Token | s |
Token ID | 82 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
"$:/ | +2.079 |
Malf | +1.873 |
LW | +1.819 |
Nanto | +1.799 |
Pyro | +1.727 |
Neg logit contributions | |
subsequ | -1.200 |
exhib | -1.104 |
momentum | -1.086 |
therap | -1.085 |
upgr | -1.081 |
Token | top |
Token ID | 1353 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
"$:/ | +2.619 |
Malf | +2.530 |
LW | +2.384 |
Pyro | +2.192 |
LG | +2.173 |
Neg logit contributions | |
subsequ | -2.016 |
therap | -1.978 |
mosqu | -1.918 |
enthusi | -1.860 |
obser | -1.804 |
Token | discipline |
Token ID | 12883 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
"$:/ | +2.595 |
Malf | +2.437 |
LW | +2.408 |
Nanto | +2.215 |
spection | +2.190 |
Neg logit contributions | |
subsequ | -1.963 |
therap | -1.814 |
ammy | -1.775 |
mosqu | -1.744 |
upgr | -1.742 |
Token | watchdog |
Token ID | 26856 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
subsequ | +0.522 |
therap | +0.462 |
ammy | +0.456 |
exhib | +0.446 |
mosqu | +0.442 |
Neg logit contributions | |
"$:/ | -0.681 |
LW | -0.621 |
Malf | -0.615 |
Nanto | -0.580 |
Pyro | -0.569 |
Token | , |
Token ID | 11 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
"$:/ | +5.010 |
LW | +4.847 |
Malf | +4.796 |
Nanto | +4.523 |
spection | +4.309 |
Neg logit contributions | |
subsequ | -5.534 |
therap | -5.135 |
exhib | -4.958 |
upgr | -4.954 |
enthusi | -4.942 |
Token | the |
Token ID | 262 |
Feature activation | +8.99e-01 |
Pos logit contributions | |
"$:/ | +6.398 |
LW | +6.284 |
Malf | +6.012 |
Nanto | +5.694 |
LG | +5.487 |
Neg logit contributions | |
subsequ | -7.440 |
therap | -7.148 |
exhib | -6.797 |
mosqu | -6.773 |
upgr | -6.760 |
Token | Central |
Token ID | 5694 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
LW | +7.682 |
"$:/ | +7.358 |
Malf | +7.106 |
LG | +6.654 |
spection | +6.409 |
Neg logit contributions | |
subsequ | -8.971 |
therap | -8.773 |
mosqu | -8.431 |
enthusi | -8.142 |
upgr | -8.101 |
Token | Commission |
Token ID | 4513 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
LW | +4.914 |
"$:/ | +4.804 |
Malf | +4.649 |
spection | +4.348 |
LG | +4.311 |
Neg logit contributions | |
subsequ | -7.037 |
therap | -6.897 |
mosqu | -6.814 |
enthusi | -6.561 |
exhib | -6.372 |
Token | for |
Token ID | 329 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
"$:/ | +3.692 |
Malf | +3.282 |
LW | +3.168 |
Nanto | +3.128 |
Pyro | +3.011 |
Neg logit contributions | |
subsequ | -2.621 |
therap | -2.416 |
exhib | -2.394 |
upgr | -2.378 |
momentum | -2.360 |
Token | Discipline |
Token ID | 48532 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
"$:/ | +2.097 |
Malf | +1.849 |
LW | +1.826 |
spection | +1.521 |
Pyro | +1.509 |
Neg logit contributions | |
subsequ | -6.294 |
therap | -6.270 |
mosqu | -6.016 |
exhib | -5.888 |
inertia | -5.834 |
Token | Inspection |
Token ID | 47115 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
"$:/ | +1.109 |
Malf | +0.877 |
LW | +0.829 |
spection | +0.741 |
Nanto | +0.694 |
Neg logit contributions | |
subsequ | -4.431 |
therap | -4.284 |
ammy | -4.216 |
exhib | -4.196 |
mosqu | -4.145 |
Token | that |
Token ID | 326 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
"$:/ | +2.146 |
Malf | +1.908 |
LW | +1.848 |
Nanto | +1.757 |
Pyro | +1.705 |
Neg logit contributions | |
subsequ | -2.511 |
therap | -2.390 |
upgr | -2.284 |
exhib | -2.278 |
ammy | -2.259 |
Token | inward |
Token ID | 29879 |
Feature activation | +4.08e-02 |
Pos logit contributions | |
"$:/ | +0.567 |
Malf | +0.547 |
LW | +0.527 |
Pyro | +0.472 |
Boyle | +0.469 |
Neg logit contributions | |
subsequ | -0.610 |
therap | -0.588 |
mosqu | -0.572 |
ammy | -0.570 |
obser | -0.554 |
Token | migration |
Token ID | 13472 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
"$:/ | +0.511 |
LW | +0.474 |
Malf | +0.473 |
spection | +0.467 |
Nanto | +0.445 |
Neg logit contributions | |
subsequ | -0.338 |
therap | -0.314 |
unden | -0.303 |
ammy | -0.298 |
mosqu | -0.287 |
Token | has |
Token ID | 468 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
subsequ | +0.304 |
therap | +0.285 |
mosqu | +0.274 |
ammy | +0.274 |
exhib | +0.274 |
Neg logit contributions | |
"$:/ | -0.269 |
Malf | -0.231 |
LW | -0.229 |
Pyro | -0.210 |
Nanto | -0.205 |
Token | taken |
Token ID | 2077 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
"$:/ | +1.583 |
Malf | +1.438 |
LW | +1.409 |
Pyro | +1.267 |
Boyle | +1.236 |
Neg logit contributions | |
subsequ | -1.423 |
mosqu | -1.302 |
therap | -1.263 |
ammy | -1.240 |
obser | -1.230 |
Token | the |
Token ID | 262 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
"$:/ | +1.461 |
Malf | +1.353 |
LW | +1.302 |
Pyro | +1.158 |
Nanto | +1.132 |
Neg logit contributions | |
subsequ | -1.669 |
therap | -1.583 |
mosqu | -1.521 |
obser | -1.511 |
exhib | -1.508 |
Token | jobs |
Token ID | 3946 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
Malf | +7.600 |
"$:/ | +7.398 |
LW | +7.267 |
Pyro | +6.740 |
Haz | +6.452 |
Neg logit contributions | |
subsequ | -7.643 |
mosqu | -7.382 |
therap | -7.299 |
oshenko | -7.282 |
ammy | -7.142 |
Token | of |
Token ID | 286 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
"$:/ | +1.678 |
Malf | +1.470 |
LW | +1.426 |
Nanto | +1.327 |
Pyro | +1.314 |
Neg logit contributions | |
subsequ | -1.945 |
therap | -1.858 |
mosqu | -1.796 |
exhib | -1.780 |
upgr | -1.745 |
Token | many |
Token ID | 867 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
subsequ | +0.376 |
therap | +0.356 |
mosqu | +0.349 |
ammy | +0.336 |
obser | +0.336 |
Neg logit contributions | |
"$:/ | -0.379 |
Malf | -0.375 |
LW | -0.361 |
Pyro | -0.326 |
Nanto | -0.318 |
Token | local |
Token ID | 1957 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
"$:/ | +0.107 |
Malf | +0.105 |
LW | +0.102 |
Pyro | +0.092 |
Nanto | +0.090 |
Neg logit contributions | |
subsequ | -0.116 |
therap | -0.114 |
mosqu | -0.110 |
ammy | -0.109 |
upgr | -0.106 |
Token | people |
Token ID | 661 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
subsequ | +0.134 |
therap | +0.131 |
mosqu | +0.127 |
ammy | +0.126 |
sugg | +0.122 |
Neg logit contributions | |
"$:/ | -0.155 |
Malf | -0.147 |
LW | -0.142 |
Nanto | -0.130 |
abouts | -0.130 |
Token | 1 |
Token ID | 16 |
Feature activation | +6.31e-02 |
Pos logit contributions | |
"$:/ | +2.285 |
Malf | +2.038 |
LW | +2.015 |
Nanto | +2.012 |
Pyro | +1.880 |
Neg logit contributions | |
subsequ | -1.549 |
exhib | -1.445 |
therap | -1.405 |
momentum | -1.388 |
upgr | -1.380 |
Token | f |
Token ID | 69 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
"$:/ | +0.492 |
Malf | +0.400 |
lf | +0.388 |
LW | +0.385 |
spection | +0.341 |
Neg logit contributions | |
therap | -0.934 |
metic | -0.917 |
unden | -0.915 |
subsequ | -0.912 |
mosqu | -0.903 |
Token | 4 |
Token ID | 19 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
"$:/ | +2.179 |
Malf | +1.934 |
LW | +1.913 |
Nanto | +1.890 |
Pyro | +1.787 |
Neg logit contributions | |
subsequ | -1.522 |
exhib | -1.416 |
therap | -1.409 |
upgr | -1.395 |
momentum | -1.364 |
Token | b |
Token ID | 65 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
"$:/ | +3.590 |
Malf | +3.137 |
LW | +3.011 |
Nanto | +2.802 |
Pyro | +2.785 |
Neg logit contributions | |
subsequ | -3.843 |
therap | -3.625 |
exhib | -3.531 |
upgr | -3.524 |
momentum | -3.499 |
Token | be |
Token ID | 1350 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
"$:/ | +1.173 |
lf | +0.693 |
Malf | +0.436 |
915 | +0.420 |
LW | +0.399 |
Neg logit contributions | |
mosqu | -6.753 |
therap | -6.715 |
enthusi | -6.671 |
millenn | -6.501 |
subsequ | -6.483 |
Token | 66 |
Token ID | 2791 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
"$:/ | +4.801 |
Malf | +4.196 |
LW | +4.118 |
Nanto | +4.114 |
Pyro | +3.869 |
Neg logit contributions | |
subsequ | -3.154 |
exhib | -2.937 |
therap | -2.927 |
upgr | -2.887 |
momentum | -2.854 |
Token | IEEE |
Token ID | 40552 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
"$:/ | +7.708 |
Malf | +7.205 |
LW | +7.031 |
Nanto | +6.225 |
Pyro | +6.153 |
Neg logit contributions | |
subsequ | -8.556 |
therap | -8.014 |
ammy | -7.748 |
exhib | -7.748 |
upgr | -7.635 |
Token | 802 |
Token ID | 33121 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
"$:/ | +2.003 |
LW | +1.865 |
Malf | +1.375 |
RJ | +1.138 |
Pyro | +1.041 |
Neg logit contributions | |
therap | -7.883 |
enthusi | -7.783 |
unden | -7.763 |
subsequ | -7.733 |
undermin | -7.672 |
Token | . |
Token ID | 13 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
"$:/ | +1.981 |
LW | +1.848 |
Malf | +1.653 |
Pyro | +1.545 |
spection | +1.539 |
Neg logit contributions | |
enthusi | -3.326 |
therap | -3.322 |
metic | -3.287 |
mosqu | -3.236 |
millenn | -3.171 |
Token | 11 |
Token ID | 1157 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
therap | +0.684 |
subsequ | +0.677 |
metic | +0.666 |
enthusi | +0.664 |
mosqu | +0.656 |
Neg logit contributions | |
"$:/ | -0.322 |
LW | -0.268 |
915 | -0.247 |
Malf | -0.239 |
Pyro | -0.222 |
Token | b |
Token ID | 65 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
"$:/ | +4.487 |
Malf | +3.885 |
Nanto | +3.870 |
LW | +3.761 |
Pyro | +3.597 |
Neg logit contributions | |
subsequ | -3.241 |
upgr | -3.045 |
therap | -3.002 |
exhib | -2.995 |
momentum | -2.970 |
Token | 1 |
Token ID | 16 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
"$:/ | +2.451 |
LW | +2.152 |
Malf | +2.096 |
lf | +1.954 |
Py | +1.912 |
Neg logit contributions | |
subsequ | -3.802 |
therap | -3.674 |
mosqu | -3.574 |
exhib | -3.531 |
upgr | -3.524 |
Token | G |
Token ID | 38 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
"$:/ | +1.709 |
LW | +1.550 |
Malf | +1.526 |
lf | +1.520 |
Py | +1.398 |
Neg logit contributions | |
subsequ | -2.313 |
therap | -2.262 |
exhib | -2.145 |
mosqu | -2.137 |
momentum | -2.128 |
Token | q |
Token ID | 80 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
"$:/ | +1.933 |
lf | +1.720 |
LW | +1.711 |
Malf | +1.691 |
Py | +1.544 |
Neg logit contributions | |
subsequ | -2.655 |
therap | -2.589 |
upgr | -2.457 |
mosqu | -2.457 |
metic | -2.448 |
Token | FP |
Token ID | 5837 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
subsequ | +2.495 |
therap | +2.399 |
exhib | +2.292 |
upgr | +2.282 |
momentum | +2.278 |
Neg logit contributions | |
"$:/ | -1.912 |
Malf | -1.681 |
LW | -1.680 |
lf | -1.570 |
Pyro | -1.500 |
Token | P |
Token ID | 47 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
"$:/ | +0.301 |
LW | +0.267 |
Malf | +0.264 |
lf | +0.257 |
Pyro | +0.235 |
Neg logit contributions | |
subsequ | -0.407 |
therap | -0.397 |
mosqu | -0.374 |
exhib | -0.373 |
upgr | -0.371 |
Token | 9 |
Token ID | 24 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
"$:/ | +3.543 |
lf | +3.227 |
Malf | +3.182 |
LW | +3.175 |
Py | +2.903 |
Neg logit contributions | |
subsequ | -5.275 |
therap | -5.148 |
metic | -4.945 |
mosqu | -4.885 |
obser | -4.815 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
LW | +4.813 |
Malf | +4.320 |
"$:/ | +4.305 |
lf | +4.230 |
Pyro | +3.822 |
Neg logit contributions | |
subsequ | -11.517 |
therap | -11.489 |
metic | -11.304 |
mosqu | -11.106 |
obser | -10.963 |
Token | Donald |
Token ID | 7371 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
"$:/ | +1.385 |
Malf | +1.199 |
LW | +1.158 |
spection | +1.093 |
lf | +1.075 |
Neg logit contributions | |
subsequ | -1.654 |
therap | -1.597 |
mosqu | -1.528 |
exhib | -1.518 |
upgr | -1.507 |
Token | Trump |
Token ID | 1301 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
subsequ | +1.167 |
exhib | +1.030 |
therap | +1.030 |
momentum | +1.023 |
upgr | +0.997 |
Neg logit contributions | |
"$:/ | -2.515 |
Malf | -2.258 |
LW | -2.203 |
Nanto | -2.198 |
spection | -2.114 |
Token | will |
Token ID | 481 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
"$:/ | +2.808 |
Malf | +2.544 |
LW | +2.470 |
Nanto | +2.383 |
Pyro | +2.291 |
Neg logit contributions | |
subsequ | -2.337 |
therap | -2.158 |
upgr | -2.141 |
ammy | -2.118 |
exhib | -2.115 |
Token | not |
Token ID | 407 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
"$:/ | +2.506 |
Malf | +2.160 |
LW | +2.124 |
Nanto | +2.025 |
Pyro | +1.988 |
Neg logit contributions | |
subsequ | -2.201 |
therap | -2.146 |
mosqu | -2.145 |
enthusi | -2.094 |
pestic | -1.979 |
Token | green |
Token ID | 4077 |
Feature activation | -3.58e-03 |
Pos logit contributions | |
"$:/ | +1.658 |
Malf | +1.541 |
LW | +1.496 |
Pyro | +1.396 |
Nanto | +1.343 |
Neg logit contributions | |
subsequ | -1.599 |
artments | -1.523 |
ammy | -1.500 |
therap | -1.490 |
exhib | -1.465 |
Token | still |
Token ID | 991 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
subsequ | +0.038 |
therap | +0.037 |
exhib | +0.035 |
ammy | +0.034 |
artments | +0.034 |
Neg logit contributions | |
"$:/ | -0.037 |
Malf | -0.034 |
LW | -0.034 |
Nanto | -0.032 |
Pyro | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +8.38e-03 |
Pos logit contributions | |
subsequ | +1.208 |
therap | +1.175 |
exhib | +1.141 |
upgr | +1.107 |
ammy | +1.091 |
Neg logit contributions | |
"$:/ | -1.210 |
LW | -1.087 |
Malf | -1.073 |
spection | -1.015 |
Nanto | -1.003 |
Token | but |
Token ID | 475 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
"$:/ | +0.079 |
Malf | +0.071 |
LW | +0.070 |
Pyro | +0.064 |
Nanto | +0.062 |
Neg logit contributions | |
subsequ | -0.097 |
therap | -0.095 |
ammy | -0.093 |
exhib | -0.091 |
mosqu | -0.091 |
Token | by |
Token ID | 416 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
subsequ | +0.577 |
ammy | +0.573 |
therap | +0.561 |
artments | +0.537 |
exhib | +0.535 |
Neg logit contributions | |
"$:/ | -0.524 |
Malf | -0.480 |
LW | -0.474 |
Pyro | -0.432 |
Nanto | -0.405 |
Token | any |
Token ID | 597 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
"$:/ | +1.733 |
LW | +1.640 |
Malf | +1.609 |
Pyro | +1.489 |
Nanto | +1.481 |
Neg logit contributions | |
subsequ | -1.905 |
therap | -1.826 |
ammy | -1.824 |
upgr | -1.810 |
artments | -1.785 |
Token | standard |
Token ID | 3210 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
"$:/ | +8.270 |
LW | +7.316 |
Malf | +7.206 |
Pyro | +6.977 |
spection | +6.752 |
Neg logit contributions | |
ammy | -6.963 |
subsequ | -6.615 |
artments | -6.589 |
therap | -6.442 |
upgr | -6.391 |
Token | it |
Token ID | 340 |
Feature activation | +8.81e-02 |
Pos logit contributions | |
subsequ | +0.414 |
ammy | +0.401 |
therap | +0.388 |
exhib | +0.377 |
artments | +0.371 |
Neg logit contributions | |
"$:/ | -0.367 |
LW | -0.349 |
Malf | -0.346 |
Pyro | -0.312 |
Nanto | -0.297 |
Token | 's |
Token ID | 338 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
"$:/ | +0.970 |
Malf | +0.859 |
LW | +0.856 |
Pyro | +0.784 |
lf | +0.779 |
Neg logit contributions | |
subsequ | -0.907 |
therap | -0.893 |
mosqu | -0.846 |
ammy | -0.835 |
pestic | -0.829 |
Token | an |
Token ID | 281 |
Feature activation | +6.96e-02 |
Pos logit contributions | |
"$:/ | +1.458 |
Malf | +1.434 |
LW | +1.378 |
Pyro | +1.268 |
LG | +1.189 |
Neg logit contributions | |
subsequ | -1.574 |
artments | -1.518 |
ammy | -1.513 |
therap | -1.468 |
mosqu | -1.439 |
Token | excellent |
Token ID | 6275 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
"$:/ | +0.770 |
LW | +0.748 |
Malf | +0.738 |
Pyro | +0.667 |
Nanto | +0.641 |
Neg logit contributions | |
subsequ | -0.680 |
artments | -0.616 |
therap | -0.616 |
ammy | -0.603 |
mosqu | -0.586 |
Token | reement |
Token ID | 10237 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
subsequ | +5.289 |
therap | +5.043 |
upgr | +4.978 |
exhib | +4.927 |
momentum | +4.835 |
Neg logit contributions | |
"$:/ | -3.973 |
Malf | -3.461 |
LW | -3.360 |
Nanto | -3.181 |
Pyro | -3.072 |
Token | number |
Token ID | 1271 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
"$:/ | +1.822 |
LW | +1.693 |
Malf | +1.602 |
Nanto | +1.523 |
Pyro | +1.475 |
Neg logit contributions | |
subsequ | -2.107 |
therap | -1.984 |
upgr | -1.972 |
ammy | -1.951 |
exhib | -1.893 |
Token | : |
Token ID | 25 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
"$:/ | +4.474 |
LW | +4.370 |
Malf | +4.108 |
LG | +3.745 |
Pyro | +3.692 |
Neg logit contributions | |
subsequ | -5.493 |
therap | -5.223 |
ammy | -5.137 |
obser | -5.026 |
mosqu | -4.984 |
Token | 501 |
Token ID | 24555 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
"$:/ | +2.164 |
LW | +2.087 |
Malf | +1.974 |
LG | +1.784 |
Nanto | +1.763 |
Neg logit contributions | |
subsequ | -2.639 |
therap | -2.463 |
upgr | -2.414 |
ammy | -2.411 |
momentum | -2.349 |
Token | 20 |
Token ID | 1238 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
"$:/ | +4.259 |
LW | +3.830 |
Malf | +3.648 |
Nanto | +3.300 |
Pyro | +3.295 |
Neg logit contributions | |
subsequ | -4.999 |
therap | -4.940 |
mosqu | -4.657 |
upgr | -4.627 |
ammy | -4.619 |
Token | 78 |
Token ID | 3695 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
"$:/ | +6.356 |
LW | +5.700 |
Malf | +5.475 |
Nanto | +5.045 |
Pyro | +4.972 |
Neg logit contributions | |
subsequ | -7.809 |
therap | -7.579 |
ammy | -7.346 |
mosqu | -7.309 |
exhib | -7.229 |
Token | - |
Token ID | 12 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
"$:/ | +6.504 |
LW | +5.985 |
Malf | +5.719 |
Nanto | +5.072 |
Pyro | +5.063 |
Neg logit contributions | |
subsequ | -8.852 |
therap | -8.591 |
mosqu | -8.337 |
ammy | -8.332 |
traged | -8.236 |
Token | A |
Token ID | 32 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
"$:/ | +3.857 |
LW | +3.439 |
Malf | +3.375 |
lf | +2.998 |
Pyro | +2.968 |
Neg logit contributions | |
subsequ | -5.520 |
therap | -5.280 |
mosqu | -5.092 |
upgr | -5.091 |
erosion | -5.026 |
Token | ). |
Token ID | 737 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
"$:/ | +3.972 |
LW | +3.556 |
Malf | +3.502 |
Nanto | +3.268 |
spection | +3.215 |
Neg logit contributions | |
subsequ | -5.098 |
therap | -4.886 |
upgr | -4.814 |
ammy | -4.728 |
exhib | -4.660 |
Token | The |
Token ID | 383 |
Feature activation | -4.77e-03 |
Pos logit contributions | |
subsequ | +0.604 |
therap | +0.564 |
ammy | +0.548 |
upgr | +0.546 |
undermin | +0.538 |
Neg logit contributions | |
"$:/ | -0.471 |
Malf | -0.431 |
LW | -0.428 |
Pyro | -0.402 |
Nanto | -0.386 |
Token | monkeys |
Token ID | 26697 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
subsequ | +0.045 |
ammy | +0.042 |
therap | +0.041 |
upgr | +0.041 |
erosion | +0.040 |
Neg logit contributions | |
"$:/ | -0.055 |
LW | -0.051 |
Malf | -0.050 |
Pyro | -0.047 |
Nanto | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +9.92e-02 |
Pos logit contributions | |
"$:/ | +1.829 |
LW | +1.775 |
Malf | +1.681 |
Pyro | +1.542 |
spection | +1.531 |
Neg logit contributions | |
subsequ | -2.405 |
mosqu | -2.288 |
therap | -2.282 |
pestic | -2.268 |
obser | -2.265 |
Token | Here |
Token ID | 3423 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
"$:/ | +1.018 |
LW | +0.893 |
Malf | +0.877 |
Nanto | +0.861 |
Pyro | +0.822 |
Neg logit contributions | |
subsequ | -1.126 |
therap | -1.069 |
exhib | -1.051 |
upgr | -1.032 |
mosqu | -1.001 |
Token | is |
Token ID | 318 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
"$:/ | +2.777 |
LW | +2.571 |
Malf | +2.438 |
Pyro | +2.303 |
abouts | +2.260 |
Neg logit contributions | |
subsequ | -3.077 |
upgr | -2.987 |
therap | -2.876 |
exhib | -2.854 |
momentum | -2.781 |
Token | the |
Token ID | 262 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
"$:/ | +3.052 |
LW | +2.734 |
Malf | +2.730 |
Nanto | +2.634 |
Pyro | +2.500 |
Neg logit contributions | |
subsequ | -2.712 |
therap | -2.611 |
exhib | -2.530 |
upgr | -2.516 |
sugg | -2.413 |
Token | link |
Token ID | 2792 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
"$:/ | +4.468 |
Malf | +3.957 |
LW | +3.944 |
Pyro | +3.644 |
Nanto | +3.597 |
Neg logit contributions | |
subsequ | -5.418 |
therap | -5.158 |
upgr | -5.003 |
mosqu | -4.999 |
exhib | -4.987 |
Token | for |
Token ID | 329 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
"$:/ | +3.636 |
LW | +3.546 |
Malf | +3.413 |
Pyro | +3.274 |
spection | +3.208 |
Neg logit contributions | |
subsequ | -7.291 |
therap | -7.255 |
upgr | -7.131 |
enthusi | -6.817 |
exhib | -6.775 |
Token | buying |
Token ID | 7067 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
"$:/ | +8.874 |
LW | +7.862 |
Malf | +7.798 |
Nanto | +7.581 |
Pyro | +7.370 |
Neg logit contributions | |
subsequ | -7.689 |
exhib | -7.068 |
therap | -7.061 |
upgr | -6.996 |
mosqu | -6.931 |
Token | the |
Token ID | 262 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
"$:/ | +3.217 |
LW | +2.849 |
Malf | +2.847 |
Nanto | +2.706 |
Pyro | +2.630 |
Neg logit contributions | |
subsequ | -2.644 |
therap | -2.448 |
exhib | -2.425 |
upgr | -2.367 |
momentum | -2.299 |
Token | second |
Token ID | 1218 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
"$:/ | +3.017 |
LW | +2.631 |
Malf | +2.598 |
Nanto | +2.513 |
Pyro | +2.446 |
Neg logit contributions | |
subsequ | -3.029 |
therap | -2.894 |
exhib | -2.815 |
mosqu | -2.811 |
upgr | -2.794 |
Token | edition |
Token ID | 8313 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
"$:/ | +7.126 |
Pyro | +5.950 |
Malf | +5.949 |
LW | +5.913 |
Nanto | +5.702 |
Neg logit contributions | |
subsequ | -7.389 |
mosqu | -7.214 |
therap | -6.852 |
unden | -6.804 |
enthusi | -6.658 |
Token | of |
Token ID | 286 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
"$:/ | +2.975 |
Malf | +2.654 |
LW | +2.561 |
Nanto | +2.528 |
Pyro | +2.425 |
Neg logit contributions | |
subsequ | -1.932 |
therap | -1.747 |
upgr | -1.746 |
momentum | -1.743 |
exhib | -1.734 |
Token | REF |
Token ID | 31688 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
"$:/ | +1.987 |
LW | +1.757 |
Malf | +1.753 |
spection | +1.626 |
lf | +1.621 |
Neg logit contributions | |
subsequ | -2.897 |
therap | -2.779 |
mosqu | -2.701 |
exhib | -2.689 |
erosion | -2.687 |
Token | LECT |
Token ID | 16779 |
Feature activation | +7.22e-02 |
Pos logit contributions | |
"$:/ | +2.360 |
LW | +2.024 |
Malf | +1.949 |
spection | +1.797 |
Pyro | +1.776 |
Neg logit contributions | |
subsequ | -2.739 |
therap | -2.571 |
ammy | -2.550 |
upgr | -2.543 |
exhib | -2.534 |
Token | _ |
Token ID | 62 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
"$:/ | +0.759 |
LW | +0.657 |
Malf | +0.650 |
Pyro | +0.594 |
spection | +0.589 |
Neg logit contributions | |
subsequ | -0.778 |
therap | -0.734 |
exhib | -0.732 |
upgr | -0.727 |
ammy | -0.724 |
Token | CLASS |
Token ID | 31631 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
"$:/ | +1.479 |
LW | +1.304 |
Malf | +1.274 |
spection | +1.149 |
lf | +1.123 |
Neg logit contributions | |
subsequ | -1.846 |
therap | -1.745 |
exhib | -1.699 |
upgr | -1.677 |
erosion | -1.677 |
Token | _ |
Token ID | 62 |
Feature activation | +9.18e-03 |
Pos logit contributions | |
"$:/ | +0.953 |
Malf | +0.852 |
LW | +0.846 |
Pyro | +0.761 |
spection | +0.756 |
Neg logit contributions | |
subsequ | -1.191 |
ammy | -1.116 |
therap | -1.112 |
exhib | -1.103 |
upgr | -1.095 |
Token | ST |
Token ID | 2257 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
"$:/ | +0.086 |
Malf | +0.076 |
LW | +0.075 |
spection | +0.067 |
