Token | evidence |
Token ID | 2370 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
Carbuncle | +0.189 |
certific | +0.188 |
Hungry | +0.185 |
leans | +0.183 |
Telescope | +0.181 |
Neg logit contributions | |
chief | -0.196 |
harness | -0.188 |
lex | -0.184 |
TO | -0.183 |
mins | -0.178 |
Token | Trump |
Token ID | 1301 |
Feature activation | +9.39e-01 |
Pos logit contributions | |
Carbuncle | +0.117 |
certific | +0.117 |
Hungry | +0.115 |
leans | +0.114 |
Telescope | +0.113 |
Neg logit contributions | |
chief | -0.138 |
harness | -0.132 |
lex | -0.129 |
TO | -0.129 |
mins | -0.126 |
Token | requested |
Token ID | 9167 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
certific | +0.571 |
Carbuncle | +0.570 |
Hungry | +0.559 |
leans | +0.559 |
Telescope | +0.552 |
Neg logit contributions | |
chief | -0.707 |
harness | -0.679 |
lex | -0.669 |
TO | -0.664 |
mins | -0.648 |
Token | putting |
Token ID | 5137 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
Carbuncle | +0.408 |
certific | +0.405 |
Hungry | +0.400 |
leans | +0.394 |
Telescope | +0.389 |
Neg logit contributions | |
chief | -0.426 |
harness | -0.410 |
lex | -0.402 |
TO | -0.397 |
mins | -0.388 |
Token | ally |
Token ID | 12525 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
chief | +0.460 |
harness | +0.439 |
lex | +0.434 |
TO | +0.431 |
mins | +0.425 |
Neg logit contributions | |
Carbuncle | -0.356 |
certific | -0.355 |
Hungry | -0.351 |
leans | -0.346 |
Telescope | -0.342 |
Token | in |
Token ID | 287 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
chief | +0.469 |
harness | +0.447 |
lex | +0.440 |
TO | +0.439 |
mins | +0.431 |
Neg logit contributions | |
Carbuncle | -0.403 |
certific | -0.401 |
Hungry | -0.395 |
leans | -0.392 |
Telescope | -0.388 |
Token | charge |
Token ID | 3877 |
Feature activation | +7.17e-01 |
Pos logit contributions | |
certific | +1.880 |
Carbuncle | +1.866 |
Hungry | +1.828 |
clair | +1.815 |
leans | +1.814 |
Neg logit contributions | |
chief | -2.168 |
harness | -2.117 |
lex | -2.054 |
TO | -1.962 |
mins | -1.933 |
Token | of |
Token ID | 286 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
certific | +0.478 |
Carbuncle | +0.478 |
Hungry | +0.468 |
leans | +0.462 |
Telescope | +0.456 |
Neg logit contributions | |
chief | -0.502 |
harness | -0.480 |
lex | -0.471 |
TO | -0.467 |
mins | -0.455 |
Token | Cohen |
Token ID | 17154 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
chief | +0.489 |
harness | +0.467 |
lex | +0.461 |
TO | +0.459 |
mins | +0.451 |
Neg logit contributions | |
Carbuncle | -0.367 |
certific | -0.367 |
Hungry | -0.363 |
leans | -0.357 |
Telescope | -0.351 |
Token | probe |
Token ID | 12774 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
chief | +0.352 |
harness | +0.336 |
lex | +0.332 |
TO | +0.331 |
mins | +0.324 |
Neg logit contributions | |
Carbuncle | -0.267 |
certific | -0.266 |
Hungry | -0.262 |
leans | -0.259 |
Telescope | -0.255 |
Token | : |
Token ID | 25 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
Carbuncle | +0.085 |
certific | +0.085 |
Hungry | +0.083 |
leans | +0.082 |
Telescope | +0.081 |
Neg logit contributions | |
chief | -0.131 |
harness | -0.126 |
lex | -0.125 |
TO | -0.124 |
mins | -0.121 |
Token | evidence |
Token ID | 2370 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
Carbuncle | +0.118 |
certific | +0.118 |
Hungry | +0.116 |
leans | +0.115 |
Telescope | +0.114 |
Neg logit contributions | |
chief | -0.124 |
harness | -0.118 |
lex | -0.116 |
TO | -0.116 |
mins | -0.113 |
Token | Trump |
Token ID | 1301 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Carbuncle | +0.152 |
certific | +0.152 |
Hungry | +0.149 |
leans | +0.147 |
Telescope | +0.146 |
Neg logit contributions | |
chief | -0.182 |
harness | -0.174 |
lex | -0.171 |
TO | -0.170 |
mins | -0.166 |
Token | requested |
Token ID | 9167 |
Feature activation | +7.21e-01 |
Pos logit contributions | |
certific | +0.624 |
Carbuncle | +0.621 |
Hungry | +0.611 |
leans | +0.609 |
Telescope | +0.603 |
Neg logit contributions | |
chief | -0.779 |
harness | -0.746 |
lex | -0.736 |
TO | -0.732 |
mins | -0.713 |
Token | putting |
Token ID | 5137 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
Carbuncle | +0.489 |
certific | +0.485 |
Hungry | +0.480 |
leans | +0.472 |
clair | +0.467 |
Neg logit contributions | |
chief | -0.494 |
harness | -0.476 |
lex | -0.466 |
TO | -0.460 |
mins | -0.448 |
Token | ally |
Token ID | 12525 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
chief | +0.410 |
harness | +0.392 |
lex | +0.387 |
TO | +0.385 |
mins | +0.379 |
Neg logit contributions | |
Carbuncle | -0.317 |
certific | -0.316 |
Hungry | -0.312 |
leans | -0.308 |
Telescope | -0.304 |
Token | in |
Token ID | 287 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
chief | +0.413 |
harness | +0.394 |
lex | +0.388 |
TO | +0.387 |
mins | +0.380 |
Neg logit contributions | |
Carbuncle | -0.358 |
certific | -0.357 |
Hungry | -0.352 |
leans | -0.349 |
Telescope | -0.345 |
Token | charge |
Token ID | 3877 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
certific | +1.817 |
Carbuncle | +1.807 |
Hungry | +1.768 |
leans | +1.743 |
clair | +1.739 |
Neg logit contributions | |
chief | -1.998 |
harness | -1.944 |
lex | -1.895 |
TO | -1.821 |
mins | -1.786 |
Token | of |
Token ID | 286 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
Carbuncle | +0.387 |
certific | +0.387 |
Hungry | +0.379 |
leans | +0.374 |
Telescope | +0.370 |
Neg logit contributions | |
chief | -0.421 |
harness | -0.403 |
lex | -0.396 |
TO | -0.393 |
mins | -0.383 |
Token | Cohen |
Token ID | 17154 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
chief | +0.381 |
harness | +0.364 |
lex | +0.359 |
TO | +0.356 |
mins | +0.350 |
Neg logit contributions | |
Carbuncle | -0.300 |
certific | -0.299 |
Hungry | -0.296 |
leans | -0.291 |
Telescope | -0.287 |
Token | probe |
Token ID | 12774 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
chief | +0.281 |
harness | +0.269 |
lex | +0.265 |
TO | +0.264 |
mins | +0.259 |
Neg logit contributions | |
Carbuncle | -0.224 |
certific | -0.222 |
Hungry | -0.219 |
leans | -0.217 |
Telescope | -0.214 |
Token | : |
Token ID | 25 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
Carbuncle | +0.211 |
certific | +0.210 |
Hungry | +0.206 |
leans | +0.203 |
Telescope | +0.199 |
Neg logit contributions | |
chief | -0.314 |
harness | -0.301 |
lex | -0.297 |
TO | -0.296 |
mins | -0.289 |
Token | , |
Token ID | 11 |
Feature activation | +1.84e-03 |
Pos logit contributions | |
chief | +0.139 |
harness | +0.133 |
lex | +0.131 |
TO | +0.130 |
mins | +0.127 |
Neg logit contributions | |
Carbuncle | -0.123 |
certific | -0.122 |
Hungry | -0.121 |
leans | -0.119 |
Telescope | -0.118 |
Token | 2014 |
Token ID | 1946 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
Carbuncle | +0.001 |
certific | +0.001 |
Hungry | +0.001 |
leans | +0.001 |
Telescope | +0.001 |
Neg logit contributions | |
chief | -0.001 |
harness | -0.001 |
lex | -0.001 |
TO | -0.001 |
mins | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
chief | +0.329 |
harness | +0.314 |
lex | +0.310 |
TO | +0.307 |
mins | +0.301 |
Neg logit contributions | |
Carbuncle | -0.278 |
certific | -0.276 |
Hungry | -0.273 |
leans | -0.269 |
Telescope | -0.266 |
Token | REUTERS |
Token ID | 15862 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
certific | +0.374 |
Carbuncle | +0.373 |
Hungry | +0.366 |
leans | +0.363 |
Telescope | +0.357 |
Neg logit contributions | |
chief | -0.432 |
harness | -0.412 |
lex | -0.408 |
TO | -0.406 |
mins | -0.396 |
Token | / |
Token ID | 14 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
certific | +0.237 |
Carbuncle | +0.236 |
Hungry | +0.231 |
leans | +0.229 |
Telescope | +0.225 |
Neg logit contributions | |
chief | -0.329 |
harness | -0.315 |
lex | -0.312 |
TO | -0.310 |
mins | -0.303 |
Token | En |
Token ID | 4834 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
Carbuncle | +0.748 |
certific | +0.748 |
Hungry | +0.735 |
leans | +0.722 |
Telescope | +0.719 |
Neg logit contributions | |
chief | -0.715 |
harness | -0.672 |
lex | -0.669 |
TO | -0.667 |
mins | -0.645 |
Token | rique |
Token ID | 33865 |
Feature activation | +5.96e-01 |
Pos logit contributions | |
certific | +2.628 |
Carbuncle | +2.594 |
Hungry | +2.576 |
leans | +2.561 |
Telescope | +2.550 |
Neg logit contributions | |
chief | -0.958 |
lex | -0.893 |
harness | -0.848 |
TO | -0.830 |
mins | -0.781 |
Token | de |
Token ID | 390 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
Carbuncle | +0.405 |
certific | +0.405 |
Hungry | +0.398 |
leans | +0.396 |
Telescope | +0.390 |
Neg logit contributions | |
chief | -0.406 |
harness | -0.388 |
lex | -0.382 |
TO | -0.378 |
mins | -0.368 |
Token | la |
Token ID | 8591 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
chief | +0.499 |
harness | +0.475 |
lex | +0.470 |
TO | +0.466 |
mins | +0.457 |
Neg logit contributions | |
Carbuncle | -0.420 |
certific | -0.419 |
Hungry | -0.415 |
leans | -0.405 |
Telescope | -0.404 |
Token | Os |
Token ID | 8834 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
certific | +0.802 |
Carbuncle | +0.801 |
leans | +0.785 |
Hungry | +0.780 |
Telescope | +0.769 |
Neg logit contributions | |
chief | -1.009 |
harness | -0.965 |
lex | -0.950 |
TO | -0.949 |
mins | -0.922 |
Token | a |
Token ID | 64 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
Carbuncle | +0.367 |
certific | +0.366 |
Hungry | +0.359 |
leans | +0.356 |
Telescope | +0.352 |
Neg logit contributions | |
chief | -0.373 |
harness | -0.355 |
lex | -0.351 |
TO | -0.347 |
mins | -0.339 |
Token | listening |
Token ID | 8680 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
chief | +0.027 |
harness | +0.026 |
lex | +0.026 |
TO | +0.025 |
mins | +0.025 |
Neg logit contributions | |
Carbuncle | -0.023 |
certific | -0.023 |
Hungry | -0.023 |
leans | -0.022 |
Telescope | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
chief | +0.396 |
harness | +0.378 |
lex | +0.372 |
TO | +0.369 |
mins | +0.362 |
Neg logit contributions | |
Carbuncle | -0.344 |
certific | -0.343 |
Hungry | -0.339 |
leans | -0.334 |
Telescope | -0.331 |
Token | speeches |
Token ID | 18138 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
Carbuncle | +0.110 |
certific | +0.109 |
Hungry | +0.108 |
leans | +0.106 |
Telescope | +0.105 |
Neg logit contributions | |
chief | -0.126 |
harness | -0.120 |
lex | -0.119 |
TO | -0.118 |
mins | -0.115 |
Token | in |
Token ID | 287 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
Carbuncle | +0.147 |
certific | +0.146 |
Hungry | +0.144 |
leans | +0.142 |
Telescope | +0.140 |
Neg logit contributions | |
chief | -0.174 |
harness | -0.166 |
lex | -0.164 |
TO | -0.163 |
mins | -0.159 |
Token | Tra |
Token ID | 4759 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
Carbuncle | +0.143 |
certific | +0.143 |
Hungry | +0.141 |
leans | +0.139 |
Telescope | +0.137 |
Neg logit contributions | |
chief | -0.178 |
harness | -0.171 |
lex | -0.168 |
TO | -0.167 |
mins | -0.163 |
Token | fal |
Token ID | 42932 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
Carbuncle | +0.177 |
certific | +0.176 |
Hungry | +0.174 |
leans | +0.172 |
Telescope | +0.170 |
Neg logit contributions | |
chief | -0.147 |
harness | -0.139 |
lex | -0.137 |
TO | -0.136 |
mins | -0.133 |
Token | gar |
Token ID | 4563 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
certific | +1.635 |
Carbuncle | +1.616 |
Hungry | +1.589 |
leans | +1.560 |
Telescope | +1.548 |
Neg logit contributions | |
chief | -1.890 |
lex | -1.818 |
harness | -1.795 |
TO | -1.780 |
mins | -1.756 |
Token | Square |
Token ID | 9276 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
Carbuncle | +0.080 |
certific | +0.080 |
Hungry | +0.079 |
leans | +0.078 |
Telescope | +0.076 |
Neg logit contributions | |
chief | -0.119 |
harness | -0.114 |
lex | -0.112 |
TO | -0.112 |
mins | -0.109 |
Token | following |
Token ID | 1708 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
Carbuncle | +0.014 |
certific | +0.014 |
Hungry | +0.014 |
leans | +0.014 |
Telescope | +0.014 |
Neg logit contributions | |
chief | -0.017 |
harness | -0.016 |
lex | -0.016 |
TO | -0.016 |
mins | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
Carbuncle | +0.259 |
certific | +0.259 |
Hungry | +0.254 |
leans | +0.252 |
Telescope | +0.248 |
Neg logit contributions | |
chief | -0.313 |
harness | -0.299 |
lex | -0.295 |
TO | -0.293 |
mins | -0.287 |
Token | march |
Token ID | 9960 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
Carbuncle | +0.366 |
certific | +0.366 |
Hungry | +0.360 |
leans | +0.356 |
Telescope | +0.351 |
Neg logit contributions | |
chief | -0.409 |
harness | -0.391 |
lex | -0.385 |
TO | -0.382 |
mins | -0.373 |
Token | evidence |
Token ID | 2370 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Carbuncle | +0.143 |
certific | +0.142 |
Hungry | +0.140 |
leans | +0.139 |
Telescope | +0.137 |
Neg logit contributions | |
chief | -0.150 |
harness | -0.143 |
lex | -0.141 |
TO | -0.140 |
mins | -0.136 |
Token | Trump |
Token ID | 1301 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
chief | +0.192 |
harness | +0.184 |
lex | +0.182 |
TO | +0.180 |
mins | +0.177 |
Neg logit contributions | |
Carbuncle | -0.152 |
certific | -0.151 |
Hungry | -0.149 |
leans | -0.147 |
Telescope | -0.145 |
Token | requested |
Token ID | 9167 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
Carbuncle | +0.495 |
certific | +0.495 |
Hungry | +0.484 |
leans | +0.484 |
Telescope | +0.477 |
Neg logit contributions | |
chief | -0.613 |
harness | -0.588 |
lex | -0.579 |
TO | -0.575 |
mins | -0.561 |
Token | putting |
Token ID | 5137 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
Carbuncle | +0.389 |
certific | +0.386 |
Hungry | +0.382 |
leans | +0.377 |
clair | +0.372 |
Neg logit contributions | |
chief | -0.408 |
harness | -0.392 |
lex | -0.385 |
TO | -0.381 |
mins | -0.371 |
Token | ally |
Token ID | 12525 |
Feature activation | -7.83e-01 |
Pos logit contributions | |
chief | +0.856 |
harness | +0.822 |
lex | +0.812 |
TO | +0.809 |
mins | +0.801 |
Neg logit contributions | |
certific | -0.421 |
Carbuncle | -0.419 |
Hungry | -0.414 |
leans | -0.404 |
Telescope | -0.400 |
Token | in |
Token ID | 287 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
chief | +0.579 |
harness | +0.551 |
lex | +0.542 |
TO | +0.541 |
mins | +0.532 |
Neg logit contributions | |
Carbuncle | -0.489 |
certific | -0.487 |
Hungry | -0.480 |
leans | -0.476 |
Telescope | -0.471 |
Token | charge |
Token ID | 3877 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
certific | +1.659 |
Carbuncle | +1.653 |
Hungry | +1.617 |
leans | +1.617 |
clair | +1.604 |
Neg logit contributions | |
chief | -1.807 |
harness | -1.761 |
lex | -1.709 |
TO | -1.646 |
mins | -1.619 |
Token | of |
Token ID | 286 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
Carbuncle | +0.445 |
certific | +0.445 |
Hungry | +0.435 |
leans | +0.430 |
Telescope | +0.424 |
Neg logit contributions | |
chief | -0.467 |
harness | -0.446 |
lex | -0.438 |
TO | -0.434 |
mins | -0.423 |
Token | Cohen |
Token ID | 17154 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
chief | +0.731 |
harness | +0.700 |
lex | +0.690 |
TO | +0.687 |
mins | +0.678 |
Neg logit contributions | |
Carbuncle | -0.483 |
certific | -0.482 |
Hungry | -0.478 |
leans | -0.469 |
Telescope | -0.460 |
Token | probe |
Token ID | 12774 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
chief | +0.340 |
harness | +0.324 |
lex | +0.320 |
TO | +0.319 |
mins | +0.313 |
Neg logit contributions | |
Carbuncle | -0.266 |
certific | -0.265 |
Hungry | -0.261 |
leans | -0.258 |
Telescope | -0.254 |
Token | : |
Token ID | 25 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
Carbuncle | +0.030 |
certific | +0.030 |
Hungry | +0.030 |
leans | +0.029 |
Telescope | +0.029 |
Neg logit contributions | |
chief | -0.038 |
harness | -0.036 |
lex | -0.036 |
TO | -0.036 |
mins | -0.035 |
Token | of |
Token ID | 286 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Carbuncle | +0.133 |
certific | +0.133 |
Hungry | +0.131 |
leans | +0.129 |
Telescope | +0.128 |
Neg logit contributions | |
chief | -0.148 |
harness | -0.141 |
lex | -0.139 |
TO | -0.138 |
mins | -0.135 |
Token | books |
Token ID | 3835 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Carbuncle | +0.054 |
certific | +0.054 |
Hungry | +0.052 |
leans | +0.051 |
Telescope | +0.050 |
Neg logit contributions | |
chief | -0.172 |
harness | -0.166 |
lex | -0.165 |
TO | -0.164 |
mins | -0.161 |
Token | , |
Token ID | 11 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
chief | +0.175 |
harness | +0.167 |
lex | +0.165 |
TO | +0.164 |
mins | +0.161 |
Neg logit contributions | |
Carbuncle | -0.154 |
certific | -0.154 |
Hungry | -0.152 |
leans | -0.150 |
Telescope | -0.148 |
Token | which |
Token ID | 543 |
Feature activation | +9.42e-01 |
Pos logit contributions | |
certific | +0.521 |
Carbuncle | +0.519 |
Hungry | +0.507 |
leans | +0.502 |
Telescope | +0.496 |
Neg logit contributions | |
chief | -0.558 |
harness | -0.533 |
lex | -0.522 |
TO | -0.519 |
mins | -0.507 |
Token | you |
Token ID | 345 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
Carbuncle | +0.590 |
certific | +0.585 |
Hungry | +0.575 |
leans | +0.570 |
Telescope | +0.561 |
Neg logit contributions | |
chief | -0.697 |
harness | -0.668 |
lex | -0.658 |
TO | -0.652 |
mins | -0.635 |
Token | might |
Token ID | 1244 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
Carbuncle | +0.371 |
certific | +0.370 |
Hungry | +0.364 |
leans | +0.359 |
Telescope | +0.356 |
Neg logit contributions | |
chief | -0.310 |
harness | -0.294 |
lex | -0.289 |
TO | -0.289 |
mins | -0.280 |
Token | want |
Token ID | 765 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
Carbuncle | +1.378 |
certific | +1.364 |
Hungry | +1.337 |
leans | +1.328 |
clair | +1.316 |
Neg logit contributions | |
chief | -1.764 |
harness | -1.717 |
lex | -1.667 |
TO | -1.648 |
mins | -1.631 |
Token | to |
Token ID | 284 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
Carbuncle | +0.116 |
certific | +0.116 |
Hungry | +0.114 |
leans | +0.113 |
Telescope | +0.111 |
Neg logit contributions | |
chief | -0.129 |
harness | -0.123 |
lex | -0.121 |
TO | -0.120 |
mins | -0.117 |
Token | per |
Token ID | 583 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
Carbuncle | +0.573 |
certific | +0.573 |
Hungry | +0.564 |
leans | +0.559 |
Telescope | +0.550 |
Neg logit contributions | |
chief | -0.577 |
harness | -0.549 |
lex | -0.543 |
TO | -0.536 |
mins | -0.525 |
Token | use |
Token ID | 1904 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
chief | +0.523 |
harness | +0.503 |
lex | +0.494 |
TO | +0.493 |
mins | +0.482 |
Neg logit contributions | |
Carbuncle | -0.508 |
certific | -0.504 |
Hungry | -0.500 |
leans | -0.495 |
Telescope | -0.490 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
certific | +0.940 |
Carbuncle | +0.940 |
Hungry | +0.917 |
leans | +0.915 |
Telescope | +0.896 |
Neg logit contributions | |
chief | -0.993 |
harness | -0.945 |
lex | -0.930 |
TO | -0.921 |
mins | -0.899 |
Token | the |
Token ID | 262 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
chief | +0.093 |
harness | +0.089 |
lex | +0.088 |
TO | +0.087 |
mins | +0.086 |
Neg logit contributions | |
Carbuncle | -0.076 |
certific | -0.075 |
Hungry | -0.074 |
leans | -0.073 |
Telescope | -0.072 |
Token | Q |
Token ID | 1195 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
chief | +0.170 |
harness | +0.162 |
lex | +0.160 |
TO | +0.159 |
mins | +0.156 |
Neg logit contributions | |
Carbuncle | -0.146 |
certific | -0.145 |
Hungry | -0.143 |
leans | -0.141 |
Telescope | -0.139 |
Token | E |
Token ID | 36 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
Carbuncle | +0.300 |
certific | +0.299 |
Hungry | +0.295 |
leans | +0.291 |
Telescope | +0.287 |
Neg logit contributions | |
chief | -0.375 |
harness | -0.359 |
lex | -0.355 |
TO | -0.353 |
mins | -0.344 |
Token | and |
Token ID | 290 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
Carbuncle | +0.164 |
certific | +0.163 |
Hungry | +0.161 |
leans | +0.159 |
Telescope | +0.157 |
Neg logit contributions | |
chief | -0.198 |
harness | -0.189 |
lex | -0.186 |
TO | -0.185 |
mins | -0.181 |
Token | Z |
Token ID | 1168 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
Carbuncle | +0.077 |
certific | +0.077 |
Hungry | +0.075 |
leans | +0.075 |
Telescope | +0.073 |
Neg logit contributions | |
chief | -0.092 |
harness | -0.088 |
lex | -0.087 |
TO | -0.086 |
mins | -0.084 |
Token | IR |
Token ID | 4663 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
certific | +0.306 |
Carbuncle | +0.306 |
Hungry | +0.301 |
leans | +0.297 |
Telescope | +0.294 |
Neg logit contributions | |
chief | -0.329 |
harness | -0.314 |
lex | -0.310 |
TO | -0.308 |
mins | -0.300 |
Token | P |
Token ID | 47 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
Carbuncle | +0.965 |
certific | +0.964 |
Hungry | +0.945 |
leans | +0.938 |
ensis | +0.906 |
Neg logit contributions | |
chief | -2.020 |
harness | -1.953 |
TO | -1.936 |
lex | -1.927 |
mins | -1.881 |
Token | save |
Token ID | 3613 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
Carbuncle | +0.074 |
certific | +0.074 |
Hungry | +0.073 |
leans | +0.072 |
Telescope | +0.071 |
Neg logit contributions | |
chief | -0.090 |
harness | -0.087 |
lex | -0.085 |
TO | -0.085 |
mins | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
Carbuncle | +0.182 |
certific | +0.181 |
Hungry | +0.179 |
leans | +0.176 |
Telescope | +0.174 |
Neg logit contributions | |
chief | -0.212 |
harness | -0.203 |
lex | -0.200 |
TO | -0.198 |
mins | -0.194 |
Token | banker |
Token ID | 33371 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
chief | +0.028 |
harness | +0.027 |
lex | +0.027 |
TO | +0.026 |
mins | +0.026 |
Neg logit contributions | |
Carbuncle | -0.023 |
certific | -0.023 |
Hungry | -0.022 |
leans | -0.022 |
Telescope | -0.022 |
Token | programs |
Token ID | 4056 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
Carbuncle | +0.020 |
certific | +0.019 |
Hungry | +0.019 |
leans | +0.019 |
Telescope | +0.019 |
Neg logit contributions | |
chief | -0.023 |
harness | -0.022 |
lex | -0.022 |
TO | -0.022 |
mins | -0.022 |
Token | Van |
Token ID | 6656 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
chief | +0.130 |
harness | +0.124 |
lex | +0.122 |
TO | +0.121 |
mins | +0.119 |
Neg logit contributions | |
Carbuncle | -0.112 |
certific | -0.111 |
Hungry | -0.110 |
leans | -0.108 |
Telescope | -0.107 |
Token | Go |
Token ID | 1514 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
chief | +0.073 |
harness | +0.069 |
lex | +0.068 |
TO | +0.068 |
mins | +0.066 |
Neg logit contributions | |
Carbuncle | -0.072 |
certific | -0.072 |
Hungry | -0.071 |
leans | -0.070 |
Telescope | -0.069 |
Token | gh |
Token ID | 456 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
Carbuncle | +0.454 |
certific | +0.453 |
Hungry | +0.447 |
leans | +0.442 |
Telescope | +0.440 |
Neg logit contributions | |
chief | -0.273 |
harness | -0.255 |
lex | -0.252 |
TO | -0.247 |
mins | -0.239 |
Token | , |
Token ID | 11 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
chief | +0.069 |
harness | +0.066 |
lex | +0.065 |
TO | +0.064 |
mins | +0.063 |
Neg logit contributions | |
Carbuncle | -0.057 |
certific | -0.057 |
Hungry | -0.056 |
leans | -0.056 |
Telescope | -0.055 |
Token | Ing |
Token ID | 17589 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
chief | +0.162 |
harness | +0.155 |
lex | +0.153 |
TO | +0.152 |
mins | +0.148 |
Neg logit contributions | |
Carbuncle | -0.141 |
certific | -0.141 |
Hungry | -0.139 |
leans | -0.137 |
Telescope | -0.136 |
Token | mar |
Token ID | 3876 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
