Token | We |
Token ID | 775 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
Magical | +3.202 |
adesh | +2.992 |
ower | +2.882 |
deen | +2.851 |
Pavilion | +2.744 |
Neg logit contributions | |
opoly | -5.072 |
roc | -5.068 |
ouched | -4.736 |
upfront | -4.716 |
Msg | -4.710 |
Token | need |
Token ID | 761 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
Magical | +6.879 |
shed | +6.208 |
ing | +5.976 |
ower | +5.963 |
adesh | +5.906 |
Neg logit contributions | |
opoly | -11.279 |
roc | -10.570 |
Times | -10.463 |
ouched | -10.147 |
Concord | -10.059 |
Token | the |
Token ID | 262 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
Magical | +5.764 |
ing | +5.192 |
adesh | +5.112 |
ower | +5.032 |
shed | +4.903 |
Neg logit contributions | |
opoly | -8.552 |
roc | -8.154 |
ouched | -7.771 |
Times | -7.756 |
ustain | -7.709 |
Token | Travel |
Token ID | 13524 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
roc | +1.277 |
opoly | +1.257 |
Concord | +1.164 |
Times | +1.164 |
upfront | +1.140 |
Neg logit contributions | |
Magical | -1.570 |
adesh | -1.554 |
deen | -1.522 |
ower | -1.509 |
Piercing | -1.438 |
Token | Ban |
Token ID | 10274 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
Magical | +6.601 |
adesh | +6.467 |
ing | +6.295 |
ower | +6.241 |
Pavilion | +6.115 |
Neg logit contributions | |
roc | -5.499 |
opoly | -5.425 |
Times | -5.011 |
TEXT | -4.952 |
Concord | -4.900 |
Token | as |
Token ID | 355 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
Magical | +6.379 |
adesh | +5.992 |
ing | +5.837 |
ower | +5.710 |
deen | +5.657 |
Neg logit contributions | |
opoly | -9.639 |
roc | -9.554 |
Times | -8.861 |
Bee | -8.791 |
ailability | -8.776 |
Token | an |
Token ID | 281 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
ing | +7.356 |
Magical | +6.973 |
(@ | +6.481 |
shed | +5.474 |
Pavilion | +5.373 |
Neg logit contributions | |
ailability | -24.380 |
opoly | -24.257 |
hovah | -23.321 |
neighb | -23.160 |
ccording | -22.616 |
Token | extra |
Token ID | 3131 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Magical | +5.465 |
adesh | +5.229 |
ower | +5.069 |
shed | +5.007 |
ing | +4.953 |
Neg logit contributions | |
opoly | -4.887 |
roc | -4.690 |
Concord | -4.472 |
Times | -4.414 |
ustain | -4.386 |
Token | level |
Token ID | 1241 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
Magical | +6.627 |
ing | +6.171 |
shed | +5.616 |
ower | +5.229 |
adesh | +4.988 |
Neg logit contributions | |
ailability | -21.000 |
opoly | -20.746 |
hovah | -19.835 |
Times | -19.587 |
roc | -19.568 |
Token | of |
Token ID | 286 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Magical | +4.858 |
ing | +4.503 |
ower | +4.377 |
adesh | +4.300 |
Piercing | +4.166 |
Neg logit contributions | |
opoly | -7.890 |
roc | -7.606 |
ailability | -7.345 |
Concord | -7.191 |
Times | -7.172 |
Token | safety |
Token ID | 3747 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
Magical | +10.474 |
ing | +9.034 |
shed | +8.908 |
adesh | +8.692 |
Magic | +8.674 |
Neg logit contributions | |
opoly | -14.765 |
Times | -13.642 |
roc | -13.513 |
ailability | -13.383 |
TEXT | -13.285 |
Token | the |
Token ID | 262 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
roc | +2.887 |
opoly | +2.873 |
Concord | +2.695 |
Times | +2.660 |
upfront | +2.647 |
Neg logit contributions | |
adesh | -3.108 |
Magical | -3.100 |
deen | -2.990 |
ower | -2.920 |
Piercing | -2.844 |
Token | room |
Token ID | 2119 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
roc | +1.053 |
opoly | +1.041 |
Concord | +0.969 |
Times | +0.965 |
upfront | +0.957 |
