Token | k |
Token ID | 74 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
王 | +0.052 |
dullah | +0.047 |
avis | +0.044 |
alky | +0.044 |
alde | +0.044 |
Neg logit contributions | |
downstream | -0.050 |
ervation | -0.043 |
sonian | -0.042 |
reservation | -0.041 |
Epstein | -0.041 |
Token | icker |
Token ID | 15799 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
downstream | +0.312 |
ervation | +0.276 |
reservation | +0.269 |
sonian | +0.267 |
Epstein | +0.267 |
Neg logit contributions | |
王 | -0.226 |
dullah | -0.199 |
avis | -0.181 |
alky | -0.180 |
alde | -0.180 |
Token | for |
Token ID | 329 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
王 | +1.317 |
dullah | +1.182 |
avis | +1.106 |
alky | +1.096 |
alde | +1.095 |
Neg logit contributions | |
downstream | -1.142 |
ervation | -0.995 |
sonian | -0.967 |
reservation | -0.955 |
GMT | -0.943 |
Token | the |
Token ID | 262 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
王 | +0.150 |
dullah | +0.134 |
avis | +0.124 |
alky | +0.123 |
alde | +0.123 |
Neg logit contributions | |
downstream | -0.162 |
ervation | -0.141 |
reservation | -0.137 |
sonian | -0.136 |
Epstein | -0.136 |
Token | Cincinnati |
Token ID | 16137 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
downstream | +0.101 |
ervation | +0.088 |
reservation | +0.085 |
Epstein | +0.084 |
sonian | +0.084 |
Neg logit contributions | |
王 | -0.096 |
dullah | -0.086 |
alde | -0.079 |
avis | -0.079 |
alky | -0.079 |
Token | Bengals |
Token ID | 24037 |
Feature activation | +4.61e+00 |
Pos logit contributions | |
王 | +0.935 |
dullah | +0.832 |
avis | +0.761 |
alky | +0.759 |
alde | +0.756 |
Neg logit contributions | |
downstream | -1.198 |
ervation | -1.059 |
sonian | -1.029 |
reservation | -1.022 |
Epstein | -1.018 |
Token | out |
Token ID | 503 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
王 | +3.808 |
dullah | +3.373 |
alky | +3.170 |
avis | +3.137 |
alde | +3.134 |
Neg logit contributions | |
downstream | -3.443 |
ervation | -3.019 |
sonian | -2.983 |
GMT | -2.933 |
Epstein | -2.889 |
Token | of |
Token ID | 286 |
Feature activation | -9.41e-02 |
Pos logit contributions | |
downstream | +0.016 |
ervation | +0.013 |
reservation | +0.013 |
sonian | +0.013 |
Epstein | +0.013 |
Neg logit contributions | |
王 | -0.018 |
dullah | -0.016 |
alky | -0.015 |
alde | -0.015 |
avis | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
downstream | +0.076 |
ervation | +0.066 |
reservation | +0.064 |
sonian | +0.064 |
Epstein | +0.064 |
Neg logit contributions | |
王 | -0.075 |
dullah | -0.067 |
avis | -0.062 |
alky | -0.062 |
alde | -0.062 |
Token | University |
Token ID | 2059 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
王 | +0.060 |
dullah | +0.055 |
avis | +0.051 |
alky | +0.050 |
alde | +0.050 |
Neg logit contributions | |
downstream | -0.056 |
ervation | -0.048 |
reservation | -0.046 |
sonian | -0.046 |
Epstein | -0.046 |
Token | of |
Token ID | 286 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
downstream | +1.323 |
ervation | +1.134 |
reservation | +1.096 |
Epstein | +1.084 |
sonian | +1.082 |
Neg logit contributions | |
王 | -1.554 |
dullah | -1.408 |
alky | -1.306 |
alde | -1.306 |
avis | -1.305 |
Token | , |
Token ID | 11 |
Feature activation | -9.53e-03 |
Pos logit contributions | |
王 | +0.342 |
dullah | +0.308 |
avis | +0.288 |
alky | +0.287 |
alde | +0.286 |
Neg logit contributions | |
downstream | -0.295 |
ervation | -0.255 |
sonian | -0.246 |
reservation | -0.245 |
Epstein | -0.243 |
Token | as |
Token ID | 355 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
downstream | +0.007 |
ervation | +0.006 |
reservation | +0.006 |
sonian | +0.006 |
Epstein | +0.006 |
Neg logit contributions | |
王 | -0.008 |
dullah | -0.007 |
alky | -0.007 |
avis | -0.007 |
alde | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
downstream | +0.109 |
ervation | +0.095 |
reservation | +0.092 |
sonian | +0.091 |
Epstein | +0.091 |
Neg logit contributions | |
王 | -0.107 |
dullah | -0.096 |
alky | -0.088 |
avis | -0.088 |
alde | -0.088 |
Token | original |
Token ID | 2656 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
王 | +0.078 |
dullah | +0.070 |
alky | +0.064 |
avis | +0.064 |
alde | +0.064 |
Neg logit contributions | |
downstream | -0.082 |
ervation | -0.072 |
reservation | -0.069 |
sonian | -0.069 |
Epstein | -0.069 |
Token | Cleveland |
Token ID | 10306 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
王 | +1.082 |
dullah | +0.968 |
alky | +0.903 |
alde | +0.894 |
avis | +0.893 |
Neg logit contributions | |
downstream | -1.100 |
ervation | -0.960 |
sonian | -0.923 |
Epstein | -0.919 |
reservation | -0.918 |
Token | Browns |
Token ID | 17671 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
王 | +0.660 |
dullah | +0.588 |
alky | +0.546 |
avis | +0.544 |
alde | +0.543 |
Neg logit contributions | |
downstream | -0.711 |
ervation | -0.623 |
Epstein | -0.599 |
sonian | -0.598 |
reservation | -0.597 |
Token | played |
Token ID | 2826 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
王 | +3.572 |
dullah | +3.129 |
alky | +2.993 |
avis | +2.952 |
bin | +2.942 |
Neg logit contributions | |
downstream | -3.514 |
ervation | -3.089 |
sonian | -3.009 |
GMT | -2.984 |
Epstein | -2.961 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
downstream | +0.675 |
ervation | +0.581 |
reservation | +0.563 |
Epstein | +0.557 |
sonian | +0.554 |
Neg logit contributions | |
王 | -0.733 |
dullah | -0.661 |
alde | -0.613 |
avis | -0.611 |
alky | -0.610 |
Token | won |
Token ID | 1839 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
王 | +0.125 |
dullah | +0.112 |
avis | +0.103 |
alde | +0.103 |
alky | +0.103 |
Neg logit contributions | |
downstream | -0.134 |
ervation | -0.117 |
reservation | -0.113 |
sonian | -0.112 |
Epstein | -0.112 |
Token | their |
Token ID | 511 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
王 | +0.631 |
dullah | +0.567 |
alky | +0.526 |
alde | +0.525 |
avis | +0.525 |
Neg logit contributions | |
downstream | -0.591 |
ervation | -0.511 |
reservation | -0.493 |
sonian | -0.490 |
Epstein | -0.489 |
Token | final |
Token ID | 2457 |
Feature activation | +9.42e-01 |
Pos logit contributions | |
王 | +0.045 |
dullah | +0.041 |
alky | +0.038 |
avis | +0.037 |
alde | +0.037 |
Neg logit contributions | |
downstream | -0.044 |
ervation | -0.038 |
reservation | -0.037 |
sonian | -0.036 |
Epstein | -0.036 |
Token | of |
Token ID | 286 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
王 | +0.875 |
dullah | +0.787 |
avis | +0.734 |
alky | +0.732 |
alde | +0.729 |
Neg logit contributions | |
downstream | -0.804 |
ervation | -0.700 |
sonian | -0.678 |
reservation | -0.673 |
Epstein | -0.668 |
Token | visitors |
Token ID | 9692 |
Feature activation | -4.22e-02 |
Pos logit contributions | |
downstream | +0.258 |
ervation | +0.226 |
reservation | +0.218 |
sonian | +0.217 |
Epstein | +0.217 |
Neg logit contributions | |
王 | -0.239 |
dullah | -0.213 |
avis | -0.196 |
alky | -0.196 |
alde | -0.196 |
Token | and |
Token ID | 290 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
downstream | +0.031 |
ervation | +0.027 |
sonian | +0.026 |
reservation | +0.026 |
Epstein | +0.025 |
Neg logit contributions | |
王 | -0.036 |
dullah | -0.032 |
avis | -0.030 |
alky | -0.030 |
bin | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
downstream | +0.087 |
ervation | +0.076 |
reservation | +0.073 |
sonian | +0.072 |
Epstein | +0.072 |
Neg logit contributions | |
王 | -0.088 |
dullah | -0.079 |
avis | -0.073 |
alky | -0.073 |
alde | -0.073 |
Token | residence |
Token ID | 11566 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
王 | +0.250 |
dullah | +0.225 |
avis | +0.209 |
alky | +0.209 |
alde | +0.208 |
Neg logit contributions | |
downstream | -0.237 |
ervation | -0.206 |
sonian | -0.198 |
reservation | -0.197 |
Epstein | -0.196 |
Token | resembles |
Token ID | 22960 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
王 | +1.338 |
dullah | +1.190 |
alde | +1.107 |
avis | +1.106 |
bin | +1.104 |
Neg logit contributions | |
downstream | -1.260 |
ervation | -1.101 |
sonian | -1.061 |
GMT | -1.049 |
Epstein | -1.046 |
Token | a |
Token ID | 257 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
王 | +3.599 |
dullah | +3.186 |
alky | +3.006 |
bin | +2.988 |
avis | +2.984 |
Neg logit contributions | |
downstream | -3.402 |
ervation | -2.959 |
sonian | -2.836 |
reservation | -2.787 |
GMT | -2.784 |
Token | Hollywood |
Token ID | 8502 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
王 | +0.045 |
dullah | +0.041 |
avis | +0.037 |
alky | +0.037 |
alde | +0.037 |
Neg logit contributions | |
downstream | -0.049 |
ervation | -0.043 |
reservation | -0.041 |
sonian | -0.041 |
Epstein | -0.041 |
Token | - |
Token ID | 12 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
downstream | +0.573 |
ervation | +0.496 |
reservation | +0.483 |
Epstein | +0.478 |
sonian | +0.477 |
Neg logit contributions | |
王 | -0.584 |
dullah | -0.525 |
avis | -0.483 |
alde | -0.483 |
alky | -0.483 |
Token | like |
Token ID | 2339 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
王 | +0.031 |
dullah | +0.028 |
avis | +0.026 |
alky | +0.026 |
alde | +0.026 |
Neg logit contributions | |
downstream | -0.030 |
ervation | -0.026 |
reservation | -0.025 |
Epstein | -0.025 |
sonian | -0.025 |
Token | Christmas |
Token ID | 6786 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
王 | +1.571 |
dullah | +1.395 |
alky | +1.306 |
avis | +1.302 |
bin | +1.298 |
Neg logit contributions | |
downstream | -1.567 |
ervation | -1.370 |
sonian | -1.311 |
Epstein | -1.294 |
reservation | -1.292 |
Token | fight |
Token ID | 1907 |
Feature activation | -3.78e-04 |
Pos logit contributions | |
downstream | +0.823 |
ervation | +0.710 |
reservation | +0.681 |
sonian | +0.679 |
Epstein | +0.677 |
Neg logit contributions | |
王 | -0.948 |
dullah | -0.856 |
alky | -0.795 |
avis | -0.794 |
alde | -0.792 |
Token | broke |
Token ID | 6265 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
downstream | +0.000 |
ervation | +0.000 |
reservation | +0.000 |
sonian | +0.000 |
Epstein | +0.000 |
Neg logit contributions | |
王 | -0.000 |
dullah | -0.000 |
bin | -0.000 |
alky | -0.000 |
alde | -0.000 |
Token | out |
Token ID | 503 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
王 | +1.127 |
dullah | +1.005 |
alky | +0.926 |
bin | +0.924 |
alde | +0.918 |
Neg logit contributions | |
downstream | -1.184 |
ervation | -1.047 |
sonian | -1.010 |
reservation | -1.004 |
Epstein | -1.000 |
Token | , |
Token ID | 11 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
downstream | +0.094 |
ervation | +0.082 |
sonian | +0.079 |
reservation | +0.079 |
Epstein | +0.077 |
Neg logit contributions | |
王 | -0.119 |
dullah | -0.107 |
bin | -0.101 |
alky | -0.100 |
alde | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | +8.32e-02 |
Pos logit contributions | |
王 | +0.130 |
dullah | +0.117 |
alky | +0.108 |
bin | +0.108 |
alde | +0.108 |
Neg logit contributions | |
downstream | -0.115 |
ervation | -0.101 |
sonian | -0.097 |
reservation | -0.097 |
Epstein | -0.095 |
Token | Spl |
Token ID | 13341 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
王 | +0.063 |
dullah | +0.056 |
alky | +0.052 |
alde | +0.052 |
avis | +0.052 |
Neg logit contributions | |
downstream | -0.070 |
ervation | -0.062 |
reservation | -0.060 |
sonian | -0.060 |
Epstein | -0.059 |
Token | inter |
Token ID | 3849 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
王 | +2.192 |
dullah | +1.919 |
avis | +1.724 |
alde | +1.699 |
bin | +1.695 |
Neg logit contributions | |
downstream | -4.690 |
ervation | -4.212 |
reservation | -4.181 |
Duffy | -4.099 |
Epstein | -4.097 |
Token | 's |
Token ID | 338 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
downstream | +0.059 |
ervation | +0.051 |
reservation | +0.049 |
sonian | +0.049 |
Epstein | +0.049 |
Neg logit contributions | |
王 | -0.063 |
dullah | -0.057 |
alky | -0.053 |
avis | -0.053 |
alde | -0.053 |
Token | cage |
Token ID | 17186 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
王 | +0.308 |
dullah | +0.274 |
alky | +0.256 |
bin | +0.255 |
avis | +0.252 |
Neg logit contributions | |
downstream | -0.303 |
ervation | -0.265 |
sonian | -0.258 |
reservation | -0.253 |
GMT | -0.253 |
Token | was |
Token ID | 373 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
王 | +0.262 |
dullah | +0.232 |
alde | +0.213 |
bin | +0.213 |
alky | +0.212 |
Neg logit contributions | |
downstream | -0.328 |
ervation | -0.290 |
sonian | -0.282 |
Epstein | -0.281 |
reservation | -0.280 |
Token | broken |
Token ID | 5445 |
Feature activation | +8.51e-01 |
Pos logit contributions | |
downstream | +0.041 |
ervation | +0.036 |
sonian | +0.035 |
reservation | +0.035 |
Epstein | +0.035 |
Neg logit contributions | |
王 | -0.039 |
dullah | -0.035 |
avis | -0.032 |
bin | -0.032 |
alde | -0.032 |
Token | was |
Token ID | 373 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
王 | +0.327 |
dullah | +0.293 |
alky | +0.271 |
avis | +0.271 |
bin | +0.271 |
Neg logit contributions | |
downstream | -0.341 |
ervation | -0.301 |
reservation | -0.289 |
sonian | -0.288 |
GMT | -0.287 |
Token | traded |
Token ID | 14018 |
Feature activation | +9.80e-01 |
Pos logit contributions | |
downstream | +0.238 |
ervation | +0.209 |
sonian | +0.200 |
reservation | +0.200 |
GMT | +0.197 |
Neg logit contributions | |
王 | -0.246 |
dullah | -0.222 |
alky | -0.205 |
avis | -0.205 |
bin | -0.205 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
王 | +0.815 |
dullah | +0.731 |
bin | +0.683 |
avis | +0.682 |
alky | +0.680 |
Neg logit contributions | |
downstream | -0.735 |
ervation | -0.652 |
sonian | -0.629 |
reservation | -0.623 |
GMT | -0.616 |
Token | the |
Token ID | 262 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
downstream | +0.084 |
ervation | +0.073 |
reservation | +0.071 |
Epstein | +0.071 |
sonian | +0.070 |
Neg logit contributions | |
王 | -0.079 |
dullah | -0.070 |
alde | -0.065 |
avis | -0.065 |
alky | -0.064 |
Token | Cleveland |
Token ID | 10306 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
downstream | +0.133 |
ervation | +0.117 |
reservation | +0.114 |
Epstein | +0.113 |
sonian | +0.113 |
Neg logit contributions | |
王 | -0.110 |
dullah | -0.098 |
alde | -0.089 |
avis | -0.089 |
alky | -0.089 |
Token | Browns |
Token ID | 17671 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
王 | +0.564 |
dullah | +0.506 |
avis | +0.468 |
alky | +0.467 |
bin | +0.465 |
Neg logit contributions | |
downstream | -0.650 |
ervation | -0.575 |
sonian | -0.559 |
Epstein | -0.555 |
reservation | -0.551 |
Token | just |
Token ID | 655 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
王 | +3.418 |
dullah | +3.014 |
bin | +2.907 |
alky | +2.842 |
avis | +2.834 |
Neg logit contributions | |
downstream | -3.285 |
ervation | -2.940 |
sonian | -2.889 |
GMT | -2.842 |
reservation | -2.788 |
Token | so |
Token ID | 523 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
downstream | +0.016 |
ervation | +0.014 |
sonian | +0.013 |
reservation | +0.013 |
Epstein | +0.013 |
Neg logit contributions | |
王 | -0.019 |
dullah | -0.017 |
avis | -0.016 |
alky | -0.016 |
bin | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +1.74e-03 |
Pos logit contributions | |
downstream | +0.701 |
ervation | +0.601 |
reservation | +0.580 |
Epstein | +0.577 |
sonian | +0.574 |
Neg logit contributions | |
王 | -0.796 |
dullah | -0.719 |
alde | -0.668 |
avis | -0.668 |
alky | -0.667 |
Token | Texans |
Token ID | 21012 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
王 | +0.001 |
dullah | +0.001 |
alky | +0.001 |
avis | +0.001 |
alde | +0.001 |
Neg logit contributions | |
downstream | -0.002 |
ervation | -0.001 |
sonian | -0.001 |
reservation | -0.001 |
Epstein | -0.001 |
Token | could |
Token ID | 714 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
王 | +1.729 |
dullah | +1.497 |
bin | +1.388 |
alky | +1.381 |
alde | +1.371 |
Neg logit contributions | |
downstream | -2.303 |
ervation | -2.078 |
sonian | -2.028 |
GMT | -1.992 |
reservation | -1.986 |
Token | death |
Token ID | 1918 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
王 | +0.204 |
dullah | +0.181 |
alky | +0.166 |
avis | +0.166 |
alde | +0.165 |
Neg logit contributions | |
downstream | -0.239 |
ervation | -0.211 |
sonian | -0.204 |
reservation | -0.204 |
Epstein | -0.203 |
Token | of |
Token ID | 286 |
Feature activation | -9.31e-02 |
Pos logit contributions | |
王 | +0.205 |
dullah | +0.185 |
alky | +0.172 |
avis | +0.171 |
alde | +0.171 |
Neg logit contributions | |
downstream | -0.179 |
ervation | -0.155 |
reservation | -0.149 |
sonian | -0.149 |
Epstein | -0.148 |
Token | his |
Token ID | 465 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
downstream | +0.075 |
ervation | +0.065 |
reservation | +0.063 |
sonian | +0.063 |
Epstein | +0.062 |
Neg logit contributions | |
王 | -0.075 |
dullah | -0.067 |
alky | -0.062 |
avis | -0.062 |
alde | -0.062 |
Token | Master |
Token ID | 5599 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
downstream | +0.010 |
ervation | +0.009 |
sonian | +0.009 |
reservation | +0.008 |
Epstein | +0.008 |
Neg logit contributions | |
王 | -0.011 |
dullah | -0.010 |
alky | -0.010 |
alde | -0.010 |
avis | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
王 | +0.954 |
dullah | +0.857 |
alky | +0.798 |
alde | +0.797 |
avis | +0.794 |
Neg logit contributions | |
downstream | -0.827 |
ervation | -0.715 |
reservation | -0.690 |
sonian | -0.684 |
Epstein | -0.681 |
Token | Spl |
Token ID | 13341 |
Feature activation | +4.29e+00 |
Pos logit contributions | |
王 | +0.143 |
dullah | +0.127 |
avis | +0.116 |
alde | +0.116 |
alky | +0.115 |
Neg logit contributions | |
downstream | -0.178 |
ervation | -0.156 |
reservation | -0.152 |
Epstein | -0.151 |
sonian | -0.150 |
Token | inter |
Token ID | 3849 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
王 | +2.147 |
dullah | +1.890 |
avis | +1.681 |
bin | +1.672 |
alde | +1.666 |
Neg logit contributions | |
downstream | -4.634 |
ervation | -4.156 |
reservation | -4.123 |
Duffy | -4.060 |
Epstein | -4.054 |
Token | fled |
Token ID | 11468 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
王 | +0.043 |
dullah | +0.039 |
avis | +0.036 |
alky | +0.036 |
alde | +0.036 |
Neg logit contributions | |
downstream | -0.041 |
ervation | -0.036 |
reservation | -0.034 |
sonian | -0.034 |
Epstein | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
downstream | +0.313 |
ervation | +0.271 |
reservation | +0.261 |
sonian | +0.261 |
Epstein | +0.258 |
Neg logit contributions | |
王 | -0.349 |
dullah | -0.314 |
alky | -0.291 |
alde | -0.291 |
avis | -0.291 |
Token | the |
Token ID | 262 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
王 | +0.256 |
dullah | +0.230 |
avis | +0.212 |
alky | +0.212 |
alde | +0.212 |
Neg logit contributions | |
downstream | -0.250 |
ervation | -0.217 |
reservation | -0.210 |
sonian | -0.209 |
Epstein | -0.208 |
Token | al |
Token ID | 435 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
王 | +0.196 |
dullah | +0.175 |
avis | +0.162 |
alky | +0.162 |
alde | +0.161 |
Neg logit contributions | |
downstream | -0.203 |
ervation | -0.177 |
sonian | -0.171 |
reservation | -0.171 |
Epstein | -0.170 |
Token | Nick |
Token ID | 8047 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
downstream | +0.232 |
ervation | +0.201 |
reservation | +0.195 |
Epstein | +0.194 |
sonian | +0.193 |
Neg logit contributions | |
王 | -0.243 |
dullah | -0.218 |
alde | -0.201 |
alky | -0.201 |
avis | -0.201 |
Token | all |
Token ID | 439 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
downstream | +0.091 |
ervation | +0.079 |
reservation | +0.077 |
Epstein | +0.076 |
sonian | +0.076 |
Neg logit contributions | |
王 | -0.092 |
dullah | -0.082 |
alky | -0.076 |
avis | -0.076 |
alde | -0.076 |
Token | ini |
Token ID | 5362 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
王 | +1.114 |
dullah | +1.001 |
avis | +0.928 |
alde | +0.922 |
alky | +0.921 |
Neg logit contributions | |
downstream | -1.173 |
ervation | -1.021 |
reservation | -0.987 |
sonian | -0.981 |
Epstein | -0.974 |
Token | - |
Token ID | 12 |
Feature activation | +8.79e-01 |
Pos logit contributions | |
downstream | +0.178 |
ervation | +0.155 |
reservation | +0.151 |
Epstein | +0.150 |
sonian | +0.149 |
Neg logit contributions | |
王 | -0.170 |
dullah | -0.152 |
alky | -0.140 |
alde | -0.140 |
avis | -0.139 |
Token | Jim |
Token ID | 18050 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
王 | +0.671 |
dullah | +0.597 |
avis | +0.549 |
alky | +0.548 |
alde | +0.547 |
Neg logit contributions | |
downstream | -0.744 |
ervation | -0.648 |
reservation | -0.630 |
sonian | -0.623 |
Epstein | -0.621 |
Token | Vs |
Token ID | 31280 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
王 | +0.192 |
dullah | +0.173 |
avis | +0.160 |
alde | +0.159 |
alky | +0.159 |
Neg logit contributions | |
downstream | -0.198 |
ervation | -0.173 |
reservation | -0.167 |
sonian | -0.165 |
Epstein | -0.164 |
Token | Colin |
Token ID | 18373 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
王 | +3.312 |
dullah | +2.983 |
avis | +2.806 |
bin | +2.792 |
alky | +2.781 |
Neg logit contributions | |
downstream | -3.431 |
ervation | -2.965 |
reservation | -2.881 |
sonian | -2.845 |
GMT | -2.819 |
Token | - |
Token ID | 12 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
downstream | +0.642 |
ervation | +0.556 |
reservation | +0.537 |
Epstein | +0.533 |
sonian | +0.533 |
Neg logit contributions | |
王 | -0.671 |
dullah | -0.603 |
alde | -0.558 |
alky | -0.558 |
avis | -0.558 |
Token | Ant |
Token ID | 13217 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
王 | +0.788 |
dullah | +0.701 |
avis | +0.645 |
alky | +0.643 |
alde | +0.642 |
Neg logit contributions | |
downstream | -0.870 |
ervation | -0.758 |
reservation | -0.736 |
sonian | -0.728 |
Epstein | -0.726 |
Token | W |
Token ID | 370 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
王 | +0.439 |
dullah | +0.394 |
avis | +0.364 |
alde | +0.362 |
alky | +0.362 |
Neg logit contributions | |
downstream | -0.467 |
ervation | -0.406 |
reservation | -0.394 |
sonian | -0.389 |
Epstein | -0.389 |
Token | ears |
Token ID | 4127 |
Feature activation | -7.12e-01 |
Pos logit contributions | |
王 | +1.186 |
dullah | +1.085 |
avis | +1.024 |
alde | +1.015 |
alky | +1.015 |
Neg logit contributions | |
downstream | -1.029 |
ervation | -0.870 |
reservation | -0.842 |
sonian | -0.834 |
GMT | -0.827 |
Token | the |
Token ID | 262 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
王 | +1.181 |
dullah | +1.058 |
alky | +0.973 |
avis | +0.973 |
alde | +0.968 |
Neg logit contributions | |
downstream | -1.199 |
ervation | -1.052 |
sonian | -1.015 |
reservation | -1.011 |
Epstein | -0.994 |
Token | Din |
Token ID | 23448 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
王 | +0.338 |
dullah | +0.303 |
avis | +0.280 |
alky | +0.280 |
alde | +0.279 |
Neg logit contributions | |
downstream | -0.333 |
ervation | -0.291 |
sonian | -0.281 |
reservation | -0.279 |
Epstein | -0.277 |
Token | ah |
Token ID | 993 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
王 | +0.761 |
dullah | +0.673 |
alde | +0.613 |
avis | +0.613 |
alky | +0.612 |
Neg logit contributions | |
downstream | -1.019 |
ervation | -0.898 |
reservation | -0.874 |
Epstein | -0.872 |
sonian | -0.867 |
Token | story |
Token ID | 1621 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
王 | +1.375 |
dullah | +1.237 |
avis | +1.150 |
alde | +1.145 |
bin | +1.141 |
Neg logit contributions | |
downstream | -1.277 |
ervation | -1.117 |
sonian | -1.075 |
Epstein | -1.062 |
GMT | -1.060 |
Token | like |
Token ID | 588 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
downstream | +0.554 |
ervation | +0.478 |
reservation | +0.462 |
sonian | +0.458 |
Epstein | +0.458 |
Neg logit contributions | |
王 | -0.599 |
dullah | -0.539 |
avis | -0.499 |
alky | -0.499 |
alde | -0.499 |
Token | Nah |
Token ID | 35201 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
王 | +1.631 |
dullah | +1.463 |
alky | +1.356 |
avis | +1.356 |
alde | +1.355 |
Neg logit contributions | |
downstream | -1.531 |