Carbuncle | +0.007 |
certific | +0.007 |
Hungry | +0.007 |
leans | +0.007 |
Telescope | +0.007 |
Neg logit contributions | |
chief | -0.008 |
harness | -0.008 |
lex | -0.008 |
TO | -0.008 |
mins | -0.007 |
Token | Berg |
Token ID | 24626 |
Feature activation | -9.42e-01 |
Pos logit contributions | |
certific | +1.492 |
Carbuncle | +1.476 |
clair | +1.424 |
Hungry | +1.422 |
leans | +1.421 |
Neg logit contributions | |
chief | -1.466 |
harness | -1.389 |
lex | -1.372 |
TO | -1.358 |
mins | -1.326 |
Token | man |
Token ID | 805 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
chief | +0.618 |
harness | +0.587 |
lex | +0.579 |
TO | +0.576 |
mins | +0.558 |
Neg logit contributions | |
certific | -0.657 |
Carbuncle | -0.657 |
Hungry | -0.652 |
leans | -0.643 |
Telescope | -0.638 |
Token | , |
Token ID | 11 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
Carbuncle | +0.072 |
certific | +0.072 |
Hungry | +0.071 |
leans | +0.070 |
Telescope | +0.069 |
Neg logit contributions | |
chief | -0.082 |
harness | -0.078 |
lex | -0.077 |
TO | -0.077 |
mins | -0.075 |
Token | and |
Token ID | 290 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
chief | +0.217 |
harness | +0.207 |
lex | +0.204 |
TO | +0.203 |
mins | +0.198 |
Neg logit contributions | |
Carbuncle | -0.180 |
certific | -0.179 |
Hungry | -0.177 |
leans | -0.174 |
Telescope | -0.172 |
Token | F |
Token ID | 376 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
chief | +0.244 |
harness | +0.233 |
lex | +0.230 |
TO | +0.228 |
mins | +0.223 |
Neg logit contributions | |
Carbuncle | -0.220 |
certific | -0.219 |
Hungry | -0.216 |
leans | -0.213 |
Telescope | -0.211 |
Token | that |
Token ID | 326 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Carbuncle | +0.063 |
certific | +0.062 |
Hungry | +0.062 |
leans | +0.061 |
Telescope | +0.060 |
Neg logit contributions | |
chief | -0.071 |
harness | -0.068 |
lex | -0.067 |
TO | -0.067 |
mins | -0.065 |
Token | matter |
Token ID | 2300 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
chief | +0.155 |
harness | +0.146 |
lex | +0.144 |
TO | +0.143 |
mins | +0.140 |
Neg logit contributions | |
Carbuncle | -0.190 |
certific | -0.190 |
Hungry | -0.188 |
leans | -0.185 |
Telescope | -0.183 |
Token | . |
Token ID | 13 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
Carbuncle | +0.021 |
certific | +0.021 |
Hungry | +0.021 |
leans | +0.021 |
Telescope | +0.021 |
Neg logit contributions | |
chief | -0.025 |
harness | -0.024 |
lex | -0.023 |
TO | -0.023 |
mins | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -9.99e-02 |
Pos logit contributions | |
certific | +0.184 |
Carbuncle | +0.184 |
Hungry | +0.180 |
leans | +0.179 |
Telescope | +0.176 |
Neg logit contributions | |
chief | -0.228 |
harness | -0.218 |
lex | -0.215 |
TO | -0.214 |
mins | -0.209 |
Token |
|
Token ID | 198 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
chief | +0.076 |
harness | +0.073 |
lex | +0.072 |
TO | +0.071 |
mins | +0.070 |
Neg logit contributions | |
Carbuncle | -0.060 |
certific | -0.060 |
Hungry | -0.059 |
leans | -0.058 |
Telescope | -0.057 |
Token | Fight |
Token ID | 27365 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
chief | +0.407 |
harness | +0.390 |
lex | +0.383 |
TO | +0.379 |
mins | +0.373 |
Neg logit contributions | |
Carbuncle | -0.399 |
certific | -0.396 |
Hungry | -0.394 |
leans | -0.389 |
Telescope | -0.385 |
Token | Back |
Token ID | 5157 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
Carbuncle | +1.700 |
certific | +1.695 |
leans | +1.660 |
clair | +1.637 |
Hungry | +1.636 |
Neg logit contributions | |
chief | -1.217 |
harness | -1.173 |
lex | -1.159 |
TO | -1.134 |
mins | -1.090 |
Token | ! |
Token ID | 0 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
chief | +0.439 |
harness | +0.419 |
lex | +0.412 |
TO | +0.408 |
mins | +0.402 |
Neg logit contributions | |
Carbuncle | -0.419 |
certific | -0.418 |
Hungry | -0.415 |
leans | -0.406 |
Telescope | -0.405 |
Token | Sign |
Token ID | 5865 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
certific | +0.199 |
Carbuncle | +0.199 |
Hungry | +0.195 |
leans | +0.194 |
Telescope | +0.191 |
Neg logit contributions | |
chief | -0.172 |
harness | -0.164 |
lex | -0.161 |
TO | -0.160 |
mins | -0.155 |
Token | up |
Token ID | 510 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
certific | +0.327 |
Carbuncle | +0.326 |
Hungry | +0.319 |
leans | +0.316 |
Telescope | +0.312 |
Neg logit contributions | |
chief | -0.425 |
harness | -0.409 |
lex | -0.401 |
TO | -0.399 |
mins | -0.391 |
Token | for |
Token ID | 329 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
Carbuncle | +0.061 |
certific | +0.061 |
Hungry | +0.060 |
leans | +0.059 |
Telescope | +0.059 |
Neg logit contributions | |
chief | -0.069 |
harness | -0.065 |
lex | -0.064 |
TO | -0.064 |
mins | -0.063 |
Token | spokesman |
Token ID | 6523 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
Carbuncle | +0.208 |
certific | +0.207 |
Hungry | +0.202 |
leans | +0.197 |
Telescope | +0.193 |
Neg logit contributions | |
chief | -0.515 |
harness | -0.497 |
lex | -0.490 |
TO | -0.489 |
mins | -0.479 |
Token | : |
Token ID | 25 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
chief | +0.242 |
harness | +0.231 |
lex | +0.228 |
TO | +0.226 |
mins | +0.222 |
Neg logit contributions | |
Carbuncle | -0.215 |
certific | -0.214 |
Hungry | -0.212 |
leans | -0.209 |
Telescope | -0.207 |
Token | Clinton |
Token ID | 2605 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
certific | +0.673 |
Carbuncle | +0.670 |
Hungry | +0.655 |
leans | +0.644 |
Telescope | +0.636 |
Neg logit contributions | |
chief | -0.712 |
harness | -0.674 |
TO | -0.670 |
lex | -0.669 |
mins | -0.651 |
Token | staff |
Token ID | 3085 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Carbuncle | +0.405 |
certific | +0.402 |
Hungry | +0.395 |
leans | +0.388 |
Telescope | +0.386 |
Neg logit contributions | |
chief | -0.512 |
harness | -0.491 |
lex | -0.483 |
TO | -0.481 |
mins | -0.468 |
Token | are |
Token ID | 389 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
chief | +0.008 |
harness | +0.007 |
lex | +0.007 |
TO | +0.007 |
mins | +0.007 |
Neg logit contributions | |
Carbuncle | -0.006 |
certific | -0.006 |
Hungry | -0.006 |
leans | -0.006 |
Telescope | -0.006 |
Token | ' |
Token ID | 705 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
chief | +0.009 |
harness | +0.009 |
lex | +0.009 |
TO | +0.009 |
mins | +0.008 |
Neg logit contributions | |
Carbuncle | -0.008 |
certific | -0.008 |
Hungry | -0.008 |
leans | -0.008 |
Telescope | -0.008 |
Token | big |
Token ID | 14261 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
Carbuncle | +1.755 |
certific | +1.736 |
Hungry | +1.694 |
Telescope | +1.687 |
ayne | +1.659 |
Neg logit contributions | |
chief | -1.227 |
harness | -1.118 |
TO | -1.112 |
lex | -1.104 |
mins | -1.045 |
Token | gest |
Token ID | 3495 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Carbuncle | +0.296 |
certific | +0.294 |
Hungry | +0.289 |
leans | +0.285 |
Telescope | +0.282 |
Neg logit contributions | |
chief | -0.395 |
harness | -0.380 |
lex | -0.374 |
TO | -0.373 |
mins | -0.363 |
Token | a |
Token ID | 257 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
Carbuncle | +0.109 |
certific | +0.109 |
Hungry | +0.107 |
leans | +0.106 |
Telescope | +0.105 |
Neg logit contributions | |
chief | -0.130 |
harness | -0.125 |
lex | -0.123 |
TO | -0.122 |
mins | -0.119 |
Token | -- |
Token ID | 438 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
chief | +0.271 |
harness | +0.256 |
lex | +0.251 |
TO | +0.250 |
mins | +0.245 |
Neg logit contributions | |
Carbuncle | -0.312 |
certific | -0.311 |
Hungry | -0.308 |
leans | -0.305 |
Telescope | -0.301 |
Token | holes |
Token ID | 28439 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
Carbuncle | +0.042 |
certific | +0.042 |
Hungry | +0.042 |
leans | +0.041 |
Telescope | +0.041 |
Neg logit contributions | |
chief | -0.047 |
harness | -0.045 |
lex | -0.045 |
TO | -0.044 |
mins | -0.043 |
Token | week |
Token ID | 1285 |
Feature activation | +1.71e-03 |
Pos logit contributions | |
chief | +0.084 |
harness | +0.080 |
lex | +0.079 |
TO | +0.078 |
mins | +0.077 |
Neg logit contributions | |
Carbuncle | -0.085 |
certific | -0.085 |
Hungry | -0.084 |
leans | -0.083 |
Telescope | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Carbuncle | +0.001 |
certific | +0.001 |
Hungry | +0.001 |
leans | +0.001 |
Telescope | +0.001 |
Neg logit contributions | |
chief | -0.001 |
harness | -0.001 |
lex | -0.001 |
TO | -0.001 |
mins | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
chief | +0.075 |
harness | +0.072 |
lex | +0.071 |
TO | +0.070 |
mins | +0.069 |
Neg logit contributions | |
Carbuncle | -0.062 |
certific | -0.062 |
Hungry | -0.061 |
leans | -0.060 |
Telescope | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
chief | +0.060 |
harness | +0.057 |
lex | +0.056 |
TO | +0.056 |
mins | +0.055 |
Neg logit contributions | |
Carbuncle | -0.047 |
certific | -0.047 |
Hungry | -0.046 |
leans | -0.046 |
Telescope | -0.045 |
Token | Coach |
Token ID | 40677 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
Carbuncle | +0.079 |
certific | +0.079 |
Hungry | +0.077 |
leans | +0.076 |
Telescope | +0.076 |
Neg logit contributions | |
chief | -0.084 |
harness | -0.080 |
lex | -0.079 |
TO | -0.078 |
mins | -0.077 |
Token | Jay |
Token ID | 9180 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
Carbuncle | +0.365 |
certific | +0.363 |
Hungry | +0.356 |
leans | +0.352 |
Telescope | +0.348 |
Neg logit contributions | |
chief | -0.458 |
harness | -0.438 |
lex | -0.433 |
TO | -0.428 |
mins | -0.420 |
Token | Gr |
Token ID | 1902 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
Carbuncle | +1.849 |
certific | +1.847 |
Hungry | +1.787 |
leans | +1.778 |
ensis | +1.778 |
Neg logit contributions | |
chief | -1.055 |
harness | -0.970 |
lex | -0.964 |
TO | -0.939 |
mins | -0.911 |
Token | uden |
Token ID | 44452 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
chief | +0.257 |
harness | +0.249 |
lex | +0.246 |
TO | +0.244 |
mins | +0.241 |
Neg logit contributions | |
Carbuncle | -0.107 |
certific | -0.106 |
Hungry | -0.104 |
leans | -0.102 |
Telescope | -0.100 |
Token | praised |
Token ID | 15342 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
chief | +0.084 |
harness | +0.080 |
lex | +0.079 |
TO | +0.078 |
mins | +0.077 |
Neg logit contributions | |
Carbuncle | -0.066 |
certific | -0.066 |
Hungry | -0.065 |
leans | -0.064 |
Telescope | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
Carbuncle | +0.118 |
certific | +0.118 |
Hungry | +0.116 |
leans | +0.115 |
Telescope | +0.113 |
Neg logit contributions | |
chief | -0.146 |
harness | -0.140 |
lex | -0.138 |
TO | -0.137 |
mins | -0.134 |
Token | tempo |
Token ID | 28691 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
chief | +0.015 |
harness | +0.015 |
lex | +0.014 |
TO | +0.014 |
mins | +0.014 |
Neg logit contributions | |
Carbuncle | -0.014 |
certific | -0.014 |
Hungry | -0.014 |
leans | -0.014 |
Telescope | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
chief | +0.060 |
harness | +0.058 |
lex | +0.057 |
TO | +0.056 |
mins | +0.055 |
Neg logit contributions | |
Carbuncle | -0.048 |
certific | -0.047 |
Hungry | -0.047 |
leans | -0.046 |
Telescope | -0.046 |
Token | Last |
Token ID | 5956 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
chief | +0.184 |
harness | +0.176 |
lex | +0.173 |
TO | +0.171 |
mins | +0.168 |
Neg logit contributions | |
Carbuncle | -0.167 |
certific | -0.167 |
Hungry | -0.165 |
leans | -0.163 |
Telescope | -0.160 |
Token | month |
Token ID | 1227 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
chief | +0.205 |
harness | +0.196 |
lex | +0.194 |
TO | +0.192 |
mins | +0.188 |
Neg logit contributions | |
Carbuncle | -0.158 |
certific | -0.158 |
Hungry | -0.155 |
leans | -0.153 |
Telescope | -0.151 |
Token | , |
Token ID | 11 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
chief | +0.144 |
harness | +0.137 |
lex | +0.135 |
TO | +0.134 |
mins | +0.131 |
Neg logit contributions | |
Carbuncle | -0.126 |
certific | -0.125 |
Hungry | -0.124 |
leans | -0.122 |
Telescope | -0.120 |
Token | Supervisor |
Token ID | 45673 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
chief | +0.272 |
harness | +0.260 |
lex | +0.257 |
TO | +0.255 |
mins | +0.250 |
Neg logit contributions | |
Carbuncle | -0.220 |
certific | -0.220 |
Hungry | -0.217 |
leans | -0.214 |
Telescope | -0.211 |
Token | Aaron |
Token ID | 12139 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
chief | +0.209 |
harness | +0.200 |
lex | +0.197 |
TO | +0.195 |
mins | +0.191 |
Neg logit contributions | |
Carbuncle | -0.202 |
certific | -0.201 |
Hungry | -0.199 |
leans | -0.196 |
Telescope | -0.194 |
Token | Pes |
Token ID | 38434 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
certific | +0.956 |
Carbuncle | +0.909 |
Hungry | +0.896 |
leans | +0.887 |
Telescope | +0.866 |
Neg logit contributions | |
chief | -1.920 |
harness | -1.847 |
lex | -1.824 |
TO | -1.806 |
mins | -1.774 |
Token | kin |
Token ID | 5116 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Carbuncle | +0.370 |
certific | +0.369 |
Hungry | +0.364 |
leans | +0.359 |
Telescope | +0.356 |
Neg logit contributions | |
chief | -0.314 |
harness | -0.296 |
lex | -0.293 |
TO | -0.289 |
mins | -0.282 |
Token | called |
Token ID | 1444 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
Carbuncle | +0.087 |
certific | +0.087 |
Hungry | +0.086 |
leans | +0.085 |
Telescope | +0.084 |
Neg logit contributions | |
chief | -0.098 |
harness | -0.094 |
lex | -0.093 |
TO | -0.092 |
mins | -0.090 |
Token | 911 |
Token ID | 16679 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
chief | +0.030 |
harness | +0.029 |
lex | +0.028 |
TO | +0.028 |
mins | +0.027 |
Neg logit contributions | |
Carbuncle | -0.026 |
certific | -0.026 |
Hungry | -0.026 |
leans | -0.025 |
Telescope | -0.025 |
Token | because |
Token ID | 780 |
Feature activation | +5.75e-02 |
Pos logit contributions | |
chief | +0.075 |
harness | +0.072 |
lex | +0.071 |
TO | +0.070 |
mins | +0.069 |
Neg logit contributions | |
Carbuncle | -0.065 |
certific | -0.065 |
Hungry | -0.064 |
leans | -0.063 |
Telescope | -0.062 |
Token | spokesman |
Token ID | 6523 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
Carbuncle | +0.315 |
certific | +0.314 |
Hungry | +0.305 |
leans | +0.298 |
Telescope | +0.292 |
Neg logit contributions | |
chief | -0.759 |
harness | -0.732 |
TO | -0.721 |
lex | -0.721 |
mins | -0.703 |
Token | : |
Token ID | 25 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
chief | +0.373 |
harness | +0.356 |
lex | +0.352 |
TO | +0.349 |
mins | +0.343 |
Neg logit contributions | |
certific | -0.329 |
Carbuncle | -0.328 |
Hungry | -0.326 |
leans | -0.320 |
Telescope | -0.317 |
Token | Clinton |
Token ID | 2605 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
certific | +0.533 |
Carbuncle | +0.531 |
Hungry | +0.520 |
leans | +0.512 |
Telescope | +0.505 |
Neg logit contributions | |
chief | -0.573 |
harness | -0.543 |
lex | -0.538 |
TO | -0.538 |
mins | -0.524 |
Token | staff |
Token ID | 3085 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
Carbuncle | +0.423 |
certific | +0.420 |
Hungry | +0.412 |
leans | +0.406 |
Telescope | +0.403 |
Neg logit contributions | |
chief | -0.530 |
harness | -0.508 |
lex | -0.500 |
TO | -0.498 |
mins | -0.485 |
Token | are |
Token ID | 389 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
Carbuncle | +0.051 |
certific | +0.051 |
Hungry | +0.050 |
leans | +0.049 |
Telescope | +0.049 |
Neg logit contributions | |
chief | -0.062 |
harness | -0.060 |
lex | -0.059 |
TO | -0.058 |
mins | -0.057 |
Token | ' |
Token ID | 705 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Carbuncle | +0.070 |
certific | +0.070 |
Hungry | +0.069 |
leans | +0.068 |
Telescope | +0.067 |
Neg logit contributions | |
chief | -0.078 |
harness | -0.074 |
lex | -0.073 |
TO | -0.072 |
mins | -0.071 |
Token | big |
Token ID | 14261 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
Carbuncle | +1.687 |
certific | +1.672 |
Hungry | +1.629 |
Telescope | +1.623 |
ayne | +1.594 |
Neg logit contributions | |
chief | -1.182 |
harness | -1.082 |
TO | -1.073 |
lex | -1.064 |
mins | -1.012 |
Token | gest |
Token ID | 3495 |
Feature activation | +7.24e-02 |
Pos logit contributions | |
Carbuncle | +0.219 |
certific | +0.218 |
Hungry | +0.213 |
leans | +0.210 |
Telescope | +0.208 |
Neg logit contributions | |
chief | -0.320 |
harness | -0.307 |
lex | -0.303 |
TO | -0.302 |
mins | -0.295 |
Token | a |
Token ID | 257 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
Carbuncle | +0.045 |
certific | +0.045 |
Hungry | +0.044 |
leans | +0.043 |
Telescope | +0.043 |
Neg logit contributions | |
chief | -0.054 |
harness | -0.052 |
lex | -0.051 |
TO | -0.051 |
mins | -0.050 |
Token | -- |
Token ID | 438 |
Feature activation | -6.06e-02 |
Pos logit contributions | |
chief | +0.321 |
harness | +0.304 |
lex | +0.298 |
TO | +0.297 |
mins | +0.291 |
Neg logit contributions | |
Carbuncle | -0.354 |
certific | -0.353 |
Hungry | -0.349 |
leans | -0.345 |
Telescope | -0.341 |
Token | holes |
Token ID | 28439 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
chief | +0.044 |
harness | +0.042 |
lex | +0.041 |
TO | +0.041 |
mins | +0.040 |
Neg logit contributions | |
Carbuncle | -0.039 |
certific | -0.039 |
Hungry | -0.038 |
leans | -0.037 |
Telescope | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Carbuncle | +0.048 |
certific | +0.048 |
Hungry | +0.047 |
leans | +0.047 |
Telescope | +0.046 |
Neg logit contributions | |
chief | -0.060 |
harness | -0.058 |
lex | -0.057 |
TO | -0.057 |
mins | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
chief | +0.097 |
harness | +0.093 |
lex | +0.091 |
TO | +0.091 |
mins | +0.089 |
Neg logit contributions | |
Carbuncle | -0.076 |
certific | -0.076 |
Hungry | -0.075 |
leans | -0.074 |
Telescope | -0.073 |
Token | Travel |
Token ID | 33074 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
chief | +0.385 |
harness | +0.368 |
lex | +0.361 |
TO | +0.357 |
mins | +0.352 |
Neg logit contributions | |
Carbuncle | -0.384 |
certific | -0.382 |
Hungry | -0.378 |
leans | -0.375 |
Telescope | -0.369 |
Token | With |
Token ID | 2080 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
certific | +0.912 |
Carbuncle | +0.899 |
leans | +0.881 |
Hungry | +0.878 |
Telescope | +0.869 |
Neg logit contributions | |
chief | -0.969 |
harness | -0.923 |
lex | -0.915 |
TO | -0.903 |
mins | -0.873 |
Token | The |
Token ID | 383 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
certific | +0.332 |
Carbuncle | +0.331 |
Hungry | +0.326 |
leans | +0.325 |
Telescope | +0.319 |
Neg logit contributions | |
chief | -0.309 |
harness | -0.294 |
lex | -0.289 |
TO | -0.287 |
mins | -0.279 |
Token | Nation |
Token ID | 8741 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
certific | +0.179 |
Carbuncle | +0.178 |
Hungry | +0.173 |
leans | +0.172 |
Telescope | +0.168 |
Neg logit contributions | |
chief | -0.347 |
harness | -0.334 |
lex | -0.330 |
TO | -0.329 |
mins | -0.322 |
Token | Be |
Token ID | 1355 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
certific | +1.304 |
Carbuncle | +1.277 |
Hungry | +1.250 |
leans | +1.250 |
clair | +1.238 |
Neg logit contributions | |
chief | -1.497 |
harness | -1.448 |
lex | -1.406 |
TO | -1.400 |
mins | -1.367 |
Token | the |
Token ID | 262 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
certific | +0.370 |
Carbuncle | +0.369 |
Hungry | +0.363 |
leans | +0.358 |
Telescope | +0.352 |
Neg logit contributions | |
chief | -0.458 |
harness | -0.440 |
lex | -0.433 |
TO | -0.429 |
mins | -0.420 |
Token | first |
Token ID | 717 |
Feature activation | -9.12e-01 |
Pos logit contributions | |
Carbuncle | +0.110 |
certific | +0.110 |
Hungry | +0.108 |
leans | +0.107 |
Telescope | +0.106 |
Neg logit contributions | |
chief | -0.107 |
harness | -0.102 |
lex | -0.100 |
TO | -0.099 |
mins | -0.097 |
Token | to |
Token ID | 284 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
chief | +0.679 |
harness | +0.643 |
lex | +0.637 |
TO | +0.631 |
mins | +0.624 |
Neg logit contributions | |
certific | -0.562 |
Carbuncle | -0.562 |
Hungry | -0.558 |
leans | -0.545 |
Telescope | -0.541 |
Token | hear |
Token ID | 3285 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
chief | +0.066 |
harness | +0.063 |
lex | +0.062 |
TO | +0.061 |
mins | +0.060 |
Neg logit contributions | |
Carbuncle | -0.077 |
certific | -0.077 |
Hungry | -0.076 |
leans | -0.075 |
Telescope | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
Carbuncle | +0.099 |
certific | +0.099 |
Hungry | +0.097 |
leans | +0.096 |
Telescope | +0.095 |
Neg logit contributions | |
chief | -0.121 |
harness | -0.116 |
lex | -0.114 |
TO | -0.113 |
mins | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
chief | +0.044 |
harness | +0.042 |
lex | +0.042 |
TO | +0.042 |
mins | +0.041 |
Neg logit contributions | |
Carbuncle | -0.035 |
certific | -0.035 |
Hungry | -0.034 |
leans | -0.034 |
Telescope | -0.034 |
Token | O |
Token ID | 46 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
Carbuncle | +0.144 |
certific | +0.144 |
Hungry | +0.142 |
leans | +0.140 |
Telescope | +0.138 |
Neg logit contributions | |
chief | -0.158 |
harness | -0.150 |
lex | -0.148 |
TO | -0.147 |
mins | -0.144 |
Token | tto |
Token ID | 33955 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
chief | +0.310 |
harness | +0.297 |
lex | +0.291 |
TO | +0.289 |
mins | +0.283 |
Neg logit contributions | |
Carbuncle | -0.277 |
certific | -0.277 |
Hungry | -0.274 |
leans | -0.270 |
Telescope | -0.266 |
Token | Warm |
Token ID | 25692 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
chief | +0.102 |
harness | +0.098 |
lex | +0.096 |
TO | +0.096 |
mins | +0.093 |
Neg logit contributions | |
Carbuncle | -0.095 |
certific | -0.094 |
Hungry | -0.093 |
leans | -0.092 |
Telescope | -0.091 |
Token | b |
Token ID | 65 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
Carbuncle | +0.023 |
certific | +0.023 |
Hungry | +0.023 |
leans | +0.022 |
Telescope | +0.022 |
Neg logit contributions | |
chief | -0.034 |
harness | -0.033 |
lex | -0.033 |
TO | -0.032 |
mins | -0.032 |
Token | ier |
Token ID | 959 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
certific | +1.408 |
Carbuncle | +1.399 |
Hungry | +1.356 |
leans | +1.350 |
Telescope | +1.330 |
Neg logit contributions | |
chief | -1.387 |
harness | -1.312 |
lex | -1.292 |
TO | -1.285 |
mins | -1.249 |
Token | , |
Token ID | 11 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
chief | +0.109 |
harness | +0.104 |
lex | +0.103 |
TO | +0.102 |
mins | +0.100 |
Neg logit contributions | |
Carbuncle | -0.093 |
certific | -0.093 |
Hungry | -0.092 |
leans | -0.090 |
Telescope | -0.089 |
Token | a |
Token ID | 257 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
Carbuncle | +0.164 |
certific | +0.164 |
Hungry | +0.161 |
leans | +0.160 |
Telescope | +0.158 |
Neg logit contributions | |
chief | -0.161 |
harness | -0.154 |
lex | -0.151 |
TO | -0.150 |
mins | -0.146 |
Token | 22 |
Token ID | 2534 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
Carbuncle | +0.149 |
certific | +0.149 |
Hungry | +0.147 |
leans | +0.145 |
Telescope | +0.143 |
Neg logit contributions | |
chief | -0.159 |
harness | -0.152 |
lex | -0.149 |
TO | -0.148 |
mins | -0.144 |
Token | - |
Token ID | 12 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
chief | +0.032 |
harness | +0.030 |
lex | +0.030 |
TO | +0.030 |
mins | +0.029 |
Neg logit contributions | |
Carbuncle | -0.027 |
certific | -0.026 |
Hungry | -0.026 |
leans | -0.026 |
Telescope | -0.025 |
Token | spokesman |
Token ID | 6523 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