Neg logit contributions | |
Magical | -1.202 |
adesh | -1.186 |
deen | -1.144 |
ower | -1.139 |
Piercing | -1.090 |
Token | , |
Token ID | 11 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
Magical | +2.407 |
adesh | +2.313 |
ower | +2.296 |
deen | +2.272 |
shed | +2.154 |
Neg logit contributions | |
opoly | -2.741 |
roc | -2.735 |
Times | -2.546 |
Concord | -2.533 |
Msg | -2.489 |
Token | � |
Token ID | 447 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
Magical | +3.673 |
adesh | +3.511 |
ower | +3.506 |
deen | +3.422 |
shed | +3.329 |
Neg logit contributions | |
opoly | -4.366 |
roc | -4.343 |
Times | -4.073 |
Concord | -4.037 |
ouched | -3.977 |
Token | � |
Token ID | 251 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
roc | +2.017 |
opoly | +2.003 |
Concord | +1.878 |
Times | +1.876 |
upfront | +1.861 |
Neg logit contributions | |
Magical | -1.789 |
adesh | -1.754 |
deen | -1.708 |
ower | -1.682 |
Piercing | -1.607 |
Token | H |
Token ID | 367 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
Magical | +0.200 |
adesh | +0.195 |
deen | +0.189 |
ower | +0.189 |
Piercing | +0.180 |
Neg logit contributions | |
roc | -0.203 |
opoly | -0.202 |
Times | -0.187 |
Concord | -0.187 |
upfront | -0.184 |
Token | ower |
Token ID | 789 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
ower | +13.176 |
owers | +8.756 |
adesh | +7.705 |
room | +7.647 |
ing | +7.601 |
Neg logit contributions | |
Sagan | -21.642 |
unbeliev | -21.231 |
ゴ | -21.077 |
Concord | -20.919 |
Homs | -20.851 |
Token | ton |
Token ID | 1122 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
Magical | +1.265 |
adesh | +1.242 |
ower | +1.222 |
deen | +1.207 |
shed | +1.122 |
Neg logit contributions | |
opoly | -1.692 |
roc | -1.683 |
Concord | -1.588 |
Times | -1.574 |
upfront | -1.551 |
Token | said |
Token ID | 531 |
Feature activation | +7.36e-02 |
Pos logit contributions | |
roc | +1.607 |
opoly | +1.586 |
Times | +1.446 |
Concord | +1.439 |
upfront | +1.418 |
Neg logit contributions | |
Magical | -2.948 |
adesh | -2.900 |
deen | -2.816 |
ower | -2.778 |
Piercing | -2.722 |
Token | . |
Token ID | 13 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
Magical | +0.749 |
adesh | +0.726 |
ower | +0.710 |
deen | +0.708 |
Piercing | +0.670 |
Neg logit contributions | |
roc | -0.817 |
opoly | -0.813 |
Concord | -0.756 |
Times | -0.755 |
upfront | -0.740 |
Token | � |
Token ID | 564 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
Magical | +0.153 |
adesh | +0.149 |
deen | +0.145 |
ower | +0.145 |
Piercing | +0.138 |
Neg logit contributions | |
roc | -0.158 |
opoly | -0.157 |
Concord | -0.146 |
Times | -0.146 |
upfront | -0.143 |
Token | fantastic |
Token ID | 9623 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
Magical | +3.849 |
adesh | +3.674 |
ower | +3.588 |
deen | +3.558 |
shed | +3.489 |
Neg logit contributions | |
roc | -3.629 |
opoly | -3.564 |
Times | -3.355 |
Concord | -3.318 |
Msg | -3.284 |
Token | location |
Token ID | 4067 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
Magical | +4.658 |
adesh | +4.411 |
ower | +4.380 |
shed | +4.295 |
Pavilion | +4.204 |
Neg logit contributions | |
opoly | -4.621 |
roc | -4.620 |
Times | -4.231 |
Msg | -4.220 |
TEXT | -4.185 |
Token | , |
Token ID | 11 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
roc | +1.561 |
opoly | +1.552 |
Concord | +1.444 |
upfront | +1.442 |
Times | +1.440 |
Neg logit contributions | |
Magical | -1.580 |
adesh | -1.552 |
deen | -1.504 |
ower | -1.489 |
Piercing | -1.425 |
Token | � |
Token ID | 447 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
