
Token | the |
Token ID | 262 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
isa | +0.345 |
favour | +0.336 |
awa | +0.336 |
ophon | +0.331 |
swat | +0.312 |
Neg logit contributions | |
DEFENSE | -0.390 |
Vert | -0.348 |
HQ | -0.296 |
Administrator | -0.291 |
Crest | -0.290 |
Token | sp |
Token ID | 599 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
DEFENSE | +0.177 |
Vert | +0.158 |
HQ | +0.134 |
Administrator | +0.134 |
Crest | +0.133 |
Neg logit contributions | |
isa | -0.151 |
awa | -0.146 |
favour | -0.146 |
ophon | -0.145 |
swat | -0.135 |
Token | ines |
Token ID | 1127 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
isa | +0.091 |
awa | +0.088 |
favour | +0.087 |
ophon | +0.087 |
ethical | +0.081 |
Neg logit contributions | |
DEFENSE | -0.115 |
Vert | -0.104 |
HQ | -0.089 |
Crest | -0.088 |
Administrator | -0.088 |
Token | on |
Token ID | 319 |
Feature activation | +6.44e-02 |
Pos logit contributions | |
isa | +0.156 |
awa | +0.151 |
favour | +0.150 |
ophon | +0.149 |
ethical | +0.138 |
Neg logit contributions | |
DEFENSE | -0.210 |
Vert | -0.190 |
HQ | -0.163 |
Administrator | -0.162 |
Crest | -0.162 |
Token | s |
Token ID | 82 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
DEFENSE | +0.821 |
Vert | +0.738 |
HQ | +0.629 |
Administrator | +0.627 |
Crest | +0.626 |
Neg logit contributions | |
isa | -0.659 |
awa | -0.639 |
favour | -0.636 |
ophon | -0.631 |
ethical | -0.588 |
Token | )} |
Token ID | 38165 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
isa | +0.360 |
favour | +0.346 |
awa | +0.346 |
ophon | +0.344 |
ethical | +0.320 |
Neg logit contributions | |
DEFENSE | -0.456 |
Vert | -0.410 |
HQ | -0.350 |
Administrator | -0.349 |
Crest | -0.349 |
Token | 57 |
Token ID | 7632 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
isa | +0.493 |
awa | +0.479 |
favour | +0.476 |
ophon | +0.473 |
ethical | +0.444 |
Neg logit contributions | |
DEFENSE | -0.524 |
Vert | -0.467 |
HQ | -0.392 |
Crest | -0.391 |
Administrator | -0.391 |
Token | . |
Token ID | 13 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
DEFENSE | +1.509 |
Vert | +1.353 |
HQ | +1.142 |
Crest | +1.137 |
Administrator | +1.134 |
Neg logit contributions | |
isa | -1.375 |
awa | -1.370 |
favour | -1.334 |
ophon | -1.331 |
ethical | -1.250 |
Token | f |
Token ID | 277 |
Feature activation | -6.78e-02 |
Pos logit contributions | |
isa | +0.545 |
favour | +0.529 |
awa | +0.528 |
ophon | +0.520 |
swat | +0.486 |
Neg logit contributions | |
DEFENSE | -0.665 |
Vert | -0.597 |
HQ | -0.507 |
Administrator | -0.505 |
Crest | -0.503 |
Token | 6 |
Token ID | 21 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
DEFENSE | +0.234 |
Vert | +0.208 |
HQ | +0.172 |
Crest | +0.171 |
Administrator | +0.171 |
Neg logit contributions | |
isa | -0.244 |
awa | -0.239 |
favour | -0.237 |
ophon | -0.235 |
ethical | -0.222 |
Token | { |
Token ID | 1391 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
isa | +2.386 |
favour | +2.298 |
awa | +2.275 |
ophon | +2.246 |
swat | +2.074 |
Neg logit contributions | |
DEFENSE | -3.845 |
Vert | -3.500 |
HQ | -3.028 |
Administrator | -3.025 |
Crest | -3.013 |
Token | ( |
Token ID | 7 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
DEFENSE | +0.391 |
Vert | +0.350 |
HQ | +0.297 |
Administrator | +0.296 |
Crest | +0.296 |
Neg logit contributions | |
isa | -0.326 |
awa | -0.315 |
favour | -0.314 |
ophon | -0.312 |
ethical | -0.291 |
Token | 3 |
Token ID | 18 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
DEFENSE | +1.192 |
Vert | +1.042 |
HQ | +0.844 |
Administrator | +0.840 |
Crest | +0.840 |
Neg logit contributions | |
isa | -1.488 |
awa | -1.451 |
favour | -1.444 |
ophon | -1.436 |
ethical | -1.358 |
Token | s |
Token ID | 82 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
DEFENSE | +0.423 |
Vert | +0.379 |
HQ | +0.323 |
Administrator | +0.321 |
Crest | +0.320 |
Neg logit contributions | |
isa | -0.357 |
awa | -0.347 |
favour | -0.344 |
ophon | -0.342 |
ethical | -0.320 |
Token | )} |
Token ID | 38165 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
isa | +0.543 |
favour | +0.521 |
awa | +0.518 |
ophon | +0.518 |
ethical | +0.480 |
Neg logit contributions | |
DEFENSE | -0.698 |
Vert | -0.628 |
HQ | -0.537 |
Administrator | -0.535 |
Crest | -0.535 |
Token | < |
Token ID | 1279 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
isa | +1.872 |
awa | +1.808 |
favour | +1.801 |
ophon | +1.764 |
ethical | +1.644 |
Neg logit contributions | |
DEFENSE | -2.788 |
Vert | -2.473 |
Crest | -2.156 |
Administrator | -2.153 |
7601 | -2.151 |
Token | T |
Token ID | 309 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
DEFENSE | +1.095 |
Vert | +0.982 |
Crest | +0.834 |
HQ | +0.834 |
Administrator | +0.832 |
Neg logit contributions | |
isa | -0.914 |
awa | -0.886 |
favour | -0.881 |
ophon | -0.877 |
ethical | -0.819 |
Token | > |
Token ID | 1875 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
isa | +0.704 |
awa | +0.673 |
favour | +0.669 |
ophon | +0.657 |
ethical | +0.595 |
Neg logit contributions | |
DEFENSE | -1.482 |
Vert | -1.357 |
HQ | -1.198 |
Crest | -1.193 |
Administrator | -1.193 |
Token | { |
Token ID | 1391 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
isa | +1.605 |
favour | +1.563 |
awa | +1.556 |
ophon | +1.529 |
ethical | +1.419 |
Neg logit contributions | |
DEFENSE | -2.016 |
Vert | -1.807 |
HQ | -1.548 |
Administrator | -1.538 |
7601 | -1.537 |
Token | pub |
Token ID | 2240 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
isa | +0.817 |
awa | +0.799 |
favour | +0.798 |
ophon | +0.778 |
swat | +0.727 |
Neg logit contributions | |
DEFENSE | -0.996 |
Vert | -0.896 |
HQ | -0.763 |
7601 | -0.754 |
Administrator | -0.754 |
Token | fn |
Token ID | 24714 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
isa | +0.637 |
awa | +0.616 |
favour | +0.614 |
ophon | +0.601 |
swat | +0.553 |
Neg logit contributions | |
DEFENSE | -1.003 |
Vert | -0.905 |
HQ | -0.787 |
Administrator | -0.783 |
7601 | -0.779 |
Token | new |
Token ID | 649 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
isa | +3.164 |
awa | +3.133 |
favour | +3.123 |
ophon | +2.962 |
swat | +2.851 |
Neg logit contributions | |
DEFENSE | -3.732 |
Vert | -3.334 |
HQ | -2.867 |
7601 | -2.826 |
Ground | -2.799 |
Token | ( |
Token ID | 357 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
isa | +0.087 |
awa | +0.085 |
favour | +0.084 |
ophon | +0.083 |
ethical | +0.078 |
Neg logit contributions | |
DEFENSE | -0.109 |
Vert | -0.098 |
HQ | -0.083 |
Crest | -0.083 |
Administrator | -0.083 |
Token | t |
Token ID | 256 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
isa | +1.526 |
awa | +1.476 |
favour | +1.471 |
ophon | +1.451 |
swat | +1.355 |
Neg logit contributions | |
DEFENSE | -1.819 |
Vert | -1.621 |
HQ | -1.377 |
Administrator | -1.369 |
Crest | -1.368 |
Token | : |
Token ID | 1058 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
isa | +0.485 |
awa | +0.464 |
favour | +0.461 |
ophon | +0.456 |
ethical | +0.412 |
Neg logit contributions | |
DEFENSE | -0.991 |
Vert | -0.907 |
HQ | -0.799 |
Administrator | -0.796 |
Crest | -0.796 |
Token | T |
Token ID | 309 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