Carbuncle | +0.324 |
certific | +0.323 |
Hungry | +0.314 |
leans | +0.308 |
Telescope | +0.303 |
Neg logit contributions | |
chief | -0.713 |
harness | -0.688 |
lex | -0.677 |
TO | -0.677 |
mins | -0.659 |
Token | : |
Token ID | 25 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
chief | +0.282 |
harness | +0.269 |
lex | +0.266 |
TO | +0.264 |
mins | +0.259 |
Neg logit contributions | |
certific | -0.250 |
Carbuncle | -0.250 |
Hungry | -0.248 |
leans | -0.244 |
Telescope | -0.242 |
Token | Clinton |
Token ID | 2605 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
certific | +0.590 |
Carbuncle | +0.588 |
Hungry | +0.574 |
leans | +0.566 |
Telescope | +0.559 |
Neg logit contributions | |
chief | -0.630 |
harness | -0.597 |
lex | -0.592 |
TO | -0.591 |
mins | -0.576 |
Token | staff |
Token ID | 3085 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
Carbuncle | +0.486 |
certific | +0.481 |
Hungry | +0.472 |
leans | +0.464 |
Telescope | +0.462 |
Neg logit contributions | |
chief | -0.601 |
harness | -0.576 |
lex | -0.567 |
TO | -0.565 |
mins | -0.549 |
Token | are |
Token ID | 389 |
Feature activation | -9.01e-02 |
Pos logit contributions | |
Carbuncle | +0.021 |
certific | +0.021 |
Hungry | +0.021 |
leans | +0.020 |
Telescope | +0.020 |
Neg logit contributions | |
chief | -0.024 |
harness | -0.023 |
lex | -0.023 |
TO | -0.023 |
mins | -0.022 |
Token | ' |
Token ID | 705 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
chief | +0.065 |
harness | +0.062 |
lex | +0.061 |
TO | +0.061 |
mins | +0.059 |
Neg logit contributions | |
Carbuncle | -0.058 |
certific | -0.058 |
Hungry | -0.057 |
leans | -0.056 |
Telescope | -0.056 |
Token | big |
Token ID | 14261 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
Carbuncle | +1.661 |
certific | +1.642 |
Hungry | +1.605 |
Telescope | +1.599 |
ayne | +1.571 |
Neg logit contributions | |
chief | -1.161 |
harness | -1.058 |
lex | -1.054 |
TO | -1.051 |
mins | -0.995 |
Token | gest |
Token ID | 3495 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
Carbuncle | +0.272 |
certific | +0.271 |
Hungry | +0.266 |
leans | +0.263 |
Telescope | +0.260 |
Neg logit contributions | |
chief | -0.336 |
harness | -0.322 |
lex | -0.317 |
TO | -0.317 |
mins | -0.308 |
Token | a |
Token ID | 257 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
Carbuncle | +0.116 |
certific | +0.116 |
Hungry | +0.114 |
leans | +0.112 |
Telescope | +0.111 |
Neg logit contributions | |
chief | -0.133 |
harness | -0.127 |
lex | -0.125 |
TO | -0.124 |
mins | -0.121 |
Token | -- |
Token ID | 438 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
chief | +0.232 |
harness | +0.222 |
lex | +0.218 |
TO | +0.217 |
mins | +0.212 |
Neg logit contributions | |
Carbuncle | -0.212 |
certific | -0.211 |
Hungry | -0.208 |
leans | -0.205 |
Telescope | -0.203 |
Token | holes |
Token ID | 28439 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
chief | +0.044 |
harness | +0.042 |
lex | +0.042 |
TO | +0.041 |
mins | +0.040 |
Neg logit contributions | |
Carbuncle | -0.039 |
certific | -0.039 |
Hungry | -0.039 |
leans | -0.038 |
Telescope | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +5.43e-01 |
Pos logit contributions | |
Carbuncle | +0.005 |
certific | +0.005 |
Hungry | +0.005 |
leans | +0.004 |
Telescope | +0.004 |
Neg logit contributions | |
chief | -0.005 |
harness | -0.005 |
lex | -0.005 |
TO | -0.005 |
mins | -0.005 |
Token | ( |
Token ID | 357 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
certific | +0.362 |
Carbuncle | +0.360 |
Hungry | +0.354 |
leans | +0.351 |
Telescope | +0.346 |
Neg logit contributions | |
chief | -0.380 |
harness | -0.362 |
lex | -0.357 |
TO | -0.355 |
mins | -0.346 |
Token | Photo |
Token ID | 6191 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
certific | +0.555 |
Carbuncle | +0.554 |
Hungry | +0.543 |
leans | +0.535 |
Telescope | +0.530 |
Neg logit contributions | |
chief | -0.650 |
harness | -0.618 |
lex | -0.611 |
TO | -0.610 |
mins | -0.594 |
Token | : |
Token ID | 25 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
Carbuncle | +0.097 |
certific | +0.097 |
Hungry | +0.096 |
leans | +0.094 |
Telescope | +0.093 |
Neg logit contributions | |
chief | -0.107 |
harness | -0.102 |
lex | -0.100 |
TO | -0.099 |
mins | -0.097 |
Token | Kare |
Token ID | 49396 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
certific | +0.267 |
Carbuncle | +0.267 |
Hungry | +0.262 |
leans | +0.259 |
Telescope | +0.256 |
Neg logit contributions | |
chief | -0.286 |
harness | -0.273 |
lex | -0.269 |
TO | -0.267 |
mins | -0.261 |
Token | em |
Token ID | 368 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
chief | +0.013 |
harness | +0.012 |
lex | +0.012 |
TO | +0.012 |
mins | +0.012 |
Neg logit contributions | |
Carbuncle | -0.013 |
certific | -0.013 |
Hungry | -0.013 |
leans | -0.013 |
Telescope | -0.013 |
Token | El |
Token ID | 2574 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
certific | +1.356 |
leans | +1.330 |
Carbuncle | +1.315 |
Hungry | +1.300 |
clair | +1.288 |
Neg logit contributions | |
chief | -1.394 |
harness | -1.311 |
lex | -1.304 |
TO | -1.282 |
mins | -1.255 |
Token | g |
Token ID | 70 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
Carbuncle | +0.935 |
certific | +0.933 |
Hungry | +0.915 |
leans | +0.910 |
Telescope | +0.910 |
Neg logit contributions | |
chief | -0.654 |
harness | -0.612 |
lex | -0.601 |
TO | -0.594 |
mins | -0.578 |
Token | azz |
Token ID | 8101 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
certific | +0.337 |
Carbuncle | +0.336 |
Hungry | +0.330 |
leans | +0.324 |
Telescope | +0.321 |
Neg logit contributions | |
chief | -0.473 |
harness | -0.453 |
lex | -0.450 |
TO | -0.445 |
mins | -0.436 |
Token | ar |
Token ID | 283 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
Carbuncle | +0.021 |
certific | +0.021 |
Hungry | +0.021 |
leans | +0.020 |
Telescope | +0.020 |
Neg logit contributions | |
chief | -0.020 |
harness | -0.019 |
lex | -0.019 |
TO | -0.018 |
mins | -0.018 |
Token | ) |
Token ID | 8 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
Carbuncle | +0.220 |
certific | +0.220 |
Hungry | +0.216 |
leans | +0.213 |
Telescope | +0.210 |
Neg logit contributions | |
chief | -0.251 |
harness | -0.240 |
lex | -0.237 |
TO | -0.235 |
mins | -0.230 |
Token | BBC |
Token ID | 7823 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
chief | +0.184 |
harness | +0.175 |
lex | +0.173 |
TO | +0.172 |
mins | +0.168 |
Neg logit contributions | |
Carbuncle | -0.156 |
certific | -0.155 |
Hungry | -0.153 |
leans | -0.151 |
Telescope | -0.149 |
Token | 's |
Token ID | 338 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
Carbuncle | +0.098 |
certific | +0.098 |
Hungry | +0.096 |
leans | +0.095 |
Telescope | +0.093 |
Neg logit contributions | |
chief | -0.123 |
harness | -0.118 |
lex | -0.116 |
TO | -0.116 |
mins | -0.113 |
Token | Im |
Token ID | 1846 |
Feature activation | -8.03e-02 |
Pos logit contributions | |
Carbuncle | +0.026 |
certific | +0.026 |
Hungry | +0.025 |
leans | +0.025 |
Telescope | +0.025 |
Neg logit contributions | |
chief | -0.028 |
harness | -0.026 |
lex | -0.026 |
TO | -0.026 |
mins | -0.025 |
Token | ogen |
Token ID | 6644 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
chief | +0.020 |
harness | +0.017 |
lex | +0.017 |
TO | +0.016 |
mins | +0.015 |
Neg logit contributions | |
Carbuncle | -0.090 |
certific | -0.089 |
Hungry | -0.089 |
leans | -0.088 |
Telescope | -0.087 |
Token | F |
Token ID | 376 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
Carbuncle | +0.172 |
certific | +0.172 |
Hungry | +0.168 |
leans | +0.166 |
Telescope | +0.164 |
Neg logit contributions | |
chief | -0.232 |
harness | -0.222 |
lex | -0.219 |
TO | -0.218 |
mins | -0.213 |
Token | oul |
Token ID | 2852 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
Carbuncle | +0.562 |
certific | +0.562 |
Hungry | +0.550 |
leans | +0.545 |
Telescope | +0.535 |
Neg logit contributions | |
chief | -0.645 |
harness | -0.616 |
lex | -0.610 |
TO | -0.608 |
mins | -0.588 |
Token | kes |
Token ID | 5209 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Carbuncle | +0.982 |
certific | +0.970 |
Hungry | +0.942 |
leans | +0.928 |
Telescope | +0.922 |
Neg logit contributions | |
chief | -1.707 |
harness | -1.640 |
lex | -1.609 |
TO | -1.602 |
mins | -1.578 |
Token | , |
Token ID | 11 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
Carbuncle | +0.125 |
certific | +0.124 |
Hungry | +0.122 |
leans | +0.121 |
Telescope | +0.119 |
Neg logit contributions | |
chief | -0.142 |
harness | -0.136 |
lex | -0.134 |
TO | -0.133 |
mins | -0.130 |
Token | in |
Token ID | 287 |
Feature activation | +9.01e-02 |
Pos logit contributions | |
chief | +0.042 |
harness | +0.040 |
lex | +0.039 |
TO | +0.039 |
mins | +0.038 |
Neg logit contributions | |
Carbuncle | -0.035 |
certific | -0.035 |
Hungry | -0.034 |
leans | -0.034 |
Telescope | -0.033 |
Token | Geneva |
Token ID | 20552 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
Carbuncle | +0.052 |
certific | +0.052 |
Hungry | +0.051 |
leans | +0.050 |
Telescope | +0.050 |
Neg logit contributions | |
chief | -0.071 |
harness | -0.068 |
lex | -0.067 |
TO | -0.067 |
mins | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
Carbuncle | +0.047 |
certific | +0.047 |
Hungry | +0.046 |
leans | +0.046 |
Telescope | +0.045 |
Neg logit contributions | |
chief | -0.053 |
harness | -0.051 |
lex | -0.050 |
TO | -0.050 |
mins | -0.049 |
Token | of |
Token ID | 286 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
Carbuncle | +0.073 |
certific | +0.073 |
Hungry | +0.072 |
leans | +0.071 |
Telescope | +0.070 |
Neg logit contributions | |
chief | -0.080 |
harness | -0.076 |
lex | -0.075 |
TO | -0.074 |
mins | -0.073 |
Token | cerebral |
Token ID | 31169 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
Carbuncle | +0.351 |
certific | +0.349 |
Hungry | +0.344 |
leans | +0.340 |
Telescope | +0.336 |
Neg logit contributions | |
chief | -0.420 |
harness | -0.403 |
lex | -0.396 |
TO | -0.393 |
mins | -0.384 |
Token | ha |
Token ID | 387 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
Carbuncle | +0.035 |
certific | +0.034 |
Hungry | +0.034 |
leans | +0.034 |
Telescope | +0.033 |
Neg logit contributions | |
chief | -0.028 |
harness | -0.027 |
lex | -0.026 |
TO | -0.026 |
mins | -0.025 |
Token | em |
Token ID | 368 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
Carbuncle | +0.035 |
certific | +0.035 |
Hungry | +0.035 |
leans | +0.034 |
Telescope | +0.034 |
Neg logit contributions | |
chief | -0.034 |
harness | -0.032 |
lex | -0.032 |
TO | -0.032 |
mins | -0.031 |
Token | or |
Token ID | 273 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
certific | +0.458 |
Carbuncle | +0.458 |
Hungry | +0.449 |
leans | +0.443 |
Telescope | +0.435 |
Neg logit contributions | |
chief | -0.624 |
harness | -0.596 |
lex | -0.593 |
TO | -0.586 |
mins | -0.576 |
Token | rh |
Token ID | 17179 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
certific | +0.634 |
Carbuncle | +0.631 |
Hungry | +0.618 |
leans | +0.609 |
clair | +0.604 |
Neg logit contributions | |
chief | -0.723 |
harness | -0.687 |
lex | -0.682 |
TO | -0.669 |
mins | -0.656 |
Token | age |
Token ID | 496 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
certific | +1.338 |
Carbuncle | +1.318 |
Hungry | +1.299 |
Telescope | +1.252 |
leans | +1.250 |
Neg logit contributions | |
chief | -1.348 |
harness | -1.278 |
lex | -1.277 |
TO | -1.253 |
mins | -1.235 |
Token | ," |
Token ID | 553 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
Carbuncle | +0.022 |
certific | +0.022 |
Hungry | +0.022 |
leans | +0.021 |
Telescope | +0.021 |
Neg logit contributions | |
chief | -0.024 |
harness | -0.023 |
lex | -0.022 |
TO | -0.022 |
mins | -0.022 |
Token | said |
Token ID | 531 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
chief | +0.206 |
harness | +0.197 |
lex | +0.194 |
TO | +0.193 |
mins | +0.189 |
Neg logit contributions | |
Carbuncle | -0.171 |
certific | -0.170 |
Hungry | -0.168 |
leans | -0.165 |
Telescope | -0.163 |
Token | Raj |
Token ID | 13308 |
Feature activation | -7.91e-04 |
Pos logit contributions | |
chief | +0.272 |
harness | +0.260 |
lex | +0.257 |
TO | +0.255 |
mins | +0.249 |
Neg logit contributions | |
Carbuncle | -0.233 |
certific | -0.232 |
Hungry | -0.229 |
leans | -0.225 |
Telescope | -0.223 |
Token | k |
Token ID | 74 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
chief | +0.001 |
harness | +0.001 |
lex | +0.001 |
TO | +0.001 |
mins | +0.001 |
Neg logit contributions | |
Carbuncle | -0.001 |
certific | -0.000 |
Hungry | -0.000 |
leans | -0.000 |
Telescope | -0.000 |
Token | ES |
Token ID | 1546 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
Carbuncle | +0.179 |
certific | +0.178 |
Hungry | +0.175 |
leans | +0.173 |
Telescope | +0.171 |
Neg logit contributions | |
chief | -0.215 |
harness | -0.205 |
lex | -0.203 |
TO | -0.202 |
mins | -0.197 |
Token | ) |
Token ID | 8 |
Feature activation | +9.55e-02 |
Pos logit contributions | |
Carbuncle | +0.067 |
certific | +0.067 |
Hungry | +0.066 |
leans | +0.065 |
Telescope | +0.065 |
Neg logit contributions | |
chief | -0.069 |
harness | -0.066 |
lex | -0.065 |
TO | -0.064 |
mins | -0.063 |
Token | US |
Token ID | 2937 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
Carbuncle | +0.061 |
certific | +0.061 |
Hungry | +0.060 |
leans | +0.059 |
Telescope | +0.059 |
Neg logit contributions | |
chief | -0.069 |
harness | -0.066 |
lex | -0.065 |
TO | -0.064 |
mins | -0.063 |
Token | Treasury |
Token ID | 13419 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
Carbuncle | +0.219 |
certific | +0.218 |
Hungry | +0.215 |
leans | +0.213 |
Telescope | +0.210 |
Neg logit contributions | |
chief | -0.249 |
harness | -0.238 |
lex | -0.235 |
TO | -0.233 |
mins | -0.228 |
Token | Secretary |
Token ID | 4986 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
chief | +0.148 |
harness | +0.141 |
lex | +0.139 |
TO | +0.138 |
mins | +0.135 |
Neg logit contributions | |
Carbuncle | -0.136 |
certific | -0.136 |
Hungry | -0.134 |
leans | -0.132 |
Telescope | -0.131 |
Token | Jacob |
Token ID | 12806 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
Carbuncle | +0.065 |
certific | +0.065 |
Hungry | +0.064 |
leans | +0.063 |
Telescope | +0.062 |
Neg logit contributions | |
chief | -0.091 |
harness | -0.087 |
lex | -0.086 |
TO | -0.085 |
mins | -0.084 |
Token | Lew |
Token ID | 8260 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
certific | +1.252 |
Carbuncle | +1.246 |
Hungry | +1.215 |
leans | +1.203 |
Telescope | +1.188 |
Neg logit contributions | |
chief | -1.378 |
harness | -1.320 |
lex | -1.293 |
TO | -1.274 |
mins | -1.270 |
Token | speaks |
Token ID | 9209 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
Carbuncle | +0.041 |
certific | +0.041 |
Hungry | +0.040 |
leans | +0.039 |
Telescope | +0.039 |
Neg logit contributions | |
chief | -0.048 |
harness | -0.046 |
lex | -0.046 |
TO | -0.045 |
mins | -0.044 |
Token | on |
Token ID | 319 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
Carbuncle | +0.229 |
certific | +0.228 |
Hungry | +0.225 |
leans | +0.222 |
Telescope | +0.219 |
Neg logit contributions | |
chief | -0.267 |
harness | -0.255 |
lex | -0.251 |
TO | -0.250 |
mins | -0.244 |
Token | the |
Token ID | 262 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
Carbuncle | +0.126 |
certific | +0.126 |
Hungry | +0.124 |
leans | +0.123 |
Telescope | +0.121 |
Neg logit contributions | |
chief | -0.153 |
harness | -0.146 |
lex | -0.144 |
TO | -0.143 |
mins | -0.140 |
Token | state |
Token ID | 1181 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
Carbuncle | +0.222 |
certific | +0.221 |
Hungry | +0.218 |
leans | +0.215 |
Telescope | +0.212 |
Neg logit contributions | |
chief | -0.247 |
harness | -0.236 |
lex | -0.233 |
TO | -0.231 |
mins | -0.225 |
Token | Obama |
Token ID | 2486 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Carbuncle | +0.901 |
certific | +0.897 |
Hungry | +0.883 |
leans | +0.873 |
Telescope | +0.861 |
Neg logit contributions | |
chief | -1.110 |
harness | -1.063 |
TO | -1.043 |
lex | -1.041 |
mins | -1.025 |
Token | With |
Token ID | 3152 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
Carbuncle | +0.689 |
certific | +0.684 |
Hungry | +0.675 |
leans | +0.658 |
Telescope | +0.653 |
Neg logit contributions | |
chief | -0.898 |
harness | -0.855 |
lex | -0.848 |
TO | -0.846 |
mins | -0.818 |
Token | low |
Token ID | 1877 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
Carbuncle | +0.595 |
certific | +0.587 |
Hungry | +0.584 |
leans | +0.574 |
clair | +0.566 |
Neg logit contributions | |
chief | -0.679 |
harness | -0.662 |
lex | -0.646 |
TO | -0.637 |
mins | -0.622 |
Token | birth |
Token ID | 4082 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
chief | +0.285 |
harness | +0.270 |
lex | +0.267 |
TO | +0.265 |
mins | +0.260 |
Neg logit contributions | |
Carbuncle | -0.271 |
certific | -0.271 |
Hungry | -0.266 |
leans | -0.264 |
Telescope | -0.260 |
Token | rate |
Token ID | 2494 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
chief | +0.329 |
harness | +0.313 |
lex | +0.309 |
TO | +0.306 |
mins | +0.300 |
Neg logit contributions | |
certific | -0.336 |
Carbuncle | -0.336 |
Hungry | -0.330 |
leans | -0.328 |
Telescope | -0.322 |
Token | , |
Token ID | 11 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
chief | +1.010 |
harness | +0.966 |
lex | +0.953 |
TO | +0.942 |
mins | +0.937 |
Neg logit contributions | |
Carbuncle | -0.826 |
certific | -0.826 |
Hungry | -0.817 |
leans | -0.807 |
Telescope | -0.794 |
Token | America |
Token ID | 2253 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
chief | +2.238 |
harness | +2.142 |
TO | +2.115 |
mins | +2.111 |
lex | +2.090 |
Neg logit contributions | |
certific | -1.912 |
Carbuncle | -1.862 |
leans | -1.837 |
Hungry | -1.837 |
reating | -1.794 |
Token | needs |
Token ID | 2476 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
chief | +0.395 |
harness | +0.377 |
lex | +0.374 |
TO | +0.372 |
mins | +0.364 |
Neg logit contributions | |
Carbuncle | -0.290 |
certific | -0.290 |
Hungry | -0.286 |
leans | -0.282 |
Telescope | -0.277 |
Token | future |
Token ID | 2003 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
chief | +0.105 |
harness | +0.100 |
lex | +0.099 |
TO | +0.098 |
mins | +0.096 |
Neg logit contributions | |
certific | -0.092 |
Carbuncle | -0.092 |
Hungry | -0.091 |
leans | -0.089 |
Telescope | -0.088 |
Token | migrants |
Token ID | 12783 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
chief | +0.683 |
harness | +0.637 |
lex | +0.634 |
TO | +0.628 |
mins | +0.614 |
Neg logit contributions | |
certific | -0.847 |
Carbuncle | -0.835 |
Hungry | -0.833 |
leans | -0.821 |
Telescope | -0.813 |
Token | 4 |
Token ID | 604 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Carbuncle | +0.433 |
certific | +0.431 |
Hungry | +0.422 |
leans | +0.420 |
clair | +0.412 |
Neg logit contributions | |
chief | -0.464 |
harness | -0.439 |
lex | -0.436 |
TO | -0.433 |
mins | -0.420 |
Token | deal |
Token ID | 1730 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
chief | +0.660 |
harness | +0.622 |
lex | +0.615 |
TO | +0.611 |
mins | +0.600 |
Neg logit contributions | |
certific | -0.684 |
Carbuncle | -0.683 |
Hungry | -0.672 |
leans | -0.661 |
Telescope | -0.656 |
Token | between |
Token ID | 1022 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
chief | +0.181 |
harness | +0.173 |
lex | +0.171 |
TO | +0.169 |
mins | +0.166 |
Neg logit contributions | |
Carbuncle | -0.153 |
certific | -0.153 |
Hungry | -0.151 |
leans | -0.149 |
Telescope | -0.147 |
Token | Iran |
Token ID | 4068 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
chief | +0.029 |
harness | +0.028 |
lex | +0.028 |
TO | +0.027 |
mins | +0.027 |
Neg logit contributions | |
Carbuncle | -0.023 |
certific | -0.023 |
Hungry | -0.023 |
leans | -0.023 |
Telescope | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
chief | +0.010 |
harness | +0.010 |
lex | +0.010 |
TO | +0.010 |
mins | +0.009 |
Neg logit contributions | |
Carbuncle | -0.008 |
certific | -0.008 |
Hungry | -0.008 |
leans | -0.008 |
Telescope | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
chief | +0.663 |
harness | +0.631 |
lex | +0.621 |
TO | +0.618 |
mins | +0.607 |
Neg logit contributions | |
certific | -0.583 |
Carbuncle | -0.581 |
Hungry | -0.576 |
leans | -0.565 |
Telescope | -0.559 |
Token | P |
Token ID | 350 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
chief | +0.908 |
harness | +0.863 |
lex | +0.850 |
TO | +0.845 |
mins | +0.835 |
Neg logit contributions | |
certific | -0.738 |
Carbuncle | -0.732 |
Hungry | -0.726 |
leans | -0.712 |
Telescope | -0.706 |
Token | 5 |
Token ID | 20 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
chief | +1.742 |
harness | +1.663 |
lex | +1.626 |
mins | +1.615 |
TO | +1.590 |
Neg logit contributions | |
certific | -1.129 |
Carbuncle | -1.129 |
Hungry | -1.120 |
leans | -1.109 |
Telescope | -1.074 |
Token | + |
Token ID | 10 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
chief | +0.640 |
harness | +0.612 |
lex | +0.604 |
TO | +0.597 |
mins | +0.587 |
Neg logit contributions | |
Carbuncle | -0.509 |
certific | -0.507 |
Hungry | -0.500 |
leans | -0.492 |
Telescope | -0.488 |
Token | 1 |
Token ID | 16 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
chief | +0.087 |
harness | +0.083 |
lex | +0.082 |
TO | +0.082 |
mins | +0.080 |
Neg logit contributions | |
Carbuncle | -0.061 |
certific | -0.061 |
Hungry | -0.060 |
leans | -0.059 |
Telescope | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
chief | +0.575 |
harness | +0.548 |
lex | +0.542 |
TO | +0.535 |
mins | +0.526 |
Neg logit contributions | |
Carbuncle | -0.483 |
certific | -0.483 |
Hungry | -0.476 |
leans | -0.469 |
Telescope | -0.464 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
chief | +0.014 |
harness | +0.014 |
lex | +0.014 |
TO | +0.013 |
mins | +0.013 |
Neg logit contributions | |
Carbuncle | -0.012 |
certific | -0.012 |
Hungry | -0.012 |
leans | -0.012 |
Telescope | -0.011 |
Token | sometimes |
Token ID | 3360 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
Carbuncle | +0.065 |
certific | +0.065 |
Hungry | +0.064 |
leans | +0.063 |
Telescope | +0.063 |
Neg logit contributions | |
chief | -0.073 |
harness | -0.070 |
lex | -0.069 |
TO | -0.069 |
mins | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
Carbuncle | +0.228 |
certific | +0.227 |
Hungry | +0.224 |
leans | +0.221 |
Telescope | +0.218 |
Neg logit contributions | |
chief | -0.263 |
harness | -0.252 |
lex | -0.248 |
TO | -0.246 |
mins | -0.241 |
Token | goalkeeper |
Token ID | 26318 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
chief | +0.013 |
harness | +0.012 |
lex | +0.012 |
TO | +0.012 |
mins | +0.011 |
Neg logit contributions | |
Carbuncle | -0.012 |
certific | -0.012 |
Hungry | -0.011 |
leans | -0.011 |
Telescope | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
chief | +0.020 |
harness | +0.019 |
lex | +0.019 |
TO | +0.019 |
mins | +0.018 |
Neg logit contributions | |
Carbuncle | -0.017 |
certific | -0.017 |
Hungry | -0.017 |
leans | -0.017 |
Telescope | -0.016 |
Token | Jasper |
Token ID | 42992 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
chief | +0.057 |
harness | +0.055 |
lex | +0.054 |
TO | +0.053 |
mins | +0.052 |
Neg logit contributions | |
Carbuncle | -0.048 |
certific | -0.048 |
Hungry | -0.048 |
leans | -0.047 |
Telescope | -0.046 |
Token | C |
Token ID | 327 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
chief | +0.451 |
harness | +0.429 |
lex | +0.424 |
TO | +0.419 |
mins | +0.411 |
Neg logit contributions | |
Carbuncle | -0.444 |
certific | -0.438 |
Hungry | -0.436 |
leans | -0.430 |