Magical | +1.957 |
adesh | +1.875 |
ower | +1.810 |
deen | +1.801 |
shed | +1.705 |
Neg logit contributions | |
roc | -2.803 |
opoly | -2.793 |
Times | -2.624 |
Concord | -2.618 |
ouched | -2.560 |
Token | � |
Token ID | 251 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
roc | +0.955 |
opoly | +0.950 |
Concord | +0.890 |
Times | +0.890 |
upfront | +0.880 |
Neg logit contributions | |
Magical | -0.764 |
adesh | -0.743 |
deen | -0.722 |
ower | -0.716 |
Piercing | -0.679 |
Token | H |
Token ID | 367 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Magical | +0.549 |
adesh | +0.535 |
deen | +0.521 |
ower | +0.521 |
Piercing | +0.497 |
Neg logit contributions | |
roc | -0.468 |
opoly | -0.465 |
Times | -0.429 |
Concord | -0.427 |
upfront | -0.420 |
Token | ower |
Token ID | 789 |
Feature activation | +7.24e-02 |
Pos logit contributions | |
ower | +12.876 |
owers | +8.454 |
adesh | +7.517 |
ing | +7.310 |
room | +7.160 |
Neg logit contributions | |
Sagan | -21.482 |
unbeliev | -21.372 |
ゴ | -21.282 |
Concord | -21.160 |
Homs | -20.849 |
Token | ton |
Token ID | 1122 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
Magical | +0.678 |
adesh | +0.662 |
ower | +0.646 |
deen | +0.642 |
Piercing | +0.600 |
Neg logit contributions | |
opoly | -0.857 |
roc | -0.856 |
Concord | -0.803 |
Times | -0.798 |
upfront | -0.787 |
Token | said |
Token ID | 531 |
Feature activation | -4.86e-03 |
Pos logit contributions | |
roc | +1.745 |
opoly | +1.722 |
Concord | +1.577 |
Times | +1.574 |
upfront | +1.556 |
Neg logit contributions | |
Magical | -2.807 |
adesh | -2.756 |
deen | -2.672 |
ower | -2.639 |
Piercing | -2.583 |
Token | . |
Token ID | 13 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
roc | +0.053 |
opoly | +0.053 |
Times | +0.049 |
Concord | +0.049 |
upfront | +0.048 |
Neg logit contributions | |
Magical | -0.051 |
adesh | -0.049 |
deen | -0.048 |
ower | -0.048 |
Piercing | -0.045 |
Token | � |
Token ID | 564 |
Feature activation | -7.69e-02 |
Pos logit contributions | |
Magical | +2.802 |
adesh | +2.663 |
ower | +2.607 |
deen | +2.596 |
Piercing | +2.477 |
Neg logit contributions | |
roc | -3.161 |
opoly | -3.142 |
Times | -2.919 |
Concord | -2.900 |
Msg | -2.865 |
Token | Email |
Token ID | 15333 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
roc | +0.352 |
opoly | +0.348 |
Times | +0.319 |
Concord | +0.314 |
upfront | +0.307 |
Neg logit contributions | |
Magical | -0.606 |
adesh | -0.595 |
deen | -0.582 |
ower | -0.578 |
Piercing | -0.558 |
Token |
|
Token ID | 198 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
roc | +4.633 |
opoly | +4.497 |
Times | +4.385 |
hello | +4.285 |
Concord | +4.269 |
Neg logit contributions | |
adesh | -7.179 |
Magical | -7.073 |
deen | -7.041 |
ower | -6.820 |
shed | -6.629 |
Token |
|
Token ID | 198 |
Feature activation | +6.80e-01 |
Pos logit contributions | |
Magical | +2.980 |
ower | +2.825 |
adesh | +2.813 |
deen | +2.757 |
ing | +2.675 |
Neg logit contributions | |
opoly | -3.962 |
roc | -3.909 |
Concord | -3.684 |
tac | -3.583 |
ouched | -3.572 |
Token | Dis |
Token ID | 7279 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
Magical | +4.466 |
ower | +4.450 |
deen | +4.311 |
ing | +4.165 |
adesh | +4.134 |
Neg logit contributions | |
opoly | -9.139 |
Concord | -9.072 |
roc | -9.033 |
Sagan | -8.647 |
tac | -8.627 |
Token | ruption |
Token ID | 6417 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
Magical | +3.022 |
adesh | +2.980 |
ower | +2.964 |
deen | +2.909 |
Piercing | +2.780 |
Neg logit contributions | |
roc | -1.710 |
opoly | -1.685 |
Concord | -1.596 |