ervation | -1.333 |
sonian | -1.295 |
reservation | -1.277 |
Epstein | -1.256 |
Token | man |
Token ID | 805 |
Feature activation | +5.58e-02 |
Pos logit contributions | |
王 | +3.202 |
dullah | +2.940 |
bin | +2.693 |
avis | +2.671 |
alde | +2.667 |
Neg logit contributions | |
downstream | -3.515 |
ervation | -3.043 |
reservation | -2.983 |
GMT | -2.937 |
Epstein | -2.929 |
Token | ides |
Token ID | 1460 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
王 | +0.044 |
dullah | +0.039 |
alky | +0.036 |
alde | +0.036 |
avis | +0.036 |
Neg logit contributions | |
downstream | -0.046 |
ervation | -0.040 |
reservation | -0.039 |
Epstein | -0.038 |
sonian | -0.038 |
Token | does |
Token ID | 857 |
Feature activation | +8.78e-01 |
Pos logit contributions | |
downstream | +0.032 |
ervation | +0.028 |
reservation | +0.027 |
sonian | +0.027 |
Epstein | +0.027 |
Neg logit contributions | |
王 | -0.034 |
dullah | -0.031 |
avis | -0.029 |
alde | -0.029 |
alky | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
王 | +0.737 |
dullah | +0.666 |
avis | +0.617 |
alde | +0.616 |
alky | +0.615 |
Neg logit contributions | |
downstream | -0.668 |
ervation | -0.578 |
sonian | -0.556 |
reservation | -0.555 |
Epstein | -0.549 |
Token | He |
Token ID | 679 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
王 | +0.194 |
dullah | +0.174 |
avis | +0.160 |
alky | +0.160 |
alde | +0.160 |
Neg logit contributions | |
downstream | -0.203 |
ervation | -0.177 |
reservation | -0.171 |
sonian | -0.170 |
Epstein | -0.169 |
Token | said |
Token ID | 531 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
王 | +0.605 |
dullah | +0.542 |
alde | +0.501 |
avis | +0.500 |
alky | +0.500 |
Neg logit contributions | |
downstream | -0.602 |
ervation | -0.525 |
reservation | -0.507 |
sonian | -0.502 |
Epstein | -0.499 |
Token | that |
Token ID | 326 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
王 | +0.115 |
dullah | +0.104 |
alky | +0.097 |
avis | +0.096 |
alde | +0.096 |
Neg logit contributions | |
downstream | -0.107 |
ervation | -0.093 |
sonian | -0.090 |
reservation | -0.089 |
Epstein | -0.088 |
Token | he |
Token ID | 339 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
王 | +0.027 |
dullah | +0.024 |
alky | +0.022 |
alde | +0.022 |
avis | +0.022 |
Neg logit contributions | |
downstream | -0.026 |
ervation | -0.023 |
sonian | -0.022 |
reservation | -0.022 |
Epstein | -0.022 |
Token | never |
Token ID | 1239 |
Feature activation | +4.41e-01 |
Pos logit contributions | |
王 | +0.116 |
dullah | +0.104 |
avis | +0.095 |
alde | +0.095 |
alky | +0.095 |
Neg logit contributions | |
downstream | -0.123 |
ervation | -0.108 |
reservation | -0.104 |
sonian | -0.104 |
Epstein | -0.103 |
Token | liked |
Token ID | 8288 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
王 | +0.346 |
dullah | +0.310 |
avis | +0.285 |
alky | +0.285 |
alde | +0.285 |
Neg logit contributions | |
downstream | -0.359 |
ervation | -0.314 |
reservation | -0.303 |
sonian | -0.301 |
Epstein | -0.300 |
Token | literal |
Token ID | 18875 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
王 | +0.733 |
dullah | +0.659 |
alky | +0.611 |
avis | +0.610 |
alde | +0.609 |
Neg logit contributions | |
downstream | -0.721 |
ervation | -0.627 |
sonian | -0.604 |
reservation | -0.603 |
Epstein | -0.597 |
Token | things |
Token ID | 1243 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
王 | +3.288 |
dullah | +2.932 |
alky | +2.777 |
bin | +2.740 |
avis | +2.730 |
Neg logit contributions | |
downstream | -3.329 |
ervation | -2.921 |
sonian | -2.869 |
GMT | -2.772 |
reservation | -2.763 |
Token | . |
Token ID | 13 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
downstream | +1.199 |
ervation | +1.033 |
reservation | +0.998 |
Epstein | +0.990 |
sonian | +0.988 |
Neg logit contributions | |
王 | -1.304 |
dullah | -1.174 |
alde | -1.088 |
avis | -1.087 |
alky | -1.087 |
Token | He |
Token ID | 679 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
王 | +0.156 |
dullah | +0.139 |
alky | +0.128 |
avis | +0.128 |
alde | +0.128 |
Neg logit contributions | |
downstream | -0.170 |
ervation | -0.148 |
reservation | -0.143 |
sonian | -0.143 |
Epstein | -0.141 |
Token | liked |
Token ID | 8288 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
downstream | +0.326 |
ervation | +0.283 |
reservation | +0.274 |
sonian | +0.272 |
Epstein | +0.272 |
Neg logit contributions | |
王 | -0.324 |
dullah | -0.291 |
avis | -0.268 |
alde | -0.268 |
alky | -0.268 |
Token | cryptic |
Token ID | 47774 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
王 | +0.704 |
dullah | +0.632 |
alky | +0.585 |
avis | +0.583 |
alde | +0.582 |
Neg logit contributions | |
downstream | -0.717 |
ervation | -0.629 |
sonian | -0.605 |
reservation | -0.603 |
Epstein | -0.599 |
Token | - |
Token ID | 12 |
Feature activation | +6.55e-01 |
Pos logit contributions | |
王 | +0.560 |
dullah | +0.501 |
avis | +0.463 |
bin | +0.463 |
alky | +0.460 |
Neg logit contributions | |
downstream | -0.579 |
ervation | -0.503 |
reservation | -0.488 |
sonian | -0.480 |
Epstein | -0.476 |
Token | Black |
Token ID | 9915 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
王 | +0.501 |
dullah | +0.446 |
avis | +0.411 |
alky | +0.410 |
alde | +0.409 |
Neg logit contributions | |
downstream | -0.553 |
ervation | -0.482 |
reservation | -0.468 |
sonian | -0.463 |
Epstein | -0.462 |
Token | President |
Token ID | 1992 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
downstream | +0.218 |
ervation | +0.191 |
reservation | +0.185 |
Epstein | +0.184 |
sonian | +0.184 |
Neg logit contributions | |
王 | -0.198 |
dullah | -0.177 |
avis | -0.162 |
alky | -0.162 |
alde | -0.162 |
Token | - |
Token ID | 12 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
王 | +1.243 |
dullah | +1.121 |
avis | +1.036 |
alky | +1.031 |
alde | +1.031 |
Neg logit contributions | |
downstream | -1.282 |
ervation | -1.112 |
reservation | -1.078 |
sonian | -1.066 |
Epstein | -1.059 |
Token | White |
Token ID | 12256 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
王 | +0.483 |
dullah | +0.429 |
avis | +0.395 |
alky | +0.394 |
alde | +0.393 |
Neg logit contributions | |
downstream | -0.543 |
ervation | -0.474 |
reservation | -0.461 |
sonian | -0.456 |
Epstein | -0.455 |
Token | Vs |
Token ID | 31280 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
王 | +0.534 |
dullah | +0.478 |
avis | +0.440 |
alde | +0.440 |
alky | +0.440 |
Neg logit contributions | |
downstream | -0.562 |
ervation | -0.490 |
reservation | -0.474 |
sonian | -0.470 |
Epstein | -0.469 |
Token | Black |
Token ID | 2619 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
王 | +3.364 |
dullah | +3.057 |
bin | +2.869 |
alky | +2.839 |
alde | +2.835 |
Neg logit contributions | |
downstream | -3.272 |
ervation | -2.784 |
reservation | -2.705 |
GMT | -2.663 |
sonian | -2.658 |
Token | - |
Token ID | 12 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
downstream | +0.162 |
ervation | +0.140 |
reservation | +0.136 |
sonian | +0.135 |
Epstein | +0.135 |
Neg logit contributions | |
王 | -0.160 |
dullah | -0.144 |
avis | -0.132 |
alky | -0.132 |
alde | -0.132 |
Token | Jer |
Token ID | 36134 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
王 | +0.548 |
dullah | +0.487 |
avis | +0.448 |
alky | +0.446 |
alde | +0.446 |
Neg logit contributions | |
downstream | -0.611 |
ervation | -0.532 |
reservation | -0.517 |
sonian | -0.511 |
Epstein | -0.510 |
Token | kin |
Token ID | 5116 |
Feature activation | -9.15e-01 |
Pos logit contributions | |
王 | +2.656 |
dullah | +2.451 |
avis | +2.308 |
bin | +2.273 |
alky | +2.251 |
Neg logit contributions | |
downstream | -2.473 |
ervation | -2.113 |
reservation | -2.082 |
GMT | -2.037 |
sonian | -2.002 |
Token | Off |
Token ID | 3242 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
downstream | +0.751 |
ervation | +0.657 |
reservation | +0.636 |
Epstein | +0.634 |
sonian | +0.632 |
Neg logit contributions | |
王 | -0.716 |
dullah | -0.637 |
alde | -0.587 |
alky | -0.586 |
avis | -0.585 |
Token | Qu |
Token ID | 2264 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
downstream | +0.276 |
ervation | +0.240 |
reservation | +0.232 |
sonian | +0.230 |
Epstein | +0.230 |
Neg logit contributions | |
王 | -0.275 |
dullah | -0.246 |
avis | -0.227 |
alde | -0.227 |
alky | -0.227 |
Token | otes |
Token ID | 6421 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
王 | +1.011 |
dullah | +0.923 |
avis | +0.859 |
alky | +0.855 |
alde | +0.854 |
Neg logit contributions | |
downstream | -1.025 |
ervation | -0.882 |
reservation | -0.859 |
sonian | -0.855 |
Epstein | -0.845 |
Token | - |
Token ID | 12 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
王 | +0.821 |
dullah | +0.736 |
avis | +0.679 |
alky | +0.678 |
alde | +0.677 |
Neg logit contributions | |
downstream | -0.843 |
ervation | -0.733 |
reservation | -0.710 |
sonian | -0.705 |
Epstein | -0.702 |
Token | Fat |
Token ID | 33804 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
王 | +0.509 |
dullah | +0.453 |
avis | +0.417 |
alky | +0.416 |
alde | +0.415 |
Neg logit contributions | |
downstream | -0.556 |
ervation | -0.484 |
reservation | -0.471 |
sonian | -0.465 |
Epstein | -0.464 |
Token | Girl |
Token ID | 7430 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
王 | +2.674 |
dullah | +2.413 |
avis | +2.240 |
bin | +2.216 |
alky | +2.209 |
Neg logit contributions | |
downstream | -2.926 |
ervation | -2.551 |
reservation | -2.470 |
sonian | -2.448 |
GMT | -2.428 |
Token | Vs |
Token ID | 31280 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
王 | +1.228 |
dullah | +1.105 |
avis | +1.021 |
alky | +1.017 |
alde | +1.014 |
Neg logit contributions | |
downstream | -1.311 |
ervation | -1.141 |
reservation | -1.105 |
sonian | -1.096 |
Epstein | -1.085 |
Token | U |
Token ID | 471 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
王 | +3.284 |
dullah | +2.964 |
avis | +2.766 |
bin | +2.759 |
alky | +2.743 |
Neg logit contributions | |
downstream | -3.303 |
ervation | -2.844 |
reservation | -2.765 |
sonian | -2.735 |
GMT | -2.715 |
Token | gly |
Token ID | 10853 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
王 | +0.806 |
dullah | +0.731 |
avis | +0.676 |
bin | +0.672 |
alky | +0.669 |
Neg logit contributions | |
downstream | -0.865 |
ervation | -0.756 |
reservation | -0.734 |
Epstein | -0.719 |
GMT | -0.718 |
Token | Girl |
Token ID | 7430 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
downstream | +0.384 |
ervation | +0.338 |
reservation | +0.328 |
Epstein | +0.328 |
sonian | +0.326 |
Neg logit contributions | |
王 | -0.319 |
dullah | -0.282 |
alde | -0.258 |
alky | -0.258 |
avis | -0.258 |
Token | - |
Token ID | 12 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
王 | +1.411 |
dullah | +1.265 |
avis | +1.166 |
alky | +1.164 |
alde | +1.162 |
Neg logit contributions | |
downstream | -1.482 |
ervation | -1.291 |
reservation | -1.251 |
sonian | -1.240 |
Epstein | -1.236 |
Token | False |
Token ID | 25101 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
王 | +0.731 |
dullah | +0.650 |
avis | +0.599 |
alky | +0.596 |
alde | +0.596 |
Neg logit contributions | |
downstream | -0.804 |
ervation | -0.700 |
reservation | -0.681 |
sonian | -0.673 |
Epstein | -0.671 |
Token | finished |
Token ID | 5201 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
王 | +0.126 |
dullah | +0.114 |
avis | +0.106 |
alky | +0.105 |
alde | +0.105 |
Neg logit contributions | |
downstream | -0.115 |
ervation | -0.099 |
reservation | -0.095 |
sonian | -0.095 |
Epstein | -0.095 |
Token | product |
Token ID | 1720 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
downstream | +0.177 |
ervation | +0.151 |
Epstein | +0.144 |
reservation | +0.144 |
sonian | +0.143 |
Neg logit contributions | |
王 | -0.237 |
dullah | -0.216 |
alky | -0.203 |
alde | -0.202 |
avis | -0.202 |
Token | wasn |
Token ID | 2492 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
downstream | +0.291 |
ervation | +0.254 |
reservation | +0.246 |
Epstein | +0.244 |
sonian | +0.244 |
Neg logit contributions | |
王 | -0.276 |
dullah | -0.247 |
alky | -0.228 |
avis | -0.228 |
alde | -0.227 |
Token | 't |
Token ID | 470 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
downstream | +0.156 |
ervation | +0.129 |
reservation | +0.125 |
Epstein | +0.122 |
sonian | +0.121 |
Neg logit contributions | |
王 | -0.233 |
dullah | -0.213 |
bin | -0.200 |
avis | -0.200 |
alky | -0.199 |
Token | a |
Token ID | 257 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
downstream | +0.749 |
ervation | +0.649 |
reservation | +0.627 |
sonian | +0.622 |
Epstein | +0.622 |
Neg logit contributions | |
王 | -0.780 |
dullah | -0.702 |
avis | -0.649 |
alde | -0.649 |
alky | -0.649 |
Token | literal |
Token ID | 18875 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
downstream | +0.025 |
ervation | +0.022 |
reservation | +0.021 |
Epstein | +0.021 |
sonian | +0.021 |
Neg logit contributions | |
王 | -0.026 |
dullah | -0.023 |
alky | -0.021 |
avis | -0.021 |
alde | -0.021 |
Token | translation |
Token ID | 11059 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
王 | +3.360 |
dullah | +3.049 |
alky | +2.869 |
alde | +2.858 |
avis | +2.827 |
Neg logit contributions | |
downstream | -3.135 |
ervation | -2.709 |
Epstein | -2.610 |
sonian | -2.602 |
reservation | -2.580 |
Token | of |
Token ID | 286 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
王 | +1.169 |
dullah | +1.057 |
alky | +0.981 |
alde | +0.977 |
bin | +0.976 |
Neg logit contributions | |
downstream | -1.079 |
ervation | -0.929 |
reservation | -0.898 |
sonian | -0.893 |
Epstein | -0.893 |
Token | Bass |
Token ID | 19829 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
downstream | +0.023 |
ervation | +0.020 |
reservation | +0.019 |
sonian | +0.019 |
Epstein | +0.019 |
Neg logit contributions | |
王 | -0.023 |
dullah | -0.020 |
alky | -0.019 |
avis | -0.019 |
alde | -0.019 |
Token | ' |
Token ID | 6 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
downstream | +0.379 |
ervation | +0.328 |
reservation | +0.317 |
Epstein | +0.315 |
sonian | +0.315 |
Neg logit contributions | |
王 | -0.396 |
dullah | -0.356 |
avis | -0.329 |
alde | -0.329 |
alky | -0.329 |
Token | concept |
Token ID | 3721 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
王 | +0.978 |
dullah | +0.883 |
alky | +0.821 |
alde | +0.819 |
avis | +0.819 |
Neg logit contributions | |
downstream | -0.892 |
ervation | -0.767 |
sonian | -0.740 |
reservation | -0.740 |
Epstein | -0.738 |
Token | Said |
Token ID | 23885 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
downstream | +0.106 |
ervation | +0.092 |
reservation | +0.089 |
sonian | +0.089 |
Epstein | +0.088 |
Neg logit contributions | |
王 | -0.097 |
dullah | -0.087 |
avis | -0.080 |
alky | -0.079 |
alde | -0.079 |
Token | South |
Token ID | 2520 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
王 | +0.194 |
dullah | +0.171 |
avis | +0.156 |
alde | +0.155 |
alky | +0.155 |
Neg logit contributions | |
downstream | -0.263 |
ervation | -0.233 |
reservation | -0.226 |
sonian | -0.225 |
Epstein | -0.223 |
Token | Carolina |
Token ID | 5913 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
downstream | +0.145 |
ervation | +0.129 |
reservation | +0.125 |
sonian | +0.125 |
Epstein | +0.124 |
Neg logit contributions | |
王 | -0.098 |
dullah | -0.086 |
avis | -0.077 |
alde | -0.077 |
alky | -0.077 |
Token | coach |
Token ID | 3985 |
Feature activation | +7.88e-01 |
Pos logit contributions | |
王 | +1.395 |
dullah | +1.244 |
alky | +1.155 |
alde | +1.153 |
avis | +1.153 |
Neg logit contributions | |
downstream | -1.424 |
ervation | -1.245 |
sonian | -1.204 |
Epstein | -1.192 |
reservation | -1.191 |
Token | Steve |
Token ID | 6542 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
王 | +0.617 |
dullah | +0.552 |
avis | +0.512 |
alky | +0.512 |
alde | +0.509 |
Neg logit contributions | |
downstream | -0.645 |
ervation | -0.562 |
reservation | -0.545 |
sonian | -0.541 |
Epstein | -0.537 |
Token | Spur |
Token ID | 47788 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
downstream | +0.318 |
ervation | +0.287 |
Epstein | +0.280 |
reservation | +0.278 |
Duffy | +0.278 |
Neg logit contributions | |
王 | -0.181 |
dullah | -0.153 |
alky | -0.135 |
alde | -0.134 |
avis | -0.134 |
Token | rier |
Token ID | 5277 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
王 | +3.093 |
dullah | +2.812 |
bin | +2.645 |
alde | +2.606 |
avis | +2.542 |
Neg logit contributions | |
downstream | -3.408 |
ervation | -3.018 |
reservation | -2.925 |
GMT | -2.889 |
Epstein | -2.865 |
Token | , |
Token ID | 11 |
Feature activation | -8.12e-02 |
Pos logit contributions | |
downstream | +0.660 |
ervation | +0.570 |
reservation | +0.551 |
sonian | +0.546 |
Epstein | +0.546 |
Neg logit contributions | |
王 | -0.704 |
dullah | -0.634 |
avis | -0.586 |
alde | -0.586 |
alky | -0.586 |
Token | according |
Token ID | 1864 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
downstream | +0.072 |
ervation | +0.064 |
reservation | +0.062 |
sonian | +0.061 |
Epstein | +0.061 |
Neg logit contributions | |
王 | -0.058 |
dullah | -0.052 |
avis | -0.047 |
alky | -0.047 |
alde | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
王 | +0.063 |
dullah | +0.057 |
avis | +0.053 |
alky | +0.053 |
alde | +0.053 |
Neg logit contributions | |
downstream | -0.057 |
ervation | -0.049 |
reservation | -0.047 |
sonian | -0.047 |
Epstein | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
王 | +0.076 |
dullah | +0.068 |
avis | +0.062 |
alky | +0.062 |
alde | +0.062 |
Neg logit contributions | |
downstream | -0.082 |
ervation | -0.071 |
reservation | -0.069 |
sonian | -0.069 |
Epstein | -0.068 |
Token | a |
Token ID | 257 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
王 | +0.096 |
dullah | +0.087 |
avis | +0.081 |
alky | +0.080 |
alde | +0.080 |
Neg logit contributions | |
downstream | -0.087 |
ervation | -0.075 |
sonian | -0.073 |
reservation | -0.072 |
Epstein | -0.071 |
Token | break |
Token ID | 2270 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
downstream | +0.111 |
ervation | +0.096 |
sonian | +0.092 |
reservation | +0.092 |
Epstein | +0.092 |
Neg logit contributions | |
王 | -0.119 |
dullah | -0.107 |
alky | -0.099 |
avis | -0.099 |
alde | -0.099 |
Token | neck |
Token ID | 27235 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
王 | +0.995 |
dullah | +0.904 |
alde | +0.838 |
bin | +0.834 |
alky | +0.834 |
Neg logit contributions | |
downstream | -0.831 |
ervation | -0.707 |
reservation | -0.677 |
sonian | -0.677 |
Epstein | -0.671 |
Token | schedule |
Token ID | 7269 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
downstream | +1.000 |
ervation | +0.865 |
reservation | +0.832 |
sonian | +0.829 |
Epstein | +0.825 |
Neg logit contributions | |
王 | -1.093 |
dullah | -0.984 |
avis | -0.912 |
alky | -0.912 |
alde | -0.911 |
Token | that |
Token ID | 326 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
downstream | +0.967 |
ervation | +0.829 |
sonian | +0.797 |
reservation | +0.797 |
Epstein | +0.786 |
Neg logit contributions | |
王 | -1.182 |
dullah | -1.070 |
avis | -0.996 |
alde | -0.996 |
alky | -0.995 |
Token | resembles |
Token ID | 22960 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
downstream | +0.228 |
ervation | +0.196 |
reservation | +0.188 |
sonian | +0.188 |
Epstein | +0.185 |
Neg logit contributions | |
王 | -0.263 |
dullah | -0.238 |
alky | -0.221 |
avis | -0.221 |
alde | -0.221 |
Token | a |
Token ID | 257 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
王 | +3.325 |
dullah | +2.964 |
avis | +2.797 |
alky | +2.787 |
bin | +2.749 |
Neg logit contributions | |
downstream | -3.131 |
ervation | -2.709 |
sonian | -2.657 |
reservation | -2.566 |
GMT | -2.566 |
Token | road |
Token ID | 2975 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
downstream | +0.102 |
ervation | +0.089 |
reservation | +0.086 |
sonian | +0.085 |
Epstein | +0.085 |
Neg logit contributions | |
王 | -0.102 |
dullah | -0.091 |
avis | -0.084 |
alde | -0.084 |
alky | -0.084 |
Token | trip |
Token ID | 5296 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
王 | +0.042 |
dullah | +0.038 |
alky | +0.035 |
avis | +0.035 |
alde | +0.035 |
Neg logit contributions | |
downstream | -0.041 |
ervation | -0.035 |
reservation | -0.034 |
sonian | -0.034 |
Epstein | -0.034 |
Token | for |
Token ID | 329 |
Feature activation | -3.00e-01 |
Pos logit contributions | |
downstream | +0.361 |
ervation | +0.311 |
sonian | +0.302 |
reservation | +0.296 |
Epstein | +0.295 |
Neg logit contributions | |
王 | -0.429 |
dullah | -0.388 |
avis | -0.362 |
alky | -0.361 |
alde | -0.360 |
Token | a |
Token ID | 257 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
downstream | +0.241 |
ervation | +0.210 |
sonian | +0.202 |
reservation | +0.202 |
Epstein | +0.200 |
Neg logit contributions | |
王 | -0.240 |
dullah | -0.216 |
alky | -0.199 |
avis | -0.199 |
alde | -0.199 |
Token | - |
Token ID | 12 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
downstream | +0.186 |
ervation | +0.165 |
Epstein | +0.161 |
reservation | +0.161 |
sonian | +0.160 |
Neg logit contributions | |
王 | -0.138 |
dullah | -0.121 |
alky | -0.110 |
avis | -0.110 |
alde | -0.110 |
Token | G |
Token ID | 38 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
王 | +0.679 |
dullah | +0.604 |
avis | +0.556 |
alky | +0.555 |
alde | +0.554 |
Neg logit contributions | |
downstream | -0.749 |
ervation | -0.653 |
reservation | -0.634 |
sonian | -0.627 |
Epstein | -0.626 |
Token | ambling |
Token ID | 15366 |
Feature activation | +8.98e-01 |
Pos logit contributions | |
王 | +0.433 |
dullah | +0.384 |
avis | +0.352 |
alky | +0.351 |
alde | +0.351 |
Neg logit contributions | |
downstream | -0.522 |
ervation | -0.459 |
reservation | -0.446 |
sonian | -0.442 |
Epstein | -0.442 |
Token | - |
Token ID | 12 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
王 | +0.706 |
dullah | +0.633 |
avis | +0.584 |
alky | +0.582 |
alde | +0.582 |
Neg logit contributions | |
downstream | -0.735 |
ervation | -0.639 |
reservation | -0.619 |
sonian | -0.613 |
Epstein | -0.611 |
Token | Super |
Token ID | 12442 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
王 | +0.661 |
dullah | +0.588 |
avis | +0.541 |
alky | +0.540 |
alde | +0.539 |
Neg logit contributions | |
downstream | -0.733 |
ervation | -0.639 |
reservation | -0.621 |
sonian | -0.614 |
Epstein | -0.612 |
Token | bowl |
Token ID | 36859 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
downstream | +0.840 |
ervation | +0.753 |
reservation | +0.732 |
Epstein | +0.730 |
sonian | +0.728 |
Neg logit contributions | |
王 | -0.553 |
dullah | -0.478 |
alky | -0.433 |
alde | -0.429 |
avis | -0.428 |
Token | Advertising |
Token ID | 40113 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
王 | +3.034 |
dullah | +2.730 |
bin | +2.532 |
avis | +2.512 |
alky | +2.509 |
Neg logit contributions | |
downstream | -3.467 |
ervation | -3.023 |
reservation | -2.907 |
sonian | -2.866 |
Epstein | -2.851 |
Token | - |
Token ID | 12 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
王 | +0.769 |
dullah | +0.690 |
avis | +0.637 |
alky | +0.636 |
alde | +0.635 |
Neg logit contributions | |
downstream | -0.775 |
ervation | -0.673 |
reservation | -0.651 |
sonian | -0.646 |
Epstein | -0.645 |
Token | Aust |
Token ID | 15160 |
Feature activation | +7.00e-01 |
Pos logit contributions | |
王 | +0.588 |
dullah | +0.523 |
avis | +0.481 |
alky | +0.480 |
alde | +0.479 |
Neg logit contributions | |
downstream | -0.654 |
ervation | -0.571 |
reservation | -0.554 |
sonian | -0.548 |
Epstein | -0.547 |
Token | rail |
Token ID | 30224 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
王 | +0.454 |
dullah | +0.403 |
avis | +0.366 |
alky | +0.364 |
alde | +0.362 |
Neg logit contributions | |
downstream | -0.669 |
ervation | -0.591 |
reservation | -0.578 |
sonian | -0.568 |
Epstein | -0.567 |
Token | ian |
Token ID | 666 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
王 | +0.311 |
dullah | +0.283 |
avis | +0.263 |
alde | +0.263 |
alky | +0.262 |
Neg logit contributions | |
downstream | -0.261 |
ervation | -0.223 |
reservation | -0.214 |
sonian | -0.212 |