isa | +1.348 |
awa | +1.318 |
favour | +1.310 |
ophon | +1.276 |
swat | +1.204 |
Neg logit contributions | |
DEFENSE | -1.721 |
Vert | -1.548 |
HQ | -1.325 |
7601 | -1.311 |
Administrator | -1.305 |
Token | ( |
Token ID | 7 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
DEFENSE | +0.879 |
Vert | +0.789 |
HQ | +0.669 |
Administrator | +0.666 |
Crest | +0.665 |
Neg logit contributions | |
isa | -0.747 |
awa | -0.727 |
favour | -0.721 |
ophon | -0.716 |
ethical | -0.669 |
Token | 0 |
Token ID | 15 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
DEFENSE | +1.764 |
Vert | +1.550 |
HQ | +1.263 |
Administrator | +1.247 |
Crest | +1.245 |
Neg logit contributions | |
isa | -2.168 |
awa | -2.126 |
favour | -2.108 |
ophon | -2.099 |
abet | -1.979 |
Token | s |
Token ID | 82 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
DEFENSE | +1.207 |
Vert | +1.086 |
HQ | +0.926 |
Administrator | +0.921 |
Crest | +0.920 |
Neg logit contributions | |
isa | -0.979 |
awa | -0.957 |
favour | -0.946 |
ophon | -0.940 |
ethical | -0.876 |
Token | )} |
Token ID | 38165 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
DEFENSE | +0.891 |
Vert | +0.801 |
HQ | +0.680 |
Administrator | +0.676 |
Crest | +0.676 |
Neg logit contributions | |
isa | -0.742 |
awa | -0.733 |
favour | -0.719 |
ophon | -0.713 |
ethical | -0.667 |
Token | c |
Token ID | 269 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
isa | +1.308 |
favour | +1.272 |
awa | +1.263 |
ophon | +1.240 |
swat | +1.168 |
Neg logit contributions | |
DEFENSE | -1.563 |
Vert | -1.406 |
HQ | -1.191 |
Administrator | -1.184 |
7601 | -1.178 |
Token | 5 |
Token ID | 20 |
Feature activation | +8.25e-01 |
Pos logit contributions | |
DEFENSE | +0.231 |
Vert | +0.204 |
HQ | +0.168 |
Administrator | +0.168 |
Crest | +0.167 |
Neg logit contributions | |
isa | -0.252 |
awa | -0.244 |
favour | -0.243 |
ophon | -0.242 |
ethical | -0.228 |
Token | { |
Token ID | 1391 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
isa | +2.250 |
favour | +2.142 |
awa | +2.132 |
ophon | +2.087 |
swat | +1.946 |
Neg logit contributions | |
DEFENSE | -3.535 |
Vert | -3.246 |
HQ | -2.808 |
Administrator | -2.790 |
7601 | -2.789 |
Token | ( |
Token ID | 7 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
DEFENSE | +0.167 |
Vert | +0.150 |
Administrator | +0.127 |
Crest | +0.127 |
HQ | +0.127 |
Neg logit contributions | |
isa | -0.138 |
favour | -0.132 |
awa | -0.132 |
ophon | -0.131 |
ethical | -0.123 |
Token | 60 |
Token ID | 1899 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
DEFENSE | +1.613 |
Vert | +1.415 |
HQ | +1.152 |
Administrator | +1.139 |
Crest | +1.137 |
Neg logit contributions | |
isa | -1.989 |
awa | -1.950 |
favour | -1.935 |
ophon | -1.927 |
ethical | -1.817 |
Token | s |
Token ID | 82 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
DEFENSE | +0.327 |
Vert | +0.294 |
HQ | +0.250 |
Administrator | +0.250 |
Crest | +0.249 |
Neg logit contributions | |
isa | -0.262 |
awa | -0.253 |
favour | -0.252 |
ophon | -0.250 |
ethical | -0.233 |
Token | )} |
Token ID | 38165 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
isa | +0.137 |
favour | +0.131 |
awa | +0.131 |
ophon | +0.130 |
ethical | +0.121 |
Neg logit contributions | |
DEFENSE | -0.173 |
Vert | -0.155 |
HQ | -0.133 |
Administrator | -0.132 |
Crest | -0.132 |
Token | have |
Token ID | 423 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
isa | +0.509 |
favour | +0.500 |
awa | +0.493 |
ophon | +0.483 |
swat | +0.460 |
Neg logit contributions | |
DEFENSE | -0.662 |
Vert | -0.604 |
HQ | -0.515 |
Administrator | -0.506 |
Crest | -0.506 |
Token | a |
Token ID | 257 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
isa | +1.287 |
favour | +1.250 |
awa | +1.243 |
ophon | +1.220 |
swat | +1.155 |
Neg logit contributions | |
DEFENSE | -1.542 |
Vert | -1.390 |
HQ | -1.178 |
7601 | -1.157 |
onel | -1.154 |
Token | backup |
Token ID | 11559 |
Feature activation | +4.98e-04 |
Pos logit contributions | |
isa | +1.439 |
favour | +1.406 |
awa | +1.397 |
ophon | +1.368 |
swat | +1.300 |
Neg logit contributions | |
DEFENSE | -1.566 |
Vert | -1.411 |
HQ | -1.175 |
Administrator | -1.165 |
7601 | -1.162 |
Token | battery |
Token ID | 6555 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
isa | +0.001 |
awa | +0.001 |
favour | +0.001 |
ophon | +0.001 |
ethical | +0.001 |
Neg logit contributions | |
DEFENSE | -0.002 |
Vert | -0.002 |
HQ | -0.002 |
Administrator | -0.002 |
Crest | -0.002 |
Token | equip |
Token ID | 25725 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
isa | +1.006 |
favour | +0.976 |
awa | +0.972 |
ophon | +0.952 |
swat | +0.900 |
Neg logit contributions | |
DEFENSE | -1.186 |
Vert | -1.071 |
HQ | -0.898 |
Administrator | -0.892 |
Crest | -0.890 |
Token | ed |
Token ID | 276 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
DEFENSE | +0.567 |
Vert | +0.506 |
HQ | +0.426 |
Administrator | +0.425 |
Crest | +0.425 |
Neg logit contributions | |
isa | -0.522 |
awa | -0.508 |
favour | -0.504 |
ophon | -0.501 |
ethical | -0.470 |
Token | dock |
Token ID | 23423 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
isa | +0.856 |
favour | +0.830 |
awa | +0.829 |
ophon | +0.815 |
swat | +0.766 |
Neg logit contributions | |
DEFENSE | -1.000 |
Vert | -0.901 |
HQ | -0.760 |
Administrator | -0.755 |
Crest | -0.752 |
Token | , |
Token ID | 11 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
isa | +1.008 |
favour | +0.975 |
awa | +0.973 |
ophon | +0.958 |
swat | +0.896 |
Neg logit contributions | |
DEFENSE | -1.167 |
Vert | -1.047 |
HQ | -0.880 |
Crest | -0.877 |
Administrator | -0.877 |
Token | have |
Token ID | 423 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
isa | +0.347 |
favour | +0.340 |
awa | +0.337 |
ophon | +0.329 |
swat | +0.315 |
Neg logit contributions | |
DEFENSE | -0.395 |
Vert | -0.357 |
HQ | -0.301 |
Crest | -0.296 |
7601 | -0.296 |
Token | it |
Token ID | 340 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
isa | +0.643 |
awa | +0.626 |
favour | +0.626 |
ophon | +0.612 |
swat | +0.578 |
Neg logit contributions | |
DEFENSE | -0.725 |
Vert | -0.654 |
HQ | -0.551 |
Crest | -0.541 |
7601 | -0.541 |
Token | with |
Token ID | 351 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
isa | +0.397 |
awa | +0.388 |
favour | +0.387 |
ophon | +0.377 |
swat | +0.359 |
Neg logit contributions | |
DEFENSE | -0.459 |
Vert | -0.418 |
HQ | -0.352 |
Administrator | -0.345 |
Crest | -0.345 |
Token | pixels |
Token ID | 17848 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
DEFENSE | +0.084 |
Vert | +0.075 |
HQ | +0.063 |
Crest | +0.063 |
Administrator | +0.063 |
Neg logit contributions | |
isa | -0.079 |
awa | -0.077 |
favour | -0.076 |
ophon | -0.076 |
swat | -0.071 |
Token | but |
Token ID | 475 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
DEFENSE | +0.429 |
Vert | +0.384 |
HQ | +0.324 |
Crest | +0.323 |
Administrator | +0.323 |
Neg logit contributions | |
isa | -0.379 |
awa | -0.369 |
favour | -0.367 |
ophon | -0.363 |
ethical | -0.340 |
Token | you |
Token ID | 345 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
isa | +0.577 |
favour | +0.566 |
awa | +0.561 |
ophon | +0.545 |
swat | +0.523 |
