Token | scrap |
Token ID | 15881 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
Shinra | +0.327 |
phrine | +0.310 |
Lowe | +0.308 |
cens | +0.299 |
AFB | +0.294 |
Neg logit contributions | |
️ | -0.349 |
ilan | -0.340 |
andowski | -0.326 |
ulhu | -0.325 |
otle | -0.315 |
Token | ed |
Token ID | 276 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
️ | +0.258 |
ilan | +0.249 |
andowski | +0.242 |
ulhu | +0.241 |
otle | +0.235 |
Neg logit contributions | |
Shinra | -0.184 |
phrine | -0.173 |
Lowe | -0.171 |
cens | -0.168 |
AFB | -0.163 |
Token | by |
Token ID | 416 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
Shinra | +1.737 |
Lowe | +1.636 |
phrine | +1.616 |
cens | +1.566 |
AFB | +1.527 |
Neg logit contributions | |
️ | -2.455 |
ilan | -2.368 |
andowski | -2.318 |
ulhu | -2.310 |
otle | -2.242 |
Token | from |
Token ID | 422 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
Shinra | +2.005 |
Lowe | +1.908 |
phrine | +1.908 |
cens | +1.817 |
AFB | +1.793 |
Neg logit contributions | |
️ | -2.686 |
ilan | -2.591 |
andowski | -2.538 |
ulhu | -2.525 |
otle | -2.444 |
Token | deal |
Token ID | 1730 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
️ | +1.245 |
ilan | +1.208 |
andowski | +1.167 |
ulhu | +1.151 |
otle | +1.131 |
Neg logit contributions | |
Shinra | -1.103 |
phrine | -1.048 |
Lowe | -1.028 |
cens | -1.022 |
AFB | -0.988 |
Token | to |
Token ID | 284 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Shinra | +0.180 |
phrine | +0.170 |
Lowe | +0.168 |
cens | +0.165 |
AFB | +0.160 |
Neg logit contributions | |
️ | -0.239 |
ilan | -0.231 |
andowski | -0.224 |
ulhu | -0.221 |
otle | -0.217 |
Token | deal |
Token ID | 1730 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Lowe | +7.095 |
clipping | +6.820 |
Shinra | +6.567 |
premier | +6.450 |
phrine | +6.344 |
Neg logit contributions | |
andowski | -11.896 |
ulhu | -11.606 |
️ | -11.361 |
ilan | -11.312 |
otle | -10.665 |
Token | , |
Token ID | 11 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
Shinra | +0.952 |
phrine | +0.901 |
Lowe | +0.893 |
cens | +0.876 |
AFB | +0.850 |
Neg logit contributions | |
️ | -1.188 |
ilan | -1.152 |
andowski | -1.117 |
ulhu | -1.110 |
otle | -1.085 |
Token | drilling |
Token ID | 19747 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
Shinra | +2.832 |
Lowe | +2.723 |
phrine | +2.683 |
clipping | +2.595 |
cens | +2.559 |
Neg logit contributions | |
️ | -3.705 |
ilan | -3.591 |
ulhu | -3.511 |
andowski | -3.479 |
otle | -3.382 |
Token | hundreds |
Token ID | 5179 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
Shinra | +1.815 |
Lowe | +1.721 |
phrine | +1.710 |
cens | +1.653 |
AFB | +1.620 |
Neg logit contributions | |
️ | -2.325 |
ilan | -2.254 |
ulhu | -2.188 |
andowski | -2.184 |
otle | -2.115 |
Token | of |
Token ID | 286 |
Feature activation | -1.08e-04 |
Pos logit contributions | |
️ | +1.960 |
ilan | +1.902 |
andowski | +1.844 |
ulhu | +1.816 |
otle | +1.785 |
Neg logit contributions | |
Shinra | -1.665 |
phrine | -1.581 |
Lowe | -1.557 |
cens | -1.537 |
AFB | -1.489 |
Token | took |
Token ID | 1718 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
️ | +0.530 |
ilan | +0.513 |
andowski | +0.495 |
ulhu | +0.490 |
otle | +0.477 |
Neg logit contributions | |
Shinra | -0.538 |
phrine | -0.512 |
Lowe | -0.507 |
cens | -0.500 |
AFB | -0.486 |
Token | Michael |
Token ID | 3899 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
️ | +0.135 |
ilan | +0.131 |
ulhu | +0.127 |
andowski | +0.127 |
otle | +0.122 |
Neg logit contributions | |
Shinra | -0.112 |
Lowe | -0.105 |
phrine | -0.104 |
cens | -0.101 |
AFB | -0.099 |
Token | on |
Token ID | 319 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
️ | +2.963 |
ilan | +2.839 |
andowski | +2.761 |
ulhu | +2.684 |
otle | +2.661 |
Neg logit contributions | |
Shinra | -3.040 |
phrine | -2.910 |
cens | -2.849 |
Lowe | -2.837 |
AFB | -2.747 |
Token | point |
Token ID | 966 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
️ | +3.219 |
ilan | +3.104 |
andowski | +3.012 |
ulhu | +2.930 |
otle | +2.914 |
Neg logit contributions | |
Shinra | -2.949 |
phrine | -2.818 |
cens | -2.734 |
Lowe | -2.726 |
AFB | -2.622 |
Token | - |
Token ID | 12 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Shinra | +1.901 |
Lowe | +1.804 |
phrine | +1.800 |
cens | +1.760 |
AFB | +1.706 |
Neg logit contributions | |
️ | -2.524 |
ilan | -2.478 |
ulhu | -2.420 |
andowski | -2.390 |
otle | -2.314 |
Token | by |
Token ID | 1525 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
️ | +1.493 |
ilan | +1.447 |
andowski | +1.398 |
ulhu | +1.371 |
otle | +1.360 |
Neg logit contributions | |
Shinra | -1.410 |
phrine | -1.346 |
Lowe | -1.316 |
cens | -1.301 |
AFB | -1.271 |
Token | - |
Token ID | 12 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
Lowe | +5.493 |
clipping | +5.359 |
phrine | +5.323 |
AFB | +5.270 |
Shinra | +5.253 |
Neg logit contributions | |
️ | -12.132 |
ulhu | -12.037 |
ilan | -12.014 |
andowski | -11.976 |
otle | -11.194 |
Token | point |
Token ID | 4122 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
cens | +4.528 |
Shinra | +4.394 |
phrine | +4.348 |
bj | +4.188 |
Lowe | +4.176 |
Neg logit contributions | |
️ | -7.075 |
ulhu | -6.762 |
ilan | -6.718 |
andowski | -6.525 |
otle | -6.358 |
Token | . |
Token ID | 13 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
️ | +0.364 |
ilan | +0.354 |
andowski | +0.342 |
ulhu | +0.342 |
otle | +0.331 |
Neg logit contributions | |
Shinra | -0.294 |
phrine | -0.278 |
Lowe | -0.276 |
cens | -0.271 |
AFB | -0.262 |
Token | Gary |
Token ID | 10936 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
️ | +2.064 |
ilan | +1.987 |
andowski | +1.924 |
ulhu | +1.880 |
otle | +1.867 |
Neg logit contributions | |
Shinra | -1.785 |
phrine | -1.738 |
cens | -1.676 |
Lowe | -1.657 |
AFB | -1.605 |
Token | � |
Token ID | 447 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
️ | +2.373 |
ilan | +2.275 |
andowski | +2.212 |
ulhu | +2.153 |
otle | +2.136 |
Neg logit contributions | |
Shinra | -2.378 |
phrine | -2.274 |
Lowe | -2.222 |
cens | -2.220 |
AFB | -2.146 |
Token | on |
Token ID | 319 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
️ | +1.463 |
ilan | +1.417 |
andowski | +1.375 |
ulhu | +1.348 |
otle | +1.331 |
Neg logit contributions | |
Shinra | -1.174 |
phrine | -1.113 |
Lowe | -1.091 |
cens | -1.080 |
AFB | -1.047 |
Token | Division |
Token ID | 7458 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
️ | +2.091 |
ilan | +2.027 |
andowski | +1.970 |
ulhu | +1.949 |
otle | +1.909 |
Neg logit contributions | |
Shinra | -1.663 |
phrine | -1.564 |
Lowe | -1.555 |
cens | -1.526 |
AFB | -1.471 |
Token | Street |
Token ID | 3530 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
️ | +0.297 |
ilan | +0.289 |
andowski | +0.277 |
ulhu | +0.276 |
otle | +0.266 |
Neg logit contributions | |
Shinra | -0.323 |
Lowe | -0.306 |
phrine | -0.305 |
cens | -0.300 |
AFB | -0.291 |
Token | . |
Token ID | 13 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
Shinra | +0.310 |
phrine | +0.291 |
Lowe | +0.290 |
cens | +0.283 |
AFB | +0.274 |
Neg logit contributions | |
️ | -0.390 |
ilan | -0.381 |
andowski | -0.369 |
ulhu | -0.368 |
otle | -0.358 |
Token | Block |
Token ID | 9726 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
Shinra | +2.068 |
Lowe | +1.986 |
cens | +1.874 |
phrine | +1.866 |
Unch | +1.837 |
Neg logit contributions | |
️ | -2.622 |
ilan | -2.603 |
andowski | -2.551 |
ulhu | -2.544 |
otle | -2.423 |
Token | after |
Token ID | 706 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
Shinra | +1.377 |
Lowe | +1.282 |
phrine | +1.276 |
cens | +1.237 |
AFB | +1.206 |
Neg logit contributions | |
️ | -1.706 |
ilan | -1.659 |
ulhu | -1.613 |
andowski | -1.597 |
otle | -1.532 |
Token | block |
Token ID | 2512 |
Feature activation | -5.38e-02 |
Pos logit contributions | |
clipping | +5.715 |
Shinra | +5.646 |
cens | +5.533 |
Lowe | +5.368 |
Unch | +5.189 |
Neg logit contributions | |
andowski | -11.762 |
️ | -11.720 |
ilan | -11.563 |
ulhu | -11.217 |
otle | -10.925 |
Token | is |
Token ID | 318 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
️ | +0.406 |
ilan | +0.395 |
andowski | +0.382 |
ulhu | +0.379 |
otle | +0.370 |
Neg logit contributions | |
Shinra | -0.334 |
phrine | -0.315 |
Lowe | -0.313 |
cens | -0.307 |
AFB | -0.297 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.49e-03 |
Pos logit contributions | |
Shinra | +1.446 |
phrine | +1.358 |
Lowe | +1.356 |
cens | +1.333 |
clipping | +1.288 |
Neg logit contributions | |
️ | -1.697 |
ilan | -1.645 |
ulhu | -1.599 |
andowski | -1.595 |
otle | -1.545 |
Token | with |
Token ID | 351 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
️ | +0.012 |
ilan | +0.011 |
andowski | +0.011 |
ulhu | +0.011 |
otle | +0.011 |
Neg logit contributions | |
Shinra | -0.009 |
phrine | -0.008 |
Lowe | -0.008 |
cens | -0.008 |
AFB | -0.008 |
Token | tents |
Token ID | 29804 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Shinra | +1.227 |
phrine | +1.165 |
Lowe | +1.164 |
cens | +1.121 |
AFB | +1.096 |
Neg logit contributions | |
️ | -1.274 |
ilan | -1.225 |
andowski | -1.200 |
ulhu | -1.198 |
otle | -1.148 |
Token | and |
Token ID | 290 |
Feature activation | +9.53e-02 |
Pos logit contributions | |
️ | +0.204 |
ilan | +0.198 |
andowski | +0.192 |
ulhu | +0.190 |
otle | +0.186 |
Neg logit contributions | |
Shinra | -0.167 |
phrine | -0.158 |
Lowe | -0.156 |
cens | -0.154 |
AFB | -0.149 |
Token | learn |
Token ID | 2193 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
Shinra | +0.611 |
Lowe | +0.573 |
phrine | +0.569 |
cens | +0.551 |
AFB | +0.541 |
Neg logit contributions | |
️ | -0.704 |
ilan | -0.690 |
andowski | -0.667 |
ulhu | -0.655 |
otle | -0.635 |
Token | things |
Token ID | 1243 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
Shinra | +0.728 |
Lowe | +0.677 |
phrine | +0.676 |
cens | +0.657 |
AFB | +0.643 |
Neg logit contributions | |
️ | -0.939 |
ilan | -0.914 |
andowski | -0.885 |
ulhu | -0.878 |
otle | -0.854 |
Token | piece |
Token ID | 3704 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
Shinra | +0.570 |
phrine | +0.535 |
Lowe | +0.530 |
cens | +0.522 |
AFB | +0.505 |
Neg logit contributions | |
️ | -0.709 |
ilan | -0.691 |
andowski | -0.668 |
ulhu | -0.662 |
otle | -0.646 |
Token | - |
Token ID | 12 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
Shinra | +0.331 |
phrine | +0.311 |
Lowe | +0.304 |
cens | +0.300 |
AFB | +0.290 |
Neg logit contributions | |
️ | -0.449 |
ilan | -0.435 |
andowski | -0.425 |
ulhu | -0.417 |
otle | -0.407 |
Token | by |
Token ID | 1525 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
️ | +0.238 |
ilan | +0.228 |
andowski | +0.222 |
ulhu | +0.219 |
otle | +0.213 |
Neg logit contributions | |
Shinra | -0.250 |
phrine | -0.238 |
Lowe | -0.234 |
cens | -0.233 |
AFB | -0.225 |
Token | - |
Token ID | 12 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
Lowe | +4.939 |
clipping | +4.876 |
AFB | +4.776 |
cens | +4.730 |
Shinra | +4.676 |
Neg logit contributions | |
️ | -11.773 |
ulhu | -11.480 |
andowski | -11.449 |
ilan | -11.372 |
otle | -10.769 |
Token | piece |
Token ID | 12239 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
cens | +3.494 |
phrine | +3.399 |
Shinra | +3.383 |
bj | +3.235 |
clipping | +3.232 |
Neg logit contributions | |
️ | -5.374 |
ulhu | -5.056 |
ilan | -5.025 |
andowski | -4.952 |
otle | -4.841 |
Token | , |
Token ID | 11 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
Shinra | +2.047 |
phrine | +1.915 |
Lowe | +1.906 |
cens | +1.881 |
AFB | +1.813 |
Neg logit contributions | |
️ | -2.483 |
ilan | -2.426 |
andowski | -2.340 |
ulhu | -2.319 |
otle | -2.253 |
Token | and |
Token ID | 290 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
Shinra | +1.934 |
Lowe | +1.814 |
phrine | +1.792 |
cens | +1.745 |
clipping | +1.724 |
Neg logit contributions | |
️ | -2.544 |
ilan | -2.492 |
andowski | -2.417 |
ulhu | -2.405 |
otle | -2.315 |
Token | that |
Token ID | 326 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
️ | +0.274 |
ilan | +0.267 |
andowski | +0.258 |
ulhu | +0.254 |
otle | +0.248 |
Neg logit contributions | |
Shinra | -0.236 |
Lowe | -0.221 |
phrine | -0.221 |
cens | -0.214 |
AFB | -0.209 |
Token | and |
Token ID | 290 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
️ | +0.123 |
ilan | +0.119 |
andowski | +0.115 |
ulhu | +0.114 |
otle | +0.112 |
Neg logit contributions | |
Shinra | -0.105 |
phrine | -0.099 |
Lowe | -0.098 |
cens | -0.097 |
AFB | -0.094 |
Token | additional |
Token ID | 3224 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
Shinra | +3.620 |
Lowe | +3.473 |
phrine | +3.462 |
cens | +3.339 |
clipping | +3.316 |
Neg logit contributions | |
ilan | -4.266 |
️ | -4.245 |
ulhu | -4.160 |
andowski | -4.141 |
otle | -3.930 |
Token | variation |
Token ID | 12291 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
Shinra | +1.598 |
phrine | +1.518 |
Lowe | +1.498 |
cens | +1.455 |
AFB | +1.416 |
Neg logit contributions | |
️ | -2.155 |
ilan | -2.108 |
andowski | -2.057 |
ulhu | -2.054 |
otle | -1.992 |
Token | from |
Token ID | 422 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
Shinra | +1.270 |
phrine | +1.192 |
Lowe | +1.181 |
cens | +1.162 |
AFB | +1.110 |
Neg logit contributions | |
️ | -2.147 |
ilan | -2.133 |
andowski | -2.065 |
ulhu | -2.056 |
otle | -1.995 |
Token | year |
Token ID | 614 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Shinra | +2.456 |
Lowe | +2.358 |
phrine | +2.330 |
cens | +2.272 |
clipping | +2.238 |
Neg logit contributions | |
️ | -2.999 |
ilan | -2.962 |
andowski | -2.902 |
ulhu | -2.889 |
otle | -2.782 |
Token | to |
Token ID | 284 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Shinra | +0.665 |
phrine | +0.620 |
Lowe | +0.611 |
cens | +0.604 |
AFB | +0.585 |
Neg logit contributions | |
️ | -0.962 |
ilan | -0.940 |
andowski | -0.912 |
ulhu | -0.909 |
otle | -0.882 |
Token | year |
Token ID | 614 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
Shinra | +4.932 |
clipping | +4.835 |
Lowe | +4.728 |
AFB | +4.655 |
cull | +4.601 |
Neg logit contributions | |
ulhu | -12.108 |
ilan | -11.846 |
️ | -11.646 |
andowski | -11.630 |
otle | -10.978 |
Token | . |
Token ID | 13 |
Feature activation | -6.31e-02 |
Pos logit contributions | |
Shinra | +2.285 |
phrine | +2.129 |
Lowe | +2.122 |
cens | +2.103 |
AFB | +2.018 |
Neg logit contributions | |
️ | -3.126 |
ilan | -3.095 |
ulhu | -3.003 |
andowski | -3.001 |
otle | -2.911 |
Token | In |
Token ID | 554 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
️ | +0.483 |
ilan | +0.469 |
andowski | +0.455 |
ulhu | +0.449 |
otle | +0.441 |
Neg logit contributions | |
Shinra | -0.384 |
phrine | -0.364 |
Lowe | -0.358 |
cens | -0.354 |
AFB | -0.342 |
Token | the |
Token ID | 262 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
Shinra | +0.092 |
phrine | +0.087 |
Lowe | +0.086 |
cens | +0.085 |
AFB | +0.082 |
Neg logit contributions | |
️ | -0.103 |
ilan | -0.100 |
andowski | -0.097 |
ulhu | -0.096 |
otle | -0.093 |
Token | summer |
Token ID | 3931 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
Shinra | +1.013 |
Lowe | +0.950 |
phrine | +0.943 |
cens | +0.927 |
AFB | +0.899 |
Neg logit contributions | |
️ | -1.215 |
ilan | -1.187 |
andowski | -1.156 |
ulhu | -1.150 |
otle | -1.113 |
Token | ritch |
Token ID | 46510 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
️ | +1.331 |
ilan | +1.301 |
andowski | +1.256 |
ulhu | +1.245 |
otle | +1.226 |
Neg logit contributions | |
Shinra | -0.888 |
phrine | -0.837 |
Lowe | -0.823 |
cens | -0.810 |
AFB | -0.779 |
Token | mon |
Token ID | 937 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
️ | +1.566 |
ilan | +1.515 |
andowski | +1.464 |
ulhu | +1.444 |
otle | +1.418 |
Neg logit contributions | |
Shinra | -1.460 |
phrine | -1.396 |
Lowe | -1.372 |
cens | -1.363 |
AFB | -1.313 |
Token | stros |
Token ID | 48288 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
️ | +0.621 |
ilan | +0.600 |
andowski | +0.582 |
ulhu | +0.575 |
otle | +0.567 |
Neg logit contributions | |
Shinra | -0.440 |
phrine | -0.417 |
Lowe | -0.408 |
cens | -0.404 |
AFB | -0.388 |
Token | ities |
Token ID | 871 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
Shinra | +0.994 |
phrine | +0.955 |
Lowe | +0.945 |
cens | +0.942 |
AFB | +0.918 |
Neg logit contributions | |
️ | -0.562 |
ilan | -0.531 |
andowski | -0.511 |
ulhu | -0.501 |
otle | -0.488 |
Token | year |
Token ID | 614 |
Feature activation | +9.93e-02 |
Pos logit contributions | |
️ | +0.416 |
ilan | +0.404 |
andowski | +0.392 |
ulhu | +0.386 |
otle | +0.380 |
Neg logit contributions | |
Shinra | -0.309 |
phrine | -0.291 |
Lowe | -0.287 |
cens | -0.283 |
AFB | -0.273 |
Token | after |
Token ID | 706 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
Shinra | +0.546 |
phrine | +0.509 |
Lowe | +0.498 |
cens | +0.496 |
AFB | +0.478 |
Neg logit contributions | |
️ | -0.825 |
ilan | -0.796 |
andowski | -0.779 |
ulhu | -0.767 |
otle | -0.754 |
Token | year |
Token ID | 614 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
Shinra | +5.111 |
Lowe | +4.806 |
clipping | +4.796 |
cull | +4.709 |
AFB | +4.492 |
Neg logit contributions | |
️ | -11.515 |
andowski | -11.457 |
ulhu | -11.282 |
ilan | -11.263 |
otle | -11.043 |
Token | and |
Token ID | 290 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
️ | +1.031 |
ilan | +0.999 |
andowski | +0.967 |
ulhu | +0.953 |
otle | +0.937 |
Neg logit contributions | |
Shinra | -0.879 |
phrine | -0.834 |
Lowe | -0.824 |
cens | -0.812 |
AFB | -0.786 |
Token | held |
Token ID | 2714 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
Shinra | +0.275 |
phrine | +0.261 |
Lowe | +0.260 |
cens | +0.255 |
AFB | +0.248 |
Neg logit contributions | |
️ | -0.294 |
ilan | -0.285 |
andowski | -0.274 |
ulhu | -0.272 |
otle | -0.265 |
Token | them |
Token ID | 606 |
Feature activation | +9.49e-03 |
Pos logit contributions | |
Shinra | +0.140 |
phrine | +0.131 |
Lowe | +0.130 |
cens | +0.128 |
AFB | +0.124 |
Neg logit contributions | |
️ | -0.182 |
ilan | -0.175 |
andowski | -0.171 |
ulhu | -0.169 |
otle | -0.165 |
Token | at |
Token ID | 379 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
Shinra | +0.054 |
phrine | +0.051 |
Lowe | +0.050 |
cens | +0.049 |
AFB | +0.048 |
Neg logit contributions | |
️ | -0.077 |
ilan | -0.074 |
andowski | -0.072 |
ulhu | -0.072 |
otle | -0.070 |
Token | catchy |
Token ID | 46124 |
Feature activation | +7.07e-02 |
Pos logit contributions | |
️ | +0.799 |
ilan | +0.778 |
andowski | +0.757 |
ulhu | +0.742 |
otle | +0.726 |
Neg logit contributions | |
Shinra | -0.757 |
phrine | -0.719 |
Lowe | -0.715 |
cens | -0.702 |
AFB | -0.680 |
Token | on |
Token ID | 319 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
Shinra | +0.436 |
Lowe | +0.410 |
phrine | +0.410 |
cens | +0.400 |
AFB | +0.390 |
Neg logit contributions | |
️ | -0.529 |
ilan | -0.519 |
andowski | -0.507 |
ulhu | -0.499 |
otle | -0.486 |
Token | a |
Token ID | 257 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
Shinra | +3.945 |
Lowe | +3.792 |
phrine | +3.619 |
clipping | +3.568 |
AFB | +3.537 |
Neg logit contributions | |
️ | -5.414 |
ilan | -5.410 |
ulhu | -5.288 |
andowski | -5.282 |
otle | -5.041 |
Token | song |
Token ID | 3496 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
Shinra | +1.115 |
Lowe | +1.070 |
phrine | +1.054 |
cens | +1.018 |
AFB | +1.004 |
Neg logit contributions | |
️ | -1.234 |
ilan | -1.191 |
andowski | -1.187 |
ulhu | -1.141 |
otle | -1.118 |
Token | - |
Token ID | 12 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
️ | +1.347 |
ilan | +1.304 |
andowski | +1.263 |
ulhu | +1.248 |
otle | +1.227 |
Neg logit contributions | |
Shinra | -1.102 |
phrine | -1.042 |
Lowe | -1.024 |
cens | -1.017 |
AFB | -0.980 |
Token | by |
Token ID | 1525 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Shinra | +0.302 |
phrine | +0.288 |
Lowe | +0.285 |
cens | +0.282 |
AFB | +0.273 |
Neg logit contributions | |
️ | -0.284 |
ilan | -0.274 |
andowski | -0.264 |
ulhu | -0.261 |
otle | -0.255 |
Token | - |
Token ID | 12 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Lowe | +4.978 |
phrine | +4.876 |
AFB | +4.745 |
Shinra | +4.699 |
cens | +4.699 |
Neg logit contributions | |
️ | -10.786 |
andowski | -10.722 |
ilan | -10.665 |
ulhu | -10.591 |
otle | -10.099 |
Token | song |
Token ID | 34050 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
cens | +5.518 |
phrine | +5.374 |
bj | +5.249 |
ife | +5.096 |
Lowe | +5.057 |
Neg logit contributions | |
️ | -7.321 |
ulhu | -7.264 |
andowski | -7.057 |
ilan | -7.036 |
otle | -6.787 |
Token | basis |
Token ID | 4308 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Shinra | +0.734 |
Lowe | +0.697 |
phrine | +0.695 |
cens | +0.673 |
AFB | +0.667 |
Neg logit contributions | |
️ | -0.787 |
ilan | -0.761 |
andowski | -0.743 |
ulhu | -0.729 |
otle | -0.710 |
Token | , |
Token ID | 11 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Shinra | +1.079 |
phrine | +1.012 |
Lowe | +1.011 |
cens | +0.995 |
AFB | +0.964 |
Neg logit contributions | |
️ | -1.288 |
ilan | -1.267 |
andowski | -1.229 |
ulhu | -1.210 |
otle | -1.188 |
Token | and |
Token ID | 290 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
Shinra | +0.755 |
phrine | +0.715 |
Lowe | +0.713 |
cens | +0.701 |
AFB | +0.676 |
Neg logit contributions | |
️ | -0.913 |
ilan | -0.890 |
andowski | -0.865 |
ulhu | -0.856 |
otle | -0.834 |
Token | we |
Token ID | 356 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Shinra | +0.331 |
Lowe | +0.313 |
phrine | +0.312 |
cens | +0.304 |
AFB | +0.297 |
Neg logit contributions | |
️ | -0.391 |
ilan | -0.381 |
andowski | -0.372 |
ulhu | -0.365 |
otle | -0.354 |
Token | can |
Token ID | 460 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
️ | +0.771 |
ilan | +0.747 |
andowski | +0.726 |
ulhu | +0.718 |
otle | +0.699 |
Neg logit contributions | |
Shinra | -0.670 |
phrine | -0.637 |
Lowe | -0.629 |
cens | -0.619 |
AFB | -0.601 |
Token | work |
Token ID | 670 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Shinra | +0.133 |
phrine | +0.128 |
Lowe | +0.125 |
cens | +0.124 |
AFB | +0.121 |
Neg logit contributions | |
️ | -0.144 |
ilan | -0.140 |
andowski | -0.135 |
ulhu | -0.134 |
otle | -0.130 |
Token | on |
Token ID | 319 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
Shinra | +0.869 |
phrine | +0.815 |
Lowe | +0.807 |
cens | +0.785 |
AFB | +0.770 |
Neg logit contributions | |
️ | -1.358 |
ilan | -1.325 |
ulhu | -1.290 |
andowski | -1.290 |
otle | -1.248 |
Token | piece |
Token ID | 3704 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
Shinra | +0.360 |
phrine | +0.341 |
Lowe | +0.337 |
cens | +0.333 |
AFB | +0.322 |
Neg logit contributions | |
️ | -0.446 |
ilan | -0.434 |
andowski | -0.421 |
ulhu | -0.416 |
otle | -0.407 |
Token | by |