Telescope | -0.426 |
Token | ill |
Token ID | 359 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
chief | +1.728 |
harness | +1.656 |
lex | +1.608 |
TO | +1.599 |
mins | +1.586 |
Neg logit contributions | |
Carbuncle | -1.110 |
certific | -1.077 |
Hungry | -1.071 |
clair | -1.060 |
leans | -1.056 |
Token | essen |
Token ID | 44483 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
chief | +1.343 |
harness | +1.299 |
TO | +1.259 |
lex | +1.247 |
mins | +1.227 |
Neg logit contributions | |
Carbuncle | -0.916 |
certific | -0.898 |
Hungry | -0.888 |
leans | -0.885 |
clair | -0.872 |
Token | , |
Token ID | 11 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Carbuncle | +0.131 |
certific | +0.131 |
Hungry | +0.129 |
leans | +0.127 |
Telescope | +0.125 |
Neg logit contributions | |
chief | -0.154 |
harness | -0.148 |
lex | -0.146 |
TO | -0.144 |
mins | -0.141 |
Token | who |
Token ID | 508 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
Carbuncle | +0.088 |
certific | +0.087 |
Hungry | +0.086 |
leans | +0.085 |
Telescope | +0.084 |
Neg logit contributions | |
chief | -0.105 |
harness | -0.100 |
lex | -0.099 |
TO | -0.098 |
mins | -0.096 |
Token | started |
Token ID | 2067 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
Carbuncle | +0.019 |
certific | +0.019 |
Hungry | +0.019 |
leans | +0.018 |
Telescope | +0.018 |
Neg logit contributions | |
chief | -0.023 |
harness | -0.022 |
lex | -0.022 |
TO | -0.022 |
mins | -0.022 |
Token | very |
Token ID | 845 |
Feature activation | -2.04e-03 |
Pos logit contributions | |
chief | +0.224 |
harness | +0.214 |
lex | +0.211 |
TO | +0.210 |
mins | +0.205 |
Neg logit contributions | |
Carbuncle | -0.186 |
certific | -0.186 |
Hungry | -0.183 |
leans | -0.180 |
Telescope | -0.178 |
Token | popular |
Token ID | 2968 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
chief | +0.001 |
harness | +0.001 |
lex | +0.001 |
TO | +0.001 |
mins | +0.001 |
Neg logit contributions | |
Carbuncle | -0.001 |
certific | -0.001 |
Hungry | -0.001 |
leans | -0.001 |
Telescope | -0.001 |
Token | Asus |
Token ID | 46301 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
chief | +0.180 |
harness | +0.172 |
lex | +0.169 |
TO | +0.168 |
mins | +0.165 |
Neg logit contributions | |
Carbuncle | -0.154 |
certific | -0.154 |
Hungry | -0.152 |
leans | -0.149 |
Telescope | -0.148 |
Token | R |
Token ID | 371 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
chief | +0.200 |
harness | +0.191 |
lex | +0.188 |
TO | +0.187 |
mins | +0.183 |
Neg logit contributions | |
Carbuncle | -0.178 |
certific | -0.177 |
Hungry | -0.175 |
leans | -0.172 |
Telescope | -0.171 |
Token | OG |
Token ID | 7730 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
chief | +0.233 |
harness | +0.223 |
lex | +0.220 |
TO | +0.218 |
mins | +0.215 |
Neg logit contributions | |
Carbuncle | -0.157 |
certific | -0.156 |
Hungry | -0.154 |
leans | -0.152 |
Telescope | -0.149 |
Token | Swift |
Token ID | 15608 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
chief | +0.401 |
harness | +0.381 |
lex | +0.376 |
TO | +0.373 |
mins | +0.365 |
Neg logit contributions | |
Carbuncle | -0.384 |
certific | -0.382 |
Hungry | -0.379 |
leans | -0.372 |
Telescope | -0.370 |
Token | PG |
Token ID | 23842 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
chief | +1.351 |
harness | +1.300 |
TO | +1.264 |
lex | +1.248 |
mins | +1.242 |
Neg logit contributions | |
Carbuncle | -1.426 |
certific | -1.391 |
Hungry | -1.384 |
leans | -1.369 |
Telescope | -1.358 |
Token | 278 |
Token ID | 25870 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
Carbuncle | +0.317 |
certific | +0.315 |
Hungry | +0.310 |
leans | +0.307 |
Telescope | +0.303 |
Neg logit contributions | |
chief | -0.363 |
harness | -0.347 |
lex | -0.343 |
TO | -0.340 |
mins | -0.332 |
Token | Q |
Token ID | 48 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
chief | +0.403 |
harness | +0.380 |
lex | +0.371 |
TO | +0.364 |
mins | +0.357 |
Neg logit contributions | |
Carbuncle | -0.641 |
certific | -0.640 |
Hungry | -0.635 |
leans | -0.627 |
Telescope | -0.622 |
Token | with |
Token ID | 351 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
chief | +0.074 |
harness | +0.071 |
lex | +0.070 |
TO | +0.069 |
mins | +0.068 |
Neg logit contributions | |
Carbuncle | -0.065 |
certific | -0.065 |
Hungry | -0.064 |
leans | -0.063 |
Telescope | -0.062 |
Token | its |
Token ID | 663 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
chief | +0.296 |
harness | +0.283 |
lex | +0.279 |
TO | +0.277 |
mins | +0.271 |
Neg logit contributions | |
Carbuncle | -0.238 |
certific | -0.238 |
Hungry | -0.235 |
leans | -0.231 |
Telescope | -0.229 |
Token | K |
Token ID | 42 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
chief | +0.676 |
harness | +0.648 |
lex | +0.639 |
TO | +0.635 |
mins | +0.629 |
Neg logit contributions | |
Carbuncle | -0.405 |
certific | -0.402 |
Hungry | -0.399 |
leans | -0.386 |
Telescope | -0.381 |
Token | . |
Token ID | 13 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
chief | +0.258 |
harness | +0.247 |
lex | +0.243 |
TO | +0.242 |
mins | +0.237 |
Neg logit contributions | |
Carbuncle | -0.221 |
certific | -0.219 |
Hungry | -0.217 |
leans | -0.214 |
Telescope | -0.211 |
Token | Louie |
Token ID | 48352 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
chief | +0.047 |
harness | +0.044 |
lex | +0.044 |
TO | +0.043 |
mins | +0.042 |
Neg logit contributions | |
Carbuncle | -0.043 |
certific | -0.042 |
Hungry | -0.042 |
leans | -0.041 |
Telescope | -0.041 |
Token | FX |
Token ID | 19534 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
Carbuncle | +0.016 |
certific | +0.016 |
Hungry | +0.016 |
leans | +0.016 |
Telescope | +0.015 |
Neg logit contributions | |
chief | -0.019 |
harness | -0.018 |
lex | -0.018 |
TO | -0.017 |
mins | -0.017 |
Token | Mind |
Token ID | 10175 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
chief | +0.033 |
harness | +0.032 |
lex | +0.031 |
TO | +0.031 |
mins | +0.030 |
Neg logit contributions | |
Carbuncle | -0.030 |
certific | -0.030 |
Hungry | -0.030 |
leans | -0.029 |
Telescope | -0.029 |
Token | y |
Token ID | 88 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
chief | +0.167 |
harness | +0.159 |
lex | +0.156 |
TO | +0.155 |
mins | +0.152 |
Neg logit contributions | |
Carbuncle | -0.152 |
certific | -0.152 |
Hungry | -0.150 |
leans | -0.148 |
Telescope | -0.146 |
Token | K |
Token ID | 509 |
Feature activation | -8.62e-01 |
Pos logit contributions | |
chief | +1.544 |
harness | +1.496 |
TO | +1.458 |
lex | +1.452 |
mins | +1.446 |
Neg logit contributions | |
Carbuncle | -1.177 |
Hungry | -1.156 |
certific | -1.140 |
Telescope | -1.131 |
leans | -1.104 |
Token | aling |
Token ID | 4272 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
chief | +0.685 |
harness | +0.658 |
lex | +0.649 |
TO | +0.643 |
mins | +0.634 |
Neg logit contributions | |
certific | -0.490 |
Carbuncle | -0.486 |
Hungry | -0.483 |
leans | -0.475 |
Telescope | -0.468 |
Token | The |
Token ID | 383 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
chief | +0.017 |
harness | +0.016 |
lex | +0.016 |
TO | +0.016 |
mins | +0.015 |
Neg logit contributions | |
Carbuncle | -0.015 |
certific | -0.015 |
Hungry | -0.015 |
leans | -0.015 |
Telescope | -0.015 |
Token | Mind |
Token ID | 10175 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
chief | +0.323 |
harness | +0.309 |
lex | +0.304 |
TO | +0.302 |
mins | +0.296 |
Neg logit contributions | |
Carbuncle | -0.267 |
certific | -0.265 |
Hungry | -0.264 |
leans | -0.259 |
Telescope | -0.256 |
Token | y |
Token ID | 88 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
Carbuncle | +0.103 |
certific | +0.102 |
Hungry | +0.100 |
leans | +0.099 |
Telescope | +0.098 |
Neg logit contributions | |
chief | -0.148 |
harness | -0.142 |
lex | -0.141 |
TO | -0.140 |
mins | -0.137 |
Token | takeover |
Token ID | 27011 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
Carbuncle | +0.313 |
certific | +0.312 |
Hungry | +0.307 |
leans | +0.304 |
Telescope | +0.300 |
Neg logit contributions | |
chief | -0.371 |
harness | -0.355 |
lex | -0.350 |
TO | -0.347 |
mins | -0.339 |
Token | of |
Token ID | 286 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
Carbuncle | +0.050 |
certific | +0.050 |
Hungry | +0.049 |
leans | +0.049 |
Telescope | +0.048 |
Neg logit contributions | |
chief | -0.056 |
harness | -0.054 |
lex | -0.053 |
TO | -0.053 |
mins | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Carbuncle | +0.029 |
certific | +0.029 |
Hungry | +0.029 |
leans | +0.028 |
Telescope | +0.028 |
Neg logit contributions | |
chief | -0.042 |
harness | -0.040 |
lex | -0.040 |
TO | -0.039 |
mins | -0.039 |
Token | M |
Token ID | 337 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
Carbuncle | +0.059 |
certific | +0.059 |
Hungry | +0.058 |
leans | +0.057 |
Telescope | +0.056 |
Neg logit contributions | |
chief | -0.099 |
harness | -0.095 |
lex | -0.094 |
TO | -0.094 |
mins | -0.092 |
Token | avi |
Token ID | 15820 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
chief | +0.262 |
harness | +0.250 |
lex | +0.246 |
TO | +0.243 |
mins | +0.239 |
Neg logit contributions | |
Carbuncle | -0.237 |
certific | -0.237 |
Hungry | -0.234 |
leans | -0.231 |
Telescope | -0.227 |
Token | Mar |
Token ID | 1526 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
Carbuncle | +0.095 |
certific | +0.095 |
Hungry | +0.093 |
leans | +0.092 |
Telescope | +0.090 |
Neg logit contributions | |
chief | -0.165 |
harness | -0.159 |
lex | -0.157 |
TO | -0.156 |
mins | -0.153 |
Token | m |
Token ID | 76 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
chief | +1.075 |
harness | +1.024 |
TO | +1.007 |
lex | +1.004 |
ODUCT | +0.957 |
Neg logit contributions | |
Hungry | -1.510 |
leans | -1.491 |
Carbuncle | -1.488 |
certific | -1.484 |
Telescope | -1.471 |
Token | ara |
Token ID | 3301 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
chief | +0.362 |
harness | +0.345 |
lex | +0.341 |
TO | +0.339 |
mins | +0.332 |
Neg logit contributions | |
Carbuncle | -0.289 |
certific | -0.287 |
Hungry | -0.284 |
leans | -0.280 |
Telescope | -0.277 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
chief | +0.141 |
harness | +0.134 |
lex | +0.132 |
TO | +0.131 |
mins | +0.129 |
Neg logit contributions | |
Carbuncle | -0.120 |
certific | -0.119 |
Hungry | -0.118 |
leans | -0.116 |
Telescope | -0.115 |
Token | Turkey |
Token ID | 7137 |
Feature activation | -7.92e-02 |
Pos logit contributions | |
chief | +0.136 |
harness | +0.130 |
lex | +0.128 |
TO | +0.127 |
mins | +0.124 |
Neg logit contributions | |
Carbuncle | -0.110 |
certific | -0.110 |
Hungry | -0.108 |
leans | -0.107 |
Telescope | -0.105 |
Token | and |
Token ID | 290 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
chief | +0.061 |
harness | +0.058 |
lex | +0.057 |
TO | +0.057 |
mins | +0.056 |
Neg logit contributions | |
Carbuncle | -0.047 |
certific | -0.047 |
Hungry | -0.047 |
leans | -0.046 |
Telescope | -0.045 |
Token | ), |
Token ID | 828 |
Feature activation | -1.38e-04 |
Pos logit contributions | |
Carbuncle | +0.109 |
certific | +0.109 |
Hungry | +0.108 |
leans | +0.106 |
Telescope | +0.105 |
Neg logit contributions | |
chief | -0.133 |
harness | -0.127 |
lex | -0.125 |
TO | -0.124 |
mins | -0.122 |
Token | – |
Token ID | 784 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
chief | +0.000 |
harness | +0.000 |
lex | +0.000 |
TO | +0.000 |
mins | +0.000 |
Neg logit contributions | |
Carbuncle | -0.000 |
certific | -0.000 |
Hungry | -0.000 |
leans | -0.000 |
Telescope | -0.000 |
Token | one |
Token ID | 530 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
chief | +0.114 |
harness | +0.109 |
lex | +0.108 |
TO | +0.107 |
mins | +0.105 |
Neg logit contributions | |
Carbuncle | -0.097 |
certific | -0.097 |
Hungry | -0.096 |
leans | -0.094 |
Telescope | -0.093 |
Token | from |
Token ID | 422 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
Carbuncle | +0.085 |
certific | +0.084 |
Hungry | +0.083 |
leans | +0.082 |
Telescope | +0.081 |
Neg logit contributions | |
chief | -0.096 |
harness | -0.092 |
lex | -0.090 |
TO | -0.090 |
mins | -0.088 |
Token | the |
Token ID | 262 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
chief | +0.099 |
harness | +0.095 |
lex | +0.093 |
TO | +0.093 |
mins | +0.091 |
Neg logit contributions | |
Carbuncle | -0.077 |
certific | -0.077 |
Hungry | -0.076 |
leans | -0.075 |
Telescope | -0.074 |
Token | Kimber |
Token ID | 36015 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
chief | +0.169 |
harness | +0.161 |
lex | +0.159 |
TO | +0.158 |
mins | +0.155 |
Neg logit contributions | |
Carbuncle | -0.140 |
certific | -0.139 |
Hungry | -0.138 |
leans | -0.135 |
Telescope | -0.134 |
Token | ley |
Token ID | 1636 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
chief | +1.141 |
harness | +1.077 |
lex | +1.057 |
TO | +1.049 |
mins | +1.025 |
Neg logit contributions | |
Carbuncle | -1.450 |
certific | -1.442 |
Hungry | -1.438 |
leans | -1.427 |
Telescope | -1.408 |
Token | and |
Token ID | 290 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
chief | +0.473 |
harness | +0.451 |
lex | +0.446 |
TO | +0.442 |
mins | +0.433 |
Neg logit contributions | |
Carbuncle | -0.393 |
certific | -0.391 |
Hungry | -0.387 |
leans | -0.380 |
Telescope | -0.377 |
Token | western |
Token ID | 8830 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
chief | +0.113 |
harness | +0.108 |
lex | +0.107 |
TO | +0.106 |
mins | +0.104 |
Neg logit contributions | |
Carbuncle | -0.090 |
certific | -0.089 |
Hungry | -0.088 |
leans | -0.087 |
Telescope | -0.086 |
Token | Northern |
Token ID | 8342 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
chief | +0.233 |
harness | +0.222 |
lex | +0.219 |
TO | +0.218 |
mins | +0.213 |
Neg logit contributions | |
Carbuncle | -0.185 |
certific | -0.184 |
Hungry | -0.182 |
leans | -0.178 |
Telescope | -0.176 |
Token | Territory |
Token ID | 25219 |
Feature activation | +7.75e-02 |
Pos logit contributions | |
Carbuncle | +0.185 |
certific | +0.185 |
Hungry | +0.181 |
leans | +0.179 |
Telescope | +0.176 |
Neg logit contributions | |
chief | -0.259 |
harness | -0.249 |
lex | -0.245 |
TO | -0.244 |
mins | -0.239 |
Token | Manafort |
Token ID | 26351 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Carbuncle | +0.103 |
certific | +0.102 |
Hungry | +0.101 |
leans | +0.100 |
Telescope | +0.098 |
Neg logit contributions | |
chief | -0.109 |
harness | -0.104 |
lex | -0.102 |
TO | -0.101 |
mins | -0.099 |
Token | 's |
Token ID | 338 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
Carbuncle | +0.655 |
certific | +0.650 |
Hungry | +0.639 |
leans | +0.633 |
clair | +0.622 |
Neg logit contributions | |
chief | -0.810 |
harness | -0.775 |
lex | -0.752 |
TO | -0.752 |
mins | -0.729 |
Token | attorneys |
Token ID | 14449 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
Carbuncle | +0.589 |
certific | +0.585 |
Hungry | +0.578 |
leans | +0.574 |
Telescope | +0.565 |
Neg logit contributions | |
chief | -0.648 |
harness | -0.622 |
lex | -0.610 |
TO | -0.606 |
mins | -0.583 |
Token | say |
Token ID | 910 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Carbuncle | +0.326 |
certific | +0.325 |
Hungry | +0.320 |
leans | +0.316 |
Telescope | +0.310 |
Neg logit contributions | |
chief | -0.437 |
harness | -0.421 |
lex | -0.412 |
TO | -0.411 |
mins | -0.399 |
Token | he |
Token ID | 339 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
Carbuncle | +0.874 |
certific | +0.864 |
Hungry | +0.847 |
leans | +0.838 |
Telescope | +0.836 |
Neg logit contributions | |
chief | -0.963 |
harness | -0.920 |
lex | -0.904 |
TO | -0.900 |
mins | -0.871 |
Token | should |
Token ID | 815 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
Carbuncle | +1.043 |
certific | +1.034 |
Hungry | +1.012 |
leans | +1.005 |
clair | +1.002 |
Neg logit contributions | |
chief | -1.203 |
harness | -1.164 |
lex | -1.141 |
TO | -1.131 |
mins | -1.096 |
Token | get |
Token ID | 651 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
chief | +1.470 |
harness | +1.385 |
TO | +1.373 |
lex | +1.362 |
mins | +1.357 |
Neg logit contributions | |
Carbuncle | -1.112 |
leans | -1.101 |
Hungry | -1.094 |
certific | -1.092 |
Telescope | -1.068 |
Token | less |
Token ID | 1342 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
Carbuncle | +0.560 |
certific | +0.560 |
Hungry | +0.553 |
leans | +0.543 |
clair | +0.539 |
Neg logit contributions | |
chief | -0.593 |
harness | -0.572 |
lex | -0.561 |
TO | -0.551 |
mins | -0.536 |
Token | than |
Token ID | 621 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
chief | +0.119 |
harness | +0.113 |
lex | +0.111 |
TO | +0.111 |
mins | +0.108 |
Neg logit contributions | |
Carbuncle | -0.120 |
certific | -0.119 |
Hungry | -0.118 |
leans | -0.116 |
Telescope | -0.115 |
Token | 10 |
Token ID | 838 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
chief | +0.064 |
harness | +0.061 |
lex | +0.060 |
TO | +0.060 |
mins | +0.058 |
Neg logit contributions | |
Carbuncle | -0.057 |
certific | -0.057 |
Hungry | -0.056 |
leans | -0.056 |
Telescope | -0.055 |
Token | years |
Token ID | 812 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
chief | +0.213 |
harness | +0.201 |
lex | +0.198 |
TO | +0.196 |
mins | +0.190 |
Neg logit contributions | |
Carbuncle | -0.257 |
certific | -0.257 |
Hungry | -0.253 |
leans | -0.250 |
Telescope | -0.248 |
Token | ; |
Token ID | 26 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Carbuncle | +0.130 |
certific | +0.129 |
Hungry | +0.128 |
leans | +0.126 |
Telescope | +0.125 |
Neg logit contributions | |
chief | -0.129 |
harness | -0.122 |
lex | -0.121 |
TO | -0.120 |
mins | -0.117 |
Token | Coral |
Token ID | 41390 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
chief | +0.082 |
harness | +0.078 |
lex | +0.077 |
TO | +0.077 |
mins | +0.075 |
Neg logit contributions | |
Carbuncle | -0.068 |
certific | -0.068 |
Hungry | -0.067 |
leans | -0.066 |
Telescope | -0.065 |
Token | Sea |
Token ID | 6896 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
chief | +0.706 |
harness | +0.671 |
lex | +0.658 |
TO | +0.652 |
mins | +0.637 |
Neg logit contributions | |
Carbuncle | -0.913 |
certific | -0.903 |
Hungry | -0.900 |
leans | -0.889 |
Telescope | -0.881 |
Token | Islands |
Token ID | 12010 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
chief | +0.640 |
harness | +0.610 |
lex | +0.602 |
TO | +0.598 |
mins | +0.587 |
Neg logit contributions | |
Carbuncle | -0.586 |
certific | -0.585 |
Hungry | -0.579 |
leans | -0.569 |
Telescope | -0.566 |
Token | ; |
Token ID | 26 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Carbuncle | +0.074 |
certific | +0.074 |
Hungry | +0.073 |
leans | +0.072 |
Telescope | +0.072 |
Neg logit contributions | |
chief | -0.074 |
harness | -0.071 |
lex | -0.069 |
TO | -0.069 |
mins | -0.067 |
Token | Heard |
Token ID | 49417 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
chief | +0.080 |
harness | +0.077 |
lex | +0.076 |
TO | +0.075 |
mins | +0.074 |
Neg logit contributions | |
Carbuncle | -0.067 |
certific | -0.067 |
Hungry | -0.066 |
leans | -0.065 |
Telescope | -0.064 |
Token | Island |
Token ID | 5451 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
chief | +1.669 |
harness | +1.635 |
lex | +1.604 |
TO | +1.602 |
mins | +1.560 |
Neg logit contributions | |
Carbuncle | -0.871 |
Hungry | -0.863 |
certific | -0.847 |
leans | -0.837 |
Telescope | -0.827 |
Token | and |
Token ID | 290 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
chief | +0.172 |
harness | +0.164 |
lex | +0.162 |
TO | +0.161 |
mins | +0.157 |
Neg logit contributions | |
Carbuncle | -0.136 |
certific | -0.136 |
Hungry | -0.134 |
leans | -0.132 |
Telescope | -0.131 |
Token | McDonald |
Token ID | 14115 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
chief | +0.823 |
harness | +0.780 |
lex | +0.774 |
TO | +0.765 |
mins | +0.738 |
Neg logit contributions | |
Hungry | -1.129 |
Carbuncle | -1.121 |
certific | -1.120 |
leans | -1.096 |
Telescope | -1.089 |
Token | Islands |
Token ID | 12010 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
chief | +0.034 |
harness | +0.032 |
lex | +0.032 |
TO | +0.032 |
mins | +0.031 |
Neg logit contributions | |
Carbuncle | -0.025 |
certific | -0.025 |
Hungry | -0.024 |
leans | -0.024 |
Telescope | -0.024 |
Token | ; |
Token ID | 26 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
Carbuncle | +0.026 |
certific | +0.026 |
Hungry | +0.026 |
leans | +0.025 |
Telescope | +0.025 |
Neg logit contributions | |
chief | -0.026 |
harness | -0.025 |
lex | -0.024 |
TO | -0.024 |
mins | -0.024 |
Token | by |
Token ID | 416 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
chief | +0.040 |
harness | +0.038 |
lex | +0.038 |
TO | +0.038 |
mins | +0.037 |
Neg logit contributions | |
Carbuncle | -0.032 |
certific | -0.032 |
Hungry | -0.032 |
leans | -0.031 |
Telescope | -0.031 |
Token | going |
Token ID | 1016 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
chief | +0.162 |
harness | +0.154 |
lex | +0.153 |
TO | +0.151 |
mins | +0.148 |
Neg logit contributions | |
Carbuncle | -0.136 |
certific | -0.136 |
Hungry | -0.134 |
leans | -0.132 |
Telescope | -0.130 |
Token | here |
Token ID | 994 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
chief | +0.083 |
harness | +0.079 |
lex | +0.078 |
TO | +0.077 |
mins | +0.076 |
Neg logit contributions | |
Carbuncle | -0.074 |
certific | -0.074 |
Hungry | -0.073 |
leans | -0.072 |
Telescope | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
chief | +0.148 |
harness | +0.141 |
lex | +0.139 |
TO | +0.138 |
mins | +0.135 |
Neg logit contributions | |
Carbuncle | -0.126 |
certific | -0.125 |
Hungry | -0.123 |
leans | -0.122 |
Telescope | -0.120 |
Token | But |
Token ID | 887 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
Carbuncle | +0.516 |
certific | +0.515 |
leans | +0.504 |
Hungry | +0.501 |
Telescope | +0.494 |
Neg logit contributions | |
chief | -0.510 |
harness | -0.485 |
lex | -0.479 |
TO | -0.477 |
mins | -0.462 |
Token | please |
Token ID | 3387 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
chief | +0.523 |
harness | +0.495 |
lex | +0.491 |
TO | +0.490 |
mins | +0.474 |
Neg logit contributions | |
certific | -0.522 |
Carbuncle | -0.517 |
Hungry | -0.514 |
leans | -0.501 |
Telescope | -0.499 |
Token | note |
Token ID | 3465 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
chief | +1.038 |
harness | +0.968 |
TO | +0.957 |
lex | +0.944 |
mins | +0.935 |
Neg logit contributions | |
certific | -1.511 |
Carbuncle | -1.509 |
Hungry | -1.468 |
Telescope | -1.440 |
ensis | -1.434 |
Token | that |
Token ID | 326 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
chief | +0.288 |
harness | +0.275 |
lex | +0.271 |
TO | +0.269 |
mins | +0.264 |
Neg logit contributions | |
Carbuncle | -0.245 |
certific | -0.245 |
Hungry | -0.240 |
leans | -0.237 |
Telescope | -0.235 |
Token | my |
Token ID | 616 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
chief | +0.091 |
harness | +0.087 |
lex | +0.086 |
TO | +0.086 |
mins | +0.084 |
Neg logit contributions | |
Carbuncle | -0.075 |
certific | -0.075 |
Hungry | -0.074 |
leans | -0.073 |
Telescope | -0.072 |
Token | tweets |
Token ID | 12665 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
chief | +0.753 |
lex | +0.704 |
harness | +0.701 |
TO | +0.699 |
mins | +0.672 |
Neg logit contributions | |
certific | -1.015 |
Carbuncle | -1.014 |
Hungry | -1.006 |
leans | -0.993 |
Telescope | -0.984 |
Token | are |
Token ID | 389 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
chief | +0.312 |
harness | +0.298 |
lex | +0.294 |
TO | +0.292 |
mins | +0.286 |
Neg logit contributions | |
Carbuncle | -0.238 |
certific | -0.237 |
Hungry | -0.234 |
leans | -0.231 |
Telescope | -0.228 |
Token | Manafort |