Neg logit contributions | |
DEFENSE | -0.628 |
Vert | -0.566 |
HQ | -0.474 |
Crest | -0.473 |
Administrator | -0.469 |
Token | will |
Token ID | 481 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
isa | +0.080 |
favour | +0.078 |
awa | +0.078 |
ophon | +0.076 |
swat | +0.072 |
Neg logit contributions | |
DEFENSE | -0.097 |
Vert | -0.088 |
HQ | -0.075 |
Administrator | -0.074 |
Crest | -0.074 |
Token | not |
Token ID | 407 |
Feature activation | +8.35e-02 |
Pos logit contributions | |
DEFENSE | +1.154 |
Vert | +1.034 |
HQ | +0.873 |
Administrator | +0.865 |
Crest | +0.865 |
Neg logit contributions | |
isa | -1.052 |
awa | -1.024 |
favour | -1.023 |
ophon | -1.004 |
swat | -0.948 |
Token | be |
Token ID | 307 |
Feature activation | +7.12e-02 |
Pos logit contributions | |
isa | +0.274 |
favour | +0.268 |
awa | +0.268 |
ophon | +0.260 |
swat | +0.248 |
Neg logit contributions | |
DEFENSE | -0.317 |
Vert | -0.284 |
HQ | -0.242 |
Crest | -0.239 |
Administrator | -0.239 |
Token | able |
Token ID | 1498 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
isa | +0.221 |
awa | +0.215 |
favour | +0.215 |
ophon | +0.208 |
swat | +0.198 |
Neg logit contributions | |
DEFENSE | -0.283 |
Vert | -0.255 |
HQ | -0.219 |
Crest | -0.217 |
Administrator | -0.215 |
Token | to |
Token ID | 284 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
DEFENSE | +0.094 |
Vert | +0.084 |
HQ | +0.072 |
Crest | +0.071 |
Administrator | +0.071 |
Neg logit contributions | |
isa | -0.079 |
favour | -0.076 |
awa | -0.076 |
ophon | -0.075 |
swat | -0.070 |
Token | see |
Token ID | 766 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
isa | +1.062 |
favour | +1.040 |
awa | +1.040 |
ophon | +1.012 |
swat | +0.971 |
Neg logit contributions | |
DEFENSE | -1.034 |
Vert | -0.918 |
HQ | -0.769 |
Crest | -0.760 |
Administrator | -0.757 |
Token | the |
Token ID | 262 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
isa | +1.095 |
awa | +1.078 |
favour | +1.077 |
ophon | +1.038 |
swat | +1.000 |
Neg logit contributions | |
DEFENSE | -1.216 |
Vert | -1.092 |
HQ | -0.920 |
Crest | -0.915 |
Administrator | -0.910 |
Token | screen |
Token ID | 3159 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
isa | +0.499 |
awa | +0.488 |
favour | +0.488 |
ophon | +0.475 |
swat | +0.453 |
Neg logit contributions | |
DEFENSE | -0.552 |
Vert | -0.495 |
HQ | -0.418 |
Crest | -0.414 |
Administrator | -0.413 |
Token | suspended |
Token ID | 9951 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
DEFENSE | +0.148 |
Vert | +0.133 |
HQ | +0.111 |
Administrator | +0.110 |
Crest | +0.110 |
Neg logit contributions | |
isa | -0.140 |
awa | -0.136 |
favour | -0.136 |
ophon | -0.134 |
swat | -0.126 |
Token | for |
Token ID | 329 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
isa | +0.365 |
favour | +0.356 |
awa | +0.355 |
ophon | +0.349 |
swat | +0.331 |
Neg logit contributions | |
DEFENSE | -0.402 |
Vert | -0.360 |
HQ | -0.303 |
Administrator | -0.300 |
Crest | -0.300 |
Token | two |
Token ID | 734 |
Feature activation | -2.34e-03 |
Pos logit contributions | |
isa | +0.325 |
favour | +0.316 |
awa | +0.314 |
ophon | +0.309 |
swat | +0.295 |
Neg logit contributions | |
DEFENSE | -0.370 |
Vert | -0.332 |
HQ | -0.282 |
7601 | -0.279 |
Administrator | -0.278 |
Token | games |
Token ID | 1830 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
DEFENSE | +0.008 |
Vert | +0.007 |
HQ | +0.006 |
Administrator | +0.005 |
Crest | +0.005 |
Neg logit contributions | |
isa | -0.009 |
favour | -0.009 |
awa | -0.009 |
ophon | -0.009 |
swat | -0.008 |
Token | for |
Token ID | 329 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
isa | +0.112 |
favour | +0.109 |
awa | +0.108 |
ophon | +0.106 |
swat | +0.100 |
Neg logit contributions | |
DEFENSE | -0.126 |
Vert | -0.113 |
HQ | -0.095 |
Crest | -0.094 |
Administrator | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
isa | +0.288 |
awa | +0.280 |
favour | +0.279 |
ophon | +0.275 |
swat | +0.260 |
Neg logit contributions | |
DEFENSE | -0.314 |
Vert | -0.281 |
HQ | -0.237 |
Administrator | -0.235 |
Crest | -0.234 |
Token | incident |
Token ID | 4519 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
isa | +0.522 |
favour | +0.507 |
awa | +0.507 |
ophon | +0.498 |
swat | +0.473 |
Neg logit contributions | |
DEFENSE | -0.598 |
Vert | -0.536 |
HQ | -0.452 |
Administrator | -0.450 |
Crest | -0.447 |
Token | , |
Token ID | 11 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
isa | +0.953 |
favour | +0.938 |
awa | +0.925 |
ophon | +0.914 |
swat | +0.867 |
Neg logit contributions | |
DEFENSE | -1.137 |
Vert | -1.030 |
HQ | -0.872 |
7601 | -0.861 |
Administrator | -0.857 |
Token | but |
Token ID | 475 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
DEFENSE | +0.830 |
Vert | +0.739 |
HQ | +0.619 |
Administrator | +0.614 |
Crest | +0.613 |
Neg logit contributions | |
isa | -0.810 |
awa | -0.788 |
favour | -0.787 |
ophon | -0.778 |
swat | -0.733 |
Token | has |
Token ID | 468 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
isa | +0.526 |
favour | +0.513 |
awa | +0.511 |
ophon | +0.503 |
swat | +0.475 |
Neg logit contributions | |
DEFENSE | -0.603 |
Vert | -0.543 |
HQ | -0.459 |
Administrator | -0.453 |
7601 | -0.453 |
Token | apparently |
Token ID | 5729 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
isa | +0.120 |
favour | +0.117 |
awa | +0.117 |
ophon | +0.115 |
swat | +0.109 |
Neg logit contributions | |
DEFENSE | -0.131 |
Vert | -0.117 |
HQ | -0.099 |
Administrator | -0.098 |
7601 | -0.098 |
Token | wireless |
Token ID | 12521 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
DEFENSE | +0.347 |
Vert | +0.312 |
HQ | +0.262 |
7601 | +0.257 |
Administrator | +0.256 |
Neg logit contributions | |
isa | -0.319 |
favour | -0.314 |
awa | -0.309 |
ophon | -0.304 |
swat | -0.293 |
Token | service |
Token ID | 2139 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
DEFENSE | +0.059 |
Vert | +0.053 |
HQ | +0.044 |
onel | +0.043 |
Administrator | +0.043 |
Neg logit contributions | |
isa | -0.056 |
favour | -0.055 |
awa | -0.054 |
ophon | -0.054 |
swat | -0.052 |
Token | for |
Token ID | 329 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
DEFENSE | +0.638 |
Vert | +0.572 |
HQ | +0.481 |
Administrator | +0.477 |
Crest | +0.474 |
Neg logit contributions | |
isa | -0.572 |
favour | -0.558 |
awa | -0.553 |
ophon | -0.548 |
swat | -0.516 |
Token | low |
Token ID | 1877 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
DEFENSE | +0.284 |
Vert | +0.255 |
HQ | +0.215 |
Administrator | +0.212 |
Crest | +0.211 |
Neg logit contributions | |
isa | -0.253 |
favour | -0.247 |
awa | -0.245 |
ophon | -0.242 |
swat | -0.230 |
Token | - |
Token ID | 12 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
isa | +0.137 |
awa | +0.133 |
favour | +0.132 |
ophon | +0.131 |
ethical | +0.122 |
Neg logit contributions | |
DEFENSE | -0.174 |
Vert | -0.156 |
HQ | -0.134 |
Crest | -0.133 |
Administrator | -0.133 |
Token | income |
Token ID | 12519 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
DEFENSE | +0.214 |
Vert | +0.194 |
HQ | +0.167 |
Crest | +0.166 |
Administrator | +0.166 |
Neg logit contributions | |
isa | -0.152 |
awa | -0.148 |
favour | -0.146 |