Pyro | +0.066 |
Neg logit contributions | |
subsequ | -0.112 |
therap | -0.106 |
exhib | -0.103 |
mosqu | -0.102 |
momentum | -0.102 |
Token | ART |
Token ID | 7227 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
"$:/ | +6.057 |
LW | +5.476 |
Malf | +5.290 |
lf | +4.818 |
KT | +4.647 |
Neg logit contributions | |
subsequ | -9.306 |
therap | -9.255 |
obser | -8.954 |
mosqu | -8.925 |
exhib | -8.771 |
Token | ( |
Token ID | 7 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
subsequ | +0.171 |
therap | +0.163 |
exhib | +0.161 |
ammy | +0.160 |
upgr | +0.158 |
Neg logit contributions | |
"$:/ | -0.154 |
Malf | -0.134 |
LW | -0.131 |
Pyro | -0.120 |
spection | -0.119 |
Token | mem |
Token ID | 11883 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
"$:/ | +3.371 |
LW | +2.958 |
Malf | +2.891 |
spection | +2.669 |
Nanto | +2.650 |
Neg logit contributions | |
subsequ | -3.856 |
therap | -3.620 |
exhib | -3.525 |
ammy | -3.513 |
erosion | -3.489 |
Token | c |
Token ID | 66 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
"$:/ | +1.780 |
Malf | +1.532 |
LW | +1.520 |
Nanto | +1.458 |
spection | +1.406 |
Neg logit contributions | |
subsequ | -1.890 |
therap | -1.768 |
exhib | -1.755 |
upgr | -1.733 |
momentum | -1.718 |
Token | py |
Token ID | 9078 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
"$:/ | +0.564 |
Py | +0.313 |
spection | +0.313 |
lf | +0.306 |
LW | +0.290 |
Neg logit contributions | |
mosqu | -2.595 |
enthusi | -2.585 |
subsequ | -2.541 |
unden | -2.528 |
therap | -2.511 |
Token | Height |
Token ID | 23106 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
"$:/ | +0.035 |
Malf | +0.030 |
LW | +0.029 |
Nanto | +0.027 |
Pyro | +0.026 |
Neg logit contributions | |
subsequ | -0.047 |
therap | -0.045 |
exhib | -0.044 |
upgr | -0.043 |
momentum | -0.043 |
Token | ) |
Token ID | 8 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
"$:/ | +0.871 |
Malf | +0.758 |
LW | +0.737 |
Nanto | +0.718 |
Pyro | +0.686 |
Neg logit contributions | |
subsequ | -0.846 |
therap | -0.790 |
exhib | -0.782 |
upgr | -0.777 |
momentum | -0.769 |
Token | $ |
Token ID | 720 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
subsequ | +3.269 |
therap | +3.097 |
exhib | +3.048 |
upgr | +3.003 |
momentum | +2.976 |
Neg logit contributions | |
"$:/ | -2.970 |
Malf | -2.557 |
LW | -2.525 |
Nanto | -2.403 |
Pyro | -2.323 |
Token | image |
Token ID | 9060 |
Feature activation | -8.10e-02 |
Pos logit contributions | |
subsequ | +2.938 |
therap | +2.780 |
exhib | +2.731 |
upgr | +2.710 |
momentum | +2.686 |
Neg logit contributions | |
"$:/ | -2.585 |
Malf | -2.215 |
LW | -2.204 |
Nanto | -2.063 |
spection | -2.003 |
Token | = |
Token ID | 796 |
Feature activation | +8.28e-02 |
Pos logit contributions | |
subsequ | +0.724 |
therap | +0.660 |
exhib | +0.657 |
upgr | +0.655 |
momentum | +0.649 |
Neg logit contributions | |
"$:/ | -1.005 |
Malf | -0.891 |
LW | -0.863 |
Nanto | -0.853 |
Pyro | -0.813 |
Token | [ |
Token ID | 685 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
"$:/ | +0.986 |
Malf | +0.873 |
LW | +0.845 |
Nanto | +0.831 |
Pyro | +0.794 |
Neg logit contributions | |
subsequ | -0.781 |
therap | -0.719 |
exhib | -0.716 |
upgr | -0.714 |
momentum | -0.705 |
Token | System |
Token ID | 11964 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
"$:/ | +7.507 |
Malf | +6.325 |
LW | +6.193 |
Nanto | +5.934 |
spection | +5.693 |
Neg logit contributions | |
subsequ | -8.866 |
therap | -8.324 |
exhib | -8.229 |
upgr | -8.180 |
momentum | -8.116 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
"$:/ | +0.449 |
LW | +0.389 |
Malf | +0.362 |
lf | +0.346 |
Pyro | +0.341 |
Neg logit contributions | |
therap | -0.657 |
mosqu | -0.654 |
enthusi | -0.641 |
metic | -0.630 |
unden | -0.623 |
Token | Draw |
Token ID | 25302 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
"$:/ | +1.319 |
Malf | +1.151 |
Nanto | +1.053 |
LW | +1.005 |
Pengu | +0.996 |
Neg logit contributions | |
subsequ | -1.987 |
exhib | -1.911 |
upgr | -1.903 |
momentum | -1.874 |
erosion | -1.864 |
Token | ing |
Token ID | 278 |
Feature activation | -8.88e-02 |
Pos logit contributions | |
therap | +5.086 |
mosqu | +4.964 |
enthusi | +4.911 |
unden | +4.905 |
subsequ | +4.858 |
Neg logit contributions | |
"$:/ | -3.367 |
LW | -2.852 |
Malf | -2.678 |
spection | -2.544 |
lf | -2.508 |
Token | . |
Token ID | 13 |
Feature activation | -5.64e-03 |
Pos logit contributions | |
therap | +1.207 |
mosqu | +1.179 |
enthusi | +1.159 |
subsequ | +1.151 |
unden | +1.140 |
Neg logit contributions | |
"$:/ | -0.941 |
LW | -0.816 |
Malf | -0.775 |
Pyro | -0.737 |
spection | -0.723 |
Token | little |
Token ID | 1310 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
subsequ | +2.789 |
mosqu | +2.778 |
therap | +2.740 |
unden | +2.666 |
enthusi | +2.621 |
Neg logit contributions | |
"$:/ | -2.676 |
LW | -2.423 |
Malf | -2.353 |
Pyro | -2.152 |
spection | -2.103 |
Token | gag |
Token ID | 32261 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
"$:/ | +1.110 |
LW | +0.952 |
Pyro | +0.870 |
Malf | +0.849 |
LG | +0.846 |
Neg logit contributions | |
mosqu | -1.931 |
subsequ | -1.833 |
therap | -1.800 |
unden | -1.797 |
metic | -1.723 |
Token | with |
Token ID | 351 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
"$:/ | +1.636 |
LW | +1.455 |
Pyro | +1.289 |
Malf | +1.273 |
LG | +1.202 |
Neg logit contributions | |
enthusi | -3.027 |
therap | -3.017 |
mosqu | -3.001 |
unden | -2.998 |
metic | -2.980 |
Token | Erin |
Token ID | 28894 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
subsequ | +1.335 |
therap | +1.256 |
exhib | +1.245 |
upgr | +1.244 |
momentum | +1.226 |
Neg logit contributions | |
"$:/ | -1.564 |
Malf | -1.350 |
LW | -1.310 |
Nanto | -1.302 |
Pyro | -1.226 |
Token | and |
Token ID | 290 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
"$:/ | +2.182 |
LW | +1.890 |
Malf | +1.788 |
Pyro | +1.685 |
spection | +1.490 |
Neg logit contributions | |
mosqu | -4.293 |
metic | -4.281 |
therap | -4.266 |
enthusi | -4.254 |
unden | -4.194 |
Token | Jill |
Token ID | 22772 |
Feature activation | +7.69e-01 |
Pos logit contributions | |
subsequ | +2.949 |
therap | +2.884 |
upgr | +2.768 |
exhib | +2.761 |
mosqu | +2.744 |
Neg logit contributions | |
"$:/ | -2.237 |
Malf | -1.993 |
LW | -1.962 |
Pyro | -1.773 |
Nanto | -1.748 |
Token | ian |
Token ID | 666 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
"$:/ | +3.898 |
LW | +3.300 |
lf | +2.662 |
Malf | +2.571 |
Pyro | +2.372 |
Neg logit contributions | |
therap | -13.002 |
mosqu | -12.745 |
enthusi | -12.568 |
unden | -12.301 |
metic | -12.301 |
Token | , |
Token ID | 11 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
"$:/ | +1.555 |
LW | +1.324 |
Malf | +1.223 |
Pyro | +1.203 |
Nanto | +1.097 |
Neg logit contributions | |
mosqu | -2.617 |
therap | -2.603 |
enthusi | -2.587 |
subsequ | -2.558 |
metic | -2.548 |
Token | we |
Token ID | 356 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
enthusi | +0.820 |
therap | +0.820 |
mosqu | +0.819 |
unden | +0.818 |
metic | +0.815 |
Neg logit contributions | |
"$:/ | -0.516 |
LW | -0.453 |
Malf | -0.418 |
Pyro | -0.400 |
spection | -0.370 |
Token | couldn |
Token ID | 3521 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
subsequ | +1.331 |
therap | +1.296 |
mosqu | +1.284 |
metic | +1.276 |
enthusi | +1.228 |
Neg logit contributions | |
"$:/ | -1.212 |
Malf | -1.020 |
LW | -0.974 |
lf | -0.906 |
Pyro | -0.888 |
Token | 't |
Token ID | 470 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
"$:/ | +0.276 |
LW | +0.230 |
Malf | +0.219 |
lf | +0.204 |
spection | +0.197 |
Neg logit contributions | |
mosqu | -0.412 |
metic | -0.411 |
therap | -0.411 |
enthusi | -0.409 |
subsequ | -0.408 |
Token | decide |
Token ID | 5409 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
subsequ | +0.730 |
therap | +0.677 |
mosqu | +0.672 |
enthusi | +0.667 |
exhib | +0.644 |
Neg logit contributions | |
"$:/ | -0.876 |
Malf | -0.748 |
LW | -0.744 |
Pyro | -0.731 |
Nanto | -0.676 |
Token | what |
Token ID | 644 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
"$:/ | +1.211 |
Malf | +1.077 |
LW | +1.075 |
Pyro | +1.006 |
spection | +0.935 |
Neg logit contributions | |
subsequ | -1.511 |
therap | -1.392 |
upgr | -1.356 |
ammy | -1.344 |
mosqu | -1.344 |
Token | we |
Token ID | 356 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
"$:/ | +2.089 |
Malf | +1.818 |
LW | +1.813 |
Pyro | +1.703 |
Nanto | +1.686 |
Neg logit contributions | |
subsequ | -2.056 |
therap | -1.917 |
upgr | -1.879 |
ammy | -1.838 |
erosion | -1.803 |
Token | can |
Token ID | 460 |
Feature activation | +4.79e-02 |
Pos logit contributions | |
"$:/ | +1.125 |
LW | +0.944 |
Malf | +0.931 |
Pyro | +0.884 |
lf | +0.850 |
Neg logit contributions | |
subsequ | -1.199 |
therap | -1.096 |
ammy | -1.070 |
upgr | -1.055 |
unden | -1.052 |
Token | say |
Token ID | 910 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
"$:/ | +0.558 |
LW | +0.468 |
Malf | +0.459 |
Pyro | +0.453 |
Nanto | +0.445 |
Neg logit contributions | |
subsequ | -0.488 |
therap | -0.449 |
upgr | -0.437 |
momentum | -0.430 |
enthusi | -0.428 |
Token | , |
Token ID | 11 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
"$:/ | +2.032 |
LW | +1.741 |
Malf | +1.734 |
Pyro | +1.629 |
spection | +1.626 |
Neg logit contributions | |
subsequ | -2.239 |
therap | -2.071 |
upgr | -2.041 |
exhib | -1.987 |
mosqu | -1.987 |
Token | see |
Token ID | 766 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
"$:/ | +7.438 |
LW | +6.491 |
Malf | +6.242 |
Pyro | +6.096 |
Nanto | +5.877 |
Neg logit contributions | |
subsequ | -7.648 |
therap | -7.206 |
upgr | -7.003 |
ammy | -6.716 |
unden | -6.610 |
Token | or |
Token ID | 393 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
"$:/ | +5.555 |
Malf | +5.078 |
LW | +4.879 |
Nanto | +4.831 |
Pyro | +4.625 |
Neg logit contributions | |
subsequ | -3.085 |
exhib | -2.934 |
momentum | -2.883 |
therap | -2.872 |
upgr | -2.829 |
Token | hear |
Token ID | 3285 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
"$:/ | +0.572 |
LW | +0.509 |
Malf | +0.465 |
Nanto | +0.446 |
lf | +0.444 |
Neg logit contributions | |
subsequ | -0.633 |
therap | -0.627 |
unden | -0.589 |
pestic | -0.580 |
enthusi | -0.580 |
Token | ? |
Token ID | 30 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
"$:/ | +3.345 |
LW | +2.896 |
Malf | +2.839 |
Pyro | +2.688 |
abouts | +2.581 |
Neg logit contributions | |
subsequ | -4.507 |
therap | -4.171 |
upgr | -4.093 |
mosqu | -4.092 |
enthusi | -4.007 |
Token | Who |
Token ID | 5338 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
subsequ | +1.504 |
therap | +1.413 |
exhib | +1.361 |
mosqu | +1.356 |
enthusi | +1.349 |
Neg logit contributions | |
"$:/ | -1.103 |
Malf | -1.030 |
LW | -1.008 |
Pyro | -0.945 |
Nanto | -0.889 |
Token | and |
Token ID | 290 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
"$:/ | +3.271 |
Malf | +2.882 |
LW | +2.801 |
Nanto | +2.752 |
Pyro | +2.635 |
Neg logit contributions | |
subsequ | -2.520 |
therap | -2.304 |
exhib | -2.300 |
upgr | -2.288 |
momentum | -2.271 |
Token | what |
Token ID | 644 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
"$:/ | +1.428 |
Malf | +1.256 |
LW | +1.239 |
Nanto | +1.196 |
Pyro | +1.155 |
Neg logit contributions | |
subsequ | -1.135 |
therap | -1.055 |
exhib | -1.035 |
upgr | -1.026 |
momentum | -1.013 |
Token | scope |
Token ID | 8354 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
"$:/ | +3.478 |
LW | +3.408 |
Pyro | +2.937 |
Malf | +2.818 |
spection | +2.644 |
Neg logit contributions | |
enthusi | -4.577 |
therap | -4.556 |
undermin | -4.524 |
mosqu | -4.510 |
unden | -4.494 |
Token | do |
Token ID | 466 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
"$:/ | +1.453 |
LW | +1.374 |
Pyro | +1.179 |
AF | +1.095 |
Nanto | +1.087 |
Neg logit contributions | |
enthusi | -2.232 |
therap | -2.180 |
unden | -2.174 |
mosqu | -2.119 |
metic | -2.100 |
Token | those |
Token ID | 883 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
"$:/ | +1.124 |
LW | +0.950 |
Malf | +0.851 |
Pyro | +0.842 |
Nanto | +0.770 |
Neg logit contributions | |
enthusi | -1.699 |
therap | -1.682 |
unden | -1.645 |
subsequ | -1.634 |
metic | -1.632 |
Token | parameters |
Token ID | 10007 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
"$:/ | +2.561 |
LW | +2.328 |
Malf | +2.052 |
Pyro | +1.863 |
abouts | +1.732 |
Neg logit contributions | |
therap | -3.661 |
unden | -3.532 |
enthusi | -3.519 |
metic | -3.489 |
undermin | -3.435 |
Token | have |
Token ID | 423 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
"$:/ | +7.249 |
LW | +6.548 |
AF | +5.562 |
lf | +5.485 |
Malf | +5.329 |
Neg logit contributions | |
enthusi | -10.163 |
mosqu | -10.043 |
millenn | -9.947 |
therap | -9.699 |
unden | -9.637 |
Token | ? |
Token ID | 30 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
therap | +1.795 |
subsequ | +1.743 |
mosqu | +1.713 |
enthusi | +1.700 |
unden | +1.681 |
Neg logit contributions | |
"$:/ | -0.872 |
LW | -0.766 |
Malf | -0.736 |
Pyro | -0.629 |
spection | -0.617 |
Token | can |
Token ID | 460 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
"$:/ | +0.659 |
LW | +0.630 |
Malf | +0.598 |
Pyro | +0.557 |
Nanto | +0.515 |
Neg logit contributions | |
therap | -0.770 |
subsequ | -0.730 |
ammy | -0.712 |
exhib | -0.708 |
enthusi | -0.704 |
Token | a |
Token ID | 257 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
"$:/ | +0.560 |
LW | +0.497 |
Malf | +0.487 |
Nanto | +0.452 |
Pyro | +0.451 |
Neg logit contributions | |
subsequ | -0.496 |
therap | -0.485 |
exhib | -0.476 |
ammy | -0.464 |
upgr | -0.452 |
Token | block |
Token ID | 2512 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
therap | +0.323 |
subsequ | +0.321 |
exhib | +0.320 |
ammy | +0.311 |
erosion | +0.308 |
Neg logit contributions | |
"$:/ | -0.352 |
LW | -0.310 |
Malf | -0.302 |
Pyro | -0.280 |
spection | -0.269 |
Token | Recommend |
Token ID | 30361 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
"$:/ | +2.294 |
Malf | +1.851 |
LW | +1.815 |
Pyro | +1.749 |
Nanto | +1.691 |
Neg logit contributions | |
subsequ | -5.276 |
therap | -5.208 |
mosqu | -4.922 |
traged | -4.882 |
enthusi | -4.831 |
Token |
|
Token ID | 198 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
therap | +4.788 |
enthusi | +4.587 |
mosqu | +4.526 |
subsequ | +4.479 |
unden | +4.466 |
Neg logit contributions | |
"$:/ | -3.312 |
LW | -2.838 |
Malf | -2.722 |
Pyro | -2.526 |
Nanto | -2.432 |
Token |
|
Token ID | 198 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
"$:/ | +1.339 |
LW | +1.114 |
Malf | +1.010 |
lf | +0.958 |
Pyro | +0.952 |
Neg logit contributions | |
therap | -2.831 |
mosqu | -2.811 |
enthusi | -2.809 |
metic | -2.802 |
unden | -2.721 |
Token | Recommend |
Token ID | 41248 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
"$:/ | +0.597 |
Malf | +0.534 |
LW | +0.515 |
Py | +0.481 |
lf | +0.474 |
Neg logit contributions | |
therap | -1.186 |
mosqu | -1.140 |
subsequ | -1.127 |
metic | -1.114 |
traged | -1.086 |
Token | Share |
Token ID | 8734 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
subsequ | +8.775 |
therap | +8.388 |
exhib | +8.247 |
upgr | +8.194 |
momentum | +7.973 |
Neg logit contributions | |
"$:/ | -7.004 |
Malf | -5.973 |
LW | -5.892 |
Nanto | -5.610 |
Pyro | -5.386 |
Token | this |
Token ID | 428 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
"$:/ | +2.594 |
LW | +2.172 |
Malf | +2.086 |
Pyro | +1.926 |
Nanto | +1.896 |
Neg logit contributions | |
enthusi | -4.951 |
mosqu | -4.922 |
metic | -4.854 |
obser | -4.842 |
therap | -4.771 |
Token | comment |
Token ID | 2912 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
"$:/ | +3.868 |
LW | +3.843 |
Pyro | +3.570 |
Malf | +3.481 |
LG | +3.446 |
Neg logit contributions | |
mosqu | -10.904 |
unden | -10.818 |
therap | -10.472 |
subsequ | -10.312 |
metic | -10.297 |
Token | on |
Token ID | 319 |
Feature activation | +6.58e-02 |
Pos logit contributions | |
"$:/ | +0.098 |
LW | +0.085 |
Malf | +0.081 |
Pyro | +0.073 |
Nanto | +0.069 |
Neg logit contributions | |
subsequ | -0.133 |
therap | -0.132 |
enthusi | -0.129 |
mosqu | -0.129 |
obser | -0.128 |
Token | Facebook |
Token ID | 3203 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
LW | +0.416 |
"$:/ | +0.404 |
Malf | +0.347 |
Pyro | +0.326 |
LG | +0.307 |
Neg logit contributions | |
subsequ | -0.974 |
therap | -0.944 |
enthusi | -0.936 |
traged | -0.928 |
upgr | -0.925 |
Token | Share |
Token ID | 8734 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
subsequ | +1.051 |
upgr | +1.007 |
therap | +1.002 |
exhib | +0.969 |
mosqu | +0.954 |
Neg logit contributions | |
"$:/ | -0.530 |
LW | -0.453 |
Malf | -0.412 |
Pyro | -0.406 |
Nanto | -0.394 |
Token | this |
Token ID | 428 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
"$:/ | +0.629 |
LW | +0.509 |
Malf | +0.496 |
Pyro | +0.473 |
spection | +0.464 |
Neg logit contributions | |
enthusi | -1.029 |
metic | -1.001 |
mosqu | -0.998 |
unden | -0.997 |
therap | -0.987 |
Token | always |
Token ID | 1464 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
subsequ | +0.451 |
therap | +0.425 |
upgr | +0.409 |
mosqu | +0.409 |
exhib | +0.406 |
Neg logit contributions | |
"$:/ | -0.538 |
LW | -0.463 |
Malf | -0.463 |
Nanto | -0.451 |
lf | -0.432 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.72e-03 |
Pos logit contributions | |
subsequ | +0.476 |
therap | +0.447 |
mosqu | +0.438 |
unden | +0.436 |
pestic | +0.432 |
Neg logit contributions | |
"$:/ | -0.551 |
Malf | -0.489 |
LW | -0.488 |
Nanto | -0.463 |
abouts | -0.436 |
Token | to |
Token ID | 284 |
Feature activation | +6.73e-02 |
Pos logit contributions | |
subsequ | +0.092 |
therap | +0.087 |
exhib | +0.084 |
upgr | +0.084 |
erosion | +0.082 |
Neg logit contributions | |
"$:/ | -0.094 |
Malf | -0.085 |
LW | -0.081 |
Nanto | -0.080 |
Pyro | -0.076 |
Token | be |
Token ID | 307 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
"$:/ | +0.742 |
Malf | +0.650 |
LW | +0.635 |
Pyro | +0.604 |
Nanto | +0.603 |
Neg logit contributions | |
subsequ | -0.699 |
mosqu | -0.668 |
therap | -0.659 |
enthusi | -0.642 |
unden | -0.641 |
Token | by |
Token ID | 416 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
"$:/ | +2.488 |
Malf | +2.392 |
LW | +2.214 |
Pyro | +2.192 |
Nanto | +2.133 |
Neg logit contributions | |
subsequ | -2.826 |
therap | -2.567 |
mosqu | -2.528 |
upgr | -2.485 |
artments | -2.458 |
Token | my |
Token ID | 616 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
subsequ | +0.989 |
therap | +0.933 |
upgr | +0.909 |
exhib | +0.900 |
momentum | +0.876 |
Neg logit contributions | |
"$:/ | -1.167 |
Malf | -1.052 |
LW | -1.020 |
Nanto | -1.001 |
Pyro | -0.976 |
Token | side |
Token ID | 1735 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
"$:/ | +3.566 |
LW | +3.317 |
Malf | +3.308 |
Pyro | +2.830 |
Nanto | +2.579 |
Neg logit contributions | |
subsequ | -10.563 |
therap | -10.277 |
mosqu | -10.276 |
upgr | -10.172 |
enthusi | -9.909 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
subsequ | +0.206 |
therap | +0.192 |
upgr | +0.188 |
exhib | +0.185 |
mosqu | +0.181 |
Neg logit contributions | |
"$:/ | -0.218 |
Malf | -0.194 |
LW | -0.188 |
Nanto | -0.182 |
spection | -0.180 |
Token |
|
Token ID | 198 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
"$:/ | +1.644 |
Malf | +1.480 |
LW | +1.415 |
Nanto | +1.388 |
Pyro | +1.309 |
Neg logit contributions | |
subsequ | -2.113 |
therap | -2.030 |
mosqu | -1.920 |
exhib | -1.909 |
upgr | -1.900 |
Token |
|
Token ID | 198 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
"$:/ | +0.565 |
LW | +0.484 |
Malf | +0.448 |
Pyro | +0.410 |
lf | +0.401 |
Neg logit contributions | |
therap | -1.181 |
enthusi | -1.173 |
mosqu | -1.171 |
metic | -1.149 |
undermin | -1.139 |
Token | � |
Token ID | 447 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
"$:/ | +0.255 |
Malf | +0.220 |
LW | +0.210 |
spection | +0.198 |
Py | +0.195 |
Neg logit contributions | |
subsequ | -0.383 |
therap | -0.377 |
mosqu | -0.364 |
enthusi | -0.356 |
metic | -0.352 |
Token |
|
Token ID | 198 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
"$:/ | +1.411 |
LW | +1.257 |
Malf | +1.254 |
Nanto | +1.178 |
spection | +1.151 |
Neg logit contributions | |
subsequ | -1.240 |
therap | -1.141 |
exhib | -1.133 |
upgr | -1.127 |
momentum | -1.091 |
Token |
|
Token ID | 198 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
therap | +3.212 |
enthusi | +3.193 |
mosqu | +3.170 |
metic | +3.099 |
unden | +3.089 |
Neg logit contributions | |
"$:/ | -2.002 |
LW | -1.714 |
Malf | -1.627 |
Pyro | -1.483 |
lf | -1.465 |
Token | Read |
Token ID | 5569 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
subsequ | +0.767 |
therap | +0.736 |
upgr | +0.725 |
exhib | +0.709 |
enthusi | +0.708 |
Neg logit contributions | |
"$:/ | -0.511 |
Malf | -0.442 |
LW | -0.433 |
spection | -0.409 |
lf | -0.397 |
Token | or |
Token ID | 393 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
subsequ | +1.148 |
therap | +1.102 |
upgr | +1.084 |
exhib | +1.063 |
momentum | +1.033 |
Neg logit contributions | |
"$:/ | -1.140 |
LW | -1.047 |
Malf | -1.026 |
Nanto | -0.957 |
spection | -0.941 |
Token | Share |
Token ID | 8734 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
"$:/ | +2.908 |
LW | +2.381 |
Malf | +2.097 |
LG | +1.584 |
Nanto | +1.502 |
Neg logit contributions | |
mosqu | -10.144 |
therap | -9.747 |
subsequ | -9.584 |
enthusi | -9.498 |
metic | -9.247 |
Token | this |
Token ID | 428 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
"$:/ | +3.158 |
LW | +2.977 |
Malf | +2.597 |
LG | +2.333 |
Pyro | +2.330 |
Neg logit contributions | |
enthusi | -6.423 |
mosqu | -6.139 |
therap | -5.984 |
unden | -5.966 |
metic | -5.905 |
Token | story |
Token ID | 1621 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
LW | +4.510 |
"$:/ | +4.131 |
LG | +3.920 |
Malf | +3.675 |
Nanto | +3.653 |
Neg logit contributions | |
mosqu | -10.813 |
enthusi | -10.442 |
metic | -10.281 |
unden | -10.130 |
therap | -10.087 |
Token | : |
Token ID | 25 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
"$:/ | +2.482 |
LW | +2.155 |
Malf | +1.852 |
lf | +1.847 |
Pyro | +1.731 |
Neg logit contributions | |
mosqu | -5.513 |
enthusi | -5.343 |
metic | -5.336 |
therap | -5.306 |
unden | -5.188 |
Token | http |
Token ID | 2638 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
"$:/ | +2.150 |
LW | +1.823 |
Malf | +1.814 |
Nanto | +1.646 |
Pyro | +1.515 |
Neg logit contributions | |
subsequ | -3.972 |
exhib | -3.699 |
upgr | -3.685 |
therap | -3.659 |
momentum | -3.603 |
Token | :// |
Token ID | 1378 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
"$:/ | +2.984 |
LW | +2.664 |
lf | +2.408 |
Malf | +2.368 |
spection | +2.251 |
Neg logit contributions | |
enthusi | -5.337 |
therap | -5.313 |
mosqu | -5.271 |
unden | -5.198 |
metic | -5.170 |
Token | usat |
Token ID | 37937 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
"$:/ | +3.181 |
LW | +2.728 |
Malf | +2.569 |
spection | +2.464 |
lf | +2.414 |
Neg logit contributions | |
subsequ | -5.228 |
upgr | -5.126 |
exhib | -5.073 |
therap | -5.019 |
momentum | -4.857 |
Token | the |