Token ID | 416 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Shinra | +2.058 |
phrine | +1.935 |
Lowe | +1.916 |
cens | +1.885 |
AFB | +1.833 |
Neg logit contributions | |
️ | -2.375 |
ilan | -2.278 |
andowski | -2.247 |
ulhu | -2.185 |
otle | -2.126 |
Token | piece |
Token ID | 3704 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
clipping | +7.885 |
Lowe | +7.433 |
cull | +7.344 |
Shinra | +7.189 |
AFB | +7.178 |
Neg logit contributions | |
andowski | -8.458 |
ulhu | -8.401 |
️ | -8.392 |
ilan | -8.118 |
otle | -7.637 |
Token | ." |
Token ID | 526 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
Shinra | +0.381 |
phrine | +0.362 |
Lowe | +0.358 |
cens | +0.354 |
AFB | +0.343 |
Neg logit contributions | |
️ | -0.415 |
ilan | -0.403 |
andowski | -0.391 |
ulhu | -0.386 |
otle | -0.377 |
Token |
|
Token ID | 198 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
Shinra | +1.814 |
Lowe | +1.742 |
phrine | +1.713 |
cens | +1.688 |
AFB | +1.645 |
Neg logit contributions | |
️ | -2.094 |
ilan | -2.046 |
andowski | -2.004 |
ulhu | -1.963 |
otle | -1.898 |
Token |
|
Token ID | 198 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
Shinra | +1.671 |
Lowe | +1.586 |
phrine | +1.582 |
cens | +1.563 |
AFB | +1.519 |
Neg logit contributions | |
️ | -2.014 |
ilan | -1.956 |
andowski | -1.911 |
ulhu | -1.900 |
otle | -1.828 |
Token | She |
Token ID | 3347 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
Shinra | +1.435 |
cens | +1.387 |
phrine | +1.377 |
Lowe | +1.367 |
AFB | +1.309 |
Neg logit contributions | |
️ | -1.507 |
ilan | -1.460 |
andowski | -1.439 |
ulhu | -1.427 |
otle | -1.371 |
Token | b |
Token ID | 65 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
Shinra | +1.739 |
cens | +1.686 |
phrine | +1.646 |
Lowe | +1.617 |
AFB | +1.550 |
Neg logit contributions | |
️ | -2.306 |
ilan | -2.218 |
andowski | -2.175 |
ulhu | -2.129 |
otle | -2.105 |
Token | ] |
Token ID | 60 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
️ | +0.764 |
ilan | +0.724 |
andowski | +0.676 |
ulhu | +0.650 |
otle | +0.637 |
Neg logit contributions | |
Shinra | -1.960 |
phrine | -1.894 |
Lowe | -1.876 |
cens | -1.857 |
AFB | -1.821 |
Token | Personal |
Token ID | 30228 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
Shinra | +0.240 |
phrine | +0.229 |
Lowe | +0.226 |
cens | +0.225 |
AFB | +0.217 |
Neg logit contributions | |
️ | -0.204 |
ilan | -0.197 |
andowski | -0.190 |
ulhu | -0.188 |
otle | -0.183 |
Token | best |
Token ID | 1266 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
Shinra | +1.843 |
Lowe | +1.739 |
cens | +1.671 |
phrine | +1.670 |
AFB | +1.614 |
Neg logit contributions | |
ilan | -3.022 |
️ | -2.993 |
ulhu | -2.945 |
andowski | -2.925 |
otle | -2.789 |
Token | : |
Token ID | 25 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
️ | +1.389 |
ilan | +1.334 |
andowski | +1.282 |
ulhu | +1.245 |
otle | +1.238 |
Neg logit contributions | |
Shinra | -1.725 |
phrine | -1.658 |
Lowe | -1.624 |
cens | -1.613 |
AFB | -1.561 |
Token | [/ |
Token ID | 13412 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Shinra | +4.010 |
Lowe | +3.819 |
cens | +3.754 |
phrine | +3.678 |
AFB | +3.677 |
Neg logit contributions | |
️ | -5.247 |
ilan | -5.230 |
andowski | -5.062 |
ulhu | -5.048 |
otle | -4.879 |
Token | b |
Token ID | 65 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
cens | +5.571 |
bj | +5.390 |
fg | +5.127 |
Shinra | +5.063 |
phrine | +5.057 |
Neg logit contributions | |
️ | -10.626 |
ulhu | -10.202 |
ilan | -10.083 |
andowski | -9.809 |
otle | -9.603 |
Token | ] |
Token ID | 60 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
️ | +0.493 |
ilan | +0.465 |
andowski | +0.437 |
ulhu | +0.423 |
otle | +0.411 |
Neg logit contributions | |
Shinra | -1.201 |
phrine | -1.162 |
Lowe | -1.150 |
cens | -1.142 |
AFB | -1.118 |
Token | 00 |
Token ID | 3571 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
Shinra | +3.391 |
Lowe | +3.220 |
cens | +3.137 |
phrine | +3.108 |
AFB | +3.101 |
Neg logit contributions | |
️ | -3.239 |
ilan | -3.228 |
andowski | -3.123 |
ulhu | -3.079 |
otle | -2.967 |
Token | : |
Token ID | 25 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
️ | +2.096 |
ilan | +2.008 |
andowski | +1.914 |
otle | +1.864 |
ulhu | +1.824 |
Neg logit contributions | |
Shinra | -2.967 |
phrine | -2.862 |
Lowe | -2.781 |
exting | -2.750 |
cens | -2.735 |
Token | 47 |
Token ID | 2857 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
️ | +0.820 |
ilan | +0.792 |
andowski | +0.766 |
ulhu | +0.746 |
otle | +0.739 |
Neg logit contributions | |
Shinra | -0.873 |
phrine | -0.838 |
Lowe | -0.819 |
cens | -0.807 |
AFB | -0.787 |
Token | 62 |
Token ID | 8190 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
Shinra | +2.587 |
Lowe | +2.481 |
phrine | +2.451 |
cens | +2.409 |
Unch | +2.343 |
Neg logit contributions | |
️ | -3.265 |
ilan | -3.222 |
ulhu | -3.121 |
andowski | -3.111 |
otle | -2.995 |
Token | ( |
Token ID | 357 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
️ | +0.319 |
ilan | +0.309 |
andowski | +0.297 |
ulhu | +0.294 |
otle | +0.288 |
Neg logit contributions | |
Shinra | -0.345 |
phrine | -0.329 |
Lowe | -0.327 |
cens | -0.323 |
AFB | -0.313 |
Token | 5 |
Token ID | 20 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
️ | +1.103 |
ilan | +1.070 |
andowski | +1.035 |
ulhu | +1.017 |
otle | +1.003 |
Neg logit contributions | |
Shinra | -0.849 |
phrine | -0.801 |
Lowe | -0.789 |
cens | -0.772 |
AFB | -0.753 |
Token | ) |
Token ID | 8 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
️ | +0.378 |
ilan | +0.364 |
andowski | +0.352 |
ulhu | +0.345 |
otle | +0.339 |
Neg logit contributions | |
Shinra | -0.388 |
phrine | -0.369 |
Lowe | -0.364 |
cens | -0.359 |
AFB | -0.350 |
Token | 401 |
Token ID | 22219 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Shinra | +1.946 |
Lowe | +1.852 |
phrine | +1.835 |
cens | +1.822 |
AFB | +1.752 |
Neg logit contributions | |
️ | -2.025 |
ilan | -1.980 |
andowski | -1.915 |
ulhu | -1.894 |
otle | -1.853 |
Token | - |
Token ID | 12 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Shinra | +1.028 |
phrine | +0.968 |
Lowe | +0.959 |
cens | +0.949 |
AFB | +0.910 |
Neg logit contributions | |
️ | -1.290 |
ilan | -1.254 |
andowski | -1.226 |
ulhu | -1.208 |
otle | -1.181 |
Token | 414 |
Token ID | 37309 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
Shinra | +6.609 |
cens | +6.222 |
Lowe | +6.196 |
phrine | +6.044 |
AFB | +5.900 |
Neg logit contributions | |
️ | -8.808 |
ilan | -8.655 |
ulhu | -8.531 |
andowski | -8.506 |
imar | -7.884 |
Token | ; |
Token ID | 26 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
Shinra | +3.607 |
Lowe | +3.357 |
phrine | +3.338 |
cens | +3.277 |
AFB | +3.116 |
Neg logit contributions | |
️ | -4.526 |
ilan | -4.496 |
andowski | -4.469 |
ulhu | -4.412 |
otle | -4.126 |
Token | doi |
Token ID | 23899 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
Shinra | +1.474 |
Lowe | +1.378 |
phrine | +1.308 |
cens | +1.273 |
AFB | +1.205 |
Neg logit contributions | |
️ | -3.895 |
ilan | -3.843 |
ulhu | -3.795 |
andowski | -3.779 |
otle | -3.645 |
Token | : |
Token ID | 25 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
️ | +0.353 |
ilan | +0.340 |
andowski | +0.328 |
ulhu | +0.322 |
otle | +0.316 |
Neg logit contributions | |
Shinra | -0.409 |
phrine | -0.391 |
Lowe | -0.386 |
cens | -0.382 |
AFB | -0.371 |
Token | 10 |
Token ID | 838 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
️ | +2.078 |
ilan | +1.996 |
andowski | +1.914 |
ulhu | +1.879 |
otle | +1.833 |
Neg logit contributions | |
Shinra | -2.946 |
phrine | -2.825 |
Lowe | -2.795 |
cens | -2.769 |
AFB | -2.700 |
Token | I |
Token ID | 40 |
Feature activation | +9.67e-02 |
Pos logit contributions | |
Shinra | +2.457 |
Lowe | +2.265 |
cens | +2.238 |
phrine | +2.224 |
AFB | +2.190 |
Neg logit contributions | |
️ | -4.167 |
ilan | -4.030 |
andowski | -3.950 |
ulhu | -3.934 |
otle | -3.825 |
Token | ANS |
Token ID | 15037 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
Shinra | +0.477 |
phrine | +0.450 |
Lowe | +0.438 |
cens | +0.431 |
AFB | +0.408 |
Neg logit contributions | |
️ | -0.854 |
ilan | -0.830 |
andowski | -0.809 |
ulhu | -0.797 |
otle | -0.788 |
Token | OF |
Token ID | 3963 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
️ | +1.113 |
ilan | +1.084 |
andowski | +1.054 |
ulhu | +1.038 |
otle | +1.018 |
Neg logit contributions | |
Shinra | -0.929 |
phrine | -0.871 |
Lowe | -0.867 |
cens | -0.858 |
AFB | -0.833 |
Token | THE |
Token ID | 3336 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
️ | +0.929 |
ilan | +0.892 |
andowski | +0.851 |
ulhu | +0.841 |
otle | +0.832 |
Neg logit contributions | |
Shinra | -0.979 |
phrine | -0.948 |
Lowe | -0.924 |
cens | -0.919 |
AFB | -0.880 |
Token | G |
Token ID | 402 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
️ | +0.088 |
ilan | +0.085 |
andowski | +0.082 |
ulhu | +0.081 |
otle | +0.080 |
Neg logit contributions | |
Shinra | -0.080 |
phrine | -0.076 |
Lowe | -0.075 |
cens | -0.074 |
AFB | -0.072 |
Token | AL |
Token ID | 1847 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Shinra | +1.500 |
phrine | +1.428 |
Lowe | +1.425 |
cens | +1.413 |
AFB | +1.385 |
Neg logit contributions | |
️ | -1.189 |
ilan | -1.140 |
andowski | -1.112 |
ulhu | -1.103 |
otle | -1.053 |
Token | AX |
Token ID | 25922 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
Shinra | +1.375 |
AFB | +1.268 |
cens | +1.115 |
phrine | +0.966 |
Lowe | +0.935 |
Neg logit contributions | |
️ | -14.073 |
ilan | -13.978 |
andowski | -13.772 |
ulhu | -13.556 |
otle | -13.434 |
Token | Y |
Token ID | 56 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
Shinra | +2.030 |
Lowe | +1.923 |
phrine | +1.915 |
cens | +1.905 |
AFB | +1.869 |
Neg logit contributions | |
️ | -2.155 |
ilan | -2.140 |
ulhu | -2.070 |
andowski | -2.064 |
otle | -1.965 |
Token | VOL |
Token ID | 38570 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
Shinra | +1.982 |
Lowe | +1.853 |
cens | +1.838 |
phrine | +1.831 |
Unch | +1.785 |
Neg logit contributions | |
️ | -2.731 |
ilan | -2.705 |
andowski | -2.645 |
ulhu | -2.583 |
otle | -2.546 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
️ | +1.177 |
ilan | +1.128 |
andowski | +1.087 |
ulhu | +1.071 |
otle | +1.056 |
Neg logit contributions | |
Shinra | -1.052 |
phrine | -1.012 |
Lowe | -0.982 |
cens | -0.977 |
AFB | -0.938 |
Token | 2 |
Token ID | 362 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
Shinra | +1.077 |
Lowe | +1.010 |
cens | +0.990 |
phrine | +0.990 |
AFB | +0.959 |
Neg logit contributions | |
️ | -1.193 |
ilan | -1.168 |
andowski | -1.141 |
ulhu | -1.127 |
otle | -1.099 |
Token | up |
Token ID | 510 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
Shinra | +0.802 |
phrine | +0.753 |
Lowe | +0.749 |
cens | +0.736 |
AFB | +0.709 |
Neg logit contributions | |
️ | -1.088 |
ilan | -1.071 |
andowski | -1.039 |
ulhu | -1.029 |
otle | -1.006 |
Token | to |
Token ID | 284 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
Shinra | +0.595 |
Lowe | +0.556 |
phrine | +0.555 |
cens | +0.543 |
AFB | +0.525 |
Neg logit contributions | |
️ | -0.790 |
ilan | -0.778 |
andowski | -0.754 |
ulhu | -0.748 |
otle | -0.728 |
Token | the |
Token ID | 262 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
Shinra | +2.951 |
Lowe | +2.854 |
phrine | +2.668 |
cens | +2.658 |
clipping | +2.623 |
Neg logit contributions | |
ilan | -3.518 |
️ | -3.509 |
andowski | -3.417 |
ulhu | -3.395 |
otle | -3.237 |
Token | side |
Token ID | 1735 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
Shinra | +1.715 |
Lowe | +1.623 |
phrine | +1.586 |
cens | +1.562 |
AFB | +1.534 |
Neg logit contributions | |
️ | -1.877 |
ilan | -1.834 |
andowski | -1.780 |
ulhu | -1.776 |
otle | -1.723 |
Token | - |
Token ID | 12 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
Shinra | +0.378 |
phrine | +0.357 |
Lowe | +0.354 |
cens | +0.349 |
AFB | +0.337 |
Neg logit contributions | |
️ | -0.453 |
ilan | -0.441 |
andowski | -0.427 |
ulhu | -0.422 |
otle | -0.414 |
Token | by |
Token ID | 1525 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Shinra | +0.898 |
phrine | +0.850 |
Lowe | +0.845 |
cens | +0.844 |
AFB | +0.806 |
Neg logit contributions | |
️ | -0.864 |
ilan | -0.831 |
andowski | -0.803 |
ulhu | -0.801 |
otle | -0.778 |
Token | - |
Token ID | 12 |
Feature activation | +7.74e-01 |
Pos logit contributions | |
Shinra | +4.967 |
Lowe | +4.866 |
cens | +4.734 |
phrine | +4.725 |
AFB | +4.678 |
Neg logit contributions | |
️ | -9.941 |
ulhu | -9.937 |
ilan | -9.781 |
andowski | -9.698 |
otle | -9.284 |
Token | side |
Token ID | 1589 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
cens | +5.110 |
Shinra | +5.036 |
phrine | +4.873 |
Lowe | +4.654 |
clipping | +4.621 |
Neg logit contributions | |
️ | -5.161 |
ulhu | -4.965 |
ilan | -4.831 |
andowski | -4.700 |
otle | -4.650 |
Token | M |
Token ID | 337 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
Shinra | +1.349 |
Lowe | +1.268 |
phrine | +1.252 |
cens | +1.232 |
AFB | +1.206 |
Neg logit contributions | |
️ | -1.670 |
ilan | -1.638 |
ulhu | -1.585 |
andowski | -1.583 |
otle | -1.544 |
Token | TF |
Token ID | 10234 |
Feature activation | -9.53e-02 |
Pos logit contributions | |
️ | +1.028 |
ilan | +0.976 |
andowski | +0.922 |
ulhu | +0.900 |
otle | +0.869 |
Neg logit contributions | |
Shinra | -2.249 |
phrine | -2.170 |
Lowe | -2.152 |
cens | -2.132 |
AFB | -2.088 |
Token | charts |
Token ID | 15907 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
️ | +0.699 |
ilan | +0.676 |
andowski | +0.655 |
ulhu | +0.646 |
otle | +0.634 |
Neg logit contributions | |
Shinra | -0.612 |
phrine | -0.581 |
Lowe | -0.573 |
cens | -0.566 |
AFB | -0.548 |
Token | B |
Token ID | 33 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
️ | +0.043 |
ilan | +0.041 |
andowski | +0.040 |
ulhu | +0.039 |
otle | +0.039 |
Neg logit contributions | |
Shinra | -0.040 |
phrine | -0.038 |
Lowe | -0.037 |
cens | -0.037 |
AFB | -0.035 |
Token | ionic |
Token ID | 26523 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
Shinra | +2.390 |
Lowe | +2.265 |
phrine | +2.261 |
cens | +2.228 |
AFB | +2.158 |
Neg logit contributions | |
️ | -2.456 |
ilan | -2.351 |
andowski | -2.316 |
ulhu | -2.248 |
otle | -2.198 |
Token | Commando |
Token ID | 47150 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
️ | +2.526 |
ilan | +2.447 |
andowski | +2.376 |
ulhu | +2.347 |
otle | +2.306 |
Neg logit contributions | |
Shinra | -2.034 |
phrine | -1.925 |
Lowe | -1.894 |
cens | -1.877 |
AFB | -1.811 |
Token | Rear |
Token ID | 30144 |
Feature activation | -9.93e-02 |
Pos logit contributions | |
️ | +2.528 |
ilan | +2.447 |
andowski | +2.362 |
ulhu | +2.332 |
otle | +2.283 |
Neg logit contributions | |
Shinra | -2.476 |
phrine | -2.368 |
Lowe | -2.330 |
cens | -2.301 |
AFB | -2.235 |
Token | med |
Token ID | 1150 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
️ | +0.601 |
ilan | +0.576 |
andowski | +0.555 |
ulhu | +0.546 |
otle | +0.533 |
Neg logit contributions | |
Shinra | -0.768 |
phrine | -0.731 |
Lowe | -0.726 |
cens | -0.720 |
AFB | -0.700 |
Token | [/ |
Token ID | 13412 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
️ | +0.101 |
ilan | +0.097 |
andowski | +0.094 |
ulhu | +0.093 |
otle | +0.091 |
Neg logit contributions | |
Shinra | -0.093 |
phrine | -0.088 |
Lowe | -0.087 |
cens | -0.086 |
AFB | -0.083 |
Token | b |
Token ID | 65 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
cens | +5.738 |
Shinra | +5.300 |
fg | +5.297 |
bj | +5.120 |
phrine | +4.973 |
Neg logit contributions | |
️ | -9.526 |
andowski | -9.075 |
ilan | -9.004 |
ulhu | -8.785 |
otle | -8.700 |
Token | ][/ |
Token ID | 44926 |
Feature activation | +4.41e-01 |
Pos logit contributions | |
Shinra | +0.862 |
phrine | +0.823 |
Lowe | +0.817 |
cens | +0.815 |
AFB | +0.795 |
Neg logit contributions | |
️ | -0.405 |
ilan | -0.379 |
andowski | -0.361 |
ulhu | -0.350 |
otle | -0.338 |
Token | color |
Token ID | 8043 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
Shinra | +2.677 |
cens | +2.662 |
phrine | +2.549 |
Lowe | +2.497 |
fg | +2.426 |
Neg logit contributions | |
️ | -3.221 |
andowski | -3.061 |
ilan | -3.039 |
ulhu | -2.942 |
otle | -2.918 |
Token | ] |
Token ID | 60 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
️ | +1.361 |
ilan | +1.286 |
andowski | +1.211 |
ulhu | +1.179 |
otle | +1.137 |
Neg logit contributions | |
Shinra | -3.213 |
phrine | -3.104 |
Lowe | -3.076 |
cens | -3.052 |
AFB | -2.990 |
Token | - |
Token ID | 532 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
️ | +0.712 |
ilan | +0.690 |
andowski | +0.668 |
ulhu | +0.659 |
otle | +0.646 |
Neg logit contributions | |
Shinra | -0.656 |
phrine | -0.623 |
Lowe | -0.615 |
cens | -0.608 |
AFB | -0.589 |
Token | Haitian |
Token ID | 48723 |
Feature activation | -2.44e-03 |
Pos logit contributions | |
️ | +0.203 |
ilan | +0.198 |
andowski | +0.193 |
ulhu | +0.190 |
otle | +0.185 |
Neg logit contributions | |
Shinra | -0.169 |
Lowe | -0.160 |
phrine | -0.159 |
cens | -0.156 |
AFB | -0.151 |
Token | government |
Token ID | 1230 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
️ | +0.017 |
ilan | +0.017 |
andowski | +0.016 |
ulhu | +0.016 |
otle | +0.015 |
Neg logit contributions | |
Shinra | -0.016 |
Lowe | -0.016 |
phrine | -0.016 |
cens | -0.015 |
AFB | -0.015 |
Token | officials |
Token ID | 2828 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
️ | +1.127 |
ilan | +1.087 |
andowski | +1.048 |
ulhu | +1.031 |
otle | +1.009 |
Neg logit contributions | |
Shinra | -1.271 |
phrine | -1.213 |
Lowe | -1.198 |
cens | -1.186 |
AFB | -1.153 |
Token | went |
Token ID | 1816 |
Feature activation | -1.52e-03 |
Pos logit contributions | |
Shinra | +0.109 |
phrine | +0.104 |
Lowe | +0.103 |
cens | +0.101 |
AFB | +0.099 |
Neg logit contributions | |
️ | -0.112 |
ilan | -0.109 |
ulhu | -0.104 |
andowski | -0.104 |
otle | -0.100 |
Token | door |
Token ID | 3420 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
️ | +0.012 |
ilan | +0.012 |
andowski | +0.011 |
ulhu | +0.011 |
otle | +0.011 |
Neg logit contributions | |
Shinra | -0.009 |
phrine | -0.008 |
Lowe | -0.008 |
cens | -0.008 |
AFB | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Shinra | +1.084 |
Lowe | +1.025 |
phrine | +1.019 |
cens | +0.995 |
AFB | +0.966 |
Neg logit contributions | |
️ | -1.579 |
ilan | -1.537 |
andowski | -1.489 |
ulhu | -1.473 |
otle | -1.428 |
Token | door |
Token ID | 3420 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
Lowe | +3.593 |
Shinra | +3.450 |
phrine | +3.321 |
clipping | +3.318 |
AFB | +3.302 |
Neg logit contributions | |
ilan | -11.542 |
️ | -11.354 |
andowski | -11.305 |
ulhu | -10.943 |
otle | -10.678 |
Token | warning |
Token ID | 6509 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
Shinra | +0.140 |
phrine | +0.133 |
Lowe | +0.131 |
cens | +0.129 |
AFB | +0.125 |
Neg logit contributions | |
️ | -0.185 |
ilan | -0.180 |
andowski | -0.175 |
ulhu | -0.173 |
otle | -0.170 |
Token | residents |
Token ID | 5085 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
Shinra | +1.245 |
Lowe | +1.190 |
phrine | +1.160 |
cens | +1.128 |
AFB | +1.079 |
Neg logit contributions | |
️ | -2.064 |
ilan | -2.001 |
andowski | -1.981 |
ulhu | -1.967 |
otle | -1.903 |
Token | . |
Token ID | 13 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
Shinra | +1.500 |
Lowe | +1.418 |
phrine | +1.401 |
cens | +1.379 |
AFB | +1.322 |
Neg logit contributions | |
️ | -2.150 |
ilan | -2.103 |
ulhu | -2.034 |
andowski | -2.033 |
otle | -1.964 |
Token | A |
Token ID | 317 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
️ | +0.423 |
ilan | +0.412 |
andowski | +0.401 |
ulhu | +0.395 |
otle | +0.386 |
Neg logit contributions | |
Shinra | -0.340 |
Lowe | -0.319 |
phrine | -0.319 |
cens | -0.312 |
AFB | -0.304 |
Token | between |
Token ID | 1022 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
️ | +0.042 |
ilan | +0.041 |
andowski | +0.040 |
ulhu | +0.039 |
otle | +0.039 |
Neg logit contributions | |
Shinra | -0.035 |
phrine | -0.034 |
Lowe | -0.033 |
cens | -0.033 |
AFB | -0.032 |
Token | economies |
Token ID | 16533 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
Shinra | +0.529 |
phrine | +0.500 |
Lowe | +0.499 |
cens | +0.490 |
AFB | +0.473 |
Neg logit contributions | |
️ | -0.638 |
ilan | -0.620 |
andowski | -0.603 |
ulhu | -0.595 |
otle | -0.583 |
Token | that |
Token ID | 326 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
️ | +3.361 |
ilan | +3.218 |
andowski | +3.136 |
ulhu | +3.066 |
otle | +3.049 |
Neg logit contributions | |
Shinra | -2.988 |
phrine | -2.859 |
Lowe | -2.784 |
cens | -2.745 |
AFB | -2.687 |
Token | go |
Token ID | 467 |
Feature activation | -9.98e-02 |
Pos logit contributions | |
️ | +1.496 |
ilan | +1.450 |
andowski | +1.404 |
ulhu | +1.385 |
otle | +1.358 |
Neg logit contributions | |
Shinra | -1.310 |
phrine | -1.241 |
Lowe | -1.226 |
cens | -1.210 |
AFB | -1.172 |
Token | up |
Token ID | 510 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
️ | +0.768 |
ilan | +0.746 |
andowski | +0.723 |
ulhu | +0.714 |
otle | +0.701 |
Neg logit contributions | |
Shinra | -0.605 |
phrine | -0.572 |
Lowe | -0.564 |
cens | -0.556 |
AFB | -0.538 |
Token | and |
Token ID | 290 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Shinra | +0.608 |
phrine | +0.575 |
Lowe | +0.570 |
cens | +0.562 |
AFB | +0.542 |
Neg logit contributions | |
️ | -0.746 |
ilan | -0.725 |
andowski | -0.702 |
ulhu | -0.699 |
otle | -0.681 |
Token | down |
Token ID | 866 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
Lowe | +5.000 |
Shinra | +4.913 |
cens | +4.862 |
Unch | +4.653 |
phrine | +4.642 |
Neg logit contributions | |
ilan | -9.830 |
️ | -9.607 |
ulhu | -9.234 |
andowski | -9.188 |
otle | -9.042 |
Token | , |
Token ID | 11 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
️ | +1.867 |
ilan | +1.800 |
andowski | +1.744 |
ulhu | +1.704 |
otle | +1.690 |
Neg logit contributions | |
Shinra | -1.647 |
phrine | -1.556 |
Lowe | -1.530 |
cens | -1.516 |
AFB | -1.471 |
Token | and |
Token ID | 290 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
Shinra | +2.036 |
Lowe | +1.934 |
phrine | +1.910 |
cens | +1.883 |
AFB | +1.805 |
Neg logit contributions | |
️ | -2.597 |
ilan | -2.534 |
ulhu | -2.439 |
andowski | -2.436 |
otle | -2.363 |
Token | the |
Token ID | 262 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
Shinra | +1.277 |
Lowe | +1.204 |
phrine | +1.194 |
cens | +1.169 |
AFB | +1.133 |
Neg logit contributions | |
️ | -1.554 |
ilan | -1.516 |
andowski | -1.461 |
ulhu | -1.452 |
otle | -1.415 |
Token | numbers |
Token ID | 3146 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
Shinra | +1.942 |
Lowe | +1.842 |
phrine | +1.827 |
cens | +1.786 |
AFB | +1.730 |
Neg logit contributions | |