ophon | -0.145 |
ethical | -0.134 |
Token | Americans |
Token ID | 3399 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
isa | +0.365 |
favour | +0.357 |
awa | +0.349 |
ophon | +0.346 |
swat | +0.334 |
Neg logit contributions | |
DEFENSE | -0.434 |
Vert | -0.391 |
HQ | -0.330 |
7601 | -0.325 |
Administrator | -0.323 |
Token | . |
Token ID | 13 |
Feature activation | -5.77e-02 |
Pos logit contributions | |
isa | +1.354 |
favour | +1.313 |
awa | +1.278 |
ophon | +1.274 |
abet | +1.212 |
Neg logit contributions | |
DEFENSE | -1.717 |
Vert | -1.557 |
HQ | -1.327 |
Administrator | -1.304 |
Crest | -1.303 |
Token |
|
Token ID | 198 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
DEFENSE | +0.221 |
Vert | +0.198 |
HQ | +0.169 |
Crest | +0.167 |
Administrator | +0.166 |
Neg logit contributions | |
isa | -0.186 |
awa | -0.178 |
favour | -0.178 |
ophon | -0.177 |
abet | -0.165 |
Token |
|
Token ID | 198 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
isa | +0.596 |
favour | +0.570 |
ophon | +0.562 |
awa | +0.551 |
abet | +0.524 |
Neg logit contributions | |
DEFENSE | -0.755 |
Vert | -0.674 |
PDATE | -0.575 |
Crest | -0.574 |
HQ | -0.573 |
Token | The |
Token ID | 464 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
isa | +0.453 |
ophon | +0.432 |
favour | +0.427 |
awa | +0.424 |
abet | +0.407 |
Neg logit contributions | |
DEFENSE | -0.446 |
Vert | -0.386 |
Crest | -0.329 |
Administrator | -0.325 |
Ft | -0.325 |
Token | thought |
Token ID | 1807 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
isa | +0.305 |
favour | +0.298 |
awa | +0.295 |
ophon | +0.292 |
swat | +0.276 |
Neg logit contributions | |
DEFENSE | -0.335 |
Vert | -0.300 |
HQ | -0.253 |
Administrator | -0.249 |
Crest | -0.249 |
Token | the |
Token ID | 262 |
Feature activation | -4.30e-02 |
Pos logit contributions | |
isa | +0.359 |
favour | +0.350 |
awa | +0.350 |
ophon | +0.344 |
swat | +0.326 |
Neg logit contributions | |
DEFENSE | -0.360 |
Vert | -0.320 |
HQ | -0.267 |
Administrator | -0.265 |
Crest | -0.265 |
Token | already |
Token ID | 1541 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
DEFENSE | +0.162 |
Vert | +0.145 |
HQ | +0.123 |
Administrator | +0.122 |
Crest | +0.122 |
Neg logit contributions | |
isa | -0.142 |
awa | -0.138 |
favour | -0.137 |
ophon | -0.136 |
swat | -0.127 |
Token | booked |
Token ID | 21765 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
isa | +0.197 |
favour | +0.193 |
awa | +0.192 |
ophon | +0.189 |
swat | +0.178 |
Neg logit contributions | |
DEFENSE | -0.220 |
Vert | -0.196 |
HQ | -0.165 |
Administrator | -0.164 |
Crest | -0.164 |
Token | Can |
Token ID | 1680 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
DEFENSE | +1.698 |
Vert | +1.519 |
HQ | +1.287 |
Crest | +1.287 |
Administrator | +1.283 |
Neg logit contributions | |
isa | -1.478 |
awa | -1.435 |
ophon | -1.427 |
favour | -1.412 |
ethical | -1.328 |
Token | a |
Token ID | 64 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
isa | +0.458 |
awa | +0.444 |
favour | +0.441 |
ophon | +0.438 |
ethical | +0.408 |
Neg logit contributions | |
DEFENSE | -0.585 |
Vert | -0.526 |
HQ | -0.449 |
Administrator | -0.448 |
Crest | -0.447 |
Token | should |
Token ID | 815 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
isa | +0.265 |
favour | +0.258 |
awa | +0.256 |
ophon | +0.253 |
swat | +0.239 |
Neg logit contributions | |
DEFENSE | -0.297 |
Vert | -0.266 |
HQ | -0.225 |
Administrator | -0.223 |
Crest | -0.223 |
Token | have |
Token ID | 423 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
isa | +0.972 |
favour | +0.951 |
awa | +0.945 |
ophon | +0.929 |
swat | +0.873 |
Neg logit contributions | |
DEFENSE | -1.246 |
Vert | -1.126 |
HQ | -0.969 |
Administrator | -0.957 |
Crest | -0.956 |
Token | been |
Token ID | 587 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
DEFENSE | +0.109 |
Vert | +0.097 |
HQ | +0.083 |
Crest | +0.082 |
Administrator | +0.082 |
Neg logit contributions | |
isa | -0.096 |
favour | -0.094 |
awa | -0.093 |
ophon | -0.092 |
swat | -0.087 |
Token | sent |
Token ID | 1908 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
isa | +0.470 |
favour | +0.460 |
awa | +0.457 |
ophon | +0.449 |
swat | +0.426 |
Neg logit contributions | |
DEFENSE | -0.494 |
Vert | -0.441 |
HQ | -0.370 |
Administrator | -0.366 |
Crest | -0.366 |
Token | off |
Token ID | 572 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
isa | +0.178 |
awa | +0.174 |
favour | +0.174 |
ophon | +0.172 |
swat | +0.163 |
Neg logit contributions | |
DEFENSE | -0.145 |
Vert | -0.127 |
HQ | -0.103 |
Administrator | -0.102 |
Crest | -0.102 |
Token | ate |
Token ID | 378 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
DEFENSE | +0.950 |
Vert | +0.874 |
HQ | +0.775 |
Crest | +0.769 |
Administrator | +0.769 |
Neg logit contributions | |
isa | -0.447 |
awa | -0.442 |
favour | -0.432 |
ophon | -0.422 |
ethical | -0.386 |
Token | , |
Token ID | 11 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
DEFENSE | +0.610 |
Vert | +0.545 |
HQ | +0.459 |
Administrator | +0.456 |
Crest | +0.456 |
Neg logit contributions | |
isa | -0.555 |
awa | -0.540 |
favour | -0.537 |
ophon | -0.532 |
ethical | -0.499 |
Token | ur |
Token ID | 2956 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
DEFENSE | +0.434 |
Vert | +0.387 |
HQ | +0.326 |
Administrator | +0.324 |
Crest | +0.324 |
Neg logit contributions | |
isa | -0.401 |
awa | -0.391 |
favour | -0.390 |
ophon | -0.384 |
swat | -0.364 |
Token | inate |
Token ID | 4559 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
isa | +0.462 |
awa | +0.429 |
favour | +0.426 |
ophon | +0.421 |
abet | +0.374 |
Neg logit contributions | |
DEFENSE | -1.457 |
Vert | -1.353 |
Administrator | -1.211 |
HQ | -1.209 |
Crest | -1.207 |
Token | and |
Token ID | 290 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
isa | +0.684 |
awa | +0.667 |
favour | +0.665 |
ophon | +0.653 |
swat | +0.617 |
Neg logit contributions | |
DEFENSE | -0.771 |
Vert | -0.688 |
HQ | -0.578 |
Administrator | -0.575 |
Crest | -0.575 |
Token | pro |
Token ID | 386 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
DEFENSE | +0.699 |
Vert | +0.629 |
HQ | +0.537 |
Administrator | +0.535 |
Crest | +0.535 |
Neg logit contributions | |
isa | -0.548 |
awa | -0.531 |
favour | -0.529 |
ophon | -0.523 |
swat | -0.488 |
Token | create |
Token ID | 17953 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
DEFENSE | +1.077 |
Vert | +0.958 |
HQ | +0.807 |
Crest | +0.804 |
Administrator | +0.803 |
Neg logit contributions | |
isa | -0.985 |
awa | -0.960 |
favour | -0.955 |
ophon | -0.946 |
ethical | -0.888 |
Token | over |
Token ID | 625 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
DEFENSE | +0.884 |
Vert | +0.787 |
HQ | +0.659 |
Administrator | +0.657 |
Crest | +0.656 |
Neg logit contributions | |
isa | -0.854 |
awa | -0.830 |
favour | -0.825 |
ophon | -0.821 |
ethical | -0.770 |
Token | there |
Token ID | 612 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
isa | +1.074 |
favour | +1.047 |
awa | +1.046 |
ophon | +1.022 |
swat | +0.983 |
Neg logit contributions | |
DEFENSE | -1.160 |
Vert | -1.036 |
HQ | -0.863 |
Administrator | -0.862 |
7601 | -0.857 |
Token | ," |