Times | -1.569 |
Sagan | -1.539 |
Token | is |
Token ID | 318 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ing | +6.822 |
Magical | +6.261 |
ower | +5.687 |
shed | +5.362 |
adesh | +5.323 |
Neg logit contributions | |
ailability | -16.801 |
hovah | -15.749 |
unbeliev | -15.504 |
opoly | -15.448 |
neighb | -15.214 |
Token | perhaps |
Token ID | 3737 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
Pavilion | +4.626 |
shed | +4.436 |
ing | +4.340 |
Magical | +4.121 |
illuminate | +3.996 |
Neg logit contributions | |
Homs | -24.659 |
ailability | -24.572 |
hello | -23.909 |
roc | -23.903 |
CCC | -23.835 |
Token | one |
Token ID | 530 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
Magical | +3.957 |
ower | +3.751 |
adesh | +3.693 |
deen | +3.672 |
shed | +3.653 |
Neg logit contributions | |
opoly | -3.393 |
roc | -3.360 |
Times | -3.137 |
Concord | -3.131 |
ustain | -3.055 |
Token | of |
Token ID | 286 |
Feature activation | +8.07e-01 |
Pos logit contributions | |
Magical | +3.624 |
ower | +3.280 |
adesh | +3.276 |
ing | +3.268 |
shed | +3.170 |
Neg logit contributions | |
opoly | -5.107 |
roc | -4.986 |
Concord | -4.717 |
Times | -4.705 |
ustain | -4.662 |
Token | the |
Token ID | 262 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
Magical | +6.036 |
ing | +5.697 |
ower | +5.195 |
shed | +5.079 |
Piercing | +5.049 |
Neg logit contributions | |
ailability | -10.781 |
opoly | -10.297 |
roc | -9.993 |
neighb | -9.871 |
hovah | -9.781 |
Token | most |
Token ID | 749 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
Magical | +5.456 |
ing | +5.088 |
shed | +4.957 |
ower | +4.886 |
adesh | +4.628 |
Neg logit contributions | |
opoly | -10.650 |
ailability | -10.648 |
roc | -10.461 |
Times | -10.286 |
TEXT | -10.185 |
Token | OL |
Token ID | 3535 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
roc | +2.999 |
opoly | +2.993 |
Times | +2.856 |
Concord | +2.718 |
Bee | +2.698 |
Neg logit contributions | |
adesh | -5.314 |
Magical | -5.168 |
deen | -5.156 |
ower | -4.997 |
shed | -4.916 |
Token | TO |
Token ID | 5390 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
Magical | +5.468 |
adesh | +4.910 |
ower | +4.882 |
Piercing | +4.879 |
Pavilion | +4.789 |
Neg logit contributions | |
roc | -6.592 |
opoly | -6.465 |
ouched | -6.086 |
etition | -6.065 |
hello | -6.033 |
Token | MAKE |
Token ID | 39134 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
Magical | +3.697 |
ower | +3.465 |
adesh | +3.451 |
deen | +3.391 |
Piercing | +3.348 |
Neg logit contributions | |
roc | -3.079 |
opoly | -3.063 |
Concord | -2.789 |
Times | -2.776 |
ouched | -2.760 |
Token | THE |
Token ID | 3336 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
Magical | +9.682 |
Piercing | +8.560 |
ing | +8.358 |
Pavilion | +8.331 |
Rated | +8.098 |
Neg logit contributions | |
roc | -8.927 |
opoly | -8.926 |
etition | -8.513 |
ouched | -8.193 |
imity | -8.029 |
Token | BE |
Token ID | 9348 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
Magical | +4.317 |
adesh | +3.969 |
ower | +3.924 |
deen | +3.876 |
Pavilion | +3.853 |
Neg logit contributions | |
roc | -3.715 |
opoly | -3.691 |
ouched | -3.356 |
ustain | -3.355 |
Times | -3.350 |
Token | ER |
Token ID | 1137 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
Magical | +1.844 |
adesh | +1.665 |
ower | +1.643 |
deen | +1.584 |
Piercing | +1.502 |
Neg logit contributions | |
roc | -4.567 |
opoly | -4.563 |
Concord | -4.320 |
Times | -4.310 |
ouched | -4.255 |
Token | T |
Token ID | 309 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
Magical | +15.602 |
ing | +13.953 |
Rated | +13.822 |