️ | -2.233 |
ilan | -2.189 |
andowski | -2.107 |
ulhu | -2.100 |
otle | -2.035 |
Token | the |
Token ID | 262 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
️ | +2.413 |
ilan | +2.321 |
andowski | +2.249 |
ulhu | +2.186 |
otle | +2.181 |
Neg logit contributions | |
Shinra | -2.143 |
phrine | -2.034 |
Lowe | -1.999 |
cens | -1.995 |
AFB | -1.923 |
Token | impression |
Token ID | 10647 |
Feature activation | -6.30e-02 |
Pos logit contributions | |
️ | +0.489 |
ilan | +0.475 |
andowski | +0.460 |
ulhu | +0.454 |
otle | +0.445 |
Neg logit contributions | |
Shinra | -0.436 |
phrine | -0.413 |
Lowe | -0.408 |
cens | -0.403 |
AFB | -0.390 |
Token | that |
Token ID | 326 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
️ | +0.467 |
ilan | +0.454 |
andowski | +0.440 |
ulhu | +0.434 |
otle | +0.426 |
Neg logit contributions | |
Shinra | -0.399 |
phrine | -0.377 |
Lowe | -0.373 |
cens | -0.369 |
AFB | -0.356 |
Token | , |
Token ID | 11 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Shinra | +1.060 |
phrine | +0.998 |
Lowe | +0.996 |
cens | +0.974 |
AFB | +0.943 |
Neg logit contributions | |
️ | -1.278 |
ilan | -1.255 |
andowski | -1.218 |
ulhu | -1.205 |
otle | -1.168 |
Token | little |
Token ID | 1310 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
️ | +0.792 |
ilan | +0.767 |
andowski | +0.743 |
ulhu | +0.733 |
otle | +0.720 |
Neg logit contributions | |
Shinra | -0.681 |
phrine | -0.647 |
Lowe | -0.636 |
cens | -0.629 |
AFB | -0.609 |
Token | by |
Token ID | 416 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Shinra | +0.786 |
phrine | +0.746 |
Lowe | +0.736 |
cens | +0.716 |
clipping | +0.698 |
Neg logit contributions | |
️ | -1.085 |
ilan | -1.064 |
andowski | -1.036 |
ulhu | -1.024 |
otle | -0.995 |
Token | little |
Token ID | 1310 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
Shinra | +3.292 |
Lowe | +3.252 |
clipping | +3.205 |
phrine | +2.925 |
AFB | +2.779 |
Neg logit contributions | |
️ | -11.085 |
andowski | -10.976 |
ilan | -10.936 |
ulhu | -10.765 |
otle | -10.243 |
Token | , |
Token ID | 11 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Shinra | +2.199 |
Lowe | +2.080 |
phrine | +2.070 |
cens | +1.984 |
clipping | +1.978 |
Neg logit contributions | |
️ | -2.967 |
ilan | -2.948 |
ulhu | -2.869 |
andowski | -2.853 |
otle | -2.746 |
Token | you |
Token ID | 345 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
Shinra | +0.730 |
phrine | +0.686 |
Lowe | +0.682 |
cens | +0.666 |
AFB | +0.647 |
Neg logit contributions | |
️ | -0.882 |
ilan | -0.863 |
andowski | -0.836 |
ulhu | -0.830 |
otle | -0.805 |
Token | freed |
Token ID | 13459 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
Shinra | +0.140 |
phrine | +0.134 |
Lowe | +0.132 |
cens | +0.130 |
AFB | +0.126 |
Neg logit contributions | |
️ | -0.156 |
ilan | -0.152 |
andowski | -0.147 |
ulhu | -0.146 |
otle | -0.141 |
Token | yourself |
Token ID | 3511 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
Shinra | +0.429 |
phrine | +0.399 |
Lowe | +0.395 |
cens | +0.388 |
AFB | +0.372 |
Neg logit contributions | |
️ | -0.695 |
ilan | -0.679 |
andowski | -0.663 |
ulhu | -0.658 |
otle | -0.640 |
Token | International |
Token ID | 4037 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
️ | +0.665 |
ilan | +0.643 |
andowski | +0.623 |
ulhu | +0.615 |
otle | +0.600 |
Neg logit contributions | |
Shinra | -0.578 |
phrine | -0.549 |
Lowe | -0.547 |
cens | -0.538 |
AFB | -0.521 |
Token | , |
Token ID | 11 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
️ | +1.109 |
ilan | +1.074 |
andowski | +1.039 |
ulhu | +1.026 |
otle | +1.004 |
Neg logit contributions | |
Shinra | -1.006 |
phrine | -0.955 |
Lowe | -0.945 |
cens | -0.934 |
AFB | -0.906 |
Token | 25 |
Token ID | 1679 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
Shinra | +0.249 |
phrine | +0.238 |
Lowe | +0.235 |
cens | +0.233 |
AFB | +0.227 |
Neg logit contributions | |
️ | -0.222 |
ilan | -0.214 |
andowski | -0.207 |
ulhu | -0.204 |
otle | -0.199 |
Token | , |
Token ID | 11 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
️ | +0.352 |
ilan | +0.341 |
andowski | +0.329 |
ulhu | +0.326 |
otle | +0.320 |
Neg logit contributions | |
Shinra | -0.320 |
phrine | -0.304 |
Lowe | -0.301 |
cens | -0.298 |
AFB | -0.289 |
Token | 287 |
Token ID | 38721 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
️ | +0.118 |
ilan | +0.114 |
andowski | +0.110 |
ulhu | +0.108 |
otle | +0.106 |
Neg logit contributions | |
Shinra | -0.116 |
phrine | -0.110 |
Lowe | -0.109 |
cens | -0.108 |
AFB | -0.105 |
Token | - |
Token ID | 532 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Shinra | +1.244 |
phrine | +1.171 |
Lowe | +1.167 |
cens | +1.149 |
AFB | +1.121 |
Neg logit contributions | |
️ | -1.405 |
ilan | -1.370 |
andowski | -1.327 |
ulhu | -1.315 |
otle | -1.280 |
Token | 300 |
Token ID | 5867 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
Lowe | +5.815 |
Shinra | +5.809 |
phrine | +5.452 |
cens | +5.431 |
Unch | +5.207 |
Neg logit contributions | |
️ | -8.974 |
ilan | -8.891 |
andowski | -8.721 |
ulhu | -8.658 |
otle | -8.153 |
Token | . |
Token ID | 764 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
Shinra | +0.681 |
phrine | +0.632 |
Lowe | +0.622 |
cens | +0.611 |
AFB | +0.585 |
Neg logit contributions | |
️ | -1.319 |
ilan | -1.287 |
andowski | -1.254 |
ulhu | -1.242 |
otle | -1.221 |
Token |
|
Token ID | 198 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
Shinra | +2.469 |
Lowe | +2.350 |
cens | +2.317 |
phrine | +2.311 |
AFB | +2.262 |
Neg logit contributions | |
️ | -3.210 |
ilan | -3.110 |
ulhu | -3.062 |
andowski | -3.030 |
otle | -2.896 |
Token |
|
Token ID | 198 |
Feature activation | +4.79e-03 |
Pos logit contributions | |
Shinra | +1.384 |
Lowe | +1.308 |
phrine | +1.308 |
cens | +1.294 |
AFB | +1.255 |
Neg logit contributions | |
️ | -1.636 |
ilan | -1.585 |
andowski | -1.542 |
ulhu | -1.536 |
otle | -1.482 |
Token | Google |
Token ID | 11708 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
Shinra | +0.042 |
phrine | +0.041 |
Lowe | +0.040 |
cens | +0.040 |
AFB | +0.039 |
Neg logit contributions | |
️ | -0.023 |
ilan | -0.022 |
andowski | -0.021 |
ulhu | -0.021 |
otle | -0.020 |
Token | arguments |
Token ID | 7159 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
Shinra | +1.481 |
phrine | +1.406 |
Lowe | +1.390 |
cens | +1.376 |
AFB | +1.339 |
Neg logit contributions | |
️ | -1.412 |
ilan | -1.359 |
andowski | -1.310 |
ulhu | -1.301 |
otle | -1.269 |
Token | ap |
Token ID | 2471 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
️ | +1.732 |
ilan | +1.685 |
andowski | +1.634 |
ulhu | +1.615 |
otle | +1.585 |
Neg logit contributions | |
Shinra | -1.277 |
phrine | -1.201 |
Lowe | -1.185 |
cens | -1.170 |
AFB | -1.129 |
Token | = |
Token ID | 796 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
️ | +0.228 |
ilan | +0.222 |
andowski | +0.214 |
ulhu | +0.210 |
otle | +0.208 |
Neg logit contributions | |
Shinra | -0.186 |
phrine | -0.175 |
Lowe | -0.173 |
cens | -0.169 |
AFB | -0.165 |
Token | arg |
Token ID | 1822 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
Shinra | +2.805 |
phrine | +2.597 |
Lowe | +2.589 |
cens | +2.571 |
AFB | +2.532 |
Neg logit contributions | |
️ | -3.605 |
ilan | -3.487 |
ulhu | -3.404 |
andowski | -3.387 |
otle | -3.278 |
Token | parse |
Token ID | 29572 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
Shinra | +1.549 |
phrine | +1.461 |
cens | +1.439 |
Lowe | +1.433 |
AFB | +1.368 |
Neg logit contributions | |
️ | -2.157 |
ilan | -2.107 |
andowski | -2.030 |
ulhu | -2.021 |
otle | -1.964 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Shinra | +1.776 |
phrine | +1.665 |
Lowe | +1.657 |
cens | +1.654 |
AFB | +1.593 |
Neg logit contributions | |
️ | -2.449 |
ilan | -2.399 |
andowski | -2.323 |
ulhu | -2.322 |
otle | -2.238 |
Token | Arg |
Token ID | 28100 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
Shinra | +2.394 |
cens | +1.955 |
AFB | +1.759 |
Lowe | +1.630 |
IAS | +1.596 |
Neg logit contributions | |
️ | -12.513 |
ulhu | -12.197 |
ilan | -11.982 |
andowski | -11.867 |
otle | -11.715 |
Token | ument |
Token ID | 1713 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
Shinra | +3.318 |
phrine | +3.144 |
cens | +3.131 |
Lowe | +3.067 |
clipping | +2.967 |
Neg logit contributions | |
️ | -3.257 |
ilan | -3.125 |
ulhu | -3.045 |
andowski | -2.993 |
otle | -2.929 |
Token | Parser |
Token ID | 46677 |
Feature activation | +8.99e-03 |
Pos logit contributions | |
Shinra | +1.578 |
phrine | +1.519 |
Lowe | +1.512 |
cens | +1.502 |
AFB | +1.470 |
Neg logit contributions | |
️ | -0.637 |
ilan | -0.603 |
andowski | -0.565 |
ulhu | -0.555 |
otle | -0.529 |
Token | () |
Token ID | 3419 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
Shinra | +0.082 |
phrine | +0.079 |
Lowe | +0.078 |
cens | +0.077 |
AFB | +0.076 |
Neg logit contributions | |
️ | -0.042 |
ilan | -0.040 |
andowski | -0.038 |
ulhu | -0.037 |
otle | -0.036 |
Token | ap |
Token ID | 2471 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
Shinra | +0.869 |
phrine | +0.803 |
Lowe | +0.795 |
cens | +0.791 |
AFB | +0.768 |
Neg logit contributions | |
️ | -1.292 |
ilan | -1.263 |
ulhu | -1.221 |
andowski | -1.220 |
otle | -1.186 |
Token | his |
Token ID | 465 |
Feature activation | -9.92e-02 |
Pos logit contributions | |
Shinra | +0.657 |
Lowe | +0.623 |
phrine | +0.614 |
cens | +0.600 |
AFB | +0.592 |
Neg logit contributions | |
️ | -0.735 |
ilan | -0.710 |
andowski | -0.690 |
ulhu | -0.688 |
otle | -0.669 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
️ | +0.732 |
ilan | +0.710 |
andowski | +0.688 |
ulhu | +0.680 |
otle | +0.666 |
Neg logit contributions | |
Shinra | -0.632 |
phrine | -0.598 |
Lowe | -0.591 |
cens | -0.583 |
AFB | -0.566 |
Token | page |
Token ID | 2443 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Shinra | +2.463 |
Lowe | +2.376 |
phrine | +2.356 |
cens | +2.328 |
AFB | +2.295 |
Neg logit contributions | |
️ | -1.669 |
ilan | -1.623 |
ulhu | -1.579 |
andowski | -1.551 |
otle | -1.504 |
Token | to |
Token ID | 284 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
️ | +0.961 |
ilan | +0.933 |
andowski | +0.904 |
ulhu | +0.893 |
otle | +0.876 |
Neg logit contributions | |
Shinra | -0.801 |
phrine | -0.759 |
Lowe | -0.749 |
cens | -0.740 |
AFB | -0.716 |
Token | thousands |
Token ID | 4138 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
️ | +1.028 |
ilan | +0.995 |
andowski | +0.963 |
ulhu | +0.953 |
otle | +0.927 |
Neg logit contributions | |
Shinra | -1.008 |
phrine | -0.956 |
Lowe | -0.953 |
cens | -0.936 |
AFB | -0.911 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
️ | +1.763 |
ilan | +1.718 |
andowski | +1.664 |
ulhu | +1.644 |
otle | +1.611 |
Neg logit contributions | |
Shinra | -1.434 |
phrine | -1.354 |
Lowe | -1.339 |
cens | -1.323 |
AFB | -1.276 |
Token | thousands |
Token ID | 4138 |
Feature activation | -5.88e-01 |
Pos logit contributions | |
Shinra | +5.014 |
Lowe | +4.621 |
phrine | +4.561 |
cens | +4.415 |
jumps | +4.374 |
Neg logit contributions | |
ulhu | -9.477 |
️ | -9.402 |
ilan | -9.259 |
andowski | -9.075 |
otle | -8.843 |
Token | of |
Token ID | 286 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
️ | +4.270 |
ilan | +4.081 |
andowski | +3.964 |
ulhu | +3.905 |
otle | +3.851 |
Neg logit contributions | |
Shinra | -3.800 |
phrine | -3.621 |
Lowe | -3.560 |
cens | -3.512 |
AFB | -3.417 |
Token | Trump |
Token ID | 1301 |
Feature activation | -9.42e-02 |
Pos logit contributions | |
️ | +1.623 |
ilan | +1.575 |
andowski | +1.523 |
ulhu | +1.502 |
otle | +1.480 |
Neg logit contributions | |
Shinra | -1.253 |
phrine | -1.188 |
Lowe | -1.163 |
cens | -1.154 |
AFB | -1.113 |
Token | supporters |
Token ID | 5941 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
️ | +0.869 |
ilan | +0.847 |
andowski | +0.826 |
ulhu | +0.817 |
otle | +0.805 |
Neg logit contributions | |
Shinra | -0.426 |
phrine | -0.395 |
Lowe | -0.387 |
cens | -0.381 |
AFB | -0.363 |
Token | who |
Token ID | 508 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
️ | +0.136 |
ilan | +0.133 |
andowski | +0.128 |
ulhu | +0.127 |
otle | +0.124 |
Neg logit contributions | |
Shinra | -0.112 |
phrine | -0.105 |
Lowe | -0.104 |
cens | -0.103 |
AFB | -0.100 |
Token | – |
Token ID | 784 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
Shinra | +0.251 |
phrine | +0.237 |
Lowe | +0.234 |
cens | +0.231 |
AFB | +0.224 |
Neg logit contributions | |
️ | -0.307 |
ilan | -0.299 |
andowski | -0.289 |
ulhu | -0.286 |
otle | -0.280 |
Token | f |
Token ID | 277 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
Shinra | +1.222 |
phrine | +1.148 |
Lowe | +1.145 |
cens | +1.125 |
AFB | +1.088 |
Neg logit contributions | |
️ | -1.513 |
ilan | -1.471 |
andowski | -1.426 |
ulhu | -1.420 |
otle | -1.383 |
Token | 1 |
Token ID | 16 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
Shinra | +2.638 |
Lowe | +2.482 |
phrine | +2.458 |
cens | +2.443 |
AFB | +2.375 |
Neg logit contributions | |
️ | -2.530 |
ilan | -2.502 |
andowski | -2.417 |
ulhu | -2.410 |
otle | -2.331 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
️ | +0.379 |
ilan | +0.371 |
andowski | +0.359 |
ulhu | +0.356 |
otle | +0.348 |
Neg logit contributions | |
Shinra | -0.308 |
phrine | -0.288 |
Lowe | -0.288 |
cens | -0.283 |
AFB | -0.274 |
Token | 4 |
Token ID | 19 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
Shinra | +2.657 |
Lowe | +2.475 |
phrine | +2.431 |
cens | +2.421 |
AFB | +2.312 |
Neg logit contributions | |
️ | -4.328 |
ilan | -4.220 |
andowski | -4.143 |
ulhu | -4.098 |
otle | -3.997 |
Token | - |
Token ID | 12 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Shinra | +2.924 |
Lowe | +2.736 |
phrine | +2.681 |
cens | +2.667 |
AFB | +2.588 |
Neg logit contributions | |
️ | -3.595 |
ilan | -3.540 |
andowski | -3.449 |
ulhu | -3.426 |
otle | -3.317 |
Token | 2 |
Token ID | 17 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
Shinra | +5.732 |
Lowe | +5.356 |
cens | +5.287 |
phrine | +5.241 |
fg | +5.052 |
Neg logit contributions | |
ulhu | -8.755 |
️ | -8.744 |
ilan | -8.501 |
andowski | -8.392 |
otle | -8.233 |
Token | . |
Token ID | 13 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
Shinra | +2.556 |
Lowe | +2.408 |
phrine | +2.309 |
cens | +2.291 |
AFB | +2.224 |
Neg logit contributions | |
ilan | -3.604 |
️ | -3.571 |
ulhu | -3.487 |
andowski | -3.475 |
otle | -3.388 |
Token | 3 |
Token ID | 18 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
Shinra | +1.451 |
Lowe | +1.351 |
phrine | +1.345 |
cens | +1.313 |
AFB | +1.272 |
Neg logit contributions | |
️ | -2.184 |
ilan | -2.128 |
andowski | -2.074 |
ulhu | -2.053 |
otle | -2.011 |
Token | like |
Token ID | 588 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
Shinra | +2.632 |
Lowe | +2.448 |
phrine | +2.429 |
cens | +2.386 |
AFB | +2.327 |
Neg logit contributions | |
️ | -3.231 |
ilan | -3.174 |
andowski | -3.097 |
ulhu | -3.081 |
otle | -2.994 |
Token | the |
Token ID | 262 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
️ | +0.420 |
ilan | +0.407 |
andowski | +0.394 |
ulhu | +0.388 |
otle | +0.381 |
Neg logit contributions | |
Shinra | -0.360 |
phrine | -0.342 |
Lowe | -0.336 |
cens | -0.333 |
AFB | -0.322 |
Token | of |
Token ID | 286 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
️ | +0.598 |
ilan | +0.580 |
andowski | +0.563 |
ulhu | +0.552 |
otle | +0.544 |
Neg logit contributions | |
Shinra | -0.516 |
phrine | -0.490 |
Lowe | -0.481 |
cens | -0.475 |
AFB | -0.460 |
Token | physical |
Token ID | 3518 |
Feature activation | -7.53e-02 |
Pos logit contributions | |
Shinra | +0.696 |
Lowe | +0.654 |
phrine | +0.654 |
cens | +0.636 |
AFB | +0.622 |
Neg logit contributions | |
️ | -0.793 |
ilan | -0.772 |
andowski | -0.747 |
ulhu | -0.736 |
otle | -0.721 |
Token | ity |
Token ID | 414 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
️ | +0.548 |
ilan | +0.531 |
andowski | +0.516 |
ulhu | +0.508 |
otle | +0.499 |
Neg logit contributions | |
Shinra | -0.489 |
phrine | -0.463 |
Lowe | -0.460 |
cens | -0.452 |
AFB | -0.439 |
Token | as |
Token ID | 355 |
Feature activation | +8.98e-02 |
Pos logit contributions | |
️ | +2.051 |
ilan | +1.992 |
andowski | +1.929 |
ulhu | +1.905 |
otle | +1.869 |
Neg logit contributions | |
Shinra | -1.696 |
phrine | -1.605 |
Lowe | -1.584 |
cens | -1.562 |
AFB | -1.512 |
Token | System |
Token ID | 4482 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Shinra | +0.597 |
Lowe | +0.557 |
phrine | +0.557 |
cens | +0.543 |
AFB | +0.531 |
Neg logit contributions | |
️ | -0.645 |
ilan | -0.627 |
andowski | -0.609 |
ulhu | -0.597 |
otle | -0.588 |
Token | Shock |
Token ID | 22763 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Shinra | +0.784 |
Lowe | +0.721 |
phrine | +0.709 |
cens | +0.705 |
AFB | +0.662 |
Neg logit contributions | |
️ | -1.640 |
ilan | -1.593 |
andowski | -1.564 |
ulhu | -1.551 |
otle | -1.509 |
Token | 2 |
Token ID | 362 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
️ | +7.429 |
ilan | +6.953 |
andowski | +6.869 |
ulhu | +6.553 |
otle | +6.457 |
Neg logit contributions | |
Shinra | -8.154 |
phrine | -7.857 |
cens | -7.614 |
AFB | -7.589 |
Lowe | -7.538 |
Token | � |
Token ID | 447 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
️ | +3.284 |
ilan | +3.127 |
andowski | +3.047 |
ulhu | +2.957 |
otle | +2.903 |
Neg logit contributions | |
Shinra | -3.322 |
phrine | -3.179 |
Lowe | -3.106 |
cens | -3.087 |
AFB | -3.012 |
Token | � |
Token ID | 247 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
️ | +2.671 |
ilan | +2.519 |
andowski | +2.414 |
ulhu | +2.373 |
otle | +2.322 |
Neg logit contributions | |
Shinra | -3.992 |
phrine | -3.847 |
Lowe | -3.786 |
cens | -3.782 |
AFB | -3.678 |
Token | s |
Token ID | 82 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
️ | +2.822 |
ilan | +2.683 |
andowski | +2.595 |
ulhu | +2.530 |
otle | +2.492 |
Neg logit contributions | |
Shinra | -3.149 |
phrine | -3.005 |
Lowe | -2.971 |
cens | -2.918 |
AFB | -2.862 |
Token | Von |
Token ID | 26985 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
️ | +0.796 |
ilan | +0.771 |
andowski | +0.746 |
ulhu | +0.736 |
otle | +0.722 |
Neg logit contributions | |
Shinra | -0.719 |
phrine | -0.681 |
Lowe | -0.673 |
cens | -0.665 |
AFB | -0.644 |
Token | achievement |
Token ID | 13293 |
Feature activation | -1.14e-03 |
Pos logit contributions | |
Shinra | +1.121 |
Lowe | +1.055 |
phrine | +1.046 |
cens | +1.034 |
AFB | +1.014 |
Neg logit contributions | |
️ | -1.054 |
ilan | -1.026 |
andowski | -0.987 |
ulhu | -0.985 |
otle | -0.939 |
Token | . |
Token ID | 13 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
️ | +0.009 |
ilan | +0.009 |
andowski | +0.008 |
ulhu | +0.008 |
otle | +0.008 |
Neg logit contributions | |
Shinra | -0.007 |
phrine | -0.006 |
Lowe | -0.006 |
cens | -0.006 |
AFB | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
Shinra | +0.316 |
Lowe | +0.298 |
phrine | +0.290 |
cens | +0.288 |
AFB | +0.283 |
Neg logit contributions | |
️ | -0.370 |
ilan | -0.362 |
andowski | -0.355 |
ulhu | -0.348 |
otle | -0.337 |
Token |
|
Token ID | 198 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
Shinra | +0.044 |
phrine | +0.042 |
Lowe | +0.041 |
cens | +0.041 |
AFB | +0.039 |
Neg logit contributions | |
️ | -0.045 |
ilan | -0.044 |
andowski | -0.042 |
ulhu | -0.042 |
otle | -0.041 |
Token | 24 |
Token ID | 1731 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
Shinra | +0.106 |
phrine | +0.100 |
Lowe | +0.099 |
cens | +0.098 |
AFB | +0.095 |
Neg logit contributions | |
️ | -0.122 |
ilan | -0.118 |
andowski | -0.115 |
ulhu | -0.113 |
otle | -0.111 |
Token | 11 |
Token ID | 1157 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
️ | +3.115 |
ilan | +2.977 |
andowski | +2.879 |
ulhu | +2.834 |
otle | +2.788 |
Neg logit contributions | |
Shinra | -2.804 |
phrine | -2.695 |
cens | -2.622 |
Lowe | -2.590 |
AFB | -2.498 |
Token | S |
Token ID | 311 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
️ | +8.248 |
ilan | +7.817 |
otle | +7.287 |
andowski | +7.274 |
ulhu | +7.192 |
Neg logit contributions | |
phrine | -6.954 |
Shinra | -6.779 |
cens | -6.582 |
cffffcc | -6.361 |
clipping | -6.336 |
Token | . |
Token ID | 13 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
️ | +1.576 |
ilan | +1.529 |
andowski | +1.479 |
ulhu | +1.460 |
otle | +1.434 |
Neg logit contributions | |
Shinra | -1.338 |
phrine | -1.268 |
Lowe | -1.249 |
cens | -1.236 |
AFB | -1.195 |
Token | Shepherd |
Token ID | 30890 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
️ | +2.201 |
ilan | +2.123 |
andowski | +2.048 |
ulhu | +2.013 |
otle | +1.988 |
Neg logit contributions | |
Shinra | -1.972 |
phrine | -1.892 |
Lowe | -1.834 |
cens | -1.831 |
AFB | -1.764 |
Token | Today |
Token ID | 6288 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
️ | +1.750 |
ilan | +1.697 |
andowski | +1.644 |
ulhu | +1.620 |
otle | +1.592 |
Neg logit contributions | |
Shinra | -1.486 |
phrine | -1.409 |
Lowe | -1.389 |
cens | -1.372 |
AFB | -1.327 |
Token | : |
Token ID | 25 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Shinra | +0.338 |
phrine | +0.322 |
Lowe | +0.319 |
cens | +0.315 |
AFB | +0.307 |
Neg logit contributions | |
️ | -0.300 |
ilan | -0.289 |
andowski | -0.279 |
ulhu | -0.275 |
otle | -0.268 |
Token | attend |
Token ID | 5262 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Shinra | +1.398 |
Lowe | +1.309 |
phrine | +1.298 |
cens | +1.272 |
AFB | +1.243 |
Neg logit contributions | |
️ | -1.846 |
ilan | -1.805 |
ulhu | -1.747 |
andowski | -1.740 |
otle | -1.696 |
Token | a |
Token ID | 257 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
️ | +1.015 |
ilan | +0.983 |
andowski | +0.950 |
ulhu | +0.946 |
otle | +0.924 |
Neg logit contributions | |
Shinra | -0.867 |
phrine | -0.815 |
Lowe | -0.810 |
cens | -0.795 |
AFB | -0.774 |
Token | meeting |
Token ID | 3249 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
Shinra | +0.219 |
Lowe | +0.204 |
phrine | +0.204 |
cens | +0.201 |
AFB | +0.195 |
Neg logit contributions | |
️ | -0.240 |
ilan | -0.232 |
andowski | -0.224 |
ulhu | -0.224 |
otle | -0.217 |
Token | , |
Token ID | 11 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
Shinra | +0.362 |
phrine | +0.338 |
Lowe | +0.337 |
cens | +0.333 |
AFB | +0.320 |
Neg logit contributions | |
️ | -0.498 |
ilan | -0.485 |
andowski | -0.470 |
ulhu | -0.468 |
otle | -0.459 |
Token | Pu |
Token ID | 11634 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
️ | +0.409 |
ilan | +0.395 |
andowski | +0.382 |
ulhu | +0.379 |
otle | +0.370 |
Neg logit contributions | |
Shinra | -0.435 |
phrine | -0.413 |
Lowe | -0.411 |
cens | -0.404 |
AFB | -0.394 |
Token | ig |
Token ID | 328 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
️ | +1.645 |
ilan | +1.644 |
andowski | +1.568 |
ulhu | +1.540 |