Token ID | 553 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
DEFENSE | +0.091 |
Vert | +0.081 |
HQ | +0.067 |
Administrator | +0.066 |
Crest | +0.066 |
Neg logit contributions | |
isa | -0.095 |
awa | -0.093 |
favour | -0.093 |
ophon | -0.091 |
swat | -0.087 |
Token | said |
Token ID | 531 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
DEFENSE | +0.172 |
Vert | +0.150 |
HQ | +0.121 |
Administrator | +0.120 |
Crest | +0.120 |
Neg logit contributions | |
isa | -0.220 |
awa | -0.215 |
favour | -0.214 |
ophon | -0.212 |
ethical | -0.201 |
Token | facebook |
Token ID | 19024 |
Feature activation | -7.78e-03 |
Pos logit contributions | |
isa | +0.496 |
awa | +0.472 |
favour | +0.471 |
ophon | +0.470 |
ethical | +0.438 |
Neg logit contributions | |
DEFENSE | -0.758 |
Vert | -0.681 |
Crest | -0.592 |
Administrator | -0.592 |
HQ | -0.591 |
Token | . |
Token ID | 13 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
DEFENSE | +0.030 |
Vert | +0.027 |
HQ | +0.023 |
Administrator | +0.023 |
Crest | +0.023 |
Neg logit contributions | |
isa | -0.025 |
favour | -0.024 |
awa | -0.024 |
ophon | -0.024 |
ethical | -0.022 |
Token | com |
Token ID | 785 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
isa | +1.273 |
favour | +1.215 |
ophon | +1.213 |
awa | +1.187 |
abet | +1.147 |
Neg logit contributions | |
DEFENSE | -1.248 |
Vert | -1.088 |
PDATE | -0.912 |
Administrator | -0.910 |
Crest | -0.907 |
Token | / |
Token ID | 14 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
isa | +1.653 |
favour | +1.577 |
ophon | +1.565 |
awa | +1.534 |
abet | +1.471 |
Neg logit contributions | |
DEFENSE | -2.101 |
Vert | -1.878 |
PDATE | -1.611 |
HQ | -1.596 |
Administrator | -1.594 |
Token | iv |
Token ID | 452 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
DEFENSE | +0.102 |
Vert | +0.090 |
Crest | +0.076 |
Administrator | +0.076 |
HQ | +0.076 |
Neg logit contributions | |
isa | -0.092 |
awa | -0.089 |
favour | -0.089 |
ophon | -0.088 |
ethical | -0.083 |
Token | xx |
Token ID | 5324 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
DEFENSE | +0.878 |
Vert | +0.790 |
HQ | +0.672 |
Administrator | +0.665 |
Crest | +0.665 |
Neg logit contributions | |
isa | -0.727 |
awa | -0.715 |
favour | -0.707 |
ophon | -0.698 |
swat | -0.655 |
Token | brand |
Token ID | 17938 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
DEFENSE | +2.483 |
Vert | +2.239 |
HQ | +1.917 |
Administrator | +1.894 |
Crest | +1.882 |
Neg logit contributions | |
isa | -2.130 |
awa | -2.103 |
favour | -2.068 |
ophon | -2.055 |
swat | -1.926 |
Token | ? |
Token ID | 30 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
DEFENSE | +1.606 |
Vert | +1.445 |
HQ | +1.232 |
Administrator | +1.223 |
Crest | +1.219 |
Neg logit contributions | |
isa | -1.344 |
awa | -1.313 |
favour | -1.301 |
ophon | -1.289 |
ethical | -1.204 |
Token | f |
Token ID | 69 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
isa | +1.762 |
awa | +1.692 |
favour | +1.685 |
ophon | +1.649 |
abet | +1.573 |
Neg logit contributions | |
DEFENSE | -2.328 |
Vert | -2.056 |
Crest | -1.763 |
Administrator | -1.760 |
7601 | -1.755 |
Token | ref |
Token ID | 5420 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
isa | +0.296 |
awa | +0.283 |
favour | +0.282 |
ophon | +0.279 |
ethical | +0.259 |
Neg logit contributions | |
DEFENSE | -0.475 |
Vert | -0.429 |
HQ | -0.374 |
Crest | -0.373 |
Administrator | -0.372 |
Token | = |
Token ID | 28 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
DEFENSE | +2.098 |
Vert | +1.933 |
HQ | +1.685 |
Administrator | +1.671 |
Crest | +1.667 |
Neg logit contributions | |
isa | -1.281 |
awa | -1.265 |
favour | -1.232 |
ophon | -1.226 |
ethical | -1.124 |
Token | rying |
Token ID | 14992 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
DEFENSE | +1.479 |
Vert | +1.345 |
HQ | +1.153 |
Administrator | +1.137 |
Crest | +1.129 |
Neg logit contributions | |
awa | -1.184 |
isa | -1.183 |
favour | -1.172 |
ophon | -1.164 |
ethical | -1.081 |
Token | the |
Token ID | 262 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
DEFENSE | +1.479 |
Vert | +1.327 |
Crest | +1.116 |
Administrator | +1.115 |
HQ | +1.114 |
Neg logit contributions | |
isa | -1.336 |
awa | -1.294 |
favour | -1.284 |
ophon | -1.283 |
ethical | -1.206 |
Token | mushrooms |
Token ID | 23452 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
DEFENSE | +1.626 |
Vert | +1.468 |
Crest | +1.256 |
Administrator | +1.256 |
HQ | +1.254 |
Neg logit contributions | |
isa | -1.234 |
awa | -1.192 |
ophon | -1.181 |
favour | -1.180 |
ethical | -1.100 |
Token | as |
Token ID | 355 |
Feature activation | -9.69e-02 |
Pos logit contributions | |
DEFENSE | +0.350 |
Vert | +0.313 |
HQ | +0.265 |
Administrator | +0.263 |
Crest | +0.263 |
Neg logit contributions | |
isa | -0.304 |
awa | -0.295 |
favour | -0.295 |
ophon | -0.291 |
ethical | -0.272 |
Token | described |
Token ID | 3417 |
Feature activation | -3.01e-03 |
Pos logit contributions | |
DEFENSE | +0.354 |
Vert | +0.315 |
HQ | +0.264 |
Crest | +0.262 |
Administrator | +0.261 |
Neg logit contributions | |
isa | -0.332 |
favour | -0.323 |
awa | -0.321 |
ophon | -0.319 |
swat | -0.299 |
Token | in |
Token ID | 287 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
DEFENSE | +0.011 |
Vert | +0.010 |
HQ | +0.008 |
Administrator | +0.008 |
Crest | +0.008 |
Neg logit contributions | |
isa | -0.010 |
awa | -0.010 |
favour | -0.010 |
ophon | -0.010 |
swat | -0.009 |
Token | this |
Token ID | 428 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
DEFENSE | +0.832 |
Vert | +0.736 |
HQ | +0.609 |
Administrator | +0.607 |
Crest | +0.607 |
Neg logit contributions | |
isa | -0.886 |
awa | -0.863 |
favour | -0.858 |
ophon | -0.854 |
ethical | -0.803 |
Token | document |
Token ID | 3188 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
DEFENSE | +1.669 |
Vert | +1.477 |
Administrator | +1.230 |
HQ | +1.230 |
Crest | +1.227 |
Neg logit contributions | |
isa | -1.709 |
awa | -1.666 |
favour | -1.651 |
ophon | -1.650 |
ethical | -1.548 |
Token | will |
Token ID | 481 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
DEFENSE | +0.234 |
Vert | +0.209 |
HQ | +0.178 |
Crest | +0.177 |
Administrator | +0.177 |
Neg logit contributions | |
isa | -0.199 |
awa | -0.193 |
favour | -0.192 |
ophon | -0.190 |
swat | -0.177 |
Token | have |
Token ID | 423 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
DEFENSE | +0.978 |
Vert | +0.882 |
HQ | +0.753 |
Crest | +0.748 |
Administrator | +0.747 |
Neg logit contributions | |
isa | -0.785 |
awa | -0.763 |
favour | -0.763 |
ophon | -0.750 |
swat | -0.702 |
Token | minimal |
Token ID | 10926 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
DEFENSE | +0.290 |
Vert | +0.260 |
HQ | +0.219 |
Crest | +0.217 |
Administrator | +0.217 |
Neg logit contributions | |
isa | -0.252 |
awa | -0.245 |
favour | -0.245 |
ophon | -0.240 |
swat | -0.227 |
Token | Talk |
Token ID | 12167 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
DEFENSE | +0.292 |
Vert | +0.261 |
HQ | +0.222 |
Administrator | +0.220 |
Crest | +0.220 |
Neg logit contributions | |
isa | -0.241 |
awa | -0.232 |
favour | -0.231 |
ophon | -0.230 |
abet | -0.215 |
Token | To |