Token ID | 262 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
"$:/ | +4.630 |
Malf | +4.143 |
LW | +4.091 |
Nanto | +4.012 |
Pyro | +3.800 |
Neg logit contributions | |
subsequ | -3.989 |
therap | -3.709 |
exhib | -3.660 |
upgr | -3.588 |
momentum | -3.502 |
Token | river |
Token ID | 7850 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
"$:/ | +3.986 |
Malf | +3.760 |
LW | +3.696 |
Nanto | +3.555 |
Pyro | +3.487 |
Neg logit contributions | |
subsequ | -3.629 |
therap | -3.433 |
enthusi | -3.214 |
ammy | -3.199 |
upgr | -3.194 |
Token | for |
Token ID | 329 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
"$:/ | +2.841 |
Malf | +2.522 |
LW | +2.486 |
Nanto | +2.421 |
Pyro | +2.339 |
Neg logit contributions | |
subsequ | -2.607 |
therap | -2.486 |
exhib | -2.259 |
upgr | -2.251 |
inertia | -2.224 |
Token | emergency |
Token ID | 6334 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
"$:/ | +2.982 |
Malf | +2.817 |
LW | +2.811 |
Nanto | +2.651 |
Pyro | +2.589 |
Neg logit contributions | |
subsequ | -2.473 |
therap | -2.350 |
constitu | -2.163 |
enthusi | -2.154 |
mosqu | -2.151 |
Token | evacuation |
Token ID | 24663 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
"$:/ | +1.665 |
LW | +1.528 |
Malf | +1.509 |
Nanto | +1.396 |
Pyro | +1.394 |
Neg logit contributions | |
subsequ | -1.196 |
ammy | -1.110 |
therap | -1.104 |
mosqu | -1.064 |
unden | -1.054 |
Token | ! |
Token ID | 0 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
"$:/ | +1.760 |
LW | +1.543 |
Malf | +1.542 |
Nanto | +1.430 |
spection | +1.414 |
Neg logit contributions | |
subsequ | -1.592 |
therap | -1.509 |
ammy | -1.480 |
exhib | -1.432 |
mosqu | -1.407 |
Token |
|
Token ID | 198 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
"$:/ | +5.876 |
Malf | +5.616 |
LW | +5.317 |
Pyro | +5.030 |
Nanto | +4.870 |
Neg logit contributions | |
subsequ | -9.735 |
therap | -9.646 |
mosqu | -9.012 |
satell | -8.961 |
undermin | -8.942 |
Token |
|
Token ID | 198 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
"$:/ | +2.754 |
LW | +2.373 |
Malf | +2.012 |
lf | +1.880 |
Pyro | +1.856 |
Neg logit contributions | |
mosqu | -7.414 |
enthusi | -7.403 |
therap | -7.395 |
metic | -7.283 |
undermin | -7.235 |
Token | Rec |
Token ID | 6690 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
"$:/ | +3.253 |
Malf | +2.733 |
LW | +2.731 |
Py | +2.509 |
lf | +2.496 |
Neg logit contributions | |
therap | -5.244 |
subsequ | -5.237 |
mosqu | -4.957 |
metic | -4.850 |
enthusi | -4.822 |
Token | alling |
Token ID | 9221 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
"$:/ | +1.130 |
LW | +1.040 |
Malf | +1.017 |
Nanto | +0.964 |
spection | +0.957 |
Neg logit contributions | |
subsequ | -1.097 |
therap | -1.041 |
exhib | -1.021 |
momentum | -0.995 |
mosqu | -0.992 |
Token | that |
Token ID | 326 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
"$:/ | +3.805 |
Malf | +3.726 |
LW | +3.596 |
Nanto | +3.439 |
Pyro | +3.325 |
Neg logit contributions | |
subsequ | -4.065 |
therap | -4.007 |
upgr | -3.790 |
exhib | -3.788 |
ammy | -3.679 |
Token | again |
Token ID | 757 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
"$:/ | +0.070 |
Malf | +0.063 |
LW | +0.061 |
spection | +0.057 |
Nanto | +0.057 |
Neg logit contributions | |
subsequ | -0.079 |
therap | -0.077 |
mosqu | -0.072 |
upgr | -0.072 |
pestic | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
"$:/ | +1.685 |
Malf | +1.468 |
LW | +1.434 |
spection | +1.380 |
Nanto | +1.365 |
Neg logit contributions | |
subsequ | -1.888 |
therap | -1.799 |
upgr | -1.713 |
exhib | -1.694 |
mosqu | -1.685 |
Token | The |
Token ID | 383 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
"$:/ | +2.228 |
Malf | +2.172 |
LW | +2.147 |
Pyro | +1.984 |
Nanto | +1.838 |
Neg logit contributions | |
subsequ | -3.163 |
therap | -3.023 |
mosqu | -2.871 |
exhib | -2.785 |
undermin | -2.770 |
Token | wing |
Token ID | 8539 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
"$:/ | +2.440 |
Malf | +2.352 |
LW | +2.272 |
Nanto | +2.096 |
Pyro | +2.073 |
Neg logit contributions | |
subsequ | -1.942 |
therap | -1.775 |
exhib | -1.732 |
ammy | -1.728 |
upgr | -1.721 |
Token | tip |
Token ID | 8171 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
"$:/ | +1.665 |
Malf | +1.523 |
LW | +1.486 |
Nanto | +1.406 |
Pyro | +1.388 |
Neg logit contributions | |
subsequ | -1.222 |
therap | -1.146 |
upgr | -1.115 |
exhib | -1.106 |
ammy | -1.084 |
Token | position |
Token ID | 2292 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
"$:/ | +3.713 |
Malf | +3.504 |
LW | +3.415 |
spection | +3.108 |
Nanto | +3.036 |
Neg logit contributions | |
subsequ | -3.130 |
therap | -3.091 |
ammy | -3.026 |
upgr | -2.893 |
exhib | -2.873 |
Token | lights |
Token ID | 7588 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
"$:/ | +6.398 |
Malf | +6.040 |
LW | +5.969 |
spection | +5.166 |
Pyro | +5.014 |
Neg logit contributions | |
subsequ | -7.653 |
therap | -7.469 |
ammy | -7.223 |
upgr | -7.148 |
exhib | -7.021 |
Token | were |
Token ID | 547 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
subsequ | +1.347 |
therap | +1.275 |
mosqu | +1.198 |
upgr | +1.195 |
exhib | +1.194 |
Neg logit contributions | |
"$:/ | -1.638 |
Malf | -1.507 |
LW | -1.480 |
Nanto | -1.352 |
Pyro | -1.334 |
Token | now |
Token ID | 783 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
"$:/ | +0.164 |
Malf | +0.157 |
LW | +0.153 |
Pyro | +0.135 |
Nanto | +0.134 |
Neg logit contributions | |
subsequ | -0.161 |
therap | -0.150 |
mosqu | -0.142 |
ammy | -0.142 |
exhib | -0.141 |
Token | c |
Token ID | 269 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
"$:/ | +0.370 |
Malf | +0.352 |
LW | +0.346 |
Pyro | +0.307 |
Nanto | +0.303 |
Neg logit contributions | |
subsequ | -0.367 |
therap | -0.345 |
ammy | -0.327 |
mosqu | -0.326 |
exhib | -0.325 |
Token | anted |
Token ID | 4126 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
subsequ | +1.392 |
therap | +1.332 |
exhib | +1.303 |
upgr | +1.267 |
momentum | +1.253 |
Neg logit contributions | |
"$:/ | -1.415 |
Malf | -1.283 |
LW | -1.240 |
Nanto | -1.161 |
Pyro | -1.144 |
Token | Ag |
Token ID | 10262 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
"$:/ | +0.576 |
Malf | +0.504 |
LW | +0.499 |
spection | +0.471 |
Nanto | +0.449 |
Neg logit contributions | |
subsequ | -0.736 |
therap | -0.682 |
upgr | -0.678 |
erosion | -0.667 |
exhib | -0.665 |
Token | reement |
Token ID | 10237 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
subsequ | +5.289 |
therap | +5.043 |
upgr | +4.978 |
exhib | +4.927 |
momentum | +4.835 |
Neg logit contributions | |
"$:/ | -3.973 |
Malf | -3.461 |
LW | -3.360 |
Nanto | -3.181 |
Pyro | -3.072 |
Token | number |
Token ID | 1271 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
"$:/ | +1.822 |
LW | +1.693 |
Malf | +1.602 |
Nanto | +1.523 |
Pyro | +1.475 |
Neg logit contributions | |
subsequ | -2.107 |
therap | -1.984 |
upgr | -1.972 |
ammy | -1.951 |
exhib | -1.893 |
Token | : |
Token ID | 25 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
"$:/ | +4.474 |
LW | +4.370 |
Malf | +4.108 |
LG | +3.745 |
Pyro | +3.692 |
Neg logit contributions | |
subsequ | -5.493 |
therap | -5.223 |
ammy | -5.137 |
obser | -5.026 |
mosqu | -4.984 |
Token | 501 |
Token ID | 24555 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
"$:/ | +2.164 |
LW | +2.087 |
Malf | +1.974 |
LG | +1.784 |
Nanto | +1.763 |
Neg logit contributions | |
subsequ | -2.639 |
therap | -2.463 |
upgr | -2.414 |
ammy | -2.411 |
momentum | -2.349 |
Token | 20 |
Token ID | 1238 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
"$:/ | +4.259 |
LW | +3.830 |
Malf | +3.648 |
Nanto | +3.300 |
Pyro | +3.295 |
Neg logit contributions | |
subsequ | -4.999 |
therap | -4.940 |
mosqu | -4.657 |
upgr | -4.627 |
ammy | -4.619 |
Token | 78 |
Token ID | 3695 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
"$:/ | +6.356 |
LW | +5.700 |
Malf | +5.475 |
Nanto | +5.045 |
Pyro | +4.972 |
Neg logit contributions | |
subsequ | -7.809 |
therap | -7.579 |
ammy | -7.346 |
mosqu | -7.309 |
exhib | -7.229 |
Token | - |
Token ID | 12 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
"$:/ | +6.504 |
LW | +5.985 |
Malf | +5.719 |
Nanto | +5.072 |
Pyro | +5.063 |
Neg logit contributions | |
subsequ | -8.852 |
therap | -8.591 |
mosqu | -8.337 |
ammy | -8.332 |
traged | -8.236 |
Token | A |
Token ID | 32 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
"$:/ | +3.857 |
LW | +3.439 |
Malf | +3.375 |
lf | +2.998 |
Pyro | +2.968 |
Neg logit contributions | |
subsequ | -5.520 |
therap | -5.280 |
mosqu | -5.092 |
upgr | -5.091 |
erosion | -5.026 |
Token | ). |
Token ID | 737 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
"$:/ | +3.972 |
LW | +3.556 |
Malf | +3.502 |
Nanto | +3.268 |
spection | +3.215 |
Neg logit contributions | |
subsequ | -5.098 |
therap | -4.886 |
upgr | -4.814 |
ammy | -4.728 |
exhib | -4.660 |
Token | The |
Token ID | 383 |
Feature activation | -4.77e-03 |
Pos logit contributions | |
subsequ | +0.604 |
therap | +0.564 |
ammy | +0.548 |
upgr | +0.546 |
undermin | +0.538 |
Neg logit contributions | |
"$:/ | -0.471 |
Malf | -0.431 |
LW | -0.428 |
Pyro | -0.402 |
Nanto | -0.386 |
Token | � |
Token ID | 247 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
"$:/ | +0.140 |
LW | +0.134 |
Malf | +0.129 |
Pyro | +0.116 |
spection | +0.116 |
Neg logit contributions | |
subsequ | -0.156 |
therap | -0.154 |
mosqu | -0.149 |
enthusi | -0.145 |
unden | -0.140 |
Token | s |
Token ID | 82 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
"$:/ | +2.079 |
Malf | +1.873 |
LW | +1.819 |
Nanto | +1.799 |
Pyro | +1.727 |
Neg logit contributions | |
subsequ | -1.200 |
exhib | -1.104 |
momentum | -1.086 |
therap | -1.085 |
upgr | -1.081 |
Token | top |
Token ID | 1353 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
"$:/ | +2.619 |
Malf | +2.530 |
LW | +2.384 |
Pyro | +2.192 |
LG | +2.173 |
Neg logit contributions | |
subsequ | -2.016 |
therap | -1.978 |
mosqu | -1.918 |
enthusi | -1.860 |
obser | -1.804 |
Token | discipline |
Token ID | 12883 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
"$:/ | +2.595 |
Malf | +2.437 |
LW | +2.408 |
Nanto | +2.215 |
spection | +2.190 |
Neg logit contributions | |
subsequ | -1.963 |
therap | -1.814 |
ammy | -1.775 |
mosqu | -1.744 |
upgr | -1.742 |
Token | watchdog |
Token ID | 26856 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
subsequ | +0.522 |
therap | +0.462 |
ammy | +0.456 |
exhib | +0.446 |
mosqu | +0.442 |
Neg logit contributions | |
"$:/ | -0.681 |
LW | -0.621 |
Malf | -0.615 |
Nanto | -0.580 |
Pyro | -0.569 |
Token | , |
Token ID | 11 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
"$:/ | +5.010 |
LW | +4.847 |
Malf | +4.796 |
Nanto | +4.523 |
spection | +4.309 |
Neg logit contributions | |
subsequ | -5.534 |
therap | -5.135 |
exhib | -4.958 |
upgr | -4.954 |
enthusi | -4.942 |
Token | the |
Token ID | 262 |
Feature activation | +8.99e-01 |
Pos logit contributions | |
"$:/ | +6.398 |
LW | +6.284 |
Malf | +6.012 |
Nanto | +5.694 |
LG | +5.487 |
Neg logit contributions | |
subsequ | -7.440 |
therap | -7.148 |
exhib | -6.797 |
mosqu | -6.773 |
upgr | -6.760 |
Token | Central |
Token ID | 5694 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
LW | +7.682 |
"$:/ | +7.358 |
Malf | +7.106 |
LG | +6.654 |
spection | +6.409 |
Neg logit contributions | |
subsequ | -8.971 |
therap | -8.773 |
mosqu | -8.431 |
enthusi | -8.142 |
upgr | -8.101 |
Token | Commission |
Token ID | 4513 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
LW | +4.914 |
"$:/ | +4.804 |
Malf | +4.649 |
spection | +4.348 |
LG | +4.311 |
Neg logit contributions | |
subsequ | -7.037 |
therap | -6.897 |
mosqu | -6.814 |
enthusi | -6.561 |
exhib | -6.372 |
Token | for |
Token ID | 329 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
"$:/ | +3.692 |
Malf | +3.282 |
LW | +3.168 |
Nanto | +3.128 |
Pyro | +3.011 |
Neg logit contributions | |
subsequ | -2.621 |
therap | -2.416 |
exhib | -2.394 |
upgr | -2.378 |
momentum | -2.360 |
Token | Discipline |
Token ID | 48532 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
"$:/ | +2.097 |
Malf | +1.849 |
LW | +1.826 |
spection | +1.521 |
Pyro | +1.509 |
Neg logit contributions | |
subsequ | -6.294 |
therap | -6.270 |
mosqu | -6.016 |
exhib | -5.888 |
inertia | -5.834 |
Token |
|
Token ID | 198 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
subsequ | +0.630 |
exhib | +0.589 |
momentum | +0.582 |
therap | +0.581 |
upgr | +0.579 |
Neg logit contributions | |
"$:/ | -0.923 |
Malf | -0.806 |
LW | -0.785 |
Nanto | -0.783 |
spection | -0.743 |
Token |
|
Token ID | 198 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
"$:/ | +1.609 |
LW | +1.368 |
Malf | +1.256 |
lf | +1.175 |
Pyro | +1.160 |
Neg logit contributions | |
enthusi | -3.755 |
therap | -3.746 |
metic | -3.728 |
mosqu | -3.717 |
undermin | -3.640 |
Token | Read |
Token ID | 5569 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
"$:/ | +0.983 |
Malf | +0.847 |
LW | +0.824 |
lf | +0.771 |
Py | +0.767 |
Neg logit contributions | |
subsequ | -1.370 |
therap | -1.321 |
mosqu | -1.297 |
metic | -1.266 |
enthusi | -1.242 |
Token | or |
Token ID | 393 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
subsequ | +1.791 |
therap | +1.687 |
exhib | +1.667 |
upgr | +1.657 |
momentum | +1.607 |
Neg logit contributions | |
"$:/ | -2.062 |
Malf | -1.838 |
LW | -1.813 |
Nanto | -1.724 |
spection | -1.657 |
Token | Share |
Token ID | 8734 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
"$:/ | +2.951 |
LW | +2.186 |
Malf | +1.809 |
Nanto | +1.421 |
LG | +1.408 |
Neg logit contributions | |
mosqu | -9.745 |
therap | -9.445 |
enthusi | -9.242 |
subsequ | -9.121 |
metic | -9.003 |
Token | this |
Token ID | 428 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
"$:/ | +2.336 |
LW | +2.153 |
Malf | +1.895 |
Pyro | +1.689 |
lf | +1.677 |
Neg logit contributions | |
enthusi | -4.408 |
mosqu | -4.282 |
unden | -4.182 |
therap | -4.152 |
metic | -4.124 |
Token | story |
Token ID | 1621 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
LW | +4.335 |
"$:/ | +4.193 |
Nanto | +3.705 |
LG | +3.607 |
Malf | +3.498 |
Neg logit contributions | |
mosqu | -10.807 |
enthusi | -10.371 |
metic | -10.136 |
unden | -10.008 |
therap | -9.959 |
Token | : |
Token ID | 25 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
"$:/ | +2.080 |
LW | +1.777 |
Malf | +1.570 |
lf | +1.539 |
Pyro | +1.451 |
Neg logit contributions | |
mosqu | -4.359 |
enthusi | -4.283 |
metic | -4.262 |
therap | -4.247 |
unden | -4.157 |
Token | http |
Token ID | 2638 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
"$:/ | +1.952 |
Malf | +1.681 |
LW | +1.679 |
Nanto | +1.510 |
Pyro | +1.402 |
Neg logit contributions | |
subsequ | -3.570 |
therap | -3.349 |
upgr | -3.338 |
exhib | -3.322 |
unden | -3.255 |
Token | :// |
Token ID | 1378 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
"$:/ | +2.500 |
LW | +2.201 |
lf | +1.994 |
Malf | +1.988 |
spection | +1.902 |
Neg logit contributions | |
enthusi | -4.310 |
therap | -4.287 |
unden | -4.234 |
mosqu | -4.185 |
metic | -4.165 |
Token | cin |
Token ID | 17879 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
"$:/ | +3.243 |
LW | +2.813 |
Malf | +2.654 |
Nanto | +2.534 |
spection | +2.527 |
Neg logit contributions | |
subsequ | -4.827 |
upgr | -4.716 |
exhib | -4.660 |
therap | -4.641 |
erosion | -4.497 |
Token | Chandler |
Token ID | 28346 |
Feature activation | +5.77e-03 |
Pos logit contributions | |
"$:/ | +0.617 |
Malf | +0.530 |
LW | +0.523 |
Pyro | +0.451 |
Nanto | +0.433 |
Neg logit contributions | |
subsequ | -1.599 |
therap | -1.501 |
exhib | -1.451 |
upgr | -1.449 |
sugg | -1.428 |
Token | , |
Token ID | 11 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
"$:/ | +0.067 |
LW | +0.059 |
Malf | +0.059 |
Nanto | +0.057 |
Pyro | +0.055 |
Neg logit contributions | |
subsequ | -0.056 |
therap | -0.052 |
exhib | -0.051 |
upgr | -0.050 |
momentum | -0.049 |
Token | Al |
Token ID | 978 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
subsequ | +0.703 |
therap | +0.700 |
upgr | +0.638 |
exhib | +0.635 |
ammy | +0.635 |
Neg logit contributions | |
"$:/ | -0.596 |
LW | -0.557 |
Malf | -0.539 |
Nanto | -0.505 |
Pyro | -0.500 |
Token | - |
Token ID | 12 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
subsequ | +1.207 |
therap | +1.144 |
upgr | +1.119 |
exhib | +1.091 |
momentum | +1.079 |
Neg logit contributions | |
"$:/ | -1.378 |
Malf | -1.217 |
LW | -1.189 |
Nanto | -1.143 |
Pyro | -1.097 |
Token | Far |
Token ID | 21428 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
subsequ | +4.068 |
therap | +3.833 |
upgr | +3.766 |
exhib | +3.754 |
momentum | +3.709 |
Neg logit contributions | |
"$:/ | -3.222 |
Malf | -2.745 |
LW | -2.641 |
Nanto | -2.551 |
spection | -2.416 |
Token | ou |
Token ID | 280 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
subsequ | +5.743 |
therap | +5.401 |
upgr | +5.261 |
exhib | +5.189 |
erosion | +5.121 |
Neg logit contributions | |
"$:/ | -4.553 |
Malf | -3.961 |
LW | -3.708 |
Nanto | -3.698 |
Pyro | -3.437 |
Token | q |
Token ID | 80 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
subsequ | +10.784 |
therap | +10.030 |
exhib | +10.011 |
upgr | +9.923 |
momentum | +9.862 |
Neg logit contributions | |
"$:/ | -11.309 |
Malf | -9.844 |
LW | -9.570 |
Nanto | -9.351 |
Pyro | -8.894 |
Token | Amin |
Token ID | 39869 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
subsequ | +2.425 |
therap | +2.390 |
exhib | +2.193 |
mosqu | +2.170 |
obser | +2.157 |
Neg logit contributions | |
"$:/ | -2.377 |
Malf | -2.099 |
LW | -2.030 |
Pyro | -1.947 |
Nanto | -1.912 |
Token | u |
Token ID | 84 |
Feature activation | -9.83e-02 |
Pos logit contributions | |
subsequ | +1.898 |
therap | +1.830 |
exhib | +1.786 |
upgr | +1.777 |
mosqu | +1.771 |
Neg logit contributions | |
"$:/ | -1.093 |
Malf | -0.909 |
LW | -0.895 |
Nanto | -0.819 |
Pyro | -0.782 |
Token | , |
Token ID | 11 |
Feature activation | -8.34e-02 |
Pos logit contributions | |
subsequ | +0.630 |
momentum | +0.588 |
erosion | +0.587 |
upgr | +0.569 |
egg | +0.556 |
Neg logit contributions | |
"$:/ | -1.407 |
Malf | -1.268 |
Nanto | -1.212 |
spection | -1.190 |
LW | -1.189 |
Token | and |
Token ID | 290 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
subsequ | +0.884 |
therap | +0.842 |
exhib | +0.809 |
upgr | +0.804 |
erosion | +0.784 |
Neg logit contributions | |
"$:/ | -0.897 |
Malf | -0.795 |
LW | -0.790 |
Nanto | -0.753 |
Pyro | -0.728 |
Token | to |
Token ID | 284 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
subsequ | +2.587 |
therap | +2.385 |
upgr | +2.363 |
exhib | +2.359 |
inertia | +2.332 |
Neg logit contributions | |
"$:/ | -2.558 |
Malf | -2.284 |
LW | -2.270 |
Nanto | -2.097 |
Pyro | -2.075 |
Token | Recap |
Token ID | 46585 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
subsequ | +1.299 |
upgr | +1.190 |
exhib | +1.163 |
ammy | +1.157 |
therap | +1.157 |
Neg logit contributions | |
"$:/ | -1.407 |
LW | -1.325 |
Malf | -1.321 |
Pyro | -1.235 |
Nanto | -1.222 |
Token | San |
Token ID | 2986 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
subsequ | +4.359 |
therap | +4.094 |
exhib | +4.008 |
upgr | +3.998 |
inertia | +3.931 |
Neg logit contributions | |
"$:/ | -4.611 |
Malf | -4.088 |
LW | -4.038 |
Nanto | -3.875 |
spection | -3.747 |
Token | Juan |
Token ID | 16852 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
subsequ | +1.618 |
therap | +1.476 |
exhib | +1.475 |
upgr | +1.466 |
momentum | +1.453 |
Neg logit contributions | |
"$:/ | -2.413 |
Malf | -2.131 |
LW | -2.085 |
Nanto | -2.050 |
Pyro | -1.969 |
Token | Del |
Token ID | 4216 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
subsequ | +1.986 |
therap | +1.796 |
exhib | +1.768 |
upgr | +1.758 |
momentum | +1.725 |
Neg logit contributions | |
"$:/ | -3.553 |
Malf | -3.189 |
LW | -3.122 |
Nanto | -3.040 |
Pyro | -2.947 |
Token | Sur |
Token ID | 4198 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
therap | +5.399 |
subsequ | +5.281 |
enthusi | +5.181 |
mosqu | +5.015 |
unden | +4.974 |
Neg logit contributions | |
"$:/ | -3.131 |
Malf | -2.725 |
LW | -2.531 |
Pyro | -2.440 |
Nanto | -2.435 |
Token | Episode |
Token ID | 7922 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
subsequ | +8.776 |
therap | +8.154 |
exhib | +8.078 |
upgr | +8.039 |
momentum | +7.923 |
Neg logit contributions | |
"$:/ | -9.618 |
Malf | -8.427 |
LW | -8.249 |
Nanto | -7.967 |
Pyro | -7.652 |
Token | 5 |
Token ID | 642 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
subsequ | +5.852 |
therap | +5.466 |
exhib | +5.359 |
upgr | +5.312 |
erosion | +5.188 |
Neg logit contributions | |
"$:/ | -5.956 |
Malf | -5.224 |
LW | -5.152 |
Nanto | -4.887 |
Pyro | -4.736 |
Token |
|
Token ID | 198 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
subsequ | +1.739 |
therap | +1.616 |
exhib | +1.588 |
upgr | +1.588 |
inertia | +1.579 |
Neg logit contributions | |
"$:/ | -1.714 |
LW | -1.573 |
Malf | -1.529 |
Pyro | -1.427 |
Nanto | -1.421 |
Token |
|
Token ID | 198 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
therap | +0.545 |
enthusi | +0.543 |
mosqu | +0.543 |
metic | +0.532 |
unden | +0.528 |
Neg logit contributions | |
"$:/ | -0.286 |
LW | -0.249 |
Malf | -0.228 |
Pyro | -0.210 |
lf | -0.207 |
Token | Coach |
Token ID | 40677 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
subsequ | +0.858 |
therap | +0.810 |
mosqu | +0.786 |
upgr | +0.782 |
exhib | +0.781 |
Neg logit contributions | |
"$:/ | -0.636 |
LW | -0.566 |
Malf | -0.560 |
spection | -0.510 |
lf | -0.499 |
Token | vi |
Token ID | 8903 |
Feature activation | +7.38e-02 |
Pos logit contributions | |
subsequ | +5.264 |
therap | +4.956 |
exhib | +4.941 |
upgr | +4.919 |
momentum | +4.877 |
Neg logit contributions | |
"$:/ | -3.557 |
Malf | -2.985 |
LW | -2.869 |
Nanto | -2.781 |
Pyro | -2.618 |
Token | : |
Token ID | 25 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
"$:/ | +0.948 |
Malf | +0.849 |
LW | +0.830 |
Nanto | +0.819 |
Pyro | +0.788 |
Neg logit contributions | |
subsequ | -0.619 |
exhib | -0.565 |
upgr | -0.558 |
therap | -0.557 |
momentum | -0.557 |
Token | | |
Token ID | 91 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
subsequ | +4.021 |
exhib | +3.766 |
upgr | +3.718 |
therap | +3.706 |
momentum | +3.629 |
Neg logit contributions | |
"$:/ | -5.261 |
Malf | -4.653 |
LW | -4.512 |
Nanto | -4.472 |
Pyro | -4.303 |
Token | vim |
Token ID | 31124 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
subsequ | +3.351 |
exhib | +3.090 |
therap | +3.071 |
upgr | +3.065 |
momentum | +3.031 |
Neg logit contributions | |
"$:/ | -4.488 |
Malf | -3.991 |
LW | -3.862 |
Nanto | -3.820 |
Pyro | -3.685 |
Token | : |
Token ID | 25 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
therap | +2.328 |
mosqu | +2.315 |
enthusi | +2.267 |
subsequ | +2.263 |
unden | +2.260 |
Neg logit contributions | |
"$:/ | -1.685 |
LW | -1.365 |
Malf | -1.333 |
spection | -1.222 |
lf | -1.208 |
Token | | |
Token ID | 91 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
subsequ | +4.319 |
exhib | +4.142 |
upgr | +4.059 |
therap | +4.022 |
hol | +3.886 |
Neg logit contributions | |