otle | +1.526 |
Neg logit contributions | |
Shinra | -0.969 |
phrine | -0.907 |
Lowe | -0.880 |
cens | -0.865 |
AFB | -0.836 |
Token | dem |
Token ID | 9536 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
️ | +7.999 |
ilan | +7.695 |
andowski | +7.540 |
ulhu | +7.204 |
otle | +7.118 |
Neg logit contributions | |
Shinra | -7.458 |
phrine | -6.916 |
Lowe | -6.745 |
cens | -6.735 |
AFB | -6.597 |
Token | ont |
Token ID | 756 |
Feature activation | +9.76e-02 |
Pos logit contributions | |
Shinra | +0.705 |
phrine | +0.673 |
Lowe | +0.665 |
cens | +0.659 |
AFB | +0.640 |
Neg logit contributions | |
️ | -0.666 |
ilan | -0.643 |
andowski | -0.620 |
ulhu | -0.611 |
otle | -0.598 |
Token | said |
Token ID | 531 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
Shinra | +0.698 |
phrine | +0.665 |
Lowe | +0.663 |
cens | +0.650 |
AFB | +0.635 |
Neg logit contributions | |
️ | -0.640 |
ilan | -0.620 |
ulhu | -0.595 |
andowski | -0.593 |
otle | -0.578 |
Token | : |
Token ID | 25 |
Feature activation | -2.32e-03 |
Pos logit contributions | |
Shinra | +0.583 |
Lowe | +0.554 |
phrine | +0.545 |
cens | +0.541 |
AFB | +0.526 |
Neg logit contributions | |
️ | -0.595 |
ilan | -0.576 |
ulhu | -0.558 |
andowski | -0.556 |
otle | -0.542 |
Token | � |
Token ID | 564 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
️ | +0.015 |
ilan | +0.015 |
andowski | +0.014 |
ulhu | +0.014 |
otle | +0.014 |
Neg logit contributions | |
Shinra | -0.017 |
Lowe | -0.016 |
phrine | -0.016 |
cens | -0.016 |
AFB | -0.015 |
Token | s |
Token ID | 264 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Shinra | +2.286 |
Lowe | +2.172 |
phrine | +2.117 |
AFB | +2.040 |
cens | +2.040 |
Neg logit contributions | |
ilan | -2.645 |
️ | -2.638 |
ulhu | -2.579 |
andowski | -2.548 |
otle | -2.487 |
Token | . |
Token ID | 13 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
️ | +0.220 |
ilan | +0.213 |
andowski | +0.206 |
ulhu | +0.202 |
otle | +0.200 |
Neg logit contributions | |
Shinra | -0.193 |
phrine | -0.184 |
Lowe | -0.180 |
cens | -0.178 |
AFB | -0.172 |
Token | t |
Token ID | 83 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
Shinra | +0.546 |
phrine | +0.526 |
Lowe | +0.521 |
cens | +0.520 |
AFB | +0.506 |
Neg logit contributions | |
️ | -0.317 |
ilan | -0.303 |
andowski | -0.289 |
ulhu | -0.286 |
otle | -0.275 |
Token | . |
Token ID | 13 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
️ | +2.737 |
ilan | +2.601 |
andowski | +2.506 |
otle | +2.463 |
ulhu | +2.426 |
Neg logit contributions | |
Shinra | -2.786 |
phrine | -2.682 |
exting | -2.579 |
Lowe | -2.576 |
cens | -2.559 |
Token | F |
Token ID | 376 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
Shinra | +2.120 |
Lowe | +2.022 |
phrine | +1.891 |
AFB | +1.882 |
cens | +1.872 |
Neg logit contributions | |
️ | -3.088 |
ilan | -3.049 |
ulhu | -2.999 |
andowski | -2.939 |
otle | -2.855 |
Token | RA |
Token ID | 3861 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
️ | +1.868 |
ilan | +1.811 |
andowski | +1.745 |
ulhu | +1.731 |
otle | +1.714 |
Neg logit contributions | |
Shinra | -1.403 |
phrine | -1.319 |
Lowe | -1.291 |
cens | -1.282 |
AFB | -1.244 |
Token | 13 |
Token ID | 1511 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
️ | +5.633 |
ilan | +5.252 |
andowski | +4.983 |
otle | +4.857 |
ulhu | +4.696 |
Neg logit contributions | |
Shinra | -9.314 |
phrine | -9.081 |
Lowe | -8.945 |
cens | -8.839 |
AFB | -8.715 |
Token | Rom |
Token ID | 3570 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
️ | +2.319 |
ilan | +2.225 |
andowski | +2.148 |
ulhu | +2.097 |
otle | +2.072 |
Neg logit contributions | |
Shinra | -2.415 |
phrine | -2.328 |
cens | -2.260 |
Lowe | -2.258 |
AFB | -2.191 |
Token | ain |
Token ID | 391 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
️ | +0.209 |
ilan | +0.203 |
andowski | +0.198 |
ulhu | +0.196 |
otle | +0.193 |
Neg logit contributions | |
Shinra | -0.120 |
phrine | -0.112 |
Lowe | -0.110 |
cens | -0.108 |
AFB | -0.104 |
Token | Fe |
Token ID | 5452 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
Shinra | +0.340 |
Lowe | +0.323 |
phrine | +0.322 |
cens | +0.319 |
AFB | +0.309 |
Neg logit contributions | |
️ | -0.285 |
ilan | -0.278 |
andowski | -0.264 |
ulhu | -0.263 |
otle | -0.254 |
Token | ill |
Token ID | 359 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Shinra | +4.669 |
phrine | +4.606 |
Lowe | +4.551 |
cens | +4.494 |
ife | +4.325 |
Neg logit contributions | |
️ | -3.498 |
andowski | -3.134 |
ilan | -3.055 |
ulhu | -3.020 |
otle | -2.914 |
Token | for |
Token ID | 329 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
️ | +2.459 |
ilan | +2.378 |
andowski | +2.304 |
ulhu | +2.252 |
otle | +2.238 |
Neg logit contributions | |
Shinra | -2.115 |
phrine | -2.033 |
Lowe | -1.970 |
cens | -1.953 |
AFB | -1.893 |
Token | nations |
Token ID | 7027 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Shinra | +1.099 |
Lowe | +1.041 |
phrine | +1.019 |
cens | +1.010 |
AFB | +0.987 |
Neg logit contributions | |
️ | -1.240 |
ilan | -1.197 |
andowski | -1.165 |
ulhu | -1.156 |
otle | -1.122 |
Token | in |
Token ID | 287 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
️ | +1.019 |
ilan | +0.990 |
andowski | +0.960 |
ulhu | +0.948 |
otle | +0.931 |
Neg logit contributions | |
Shinra | -0.780 |
phrine | -0.735 |
Lowe | -0.726 |
cens | -0.717 |
AFB | -0.692 |
Token | the |
Token ID | 262 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
️ | +1.332 |
ilan | +1.291 |
andowski | +1.250 |
ulhu | +1.232 |
otle | +1.209 |
Neg logit contributions | |
Shinra | -1.188 |
phrine | -1.128 |
Lowe | -1.112 |
cens | -1.099 |
AFB | -1.065 |
Token | Organisation |
Token ID | 30801 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
Shinra | +0.213 |
Lowe | +0.202 |
phrine | +0.200 |
cens | +0.195 |
AFB | +0.192 |
Neg logit contributions | |
️ | -0.217 |
ilan | -0.208 |
andowski | -0.204 |
ulhu | -0.201 |
otle | -0.194 |
Token | for |
Token ID | 329 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
️ | +0.109 |
ilan | +0.106 |
andowski | +0.104 |
ulhu | +0.103 |
otle | +0.101 |
Neg logit contributions | |
Shinra | -0.060 |
phrine | -0.055 |
Lowe | -0.054 |
cens | -0.053 |
AFB | -0.051 |
Token | Economic |
Token ID | 11279 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
️ | +6.220 |
ilan | +6.093 |
andowski | +6.061 |
Tea | +5.705 |
/ | +5.419 |
Neg logit contributions | |
phrine | -7.864 |
cens | -7.545 |
Shinra | -7.388 |
� | -7.267 |
bj | -7.247 |
Token | Co |
Token ID | 1766 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
️ | +4.314 |
ilan | +4.136 |
andowski | +3.973 |
otle | +3.644 |
ulhu | +3.599 |
Neg logit contributions | |
Shinra | -4.937 |
phrine | -4.696 |
Lowe | -4.527 |
cens | -4.513 |
� | -4.467 |
Token | - |
Token ID | 12 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
️ | +0.179 |
ilan | +0.174 |
andowski | +0.168 |
ulhu | +0.164 |
otle | +0.161 |
Neg logit contributions | |
Shinra | -0.163 |
phrine | -0.155 |
cens | -0.152 |
Lowe | -0.152 |
AFB | -0.146 |
Token | operation |
Token ID | 27184 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
️ | +2.772 |
ilan | +2.687 |
andowski | +2.602 |
ulhu | +2.509 |
otle | +2.432 |
Neg logit contributions | |
Shinra | -2.840 |
phrine | -2.723 |
Lowe | -2.661 |
cens | -2.627 |
AFB | -2.552 |
Token | and |
Token ID | 290 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
️ | +2.659 |
ilan | +2.563 |
andowski | +2.477 |
ulhu | +2.427 |
otle | +2.380 |
Neg logit contributions | |
Shinra | -2.583 |
phrine | -2.465 |
Lowe | -2.418 |
cens | -2.411 |
AFB | -2.328 |
Token | { |
Token ID | 1391 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Shinra | +0.421 |
phrine | +0.402 |
Lowe | +0.397 |
cens | +0.393 |
AFB | +0.382 |
Neg logit contributions | |
️ | -0.384 |
ilan | -0.371 |
andowski | -0.358 |
ulhu | -0.352 |
otle | -0.345 |
Token | // |
Token ID | 3373 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
️ | +0.919 |
ilan | +0.902 |
andowski | +0.871 |
ulhu | +0.865 |
otle | +0.837 |
Neg logit contributions | |
Shinra | -0.851 |
phrine | -0.796 |
Lowe | -0.796 |
cens | -0.775 |
AFB | -0.761 |
Token | Time |
Token ID | 3862 |
Feature activation | +2.42e-03 |
Pos logit contributions | |
️ | +2.335 |
ilan | +2.165 |
andowski | +2.043 |
ulhu | +1.981 |
otle | +1.933 |
Neg logit contributions | |
Shinra | -4.958 |
phrine | -4.834 |
Lowe | -4.754 |
cens | -4.739 |
AFB | -4.628 |
Token | component |
Token ID | 7515 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
Shinra | +0.015 |
phrine | +0.014 |
Lowe | +0.014 |
cens | +0.014 |
AFB | +0.013 |
Neg logit contributions | |
️ | -0.018 |
ilan | -0.018 |
andowski | -0.017 |
ulhu | -0.017 |
otle | -0.017 |
Token | const |
Token ID | 1500 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
️ | +2.815 |
ilan | +2.680 |
andowski | +2.581 |
ulhu | +2.494 |
otle | +2.478 |
Neg logit contributions | |
Shinra | -3.794 |
phrine | -3.629 |
Lowe | -3.604 |
cens | -3.563 |
AFB | -3.442 |
Token | float |
Token ID | 12178 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
️ | +1.570 |
ilan | +1.495 |
andowski | +1.441 |
ulhu | +1.402 |
otle | +1.388 |
Neg logit contributions | |
Shinra | -2.084 |
phrine | -2.016 |
Lowe | -1.972 |
cens | -1.964 |
AFB | -1.905 |
Token | TA |
Token ID | 21664 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
️ | +5.644 |
andowski | +5.040 |
ilan | +4.985 |
otle | +4.869 |
ulhu | +4.534 |
Neg logit contributions | |
Shinra | -8.659 |
phrine | -8.615 |
cens | -8.515 |
� | -8.260 |
Lowe | -8.144 |
Token | U |
Token ID | 52 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
️ | +1.911 |
ilan | +1.827 |
andowski | +1.786 |
ulhu | +1.748 |
otle | +1.743 |
Neg logit contributions | |
Shinra | -1.542 |
phrine | -1.461 |
Lowe | -1.427 |
cens | -1.413 |
clipping | -1.352 |
Token | = |
Token ID | 796 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
️ | +0.157 |
ilan | +0.152 |
andowski | +0.147 |
ulhu | +0.145 |
otle | +0.143 |
Neg logit contributions | |
Shinra | -0.122 |
phrine | -0.115 |
Lowe | -0.113 |
cens | -0.112 |
AFB | -0.108 |
Token | 6 |
Token ID | 718 |
Feature activation | +9.54e-02 |
Pos logit contributions | |
️ | +1.006 |
ilan | +0.972 |
andowski | +0.938 |
ulhu | +0.924 |
otle | +0.899 |
Neg logit contributions | |
Shinra | -1.096 |
phrine | -1.043 |
Lowe | -1.033 |
cens | -1.022 |
AFB | -0.995 |
Token | . |
Token ID | 13 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
Shinra | +0.564 |
phrine | +0.532 |
Lowe | +0.524 |
cens | +0.517 |
AFB | +0.499 |
Neg logit contributions | |
️ | -0.749 |
ilan | -0.727 |
andowski | -0.705 |
ulhu | -0.696 |
otle | -0.684 |
Token | conceivable |
Token ID | 40324 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
Shinra | +0.206 |
phrine | +0.195 |
Lowe | +0.194 |
cens | +0.192 |
AFB | +0.185 |
Neg logit contributions | |
️ | -0.210 |
ilan | -0.204 |
andowski | -0.198 |
ulhu | -0.194 |
otle | -0.191 |
Token | way |
Token ID | 835 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
Shinra | +0.469 |
phrine | +0.443 |
Lowe | +0.442 |
cens | +0.437 |
AFB | +0.421 |
Neg logit contributions | |
️ | -0.483 |
ilan | -0.471 |
andowski | -0.456 |
ulhu | -0.448 |
otle | -0.441 |
Token | . |
Token ID | 13 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
Shinra | +1.296 |
phrine | +1.212 |
Lowe | +1.204 |
cens | +1.200 |
AFB | +1.150 |
Neg logit contributions | |
️ | -1.598 |
ilan | -1.565 |
andowski | -1.516 |
ulhu | -1.500 |
otle | -1.472 |
Token | 9 |
Token ID | 24 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Shinra | +0.351 |
phrine | +0.329 |
Lowe | +0.327 |
cens | +0.323 |
AFB | +0.312 |
Neg logit contributions | |
️ | -0.424 |
ilan | -0.412 |
andowski | -0.401 |
ulhu | -0.395 |
otle | -0.388 |
Token | .- |
Token ID | 7874 |
Feature activation | -8.90e-02 |
Pos logit contributions | |
️ | +1.724 |
ilan | +1.668 |
andowski | +1.614 |
ulhu | +1.592 |
otle | +1.560 |
Neg logit contributions | |
Shinra | -1.571 |
phrine | -1.493 |
Lowe | -1.473 |
cens | -1.455 |
AFB | -1.411 |
Token | any |
Token ID | 597 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
️ | +0.622 |
ilan | +0.602 |
andowski | +0.582 |
ulhu | +0.574 |
otle | +0.563 |
Neg logit contributions | |
Shinra | -0.604 |
phrine | -0.573 |
Lowe | -0.566 |
cens | -0.560 |
AFB | -0.543 |
Token | % |
Token ID | 4 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
️ | +6.392 |
ilan | +6.181 |
andowski | +5.927 |
ulhu | +5.893 |
otle | +5.709 |
Neg logit contributions | |
Shinra | -7.848 |
phrine | -7.542 |
Lowe | -7.469 |
cens | -7.307 |
AFB | -7.067 |
Token | 0 |
Token ID | 15 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
Shinra | +2.219 |
phrine | +2.041 |
cens | +2.023 |
Lowe | +2.016 |
AFB | +1.938 |
Neg logit contributions | |
️ | -3.480 |
ilan | -3.379 |
andowski | -3.326 |
ulhu | -3.253 |
otle | -3.240 |
Token | : |
Token ID | 25 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
️ | +0.090 |
ilan | +0.088 |
andowski | +0.085 |
ulhu | +0.084 |
otle | +0.082 |
Neg logit contributions | |
Shinra | -0.096 |
phrine | -0.091 |
Lowe | -0.091 |
cens | -0.090 |
AFB | -0.087 |
Token | 20 |
Token ID | 1238 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
Shinra | +0.168 |
phrine | +0.160 |
Lowe | +0.158 |
cens | +0.156 |
AFB | +0.151 |
Neg logit contributions | |
️ | -0.175 |
ilan | -0.170 |
andowski | -0.164 |
ulhu | -0.162 |
otle | -0.159 |
Token | : |
Token ID | 25 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
️ | +0.563 |
ilan | +0.546 |
andowski | +0.528 |
ulhu | +0.519 |
otle | +0.507 |
Neg logit contributions | |
Shinra | -0.608 |
phrine | -0.578 |
Lowe | -0.572 |
cens | -0.568 |
AFB | -0.552 |
Token | d |
Token ID | 67 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
️ | +1.817 |
ilan | +1.766 |
andowski | +1.714 |
ulhu | +1.692 |
otle | +1.665 |
Neg logit contributions | |
Shinra | -1.300 |
phrine | -1.226 |
Lowe | -1.207 |
cens | -1.192 |
AFB | -1.148 |
Token | 3 |
Token ID | 513 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
️ | +1.719 |
ilan | +1.653 |
andowski | +1.587 |
ulhu | +1.560 |
otle | +1.523 |
Neg logit contributions | |
Shinra | -2.285 |
phrine | -2.189 |
Lowe | -2.164 |
cens | -2.144 |
AFB | -2.089 |
Token | . |
Token ID | 13 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
️ | +3.038 |
ilan | +2.893 |
andowski | +2.766 |
otle | +2.692 |
ulhu | +2.680 |
Neg logit contributions | |
Shinra | -3.240 |
phrine | -3.099 |
exting | -3.023 |
Lowe | -2.980 |
cens | -2.932 |
Token | Count |
Token ID | 2764 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
️ | +2.257 |
ilan | +2.183 |
andowski | +2.113 |
ulhu | +2.085 |
otle | +2.044 |
Neg logit contributions | |
Shinra | -1.986 |
phrine | -1.901 |
Lowe | -1.855 |
cens | -1.850 |
AFB | -1.782 |
Token | below |
Token ID | 2174 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
️ | +1.297 |
ilan | +1.262 |
andowski | +1.226 |
ulhu | +1.212 |
otle | +1.192 |
Neg logit contributions | |
Shinra | -0.866 |
phrine | -0.814 |
Lowe | -0.801 |
cens | -0.790 |
AFB | -0.761 |
Token | one |
Token ID | 530 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
️ | +2.634 |
ilan | +2.550 |
andowski | +2.468 |
ulhu | +2.433 |
otle | +2.386 |
Neg logit contributions | |
Shinra | -2.396 |
phrine | -2.278 |
Lowe | -2.243 |
cens | -2.221 |
AFB | -2.149 |
Token | thousand |
Token ID | 7319 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
️ | +6.308 |
ilan | +5.983 |
andowski | +5.698 |
ulhu | +5.618 |
otle | +5.556 |
Neg logit contributions | |
Shinra | -7.934 |
phrine | -7.680 |
Lowe | -7.552 |
cens | -7.534 |
AFB | -7.295 |
Token | is |
Token ID | 318 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
️ | +0.512 |
ilan | +0.497 |
andowski | +0.483 |
ulhu | +0.477 |
otle | +0.467 |
Neg logit contributions | |
Shinra | -0.414 |
phrine | -0.389 |
Lowe | -0.387 |
cens | -0.381 |
AFB | -0.370 |
Token | low |
Token ID | 1877 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
️ | +2.253 |
ilan | +2.189 |
andowski | +2.126 |
ulhu | +2.100 |
otle | +2.065 |
Neg logit contributions | |
Shinra | -1.567 |
phrine | -1.477 |
Lowe | -1.449 |
cens | -1.433 |
AFB | -1.379 |
Token | ( |
Token ID | 357 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
️ | +0.366 |
ilan | +0.359 |
andowski | +0.345 |
ulhu | +0.339 |
otle | +0.333 |
Neg logit contributions | |
Shinra | -0.366 |
phrine | -0.350 |
Lowe | -0.348 |
cens | -0.342 |
AFB | -0.333 |
Token | 7 |
Token ID | 22 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
️ | +0.108 |
ilan | +0.105 |
andowski | +0.102 |
ulhu | +0.101 |
otle | +0.099 |
Neg logit contributions | |
Shinra | -0.071 |
phrine | -0.067 |
Lowe | -0.066 |
cens | -0.065 |
AFB | -0.062 |
Token | desolate |
Token ID | 50244 |
Feature activation | -8.63e-02 |
Pos logit contributions | |
️ | +0.174 |
ilan | +0.168 |
andowski | +0.162 |
ulhu | +0.161 |
otle | +0.158 |
Neg logit contributions | |
Shinra | -0.151 |
phrine | -0.142 |
Lowe | -0.142 |
cens | -0.140 |
AFB | -0.135 |
Token | - |
Token ID | 532 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
️ | +0.653 |
ilan | +0.635 |
andowski | +0.617 |
ulhu | +0.611 |
otle | +0.596 |
Neg logit contributions | |
Shinra | -0.533 |
phrine | -0.505 |
Lowe | -0.500 |
cens | -0.492 |
AFB | -0.475 |
Token | like |
Token ID | 588 |
Feature activation | -7.35e-02 |
Pos logit contributions | |
️ | +0.873 |
ilan | +0.846 |
andowski | +0.820 |
ulhu | +0.809 |
otle | +0.793 |
Neg logit contributions | |
Shinra | -0.746 |
phrine | -0.707 |
Lowe | -0.698 |
cens | -0.690 |
AFB | -0.667 |
Token | James |
Token ID | 3700 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
️ | +0.542 |
ilan | +0.526 |
andowski | +0.510 |
ulhu | +0.503 |
otle | +0.493 |
Neg logit contributions | |
Shinra | -0.470 |
phrine | -0.444 |
Lowe | -0.441 |
cens | -0.434 |
AFB | -0.420 |
Token | Tro |
Token ID | 8498 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
️ | +0.517 |
ilan | +0.501 |
andowski | +0.484 |
ulhu | +0.478 |
otle | +0.468 |
Neg logit contributions | |
Shinra | -0.475 |
phrine | -0.450 |
Lowe | -0.448 |
cens | -0.441 |
AFB | -0.428 |
Token | t |
Token ID | 83 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
️ | +2.078 |
ilan | +2.013 |
andowski | +1.903 |
ulhu | +1.816 |
otle | +1.813 |
Neg logit contributions | |
Shinra | -4.535 |
phrine | -4.317 |
Lowe | -4.299 |
cens | -4.199 |
AFB | -4.147 |
Token | ter |
Token ID | 353 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
️ | +1.413 |
andowski | +1.220 |
ilan | +1.218 |
ulhu | +0.887 |
otle | +0.863 |
Neg logit contributions | |
exting | -13.190 |
Shinra | -13.081 |
mosqu | -12.833 |
eleph | -12.652 |
phrine | -12.389 |
Token | , |
Token ID | 11 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Shinra | +0.349 |
phrine | +0.329 |
Lowe | +0.328 |
cens | +0.324 |
AFB | +0.314 |
Neg logit contributions | |
️ | -0.384 |
ilan | -0.372 |
andowski | -0.360 |
ulhu | -0.355 |
otle | -0.348 |
Token | his |
Token ID | 465 |
Feature activation | +1.51e-03 |
Pos logit contributions | |
Shinra | +0.677 |
Lowe | +0.642 |
phrine | +0.638 |
cens | +0.632 |
AFB | +0.606 |
Neg logit contributions | |
️ | -0.754 |
ilan | -0.730 |
andowski | -0.711 |
ulhu | -0.706 |
otle | -0.688 |
Token | happiness |
Token ID | 12157 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
Shinra | +0.009 |
phrine | +0.009 |
Lowe | +0.009 |
cens | +0.009 |
AFB | +0.008 |
Neg logit contributions | |
️ | -0.011 |
ilan | -0.011 |
andowski | -0.011 |
ulhu | -0.011 |
otle | -0.010 |
Token | consists |
Token ID | 10874 |
Feature activation | -5.10e-02 |
Pos logit contributions | |
️ | +0.305 |
ilan | +0.296 |
andowski | +0.286 |
ulhu | +0.284 |
otle | +0.277 |
Neg logit contributions | |
Shinra | -0.249 |
phrine | -0.234 |
Lowe | -0.234 |
cens | -0.230 |
AFB | -0.223 |
Token | is |
Token ID | 318 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
Shinra | +2.687 |
phrine | +2.514 |
Lowe | +2.486 |
cens | +2.463 |
AFB | +2.384 |
Neg logit contributions | |
️ | -3.392 |
ilan | -3.291 |
andowski | -3.256 |
ulhu | -3.173 |
otle | -3.101 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
Shinra | +0.645 |
phrine | +0.604 |
Lowe | +0.602 |
cens | +0.598 |
AFB | +0.577 |
Neg logit contributions | |
️ | -0.726 |
ilan | -0.703 |
andowski | -0.687 |
ulhu | -0.674 |
otle | -0.660 |
Token | . |
Token ID | 13 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
️ | +0.231 |
ilan | +0.224 |
andowski | +0.218 |
ulhu | +0.215 |
otle | +0.211 |
Neg logit contributions | |
Shinra | -0.185 |
phrine | -0.175 |
Lowe | -0.173 |
cens | -0.171 |
AFB | -0.165 |
Token | 11 |
Token ID | 1157 |
Feature activation | -7.27e-02 |
Pos logit contributions | |
Shinra | +0.256 |
phrine | +0.241 |
Lowe | +0.240 |
cens | +0.238 |
AFB | +0.230 |
Neg logit contributions | |
️ | -0.274 |
ilan | -0.267 |
andowski | -0.259 |
ulhu | -0.255 |
otle | -0.250 |
Token | .- |
Token ID | 7874 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
️ | +0.533 |
ilan | +0.520 |
andowski | +0.502 |
ulhu | +0.495 |
otle | +0.487 |
Neg logit contributions | |
Shinra | -0.466 |
phrine | -0.440 |
Lowe | -0.435 |
cens | -0.431 |
AFB | -0.417 |
Token | any |
Token ID | 597 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Shinra | +0.151 |
phrine | +0.143 |
Lowe | +0.141 |
cens | +0.139 |
AFB | +0.135 |
Neg logit contributions | |
️ | -0.155 |
ilan | -0.150 |
andowski | -0.146 |
ulhu | -0.144 |
otle | -0.141 |
Token | % |
Token ID | 4 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
️ | +6.231 |
ilan | +6.026 |
andowski | +5.788 |
ulhu | +5.767 |
otle | +5.574 |
Neg logit contributions | |
Shinra | -7.864 |
phrine | -7.563 |
Lowe | -7.509 |
cens | -7.375 |
AFB | -7.108 |
Token | 0 |
Token ID | 15 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
Shinra | +2.163 |
phrine | +2.004 |
Lowe | +1.990 |
cens | +1.988 |
AFB | +1.898 |
Neg logit contributions | |
️ | -3.252 |
ilan | -3.149 |
andowski | -3.097 |
ulhu | -3.027 |
otle | -3.010 |
Token | : |
Token ID | 25 |
Feature activation | -7.60e-04 |
Pos logit contributions | |
️ | +0.747 |
ilan | +0.725 |
andowski | +0.701 |
ulhu | +0.690 |
otle | +0.675 |
Neg logit contributions | |
Shinra | -0.815 |
phrine | -0.776 |
Lowe | -0.769 |
cens | -0.763 |
AFB | -0.740 |
Token | 05 |
Token ID | 2713 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
️ | +0.005 |
ilan | +0.005 |
andowski | +0.005 |
ulhu | +0.005 |
otle | +0.005 |
Neg logit contributions | |
Shinra | -0.005 |
phrine | -0.005 |
Lowe | -0.005 |
cens | -0.005 |
AFB | -0.005 |
Token | : |
Token ID | 25 |
Feature activation | +7.04e-03 |
Pos logit contributions | |
️ | +1.695 |
ilan | +1.635 |
andowski | +1.576 |
ulhu | +1.551 |
otle | +1.517 |
Neg logit contributions | |
Shinra | -1.935 |
phrine | -1.847 |
Lowe | -1.825 |
cens | -1.807 |
AFB | -1.757 |
Token | N |