Token ID | 26351 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
chief | +0.134 |
harness | +0.128 |
lex | +0.126 |
TO | +0.125 |
mins | +0.123 |
Neg logit contributions | |
Carbuncle | -0.124 |
certific | -0.123 |
Hungry | -0.122 |
leans | -0.120 |
Telescope | -0.119 |
Token | 's |
Token ID | 338 |
Feature activation | +9.11e-01 |
Pos logit contributions | |
Carbuncle | +0.695 |
certific | +0.690 |
Hungry | +0.678 |
leans | +0.671 |
clair | +0.660 |
Neg logit contributions | |
chief | -0.866 |
harness | -0.830 |
lex | -0.804 |
TO | -0.804 |
mins | -0.780 |
Token | attorneys |
Token ID | 14449 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
Carbuncle | +0.592 |
certific | +0.587 |
Hungry | +0.581 |
leans | +0.576 |
Telescope | +0.567 |
Neg logit contributions | |
chief | -0.653 |
harness | -0.627 |
lex | -0.614 |
TO | -0.610 |
mins | -0.586 |
Token | say |
Token ID | 910 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Carbuncle | +0.336 |
certific | +0.336 |
Hungry | +0.331 |
leans | +0.327 |
Telescope | +0.321 |
Neg logit contributions | |
chief | -0.442 |
harness | -0.425 |
lex | -0.416 |
TO | -0.414 |
mins | -0.403 |
Token | he |
Token ID | 339 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
Carbuncle | +0.751 |
certific | +0.742 |
Hungry | +0.729 |
leans | +0.720 |
Telescope | +0.717 |
Neg logit contributions | |
chief | -0.845 |
harness | -0.807 |
lex | -0.794 |
TO | -0.790 |
mins | -0.766 |
Token | should |
Token ID | 815 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
Carbuncle | +1.102 |
certific | +1.092 |
Hungry | +1.068 |
leans | +1.060 |
clair | +1.059 |
Neg logit contributions | |
chief | -1.235 |
harness | -1.195 |
lex | -1.169 |
TO | -1.156 |
mins | -1.120 |
Token | get |
Token ID | 651 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
chief | +1.424 |
harness | +1.339 |
TO | +1.329 |
lex | +1.318 |
mins | +1.312 |
Neg logit contributions | |
Carbuncle | -1.110 |
leans | -1.100 |
Hungry | -1.095 |
certific | -1.091 |
Telescope | -1.069 |
Token | less |
Token ID | 1342 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Carbuncle | +0.556 |
certific | +0.555 |
Hungry | +0.548 |
leans | +0.539 |
clair | +0.535 |
Neg logit contributions | |
chief | -0.604 |
harness | -0.583 |
lex | -0.572 |
TO | -0.562 |
mins | -0.547 |
Token | than |
Token ID | 621 |
Feature activation | -8.32e-02 |
Pos logit contributions | |
chief | +0.167 |
harness | +0.159 |
lex | +0.156 |
TO | +0.155 |
mins | +0.152 |
Neg logit contributions | |
Carbuncle | -0.172 |
certific | -0.172 |
Hungry | -0.170 |
leans | -0.168 |
Telescope | -0.166 |
Token | 10 |
Token ID | 838 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
chief | +0.059 |
harness | +0.056 |
lex | +0.055 |
TO | +0.055 |
mins | +0.054 |
Neg logit contributions | |
Carbuncle | -0.055 |
certific | -0.054 |
Hungry | -0.054 |
leans | -0.053 |
Telescope | -0.052 |
Token | years |
Token ID | 812 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
chief | +0.190 |
harness | +0.180 |
lex | +0.177 |
TO | +0.175 |
mins | +0.170 |
Neg logit contributions | |
Carbuncle | -0.226 |
certific | -0.226 |
Hungry | -0.223 |
leans | -0.220 |
Telescope | -0.218 |
Token | shareholders |
Token ID | 19195 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Carbuncle | +0.071 |
certific | +0.071 |
Hungry | +0.069 |
leans | +0.069 |
Telescope | +0.068 |
Neg logit contributions | |
chief | -0.089 |
harness | -0.085 |
lex | -0.084 |
TO | -0.083 |
mins | -0.082 |
Token | ," |
Token ID | 553 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
Carbuncle | +0.106 |
certific | +0.106 |
Hungry | +0.104 |
leans | +0.103 |
Telescope | +0.102 |
Neg logit contributions | |
chief | -0.120 |
harness | -0.115 |
lex | -0.113 |
TO | -0.112 |
mins | -0.110 |
Token | deputy |
Token ID | 10636 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
Carbuncle | +0.157 |
certific | +0.156 |
Hungry | +0.154 |
leans | +0.152 |
Telescope | +0.150 |
Neg logit contributions | |
chief | -0.194 |
harness | -0.186 |
lex | -0.184 |
TO | -0.182 |
mins | -0.178 |
Token | Conservative |
Token ID | 11132 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
Carbuncle | +0.164 |
certific | +0.163 |
Hungry | +0.161 |
leans | +0.159 |
Telescope | +0.157 |
Neg logit contributions | |
chief | -0.198 |
harness | -0.189 |
lex | -0.187 |
TO | -0.185 |
mins | -0.181 |
Token | leader |
Token ID | 3554 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
chief | +0.284 |
harness | +0.271 |
lex | +0.267 |
TO | +0.264 |
mins | +0.259 |
Neg logit contributions | |
Carbuncle | -0.282 |
certific | -0.281 |
Hungry | -0.278 |
leans | -0.274 |
Telescope | -0.271 |
Token | Lisa |
Token ID | 15378 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
chief | +0.024 |
harness | +0.023 |
lex | +0.022 |
TO | +0.022 |
mins | +0.022 |
Neg logit contributions | |
Carbuncle | -0.021 |
certific | -0.021 |
Hungry | -0.021 |
leans | -0.020 |
Telescope | -0.020 |
Token | Ra |
Token ID | 7567 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
chief | +1.051 |
harness | +0.991 |
lex | +0.971 |
TO | +0.962 |
mins | +0.930 |
Neg logit contributions | |
Hungry | -1.491 |
Carbuncle | -1.485 |
certific | -1.449 |
Telescope | -1.442 |
leans | -1.437 |
Token | itt |
Token ID | 715 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
Carbuncle | +0.269 |
certific | +0.268 |
Hungry | +0.262 |
leans | +0.257 |
Telescope | +0.254 |
Neg logit contributions | |
chief | -0.498 |
harness | -0.478 |
lex | -0.475 |
TO | -0.471 |
mins | -0.462 |
Token | said |
Token ID | 531 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
Carbuncle | +0.194 |
certific | +0.194 |
Hungry | +0.191 |
leans | +0.188 |
Telescope | +0.186 |
Neg logit contributions | |
chief | -0.240 |
harness | -0.230 |
lex | -0.227 |
TO | -0.225 |
mins | -0.220 |
Token | . |
Token ID | 13 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
Carbuncle | +0.178 |
certific | +0.177 |
Hungry | +0.174 |
leans | +0.172 |
Telescope | +0.170 |
Neg logit contributions | |
chief | -0.203 |
harness | -0.194 |
lex | -0.191 |
TO | -0.190 |
mins | -0.186 |
Token | " |
Token ID | 366 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Carbuncle | +0.023 |
certific | +0.023 |
Hungry | +0.023 |
leans | +0.023 |
Telescope | +0.022 |
Neg logit contributions | |
chief | -0.029 |
harness | -0.027 |
lex | -0.027 |
TO | -0.027 |
mins | -0.026 |
Token | at |
Token ID | 379 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
chief | +0.037 |
harness | +0.035 |
lex | +0.034 |
TO | +0.034 |
mins | +0.033 |
Neg logit contributions | |
Carbuncle | -0.031 |
certific | -0.031 |
Hungry | -0.031 |
leans | -0.030 |
Telescope | -0.030 |
Token | home |
Token ID | 1363 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
chief | +0.148 |
harness | +0.139 |
lex | +0.137 |
TO | +0.136 |
mins | +0.132 |
Neg logit contributions | |
Carbuncle | -0.189 |
certific | -0.188 |
Hungry | -0.186 |
leans | -0.184 |
Telescope | -0.182 |
Token | . |
Token ID | 13 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
Carbuncle | +0.088 |
certific | +0.087 |
Hungry | +0.086 |
leans | +0.085 |
Telescope | +0.084 |
Neg logit contributions | |
chief | -0.106 |
harness | -0.101 |
lex | -0.100 |
TO | -0.099 |
mins | -0.097 |
Token |
|
Token ID | 198 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
chief | +0.025 |
harness | +0.024 |
lex | +0.024 |
TO | +0.023 |
mins | +0.023 |
Neg logit contributions | |
Carbuncle | -0.020 |
certific | -0.020 |
Hungry | -0.020 |
leans | -0.019 |
Telescope | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
Carbuncle | +0.042 |
certific | +0.042 |
Hungry | +0.041 |
leans | +0.040 |
Telescope | +0.040 |
Neg logit contributions | |
chief | -0.052 |
harness | -0.050 |
lex | -0.049 |
TO | -0.049 |
mins | -0.048 |
Token | Taylor |
Token ID | 29907 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
chief | +0.008 |
harness | +0.008 |
lex | +0.007 |
TO | +0.007 |
mins | +0.007 |
Neg logit contributions | |
Carbuncle | -0.007 |
certific | -0.007 |
Hungry | -0.007 |
leans | -0.007 |
Telescope | -0.007 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
chief | +1.073 |
harness | +1.006 |
lex | +0.983 |
TO | +0.972 |
mins | +0.947 |
Neg logit contributions | |
Carbuncle | -1.430 |
Hungry | -1.421 |
certific | -1.408 |
leans | -1.388 |
Telescope | -1.375 |
Token | 't |
Token ID | 470 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
chief | +0.155 |
harness | +0.148 |
lex | +0.145 |
TO | +0.144 |
mins | +0.141 |
Neg logit contributions | |
Carbuncle | -0.147 |
certific | -0.147 |
Hungry | -0.145 |
leans | -0.143 |
Telescope | -0.142 |
Token | have |
Token ID | 423 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
chief | +0.225 |
harness | +0.214 |
lex | +0.212 |
TO | +0.210 |
mins | +0.206 |
Neg logit contributions | |
Carbuncle | -0.194 |
certific | -0.193 |
Hungry | -0.191 |
leans | -0.188 |
Telescope | -0.186 |
Token | the |
Token ID | 262 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
chief | +0.376 |
harness | +0.359 |
lex | +0.354 |
TO | +0.352 |
mins | +0.345 |
Neg logit contributions | |
Carbuncle | -0.317 |
certific | -0.316 |
Hungry | -0.311 |
leans | -0.307 |
Telescope | -0.303 |
Token | big |
Token ID | 1263 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
chief | +0.288 |
harness | +0.273 |
lex | +0.270 |
TO | +0.268 |
mins | +0.262 |
Neg logit contributions | |
Carbuncle | -0.278 |
certific | -0.277 |
Hungry | -0.274 |
leans | -0.270 |
Telescope | -0.267 |
Token | and |
Token ID | 290 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
Carbuncle | +0.117 |
certific | +0.117 |
Hungry | +0.115 |
leans | +0.113 |
Telescope | +0.112 |
Neg logit contributions | |
chief | -0.144 |
harness | -0.137 |
lex | -0.135 |
TO | -0.134 |
mins | -0.132 |
Token | rookie |
Token ID | 12302 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
Carbuncle | +0.135 |
certific | +0.135 |
Hungry | +0.133 |
leans | +0.131 |
Telescope | +0.129 |
Neg logit contributions | |
chief | -0.148 |
harness | -0.142 |
lex | -0.140 |
TO | -0.139 |
mins | -0.135 |
Token | De |
Token ID | 1024 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
Carbuncle | +0.292 |
certific | +0.292 |
Hungry | +0.287 |
leans | +0.284 |
Telescope | +0.281 |
Neg logit contributions | |
chief | -0.330 |
harness | -0.315 |
lex | -0.310 |
TO | -0.308 |
mins | -0.301 |
Token | � |
Token ID | 447 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
chief | +0.141 |
harness | +0.134 |
lex | +0.132 |
TO | +0.131 |
mins | +0.128 |
Neg logit contributions | |
Carbuncle | -0.160 |
certific | -0.159 |
Hungry | -0.158 |
leans | -0.156 |
Telescope | -0.154 |
Token | � |
Token ID | 247 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
Carbuncle | +0.153 |
certific | +0.153 |
Hungry | +0.151 |
leans | +0.149 |
Telescope | +0.147 |
Neg logit contributions | |
chief | -0.175 |
harness | -0.167 |
lex | -0.165 |
TO | -0.164 |
mins | -0.160 |
Token | Angelo |
Token ID | 45585 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
chief | +0.057 |
harness | +0.055 |
lex | +0.054 |
TO | +0.053 |
mins | +0.052 |
Neg logit contributions | |
Carbuncle | -0.052 |
certific | -0.052 |
Hungry | -0.052 |
leans | -0.051 |
Telescope | -0.050 |
Token | Henderson |
Token ID | 22016 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
chief | +1.301 |
harness | +1.236 |
lex | +1.224 |
TO | +1.203 |
mins | +1.184 |
Neg logit contributions | |
Carbuncle | -1.204 |
Hungry | -1.193 |
certific | -1.177 |
leans | -1.148 |
Telescope | -1.145 |
Token | also |
Token ID | 635 |
Feature activation | +9.12e-02 |
Pos logit contributions | |
Carbuncle | +0.206 |
certific | +0.206 |
Hungry | +0.202 |
leans | +0.200 |
Telescope | +0.197 |
Neg logit contributions | |
chief | -0.256 |
harness | -0.246 |
lex | -0.242 |
TO | -0.240 |
mins | -0.235 |
Token | produced |
Token ID | 4635 |
Feature activation | +7.29e-02 |
Pos logit contributions | |
Carbuncle | +0.055 |
certific | +0.055 |
Hungry | +0.054 |
leans | +0.053 |
Telescope | +0.053 |
Neg logit contributions | |
chief | -0.069 |
harness | -0.066 |
lex | -0.065 |
TO | -0.065 |
mins | -0.064 |
Token | strong |
Token ID | 1913 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Carbuncle | +0.042 |
certific | +0.042 |
Hungry | +0.041 |
leans | +0.040 |
Telescope | +0.040 |
Neg logit contributions | |
chief | -0.058 |
harness | -0.055 |
lex | -0.055 |
TO | -0.054 |
mins | -0.053 |
Token | runs |
Token ID | 4539 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
chief | +0.081 |
harness | +0.077 |
lex | +0.076 |
TO | +0.076 |
mins | +0.074 |
Neg logit contributions | |
Carbuncle | -0.073 |
certific | -0.073 |
Hungry | -0.072 |
leans | -0.071 |
Telescope | -0.070 |
Token | is |
Token ID | 318 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
Carbuncle | +0.106 |
certific | +0.105 |
Hungry | +0.104 |
leans | +0.103 |
Telescope | +0.101 |
Neg logit contributions | |
chief | -0.123 |
harness | -0.117 |
lex | -0.115 |
TO | -0.114 |
mins | -0.112 |
Token | hell |
Token ID | 5968 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
Carbuncle | +0.023 |
certific | +0.023 |
Hungry | +0.023 |
leans | +0.023 |
Telescope | +0.022 |
Neg logit contributions | |
chief | -0.028 |
harness | -0.027 |
lex | -0.026 |
TO | -0.026 |
mins | -0.026 |
Token | ," |
Token ID | 553 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
Carbuncle | +0.187 |
certific | +0.186 |
Hungry | +0.183 |
leans | +0.181 |
Telescope | +0.179 |
Neg logit contributions | |
chief | -0.213 |
harness | -0.203 |
lex | -0.200 |
TO | -0.199 |
mins | -0.195 |
Token | William |
Token ID | 3977 |
Feature activation | +9.06e-02 |
Pos logit contributions | |
Carbuncle | +0.034 |
certific | +0.034 |
Hungry | +0.034 |
leans | +0.033 |
Telescope | +0.033 |
Neg logit contributions | |
chief | -0.041 |
harness | -0.039 |
lex | -0.038 |
TO | -0.038 |
mins | -0.037 |
Token | Tec |
Token ID | 48257 |
Feature activation | +9.40e-01 |
Pos logit contributions | |
Carbuncle | +0.060 |
certific | +0.060 |
Hungry | +0.059 |
leans | +0.058 |
Telescope | +0.058 |
Neg logit contributions | |
chief | -0.064 |
harness | -0.061 |
lex | -0.060 |
TO | -0.059 |
mins | -0.058 |
Token | um |
Token ID | 388 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
Carbuncle | +0.611 |
certific | +0.610 |
Hungry | +0.600 |
Telescope | +0.586 |
leans | +0.586 |
Neg logit contributions | |
chief | -0.680 |
harness | -0.640 |
lex | -0.639 |
TO | -0.631 |
mins | -0.618 |
Token | se |
Token ID | 325 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
chief | +1.059 |
harness | +1.000 |
TO | +0.969 |
lex | +0.963 |
ODUCT | +0.935 |
Neg logit contributions | |
Carbuncle | -1.406 |
certific | -1.395 |
Hungry | -1.389 |
leans | -1.368 |
clair | -1.353 |
Token | h |
Token ID | 71 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
chief | +0.260 |
harness | +0.247 |
lex | +0.242 |
TO | +0.242 |
mins | +0.235 |
Neg logit contributions | |
Carbuncle | -0.281 |
certific | -0.279 |
Hungry | -0.277 |
leans | -0.274 |
Telescope | -0.270 |
Token | Sherman |
Token ID | 23198 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
Carbuncle | +0.352 |
certific | +0.352 |
Hungry | +0.345 |
leans | +0.341 |
Telescope | +0.336 |
Neg logit contributions | |
chief | -0.395 |
harness | -0.380 |
lex | -0.374 |
TO | -0.369 |
mins | -0.362 |
Token | once |
Token ID | 1752 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
Carbuncle | +0.041 |
certific | +0.041 |
Hungry | +0.040 |
leans | +0.040 |
Telescope | +0.039 |
Neg logit contributions | |
chief | -0.053 |
harness | -0.051 |
lex | -0.050 |
TO | -0.050 |
mins | -0.049 |
Token | said |
Token ID | 531 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
Carbuncle | +0.013 |
certific | +0.013 |
Hungry | +0.013 |
leans | +0.013 |
Telescope | +0.013 |
Neg logit contributions | |
chief | -0.017 |
harness | -0.016 |
lex | -0.016 |
TO | -0.016 |
mins | -0.016 |
Token | Fund |
Token ID | 7557 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
chief | +0.103 |
harness | +0.098 |
lex | +0.097 |
TO | +0.096 |
mins | +0.094 |
Neg logit contributions | |
Carbuncle | -0.087 |
certific | -0.087 |
Hungry | -0.086 |
leans | -0.085 |
Telescope | -0.084 |
Token | chief |
Token ID | 4039 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
chief | +0.207 |
harness | +0.199 |
lex | +0.196 |
TO | +0.195 |
mins | +0.191 |
Neg logit contributions | |
Carbuncle | -0.156 |
certific | -0.155 |
Hungry | -0.153 |
leans | -0.150 |
Telescope | -0.149 |
Token | Domin |
Token ID | 11817 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
chief | +0.272 |
harness | +0.256 |
lex | +0.252 |
TO | +0.250 |
mins | +0.242 |
Neg logit contributions | |
Carbuncle | -0.410 |
certific | -0.409 |
Hungry | -0.405 |
leans | -0.401 |
Telescope | -0.397 |
Token | ique |
Token ID | 2350 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
Carbuncle | +0.670 |
certific | +0.669 |
Hungry | +0.658 |
leans | +0.650 |
Telescope | +0.640 |
Neg logit contributions | |
chief | -0.632 |
harness | -0.599 |
lex | -0.597 |
TO | -0.587 |
mins | -0.570 |
Token | Strauss |
Token ID | 42971 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
Carbuncle | +0.128 |
certific | +0.127 |
Hungry | +0.126 |
leans | +0.124 |
Telescope | +0.122 |
Neg logit contributions | |
chief | -0.130 |
harness | -0.124 |
lex | -0.122 |
TO | -0.121 |
mins | -0.118 |
Token | - |
Token ID | 12 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
Carbuncle | +0.492 |
certific | +0.489 |
Hungry | +0.480 |
leans | +0.477 |
Telescope | +0.468 |
Neg logit contributions | |
chief | -0.574 |
harness | -0.551 |
lex | -0.538 |
TO | -0.536 |
mins | -0.523 |
Token | K |
Token ID | 42 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
chief | +1.566 |
harness | +1.533 |
lex | +1.502 |
TO | +1.482 |
mins | +1.471 |
Neg logit contributions | |
certific | -0.870 |
Hungry | -0.869 |
Carbuncle | -0.863 |
leans | -0.846 |
Telescope | -0.842 |
Token | ahn |
Token ID | 15386 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
chief | +0.386 |
harness | +0.373 |
lex | +0.365 |
TO | +0.364 |
mins | +0.357 |
Neg logit contributions | |
Carbuncle | -0.265 |
certific | -0.264 |
Hungry | -0.261 |
leans | -0.259 |
Telescope | -0.254 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
Carbuncle | +0.025 |
certific | +0.025 |
Hungry | +0.025 |
leans | +0.024 |
Telescope | +0.024 |
Neg logit contributions | |
chief | -0.029 |
harness | -0.028 |
lex | -0.027 |
TO | -0.027 |
mins | -0.026 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
chief | +0.151 |
harness | +0.145 |
lex | +0.143 |
TO | +0.142 |
mins | +0.139 |
Neg logit contributions | |
Carbuncle | -0.118 |
certific | -0.117 |
Hungry | -0.116 |
leans | -0.114 |
Telescope | -0.113 |
Token | SAN |
Token ID | 36753 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
Carbuncle | +0.010 |
certific | +0.010 |
Hungry | +0.010 |
leans | +0.010 |
Telescope | +0.010 |
Neg logit contributions | |
chief | -0.011 |
harness | -0.010 |
lex | -0.010 |
TO | -0.010 |
mins | -0.010 |
Token | in |
Token ID | 287 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
chief | +0.081 |
harness | +0.077 |
lex | +0.076 |
TO | +0.076 |
mins | +0.074 |
Neg logit contributions | |
Carbuncle | -0.070 |
certific | -0.069 |
Hungry | -0.068 |
leans | -0.067 |
Telescope | -0.067 |
Token | third |
Token ID | 2368 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
chief | +0.575 |
harness | +0.548 |
lex | +0.543 |
TO | +0.539 |
mins | +0.526 |
Neg logit contributions | |
Carbuncle | -0.482 |
certific | -0.481 |
Hungry | -0.476 |
leans | -0.466 |
Telescope | -0.462 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
chief | +0.299 |
harness | +0.285 |
lex | +0.282 |
TO | +0.280 |
mins | +0.274 |
Neg logit contributions | |
Carbuncle | -0.255 |
certific | -0.254 |
Hungry | -0.250 |
leans | -0.247 |
Telescope | -0.244 |
Token | Sky |
Token ID | 5274 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
chief | +0.114 |
harness | +0.109 |
lex | +0.107 |
TO | +0.106 |
mins | +0.104 |
Neg logit contributions | |
Carbuncle | -0.092 |
certific | -0.091 |
Hungry | -0.090 |
leans | -0.089 |
Telescope | -0.088 |
Token | Pro |
Token ID | 1041 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
chief | +0.157 |
harness | +0.150 |
lex | +0.148 |
TO | +0.147 |
mins | +0.144 |
Neg logit contributions | |
Carbuncle | -0.123 |
certific | -0.123 |
Hungry | -0.121 |
leans | -0.119 |
Telescope | -0.118 |
Token | cy |
Token ID | 948 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
chief | +0.168 |
harness | +0.163 |
lex | +0.162 |
TO | +0.161 |
mins | +0.159 |
Neg logit contributions | |
Carbuncle | -0.039 |
certific | -0.039 |
Hungry | -0.038 |
leans | -0.036 |
Telescope | -0.035 |
Token | cling |
Token ID | 8493 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
chief | +1.376 |
harness | +1.327 |
lex | +1.303 |
TO | +1.292 |
mins | +1.275 |
Neg logit contributions | |
Carbuncle | -1.043 |
certific | -1.033 |
Hungry | -1.029 |
leans | -1.027 |
Telescope | -1.000 |
Token | team |
Token ID | 1074 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
Carbuncle | +0.120 |
certific | +0.119 |
Hungry | +0.118 |
leans | +0.116 |
Telescope | +0.114 |
Neg logit contributions | |
chief | -0.154 |
harness | -0.147 |
lex | -0.145 |
TO | -0.144 |
mins | -0.141 |
Token | - |
Token ID | 12 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
chief | +0.016 |
harness | +0.015 |
lex | +0.015 |
TO | +0.015 |
mins | +0.015 |
Neg logit contributions | |
Carbuncle | -0.013 |
certific | -0.013 |
Hungry | -0.013 |
leans | -0.012 |
Telescope | -0.012 |
Token | mate |
Token ID | 9830 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
chief | +0.189 |
harness | +0.180 |
lex | +0.178 |
TO | +0.177 |
mins | +0.173 |
Neg logit contributions | |
Carbuncle | -0.165 |
certific | -0.164 |
Hungry | -0.162 |
leans | -0.160 |
Telescope | -0.158 |
Token | Chris |
Token ID | 5180 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
chief | +0.135 |
harness | +0.129 |
lex | +0.127 |
TO | +0.126 |
mins | +0.124 |
Neg logit contributions | |
Carbuncle | -0.123 |
certific | -0.123 |
Hungry | -0.121 |
leans | -0.120 |
Telescope | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
chief | +0.357 |
harness | +0.341 |
lex | +0.336 |
TO | +0.334 |
mins | +0.327 |
Neg logit contributions | |
Carbuncle | -0.306 |
certific | -0.305 |
Hungry | -0.302 |
leans | -0.296 |
Telescope | -0.294 |
Token | conservative |
Token ID | 5940 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
chief | +0.197 |
harness | +0.188 |
lex | +0.185 |
TO | +0.184 |
mins | +0.180 |
Neg logit contributions | |
Carbuncle | -0.177 |
certific | -0.177 |
Hungry | -0.174 |
leans | -0.172 |
Telescope | -0.170 |
Token | lobbying |
Token ID | 17502 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
chief | +0.250 |
harness | +0.240 |
lex | +0.237 |
TO | +0.235 |
mins | +0.231 |
Neg logit contributions | |
Carbuncle | -0.189 |
certific | -0.189 |
Hungry | -0.186 |
leans | -0.183 |
Telescope | -0.181 |
Token | group |
Token ID | 1448 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
chief | +0.051 |
harness | +0.049 |
lex | +0.048 |
TO | +0.048 |
mins | +0.047 |
Neg logit contributions | |
Carbuncle | -0.041 |
certific | -0.040 |
Hungry | -0.040 |
leans | -0.039 |
Telescope | -0.039 |
Token | Family |
Token ID | 7884 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
chief | +0.190 |
harness | +0.181 |
lex | +0.179 |
TO | +0.178 |
mins | +0.174 |
Neg logit contributions | |
Carbuncle | -0.159 |
certific | -0.159 |
Hungry | -0.157 |
leans | -0.154 |
Telescope | -0.153 |
Token | Research |
Token ID | 4992 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Carbuncle | +0.132 |