Piercing | +13.539 |
Magic | +13.403 |
Neg logit contributions | |
opoly | -12.858 |
roc | -12.830 |
achev | -12.445 |
etition | -12.186 |
auntlet | -11.969 |
Token | AS |
Token ID | 1921 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
Magical | +5.805 |
ower | +5.664 |
adesh | +5.366 |
ing | +5.353 |
deen | +5.196 |
Neg logit contributions | |
opoly | -6.766 |
roc | -6.560 |
Concord | -6.309 |
upfront | -6.181 |
Times | -6.099 |
Token | TE |
Token ID | 9328 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
Magical | +5.346 |
ower | +4.805 |
Piercing | +4.640 |
ing | +4.585 |
adesh | +4.585 |
Neg logit contributions | |
opoly | -5.615 |
roc | -5.576 |
upfront | -5.154 |
ouched | -5.123 |
Concord | -5.120 |
Token | AS |
Token ID | 7054 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
Magical | +7.347 |
Piercing | +6.660 |
ower | +6.629 |
ing | +6.550 |
Pavilion | +6.469 |
Neg logit contributions | |
opoly | -7.314 |
roc | -7.302 |
ouched | -6.721 |
Times | -6.712 |
Concord | -6.675 |
Token | N |
Token ID | 399 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
Magical | +7.996 |
Piercing | +7.273 |
adesh | +7.254 |
ower | +7.131 |
deen | +7.025 |
Neg logit contributions | |
roc | -5.707 |
opoly | -5.590 |
etition | -5.085 |
Concord | -5.070 |
Times | -5.065 |
Token | of |
Token ID | 286 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
roc | +2.501 |
opoly | +2.457 |
upfront | +2.320 |
Concord | +2.316 |
Times | +2.287 |
Neg logit contributions | |
adesh | -3.030 |
Magical | -3.014 |
deen | -2.943 |
ower | -2.876 |
Piercing | -2.784 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
roc | +3.087 |
opoly | +3.022 |
upfront | +2.880 |
Concord | +2.828 |
Times | +2.811 |
Neg logit contributions | |
Magical | -3.564 |
adesh | -3.550 |
deen | -3.464 |
ower | -3.408 |
Piercing | -3.293 |
Token | In |
Token ID | 554 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
roc | +0.172 |
opoly | +0.171 |
Concord | +0.159 |
Times | +0.159 |
upfront | +0.157 |
Neg logit contributions | |
Magical | -0.162 |
adesh | -0.158 |
deen | -0.153 |
ower | -0.153 |
Piercing | -0.145 |
Token | this |
Token ID | 428 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
Magical | +0.482 |
adesh | +0.469 |
ower | +0.457 |
deen | +0.456 |
Piercing | +0.438 |
Neg logit contributions | |
roc | -0.396 |
opoly | -0.395 |
Times | -0.363 |
Concord | -0.362 |
upfront | -0.354 |
Token | post |
Token ID | 1281 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
Magical | +4.464 |
adesh | +4.198 |
ower | +4.183 |
shed | +4.147 |
deen | +4.091 |
Neg logit contributions | |
opoly | -3.600 |
roc | -3.597 |
Times | -3.280 |
Concord | -3.265 |
ustain | -3.259 |
Token | I |
Token ID | 314 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Magical | +3.802 |
adesh | +3.697 |
ower | +3.633 |
shed | +3.526 |
deen | +3.495 |
Neg logit contributions | |
roc | -3.647 |
opoly | -3.640 |
Times | -3.392 |
Concord | -3.373 |
ustain | -3.305 |
Token | 'll |
Token ID | 1183 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ower | +6.220 |
shed | +5.630 |
Magical | +5.033 |
ing | +4.687 |
illuminate | +4.420 |
Neg logit contributions | |
ailability | -23.671 |
Ambro | -23.225 |
Bee | -23.113 |
Concord | -22.809 |
Times | -22.703 |
Token | round |
Token ID | 2835 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
shed | +8.583 |
Magical | +8.321 |
ower | +7.617 |
ing | +7.532 |
illuminate | +7.450 |
Neg logit contributions | |
ailability | -18.997 |
Times | -18.754 |
Bee | -18.638 |
opoly | -18.631 |
unbeliev | -18.153 |
Token | up |