"$:/ | -7.399 |
Malf | -6.550 |
Nanto | -6.359 |
LW | -6.354 |
Pyro | -6.185 |
Token | ex |
Token ID | 1069 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
subsequ | +6.758 |
exhib | +6.309 |
upgr | +6.172 |
therap | +6.166 |
momentum | +6.043 |
Neg logit contributions | |
"$:/ | -10.254 |
Malf | -9.224 |
LW | -8.853 |
Nanto | -8.792 |
Pyro | -8.510 |
Token | : |
Token ID | 25 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
subsequ | +0.129 |
exhib | +0.118 |
momentum | +0.116 |
therap | +0.115 |
upgr | +0.115 |
Neg logit contributions | |
"$:/ | -0.247 |
Malf | -0.225 |
LW | -0.219 |
Nanto | -0.218 |
Pyro | -0.212 |
Token | } |
Token ID | 92 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
subsequ | +6.337 |
exhib | +5.952 |
therap | +5.849 |
upgr | +5.842 |
momentum | +5.822 |
Neg logit contributions | |
"$:/ | -7.659 |
Malf | -6.802 |
Nanto | -6.556 |
LW | -6.493 |
Pyro | -6.299 |
Token | [ |
Token ID | 58 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
subsequ | +2.258 |
therap | +2.137 |
exhib | +2.094 |
upgr | +2.083 |
momentum | +2.065 |
Neg logit contributions | |
"$:/ | -2.093 |
Malf | -1.767 |
LW | -1.741 |
Nanto | -1.650 |
Pyro | -1.602 |
Token | white |
Token ID | 11186 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
subsequ | +1.710 |
therap | +1.566 |
exhib | +1.564 |
upgr | +1.544 |
momentum | +1.534 |
Neg logit contributions | |
"$:/ | -2.457 |
Malf | -2.217 |
LW | -2.139 |
Nanto | -2.119 |
Pyro | -2.029 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
subsequ | +0.578 |
therap | +0.554 |
upgr | +0.531 |
mosqu | +0.523 |
exhib | +0.518 |
Neg logit contributions | |
"$:/ | -0.486 |
LW | -0.430 |
Malf | -0.428 |
Pyro | -0.380 |
spection | -0.368 |
Token | K |
Token ID | 42 |
Feature activation | +8.51e-02 |
Pos logit contributions | |
"$:/ | +2.087 |
Malf | +1.809 |
LW | +1.747 |
spection | +1.653 |
lf | +1.633 |
Neg logit contributions | |
subsequ | -2.502 |
therap | -2.413 |
mosqu | -2.310 |
exhib | -2.288 |
upgr | -2.274 |
Token | UAL |
Token ID | 25620 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
"$:/ | +0.848 |
Malf | +0.757 |
LW | +0.738 |
Nanto | +0.733 |
lf | +0.682 |
Neg logit contributions | |
subsequ | -0.915 |
therap | -0.894 |
upgr | -0.867 |
exhib | -0.861 |
momentum | -0.852 |
Token | A |
Token ID | 32 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
subsequ | +1.870 |
therap | +1.769 |
mosqu | +1.742 |
upgr | +1.702 |
ammy | +1.693 |
Neg logit contributions | |
"$:/ | -1.095 |
LW | -0.960 |
AF | -0.895 |
Malf | -0.889 |
Pyro | -0.810 |
Token | L |
Token ID | 406 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
subsequ | +4.126 |
therap | +3.988 |
upgr | +3.827 |
exhib | +3.783 |
ammy | +3.755 |
Neg logit contributions | |
"$:/ | -3.109 |
LW | -2.694 |
Malf | -2.636 |
Nanto | -2.475 |
Pyro | -2.338 |
Token | UMP |
Token ID | 20476 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
subsequ | +4.951 |
therap | +4.856 |
sugg | +4.779 |
upgr | +4.724 |
mosqu | +4.720 |
Neg logit contributions | |
"$:/ | -2.663 |
Malf | -2.132 |
LW | -2.066 |
Nanto | -1.825 |
Pyro | -1.775 |
Token | UR |
Token ID | 4261 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
subsequ | +12.601 |
therap | +11.971 |
exhib | +11.806 |
upgr | +11.751 |
mosqu | +11.553 |
Neg logit contributions | |
"$:/ | -5.671 |
Malf | -4.497 |
LW | -4.380 |
Nanto | -3.930 |
Pyro | -3.740 |
Token | -- |
Token ID | 1377 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
subsequ | +2.085 |
therap | +1.961 |
exhib | +1.907 |
upgr | +1.892 |
momentum | +1.858 |
Neg logit contributions | |
"$:/ | -2.173 |
Malf | -1.941 |
LW | -1.917 |
Nanto | -1.800 |
Pyro | -1.736 |
Token | With |
Token ID | 2080 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
"$:/ | +0.463 |
Malf | +0.417 |
LW | +0.408 |
Nanto | +0.387 |
Pyro | +0.380 |
Neg logit contributions | |
subsequ | -0.478 |
therap | -0.456 |
exhib | -0.426 |
upgr | -0.422 |
mosqu | -0.422 |
Token | barely |
Token ID | 8523 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
subsequ | +0.329 |
therap | +0.315 |
upgr | +0.297 |
exhib | +0.295 |
unden | +0.294 |
Neg logit contributions | |
"$:/ | -0.371 |
Malf | -0.334 |
LW | -0.328 |
Nanto | -0.311 |
Pyro | -0.300 |
Token | a |
Token ID | 257 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
subsequ | +2.018 |
therap | +1.959 |
ammy | +1.889 |
mosqu | +1.864 |
artments | +1.826 |
Neg logit contributions | |
"$:/ | -1.840 |
LW | -1.678 |
Malf | -1.659 |
Nanto | -1.532 |
Pyro | -1.499 |
Token | seek |
Token ID | 5380 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
subsequ | +4.315 |
upgr | +3.936 |
therap | +3.909 |
exhib | +3.859 |
mosqu | +3.840 |
Neg logit contributions | |
"$:/ | -4.184 |
Malf | -3.812 |
LW | -3.659 |
Pyro | -3.402 |
Nanto | -3.333 |
Token | professional |
Token ID | 4708 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
subsequ | +3.844 |
upgr | +3.428 |
ammy | +3.412 |
exhib | +3.374 |
therap | +3.299 |
Neg logit contributions | |
"$:/ | -5.472 |
Malf | -4.990 |
LW | -4.941 |
Nanto | -4.700 |
Pyro | -4.611 |
Token | help |
Token ID | 1037 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
subsequ | +0.280 |
ammy | +0.271 |
mosqu | +0.252 |
oshenko | +0.252 |
upgr | +0.246 |
Neg logit contributions | |
"$:/ | -0.318 |
LW | -0.295 |
Malf | -0.291 |
Pyro | -0.274 |
Nanto | -0.273 |
Token | . |
Token ID | 13 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
"$:/ | +1.770 |
Malf | +1.499 |
LW | +1.491 |
spection | +1.439 |
Pyro | +1.369 |
Neg logit contributions | |
subsequ | -2.064 |
ammy | -1.944 |
mosqu | -1.886 |
upgr | -1.864 |
therap | -1.858 |
Token | Continue |
Token ID | 10054 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
subsequ | +0.322 |
mosqu | +0.294 |
therap | +0.292 |
upgr | +0.286 |
unden | +0.282 |
Neg logit contributions | |
"$:/ | -0.234 |
Malf | -0.215 |
LW | -0.210 |
Pyro | -0.196 |
Nanto | -0.185 |
Token | Reading |
Token ID | 11725 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
subsequ | +1.324 |
therap | +1.221 |
upgr | +1.217 |
ammy | +1.187 |
exhib | +1.173 |
Neg logit contributions | |
"$:/ | -1.198 |
Malf | -1.054 |
LW | -1.044 |
Nanto | -0.966 |
spection | -0.928 |
Token | Below |
Token ID | 10383 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
subsequ | +9.307 |
therap | +8.725 |
exhib | +8.683 |
upgr | +8.627 |
momentum | +8.582 |
Neg logit contributions | |
"$:/ | -7.851 |
Malf | -6.694 |
LW | -6.450 |
Nanto | -6.317 |
Pyro | -5.969 |
Token | Advertisement |
Token ID | 39711 |
Feature activation | -9.51e-02 |
Pos logit contributions | |
subsequ | +6.909 |
therap | +6.557 |
exhib | +6.496 |
upgr | +6.482 |
momentum | +6.380 |
Neg logit contributions | |
"$:/ | -3.012 |
Malf | -2.393 |
LW | -2.307 |
Nanto | -2.137 |
Pyro | -1.995 |
Token |
|
Token ID | 198 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
subsequ | +1.053 |
therap | +0.976 |
upgr | +0.966 |
exhib | +0.959 |
momentum | +0.937 |
Neg logit contributions | |
"$:/ | -1.023 |
Malf | -0.900 |
LW | -0.877 |
Nanto | -0.833 |
Pyro | -0.810 |
Token |
|
Token ID | 198 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
"$:/ | +1.576 |
LW | +1.319 |
Malf | +1.175 |
lf | +1.119 |
Pyro | +1.082 |
Neg logit contributions | |
enthusi | -3.634 |
therap | -3.602 |
mosqu | -3.592 |
metic | -3.579 |
unden | -3.517 |
Token | 9 |
Token ID | 24 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
"$:/ | +0.307 |
Malf | +0.254 |
LW | +0.252 |
lf | +0.237 |
spection | +0.233 |
Neg logit contributions | |
subsequ | -0.406 |
therap | -0.391 |
mosqu | -0.383 |
enthusi | -0.377 |
upgr | -0.374 |
Token | ij |
Token ID | 2926 |
Feature activation | +8.48e-02 |
Pos logit contributions | |
subsequ | +1.750 |
therap | +1.684 |
exhib | +1.620 |
upgr | +1.619 |
momentum | +1.607 |
Neg logit contributions | |
"$:/ | -1.666 |
Malf | -1.434 |
LW | -1.416 |
Nanto | -1.326 |
Pyro | -1.302 |
Token | ii |
Token ID | 4178 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
"$:/ | +0.922 |
LW | +0.797 |
Malf | +0.796 |
Nanto | +0.763 |
spection | +0.748 |
Neg logit contributions | |
subsequ | -0.894 |
therap | -0.849 |
exhib | -0.825 |
upgr | -0.818 |
momentum | -0.801 |
Token | > |
Token ID | 29 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
"$:/ | +0.572 |
Malf | +0.520 |
LW | +0.503 |
Nanto | +0.477 |
spection | +0.467 |
Neg logit contributions | |
subsequ | -0.486 |
therap | -0.459 |
upgr | -0.450 |
exhib | -0.449 |
momentum | -0.442 |
Token | L |
Token ID | 406 |
Feature activation | +2.99e-03 |
Pos logit contributions | |
subsequ | +1.451 |
therap | +1.360 |
exhib | +1.338 |
upgr | +1.311 |
ammy | +1.291 |
Neg logit contributions | |
"$:/ | -1.398 |
Malf | -1.291 |
LW | -1.266 |
Pyro | -1.183 |
Nanto | -1.152 |
Token | EL |
Token ID | 3698 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
"$:/ | +0.029 |
Malf | +0.026 |
LW | +0.026 |
Nanto | +0.023 |
Pyro | +0.023 |
Neg logit contributions | |
subsequ | -0.034 |
therap | -0.032 |
exhib | -0.032 |
upgr | -0.032 |
momentum | -0.031 |
Token | M |
Token ID | 337 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
subsequ | +4.026 |
exhib | +3.680 |
therap | +3.660 |
upgr | +3.640 |
momentum | +3.589 |
Neg logit contributions | |
"$:/ | -4.076 |
Malf | -3.628 |
LW | -3.590 |
Nanto | -3.335 |
Pyro | -3.312 |
Token | 2 |
Token ID | 17 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
subsequ | +8.875 |
therap | +8.292 |
exhib | +8.235 |
upgr | +8.190 |
momentum | +8.116 |
Neg logit contributions | |
"$:/ | -8.370 |
Malf | -7.220 |
LW | -7.007 |
Nanto | -6.820 |
Pyro | -6.490 |
Token | K |
Token ID | 42 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
subsequ | +1.889 |
therap | +1.783 |
exhib | +1.767 |
upgr | +1.753 |
momentum | +1.738 |
Neg logit contributions | |
"$:/ | -1.385 |
Malf | -1.181 |
LW | -1.146 |
Nanto | -1.094 |
Pyro | -1.045 |
Token | [ |
Token ID | 685 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
subsequ | +4.744 |
therap | +4.414 |
exhib | +4.368 |
upgr | +4.363 |
momentum | +4.280 |
Neg logit contributions | |
"$:/ | -4.781 |
Malf | -4.225 |
LW | -4.120 |
Nanto | -3.901 |
Pyro | -3.812 |
Token | 10 |
Token ID | 940 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
mosqu | +0.217 |
metic | +0.215 |
enthusi | +0.213 |
unden | +0.213 |
therap | +0.212 |
Neg logit contributions | |
"$:/ | -0.101 |
LW | -0.092 |
Malf | -0.078 |
lf | -0.075 |
Pyro | -0.071 |
Token | : |
Token ID | 25 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
therap | +2.256 |
enthusi | +2.208 |
mosqu | +2.205 |
unden | +2.154 |
subsequ | +2.150 |
Neg logit contributions | |
"$:/ | -1.506 |
LW | -1.323 |
Malf | -1.250 |
Pyro | -1.153 |
lf | -1.144 |
Token | Berger |
Token ID | 33337 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
"$:/ | +0.954 |
Malf | +0.853 |
LW | +0.824 |
Nanto | +0.795 |
Pyro | +0.775 |
Neg logit contributions | |
subsequ | -0.934 |
therap | -0.878 |
upgr | -0.857 |
exhib | -0.856 |
momentum | -0.833 |
Token | on |
Token ID | 261 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
subsequ | +0.856 |
therap | +0.814 |
momentum | +0.777 |
inertia | +0.775 |
exhib | +0.771 |
Neg logit contributions | |
"$:/ | -0.823 |
LW | -0.728 |
Malf | -0.720 |
Pyro | -0.677 |
Nanto | -0.676 |
Token | under |
Token ID | 739 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
"$:/ | +1.917 |
LW | +1.712 |
Malf | +1.686 |
Nanto | +1.616 |
Pyro | +1.589 |
Neg logit contributions | |
subsequ | -2.029 |
therap | -1.922 |
inertia | -1.829 |
momentum | -1.804 |
upgr | -1.796 |
Token | the |
Token ID | 262 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
subsequ | +0.689 |
therap | +0.644 |
exhib | +0.625 |
upgr | +0.620 |
momentum | +0.616 |
Neg logit contributions | |
"$:/ | -0.740 |
Malf | -0.653 |
LW | -0.644 |
Nanto | -0.619 |
Pyro | -0.597 |
Token | St |
Token ID | 520 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
"$:/ | +0.584 |
LW | +0.540 |
Malf | +0.534 |
Nanto | +0.506 |
Pyro | +0.501 |
Neg logit contributions | |
subsequ | -0.507 |
therap | -0.470 |
ammy | -0.452 |
upgr | -0.448 |
exhib | -0.446 |
Token | . |
Token ID | 13 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
subsequ | +2.098 |
therap | +1.954 |
exhib | +1.926 |
upgr | +1.908 |
momentum | +1.879 |
Neg logit contributions | |
"$:/ | -2.700 |
Malf | -2.415 |
LW | -2.357 |
Nanto | -2.291 |
Pyro | -2.218 |
Token | Phillips |
Token ID | 17630 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
subsequ | +8.660 |
therap | +8.028 |
exhib | +8.008 |
upgr | +7.980 |
momentum | +7.932 |
Neg logit contributions | |
"$:/ | -8.161 |
Malf | -6.910 |
LW | -6.749 |
Nanto | -6.597 |
spection | -6.253 |
Token | Bridge |
Token ID | 10290 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
subsequ | +2.422 |
therap | +2.283 |
exhib | +2.236 |
upgr | +2.194 |
inertia | +2.191 |
Neg logit contributions | |
"$:/ | -2.206 |
Malf | -1.950 |
LW | -1.927 |
Pyro | -1.802 |
Nanto | -1.801 |
Token | in |
Token ID | 287 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
subsequ | +0.880 |
therap | +0.828 |
exhib | +0.800 |
inertia | +0.794 |
momentum | +0.788 |
Neg logit contributions | |
"$:/ | -0.851 |
Malf | -0.766 |
LW | -0.754 |
Nanto | -0.728 |
Pyro | -0.705 |
Token | Weston |
Token ID | 46949 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
subsequ | +0.692 |
therap | +0.656 |
exhib | +0.639 |
upgr | +0.628 |
mosqu | +0.611 |
Neg logit contributions | |
"$:/ | -0.678 |
Malf | -0.611 |
LW | -0.603 |
Nanto | -0.588 |
Pyro | -0.566 |
Token | exploring |
Token ID | 13504 |
Feature activation | -4.40e-03 |
Pos logit contributions | |
subsequ | +0.228 |
therap | +0.218 |
exhib | +0.209 |
momentum | +0.206 |
ammy | +0.205 |
Neg logit contributions | |
"$:/ | -0.228 |
Malf | -0.205 |
LW | -0.202 |
Nanto | -0.196 |
Pyro | -0.189 |
Token | MAKE |
Token ID | 39134 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
subsequ | +1.489 |
therap | +1.432 |
mosqu | +1.392 |
exhib | +1.374 |
upgr | +1.361 |
Neg logit contributions | |
"$:/ | -1.131 |
LW | -1.019 |
Malf | -0.969 |
Pyro | -0.888 |
LG | -0.835 |
Token | THE |
Token ID | 3336 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
subsequ | +2.324 |
therap | +2.219 |
exhib | +2.085 |
mosqu | +2.076 |
ammy | +2.057 |
Neg logit contributions | |
"$:/ | -2.289 |
LW | -2.038 |
Malf | -2.020 |
Pyro | -1.844 |
Nanto | -1.777 |
Token | BE |
Token ID | 9348 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
"$:/ | +0.564 |
LW | +0.505 |
Malf | +0.492 |
Pyro | +0.443 |
LG | +0.416 |
Neg logit contributions | |
subsequ | -0.729 |
therap | -0.702 |
exhib | -0.668 |
ammy | -0.668 |
mosqu | -0.664 |
Token | ER |
Token ID | 1137 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
subsequ | +0.146 |
therap | +0.139 |
upgr | +0.133 |
exhib | +0.133 |
mosqu | +0.133 |
Neg logit contributions | |
"$:/ | -0.106 |
Malf | -0.094 |
LW | -0.091 |
Pyro | -0.081 |
Nanto | -0.080 |
Token | T |
Token ID | 309 |
Feature activation | -8.07e-02 |
Pos logit contributions | |
"$:/ | +1.977 |
LW | +1.796 |
Malf | +1.761 |
Pyro | +1.555 |
LG | +1.501 |
Neg logit contributions | |
subsequ | -2.583 |
therap | -2.456 |
mosqu | -2.391 |
exhib | -2.362 |
erosion | -2.346 |
Token | AS |
Token ID | 1921 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
subsequ | +1.023 |
therap | +0.963 |
exhib | +0.956 |
erosion | +0.940 |
upgr | +0.935 |
Neg logit contributions | |
"$:/ | -0.736 |
Malf | -0.636 |
LW | -0.617 |
Pyro | -0.554 |
Nanto | -0.551 |
Token | TE |
Token ID | 9328 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
subsequ | +8.138 |
therap | +7.557 |
exhib | +7.548 |
upgr | +7.518 |
momentum | +7.441 |
Neg logit contributions | |
"$:/ | -8.785 |
Malf | -7.627 |
LW | -7.397 |
Nanto | -7.279 |
spection | -6.919 |
Token | AS |
Token ID | 7054 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
subsequ | +0.767 |
therap | +0.716 |
exhib | +0.686 |
mosqu | +0.684 |
momentum | +0.676 |
Neg logit contributions | |
"$:/ | -0.639 |
LW | -0.564 |
Malf | -0.564 |
Pyro | -0.514 |
Nanto | -0.487 |
Token | N |
Token ID | 399 |
Feature activation | +8.22e-03 |
Pos logit contributions | |
"$:/ | +1.667 |
LW | +1.496 |
Malf | +1.449 |
Pyro | +1.327 |
LG | +1.242 |
Neg logit contributions | |
subsequ | -2.084 |
therap | -1.944 |
mosqu | -1.881 |
exhib | -1.872 |
upgr | -1.831 |
Token | ICE |
Token ID | 8476 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
"$:/ | +0.076 |
Malf | +0.066 |
LW | +0.064 |
Nanto | +0.058 |
Pyro | +0.058 |
Neg logit contributions | |
subsequ | -0.101 |
therap | -0.099 |
upgr | -0.096 |
exhib | -0.095 |
mosqu | -0.095 |
Token | AS |
Token ID | 7054 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
subsequ | +1.247 |
therap | +1.181 |
exhib | +1.142 |
upgr | +1.125 |
momentum | +1.122 |
Neg logit contributions | |
"$:/ | -0.940 |
Malf | -0.808 |
LW | -0.802 |
Nanto | -0.725 |
Pyro | -0.723 |
Token | of |
Token ID | 286 |
Feature activation | -4.18e-02 |
Pos logit contributions | |
subsequ | +0.134 |
therap | +0.126 |
upgr | +0.124 |
exhib | +0.123 |
mosqu | +0.120 |
Neg logit contributions | |
"$:/ | -0.136 |
LW | -0.120 |
Malf | -0.117 |
Nanto | -0.111 |
Pyro | -0.110 |
Token | Survivor |
Token ID | 23740 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
subsequ | +0.426 |
therap | +0.398 |
exhib | +0.392 |
upgr | +0.390 |
inertia | +0.380 |
Neg logit contributions | |
"$:/ | -0.462 |
LW | -0.426 |
Malf | -0.421 |
Pyro | -0.392 |
Nanto | -0.390 |
Token | San |
Token ID | 2986 |
Feature activation | -6.41e-03 |
Pos logit contributions | |
subsequ | +0.144 |
therap | +0.136 |
upgr | +0.135 |
exhib | +0.133 |
unden | +0.130 |
Neg logit contributions | |
"$:/ | -0.131 |
LW | -0.125 |
Malf | -0.124 |
Pyro | -0.114 |
Nanto | -0.111 |
Token | Juan |
Token ID | 16852 |
Feature activation | +9.21e-03 |
Pos logit contributions | |
subsequ | +0.079 |
therap | +0.075 |
exhib | +0.073 |
upgr | +0.072 |
inertia | +0.071 |
Neg logit contributions | |
"$:/ | -0.058 |
Malf | -0.053 |
LW | -0.051 |
Nanto | -0.046 |
Pyro | -0.046 |
Token | Del |
Token ID | 4216 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
"$:/ | +0.101 |
Malf | +0.090 |
LW | +0.088 |
Pyro | +0.081 |
spection | +0.079 |
Neg logit contributions | |
subsequ | -0.098 |
upgr | -0.091 |
therap | -0.088 |
ammy | -0.088 |
exhib | -0.087 |
Token | Sur |
Token ID | 4198 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
subsequ | +1.817 |
therap | +1.718 |
upgr | +1.692 |
exhib | +1.674 |
momentum | +1.647 |
Neg logit contributions | |
"$:/ | -1.546 |
Malf | -1.394 |
LW | -1.320 |
Nanto | -1.248 |
Pyro | -1.205 |
Token | featuring |
Token ID | 9593 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
subsequ | +7.789 |
therap | +7.198 |
exhib | +7.142 |
upgr | +7.115 |
momentum | +6.974 |
Neg logit contributions | |
"$:/ | -9.359 |
Malf | -8.288 |
LW | -8.122 |
Nanto | -7.858 |
Pyro | -7.563 |
Token | Coach |
Token ID | 16393 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
subsequ | +0.595 |
therap | +0.555 |
upgr | +0.546 |
exhib | +0.536 |
inertia | +0.530 |
Neg logit contributions | |
"$:/ | -0.622 |
Malf | -0.576 |
LW | -0.572 |
Pyro | -0.527 |
Nanto | -0.517 |
Token | ( |
Token ID | 357 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
subsequ | +3.045 |
therap | +2.859 |
upgr | +2.816 |
exhib | +2.809 |
momentum | +2.722 |
Neg logit contributions | |
"$:/ | -2.813 |
Malf | -2.498 |
LW | -2.429 |
Nanto | -2.314 |
Pyro | -2.264 |
Token | Surv |
Token ID | 34652 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
subsequ | +1.587 |
therap | +1.466 |
upgr | +1.438 |
exhib | +1.435 |
inertia | +1.411 |
Neg logit contributions | |
"$:/ | -1.252 |
Malf | -1.123 |
LW | -1.115 |
Pyro | -1.020 |
Nanto | -1.005 |
Token | iv |
Token ID | 452 |
Feature activation | +1.01e-03 |
Pos logit contributions | |
"$:/ | +0.170 |
LW | +0.158 |
Malf | +0.156 |
Pyro | +0.148 |
Nanto | +0.145 |
Neg logit contributions | |
subsequ | -0.166 |
therap | -0.157 |
exhib | -0.152 |
upgr | -0.150 |
mosqu | -0.147 |
Token | version |
Token ID | 2196 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
subsequ | +3.002 |
therap | +2.852 |
exhib | +2.759 |
erosion | +2.687 |
momentum | +2.673 |
Neg logit contributions | |
"$:/ | -3.695 |
Malf | -3.218 |
LW | -3.168 |
Nanto | -3.008 |
Pyro | -2.931 |
Token | before |
Token ID | 878 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
subsequ | +3.886 |
therap | +3.698 |
exhib | +3.612 |
upgr | +3.500 |
erosion | +3.498 |
Neg logit contributions | |
"$:/ | -4.431 |
Malf | -3.823 |
LW | -3.749 |
Nanto | -3.596 |
Pyro | -3.484 |
Token | { |
Token ID | 1391 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
subsequ | +2.278 |
therap | +2.192 |
exhib | +2.120 |
ammy | +2.068 |
momentum | +2.059 |
Neg logit contributions | |
"$:/ | -2.788 |
Malf | -2.364 |
LW | -2.361 |
Nanto | -2.258 |
spection | -2.165 |
Token | vers |
Token ID | 690 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
subsequ | +1.495 |
therap | +1.404 |
exhib | +1.394 |
upgr | +1.387 |
momentum | +1.375 |
Neg logit contributions | |
"$:/ | -1.228 |
Malf | -1.046 |
LW | -1.012 |
Nanto | -0.982 |
Pyro | -0.930 |
Token | } |
Token ID | 92 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
subsequ | +2.866 |
therap | +2.673 |
exhib | +2.672 |
upgr | +2.660 |
momentum | +2.652 |
Neg logit contributions | |
"$:/ | -2.448 |
Malf | -2.115 |
LW | -2.047 |
Nanto | -2.020 |
Pyro | -1.901 |
Token | vim |
Token ID | 43907 |
Feature activation | -7.92e-01 |
Pos logit contributions | |
subsequ | +3.699 |
therap | +3.623 |
exhib | +3.475 |
ammy | +3.466 |
erosion | +3.390 |
Neg logit contributions | |
"$:/ | -3.937 |
Malf | -3.363 |
LW | -3.322 |
Nanto | -3.134 |
Pyro | -3.051 |
Token | ={ |
Token ID | 34758 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
subsequ | +7.573 |
therap | +6.979 |
exhib | +6.941 |
upgr | +6.905 |
momentum | +6.825 |
Neg logit contributions | |
"$:/ | -9.248 |
Malf | -8.141 |
LW | -7.933 |
Nanto | -7.764 |
Pyro | -7.441 |
Token | vers |
Token ID | 690 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
subsequ | +1.228 |
therap | +1.140 |
exhib | +1.136 |
upgr | +1.130 |
momentum | +1.119 |
Neg logit contributions | |
"$:/ | -1.260 |
Malf | -1.097 |
LW | -1.066 |
Nanto | -1.041 |
Pyro | -0.991 |
Token | }: |
Token ID | 38362 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
subsequ | +4.121 |
exhib | +3.886 |
therap | +3.880 |
upgr | +3.869 |
momentum | +3.846 |
Neg logit contributions | |
"$:/ | -2.460 |
Malf | -2.045 |
LW | -1.957 |
Nanto | -1.903 |
Pyro | -1.773 |
Token | version |
Token ID | 2196 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
subsequ | +2.828 |
therap | +2.658 |
exhib | +2.602 |
upgr | +2.515 |
momentum | +2.514 |
Neg logit contributions | |
"$:/ | -3.766 |
Malf | -3.292 |
LW | -3.237 |
Nanto | -3.115 |
Pyro | -3.019 |
Token | { |
Token ID | 1391 |
Feature activation | -9.58e-02 |
Pos logit contributions | |
subsequ | +5.665 |
therap | +5.450 |
exhib | +5.305 |
ammy | +5.177 |
upgr | +5.170 |
Neg logit contributions | |
"$:/ | -6.485 |
Malf | -5.603 |
LW | -5.557 |
Nanto | -5.314 |
Pyro | -5.096 |