Epstein | -0.212 |
Token | Ads |
Token ID | 47442 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
王 | +0.508 |
dullah | +0.455 |
avis | +0.420 |
alky | +0.419 |
alde | +0.419 |
Neg logit contributions | |
downstream | -0.522 |
ervation | -0.454 |
reservation | -0.439 |
sonian | -0.436 |
Epstein | -0.435 |
Token | - |
Token ID | 12 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
王 | +0.907 |
dullah | +0.816 |
avis | +0.755 |
alky | +0.752 |
alde | +0.752 |
Neg logit contributions | |
downstream | -0.893 |
ervation | -0.774 |
reservation | -0.750 |
sonian | -0.743 |
Epstein | -0.741 |
Token | R |
Token ID | 49 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
王 | +0.437 |
dullah | +0.390 |
avis | +0.359 |
alky | +0.358 |
alde | +0.357 |
Neg logit contributions | |
downstream | -0.480 |
ervation | -0.418 |
reservation | -0.406 |
sonian | -0.402 |
Epstein | -0.401 |
Token | ac |
Token ID | 330 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
王 | +0.644 |
dullah | +0.578 |
avis | +0.535 |
alky | +0.532 |
alde | +0.532 |
Neg logit contributions | |
downstream | -0.711 |
ervation | -0.620 |
reservation | -0.601 |
sonian | -0.597 |
Epstein | -0.594 |
Token | ism |
Token ID | 1042 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
downstream | +0.580 |
ervation | +0.499 |
reservation | +0.483 |
sonian | +0.479 |
Epstein | +0.478 |
Neg logit contributions | |
王 | -0.639 |
dullah | -0.577 |
avis | -0.536 |
alde | -0.535 |
alky | -0.535 |
Token | Vs |
Token ID | 31280 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
downstream | +0.615 |
ervation | +0.537 |
reservation | +0.519 |
Epstein | +0.517 |
sonian | +0.516 |
Neg logit contributions | |
王 | -0.586 |
dullah | -0.522 |
alde | -0.481 |
alky | -0.481 |
avis | -0.481 |
Token | Xbox |
Token ID | 9445 |
Feature activation | +9.16e-05 |
Pos logit contributions | |
王 | +3.241 |
dullah | +2.943 |
bin | +2.760 |
avis | +2.756 |
alky | +2.712 |
Neg logit contributions | |
downstream | -3.157 |
ervation | -2.696 |
sonian | -2.598 |
reservation | -2.597 |
GMT | -2.564 |
Token | - |
Token ID | 12 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
王 | +0.000 |
dullah | +0.000 |
avis | +0.000 |
alky | +0.000 |
alde | +0.000 |
Neg logit contributions | |
downstream | -0.000 |
ervation | -0.000 |
reservation | -0.000 |
sonian | -0.000 |
Epstein | -0.000 |
Token | Lou |
Token ID | 24016 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
王 | +0.552 |
dullah | +0.491 |
avis | +0.452 |
alky | +0.451 |
alde | +0.451 |
Neg logit contributions | |
downstream | -0.600 |
ervation | -0.523 |
reservation | -0.508 |
sonian | -0.502 |
Epstein | -0.501 |
Token | C |
Token ID | 327 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
王 | +1.311 |
dullah | +1.136 |
avis | +1.027 |
alde | +1.020 |
bin | +1.019 |
Neg logit contributions | |
downstream | -2.132 |
ervation | -1.905 |
reservation | -1.852 |
sonian | -1.827 |
Epstein | -1.819 |
Token | k |
Token ID | 74 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
downstream | +0.089 |
ervation | +0.078 |
reservation | +0.075 |
sonian | +0.075 |
Epstein | +0.074 |
Neg logit contributions | |
王 | -0.085 |
dullah | -0.076 |
avis | -0.070 |
alde | -0.070 |
alky | -0.070 |
Token | � |
Token ID | 247 |
Feature activation | -1.18e-03 |
Pos logit contributions | |
downstream | +0.124 |
ervation | +0.108 |
reservation | +0.106 |
Epstein | +0.105 |
sonian | +0.104 |
Neg logit contributions | |
王 | -0.113 |
dullah | -0.101 |
alky | -0.092 |
alde | -0.092 |
avis | -0.092 |
Token | s |
Token ID | 82 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
downstream | +0.001 |
ervation | +0.001 |
reservation | +0.001 |
Epstein | +0.001 |
sonian | +0.001 |
Neg logit contributions | |
王 | -0.001 |
dullah | -0.001 |
alde | -0.001 |
avis | -0.001 |
alky | -0.001 |
Token | best |
Token ID | 1266 |
Feature activation | -8.99e-01 |
Pos logit contributions | |
downstream | +0.015 |
ervation | +0.013 |
reservation | +0.013 |
sonian | +0.013 |
Epstein | +0.013 |
Neg logit contributions | |
王 | -0.015 |
dullah | -0.014 |
alky | -0.013 |
avis | -0.013 |
alde | -0.013 |
Token | wel |
Token ID | 5029 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
downstream | +0.728 |
ervation | +0.633 |
reservation | +0.612 |
Epstein | +0.608 |
sonian | +0.607 |
Neg logit contributions | |
王 | -0.714 |
dullah | -0.640 |
avis | -0.590 |
alde | -0.590 |
alky | -0.589 |
Token | ter |
Token ID | 353 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
王 | +0.956 |
dullah | +0.864 |
avis | +0.803 |
alky | +0.803 |
alde | +0.803 |
Neg logit contributions | |
downstream | -0.812 |
ervation | -0.695 |
reservation | -0.671 |
sonian | -0.665 |
Epstein | -0.665 |
Token | weight |
Token ID | 6551 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
downstream | +0.541 |
ervation | +0.494 |
reservation | +0.484 |
Epstein | +0.483 |
sonian | +0.482 |
Neg logit contributions | |
王 | -0.151 |
dullah | -0.114 |
alde | -0.091 |
avis | -0.091 |
alky | -0.090 |
Token | while |
Token ID | 981 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
王 | +3.325 |
dullah | +2.949 |
alde | +2.761 |
alky | +2.760 |
bin | +2.759 |
Neg logit contributions | |
downstream | -3.100 |
ervation | -2.686 |
sonian | -2.627 |
reservation | -2.566 |
GMT | -2.561 |
Token | suffering |
Token ID | 7195 |
Feature activation | +9.26e-01 |
Pos logit contributions | |
downstream | +0.378 |
ervation | +0.326 |
sonian | +0.314 |
reservation | +0.313 |
Epstein | +0.309 |
Neg logit contributions | |
王 | -0.435 |
dullah | -0.393 |
alky | -0.366 |
avis | -0.364 |
alde | -0.364 |
Token | double |
Token ID | 4274 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
王 | +0.801 |
dullah | +0.718 |
alky | +0.671 |
bin | +0.670 |
alde | +0.665 |
Neg logit contributions | |
downstream | -0.680 |
ervation | -0.586 |
sonian | -0.569 |
reservation | -0.564 |
Epstein | -0.556 |
Token | vision |
Token ID | 5761 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
王 | +0.489 |
dullah | +0.432 |
alky | +0.401 |
alde | +0.399 |
avis | +0.398 |
Neg logit contributions | |
downstream | -0.531 |
ervation | -0.468 |
sonian | -0.450 |
reservation | -0.450 |
Epstein | -0.448 |
Token | and |
Token ID | 290 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
downstream | +0.171 |
ervation | +0.147 |
sonian | +0.142 |
reservation | +0.141 |
Epstein | +0.141 |
Neg logit contributions | |
王 | -0.204 |
dullah | -0.185 |
alde | -0.172 |
avis | -0.171 |
bin | -0.171 |
Token | s |
Token ID | 82 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
王 | +0.206 |
dullah | +0.185 |
avis | +0.170 |
alde | +0.170 |
alky | +0.169 |
Neg logit contributions | |
downstream | -0.224 |
ervation | -0.195 |
reservation | -0.189 |
sonian | -0.188 |
Epstein | -0.188 |
Token | unlikely |
Token ID | 7485 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
王 | +0.066 |
dullah | +0.060 |
avis | +0.055 |
alky | +0.055 |
alde | +0.055 |
Neg logit contributions | |
downstream | -0.065 |
ervation | -0.057 |
reservation | -0.054 |
sonian | -0.054 |
Epstein | -0.054 |
Token | that |
Token ID | 326 |
Feature activation | -8.48e-02 |
Pos logit contributions | |
王 | +1.203 |
dullah | +1.094 |
alky | +1.013 |
avis | +1.011 |
alde | +1.010 |
Neg logit contributions | |
downstream | -1.085 |
ervation | -0.933 |
reservation | -0.897 |
sonian | -0.892 |
GMT | -0.884 |
Token | Allen |
Token ID | 9659 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
downstream | +0.068 |
ervation | +0.059 |
reservation | +0.057 |
sonian | +0.057 |
Epstein | +0.057 |
Neg logit contributions | |
王 | -0.067 |
dullah | -0.061 |
avis | -0.056 |
alky | -0.056 |
alde | -0.056 |
Token | will |
Token ID | 481 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
downstream | +0.201 |
ervation | +0.171 |
reservation | +0.165 |
sonian | +0.163 |
Epstein | +0.163 |
Neg logit contributions | |
王 | -0.247 |
dullah | -0.224 |
avis | -0.209 |
alde | -0.209 |
alky | -0.209 |
Token | tread |
Token ID | 23153 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
downstream | +0.447 |
ervation | +0.396 |
reservation | +0.385 |
Epstein | +0.382 |
sonian | +0.382 |
Neg logit contributions | |
王 | -0.335 |
dullah | -0.294 |
alde | -0.267 |
alky | -0.267 |
avis | -0.267 |
Token | down |
Token ID | 866 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
王 | +3.334 |
dullah | +3.004 |
avis | +2.870 |
bin | +2.836 |
alky | +2.800 |
Neg logit contributions | |
downstream | -2.968 |
ervation | -2.620 |
sonian | -2.499 |
reservation | -2.482 |
Epstein | -2.479 |
Token | this |
Token ID | 428 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
downstream | +0.105 |
ervation | +0.091 |
reservation | +0.087 |
sonian | +0.087 |
Epstein | +0.086 |
Neg logit contributions | |
王 | -0.114 |
dullah | -0.102 |
avis | -0.095 |
alky | -0.095 |
alde | -0.095 |
Token | cinematic |
Token ID | 29932 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
王 | +0.554 |
dullah | +0.496 |
avis | +0.467 |
alky | +0.466 |
alde | +0.463 |
Neg logit contributions | |
downstream | -0.460 |
ervation | -0.392 |
sonian | -0.378 |
reservation | -0.374 |
Epstein | -0.374 |
Token | path |
Token ID | 3108 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
downstream | +0.468 |
ervation | +0.403 |
sonian | +0.387 |
reservation | +0.387 |
Epstein | +0.385 |
Neg logit contributions | |
王 | -0.528 |
dullah | -0.475 |
avis | -0.442 |
alky | -0.442 |
alde | -0.441 |
Token | again |
Token ID | 757 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
downstream | +1.005 |
ervation | +0.881 |
reservation | +0.854 |
Epstein | +0.847 |
sonian | +0.847 |
Neg logit contributions | |
王 | -0.891 |
dullah | -0.793 |
alky | -0.727 |
alde | -0.726 |
avis | -0.726 |
Token | Mad |
Token ID | 18454 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
downstream | +0.168 |
ervation | +0.147 |
reservation | +0.143 |
sonian | +0.142 |
Epstein | +0.142 |
Neg logit contributions | |
王 | -0.156 |
dullah | -0.139 |
alde | -0.128 |
alky | -0.128 |
avis | -0.128 |
Token | Men |
Token ID | 6065 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
王 | +1.223 |
dullah | +1.098 |
avis | +1.020 |
alky | +1.016 |
alde | +1.014 |
Neg logit contributions | |
downstream | -1.206 |
ervation | -1.049 |
reservation | -1.013 |
sonian | -1.006 |
Epstein | -1.002 |
Token | " |
Token ID | 1 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
王 | +1.626 |
dullah | +1.455 |
avis | +1.346 |
alky | +1.342 |
bin | +1.334 |
Neg logit contributions | |
downstream | -1.729 |
ervation | -1.507 |
reservation | -1.457 |
sonian | -1.449 |
Epstein | -1.442 |
Token | or |
Token ID | 393 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
王 | +1.542 |
dullah | +1.394 |
avis | +1.296 |
bin | +1.290 |
alky | +1.288 |
Neg logit contributions | |
downstream | -1.508 |
ervation | -1.307 |
sonian | -1.273 |
reservation | -1.259 |
Epstein | -1.255 |
Token | " |
Token ID | 366 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
王 | +0.401 |
dullah | +0.354 |
avis | +0.324 |
alky | +0.324 |
alde | +0.323 |
Neg logit contributions | |
downstream | -0.498 |
ervation | -0.439 |
reservation | -0.426 |
sonian | -0.424 |
Epstein | -0.422 |
Token | Breaking |
Token ID | 29449 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
王 | +0.424 |
dullah | +0.378 |
avis | +0.348 |
alky | +0.347 |
alde | +0.347 |
Neg logit contributions | |
downstream | -0.464 |
ervation | -0.405 |
reservation | -0.393 |
sonian | -0.390 |
Epstein | -0.389 |
Token | Bad |
Token ID | 7772 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
王 | +1.643 |
dullah | +1.327 |
avis | +1.146 |
alky | +1.112 |
bin | +1.088 |
Neg logit contributions | |
downstream | -4.732 |
ervation | -4.343 |
sonian | -4.233 |
reservation | -4.220 |
GMT | -4.195 |
Token | " |
Token ID | 1 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
王 | +2.656 |
dullah | +2.353 |
alky | +2.163 |
avis | +2.163 |
bin | +2.163 |
Neg logit contributions | |
downstream | -3.100 |
ervation | -2.725 |
sonian | -2.623 |
reservation | -2.623 |
Epstein | -2.612 |
Token | on |
Token ID | 319 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
王 | +1.856 |
dullah | +1.687 |
bin | +1.569 |
avis | +1.566 |
alky | +1.557 |
Neg logit contributions | |
downstream | -1.762 |
ervation | -1.532 |
sonian | -1.496 |
reservation | -1.472 |
Epstein | -1.466 |
Token | AMC |
Token ID | 36239 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
王 | +0.178 |
dullah | +0.161 |
avis | +0.152 |
alky | +0.151 |
alde | +0.150 |
Neg logit contributions | |
downstream | -0.146 |
ervation | -0.125 |
sonian | -0.120 |
reservation | -0.119 |
Epstein | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
王 | +1.081 |
dullah | +0.970 |
avis | +0.910 |
alde | +0.908 |
alky | +0.906 |
Neg logit contributions | |
downstream | -0.993 |
ervation | -0.855 |
sonian | -0.824 |
Epstein | -0.820 |
reservation | -0.815 |
Token | ise |
Token ID | 786 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
王 | +0.257 |
dullah | +0.205 |
alky | +0.170 |
alde | +0.170 |
avis | +0.169 |
Neg logit contributions | |
downstream | -0.773 |
ervation | -0.705 |
reservation | -0.690 |
sonian | -0.687 |
Epstein | -0.686 |
Token | � |
Token ID | 447 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
王 | +1.227 |
dullah | +1.104 |
alde | +1.026 |
avis | +1.026 |
alky | +1.026 |
Neg logit contributions | |
downstream | -1.104 |
ervation | -0.951 |
sonian | -0.917 |
reservation | -0.915 |
Epstein | -0.901 |
Token | � |
Token ID | 247 |
Feature activation | -9.98e-02 |
Pos logit contributions | |
王 | +0.125 |
dullah | +0.111 |
alky | +0.102 |
alde | +0.102 |
avis | +0.102 |
Neg logit contributions | |
downstream | -0.139 |
ervation | -0.122 |
reservation | -0.118 |
sonian | -0.117 |
Epstein | -0.117 |
Token | s |
Token ID | 82 |
Feature activation | +7.23e-02 |
Pos logit contributions | |
downstream | +0.078 |
ervation | +0.068 |
reservation | +0.065 |
sonian | +0.065 |
Epstein | +0.065 |
Neg logit contributions | |
王 | -0.082 |
dullah | -0.074 |
avis | -0.068 |
alky | -0.068 |
alde | -0.068 |
Token | baby |
Token ID | 5156 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
王 | +0.058 |
dullah | +0.052 |
alky | +0.048 |
avis | +0.048 |
alde | +0.048 |
Neg logit contributions | |
downstream | -0.058 |
ervation | -0.050 |
reservation | -0.048 |
sonian | -0.048 |
Epstein | -0.048 |
Token | bump |
Token ID | 13852 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
downstream | +0.014 |
ervation | +0.012 |
sonian | +0.011 |
reservation | +0.011 |
Epstein | +0.011 |
Neg logit contributions | |
王 | -0.015 |
dullah | -0.014 |
avis | -0.013 |
alky | -0.013 |
alde | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
王 | +3.272 |
dullah | +2.934 |
bin | +2.744 |
alky | +2.718 |
avis | +2.711 |
Neg logit contributions | |
downstream | -3.013 |
ervation | -2.625 |
GMT | -2.516 |
Epstein | -2.509 |
reservation | -2.509 |
Token | the |
Token ID | 262 |
Feature activation | +8.90e-02 |
Pos logit contributions | |
王 | +0.058 |
dullah | +0.052 |
alky | +0.048 |
avis | +0.048 |
alde | +0.048 |
Neg logit contributions | |
downstream | -0.056 |
ervation | -0.049 |
reservation | -0.047 |
sonian | -0.047 |
Epstein | -0.046 |
Token | shocking |
Token ID | 14702 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
王 | +0.071 |
dullah | +0.064 |
alky | +0.059 |
avis | +0.059 |
alde | +0.059 |
Neg logit contributions | |
downstream | -0.071 |
ervation | -0.062 |
reservation | -0.060 |
sonian | -0.060 |
Epstein | -0.059 |
Token | sex |
Token ID | 1714 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
王 | +1.419 |
dullah | +1.265 |
alky | +1.185 |
avis | +1.174 |
bin | +1.173 |
Neg logit contributions | |
downstream | -1.366 |
ervation | -1.198 |
sonian | -1.154 |
reservation | -1.138 |
GMT | -1.127 |
Token | scene |
Token ID | 3715 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
downstream | +0.860 |
ervation | +0.749 |
reservation | +0.726 |
Epstein | +0.721 |
sonian | +0.720 |
Neg logit contributions | |
王 | -0.820 |
dullah | -0.734 |
alde | -0.675 |
alky | -0.675 |
avis | -0.675 |
Token | now |
Token ID | 783 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
downstream | +0.195 |
ervation | +0.164 |
sonian | +0.156 |
reservation | +0.156 |
Epstein | +0.155 |
Neg logit contributions | |
王 | -0.293 |
dullah | -0.268 |
avis | -0.251 |
alde | -0.251 |
alky | -0.251 |
Token | than |
Token ID | 621 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
downstream | +0.082 |
ervation | +0.070 |
reservation | +0.067 |
sonian | +0.067 |
Epstein | +0.067 |
Neg logit contributions | |
王 | -0.103 |
dullah | -0.093 |
avis | -0.087 |
alde | -0.087 |
alky | -0.087 |
Token | under |
Token ID | 739 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
downstream | +0.066 |
ervation | +0.058 |
reservation | +0.056 |
sonian | +0.056 |
Epstein | +0.055 |
Neg logit contributions | |
王 | -0.064 |
dullah | -0.057 |
alde | -0.053 |
alky | -0.053 |
avis | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +4.33e-02 |
Pos logit contributions | |
downstream | +1.157 |
ervation | +1.010 |
reservation | +0.982 |
Epstein | +0.977 |
sonian | +0.965 |
Neg logit contributions | |
王 | -1.100 |
dullah | -0.983 |
avis | -0.908 |
alde | -0.906 |
alky | -0.903 |
Token | status |
Token ID | 3722 |
Feature activation | +9.75e-01 |
Pos logit contributions | |
王 | +0.034 |
dullah | +0.031 |
alky | +0.028 |
avis | +0.028 |
bin | +0.028 |
Neg logit contributions | |
downstream | -0.035 |
ervation | -0.031 |
sonian | -0.030 |
reservation | -0.029 |
Epstein | -0.029 |
Token | quo |
Token ID | 18658 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
王 | +1.211 |
dullah | +1.130 |
alky | +1.076 |
alde | +1.076 |
avis | +1.075 |
Neg logit contributions | |
downstream | -0.354 |
ervation | -0.250 |
reservation | -0.227 |
sonian | -0.225 |
Epstein | -0.223 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
downstream | +4.064 |
ervation | +3.420 |
reservation | +3.302 |
Epstein | +3.248 |
Duffy | +3.191 |
Neg logit contributions | |
王 | -5.790 |
dullah | -5.308 |
alde | -4.979 |
alky | -4.969 |
avis | -4.948 |
Token | � |
Token ID | 447 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
王 | +0.099 |
dullah | +0.089 |
avis | +0.082 |
alky | +0.082 |
alde | +0.082 |
Neg logit contributions | |
downstream | -0.096 |
ervation | -0.083 |
reservation | -0.080 |
sonian | -0.080 |
Epstein | -0.079 |
Token | � |
Token ID | 251 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
downstream | +0.260 |
ervation | +0.230 |
reservation | +0.223 |
sonian | +0.222 |
Epstein | +0.221 |
Neg logit contributions | |
王 | -0.209 |
dullah | -0.184 |
alky | -0.168 |
avis | -0.168 |
alde | -0.167 |
Token | Here |
Token ID | 3423 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
王 | +0.093 |
dullah | +0.084 |
avis | +0.077 |
alky | +0.077 |
alde | +0.077 |
Neg logit contributions | |
downstream | -0.094 |
ervation | -0.082 |
reservation | -0.079 |
sonian | -0.079 |
Epstein | -0.078 |
Token | , |
Token ID | 11 |
Feature activation | +6.13e-02 |
Pos logit contributions | |
王 | +0.706 |
dullah | +0.639 |
alky | +0.591 |
alde | +0.591 |
avis | +0.591 |
Neg logit contributions | |
downstream | -0.631 |
ervation | -0.542 |
reservation | -0.522 |
sonian | -0.520 |
Epstein | -0.516 |
Token | president |
Token ID | 1893 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
downstream | +0.337 |
ervation | +0.278 |
reservation | +0.263 |
Epstein | +0.263 |
sonian | +0.263 |
Neg logit contributions | |
王 | -0.563 |
dullah | -0.516 |
alky | -0.486 |
avis | -0.486 |
alde | -0.485 |
Token | and |
Token ID | 290 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
downstream | +0.009 |
ervation | +0.007 |
reservation | +0.007 |
sonian | +0.007 |
Epstein | +0.007 |
Neg logit contributions | |
王 | -0.010 |
dullah | -0.009 |
avis | -0.009 |
alde | -0.009 |
alky | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
王 | +0.063 |
dullah | +0.056 |
alky | +0.052 |
avis | +0.052 |
alde | +0.052 |
Neg logit contributions | |
downstream | -0.058 |
ervation | -0.050 |
sonian | -0.048 |
reservation | -0.048 |
Epstein | -0.048 |
Token | Washington |
Token ID | 2669 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
王 | +0.323 |
dullah | +0.291 |
avis | +0.271 |
alky | +0.271 |
alde | +0.270 |
Neg logit contributions | |
downstream | -0.287 |
ervation | -0.248 |
sonian | -0.238 |
reservation | -0.237 |
Epstein | -0.237 |
Token | status |
Token ID | 3722 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
王 | +0.681 |
dullah | +0.612 |
alky | +0.573 |
avis | +0.573 |
alde | +0.572 |
Neg logit contributions | |
downstream | -0.599 |
ervation | -0.517 |
Epstein | -0.494 |
reservation | -0.493 |
sonian | -0.489 |
Token | quo |
Token ID | 18658 |
Feature activation | -5.91e+00 |
Pos logit contributions | |
王 | +1.490 |
dullah | +1.385 |
alde | +1.326 |
alky | +1.324 |
avis | +1.319 |
Neg logit contributions | |
downstream | -0.475 |
ervation | -0.345 |
sonian | -0.315 |
Epstein | -0.311 |
reservation | -0.311 |
Token | , |
Token ID | 11 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
downstream | +4.057 |
ervation | +3.392 |
reservation | +3.279 |
Epstein | +3.232 |
Duffy | +3.186 |
Neg logit contributions | |
王 | -5.408 |
dullah | -4.940 |
alde | -4.581 |
alky | -4.572 |
avis | -4.548 |
Token | the |
Token ID | 262 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
王 | +0.171 |
dullah | +0.153 |
alky | +0.142 |
alde | +0.142 |
avis | +0.142 |
Neg logit contributions | |
downstream | -0.169 |
ervation | -0.147 |
sonian | -0.142 |
reservation | -0.141 |
Epstein | -0.141 |
Token | ga |
Token ID | 31986 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
王 | +0.129 |
dullah | +0.115 |
avis | +0.108 |
alky | +0.108 |
alde | +0.107 |
Neg logit contributions | |
downstream | -0.119 |
ervation | -0.103 |
sonian | -0.100 |
Epstein | -0.099 |
reservation | -0.099 |
Token | vel |
Token ID | 626 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
downstream | +0.253 |
ervation | +0.226 |
reservation | +0.221 |
sonian | +0.220 |
Epstein | +0.220 |
Neg logit contributions | |
王 | -0.148 |
dullah | -0.127 |
avis | -0.114 |
alde | -0.114 |
alky | -0.113 |
Token | may |
Token ID | 743 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
downstream | +0.787 |
ervation | +0.682 |
sonian | +0.656 |
reservation | +0.656 |
Epstein | +0.656 |
Neg logit contributions | |
王 | -0.804 |
dullah | -0.722 |
avis | -0.669 |
alky | -0.668 |
alde | -0.668 |
Token | good |
Token ID | 922 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
王 | +0.043 |
dullah | +0.038 |
avis | +0.035 |
alky | +0.035 |
alde | +0.035 |
Neg logit contributions | |
downstream | -0.044 |
ervation | -0.039 |
sonian | -0.038 |
reservation | -0.038 |
Epstein | -0.037 |
Token | about |
Token ID | 546 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
downstream | +0.894 |
ervation | +0.781 |
sonian | +0.749 |
reservation | +0.748 |
Epstein | +0.744 |
Neg logit contributions | |
王 | -0.975 |
dullah | -0.877 |
avis | -0.817 |
alky | -0.813 |
bin | -0.813 |
Token | a |
Token ID | 257 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
王 | +0.385 |
dullah | +0.344 |
avis | +0.318 |
alky | +0.317 |
bin | +0.317 |
Neg logit contributions | |
downstream | -0.413 |
ervation | -0.362 |
sonian | -0.351 |
reservation | -0.347 |
Epstein | -0.344 |
Token | present |
Token ID | 1944 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
downstream | +0.133 |
ervation | +0.116 |
sonian | +0.111 |
Epstein | +0.109 |
reservation | +0.109 |
Neg logit contributions | |
王 | -0.153 |
dullah | -0.138 |
avis | -0.129 |
alky | -0.128 |
bin | -0.128 |
Token | status |
Token ID | 3722 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
downstream | +0.451 |
ervation | +0.389 |
sonian | +0.374 |
reservation | +0.373 |
Epstein | +0.372 |
Neg logit contributions | |
王 | -0.496 |
dullah | -0.447 |
avis | -0.416 |
alky | -0.415 |
alde | -0.415 |
Token | quo |
Token ID | 18658 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
王 | +1.430 |
dullah | +1.327 |