certific | +0.132 |
Hungry | +0.130 |
leans | +0.128 |
Telescope | +0.126 |
Neg logit contributions | |
chief | -0.187 |
harness | -0.180 |
lex | -0.177 |
TO | -0.176 |
mins | -0.173 |
Token | Council |
Token ID | 4281 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
chief | +1.437 |
harness | +1.380 |
lex | +1.367 |
TO | +1.352 |
mins | +1.341 |
Neg logit contributions | |
Hungry | -0.947 |
Carbuncle | -0.940 |
certific | -0.938 |
Telescope | -0.914 |
leans | -0.886 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
chief | +0.119 |
harness | +0.114 |
lex | +0.112 |
TO | +0.111 |
mins | +0.109 |
Neg logit contributions | |
Carbuncle | -0.103 |
certific | -0.103 |
Hungry | -0.101 |
leans | -0.100 |
Telescope | -0.099 |
Token |
|
Token ID | 198 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
chief | +0.015 |
harness | +0.014 |
lex | +0.014 |
TO | +0.014 |
mins | +0.014 |
Neg logit contributions | |
Carbuncle | -0.013 |
certific | -0.013 |
Hungry | -0.013 |
leans | -0.013 |
Telescope | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
Carbuncle | +0.010 |
certific | +0.010 |
Hungry | +0.010 |
leans | +0.009 |
Telescope | +0.009 |
Neg logit contributions | |
chief | -0.012 |
harness | -0.012 |
lex | -0.012 |
TO | -0.012 |
mins | -0.011 |
Token | In |
Token ID | 818 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
chief | +0.044 |
harness | +0.042 |
lex | +0.041 |
TO | +0.041 |
mins | +0.040 |
Neg logit contributions | |
Carbuncle | -0.041 |
certific | -0.041 |
Hungry | -0.040 |
leans | -0.040 |
Telescope | -0.039 |
Token | right |
Token ID | 826 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
chief | +0.341 |
harness | +0.324 |
lex | +0.321 |
TO | +0.318 |
mins | +0.312 |
Neg logit contributions | |
Carbuncle | -0.298 |
certific | -0.297 |
Hungry | -0.292 |
leans | -0.288 |
Telescope | -0.285 |
Token | to |
Token ID | 284 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
chief | +0.591 |
harness | +0.563 |
lex | +0.558 |
TO | +0.551 |
mins | +0.542 |
Neg logit contributions | |
Carbuncle | -0.507 |
certific | -0.506 |
Hungry | -0.498 |
leans | -0.492 |
Telescope | -0.485 |
Token | the |
Token ID | 262 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
chief | +0.598 |
harness | +0.570 |
lex | +0.566 |
TO | +0.561 |
mins | +0.551 |
Neg logit contributions | |
certific | -0.500 |
Carbuncle | -0.499 |
Hungry | -0.490 |
leans | -0.481 |
Telescope | -0.478 |
Token | equal |
Token ID | 4961 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
chief | +0.246 |
harness | +0.234 |
lex | +0.231 |
TO | +0.229 |
mins | +0.225 |
Neg logit contributions | |
Carbuncle | -0.227 |
certific | -0.227 |
Hungry | -0.224 |
leans | -0.220 |
Telescope | -0.218 |
Token | protection |
Token ID | 4800 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
chief | +0.301 |
harness | +0.285 |
lex | +0.282 |
TO | +0.280 |
mins | +0.274 |
Neg logit contributions | |
certific | -0.312 |
Carbuncle | -0.311 |
Hungry | -0.307 |
leans | -0.302 |
Telescope | -0.300 |
Token | and |
Token ID | 290 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
chief | +0.487 |
harness | +0.465 |
lex | +0.461 |
TO | +0.456 |
mins | +0.447 |
Neg logit contributions | |
Carbuncle | -0.418 |
certific | -0.418 |
Hungry | -0.411 |
leans | -0.406 |
Telescope | -0.400 |
Token | equal |
Token ID | 4961 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
chief | +1.239 |
harness | +1.171 |
lex | +1.168 |
TO | +1.168 |
mins | +1.145 |
Neg logit contributions | |
certific | -1.142 |
Carbuncle | -1.125 |
Hungry | -1.106 |
leans | -1.089 |
Telescope | -1.076 |
Token | benefit |
Token ID | 4414 |
Feature activation | -9.35e-01 |
Pos logit contributions | |
chief | +0.524 |
harness | +0.496 |
lex | +0.492 |
TO | +0.489 |
mins | +0.477 |
Neg logit contributions | |
certific | -0.528 |
Carbuncle | -0.524 |
Hungry | -0.516 |
leans | -0.509 |
Telescope | -0.504 |
Token | of |
Token ID | 286 |
Feature activation | -8.71e-01 |
Pos logit contributions | |
chief | +0.688 |
harness | +0.654 |
lex | +0.651 |
TO | +0.644 |
mins | +0.632 |
Neg logit contributions | |
certific | -0.587 |
Carbuncle | -0.586 |
Hungry | -0.576 |
leans | -0.568 |
Telescope | -0.560 |
Token | the |
Token ID | 262 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
chief | +0.671 |
harness | +0.641 |
lex | +0.634 |
TO | +0.630 |
mins | +0.618 |
Neg logit contributions | |
certific | -0.517 |
Carbuncle | -0.515 |
Hungry | -0.507 |
leans | -0.498 |
Telescope | -0.494 |
Token | law |
Token ID | 1099 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
chief | +0.684 |
harness | +0.652 |
lex | +0.644 |
TO | +0.643 |
mins | +0.632 |
Neg logit contributions | |
Carbuncle | -0.503 |
certific | -0.502 |
Hungry | -0.495 |
leans | -0.486 |
Telescope | -0.481 |
Token | , |
Token ID | 11 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
chief | +0.046 |
harness | +0.044 |
lex | +0.043 |
TO | +0.043 |
mins | +0.042 |
Neg logit contributions | |
Carbuncle | -0.040 |
certific | -0.040 |
Hungry | -0.039 |
leans | -0.039 |
Telescope | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
chief | +0.116 |
harness | +0.111 |
lex | +0.109 |
TO | +0.109 |
mins | +0.106 |
Neg logit contributions | |
Carbuncle | -0.099 |
certific | -0.099 |
Hungry | -0.098 |
leans | -0.096 |
Telescope | -0.095 |
Token | San |
Token ID | 2986 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
Carbuncle | +0.139 |
certific | +0.138 |
Hungry | +0.136 |
leans | +0.134 |
Telescope | +0.132 |
Neg logit contributions | |
chief | -0.180 |
harness | -0.172 |
lex | -0.170 |
TO | -0.169 |
mins | -0.165 |
Token | Antonio |
Token ID | 11366 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
chief | +0.195 |
harness | +0.183 |
lex | +0.179 |
TO | +0.178 |
mins | +0.172 |
Neg logit contributions | |
Carbuncle | -0.307 |
certific | -0.306 |
Hungry | -0.303 |
leans | -0.300 |
Telescope | -0.297 |
Token | Express |
Token ID | 10604 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
chief | +0.777 |
harness | +0.745 |
lex | +0.735 |
TO | +0.730 |
mins | +0.719 |
Neg logit contributions | |
Carbuncle | -0.562 |
certific | -0.558 |
Hungry | -0.550 |
leans | -0.539 |
Telescope | -0.536 |
Token | - |
Token ID | 12 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
chief | +0.365 |
harness | +0.349 |
lex | +0.343 |
TO | +0.342 |
mins | +0.335 |
Neg logit contributions | |
Carbuncle | -0.302 |
certific | -0.300 |
Hungry | -0.297 |
leans | -0.291 |
Telescope | -0.288 |
Token | News |
Token ID | 9980 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
chief | +0.543 |
harness | +0.504 |
lex | +0.458 |
TO | +0.442 |
mins | +0.437 |
Neg logit contributions | |
Carbuncle | -1.800 |
certific | -1.798 |
Hungry | -1.781 |
leans | -1.763 |
clair | -1.755 |
Token | , |
Token ID | 11 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
chief | +0.097 |
harness | +0.092 |
lex | +0.091 |
TO | +0.090 |
mins | +0.089 |
Neg logit contributions | |
Carbuncle | -0.084 |
certific | -0.084 |
Hungry | -0.083 |
leans | -0.082 |
Telescope | -0.081 |
Token | Go |
Token ID | 1514 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
chief | +0.159 |
harness | +0.152 |
lex | +0.150 |
TO | +0.149 |
mins | +0.145 |
Neg logit contributions | |
Carbuncle | -0.137 |
certific | -0.137 |
Hungry | -0.135 |
leans | -0.133 |
Telescope | -0.132 |
Token | N |
Token ID | 45 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
chief | +0.188 |
harness | +0.179 |
lex | +0.176 |
TO | +0.175 |
mins | +0.172 |
Neg logit contributions | |
Carbuncle | -0.168 |
certific | -0.167 |
Hungry | -0.165 |
leans | -0.163 |
Telescope | -0.161 |
Token | om |
Token ID | 296 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
chief | +0.128 |
harness | +0.122 |
lex | +0.120 |
TO | +0.119 |
mins | +0.117 |
Neg logit contributions | |
Carbuncle | -0.122 |
certific | -0.122 |
Hungry | -0.120 |
leans | -0.119 |
Telescope | -0.118 |
Token |
|
Token ID | 198 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
Carbuncle | +0.099 |
certific | +0.099 |
Hungry | +0.097 |
leans | +0.096 |
Telescope | +0.095 |
Neg logit contributions | |
chief | -0.121 |
harness | -0.116 |
lex | -0.114 |
TO | -0.113 |
mins | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
chief | +0.044 |
harness | +0.042 |
lex | +0.042 |
TO | +0.042 |
mins | +0.041 |
Neg logit contributions | |
Carbuncle | -0.035 |
certific | -0.035 |
Hungry | -0.034 |
leans | -0.034 |
Telescope | -0.034 |
Token | O |
Token ID | 46 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
Carbuncle | +0.144 |
certific | +0.144 |
Hungry | +0.142 |
leans | +0.140 |
Telescope | +0.138 |
Neg logit contributions | |
chief | -0.158 |
harness | -0.150 |
lex | -0.148 |
TO | -0.147 |
mins | -0.144 |
Token | tto |
Token ID | 33955 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
chief | +0.310 |
harness | +0.297 |
lex | +0.291 |
TO | +0.289 |
mins | +0.283 |
Neg logit contributions | |
Carbuncle | -0.277 |
certific | -0.277 |
Hungry | -0.274 |
leans | -0.270 |
Telescope | -0.266 |
Token | Warm |
Token ID | 25692 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
chief | +0.102 |
harness | +0.098 |
lex | +0.096 |
TO | +0.096 |
mins | +0.093 |
Neg logit contributions | |
Carbuncle | -0.095 |
certific | -0.094 |
Hungry | -0.093 |
leans | -0.092 |
Telescope | -0.091 |
Token | b |
Token ID | 65 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
Carbuncle | +0.023 |
certific | +0.023 |
Hungry | +0.023 |
leans | +0.022 |
Telescope | +0.022 |
Neg logit contributions | |
chief | -0.034 |
harness | -0.033 |
lex | -0.033 |
TO | -0.032 |
mins | -0.032 |
Token | ier |
Token ID | 959 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
certific | +1.408 |
Carbuncle | +1.399 |
Hungry | +1.356 |
leans | +1.350 |
Telescope | +1.330 |
Neg logit contributions | |
chief | -1.387 |
harness | -1.312 |
lex | -1.292 |
TO | -1.285 |
mins | -1.249 |
Token | , |
Token ID | 11 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
chief | +0.109 |
harness | +0.104 |
lex | +0.103 |
TO | +0.102 |
mins | +0.100 |
Neg logit contributions | |
Carbuncle | -0.093 |
certific | -0.093 |
Hungry | -0.092 |
leans | -0.090 |
Telescope | -0.089 |
Token | a |
Token ID | 257 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
Carbuncle | +0.164 |
certific | +0.164 |
Hungry | +0.161 |
leans | +0.160 |
Telescope | +0.158 |
Neg logit contributions | |
chief | -0.161 |
harness | -0.154 |
lex | -0.151 |
TO | -0.150 |
mins | -0.146 |
Token | 22 |
Token ID | 2534 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
Carbuncle | +0.149 |
certific | +0.149 |
Hungry | +0.147 |
leans | +0.145 |
Telescope | +0.143 |
Neg logit contributions | |
chief | -0.159 |
harness | -0.152 |
lex | -0.149 |
TO | -0.148 |
mins | -0.144 |
Token | - |
Token ID | 12 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
chief | +0.032 |
harness | +0.030 |
lex | +0.030 |
TO | +0.030 |
mins | +0.029 |
Neg logit contributions | |
Carbuncle | -0.027 |
certific | -0.026 |
Hungry | -0.026 |
leans | -0.026 |
Telescope | -0.025 |
Token | Night |
Token ID | 5265 |
Feature activation | +9.51e-01 |
Pos logit contributions | |
chief | +0.141 |
harness | +0.135 |
lex | +0.133 |
TO | +0.132 |
mins | +0.129 |
Neg logit contributions | |
Carbuncle | -0.122 |
certific | -0.121 |
Hungry | -0.120 |
leans | -0.118 |
Telescope | -0.117 |
Token | c |
Token ID | 66 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
certific | +0.569 |
Carbuncle | +0.561 |
Hungry | +0.552 |
leans | +0.546 |
Telescope | +0.539 |
Neg logit contributions | |
chief | -0.739 |
harness | -0.702 |
lex | -0.693 |
TO | -0.689 |
mins | -0.673 |
Token | rawler |
Token ID | 39464 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
certific | +0.322 |
Carbuncle | +0.321 |
Hungry | +0.317 |
leans | +0.313 |
Telescope | +0.311 |
Neg logit contributions | |
chief | -0.215 |
harness | -0.201 |
lex | -0.199 |
TO | -0.196 |
mins | -0.190 |
Token | , |
Token ID | 11 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
Carbuncle | +0.112 |
certific | +0.111 |
Hungry | +0.110 |
leans | +0.108 |
Telescope | +0.107 |
Neg logit contributions | |
chief | -0.127 |
harness | -0.121 |
lex | -0.119 |
TO | -0.119 |
mins | -0.116 |
Token | and |
Token ID | 290 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
chief | +0.201 |
harness | +0.192 |
lex | +0.189 |
TO | +0.188 |
mins | +0.184 |
Neg logit contributions | |
Carbuncle | -0.167 |
certific | -0.167 |
Hungry | -0.165 |
leans | -0.162 |
Telescope | -0.160 |
Token | Jub |
Token ID | 38075 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
chief | +0.210 |
harness | +0.200 |
lex | +0.197 |
TO | +0.196 |
mins | +0.192 |
Neg logit contributions | |
Carbuncle | -0.186 |
certific | -0.185 |
Hungry | -0.183 |
leans | -0.180 |
Telescope | -0.179 |
Token | ilee |
Token ID | 40626 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
certific | +1.293 |
Carbuncle | +1.279 |
Hungry | +1.275 |
leans | +1.262 |
Telescope | +1.252 |
Neg logit contributions | |
chief | -0.880 |
harness | -0.828 |
lex | -0.828 |
TO | -0.810 |
mins | -0.777 |
Token | hanging |
Token ID | 10938 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
Carbuncle | +0.024 |
certific | +0.024 |
Hungry | +0.024 |
leans | +0.023 |
Telescope | +0.023 |
Neg logit contributions | |
chief | -0.029 |
harness | -0.028 |
lex | -0.028 |
TO | -0.027 |
mins | -0.027 |
Token | out |
Token ID | 503 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
chief | +0.398 |
harness | +0.379 |
lex | +0.376 |
TO | +0.373 |
mins | +0.366 |
Neg logit contributions | |
Carbuncle | -0.320 |
certific | -0.318 |
Hungry | -0.315 |
leans | -0.310 |
Telescope | -0.306 |
Token | in |
Token ID | 287 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
chief | +0.221 |
harness | +0.210 |
lex | +0.208 |
TO | +0.206 |
mins | +0.202 |
Neg logit contributions | |
Carbuncle | -0.191 |
certific | -0.190 |
Hungry | -0.188 |
leans | -0.186 |
Telescope | -0.183 |
Token | a |
Token ID | 257 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
chief | +0.234 |
harness | +0.224 |
lex | +0.221 |
TO | +0.219 |
mins | +0.215 |
Neg logit contributions | |
Carbuncle | -0.186 |
certific | -0.186 |
Hungry | -0.183 |
leans | -0.180 |
Telescope | -0.178 |
Token | spokesman |
Token ID | 6523 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
Carbuncle | +0.208 |
certific | +0.207 |
Hungry | +0.202 |
leans | +0.197 |
Telescope | +0.193 |
Neg logit contributions | |
chief | -0.515 |
harness | -0.497 |
lex | -0.490 |
TO | -0.489 |
mins | -0.479 |
Token | : |
Token ID | 25 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
chief | +0.242 |
harness | +0.231 |
lex | +0.228 |
TO | +0.226 |
mins | +0.222 |
Neg logit contributions | |
Carbuncle | -0.215 |
certific | -0.214 |
Hungry | -0.212 |
leans | -0.209 |
Telescope | -0.207 |
Token | Clinton |
Token ID | 2605 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
certific | +0.673 |
Carbuncle | +0.670 |
Hungry | +0.655 |
leans | +0.644 |
Telescope | +0.636 |
Neg logit contributions | |
chief | -0.712 |
harness | -0.674 |
TO | -0.670 |
lex | -0.669 |
mins | -0.651 |
Token | staff |
Token ID | 3085 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Carbuncle | +0.405 |
certific | +0.402 |
Hungry | +0.395 |
leans | +0.388 |
Telescope | +0.386 |
Neg logit contributions | |
chief | -0.512 |
harness | -0.491 |
lex | -0.483 |
TO | -0.481 |
mins | -0.468 |
Token | are |
Token ID | 389 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
chief | +0.008 |
harness | +0.007 |
lex | +0.007 |
TO | +0.007 |
mins | +0.007 |
Neg logit contributions | |
Carbuncle | -0.006 |
certific | -0.006 |
Hungry | -0.006 |
leans | -0.006 |
Telescope | -0.006 |
Token | ' |
Token ID | 705 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
chief | +0.009 |
harness | +0.009 |
lex | +0.009 |
TO | +0.009 |
mins | +0.008 |
Neg logit contributions | |
Carbuncle | -0.008 |
certific | -0.008 |
Hungry | -0.008 |
leans | -0.008 |
Telescope | -0.008 |
Token | big |
Token ID | 14261 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
Carbuncle | +1.755 |
certific | +1.736 |
Hungry | +1.694 |
Telescope | +1.687 |
ayne | +1.659 |
Neg logit contributions | |
chief | -1.227 |
harness | -1.118 |
TO | -1.112 |
lex | -1.104 |
mins | -1.045 |
Token | gest |
Token ID | 3495 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Carbuncle | +0.296 |
certific | +0.294 |
Hungry | +0.289 |
leans | +0.285 |
Telescope | +0.282 |
Neg logit contributions | |
chief | -0.395 |
harness | -0.380 |
lex | -0.374 |
TO | -0.373 |
mins | -0.363 |
Token | a |
Token ID | 257 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
Carbuncle | +0.109 |
certific | +0.109 |
Hungry | +0.107 |
leans | +0.106 |
Telescope | +0.105 |
Neg logit contributions | |
chief | -0.130 |
harness | -0.125 |
lex | -0.123 |
TO | -0.122 |
mins | -0.119 |
Token | -- |
Token ID | 438 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
chief | +0.271 |
harness | +0.256 |
lex | +0.251 |
TO | +0.250 |
mins | +0.245 |
Neg logit contributions | |
Carbuncle | -0.312 |
certific | -0.311 |
Hungry | -0.308 |
leans | -0.305 |
Telescope | -0.301 |
Token | holes |
Token ID | 28439 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
Carbuncle | +0.042 |
certific | +0.042 |
Hungry | +0.042 |
leans | +0.041 |
Telescope | +0.041 |
Neg logit contributions | |
chief | -0.047 |
harness | -0.045 |
lex | -0.045 |
TO | -0.044 |
mins | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
chief | +0.181 |
harness | +0.173 |
lex | +0.171 |
TO | +0.169 |
mins | +0.166 |
Neg logit contributions | |
Carbuncle | -0.158 |
certific | -0.158 |
Hungry | -0.155 |
leans | -0.153 |
Telescope | -0.151 |
Token | Future |
Token ID | 10898 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
chief | +0.267 |
harness | +0.255 |
lex | +0.252 |
TO | +0.249 |
mins | +0.244 |
Neg logit contributions | |
Carbuncle | -0.235 |
certific | -0.234 |
Hungry | -0.232 |
leans | -0.227 |
Telescope | -0.225 |
Token | Hall |
Token ID | 4789 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
chief | +0.197 |
harness | +0.188 |
lex | +0.185 |
TO | +0.184 |
mins | +0.180 |
Neg logit contributions | |
Carbuncle | -0.169 |
certific | -0.169 |
Hungry | -0.167 |
leans | -0.164 |
Telescope | -0.162 |
Token | of |
Token ID | 286 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
chief | +0.209 |
harness | +0.200 |
lex | +0.197 |
TO | +0.196 |
mins | +0.191 |
Neg logit contributions | |
Carbuncle | -0.184 |
certific | -0.184 |
Hungry | -0.181 |
leans | -0.179 |
Telescope | -0.177 |
Token | Fame |
Token ID | 18864 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
Carbuncle | +0.072 |
certific | +0.072 |
Hungry | +0.071 |
leans | +0.070 |
Telescope | +0.069 |
Neg logit contributions | |
chief | -0.082 |
harness | -0.078 |
lex | -0.077 |
TO | -0.076 |
mins | -0.075 |
Token | slug |
Token ID | 31065 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Carbuncle | +0.540 |
certific | +0.536 |
Hungry | +0.528 |
leans | +0.522 |
Telescope | +0.515 |
Neg logit contributions | |
chief | -0.651 |
harness | -0.623 |
lex | -0.611 |
TO | -0.608 |
mins | -0.594 |
Token | ger |
Token ID | 1362 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
Carbuncle | +0.766 |
certific | +0.759 |
Hungry | +0.738 |
leans | +0.728 |
Telescope | +0.723 |
Neg logit contributions | |
chief | -1.375 |
harness | -1.312 |
lex | -1.298 |
TO | -1.295 |
mins | -1.270 |
Token | Stan |
Token ID | 7299 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
Carbuncle | +0.203 |
certific | +0.202 |
Hungry | +0.199 |
leans | +0.197 |
Telescope | +0.194 |
Neg logit contributions | |
chief | -0.228 |
harness | -0.218 |
lex | -0.215 |
TO | -0.213 |
mins | -0.208 |
Token | Mus |
Token ID | 2629 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
chief | +0.190 |
harness | +0.181 |
lex | +0.178 |
TO | +0.177 |
mins | +0.173 |
Neg logit contributions | |
Carbuncle | -0.190 |
certific | -0.189 |
Hungry | -0.187 |
leans | -0.185 |
Telescope | -0.183 |
Token | ial |
Token ID | 498 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
Carbuncle | +0.366 |
certific | +0.365 |
Hungry | +0.360 |
leans | +0.357 |
Telescope | +0.353 |
Neg logit contributions | |
chief | -0.251 |
harness | -0.235 |
lex | -0.231 |
TO | -0.228 |
mins | -0.223 |
Token | , |
Token ID | 11 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
Carbuncle | +0.253 |
certific | +0.252 |
Hungry | +0.248 |
leans | +0.245 |
Telescope | +0.242 |
Neg logit contributions | |
chief | -0.304 |
harness | -0.291 |
lex | -0.286 |
TO | -0.284 |
mins | -0.278 |
Token | provocative |
Token ID | 28695 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
Carbuncle | +0.027 |
certific | +0.027 |
Hungry | +0.026 |
leans | +0.026 |
Telescope | +0.026 |
Neg logit contributions | |
chief | -0.040 |
harness | -0.038 |
lex | -0.038 |
TO | -0.037 |
mins | -0.037 |
Token | , |
Token ID | 11 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
chief | +0.362 |
harness | +0.346 |
lex | +0.342 |
TO | +0.339 |
mins | +0.333 |
Neg logit contributions | |
Carbuncle | -0.311 |
certific | -0.310 |
Hungry | -0.306 |
leans | -0.302 |
Telescope | -0.298 |
Token | often |
Token ID | 1690 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
certific | +0.392 |
Carbuncle | +0.390 |
Hungry | +0.383 |
leans | +0.382 |
Telescope | +0.375 |
Neg logit contributions | |
chief | -0.536 |
harness | -0.512 |
lex | -0.504 |
TO | -0.501 |
mins | -0.490 |
Token | made |
Token ID | 925 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Carbuncle | +0.597 |
certific | +0.595 |
Hungry | +0.589 |
leans | +0.584 |
Telescope | +0.576 |
Neg logit contributions | |
chief | -0.677 |
harness | -0.647 |
lex | -0.641 |
TO | -0.632 |
mins | -0.615 |
Token | in |
Token ID | 287 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
Carbuncle | +0.734 |
certific | +0.723 |
Hungry | +0.718 |
leans | +0.707 |
Telescope | +0.702 |
Neg logit contributions | |
chief | -0.790 |
harness | -0.755 |
lex | -0.746 |
TO | -0.735 |
mins | -0.708 |
Token | j |
Token ID | 474 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
leans | +1.165 |
certific | +1.160 |
Carbuncle | +1.157 |
Hungry | +1.156 |
Telescope | +1.128 |
Neg logit contributions | |
chief | -1.405 |
harness | -1.350 |
lex | -1.345 |
TO | -1.334 |
mins | -1.277 |
Token | est |
Token ID | 395 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
certific | +1.356 |
Hungry | +1.341 |
Carbuncle | +1.339 |
leans | +1.318 |
Telescope | +1.314 |
Neg logit contributions | |
chief | -1.141 |
lex | -1.065 |
harness | -1.055 |
TO | -1.039 |
mins | -0.995 |
Token | , |
Token ID | 11 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
Carbuncle | +0.785 |
certific | +0.784 |
Hungry | +0.768 |
leans | +0.755 |
Telescope | +0.751 |
Neg logit contributions | |
chief | -0.813 |
harness | -0.771 |
TO | -0.760 |
lex | -0.756 |
mins | -0.731 |
Token | and |
Token ID | 290 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
certific | +0.549 |
Carbuncle | +0.546 |
Hungry | +0.537 |
leans | +0.534 |
Telescope | +0.526 |
Neg logit contributions | |
chief | -0.661 |
harness | -0.631 |
lex | -0.619 |
TO | -0.618 |
mins | -0.604 |
Token | largely |
Token ID | 5688 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
chief | +0.489 |
harness | +0.462 |
lex | +0.456 |
TO | +0.452 |
mins | +0.443 |
Neg logit contributions | |
Carbuncle | -0.527 |
certific | -0.523 |
Hungry | -0.518 |
leans | -0.510 |
Telescope | -0.507 |
Token | consist |
Token ID | 3473 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
Carbuncle | +0.191 |
certific | +0.190 |
Hungry | +0.188 |
leans | +0.186 |
Telescope | +0.183 |
Neg logit contributions | |
chief | -0.198 |
harness | -0.189 |