Token | =" |
Token ID | 2625 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
therap | +1.638 |
subsequ | +1.596 |
mosqu | +1.595 |
enthusi | +1.560 |
unden | +1.533 |
Neg logit contributions | |
"$:/ | -0.968 |
LW | -0.725 |
Malf | -0.693 |
Pyro | -0.646 |
lf | -0.626 |
Token | phone |
Token ID | 4862 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
subsequ | +3.321 |
therap | +3.107 |
exhib | +3.104 |
upgr | +3.088 |
momentum | +3.066 |
Neg logit contributions | |
"$:/ | -2.735 |
Malf | -2.354 |
LW | -2.280 |
Nanto | -2.216 |
Pyro | -2.087 |
Token | in |
Token ID | 287 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
subsequ | +0.424 |
momentum | +0.388 |
upgr | +0.387 |
exhib | +0.384 |
hol | +0.379 |
Neg logit contributions | |
"$:/ | -0.683 |
Malf | -0.627 |
Nanto | -0.596 |
LW | -0.595 |
spection | -0.569 |
Token | phones |
Token ID | 9512 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
subsequ | +0.243 |
therap | +0.226 |
exhib | +0.220 |
upgr | +0.220 |
momentum | +0.215 |
Neg logit contributions | |
"$:/ | -0.338 |
Malf | -0.293 |
LW | -0.290 |
Nanto | -0.280 |
Pyro | -0.272 |
Token | "> |
Token ID | 5320 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
subsequ | +4.469 |
momentum | +4.182 |
exhib | +4.138 |
upgr | +4.116 |
inertia | +4.033 |
Neg logit contributions | |
"$:/ | -7.300 |
Malf | -7.007 |
LW | -6.829 |
Nanto | -6.737 |
Pengu | -6.533 |
Token | < |
Token ID | 1279 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
subsequ | +2.820 |
momentum | +2.701 |
upgr | +2.677 |
therap | +2.668 |
erosion | +2.631 |
Neg logit contributions | |
"$:/ | -2.085 |
LW | -1.816 |
Malf | -1.803 |
Nanto | -1.716 |
spection | -1.598 |
Token | span |
Token ID | 12626 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
subsequ | +5.741 |
momentum | +5.542 |
piv | +5.250 |
upgr | +5.240 |
therap | +5.235 |
Neg logit contributions | |
"$:/ | -9.640 |
Malf | -8.690 |
LW | -8.607 |
Nanto | -8.606 |
Pyro | -7.798 |
Token | ng |
Token ID | 23370 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
"$:/ | +1.276 |
Malf | +1.087 |
LW | +1.082 |
Nanto | +1.014 |
Pyro | +1.000 |
Neg logit contributions | |
subsequ | -1.121 |
therap | -1.048 |
exhib | -1.027 |
upgr | -1.001 |
ammy | -0.997 |
Token | - |
Token ID | 12 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
therap | +1.792 |
subsequ | +1.772 |
mosqu | +1.678 |
exhib | +1.659 |
upgr | +1.626 |
Neg logit contributions | |
"$:/ | -1.938 |
Malf | -1.629 |
LW | -1.616 |
Pyro | -1.507 |
spection | -1.501 |
Token | bind |
Token ID | 21653 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
subsequ | +1.309 |
therap | +1.255 |
exhib | +1.219 |
erosion | +1.200 |
upgr | +1.198 |
Neg logit contributions | |
"$:/ | -1.135 |
Malf | -0.950 |
LW | -0.930 |
Nanto | -0.897 |
spection | -0.864 |
Token | =" |
Token ID | 2625 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
subsequ | +2.377 |
therap | +2.201 |
exhib | +2.188 |
upgr | +2.177 |
momentum | +2.158 |
Neg logit contributions | |
"$:/ | -2.662 |
Malf | -2.331 |
LW | -2.269 |
Nanto | -2.219 |
Pyro | -2.117 |
Token | 08 |
Token ID | 2919 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
"$:/ | +1.548 |
Malf | +1.271 |
LW | +1.249 |
Nanto | +1.197 |
spection | +1.148 |
Neg logit contributions | |
subsequ | -2.566 |
therap | -2.393 |
upgr | -2.392 |
momentum | -2.336 |
exhib | -2.328 |
Token | 71 |
Token ID | 4869 |
Feature activation | +8.63e-02 |
Pos logit contributions | |
subsequ | +0.189 |
therap | +0.174 |
exhib | +0.173 |
upgr | +0.172 |
momentum | +0.170 |
Neg logit contributions | |
"$:/ | -0.256 |
Malf | -0.227 |
LW | -0.221 |
Nanto | -0.216 |
Pyro | -0.208 |
Token | 231 |
Token ID | 34598 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
"$:/ | +0.784 |
LW | +0.711 |
Malf | +0.666 |
Pyro | +0.628 |
Nanto | +0.598 |
Neg logit contributions | |
subsequ | -1.071 |
therap | -1.029 |
mosqu | -0.981 |
upgr | -0.979 |
enthusi | -0.972 |
Token | 4 |
Token ID | 604 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
"$:/ | +0.573 |
LW | +0.498 |
Malf | +0.480 |
Pyro | +0.443 |
Nanto | +0.419 |
Neg logit contributions | |
subsequ | -0.781 |
therap | -0.778 |
mosqu | -0.738 |
enthusi | -0.730 |
ammy | -0.720 |
Token | 69 |
Token ID | 3388 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
"$:/ | +4.112 |
LW | +3.312 |
Malf | +3.259 |
Pyro | +2.951 |
lf | +2.892 |
Neg logit contributions | |
therap | -7.417 |
subsequ | -7.374 |
mosqu | -7.153 |
enthusi | -7.036 |
unden | -7.008 |
Token | ; |
Token ID | 26 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
"$:/ | +2.183 |
LW | +1.945 |
Malf | +1.873 |
Pyro | +1.709 |
Nanto | +1.628 |
Neg logit contributions | |
subsequ | -2.829 |
therap | -2.738 |
mosqu | -2.604 |
unden | -2.602 |
enthusi | -2.566 |
Token | th |
Token ID | 294 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
subsequ | +6.976 |
therap | +6.459 |
exhib | +6.453 |
upgr | +6.422 |
momentum | +6.356 |
Neg logit contributions | |
"$:/ | -9.249 |
Malf | -8.372 |
LW | -7.954 |
Nanto | -7.877 |
Pyro | -7.604 |
Token | om |
Token ID | 296 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
subsequ | +2.162 |
therap | +1.936 |
exhib | +1.921 |
upgr | +1.913 |
momentum | +1.884 |
Neg logit contributions | |
"$:/ | -4.089 |
Malf | -3.690 |
LW | -3.581 |
Nanto | -3.541 |
Pyro | -3.423 |
Token | son |
Token ID | 1559 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
subsequ | +4.927 |
ammy | +4.658 |
exhib | +4.488 |
upgr | +4.371 |
therap | +4.360 |
Neg logit contributions | |
"$:/ | -6.930 |
Malf | -6.498 |
Nanto | -6.042 |
LW | -5.972 |
Pyro | -5.867 |
Token | . |
Token ID | 13 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
subsequ | +1.447 |
therap | +1.339 |
exhib | +1.329 |
upgr | +1.320 |
momentum | +1.307 |
Neg logit contributions | |
"$:/ | -1.738 |
Malf | -1.526 |
LW | -1.486 |
Nanto | -1.452 |
Pyro | -1.391 |
Token | co |
Token ID | 1073 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
subsequ | +0.601 |
therap | +0.558 |
exhib | +0.550 |
upgr | +0.550 |
momentum | +0.545 |
Neg logit contributions | |
"$:/ | -0.642 |
Malf | -0.557 |
LW | -0.544 |
Nanto | -0.530 |
spection | -0.507 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
therap | +2.027 |
subsequ | +1.973 |
mosqu | +1.966 |
enthusi | +1.949 |
unden | +1.893 |
Neg logit contributions | |
"$:/ | -1.600 |
LW | -1.478 |
Malf | -1.405 |
Pyro | -1.352 |
lf | -1.266 |
Token | Heat |
Token ID | 12308 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
"$:/ | +0.786 |
Malf | +0.649 |
LW | +0.624 |
Pyro | +0.567 |
Nanto | +0.481 |
Neg logit contributions | |
subsequ | -2.399 |
therap | -2.336 |
exhib | -2.286 |
upgr | -2.225 |
enthusi | -2.218 |
Token | about |
Token ID | 546 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
subsequ | +6.965 |
therap | +6.795 |
enthusi | +6.510 |
mosqu | +6.410 |
traged | +6.350 |
Neg logit contributions | |
"$:/ | -5.862 |
Malf | -5.070 |
LW | -5.064 |
Pyro | -4.880 |
abouts | -4.539 |
Token | an |
Token ID | 281 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
therap | +6.339 |
subsequ | +6.318 |
mosqu | +5.996 |
traged | +5.936 |
enthusi | +5.909 |
Neg logit contributions | |
"$:/ | -5.484 |
LW | -4.716 |
Malf | -4.710 |
Nanto | -4.293 |
Pyro | -4.262 |
Token | inch |
Token ID | 11111 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
therap | +2.203 |
subsequ | +2.174 |
mosqu | +2.057 |
enthusi | +2.026 |
traged | +2.006 |
Neg logit contributions | |
"$:/ | -1.728 |
LW | -1.493 |
Malf | -1.405 |
LG | -1.308 |
spection | -1.276 |
Token | of |
Token ID | 286 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
therap | +4.346 |
enthusi | +4.291 |
mosqu | +4.278 |
subsequ | +4.257 |
unden | +4.210 |
Neg logit contributions | |
"$:/ | -3.507 |
LW | -2.928 |
Malf | -2.791 |
Pyro | -2.654 |
Nanto | -2.579 |
Token | oil |
Token ID | 3056 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
subsequ | +5.902 |
therap | +5.516 |
enthusi | +5.153 |
mosqu | +5.151 |
obser | +5.003 |
Neg logit contributions | |
"$:/ | -11.448 |
LW | -10.667 |
Malf | -10.365 |
Pyro | -9.986 |
Nanto | -9.652 |
Token | in |
Token ID | 287 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
therap | +5.269 |
subsequ | +5.217 |
mosqu | +5.127 |
enthusi | +5.124 |
metic | +5.009 |
Neg logit contributions | |
"$:/ | -3.742 |
LW | -3.165 |
Malf | -3.047 |
Pyro | -2.893 |
Nanto | -2.800 |
Token | a |
Token ID | 257 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
enthusi | +2.127 |
therap | +2.088 |
mosqu | +2.033 |
subsequ | +2.019 |
metic | +2.001 |
Neg logit contributions | |
"$:/ | -1.531 |
LW | -1.400 |
Pyro | -1.316 |
Malf | -1.299 |
LG | -1.185 |
Token | skillet |
Token ID | 41306 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
subsequ | +7.382 |
therap | +6.790 |
enthusi | +6.561 |
mosqu | +6.502 |
exhib | +6.496 |
Neg logit contributions | |
"$:/ | -6.572 |
LW | -5.964 |
Malf | -5.546 |
Pyro | -5.457 |
spection | -5.286 |
Token | over |
Token ID | 625 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
therap | +4.290 |
subsequ | +4.192 |
enthusi | +4.180 |
mosqu | +4.153 |
unden | +4.024 |
Neg logit contributions | |
"$:/ | -3.214 |
LW | -2.650 |
Pyro | -2.597 |
Malf | -2.549 |
spection | -2.301 |
Token | left |
Token ID | 1364 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
"$:/ | +0.616 |
Malf | +0.555 |
LW | +0.532 |
Nanto | +0.504 |
LG | +0.491 |
Neg logit contributions | |
subsequ | -0.489 |
therap | -0.463 |
mosqu | -0.452 |
artments | -0.435 |
traged | -0.431 |
Token | unable |
Token ID | 5906 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
subsequ | +1.142 |
therap | +1.072 |
upgr | +1.033 |
exhib | +1.009 |
mosqu | +1.008 |
Neg logit contributions | |
"$:/ | -1.167 |
Malf | -1.024 |
LW | -0.970 |
Nanto | -0.934 |
spection | -0.886 |
Token | to |
Token ID | 284 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
"$:/ | +0.561 |
Malf | +0.539 |
LW | +0.524 |
Pyro | +0.503 |
Pengu | +0.494 |
Neg logit contributions | |
subsequ | -0.271 |
therap | -0.266 |
momentum | -0.261 |
hol | -0.251 |
exhib | -0.250 |
Token | afford |
Token ID | 5368 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
subsequ | +2.431 |
therap | +2.296 |
mosqu | +2.276 |
enthusi | +2.261 |
unden | +2.224 |
Neg logit contributions | |
"$:/ | -2.380 |
LW | -2.015 |
Malf | -1.968 |
Nanto | -1.937 |
Pyro | -1.824 |
Token | life |
Token ID | 1204 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
subsequ | +3.794 |
exhib | +3.498 |
therap | +3.491 |
upgr | +3.435 |
ammy | +3.426 |
Neg logit contributions | |
"$:/ | -4.781 |
Malf | -4.239 |
LW | -4.169 |
Nanto | -4.020 |
Pyro | -3.828 |
Token | - |
Token ID | 12 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
subsequ | +3.346 |
exhib | +3.074 |
therap | +3.038 |
upgr | +2.962 |
momentum | +2.908 |
Neg logit contributions | |
"$:/ | -4.441 |
Malf | -3.924 |
LW | -3.855 |
Nanto | -3.755 |
spection | -3.569 |
Token | saving |
Token ID | 29336 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
subsequ | +7.182 |
therap | +6.658 |
upgr | +6.579 |
exhib | +6.573 |
momentum | +6.498 |
Neg logit contributions | |
"$:/ | -9.225 |
Malf | -8.134 |
LW | -7.893 |
Nanto | -7.747 |
Pyro | -7.452 |
Token | medication |
Token ID | 14103 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
"$:/ | +1.692 |
LW | +1.550 |
Malf | +1.520 |
Nanto | +1.418 |
Pyro | +1.365 |
Neg logit contributions | |
subsequ | -1.317 |
ammy | -1.304 |
mosqu | -1.252 |
exhib | -1.229 |
artments | -1.190 |
Token | , |
Token ID | 11 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
"$:/ | +1.086 |
Malf | +0.995 |
LW | +0.967 |
Nanto | +0.893 |
spection | +0.866 |
Neg logit contributions | |
subsequ | -1.224 |
ammy | -1.167 |
therap | -1.149 |
mosqu | -1.149 |
exhib | -1.137 |
Token | while |
Token ID | 981 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
"$:/ | +1.460 |
Malf | +1.338 |
LW | +1.303 |
Pyro | +1.173 |
Nanto | +1.163 |
Neg logit contributions | |
subsequ | -1.485 |
ammy | -1.429 |
therap | -1.410 |
mosqu | -1.392 |
exhib | -1.371 |
Token | at |
Token ID | 379 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
subsequ | +1.446 |
mosqu | +1.350 |
therap | +1.348 |
ammy | +1.342 |
upgr | +1.298 |
Neg logit contributions | |
"$:/ | -1.623 |
Malf | -1.516 |
LW | -1.481 |
Nanto | -1.340 |
Pyro | -1.327 |
Token | download |
Token ID | 4321 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
subsequ | +1.939 |
therap | +1.878 |
mosqu | +1.797 |
ammy | +1.794 |
sugg | +1.738 |
Neg logit contributions | |
"$:/ | -2.391 |
LW | -2.131 |
Malf | -2.092 |
Nanto | -2.047 |
Pyro | -1.997 |
Token | , |
Token ID | 11 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
subsequ | +2.011 |
therap | +1.908 |
exhib | +1.809 |
mosqu | +1.801 |
ammy | +1.783 |
Neg logit contributions | |
"$:/ | -2.177 |
Malf | -1.846 |
LW | -1.819 |
Nanto | -1.741 |
Pyro | -1.713 |
Token | it |
Token ID | 340 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
subsequ | +2.009 |
therap | +1.995 |
mosqu | +1.881 |
ammy | +1.852 |
exhib | +1.826 |
Neg logit contributions | |
"$:/ | -2.046 |
Malf | -1.899 |
LW | -1.858 |
Pyro | -1.714 |
Nanto | -1.689 |
Token | certainly |
Token ID | 3729 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
subsequ | +1.914 |
therap | +1.805 |
exhib | +1.722 |
upgr | +1.718 |
momentum | +1.706 |
Neg logit contributions | |
"$:/ | -2.354 |
Malf | -2.071 |
LW | -2.044 |
Nanto | -1.958 |
spection | -1.911 |
Token | holds |
Token ID | 6622 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
"$:/ | +1.393 |
Malf | +1.237 |
LW | +1.228 |
Nanto | +1.150 |
Pyro | +1.124 |
Neg logit contributions | |
subsequ | -1.160 |
therap | -1.103 |
mosqu | -1.088 |
pestic | -1.044 |
unden | -1.007 |
Token | up |
Token ID | 510 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
subsequ | +3.712 |
therap | +3.537 |
exhib | +3.454 |
upgr | +3.365 |
ammy | +3.345 |
Neg logit contributions | |
"$:/ | -4.255 |
Malf | -3.691 |
LW | -3.652 |
Nanto | -3.465 |
Pyro | -3.350 |
Token | the |
Token ID | 262 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
subsequ | +7.521 |
therap | +6.996 |
exhib | +6.889 |
upgr | +6.846 |
momentum | +6.718 |
Neg logit contributions | |
"$:/ | -9.134 |
Malf | -8.028 |
LW | -7.834 |
Nanto | -7.637 |
Pyro | -7.328 |
Token | best |
Token ID | 1266 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
subsequ | +3.086 |
therap | +3.042 |
mosqu | +2.924 |
exhib | +2.847 |
ammy | +2.837 |
Neg logit contributions | |
"$:/ | -4.088 |
Malf | -3.695 |
LW | -3.620 |
Nanto | -3.423 |
Pyro | -3.412 |
Token | , |
Token ID | 11 |
Feature activation | -8.74e-02 |
Pos logit contributions | |
therap | +2.320 |
subsequ | +2.265 |
mosqu | +2.217 |
ammy | +2.193 |
exhib | +2.165 |
Neg logit contributions | |
"$:/ | -2.441 |
Malf | -2.167 |
LW | -2.137 |
Nanto | -2.037 |
Pyro | -1.984 |
Token | but |
Token ID | 475 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
subsequ | +0.961 |
therap | +0.937 |
mosqu | +0.907 |
ammy | +0.889 |
exhib | +0.887 |
Neg logit contributions | |
"$:/ | -0.888 |
Malf | -0.780 |
LW | -0.769 |
Nanto | -0.743 |
Pyro | -0.718 |
Token | if |
Token ID | 611 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
therap | +2.216 |
subsequ | +2.177 |
mosqu | +2.082 |
ammy | +2.054 |
exhib | +2.015 |
Neg logit contributions | |
"$:/ | -2.108 |
Malf | -1.999 |
LW | -1.952 |
Pyro | -1.790 |
Nanto | -1.772 |
Token | . |
Token ID | 13 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
"$:/ | +0.937 |
LW | +0.824 |
Malf | +0.809 |
spection | +0.770 |
Nanto | +0.757 |
Neg logit contributions | |
therap | -1.009 |
subsequ | -1.008 |
exhib | -0.948 |
upgr | -0.938 |
ammy | -0.931 |
Token | ( |
Token ID | 357 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
"$:/ | +0.451 |
Malf | +0.419 |
LW | +0.411 |
Pyro | +0.395 |
Nanto | +0.366 |
Neg logit contributions | |
therap | -0.557 |
subsequ | -0.556 |
undermin | -0.516 |
ammy | -0.515 |
enthusi | -0.513 |
Token | A |
Token ID | 32 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
subsequ | +1.449 |
therap | +1.395 |
upgr | +1.344 |
erosion | +1.319 |
exhib | +1.302 |
Neg logit contributions | |
"$:/ | -1.161 |
Malf | -1.044 |
LW | -1.020 |
Nanto | -0.931 |
Pyro | -0.920 |
Token | ; |
Token ID | 26 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
subsequ | +3.211 |
therap | +3.024 |
upgr | +2.977 |
exhib | +2.963 |
momentum | +2.918 |
Neg logit contributions | |
"$:/ | -3.178 |
Malf | -2.761 |
LW | -2.702 |
Nanto | -2.607 |
Pyro | -2.505 |
Token | C |
Token ID | 327 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
"$:/ | +0.705 |
LW | +0.649 |
Malf | +0.642 |
Pyro | +0.588 |
Nanto | +0.574 |
Neg logit contributions | |
subsequ | -0.623 |
therap | -0.597 |
upgr | -0.580 |
ammy | -0.562 |
exhib | -0.545 |
Token | ; |
Token ID | 26 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
subsequ | +1.496 |
therap | +1.423 |
upgr | +1.392 |
exhib | +1.373 |
momentum | +1.360 |
Neg logit contributions | |
"$:/ | -1.438 |
Malf | -1.256 |
LW | -1.226 |
Nanto | -1.170 |
Pyro | -1.134 |
Token | E |
Token ID | 412 |
Feature activation | -5.78e-03 |
Pos logit contributions | |
subsequ | +7.399 |
therap | +6.879 |
exhib | +6.744 |
upgr | +6.737 |
momentum | +6.639 |
Neg logit contributions | |
"$:/ | -9.088 |
Malf | -8.031 |
LW | -7.865 |
Nanto | -7.631 |
Pyro | -7.347 |
Token | ; |
Token ID | 26 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
subsequ | +0.064 |
therap | +0.061 |
upgr | +0.059 |
exhib | +0.058 |
momentum | +0.058 |
Neg logit contributions | |
"$:/ | -0.059 |
Malf | -0.052 |
LW | -0.051 |
Nanto | -0.049 |
Pyro | -0.047 |
Token | G |
Token ID | 402 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
subsequ | +3.628 |
therap | +3.511 |
upgr | +3.357 |
exhib | +3.289 |
ammy | +3.287 |
Neg logit contributions | |
"$:/ | -3.453 |
Malf | -3.097 |
LW | -3.096 |
Nanto | -2.839 |
Pyro | -2.785 |
Token | ) |
Token ID | 8 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
subsequ | +2.208 |
exhib | +2.044 |
therap | +2.025 |
upgr | +2.009 |
momentum | +2.004 |
Neg logit contributions | |
"$:/ | -2.532 |
Malf | -2.210 |
LW | -2.143 |
Nanto | -2.101 |
Pyro | -2.006 |
Token | Acquisition |
Token ID | 44564 |
Feature activation | +5.63e-02 |
Pos logit contributions | |
"$:/ | +0.700 |
Malf | +0.670 |
LW | +0.653 |
Pyro | +0.621 |
Nanto | +0.596 |
Neg logit contributions | |
subsequ | -0.795 |
therap | -0.787 |
ammy | -0.737 |
upgr | -0.737 |
exhib | -0.731 |
Token | � |
Token ID | 250 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
"$:/ | +0.966 |
LW | +0.843 |
Malf | +0.810 |
Nanto | +0.755 |
spection | +0.747 |
Neg logit contributions | |
subsequ | -1.355 |
therap | -1.298 |
upgr | -1.245 |
chromos | -1.233 |
momentum | -1.230 |
Token | Res |
Token ID | 4965 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
subsequ | +2.237 |
therap | +2.081 |
exhib | +2.076 |
momentum | +2.036 |
upgr | +2.035 |
Neg logit contributions | |
"$:/ | -1.892 |
Malf | -1.665 |
LW | -1.582 |
Nanto | -1.531 |
Pyro | -1.486 |
Token | ign |
Token ID | 570 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
"$:/ | +3.118 |
Malf | +2.790 |
LW | +2.782 |
spection | +2.625 |
Nanto | +2.621 |
Neg logit contributions | |
subsequ | -5.366 |
therap | -5.186 |
inertia | -5.007 |
exhib | -4.946 |
upgr | -4.883 |
Token | ! |
Token ID | 0 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
subsequ | +4.541 |
therap | +4.180 |
exhib | +4.155 |
upgr | +4.112 |
momentum | +4.104 |
Neg logit contributions | |
"$:/ | -5.852 |
Malf | -5.121 |
LW | -5.035 |
Nanto | -4.918 |
Pyro | -4.703 |
Token | Res |
Token ID | 1874 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
subsequ | +1.455 |
therap | +1.370 |
exhib | +1.344 |
upgr | +1.328 |
momentum | +1.324 |
Neg logit contributions | |
"$:/ | -1.232 |
Malf | -1.105 |
LW | -1.032 |
Nanto | -1.007 |
Pyro | -0.975 |
Token | ign |
Token ID | 570 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
"$:/ | +2.723 |
Malf | +2.422 |
LW | +2.283 |
Nanto | +2.197 |
spection | +2.176 |
Neg logit contributions | |
subsequ | -3.318 |
therap | -3.238 |
upgr | -3.081 |
exhib | -3.069 |
sugg | -3.019 |
Token | ! |
Token ID | 0 |
Feature activation | -8.86e-03 |
Pos logit contributions | |
subsequ | +5.474 |
exhib | +4.978 |
momentum | +4.965 |
therap | +4.963 |
upgr | +4.853 |
Neg logit contributions | |
"$:/ | -10.306 |
LW | -8.961 |
Malf | -8.950 |
Nanto | -8.839 |
Pyro | -8.429 |
Token | � |
Token ID | 447 |
Feature activation | -6.72e-02 |
Pos logit contributions | |
subsequ | +0.096 |
therap | +0.090 |
exhib | +0.088 |
upgr | +0.087 |
momentum | +0.087 |
Neg logit contributions | |
"$:/ | -0.091 |
Malf | -0.083 |
LW | -0.078 |
Nanto | -0.076 |
Pyro | -0.074 |
Token | � |
Token ID | 251 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
subsequ | +0.772 |
therap | +0.732 |
exhib | +0.719 |
upgr | +0.706 |
ammy | +0.690 |
Neg logit contributions | |
"$:/ | -0.675 |
Malf | -0.599 |
LW | -0.599 |
spection | -0.541 |
Nanto | -0.539 |
Token | could |
Token ID | 714 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
subsequ | +3.368 |
therap | +3.071 |
exhib | +3.028 |
upgr | +2.942 |
inertia | +2.906 |
Neg logit contributions | |
"$:/ | -4.443 |
Malf | -4.020 |
LW | -3.903 |
Pyro | -3.746 |
Nanto | -3.722 |
Token | be |
Token ID | 307 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
"$:/ | +2.997 |
Malf | +2.590 |
LW | +2.543 |
Nanto | +2.379 |
Pyro | +2.375 |
Neg logit contributions | |
subsequ | -2.609 |
therap | -2.459 |
exhib | -2.375 |
mosqu | -2.346 |
upgr | -2.344 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
therap | +1.626 |
mosqu | +1.611 |
enthusi | +1.601 |
metic | +1.579 |
subsequ | +1.562 |
Neg logit contributions | |
"$:/ | -0.918 |
LW | -0.779 |
Malf | -0.736 |
Pyro | -0.669 |
lf | -0.661 |
Token | � |
Token ID | 447 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
subsequ | +2.542 |
therap | +2.432 |
mosqu | +2.315 |
erosion | +2.309 |
exhib | +2.271 |
Neg logit contributions | |
"$:/ | -1.989 |
Malf | -1.712 |
LW | -1.660 |
spection | -1.559 |
lf | -1.534 |
Token | � |
Token ID | 250 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
"$:/ | +0.545 |
Malf | +0.487 |
LW | +0.477 |
Nanto | +0.466 |
Pyro | +0.450 |
Neg logit contributions | |
subsequ | -0.336 |
therap | -0.306 |
exhib | -0.303 |
upgr | -0.300 |
momentum | -0.297 |
Token | T |
Token ID | 51 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
"$:/ | +1.243 |
Malf | +1.084 |
LW | +1.054 |
Pyro | +0.998 |
Nanto | +0.967 |
Neg logit contributions | |
subsequ | -1.473 |
therap | -1.363 |
exhib | -1.332 |
erosion | -1.311 |
upgr | -1.301 |
Token | il |
Token ID | 346 |
Feature activation | +6.97e-02 |
Pos logit contributions | |
"$:/ | +2.065 |
Malf | +1.864 |
LW | +1.782 |
Pyro | +1.676 |
spection | +1.649 |
Neg logit contributions | |
subsequ | -2.318 |
exhib | -2.172 |
therap | -2.150 |
momentum | -2.124 |
upgr | -2.108 |
Token | ap |
Token ID | 499 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
"$:/ | +0.728 |
Malf | +0.655 |
LW | +0.646 |
Nanto | +0.603 |
Pyro | +0.600 |
Neg logit contributions | |