Token ID | 2514 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
isa | +0.455 |
awa | +0.439 |
favour | +0.436 |
ophon | +0.433 |
ethical | +0.406 |
Neg logit contributions | |
DEFENSE | -0.516 |
Vert | -0.459 |
Crest | -0.389 |
Administrator | -0.386 |
HQ | -0.386 |
Token | founder |
Token ID | 9119 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
isa | +0.327 |
awa | +0.316 |
favour | +0.315 |
ophon | +0.313 |
swat | +0.293 |
Neg logit contributions | |
DEFENSE | -0.358 |
Vert | -0.318 |
HQ | -0.266 |
Crest | -0.266 |
7601 | -0.265 |
Token | Stuart |
Token ID | 22559 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
DEFENSE | +1.653 |
Vert | +1.476 |
Crest | +1.247 |
Administrator | +1.244 |
HQ | +1.239 |
Neg logit contributions | |
isa | -1.563 |
awa | -1.530 |
favour | -1.520 |
ophon | -1.511 |
ethical | -1.416 |
Token | Lev |
Token ID | 16042 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
isa | +1.367 |
favour | +1.317 |
awa | +1.316 |
ophon | +1.303 |
abet | +1.229 |
Neg logit contributions | |
DEFENSE | -1.334 |
Vert | -1.178 |
HQ | -0.982 |
7601 | -0.969 |
Administrator | -0.967 |
Token | inson |
Token ID | 7899 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
DEFENSE | +0.770 |
Vert | +0.692 |
HQ | +0.589 |
Crest | +0.587 |
Administrator | +0.587 |
Neg logit contributions | |
isa | -0.623 |
awa | -0.605 |
favour | -0.601 |
ophon | -0.596 |
ethical | -0.556 |
Token | wouldn |
Token ID | 3636 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
DEFENSE | +1.311 |
Vert | +1.159 |
HQ | +0.954 |
Crest | +0.953 |
Administrator | +0.953 |
Neg logit contributions | |
isa | -1.455 |
awa | -1.426 |
favour | -1.415 |
ophon | -1.405 |
ethical | -1.329 |
Token | � |
Token ID | 447 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
DEFENSE | +1.727 |
Vert | +1.562 |
HQ | +1.342 |
Administrator | +1.336 |
Crest | +1.335 |
Neg logit contributions | |
isa | -1.261 |
awa | -1.232 |
favour | -1.218 |
ophon | -1.208 |
ethical | -1.122 |
Token | � |
Token ID | 247 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
DEFENSE | +1.113 |
Vert | +1.007 |
HQ | +0.867 |
Administrator | +0.864 |
Crest | +0.863 |
Neg logit contributions | |
isa | -0.787 |
awa | -0.764 |
favour | -0.758 |
ophon | -0.752 |
ethical | -0.697 |
Token | t |
Token ID | 83 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
DEFENSE | +0.998 |
Vert | +0.905 |
HQ | +0.785 |
Administrator | +0.783 |
Crest | +0.782 |
Neg logit contributions | |
isa | -0.647 |
awa | -0.627 |
favour | -0.623 |
ophon | -0.617 |
ethical | -0.569 |
Token | say |
Token ID | 910 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
DEFENSE | +1.452 |
Vert | +1.280 |
Administrator | +1.087 |
HQ | +1.083 |
Crest | +1.081 |
Neg logit contributions | |
isa | -1.299 |
awa | -1.261 |
ophon | -1.252 |
favour | -1.246 |
ethical | -1.176 |
Token | wherever |
Token ID | 14530 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
isa | +1.126 |
favour | +1.105 |
awa | +1.098 |
ophon | +1.064 |
swat | +1.017 |
Neg logit contributions | |
DEFENSE | -1.326 |
Vert | -1.195 |
HQ | -1.013 |
7601 | -0.998 |
Crest | -0.998 |
Token | it |
Token ID | 340 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
DEFENSE | +1.280 |
Vert | +1.143 |
HQ | +0.956 |
Crest | +0.950 |
Administrator | +0.946 |
Neg logit contributions | |
isa | -1.217 |
awa | -1.186 |
favour | -1.181 |
ophon | -1.166 |
swat | -1.096 |
Token | came |
Token ID | 1625 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
DEFENSE | +1.857 |
Vert | +1.669 |
HQ | +1.421 |
Crest | +1.415 |
Administrator | +1.414 |
Neg logit contributions | |
isa | -1.502 |
awa | -1.455 |
favour | -1.448 |
ophon | -1.436 |
ethical | -1.338 |
Token | from |
Token ID | 422 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
DEFENSE | +1.738 |
Vert | +1.534 |
HQ | +1.267 |
Administrator | +1.261 |
Crest | +1.260 |
Neg logit contributions | |
isa | -1.909 |
awa | -1.860 |
favour | -1.850 |
ophon | -1.838 |
ethical | -1.735 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
DEFENSE | +2.065 |
Vert | +1.843 |
HQ | +1.557 |
Administrator | +1.554 |
Crest | +1.549 |
Neg logit contributions | |
isa | -1.879 |
awa | -1.827 |
favour | -1.811 |
ophon | -1.805 |
ethical | -1.690 |
Token | Make |
Token ID | 6889 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
DEFENSE | +1.011 |
Vert | +0.909 |
HQ | +0.771 |
Crest | +0.763 |
Administrator | +0.762 |
Neg logit contributions | |
isa | -0.847 |
awa | -0.818 |
favour | -0.814 |
ophon | -0.803 |
swat | -0.753 |
Token | sure |
Token ID | 1654 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
DEFENSE | +0.639 |
Vert | +0.561 |
HQ | +0.453 |
Crest | +0.448 |
Administrator | +0.447 |
Neg logit contributions | |
isa | -0.807 |
awa | -0.787 |
favour | -0.784 |
ophon | -0.776 |
swat | -0.738 |
Token | the |
Token ID | 262 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
isa | +0.305 |
awa | +0.297 |
favour | +0.297 |
ophon | +0.291 |
swat | +0.277 |
Neg logit contributions | |
DEFENSE | -0.302 |
Vert | -0.271 |
HQ | -0.225 |
Crest | -0.222 |
onel | -0.221 |
Token | data |
Token ID | 1366 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
isa | +0.097 |
awa | +0.094 |
favour | +0.094 |
ophon | +0.092 |
swat | +0.088 |
Neg logit contributions | |
DEFENSE | -0.084 |
Vert | -0.074 |
HQ | -0.060 |
Crest | -0.060 |
7601 | -0.059 |
Token | can |
Token ID | 460 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
DEFENSE | +0.696 |
Vert | +0.629 |
HQ | +0.538 |
Crest | +0.536 |
Administrator | +0.533 |
Neg logit contributions | |
isa | -0.525 |
awa | -0.506 |
favour | -0.505 |
ophon | -0.498 |
ethical | -0.465 |
Token | do |
Token ID | 466 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
DEFENSE | +0.936 |
Vert | +0.839 |
HQ | +0.712 |
Administrator | +0.709 |
Crest | +0.709 |
Neg logit contributions | |
isa | -0.791 |
awa | -0.768 |
favour | -0.763 |
ophon | -0.758 |
ethical | -0.707 |
Token | pretty |
Token ID | 2495 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
isa | +0.160 |
awa | +0.156 |
favour | +0.155 |
ophon | +0.154 |
ethical | +0.144 |
Neg logit contributions | |
DEFENSE | -0.183 |
Vert | -0.164 |
HQ | -0.138 |
Administrator | -0.138 |
Crest | -0.137 |
Token | heavy |
Token ID | 4334 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
DEFENSE | +0.439 |
Vert | +0.391 |
HQ | +0.328 |
Administrator | +0.327 |
Crest | +0.327 |
Neg logit contributions | |
isa | -0.411 |
awa | -0.400 |
favour | -0.399 |
ophon | -0.395 |
swat | -0.370 |
Token | software |
Token ID | 3788 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
isa | +0.562 |
favour | +0.549 |
awa | +0.549 |
ophon | +0.539 |
swat | +0.511 |
Neg logit contributions | |
DEFENSE | -0.602 |
Vert | -0.536 |
HQ | -0.450 |
Administrator | -0.448 |
Crest | -0.447 |
Token | correction |
Token ID | 17137 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
DEFENSE | +0.238 |
Vert | +0.213 |
HQ | +0.178 |
Crest | +0.177 |
Administrator | +0.177 |
Neg logit contributions | |
isa | -0.225 |
awa | -0.219 |
favour | -0.219 |
ophon | -0.216 |
swat | -0.203 |
Token | , |
Token ID | 11 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
DEFENSE | +1.376 |
Vert | +1.228 |