Token ID | 399 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
Shinra | +0.289 |
Lowe | +0.275 |
phrine | +0.272 |
cens | +0.263 |
AFB | +0.262 |
Neg logit contributions | |
️ | -0.297 |
ilan | -0.289 |
andowski | -0.280 |
ulhu | -0.268 |
otle | -0.267 |
Token | ages |
Token ID | 1095 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
️ | +0.459 |
ilan | +0.451 |
andowski | +0.437 |
ulhu | +0.429 |
otle | +0.423 |
Neg logit contributions | |
Shinra | -0.349 |
phrine | -0.324 |
Lowe | -0.320 |
cens | -0.316 |
AFB | -0.303 |
Token | h |
Token ID | 71 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
️ | +1.069 |
ilan | +1.049 |
andowski | +1.010 |
ulhu | +0.987 |
otle | +0.978 |
Neg logit contributions | |
Shinra | -0.986 |
phrine | -0.929 |
Lowe | -0.924 |
cens | -0.902 |
AFB | -0.879 |
Token | war |
Token ID | 5767 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
️ | +2.542 |
ilan | +2.465 |
andowski | +2.381 |
ulhu | +2.322 |
otle | +2.292 |
Neg logit contributions | |
Shinra | -2.997 |
phrine | -2.852 |
Lowe | -2.818 |
cens | -2.763 |
AFB | -2.674 |
Token | J |
Token ID | 449 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
️ | +1.371 |
ilan | +1.328 |
andowski | +1.287 |
ulhu | +1.268 |
otle | +1.246 |
Neg logit contributions | |
Shinra | -1.170 |
phrine | -1.110 |
Lowe | -1.093 |
cens | -1.082 |
AFB | -1.045 |
Token | y |
Token ID | 88 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
️ | +1.909 |
ilan | +1.874 |
andowski | +1.823 |
ulhu | +1.790 |
otle | +1.783 |
Neg logit contributions | |
Shinra | -1.171 |
phrine | -1.080 |
Lowe | -1.059 |
cens | -1.051 |
AFB | -1.004 |
Token | ot |
Token ID | 313 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
ilan | +3.447 |
️ | +3.388 |
otle | +3.329 |
andowski | +3.130 |
ulhu | +2.994 |
Neg logit contributions | |
Shinra | -10.330 |
exting | -10.150 |
phrine | -9.927 |
Lowe | -9.863 |
cens | -9.826 |
Token | irling |
Token ID | 24297 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
ilan | +1.130 |
️ | +1.109 |
andowski | +1.063 |
ulhu | +1.054 |
otle | +1.025 |
Neg logit contributions | |
Shinra | -1.249 |
Lowe | -1.198 |
phrine | -1.193 |
cens | -1.162 |
AFB | -1.138 |
Token | a |
Token ID | 64 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
Shinra | +0.843 |
phrine | +0.797 |
Lowe | +0.784 |
cens | +0.773 |
AFB | +0.745 |
Neg logit contributions | |
️ | -1.110 |
ilan | -1.079 |
andowski | -1.048 |
ulhu | -1.033 |
otle | -1.017 |
Token | Temple |
Token ID | 10857 |
Feature activation | +6.86e-02 |
Pos logit contributions | |
️ | +0.977 |
ilan | +0.927 |
andowski | +0.878 |
ulhu | +0.862 |
otle | +0.844 |
Neg logit contributions | |
Shinra | -1.693 |
phrine | -1.645 |
Lowe | -1.611 |
cens | -1.607 |
AFB | -1.557 |
Token |
|
Token ID | 198 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
Shinra | +0.359 |
Lowe | +0.333 |
phrine | +0.329 |
cens | +0.322 |
AFB | +0.313 |
Neg logit contributions | |
️ | -0.585 |
ilan | -0.571 |
andowski | -0.557 |
ulhu | -0.550 |
otle | -0.538 |
Token | Luke |
Token ID | 11336 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
️ | +0.383 |
ilan | +0.371 |
andowski | +0.360 |
ulhu | +0.356 |
otle | +0.347 |
Neg logit contributions | |
Shinra | -0.342 |
Lowe | -0.322 |
phrine | -0.320 |
cens | -0.316 |
AFB | -0.305 |
Token | Skywalker |
Token ID | 29715 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
️ | +1.321 |
ilan | +1.285 |
andowski | +1.242 |
ulhu | +1.230 |
otle | +1.205 |
Neg logit contributions | |
Shinra | -1.107 |
phrine | -1.046 |
Lowe | -1.037 |
cens | -1.018 |
AFB | -0.989 |
Token | in |
Token ID | 287 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
️ | +4.290 |
ilan | +4.142 |
andowski | +4.020 |
ulhu | +3.941 |
otle | +3.909 |
Neg logit contributions | |
Shinra | -3.155 |
phrine | -2.997 |
Lowe | -2.899 |
cens | -2.895 |
AFB | -2.791 |
Token | the |
Token ID | 262 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
️ | +2.255 |
ilan | +2.184 |
andowski | +2.114 |
ulhu | +2.087 |
otle | +2.046 |
Neg logit contributions | |
Shinra | -2.006 |
phrine | -1.907 |
Lowe | -1.878 |
cens | -1.858 |
AFB | -1.799 |
Token | Star |
Token ID | 2907 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
️ | +1.246 |
ilan | +1.207 |
andowski | +1.168 |
ulhu | +1.151 |
otle | +1.129 |
Neg logit contributions | |
Shinra | -1.157 |
phrine | -1.099 |
Lowe | -1.085 |
cens | -1.073 |
AFB | -1.040 |
Token | Wars |
Token ID | 6176 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
️ | +1.514 |
ilan | +1.459 |
andowski | +1.407 |
ulhu | +1.385 |
otle | +1.360 |
Neg logit contributions | |
Shinra | -1.745 |
phrine | -1.672 |
Lowe | -1.646 |
cens | -1.627 |
AFB | -1.585 |
Token | films |
Token ID | 7328 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
️ | +6.159 |
ilan | +5.695 |
andowski | +5.532 |
ulhu | +5.409 |
otle | +5.314 |
Neg logit contributions | |
Shinra | -7.569 |
phrine | -7.289 |
Lowe | -7.133 |
cens | -6.972 |
AFB | -6.897 |
Token | — |
Token ID | 960 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
️ | +3.868 |
ilan | +3.715 |
andowski | +3.604 |
ulhu | +3.520 |
otle | +3.470 |
Neg logit contributions | |
Shinra | -3.709 |
phrine | -3.548 |
Lowe | -3.481 |
cens | -3.435 |
AFB | -3.382 |
Token | including |
Token ID | 8201 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
️ | +0.321 |
ilan | +0.309 |
andowski | +0.297 |
ulhu | +0.294 |
otle | +0.287 |
Neg logit contributions | |
Shinra | -0.384 |
phrine | -0.367 |
Lowe | -0.363 |
cens | -0.361 |
AFB | -0.349 |
Token | this |
Token ID | 428 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
Shinra | +0.134 |
phrine | +0.126 |
Lowe | +0.126 |
cens | +0.123 |
AFB | +0.120 |
Neg logit contributions | |
️ | -0.161 |
ilan | -0.157 |
andowski | -0.151 |
ulhu | -0.150 |
otle | -0.148 |
Token | weekend |
Token ID | 5041 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
️ | +0.886 |
ilan | +0.862 |
andowski | +0.835 |
ulhu | +0.825 |
otle | +0.808 |
Neg logit contributions | |
Shinra | -0.808 |
phrine | -0.766 |
Lowe | -0.759 |
cens | -0.748 |
AFB | -0.725 |
Token | stay |
Token ID | 2652 |
Feature activation | -7.18e-02 |
Pos logit contributions | |
Shinra | +0.243 |
phrine | +0.227 |
Lowe | +0.226 |
cens | +0.215 |
AFB | +0.211 |
Neg logit contributions | |
️ | -0.478 |
ilan | -0.465 |
andowski | -0.456 |
ulhu | -0.453 |
otle | -0.441 |
Token | . |
Token ID | 13 |
Feature activation | -9.54e-02 |
Pos logit contributions | |
️ | +0.541 |
ilan | +0.526 |
andowski | +0.510 |
ulhu | +0.503 |
otle | +0.493 |
Neg logit contributions | |
Shinra | -0.446 |
phrine | -0.423 |
Lowe | -0.417 |
cens | -0.412 |
AFB | -0.399 |
Token |
|
Token ID | 198 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
️ | +0.690 |
ilan | +0.674 |
andowski | +0.656 |
ulhu | +0.641 |
otle | +0.625 |
Neg logit contributions | |
Shinra | -0.622 |
Lowe | -0.588 |
phrine | -0.582 |
cens | -0.574 |
AFB | -0.560 |
Token |
|
Token ID | 198 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
️ | +0.556 |
ilan | +0.537 |
andowski | +0.517 |
ulhu | +0.506 |
otle | +0.503 |
Neg logit contributions | |
Shinra | -0.564 |
phrine | -0.538 |
Lowe | -0.527 |
cens | -0.521 |
AFB | -0.504 |
Token | 27 |
Token ID | 1983 |
Feature activation | -4.93e-01 |
Pos logit contributions | |
️ | +1.183 |
ilan | +1.146 |
andowski | +1.107 |
ulhu | +1.090 |
otle | +1.072 |
Neg logit contributions | |
Shinra | -1.044 |
phrine | -0.991 |
Lowe | -0.976 |
cens | -0.960 |
AFB | -0.933 |
Token | 10 |
Token ID | 940 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
️ | +3.511 |
ilan | +3.344 |
andowski | +3.235 |
ulhu | +3.182 |
otle | +3.125 |
Neg logit contributions | |
Shinra | -3.220 |
phrine | -3.106 |
cens | -3.018 |
Lowe | -2.978 |
AFB | -2.873 |
Token | Mont |
Token ID | 5575 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
️ | +7.177 |
ilan | +6.798 |
andowski | +6.379 |
otle | +6.324 |
ulhu | +6.309 |
Neg logit contributions | |
phrine | -6.300 |
Shinra | -6.199 |
cens | -5.998 |
clipping | -5.764 |
cffffcc | -5.738 |
Token | rose |
Token ID | 13698 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
️ | +2.443 |
ilan | +2.373 |
andowski | +2.291 |
ulhu | +2.258 |
otle | +2.220 |
Neg logit contributions | |
Shinra | -2.249 |
phrine | -2.138 |
Lowe | -2.099 |
cens | -2.079 |
AFB | -2.015 |
Token | Today |
Token ID | 6288 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
Shinra | +0.121 |
Lowe | +0.114 |
phrine | +0.113 |
cens | +0.110 |
AFB | +0.109 |
Neg logit contributions | |
️ | -0.133 |
ilan | -0.129 |
andowski | -0.126 |
ulhu | -0.124 |
otle | -0.121 |
Token | : |
Token ID | 25 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
Shinra | +2.266 |
Lowe | +2.149 |
phrine | +2.131 |
cens | +2.092 |
AFB | +2.067 |
Neg logit contributions | |
️ | -2.275 |
ilan | -2.204 |
andowski | -2.135 |
ulhu | -2.123 |
otle | -2.045 |
Token | BB |
Token ID | 12597 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
️ | +0.219 |
ilan | +0.213 |
andowski | +0.208 |
ulhu | +0.204 |
otle | +0.198 |
Neg logit contributions | |
Shinra | -0.210 |
Lowe | -0.197 |
phrine | -0.196 |
cens | -0.192 |
AFB | -0.188 |
Token | on |
Token ID | 319 |
Feature activation | -8.72e-02 |
Pos logit contributions | |
️ | +0.890 |
ilan | +0.864 |
andowski | +0.837 |
ulhu | +0.826 |
otle | +0.811 |
Neg logit contributions | |
Shinra | -0.712 |
phrine | -0.673 |
Lowe | -0.664 |
cens | -0.655 |
AFB | -0.634 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
️ | +0.660 |
ilan | +0.640 |
andowski | +0.620 |
ulhu | +0.612 |
otle | +0.601 |
Neg logit contributions | |
Shinra | -0.540 |
phrine | -0.511 |
Lowe | -0.503 |
cens | -0.498 |
AFB | -0.481 |
Token |
|
Token ID | 198 |
Feature activation | -5.97e-02 |
Pos logit contributions | |
️ | +1.101 |
ilan | +1.067 |
andowski | +1.032 |
ulhu | +1.019 |
otle | +1.001 |
Neg logit contributions | |
Shinra | -0.944 |
phrine | -0.897 |
Lowe | -0.882 |
cens | -0.873 |
AFB | -0.844 |
Token |
|
Token ID | 198 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
️ | +0.410 |
ilan | +0.396 |
andowski | +0.381 |
ulhu | +0.373 |
otle | +0.371 |
Neg logit contributions | |
Shinra | -0.414 |
phrine | -0.394 |
Lowe | -0.386 |
cens | -0.382 |
AFB | -0.370 |
Token | 43 |
Token ID | 3559 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
️ | +2.422 |
ilan | +2.351 |
andowski | +2.265 |
ulhu | +2.225 |
otle | +2.200 |
Neg logit contributions | |
Shinra | -2.110 |
phrine | -1.998 |
Lowe | -1.964 |
cens | -1.928 |
AFB | -1.879 |
Token | 15 |
Token ID | 1314 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
️ | +2.727 |
ilan | +2.613 |
andowski | +2.524 |
ulhu | +2.481 |
otle | +2.434 |
Neg logit contributions | |
Shinra | -2.675 |
phrine | -2.573 |
cens | -2.506 |
Lowe | -2.493 |
AFB | -2.404 |
Token | Mont |
Token ID | 5575 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
️ | +7.253 |
ilan | +6.884 |
andowski | +6.452 |
otle | +6.432 |
ulhu | +6.381 |
Neg logit contributions | |
phrine | -6.144 |
Shinra | -6.072 |
cens | -5.866 |
clipping | -5.610 |
cffffcc | -5.587 |
Token | rose |
Token ID | 13698 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
️ | +1.294 |
ilan | +1.251 |
andowski | +1.215 |
ulhu | +1.199 |
otle | +1.175 |
Neg logit contributions | |
Shinra | -1.078 |
phrine | -1.022 |
Lowe | -1.011 |
cens | -0.997 |
AFB | -0.965 |
Token | Today |
Token ID | 6288 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
️ | +1.973 |
ilan | +1.912 |
andowski | +1.852 |
ulhu | +1.827 |
otle | +1.796 |
Neg logit contributions | |
Shinra | -1.640 |
phrine | -1.555 |
Lowe | -1.530 |
cens | -1.515 |
AFB | -1.462 |
Token | : |
Token ID | 25 |
Feature activation | -9.29e-02 |
Pos logit contributions | |
Shinra | +1.461 |
Lowe | +1.381 |
phrine | +1.378 |
cens | +1.343 |
AFB | +1.327 |
Neg logit contributions | |
️ | -1.474 |
ilan | -1.434 |
andowski | -1.393 |
ulhu | -1.363 |
otle | -1.326 |
Token | Brass |
Token ID | 32309 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
️ | +0.663 |
ilan | +0.645 |
andowski | +0.627 |
ulhu | +0.616 |
otle | +0.600 |
Neg logit contributions | |
Shinra | -0.617 |
Lowe | -0.581 |
phrine | -0.580 |
cens | -0.566 |
AFB | -0.554 |
Token | developing |
Token ID | 5922 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
Shinra | +0.825 |
phrine | +0.772 |
Lowe | +0.765 |
cens | +0.762 |
AFB | +0.729 |
Neg logit contributions | |
️ | -1.130 |
ilan | -1.104 |
andowski | -1.074 |
ulhu | -1.063 |
otle | -1.042 |
Token | the |
Token ID | 262 |
Feature activation | -7.70e-02 |
Pos logit contributions | |
Shinra | +0.163 |
phrine | +0.155 |
Lowe | +0.153 |
cens | +0.151 |
AFB | +0.146 |
Neg logit contributions | |
️ | -0.182 |
ilan | -0.176 |
andowski | -0.170 |
ulhu | -0.168 |
otle | -0.165 |
Token | skills |
Token ID | 4678 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
️ | +0.556 |
ilan | +0.539 |
andowski | +0.521 |
ulhu | +0.516 |
otle | +0.503 |
Neg logit contributions | |
Shinra | -0.508 |
phrine | -0.479 |
Lowe | -0.477 |
cens | -0.467 |
AFB | -0.456 |
Token | to |
Token ID | 284 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
️ | +1.035 |
ilan | +1.002 |
andowski | +0.969 |
ulhu | +0.955 |
otle | +0.937 |
Neg logit contributions | |
Shinra | -0.968 |
phrine | -0.920 |
Lowe | -0.907 |
cens | -0.897 |
AFB | -0.870 |
Token | identify |
Token ID | 5911 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
️ | +1.447 |
ilan | +1.403 |
andowski | +1.357 |
ulhu | +1.341 |
otle | +1.311 |
Neg logit contributions | |
Shinra | -1.306 |
phrine | -1.238 |
Lowe | -1.225 |
cens | -1.206 |
AFB | -1.173 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
️ | +2.759 |
ilan | +2.674 |
andowski | +2.588 |
ulhu | +2.542 |
otle | +2.509 |
Neg logit contributions | |
Shinra | -2.406 |
phrine | -2.287 |
Lowe | -2.240 |
cens | -2.230 |
AFB | -2.144 |
Token | use |
Token ID | 779 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
️ | +6.631 |
ilan | +6.191 |
andowski | +6.122 |
otle | +5.918 |
ulhu | +5.731 |
Neg logit contributions | |
Shinra | -6.995 |
phrine | -6.630 |
cens | -6.531 |
Lowe | -6.396 |
� | -6.188 |
Token | wild |
Token ID | 4295 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
️ | +4.437 |
ilan | +4.294 |
andowski | +4.163 |
ulhu | +4.068 |
otle | +4.036 |
Neg logit contributions | |
Shinra | -3.926 |
phrine | -3.734 |
cens | -3.662 |
Lowe | -3.646 |
AFB | -3.509 |
Token | and |
Token ID | 290 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
️ | +1.612 |
ilan | +1.561 |
andowski | +1.511 |
ulhu | +1.489 |
otle | +1.461 |
Neg logit contributions | |
Shinra | -1.458 |
phrine | -1.385 |
Lowe | -1.366 |
cens | -1.350 |
AFB | -1.308 |
Token | edible |
Token ID | 35988 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Shinra | +0.515 |
phrine | +0.492 |
Lowe | +0.486 |
cens | +0.478 |
AFB | +0.467 |
Neg logit contributions | |
️ | -0.525 |
ilan | -0.506 |
andowski | -0.496 |
ulhu | -0.490 |
otle | -0.475 |
Token | plants |
Token ID | 6134 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
Shinra | +1.064 |
Lowe | +1.011 |
phrine | +1.007 |
cens | +0.979 |
AFB | +0.962 |
Neg logit contributions | |
️ | -1.217 |
ilan | -1.178 |
andowski | -1.148 |
ulhu | -1.144 |
otle | -1.099 |
Token | $ |
Token ID | 720 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
Shinra | +1.067 |
Lowe | +0.989 |
phrine | +0.960 |
cens | +0.931 |
AFB | +0.898 |
Neg logit contributions | |
️ | -2.503 |
ilan | -2.502 |
andowski | -2.448 |
ulhu | -2.417 |
otle | -2.374 |
Token | 10 |
Token ID | 940 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
Shinra | +1.475 |
phrine | +1.388 |
Lowe | +1.378 |
cens | +1.371 |
AFB | +1.318 |
Neg logit contributions | |
️ | -1.726 |
ilan | -1.684 |
andowski | -1.635 |
ulhu | -1.627 |
otle | -1.579 |
Token | . |
Token ID | 13 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
️ | +1.798 |
ilan | +1.741 |
andowski | +1.686 |
ulhu | +1.662 |
otle | +1.633 |
Neg logit contributions | |
Shinra | -1.516 |
phrine | -1.438 |
Lowe | -1.415 |
cens | -1.402 |
AFB | -1.354 |
Token | 88 |
Token ID | 3459 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
Shinra | +0.191 |
phrine | +0.181 |
Lowe | +0.178 |
cens | +0.176 |
AFB | +0.169 |
Neg logit contributions | |
️ | -0.246 |
ilan | -0.239 |
andowski | -0.231 |
ulhu | -0.228 |
otle | -0.225 |
Token | million |
Token ID | 1510 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
️ | +2.146 |
ilan | +2.079 |
andowski | +2.013 |
ulhu | +1.975 |
otle | +1.947 |
Neg logit contributions | |
Shinra | -1.874 |
phrine | -1.786 |
Lowe | -1.753 |
cens | -1.735 |
AFB | -1.676 |
Token | ( |
Token ID | 357 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
️ | +2.602 |
ilan | +2.519 |
andowski | +2.429 |
ulhu | +2.392 |
otle | +2.342 |
Neg logit contributions | |
Shinra | -2.807 |
phrine | -2.675 |
Lowe | -2.646 |
cens | -2.618 |
AFB | -2.543 |
Token | 41 |
Token ID | 3901 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
️ | +6.283 |
ilan | +6.143 |
andowski | +5.805 |
ulhu | +5.507 |
otle | +5.471 |
Neg logit contributions | |
Shinra | -7.120 |
phrine | -6.672 |
Lowe | -6.611 |
AFB | -6.338 |
Unch | -6.289 |
Token | st |
Token ID | 301 |
Feature activation | -9.59e-01 |
Pos logit contributions | |
️ | +5.142 |
ilan | +4.788 |
andowski | +4.707 |
otle | +4.513 |
ulhu | +4.491 |
Neg logit contributions | |
Shinra | -5.531 |
phrine | -5.285 |
Lowe | -5.160 |
cens | -5.062 |
exting | -4.985 |
Token | among |
Token ID | 1871 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
️ | +6.050 |
ilan | +5.685 |
andowski | +5.576 |
ulhu | +5.464 |
otle | +5.399 |
Neg logit contributions | |
Shinra | -6.890 |
phrine | -6.671 |
Lowe | -6.463 |
cens | -6.458 |
AFB | -6.204 |
Token | all |
Token ID | 477 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
️ | +1.445 |
ilan | +1.402 |
andowski | +1.357 |
ulhu | +1.347 |
otle | +1.319 |
Neg logit contributions | |
Shinra | -1.193 |
phrine | -1.124 |
Lowe | -1.117 |
cens | -1.101 |
AFB | -1.067 |
Token | edge |
Token ID | 5743 |
Feature activation | +7.28e-02 |
Pos logit contributions | |
️ | +2.699 |
ilan | +2.605 |
andowski | +2.517 |
ulhu | +2.467 |
otle | +2.419 |
Neg logit contributions | |
Shinra | -2.844 |
phrine | -2.721 |
Lowe | -2.673 |
cens | -2.649 |
AFB | -2.566 |
Token | pub |
Token ID | 2240 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
️ | +0.289 |
ilan | +0.282 |
andowski | +0.275 |
ulhu | +0.273 |
otle | +0.266 |
Neg logit contributions | |
Shinra | -0.203 |
phrine | -0.190 |
Lowe | -0.189 |
cens | -0.184 |
AFB | -0.179 |
Token | fn |
Token ID | 24714 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
️ | +1.186 |
ilan | +1.163 |
andowski | +1.135 |
ulhu | +1.127 |
otle | +1.101 |
Neg logit contributions | |
Shinra | -0.522 |
phrine | -0.478 |
Lowe | -0.474 |
cens | -0.459 |
AFB | -0.438 |
Token | new |
Token ID | 649 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
️ | +2.664 |
ilan | +2.565 |
andowski | +2.476 |
ulhu | +2.411 |
otle | +2.411 |
Neg logit contributions | |
Shinra | -2.452 |
phrine | -2.357 |
cens | -2.311 |
Lowe | -2.296 |
AFB | -2.214 |
Token | ( |
Token ID | 357 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
️ | +0.369 |
ilan | +0.357 |
andowski | +0.344 |
ulhu | +0.340 |
otle | +0.332 |
Neg logit contributions | |
Shinra | -0.383 |
phrine | -0.365 |
Lowe | -0.361 |
cens | -0.357 |
AFB | -0.346 |
Token | t |
Token ID | 256 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
️ | +2.018 |
ilan | +1.951 |
andowski | +1.884 |
ulhu | +1.831 |
otle | +1.821 |
Neg logit contributions | |
Shinra | -2.151 |
phrine | -2.055 |
Lowe | -2.016 |
cens | -2.008 |
AFB | -1.947 |
Token | : |
Token ID | 1058 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
️ | +2.052 |
ilan | +1.972 |
andowski | +1.894 |
ulhu | +1.868 |
otle | +1.834 |
Neg logit contributions | |
Shinra | -2.580 |
phrine | -2.481 |
cens | -2.432 |
Lowe | -2.428 |
AFB | -2.367 |
Token | T |
Token ID | 309 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
️ | +5.707 |
ilan | +5.307 |
andowski | +4.996 |
otle | +4.936 |
ulhu | +4.710 |
Neg logit contributions | |
phrine | -7.547 |
Shinra | -7.538 |
cens | -7.340 |
Lowe | -7.103 |
� | -7.071 |
Token | ) |
Token ID | 1267 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
️ | +4.772 |
ilan | +4.570 |
andowski | +4.439 |
ulhu | +4.354 |
otle | +4.334 |
Neg logit contributions | |
Shinra | -4.095 |
phrine | -3.949 |
cens | -3.826 |
Lowe | -3.769 |
AFB | -3.656 |
Token | -> |
Token ID | 4613 |
Feature activation | -7.63e-01 |
Pos logit contributions | |
️ | +4.245 |
ilan | +3.994 |
andowski | +3.807 |
otle | +3.719 |
ulhu | +3.681 |
Neg logit contributions | |
Shinra | -4.633 |
phrine | -4.546 |
cens | -4.373 |
Lowe | -4.324 |
AFB | -4.260 |
Token | Own |
Token ID | 11744 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
️ | +5.358 |
ilan | +5.090 |
andowski | +4.892 |
otle | +4.791 |
ulhu | +4.657 |
Neg logit contributions | |
Shinra | -4.860 |
phrine | -4.813 |
cens | -4.678 |
Lowe | -4.489 |
AFB | -4.396 |
Token | ed |
Token ID | 276 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
️ | +1.183 |
ilan | +1.141 |
andowski | +1.119 |
otle | +1.071 |
ulhu | +1.070 |
Neg logit contributions | |
Shinra | -1.039 |
phrine | -0.981 |
Lowe | -0.959 |
cens | -0.954 |
AFB | -0.913 |
Token | signing |
Token ID | 8415 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
️ | +2.452 |
ilan | +2.373 |
andowski | +2.265 |
otle | +2.205 |
ulhu | +2.203 |
Neg logit contributions | |
Shinra | -3.242 |
phrine | -3.126 |
cens | -3.040 |
Lowe | -3.031 |
AFB | -2.918 |
Token | : |
Token ID | 25 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
️ | +2.762 |
ilan | +2.590 |
andowski | +2.491 |
otle | +2.428 |
ulhu | +2.363 |
Neg logit contributions | |
Shinra | -3.849 |
phrine | -3.728 |
cens | -3.609 |
Lowe | -3.589 |
exting | -3.552 |
Token | $ |
Token ID | 720 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Shinra | +0.835 |
Lowe | +0.765 |
phrine | +0.763 |
cens | +0.742 |
AFB | +0.716 |
Neg logit contributions | |
️ | -1.706 |
ilan | -1.675 |
andowski | -1.633 |
ulhu | -1.624 |
otle | -1.587 |
Token | 6 |
Token ID | 21 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
Shinra | +0.942 |
phrine | +0.883 |
Lowe | +0.874 |
cens | +0.868 |
AFB | +0.830 |
Neg logit contributions | |
️ | -1.323 |
ilan | -1.295 |
andowski | -1.259 |
ulhu | -1.250 |
otle | -1.217 |
Token | million |
Token ID | 1510 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
️ | +2.073 |
ilan | +2.004 |
andowski | +1.939 |
ulhu | +1.909 |
otle | +1.878 |
Neg logit contributions | |
Shinra | -1.782 |
phrine | -1.692 |
Lowe | -1.665 |
cens | -1.651 |
AFB | -1.596 |
Token | ( |
Token ID | 357 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
️ | +3.283 |
ilan | +3.158 |
andowski | +3.041 |