avis | +1.261 |
bin | +1.259 |
alky | +1.259 |
Neg logit contributions | |
downstream | -0.486 |
ervation | -0.370 |
GMT | -0.341 |
sonian | -0.340 |
Duffy | -0.329 |
Token | and |
Token ID | 290 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
downstream | +4.077 |
ervation | +3.407 |
reservation | +3.362 |
Epstein | +3.301 |
sonian | +3.197 |
Neg logit contributions | |
王 | -5.338 |
dullah | -4.869 |
alde | -4.559 |
alky | -4.513 |
avis | -4.502 |
Token | cast |
Token ID | 3350 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
王 | +0.233 |
dullah | +0.210 |
avis | +0.196 |
alky | +0.194 |
bin | +0.194 |
Neg logit contributions | |
downstream | -0.217 |
ervation | -0.190 |
sonian | -0.182 |
Epstein | -0.180 |
reservation | -0.180 |
Token | itself |
Token ID | 2346 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
downstream | +2.468 |
ervation | +2.119 |
reservation | +2.066 |
Epstein | +2.034 |
sonian | +2.011 |
Neg logit contributions | |
王 | -2.737 |
dullah | -2.464 |
alde | -2.293 |
alky | -2.270 |
avis | -2.261 |
Token | as |
Token ID | 355 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
downstream | +0.673 |
ervation | +0.588 |
reservation | +0.567 |
Epstein | +0.564 |
sonian | +0.564 |
Neg logit contributions | |
王 | -0.656 |
dullah | -0.587 |
avis | -0.543 |
alky | -0.542 |
alde | -0.541 |
Token | the |
Token ID | 262 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
downstream | +0.049 |
ervation | +0.043 |
sonian | +0.041 |
reservation | +0.041 |
Epstein | +0.041 |
Neg logit contributions | |
王 | -0.054 |
dullah | -0.048 |
avis | -0.045 |
alky | -0.045 |
bin | -0.045 |
Token | ; |
Token ID | 26 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
downstream | +0.294 |
ervation | +0.253 |
reservation | +0.243 |
sonian | +0.243 |
Epstein | +0.242 |
Neg logit contributions | |
王 | -0.346 |
dullah | -0.313 |
avis | -0.291 |
alde | -0.291 |
alky | -0.291 |
Token | to |
Token ID | 284 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
王 | +0.124 |
dullah | +0.112 |
alde | +0.104 |
alky | +0.103 |
avis | +0.103 |
Neg logit contributions | |
downstream | -0.122 |
ervation | -0.107 |
sonian | -0.103 |
reservation | -0.102 |
Epstein | -0.102 |
Token | maintain |
Token ID | 5529 |
Feature activation | -8.38e-01 |
Pos logit contributions | |
王 | +0.331 |
dullah | +0.299 |
avis | +0.277 |
alky | +0.277 |
alde | +0.277 |
Neg logit contributions | |
downstream | -0.289 |
ervation | -0.250 |
sonian | -0.241 |
reservation | -0.239 |
Epstein | -0.238 |
Token | the |
Token ID | 262 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
downstream | +0.638 |
ervation | +0.551 |
sonian | +0.530 |
reservation | +0.529 |
Epstein | +0.527 |
Neg logit contributions | |
王 | -0.703 |
dullah | -0.633 |
avis | -0.588 |
alde | -0.587 |
alky | -0.587 |
Token | status |
Token ID | 3722 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
王 | +0.306 |
dullah | +0.271 |
avis | +0.251 |
alky | +0.250 |
alde | +0.250 |
Neg logit contributions | |
downstream | -0.319 |
ervation | -0.280 |
sonian | -0.271 |
reservation | -0.269 |
Epstein | -0.268 |
Token | quo |
Token ID | 18658 |
Feature activation | -5.69e+00 |
Pos logit contributions | |
王 | +1.499 |
dullah | +1.396 |
avis | +1.331 |
alde | +1.330 |
alky | +1.328 |
Neg logit contributions | |
downstream | -0.502 |
ervation | -0.375 |
sonian | -0.344 |
Epstein | -0.339 |
reservation | -0.337 |
Token | , |
Token ID | 11 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
downstream | +4.150 |
ervation | +3.481 |
reservation | +3.418 |
Epstein | +3.366 |
sonian | +3.312 |
Neg logit contributions | |
王 | -5.030 |
dullah | -4.546 |
alky | -4.240 |
alde | -4.228 |
avis | -4.193 |
Token | the |
Token ID | 262 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
王 | +0.188 |
dullah | +0.171 |
alde | +0.159 |
avis | +0.159 |
alky | +0.158 |
Neg logit contributions | |
downstream | -0.168 |
ervation | -0.146 |
sonian | -0.140 |
reservation | -0.139 |
Epstein | -0.139 |
Token | capacity |
Token ID | 5339 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
王 | +0.250 |
dullah | +0.224 |
alky | +0.208 |
avis | +0.208 |
alde | +0.207 |
Neg logit contributions | |
downstream | -0.238 |
ervation | -0.208 |
sonian | -0.201 |
reservation | -0.199 |
Epstein | -0.198 |
Token | for |
Token ID | 329 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
王 | +0.240 |
dullah | +0.217 |
alde | +0.201 |
avis | +0.201 |
alky | +0.201 |
Neg logit contributions | |
downstream | -0.217 |
ervation | -0.188 |
reservation | -0.181 |
sonian | -0.180 |
Epstein | -0.180 |
Token | outrage |
Token ID | 11616 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
王 | +0.126 |
dullah | +0.113 |
avis | +0.106 |
alky | +0.105 |
bin | +0.105 |
Neg logit contributions | |
downstream | -0.111 |
ervation | -0.097 |
sonian | -0.094 |
GMT | -0.092 |
Epstein | -0.092 |
Token | fit |
Token ID | 4197 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
王 | +0.771 |
dullah | +0.697 |
bin | +0.646 |
alky | +0.643 |
avis | +0.641 |
Neg logit contributions | |
downstream | -0.757 |
ervation | -0.665 |
sonian | -0.642 |
reservation | -0.636 |
GMT | -0.633 |
Token | or |
Token ID | 393 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
王 | +1.177 |
dullah | +1.054 |
bin | +0.982 |
avis | +0.977 |
alky | +0.976 |
Neg logit contributions | |
downstream | -1.179 |
ervation | -1.047 |
sonian | -1.007 |
reservation | -0.989 |
GMT | -0.987 |
Token | match |
Token ID | 2872 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
王 | +0.197 |
dullah | +0.178 |
alky | +0.164 |
bin | +0.164 |
avis | +0.164 |
Neg logit contributions | |
downstream | -0.186 |
ervation | -0.163 |
sonian | -0.156 |
reservation | -0.155 |
Epstein | -0.155 |
Token | the |
Token ID | 262 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
王 | +0.850 |
dullah | +0.767 |
alky | +0.715 |
avis | +0.713 |
bin | +0.711 |
Neg logit contributions | |
downstream | -0.794 |
ervation | -0.695 |
sonian | -0.669 |
reservation | -0.662 |
GMT | -0.658 |
Token | status |
Token ID | 3722 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
downstream | +0.018 |
ervation | +0.016 |
sonian | +0.015 |
reservation | +0.015 |
Epstein | +0.015 |
Neg logit contributions | |
王 | -0.022 |
dullah | -0.019 |
alky | -0.018 |
avis | -0.018 |
alde | -0.018 |
Token | quo |
Token ID | 18658 |
Feature activation | -5.68e+00 |
Pos logit contributions | |
王 | +1.815 |
dullah | +1.693 |
alde | +1.614 |
avis | +1.614 |
alky | +1.612 |
Neg logit contributions | |
downstream | -0.580 |
ervation | -0.426 |
sonian | -0.395 |
Epstein | -0.384 |
reservation | -0.383 |
Token | . |
Token ID | 13 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
downstream | +3.885 |
ervation | +3.240 |
reservation | +3.171 |
Epstein | +3.100 |
Duffy | +3.057 |
Neg logit contributions | |
王 | -5.280 |
dullah | -4.774 |
alde | -4.458 |
alky | -4.445 |
avis | -4.415 |
Token | Finn |
Token ID | 15368 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
王 | +0.229 |
dullah | +0.205 |
alky | +0.189 |
avis | +0.189 |
alde | +0.188 |
Neg logit contributions | |
downstream | -0.234 |
ervation | -0.204 |
reservation | -0.197 |
sonian | -0.195 |
Epstein | -0.193 |
Token | isn |
Token ID | 2125 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
downstream | +0.339 |
ervation | +0.294 |
reservation | +0.283 |
sonian | +0.281 |
Epstein | +0.280 |
Neg logit contributions | |
王 | -0.360 |
dullah | -0.323 |
avis | -0.300 |
alky | -0.299 |
alde | -0.299 |
Token | 't |
Token ID | 470 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
downstream | +0.206 |
ervation | +0.171 |
reservation | +0.165 |
sonian | +0.161 |
Epstein | +0.161 |
Neg logit contributions | |
王 | -0.304 |
dullah | -0.278 |
bin | -0.262 |
avis | -0.261 |
alky | -0.260 |
Token | a |
Token ID | 257 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
downstream | +0.710 |
ervation | +0.616 |
reservation | +0.596 |
Epstein | +0.591 |
sonian | +0.591 |
Neg logit contributions | |
王 | -0.720 |
dullah | -0.646 |
alde | -0.596 |
avis | -0.596 |
alky | -0.596 |
Token | and |
Token ID | 290 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
王 | +0.813 |
dullah | +0.734 |
avis | +0.683 |
alde | +0.682 |
bin | +0.679 |
Neg logit contributions | |
downstream | -0.712 |
ervation | -0.616 |
sonian | -0.599 |
reservation | -0.593 |
Epstein | -0.593 |
Token | impose |
Token ID | 13551 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
王 | +0.053 |
dullah | +0.048 |
avis | +0.044 |
alde | +0.044 |
alky | +0.044 |
Neg logit contributions | |
downstream | -0.049 |
ervation | -0.043 |
reservation | -0.041 |
sonian | -0.041 |
Epstein | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
王 | +1.229 |
dullah | +1.097 |
avis | +1.013 |
bin | +1.012 |
alde | +1.008 |
Neg logit contributions | |
downstream | -1.222 |
ervation | -1.081 |
sonian | -1.047 |
Epstein | -1.032 |
reservation | -1.028 |
Token | challenging |
Token ID | 9389 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
downstream | +0.086 |
ervation | +0.075 |
sonian | +0.072 |
reservation | +0.072 |
Epstein | +0.072 |
Neg logit contributions | |
王 | -0.089 |
dullah | -0.080 |
avis | -0.074 |
alky | -0.074 |
alde | -0.074 |
Token | status |
Token ID | 3722 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
王 | +0.483 |
dullah | +0.433 |
alky | +0.400 |
bin | +0.400 |
alde | +0.400 |
Neg logit contributions | |
downstream | -0.458 |
ervation | -0.401 |
sonian | -0.382 |
reservation | -0.382 |
Epstein | -0.382 |
Token | quo |
Token ID | 18658 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
王 | +1.461 |
dullah | +1.348 |
avis | +1.285 |
alde | +1.283 |
alky | +1.279 |
Neg logit contributions | |
downstream | -0.548 |
ervation | -0.430 |
sonian | -0.396 |
GMT | -0.393 |
Epstein | -0.393 |
Token | in |
Token ID | 287 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
downstream | +3.697 |
ervation | +3.046 |
reservation | +2.983 |
Epstein | +2.913 |
sonian | +2.881 |
Neg logit contributions | |
王 | -4.973 |
dullah | -4.551 |
alde | -4.249 |
alky | -4.246 |
avis | -4.228 |
Token | the |
Token ID | 262 |
Feature activation | -4.89e-03 |
Pos logit contributions | |
downstream | +0.057 |
ervation | +0.050 |
reservation | +0.048 |
sonian | +0.048 |
Epstein | +0.047 |
Neg logit contributions | |
王 | -0.054 |
dullah | -0.049 |
avis | -0.045 |
alde | -0.045 |
alky | -0.045 |
Token | city |
Token ID | 1748 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
downstream | +0.004 |
ervation | +0.004 |
reservation | +0.004 |
Epstein | +0.004 |
sonian | +0.004 |
Neg logit contributions | |
王 | -0.004 |
dullah | -0.003 |
alde | -0.003 |
avis | -0.003 |
alky | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
王 | +0.224 |
dullah | +0.203 |
avis | +0.189 |
alde | +0.189 |
bin | +0.189 |
Neg logit contributions | |
downstream | -0.182 |
ervation | -0.157 |
sonian | -0.150 |
reservation | -0.150 |
Epstein | -0.150 |
Token | There |
Token ID | 1318 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
王 | +0.253 |
dullah | +0.226 |
avis | +0.207 |
alky | +0.207 |
alde | +0.207 |
Neg logit contributions | |
downstream | -0.270 |
ervation | -0.236 |
reservation | -0.228 |
sonian | -0.227 |
Epstein | -0.225 |
Token | AT |
Token ID | 5161 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
downstream | +0.937 |
ervation | +0.820 |
reservation | +0.794 |
sonian | +0.790 |
Epstein | +0.787 |
Neg logit contributions | |
王 | -0.854 |
dullah | -0.762 |
avis | -0.700 |
alky | -0.700 |
alde | -0.699 |
Token | & |
Token ID | 5 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
王 | +0.800 |
dullah | +0.724 |
avis | +0.683 |
alky | +0.679 |
bin | +0.677 |
Neg logit contributions | |
downstream | -0.597 |
ervation | -0.509 |
reservation | -0.488 |
sonian | -0.481 |
Epstein | -0.481 |
Token | T |
Token ID | 51 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
王 | +0.172 |
dullah | +0.150 |
bin | +0.139 |
avis | +0.138 |
alky | +0.137 |
Neg logit contributions | |
downstream | -0.214 |
ervation | -0.188 |
reservation | -0.183 |
Epstein | -0.181 |
sonian | -0.180 |
Token | 's |
Token ID | 338 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
王 | +0.086 |
dullah | +0.077 |
avis | +0.071 |
alky | +0.071 |
alde | +0.071 |
Neg logit contributions | |
downstream | -0.084 |
ervation | -0.074 |
reservation | -0.071 |
sonian | -0.071 |
Epstein | -0.070 |
Token | powerhouse |
Token ID | 34886 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
downstream | +0.025 |
ervation | +0.021 |
sonian | +0.021 |
reservation | +0.021 |
Epstein | +0.020 |
Neg logit contributions | |
王 | -0.029 |
dullah | -0.026 |
alky | -0.024 |
avis | -0.024 |
alde | -0.024 |
Token | FISA |
Token ID | 44073 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
王 | +2.055 |
dullah | +1.860 |
alky | +1.744 |
avis | +1.737 |
bin | +1.735 |
Neg logit contributions | |
downstream | -1.777 |
ervation | -1.558 |
sonian | -1.495 |
reservation | -1.474 |
GMT | -1.474 |
Token | lobbyist |
Token ID | 35196 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
downstream | +3.769 |
ervation | +3.179 |
reservation | +3.128 |
Epstein | +3.098 |
Duffy | +3.012 |
Neg logit contributions | |
王 | -4.801 |
dullah | -4.354 |
alde | -4.052 |
alky | -4.014 |
avis | -4.000 |
Token | Charlie |
Token ID | 11526 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
downstream | +1.189 |
ervation | +1.023 |
reservation | +0.989 |
Epstein | +0.983 |
sonian | +0.979 |
Neg logit contributions | |
王 | -1.329 |
dullah | -1.200 |
alde | -1.112 |
avis | -1.111 |
alky | -1.109 |
Token | Black |
Token ID | 2619 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
王 | +0.428 |
dullah | +0.386 |
avis | +0.357 |
alky | +0.356 |
alde | +0.355 |
Neg logit contributions | |
downstream | -0.428 |
ervation | -0.375 |
reservation | -0.363 |
sonian | -0.360 |
GMT | -0.353 |
Token | left |
Token ID | 1364 |
Feature activation | +8.60e-01 |
Pos logit contributions | |
downstream | +0.815 |
ervation | +0.709 |
reservation | +0.684 |
sonian | +0.678 |
Epstein | +0.677 |
Neg logit contributions | |
王 | -0.840 |
dullah | -0.753 |
alde | -0.697 |
avis | -0.697 |
alky | -0.696 |
Token | his |
Token ID | 465 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
王 | +0.665 |
dullah | +0.594 |
avis | +0.550 |
alky | +0.547 |
alde | +0.546 |
Neg logit contributions | |
downstream | -0.708 |
ervation | -0.622 |
sonian | -0.602 |
reservation | -0.599 |
Epstein | -0.590 |
Token | – |
Token ID | 784 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
王 | +0.367 |
dullah | +0.329 |
avis | +0.304 |
bin | +0.303 |
alky | +0.303 |
Neg logit contributions | |
downstream | -0.367 |
ervation | -0.321 |
reservation | -0.310 |
sonian | -0.310 |
GMT | -0.307 |
Token | were |
Token ID | 547 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
downstream | +0.377 |
ervation | +0.326 |
reservation | +0.315 |
sonian | +0.313 |
Epstein | +0.312 |
Neg logit contributions | |
王 | -0.399 |
dullah | -0.360 |
avis | -0.333 |
alde | -0.333 |
alky | -0.333 |
Token | extraordinary |
Token ID | 11359 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
downstream | +0.206 |
ervation | +0.181 |
reservation | +0.174 |
sonian | +0.174 |
Epstein | +0.172 |
Neg logit contributions | |
王 | -0.200 |
dullah | -0.179 |
avis | -0.166 |
alky | -0.165 |
alde | -0.165 |
Token | , |
Token ID | 11 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
downstream | +0.325 |
ervation | +0.280 |
sonian | +0.270 |
reservation | +0.269 |
Epstein | +0.266 |
Neg logit contributions | |
王 | -0.382 |
dullah | -0.345 |
avis | -0.321 |
alde | -0.321 |
alky | -0.321 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
王 | +0.088 |
dullah | +0.079 |
avis | +0.073 |
alky | +0.073 |
alde | +0.073 |
Neg logit contributions | |
downstream | -0.087 |
ervation | -0.076 |
reservation | -0.073 |
sonian | -0.073 |
Epstein | -0.073 |
Token | quo |
Token ID | 18658 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
王 | +0.093 |
dullah | +0.084 |
avis | +0.078 |
alky | +0.078 |
alde | +0.077 |
Neg logit contributions | |
downstream | -0.090 |
ervation | -0.079 |
sonian | -0.076 |
reservation | -0.075 |
GMT | -0.074 |
Token | – |
Token ID | 784 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
downstream | +3.821 |
ervation | +3.237 |
Epstein | +3.198 |
reservation | +3.179 |
Duffy | +3.119 |
Neg logit contributions | |
王 | -4.280 |
dullah | -3.838 |
alky | -3.562 |
alde | -3.555 |
avis | -3.520 |
Token | the |
Token ID | 262 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
downstream | +0.334 |
ervation | +0.290 |
reservation | +0.280 |
sonian | +0.278 |
Epstein | +0.277 |
Neg logit contributions | |
王 | -0.346 |
dullah | -0.311 |
avis | -0.288 |
alky | -0.287 |
alde | -0.287 |
Token | actions |
Token ID | 4028 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
王 | +0.090 |
dullah | +0.081 |
avis | +0.075 |
alky | +0.075 |
alde | +0.075 |
Neg logit contributions | |
downstream | -0.079 |
ervation | -0.069 |
sonian | -0.066 |
reservation | -0.066 |
GMT | -0.065 |
Token | taken |
Token ID | 2077 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
downstream | +0.357 |
ervation | +0.297 |
reservation | +0.284 |
sonian | +0.282 |
Epstein | +0.278 |
Neg logit contributions | |
王 | -0.573 |
dullah | -0.525 |
avis | -0.494 |
alky | -0.492 |
alde | -0.492 |
Token | by |
Token ID | 416 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
downstream | +1.370 |
ervation | +1.192 |
reservation | +1.154 |
Epstein | +1.147 |
sonian | +1.142 |
Neg logit contributions | |
王 | -1.316 |
dullah | -1.179 |
alde | -1.086 |
alky | -1.086 |
avis | -1.085 |
Token | of |
Token ID | 286 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
王 | +1.216 |
dullah | +1.089 |
bin | +1.014 |
avis | +1.008 |
alky | +1.007 |
Neg logit contributions | |
downstream | -1.174 |
ervation | -1.033 |
reservation | -0.993 |
sonian | -0.991 |
Epstein | -0.978 |
Token | a |
Token ID | 257 |
Feature activation | +6.71e-02 |
Pos logit contributions | |
王 | +0.031 |
dullah | +0.028 |
avis | +0.026 |
alky | +0.026 |
bin | +0.026 |
Neg logit contributions | |
downstream | -0.030 |
ervation | -0.026 |
sonian | -0.025 |
reservation | -0.025 |
GMT | -0.025 |
Token | qu |
Token ID | 627 |
Feature activation | -9.12e-01 |
Pos logit contributions | |
王 | +0.056 |
dullah | +0.051 |
avis | +0.047 |
alky | +0.047 |
bin | +0.047 |
Neg logit contributions | |
downstream | -0.051 |
ervation | -0.045 |
sonian | -0.043 |
reservation | -0.042 |
Epstein | -0.042 |
Token | id |
Token ID | 312 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
downstream | +0.923 |
ervation | +0.829 |
reservation | +0.805 |
Epstein | +0.802 |
sonian | +0.800 |
Neg logit contributions | |
王 | -0.541 |
dullah | -0.463 |
alde | -0.412 |
alky | -0.411 |
avis | -0.410 |
Token | pro |
Token ID | 386 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
downstream | +0.629 |
ervation | +0.551 |
reservation | +0.533 |
sonian | +0.530 |
Epstein | +0.529 |
Neg logit contributions | |
王 | -0.577 |
dullah | -0.515 |
avis | -0.474 |
alky | -0.473 |
alde | -0.472 |
Token | quo |
Token ID | 18658 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
downstream | +0.274 |
ervation | +0.188 |
reservation | +0.169 |
Epstein | +0.164 |
sonian | +0.163 |
Neg logit contributions | |
王 | -1.026 |
dullah | -0.957 |
alky | -0.916 |
alde | -0.915 |
avis | -0.912 |
Token | , |
Token ID | 11 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
downstream | +3.562 |
ervation | +2.986 |
Epstein | +2.936 |
reservation | +2.917 |
Duffy | +2.845 |
Neg logit contributions | |
王 | -4.530 |
dullah | -4.085 |
alde | -3.817 |
alky | -3.792 |
avis | -3.788 |
Token | it |
Token ID | 340 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
王 | +0.092 |
dullah | +0.083 |
avis | +0.077 |
alky | +0.077 |
alde | +0.077 |
Neg logit contributions | |
downstream | -0.088 |
ervation | -0.076 |
sonian | -0.073 |
reservation | -0.073 |
GMT | -0.072 |
Token | � |
Token ID | 447 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
王 | +0.132 |
dullah | +0.119 |
avis | +0.110 |
alky | +0.109 |
alde | +0.109 |
Neg logit contributions | |
downstream | -0.133 |
ervation | -0.117 |
reservation | -0.112 |
sonian | -0.112 |
Epstein | -0.111 |
Token | � |
Token ID | 247 |
Feature activation | +7.61e-02 |
Pos logit contributions | |
downstream | +0.377 |
ervation | +0.329 |
reservation | +0.321 |
Epstein | +0.318 |
sonian | +0.318 |
Neg logit contributions | |
王 | -0.341 |
dullah | -0.303 |
alky | -0.279 |
alde | -0.278 |
avis | -0.278 |
Token | s |
Token ID | 82 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
王 | +0.059 |
dullah | +0.053 |
avis | +0.049 |
alde | +0.049 |
alky | +0.049 |
Neg logit contributions | |
downstream | -0.063 |
ervation | -0.054 |
reservation | -0.053 |
sonian | -0.052 |
Epstein | -0.052 |
Token | new |
Token ID | 649 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
downstream | +0.251 |
ervation | +0.219 |
sonian | +0.210 |
reservation | +0.208 |
Epstein | +0.208 |
Neg logit contributions | |
王 | -0.286 |
dullah | -0.258 |
alky | -0.240 |
avis | -0.240 |
bin | -0.239 |
Token | Verizon |
Token ID | 18062 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
downstream | +0.886 |
ervation | +0.757 |
sonian | +0.721 |
reservation | +0.720 |
Epstein | +0.717 |
Neg logit contributions | |
王 | -1.198 |
dullah | -1.091 |
alky | -1.020 |
avis | -1.019 |
alde | -1.019 |
Token | IP |
Token ID | 6101 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
downstream | +1.443 |
ervation | +1.235 |
reservation | +1.186 |
sonian | +1.178 |
Epstein | +1.176 |
Neg logit contributions | |
王 | -1.777 |
dullah | -1.611 |
avis | -1.500 |
alky | -1.500 |
alde | -1.499 |
Token | address |
Token ID | 2209 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
downstream | +2.375 |
ervation | +2.023 |
reservation | +1.958 |
Epstein | +1.933 |
sonian | +1.928 |
Neg logit contributions | |
王 | -2.850 |
dullah | -2.585 |
alky | -2.402 |
avis | -2.401 |
alde | -2.400 |
Token | via |
Token ID | 2884 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
王 | +0.113 |
dullah | +0.102 |
bin | +0.096 |
avis | +0.096 |
alde | +0.096 |
Neg logit contributions | |
downstream | -0.098 |
ervation | -0.085 |
sonian | -0.083 |
Epstein | -0.082 |
Duffy | -0.081 |
Token | DHCP |
Token ID | 43729 |
Feature activation | -4.95e+00 |
Pos logit contributions | |
downstream | +0.933 |
ervation | +0.811 |
reservation | +0.779 |
sonian | +0.779 |
Epstein | +0.775 |
Neg logit contributions | |
王 | -0.996 |
dullah | -0.898 |
avis | -0.832 |
alky | -0.831 |
alde | -0.830 |
Token | . |
Token ID | 764 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
downstream | +3.522 |
ervation | +2.969 |
reservation | +2.903 |
Epstein | +2.826 |
sonian | +2.782 |
Neg logit contributions | |
王 | -4.452 |
dullah | -4.027 |
alde | -3.754 |
alky | -3.751 |
avis | -3.726 |
Token |
|
Token ID | 198 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
downstream | +0.359 |
ervation | +0.314 |
sonian | +0.303 |
reservation | +0.303 |
Epstein | +0.300 |
Neg logit contributions | |
王 | -0.348 |
dullah | -0.312 |
avis | -0.290 |
bin | -0.288 |
alky | -0.288 |
Token |
|
Token ID | 198 |
Feature activation | -4.14e-01 |
Pos logit contributions | |
downstream | +0.409 |
ervation | +0.356 |
reservation | +0.345 |
sonian | +0.343 |
Epstein | +0.342 |
Neg logit contributions | |
王 | -0.398 |
dullah | -0.356 |
avis | -0.329 |
alky | -0.328 |
alde | -0.328 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
downstream | +0.331 |
ervation | +0.286 |
reservation | +0.277 |
sonian | +0.277 |
Epstein | +0.276 |
Neg logit contributions | |
王 | -0.334 |
dullah | -0.299 |
avis | -0.277 |
alky | -0.276 |
alde | -0.276 |
Token | the |
Token ID | 262 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
downstream | +0.978 |
ervation | +0.852 |
reservation | +0.820 |
sonian | +0.819 |
Epstein | +0.814 |
Neg logit contributions | |
王 | -1.022 |
dullah | -0.918 |
alky | -0.852 |
avis | -0.852 |
alde | -0.848 |