HQ | +1.035 |
Administrator | +1.033 |
Crest | +1.032 |
Neg logit contributions | |
isa | -1.251 |
awa | -1.214 |
favour | -1.202 |
ophon | -1.199 |
ethical | -1.124 |
Token | especially |
Token ID | 2592 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
DEFENSE | +0.384 |
Vert | +0.342 |
HQ | +0.288 |
Crest | +0.286 |
Administrator | +0.285 |
Neg logit contributions | |
isa | -0.355 |
favour | -0.349 |
awa | -0.346 |
ophon | -0.341 |
swat | -0.323 |
Token | in |
Token ID | 287 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
DEFENSE | +2.729 |
Vert | +2.418 |
HQ | +2.007 |
Crest | +2.007 |
Administrator | +2.007 |
Neg logit contributions | |
isa | -2.772 |
awa | -2.701 |
ophon | -2.672 |
favour | -2.651 |
ethical | -2.519 |
Token | the |
Token ID | 262 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
DEFENSE | +0.257 |
Vert | +0.229 |
HQ | +0.191 |
Administrator | +0.190 |
Crest | +0.190 |
Neg logit contributions | |
isa | -0.252 |
favour | -0.245 |
awa | -0.245 |
ophon | -0.242 |
swat | -0.227 |
Token | corners |
Token ID | 14371 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
isa | +1.186 |
favour | +1.168 |
awa | +1.162 |
ophon | +1.139 |
swat | +1.075 |
Neg logit contributions | |
DEFENSE | -1.281 |
Vert | -1.145 |
HQ | -0.961 |
Administrator | -0.953 |
Crest | -0.951 |
Token | . |
Token ID | 13 |
Feature activation | +9.57e-02 |
Pos logit contributions | |
isa | +0.600 |
favour | +0.585 |
awa | +0.581 |
ophon | +0.576 |
swat | +0.539 |
Neg logit contributions | |
DEFENSE | -0.699 |
Vert | -0.626 |
HQ | -0.529 |
Crest | -0.527 |
Administrator | -0.526 |
Token | Everything |
Token ID | 11391 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
isa | +0.298 |
favour | +0.286 |
awa | +0.285 |
ophon | +0.283 |
swat | +0.263 |
Neg logit contributions | |
DEFENSE | -0.374 |
Vert | -0.337 |
HQ | -0.287 |
onel | -0.285 |
Administrator | -0.284 |
Token | that |
Token ID | 326 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
isa | +1.283 |
favour | +1.262 |
awa | +1.250 |
ophon | +1.218 |
swat | +1.163 |
Neg logit contributions | |
DEFENSE | -1.330 |
Vert | -1.199 |
HQ | -1.001 |
Crest | -0.994 |
7601 | -0.988 |
Token | would |
Token ID | 561 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
DEFENSE | +0.104 |
Vert | +0.093 |
HQ | +0.080 |
Crest | +0.079 |
Administrator | +0.079 |
Neg logit contributions | |
isa | -0.082 |
favour | -0.080 |
awa | -0.079 |
ophon | -0.078 |
swat | -0.073 |
Token | be |
Token ID | 307 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
isa | +0.139 |
favour | +0.135 |
awa | +0.135 |
ophon | +0.133 |
swat | +0.124 |
Neg logit contributions | |
DEFENSE | -0.162 |
Vert | -0.146 |
HQ | -0.124 |
Crest | -0.123 |
Administrator | -0.123 |
Token | a |
Token ID | 257 |
Feature activation | +6.76e-02 |
Pos logit contributions | |
isa | +0.164 |
favour | +0.160 |
awa | +0.159 |
ophon | +0.156 |
swat | +0.147 |
Neg logit contributions | |
DEFENSE | -0.198 |
Vert | -0.178 |
HQ | -0.151 |
Crest | -0.150 |
Administrator | -0.150 |
Token | bigger |
Token ID | 5749 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
isa | +0.201 |
awa | +0.195 |
favour | +0.194 |
ophon | +0.192 |
ethical | +0.178 |
Neg logit contributions | |
DEFENSE | -0.276 |
Vert | -0.249 |
HQ | -0.214 |
Administrator | -0.213 |
Crest | -0.213 |
Token | problem |
Token ID | 1917 |
Feature activation | -9.33e-01 |
Pos logit contributions | |
DEFENSE | +1.503 |
Vert | +1.358 |
HQ | +1.181 |
Administrator | +1.181 |
Crest | +1.179 |
Neg logit contributions | |
isa | -0.948 |
awa | -0.914 |
ophon | -0.907 |
favour | -0.896 |
abet | -0.832 |
Token | . |
Token ID | 13 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
DEFENSE | +3.181 |
Vert | +2.804 |
Administrator | +2.355 |
HQ | +2.333 |
Crest | +2.319 |
Neg logit contributions | |
isa | -3.358 |
awa | -3.275 |
ophon | -3.260 |
favour | -3.203 |
ethical | -3.056 |
Token | I |
Token ID | 314 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
DEFENSE | +1.233 |
Vert | +1.112 |
HQ | +0.952 |
Crest | +0.947 |
Administrator | +0.946 |
Neg logit contributions | |
isa | -0.940 |
awa | -0.908 |
favour | -0.903 |
ophon | -0.895 |
ethical | -0.831 |
Token | don |
Token ID | 836 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
DEFENSE | +0.559 |
Vert | +0.500 |
HQ | +0.423 |
Administrator | +0.421 |
Crest | +0.421 |
Neg logit contributions | |
isa | -0.491 |
awa | -0.477 |
favour | -0.474 |
ophon | -0.471 |
ethical | -0.440 |
Token | 't |
Token ID | 470 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
DEFENSE | +0.841 |
Vert | +0.762 |
HQ | +0.657 |
Administrator | +0.655 |
Crest | +0.655 |
Neg logit contributions | |
isa | -0.578 |
awa | -0.562 |
favour | -0.556 |
ophon | -0.552 |
ethical | -0.510 |
Token | think |
Token ID | 892 |
Feature activation | +5.95e-02 |
Pos logit contributions | |
DEFENSE | +0.075 |
Vert | +0.067 |
HQ | +0.056 |
Crest | +0.056 |
Administrator | +0.056 |
Neg logit contributions | |
isa | -0.071 |
favour | -0.069 |
awa | -0.069 |
ophon | -0.068 |
swat | -0.064 |
Token | � |
Token ID | 447 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
isa | +0.706 |
awa | +0.682 |
favour | +0.680 |
ophon | +0.673 |
swat | +0.636 |
Neg logit contributions | |
DEFENSE | -0.717 |
Vert | -0.636 |
HQ | -0.531 |
Crest | -0.526 |
Administrator | -0.524 |
Token | � |
Token ID | 247 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
isa | +0.323 |
favour | +0.310 |
ophon | +0.308 |
awa | +0.307 |
swat | +0.289 |
Neg logit contributions | |
DEFENSE | -0.371 |
Vert | -0.332 |
Crest | -0.281 |
HQ | -0.279 |
Administrator | -0.277 |
Token | s |
Token ID | 82 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
isa | +0.438 |
awa | +0.421 |
favour | +0.419 |
ophon | +0.415 |
ethical | +0.387 |
Neg logit contributions | |
DEFENSE | -0.570 |
Vert | -0.512 |
HQ | -0.437 |
Administrator | -0.436 |
Crest | -0.435 |
Token | just |
Token ID | 655 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
isa | +0.429 |
awa | +0.420 |
favour | +0.418 |
ophon | +0.409 |
swat | +0.388 |
Neg logit contributions | |
DEFENSE | -0.482 |
Vert | -0.430 |
HQ | -0.363 |
7601 | -0.360 |
Crest | -0.359 |
Token | sitting |
Token ID | 5586 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
isa | +0.614 |
awa | +0.601 |
favour | +0.598 |
ophon | +0.586 |
swat | +0.554 |
Neg logit contributions | |
DEFENSE | -0.705 |
Vert | -0.630 |
HQ | -0.533 |
Crest | -0.529 |
7601 | -0.526 |
Token | there |
Token ID | 612 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
isa | +0.814 |
awa | +0.801 |
favour | +0.793 |
ophon | +0.782 |
swat | +0.740 |
Neg logit contributions | |
DEFENSE | -0.958 |
Vert | -0.856 |
HQ | -0.728 |
7601 | -0.716 |
Administrator | -0.715 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
DEFENSE | +3.264 |
Vert | +2.927 |
Administrator | +2.465 |
Crest | +2.461 |
HQ | +2.457 |
Neg logit contributions | |
isa | -3.107 |
awa | -2.986 |
favour | -2.977 |
ophon | -2.975 |
ethical | -2.803 |
Token | So |
Token ID | 1406 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
DEFENSE | +0.412 |
Vert | +0.371 |
HQ | +0.319 |
Crest | +0.317 |
Administrator | +0.316 |
Neg logit contributions | |