ulhu | +2.995 |
otle | +2.931 |
Neg logit contributions | |
Shinra | -3.646 |
phrine | -3.484 |
Lowe | -3.434 |
cens | -3.401 |
AFB | -3.308 |
Token | 15 |
Token ID | 1314 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
️ | +6.362 |
ilan | +6.205 |
andowski | +5.877 |
ulhu | +5.558 |
otle | +5.530 |
Neg logit contributions | |
Shinra | -6.955 |
phrine | -6.570 |
Lowe | -6.480 |
AFB | -6.192 |
exting | -6.123 |
Token | th |
Token ID | 400 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
️ | +2.413 |
ilan | +2.217 |
andowski | +2.107 |
ulhu | +2.005 |
otle | +1.995 |
Neg logit contributions | |
Shinra | -4.903 |
phrine | -4.699 |
Lowe | -4.637 |
cens | -4.560 |
AFB | -4.464 |
Token | among |
Token ID | 1871 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
️ | +4.683 |
ilan | +4.503 |
andowski | +4.364 |
ulhu | +4.307 |
otle | +4.217 |
Neg logit contributions | |
Shinra | -4.903 |
phrine | -4.680 |
Lowe | -4.592 |
cens | -4.570 |
AFB | -4.389 |
Token | running |
Token ID | 2491 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
️ | +1.703 |
ilan | +1.652 |
andowski | +1.610 |
ulhu | +1.595 |
otle | +1.563 |
Neg logit contributions | |
Shinra | -1.199 |
phrine | -1.125 |
Lowe | -1.118 |
cens | -1.095 |
AFB | -1.063 |
Token | backs |
Token ID | 12983 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
️ | +0.939 |
ilan | +0.908 |
andowski | +0.878 |
ulhu | +0.857 |
otle | +0.845 |
Neg logit contributions | |
Shinra | -0.915 |
phrine | -0.867 |
Lowe | -0.849 |
cens | -0.846 |
AFB | -0.816 |
Token | ' |
Token ID | 6 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
️ | +0.240 |
ilan | +0.232 |
andowski | +0.224 |
ulhu | +0.220 |
otle | +0.217 |
Neg logit contributions | |
Shinra | -0.235 |
phrine | -0.224 |
Lowe | -0.219 |
cens | -0.219 |
AFB | -0.211 |
Token | Forum |
Token ID | 14867 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
Shinra | +0.400 |
phrine | +0.374 |
Lowe | +0.373 |
cens | +0.364 |
AFB | +0.356 |
Neg logit contributions | |
️ | -0.545 |
ilan | -0.532 |
andowski | -0.515 |
ulhu | -0.512 |
otle | -0.499 |
Token | for |
Token ID | 329 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
Shinra | +1.818 |
Lowe | +1.714 |
phrine | +1.696 |
cens | +1.665 |
AFB | +1.590 |
Neg logit contributions | |
️ | -3.255 |
ilan | -3.217 |
ulhu | -3.131 |
andowski | -3.097 |
otle | -3.025 |
Token | the |
Token ID | 262 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
Shinra | +0.701 |
Lowe | +0.659 |
phrine | +0.657 |
cens | +0.641 |
AFB | +0.613 |
Neg logit contributions | |
️ | -0.959 |
ilan | -0.940 |
andowski | -0.919 |
ulhu | -0.908 |
otle | -0.874 |
Token | 2013 |
Token ID | 2211 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
️ | +1.268 |
ilan | +1.227 |
andowski | +1.186 |
ulhu | +1.173 |
otle | +1.130 |
Neg logit contributions | |
Shinra | -1.518 |
phrine | -1.451 |
Lowe | -1.436 |
cens | -1.416 |
AFB | -1.374 |
Token | / |
Token ID | 14 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
Shinra | +1.266 |
phrine | +1.209 |
Lowe | +1.205 |
cens | +1.188 |
AFB | +1.162 |
Neg logit contributions | |
️ | -1.013 |
ilan | -0.985 |
andowski | -0.946 |
ulhu | -0.939 |
otle | -0.904 |
Token | 14 |
Token ID | 1415 |
Feature activation | -6.25e-02 |
Pos logit contributions | |
️ | +5.272 |
andowski | +4.953 |
ilan | +4.777 |
otle | +4.667 |
leon | +4.400 |
Neg logit contributions | |
exting | -8.458 |
mosqu | -7.866 |
eleph | -7.826 |
Shinra | -7.761 |
phrine | -7.473 |
Token | Barclays |
Token ID | 37985 |
Feature activation | +3.59e-02 |
Pos logit contributions | |
️ | +0.512 |
ilan | +0.505 |
andowski | +0.489 |
ulhu | +0.487 |
otle | +0.473 |
Neg logit contributions | |
Shinra | -0.345 |
phrine | -0.324 |
Lowe | -0.323 |
cens | -0.315 |
AFB | -0.304 |
Token | Premier |
Token ID | 9952 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
Shinra | +0.188 |
phrine | +0.176 |
Lowe | +0.176 |
cens | +0.172 |
AFB | +0.165 |
Neg logit contributions | |
️ | -0.305 |
ilan | -0.298 |
andowski | -0.290 |
ulhu | -0.288 |
otle | -0.281 |
Token | League |
Token ID | 4041 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
️ | +1.469 |
ilan | +1.401 |
andowski | +1.351 |
ulhu | +1.270 |
otle | +1.263 |
Neg logit contributions | |
Shinra | -2.368 |
phrine | -2.280 |
cens | -2.241 |
Lowe | -2.220 |
AFB | -2.179 |
Token | season |
Token ID | 1622 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Shinra | +0.077 |
phrine | +0.072 |
Lowe | +0.072 |
cens | +0.070 |
AFB | +0.067 |
Neg logit contributions | |
️ | -0.113 |
ilan | -0.111 |
andowski | -0.107 |
ulhu | -0.107 |
otle | -0.104 |
Token | s |
Token ID | 82 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
Shinra | +0.707 |
phrine | +0.676 |
Lowe | +0.667 |
cens | +0.662 |
AFB | +0.643 |
Neg logit contributions | |
️ | -0.674 |
ilan | -0.651 |
andowski | -0.628 |
ulhu | -0.621 |
otle | -0.606 |
Token | frame |
Token ID | 5739 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
Shinra | +1.801 |
phrine | +1.712 |
Lowe | +1.697 |
cens | +1.676 |
AFB | +1.636 |
Neg logit contributions | |
️ | -1.725 |
ilan | -1.667 |
andowski | -1.612 |
ulhu | -1.591 |
otle | -1.552 |
Token | | |
Token ID | 930 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
Shinra | +1.971 |
Lowe | +1.839 |
phrine | +1.839 |
cens | +1.789 |
AFB | +1.756 |
Neg logit contributions | |
️ | -2.692 |
ilan | -2.621 |
andowski | -2.546 |
ulhu | -2.519 |
otle | -2.454 |
Token | size |
Token ID | 2546 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
️ | +1.945 |
ilan | +1.872 |
andowski | +1.797 |
ulhu | +1.773 |
otle | +1.735 |
Neg logit contributions | |
Shinra | -2.217 |
phrine | -2.136 |
Lowe | -2.095 |
cens | -2.075 |
AFB | -2.011 |
Token | _ |
Token ID | 62 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
️ | +3.292 |
ilan | +3.076 |
andowski | +2.939 |
otle | +2.847 |
ulhu | +2.690 |
Neg logit contributions | |
Shinra | -6.996 |
exting | -6.954 |
mosqu | -6.811 |
phrine | -6.702 |
eleph | -6.679 |
Token | t |
Token ID | 83 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
️ | +2.967 |
ilan | +2.797 |
andowski | +2.622 |
otle | +2.479 |
leon | +2.319 |
Neg logit contributions | |
Shinra | -7.925 |
exting | -7.784 |
mosqu | -7.608 |
Lowe | -7.599 |
phrine | -7.568 |
Token | count |
Token ID | 954 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
️ | +4.489 |
ilan | +4.110 |
andowski | +3.952 |
ulhu | +3.814 |
otle | +3.715 |
Neg logit contributions | |
Shinra | -8.847 |
phrine | -8.587 |
cens | -8.387 |
Lowe | -8.303 |
AFB | -8.073 |
Token | | |
Token ID | 930 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
Shinra | +0.663 |
phrine | +0.627 |
Lowe | +0.618 |
cens | +0.611 |
AFB | +0.589 |
Neg logit contributions | |
️ | -0.813 |
ilan | -0.788 |
andowski | -0.765 |
ulhu | -0.753 |
otle | -0.741 |
Token | + |
Token ID | 1343 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
Shinra | +2.308 |
Lowe | +2.102 |
phrine | +2.058 |
AFB | +2.003 |
cens | +1.975 |
Neg logit contributions | |
️ | -4.698 |
ilan | -4.640 |
andowski | -4.526 |
ulhu | -4.444 |
otle | -4.362 |
Token | -------------------- |
Token ID | 19351 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Shinra | +1.257 |
phrine | +1.217 |
Lowe | +1.207 |
cens | +1.198 |
AFB | +1.176 |
Neg logit contributions | |
️ | -0.377 |
ilan | -0.350 |
andowski | -0.323 |
ulhu | -0.312 |
otle | -0.297 |
Token | --+ |
Token ID | 44785 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
Shinra | +0.809 |
phrine | +0.769 |
Lowe | +0.760 |
cens | +0.752 |
AFB | +0.729 |
Neg logit contributions | |
️ | -0.799 |
ilan | -0.772 |
andowski | -0.746 |
ulhu | -0.735 |
otle | -0.720 |
Token | like |
Token ID | 588 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Shinra | +2.413 |
Lowe | +2.291 |
phrine | +2.282 |
AFB | +2.204 |
cens | +2.198 |
Neg logit contributions | |
️ | -3.090 |
ilan | -3.006 |
ulhu | -2.949 |
andowski | -2.935 |
otle | -2.848 |
Token | it |
Token ID | 340 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
Shinra | +0.810 |
Lowe | +0.763 |
phrine | +0.761 |
cens | +0.745 |
AFB | +0.729 |
Neg logit contributions | |
️ | -1.064 |
ilan | -1.032 |
andowski | -1.003 |
ulhu | -0.997 |
otle | -0.969 |
Token | wanted |
Token ID | 2227 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
️ | +0.185 |
ilan | +0.180 |
andowski | +0.174 |
ulhu | +0.172 |
otle | +0.168 |
Neg logit contributions | |
Shinra | -0.185 |
phrine | -0.176 |
Lowe | -0.174 |
cens | -0.172 |
AFB | -0.168 |
Token | to |
Token ID | 284 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
Shinra | +1.192 |
phrine | +1.129 |
Lowe | +1.116 |
AFB | +1.084 |
cens | +1.079 |
Neg logit contributions | |
️ | -1.847 |
ilan | -1.769 |
andowski | -1.747 |
ulhu | -1.744 |
otle | -1.685 |
Token | stall |
Token ID | 22549 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
Shinra | +2.005 |
phrine | +1.886 |
Lowe | +1.882 |
AFB | +1.833 |
cens | +1.824 |
Neg logit contributions | |
️ | -2.261 |
ilan | -2.217 |
ulhu | -2.136 |
andowski | -2.134 |
otle | -2.055 |
Token | at |
Token ID | 379 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
Shinra | +0.678 |
phrine | +0.640 |
Lowe | +0.632 |
cens | +0.624 |
AFB | +0.608 |
Neg logit contributions | |
️ | -0.872 |
ilan | -0.847 |
andowski | -0.824 |
ulhu | -0.814 |
otle | -0.796 |
Token | the |
Token ID | 262 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
Shinra | +5.067 |
Lowe | +4.855 |
AFB | +4.564 |
clipping | +4.518 |
phrine | +4.468 |
Neg logit contributions | |
️ | -6.940 |
ilan | -6.613 |
ulhu | -6.563 |
andowski | -6.471 |
otle | -6.336 |
Token | lower |
Token ID | 2793 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
Shinra | +2.811 |
Lowe | +2.639 |
phrine | +2.591 |
AFB | +2.553 |
cens | +2.544 |
Neg logit contributions | |
️ | -3.302 |
ilan | -3.132 |
ulhu | -3.075 |
andowski | -3.063 |
otle | -2.995 |
Token | air |
Token ID | 1633 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
Shinra | +0.108 |
phrine | +0.102 |
Lowe | +0.102 |
cens | +0.100 |
AFB | +0.098 |
Neg logit contributions | |
️ | -0.108 |
ilan | -0.104 |
andowski | -0.101 |
ulhu | -0.100 |
otle | -0.097 |
Token | spe |
Token ID | 4125 |
Feature activation | +9.19e-02 |
Pos logit contributions | |
Shinra | +0.265 |
phrine | +0.253 |
Lowe | +0.249 |
cens | +0.249 |
AFB | +0.239 |
Neg logit contributions | |
️ | -0.314 |
ilan | -0.303 |
andowski | -0.296 |
ulhu | -0.294 |
otle | -0.284 |
Token | eds |
Token ID | 5379 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
Shinra | +0.594 |
phrine | +0.575 |
Lowe | +0.563 |
cens | +0.556 |
AFB | +0.534 |
Neg logit contributions | |
️ | -0.677 |
ilan | -0.639 |
andowski | -0.631 |
ulhu | -0.620 |
otle | -0.606 |
Token | your |
Token ID | 534 |
Feature activation | -6.02e-03 |
Pos logit contributions | |
️ | +2.435 |
ilan | +2.348 |
andowski | +2.259 |
ulhu | +2.218 |
otle | +2.218 |
Neg logit contributions | |
Shinra | -2.095 |
phrine | -1.993 |
Lowe | -1.957 |
cens | -1.948 |
AFB | -1.886 |
Token | inbox |
Token ID | 13734 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
️ | +0.040 |
ilan | +0.039 |
andowski | +0.037 |
ulhu | +0.037 |
otle | +0.036 |
Neg logit contributions | |
Shinra | -0.043 |
phrine | -0.041 |
Lowe | -0.040 |
cens | -0.040 |
AFB | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
️ | +0.238 |
ilan | +0.231 |
andowski | +0.223 |
ulhu | +0.223 |
otle | +0.217 |
Neg logit contributions | |
Shinra | -0.204 |
phrine | -0.194 |
Lowe | -0.190 |
cens | -0.189 |
AFB | -0.182 |
Token | get |
Token ID | 651 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
Shinra | +2.198 |
Lowe | +2.066 |
phrine | +2.047 |
cens | +1.974 |
AFB | +1.918 |
Neg logit contributions | |
️ | -2.963 |
ilan | -2.900 |
andowski | -2.765 |
ulhu | -2.753 |
otle | -2.711 |
Token | our |
Token ID | 674 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
Shinra | +2.315 |
Lowe | +2.195 |
phrine | +2.147 |
cens | +2.052 |
clipping | +2.023 |
Neg logit contributions | |
️ | -3.186 |
ilan | -3.043 |
ulhu | -3.007 |
andowski | -2.955 |
otle | -2.850 |
Token | Daily |
Token ID | 6714 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Shinra | +3.141 |
Lowe | +2.915 |
phrine | +2.905 |
cens | +2.879 |
clipping | +2.751 |
Neg logit contributions | |
️ | -3.715 |
ilan | -3.610 |
andowski | -3.545 |
ulhu | -3.455 |
otle | -3.372 |
Token | Dispatch |
Token ID | 35934 |
Feature activation | +6.09e-02 |
Pos logit contributions | |
Shinra | +4.245 |
Lowe | +4.044 |
phrine | +4.013 |
cens | +4.010 |
clipping | +3.717 |
Neg logit contributions | |
️ | -9.215 |
ilan | -8.934 |
otle | -8.668 |
andowski | -8.617 |
ulhu | -8.551 |
Token | and |
Token ID | 290 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
Shinra | +0.389 |
phrine | +0.370 |
Lowe | +0.364 |
cens | +0.360 |
AFB | +0.348 |
Neg logit contributions | |
️ | -0.449 |
ilan | -0.435 |
andowski | -0.421 |
ulhu | -0.417 |
otle | -0.407 |
Token | Editor |
Token ID | 12058 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
Shinra | +0.630 |
phrine | +0.599 |
Lowe | +0.592 |
cens | +0.582 |
AFB | +0.566 |
Neg logit contributions | |
️ | -0.631 |
ilan | -0.607 |
andowski | -0.588 |
ulhu | -0.580 |
otle | -0.570 |
Token | 's |
Token ID | 338 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
Shinra | +3.413 |
phrine | +3.280 |
Lowe | +3.213 |
cens | +3.164 |
AFB | +3.090 |
Neg logit contributions | |
️ | -3.388 |
ilan | -3.283 |
ulhu | -3.195 |
andowski | -3.148 |
otle | -3.018 |
Token | Picks |
Token ID | 33957 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
️ | +1.521 |
ilan | +1.489 |
andowski | +1.436 |
ulhu | +1.408 |
otle | +1.401 |
Neg logit contributions | |
Shinra | -1.173 |
phrine | -1.096 |
cens | -1.072 |
Lowe | -1.072 |
AFB | -1.039 |
Token | [ |
Token ID | 58 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
Shinra | +1.281 |
phrine | +1.177 |
Lowe | +1.176 |
cens | +1.151 |
AFB | +1.112 |
Neg logit contributions | |
️ | -1.920 |
ilan | -1.876 |
andowski | -1.819 |
ulhu | -1.808 |
otle | -1.769 |
Token | # |
Token ID | 2 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
Shinra | +0.945 |
phrine | +0.885 |
Lowe | +0.882 |
cens | +0.870 |
AFB | +0.841 |
Neg logit contributions | |
️ | -1.035 |
ilan | -1.006 |
andowski | -0.977 |
ulhu | -0.961 |
otle | -0.945 |
Token | == |
Token ID | 6624 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
Shinra | +2.431 |
Lowe | +2.211 |
phrine | +2.190 |
cens | +2.166 |
AFB | +2.089 |
Neg logit contributions | |
️ | -3.915 |
ilan | -3.857 |
andowski | -3.775 |
ulhu | -3.755 |
otle | -3.660 |
Token | " |
Token ID | 366 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Shinra | +1.676 |
Lowe | +1.527 |
phrine | +1.494 |
cens | +1.460 |
AFB | +1.424 |
Neg logit contributions | |
️ | -2.957 |
ilan | -2.927 |
andowski | -2.857 |
ulhu | -2.809 |
otle | -2.760 |
Token | E |
Token ID | 36 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
Shinra | +1.418 |
phrine | +1.343 |
Lowe | +1.323 |
cens | +1.319 |
AFB | +1.275 |
Neg logit contributions | |
️ | -1.528 |
ilan | -1.483 |
andowski | -1.437 |
ulhu | -1.422 |
otle | -1.390 |
Token | " |
Token ID | 1 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
Shinra | +1.761 |
phrine | +1.676 |
Lowe | +1.660 |
cens | +1.626 |
AFB | +1.586 |
Neg logit contributions | |
️ | -1.580 |
ilan | -1.522 |
andowski | -1.468 |
ulhu | -1.442 |
otle | -1.419 |
Token | || |
Token ID | 8614 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
Shinra | +4.848 |
Lowe | +4.335 |
phrine | +4.307 |
cens | +4.092 |
AFB | +4.079 |
Neg logit contributions | |
️ | -8.295 |
ilan | -8.123 |
andowski | -7.988 |
ulhu | -7.869 |
otle | -7.668 |
Token | # |
Token ID | 1303 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
Shinra | +3.442 |
Lowe | +3.129 |
cens | +2.968 |
RET | +2.954 |
AFB | +2.937 |
Neg logit contributions | |
ilan | -6.132 |
️ | -5.966 |
andowski | -5.952 |
ulhu | -5.824 |
otle | -5.669 |
Token | == |
Token ID | 6624 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
Shinra | +2.455 |
Lowe | +2.200 |
phrine | +2.150 |
cens | +2.146 |
AFB | +2.103 |
Neg logit contributions | |
️ | -5.700 |
ilan | -5.646 |
ulhu | -5.550 |
andowski | -5.520 |
otle | -5.367 |
Token | " |
Token ID | 366 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
Shinra | +2.212 |
Lowe | +2.037 |
phrine | +1.987 |
cens | +1.963 |
AFB | +1.917 |
Neg logit contributions | |
️ | -4.336 |
ilan | -4.252 |
andowski | -4.153 |
ulhu | -4.132 |
otle | -4.006 |
Token | A |
Token ID | 32 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
Shinra | +1.691 |
phrine | +1.571 |
cens | +1.550 |
Lowe | +1.546 |
AFB | +1.485 |
Neg logit contributions | |
️ | -2.800 |
ilan | -2.728 |
andowski | -2.653 |
ulhu | -2.649 |
otle | -2.574 |
Token | author |
Token ID | 1772 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
Shinra | +0.939 |
Lowe | +0.929 |
phrine | +0.923 |
cens | +0.896 |
clipping | +0.883 |
Neg logit contributions | |
️ | -0.989 |
ulhu | -0.954 |
ilan | -0.920 |
andowski | -0.902 |
otle | -0.848 |
Token | , |
Token ID | 11 |
Feature activation | +9.08e-04 |
Pos logit contributions | |
Shinra | +0.294 |
phrine | +0.280 |
cens | +0.277 |
Lowe | +0.276 |
AFB | +0.265 |
Neg logit contributions | |
️ | -0.390 |
ilan | -0.375 |
ulhu | -0.370 |
andowski | -0.364 |
otle | -0.355 |
Token | not |
Token ID | 407 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
Shinra | +0.005 |
Lowe | +0.005 |
cens | +0.005 |
phrine | +0.005 |
AFB | +0.005 |
Neg logit contributions | |
️ | -0.007 |
ulhu | -0.007 |
ilan | -0.007 |
andowski | -0.007 |
otle | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
Shinra | +2.085 |
Lowe | +1.990 |
cens | +1.916 |
Unch | +1.890 |
phrine | +1.837 |
Neg logit contributions | |
️ | -3.207 |
andowski | -3.146 |
ulhu | -3.134 |
ilan | -3.088 |
otle | -2.978 |
Token | position |
Token ID | 2292 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
Shinra | +3.180 |
Lowe | +3.055 |
phrine | +3.012 |
clipping | +2.984 |
cens | +2.964 |
Neg logit contributions | |
️ | -3.324 |
ulhu | -3.249 |
andowski | -3.139 |
ilan | -3.086 |
otle | -2.982 |
Token | of |
Token ID | 286 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
Shinra | +2.131 |
cens | +2.037 |
phrine | +2.018 |
Lowe | +1.988 |
clipping | +1.898 |
Neg logit contributions | |
️ | -3.051 |
ilan | -2.974 |
ulhu | -2.909 |
andowski | -2.901 |
otle | -2.795 |
Token | EUR |
Token ID | 27526 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
cens | +3.796 |
Unch | +3.710 |
Lowe | +3.674 |
Shinra | +3.668 |
RET | +3.464 |
Neg logit contributions | |
ulhu | -8.114 |
andowski | -7.957 |
️ | -7.947 |
uca | -7.384 |
otle | -7.267 |
Token | OP |
Token ID | 3185 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
Shinra | +1.212 |
phrine | +1.145 |
cens | +1.145 |
Lowe | +1.133 |
AFB | +1.088 |
Neg logit contributions | |
️ | -1.676 |
andowski | -1.636 |
ilan | -1.624 |
ulhu | -1.608 |
otle | -1.526 |
Token | P |
Token ID | 47 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
Shinra | +1.931 |
phrine | +1.824 |
cens | +1.802 |
Lowe | +1.793 |
AFB | +1.704 |
Neg logit contributions | |
️ | -2.300 |
ilan | -2.222 |
andowski | -2.151 |
ulhu | -2.128 |
otle | -2.092 |
Token | – |
Token ID | 784 |
Feature activation | +7.99e-01 |
Pos logit contributions | |
Shinra | +4.559 |
Lowe | +4.301 |
cens | +4.168 |
clipping | +4.114 |
RET | +4.055 |
Neg logit contributions | |
️ | -4.674 |
ulhu | -4.382 |
andowski | -4.282 |
ilan | -4.228 |
otle | -4.198 |
Token | European |
Token ID | 3427 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
Shinra | +5.098 |
Lowe | +5.048 |
cens | +4.664 |
clipping | +4.586 |
Unch | +4.533 |
Neg logit contributions | |
ulhu | -5.791 |
️ | -5.608 |
andowski | -5.546 |
otle | -5.235 |
ilan | -5.088 |
Token | 62 |
Token ID | 8190 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
Shinra | +2.587 |
Lowe | +2.481 |
phrine | +2.451 |
cens | +2.409 |
Unch | +2.343 |
Neg logit contributions | |
️ | -3.265 |
ilan | -3.222 |
ulhu | -3.121 |
andowski | -3.111 |
otle | -2.995 |
Token | ( |
Token ID | 357 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
️ | +0.319 |
ilan | +0.309 |
andowski | +0.297 |
ulhu | +0.294 |
otle | +0.288 |
Neg logit contributions | |
Shinra | -0.345 |
phrine | -0.329 |
Lowe | -0.327 |
cens | -0.323 |
AFB | -0.313 |
Token | 5 |
Token ID | 20 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
️ | +1.103 |
ilan | +1.070 |
andowski | +1.035 |
ulhu | +1.017 |
otle | +1.003 |
Neg logit contributions | |
Shinra | -0.849 |
phrine | -0.801 |
Lowe | -0.789 |
cens | -0.772 |
AFB | -0.753 |
Token | ) |
Token ID | 8 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
️ | +0.378 |
ilan | +0.364 |
andowski | +0.352 |
ulhu | +0.345 |
otle | +0.339 |
Neg logit contributions | |
Shinra | -0.388 |
phrine | -0.369 |
Lowe | -0.364 |
cens | -0.359 |
AFB | -0.350 |
Token | 401 |
Token ID | 22219 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Shinra | +1.946 |
Lowe | +1.852 |
phrine | +1.835 |
cens | +1.822 |
AFB | +1.752 |
Neg logit contributions | |
️ | -2.025 |
ilan | -1.980 |
andowski | -1.915 |
ulhu | -1.894 |
otle | -1.853 |
Token | - |
Token ID | 12 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Shinra | +1.028 |
phrine | +0.968 |
Lowe | +0.959 |
cens | +0.949 |
AFB | +0.910 |
Neg logit contributions | |
️ | -1.290 |
ilan | -1.254 |
andowski | -1.226 |
ulhu | -1.208 |
otle | -1.181 |
Token | 414 |
Token ID | 37309 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
Shinra | +6.609 |
cens | +6.222 |
Lowe | +6.196 |
phrine | +6.044 |
AFB | +5.900 |
Neg logit contributions | |
️ | -8.808 |
ilan | -8.655 |
ulhu | -8.531 |
andowski | -8.506 |
imar | -7.884 |
Token | ; |
Token ID | 26 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
Shinra | +3.607 |
Lowe | +3.357 |
phrine | +3.338 |
cens | +3.277 |
AFB | +3.116 |
Neg logit contributions | |
️ | -4.526 |
ilan | -4.496 |
andowski | -4.469 |
ulhu | -4.412 |
otle | -4.126 |
Token | doi |
Token ID | 23899 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
Shinra | +1.474 |
Lowe | +1.378 |
phrine | +1.308 |
cens | +1.273 |
AFB | +1.205 |
Neg logit contributions | |
️ | -3.895 |
ilan | -3.843 |
ulhu | -3.795 |
andowski | -3.779 |
otle | -3.645 |
Token | : |
Token ID | 25 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
️ | +0.353 |
ilan | +0.340 |
andowski | +0.328 |
ulhu | +0.322 |
otle | +0.316 |
Neg logit contributions | |
Shinra | -0.409 |
phrine | -0.391 |
Lowe | -0.386 |
cens | -0.382 |
AFB | -0.371 |
Token | 10 |
Token ID | 838 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
️ | +2.078 |
ilan | +1.996 |
andowski | +1.914 |
ulhu | +1.879 |
otle | +1.833 |
Neg logit contributions | |
Shinra | -2.946 |
phrine | -2.825 |
Lowe | -2.795 |
cens | -2.769 |
AFB | -2.700 |
Token | more |