Token | control |
Token ID | 1630 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
王 | +0.436 |
dullah | +0.391 |
avis | +0.361 |
alky | +0.361 |
alde | +0.361 |
Neg logit contributions | |
downstream | -0.428 |
ervation | -0.373 |
reservation | -0.359 |
sonian | -0.358 |
Epstein | -0.358 |
Token | over |
Token ID | 625 |
Feature activation | -8.88e-01 |
Pos logit contributions | |
downstream | +0.421 |
ervation | +0.358 |
reservation | +0.346 |
Epstein | +0.342 |
sonian | +0.341 |
Neg logit contributions | |
王 | -0.516 |
dullah | -0.469 |
avis | -0.437 |
alky | -0.436 |
alde | -0.435 |
Token | the |
Token ID | 262 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
downstream | +0.717 |
ervation | +0.624 |
reservation | +0.603 |
sonian | +0.599 |
Epstein | +0.598 |
Neg logit contributions | |
王 | -0.707 |
dullah | -0.634 |
avis | -0.584 |
alde | -0.584 |
alky | -0.584 |
Token | auto |
Token ID | 8295 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
downstream | +0.009 |
ervation | +0.008 |
reservation | +0.008 |
sonian | +0.008 |
Epstein | +0.008 |
Neg logit contributions | |
王 | -0.010 |
dullah | -0.009 |
avis | -0.009 |
alky | -0.009 |
alde | -0.009 |
Token | - |
Token ID | 12 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
downstream | +0.825 |
ervation | +0.712 |
reservation | +0.690 |
sonian | +0.683 |
Epstein | +0.682 |
Neg logit contributions | |
王 | -0.884 |
dullah | -0.797 |
avis | -0.736 |
alde | -0.736 |
alky | -0.735 |
Token | ISO |
Token ID | 40734 |
Feature activation | -4.65e+00 |
Pos logit contributions | |
downstream | +0.274 |
ervation | +0.238 |
reservation | +0.231 |
Epstein | +0.229 |
sonian | +0.229 |
Neg logit contributions | |
王 | -0.253 |
dullah | -0.226 |
alde | -0.209 |
avis | -0.209 |
alky | -0.209 |
Token | behaviour |
Token ID | 9172 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
downstream | +3.862 |
ervation | +3.334 |
reservation | +3.271 |
Epstein | +3.172 |
Duffy | +3.150 |
Neg logit contributions | |
王 | -3.629 |
dullah | -3.267 |
avis | -3.008 |
alde | -2.980 |
alky | -2.980 |
Token | : |
Token ID | 25 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
downstream | +0.124 |
ervation | +0.107 |
reservation | +0.104 |
sonian | +0.103 |
Epstein | +0.103 |
Neg logit contributions | |
王 | -0.138 |
dullah | -0.124 |
alde | -0.115 |
avis | -0.115 |
alky | -0.115 |
Token | it |
Token ID | 340 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
王 | +0.370 |
dullah | +0.333 |
alde | +0.308 |
alky | +0.307 |
avis | +0.307 |
Neg logit contributions | |
downstream | -0.359 |
ervation | -0.313 |
reservation | -0.302 |
sonian | -0.301 |
Epstein | -0.300 |
Token | � |
Token ID | 447 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
王 | +0.065 |
dullah | +0.059 |
avis | +0.055 |
alky | +0.055 |
alde | +0.055 |
Neg logit contributions | |
downstream | -0.058 |
ervation | -0.050 |
reservation | -0.048 |
sonian | -0.047 |
Epstein | -0.047 |
Token | � |
Token ID | 247 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
downstream | +0.293 |
ervation | +0.256 |
reservation | +0.249 |
Epstein | +0.247 |
sonian | +0.246 |
Neg logit contributions | |
王 | -0.273 |
dullah | -0.244 |
alky | -0.225 |
alde | -0.224 |
avis | -0.224 |
Token | propagation |
Token ID | 43594 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
downstream | +0.794 |
ervation | +0.672 |
sonian | +0.642 |
reservation | +0.640 |
Epstein | +0.637 |
Neg logit contributions | |
王 | -1.165 |
dullah | -1.063 |
alky | -0.995 |
avis | -0.995 |
alde | -0.994 |
Token | if |
Token ID | 611 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
downstream | +0.530 |
ervation | +0.464 |
sonian | +0.449 |
reservation | +0.445 |
Epstein | +0.442 |
Neg logit contributions | |
王 | -0.591 |
dullah | -0.533 |
alde | -0.494 |
alky | -0.494 |
bin | -0.494 |
Token | there |
Token ID | 612 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
王 | +1.065 |
dullah | +0.962 |
alky | +0.897 |
bin | +0.891 |
alde | +0.887 |
Neg logit contributions | |
downstream | -1.017 |
ervation | -0.904 |
sonian | -0.886 |
reservation | -0.869 |
Epstein | -0.868 |
Token | are |
Token ID | 389 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
王 | +0.383 |
dullah | +0.345 |
alky | +0.321 |
alde | +0.321 |
avis | +0.319 |
Neg logit contributions | |
downstream | -0.334 |
ervation | -0.288 |
reservation | -0.278 |
Epstein | -0.276 |
sonian | -0.276 |
Token | no |
Token ID | 645 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
downstream | +0.405 |
ervation | +0.352 |
reservation | +0.339 |
sonian | +0.339 |
Epstein | +0.338 |
Neg logit contributions | |
王 | -0.444 |
dullah | -0.401 |
alky | -0.371 |
alde | -0.371 |
avis | -0.371 |
Token | que |
Token ID | 8358 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
王 | +0.077 |
dullah | +0.070 |
alky | +0.065 |
alde | +0.064 |
avis | +0.064 |
Neg logit contributions | |
downstream | -0.067 |
ervation | -0.059 |
sonian | -0.057 |
reservation | -0.056 |
Epstein | -0.056 |
Token | uing |
Token ID | 4250 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
downstream | +2.472 |
ervation | +1.954 |
reservation | +1.801 |
Epstein | +1.760 |
GMT | +1.710 |
Neg logit contributions | |
王 | -4.962 |
dullah | -4.511 |
alde | -4.276 |
alky | -4.245 |
avis | -4.241 |
Token | delays |
Token ID | 16119 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
downstream | +1.637 |
ervation | +1.410 |
reservation | +1.371 |
Epstein | +1.351 |
sonian | +1.342 |
Neg logit contributions | |
王 | -1.653 |
dullah | -1.484 |
avis | -1.371 |
alky | -1.370 |
alde | -1.369 |
Token | and |
Token ID | 290 |
Feature activation | -8.55e-02 |
Pos logit contributions | |
downstream | +0.872 |
ervation | +0.755 |
sonian | +0.729 |
reservation | +0.724 |
Epstein | +0.721 |
Neg logit contributions | |
王 | -1.071 |
dullah | -0.969 |
avis | -0.903 |
alde | -0.902 |
alky | -0.902 |
Token | no |
Token ID | 645 |
Feature activation | +5.79e-01 |
Pos logit contributions | |
downstream | +0.065 |
ervation | +0.057 |
sonian | +0.056 |
reservation | +0.055 |
Epstein | +0.055 |
Neg logit contributions | |
王 | -0.071 |
dullah | -0.063 |
alky | -0.059 |
alde | -0.059 |
bin | -0.059 |
Token | processing |
Token ID | 7587 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
王 | +0.502 |
dullah | +0.453 |
alky | +0.424 |
alde | +0.422 |
avis | +0.421 |
Neg logit contributions | |
downstream | -0.414 |
ervation | -0.364 |
sonian | -0.353 |
Epstein | -0.348 |
reservation | -0.345 |
Token | best |
Token ID | 1266 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
王 | +0.326 |
dullah | +0.293 |
alky | +0.272 |
avis | +0.272 |
alde | +0.271 |
Neg logit contributions | |
downstream | -0.294 |
ervation | -0.255 |
sonian | -0.245 |
reservation | -0.245 |
Epstein | -0.243 |
Token | - |
Token ID | 12 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
downstream | +0.865 |
ervation | +0.748 |
reservation | +0.724 |
sonian | +0.718 |
Epstein | +0.718 |
Neg logit contributions | |
王 | -0.897 |
dullah | -0.808 |
avis | -0.746 |
alde | -0.745 |
alky | -0.744 |
Token | known |
Token ID | 4002 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
王 | +0.248 |
dullah | +0.222 |
avis | +0.204 |
alky | +0.204 |
alde | +0.204 |
Neg logit contributions | |
downstream | -0.260 |
ervation | -0.226 |
reservation | -0.219 |
sonian | -0.218 |
Epstein | -0.218 |
Token | mail |
Token ID | 6920 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
downstream | +0.398 |
ervation | +0.346 |
reservation | +0.335 |
sonian | +0.332 |
Epstein | +0.332 |
Neg logit contributions | |
王 | -0.392 |
dullah | -0.352 |
avis | -0.324 |
alde | -0.324 |
alky | -0.324 |
Token | order |
Token ID | 1502 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
downstream | +1.211 |
ervation | +1.045 |
reservation | +1.010 |
Epstein | +1.000 |
sonian | +1.000 |
Neg logit contributions | |
王 | -1.319 |
dullah | -1.190 |
alky | -1.100 |
avis | -1.100 |
alde | -1.099 |
Token | catalog |
Token ID | 18388 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
downstream | +1.255 |
ervation | +1.093 |
reservation | +1.061 |
Epstein | +1.051 |
sonian | +1.049 |
Neg logit contributions | |
王 | -1.200 |
dullah | -1.075 |
avis | -0.988 |
alky | -0.987 |
alde | -0.987 |
Token | ues |
Token ID | 947 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
downstream | +2.590 |
ervation | +2.147 |
reservation | +2.041 |
Epstein | +2.005 |
sonian | +1.992 |
Neg logit contributions | |
王 | -4.544 |
dullah | -4.145 |
alky | -3.902 |
avis | -3.866 |
alde | -3.863 |
Token | apologised |
Token ID | 46397 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
downstream | +1.647 |
ervation | +1.421 |
reservation | +1.372 |
Epstein | +1.361 |
sonian | +1.361 |
Neg logit contributions | |
王 | -1.785 |
dullah | -1.608 |
alde | -1.489 |
alky | -1.488 |
avis | -1.488 |
Token | today |
Token ID | 1909 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
王 | +0.245 |
dullah | +0.221 |
avis | +0.206 |
alky | +0.205 |
alde | +0.205 |
Neg logit contributions | |
downstream | -0.212 |
ervation | -0.184 |
reservation | -0.176 |
sonian | -0.176 |
Epstein | -0.175 |
Token | after |
Token ID | 706 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
王 | +0.291 |
dullah | +0.264 |
avis | +0.245 |
alde | +0.245 |
alky | +0.245 |
Neg logit contributions | |
downstream | -0.245 |
ervation | -0.210 |
reservation | -0.203 |
sonian | -0.201 |
Epstein | -0.201 |
Token | web |
Token ID | 3992 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
downstream | +1.065 |
ervation | +0.913 |
reservation | +0.884 |
Epstein | +0.875 |
sonian | +0.871 |
Neg logit contributions | |
王 | -1.223 |
dullah | -1.104 |
alde | -1.024 |
avis | -1.024 |
alky | -1.024 |
Token | The |
Token ID | 464 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
downstream | +0.206 |
ervation | +0.179 |
reservation | +0.175 |
Epstein | +0.173 |
sonian | +0.173 |
Neg logit contributions | |
王 | -0.193 |
dullah | -0.172 |
avis | -0.158 |
alky | -0.158 |
alde | -0.158 |
Token | unfortunate |
Token ID | 14855 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
downstream | +0.340 |
ervation | +0.295 |
reservation | +0.284 |
sonian | +0.284 |
Epstein | +0.283 |
Neg logit contributions | |
王 | -0.358 |
dullah | -0.322 |
alky | -0.298 |
avis | -0.297 |
alde | -0.297 |
Token | bulk |
Token ID | 11963 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
王 | +1.106 |
dullah | +0.995 |
alky | +0.927 |
alde | +0.919 |
avis | +0.919 |
Neg logit contributions | |
downstream | -1.050 |
ervation | -0.921 |
sonian | -0.887 |
reservation | -0.881 |
Epstein | -0.879 |
Token | of |
Token ID | 286 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
王 | +0.545 |
dullah | +0.491 |
avis | +0.455 |
alky | +0.455 |
alde | +0.454 |
Neg logit contributions | |
downstream | -0.490 |
ervation | -0.424 |
reservation | -0.407 |
Epstein | -0.406 |
sonian | -0.406 |
Token | these |
Token ID | 777 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
downstream | +0.037 |
ervation | +0.032 |
reservation | +0.031 |
sonian | +0.031 |
Epstein | +0.031 |
Neg logit contributions | |
王 | -0.034 |
dullah | -0.030 |
alky | -0.028 |
avis | -0.028 |
alde | -0.028 |
Token | cast |
Token ID | 3350 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
downstream | +0.564 |
ervation | +0.492 |
reservation | +0.479 |
Epstein | +0.474 |
sonian | +0.473 |
Neg logit contributions | |
王 | -0.502 |
dullah | -0.448 |
alde | -0.411 |
avis | -0.411 |
alky | -0.410 |
Token | way |
Token ID | 835 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
downstream | +3.558 |
ervation | +3.049 |
reservation | +2.965 |
Epstein | +2.933 |
sonian | +2.882 |
Neg logit contributions | |
王 | -3.503 |
dullah | -3.162 |
alde | -2.901 |
avis | -2.901 |
alky | -2.894 |
Token | voters |
Token ID | 4446 |
Feature activation | +8.65e-01 |
Pos logit contributions | |
王 | +0.341 |
dullah | +0.308 |
avis | +0.286 |
alky | +0.286 |
alde | +0.286 |
Neg logit contributions | |
downstream | -0.301 |
ervation | -0.259 |
reservation | -0.249 |
sonian | -0.247 |
Epstein | -0.247 |
Token | was |
Token ID | 373 |
Feature activation | +6.95e-02 |
Pos logit contributions | |
王 | +0.717 |
dullah | +0.645 |
avis | +0.597 |
alky | +0.597 |
alde | +0.595 |
Neg logit contributions | |
downstream | -0.665 |
ervation | -0.579 |
sonian | -0.557 |
reservation | -0.557 |
Epstein | -0.555 |
Token | caused |
Token ID | 4073 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
王 | +0.057 |
dullah | +0.052 |
alky | +0.048 |
alde | +0.048 |
avis | +0.048 |
Neg logit contributions | |
downstream | -0.054 |
ervation | -0.047 |
sonian | -0.045 |
reservation | -0.045 |
Epstein | -0.045 |
Token | by |
Token ID | 416 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
王 | +0.776 |
dullah | +0.691 |
avis | +0.638 |
alky | +0.636 |
alde | +0.635 |
Neg logit contributions | |
downstream | -0.826 |
ervation | -0.724 |
sonian | -0.703 |
reservation | -0.701 |
Epstein | -0.696 |
Token | two |
Token ID | 734 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
downstream | +1.296 |
ervation | +1.123 |
reservation | +1.085 |
Epstein | +1.076 |
sonian | +1.075 |
Neg logit contributions | |
王 | -1.346 |
dullah | -1.210 |
avis | -1.118 |
alde | -1.117 |
alky | -1.117 |
Token | quantum |
Token ID | 14821 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
王 | +0.960 |
dullah | +0.854 |
alky | +0.800 |
alde | +0.796 |
avis | +0.791 |
Neg logit contributions | |
downstream | -0.971 |
ervation | -0.857 |
sonian | -0.826 |
reservation | -0.821 |
Epstein | -0.817 |
Token | states |
Token ID | 2585 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
downstream | +0.251 |
ervation | +0.224 |
reservation | +0.218 |
sonian | +0.217 |
Epstein | +0.217 |
Neg logit contributions | |
王 | -0.169 |
dullah | -0.146 |
avis | -0.132 |
alky | -0.132 |
alde | -0.132 |
Token | during |
Token ID | 1141 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
王 | +0.102 |
dullah | +0.092 |
alde | +0.085 |
alky | +0.085 |
avis | +0.085 |
Neg logit contributions | |
downstream | -0.087 |
ervation | -0.075 |
sonian | -0.073 |
reservation | -0.073 |
Epstein | -0.072 |
Token | the |
Token ID | 262 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
downstream | +0.322 |
ervation | +0.281 |
sonian | +0.273 |
reservation | +0.270 |
Epstein | +0.268 |
Neg logit contributions | |
王 | -0.320 |
dullah | -0.287 |
alky | -0.267 |
alde | -0.266 |
avis | -0.266 |
Token | pseud |
Token ID | 25038 |
Feature activation | -4.42e+00 |
Pos logit contributions | |
downstream | +0.237 |
ervation | +0.206 |
sonian | +0.199 |
reservation | +0.197 |
Epstein | +0.197 |
Neg logit contributions | |
王 | -0.261 |
dullah | -0.234 |
alky | -0.219 |
alde | -0.218 |
avis | -0.218 |
Token | og |
Token ID | 519 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
downstream | +2.636 |
ervation | +2.206 |
sonian | +2.090 |
reservation | +2.041 |
Epstein | +2.028 |
Neg logit contributions | |
王 | -4.458 |
dullah | -4.044 |
alde | -3.790 |
avis | -3.789 |
alky | -3.756 |
Token | ap |
Token ID | 499 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
downstream | +1.072 |
ervation | +0.901 |
reservation | +0.861 |
sonian | +0.853 |
Epstein | +0.852 |
Neg logit contributions | |
王 | -1.566 |
dullah | -1.428 |
alde | -1.336 |
avis | -1.336 |
alky | -1.336 |
Token | , |
Token ID | 11 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
downstream | +0.204 |
ervation | +0.171 |
sonian | +0.163 |
reservation | +0.163 |
Epstein | +0.162 |
Neg logit contributions | |
王 | -0.306 |
dullah | -0.280 |
alky | -0.263 |
alde | -0.262 |
avis | -0.262 |
Token | one |
Token ID | 530 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
downstream | +0.190 |
ervation | +0.164 |
sonian | +0.158 |
reservation | +0.158 |
Epstein | +0.155 |
Neg logit contributions | |
王 | -0.232 |
dullah | -0.210 |
alky | -0.196 |
avis | -0.195 |
bin | -0.194 |
Token | of |
Token ID | 286 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
downstream | +0.992 |
ervation | +0.856 |
sonian | +0.824 |
reservation | +0.823 |
Epstein | +0.817 |
Neg logit contributions | |
王 | -1.153 |
dullah | -1.041 |
alky | -0.970 |
avis | -0.968 |
alde | -0.968 |
Token |
|
Token ID | 198 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
王 | +0.204 |
dullah | +0.182 |
avis | +0.168 |
bin | +0.168 |
alky | +0.167 |
Neg logit contributions | |
downstream | -0.222 |
ervation | -0.194 |
reservation | -0.188 |
sonian | -0.186 |
Epstein | -0.185 |
Token | ' |
Token ID | 6 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
王 | +0.236 |
dullah | +0.209 |
alky | +0.192 |
bin | +0.192 |
avis | +0.192 |
Neg logit contributions | |
downstream | -0.247 |
ervation | -0.213 |
reservation | -0.209 |
Epstein | -0.206 |
sonian | -0.205 |
Token | The |
Token ID | 464 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
王 | +0.164 |
dullah | +0.146 |
bin | +0.135 |
avis | +0.135 |
alky | +0.134 |
Neg logit contributions | |
downstream | -0.167 |
ervation | -0.144 |
reservation | -0.140 |
sonian | -0.138 |
Epstein | -0.137 |
Token | presence |
Token ID | 4931 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
downstream | +0.116 |
ervation | +0.100 |
sonian | +0.097 |
reservation | +0.096 |
Epstein | +0.095 |
Neg logit contributions | |
王 | -0.125 |
dullah | -0.112 |
avis | -0.104 |
alky | -0.104 |
alde | -0.104 |
Token | of |
Token ID | 286 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
downstream | +0.220 |
ervation | +0.189 |
reservation | +0.182 |
Epstein | +0.180 |
sonian | +0.180 |
Neg logit contributions | |
王 | -0.253 |
dullah | -0.229 |
avis | -0.212 |
alky | -0.212 |
alde | -0.212 |
Token | endemic |
Token ID | 42560 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
downstream | +0.045 |
ervation | +0.040 |
reservation | +0.038 |
sonian | +0.038 |
Epstein | +0.038 |
Neg logit contributions | |
王 | -0.048 |
dullah | -0.043 |
alky | -0.040 |
avis | -0.040 |
alde | -0.039 |
Token | new |
Token ID | 649 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
downstream | +3.328 |
ervation | +2.869 |
reservation | +2.786 |
Epstein | +2.780 |
Duffy | +2.726 |
Neg logit contributions | |
王 | -3.540 |
dullah | -3.234 |
alde | -2.947 |
avis | -2.938 |
alky | -2.928 |
Token | world |
Token ID | 995 |
Feature activation | -9.41e-02 |
Pos logit contributions | |
downstream | +0.655 |
ervation | +0.560 |
reservation | +0.540 |
sonian | +0.535 |
Epstein | +0.534 |
Neg logit contributions | |
王 | -0.795 |
dullah | -0.720 |
avis | -0.671 |
alde | -0.670 |
alky | -0.670 |
Token | monkeys |
Token ID | 26697 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
downstream | +0.073 |
ervation | +0.063 |
reservation | +0.061 |
sonian | +0.061 |
Epstein | +0.061 |
Neg logit contributions | |
王 | -0.078 |
dullah | -0.070 |
alky | -0.065 |
alde | -0.065 |
avis | -0.065 |
Token | on |
Token ID | 319 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
downstream | +0.779 |
ervation | +0.679 |
reservation | +0.655 |
sonian | +0.653 |
Epstein | +0.650 |
Neg logit contributions | |
王 | -0.811 |
dullah | -0.728 |
avis | -0.676 |
alde | -0.673 |
alky | -0.673 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
王 | +0.119 |
dullah | +0.106 |
alde | +0.099 |
avis | +0.098 |
alky | +0.098 |
Neg logit contributions | |
downstream | -0.115 |
ervation | -0.100 |
reservation | -0.096 |
sonian | -0.096 |
Epstein | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
王 | +0.086 |
dullah | +0.077 |
avis | +0.071 |
alky | +0.071 |
alde | +0.070 |
Neg logit contributions | |
downstream | -0.093 |
ervation | -0.081 |
reservation | -0.079 |
sonian | -0.078 |
Epstein | -0.077 |
Token |
|
Token ID | 198 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
王 | +0.100 |
dullah | +0.090 |
bin | +0.083 |
avis | +0.082 |
alky | +0.082 |
Neg logit contributions | |
downstream | -0.110 |
ervation | -0.096 |
reservation | -0.093 |
sonian | -0.093 |
Epstein | -0.092 |
Token | " |
Token ID | 1 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
王 | +0.157 |
dullah | +0.139 |
alky | +0.129 |
avis | +0.129 |
bin | +0.128 |
Neg logit contributions | |
downstream | -0.169 |
ervation | -0.146 |
reservation | -0.143 |
sonian | -0.141 |
Epstein | -0.140 |
Token | Understanding |
Token ID | 43467 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
downstream | +0.118 |
ervation | +0.102 |
reservation | +0.099 |
sonian | +0.099 |
Epstein | +0.098 |
Neg logit contributions | |
王 | -0.115 |
dullah | -0.103 |
avis | -0.095 |
alky | -0.095 |
bin | -0.095 |
Token | the |
Token ID | 262 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
王 | +0.386 |
dullah | +0.344 |
alky | +0.319 |
alde | +0.319 |
avis | +0.317 |
Neg logit contributions | |
downstream | -0.408 |
ervation | -0.359 |
sonian | -0.348 |
reservation | -0.345 |
Epstein | -0.342 |
Token | pseud |
Token ID | 25038 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
王 | +0.192 |
dullah | +0.174 |
alky | +0.163 |
avis | +0.162 |
alde | +0.162 |
Neg logit contributions | |
downstream | -0.163 |
ervation | -0.141 |
sonian | -0.135 |
reservation | -0.134 |
Epstein | -0.134 |
Token | og |
Token ID | 519 |
Feature activation | -9.71e-01 |
Pos logit contributions | |
downstream | +2.561 |
ervation | +2.148 |
sonian | +2.027 |
reservation | +2.014 |
Epstein | +1.998 |
Neg logit contributions | |
王 | -4.117 |
dullah | -3.741 |
alde | -3.505 |
avis | -3.495 |
alky | -3.482 |
Token | ap |
Token ID | 499 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
downstream | +0.636 |
ervation | +0.534 |
reservation | +0.512 |
sonian | +0.507 |
Epstein | +0.506 |
Neg logit contributions | |
王 | -0.921 |
dullah | -0.841 |
avis | -0.787 |
alky | -0.787 |
alde | -0.787 |
Token | is |
Token ID | 318 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
王 | +0.098 |
dullah | +0.089 |
alky | +0.084 |
avis | +0.084 |
alde | +0.084 |
Neg logit contributions | |
downstream | -0.066 |
ervation | -0.056 |
reservation | -0.053 |
sonian | -0.053 |
Epstein | -0.053 |
Token | as |
Token ID | 355 |
Feature activation | -8.39e-02 |
Pos logit contributions | |
王 | +0.300 |
dullah | +0.271 |
avis | +0.251 |
alky | +0.251 |
alde | +0.251 |
Neg logit contributions | |
downstream | -0.290 |
ervation | -0.255 |
reservation | -0.246 |
sonian | -0.245 |
Epstein | -0.242 |
Token | important |
Token ID | 1593 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
downstream | +0.070 |
ervation | +0.061 |
reservation | +0.060 |
sonian | +0.059 |
Epstein | +0.059 |
Neg logit contributions | |
王 | -0.064 |
dullah | -0.057 |
avis | -0.053 |
alky | -0.053 |
alde | -0.053 |
Token | will |
Token ID | 10594 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
downstream | +0.850 |
ervation | +0.743 |
reservation | +0.720 |
Epstein | +0.716 |
sonian | +0.713 |
Neg logit contributions | |
王 | -0.773 |
dullah | -0.691 |
alky | -0.636 |
avis | -0.635 |
alde | -0.633 |
Token | be |
Token ID | 307 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
downstream | +0.649 |
ervation | +0.558 |
reservation | +0.539 |
Epstein | +0.533 |
sonian | +0.530 |
Neg logit contributions | |
王 | -0.717 |
dullah | -0.646 |
avis | -0.599 |
alky | -0.599 |
alde | -0.597 |
Token | part |
Token ID | 636 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
downstream | +0.175 |
ervation | +0.152 |
reservation | +0.147 |
sonian | +0.146 |
Epstein | +0.145 |
Neg logit contributions | |
王 | -0.188 |
dullah | -0.169 |
avis | -0.157 |
alky | -0.157 |
alde | -0.157 |
Token | of |
Token ID | 286 |
Feature activation | -9.17e-02 |
Pos logit contributions | |
王 | +0.153 |
dullah | +0.138 |
avis | +0.128 |
alky | +0.128 |
alde | +0.127 |
Neg logit contributions | |
downstream | -0.140 |
ervation | -0.122 |
reservation | -0.117 |
Epstein | -0.116 |
sonian | -0.116 |
Token | a |
Token ID | 257 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
downstream | +0.071 |
ervation | +0.061 |
reservation | +0.059 |
sonian | +0.059 |
Epstein | +0.058 |
Neg logit contributions | |
王 | -0.076 |
dullah | -0.069 |
avis | -0.064 |
alde | -0.064 |