subsequ | -0.734 |
therap | -0.707 |
exhib | -0.694 |
upgr | -0.666 |
mosqu | -0.658 |
Token | ia |
Token ID | 544 |
Feature activation | -9.16e-04 |
Pos logit contributions | |
subsequ | +8.222 |
therap | +7.667 |
exhib | +7.613 |
upgr | +7.573 |
momentum | +7.504 |
Neg logit contributions | |
"$:/ | -8.142 |
Malf | -7.053 |
LW | -6.843 |
Nanto | -6.661 |
Pyro | -6.359 |
Token | , |
Token ID | 11 |
Feature activation | -3.92e-02 |
Pos logit contributions | |
subsequ | +0.010 |
therap | +0.009 |
exhib | +0.009 |
upgr | +0.009 |
momentum | +0.009 |
Neg logit contributions | |
"$:/ | -0.010 |
Malf | -0.009 |
LW | -0.008 |
Nanto | -0.008 |
Pyro | -0.008 |
Token | that |
Token ID | 326 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
subsequ | +0.442 |
therap | +0.412 |
exhib | +0.402 |
upgr | +0.394 |
mosqu | +0.390 |
Neg logit contributions | |
"$:/ | -0.395 |
Malf | -0.361 |
LW | -0.355 |
Pyro | -0.326 |
Nanto | -0.318 |
Token | � |
Token ID | 447 |
Feature activation | +5.05e-03 |
Pos logit contributions | |
subsequ | +3.549 |
therap | +3.248 |
exhib | +3.231 |
upgr | +3.158 |
momentum | +3.111 |
Neg logit contributions | |
"$:/ | -4.449 |
Malf | -3.977 |
LW | -3.892 |
Nanto | -3.717 |
Pyro | -3.697 |
Token | � |
Token ID | 247 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
"$:/ | +0.052 |
LW | +0.049 |
Malf | +0.049 |
spection | +0.044 |
Pyro | +0.044 |
Neg logit contributions | |
subsequ | -0.055 |
therap | -0.053 |
exhib | -0.050 |
mosqu | -0.050 |
upgr | -0.049 |
Token | / |
Token ID | 14 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
therap | +1.238 |
subsequ | +1.205 |
enthusi | +1.174 |
metic | +1.174 |
reluct | +1.171 |
Neg logit contributions | |
"$:/ | -0.640 |
LW | -0.602 |
Malf | -0.512 |
lf | -0.499 |
Pyro | -0.483 |
Token | edd |
Token ID | 6048 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
"$:/ | +1.285 |
LW | +1.091 |
Malf | +1.056 |
lf | +1.021 |
915 | +0.993 |
Neg logit contributions | |
subsequ | -1.809 |
therap | -1.785 |
mosqu | -1.737 |
metic | -1.735 |
unden | -1.726 |
Token | 75 |
Token ID | 2425 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
subsequ | +2.231 |
therap | +2.208 |
upgr | +2.103 |
unden | +2.092 |
mosqu | +2.086 |
Neg logit contributions | |
"$:/ | -1.739 |
LW | -1.487 |
Malf | -1.476 |
lf | -1.346 |
spection | -1.313 |
Token | / |
Token ID | 14 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
"$:/ | +1.239 |
LW | +1.099 |
Malf | +1.072 |
lf | +1.011 |
Pyro | +0.963 |
Neg logit contributions | |
therap | -1.868 |
mosqu | -1.834 |
subsequ | -1.829 |
unden | -1.822 |
metic | -1.818 |
Token | AUT |
Token ID | 39371 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
"$:/ | +0.279 |
LW | +0.233 |
Malf | +0.226 |
lf | +0.198 |
Py | +0.180 |
Neg logit contributions | |
subsequ | -0.653 |
therap | -0.646 |
mosqu | -0.645 |
erosion | -0.624 |
traged | -0.622 |
Token | OC |
Token ID | 4503 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
subsequ | +6.634 |
therap | +6.199 |
exhib | +6.162 |
upgr | +6.137 |
momentum | +6.089 |
Neg logit contributions | |
"$:/ | -5.819 |
Malf | -4.992 |
LW | -4.779 |
Nanto | -4.728 |
Pyro | -4.450 |
Token | R |
Token ID | 49 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
subsequ | +6.441 |
exhib | +5.930 |
therap | +5.897 |
upgr | +5.866 |
momentum | +5.808 |
Neg logit contributions | |
"$:/ | -9.610 |
Malf | -8.478 |
Nanto | -8.315 |
LW | -8.199 |
Pyro | -7.812 |
Token | OP |
Token ID | 3185 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
"$:/ | +0.881 |
LW | +0.771 |
lf | +0.699 |
Malf | +0.692 |
Pyro | +0.648 |
Neg logit contributions | |
subsequ | -3.569 |
mosqu | -3.423 |
therap | -3.391 |
oshenko | -3.317 |
enthusi | -3.293 |
Token | / |
Token ID | 14 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
subsequ | +3.616 |
therap | +3.387 |
exhib | +3.237 |
mosqu | +3.221 |
unden | +3.214 |
Neg logit contributions | |
"$:/ | -3.318 |
Malf | -2.836 |
LW | -2.821 |
Pyro | -2.580 |
spection | -2.515 |
Token | h |
Token ID | 71 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
subsequ | +3.571 |
therap | +3.426 |
mosqu | +3.240 |
unden | +3.233 |
upgr | +3.153 |
Neg logit contributions | |
"$:/ | -2.974 |
Malf | -2.629 |
LW | -2.595 |
lf | -2.426 |
spection | -2.389 |
Token | 342 |
Token ID | 31575 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
subsequ | +5.941 |
therap | +5.599 |
exhib | +5.478 |
erosion | +5.459 |
upgr | +5.459 |
Neg logit contributions | |
"$:/ | -4.484 |
Malf | -4.096 |
LW | -3.897 |
lf | -3.639 |
Nanto | -3.588 |
Token | 000 |
Token ID | 830 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
"$:/ | +0.094 |
Malf | +0.082 |
LW | +0.080 |
Nanto | +0.079 |
Pyro | +0.075 |
Neg logit contributions | |
subsequ | -0.075 |
exhib | -0.069 |
therap | -0.069 |
momentum | -0.069 |
upgr | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -4.93e-02 |
Pos logit contributions | |
subsequ | +1.150 |
therap | +1.077 |
upgr | +1.050 |
exhib | +1.039 |
mosqu | +1.020 |
Neg logit contributions | |
"$:/ | -1.187 |
Malf | -1.041 |
LW | -1.031 |
Nanto | -0.979 |
Pyro | -0.949 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
subsequ | +0.647 |
therap | +0.629 |
mosqu | +0.593 |
enthusi | +0.585 |
obser | +0.580 |
Neg logit contributions | |
"$:/ | -0.490 |
Malf | -0.426 |
LW | -0.425 |
Pyro | -0.392 |
Nanto | -0.387 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
therap | +1.689 |
enthusi | +1.655 |
mosqu | +1.636 |
unden | +1.610 |
metic | +1.610 |
Neg logit contributions | |
"$:/ | -0.954 |
LW | -0.800 |
Malf | -0.753 |
Pyro | -0.706 |
lf | -0.677 |
Token | Th |
Token ID | 817 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
subsequ | +1.681 |
therap | +1.622 |
exhib | +1.517 |
mosqu | +1.514 |
upgr | +1.502 |
Neg logit contributions | |
"$:/ | -1.417 |
Malf | -1.246 |
LW | -1.203 |
spection | -1.128 |
Pyro | -1.102 |
Token | ur |
Token ID | 333 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
"$:/ | +0.615 |
Malf | +0.571 |
LW | +0.556 |
Nanto | +0.514 |
Pyro | +0.506 |
Neg logit contributions | |
subsequ | -0.638 |
therap | -0.629 |
exhib | -0.605 |
momentum | -0.598 |
inertia | -0.598 |
Token | rock |
Token ID | 10823 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
subsequ | +8.865 |
therap | +8.308 |
exhib | +8.263 |
upgr | +8.224 |
momentum | +8.160 |
Neg logit contributions | |
"$:/ | -7.257 |
Malf | -6.173 |
LW | -5.985 |
Nanto | -5.814 |
Pyro | -5.488 |
Token | Council |
Token ID | 4281 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
subsequ | +1.885 |
therap | +1.778 |
exhib | +1.718 |
ammy | +1.709 |
momentum | +1.698 |
Neg logit contributions | |
"$:/ | -1.915 |
Malf | -1.761 |
LW | -1.711 |
Nanto | -1.627 |
Pyro | -1.594 |
Token | said |
Token ID | 531 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
subsequ | +1.992 |
therap | +1.938 |
exhib | +1.849 |
momentum | +1.821 |
inertia | +1.797 |
Neg logit contributions | |
"$:/ | -2.162 |
Malf | -1.932 |
LW | -1.915 |
Nanto | -1.841 |
spection | -1.804 |
Token | workers |
Token ID | 3259 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
subsequ | +2.195 |
therap | +2.106 |
ammy | +2.015 |
upgr | +1.986 |
exhib | +1.985 |
Neg logit contributions | |
"$:/ | -2.197 |
Malf | -2.026 |
LW | -1.973 |
Pyro | -1.817 |
spection | -1.807 |
Token | had |
Token ID | 550 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
subsequ | +1.923 |
therap | +1.839 |
unden | +1.757 |
exhib | +1.750 |
mosqu | +1.740 |
Neg logit contributions | |
"$:/ | -2.193 |
Malf | -1.927 |
LW | -1.873 |
Pyro | -1.781 |
Nanto | -1.746 |
Token | your |
Token ID | 534 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
subsequ | +0.772 |
ammy | +0.708 |
therap | +0.702 |
upgr | +0.693 |
exhib | +0.687 |
Neg logit contributions | |
"$:/ | -0.884 |
Malf | -0.816 |
LW | -0.792 |
Pyro | -0.754 |
Nanto | -0.746 |
Token | hands |
Token ID | 2832 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
"$:/ | +0.367 |
LW | +0.330 |
Malf | +0.330 |
Nanto | +0.311 |
Pyro | +0.309 |
Neg logit contributions | |
subsequ | -0.278 |
ammy | -0.258 |
therap | -0.254 |
upgr | -0.251 |
unden | -0.247 |
Token | before |
Token ID | 878 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
"$:/ | +1.766 |
LW | +1.534 |
Malf | +1.528 |
Pyro | +1.401 |
Nanto | +1.401 |
Neg logit contributions | |
subsequ | -1.900 |
therap | -1.800 |
upgr | -1.781 |
mosqu | -1.763 |
exhib | -1.758 |
Token | this |
Token ID | 428 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
"$:/ | +0.402 |
LW | +0.371 |
Malf | +0.365 |
Nanto | +0.346 |
Pyro | +0.336 |
Neg logit contributions | |
subsequ | -0.379 |
therap | -0.363 |
ammy | -0.361 |
upgr | -0.358 |
exhib | -0.348 |
Token | fish |
Token ID | 5916 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
subsequ | +1.329 |
therap | +1.229 |
exhib | +1.200 |
upgr | +1.192 |
ammy | +1.190 |
Neg logit contributions | |
"$:/ | -1.613 |
Malf | -1.485 |
LW | -1.471 |
Nanto | -1.382 |
Pyro | -1.370 |
Token | runs |
Token ID | 4539 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
subsequ | +0.103 |
therap | +0.097 |
upgr | +0.094 |
exhib | +0.093 |
unden | +0.093 |
Neg logit contributions | |
"$:/ | -0.123 |
Malf | -0.112 |
LW | -0.109 |
Nanto | -0.103 |
spection | -0.101 |
Token | out |
Token ID | 503 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
"$:/ | +4.069 |
Malf | +3.601 |
LW | +3.571 |
Pyro | +3.389 |
Nanto | +3.236 |
Neg logit contributions | |
subsequ | -4.898 |
mosqu | -4.416 |
upgr | -4.402 |
therap | -4.364 |
obser | -4.278 |
Token | of |
Token ID | 286 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
"$:/ | +3.637 |
Malf | +3.055 |
LW | +3.038 |
Nanto | +2.907 |
spection | +2.833 |
Neg logit contributions | |
subsequ | -3.577 |
upgr | -3.332 |
therap | -3.289 |
exhib | -3.275 |
erosion | -3.139 |
Token | water |
Token ID | 1660 |
Feature activation | -1.83e-03 |
Pos logit contributions | |
"$:/ | +1.911 |
LW | +1.827 |
Malf | +1.733 |
Nanto | +1.653 |
Pyro | +1.641 |
Neg logit contributions | |
subsequ | -1.429 |
ammy | -1.339 |
therap | -1.314 |
upgr | -1.284 |
exhib | -1.283 |
Token | ? |
Token ID | 30 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
subsequ | +0.022 |
exhib | +0.020 |
therap | +0.020 |
upgr | +0.020 |
ammy | +0.019 |
Neg logit contributions | |
"$:/ | -0.017 |
Malf | -0.015 |
LW | -0.015 |
Nanto | -0.014 |
spection | -0.014 |
Token | For |
Token ID | 1114 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
subsequ | +2.481 |
therap | +2.296 |
exhib | +2.250 |
upgr | +2.216 |
mosqu | +2.202 |
Neg logit contributions | |
"$:/ | -1.972 |
LW | -1.794 |
Malf | -1.791 |
Nanto | -1.638 |
Pyro | -1.636 |
Token | B |
Token ID | 347 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
subsequ | +0.198 |
exhib | +0.182 |
momentum | +0.180 |
therap | +0.174 |
upgr | +0.173 |
Neg logit contributions | |
"$:/ | -0.335 |
Malf | -0.291 |
LW | -0.285 |
Nanto | -0.281 |
spection | -0.274 |
Token | ao |
Token ID | 5488 |
Feature activation | +9.43e-02 |
Pos logit contributions | |
mosqu | +0.834 |
enthusi | +0.828 |
therap | +0.825 |
subsequ | +0.808 |
metic | +0.803 |
Neg logit contributions | |
"$:/ | -0.270 |
LW | -0.221 |
Malf | -0.200 |
Nanto | -0.192 |
Pyro | -0.186 |
Token | Shan |
Token ID | 14866 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
"$:/ | +0.924 |
Malf | +0.786 |
LW | +0.764 |
Nanto | +0.724 |
spection | +0.702 |
Neg logit contributions | |
subsequ | -1.079 |
exhib | -1.023 |
momentum | -1.016 |
therap | -1.010 |
upgr | -1.002 |
Token | , |
Token ID | 11 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
"$:/ | +0.917 |
Malf | +0.841 |
LW | +0.816 |
Nanto | +0.777 |
spection | +0.752 |
Neg logit contributions | |
subsequ | -0.871 |
therap | -0.816 |
exhib | -0.796 |
upgr | -0.778 |
momentum | -0.773 |
Token | one |
Token ID | 530 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
"$:/ | +1.866 |
Malf | +1.773 |
LW | +1.718 |
Nanto | +1.553 |
Haz | +1.498 |
Neg logit contributions | |
subsequ | -2.344 |
therap | -2.313 |
ammy | -2.209 |
enthusi | -2.156 |
mosqu | -2.143 |
Token | of |
Token ID | 286 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
"$:/ | +4.586 |
Malf | +4.069 |
LW | +3.968 |
Nanto | +3.878 |
Pyro | +3.715 |
Neg logit contributions | |
subsequ | -3.408 |
therap | -3.142 |
exhib | -3.109 |
upgr | -3.077 |
momentum | -3.031 |
Token | China |
Token ID | 2807 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
"$:/ | +4.717 |
Malf | +4.250 |
LW | +4.201 |
Nanto | +3.858 |
spection | +3.658 |
Neg logit contributions | |
subsequ | -5.529 |
therap | -5.328 |
enthusi | -5.222 |
upgr | -5.112 |
mosqu | -5.078 |
Token | 's |
Token ID | 338 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
subsequ | +0.482 |
therap | +0.471 |
enthusi | +0.463 |
mosqu | +0.455 |
unden | +0.451 |
Neg logit contributions | |
"$:/ | -0.500 |
Malf | -0.444 |
LW | -0.440 |
Pyro | -0.411 |
spection | -0.399 |
Token | largest |
Token ID | 4387 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
"$:/ | +2.288 |
Malf | +2.150 |
LW | +2.085 |
Pyro | +1.873 |
Haz | +1.858 |
Neg logit contributions | |
subsequ | -1.986 |
ammy | -1.907 |
enthusi | -1.875 |
therap | -1.870 |
artments | -1.855 |
Token | mine |
Token ID | 6164 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
Malf | +0.784 |
"$:/ | +0.777 |
LW | +0.770 |
Nanto | +0.691 |
Haz | +0.682 |
Neg logit contributions | |
subsequ | -0.698 |
ammy | -0.692 |
therap | -0.658 |
artments | -0.653 |
enthusi | -0.648 |
Token | that |
Token ID | 326 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
"$:/ | +0.491 |
Malf | +0.471 |
LW | +0.448 |
Nanto | +0.425 |
spection | +0.424 |
Neg logit contributions | |
subsequ | -0.418 |
therap | -0.387 |
ammy | -0.384 |
unden | -0.367 |
momentum | -0.365 |
Token | , |
Token ID | 11 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
"$:/ | +0.740 |
Malf | +0.666 |
LW | +0.644 |
spection | +0.608 |
Pyro | +0.603 |
Neg logit contributions | |
subsequ | -0.798 |
therap | -0.783 |
upgr | -0.764 |
exhib | -0.758 |
unden | -0.755 |
Token | because |
Token ID | 780 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
"$:/ | +1.123 |
Malf | +1.061 |
LW | +0.986 |
Pyro | +0.930 |
Nanto | +0.882 |
Neg logit contributions | |
subsequ | -1.295 |
therap | -1.283 |
ammy | -1.253 |
upgr | -1.253 |
exhib | -1.194 |
Token | something |
Token ID | 1223 |
Feature activation | +9.49e-02 |
Pos logit contributions | |
"$:/ | +3.221 |
Malf | +3.108 |
LW | +2.974 |
Pyro | +2.740 |
Nanto | +2.591 |
Neg logit contributions | |
therap | -3.757 |
subsequ | -3.707 |
ammy | -3.655 |
upgr | -3.613 |
mosqu | -3.509 |
Token | good |
Token ID | 922 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
"$:/ | +1.039 |
Malf | +0.967 |
LW | +0.926 |
Pyro | +0.866 |
Nanto | +0.829 |
Neg logit contributions | |
subsequ | -0.943 |
ammy | -0.872 |
upgr | -0.867 |
therap | -0.861 |
unden | -0.856 |
Token | or |
Token ID | 393 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
"$:/ | +0.308 |
Malf | +0.275 |
LW | +0.265 |
Nanto | +0.247 |
Pyro | +0.246 |
Neg logit contributions | |
subsequ | -0.327 |
therap | -0.311 |
unden | -0.302 |
upgr | -0.299 |
ammy | -0.297 |
Token | terrible |
Token ID | 7818 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
"$:/ | +1.531 |
Malf | +1.377 |
LW | +1.349 |
Nanto | +1.290 |
Pyro | +1.261 |
Neg logit contributions | |
subsequ | -1.063 |
therap | -0.977 |
exhib | -0.966 |
upgr | -0.966 |
erosion | -0.930 |
Token | might |
Token ID | 1244 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
"$:/ | +3.785 |
Malf | +3.472 |
LW | +3.252 |
Pyro | +3.007 |
abouts | +2.961 |
Neg logit contributions | |
subsequ | -4.560 |
pestic | -4.349 |
therap | -4.307 |
unden | -4.303 |
ammy | -4.243 |
Token | come |
Token ID | 1282 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
"$:/ | +3.036 |
Malf | +2.610 |
LW | +2.491 |
Pyro | +2.359 |
lf | +2.266 |
Neg logit contributions | |
subsequ | -3.071 |
mosqu | -3.059 |
therap | -3.034 |
unden | -3.014 |
pestic | -2.953 |
Token | . |
Token ID | 13 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
"$:/ | +0.655 |
Malf | +0.567 |
LW | +0.529 |
Pyro | +0.512 |
abouts | +0.509 |
Neg logit contributions | |
subsequ | -0.721 |
therap | -0.694 |
mosqu | -0.690 |
exhib | -0.671 |
pestic | -0.669 |
Token | For |
Token ID | 1114 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
"$:/ | +1.949 |
Malf | +1.907 |
LW | +1.798 |
Pyro | +1.651 |
Nanto | +1.575 |
Neg logit contributions | |
therap | -3.007 |
subsequ | -2.942 |
mosqu | -2.871 |
satell | -2.852 |
enthusi | -2.778 |
Token | your |
Token ID | 534 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
"$:/ | +0.821 |
Malf | +0.779 |
LW | +0.770 |
Pyro | +0.705 |
spection | +0.694 |
Neg logit contributions | |
subsequ | -0.678 |
therap | -0.660 |
mosqu | -0.636 |
upgr | -0.628 |
unden | -0.617 |
Token | doing |
Token ID | 1804 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
subsequ | +0.913 |
therap | +0.868 |
mosqu | +0.833 |
exhib | +0.827 |
momentum | +0.824 |
Neg logit contributions | |
"$:/ | -0.931 |
Malf | -0.831 |
LW | -0.814 |
Pyro | -0.785 |
Nanto | -0.772 |
Token | the |
Token ID | 262 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
"$:/ | +1.494 |
LW | +1.336 |
Malf | +1.326 |
Pyro | +1.256 |
Nanto | +1.185 |
Neg logit contributions | |
subsequ | -1.469 |
therap | -1.399 |
ammy | -1.313 |
mosqu | -1.302 |
momentum | -1.286 |
Token | pics |
Token ID | 27618 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
"$:/ | +0.297 |
Malf | +0.271 |
LW | +0.268 |
Pyro | +0.257 |
Nanto | +0.244 |
Neg logit contributions | |
subsequ | -0.260 |
therap | -0.239 |
ammy | -0.229 |
mosqu | -0.227 |
artments | -0.226 |
Token | but |
Token ID | 475 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
"$:/ | +1.298 |
Malf | +1.099 |
LW | +1.094 |
spection | +1.043 |
Pyro | +1.036 |
Neg logit contributions | |
subsequ | -1.429 |
therap | -1.336 |
mosqu | -1.322 |
unden | -1.285 |
upgr | -1.256 |
Token | what |
Token ID | 644 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
"$:/ | +1.819 |
LW | +1.676 |
Malf | +1.621 |
Pyro | +1.553 |
LG | +1.477 |
Neg logit contributions | |
subsequ | -1.950 |
therap | -1.919 |
mosqu | -1.890 |
ammy | -1.825 |
artments | -1.795 |
Token | do |
Token ID | 466 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
subsequ | +2.028 |
therap | +1.949 |
mosqu | +1.899 |
unden | +1.845 |
pestic | +1.839 |
Neg logit contributions | |
"$:/ | -1.783 |
LW | -1.577 |
Malf | -1.561 |
Pyro | -1.514 |
Nanto | -1.435 |
Token | you |
Token ID | 345 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
"$:/ | +4.271 |
LW | +3.671 |
Malf | +3.554 |
Pyro | +3.458 |
Nanto | +3.266 |
Neg logit contributions | |
subsequ | -5.921 |
therap | -5.609 |
mosqu | -5.491 |
ammy | -5.379 |
erosion | -5.319 |
Token | expect |
Token ID | 1607 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
"$:/ | +1.519 |
Malf | +1.324 |
LW | +1.314 |
Pyro | +1.245 |
LG | +1.188 |
Neg logit contributions | |
mosqu | -1.629 |
subsequ | -1.572 |
therap | -1.541 |
unden | -1.507 |
enthusi | -1.506 |
Token | using |
Token ID | 1262 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
subsequ | +1.626 |
therap | +1.614 |
mosqu | +1.536 |
ammy | +1.519 |
upgr | +1.476 |
Neg logit contributions | |
"$:/ | -1.604 |
LW | -1.422 |
Malf | -1.417 |
Pyro | -1.332 |
Nanto | -1.247 |
Token | paint |
Token ID | 7521 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
"$:/ | +3.247 |
LW | +2.866 |
Malf | +2.842 |
Pyro | +2.752 |
LG | +2.617 |
Neg logit contributions | |
subsequ | -3.233 |
therap | -2.950 |
ammy | -2.913 |
mosqu | -2.888 |
artments | -2.813 |
Token | Kind |
Token ID | 35854 |
Feature activation | -8.17e-02 |
Pos logit contributions | |
"$:/ | +0.149 |
Malf | +0.127 |
LW | +0.126 |
Pyro | +0.122 |
spection | +0.119 |
Neg logit contributions | |
subsequ | -0.158 |
therap | -0.146 |
mosqu | -0.141 |
exhib | -0.140 |
chromos | -0.139 |
Token | risk |
Token ID | 2526 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
"$:/ | +1.226 |
Malf | +1.133 |
LW | +1.125 |
Nanto | +1.087 |
Pyro | +1.066 |
Neg logit contributions | |
subsequ | -0.962 |
upgr | -0.877 |
exhib | -0.853 |
artments | -0.853 |
therap | -0.851 |
Token | - |
Token ID | 12 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
"$:/ | +1.493 |
Malf | +1.291 |
LW | +1.283 |
Nanto | +1.269 |
Pyro | +1.220 |
Neg logit contributions | |
subsequ | -1.347 |
therap | -1.209 |
upgr | -1.184 |
momentum | -1.173 |
exhib | -1.162 |
Token | taking |
Token ID | 26103 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
"$:/ | +2.783 |
Malf | +2.356 |
LW | +2.316 |
Nanto | +2.274 |
spection | +2.120 |
Neg logit contributions | |
subsequ | -3.963 |
therap | -3.607 |
upgr | -3.563 |
mosqu | -3.525 |
momentum | -3.521 |
Token | aut |
Token ID | 1960 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
"$:/ | +2.152 |
Malf | +2.076 |
LW | +2.051 |
Pyro | +1.921 |
Nanto | +1.890 |
Neg logit contributions | |
subsequ | -1.801 |
artments | -1.681 |
ammy | -1.666 |
therap | -1.665 |
upgr | -1.633 |
Token | ocrat |
Token ID | 35128 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
"$:/ | +3.974 |
LW | +3.623 |
Malf | +3.560 |
Pyro | +3.260 |
Nanto | +3.237 |
Neg logit contributions | |
subsequ | -5.044 |
therap | -4.854 |
momentum | -4.643 |
exhib | -4.633 |
upgr | -4.566 |
Token | governed |
Token ID | 21825 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
"$:/ | +1.429 |
Malf | +1.284 |
LW | +1.247 |
Nanto | +1.186 |
Pyro | +1.151 |
Neg logit contributions | |
subsequ | -1.464 |
therap | -1.369 |
upgr | -1.334 |
exhib | -1.333 |
ammy | -1.295 |
Token | by |
Token ID | 416 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
"$:/ | +4.446 |
Malf | +3.818 |
LW | +3.711 |
Nanto | +3.493 |
Pyro | +3.407 |
Neg logit contributions | |
subsequ | -5.321 |
therap | -4.884 |
ammy | -4.799 |
upgr | -4.797 |
exhib | -4.743 |
Token | real |
Token ID | 1103 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
"$:/ | +1.348 |
Malf | +1.242 |
LW | +1.197 |
Nanto | +1.164 |
Pyro | +1.122 |
Neg logit contributions | |
subsequ | -1.125 |
therap | -1.033 |
ammy | -1.007 |
exhib | -0.994 |
upgr | -0.989 |
Token | polit |
Token ID | 34470 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
"$:/ | +2.045 |
Malf | +1.858 |
LW | +1.856 |
spection | +1.767 |
Nanto | +1.720 |
Neg logit contributions | |
subsequ | -1.731 |
therap | -1.594 |
ammy | -1.555 |
exhib | -1.540 |
mosqu | -1.530 |
Token | ik |
Token ID | 1134 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
"$:/ | +0.260 |
LW | +0.223 |
Malf | +0.193 |
Nanto | +0.187 |
lf | +0.186 |
Neg logit contributions | |
therap | -0.633 |
mosqu | -0.629 |
subsequ | -0.627 |
enthusi | -0.624 |
unden | -0.612 |
Token | even |
Token ID | 772 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
"$:/ | +1.012 |
Malf | +0.924 |
LW | +0.903 |
Nanto | +0.858 |
spection | +0.839 |
Neg logit contributions | |
subsequ | -0.997 |
therap | -0.914 |
exhib | -0.899 |
ammy | -0.887 |
mosqu | -0.880 |
Token | steps |
Token ID | 20214 |
Feature activation | +5.97e-03 |