Token ID | 517 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
️ | +1.668 |
ilan | +1.614 |
andowski | +1.569 |
ulhu | +1.553 |
otle | +1.520 |
Neg logit contributions | |
Shinra | -1.358 |
phrine | -1.284 |
Lowe | -1.271 |
cens | -1.251 |
AFB | -1.209 |
Token | about |
Token ID | 546 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
️ | +1.452 |
ilan | +1.409 |
andowski | +1.368 |
ulhu | +1.355 |
otle | +1.323 |
Neg logit contributions | |
Shinra | -1.245 |
phrine | -1.180 |
Lowe | -1.167 |
cens | -1.149 |
AFB | -1.110 |
Token | self |
Token ID | 2116 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
️ | +0.603 |
ilan | +0.584 |
andowski | +0.567 |
ulhu | +0.562 |
otle | +0.549 |
Neg logit contributions | |
Shinra | -0.511 |
phrine | -0.484 |
Lowe | -0.479 |
cens | -0.472 |
AFB | -0.456 |
Token | - |
Token ID | 12 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
️ | +2.077 |
ilan | +2.021 |
andowski | +1.963 |
ulhu | +1.951 |
otle | +1.906 |
Neg logit contributions | |
Shinra | -1.669 |
phrine | -1.572 |
Lowe | -1.570 |
cens | -1.550 |
AFB | -1.495 |
Token | per |
Token ID | 525 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
️ | +1.351 |
ilan | +1.306 |
andowski | +1.263 |
ulhu | +1.247 |
otle | +1.222 |
Neg logit contributions | |
Shinra | -1.302 |
phrine | -1.238 |
Lowe | -1.223 |
cens | -1.211 |
AFB | -1.173 |
Token | ception |
Token ID | 4516 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
️ | +0.346 |
ilan | +0.327 |
andowski | +0.317 |
ulhu | +0.314 |
otle | +0.305 |
Neg logit contributions | |
Shinra | -0.476 |
phrine | -0.462 |
cens | -0.455 |
Lowe | -0.453 |
AFB | -0.438 |
Token | than |
Token ID | 621 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
Shinra | +1.184 |
phrine | +1.123 |
Lowe | +1.109 |
cens | +1.089 |
AFB | +1.049 |
Neg logit contributions | |
️ | -1.635 |
ilan | -1.620 |
ulhu | -1.565 |
andowski | -1.563 |
otle | -1.518 |
Token | a |
Token ID | 257 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
️ | +0.600 |
ilan | +0.581 |
ulhu | +0.563 |
andowski | +0.563 |
otle | +0.547 |
Neg logit contributions | |
Shinra | -0.517 |
phrine | -0.487 |
Lowe | -0.486 |
cens | -0.475 |
AFB | -0.462 |
Token | reflection |
Token ID | 14580 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
️ | +0.252 |
ilan | +0.242 |
andowski | +0.236 |
ulhu | +0.234 |
otle | +0.227 |
Neg logit contributions | |
Shinra | -0.261 |
Lowe | -0.246 |
phrine | -0.245 |
cens | -0.241 |
AFB | -0.234 |
Token | of |
Token ID | 286 |
Feature activation | -5.24e-03 |
Pos logit contributions | |
️ | +0.298 |
ilan | +0.295 |
ulhu | +0.287 |
andowski | +0.285 |
otle | +0.274 |
Neg logit contributions | |
Shinra | -0.214 |
phrine | -0.203 |
Lowe | -0.200 |
cens | -0.197 |
AFB | -0.189 |
Token | material |
Token ID | 2587 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
️ | +0.039 |
ilan | +0.038 |
ulhu | +0.037 |
andowski | +0.037 |
otle | +0.035 |
Neg logit contributions | |
Shinra | -0.033 |
Lowe | -0.032 |
phrine | -0.031 |
cens | -0.031 |
AFB | -0.030 |
Token | : |
Token ID | 25 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
️ | +0.759 |
ilan | +0.742 |
andowski | +0.718 |
ulhu | +0.713 |
otle | +0.696 |
Neg logit contributions | |
Shinra | -0.604 |
phrine | -0.569 |
Lowe | -0.565 |
cens | -0.552 |
AFB | -0.537 |
Token | homelessness |
Token ID | 28822 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
️ | +1.119 |
ilan | +1.086 |
andowski | +1.050 |
ulhu | +1.039 |
otle | +1.019 |
Neg logit contributions | |
Shinra | -0.943 |
phrine | -0.891 |
Lowe | -0.881 |
cens | -0.870 |
AFB | -0.841 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
️ | +0.753 |
ilan | +0.733 |
andowski | +0.709 |
ulhu | +0.701 |
otle | +0.686 |
Neg logit contributions | |
Shinra | -0.615 |
phrine | -0.583 |
Lowe | -0.576 |
cens | -0.568 |
AFB | -0.550 |
Token | The |
Token ID | 383 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
️ | +1.236 |
ilan | +1.198 |
andowski | +1.160 |
ulhu | +1.144 |
otle | +1.122 |
Neg logit contributions | |
Shinra | -1.093 |
phrine | -1.037 |
Lowe | -1.023 |
cens | -1.011 |
AFB | -0.979 |
Token | mayor |
Token ID | 9591 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
️ | +2.053 |
ilan | +1.984 |
andowski | +1.921 |
ulhu | +1.894 |
otle | +1.860 |
Neg logit contributions | |
Shinra | -1.870 |
phrine | -1.777 |
Lowe | -1.747 |
cens | -1.735 |
AFB | -1.679 |
Token | walked |
Token ID | 6807 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
️ | +1.645 |
ilan | +1.595 |
andowski | +1.543 |
ulhu | +1.523 |
otle | +1.493 |
Neg logit contributions | |
Shinra | -1.471 |
phrine | -1.396 |
Lowe | -1.378 |
cens | -1.361 |
AFB | -1.319 |
Token | into |
Token ID | 656 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
Shinra | +1.892 |
Lowe | +1.808 |
phrine | +1.776 |
cens | +1.711 |
AFB | +1.667 |
Neg logit contributions | |
️ | -2.842 |
ilan | -2.780 |
ulhu | -2.710 |
andowski | -2.671 |
otle | -2.605 |
Token | City |
Token ID | 2254 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
Shinra | +0.470 |
Lowe | +0.450 |
phrine | +0.442 |
cens | +0.422 |
AFB | +0.416 |
Neg logit contributions | |
️ | -0.558 |
ilan | -0.548 |
andowski | -0.526 |
ulhu | -0.521 |
otle | -0.506 |
Token | Hall |
Token ID | 4789 |
Feature activation | +5.69e-03 |
Pos logit contributions | |
Shinra | +0.343 |
Lowe | +0.326 |
phrine | +0.325 |
cens | +0.319 |
AFB | +0.311 |
Neg logit contributions | |
️ | -0.360 |
ilan | -0.351 |
andowski | -0.337 |
ulhu | -0.337 |
otle | -0.326 |
Token | in |
Token ID | 287 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
Shinra | +0.035 |
phrine | +0.033 |
Lowe | +0.033 |
cens | +0.032 |
AFB | +0.031 |
Neg logit contributions | |
️ | -0.043 |
ilan | -0.042 |
andowski | -0.041 |
ulhu | -0.040 |
otle | -0.040 |
Token | 2002 |
Token ID | 6244 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
Shinra | +0.097 |
Lowe | +0.092 |
phrine | +0.092 |
cens | +0.089 |
AFB | +0.087 |
Neg logit contributions | |
️ | -0.111 |
ilan | -0.108 |
andowski | -0.105 |
ulhu | -0.104 |
otle | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
Shinra | +2.869 |
Lowe | +2.683 |
phrine | +2.682 |
cens | +2.622 |
AFB | +2.481 |
Neg logit contributions | |
️ | -3.857 |
ilan | -3.794 |
andowski | -3.781 |
ulhu | -3.687 |
otle | -3.565 |
Token |
|
Token ID | 198 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Shinra | +0.602 |
phrine | +0.571 |
Lowe | +0.565 |
cens | +0.559 |
AFB | +0.542 |
Neg logit contributions | |
️ | -0.655 |
ilan | -0.635 |
andowski | -0.615 |
ulhu | -0.607 |
otle | -0.593 |
Token | I |
Token ID | 40 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
Shinra | +1.291 |
phrine | +1.228 |
cens | +1.217 |
Lowe | +1.197 |
AFB | +1.139 |
Neg logit contributions | |
️ | -1.629 |
ilan | -1.586 |
andowski | -1.550 |
ulhu | -1.538 |
otle | -1.501 |
Token | agree |
Token ID | 4236 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
Shinra | +1.765 |
phrine | +1.693 |
Lowe | +1.654 |
cens | +1.636 |
AFB | +1.574 |
Neg logit contributions | |
️ | -1.985 |
ilan | -1.974 |
andowski | -1.924 |
ulhu | -1.913 |
otle | -1.849 |
Token | with |
Token ID | 351 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
️ | +0.386 |
ilan | +0.376 |
andowski | +0.365 |
ulhu | +0.360 |
otle | +0.353 |
Neg logit contributions | |
Shinra | -0.311 |
phrine | -0.294 |
Lowe | -0.290 |
cens | -0.287 |
AFB | -0.277 |
Token | this |
Token ID | 428 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
Shinra | +2.324 |
Lowe | +2.181 |
phrine | +2.128 |
cens | +2.080 |
AFB | +2.021 |
Neg logit contributions | |
️ | -3.059 |
ilan | -3.019 |
andowski | -2.947 |
ulhu | -2.897 |
otle | -2.790 |
Token | , |
Token ID | 11 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
Shinra | +1.655 |
Lowe | +1.544 |
phrine | +1.543 |
cens | +1.517 |
AFB | +1.462 |
Neg logit contributions | |
️ | -1.887 |
ilan | -1.850 |
andowski | -1.803 |
ulhu | -1.790 |
otle | -1.718 |
Token | but |
Token ID | 475 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
Shinra | +0.821 |
phrine | +0.771 |
Lowe | +0.769 |
cens | +0.756 |
AFB | +0.727 |
Neg logit contributions | |
️ | -1.024 |
ilan | -0.996 |
andowski | -0.970 |
ulhu | -0.960 |
otle | -0.934 |
Token | there |
Token ID | 612 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
Shinra | +1.472 |
Lowe | +1.368 |
phrine | +1.367 |
cens | +1.338 |
AFB | +1.296 |
Neg logit contributions | |
️ | -1.887 |
ilan | -1.861 |
andowski | -1.808 |
ulhu | -1.797 |
otle | -1.740 |
Token | is |
Token ID | 318 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
Shinra | +0.501 |
phrine | +0.468 |
Lowe | +0.465 |
cens | +0.458 |
AFB | +0.443 |
Neg logit contributions | |
️ | -0.620 |
ilan | -0.609 |
andowski | -0.593 |
ulhu | -0.584 |
otle | -0.565 |
Token | not |
Token ID | 407 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
️ | +1.115 |
ilan | +1.083 |
andowski | +1.049 |
ulhu | +1.036 |
otle | +1.014 |
Neg logit contributions | |
Shinra | -0.983 |
phrine | -0.928 |
Lowe | -0.917 |
cens | -0.906 |
AFB | -0.879 |
Token | the |
Token ID | 262 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
Shinra | +1.467 |
Lowe | +1.385 |
phrine | +1.357 |
cens | +1.337 |
AFB | +1.321 |
Neg logit contributions | |
️ | -1.668 |
ilan | -1.625 |
ulhu | -1.576 |
andowski | -1.573 |
otle | -1.524 |
Token | centre |
Token ID | 7372 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
Shinra | +0.891 |
Lowe | +0.841 |
phrine | +0.828 |
cens | +0.818 |
AFB | +0.803 |
Neg logit contributions | |
️ | -0.893 |
ilan | -0.868 |
ulhu | -0.838 |
andowski | -0.837 |
otle | -0.813 |
Token | , |
Token ID | 11 |
Feature activation | -9.71e-02 |
Pos logit contributions | |
Shinra | +0.088 |
phrine | +0.082 |
Lowe | +0.082 |
cens | +0.081 |
AFB | +0.079 |
Neg logit contributions | |
️ | -0.105 |
ilan | -0.102 |
andowski | -0.099 |
ulhu | -0.098 |
otle | -0.096 |
Token | out |
Token ID | 503 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
️ | +0.713 |
ilan | +0.690 |
andowski | +0.667 |
ulhu | +0.656 |
otle | +0.646 |
Neg logit contributions | |
Shinra | -0.622 |
phrine | -0.592 |
Lowe | -0.581 |
cens | -0.575 |
AFB | -0.557 |
Token | res |
Token ID | 411 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
Shinra | +0.219 |
Lowe | +0.205 |
phrine | +0.204 |
cens | +0.201 |
AFB | +0.195 |
Neg logit contributions | |
️ | -0.282 |
ilan | -0.273 |
andowski | -0.267 |
ulhu | -0.266 |
otle | -0.258 |
Token | olving |
Token ID | 10890 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
Shinra | +0.658 |
phrine | +0.623 |
Lowe | +0.617 |
cens | +0.612 |
AFB | +0.591 |
Neg logit contributions | |
️ | -0.759 |
ilan | -0.731 |
andowski | -0.713 |
ulhu | -0.707 |
otle | -0.690 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Shinra | +2.203 |
Lowe | +2.054 |
phrine | +2.010 |
cens | +1.999 |
clipping | +1.962 |
Neg logit contributions | |
️ | -2.740 |
ilan | -2.706 |
andowski | -2.634 |
ulhu | -2.625 |
otle | -2.526 |
Token | sensor |
Token ID | 12694 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
Shinra | +0.671 |
phrine | +0.629 |
Lowe | +0.627 |
cens | +0.617 |
AFB | +0.600 |
Neg logit contributions | |
️ | -0.776 |
ilan | -0.755 |
andowski | -0.732 |
ulhu | -0.727 |
otle | -0.709 |
Token | at |
Token ID | 379 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
Shinra | +1.558 |
phrine | +1.459 |
Lowe | +1.459 |
cens | +1.426 |
AFB | +1.397 |
Neg logit contributions | |
️ | -1.861 |
ilan | -1.820 |
andowski | -1.763 |
ulhu | -1.753 |
otle | -1.700 |
Token | all |
Token ID | 477 |
Feature activation | +6.19e-02 |
Pos logit contributions | |
Shinra | +2.613 |
Lowe | +2.466 |
phrine | +2.386 |
cens | +2.349 |
clipping | +2.348 |
Neg logit contributions | |
️ | -3.145 |
ilan | -3.074 |
ulhu | -2.994 |
andowski | -2.986 |
otle | -2.872 |
Token | a |
Token ID | 257 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
Shinra | +0.407 |
phrine | +0.384 |
Lowe | +0.381 |
cens | +0.375 |
AFB | +0.365 |
Neg logit contributions | |
️ | -0.444 |
ilan | -0.434 |
andowski | -0.419 |
ulhu | -0.416 |
otle | -0.406 |
Token | mechanical |
Token ID | 12370 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Shinra | +0.063 |
phrine | +0.060 |
Lowe | +0.060 |
cens | +0.058 |
AFB | +0.057 |
Neg logit contributions | |
️ | -0.068 |
ilan | -0.066 |
andowski | -0.064 |
ulhu | -0.064 |
otle | -0.062 |
Token | - |
Token ID | 12 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
Shinra | +0.799 |
Lowe | +0.757 |
phrine | +0.748 |
cens | +0.726 |
AFB | +0.713 |
Neg logit contributions | |
️ | -0.883 |
ilan | -0.855 |
andowski | -0.833 |
ulhu | -0.829 |
otle | -0.805 |
Token | locking |
Token ID | 48331 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
Shinra | +0.707 |
phrine | +0.674 |
Lowe | +0.666 |
cens | +0.665 |
AFB | +0.634 |
Neg logit contributions | |
️ | -0.727 |
ilan | -0.699 |
andowski | -0.680 |
ulhu | -0.679 |
otle | -0.656 |
Token | four |
Token ID | 1440 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
️ | +0.359 |
ilan | +0.348 |
andowski | +0.338 |
ulhu | +0.334 |
otle | +0.326 |
Neg logit contributions | |
Shinra | -0.318 |
phrine | -0.299 |
Lowe | -0.298 |
cens | -0.292 |
AFB | -0.284 |
Token | - |
Token ID | 12 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
Shinra | +2.726 |
Lowe | +2.595 |
phrine | +2.534 |
cens | +2.474 |
AFB | +2.402 |
Neg logit contributions | |
️ | -3.846 |
ilan | -3.795 |
andowski | -3.693 |
ulhu | -3.677 |
otle | -3.622 |
Token | wheel |
Token ID | 22001 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
Shinra | +1.464 |
phrine | +1.377 |
Lowe | +1.369 |
cens | +1.362 |
AFB | +1.263 |
Neg logit contributions | |
️ | -2.263 |
ilan | -2.204 |
andowski | -2.142 |
ulhu | -2.142 |
otle | -2.099 |
Token | drive |
Token ID | 3708 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
Shinra | +3.351 |
Lowe | +3.290 |
phrine | +3.185 |
cens | +3.057 |
AFB | +3.050 |
Neg logit contributions | |
️ | -4.081 |
ilan | -3.934 |
andowski | -3.826 |
ulhu | -3.809 |
otle | -3.723 |
Token | . |
Token ID | 13 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Shinra | +1.550 |
Lowe | +1.466 |
phrine | +1.465 |
cens | +1.405 |
AFB | +1.381 |
Neg logit contributions | |
️ | -2.013 |
ilan | -1.965 |
ulhu | -1.908 |
andowski | -1.899 |
otle | -1.851 |
Token |
|
Token ID | 198 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
Shinra | +1.492 |
Lowe | +1.426 |
phrine | +1.360 |
cens | +1.329 |
AFB | +1.316 |
Neg logit contributions | |
️ | -1.879 |
ilan | -1.840 |
andowski | -1.796 |
ulhu | -1.777 |
otle | -1.726 |
Token |
|
Token ID | 198 |
Feature activation | -8.58e-02 |
Pos logit contributions | |
Shinra | +0.387 |
phrine | +0.368 |
Lowe | +0.364 |
cens | +0.359 |
AFB | +0.349 |
Neg logit contributions | |
️ | -0.414 |
ilan | -0.401 |
andowski | -0.388 |
ulhu | -0.383 |
otle | -0.375 |
Token | " |
Token ID | 1 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
️ | +0.638 |
ilan | +0.619 |
andowski | +0.600 |
ulhu | +0.593 |
otle | +0.581 |
Neg logit contributions | |
Shinra | -0.541 |
phrine | -0.512 |
Lowe | -0.506 |
cens | -0.500 |
AFB | -0.484 |
Token |
|
Token ID | 198 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Shinra | +0.079 |
phrine | +0.075 |
Lowe | +0.074 |
cens | +0.073 |
AFB | +0.071 |
Neg logit contributions | |
️ | -0.083 |
ilan | -0.081 |
andowski | -0.078 |
ulhu | -0.077 |
otle | -0.075 |
Token | California |
Token ID | 25284 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
Shinra | +0.878 |
phrine | +0.834 |
cens | +0.830 |
Lowe | +0.825 |
AFB | +0.785 |
Neg logit contributions | |
️ | -1.024 |
ilan | -0.979 |
andowski | -0.953 |
ulhu | -0.952 |
otle | -0.928 |
Token | v |
Token ID | 410 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
Shinra | +1.782 |
Lowe | +1.696 |
phrine | +1.690 |
cens | +1.660 |
AFB | +1.617 |
Neg logit contributions | |
️ | -1.969 |
ilan | -1.892 |
ulhu | -1.840 |
andowski | -1.829 |
otle | -1.771 |
Token | . |
Token ID | 13 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
️ | +1.585 |
ilan | +1.533 |
andowski | +1.477 |
ulhu | +1.442 |
otle | +1.441 |
Neg logit contributions | |
Shinra | -1.470 |
phrine | -1.426 |
cens | -1.375 |
Lowe | -1.364 |
AFB | -1.310 |
Token | Central |
Token ID | 5694 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
Shinra | +2.209 |
Lowe | +2.102 |
phrine | +1.985 |
AFB | +1.977 |
cens | +1.974 |
Neg logit contributions | |
️ | -2.209 |
ilan | -2.114 |
andowski | -2.072 |
ulhu | -2.054 |
otle | -1.971 |
Token | Pacific |
Token ID | 8211 |
Feature activation | +8.05e-03 |
Pos logit contributions | |
Shinra | +1.501 |
Lowe | +1.432 |
phrine | +1.427 |
cens | +1.401 |
AFB | +1.375 |
Neg logit contributions | |
️ | -1.204 |
ilan | -1.136 |
andowski | -1.095 |
ulhu | -1.090 |
otle | -1.055 |
Token | Railroad |
Token ID | 36289 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
Shinra | +0.058 |
Lowe | +0.054 |
phrine | +0.054 |
cens | +0.053 |
AFB | +0.052 |
Neg logit contributions | |
️ | -0.054 |
ilan | -0.052 |
andowski | -0.050 |
ulhu | -0.049 |
otle | -0.048 |
Token | Company |
Token ID | 5834 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
Shinra | +2.200 |
Lowe | +2.093 |
phrine | +2.073 |
cens | +2.037 |
AFB | +2.005 |
Neg logit contributions | |
️ | -1.570 |
ilan | -1.507 |
andowski | -1.453 |
ulhu | -1.436 |
otle | -1.374 |
Token |
|
Token ID | 198 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
Shinra | +2.862 |
Lowe | +2.677 |
cens | +2.638 |
phrine | +2.620 |
AFB | +2.552 |
Neg logit contributions | |
️ | -3.723 |
ilan | -3.585 |
andowski | -3.494 |
ulhu | -3.469 |
otle | -3.366 |
Token |
|
Token ID | 198 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
Shinra | +1.295 |
phrine | +1.224 |
Lowe | +1.222 |
cens | +1.209 |
AFB | +1.173 |
Neg logit contributions | |
️ | -1.535 |
ilan | -1.488 |
andowski | -1.449 |
ulhu | -1.443 |
otle | -1.392 |
Token | California |
Token ID | 25284 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
Shinra | +0.893 |
phrine | +0.848 |
cens | +0.845 |
Lowe | +0.842 |
AFB | +0.806 |
Neg logit contributions | |
️ | -0.890 |
ilan | -0.849 |
andowski | -0.825 |
ulhu | -0.825 |
otle | -0.803 |
Token | и |
Token ID | 18849 |
Feature activation | -6.25e-02 |
Pos logit contributions | |
Shinra | +0.903 |
phrine | +0.860 |
Lowe | +0.856 |
cens | +0.848 |
AFB | +0.822 |
Neg logit contributions | |
️ | -0.886 |
ilan | -0.867 |
andowski | -0.840 |
ulhu | -0.828 |
otle | -0.809 |
Token | т |
Token ID | 20375 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
️ | +0.431 |
ilan | +0.418 |
andowski | +0.404 |
ulhu | +0.398 |
otle | +0.390 |
Neg logit contributions | |
Shinra | -0.429 |
phrine | -0.408 |
Lowe | -0.403 |
cens | -0.398 |
AFB | -0.386 |
Token | ь |
Token ID | 45367 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
️ | +0.895 |
ilan | +0.860 |
andowski | +0.826 |
ulhu | +0.812 |
otle | +0.795 |
Neg logit contributions | |
Shinra | -1.112 |
phrine | -1.065 |
Lowe | -1.052 |
cens | -1.041 |
AFB | -1.014 |
Token | � |
Token ID | 12466 |
Feature activation | +7.60e-02 |
Pos logit contributions | |
️ | +0.458 |
ilan | +0.443 |
andowski | +0.427 |
ulhu | +0.420 |
otle | +0.410 |
Neg logit contributions | |
Shinra | -0.552 |
phrine | -0.526 |
Lowe | -0.521 |
cens | -0.516 |
AFB | -0.502 |
Token | � |
Token ID | 111 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
Shinra | +0.576 |
phrine | +0.547 |
Lowe | +0.545 |
cens | +0.537 |
AFB | +0.523 |
Neg logit contributions | |
️ | -0.467 |
ilan | -0.454 |
andowski | -0.435 |
ulhu | -0.429 |
otle | -0.418 |
Token | и |
Token ID | 18849 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
️ | +1.653 |
ilan | +1.590 |
andowski | +1.526 |
ulhu | +1.504 |
otle | +1.468 |
Neg logit contributions | |
Shinra | -1.946 |
phrine | -1.860 |
Lowe | -1.836 |
cens | -1.816 |
AFB | -1.771 |
Token | � |
Token ID | 140 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
Shinra | +0.509 |
phrine | +0.481 |
Lowe | +0.478 |
cens | +0.471 |
AFB | +0.457 |
Neg logit contributions | |
️ | -0.503 |
ilan | -0.492 |
andowski | -0.479 |
ulhu | -0.471 |
otle | -0.458 |
Token | � |
Token ID | 109 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
️ | +2.520 |
ilan | +2.423 |
andowski | +2.337 |
ulhu | +2.295 |
otle | +2.248 |
Neg logit contributions | |
Shinra | -2.946 |
phrine | -2.818 |
Lowe | -2.779 |
cens | -2.755 |
AFB | -2.679 |
Token | е |
Token ID | 16843 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
️ | +2.321 |
ilan | +2.234 |
andowski | +2.157 |
ulhu | +2.126 |
otle | +2.082 |
Neg logit contributions | |
Shinra | -2.377 |
phrine | -2.267 |
Lowe | -2.228 |
cens | -2.208 |
AFB | -2.139 |
Token | л |
Token ID | 30143 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
Shinra | +0.201 |
phrine | +0.190 |
Lowe | +0.188 |
cens | +0.185 |
AFB | +0.176 |
Neg logit contributions | |
️ | -0.301 |
ilan | -0.293 |
andowski | -0.286 |
ulhu | -0.281 |
otle | -0.277 |
Token | ь |
Token ID | 45367 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
Shinra | +1.870 |
phrine | +1.784 |
Lowe | +1.777 |
cens | +1.769 |
AFB | +1.714 |
Neg logit contributions | |
️ | -1.016 |
ilan | -0.985 |
andowski | -0.949 |
ulhu | -0.925 |
otle | -0.895 |
Token | me |
Token ID | 1326 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
️ | +1.748 |
ilan | +1.699 |
andowski | +1.645 |
ulhu | +1.618 |
otle | +1.591 |
Neg logit contributions | |
Shinra | -1.563 |
phrine | -1.483 |
Lowe | -1.461 |
cens | -1.444 |
AFB | -1.400 |
Token | er |
Token ID | 263 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Shinra | +1.689 |
phrine | +1.620 |
Lowe | +1.590 |
cens | +1.589 |
AFB | +1.529 |
Neg logit contributions | |
️ | -1.623 |
ilan | -1.515 |
andowski | -1.484 |
ulhu | -1.468 |
otle | -1.449 |
Token | hung |
Token ID | 9174 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
Shinra | +1.341 |
Lowe | +1.265 |
phrine | +1.255 |
cens | +1.240 |
AFB | +1.208 |
Neg logit contributions | |
️ | -1.450 |
ilan | -1.393 |
andowski | -1.355 |
ulhu | -1.342 |
otle | -1.315 |
Token | on |
Token ID | 319 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Shinra | +0.238 |
phrine | +0.222 |
Lowe | +0.221 |
cens | +0.218 |
AFB | +0.209 |
Neg logit contributions | |
️ | -0.317 |
ilan | -0.307 |
andowski | -0.300 |
ulhu | -0.295 |
otle | -0.289 |
Token | the |
Token ID | 262 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Shinra | +0.817 |
Lowe | +0.769 |
phrine | +0.752 |
cens | +0.742 |
AFB | +0.715 |
Neg logit contributions | |
️ | -1.024 |
ilan | -0.998 |
andowski | -0.969 |
ulhu | -0.951 |
otle | -0.932 |
Token | wall |
Token ID | 3355 |
Feature activation | -8.44e-02 |
Pos logit contributions | |
Shinra | +0.930 |
Lowe | +0.870 |
phrine | +0.858 |
cens | +0.840 |
AFB | +0.818 |
Neg logit contributions | |