alky | -0.064 |
Token | DACA |
Token ID | 38061 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
王 | +0.030 |
dullah | +0.028 |
avis | +0.026 |
alky | +0.026 |
alde | +0.026 |
Neg logit contributions | |
downstream | -0.023 |
ervation | -0.020 |
sonian | -0.019 |
reservation | -0.019 |
Epstein | -0.019 |
Token | deal |
Token ID | 1730 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
downstream | +2.844 |
ervation | +2.400 |
reservation | +2.338 |
Epstein | +2.277 |
sonian | +2.266 |
Neg logit contributions | |
王 | -3.787 |
dullah | -3.465 |
alde | -3.226 |
alky | -3.224 |
avis | -3.217 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
downstream | +0.563 |
ervation | +0.483 |
reservation | +0.465 |
sonian | +0.463 |
Epstein | +0.459 |
Neg logit contributions | |
王 | -0.681 |
dullah | -0.616 |
avis | -0.574 |
alde | -0.573 |
alky | -0.572 |
Token | � |
Token ID | 447 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
downstream | +0.094 |
ervation | +0.081 |
reservation | +0.078 |
sonian | +0.078 |
Epstein | +0.077 |
Neg logit contributions | |
王 | -0.105 |
dullah | -0.095 |
alky | -0.088 |
avis | -0.088 |
alde | -0.088 |
Token | � |
Token ID | 247 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
downstream | +0.377 |
ervation | +0.330 |
reservation | +0.320 |
sonian | +0.318 |
Epstein | +0.316 |
Neg logit contributions | |
王 | -0.357 |
dullah | -0.318 |
alky | -0.293 |
avis | -0.292 |
alde | -0.292 |
Token | Trump |
Token ID | 1301 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
downstream | +0.198 |
ervation | +0.172 |
reservation | +0.166 |
sonian | +0.165 |
Epstein | +0.164 |
Neg logit contributions | |
王 | -0.201 |
dullah | -0.181 |
alde | -0.167 |
avis | -0.167 |
alky | -0.167 |
Token | seeds |
Token ID | 11904 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
downstream | +1.247 |
ervation | +1.093 |
reservation | +1.061 |
Epstein | +1.051 |
sonian | +1.051 |
Neg logit contributions | |
王 | -1.089 |
dullah | -0.969 |
alky | -0.887 |
avis | -0.887 |
alde | -0.886 |
Token | but |
Token ID | 475 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
downstream | +0.095 |
ervation | +0.082 |
reservation | +0.079 |
sonian | +0.079 |
Epstein | +0.079 |
Neg logit contributions | |
王 | -0.106 |
dullah | -0.096 |
avis | -0.089 |
alde | -0.089 |
alky | -0.089 |
Token | these |
Token ID | 777 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
王 | +1.009 |
dullah | +0.910 |
alky | +0.851 |
bin | +0.848 |
alde | +0.847 |
Neg logit contributions | |
downstream | -0.960 |
ervation | -0.836 |
sonian | -0.810 |
reservation | -0.803 |
Epstein | -0.798 |
Token | are |
Token ID | 389 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
downstream | +0.570 |
ervation | +0.492 |
reservation | +0.474 |
sonian | +0.471 |
Epstein | +0.470 |
Neg logit contributions | |
王 | -0.630 |
dullah | -0.567 |
alde | -0.527 |
avis | -0.526 |
alky | -0.526 |
Token | a |
Token ID | 257 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
downstream | +0.113 |
ervation | +0.098 |
reservation | +0.094 |
sonian | +0.094 |
Epstein | +0.094 |
Neg logit contributions | |
王 | -0.117 |
dullah | -0.105 |
alky | -0.098 |
alde | -0.098 |
avis | -0.097 |
Token | pseud |
Token ID | 25038 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
downstream | +0.121 |
ervation | +0.105 |
reservation | +0.101 |
sonian | +0.101 |
Epstein | +0.101 |
Neg logit contributions | |
王 | -0.121 |
dullah | -0.108 |
alky | -0.100 |
avis | -0.100 |
alde | -0.100 |
Token | og |
Token ID | 519 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
downstream | +3.586 |
ervation | +3.160 |
reservation | +3.046 |
Epstein | +3.040 |
sonian | +3.031 |
Neg logit contributions | |
王 | -3.034 |
dullah | -2.676 |
alky | -2.435 |
alde | -2.430 |
avis | -2.424 |
Token | rain |
Token ID | 3201 |
Feature activation | +9.24e-01 |
Pos logit contributions | |
downstream | +0.976 |
ervation | +0.848 |
reservation | +0.814 |
sonian | +0.811 |
Epstein | +0.810 |
Neg logit contributions | |
王 | -1.085 |
dullah | -0.966 |
alde | -0.894 |
alky | -0.893 |
avis | -0.892 |
Token | , |
Token ID | 11 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
王 | +0.780 |
dullah | +0.700 |
avis | +0.659 |
alky | +0.657 |
alde | +0.656 |
Neg logit contributions | |
downstream | -0.690 |
ervation | -0.593 |
sonian | -0.580 |
reservation | -0.578 |
Epstein | -0.573 |
Token | and |
Token ID | 290 |
Feature activation | -3.52e-03 |
Pos logit contributions | |
downstream | +0.018 |
ervation | +0.015 |
reservation | +0.015 |
sonian | +0.015 |
Epstein | +0.015 |
Neg logit contributions | |
王 | -0.019 |
dullah | -0.018 |
alky | -0.016 |
avis | -0.016 |
alde | -0.016 |
Token | Prof |
Token ID | 4415 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
downstream | +0.003 |
ervation | +0.002 |
reservation | +0.002 |
sonian | +0.002 |
Epstein | +0.002 |
Neg logit contributions | |
王 | -0.003 |
dullah | -0.003 |
alky | -0.002 |
alde | -0.002 |
avis | -0.002 |
Token | on |
Token ID | 319 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
王 | +0.148 |
dullah | +0.135 |
avis | +0.126 |
alky | +0.125 |
alde | +0.125 |
Neg logit contributions | |
downstream | -0.120 |
ervation | -0.104 |
sonian | -0.100 |
reservation | -0.099 |
Epstein | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
downstream | +0.186 |
ervation | +0.161 |
reservation | +0.155 |
sonian | +0.155 |
Epstein | +0.154 |
Neg logit contributions | |
王 | -0.215 |
dullah | -0.195 |
avis | -0.181 |
alde | -0.181 |
alky | -0.181 |
Token | then |
Token ID | 788 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
downstream | +0.040 |
ervation | +0.034 |
reservation | +0.033 |
sonian | +0.033 |
Epstein | +0.033 |
Neg logit contributions | |
王 | -0.044 |
dullah | -0.040 |
alky | -0.037 |
avis | -0.037 |
alde | -0.037 |
Token | open |
Token ID | 1280 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
downstream | +0.389 |
ervation | +0.335 |
reservation | +0.322 |
sonian | +0.322 |
Epstein | +0.320 |
Neg logit contributions | |
王 | -0.446 |
dullah | -0.403 |
avis | -0.374 |
alky | -0.374 |
alde | -0.374 |
Token | the |
Token ID | 262 |
Feature activation | -7.08e-02 |
Pos logit contributions | |
王 | +0.060 |
dullah | +0.055 |
avis | +0.051 |
alky | +0.051 |
alde | +0.051 |
Neg logit contributions | |
downstream | -0.054 |
ervation | -0.046 |
sonian | -0.045 |
reservation | -0.045 |
Epstein | -0.044 |
Token | catalog |
Token ID | 18388 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
downstream | +0.057 |
ervation | +0.050 |
reservation | +0.048 |
sonian | +0.048 |
Epstein | +0.048 |
Neg logit contributions | |
王 | -0.056 |
dullah | -0.051 |
avis | -0.047 |
alky | -0.047 |
alde | -0.047 |
Token | like |
Token ID | 588 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
downstream | +3.035 |
ervation | +2.600 |
reservation | +2.514 |
Epstein | +2.445 |
sonian | +2.434 |
Neg logit contributions | |
王 | -3.608 |
dullah | -3.247 |
alky | -3.012 |
alde | -3.000 |
avis | -2.995 |
Token | you |
Token ID | 345 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
王 | +1.451 |
dullah | +1.307 |
avis | +1.211 |
alky | +1.207 |
alde | +1.201 |
Neg logit contributions | |
downstream | -1.451 |
ervation | -1.270 |
sonian | -1.238 |
reservation | -1.221 |
Epstein | -1.216 |
Token | would |
Token ID | 561 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
downstream | +0.090 |
ervation | +0.078 |
reservation | +0.075 |
sonian | +0.075 |
Epstein | +0.075 |
Neg logit contributions | |
王 | -0.092 |
dullah | -0.082 |
avis | -0.076 |
alky | -0.076 |
alde | -0.076 |
Token | any |
Token ID | 597 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
王 | +0.636 |
dullah | +0.574 |
avis | +0.532 |
alky | +0.531 |
alde | +0.529 |
Neg logit contributions | |
downstream | -0.594 |
ervation | -0.516 |
sonian | -0.498 |
reservation | -0.495 |
Epstein | -0.493 |
Token | ebook |
Token ID | 47179 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
王 | +0.130 |
dullah | +0.118 |
avis | +0.110 |
alky | +0.110 |
alde | +0.110 |
Neg logit contributions | |
downstream | -0.107 |
ervation | -0.092 |
sonian | -0.088 |
reservation | -0.087 |
Epstein | -0.087 |
Token | , |
Token ID | 11 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
downstream | +0.180 |
ervation | +0.157 |
sonian | +0.153 |
Epstein | +0.151 |
reservation | +0.151 |
Neg logit contributions | |
王 | -0.200 |
dullah | -0.181 |
avis | -0.168 |
bin | -0.167 |
alky | -0.167 |
Token | while |
Token ID | 981 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
王 | +0.028 |
dullah | +0.026 |
avis | +0.024 |
alky | +0.024 |
bin | +0.024 |
Neg logit contributions | |
downstream | -0.025 |
ervation | -0.022 |
sonian | -0.021 |
reservation | -0.021 |
Epstein | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
downstream | +0.616 |
ervation | +0.536 |
sonian | +0.515 |
reservation | +0.515 |
Epstein | +0.512 |
Neg logit contributions | |
王 | -0.640 |
dullah | -0.576 |
avis | -0.533 |
alky | -0.531 |
alde | -0.530 |
Token | Bur |
Token ID | 5481 |
Feature activation | +7.93e-02 |
Pos logit contributions | |
downstream | +0.063 |
ervation | +0.055 |
sonian | +0.053 |
reservation | +0.053 |
Epstein | +0.053 |
Neg logit contributions | |
王 | -0.062 |
dullah | -0.056 |
avis | -0.051 |
alky | -0.051 |
alde | -0.051 |
Token | m |
Token ID | 76 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
王 | +0.049 |
dullah | +0.043 |
alky | +0.038 |
avis | +0.038 |
alde | +0.038 |
Neg logit contributions | |
downstream | -0.078 |
ervation | -0.070 |
reservation | -0.068 |
sonian | -0.067 |
Epstein | -0.067 |
Token | ese |
Token ID | 2771 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
downstream | +0.810 |
ervation | +0.719 |
reservation | +0.700 |
Epstein | +0.692 |
sonian | +0.692 |
Neg logit contributions | |
王 | -0.581 |
dullah | -0.506 |
alde | -0.458 |
alky | -0.457 |
avis | -0.455 |
Token | authorities |
Token ID | 4773 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
王 | +2.495 |
dullah | +2.253 |
avis | +2.128 |
bin | +2.127 |
alde | +2.096 |
Neg logit contributions | |
downstream | -2.220 |
ervation | -1.966 |
sonian | -1.879 |
Epstein | -1.850 |
GMT | -1.840 |
Token | said |
Token ID | 531 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
王 | +1.135 |
dullah | +1.019 |
avis | +0.957 |
bin | +0.947 |
alky | +0.938 |
Neg logit contributions | |
downstream | -1.097 |
ervation | -0.971 |
sonian | -0.934 |
Epstein | -0.925 |
GMT | -0.923 |
Token | there |
Token ID | 612 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
王 | +0.303 |
dullah | +0.273 |
avis | +0.252 |
alky | +0.252 |
bin | +0.252 |
Neg logit contributions | |
downstream | -0.284 |
ervation | -0.250 |
sonian | -0.242 |
reservation | -0.237 |
Epstein | -0.237 |
Token | had |
Token ID | 550 |
Feature activation | -6.20e-02 |
Pos logit contributions | |
王 | +0.384 |
dullah | +0.343 |
avis | +0.316 |
alde | +0.316 |
alky | +0.316 |
Neg logit contributions | |
downstream | -0.400 |
ervation | -0.349 |
reservation | -0.338 |
sonian | -0.337 |
Epstein | -0.336 |
Token | been |
Token ID | 587 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
downstream | +0.045 |
ervation | +0.038 |
reservation | +0.037 |
sonian | +0.037 |
Epstein | +0.037 |
Neg logit contributions | |
王 | -0.055 |
dullah | -0.049 |
alde | -0.046 |
avis | -0.046 |
alky | -0.046 |
Token | is |
Token ID | 318 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
王 | +0.065 |
dullah | +0.058 |
avis | +0.054 |
alky | +0.054 |
alde | +0.054 |
Neg logit contributions | |
downstream | -0.062 |
ervation | -0.054 |
reservation | -0.052 |
sonian | -0.052 |
Epstein | -0.051 |
Token | born |
Token ID | 4642 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
王 | +0.213 |
dullah | +0.192 |
avis | +0.178 |
alky | +0.178 |
alde | +0.177 |
Neg logit contributions | |
downstream | -0.189 |
ervation | -0.165 |
sonian | -0.158 |
reservation | -0.158 |
Epstein | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
downstream | +0.328 |
ervation | +0.284 |
reservation | +0.274 |
sonian | +0.272 |
Epstein | +0.270 |
Neg logit contributions | |
王 | -0.361 |
dullah | -0.325 |
alky | -0.301 |
avis | -0.301 |
alde | -0.301 |
Token |
|
Token ID | 198 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
王 | +0.058 |
dullah | +0.052 |
avis | +0.048 |
alky | +0.048 |
alde | +0.048 |
Neg logit contributions | |
downstream | -0.060 |
ervation | -0.053 |
reservation | -0.051 |
sonian | -0.050 |
Epstein | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
王 | +0.198 |
dullah | +0.176 |
bin | +0.163 |
avis | +0.162 |
alky | +0.161 |
Neg logit contributions | |
downstream | -0.219 |
ervation | -0.192 |
reservation | -0.186 |
sonian | -0.184 |
Epstein | -0.183 |
Token | TIME |
Token ID | 34694 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
王 | +0.187 |
dullah | +0.167 |
alky | +0.155 |
avis | +0.155 |
alde | +0.155 |
Neg logit contributions | |
downstream | -0.183 |
ervation | -0.158 |
reservation | -0.153 |
Epstein | -0.151 |
sonian | -0.151 |
Token | TRA |
Token ID | 29125 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
王 | +2.035 |
dullah | +1.811 |
alky | +1.669 |
avis | +1.667 |
bin | +1.662 |
Neg logit contributions | |
downstream | -2.125 |
ervation | -1.862 |
reservation | -1.785 |
sonian | -1.765 |
Epstein | -1.762 |
Token | VEL |
Token ID | 18697 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
王 | +1.704 |
dullah | +1.536 |
avis | +1.437 |
bin | +1.434 |
alky | +1.424 |
Neg logit contributions | |
downstream | -1.450 |
ervation | -1.231 |
reservation | -1.193 |
sonian | -1.180 |
Epstein | -1.159 |
Token | ER |
Token ID | 1137 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
downstream | +0.845 |
ervation | +0.728 |
reservation | +0.705 |
Epstein | +0.702 |
sonian | +0.700 |
Neg logit contributions | |
王 | -0.906 |
dullah | -0.820 |
alde | -0.759 |
avis | -0.759 |
alky | -0.758 |
Token |
|
Token ID | 198 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
王 | +0.180 |
dullah | +0.162 |
avis | +0.149 |
alde | +0.149 |
alky | +0.149 |
Neg logit contributions | |
downstream | -0.181 |
ervation | -0.157 |
reservation | -0.152 |
sonian | -0.151 |
Epstein | -0.151 |
Token |
|
Token ID | 198 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
王 | +0.165 |
dullah | +0.147 |
bin | +0.136 |
avis | +0.135 |
alky | +0.135 |
Neg logit contributions | |
downstream | -0.182 |
ervation | -0.160 |
reservation | -0.155 |
sonian | -0.153 |
Epstein | -0.152 |
Token | � |
Token ID | 247 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
downstream | +0.221 |
ervation | +0.194 |
reservation | +0.189 |
sonian | +0.188 |
Epstein | +0.187 |
Neg logit contributions | |
王 | -0.192 |
dullah | -0.170 |
alky | -0.156 |
alde | -0.156 |
avis | -0.155 |
Token | s |
Token ID | 82 |
Feature activation | -7.45e-02 |
Pos logit contributions | |
downstream | +0.010 |
ervation | +0.009 |
reservation | +0.009 |
sonian | +0.009 |
Epstein | +0.009 |
Neg logit contributions | |
王 | -0.010 |
dullah | -0.009 |
alky | -0.008 |
avis | -0.008 |
alde | -0.008 |
Token | considered |
Token ID | 3177 |
Feature activation | +9.72e-01 |
Pos logit contributions | |
downstream | +0.055 |
ervation | +0.048 |
sonian | +0.046 |
reservation | +0.046 |
Epstein | +0.045 |
Neg logit contributions | |
王 | -0.064 |
dullah | -0.058 |
alky | -0.054 |
alde | -0.054 |
bin | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
王 | +0.778 |
dullah | +0.698 |
avis | +0.646 |
alky | +0.645 |
alde | +0.643 |
Neg logit contributions | |
downstream | -0.774 |
ervation | -0.676 |
reservation | -0.651 |
sonian | -0.650 |
Epstein | -0.644 |
Token | best |
Token ID | 1266 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
downstream | +0.025 |
ervation | +0.022 |
reservation | +0.021 |
sonian | +0.021 |
Epstein | +0.021 |
Neg logit contributions | |
王 | -0.025 |
dullah | -0.022 |
avis | -0.020 |
alky | -0.020 |
alde | -0.020 |
Token | spread |
Token ID | 4104 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
downstream | +0.895 |
ervation | +0.774 |
reservation | +0.749 |
Epstein | +0.743 |
sonian | +0.742 |
Neg logit contributions | |
王 | -0.938 |
dullah | -0.844 |
avis | -0.780 |
alde | -0.780 |
alky | -0.779 |
Token | in |
Token ID | 287 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
王 | +1.783 |
dullah | +1.589 |
alde | +1.490 |
bin | +1.482 |
avis | +1.473 |
Neg logit contributions | |
downstream | -1.719 |
ervation | -1.496 |
sonian | -1.447 |
reservation | -1.437 |
GMT | -1.426 |
Token | the |
Token ID | 262 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
downstream | +0.034 |
ervation | +0.030 |
reservation | +0.029 |
sonian | +0.029 |
Epstein | +0.028 |
Neg logit contributions | |
王 | -0.035 |
dullah | -0.031 |
avis | -0.029 |
alky | -0.029 |
alde | -0.029 |
Token | International |
Token ID | 4037 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
downstream | +0.023 |
ervation | +0.020 |
sonian | +0.019 |
reservation | +0.019 |
Epstein | +0.019 |
Neg logit contributions | |
王 | -0.021 |
dullah | -0.018 |
avis | -0.017 |
alky | -0.017 |
alde | -0.017 |
Token | League |
Token ID | 4041 |
Feature activation | -9.54e-04 |
Pos logit contributions | |
downstream | +1.129 |
ervation | +0.949 |
reservation | +0.905 |
sonian | +0.898 |
Epstein | +0.896 |
Neg logit contributions | |
王 | -1.643 |
dullah | -1.497 |
alde | -1.405 |
avis | -1.404 |
alky | -1.402 |
Token | , |
Token ID | 11 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
downstream | +0.001 |
ervation | +0.001 |
sonian | +0.001 |
reservation | +0.001 |
Epstein | +0.001 |
Neg logit contributions | |
王 | -0.001 |
dullah | -0.001 |
alde | -0.001 |
avis | -0.001 |
alky | -0.001 |
Token | pass |
Token ID | 1208 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
downstream | +0.044 |
ervation | +0.039 |
sonian | +0.037 |
reservation | +0.037 |
Epstein | +0.037 |
Neg logit contributions | |
王 | -0.044 |
dullah | -0.039 |
avis | -0.036 |
alky | -0.036 |
alde | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
王 | +0.514 |
dullah | +0.467 |
avis | +0.433 |
alky | +0.430 |
alde | +0.428 |
Neg logit contributions | |
downstream | -0.491 |
ervation | -0.428 |
sonian | -0.414 |
reservation | -0.410 |
Epstein | -0.410 |
Token | � |
Token ID | 447 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
downstream | +0.049 |
ervation | +0.042 |
reservation | +0.040 |
sonian | +0.040 |
Epstein | +0.040 |
Neg logit contributions | |
王 | -0.057 |
dullah | -0.051 |
avis | -0.048 |
alky | -0.047 |
alde | -0.047 |
Token | � |
Token ID | 251 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
downstream | +0.336 |
ervation | +0.298 |
reservation | +0.291 |
sonian | +0.288 |
Epstein | +0.287 |
Neg logit contributions | |
王 | -0.253 |
dullah | -0.223 |
alky | -0.202 |
alde | -0.201 |
avis | -0.201 |
Token | wide |
Token ID | 3094 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
王 | +0.200 |
dullah | +0.181 |
avis | +0.167 |
alky | +0.166 |
alde | +0.166 |
Neg logit contributions | |
downstream | -0.197 |
ervation | -0.171 |
reservation | -0.165 |
sonian | -0.164 |
Epstein | -0.163 |
Token | receiver |
Token ID | 9733 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
王 | +0.456 |
dullah | +0.403 |
alde | +0.368 |
avis | +0.367 |
alky | +0.366 |
Neg logit contributions | |
downstream | -0.583 |
ervation | -0.516 |
reservation | -0.500 |
sonian | -0.499 |
Epstein | -0.498 |
Token | Emmanuel |
Token ID | 32390 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
王 | +1.514 |
dullah | +1.357 |
avis | +1.237 |
alky | +1.233 |
bin | +1.228 |
Neg logit contributions | |
downstream | -1.797 |
ervation | -1.582 |
sonian | -1.536 |
reservation | -1.523 |
Epstein | -1.515 |
Token | Sanders |
Token ID | 5831 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
王 | +0.501 |
dullah | +0.450 |
avis | +0.413 |
alde | +0.410 |
alky | +0.409 |
Neg logit contributions | |
downstream | -0.567 |
ervation | -0.496 |
sonian | -0.481 |
reservation | -0.480 |
GMT | -0.471 |
Token | said |
Token ID | 531 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
downstream | +0.645 |
ervation | +0.563 |
reservation | +0.545 |
sonian | +0.541 |
Epstein | +0.540 |
Neg logit contributions | |
王 | -0.617 |
dullah | -0.551 |
alde | -0.508 |
avis | -0.508 |
alky | -0.507 |
Token | . |
Token ID | 13 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
downstream | +0.146 |
ervation | +0.126 |
reservation | +0.121 |
sonian | +0.120 |
Epstein | +0.119 |
Neg logit contributions | |
王 | -0.183 |
dullah | -0.166 |
alde | -0.155 |
avis | -0.154 |
alky | -0.154 |
Token | � |
Token ID | 564 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
downstream | +0.271 |
ervation | +0.236 |
reservation | +0.228 |
sonian | +0.227 |
Epstein | +0.227 |
Neg logit contributions | |
王 | -0.261 |
dullah | -0.234 |
avis | -0.215 |
alde | -0.215 |
alky | -0.215 |
Token | 4 |
Token ID | 604 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
王 | +1.196 |
dullah | +1.054 |
alky | +0.977 |
alde | +0.972 |
bin | +0.971 |
Neg logit contributions | |
downstream | -1.364 |
ervation | -1.216 |
Epstein | -1.178 |
sonian | -1.172 |
reservation | -1.156 |
Token | ," |
Token ID | 553 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
downstream | +0.201 |
ervation | +0.176 |
Epstein | +0.169 |
reservation | +0.169 |
sonian | +0.169 |
Neg logit contributions | |
王 | -0.199 |
dullah | -0.177 |
alky | -0.164 |
avis | -0.164 |
alde | -0.164 |
Token | as |
Token ID | 355 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
王 | +0.140 |
dullah | +0.124 |
alky | +0.114 |
alde | +0.114 |
avis | +0.114 |
Neg logit contributions | |
downstream | -0.167 |
ervation | -0.148 |
sonian | -0.144 |
reservation | -0.143 |
Epstein | -0.142 |
Token | well |
Token ID | 880 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
downstream | +0.408 |
ervation | +0.354 |
reservation | +0.342 |
sonian | +0.340 |
Epstein | +0.338 |
Neg logit contributions | |
王 | -0.426 |
dullah | -0.383 |
alky | -0.355 |
alde | -0.355 |
avis | -0.354 |
Token | as |
Token ID | 355 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
downstream | +0.926 |
ervation | +0.804 |
reservation | +0.780 |
Epstein | +0.774 |
sonian | +0.770 |
Neg logit contributions | |
王 | -0.902 |
dullah | -0.808 |
alde | -0.746 |
avis | -0.745 |
alky | -0.742 |
Token | Windows |
Token ID | 3964 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
王 | +0.336 |
dullah | +0.302 |
alky | +0.282 |
alde | +0.279 |
avis | +0.279 |
Neg logit contributions | |
downstream | -0.328 |
ervation | -0.288 |
sonian | -0.278 |
reservation | -0.275 |
Epstein | -0.275 |
Token | PCs |
Token ID | 21869 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
王 | +1.680 |
dullah | +1.502 |
alky | +1.435 |
avis | +1.433 |
bin | +1.426 |
Neg logit contributions | |
downstream | -1.404 |
ervation | -1.237 |
sonian | -1.182 |
GMT | -1.170 |
Epstein | -1.169 |
Token | . |
Token ID | 13 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
downstream | +0.433 |
ervation | +0.374 |
sonian | +0.359 |
reservation | +0.358 |
Epstein | +0.357 |
Neg logit contributions | |
王 | -0.520 |
dullah | -0.470 |
alky | -0.439 |
avis | -0.438 |
alde | -0.437 |
Token |
|
Token ID | 198 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
downstream | +0.167 |
ervation | +0.146 |
reservation | +0.142 |
sonian | +0.141 |
Epstein | +0.140 |
Neg logit contributions | |
王 | -0.151 |
dullah | -0.135 |
alky | -0.124 |
avis | -0.124 |
alde | -0.124 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
downstream | +0.204 |
ervation | +0.178 |
reservation | +0.172 |
sonian | +0.171 |
Epstein | +0.170 |
Neg logit contributions | |
王 | -0.189 |
dullah | -0.168 |
avis | -0.155 |
bin | -0.155 |
alky | -0.154 |
Token | The |
Token ID | 464 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
downstream | +0.177 |
ervation | +0.155 |
reservation | +0.151 |
Epstein | +0.150 |
sonian | +0.150 |
Neg logit contributions | |
王 | -0.149 |
dullah | -0.132 |
alky | -0.121 |
alde | -0.121 |
avis | -0.121 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