Pos logit contributions | |
subsequ | +0.857 |
therap | +0.789 |
upgr | +0.767 |
exhib | +0.767 |
ammy | +0.754 |
Neg logit contributions | |
"$:/ | -1.551 |
Malf | -1.403 |
LW | -1.361 |
Nanto | -1.324 |
Pyro | -1.291 |
Token | — |
Token ID | 960 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
"$:/ | +0.065 |
Malf | +0.058 |
LW | +0.056 |
spection | +0.053 |
Nanto | +0.053 |
Neg logit contributions | |
subsequ | -0.063 |
therap | -0.058 |
exhib | -0.057 |
upgr | -0.056 |
mosqu | -0.055 |
Token | ref |
Token ID | 5420 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
"$:/ | +0.320 |
Malf | +0.298 |
LW | +0.280 |
spection | +0.262 |
Pyro | +0.254 |
Neg logit contributions | |
subsequ | -0.423 |
therap | -0.394 |
exhib | -0.380 |
upgr | -0.376 |
mosqu | -0.372 |
Token | uge |
Token ID | 2217 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
"$:/ | +1.335 |
Malf | +1.180 |
LW | +1.171 |
Nanto | +1.106 |
spection | +1.069 |
Neg logit contributions | |
subsequ | -1.353 |
therap | -1.310 |
exhib | -1.263 |
upgr | -1.218 |
mosqu | -1.211 |
Token | e |
Token ID | 68 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
subsequ | +1.128 |
therap | +1.022 |
exhib | +1.009 |
upgr | +1.002 |
momentum | +0.970 |
Neg logit contributions | |
"$:/ | -1.616 |
Malf | -1.439 |
LW | -1.429 |
Nanto | -1.368 |
Pyro | -1.343 |
Token | integration |
Token ID | 11812 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
subsequ | +1.206 |
therap | +1.089 |
exhib | +1.064 |
upgr | +1.035 |
ammy | +1.022 |
Neg logit contributions | |
"$:/ | -1.564 |
LW | -1.445 |
Malf | -1.424 |
Nanto | -1.325 |
Pyro | -1.308 |
Token | , |
Token ID | 11 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
"$:/ | +2.447 |
LW | +2.168 |
Malf | +2.155 |
spection | +2.012 |
Nanto | +1.993 |
Neg logit contributions | |
subsequ | -2.562 |
exhib | -2.364 |
therap | -2.330 |
ammy | -2.285 |
mosqu | -2.279 |
Token | climate |
Token ID | 4258 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
"$:/ | +0.150 |
Malf | +0.139 |
LW | +0.136 |
Pyro | +0.127 |
Nanto | +0.126 |
Neg logit contributions | |
subsequ | -0.108 |
therap | -0.103 |
ammy | -0.101 |
exhib | -0.099 |
upgr | -0.095 |
Token | change |
Token ID | 1487 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
subsequ | +0.329 |
exhib | +0.295 |
therap | +0.293 |
ammy | +0.284 |
upgr | +0.279 |
Neg logit contributions | |
"$:/ | -0.387 |
Malf | -0.349 |
LW | -0.345 |
Nanto | -0.327 |
spection | -0.316 |
Token | adaptation |
Token ID | 16711 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
"$:/ | +1.303 |
Malf | +1.160 |
LW | +1.136 |
Nanto | +1.100 |
spection | +1.058 |
Neg logit contributions | |
subsequ | -1.076 |
exhib | -0.981 |
therap | -0.980 |
upgr | -0.964 |
momentum | -0.943 |
Token | , |
Token ID | 11 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
"$:/ | +1.984 |
Malf | +1.749 |
LW | +1.706 |
Nanto | +1.668 |
Pyro | +1.601 |
Neg logit contributions | |
subsequ | -1.534 |
therap | -1.415 |
exhib | -1.403 |
upgr | -1.394 |
momentum | -1.379 |
Token | James |
Token ID | 3700 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
"$:/ | +0.290 |
Malf | +0.273 |
LW | +0.263 |
Nanto | +0.246 |
spection | +0.245 |
Neg logit contributions | |
subsequ | -0.280 |
therap | -0.271 |
upgr | -0.262 |
exhib | -0.249 |
ammy | -0.248 |
Token | Tro |
Token ID | 8498 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
"$:/ | +0.187 |
Malf | +0.167 |
LW | +0.160 |
Pyro | +0.153 |
Nanto | +0.152 |
Neg logit contributions | |
subsequ | -0.155 |
upgr | -0.145 |
therap | -0.143 |
exhib | -0.142 |
momentum | -0.138 |
Token | t |
Token ID | 83 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
mosqu | +2.105 |
metic | +2.079 |
therap | +2.067 |
enthusi | +2.050 |
unden | +2.038 |
Neg logit contributions | |
"$:/ | -1.628 |
spection | -1.277 |
lf | -1.267 |
LW | -1.247 |
Malf | -1.222 |
Token | ter |
Token ID | 353 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
"$:/ | +1.506 |
spection | +1.135 |
LW | +1.101 |
lf | +1.083 |
Malf | +1.065 |
Neg logit contributions | |
mosqu | -2.696 |
enthusi | -2.690 |
therap | -2.669 |
metic | -2.634 |
unden | -2.591 |
Token | , |
Token ID | 11 |
Feature activation | -9.23e-02 |
Pos logit contributions | |
"$:/ | +0.626 |
Malf | +0.562 |
LW | +0.543 |
spection | +0.510 |
Nanto | +0.508 |
Neg logit contributions | |
subsequ | -0.541 |
therap | -0.501 |
upgr | -0.491 |
exhib | -0.481 |
momentum | -0.477 |
Token | his |
Token ID | 465 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
subsequ | +1.012 |
therap | +0.980 |
upgr | +0.932 |
mosqu | +0.926 |
ammy | +0.900 |
Neg logit contributions | |
"$:/ | -0.976 |
Malf | -0.893 |
LW | -0.872 |
Pyro | -0.789 |
Nanto | -0.778 |
Token | happiness |
Token ID | 12157 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
"$:/ | +0.498 |
Malf | +0.462 |
LW | +0.446 |
Pyro | +0.413 |
Nanto | +0.401 |
Neg logit contributions | |
subsequ | -0.382 |
therap | -0.366 |
mosqu | -0.365 |
upgr | -0.359 |
ammy | -0.359 |
Token | consists |
Token ID | 10874 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
"$:/ | +0.294 |
Malf | +0.260 |
LW | +0.255 |
Pyro | +0.237 |
Nanto | +0.233 |
Neg logit contributions | |
subsequ | -0.280 |
ammy | -0.277 |
therap | -0.269 |
mosqu | -0.259 |
exhib | -0.256 |
Token | of |
Token ID | 286 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
"$:/ | +0.510 |
Malf | +0.468 |
LW | +0.453 |
Nanto | +0.449 |
Pyro | +0.435 |
Neg logit contributions | |
subsequ | -0.302 |
hol | -0.280 |
therap | -0.274 |
erosion | -0.272 |
momentum | -0.272 |
Token | escape |
Token ID | 6654 |
Feature activation | +1.17e-03 |
Pos logit contributions | |
"$:/ | +1.176 |
Malf | +1.093 |
LW | +1.061 |
Nanto | +0.998 |
spection | +0.982 |
Neg logit contributions | |
subsequ | -0.947 |
therap | -0.912 |
upgr | -0.895 |
ammy | -0.894 |
exhib | -0.853 |
Token | from |
Token ID | 422 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
"$:/ | +0.013 |
Malf | +0.011 |
LW | +0.011 |
spection | +0.011 |
Nanto | +0.010 |
Neg logit contributions | |
subsequ | -0.012 |
therap | -0.012 |
ammy | -0.011 |
upgr | -0.011 |
exhib | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
subsequ | +1.283 |
therap | +1.173 |
exhib | +1.166 |
upgr | +1.162 |
momentum | +1.132 |
Neg logit contributions | |
"$:/ | -1.463 |
Malf | -1.301 |
LW | -1.268 |
Nanto | -1.225 |
Pyro | -1.188 |
Token | explode |
Token ID | 22818 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
subsequ | +0.497 |
mosqu | +0.464 |
therap | +0.463 |
unden | +0.441 |
upgr | +0.438 |
Neg logit contributions | |
"$:/ | -0.599 |
Malf | -0.545 |
LW | -0.530 |
Pyro | -0.504 |
Nanto | -0.485 |
Token | . |
Token ID | 13 |
Feature activation | +7.88e-02 |
Pos logit contributions | |
subsequ | +0.205 |
upgr | +0.189 |
therap | +0.187 |
exhib | +0.184 |
mosqu | +0.182 |
Neg logit contributions | |
"$:/ | -0.196 |
Malf | -0.182 |
LW | -0.178 |
Pyro | -0.168 |
spection | -0.161 |
Token | These |
Token ID | 2312 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
"$:/ | +0.748 |
Malf | +0.710 |
LW | +0.686 |
Pyro | +0.647 |
Nanto | +0.637 |
Neg logit contributions | |
subsequ | -1.049 |
therap | -1.008 |
mosqu | -0.991 |
unden | -0.942 |
satell | -0.928 |
Token | two |
Token ID | 734 |
Feature activation | -9.26e-02 |
Pos logit contributions | |
subsequ | +0.230 |
therap | +0.220 |
ammy | +0.213 |
exhib | +0.211 |
upgr | +0.210 |
Neg logit contributions | |
"$:/ | -0.270 |
Malf | -0.252 |
LW | -0.251 |
Nanto | -0.235 |
Pyro | -0.235 |
Token | are |
Token ID | 389 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
subsequ | +0.877 |
therap | +0.816 |
upgr | +0.802 |
ammy | +0.798 |
exhib | +0.788 |
Neg logit contributions | |
"$:/ | -1.042 |
Malf | -1.030 |
LW | -0.989 |
Nanto | -0.957 |
Pyro | -0.919 |
Token | going |
Token ID | 1016 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
"$:/ | +0.929 |
Malf | +0.917 |
LW | +0.872 |
Nanto | +0.840 |
Pyro | +0.801 |
Neg logit contributions | |
subsequ | -0.890 |
therap | -0.815 |
mosqu | -0.811 |
ammy | -0.801 |
artments | -0.795 |
Token | to |
Token ID | 284 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
subsequ | +0.574 |
therap | +0.533 |
exhib | +0.529 |
momentum | +0.526 |
upgr | +0.520 |
Neg logit contributions | |
"$:/ | -0.940 |
Malf | -0.820 |
LW | -0.801 |
Nanto | -0.791 |
Pyro | -0.758 |
Token | d |
Token ID | 288 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
"$:/ | +0.450 |
Malf | +0.429 |
LW | +0.407 |
Pyro | +0.390 |
Nanto | +0.390 |
Neg logit contributions | |
subsequ | -0.407 |
therap | -0.371 |
mosqu | -0.368 |
upgr | -0.359 |
enthusi | -0.354 |
Token | uke |
Token ID | 4649 |
Feature activation | -5.62e-02 |
Pos logit contributions | |
subsequ | +2.011 |
therap | +1.899 |
exhib | +1.859 |
momentum | +1.843 |
upgr | +1.836 |
Neg logit contributions | |
"$:/ | -1.855 |
Malf | -1.658 |
LW | -1.594 |
Nanto | -1.559 |
Pyro | -1.477 |
Token | it |
Token ID | 340 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
subsequ | +0.646 |
therap | +0.602 |
upgr | +0.593 |
ammy | +0.584 |
exhib | +0.584 |
Neg logit contributions | |
"$:/ | -0.548 |
Malf | -0.494 |
LW | -0.469 |
Nanto | -0.449 |
Pyro | -0.447 |
Token | understand |
Token ID | 1833 |
Feature activation | -7.52e-02 |
Pos logit contributions | |
"$:/ | +2.569 |
lf | +2.299 |
LW | +2.292 |
Nanto | +2.244 |
spection | +2.165 |
Neg logit contributions | |
subsequ | -2.906 |
therap | -2.818 |
mosqu | -2.801 |
undermin | -2.703 |
pestic | -2.672 |
Token | . |
Token ID | 13 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
subsequ | +0.841 |
therap | +0.778 |
exhib | +0.750 |
mosqu | +0.744 |
upgr | +0.741 |
Neg logit contributions | |
"$:/ | -0.777 |
Malf | -0.701 |
LW | -0.694 |
Nanto | -0.667 |
spection | -0.652 |
Token | I |
Token ID | 314 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
subsequ | +0.905 |
therap | +0.878 |
mosqu | +0.858 |
satell | +0.821 |
undermin | +0.812 |
Neg logit contributions | |
"$:/ | -0.545 |
Malf | -0.495 |
LW | -0.494 |
Nanto | -0.469 |
Pyro | -0.460 |
Token | 'll |
Token ID | 1183 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
"$:/ | +1.474 |
LW | +1.314 |
lf | +1.241 |
Malf | +1.215 |
Nanto | +1.163 |
Neg logit contributions | |
mosqu | -1.870 |
therap | -1.812 |
pestic | -1.812 |
subsequ | -1.798 |
undermin | -1.796 |
Token | send |
Token ID | 3758 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
"$:/ | +0.603 |
LW | +0.524 |
Malf | +0.505 |
Pyro | +0.484 |
Nanto | +0.482 |
Neg logit contributions | |
mosqu | -0.663 |
enthusi | -0.642 |
subsequ | -0.630 |
pestic | -0.628 |
therap | -0.627 |
Token | for |
Token ID | 329 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
"$:/ | +2.323 |
Malf | +2.150 |
LW | +2.126 |
Pyro | +2.047 |
Nanto | +1.991 |
Neg logit contributions | |
subsequ | -2.694 |
therap | -2.537 |
mosqu | -2.468 |
ammy | -2.443 |
exhib | -2.421 |
Token | a |
Token ID | 257 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
"$:/ | +2.046 |
Malf | +1.877 |
LW | +1.860 |
Nanto | +1.790 |
Pyro | +1.772 |
Neg logit contributions | |
subsequ | -2.118 |
therap | -1.952 |
mosqu | -1.881 |
upgr | -1.865 |
ammy | -1.864 |
Token | rescue |
Token ID | 9992 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
"$:/ | +2.820 |
LW | +2.695 |
Malf | +2.694 |
Pyro | +2.687 |
Nanto | +2.527 |
Neg logit contributions | |
subsequ | -2.618 |
exhib | -2.375 |
mosqu | -2.361 |
inertia | -2.358 |
ammy | -2.357 |
Token | immediately |
Token ID | 3393 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
"$:/ | +3.813 |
LW | +3.538 |
Pyro | +3.411 |
Malf | +3.375 |
Nanto | +3.057 |
Neg logit contributions | |
subsequ | -4.226 |
therap | -3.885 |
ammy | -3.871 |
erosion | -3.770 |
exhib | -3.751 |
Token | ." |
Token ID | 526 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
"$:/ | +2.458 |
Malf | +2.121 |
LW | +2.116 |
Nanto | +2.008 |
spection | +1.994 |
Neg logit contributions | |
subsequ | -3.216 |
therap | -3.108 |
pestic | -2.950 |
mosqu | -2.947 |
exhib | -2.940 |
Token |
|
Token ID | 198 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
"$:/ | +2.781 |
LW | +2.227 |
Malf | +2.066 |
Nanto | +1.999 |
Pyro | +1.935 |
Neg logit contributions | |
therap | -5.653 |
mosqu | -5.537 |
enthusi | -5.485 |
metic | -5.469 |
undermin | -5.447 |
Token | � |
Token ID | 447 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
subsequ | +0.631 |
exhib | +0.595 |
therap | +0.587 |
upgr | +0.572 |
ammy | +0.567 |
Neg logit contributions | |
"$:/ | -0.686 |
Malf | -0.601 |
LW | -0.591 |
Nanto | -0.564 |
Pyro | -0.552 |
Token | � |
Token ID | 247 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
"$:/ | +1.615 |
LW | +1.456 |
Malf | +1.423 |
Pyro | +1.338 |
spection | +1.333 |
Neg logit contributions | |
subsequ | -1.323 |
therap | -1.245 |
exhib | -1.241 |
ammy | -1.178 |
upgr | -1.175 |
Token | or |
Token ID | 393 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
"$:/ | +1.475 |
Malf | +1.288 |
LW | +1.278 |
spection | +1.203 |
Pyro | +1.198 |
Neg logit contributions | |
subsequ | -1.665 |
therap | -1.585 |
exhib | -1.554 |
mosqu | -1.548 |
ammy | -1.540 |
Token | the |
Token ID | 262 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
"$:/ | +3.965 |
Malf | +3.544 |
LW | +3.499 |
Pyro | +3.264 |
Nanto | +3.116 |
Neg logit contributions | |
subsequ | -3.059 |
exhib | -2.921 |
therap | -2.880 |
mosqu | -2.834 |
ammy | -2.811 |
Token | ability |
Token ID | 2694 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
"$:/ | +1.544 |
LW | +1.396 |
Malf | +1.383 |
Pyro | +1.309 |
Nanto | +1.282 |
Neg logit contributions | |
subsequ | -1.034 |
exhib | -0.956 |
therap | -0.955 |
mosqu | -0.908 |
ammy | -0.893 |
Token | to |
Token ID | 284 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
"$:/ | +2.379 |
Malf | +2.125 |
LW | +2.069 |
Nanto | +2.019 |
spection | +1.942 |
Neg logit contributions | |
subsequ | -1.684 |
therap | -1.534 |
upgr | -1.521 |
exhib | -1.515 |
momentum | -1.510 |
Token | successfully |
Token ID | 7675 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
"$:/ | +3.541 |
LW | +3.073 |
Pyro | +2.985 |
Malf | +2.970 |
Nanto | +2.815 |
Neg logit contributions | |
mosqu | -2.827 |
subsequ | -2.722 |
therap | -2.710 |
enthusi | -2.709 |
exhib | -2.573 |
Token | burn |
Token ID | 4245 |
Feature activation | -7.98e-04 |
Pos logit contributions | |
"$:/ | +3.637 |
LW | +3.206 |
Malf | +3.155 |
Pyro | +3.138 |
Nanto | +2.936 |
Neg logit contributions | |
mosqu | -2.970 |
subsequ | -2.953 |
therap | -2.864 |
enthusi | -2.841 |
exhib | -2.738 |
Token | stored |
Token ID | 8574 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
subsequ | +0.008 |
ammy | +0.008 |
exhib | +0.008 |
upgr | +0.008 |
therap | +0.008 |
Neg logit contributions | |
"$:/ | -0.008 |
LW | -0.008 |
Malf | -0.007 |
Nanto | -0.007 |
Pyro | -0.007 |
Token | body |
Token ID | 1767 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
"$:/ | +2.555 |
LW | +2.387 |
Malf | +2.275 |
Pyro | +2.191 |
Nanto | +2.042 |
Neg logit contributions | |
ammy | -2.321 |
subsequ | -2.310 |
exhib | -2.296 |
therap | -2.237 |
mosqu | -2.203 |
Token | fat |
Token ID | 3735 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
"$:/ | +1.077 |
LW | +1.000 |
Malf | +0.964 |
Nanto | +0.893 |
Pyro | +0.891 |
Neg logit contributions | |
subsequ | -1.215 |
exhib | -1.197 |
ammy | -1.156 |
therap | -1.154 |
upgr | -1.146 |
Token | During |
Token ID | 5856 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
"$:/ | +0.676 |
Malf | +0.646 |
LW | +0.644 |
Pyro | +0.577 |
Nanto | +0.574 |
Neg logit contributions | |
subsequ | -1.078 |
therap | -1.033 |
mosqu | -0.992 |
enthusi | -0.983 |
satell | -0.959 |
Token | the |
Token ID | 262 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
subsequ | +1.648 |
therap | +1.538 |
exhib | +1.507 |
upgr | +1.495 |
momentum | +1.463 |
Neg logit contributions | |
"$:/ | -2.106 |
Malf | -1.895 |
LW | -1.860 |
Nanto | -1.802 |
spection | -1.724 |
Token | � |
Token ID | 564 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
subsequ | +1.161 |
therap | +1.095 |
artments | +1.077 |
upgr | +1.070 |
exhib | +1.068 |
Neg logit contributions | |
"$:/ | -1.135 |
Malf | -1.046 |
LW | -1.033 |
Nanto | -0.960 |
spection | -0.937 |
Token | � |
Token ID | 246 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
"$:/ | +0.896 |
LW | +0.804 |
Malf | +0.790 |
Nanto | +0.733 |
spection | +0.732 |
Neg logit contributions | |
subsequ | -0.857 |
therap | -0.812 |
exhib | -0.790 |
upgr | -0.783 |
momentum | -0.776 |
Token | cart |
Token ID | 26674 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
subsequ | +1.073 |
therap | +1.019 |
exhib | +0.995 |
upgr | +0.990 |
momentum | +0.962 |
Neg logit contributions | |
"$:/ | -0.947 |
Malf | -0.810 |
LW | -0.785 |
spection | -0.759 |
Nanto | -0.748 |
Token | oon |
Token ID | 2049 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
subsequ | +1.329 |
therap | +1.238 |
exhib | +1.225 |
upgr | +1.207 |
momentum | +1.190 |
Neg logit contributions | |
"$:/ | -1.190 |
Malf | -1.020 |
LW | -0.973 |
Nanto | -0.951 |
Pyro | -0.891 |
Token | crisis |
Token ID | 4902 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
subsequ | +3.143 |
therap | +2.943 |
exhib | +2.894 |
upgr | +2.886 |
momentum | +2.803 |
Neg logit contributions | |
"$:/ | -3.282 |
Malf | -2.895 |
LW | -2.807 |
Nanto | -2.698 |
spection | -2.594 |
Token | � |
Token ID | 447 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
subsequ | +1.366 |
exhib | +1.279 |
therap | +1.275 |
upgr | +1.258 |
erosion | +1.236 |
Neg logit contributions | |
"$:/ | -1.350 |
Malf | -1.177 |
LW | -1.150 |
Nanto | -1.101 |
Pyro | -1.061 |
Token | � |
Token ID | 247 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
"$:/ | +0.274 |
LW | +0.264 |
Malf | +0.249 |
spection | +0.228 |
Pyro | +0.223 |
Neg logit contributions | |
therap | -0.330 |
subsequ | -0.321 |
mosqu | -0.315 |
enthusi | -0.302 |
exhib | -0.302 |
Token | — |
Token ID | 851 |
Feature activation | -8.62e-02 |
Pos logit contributions | |
"$:/ | +1.164 |
Malf | +1.045 |
LW | +1.037 |
Nanto | +0.939 |
spection | +0.934 |
Neg logit contributions | |
subsequ | -1.184 |
therap | -1.118 |
upgr | -1.090 |
exhib | -1.090 |
mosqu | -1.078 |
Token | a |
Token ID | 257 |
Feature activation | +8.30e-02 |
Pos logit contributions | |
subsequ | +0.921 |
therap | +0.891 |
exhib | +0.859 |
mosqu | +0.845 |
upgr | +0.840 |
Neg logit contributions | |
"$:/ | -0.911 |
Malf | -0.823 |
LW | -0.803 |
Nanto | -0.763 |
Pyro | -0.739 |
Token | lives |
Token ID | 3160 |
Feature activation | +9.22e-03 |
Pos logit contributions | |
"$:/ | +0.079 |
LW | +0.072 |
Malf | +0.071 |
Nanto | +0.065 |
lf | +0.063 |
Neg logit contributions | |
mosqu | -0.086 |
subsequ | -0.086 |
therap | -0.086 |
ammy | -0.084 |
undermin | -0.082 |
Token | in |
Token ID | 287 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
"$:/ | +0.094 |
Malf | +0.082 |
LW | +0.081 |
Nanto | +0.074 |
Pyro | +0.073 |
Neg logit contributions | |
subsequ | -0.103 |
therap | -0.096 |
exhib | -0.096 |
upgr | -0.094 |
ammy | -0.094 |
Token | Ontario |
Token ID | 10553 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
subsequ | +2.531 |
therap | +2.407 |
exhib | +2.379 |
upgr | +2.323 |
ammy | +2.287 |
Neg logit contributions | |
"$:/ | -2.027 |
Malf | -1.787 |
LW | -1.726 |
Nanto | -1.646 |
Pyro | -1.581 |
Token | . |
Token ID | 13 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
subsequ | +0.469 |
therap | +0.432 |
exhib | +0.426 |
upgr | +0.420 |
momentum | +0.415 |
Neg logit contributions | |
"$:/ | -0.571 |
Malf | -0.505 |
LW | -0.491 |
Nanto | -0.471 |
spection | -0.461 |
Token | However |
Token ID | 2102 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
subsequ | +0.278 |
therap | +0.261 |
mosqu | +0.252 |
enthusi | +0.248 |
exhib | +0.245 |
Neg logit contributions | |
"$:/ | -0.223 |
Malf | -0.204 |
LW | -0.197 |
Pyro | -0.179 |
Nanto | -0.177 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
"$:/ | +1.053 |
LW | +0.984 |
Malf | +0.931 |
spection | +0.830 |
RJ | +0.815 |
Neg logit contributions | |
therap | -1.620 |
subsequ | -1.582 |
enthusi | -1.538 |
mosqu | -1.505 |
unden | -1.493 |
Token | locating |
Token ID | 42139 |
Feature activation | -4.63e-03 |
Pos logit contributions | |
subsequ | +1.205 |
therap | +1.156 |
ammy | +1.120 |
upgr | +1.097 |
exhib | +1.096 |
Neg logit contributions | |
"$:/ | -1.105 |
Malf | -1.035 |
LW | -1.006 |
spection | -0.897 |
Nanto | -0.894 |
Token | in |
Token ID | 287 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
subsequ | +0.049 |
therap | +0.046 |
ammy | +0.045 |
exhib | +0.044 |
enthusi | +0.044 |
Neg logit contributions | |
"$:/ | -0.049 |
Malf | -0.046 |
LW | -0.044 |
Pyro | -0.039 |
Nanto | -0.039 |
Token | Alberta |
Token ID | 15555 |
Feature activation | -8.60e-02 |
Pos logit contributions | |
subsequ | +1.763 |
therap | +1.661 |
exhib | +1.641 |
upgr | +1.591 |
sugg | +1.584 |
Neg logit contributions | |
"$:/ | -1.432 |
Malf | -1.265 |
LW | -1.217 |
Nanto | -1.139 |
spection | -1.124 |
Token | has |
Token ID | 468 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
subsequ | +0.943 |
therap | +0.889 |
ammy | +0.844 |
mosqu | +0.839 |
upgr | +0.830 |
Neg logit contributions | |
"$:/ | -0.940 |
Malf | -0.837 |
LW | -0.816 |
Pyro | -0.745 |
abouts | -0.741 |
Token | in |
Token ID | 287 |
Feature activation | +3.00e-04 |
Pos logit contributions | |
subsequ | +0.498 |
therap | +0.467 |
ammy | +0.464 |
mosqu | +0.463 |
obser | +0.438 |
Neg logit contributions | |
"$:/ | -0.572 |
Malf | -0.530 |
LW | -0.524 |
Pyro | -0.467 |
abouts | -0.463 |
Token | ushing |
Token ID | 8023 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
subsequ | +0.704 |
therap | +0.682 |
exhib | +0.659 |
inertia | +0.650 |
momentum | +0.649 |
Neg logit contributions | |
"$:/ | -0.652 |
Malf | -0.579 |
LW | -0.556 |
Nanto | -0.533 |
Pyro | -0.518 |
Token | , |
Token ID | 11 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
subsequ | +1.153 |
therap | +1.099 |
exhib | +1.032 |
inertia | +1.014 |
momentum | +1.014 |
Neg logit contributions | |
"$:/ | -1.084 |
Malf | -1.008 |
LW | -0.992 |
Nanto | -0.921 |
spection | -0.905 |
Token | Oklahoma |
Token ID | 10433 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
subsequ | +4.271 |
therap | +4.034 |
exhib | +3.881 |
mosqu | +3.844 |
upgr | +3.788 |
Neg logit contributions | |
"$:/ | -3.436 |
Malf | -3.078 |
LW | -2.992 |
Nanto | -2.805 |
Pyro | -2.670 |
Token | , |
Token ID | 11 |
Feature activation | -9.20e-02 |
Pos logit contributions | |
subsequ | +2.259 |
therap | +2.086 |
exhib | +2.057 |
upgr | +2.034 |
momentum | +2.016 |
Neg logit contributions | |
"$:/ | -3.003 |
Malf | -2.666 |
LW | -2.603 |
Nanto | -2.535 |
Pyro | -2.434 |
Token | March |
Token ID | 2805 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
subsequ | +1.085 |
therap | +1.051 |
mosqu | +0.993 |
traged | +0.969 |
ammy | +0.963 |
Neg logit contributions | |
"$:/ | -0.880 |
LW | -0.826 |
Malf | -0.821 |
Nanto | -0.713 |
spection | -0.704 |
Token | 24 |
Token ID | 1987 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
LW | +0.597 |