lex | -0.186 |
TO | -0.184 |
mins | -0.180 |
Token | The |
Token ID | 464 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
chief | +0.131 |
harness | +0.125 |
lex | +0.123 |
TO | +0.122 |
mins | +0.119 |
Neg logit contributions | |
Carbuncle | -0.125 |
certific | -0.125 |
Hungry | -0.123 |
leans | -0.122 |
Telescope | -0.120 |
Token | deal |
Token ID | 1730 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
chief | +0.140 |
harness | +0.133 |
lex | +0.131 |
TO | +0.130 |
mins | +0.128 |
Neg logit contributions | |
Carbuncle | -0.133 |
certific | -0.132 |
Hungry | -0.131 |
leans | -0.129 |
Telescope | -0.127 |
Token | will |
Token ID | 481 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
chief | +0.248 |
harness | +0.236 |
lex | +0.234 |
TO | +0.232 |
mins | +0.227 |
Neg logit contributions | |
Carbuncle | -0.215 |
certific | -0.214 |
Hungry | -0.211 |
leans | -0.208 |
Telescope | -0.206 |
Token | see |
Token ID | 766 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
chief | +0.204 |
harness | +0.193 |
lex | +0.191 |
TO | +0.190 |
mins | +0.185 |
Neg logit contributions | |
Carbuncle | -0.208 |
certific | -0.207 |
Hungry | -0.204 |
leans | -0.202 |
Telescope | -0.200 |
Token | the |
Token ID | 262 |
Feature activation | +9.17e-02 |
Pos logit contributions | |
chief | +0.110 |
harness | +0.105 |
lex | +0.104 |
TO | +0.103 |
mins | +0.101 |
Neg logit contributions | |
Carbuncle | -0.091 |
certific | -0.090 |
Hungry | -0.089 |
leans | -0.088 |
Telescope | -0.087 |
Token | 12 |
Token ID | 1105 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
Carbuncle | +0.059 |
certific | +0.058 |
Hungry | +0.058 |
leans | +0.057 |
Telescope | +0.056 |
Neg logit contributions | |
chief | -0.066 |
harness | -0.063 |
lex | -0.063 |
TO | -0.062 |
mins | -0.061 |
Token | B |
Token ID | 33 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
Carbuncle | +0.122 |
certific | +0.121 |
Hungry | +0.120 |
leans | +0.118 |
Telescope | +0.117 |
Neg logit contributions | |
chief | -0.140 |
harness | -0.134 |
lex | -0.132 |
TO | -0.131 |
mins | -0.128 |
Token | ET |
Token ID | 2767 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
Carbuncle | +0.323 |
certific | +0.321 |
Hungry | +0.317 |
leans | +0.312 |
Telescope | +0.308 |
Neg logit contributions | |
chief | -0.370 |
harness | -0.354 |
lex | -0.349 |
TO | -0.347 |
mins | -0.338 |
Token | logo |
Token ID | 11112 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
chief | +0.027 |
harness | +0.025 |
lex | +0.025 |
TO | +0.025 |
mins | +0.024 |
Neg logit contributions | |
Carbuncle | -0.024 |
certific | -0.024 |
Hungry | -0.024 |
leans | -0.024 |
Telescope | -0.023 |
Token | on |
Token ID | 319 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
Carbuncle | +0.149 |
certific | +0.148 |
Hungry | +0.146 |
leans | +0.144 |
Telescope | +0.142 |
Neg logit contributions | |
chief | -0.173 |
harness | -0.166 |
lex | -0.163 |
TO | -0.162 |
mins | -0.158 |
Token | the |
Token ID | 262 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
chief | +0.224 |
harness | +0.214 |
lex | +0.211 |
TO | +0.209 |
mins | +0.205 |
Neg logit contributions | |
Carbuncle | -0.187 |
certific | -0.187 |
Hungry | -0.184 |
leans | -0.181 |
Telescope | -0.179 |
Token | ` |
Token ID | 4600 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Carbuncle | +0.001 |
certific | +0.001 |
Hungry | +0.001 |
leans | +0.001 |
Telescope | +0.001 |
Neg logit contributions | |
chief | -0.001 |
harness | -0.001 |
lex | -0.001 |
TO | -0.001 |
mins | -0.001 |
Token | & |
Token ID | 5 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
Carbuncle | +0.072 |
certific | +0.072 |
Hungry | +0.071 |
leans | +0.070 |
Telescope | +0.069 |
Neg logit contributions | |
chief | -0.081 |
harness | -0.077 |
lex | -0.076 |
TO | -0.075 |
mins | -0.074 |
Token | ` |
Token ID | 63 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
Carbuncle | +0.132 |
certific | +0.131 |
Hungry | +0.129 |
leans | +0.128 |
Telescope | +0.126 |
Neg logit contributions | |
chief | -0.149 |
harness | -0.142 |
lex | -0.140 |
TO | -0.139 |
mins | -0.136 |
Token | ) |
Token ID | 8 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
Carbuncle | +0.224 |
certific | +0.224 |
Hungry | +0.220 |
leans | +0.217 |
Telescope | +0.215 |
Neg logit contributions | |
chief | -0.258 |
harness | -0.247 |
lex | -0.243 |
TO | -0.241 |
mins | -0.236 |
Token | at |
Token ID | 379 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Carbuncle | +0.219 |
certific | +0.219 |
Hungry | +0.215 |
leans | +0.212 |
Telescope | +0.210 |
Neg logit contributions | |
chief | -0.264 |
harness | -0.253 |
lex | -0.249 |
TO | -0.247 |
mins | -0.242 |
Token | some |
Token ID | 617 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
chief | +0.076 |
harness | +0.072 |
lex | +0.071 |
TO | +0.071 |
mins | +0.069 |
Neg logit contributions | |
Carbuncle | -0.069 |
certific | -0.069 |
Hungry | -0.068 |
leans | -0.067 |
Telescope | -0.067 |
Token | point |
Token ID | 966 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
Carbuncle | +0.021 |
certific | +0.021 |
Hungry | +0.021 |
leans | +0.021 |
Telescope | +0.020 |
Neg logit contributions | |
chief | -0.024 |
harness | -0.023 |
lex | -0.023 |
TO | -0.022 |
mins | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
Carbuncle | +0.102 |
certific | +0.102 |
Hungry | +0.100 |
leans | +0.099 |
Telescope | +0.098 |
Neg logit contributions | |
chief | -0.119 |
harness | -0.114 |
lex | -0.112 |
TO | -0.111 |
mins | -0.109 |
Token | make |
Token ID | 787 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
Carbuncle | +0.033 |
certific | +0.033 |
Hungry | +0.032 |
leans | +0.032 |
Telescope | +0.032 |
Neg logit contributions | |
chief | -0.038 |
harness | -0.036 |
lex | -0.036 |
TO | -0.036 |
mins | -0.035 |
Token | ` |
Token ID | 4600 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
Carbuncle | +0.121 |
certific | +0.121 |
Hungry | +0.119 |
leans | +0.118 |
Telescope | +0.116 |
Neg logit contributions | |
chief | -0.142 |
harness | -0.136 |
lex | -0.134 |
TO | -0.133 |
mins | -0.130 |
Token | Board |
Token ID | 29828 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
Carbuncle | +0.221 |
certific | +0.221 |
Hungry | +0.217 |
leans | +0.215 |
Telescope | +0.212 |
Neg logit contributions | |
chief | -0.230 |
harness | -0.219 |
lex | -0.217 |
TO | -0.215 |
mins | -0.210 |
Token | � |
Token ID | 247 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
chief | +0.204 |
harness | +0.194 |
lex | +0.191 |
TO | +0.190 |
mins | +0.186 |
Neg logit contributions | |
Carbuncle | -0.179 |
certific | -0.179 |
Hungry | -0.176 |
leans | -0.174 |
Telescope | -0.172 |
Token | s |
Token ID | 82 |
Feature activation | +9.75e-02 |
Pos logit contributions | |
Carbuncle | +0.080 |
certific | +0.080 |
Hungry | +0.079 |
leans | +0.078 |
Telescope | +0.077 |
Neg logit contributions | |
chief | -0.093 |
harness | -0.089 |
lex | -0.088 |
TO | -0.087 |
mins | -0.085 |
Token | producing |
Token ID | 9194 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
Carbuncle | +0.062 |
certific | +0.062 |
Hungry | +0.061 |
leans | +0.061 |
Telescope | +0.060 |
Neg logit contributions | |
chief | -0.071 |
harness | -0.067 |
lex | -0.066 |
TO | -0.066 |
mins | -0.064 |
Token | results |
Token ID | 2482 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
chief | +0.125 |
harness | +0.119 |
lex | +0.118 |
TO | +0.117 |
mins | +0.114 |
Neg logit contributions | |
Carbuncle | -0.106 |
certific | -0.106 |
Hungry | -0.104 |
leans | -0.103 |
Telescope | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
chief | +0.135 |
harness | +0.128 |
lex | +0.127 |
TO | +0.126 |
mins | +0.123 |
Neg logit contributions | |
Carbuncle | -0.115 |
certific | -0.114 |
Hungry | -0.113 |
leans | -0.111 |
Telescope | -0.110 |
Token | Eight |
Token ID | 18087 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
chief | +0.022 |
harness | +0.021 |
lex | +0.021 |
TO | +0.021 |
mins | +0.020 |
Neg logit contributions | |
Carbuncle | -0.019 |
certific | -0.019 |
Hungry | -0.018 |
leans | -0.018 |
Telescope | -0.018 |
Token | city |
Token ID | 1748 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
Carbuncle | +0.118 |
certific | +0.117 |
Hungry | +0.115 |
leans | +0.114 |
Telescope | +0.112 |
Neg logit contributions | |
chief | -0.160 |
harness | -0.153 |
lex | -0.151 |
TO | -0.150 |
mins | -0.147 |
Token | departments |
Token ID | 13346 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
Carbuncle | +0.008 |
certific | +0.008 |
Hungry | +0.008 |
leans | +0.008 |
Telescope | +0.008 |
Neg logit contributions | |
chief | -0.011 |
harness | -0.010 |
lex | -0.010 |
TO | -0.010 |
mins | -0.010 |
Token | oversee |
Token ID | 19158 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
chief | +0.191 |
harness | +0.183 |
lex | +0.180 |
TO | +0.179 |
mins | +0.175 |
Neg logit contributions | |
Carbuncle | -0.161 |
certific | -0.161 |
Hungry | -0.158 |
leans | -0.156 |
Telescope | -0.154 |
Token | at |
Token ID | 379 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
Carbuncle | +0.126 |
certific | +0.126 |
Hungry | +0.124 |
leans | +0.122 |
Telescope | +0.121 |
Neg logit contributions | |
chief | -0.149 |
harness | -0.143 |
lex | -0.141 |
TO | -0.140 |
mins | -0.137 |
Token | least |
Token ID | 1551 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Carbuncle | +0.027 |
certific | +0.027 |
Hungry | +0.027 |
leans | +0.027 |
Telescope | +0.026 |
Neg logit contributions | |
chief | -0.026 |
harness | -0.025 |
lex | -0.025 |
TO | -0.024 |
mins | -0.024 |
Token | Ant |
Token ID | 13217 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
chief | +0.137 |
harness | +0.131 |
lex | +0.129 |
TO | +0.128 |
mins | +0.125 |
Neg logit contributions | |
Carbuncle | -0.120 |
certific | -0.120 |
Hungry | -0.118 |
leans | -0.117 |
Telescope | -0.115 |
Token | ip |
Token ID | 541 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
chief | +0.039 |
harness | +0.037 |
lex | +0.037 |
TO | +0.037 |
mins | +0.036 |
Neg logit contributions | |
Carbuncle | -0.040 |
certific | -0.040 |
Hungry | -0.039 |
leans | -0.039 |
Telescope | -0.039 |
Token | ov |
Token ID | 709 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
Carbuncle | +0.174 |
certific | +0.174 |
Hungry | +0.171 |
leans | +0.169 |
Telescope | +0.167 |
Neg logit contributions | |
chief | -0.170 |
harness | -0.161 |
lex | -0.159 |
TO | -0.157 |
mins | -0.154 |
Token | , |
Token ID | 11 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
chief | +0.108 |
harness | +0.104 |
lex | +0.102 |
TO | +0.101 |
mins | +0.099 |
Neg logit contributions | |
Carbuncle | -0.096 |
certific | -0.096 |
Hungry | -0.094 |
leans | -0.093 |
Telescope | -0.092 |
Token | Mikhail |
Token ID | 42040 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
chief | +0.198 |
harness | +0.189 |
lex | +0.186 |
TO | +0.184 |
mins | +0.180 |
Neg logit contributions | |
Carbuncle | -0.188 |
certific | -0.187 |
Hungry | -0.185 |
leans | -0.182 |
Telescope | -0.180 |
Token | Al |
Token ID | 978 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
chief | +0.344 |
harness | +0.327 |
lex | +0.321 |
TO | +0.319 |
mins | +0.312 |
Neg logit contributions | |
Carbuncle | -0.355 |
certific | -0.354 |
Hungry | -0.351 |
leans | -0.346 |
Telescope | -0.343 |
Token | "] |
Token ID | 8973 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Carbuncle | +0.027 |
certific | +0.027 |
Hungry | +0.026 |
leans | +0.026 |
Telescope | +0.026 |
Neg logit contributions | |
chief | -0.032 |
harness | -0.031 |
lex | -0.030 |
TO | -0.030 |
mins | -0.029 |
Token | [ |
Token ID | 685 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
Carbuncle | +0.059 |
certific | +0.059 |
Hungry | +0.058 |
leans | +0.057 |
Telescope | +0.056 |
Neg logit contributions | |
chief | -0.086 |
harness | -0.082 |
lex | -0.081 |
TO | -0.081 |
mins | -0.079 |
Token | Result |
Token ID | 23004 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
chief | +0.208 |
harness | +0.200 |
lex | +0.198 |
TO | +0.197 |
mins | +0.194 |
Neg logit contributions | |
Carbuncle | -0.107 |
certific | -0.107 |
Hungry | -0.105 |
leans | -0.103 |
Telescope | -0.102 |
Token | " |
Token ID | 366 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
chief | +0.543 |
harness | +0.519 |
lex | +0.513 |
TO | +0.509 |
mins | +0.500 |
Neg logit contributions | |
Carbuncle | -0.427 |
certific | -0.423 |
Hungry | -0.422 |
leans | -0.414 |
Telescope | -0.410 |
Token | 1 |
Token ID | 16 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
chief | +0.037 |
harness | +0.036 |
lex | +0.035 |
TO | +0.035 |
mins | +0.034 |
Neg logit contributions | |
Carbuncle | -0.026 |
certific | -0.026 |
Hungry | -0.025 |
leans | -0.025 |
Telescope | -0.024 |
Token | my |
Token ID | 616 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
chief | +0.059 |
harness | +0.056 |
lex | +0.056 |
TO | +0.055 |
mins | +0.054 |
Neg logit contributions | |
Carbuncle | -0.050 |
certific | -0.050 |
Hungry | -0.049 |
leans | -0.048 |
Telescope | -0.048 |
Token | colleague |
Token ID | 16008 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
Carbuncle | +0.096 |
certific | +0.096 |
Hungry | +0.095 |
leans | +0.093 |
Telescope | +0.092 |
Neg logit contributions | |
chief | -0.104 |
harness | -0.100 |
lex | -0.098 |
TO | -0.097 |
mins | -0.095 |
Token | Nicole |
Token ID | 25176 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
chief | +0.058 |
harness | +0.056 |
lex | +0.055 |
TO | +0.054 |
mins | +0.053 |
Neg logit contributions | |
Carbuncle | -0.055 |
certific | -0.054 |
Hungry | -0.054 |
leans | -0.053 |
Telescope | -0.052 |
Token | O |
Token ID | 440 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
chief | +0.253 |
harness | +0.242 |
lex | +0.238 |
TO | +0.237 |
mins | +0.231 |
Neg logit contributions | |
Carbuncle | -0.226 |
certific | -0.225 |
Hungry | -0.222 |
leans | -0.219 |
Telescope | -0.216 |
Token | zer |
Token ID | 9107 |
Feature activation | +5.96e-01 |
Pos logit contributions | |
chief | +0.562 |
harness | +0.541 |
lex | +0.530 |
TO | +0.528 |
mins | +0.518 |
Neg logit contributions | |
Carbuncle | -0.399 |
certific | -0.399 |
Hungry | -0.393 |
leans | -0.388 |
Telescope | -0.381 |
Token | has |
Token ID | 468 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
Carbuncle | +0.385 |
certific | +0.383 |
Hungry | +0.377 |
leans | +0.371 |
Telescope | +0.368 |
Neg logit contributions | |
chief | -0.431 |
harness | -0.410 |
lex | -0.404 |
TO | -0.401 |
mins | -0.393 |
Token | discussed |
Token ID | 6693 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Carbuncle | +0.270 |
certific | +0.269 |
Hungry | +0.265 |
leans | +0.262 |
Telescope | +0.258 |
Neg logit contributions | |
chief | -0.333 |
harness | -0.319 |
lex | -0.314 |
TO | -0.312 |
mins | -0.305 |
Token | , |
Token ID | 11 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
Carbuncle | +0.073 |
certific | +0.073 |
Hungry | +0.072 |
leans | +0.071 |
Telescope | +0.070 |
Neg logit contributions | |
chief | -0.082 |
harness | -0.078 |
lex | -0.077 |
TO | -0.076 |
mins | -0.075 |
Token | there |
Token ID | 612 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
chief | +0.019 |
harness | +0.019 |
lex | +0.018 |
TO | +0.018 |
mins | +0.018 |
Neg logit contributions | |
Carbuncle | -0.016 |
certific | -0.015 |
Hungry | -0.015 |
leans | -0.015 |
Telescope | -0.015 |
Token | is |
Token ID | 318 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
Carbuncle | +0.027 |
certific | +0.027 |
Hungry | +0.027 |
leans | +0.026 |
Telescope | +0.026 |
Neg logit contributions | |
chief | -0.035 |
harness | -0.033 |
lex | -0.033 |
TO | -0.033 |
mins | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
Carbuncle | +0.007 |
certific | +0.007 |
Hungry | +0.006 |
leans | +0.006 |
Telescope | +0.006 |
Neg logit contributions | |
chief | -0.008 |
harness | -0.008 |
lex | -0.008 |
TO | -0.007 |
mins | -0.007 |
Token | get |
Token ID | 651 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
chief | +0.332 |
harness | +0.316 |
lex | +0.313 |
TO | +0.311 |
mins | +0.304 |
Neg logit contributions | |
Carbuncle | -0.270 |
certific | -0.269 |
Hungry | -0.266 |
leans | -0.261 |
Telescope | -0.259 |
Token | out |
Token ID | 503 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
chief | +0.103 |
harness | +0.099 |
lex | +0.097 |
TO | +0.097 |
mins | +0.095 |
Neg logit contributions | |
Carbuncle | -0.087 |
certific | -0.087 |
Hungry | -0.086 |
leans | -0.085 |
Telescope | -0.083 |
Token | there |
Token ID | 612 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
chief | +0.113 |
harness | +0.108 |
lex | +0.106 |
TO | +0.105 |
mins | +0.103 |
Neg logit contributions | |
Carbuncle | -0.095 |
certific | -0.094 |
Hungry | -0.093 |
leans | -0.092 |
Telescope | -0.091 |
Token | in |
Token ID | 287 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
chief | +0.213 |
harness | +0.203 |
lex | +0.201 |
TO | +0.199 |
mins | +0.195 |
Neg logit contributions | |
Carbuncle | -0.172 |
certific | -0.172 |
Hungry | -0.169 |
leans | -0.167 |
Telescope | -0.165 |
Token | a |
Token ID | 257 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
Carbuncle | +0.139 |
certific | +0.138 |
Hungry | +0.136 |
leans | +0.134 |
Telescope | +0.133 |
Neg logit contributions | |
chief | -0.172 |
harness | -0.165 |
lex | -0.162 |
TO | -0.161 |
mins | -0.158 |
Token | game |
Token ID | 983 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
chief | +0.220 |
harness | +0.210 |
lex | +0.207 |
TO | +0.206 |
mins | +0.202 |
Neg logit contributions | |
Carbuncle | -0.186 |
certific | -0.186 |
Hungry | -0.183 |
leans | -0.180 |
Telescope | -0.178 |
Token | situation |
Token ID | 3074 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
chief | +0.252 |
harness | +0.240 |
lex | +0.237 |
TO | +0.236 |
mins | +0.231 |
Neg logit contributions | |
Carbuncle | -0.212 |
certific | -0.211 |
Hungry | -0.208 |
leans | -0.205 |
Telescope | -0.202 |
Token | and |
Token ID | 290 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
chief | +0.075 |
harness | +0.071 |
lex | +0.071 |
TO | +0.070 |
mins | +0.069 |
Neg logit contributions | |
Carbuncle | -0.063 |
certific | -0.063 |
Hungry | -0.062 |
leans | -0.062 |
Telescope | -0.061 |
Token | you |
Token ID | 345 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
chief | +0.224 |
harness | +0.214 |
lex | +0.211 |
TO | +0.210 |
mins | +0.206 |
Neg logit contributions | |
Carbuncle | -0.181 |
certific | -0.181 |
Hungry | -0.179 |
leans | -0.176 |
Telescope | -0.174 |
Token | have |
Token ID | 423 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
chief | +0.232 |
harness | +0.221 |
lex | +0.219 |
TO | +0.217 |
mins | +0.213 |
Neg logit contributions | |
Carbuncle | -0.197 |
certific | -0.197 |
Hungry | -0.194 |
leans | -0.191 |
Telescope | -0.189 |
Token | to |
Token ID | 284 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
chief | +0.179 |
harness | +0.171 |
lex | +0.169 |
TO | +0.168 |
mins | +0.164 |
Neg logit contributions | |
Carbuncle | -0.153 |
certific | -0.153 |
Hungry | -0.151 |
leans | -0.148 |
Telescope | -0.146 |
Token | promised |
Token ID | 8072 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
chief | +0.080 |
harness | +0.077 |
lex | +0.076 |
TO | +0.075 |
mins | +0.073 |
Neg logit contributions | |
Carbuncle | -0.064 |
certific | -0.064 |
Hungry | -0.063 |
leans | -0.062 |
Telescope | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
Carbuncle | +0.075 |
certific | +0.075 |
Hungry | +0.074 |
leans | +0.073 |
Telescope | +0.072 |
Neg logit contributions | |
chief | -0.089 |
harness | -0.085 |
lex | -0.083 |
TO | -0.083 |
mins | -0.081 |
Token | � |
Token ID | 564 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Carbuncle | +0.097 |
certific | +0.097 |
Hungry | +0.095 |
leans | +0.094 |
Telescope | +0.093 |
Neg logit contributions | |
chief | -0.114 |
harness | -0.109 |
lex | -0.107 |
TO | -0.106 |
mins | -0.104 |
Token | � |
Token ID | 250 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
Carbuncle | +0.010 |
certific | +0.010 |
Hungry | +0.009 |
leans | +0.009 |
Telescope | +0.009 |
Neg logit contributions | |
chief | -0.010 |
harness | -0.010 |
lex | -0.009 |
TO | -0.009 |
mins | -0.009 |
Token | There |
Token ID | 1858 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
chief | +0.041 |
harness | +0.039 |
lex | +0.038 |
TO | +0.038 |
mins | +0.037 |
Neg logit contributions | |
Carbuncle | -0.032 |
certific | -0.032 |
Hungry | -0.032 |
leans | -0.031 |
Telescope | -0.031 |
Token | will |
Token ID | 481 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
chief | +0.071 |
harness | +0.068 |
lex | +0.067 |
TO | +0.066 |
mins | +0.065 |
Neg logit contributions | |
Carbuncle | -0.054 |
certific | -0.054 |
Hungry | -0.053 |
leans | -0.052 |
Telescope | -0.052 |
Token | be |
Token ID | 307 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
chief | +0.290 |
harness | +0.277 |
lex | +0.273 |
TO | +0.271 |
mins | +0.265 |
Neg logit contributions | |
Carbuncle | -0.246 |
certific | -0.246 |
Hungry | -0.242 |
leans | -0.239 |
Telescope | -0.236 |
Token | good |
Token ID | 922 |
Feature activation | -9.33e-02 |
Pos logit contributions | |
chief | +0.180 |
harness | +0.172 |
lex | +0.170 |
TO | +0.168 |
mins | +0.165 |
Neg logit contributions | |
Carbuncle | -0.153 |
certific | -0.153 |
Hungry | -0.151 |
leans | -0.149 |
Telescope | -0.147 |
Token | times |
Token ID | 1661 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
chief | +0.067 |
harness | +0.064 |
lex | +0.063 |
TO | +0.063 |
mins | +0.061 |
Neg logit contributions | |
Carbuncle | -0.060 |
certific | -0.060 |
Hungry | -0.059 |
leans | -0.058 |
Telescope | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
chief | +0.031 |
harness | +0.029 |
lex | +0.029 |
TO | +0.029 |
mins | +0.028 |
Neg logit contributions | |
Carbuncle | -0.026 |
certific | -0.026 |
Hungry | -0.026 |
leans | -0.025 |
Telescope | -0.025 |
Token | hard |
Token ID | 1327 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
Carbuncle | +0.133 |
certific | +0.132 |
Hungry | +0.131 |
leans | +0.129 |
Telescope | +0.128 |
Neg logit contributions | |
chief | -0.120 |
harness | -0.114 |
lex | -0.112 |
TO | -0.111 |
mins | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
chief | +0.203 |
harness | +0.194 |
lex | +0.191 |
TO | +0.190 |
mins | +0.186 |
Neg logit contributions | |
Carbuncle | -0.176 |
certific | -0.176 |
Hungry | -0.173 |
leans | -0.171 |
Telescope | -0.169 |
Token | lung |
Token ID | 12317 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
chief | +0.026 |
harness | +0.025 |
lex | +0.025 |
TO | +0.024 |
mins | +0.024 |
Neg logit contributions | |
Carbuncle | -0.022 |
certific | -0.022 |
Hungry | -0.022 |
leans | -0.021 |
Telescope | -0.021 |
Token | es |
Token ID | 274 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
chief | +0.029 |
harness | +0.027 |
lex | +0.027 |
TO | +0.027 |
mins | +0.026 |
Neg logit contributions | |
Carbuncle | -0.024 |
certific | -0.024 |
Hungry | -0.024 |
leans | -0.023 |
Telescope | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
chief | +0.097 |
harness | +0.093 |
lex | +0.091 |
TO | +0.091 |
mins | +0.089 |
Neg logit contributions | |
Carbuncle | -0.084 |
certific | -0.084 |
Hungry | -0.083 |
leans | -0.082 |
Telescope | -0.081 |
Token | ( |
Token ID | 357 |
Feature activation | -3.98e-03 |
Pos logit contributions | |
chief | +0.261 |
harness | +0.249 |
lex | +0.246 |
TO | +0.244 |
mins | +0.239 |
Neg logit contributions | |
Carbuncle | -0.216 |
certific | -0.214 |
Hungry | -0.212 |
leans | -0.208 |
Telescope | -0.206 |
Token | Wa |
Token ID | 33484 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
chief | +0.003 |
harness | +0.002 |
lex | +0.002 |
TO | +0.002 |
mins | +0.002 |