王 | +0.866 |
dullah | +0.789 |
alde | +0.728 |
alky | +0.728 |
bin | +0.726 |
Neg logit contributions | |
downstream | -0.729 |
ervation | -0.623 |
reservation | -0.601 |
Epstein | -0.596 |
sonian | -0.590 |
Token | . |
Token ID | 13 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
downstream | +1.177 |
ervation | +1.027 |
reservation | +0.998 |
Epstein | +0.988 |
sonian | +0.987 |
Neg logit contributions | |
王 | -1.092 |
dullah | -0.978 |
alde | -0.898 |
avis | -0.897 |
alky | -0.896 |
Token | � |
Token ID | 564 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
王 | +0.225 |
dullah | +0.202 |
avis | +0.186 |
alde | +0.186 |
alky | +0.186 |
Neg logit contributions | |
downstream | -0.230 |
ervation | -0.200 |
reservation | -0.193 |
sonian | -0.192 |
Epstein | -0.190 |
Token | � |
Token ID | 250 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
downstream | +0.079 |
ervation | +0.071 |
reservation | +0.069 |
sonian | +0.069 |
Epstein | +0.068 |
Neg logit contributions | |
王 | -0.051 |
dullah | -0.044 |
alky | -0.040 |
avis | -0.040 |
alde | -0.040 |
Token | Six |
Token ID | 21447 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
王 | +0.248 |
dullah | +0.221 |
avis | +0.205 |
alde | +0.205 |
alky | +0.204 |
Neg logit contributions | |
downstream | -0.241 |
ervation | -0.207 |
reservation | -0.201 |
sonian | -0.200 |
Epstein | -0.199 |
Token | times |
Token ID | 1661 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
王 | +0.203 |
dullah | +0.183 |
alky | +0.171 |
alde | +0.171 |
avis | +0.171 |
Neg logit contributions | |
downstream | -0.166 |
ervation | -0.142 |
reservation | -0.136 |
sonian | -0.136 |
Epstein | -0.136 |
Token | tonight |
Token ID | 9975 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
王 | +0.544 |
dullah | +0.487 |
avis | +0.456 |
alde | +0.456 |
alky | +0.452 |
Neg logit contributions | |
downstream | -0.490 |
ervation | -0.424 |
sonian | -0.413 |
reservation | -0.408 |
Epstein | -0.405 |
Token | I |
Token ID | 314 |
Feature activation | -9.26e-03 |
Pos logit contributions | |
王 | +0.258 |
dullah | +0.233 |
alde | +0.217 |
avis | +0.217 |
alky | +0.216 |
Neg logit contributions | |
downstream | -0.219 |
ervation | -0.188 |
sonian | -0.181 |
reservation | -0.181 |
Epstein | -0.179 |
Token | have |
Token ID | 423 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
downstream | +0.007 |
ervation | +0.006 |
reservation | +0.006 |
sonian | +0.006 |
Epstein | +0.006 |
Neg logit contributions | |
王 | -0.008 |
dullah | -0.007 |
avis | -0.006 |
alde | -0.006 |
alky | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
王 | +0.533 |
dullah | +0.481 |
avis | +0.447 |
alky | +0.446 |
alde | +0.446 |
Neg logit contributions | |
downstream | -0.477 |
ervation | -0.412 |
sonian | -0.396 |
reservation | -0.394 |
Epstein | -0.392 |
Token | to |
Token ID | 284 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
王 | +0.469 |
dullah | +0.423 |
avis | +0.394 |
alde | +0.393 |
alky | +0.393 |
Neg logit contributions | |
downstream | -0.421 |
ervation | -0.361 |
sonian | -0.348 |
reservation | -0.347 |
Epstein | -0.346 |
Token | cop |
Token ID | 22163 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
downstream | +0.090 |
ervation | +0.078 |
reservation | +0.075 |
sonian | +0.075 |
Epstein | +0.075 |
Neg logit contributions | |
王 | -0.091 |
dullah | -0.081 |
alky | -0.075 |
avis | -0.075 |
alde | -0.075 |
Token | ying |
Token ID | 1112 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
王 | +0.684 |
dullah | +0.602 |
alky | +0.549 |
avis | +0.549 |
alde | +0.546 |
Neg logit contributions | |
downstream | -0.940 |
ervation | -0.833 |
reservation | -0.813 |
Epstein | -0.806 |
sonian | -0.806 |
Token | a |
Token ID | 257 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
downstream | +0.677 |
ervation | +0.586 |
reservation | +0.565 |
sonian | +0.562 |
Epstein | +0.560 |
Neg logit contributions | |
王 | -0.728 |
dullah | -0.656 |
alky | -0.607 |
avis | -0.607 |
alde | -0.606 |
Token | build |
Token ID | 1382 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
downstream | +0.466 |
ervation | +0.402 |
sonian | +0.386 |
reservation | +0.386 |
Epstein | +0.384 |
Neg logit contributions | |
王 | -0.527 |
dullah | -0.474 |
alky | -0.441 |
avis | -0.440 |
alde | -0.439 |
Token | online |
Token ID | 2691 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
downstream | +1.212 |
ervation | +1.036 |
reservation | +0.999 |
Epstein | +0.988 |
sonian | +0.988 |
Neg logit contributions | |
王 | -1.445 |
dullah | -1.307 |
alde | -1.216 |
avis | -1.215 |
alky | -1.214 |
Token | , |
Token ID | 11 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
downstream | +0.879 |
ervation | +0.756 |
reservation | +0.729 |
sonian | +0.723 |
Epstein | +0.723 |
Neg logit contributions | |
王 | -0.996 |
dullah | -0.899 |
avis | -0.834 |
alky | -0.834 |
alde | -0.834 |
Token | usually |
Token ID | 3221 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
downstream | +0.319 |
ervation | +0.275 |
sonian | +0.264 |
reservation | +0.264 |
Epstein | +0.263 |
Neg logit contributions | |
王 | -0.361 |
dullah | -0.326 |
alky | -0.303 |
avis | -0.303 |
alde | -0.302 |
Token | a |
Token ID | 257 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
downstream | +0.223 |
ervation | +0.193 |
reservation | +0.186 |
sonian | +0.186 |
Epstein | +0.184 |
Neg logit contributions | |
王 | -0.248 |
dullah | -0.224 |
alky | -0.208 |
avis | -0.208 |
alde | -0.207 |
Token | winning |
Token ID | 5442 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
downstream | +0.231 |
ervation | +0.201 |
sonian | +0.193 |
reservation | +0.193 |
Epstein | +0.192 |
Neg logit contributions | |
王 | -0.244 |
dullah | -0.219 |
alky | -0.203 |
avis | -0.202 |
alde | -0.202 |
Token | one |
Token ID | 530 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
downstream | +1.084 |
ervation | +0.925 |
reservation | +0.887 |
sonian | +0.882 |
Epstein | +0.880 |
Neg logit contributions | |
王 | -1.360 |
dullah | -1.233 |
alky | -1.149 |
avis | -1.148 |
alde | -1.148 |
Token | , |
Token ID | 11 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
downstream | +1.229 |
ervation | +1.055 |
reservation | +1.018 |
sonian | +1.009 |
Epstein | +1.009 |
Neg logit contributions | |
王 | -1.409 |
dullah | -1.273 |
avis | -1.181 |
alde | -1.181 |
alky | -1.181 |
Token | former |
Token ID | 1966 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
王 | +0.181 |
dullah | +0.164 |
alky | +0.152 |
avis | +0.152 |
alde | +0.152 |
Neg logit contributions | |
downstream | -0.159 |
ervation | -0.137 |
reservation | -0.131 |
sonian | -0.131 |
Epstein | -0.130 |
Token | Luxembourg |
Token ID | 31864 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
王 | +0.907 |
dullah | +0.815 |
avis | +0.757 |
alky | +0.755 |
bin | +0.751 |
Neg logit contributions | |
downstream | -0.864 |
ervation | -0.759 |
sonian | -0.727 |
reservation | -0.719 |
Epstein | -0.712 |
Token | premier |
Token ID | 18256 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
王 | +1.701 |
dullah | +1.549 |
avis | +1.453 |
alde | +1.452 |
alky | +1.449 |
Neg logit contributions | |
downstream | -1.239 |
ervation | -1.052 |
reservation | -1.005 |
sonian | -1.001 |
Epstein | -0.994 |
Token | Jean |
Token ID | 11320 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
downstream | +1.304 |
ervation | +1.113 |
reservation | +1.069 |
Epstein | +1.065 |
sonian | +1.058 |
Neg logit contributions | |
王 | -1.562 |
dullah | -1.406 |
alde | -1.310 |
avis | -1.301 |
alky | -1.298 |
Token | - |
Token ID | 12 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
downstream | +0.517 |
ervation | +0.447 |
reservation | +0.432 |
Epstein | +0.430 |
sonian | +0.429 |
Neg logit contributions | |
王 | -0.547 |
dullah | -0.492 |
avis | -0.454 |
alky | -0.453 |
alde | -0.453 |
Token | Cla |
Token ID | 47404 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
王 | +0.165 |
dullah | +0.152 |
avis | +0.144 |
alde | +0.143 |
alky | +0.143 |
Neg logit contributions | |
downstream | -0.081 |
ervation | -0.065 |
reservation | -0.061 |
sonian | -0.060 |
Epstein | -0.060 |
Token | ude |
Token ID | 2507 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
王 | +0.548 |
dullah | +0.478 |
alde | +0.430 |
alky | +0.430 |
bin | +0.426 |
Neg logit contributions | |
downstream | -0.806 |
ervation | -0.718 |
reservation | -0.698 |
Epstein | -0.693 |
sonian | -0.692 |
Token | Jun |
Token ID | 7653 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
downstream | +0.109 |
ervation | +0.096 |
Epstein | +0.094 |
reservation | +0.093 |
Duffy | +0.093 |
Neg logit contributions | |
王 | -0.088 |
dullah | -0.078 |
alde | -0.071 |
alky | -0.071 |
avis | -0.071 |
Token | cker |
Token ID | 15280 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
王 | +1.921 |
dullah | +1.753 |
alde | +1.666 |
avis | +1.665 |
bin | +1.656 |
Neg logit contributions | |
downstream | -1.321 |
reservation | -1.093 |
ervation | -1.092 |
sonian | -1.069 |
Epstein | -1.062 |
Token | , |
Token ID | 11 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
downstream | +0.114 |
ervation | +0.100 |
reservation | +0.096 |
sonian | +0.095 |
Epstein | +0.095 |
Neg logit contributions | |
王 | -0.115 |
dullah | -0.103 |
avis | -0.095 |
alde | -0.095 |
alky | -0.095 |
Token | she |
Token ID | 673 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
王 | +0.309 |
dullah | +0.278 |
avis | +0.258 |
alky | +0.258 |
alde | +0.257 |
Neg logit contributions | |
downstream | -0.285 |
ervation | -0.247 |
reservation | -0.238 |
sonian | -0.237 |
Epstein | -0.235 |
Token | and |
Token ID | 290 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
王 | +0.064 |
dullah | +0.057 |
alde | +0.053 |
avis | +0.053 |
alky | +0.053 |
Neg logit contributions | |
downstream | -0.064 |
ervation | -0.056 |
reservation | -0.054 |
Epstein | -0.054 |
sonian | -0.054 |
Token | hands |
Token ID | 2832 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
王 | +0.086 |
dullah | +0.074 |
alde | +0.067 |
avis | +0.067 |
alky | +0.067 |
Neg logit contributions | |
downstream | -0.127 |
ervation | -0.114 |
reservation | -0.110 |
Epstein | -0.110 |
sonian | -0.110 |
Token | ."[ |
Token ID | 18161 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
王 | +0.632 |
dullah | +0.567 |
avis | +0.528 |
alde | +0.526 |
bin | +0.524 |
Neg logit contributions | |
downstream | -0.633 |
ervation | -0.552 |
reservation | -0.535 |
Epstein | -0.533 |
sonian | -0.532 |
Token | 59 |
Token ID | 3270 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
王 | +1.268 |
dullah | +1.115 |
avis | +1.016 |
alky | +1.009 |
alde | +1.009 |
Neg logit contributions | |
downstream | -1.672 |
ervation | -1.478 |
reservation | -1.442 |
Epstein | -1.428 |
sonian | -1.425 |
Token | ] |
Token ID | 60 |
Feature activation | -6.23e-02 |
Pos logit contributions | |
王 | +0.070 |
dullah | +0.063 |
avis | +0.058 |
alde | +0.058 |
bin | +0.058 |
Neg logit contributions | |
downstream | -0.069 |
ervation | -0.060 |
reservation | -0.058 |
Epstein | -0.057 |
sonian | -0.057 |
Token | The |
Token ID | 383 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
downstream | +0.052 |
ervation | +0.045 |
reservation | +0.044 |
sonian | +0.044 |
Epstein | +0.044 |
Neg logit contributions | |
王 | -0.048 |
dullah | -0.043 |
avis | -0.039 |
alky | -0.039 |
alde | -0.039 |
Token | university |
Token ID | 6403 |
Feature activation | +8.37e-03 |
Pos logit contributions | |
王 | +0.069 |
dullah | +0.062 |
avis | +0.058 |
alky | +0.058 |
alde | +0.057 |
Neg logit contributions | |
downstream | -0.063 |
ervation | -0.054 |
reservation | -0.052 |
sonian | -0.052 |
Epstein | -0.052 |
Token | 's |
Token ID | 338 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
王 | +0.007 |
dullah | +0.006 |
avis | +0.006 |
alde | +0.006 |
alky | +0.006 |
Neg logit contributions | |
downstream | -0.006 |
ervation | -0.006 |
reservation | -0.005 |
Epstein | -0.005 |
GMT | -0.005 |
Token | vice |
Token ID | 7927 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
王 | +0.175 |
dullah | +0.157 |
avis | +0.146 |
alky | +0.146 |
alde | +0.146 |
Neg logit contributions | |
downstream | -0.155 |
ervation | -0.134 |
reservation | -0.128 |
sonian | -0.128 |
Epstein | -0.127 |
Token | president |
Token ID | 1893 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
downstream | +1.148 |
ervation | +0.984 |
reservation | +0.948 |
sonian | +0.940 |
Epstein | +0.940 |
Neg logit contributions | |
王 | -1.354 |
dullah | -1.226 |
avis | -1.139 |
alde | -1.138 |
alky | -1.138 |
Token | recalled |
Token ID | 12433 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
downstream | +0.387 |
ervation | +0.334 |
reservation | +0.322 |
sonian | +0.319 |
Epstein | +0.318 |
Neg logit contributions | |
王 | -0.439 |
dullah | -0.396 |
avis | -0.368 |
alky | -0.367 |
alde | -0.367 |
Token | . |
Token ID | 13 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
downstream | +0.692 |
ervation | +0.599 |
reservation | +0.581 |
Epstein | +0.575 |
sonian | +0.574 |
Neg logit contributions | |
王 | -0.684 |
dullah | -0.613 |
alky | -0.565 |
alde | -0.564 |
avis | -0.563 |
Token | � |
Token ID | 447 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
downstream | +0.224 |
ervation | +0.194 |
reservation | +0.187 |
Epstein | +0.187 |
sonian | +0.185 |
Neg logit contributions | |
王 | -0.236 |
dullah | -0.213 |
alde | -0.197 |
alky | -0.196 |
avis | -0.196 |
Token | � |
Token ID | 251 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
downstream | +0.020 |
ervation | +0.018 |
reservation | +0.017 |
sonian | +0.017 |
Epstein | +0.017 |
Neg logit contributions | |
王 | -0.019 |
dullah | -0.017 |
alky | -0.015 |
avis | -0.015 |
alde | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
王 | +0.441 |
dullah | +0.394 |
avis | +0.363 |
alky | +0.363 |
alde | +0.363 |
Neg logit contributions | |
downstream | -0.458 |
ervation | -0.399 |
reservation | -0.386 |
sonian | -0.384 |
Epstein | -0.382 |
Token |
|
Token ID | 198 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
downstream | +0.028 |
ervation | +0.025 |
reservation | +0.024 |
sonian | +0.024 |
Epstein | +0.024 |
Neg logit contributions | |
王 | -0.026 |
dullah | -0.023 |
bin | -0.021 |
avis | -0.021 |
alky | -0.021 |
Token | • |
Token ID | 3581 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
王 | +0.044 |
dullah | +0.039 |
avis | +0.036 |
alky | +0.036 |
alde | +0.036 |
Neg logit contributions | |
downstream | -0.050 |
ervation | -0.044 |
reservation | -0.043 |
sonian | -0.042 |
Epstein | -0.042 |
Token | Position |
Token ID | 23158 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
王 | +0.219 |
dullah | +0.195 |
alde | +0.179 |
alky | +0.179 |
avis | +0.179 |
Neg logit contributions | |
downstream | -0.243 |
ervation | -0.213 |
reservation | -0.207 |
Epstein | -0.206 |
sonian | -0.205 |
Token | on |
Token ID | 319 |
Feature activation | -8.81e-02 |
Pos logit contributions | |
王 | +0.012 |
dullah | +0.011 |
avis | +0.010 |
alde | +0.010 |
alky | +0.010 |
Neg logit contributions | |
downstream | -0.012 |
ervation | -0.010 |
reservation | -0.010 |
sonian | -0.010 |
Epstein | -0.010 |
Token | pot |
Token ID | 1787 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
downstream | +0.073 |
ervation | +0.064 |
reservation | +0.062 |
Epstein | +0.061 |
sonian | +0.061 |
Neg logit contributions | |
王 | -0.069 |
dullah | -0.061 |
alde | -0.056 |
alky | -0.056 |
avis | -0.056 |
Token | : |
Token ID | 25 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
downstream | +0.360 |
ervation | +0.314 |
reservation | +0.302 |
sonian | +0.301 |
Epstein | +0.300 |
Neg logit contributions | |
王 | -0.359 |
dullah | -0.322 |
avis | -0.298 |
alky | -0.297 |
alde | -0.297 |
Token | Personally |
Token ID | 31342 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
王 | +0.135 |
dullah | +0.120 |
avis | +0.110 |
alky | +0.110 |
alde | +0.110 |
Neg logit contributions | |
downstream | -0.154 |
ervation | -0.135 |
reservation | -0.131 |
sonian | -0.130 |
Epstein | -0.130 |
Token | 46 |
Token ID | 3510 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
王 | +0.368 |
dullah | +0.328 |
avis | +0.302 |
alky | +0.301 |
alde | +0.301 |
Neg logit contributions | |
downstream | -0.393 |
ervation | -0.341 |
reservation | -0.332 |
sonian | -0.329 |
Epstein | -0.327 |
Token | . |
Token ID | 13 |
Feature activation | -8.96e-01 |
Pos logit contributions | |
downstream | +0.628 |
ervation | +0.546 |
reservation | +0.528 |
sonian | +0.526 |
Epstein | +0.524 |
Neg logit contributions | |
王 | -0.624 |
dullah | -0.559 |
alky | -0.517 |
avis | -0.516 |
alde | -0.516 |
Token | 35 |
Token ID | 2327 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
downstream | +0.687 |
ervation | +0.592 |
reservation | +0.571 |
sonian | +0.567 |
Epstein | +0.565 |
Neg logit contributions | |
王 | -0.752 |
dullah | -0.678 |
alky | -0.628 |
avis | -0.628 |
alde | -0.627 |
Token | million |
Token ID | 1510 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
downstream | +0.224 |
ervation | +0.188 |
reservation | +0.180 |
sonian | +0.180 |
Epstein | +0.179 |
Neg logit contributions | |
王 | -0.321 |
dullah | -0.293 |
alky | -0.274 |
alde | -0.274 |
avis | -0.274 |
Token | that |
Token ID | 326 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
王 | +0.951 |
dullah | +0.856 |
alky | +0.798 |
alde | +0.795 |
avis | +0.795 |
Neg logit contributions | |
downstream | -0.848 |
ervation | -0.733 |
sonian | -0.710 |
reservation | -0.702 |
GMT | -0.697 |
Token | Melvin |
Token ID | 48834 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
downstream | +0.156 |
ervation | +0.136 |
reservation | +0.131 |
sonian | +0.130 |
Epstein | +0.130 |
Neg logit contributions | |
王 | -0.148 |
dullah | -0.132 |
avis | -0.122 |
alde | -0.122 |
alky | -0.122 |
Token | Upton |
Token ID | 47755 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
downstream | +0.989 |
ervation | +0.865 |
reservation | +0.838 |
Epstein | +0.834 |
sonian | +0.830 |
Neg logit contributions | |
王 | -0.902 |
dullah | -0.806 |
avis | -0.738 |
alde | -0.738 |
alky | -0.737 |
Token | Jr |
Token ID | 7504 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
downstream | +0.614 |
ervation | +0.530 |
reservation | +0.512 |
sonian | +0.508 |
Epstein | +0.508 |
Neg logit contributions | |
王 | -0.663 |
dullah | -0.597 |
avis | -0.553 |
alky | -0.553 |
alde | -0.553 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
downstream | +1.201 |
ervation | +1.014 |
reservation | +0.974 |
Epstein | +0.961 |
sonian | +0.958 |
Neg logit contributions | |
王 | -1.628 |
dullah | -1.483 |
alky | -1.384 |
avis | -1.381 |
alde | -1.378 |
Token | is |
Token ID | 318 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
downstream | +0.164 |
ervation | +0.142 |
reservation | +0.137 |
sonian | +0.136 |
Epstein | +0.136 |
Neg logit contributions | |
王 | -0.176 |
dullah | -0.159 |
alde | -0.147 |
alky | -0.147 |
avis | -0.147 |
Token | due |
Token ID | 2233 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
王 | +0.144 |
dullah | +0.130 |
alky | +0.122 |
bin | +0.122 |
alde | +0.122 |
Neg logit contributions | |
downstream | -0.110 |
ervation | -0.095 |
sonian | -0.092 |
reservation | -0.090 |
Epstein | -0.090 |
Token | redundant |
Token ID | 30806 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
downstream | +0.170 |
ervation | +0.147 |
reservation | +0.142 |
sonian | +0.141 |
Epstein | +0.141 |
Neg logit contributions | |
王 | -0.180 |
dullah | -0.162 |
avis | -0.150 |
alky | -0.150 |
alde | -0.149 |
Token | ) |
Token ID | 8 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
王 | +0.273 |
dullah | +0.246 |
avis | +0.228 |
alky | +0.228 |
alde | +0.228 |
Neg logit contributions | |
downstream | -0.249 |
ervation | -0.215 |
reservation | -0.207 |
sonian | -0.206 |
Epstein | -0.206 |
Token | get |
Token ID | 651 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
王 | +0.162 |
dullah | +0.146 |
avis | +0.136 |
alky | +0.136 |
alde | +0.135 |
Neg logit contributions | |
downstream | -0.150 |
ervation | -0.131 |
sonian | -0.126 |
reservation | -0.125 |
Epstein | -0.125 |
Token | out |
Token ID | 503 |
Feature activation | -7.92e-02 |
Pos logit contributions | |
downstream | +0.168 |
ervation | +0.146 |
reservation | +0.141 |
sonian | +0.141 |
Epstein | +0.140 |
Neg logit contributions | |
王 | -0.178 |
dullah | -0.160 |
avis | -0.148 |
alky | -0.148 |
alde | -0.148 |
Token | of |
Token ID | 286 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
downstream | +0.060 |
ervation | +0.052 |
reservation | +0.050 |
sonian | +0.050 |
Epstein | +0.050 |
Neg logit contributions | |
王 | -0.066 |
dullah | -0.060 |
avis | -0.056 |
alde | -0.056 |
alky | -0.056 |
Token | bed |
Token ID | 3996 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
downstream | +0.037 |
ervation | +0.032 |
sonian | +0.031 |
reservation | +0.031 |
Epstein | +0.031 |
Neg logit contributions | |
王 | -0.038 |
dullah | -0.034 |
avis | -0.032 |
alde | -0.032 |
alky | -0.032 |
Token | ready |
Token ID | 3492 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
downstream | +0.920 |
ervation | +0.789 |
reservation | +0.758 |
sonian | +0.754 |
Epstein | +0.753 |
Neg logit contributions | |
王 | -1.104 |
dullah | -1.001 |
avis | -0.932 |
alde | -0.931 |
alky | -0.930 |
Token | to |
Token ID | 284 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
downstream | +0.341 |
ervation | +0.293 |
reservation | +0.282 |
sonian | +0.281 |
Epstein | +0.280 |
Neg logit contributions | |
王 | -0.390 |
dullah | -0.352 |
alde | -0.327 |
alky | -0.327 |
avis | -0.327 |
Token | p |
Token ID | 279 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
downstream | +0.082 |
ervation | +0.072 |
sonian | +0.069 |
reservation | +0.068 |
Epstein | +0.068 |
Neg logit contributions | |
王 | -0.090 |
dullah | -0.081 |
alky | -0.075 |
avis | -0.075 |
alde | -0.075 |
Token | ounce |
Token ID | 8652 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
downstream | +0.179 |
ervation | +0.151 |
reservation | +0.146 |
sonian | +0.144 |
Epstein | +0.144 |
Neg logit contributions | |
王 | -0.243 |
dullah | -0.222 |
alky | -0.208 |
avis | -0.208 |
alde | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
downstream | +1.317 |
ervation | +1.128 |
reservation | +1.081 |
sonian | +1.077 |
Epstein | +1.072 |
Neg logit contributions | |
王 | -1.692 |
dullah | -1.536 |
avis | -1.436 |
alde | -1.431 |
alky | -1.431 |
Token | � |
Token ID | 447 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
王 | +0.212 |
dullah | +0.190 |
avis | +0.176 |
alde | +0.175 |
alky | +0.175 |
Neg logit contributions | |
downstream | -0.207 |
ervation | -0.181 |
sonian | -0.173 |
reservation | -0.173 |
Epstein | -0.172 |
Token | � |
Token ID | 247 |
Feature activation | +3.17e-01 |
Pos logit contributions | |
王 | +0.127 |
dullah | +0.113 |
alde | +0.104 |
alky | +0.103 |
avis | +0.103 |
Neg logit contributions | |
downstream | -0.143 |
ervation | -0.125 |
reservation | -0.122 |
Epstein | -0.121 |
sonian | -0.121 |
Token | s |
Token ID | 82 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
王 | +0.249 |
dullah | +0.224 |
avis | +0.206 |
alde | +0.206 |
alky | +0.206 |
Neg logit contributions | |
downstream | -0.258 |
ervation | -0.225 |
reservation | -0.217 |
sonian | -0.216 |
Epstein | -0.216 |
Token | denial |
Token ID | 14425 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
王 | +0.009 |
dullah | +0.008 |
alky | +0.007 |
avis | +0.007 |
alde | +0.007 |
Neg logit contributions | |
downstream | -0.008 |
ervation | -0.007 |
sonian | -0.007 |
reservation | -0.007 |
Epstein | -0.007 |
Token | of |
Token ID | 286 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
downstream | +0.945 |
ervation | +0.814 |
reservation | +0.786 |
sonian | +0.779 |
Epstein | +0.779 |
Neg logit contributions | |
王 | -1.042 |
dullah | -0.940 |
avis | -0.871 |
alky | -0.871 |
alde | -0.871 |
Token | business |
Token ID | 1597 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
downstream | +0.033 |
ervation | +0.029 |
sonian | +0.028 |
reservation | +0.027 |
Epstein | +0.027 |