"$:/ | +0.591 |
Malf | +0.519 |
Pyro | +0.466 |
LG | +0.465 |
Neg logit contributions | |
mosqu | -1.203 |
therap | -1.203 |
subsequ | -1.160 |
enthusi | -1.146 |
undermin | -1.143 |
Token | , |
Token ID | 11 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
therap | +3.688 |
mosqu | +3.573 |
subsequ | +3.539 |
enthusi | +3.499 |
metic | +3.403 |
Neg logit contributions | |
"$:/ | -2.495 |
LW | -2.118 |
Malf | -2.020 |
Nanto | -1.852 |
spection | -1.813 |
Token | 2016 |
Token ID | 1584 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
subsequ | +1.965 |
therap | +1.947 |
mosqu | +1.884 |
obser | +1.842 |
unden | +1.807 |
Neg logit contributions | |
"$:/ | -1.137 |
LW | -0.988 |
Malf | -0.889 |
Pyro | -0.789 |
Nanto | -0.779 |
Token | . |
Token ID | 13 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
subsequ | +2.832 |
upgr | +2.601 |
momentum | +2.555 |
exhib | +2.500 |
hol | +2.391 |
Neg logit contributions | |
"$:/ | -8.188 |
Malf | -7.197 |
Nanto | -7.142 |
LW | -6.882 |
Pengu | -6.866 |
Token | REUTERS |
Token ID | 15862 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
subsequ | +5.180 |
therap | +4.876 |
exhib | +4.829 |
upgr | +4.800 |
momentum | +4.764 |
Neg logit contributions | |
"$:/ | -3.791 |
Malf | -3.205 |
LW | -3.098 |
Nanto | -2.986 |
Pyro | -2.821 |
Token | / |
Token ID | 14 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
therap | +1.875 |
mosqu | +1.832 |
enthusi | +1.762 |
metic | +1.759 |
subsequ | +1.751 |
Neg logit contributions | |
"$:/ | -1.043 |
LW | -0.961 |
Malf | -0.858 |
Pyro | -0.779 |
LG | -0.767 |
Token | � |
Token ID | 564 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
"$:/ | +0.121 |
Malf | +0.104 |
LW | +0.101 |
Nanto | +0.093 |
Pyro | +0.090 |
Neg logit contributions | |
subsequ | -0.122 |
mosqu | -0.115 |
therap | -0.112 |
upgr | -0.110 |
exhib | -0.109 |
Token | � |
Token ID | 250 |
Feature activation | -4.47e-03 |
Pos logit contributions | |
subsequ | +1.327 |
therap | +1.269 |
mosqu | +1.226 |
exhib | +1.201 |
enthusi | +1.193 |
Neg logit contributions | |
"$:/ | -1.015 |
Malf | -0.897 |
LW | -0.895 |
spection | -0.817 |
Nanto | -0.793 |
Token | a |
Token ID | 64 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
subsequ | +0.055 |
therap | +0.051 |
mosqu | +0.051 |
upgr | +0.049 |
momentum | +0.048 |
Neg logit contributions | |
"$:/ | -0.043 |
Malf | -0.037 |
LW | -0.035 |
spection | -0.035 |
lf | -0.035 |
Token | little |
Token ID | 1310 |
Feature activation | -8.12e-02 |
Pos logit contributions | |
"$:/ | +1.992 |
Malf | +1.834 |
LW | +1.806 |
Pyro | +1.709 |
Nanto | +1.701 |
Neg logit contributions | |
subsequ | -1.681 |
therap | -1.532 |
upgr | -1.515 |
exhib | -1.505 |
mosqu | -1.495 |
Token | concerned |
Token ID | 5213 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
subsequ | +0.816 |
mosqu | +0.719 |
therap | +0.712 |
artments | +0.703 |
upgr | +0.701 |
Neg logit contributions | |
"$:/ | -0.957 |
Malf | -0.868 |
LW | -0.822 |
Nanto | -0.769 |
Pyro | -0.768 |
Token | about |
Token ID | 546 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
subsequ | +2.702 |
therap | +2.482 |
upgr | +2.439 |
exhib | +2.416 |
mosqu | +2.375 |
Neg logit contributions | |
"$:/ | -2.740 |
Malf | -2.405 |
LW | -2.387 |
Nanto | -2.274 |
Pyro | -2.201 |
Token | some |
Token ID | 617 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
subsequ | +1.181 |
therap | +1.109 |
ammy | +1.080 |
mosqu | +1.069 |
exhib | +1.062 |
Neg logit contributions | |
"$:/ | -1.304 |
Malf | -1.200 |
LW | -1.183 |
Nanto | -1.077 |
Pyro | -1.065 |
Token | of |
Token ID | 286 |
Feature activation | -8.33e-04 |
Pos logit contributions | |
subsequ | +1.807 |
therap | +1.705 |
mosqu | +1.686 |
ammy | +1.655 |
exhib | +1.649 |
Neg logit contributions | |
"$:/ | -2.005 |
LW | -1.790 |
Malf | -1.780 |
Nanto | -1.631 |
Pyro | -1.565 |
Token | the |
Token ID | 262 |
Feature activation | -8.68e-02 |
Pos logit contributions | |
subsequ | +0.008 |
therap | +0.007 |
exhib | +0.007 |
upgr | +0.007 |
ammy | +0.007 |
Neg logit contributions | |
"$:/ | -0.010 |
Malf | -0.009 |
LW | -0.009 |
Nanto | -0.008 |
Pyro | -0.008 |
Token | trade |
Token ID | 3292 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
subsequ | +0.855 |
mosqu | +0.800 |
therap | +0.798 |
ammy | +0.763 |
unden | +0.761 |
Neg logit contributions | |
"$:/ | -0.976 |
Malf | -0.911 |
LW | -0.902 |
Pyro | -0.795 |
spection | -0.794 |
Token | rhetoric |
Token ID | 13214 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
subsequ | +1.057 |
therap | +0.958 |
mosqu | +0.925 |
exhib | +0.923 |
ammy | +0.918 |
Neg logit contributions | |
"$:/ | -1.161 |
Malf | -1.078 |
LW | -1.033 |
Nanto | -1.001 |
Pyro | -0.994 |
Token | Col |
Token ID | 1623 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
subsequ | +1.714 |
therap | +1.625 |
exhib | +1.582 |
upgr | +1.571 |
momentum | +1.547 |
Neg logit contributions | |
"$:/ | -1.868 |
Malf | -1.726 |
LW | -1.652 |
Nanto | -1.594 |
Pyro | -1.567 |
Token | bie |
Token ID | 12590 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
"$:/ | +2.416 |
Malf | +2.320 |
LW | +2.222 |
Pyro | +2.104 |
Nanto | +2.099 |
Neg logit contributions | |
subsequ | -2.566 |
therap | -2.474 |
exhib | -2.401 |
upgr | -2.358 |
momentum | -2.326 |
Token | Sterling |
Token ID | 23647 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
"$:/ | +1.134 |
Malf | +1.042 |
LW | +0.994 |
Nanto | +0.959 |
Pyro | +0.945 |
Neg logit contributions | |
subsequ | -1.117 |
therap | -1.047 |
exhib | -1.016 |
upgr | -1.013 |
momentum | -0.997 |
Token | — |
Token ID | 960 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
"$:/ | +1.276 |
Malf | +1.170 |
LW | +1.132 |
Nanto | +1.103 |
Pyro | +1.082 |
Neg logit contributions | |
subsequ | -1.072 |
therap | -0.993 |
exhib | -0.974 |
momentum | -0.955 |
upgr | -0.955 |
Token | all |
Token ID | 439 |
Feature activation | -6.99e-02 |
Pos logit contributions | |
subsequ | +1.997 |
therap | +1.872 |
mosqu | +1.809 |
erosion | +1.804 |
exhib | +1.796 |
Neg logit contributions | |
"$:/ | -1.458 |
Malf | -1.336 |
LW | -1.278 |
Nanto | -1.186 |
Pyro | -1.170 |
Token | of |
Token ID | 286 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
subsequ | +0.738 |
therap | +0.689 |
exhib | +0.663 |
mosqu | +0.661 |
erosion | +0.655 |
Neg logit contributions | |
"$:/ | -0.741 |
Malf | -0.709 |
LW | -0.698 |
Nanto | -0.664 |
Pyro | -0.643 |
Token | whom |
Token ID | 4150 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
subsequ | +2.264 |
therap | +2.096 |
exhib | +2.064 |
upgr | +2.019 |
momentum | +2.003 |
Neg logit contributions | |
"$:/ | -2.607 |
Malf | -2.328 |
LW | -2.263 |
Nanto | -2.208 |
Pyro | -2.128 |
Token | can |
Token ID | 460 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
"$:/ | +0.264 |
Malf | +0.254 |
LW | +0.250 |
Pyro | +0.236 |
Nanto | +0.232 |
Neg logit contributions | |
subsequ | -0.219 |
therap | -0.206 |
mosqu | -0.201 |
chromos | -0.196 |
erosion | -0.193 |
Token | be |
Token ID | 307 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
"$:/ | +0.476 |
Malf | +0.448 |
LW | +0.441 |
Pyro | +0.424 |
Nanto | +0.420 |
Neg logit contributions | |
subsequ | -0.404 |
mosqu | -0.377 |
therap | -0.374 |
exhib | -0.370 |
erosion | -0.363 |
Token | taken |
Token ID | 2077 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
subsequ | +1.448 |
therap | +1.307 |
erosion | +1.303 |
exhib | +1.303 |
mosqu | +1.300 |
Neg logit contributions | |
"$:/ | -1.799 |
Malf | -1.754 |
LW | -1.679 |
Pyro | -1.625 |
Nanto | -1.623 |
Token | in |
Token ID | 287 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
"$:/ | +0.419 |
Malf | +0.391 |
LW | +0.375 |
Pyro | +0.365 |
Nanto | +0.347 |
Neg logit contributions | |
subsequ | -0.470 |
therap | -0.438 |
mosqu | -0.429 |
exhib | -0.415 |
enthusi | -0.409 |
Token | the |
Token ID | 262 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
subsequ | +0.436 |
therap | +0.403 |
mosqu | +0.397 |
upgr | +0.397 |
exhib | +0.397 |
Neg logit contributions | |
"$:/ | -0.468 |
Malf | -0.429 |
LW | -0.416 |
Pyro | -0.380 |
Nanto | -0.365 |
Token | prime |
Token ID | 6994 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
subsequ | +0.707 |
therap | +0.672 |
mosqu | +0.644 |
exhib | +0.642 |
upgr | +0.641 |
Neg logit contributions | |
"$:/ | -0.778 |
Malf | -0.703 |
LW | -0.702 |
Pyro | -0.635 |
Nanto | -0.599 |
Token | challenge |
Token ID | 4427 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
subsequ | +2.465 |
therap | +2.332 |
exhib | +2.265 |
ammy | +2.248 |
upgr | +2.239 |
Neg logit contributions | |
"$:/ | -2.939 |
Malf | -2.640 |
LW | -2.565 |
Nanto | -2.391 |
Pyro | -2.340 |
Token | Belfast |
Token ID | 29713 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
"$:/ | +1.723 |
Malf | +1.579 |
LW | +1.554 |
Pyro | +1.412 |
Nanto | +1.404 |
Neg logit contributions | |
subsequ | -1.957 |
therap | -1.877 |
upgr | -1.838 |
exhib | -1.806 |
ammy | -1.797 |
Token | Te |
Token ID | 6767 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
subsequ | +8.347 |
therap | +8.179 |
mosqu | +7.839 |
upgr | +7.835 |
unden | +7.698 |
Neg logit contributions | |
"$:/ | -5.407 |
Malf | -4.467 |
LW | -4.285 |
Pyro | -4.080 |
Nanto | -4.061 |
Token | legraph |
Token ID | 16606 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
subsequ | +4.400 |
therap | +4.196 |
exhib | +4.167 |
upgr | +4.151 |
mosqu | +4.134 |
Neg logit contributions | |
"$:/ | -2.155 |
LW | -1.709 |
Malf | -1.704 |
spection | -1.480 |
Nanto | -1.402 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
subsequ | +6.397 |
therap | +5.959 |
mosqu | +5.835 |
unden | +5.684 |
exhib | +5.651 |
Neg logit contributions | |
"$:/ | -6.819 |
LW | -6.154 |
Malf | -6.130 |
Nanto | -5.683 |
Pyro | -5.614 |
Token | co |
Token ID | 1073 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
mosqu | +1.655 |
subsequ | +1.617 |
therap | +1.601 |
traged | +1.568 |
unden | +1.563 |
Neg logit contributions | |
"$:/ | -0.938 |
LW | -0.808 |
Malf | -0.743 |
abouts | -0.712 |
spection | -0.697 |
Token | . |
Token ID | 13 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
therap | +2.829 |
mosqu | +2.778 |
enthusi | +2.748 |
subsequ | +2.731 |
unden | +2.693 |
Neg logit contributions | |
"$:/ | -2.230 |
LW | -2.051 |
Malf | -1.948 |
lf | -1.857 |
spection | -1.832 |
Token | uk |
Token ID | 2724 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
"$:/ | +1.368 |
LW | +1.303 |
lf | +1.203 |
Malf | +1.202 |
spection | +1.171 |
Neg logit contributions | |
mosqu | -1.525 |
subsequ | -1.495 |
therap | -1.489 |
metic | -1.463 |
unden | -1.438 |
Token | Dis |
Token ID | 3167 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
subsequ | +1.950 |
therap | +1.838 |
mosqu | +1.769 |
exhib | +1.724 |
obser | +1.709 |
Neg logit contributions | |
"$:/ | -1.690 |
Malf | -1.607 |
LW | -1.536 |
Pyro | -1.437 |
Nanto | -1.364 |
Token | that |
Token ID | 326 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
subsequ | +1.509 |
therap | +1.424 |
exhib | +1.396 |
ammy | +1.384 |
upgr | +1.374 |
Neg logit contributions | |
"$:/ | -1.476 |
Malf | -1.241 |
LW | -1.220 |
Nanto | -1.164 |
Pyro | -1.108 |
Token | has |
Token ID | 468 |
Feature activation | -2.89e-03 |
Pos logit contributions | |
"$:/ | +0.299 |
Malf | +0.263 |
LW | +0.256 |
Nanto | +0.244 |
Pyro | +0.243 |
Neg logit contributions | |
subsequ | -0.248 |
ammy | -0.240 |
therap | -0.226 |
mosqu | -0.226 |
upgr | -0.224 |
Token | greatly |
Token ID | 9257 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
subsequ | +0.027 |
ammy | +0.025 |
mosqu | +0.025 |
therap | +0.025 |
upgr | +0.024 |
Neg logit contributions | |
"$:/ | -0.035 |
Malf | -0.031 |
LW | -0.030 |
Pyro | -0.028 |
Nanto | -0.027 |
Token | harmed |
Token ID | 28517 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
subsequ | +0.156 |
mosqu | +0.146 |
ammy | +0.143 |
therap | +0.140 |
unden | +0.135 |
Neg logit contributions | |
"$:/ | -0.222 |
Malf | -0.192 |
LW | -0.184 |
Nanto | -0.175 |
Pyro | -0.175 |
Token | the |
Token ID | 262 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
subsequ | +1.696 |
therap | +1.594 |
ammy | +1.583 |
upgr | +1.569 |
exhib | +1.554 |
Neg logit contributions | |
"$:/ | -1.867 |
LW | -1.663 |
Malf | -1.660 |
Nanto | -1.548 |
spection | -1.507 |
Token | Holy |
Token ID | 7439 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
subsequ | +2.919 |
therap | +2.782 |
ammy | +2.736 |
exhib | +2.702 |
artments | +2.685 |
Neg logit contributions | |
"$:/ | -3.547 |
LW | -3.178 |
Malf | -3.126 |
Nanto | -2.865 |
Pyro | -2.846 |
Token | City |
Token ID | 2254 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
subsequ | +6.823 |
therap | +6.392 |
exhib | +6.323 |
upgr | +6.290 |
momentum | +6.213 |
Neg logit contributions | |
"$:/ | -6.521 |
Malf | -5.624 |
LW | -5.498 |
Nanto | -5.324 |
Pyro | -5.069 |
Token | because |
Token ID | 780 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
subsequ | +3.992 |
therap | +3.701 |
exhib | +3.655 |
upgr | +3.642 |
momentum | +3.540 |
Neg logit contributions | |
"$:/ | -4.725 |
Malf | -4.152 |
LW | -4.081 |
Nanto | -3.974 |
Pyro | -3.807 |
Token | it |
Token ID | 340 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
"$:/ | +0.199 |
Malf | +0.176 |
LW | +0.173 |
Nanto | +0.163 |
spection | +0.158 |
Neg logit contributions | |
subsequ | -0.195 |
therap | -0.186 |
ammy | -0.183 |
upgr | -0.181 |
exhib | -0.178 |
Token | has |
Token ID | 468 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
"$:/ | +1.595 |
Malf | +1.353 |
LW | +1.324 |
Pyro | +1.269 |
Nanto | +1.246 |
Neg logit contributions | |
subsequ | -1.366 |
therap | -1.290 |
ammy | -1.248 |
mosqu | -1.231 |
upgr | -1.213 |
Token | afforded |
Token ID | 28109 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
subsequ | +0.335 |
therap | +0.311 |
ammy | +0.308 |
mosqu | +0.307 |
upgr | +0.296 |
Neg logit contributions | |
"$:/ | -0.440 |
Malf | -0.385 |
LW | -0.374 |
Pyro | -0.347 |
Nanto | -0.344 |
Token | Stan |
Token ID | 32140 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
"$:/ | +2.867 |
LW | +2.474 |
Malf | +2.421 |
lf | +2.149 |
Nanto | +2.126 |
Neg logit contributions | |
subsequ | -4.529 |
therap | -4.338 |
upgr | -4.109 |
exhib | -4.089 |
mosqu | -4.062 |
Token | . |
Token ID | 13 |
Feature activation | -7.68e-03 |
Pos logit contributions | |
"$:/ | +1.263 |
Malf | +1.088 |
LW | +1.040 |
Nanto | +0.957 |
lf | +0.952 |
Neg logit contributions | |
subsequ | -1.692 |
therap | -1.624 |
upgr | -1.594 |
exhib | -1.542 |
inertia | -1.535 |
Token | Neil |
Token ID | 15929 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
subsequ | +0.083 |
therap | +0.078 |
exhib | +0.075 |
upgr | +0.075 |
inertia | +0.073 |
Neg logit contributions | |
"$:/ | -0.082 |
Malf | -0.073 |
LW | -0.072 |
Nanto | -0.067 |
Pyro | -0.066 |
Token | Farrell |
Token ID | 37568 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
subsequ | +1.326 |
therap | +1.259 |
exhib | +1.231 |
upgr | +1.227 |
inertia | +1.190 |
Neg logit contributions | |
"$:/ | -1.299 |
Malf | -1.145 |
LW | -1.119 |
Nanto | -1.072 |
Pyro | -1.052 |
Token | DT |
Token ID | 24311 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
subsequ | +0.863 |
therap | +0.831 |
exhib | +0.798 |
inertia | +0.790 |
upgr | +0.788 |
Neg logit contributions | |
"$:/ | -0.620 |
LW | -0.586 |
Malf | -0.554 |
Pyro | -0.498 |
Nanto | -0.496 |
Token | 6 |
Token ID | 718 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
subsequ | +1.064 |
therap | +0.972 |
upgr | +0.966 |
exhib | +0.966 |
momentum | +0.952 |
Neg logit contributions | |
"$:/ | -1.681 |
Malf | -1.490 |
Nanto | -1.432 |
LW | -1.430 |
spection | -1.367 |
Token | ' |
Token ID | 6 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
subsequ | +6.022 |
therap | +5.979 |
mosqu | +5.844 |
enthusi | +5.704 |
unden | +5.499 |
Neg logit contributions | |
"$:/ | -7.440 |
LW | -6.573 |
Malf | -6.461 |
Pyro | -6.108 |
Nanto | -6.035 |
Token | 4 |
Token ID | 19 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
subsequ | +5.990 |
therap | +5.575 |
exhib | +5.482 |
upgr | +5.463 |
momentum | +5.390 |
Neg logit contributions | |
"$:/ | -6.592 |
Malf | -5.745 |
LW | -5.617 |
Nanto | -5.432 |
Pyro | -5.230 |
Token | / |
Token ID | 14 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
therap | +3.189 |
mosqu | +3.146 |
enthusi | +3.120 |
subsequ | +3.043 |
metic | +3.013 |
Neg logit contributions | |
"$:/ | -2.084 |
LW | -1.827 |
Malf | -1.694 |
Pyro | -1.597 |
lf | -1.543 |
Token | 295 |
Token ID | 25710 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
"$:/ | +0.248 |
Malf | +0.198 |
LW | +0.194 |
lf | +0.170 |
Pyro | +0.168 |
Neg logit contributions | |
subsequ | -0.498 |
therap | -0.483 |
mosqu | -0.469 |
unden | -0.466 |
enthusi | -0.459 |
Token | 3 |
Token ID | 513 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
"$:/ | +0.344 |
LW | +0.304 |
Malf | +0.304 |
Nanto | +0.286 |
Pyro | +0.279 |
Neg logit contributions | |
subsequ | -0.307 |
therap | -0.285 |
exhib | -0.277 |
upgr | -0.276 |
inertia | -0.271 |
Token | matter |
Token ID | 47635 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
subsequ | +0.351 |
therap | +0.325 |
exhib | +0.318 |
upgr | +0.316 |
mosqu | +0.308 |
Neg logit contributions | |
"$:/ | -0.333 |
Malf | -0.309 |
LW | -0.302 |
Nanto | -0.280 |
spection | -0.278 |
Token | of |
Token ID | 286 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
subsequ | +2.040 |
therap | +1.905 |
exhib | +1.872 |
upgr | +1.827 |
ammy | +1.786 |
Neg logit contributions | |
"$:/ | -2.550 |
Malf | -2.270 |
LW | -2.209 |
Nanto | -2.107 |
spection | -2.064 |
Token | fact |
Token ID | 1109 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
"$:/ | +0.171 |
Malf | +0.150 |
LW | +0.148 |
Nanto | +0.138 |
spection | +0.137 |
Neg logit contributions | |
subsequ | -0.193 |
therap | -0.180 |
upgr | -0.177 |
exhib | -0.176 |
ammy | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | +7.22e-03 |
Pos logit contributions | |
"$:/ | +1.318 |
Malf | +1.222 |
LW | +1.180 |
Nanto | +1.122 |
spection | +1.116 |
Neg logit contributions | |
subsequ | -1.261 |
therap | -1.182 |
exhib | -1.159 |
ammy | -1.135 |
upgr | -1.133 |
Token | I |
Token ID | 314 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
"$:/ | +0.070 |
Malf | +0.066 |
LW | +0.065 |
Nanto | +0.060 |
Pyro | +0.059 |
Neg logit contributions | |
subsequ | -0.090 |
therap | -0.085 |
mosqu | -0.081 |
exhib | -0.081 |
upgr | -0.078 |
Token | sh |
Token ID | 427 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
subsequ | +1.558 |
mosqu | +1.534 |
therap | +1.476 |
pestic | +1.452 |
unden | +1.435 |
Neg logit contributions | |
"$:/ | -1.715 |
Malf | -1.571 |
LW | -1.568 |
lf | -1.432 |
Nanto | -1.408 |
Token | ivered |
Token ID | 6396 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
subsequ | +6.358 |
therap | +5.944 |
exhib | +5.780 |
upgr | +5.752 |
momentum | +5.670 |
Neg logit contributions | |
"$:/ | -7.316 |
Malf | -6.438 |
LW | -6.338 |
Nanto | -6.024 |
spection | -5.837 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
subsequ | +2.159 |
therap | +2.026 |
upgr | +1.995 |
mosqu | +1.965 |
ammy | +1.937 |
Neg logit contributions | |
"$:/ | -2.108 |
LW | -1.928 |
Malf | -1.920 |
Pyro | -1.752 |
Nanto | -1.737 |
Token | Hop |
Token ID | 9996 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
subsequ | +0.192 |
therap | +0.180 |
exhib | +0.174 |
mosqu | +0.173 |
upgr | +0.170 |
Neg logit contributions | |
"$:/ | -0.150 |
Malf | -0.137 |
LW | -0.136 |
Nanto | -0.123 |
Pyro | -0.123 |
Token | ing |
Token ID | 278 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
"$:/ | +1.813 |
Malf | +1.593 |
LW | +1.524 |
Nanto | +1.482 |
Pyro | +1.401 |
Neg logit contributions | |
subsequ | -1.932 |
therap | -1.836 |
exhib | -1.807 |
upgr | -1.806 |
mosqu | -1.761 |
Token | I |
Token ID | 314 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
subsequ | +2.375 |
therap | +2.243 |
exhib | +2.182 |
upgr | +2.165 |
ammy | +2.135 |
Neg logit contributions | |
"$:/ | -2.381 |
Malf | -2.184 |
LW | -2.137 |
Nanto | -1.994 |
Pyro | -1.948 |
Token | Head |
Token ID | 7123 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
subsequ | +1.106 |
therap | +1.036 |
upgr | +1.007 |
exhib | +1.001 |
momentum | +0.994 |
Neg logit contributions | |
"$:/ | -1.089 |
LW | -0.995 |
Malf | -0.995 |
Nanto | -0.914 |
Pyro | -0.905 |
Token | Athletic |
Token ID | 29007 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
subsequ | +2.245 |
therap | +2.095 |
exhib | +2.074 |
upgr | +2.069 |
momentum | +2.034 |
Neg logit contributions | |
"$:/ | -2.161 |
Malf | -1.903 |
LW | -1.842 |
Nanto | -1.799 |
Pyro | -1.729 |
Token | Trainer |
Token ID | 31924 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
subsequ | +0.244 |
therap | +0.225 |
upgr | +0.225 |
exhib | +0.220 |
momentum | +0.217 |
Neg logit contributions | |
"$:/ | -0.253 |
Malf | -0.236 |
LW | -0.226 |
Nanto | -0.217 |
Pyro | -0.214 |
Token | ) |
Token ID | 8 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
"$:/ | +0.952 |
Malf | +0.837 |
LW | +0.815 |
Nanto | +0.785 |
Pyro | +0.758 |
Neg logit contributions | |
subsequ | -0.997 |
therap | -0.936 |
upgr | -0.921 |
exhib | -0.920 |
momentum | -0.910 |
Token | Aff |
Token ID | 6708 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
subsequ | +0.369 |
therap | +0.349 |
upgr | +0.344 |
momentum | +0.341 |
exhib | +0.339 |
Neg logit contributions | |
"$:/ | -0.384 |
Malf | -0.362 |
LW | -0.344 |
Nanto | -0.329 |
Pyro | -0.318 |
Token | idav |
Token ID | 20331 |
Feature activation | -6.01e-01 |
Pos logit contributions | |
subsequ | +4.261 |
therap | +4.132 |
upgr | +4.056 |
exhib | +4.051 |
momentum | +4.001 |
Neg logit contributions | |
"$:/ | -2.285 |
Malf | -1.925 |
LW | -1.888 |
Nanto | -1.741 |
Pyro | -1.642 |
Token | it |
Token ID | 270 |
Feature activation | -9.18e-03 |
Pos logit contributions | |
therap | +8.416 |
mosqu | +8.338 |
subsequ | +8.205 |
upgr | +8.016 |
enthusi | +7.897 |
Neg logit contributions | |
"$:/ | -5.121 |
Malf | -4.303 |
LW | -4.234 |
spection | -4.131 |
Nanto | -3.843 |
Token |
|
Token ID | 198 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
subsequ | +0.094 |
therap | +0.088 |
upgr | +0.087 |
exhib | +0.087 |
momentum | +0.086 |
Neg logit contributions | |
"$:/ | -0.101 |
Malf | -0.091 |
LW | -0.088 |
Nanto | -0.085 |
Pyro | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
"$:/ | +0.858 |
LW | +0.743 |
Malf | +0.686 |
Pyro | +0.626 |
lf | +0.624 |
Neg logit contributions | |
mosqu | -1.840 |
therap | -1.836 |
enthusi | -1.829 |
metic | -1.802 |
undermin | -1.782 |
Token | V |
Token ID | 53 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
"$:/ | +0.697 |
Malf | +0.616 |
LW | +0.589 |
lf | +0.547 |
spection | +0.544 |
Neg logit contributions | |
subsequ | -0.919 |
therap | -0.879 |
mosqu | -0.863 |
upgr | -0.847 |
erosion | -0.838 |
Token | il |
Token ID | 346 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
"$:/ | +2.998 |
Malf | +2.680 |
LW | +2.606 |
Nanto | +2.545 |
Pyro | +2.443 |
Neg logit contributions | |
subsequ | -2.178 |
therap | -2.018 |
exhib | -1.986 |
upgr | -1.983 |
momentum | -1.946 |