isa | -0.301 |
awa | -0.291 |
favour | -0.290 |
ophon | -0.287 |
swat | -0.267 |
Token | of |
Token ID | 286 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
isa | +0.460 |
awa | +0.447 |
favour | +0.445 |
ophon | +0.438 |
swat | +0.413 |
Neg logit contributions | |
DEFENSE | -0.526 |
Vert | -0.471 |
HQ | -0.397 |
Crest | -0.395 |
Administrator | -0.393 |
Token | course |
Token ID | 1781 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
isa | +0.907 |
favour | +0.876 |
awa | +0.875 |
ophon | +0.853 |
swat | +0.802 |
Neg logit contributions | |
DEFENSE | -1.257 |
Vert | -1.135 |
HQ | -0.973 |
7601 | -0.967 |
Administrator | -0.962 |
Token | , |
Token ID | 11 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
isa | +0.414 |
awa | +0.402 |
favour | +0.400 |
ophon | +0.396 |
swat | +0.371 |
Neg logit contributions | |
DEFENSE | -0.486 |
Vert | -0.435 |
HQ | -0.368 |
Crest | -0.366 |
Administrator | -0.365 |
Token | first |
Token ID | 717 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
DEFENSE | +0.276 |
Vert | +0.246 |
HQ | +0.206 |
Administrator | +0.205 |
Crest | +0.204 |
Neg logit contributions | |
isa | -0.262 |
awa | -0.255 |
favour | -0.255 |
ophon | -0.251 |
swat | -0.237 |
Token | when |
Token ID | 618 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
isa | +1.362 |
favour | +1.349 |
awa | +1.318 |
ophon | +1.271 |
swat | +1.226 |
Neg logit contributions | |
DEFENSE | -1.697 |
Vert | -1.542 |
HQ | -1.303 |
Administrator | -1.290 |
7601 | -1.284 |
Token | Alex |
Token ID | 4422 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
DEFENSE | +0.670 |
Vert | +0.596 |
HQ | +0.499 |
Administrator | +0.498 |
Crest | +0.498 |
Neg logit contributions | |
isa | -0.642 |
awa | -0.625 |
favour | -0.621 |
ophon | -0.618 |
ethical | -0.580 |
Token | O |
Token ID | 440 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
DEFENSE | +1.253 |
Vert | +1.122 |
Crest | +0.953 |
Administrator | +0.951 |
HQ | +0.950 |
Neg logit contributions | |
isa | -1.065 |
awa | -1.045 |
favour | -1.031 |
ophon | -1.025 |
ethical | -0.965 |
Token | ve |
Token ID | 303 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
isa | +0.871 |
awa | +0.831 |
favour | +0.822 |
ophon | +0.813 |
ethical | +0.724 |
Neg logit contributions | |
DEFENSE | -2.186 |
Vert | -2.016 |
HQ | -1.789 |
Administrator | -1.784 |
Crest | -1.783 |
Token | ch |
Token ID | 354 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
DEFENSE | +1.773 |
Vert | +1.583 |
Crest | +1.335 |
onel | +1.321 |
Administrator | +1.318 |
Neg logit contributions | |
awa | -2.045 |
isa | -2.014 |
favour | -2.004 |
ophon | -1.986 |
ethical | -1.914 |
Token | kin |
Token ID | 5116 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
DEFENSE | +2.508 |
Vert | +2.247 |
Administrator | +1.838 |
VS | +1.827 |
HQ | +1.826 |
Neg logit contributions | |
isa | -2.703 |
awa | -2.686 |
favour | -2.652 |
ophon | -2.639 |
ethical | -2.505 |
Token | scored |
Token ID | 7781 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
DEFENSE | +2.262 |
Vert | +2.020 |
Crest | +1.711 |
Administrator | +1.707 |
HQ | +1.703 |
Neg logit contributions | |
isa | -1.976 |
awa | -1.921 |
ophon | -1.911 |
favour | -1.898 |
ethical | -1.782 |
Token | his |
Token ID | 465 |
Feature activation | +8.81e-05 |
Pos logit contributions | |
DEFENSE | +1.070 |
Vert | +0.955 |
HQ | +0.803 |
Administrator | +0.801 |
Crest | +0.801 |
Neg logit contributions | |
isa | -0.987 |
awa | -0.960 |
favour | -0.953 |
ophon | -0.949 |
ethical | -0.889 |
Token | 44 |
Token ID | 5846 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
isa | +0.000 |
favour | +0.000 |
awa | +0.000 |
ophon | +0.000 |
swat | +0.000 |
Neg logit contributions | |
DEFENSE | -0.000 |
Vert | -0.000 |
HQ | -0.000 |
Administrator | -0.000 |
Crest | -0.000 |
Token | th |
Token ID | 400 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
isa | +0.686 |
favour | +0.666 |
awa | +0.664 |
ophon | +0.652 |
ethical | +0.618 |
Neg logit contributions | |
DEFENSE | -0.723 |
Vert | -0.643 |
Administrator | -0.541 |
HQ | -0.539 |
Crest | -0.535 |
Token | � |
Token ID | 247 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
DEFENSE | +1.535 |
Vert | +1.377 |
HQ | +1.177 |
Administrator | +1.174 |
Crest | +1.169 |
Neg logit contributions | |
isa | -1.230 |
awa | -1.195 |
favour | -1.189 |
ophon | -1.182 |
ethical | -1.100 |
Token | t |
Token ID | 83 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
DEFENSE | +0.950 |
Vert | +0.861 |
HQ | +0.746 |
Administrator | +0.744 |
Crest | +0.743 |
Neg logit contributions | |
isa | -0.630 |
awa | -0.610 |
favour | -0.605 |
ophon | -0.601 |
ethical | -0.554 |
Token | you |
Token ID | 345 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
isa | +1.287 |
favour | +1.253 |
awa | +1.247 |
ophon | +1.227 |
swat | +1.165 |
Neg logit contributions | |
DEFENSE | -1.337 |
Vert | -1.205 |
HQ | -0.998 |
7601 | -0.990 |
Crest | -0.988 |
Token | ? |
Token ID | 30 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
DEFENSE | +0.876 |
Vert | +0.786 |
HQ | +0.668 |
Administrator | +0.665 |
Crest | +0.665 |
Neg logit contributions | |
isa | -0.728 |
awa | -0.706 |
favour | -0.703 |
ophon | -0.697 |
ethical | -0.650 |
Token | Considering |
Token ID | 40475 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
DEFENSE | +1.137 |
Vert | +1.022 |
HQ | +0.867 |
Crest | +0.863 |
Administrator | +0.862 |
Neg logit contributions | |
isa | -0.941 |
awa | -0.911 |
favour | -0.906 |
ophon | -0.899 |
swat | -0.837 |
Token | all |
Token ID | 477 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
DEFENSE | +0.781 |
Vert | +0.693 |
HQ | +0.571 |
Administrator | +0.564 |
Crest | +0.563 |
Neg logit contributions | |
isa | -0.844 |
awa | -0.823 |
favour | -0.822 |
ophon | -0.813 |
swat | -0.768 |
Token | you |
Token ID | 345 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
DEFENSE | +2.685 |
Vert | +2.378 |
Administrator | +2.012 |
HQ | +2.003 |
Crest | +2.001 |
Neg logit contributions | |
isa | -2.526 |
awa | -2.460 |
favour | -2.437 |
ophon | -2.422 |
ethical | -2.278 |
Token | get |
Token ID | 651 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
DEFENSE | +0.738 |
Vert | +0.662 |
HQ | +0.558 |
Crest | +0.556 |
Administrator | +0.555 |
Neg logit contributions | |
isa | -0.643 |
awa | -0.623 |
favour | -0.621 |
ophon | -0.614 |
swat | -0.576 |
Token | , |
Token ID | 11 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
DEFENSE | +0.195 |
Vert | +0.175 |
HQ | +0.145 |
Crest | +0.145 |
Administrator | +0.144 |
Neg logit contributions | |
isa | -0.183 |
awa | -0.178 |
favour | -0.178 |
ophon | -0.175 |
swat | -0.165 |
Token | it |
Token ID | 340 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
isa | +0.130 |
awa | +0.127 |
favour | +0.127 |
ophon | +0.125 |
swat | +0.118 |
Neg logit contributions | |
DEFENSE | -0.129 |
Vert | -0.116 |
HQ | -0.096 |
Crest | -0.095 |
7601 | -0.095 |
Token | � |
Token ID | 447 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
isa | +0.851 |
favour | +0.822 |
awa | +0.821 |
ophon | +0.808 |
swat | +0.764 |
Neg logit contributions | |
DEFENSE | -0.915 |
Vert | -0.816 |
HQ | -0.683 |
Crest | -0.679 |
Administrator | -0.678 |