Neg logit contributions | |
王 | -0.036 |
dullah | -0.032 |
avis | -0.030 |
alky | -0.030 |
alde | -0.030 |
Token | visas |
Token ID | 23727 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
downstream | +0.664 |
ervation | +0.571 |
reservation | +0.553 |
Epstein | +0.547 |
sonian | +0.546 |
Neg logit contributions | |
王 | -0.730 |
dullah | -0.658 |
alde | -0.610 |
alky | -0.609 |
avis | -0.609 |
Token | to |
Token ID | 284 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
downstream | +1.257 |
ervation | +1.078 |
reservation | +1.044 |
Epstein | +1.032 |
sonian | +1.026 |
Neg logit contributions | |
王 | -1.425 |
dullah | -1.287 |
alky | -1.194 |
alde | -1.193 |
avis | -1.193 |
Token | Swed |
Token ID | 7289 |
Feature activation | +6.54e-01 |
Pos logit contributions | |
王 | +0.011 |
dullah | +0.010 |
avis | +0.009 |
alky | +0.009 |
alde | +0.009 |
Neg logit contributions | |
downstream | -0.015 |
ervation | -0.013 |
reservation | -0.013 |
sonian | -0.013 |
Epstein | -0.013 |
Token | es |
Token ID | 274 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
王 | +0.481 |
dullah | +0.427 |
avis | +0.390 |
alde | +0.390 |
alky | +0.390 |
Neg logit contributions | |
downstream | -0.568 |
ervation | -0.499 |
reservation | -0.484 |
sonian | -0.480 |
Epstein | -0.480 |
Token | threatened |
Token ID | 8556 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
downstream | +0.482 |
ervation | +0.418 |
reservation | +0.404 |
sonian | +0.402 |
Epstein | +0.401 |
Neg logit contributions | |
王 | -0.499 |
dullah | -0.449 |
avis | -0.415 |
alde | -0.415 |
alky | -0.415 |
Token | Spl |
Token ID | 13341 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
downstream | +0.161 |
ervation | +0.141 |
reservation | +0.137 |
sonian | +0.136 |
Epstein | +0.136 |
Neg logit contributions | |
王 | -0.139 |
dullah | -0.124 |
alky | -0.113 |
alde | -0.113 |
avis | -0.113 |
Token | inter |
Token ID | 3849 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
王 | +1.934 |
dullah | +1.704 |
avis | +1.510 |
alde | +1.495 |
bin | +1.490 |
Neg logit contributions | |
downstream | -4.155 |
ervation | -3.711 |
reservation | -3.691 |
Duffy | -3.626 |
Epstein | -3.626 |
Token | 's |
Token ID | 338 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
downstream | +0.148 |
ervation | +0.127 |
reservation | +0.122 |
sonian | +0.121 |
Epstein | +0.121 |
Neg logit contributions | |
王 | -0.176 |
dullah | -0.160 |
avis | -0.148 |
alde | -0.148 |
alky | -0.148 |
Token | Master |
Token ID | 5599 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
王 | +0.359 |
dullah | +0.322 |
alky | +0.299 |
alde | +0.298 |
avis | +0.296 |
Neg logit contributions | |
downstream | -0.344 |
ervation | -0.300 |
sonian | -0.290 |
reservation | -0.288 |
Epstein | -0.287 |
Token | Yoshi |
Token ID | 38936 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
王 | +0.910 |
dullah | +0.820 |
alky | +0.763 |
alde | +0.763 |
avis | +0.761 |
Neg logit contributions | |
downstream | -0.755 |
ervation | -0.651 |
reservation | -0.630 |
sonian | -0.626 |
Epstein | -0.620 |
Token | ... |
Token ID | 986 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
王 | +0.800 |
dullah | +0.725 |
avis | +0.673 |
alde | +0.673 |
alky | +0.672 |
Neg logit contributions | |
downstream | -0.666 |
ervation | -0.574 |
reservation | -0.553 |
sonian | -0.549 |
Epstein | -0.546 |
Token | they |
Token ID | 484 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
downstream | +1.244 |
ervation | +1.073 |
reservation | +1.034 |
Epstein | +1.026 |
sonian | +1.020 |
Neg logit contributions | |
王 | -1.361 |
dullah | -1.226 |
alde | -1.138 |
avis | -1.136 |
alky | -1.136 |
Token | had |
Token ID | 550 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
王 | +0.274 |
dullah | +0.245 |
avis | +0.227 |
alde | +0.226 |
alky | +0.226 |
Neg logit contributions | |
downstream | -0.278 |
ervation | -0.243 |
reservation | -0.234 |
sonian | -0.233 |
Epstein | -0.232 |
Token | to |
Token ID | 284 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
downstream | +0.241 |
ervation | +0.207 |
reservation | +0.199 |
Epstein | +0.198 |
sonian | +0.198 |
Neg logit contributions | |
王 | -0.276 |
dullah | -0.250 |
avis | -0.232 |
alde | -0.232 |
alky | -0.232 |
Token | defeat |
Token ID | 7433 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
王 | +0.076 |
dullah | +0.069 |
avis | +0.064 |
alky | +0.063 |
alde | +0.063 |
Neg logit contributions | |
downstream | -0.068 |
ervation | -0.059 |
sonian | -0.057 |
reservation | -0.057 |
Epstein | -0.056 |
Token | the |
Token ID | 262 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
王 | +0.233 |
dullah | +0.207 |
alky | +0.190 |
avis | +0.189 |
alde | +0.189 |
Neg logit contributions | |
downstream | -0.266 |
ervation | -0.234 |
reservation | -0.227 |
sonian | -0.225 |
Epstein | -0.225 |
Token | . |
Token ID | 13 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
downstream | +0.569 |
ervation | +0.493 |
sonian | +0.475 |
reservation | +0.472 |
Epstein | +0.470 |
Neg logit contributions | |
王 | -0.687 |
dullah | -0.623 |
avis | -0.580 |
alde | -0.579 |
alky | -0.578 |
Token |
|
Token ID | 198 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
downstream | +0.351 |
ervation | +0.308 |
sonian | +0.297 |
reservation | +0.296 |
Epstein | +0.293 |
Neg logit contributions | |
王 | -0.332 |
dullah | -0.299 |
avis | -0.274 |
alky | -0.273 |
bin | -0.273 |
Token |
|
Token ID | 198 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
王 | +0.079 |
dullah | +0.071 |
bin | +0.065 |
avis | +0.065 |
alky | +0.065 |
Neg logit contributions | |
downstream | -0.087 |
ervation | -0.076 |
reservation | -0.073 |
sonian | -0.073 |
Epstein | -0.072 |
Token | Among |
Token ID | 14311 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
王 | +0.089 |
dullah | +0.079 |
bin | +0.073 |
avis | +0.072 |
alky | +0.072 |
Neg logit contributions | |
downstream | -0.098 |
ervation | -0.085 |
reservation | -0.083 |
sonian | -0.082 |
Epstein | -0.082 |
Token | other |
Token ID | 584 |
Feature activation | +8.49e-01 |
Pos logit contributions | |
downstream | +0.652 |
ervation | +0.573 |
reservation | +0.554 |
sonian | +0.553 |
Epstein | +0.550 |
Neg logit contributions | |
王 | -0.599 |
dullah | -0.538 |
alky | -0.493 |
alde | -0.493 |
avis | -0.492 |
Token | things |
Token ID | 1243 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
王 | +0.607 |
dullah | +0.542 |
alky | +0.500 |
bin | +0.497 |
avis | +0.495 |
Neg logit contributions | |
downstream | -0.719 |
ervation | -0.653 |
sonian | -0.634 |
GMT | -0.624 |
Epstein | -0.623 |
Token | , |
Token ID | 11 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
downstream | +1.191 |
ervation | +1.010 |
reservation | +0.974 |
Epstein | +0.966 |
sonian | +0.959 |
Neg logit contributions | |
王 | -1.496 |
dullah | -1.357 |
avis | -1.266 |
alky | -1.264 |
alde | -1.264 |
Token | the |
Token ID | 262 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
downstream | +0.044 |
ervation | +0.038 |
sonian | +0.037 |
reservation | +0.037 |
Epstein | +0.037 |
Neg logit contributions | |
王 | -0.042 |
dullah | -0.037 |
alky | -0.034 |
alde | -0.034 |
avis | -0.034 |
Token | In |
Token ID | 554 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
王 | +0.037 |
dullah | +0.034 |
alky | +0.032 |
avis | +0.031 |
alde | +0.031 |
Neg logit contributions | |
downstream | -0.030 |
ervation | -0.026 |
sonian | -0.025 |
GMT | -0.025 |
Epstein | -0.025 |
Token | uit |
Token ID | 5013 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
王 | +0.288 |
dullah | +0.267 |
avis | +0.251 |
alky | +0.251 |
alde | +0.251 |
Neg logit contributions | |
downstream | -0.169 |
ervation | -0.136 |
Epstein | -0.131 |
reservation | -0.131 |
sonian | -0.131 |
Token | were |
Token ID | 547 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
王 | +0.036 |
dullah | +0.032 |
alky | +0.030 |
alde | +0.030 |
bin | +0.030 |
Neg logit contributions | |
downstream | -0.033 |
ervation | -0.029 |
sonian | -0.028 |
Epstein | -0.028 |
GMT | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
王 | +0.363 |
dullah | +0.326 |
alky | +0.305 |
bin | +0.303 |
alde | +0.303 |
Neg logit contributions | |
downstream | -0.335 |
ervation | -0.292 |
sonian | -0.283 |
GMT | -0.278 |
Epstein | -0.277 |
Token | the |
Token ID | 262 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
王 | +0.311 |
dullah | +0.279 |
alky | +0.261 |
avis | +0.259 |
bin | +0.259 |
Neg logit contributions | |
downstream | -0.295 |
ervation | -0.257 |
sonian | -0.248 |
GMT | -0.245 |
Epstein | -0.244 |
Token | policy |
Token ID | 2450 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
王 | +0.480 |
dullah | +0.431 |
alky | +0.404 |
avis | +0.402 |
bin | +0.400 |
Neg logit contributions | |
downstream | -0.434 |
ervation | -0.376 |
sonian | -0.365 |
GMT | -0.359 |
Epstein | -0.358 |
Token | of |
Token ID | 286 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
王 | +0.181 |
dullah | +0.163 |
alky | +0.151 |
avis | +0.151 |
alde | +0.151 |
Neg logit contributions | |
downstream | -0.164 |
ervation | -0.142 |
reservation | -0.137 |
sonian | -0.136 |
Epstein | -0.136 |
Token | brutal |
Token ID | 12399 |
Feature activation | +5.67e-01 |
Pos logit contributions | |
王 | +0.336 |
dullah | +0.302 |
alky | +0.282 |
avis | +0.281 |
alde | +0.280 |
Neg logit contributions | |
downstream | -0.304 |
ervation | -0.264 |
sonian | -0.256 |
GMT | -0.251 |
Epstein | -0.251 |
Token | repression |
Token ID | 28279 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
王 | +0.476 |
dullah | +0.426 |
alky | +0.402 |
avis | +0.398 |
alde | +0.398 |
Neg logit contributions | |
downstream | -0.426 |
ervation | -0.368 |
sonian | -0.359 |
Epstein | -0.352 |
GMT | -0.351 |
Token | … |
Token ID | 1399 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
downstream | +2.785 |
ervation | +2.394 |
reservation | +2.320 |
Epstein | +2.300 |
sonian | +2.271 |
Neg logit contributions | |
王 | -2.939 |
dullah | -2.652 |
alde | -2.449 |
alky | -2.447 |
avis | -2.437 |
Token | In |
Token ID | 818 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
downstream | +0.379 |
ervation | +0.331 |
reservation | +0.320 |
sonian | +0.318 |
Epstein | +0.318 |
Neg logit contributions | |
王 | -0.358 |
dullah | -0.320 |
avis | -0.294 |
alky | -0.294 |
alde | -0.294 |
Token | order |
Token ID | 1502 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
王 | +0.179 |
dullah | +0.162 |
avis | +0.150 |
alde | +0.150 |
alky | +0.150 |
Neg logit contributions | |
downstream | -0.154 |
ervation | -0.132 |
sonian | -0.128 |
reservation | -0.127 |
Epstein | -0.126 |
Token | to |
Token ID | 284 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
downstream | +0.694 |
ervation | +0.595 |
reservation | +0.574 |
Epstein | +0.568 |
sonian | +0.566 |
Neg logit contributions | |
王 | -0.814 |
dullah | -0.737 |
alky | -0.684 |
alde | -0.684 |
avis | -0.683 |
Token | realize |
Token ID | 6537 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
王 | +0.246 |
dullah | +0.222 |
avis | +0.206 |
alky | +0.206 |
alde | +0.205 |
Neg logit contributions | |
downstream | -0.216 |
ervation | -0.187 |
sonian | -0.180 |
reservation | -0.179 |
Epstein | -0.178 |
Token | 91 |
Token ID | 10495 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
downstream | +0.027 |
ervation | +0.024 |
sonian | +0.023 |
reservation | +0.023 |
Epstein | +0.022 |
Neg logit contributions | |
王 | -0.029 |
dullah | -0.026 |
avis | -0.024 |
alde | -0.024 |
alky | -0.024 |
Token | % |
Token ID | 4 |
Feature activation | +6.82e-02 |
Pos logit contributions | |
王 | +0.799 |
dullah | +0.731 |
alde | +0.686 |
alky | +0.683 |
avis | +0.681 |
Neg logit contributions | |
downstream | -0.571 |
ervation | -0.478 |
sonian | -0.464 |
reservation | -0.458 |
Epstein | -0.457 |
Token | are |
Token ID | 389 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
王 | +0.059 |
dullah | +0.054 |
avis | +0.050 |
alky | +0.050 |
alde | +0.050 |
Neg logit contributions | |
downstream | -0.050 |
ervation | -0.043 |
sonian | -0.041 |
reservation | -0.041 |
Epstein | -0.041 |
Token | in |
Token ID | 287 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
downstream | +0.245 |
ervation | +0.212 |
sonian | +0.204 |
reservation | +0.203 |
Epstein | +0.202 |
Neg logit contributions | |
王 | -0.275 |
dullah | -0.248 |
avis | -0.231 |
alde | -0.230 |
alky | -0.230 |
Token | for |
Token ID | 329 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
王 | +0.094 |
dullah | +0.085 |
avis | +0.078 |
alde | +0.078 |
alky | +0.078 |
Neg logit contributions | |
downstream | -0.085 |
ervation | -0.073 |
reservation | -0.071 |
sonian | -0.070 |
Epstein | -0.070 |
Token | nonviolent |
Token ID | 42582 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
王 | +0.107 |
dullah | +0.097 |
avis | +0.090 |
alde | +0.089 |
bin | +0.089 |
Neg logit contributions | |
downstream | -0.103 |
ervation | -0.089 |
sonian | -0.086 |
reservation | -0.086 |
Epstein | -0.085 |
Token | crimes |
Token ID | 6741 |
Feature activation | -5.45e-01 |
Pos logit contributions | |
downstream | +3.192 |
ervation | +2.746 |
reservation | +2.680 |
Epstein | +2.669 |
sonian | +2.620 |
Neg logit contributions | |
王 | -3.173 |
dullah | -2.849 |
alky | -2.629 |
alde | -2.621 |
avis | -2.607 |
Token | . |
Token ID | 13 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
downstream | +0.391 |
ervation | +0.335 |
reservation | +0.323 |
sonian | +0.323 |
Epstein | +0.320 |
Neg logit contributions | |
王 | -0.481 |
dullah | -0.436 |
avis | -0.407 |
alde | -0.405 |
alky | -0.404 |
Token | No |
Token ID | 1400 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
downstream | +0.049 |
ervation | +0.043 |
reservation | +0.042 |
sonian | +0.041 |
Epstein | +0.041 |
Neg logit contributions | |
王 | -0.047 |
dullah | -0.042 |
avis | -0.039 |
alky | -0.039 |
alde | -0.039 |
Token | dictator |
Token ID | 26671 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
downstream | +0.078 |
ervation | +0.067 |
sonian | +0.064 |
Epstein | +0.064 |
reservation | +0.064 |
Neg logit contributions | |
王 | -0.088 |
dullah | -0.080 |
avis | -0.074 |
alde | -0.074 |
alky | -0.074 |
Token | in |
Token ID | 287 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
downstream | +0.824 |
ervation | +0.716 |
reservation | +0.692 |
sonian | +0.687 |
Epstein | +0.686 |
Neg logit contributions | |
王 | -0.830 |
dullah | -0.745 |
avis | -0.688 |
alde | -0.688 |
alky | -0.687 |
Token | building |
Token ID | 2615 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
王 | +0.053 |
dullah | +0.048 |
avis | +0.044 |
alky | +0.044 |
alde | +0.044 |
Neg logit contributions | |
downstream | -0.049 |
ervation | -0.043 |
sonian | -0.041 |
reservation | -0.041 |
Epstein | -0.041 |
Token | adjacent |
Token ID | 15909 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
downstream | +1.139 |
ervation | +0.978 |
reservation | +0.940 |
sonian | +0.936 |
Epstein | +0.934 |
Neg logit contributions | |
王 | -1.331 |
dullah | -1.204 |
avis | -1.119 |
alky | -1.117 |
alde | -1.116 |
Token | to |
Token ID | 284 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
downstream | +0.443 |
ervation | +0.382 |
reservation | +0.368 |
sonian | +0.366 |
Epstein | +0.365 |
Neg logit contributions | |
王 | -0.500 |
dullah | -0.452 |
avis | -0.419 |
alky | -0.418 |
alde | -0.418 |
Token | the |
Token ID | 262 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
王 | +0.188 |
dullah | +0.168 |
avis | +0.155 |
alky | +0.155 |
alde | +0.154 |
Neg logit contributions | |
downstream | -0.203 |
ervation | -0.178 |
sonian | -0.173 |
reservation | -0.172 |
Epstein | -0.170 |
Token | office |
Token ID | 2607 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
王 | +0.224 |
dullah | +0.201 |
avis | +0.187 |
alky | +0.187 |
alde | +0.186 |
Neg logit contributions | |
downstream | -0.214 |
ervation | -0.186 |
sonian | -0.179 |
reservation | -0.177 |
Epstein | -0.177 |
Token | lobby |
Token ID | 10866 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
王 | +0.183 |
dullah | +0.165 |
bin | +0.154 |
avis | +0.153 |
alde | +0.153 |
Neg logit contributions | |
downstream | -0.165 |
ervation | -0.143 |
Epstein | -0.137 |
sonian | -0.137 |
reservation | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
downstream | +2.552 |
ervation | +2.171 |
reservation | +2.095 |
Epstein | +2.063 |
sonian | +2.062 |
Neg logit contributions | |
王 | -3.209 |
dullah | -2.910 |
alde | -2.711 |
alky | -2.707 |
avis | -2.707 |
Token |
|
Token ID | 198 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
downstream | +0.113 |
ervation | +0.099 |
reservation | +0.095 |
sonian | +0.095 |
Epstein | +0.094 |
Neg logit contributions | |
王 | -0.107 |
dullah | -0.096 |
avis | -0.089 |
alky | -0.088 |
alde | -0.088 |
Token |
|
Token ID | 198 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
王 | +0.065 |
dullah | +0.058 |
avis | +0.054 |
bin | +0.053 |
alky | +0.053 |
Neg logit contributions | |
downstream | -0.071 |
ervation | -0.062 |
reservation | -0.060 |
sonian | -0.059 |
Epstein | -0.059 |
Token | Ground |
Token ID | 35539 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
王 | +0.136 |
dullah | +0.121 |
avis | +0.112 |
alde | +0.111 |
alky | +0.111 |
Neg logit contributions | |
downstream | -0.143 |
ervation | -0.123 |
reservation | -0.120 |
Epstein | -0.119 |
sonian | -0.118 |
Token | floor |
Token ID | 4314 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
downstream | +0.585 |
ervation | +0.504 |
reservation | +0.484 |
sonian | +0.484 |
Epstein | +0.483 |
Neg logit contributions | |
王 | -0.657 |
dullah | -0.595 |
avis | -0.552 |
alde | -0.551 |
alky | -0.551 |
Token | , |
Token ID | 11 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
downstream | +0.405 |
ervation | +0.348 |
reservation | +0.335 |
sonian | +0.333 |
Epstein | +0.332 |
Neg logit contributions | |
王 | -0.478 |
dullah | -0.433 |
alde | -0.403 |
alky | -0.402 |
avis | -0.402 |
Token | still |
Token ID | 991 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
downstream | +0.033 |
ervation | +0.029 |
sonian | +0.028 |
reservation | +0.028 |
Epstein | +0.028 |
Neg logit contributions | |
王 | -0.033 |
dullah | -0.029 |
avis | -0.027 |
alde | -0.027 |
bin | -0.027 |
Token | has |
Token ID | 468 |
Feature activation | +6.33e-01 |
Pos logit contributions | |
downstream | +0.467 |
ervation | +0.416 |
reservation | +0.401 |
sonian | +0.399 |
Epstein | +0.397 |
Neg logit contributions | |
王 | -0.389 |
dullah | -0.347 |
avis | -0.319 |
alde | -0.318 |
bin | -0.316 |
Token | a |
Token ID | 257 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
王 | +0.509 |
dullah | +0.460 |
alde | +0.426 |
alky | +0.426 |
avis | +0.424 |
Neg logit contributions | |
downstream | -0.494 |
ervation | -0.434 |
sonian | -0.418 |
reservation | -0.410 |
GMT | -0.409 |
Token | lawsuit |
Token ID | 8087 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
王 | +0.132 |
dullah | +0.119 |
alky | +0.110 |
alde | +0.110 |
avis | +0.110 |
Neg logit contributions | |
downstream | -0.130 |
ervation | -0.114 |
sonian | -0.110 |
reservation | -0.108 |
Epstein | -0.108 |
Token | pending |
Token ID | 13310 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
downstream | +0.511 |
ervation | +0.434 |
sonian | +0.416 |
reservation | +0.414 |
Epstein | +0.409 |
Neg logit contributions | |
王 | -0.711 |
dullah | -0.649 |
alky | -0.608 |
alde | -0.607 |
avis | -0.607 |
Token | against |
Token ID | 1028 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
downstream | +2.595 |
ervation | +2.217 |
reservation | +2.139 |
Epstein | +2.121 |
sonian | +2.110 |
Neg logit contributions | |
王 | -3.103 |
dullah | -2.810 |
avis | -2.610 |
alde | -2.609 |
alky | -2.609 |
Token | the |
Token ID | 262 |
Feature activation | +6.56e-02 |
Pos logit contributions | |
downstream | +0.847 |
ervation | +0.742 |
reservation | +0.715 |
sonian | +0.715 |
Epstein | +0.707 |
Neg logit contributions | |
王 | -0.777 |
dullah | -0.693 |
alky | -0.638 |
avis | -0.637 |
alde | -0.636 |
Token | All |
Token ID | 1439 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
王 | +0.051 |
dullah | +0.046 |
alky | +0.043 |
alde | +0.042 |
avis | +0.042 |
Neg logit contributions | |
downstream | -0.053 |
ervation | -0.047 |
sonian | -0.045 |
GMT | -0.044 |
reservation | -0.044 |
Token | ent |
Token ID | 298 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
王 | +0.916 |
dullah | +0.847 |
alde | +0.803 |
avis | +0.800 |
alky | +0.797 |
Neg logit contributions | |
downstream | -0.578 |
ervation | -0.472 |
reservation | -0.456 |
Epstein | -0.450 |
sonian | -0.447 |
Token | own |
Token ID | 593 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
王 | +0.048 |
dullah | +0.043 |
alky | +0.039 |
avis | +0.039 |
alde | +0.039 |
Neg logit contributions | |
downstream | -0.054 |
ervation | -0.047 |
reservation | -0.046 |
Epstein | -0.045 |
sonian | -0.045 |
Token | a |
Token ID | 257 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
王 | +0.268 |
dullah | +0.241 |
avis | +0.223 |
alde | +0.223 |
alky | +0.223 |
Neg logit contributions | |
downstream | -0.267 |
ervation | -0.232 |
reservation | -0.225 |
sonian | -0.223 |
Epstein | -0.223 |
Token | ck |
Token ID | 694 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
王 | +0.077 |
dullah | +0.070 |
avis | +0.064 |
alky | +0.064 |
alde | +0.064 |
Neg logit contributions | |
downstream | -0.076 |
ervation | -0.066 |
reservation | -0.064 |
Epstein | -0.064 |
sonian | -0.064 |
Token | again |
Token ID | 757 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
downstream | +0.956 |
ervation | +0.834 |
reservation | +0.807 |
sonian | +0.801 |
Epstein | +0.800 |
Neg logit contributions | |
王 | -0.931 |
dullah | -0.832 |
alky | -0.767 |
alde | -0.767 |
avis | -0.766 |
Token | st |
Token ID | 336 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
王 | +2.076 |
dullah | +1.876 |
avis | +1.738 |
alde | +1.736 |
alky | +1.735 |
Neg logit contributions | |
downstream | -1.952 |
ervation | -1.692 |
reservation | -1.629 |
sonian | -1.620 |
Epstein | -1.620 |
Token | the |
Token ID | 262 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
王 | +1.399 |
dullah | +1.262 |
alky | +1.175 |
alde | +1.171 |
avis | +1.170 |
Neg logit contributions | |
downstream | -1.445 |
ervation | -1.253 |
reservation | -1.221 |
sonian | -1.217 |
Epstein | -1.207 |
Token | pseud |
Token ID | 25038 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
王 | +0.301 |
dullah | +0.271 |
avis | +0.251 |
alky | +0.251 |
alde | +0.250 |
Neg logit contributions | |
downstream | -0.286 |
ervation | -0.249 |
reservation | -0.239 |
sonian | -0.238 |
Epstein | -0.238 |
Token | o |
Token ID | 267 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
downstream | +3.339 |
ervation | +2.940 |
reservation | +2.831 |
sonian | +2.825 |
Epstein | +2.818 |
Neg logit contributions | |
王 | -2.975 |
dullah | -2.632 |
alde | -2.399 |
alky | -2.398 |
avis | -2.395 |
Token | - |
Token ID | 12 |
Feature activation | +5.43e-01 |
Pos logit contributions | |
王 | +0.387 |
dullah | +0.348 |
avis | +0.320 |
alde | +0.320 |
alky | +0.319 |
Neg logit contributions | |
downstream | -0.420 |
ervation | -0.365 |
reservation | -0.355 |
Epstein | -0.353 |
sonian | -0.352 |
Token | cent |
Token ID | 1087 |
Feature activation | -2.72e-03 |
Pos logit contributions | |
王 | +0.416 |
dullah | +0.373 |
avis | +0.343 |
alky | +0.342 |
alde | +0.341 |
Neg logit contributions | |
downstream | -0.453 |
ervation | -0.396 |
reservation | -0.384 |
Epstein | -0.381 |
sonian | -0.380 |
Token | re |
Token ID | 260 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
downstream | +0.002 |
ervation | +0.002 |
reservation | +0.002 |
sonian | +0.002 |
Epstein | +0.002 |
Neg logit contributions | |
王 | -0.002 |
dullah | -0.002 |
avis | -0.002 |
alde | -0.002 |
alky | -0.002 |
Token | s |
Token ID | 264 |
Feature activation | +8.24e-01 |
Pos logit contributions | |
downstream | +0.153 |
ervation | +0.132 |
reservation | +0.127 |
sonian | +0.126 |
Epstein | +0.126 |
Neg logit contributions | |
王 | -0.168 |
dullah | -0.152 |
avis | -0.141 |
alde | -0.141 |
alky | -0.141 |