️ | -1.208 |
ilan | -1.175 |
andowski | -1.140 |
ulhu | -1.127 |
otle | -1.102 |
Token | . |
Token ID | 13 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
️ | +0.641 |
ilan | +0.622 |
andowski | +0.604 |
ulhu | +0.596 |
otle | +0.584 |
Neg logit contributions | |
Shinra | -0.520 |
phrine | -0.491 |
Lowe | -0.485 |
cens | -0.479 |
AFB | -0.463 |
Token | � |
Token ID | 564 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
️ | +1.377 |
ilan | +1.334 |
andowski | +1.292 |
ulhu | +1.274 |
otle | +1.251 |
Neg logit contributions | |
Shinra | -1.179 |
phrine | -1.119 |
Lowe | -1.101 |
cens | -1.089 |
AFB | -1.054 |
Token | � |
Token ID | 250 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
️ | +0.756 |
ilan | +0.739 |
andowski | +0.716 |
ulhu | +0.706 |
otle | +0.694 |
Neg logit contributions | |
Shinra | -0.597 |
phrine | -0.562 |
Lowe | -0.557 |
cens | -0.548 |
AFB | -0.529 |
Token | The |
Token ID | 464 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
️ | +0.751 |
ilan | +0.729 |
andowski | +0.705 |
ulhu | +0.695 |
otle | +0.682 |
Neg logit contributions | |
Shinra | -0.688 |
phrine | -0.652 |
Lowe | -0.645 |
cens | -0.638 |
AFB | -0.617 |
Token | oppression |
Token ID | 18345 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
️ | +0.260 |
ilan | +0.253 |
andowski | +0.244 |
ulhu | +0.241 |
otle | +0.236 |
Neg logit contributions | |
Shinra | -0.242 |
phrine | -0.229 |
Lowe | -0.227 |
cens | -0.224 |
AFB | -0.217 |
Token | minimalist |
Token ID | 44693 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
Shinra | +0.380 |
phrine | +0.363 |
Lowe | +0.360 |
cens | +0.357 |
AFB | +0.344 |
Neg logit contributions | |
️ | -0.362 |
ilan | -0.348 |
andowski | -0.337 |
ulhu | -0.333 |
otle | -0.325 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
️ | +2.751 |
ilan | +2.643 |
andowski | +2.560 |
ulhu | +2.510 |
otle | +2.476 |
Neg logit contributions | |
Shinra | -2.420 |
phrine | -2.293 |
Lowe | -2.252 |
cens | -2.213 |
AFB | -2.170 |
Token | mood |
Token ID | 10038 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
️ | +0.083 |
ilan | +0.080 |
andowski | +0.078 |
ulhu | +0.077 |
otle | +0.075 |
Neg logit contributions | |
Shinra | -0.067 |
phrine | -0.063 |
Lowe | -0.063 |
cens | -0.062 |
AFB | -0.060 |
Token | y |
Token ID | 88 |
Feature activation | +6.53e-02 |
Pos logit contributions | |
️ | +0.389 |
ilan | +0.378 |
andowski | +0.368 |
ulhu | +0.364 |
otle | +0.357 |
Neg logit contributions | |
Shinra | -0.237 |
phrine | -0.223 |
Lowe | -0.220 |
cens | -0.217 |
AFB | -0.207 |
Token | , |
Token ID | 11 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Shinra | +0.408 |
Lowe | +0.387 |
phrine | +0.385 |
cens | +0.380 |
AFB | +0.364 |
Neg logit contributions | |
️ | -0.483 |
ilan | -0.472 |
andowski | -0.459 |
ulhu | -0.454 |
otle | -0.441 |
Token | anxious |
Token ID | 18116 |
Feature activation | -7.55e-02 |
Pos logit contributions | |
Shinra | +0.100 |
phrine | +0.095 |
Lowe | +0.095 |
cens | +0.093 |
AFB | +0.090 |
Neg logit contributions | |
️ | -0.117 |
ilan | -0.114 |
andowski | -0.111 |
ulhu | -0.110 |
otle | -0.106 |
Token | to |
Token ID | 284 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
️ | +0.552 |
ilan | +0.536 |
andowski | +0.520 |
ulhu | +0.514 |
otle | +0.502 |
Neg logit contributions | |
Shinra | -0.484 |
phrine | -0.459 |
Lowe | -0.456 |
cens | -0.449 |
AFB | -0.434 |
Token | her |
Token ID | 607 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
Shinra | +2.184 |
Lowe | +2.090 |
phrine | +2.072 |
cens | +2.040 |
clipping | +1.971 |
Neg logit contributions | |
️ | -2.479 |
ilan | -2.400 |
andowski | -2.349 |
ulhu | -2.302 |
otle | -2.197 |
Token | fingertips |
Token ID | 40142 |
Feature activation | -4.28e-03 |
Pos logit contributions | |
️ | +1.558 |
ilan | +1.510 |
andowski | +1.462 |
ulhu | +1.443 |
otle | +1.415 |
Neg logit contributions | |
Shinra | -1.383 |
phrine | -1.312 |
Lowe | -1.296 |
cens | -1.281 |
AFB | -1.240 |
Token | — |
Token ID | 851 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
️ | +0.032 |
ilan | +0.031 |
andowski | +0.030 |
ulhu | +0.030 |
otle | +0.029 |
Neg logit contributions | |
Shinra | -0.027 |
phrine | -0.025 |
Lowe | -0.025 |
cens | -0.025 |
AFB | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
️ | +0.592 |
ilan | +0.574 |
andowski | +0.556 |
ulhu | +0.549 |
otle | +0.539 |
Neg logit contributions | |
Shinra | -0.484 |
phrine | -0.459 |
Lowe | -0.452 |
cens | -0.447 |
AFB | -0.432 |
Token | overall |
Token ID | 4045 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
Shinra | +0.590 |
phrine | +0.559 |
Lowe | +0.556 |
cens | +0.546 |
AFB | +0.530 |
Neg logit contributions | |
️ | -0.788 |
ilan | -0.774 |
andowski | -0.748 |
ulhu | -0.746 |
otle | -0.719 |
Token | health |
Token ID | 1535 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
️ | +0.304 |
ilan | +0.296 |
andowski | +0.286 |
ulhu | +0.284 |
otle | +0.276 |
Neg logit contributions | |
Shinra | -0.267 |
phrine | -0.254 |
Lowe | -0.252 |
cens | -0.248 |
AFB | -0.241 |
Token | of |
Token ID | 286 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
️ | +1.825 |
ilan | +1.771 |
andowski | +1.715 |
ulhu | +1.693 |
otle | +1.661 |
Neg logit contributions | |
Shinra | -1.527 |
phrine | -1.447 |
Lowe | -1.427 |
cens | -1.410 |
AFB | -1.364 |
Token | the |
Token ID | 262 |
Feature activation | +7.87e-02 |
Pos logit contributions | |
️ | +0.121 |
ilan | +0.118 |
andowski | +0.115 |
ulhu | +0.113 |
otle | +0.110 |
Neg logit contributions | |
Shinra | -0.101 |
phrine | -0.096 |
Lowe | -0.096 |
cens | -0.094 |
AFB | -0.091 |
Token | American |
Token ID | 1605 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
Shinra | +0.487 |
phrine | +0.463 |
Lowe | +0.460 |
cens | +0.453 |
AFB | +0.442 |
Neg logit contributions | |
️ | -0.593 |
ilan | -0.578 |
andowski | -0.561 |
ulhu | -0.555 |
otle | -0.535 |
Token | population |
Token ID | 3265 |
Feature activation | +8.73e-02 |
Pos logit contributions | |
️ | +0.691 |
ilan | +0.673 |
andowski | +0.653 |
ulhu | +0.646 |
otle | +0.630 |
Neg logit contributions | |
Shinra | -0.480 |
phrine | -0.454 |
Lowe | -0.448 |
cens | -0.443 |
AFB | -0.428 |
Token | . |
Token ID | 13 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Shinra | +0.521 |
phrine | +0.494 |
Lowe | +0.490 |
cens | +0.482 |
AFB | +0.467 |
Neg logit contributions | |
️ | -0.670 |
ilan | -0.658 |
andowski | -0.635 |
ulhu | -0.632 |
otle | -0.615 |
Token | I |
Token ID | 314 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
Shinra | +0.720 |
Lowe | +0.683 |
phrine | +0.671 |
cens | +0.663 |
AFB | +0.647 |
Neg logit contributions | |
️ | -0.862 |
ilan | -0.848 |
andowski | -0.821 |
ulhu | -0.813 |
otle | -0.786 |
Token | want |
Token ID | 765 |
Feature activation | +4.99e-03 |
Pos logit contributions | |
Shinra | +0.461 |
phrine | +0.444 |
Lowe | +0.434 |
cens | +0.429 |
AFB | +0.414 |
Neg logit contributions | |
️ | -0.613 |
ilan | -0.609 |
andowski | -0.589 |
ulhu | -0.587 |
otle | -0.570 |
Token | legalization |
Token ID | 23050 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
Shinra | +0.029 |
phrine | +0.028 |
Lowe | +0.028 |
cens | +0.027 |
AFB | +0.026 |
Neg logit contributions | |
️ | -0.039 |
ilan | -0.038 |
andowski | -0.037 |
ulhu | -0.037 |
otle | -0.036 |
Token | to |
Token ID | 284 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
Shinra | +1.395 |
phrine | +1.340 |
Lowe | +1.320 |
cens | +1.305 |
AFB | +1.255 |
Neg logit contributions | |
️ | -1.950 |
ilan | -1.932 |
ulhu | -1.861 |
andowski | -1.850 |
otle | -1.798 |
Token | around |
Token ID | 1088 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
Shinra | +2.974 |
Lowe | +2.774 |
phrine | +2.735 |
cens | +2.718 |
clipping | +2.635 |
Neg logit contributions | |
️ | -4.497 |
ilan | -4.410 |
andowski | -4.289 |
ulhu | -4.272 |
otle | -4.022 |
Token | Angular |
Token ID | 28147 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
Shinra | +1.621 |
Lowe | +1.528 |
phrine | +1.521 |
cens | +1.489 |
clipping | +1.479 |
Neg logit contributions | |
️ | -1.901 |
ilan | -1.892 |
ulhu | -1.841 |
andowski | -1.826 |
otle | -1.766 |
Token | , |
Token ID | 11 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
️ | +0.416 |
ilan | +0.408 |
andowski | +0.396 |
ulhu | +0.393 |
otle | +0.383 |
Neg logit contributions | |
Shinra | -0.305 |
phrine | -0.288 |
Lowe | -0.285 |
cens | -0.279 |
AFB | -0.272 |
Token | these |
Token ID | 777 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
️ | +3.214 |
ilan | +3.083 |
andowski | +2.964 |
ulhu | +2.911 |
otle | +2.894 |
Neg logit contributions | |
Shinra | -2.957 |
phrine | -2.827 |
cens | -2.767 |
Lowe | -2.758 |
AFB | -2.661 |
Token | may |
Token ID | 743 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
️ | +3.457 |
ilan | +3.325 |
andowski | +3.222 |
ulhu | +3.132 |
otle | +3.105 |
Neg logit contributions | |
Shinra | -3.240 |
phrine | -3.120 |
Lowe | -3.038 |
cens | -3.028 |
AFB | -2.919 |
Token | be |
Token ID | 307 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
️ | +3.295 |
ilan | +3.193 |
andowski | +3.093 |
ulhu | +3.027 |
otle | +3.011 |
Neg logit contributions | |
Shinra | -2.855 |
phrine | -2.699 |
Lowe | -2.672 |
cens | -2.639 |
AFB | -2.556 |
Token | helpful |
Token ID | 7613 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
️ | +3.903 |
ilan | +3.744 |
andowski | +3.610 |
ulhu | +3.495 |
otle | +3.480 |
Neg logit contributions | |
Shinra | -3.689 |
phrine | -3.552 |
cens | -3.452 |
Lowe | -3.449 |
AFB | -3.303 |
Token | . |
Token ID | 13 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
️ | +0.820 |
ilan | +0.795 |
andowski | +0.771 |
ulhu | +0.761 |
otle | +0.747 |
Neg logit contributions | |
Shinra | -0.652 |
phrine | -0.617 |
Lowe | -0.608 |
cens | -0.600 |
AFB | -0.580 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
️ | +0.620 |
ilan | +0.600 |
andowski | +0.580 |
ulhu | +0.572 |
otle | +0.562 |
Neg logit contributions | |
Shinra | -0.555 |
phrine | -0.528 |
Lowe | -0.519 |
cens | -0.514 |
AFB | -0.498 |
Token |
|
Token ID | 198 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
️ | +0.143 |
ilan | +0.138 |
andowski | +0.134 |
ulhu | +0.132 |
otle | +0.129 |
Neg logit contributions | |
Shinra | -0.136 |
phrine | -0.129 |
Lowe | -0.127 |
cens | -0.126 |
AFB | -0.122 |
Token | In |
Token ID | 818 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
Shinra | +0.156 |
phrine | +0.148 |
cens | +0.146 |
Lowe | +0.146 |
AFB | +0.140 |
Neg logit contributions | |
️ | -0.179 |
ilan | -0.174 |
andowski | -0.169 |
ulhu | -0.167 |
otle | -0.164 |
Token | ist |
Token ID | 396 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
️ | +0.515 |
ilan | +0.494 |
andowski | +0.478 |
ulhu | +0.471 |
otle | +0.460 |
Neg logit contributions | |
Shinra | -0.474 |
phrine | -0.453 |
Lowe | -0.451 |
cens | -0.443 |
AFB | -0.429 |
Token | online |
Token ID | 2691 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
️ | +1.407 |
ilan | +1.359 |
andowski | +1.311 |
ulhu | +1.290 |
otle | +1.268 |
Neg logit contributions | |
Shinra | -1.412 |
phrine | -1.349 |
Lowe | -1.326 |
cens | -1.315 |
AFB | -1.275 |
Token | persona |
Token ID | 27822 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
️ | +0.631 |
ilan | +0.609 |
andowski | +0.586 |
ulhu | +0.577 |
otle | +0.563 |
Neg logit contributions | |
Shinra | -0.770 |
phrine | -0.736 |
Lowe | -0.729 |
cens | -0.721 |
AFB | -0.702 |
Token | is |
Token ID | 318 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
️ | +2.666 |
ilan | +2.576 |
andowski | +2.494 |
ulhu | +2.449 |
otle | +2.418 |
Neg logit contributions | |
Shinra | -2.415 |
phrine | -2.312 |
Lowe | -2.258 |
cens | -2.239 |
AFB | -2.168 |
Token | another |
Token ID | 1194 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
️ | +1.441 |
ilan | +1.398 |
andowski | +1.355 |
ulhu | +1.339 |
otle | +1.312 |
Neg logit contributions | |
Shinra | -1.275 |
phrine | -1.202 |
Lowe | -1.195 |
cens | -1.180 |
AFB | -1.142 |
Token | story |
Token ID | 1621 |
Feature activation | -6.42e-01 |
Pos logit contributions | |
️ | +1.424 |
ilan | +1.381 |
andowski | +1.337 |
ulhu | +1.317 |
otle | +1.290 |
Neg logit contributions | |
Shinra | -1.351 |
phrine | -1.280 |
Lowe | -1.268 |
cens | -1.252 |
AFB | -1.216 |
Token | !) |
Token ID | 8133 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
️ | +4.486 |
ilan | +4.316 |
andowski | +4.158 |
ulhu | +4.090 |
otle | +4.055 |
Neg logit contributions | |
Shinra | -4.267 |
phrine | -4.044 |
Lowe | -3.966 |
cens | -3.888 |
AFB | -3.767 |
Token | The |
Token ID | 383 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
️ | +0.927 |
ilan | +0.905 |
andowski | +0.875 |
ulhu | +0.865 |
otle | +0.844 |
Neg logit contributions | |
Shinra | -0.818 |
phrine | -0.769 |
Lowe | -0.766 |
cens | -0.753 |
AFB | -0.730 |
Token | rant |
Token ID | 30993 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
️ | +1.328 |
ilan | +1.287 |
andowski | +1.244 |
ulhu | +1.229 |
otle | +1.202 |
Neg logit contributions | |
Shinra | -1.270 |
phrine | -1.205 |
Lowe | -1.192 |
cens | -1.178 |
AFB | -1.141 |
Token | was |
Token ID | 373 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
️ | +1.079 |
ilan | +1.044 |
andowski | +1.010 |
ulhu | +0.996 |
otle | +0.976 |
Neg logit contributions | |
Shinra | -1.018 |
phrine | -0.967 |
Lowe | -0.955 |
cens | -0.944 |
AFB | -0.915 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
️ | +0.361 |
ilan | +0.350 |
andowski | +0.339 |
ulhu | +0.337 |
otle | +0.328 |
Neg logit contributions | |
Shinra | -0.314 |
phrine | -0.295 |
Lowe | -0.295 |
cens | -0.291 |
AFB | -0.280 |
Token | recently |
Token ID | 2904 |
Feature activation | +8.34e-02 |
Pos logit contributions | |
Shinra | +2.087 |
Lowe | +1.969 |
phrine | +1.940 |
cens | +1.899 |
AFB | +1.868 |
Neg logit contributions | |
️ | -2.307 |
ilan | -2.254 |
andowski | -2.191 |
ulhu | -2.166 |
otle | -2.105 |
Token | appeared |
Token ID | 4120 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
Shinra | +0.566 |
phrine | +0.535 |
Lowe | +0.532 |
cens | +0.524 |
AFB | +0.509 |
Neg logit contributions | |
️ | -0.582 |
ilan | -0.562 |
andowski | -0.543 |
ulhu | -0.539 |
otle | -0.527 |
Token | in |
Token ID | 287 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
️ | +0.473 |
ilan | +0.461 |
andowski | +0.448 |
ulhu | +0.444 |
otle | +0.437 |
Neg logit contributions | |
Shinra | -0.288 |
phrine | -0.271 |
Lowe | -0.266 |
cens | -0.262 |
AFB | -0.251 |
Token | the |
Token ID | 262 |
Feature activation | -7.13e-02 |
Pos logit contributions | |
️ | +1.279 |
ilan | +1.241 |
andowski | +1.201 |
ulhu | +1.187 |
otle | +1.164 |
Neg logit contributions | |
Shinra | -1.131 |
phrine | -1.070 |
Lowe | -1.061 |
cens | -1.041 |
AFB | -1.015 |
Token | Journal |
Token ID | 4913 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
️ | +0.460 |
ilan | +0.444 |
andowski | +0.428 |
ulhu | +0.422 |
otle | +0.412 |
Neg logit contributions | |
Shinra | -0.522 |
phrine | -0.497 |
Lowe | -0.493 |
cens | -0.486 |
AFB | -0.474 |
Token | of |
Token ID | 286 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
️ | +1.976 |
ilan | +1.916 |
andowski | +1.857 |
ulhu | +1.831 |
otle | +1.797 |
Neg logit contributions | |
Shinra | -1.647 |
phrine | -1.561 |
Lowe | -1.538 |
cens | -1.520 |
AFB | -1.471 |
Token | Great |
Token ID | 3878 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
️ | +3.457 |
ilan | +3.324 |
andowski | +3.215 |
ulhu | +3.122 |
otle | +3.070 |
Neg logit contributions | |
Shinra | -3.874 |
phrine | -3.746 |
cens | -3.655 |
Lowe | -3.640 |
AFB | -3.497 |
Token | Lakes |
Token ID | 24153 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
Shinra | +1.237 |
Lowe | +1.171 |
phrine | +1.155 |
cens | +1.140 |
AFB | +1.112 |
Neg logit contributions | |
️ | -1.494 |
ilan | -1.455 |
andowski | -1.399 |
ulhu | -1.398 |
otle | -1.374 |
Token | Research |
Token ID | 4992 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
Shinra | +3.215 |
Lowe | +3.138 |
cens | +3.062 |
phrine | +3.051 |
AFB | +2.923 |
Neg logit contributions | |
️ | -3.667 |
ilan | -3.500 |
ulhu | -3.448 |
andowski | -3.444 |
otle | -3.330 |
Token | , |
Token ID | 11 |
Feature activation | +8.81e-02 |
Pos logit contributions | |
Shinra | +1.502 |
phrine | +1.426 |
Lowe | +1.406 |
cens | +1.382 |
AFB | +1.342 |
Neg logit contributions | |
️ | -1.904 |
ilan | -1.860 |
ulhu | -1.804 |
andowski | -1.798 |
otle | -1.765 |
Token | revealed |
Token ID | 4602 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Shinra | +0.567 |
phrine | +0.538 |
Lowe | +0.534 |
cens | +0.522 |
AFB | +0.507 |
Neg logit contributions | |
️ | -0.643 |
ilan | -0.626 |
andowski | -0.607 |
ulhu | -0.601 |
otle | -0.584 |
Token | outhern |
Token ID | 4927 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Shinra | +0.858 |
phrine | +0.820 |
Lowe | +0.809 |
cens | +0.801 |
AFB | +0.778 |
Neg logit contributions | |
️ | -0.952 |
ilan | -0.919 |
andowski | -0.890 |
ulhu | -0.886 |
otle | -0.866 |
Token | Syria |
Token ID | 4392 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
Shinra | +0.971 |
phrine | +0.928 |
Lowe | +0.918 |
cens | +0.904 |
AFB | +0.880 |
Neg logit contributions | |
️ | -1.110 |
ilan | -1.083 |
ulhu | -1.057 |
andowski | -1.048 |
otle | -1.017 |
Token | , |
Token ID | 11 |
Feature activation | +9.91e-02 |
Pos logit contributions | |
Shinra | +0.366 |
phrine | +0.348 |
Lowe | +0.345 |
cens | +0.341 |
AFB | +0.331 |
Neg logit contributions | |
️ | -0.422 |
ilan | -0.412 |
andowski | -0.398 |
ulhu | -0.396 |
otle | -0.389 |
Token | right |
Token ID | 826 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
Shinra | +0.614 |
phrine | +0.581 |
Lowe | +0.578 |
cens | +0.568 |
AFB | +0.549 |
Neg logit contributions | |
️ | -0.742 |
ilan | -0.724 |
andowski | -0.701 |
ulhu | -0.696 |
otle | -0.681 |
Token | along |
Token ID | 1863 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
️ | +1.305 |
ilan | +1.264 |
andowski | +1.225 |
ulhu | +1.207 |
otle | +1.186 |
Neg logit contributions | |
Shinra | -1.087 |
phrine | -1.030 |
Lowe | -1.014 |
cens | -1.002 |
AFB | -0.970 |
Token | the |
Token ID | 262 |
Feature activation | -7.25e-01 |
Pos logit contributions | |
️ | +1.920 |
ilan | +1.873 |
andowski | +1.816 |
ulhu | +1.803 |
otle | +1.765 |
Neg logit contributions | |
Shinra | -1.580 |
phrine | -1.492 |
Lowe | -1.476 |
cens | -1.451 |
AFB | -1.413 |
Token | borders |
Token ID | 11637 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
️ | +4.668 |
ilan | +4.582 |
andowski | +4.317 |
otle | +4.200 |
ulhu | +4.161 |
Neg logit contributions | |
Shinra | -5.095 |
phrine | -4.925 |
cens | -4.789 |
Lowe | -4.788 |
AFB | -4.586 |
Token | of |
Token ID | 286 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
️ | +1.538 |
ilan | +1.501 |
andowski | +1.455 |
ulhu | +1.439 |
otle | +1.410 |
Neg logit contributions | |
Shinra | -1.211 |
phrine | -1.143 |
Lowe | -1.131 |
cens | -1.115 |
AFB | -1.077 |
Token | Jordan |
Token ID | 8078 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
️ | +0.957 |
ilan | +0.928 |
andowski | +0.898 |
ulhu | +0.887 |
otle | +0.864 |
Neg logit contributions | |
Shinra | -0.940 |
phrine | -0.891 |
Lowe | -0.888 |
cens | -0.869 |
AFB | -0.847 |
Token | and |
Token ID | 290 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
️ | +0.777 |
ilan | +0.755 |
andowski | +0.733 |
ulhu | +0.726 |
otle | +0.712 |
Neg logit contributions | |
Shinra | -0.671 |
phrine | -0.635 |
Lowe | -0.629 |
cens | -0.621 |
AFB | -0.603 |
Token | the |
Token ID | 262 |
Feature activation | -8.47e-02 |
Pos logit contributions | |
️ | +2.748 |
ilan | +2.663 |
andowski | +2.578 |
ulhu | +2.541 |
otle | +2.497 |
Neg logit contributions | |
Shinra | -2.261 |
phrine | -2.141 |
Lowe | -2.096 |
cens | -2.072 |
AFB | -2.005 |
Token | Communities |
Token ID | 35530 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
Shinra | +1.216 |
Lowe | +1.155 |
phrine | +1.137 |
cens | +1.123 |
AFB | +1.082 |
Neg logit contributions | |
️ | -1.387 |
ilan | -1.358 |
ulhu | -1.314 |
andowski | -1.312 |
otle | -1.267 |
Token | in |
Token ID | 287 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
️ | +0.957 |
ilan | +0.930 |
andowski | +0.901 |
ulhu | +0.891 |
otle | +0.874 |
Neg logit contributions | |
Shinra | -0.762 |
phrine | -0.721 |
Lowe | -0.711 |
cens | -0.702 |
AFB | -0.678 |
Token | Finland |
Token ID | 17837 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
️ | +2.079 |
ilan | +2.014 |
andowski | +1.949 |
ulhu | +1.922 |
otle | +1.886 |
Neg logit contributions | |
Shinra | -1.848 |
phrine | -1.755 |
Lowe | -1.730 |
cens | -1.711 |
AFB | -1.656 |
Token | told |
Token ID | 1297 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
️ | +0.825 |
ilan | +0.800 |
andowski | +0.772 |
ulhu | +0.762 |
otle | +0.746 |
Neg logit contributions | |
Shinra | -0.839 |
phrine | -0.800 |
Lowe | -0.789 |
cens | -0.782 |
AFB | -0.758 |
Token | J |
Token ID | 449 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
️ | +1.067 |
ilan | +1.033 |
andowski | +1.000 |
ulhu | +0.987 |
otle | +0.967 |
Neg logit contributions | |
Shinra | -0.984 |
phrine | -0.934 |
Lowe | -0.923 |
cens | -0.912 |
AFB | -0.883 |
Token | TA |
Token ID | 5603 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
️ | +1.222 |
ilan | +1.173 |
andowski | +1.123 |
ulhu | +1.102 |
otle | +1.075 |
Neg logit contributions | |
Shinra | -1.788 |
phrine | -1.716 |
Lowe | -1.697 |
cens | -1.682 |
AFB | -1.641 |
Token | that |
Token ID | 326 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
️ | +3.770 |
ilan | +3.666 |
andowski | +3.544 |
ulhu | +3.490 |
otle | +3.419 |
Neg logit contributions | |
Shinra | -3.600 |
phrine | -3.460 |
Lowe | -3.368 |
cens | -3.359 |
AFB | -3.273 |
Token | he |
Token ID | 339 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
Shinra | +0.360 |
phrine | +0.338 |
Lowe | +0.337 |
cens | +0.329 |
AFB | +0.319 |
Neg logit contributions | |
️ | -0.424 |
ilan | -0.410 |
andowski | -0.400 |
ulhu | -0.399 |
otle | -0.387 |
Token | did |
Token ID | 750 |
Feature activation | +8.30e-02 |
Pos logit contributions | |
Shinra | +0.280 |
phrine | +0.266 |
Lowe | +0.263 |
cens | +0.260 |
AFB | +0.252 |
Neg logit contributions | |
️ | -0.301 |
ilan | -0.292 |
andowski | -0.282 |
ulhu | -0.280 |
otle | -0.272 |
Token | not |
Token ID | 407 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
Shinra | +0.452 |
phrine | +0.424 |
Lowe | +0.421 |
cens | +0.411 |
AFB | +0.392 |
Neg logit contributions | |
️ | -0.687 |
ilan | -0.671 |
andowski | -0.650 |
ulhu | -0.647 |
otle | -0.630 |
Token | foresee |
Token ID | 25511 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
️ | +0.432 |
ilan | +0.419 |
andowski | +0.405 |
ulhu | +0.399 |
otle | +0.391 |
Neg logit contributions | |
Shinra | -0.412 |
phrine | -0.392 |
Lowe | -0.387 |
cens | -0.382 |
AFB | -0.371 |