Neg logit contributions | |
Carbuncle | -0.003 |
certific | -0.003 |
Hungry | -0.003 |
leans | -0.003 |
Telescope | -0.003 |
Token | ub |
Token ID | 549 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
chief | +0.522 |
harness | +0.496 |
TO | +0.483 |
lex | +0.482 |
mins | +0.470 |
Neg logit contributions | |
Carbuncle | -0.679 |
certific | -0.678 |
Hungry | -0.672 |
leans | -0.667 |
clair | -0.658 |
Token | ges |
Token ID | 3212 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Carbuncle | +0.029 |
certific | +0.029 |
Hungry | +0.028 |
leans | +0.028 |
Telescope | +0.027 |
Neg logit contributions | |
chief | -0.034 |
harness | -0.032 |
lex | -0.032 |
TO | -0.031 |
mins | -0.031 |
Token | hig |
Token ID | 25196 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
Carbuncle | +0.096 |
certific | +0.095 |
Hungry | +0.094 |
leans | +0.092 |
Telescope | +0.091 |
Neg logit contributions | |
chief | -0.131 |
harness | -0.125 |
lex | -0.124 |
TO | -0.123 |
mins | -0.120 |
Token | Rice |
Token ID | 13823 |
Feature activation | +7.68e-03 |
Pos logit contributions | |
chief | +0.016 |
harness | +0.015 |
lex | +0.015 |
TO | +0.015 |
mins | +0.015 |
Neg logit contributions | |
Carbuncle | -0.018 |
certific | -0.018 |
Hungry | -0.018 |
leans | -0.017 |
Telescope | -0.017 |
Token | / |
Token ID | 14 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Carbuncle | +0.004 |
certific | +0.004 |
Hungry | +0.004 |
leans | +0.004 |
Telescope | +0.004 |
Neg logit contributions | |
chief | -0.006 |
harness | -0.006 |
lex | -0.006 |
TO | -0.006 |
mins | -0.006 |
Token | lost |
Token ID | 2626 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
chief | +0.380 |
harness | +0.361 |
lex | +0.356 |
TO | +0.353 |
mins | +0.345 |
Neg logit contributions | |
Carbuncle | -0.386 |
certific | -0.385 |
Hungry | -0.381 |
leans | -0.375 |
Telescope | -0.371 |
Token | leader |
Token ID | 3554 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
chief | +0.345 |
harness | +0.327 |
lex | +0.323 |
TO | +0.320 |
mins | +0.313 |
Neg logit contributions | |
Carbuncle | -0.367 |
certific | -0.365 |
Hungry | -0.361 |
leans | -0.356 |
Telescope | -0.353 |
Token | O |
Token ID | 440 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
chief | +0.184 |
harness | +0.176 |
lex | +0.173 |
TO | +0.172 |
mins | +0.168 |
Neg logit contributions | |
Carbuncle | -0.153 |
certific | -0.153 |
Hungry | -0.151 |
leans | -0.148 |
Telescope | -0.147 |
Token | der |
Token ID | 1082 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
chief | +0.219 |
harness | +0.202 |
lex | +0.194 |
TO | +0.191 |
mins | +0.184 |
Neg logit contributions | |
Carbuncle | -0.573 |
certific | -0.572 |
Hungry | -0.568 |
leans | -0.564 |
Telescope | -0.558 |
Token | us |
Token ID | 385 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
chief | +0.007 |
harness | +0.006 |
lex | +0.006 |
TO | +0.006 |
mins | +0.006 |
Neg logit contributions | |
Carbuncle | -0.007 |
certific | -0.007 |
Hungry | -0.007 |
leans | -0.007 |
Telescope | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
chief | +0.017 |
harness | +0.016 |
lex | +0.016 |
TO | +0.016 |
mins | +0.016 |
Neg logit contributions | |
Carbuncle | -0.011 |
certific | -0.010 |
Hungry | -0.010 |
leans | -0.010 |
Telescope | -0.010 |
Token | they |
Token ID | 484 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
chief | +0.141 |
harness | +0.134 |
lex | +0.133 |
TO | +0.132 |
mins | +0.129 |
Neg logit contributions | |
Carbuncle | -0.117 |
certific | -0.116 |
Hungry | -0.115 |
leans | -0.113 |
Telescope | -0.112 |
Token | pay |
Token ID | 1414 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
Carbuncle | +0.052 |
certific | +0.052 |
Hungry | +0.051 |
leans | +0.050 |
Telescope | +0.050 |
Neg logit contributions | |
chief | -0.065 |
harness | -0.062 |
lex | -0.061 |
TO | -0.061 |
mins | -0.059 |
Token | ed |
Token ID | 276 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
chief | +0.411 |
harness | +0.392 |
lex | +0.387 |
TO | +0.384 |
mins | +0.376 |
Neg logit contributions | |
Carbuncle | -0.380 |
certific | -0.379 |
Hungry | -0.374 |
leans | -0.368 |
Telescope | -0.364 |
Token | homage |
Token ID | 31044 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
chief | +0.368 |
harness | +0.351 |
lex | +0.346 |
TO | +0.344 |
mins | +0.337 |
Neg logit contributions | |
Carbuncle | -0.326 |
certific | -0.324 |
Hungry | -0.321 |
leans | -0.315 |
Telescope | -0.312 |
Token | to |
Token ID | 284 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
chief | +0.035 |
harness | +0.033 |
lex | +0.033 |
TO | +0.033 |
mins | +0.032 |
Neg logit contributions | |
Carbuncle | -0.031 |
certific | -0.031 |
Hungry | -0.031 |
leans | -0.030 |
Telescope | -0.030 |
Token | To |
Token ID | 1675 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
Carbuncle | +0.098 |
certific | +0.098 |
Hungry | +0.097 |
leans | +0.096 |
Telescope | +0.095 |
Neg logit contributions | |
chief | -0.097 |
harness | -0.092 |
lex | -0.090 |
TO | -0.090 |
mins | -0.088 |
Token | pp |
Token ID | 381 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
chief | +0.349 |
harness | +0.334 |
lex | +0.329 |
TO | +0.327 |
mins | +0.320 |
Neg logit contributions | |
Carbuncle | -0.281 |
certific | -0.279 |
Hungry | -0.275 |
leans | -0.273 |
Telescope | -0.268 |
Token | as |
Token ID | 355 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Carbuncle | +0.131 |
certific | +0.131 |
Hungry | +0.129 |
leans | +0.128 |
Telescope | +0.126 |
Neg logit contributions | |
chief | -0.142 |
harness | -0.135 |
lex | -0.133 |
TO | -0.132 |
mins | -0.129 |
Token | chief |
Token ID | 4039 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
Carbuncle | +0.088 |
certific | +0.088 |
Hungry | +0.086 |
leans | +0.085 |
Telescope | +0.084 |
Neg logit contributions | |
chief | -0.102 |
harness | -0.098 |
lex | -0.096 |
TO | -0.095 |
mins | -0.093 |
Token | of |
Token ID | 286 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
chief | +0.145 |
harness | +0.138 |
lex | +0.136 |
TO | +0.135 |
mins | +0.132 |
Neg logit contributions | |
Carbuncle | -0.124 |
certific | -0.124 |
Hungry | -0.122 |
leans | -0.120 |
Telescope | -0.119 |
Token | staff |
Token ID | 3085 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
Carbuncle | +0.224 |
certific | +0.224 |
Hungry | +0.220 |
leans | +0.217 |
Telescope | +0.214 |
Neg logit contributions | |
chief | -0.271 |
harness | -0.259 |
lex | -0.256 |
TO | -0.254 |
mins | -0.249 |
Token | reflects |
Token ID | 12497 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
chief | +0.174 |
harness | +0.166 |
lex | +0.164 |
TO | +0.163 |
mins | +0.159 |
Neg logit contributions | |
Carbuncle | -0.147 |
certific | -0.146 |
Hungry | -0.144 |
leans | -0.142 |
Telescope | -0.140 |
Token | her |
Token ID | 607 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
Carbuncle | +0.196 |
certific | +0.195 |
Hungry | +0.192 |
leans | +0.190 |
Telescope | +0.187 |
Neg logit contributions | |
chief | -0.246 |
harness | -0.235 |
lex | -0.232 |
TO | -0.230 |
mins | -0.226 |
Token | strong |
Token ID | 1913 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
chief | +0.090 |
harness | +0.086 |
lex | +0.084 |
TO | +0.084 |
mins | +0.082 |
Neg logit contributions | |
Carbuncle | -0.075 |
certific | -0.075 |
Hungry | -0.074 |
leans | -0.073 |
Telescope | -0.072 |
Token | personal |
Token ID | 2614 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
Carbuncle | +0.099 |
certific | +0.099 |
Hungry | +0.098 |
leans | +0.096 |
Telescope | +0.095 |
Neg logit contributions | |
chief | -0.116 |
harness | -0.111 |
lex | -0.109 |
TO | -0.108 |
mins | -0.106 |
Token | ties |
Token ID | 8470 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
chief | +0.044 |
harness | +0.042 |
lex | +0.042 |
TO | +0.041 |
mins | +0.041 |
Neg logit contributions | |
Carbuncle | -0.037 |
certific | -0.037 |
Hungry | -0.036 |
leans | -0.036 |
Telescope | -0.035 |
Token | slack |
Token ID | 30740 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
Carbuncle | +0.062 |
certific | +0.062 |
Hungry | +0.061 |
leans | +0.060 |
Telescope | +0.059 |
Neg logit contributions | |
chief | -0.076 |
harness | -0.073 |
lex | -0.072 |
TO | -0.072 |
mins | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -8.70e-02 |
Pos logit contributions | |
Carbuncle | +0.201 |
certific | +0.201 |
Hungry | +0.198 |
leans | +0.195 |
Telescope | +0.193 |
Neg logit contributions | |
chief | -0.231 |
harness | -0.221 |
lex | -0.218 |
TO | -0.216 |
mins | -0.212 |
Token | The |
Token ID | 383 |
Feature activation | -8.31e-03 |
Pos logit contributions | |
chief | +0.063 |
harness | +0.061 |
lex | +0.060 |
TO | +0.059 |
mins | +0.058 |
Neg logit contributions | |
Carbuncle | -0.055 |
certific | -0.055 |
Hungry | -0.054 |
leans | -0.054 |
Telescope | -0.053 |
Token | Center |
Token ID | 3337 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
chief | +0.006 |
harness | +0.006 |
lex | +0.006 |
TO | +0.006 |
mins | +0.006 |
Neg logit contributions | |
Carbuncle | -0.005 |
certific | -0.005 |
Hungry | -0.005 |
leans | -0.005 |
Telescope | -0.005 |
Token | for |
Token ID | 329 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
Carbuncle | +0.095 |
certific | +0.094 |
Hungry | +0.093 |
leans | +0.092 |
Telescope | +0.091 |
Neg logit contributions | |
chief | -0.106 |
harness | -0.102 |
lex | -0.100 |
TO | -0.099 |
mins | -0.097 |
Token | Respons |
Token ID | 20549 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
chief | +0.442 |
harness | +0.420 |
lex | +0.414 |
TO | +0.411 |
mins | +0.402 |
Neg logit contributions | |
Carbuncle | -0.424 |
certific | -0.423 |
Hungry | -0.419 |
leans | -0.412 |
Telescope | -0.410 |
Token | ive |
Token ID | 425 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
chief | +0.830 |
harness | +0.771 |
lex | +0.742 |
TO | +0.739 |
mins | +0.731 |
Neg logit contributions | |
Carbuncle | -1.505 |
certific | -1.497 |
Hungry | -1.488 |
leans | -1.487 |
reating | -1.479 |
Token | Politics |
Token ID | 17554 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
Carbuncle | +0.071 |
certific | +0.071 |
Hungry | +0.069 |
leans | +0.069 |
Telescope | +0.068 |
Neg logit contributions | |
chief | -0.094 |
harness | -0.090 |
lex | -0.088 |
TO | -0.088 |
mins | -0.086 |
Token | has |
Token ID | 468 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
Carbuncle | +0.169 |
certific | +0.168 |
Hungry | +0.166 |
leans | +0.164 |
Telescope | +0.162 |
Neg logit contributions | |
chief | -0.200 |
harness | -0.192 |
lex | -0.189 |
TO | -0.187 |
mins | -0.183 |
Token | the |
Token ID | 262 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
chief | +0.021 |
harness | +0.020 |
lex | +0.020 |
TO | +0.020 |
mins | +0.019 |
Neg logit contributions | |
Carbuncle | -0.017 |
certific | -0.017 |
Hungry | -0.017 |
leans | -0.017 |
Telescope | -0.017 |
Token | full |
Token ID | 1336 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
Carbuncle | +0.119 |
certific | +0.118 |
Hungry | +0.117 |
leans | +0.115 |
Telescope | +0.114 |
Neg logit contributions | |
chief | -0.123 |
harness | -0.117 |
lex | -0.116 |
TO | -0.115 |
mins | -0.112 |
Token | by |
Token ID | 416 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
chief | +0.040 |
harness | +0.038 |
lex | +0.038 |
TO | +0.038 |
mins | +0.037 |
Neg logit contributions | |
Carbuncle | -0.032 |
certific | -0.032 |
Hungry | -0.032 |
leans | -0.031 |
Telescope | -0.031 |
Token | going |
Token ID | 1016 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
chief | +0.162 |
harness | +0.154 |
lex | +0.153 |
TO | +0.151 |
mins | +0.148 |
Neg logit contributions | |
Carbuncle | -0.136 |
certific | -0.136 |
Hungry | -0.134 |
leans | -0.132 |
Telescope | -0.130 |
Token | here |
Token ID | 994 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
chief | +0.083 |
harness | +0.079 |
lex | +0.078 |
TO | +0.077 |
mins | +0.076 |
Neg logit contributions | |
Carbuncle | -0.074 |
certific | -0.074 |
Hungry | -0.073 |
leans | -0.072 |
Telescope | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
chief | +0.148 |
harness | +0.141 |
lex | +0.139 |
TO | +0.138 |
mins | +0.135 |
Neg logit contributions | |
Carbuncle | -0.126 |
certific | -0.125 |
Hungry | -0.123 |
leans | -0.122 |
Telescope | -0.120 |
Token | But |
Token ID | 887 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
Carbuncle | +0.516 |
certific | +0.515 |
leans | +0.504 |
Hungry | +0.501 |
Telescope | +0.494 |
Neg logit contributions | |
chief | -0.510 |
harness | -0.485 |
lex | -0.479 |
TO | -0.477 |
mins | -0.462 |
Token | please |
Token ID | 3387 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
chief | +0.523 |
harness | +0.495 |
lex | +0.491 |
TO | +0.490 |
mins | +0.474 |
Neg logit contributions | |
certific | -0.522 |
Carbuncle | -0.517 |
Hungry | -0.514 |
leans | -0.501 |
Telescope | -0.499 |
Token | note |
Token ID | 3465 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
chief | +1.038 |
harness | +0.968 |
TO | +0.957 |
lex | +0.944 |
mins | +0.935 |
Neg logit contributions | |
certific | -1.511 |
Carbuncle | -1.509 |
Hungry | -1.468 |
Telescope | -1.440 |
ensis | -1.434 |
Token | that |
Token ID | 326 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
chief | +0.288 |
harness | +0.275 |
lex | +0.271 |
TO | +0.269 |
mins | +0.264 |
Neg logit contributions | |
Carbuncle | -0.245 |
certific | -0.245 |
Hungry | -0.240 |
leans | -0.237 |
Telescope | -0.235 |
Token | my |
Token ID | 616 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
chief | +0.091 |
harness | +0.087 |
lex | +0.086 |
TO | +0.086 |
mins | +0.084 |
Neg logit contributions | |
Carbuncle | -0.075 |
certific | -0.075 |
Hungry | -0.074 |
leans | -0.073 |
Telescope | -0.072 |
Token | tweets |
Token ID | 12665 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
chief | +0.753 |
lex | +0.704 |
harness | +0.701 |
TO | +0.699 |
mins | +0.672 |
Neg logit contributions | |
certific | -1.015 |
Carbuncle | -1.014 |
Hungry | -1.006 |
leans | -0.993 |
Telescope | -0.984 |
Token | are |
Token ID | 389 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
chief | +0.312 |
harness | +0.298 |
lex | +0.294 |
TO | +0.292 |
mins | +0.286 |
Neg logit contributions | |
Carbuncle | -0.238 |
certific | -0.237 |
Hungry | -0.234 |
leans | -0.231 |
Telescope | -0.228 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
Carbuncle | +0.228 |
certific | +0.227 |
Hungry | +0.224 |
leans | +0.221 |
Telescope | +0.218 |
Neg logit contributions | |
chief | -0.263 |
harness | -0.252 |
lex | -0.248 |
TO | -0.246 |
mins | -0.241 |
Token | goalkeeper |
Token ID | 26318 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
chief | +0.013 |
harness | +0.012 |
lex | +0.012 |
TO | +0.012 |
mins | +0.011 |
Neg logit contributions | |
Carbuncle | -0.012 |
certific | -0.012 |
Hungry | -0.011 |
leans | -0.011 |
Telescope | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
chief | +0.020 |
harness | +0.019 |
lex | +0.019 |
TO | +0.019 |
mins | +0.018 |
Neg logit contributions | |
Carbuncle | -0.017 |
certific | -0.017 |
Hungry | -0.017 |
leans | -0.017 |
Telescope | -0.016 |
Token | Jasper |
Token ID | 42992 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
chief | +0.057 |
harness | +0.055 |
lex | +0.054 |
TO | +0.053 |
mins | +0.052 |
Neg logit contributions | |
Carbuncle | -0.048 |
certific | -0.048 |
Hungry | -0.048 |
leans | -0.047 |
Telescope | -0.046 |
Token | C |
Token ID | 327 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
chief | +0.451 |
harness | +0.429 |
lex | +0.424 |
TO | +0.419 |
mins | +0.411 |
Neg logit contributions | |
Carbuncle | -0.444 |
certific | -0.438 |
Hungry | -0.436 |
leans | -0.430 |
Telescope | -0.426 |
Token | ill |
Token ID | 359 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
chief | +1.728 |
harness | +1.656 |
lex | +1.608 |
TO | +1.599 |
mins | +1.586 |
Neg logit contributions | |
Carbuncle | -1.110 |
certific | -1.077 |
Hungry | -1.071 |
clair | -1.060 |
leans | -1.056 |
Token | essen |
Token ID | 44483 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
chief | +1.343 |
harness | +1.299 |
TO | +1.259 |
lex | +1.247 |
mins | +1.227 |
Neg logit contributions | |
Carbuncle | -0.916 |
certific | -0.898 |
Hungry | -0.888 |
leans | -0.885 |
clair | -0.872 |
Token | , |
Token ID | 11 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Carbuncle | +0.131 |
certific | +0.131 |
Hungry | +0.129 |
leans | +0.127 |
Telescope | +0.125 |
Neg logit contributions | |
chief | -0.154 |
harness | -0.148 |
lex | -0.146 |
TO | -0.144 |
mins | -0.141 |
Token | who |
Token ID | 508 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
Carbuncle | +0.088 |
certific | +0.087 |
Hungry | +0.086 |
leans | +0.085 |
Telescope | +0.084 |
Neg logit contributions | |
chief | -0.105 |
harness | -0.100 |
lex | -0.099 |
TO | -0.098 |
mins | -0.096 |
Token | started |
Token ID | 2067 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
Carbuncle | +0.019 |
certific | +0.019 |
Hungry | +0.019 |
leans | +0.018 |
Telescope | +0.018 |
Neg logit contributions | |
chief | -0.023 |
harness | -0.022 |
lex | -0.022 |
TO | -0.022 |
mins | -0.022 |
Token | very |
Token ID | 845 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
Carbuncle | +0.128 |
certific | +0.128 |
Hungry | +0.126 |
leans | +0.124 |
Telescope | +0.123 |
Neg logit contributions | |
chief | -0.149 |
harness | -0.142 |
lex | -0.140 |
TO | -0.139 |
mins | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
Carbuncle | +0.039 |
certific | +0.039 |
Hungry | +0.039 |
leans | +0.038 |
Telescope | +0.038 |
Neg logit contributions | |
chief | -0.044 |
harness | -0.042 |
lex | -0.042 |
TO | -0.041 |
mins | -0.041 |
Token | George |
Token ID | 4502 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
Carbuncle | +0.104 |
certific | +0.104 |
Hungry | +0.102 |
leans | +0.101 |
Telescope | +0.099 |
Neg logit contributions | |
chief | -0.128 |
harness | -0.122 |
lex | -0.120 |
TO | -0.120 |
mins | -0.117 |
Token | Stephan |
Token ID | 20800 |
Feature activation | +6.86e-02 |
Pos logit contributions | |
certific | +0.284 |
Carbuncle | +0.284 |
Hungry | +0.278 |
leans | +0.275 |
Telescope | +0.271 |
Neg logit contributions | |
chief | -0.416 |
harness | -0.400 |
lex | -0.395 |
TO | -0.392 |
mins | -0.385 |
Token | opoulos |
Token ID | 20338 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Carbuncle | +0.042 |
certific | +0.042 |
Hungry | +0.041 |
leans | +0.040 |
Telescope | +0.040 |
Neg logit contributions | |
chief | -0.052 |
harness | -0.050 |
lex | -0.049 |
TO | -0.049 |
mins | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
Carbuncle | +0.183 |
certific | +0.182 |
Hungry | +0.180 |
leans | +0.177 |
Telescope | +0.175 |
Neg logit contributions | |
chief | -0.219 |
harness | -0.210 |
lex | -0.207 |
TO | -0.205 |
mins | -0.201 |
Token | Diane |
Token ID | 26542 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
chief | +0.029 |
harness | +0.027 |
lex | +0.027 |
TO | +0.027 |
mins | +0.026 |
Neg logit contributions | |
Carbuncle | -0.026 |
certific | -0.026 |
Hungry | -0.026 |
leans | -0.025 |
Telescope | -0.025 |
Token | Sawyer |
Token ID | 42371 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
chief | +1.138 |
harness | +1.079 |
TO | +1.068 |
lex | +1.059 |
mins | +1.032 |
Neg logit contributions | |
Carbuncle | -1.083 |
Hungry | -1.066 |
certific | -1.059 |
leans | -1.050 |
Telescope | -1.034 |
Token | are |
Token ID | 389 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
Carbuncle | +0.087 |
certific | +0.087 |
Hungry | +0.085 |
leans | +0.084 |
Telescope | +0.083 |
Neg logit contributions | |
chief | -0.107 |
harness | -0.102 |
lex | -0.101 |
TO | -0.100 |
mins | -0.098 |
Token | at |
Token ID | 379 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
chief | +0.205 |
harness | +0.195 |
lex | +0.193 |
TO | +0.192 |
mins | +0.187 |
Neg logit contributions | |
Carbuncle | -0.185 |
certific | -0.185 |
Hungry | -0.182 |
leans | -0.180 |
Telescope | -0.178 |
Token | the |
Token ID | 262 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
Carbuncle | +0.051 |
certific | +0.050 |
Hungry | +0.050 |
leans | +0.049 |
Telescope | +0.048 |
Neg logit contributions | |
chief | -0.061 |
harness | -0.059 |
lex | -0.058 |
TO | -0.057 |
mins | -0.056 |
Token | helm |
Token ID | 18030 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
Carbuncle | +0.400 |
certific | +0.399 |
Hungry | +0.392 |
leans | +0.389 |
Telescope | +0.382 |
Neg logit contributions | |
chief | -0.446 |
harness | -0.426 |
lex | -0.419 |
TO | -0.415 |
mins | -0.407 |
Token | Sutton |
Token ID | 39964 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
chief | +0.229 |
harness | +0.218 |
lex | +0.215 |
TO | +0.214 |
mins | +0.210 |
Neg logit contributions | |
Carbuncle | -0.190 |
certific | -0.189 |
Hungry | -0.187 |
leans | -0.184 |
Telescope | -0.182 |
Token | , |
Token ID | 11 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
Carbuncle | +0.102 |
certific | +0.102 |
Hungry | +0.100 |
leans | +0.099 |
Telescope | +0.098 |
Neg logit contributions | |
chief | -0.116 |
harness | -0.111 |
lex | -0.110 |
TO | -0.109 |
mins | -0.106 |
Token | probably |
Token ID | 2192 |
Feature activation | -8.13e-04 |
Pos logit contributions | |
chief | +0.013 |
harness | +0.013 |
lex | +0.012 |
TO | +0.012 |
mins | +0.012 |
Neg logit contributions | |
Carbuncle | -0.011 |
certific | -0.011 |
Hungry | -0.011 |
leans | -0.011 |
Telescope | -0.011 |
Token | no |
Token ID | 645 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
chief | +0.001 |
harness | +0.001 |
lex | +0.001 |
TO | +0.001 |
mins | +0.001 |
Neg logit contributions | |
Carbuncle | -0.001 |
certific | -0.001 |
Hungry | -0.001 |
leans | -0.000 |
Telescope | -0.000 |
Token | Kenny |
Token ID | 22102 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
Carbuncle | +0.148 |
certific | +0.148 |
Hungry | +0.145 |
leans | +0.144 |
Telescope | +0.142 |
Neg logit contributions | |
chief | -0.161 |
harness | -0.154 |
lex | -0.152 |
TO | -0.150 |
mins | -0.147 |
Token | Dal |
Token ID | 12348 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
chief | +0.527 |
harness | +0.507 |
lex | +0.501 |
TO | +0.498 |
mins | +0.489 |
Neg logit contributions | |
Carbuncle | -0.299 |
certific | -0.296 |
Hungry | -0.294 |
leans | -0.287 |
Telescope | -0.283 |
Token | g |
Token ID | 70 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
chief | +0.992 |
harness | +0.966 |
lex | +0.941 |
TO | +0.931 |
mins | +0.898 |
Neg logit contributions | |
Carbuncle | -1.231 |
certific | -1.230 |
Hungry | -1.215 |
leans | -1.206 |
Telescope | -1.191 |
Token | lish |
Token ID | 1836 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
certific | +0.341 |
Carbuncle | +0.341 |
Hungry | +0.336 |
leans | +0.332 |
Telescope | +0.329 |
Neg logit contributions | |
chief | -0.230 |
harness | -0.216 |
lex | -0.213 |
TO | -0.209 |
mins | -0.204 |
Token | either |
Token ID | 2035 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
Carbuncle | +0.085 |
certific | +0.084 |
Hungry | +0.083 |
leans | +0.082 |
Telescope | +0.081 |
Neg logit contributions | |
chief | -0.097 |
harness | -0.092 |
lex | -0.091 |
TO | -0.090 |
mins | -0.088 |
Token | , |
Token ID | 11 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
chief | +0.023 |
harness | +0.022 |
lex | +0.021 |
TO | +0.021 |
mins | +0.021 |
Neg logit contributions | |
Carbuncle | -0.020 |
certific | -0.020 |
Hungry | -0.019 |
leans | -0.019 |
Telescope | -0.019 |
Token | because |
Token ID | 780 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
Carbuncle | +0.011 |
certific | +0.011 |
Hungry | +0.011 |
leans | +0.011 |
Telescope | +0.011 |
Neg logit contributions | |
chief | -0.013 |
harness | -0.012 |
lex | -0.012 |
TO | -0.012 |
mins | -0.012 |