Token ID | 510 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
Magical | +5.578 |
ower | +5.302 |
adesh | +5.249 |
ing | +5.184 |
shed | +5.157 |
Neg logit contributions | |
opoly | -5.976 |
roc | -5.873 |
Times | -5.567 |
Concord | -5.541 |
hello | -5.473 |
Token | some |
Token ID | 617 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
roc | +2.826 |
opoly | +2.588 |
upfront | +2.471 |
Times | +2.361 |
Concord | +2.342 |
Neg logit contributions | |
deen | -7.607 |
Magical | -7.535 |
adesh | -7.492 |
ower | -7.400 |
Piercing | -7.099 |
Token | got |
Token ID | 1392 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
roc | +5.253 |
upfront | +5.033 |
opoly | +4.897 |
hello | +4.797 |
Times | +4.682 |
Neg logit contributions | |
deen | -9.399 |
Magical | -9.071 |
ower | -9.067 |
adesh | -8.931 |
Piercing | -8.624 |
Token | to |
Token ID | 284 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
roc | +0.085 |
opoly | +0.084 |
Concord | +0.078 |
Times | +0.078 |
upfront | +0.077 |
Neg logit contributions | |
Magical | -0.083 |
adesh | -0.081 |
deen | -0.079 |
ower | -0.079 |
Piercing | -0.075 |
Token | 300 |
Token ID | 5867 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Magical | +7.713 |
shed | +7.657 |
deen | +7.560 |
discharge | +7.377 |
ower | +7.369 |
Neg logit contributions | |
opoly | -9.630 |
roc | -9.333 |
hello | -8.974 |
Times | -8.803 |
Msg | -8.799 |
Token | , |
Token ID | 11 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
Magical | +8.714 |
ower | +8.126 |
ing | +8.038 |
shed | +7.861 |
deen | +7.675 |
Neg logit contributions | |
opoly | -11.749 |
roc | -11.479 |
TEXT | -10.898 |
etition | -10.767 |
Sagan | -10.687 |
Token | 000 |
Token ID | 830 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
Magical | +3.207 |
ower | +3.043 |
adesh | +3.037 |
deen | +2.980 |
shed | +2.900 |
Neg logit contributions | |
roc | -3.702 |
opoly | -3.696 |
Times | -3.472 |
Concord | -3.410 |
upfront | -3.360 |
Token | barrels |
Token ID | 17907 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
Magical | +4.887 |
ing | +4.464 |
shed | +4.434 |
ower | +4.420 |
deen | +4.112 |
Neg logit contributions | |
opoly | -12.846 |
roc | -12.463 |
Times | -11.998 |
Msg | -11.995 |
TEXT | -11.928 |
Token | a |
Token ID | 257 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
Magical | +5.021 |
ower | +4.746 |
deen | +4.711 |
adesh | +4.704 |
shed | +4.648 |
Neg logit contributions | |
opoly | -6.434 |
roc | -6.207 |
Times | -6.007 |
hello | -5.797 |
Bee | -5.791 |
Token | day |
Token ID | 1110 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
shed | +4.530 |
Magical | +4.311 |
discharge | +3.879 |
ing | +3.817 |
pipeline | +2.835 |
Neg logit contributions | |
opoly | -24.214 |
ussian | -23.733 |
ouched | -23.361 |
auntlet | -23.236 |
imity | -23.094 |
Token | . |
Token ID | 13 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
Magical | +4.402 |
ower | +4.268 |
adesh | +4.151 |
deen | +4.095 |
shed | +4.024 |
Neg logit contributions | |
opoly | -5.671 |
roc | -5.553 |
Times | -5.304 |
Msg | -5.214 |
Concord | -5.187 |
Token | Virt |
Token ID | 11285 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
Magical | +3.285 |
adesh | +3.083 |
ower | +3.044 |
deen | +2.961 |
Piercing | +2.939 |
Neg logit contributions | |
roc | -3.672 |
opoly | -3.639 |
Times | -3.381 |
ouched | -3.327 |
Msg | -3.321 |
Token | ually |
Token ID | 935 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
roc | +4.720 |
opoly | +4.655 |
ouched | +4.267 |
Msg | +4.212 |
etition | +4.195 |
Neg logit contributions | |
adesh | -6.446 |
deen | -6.400 |
Magical | -6.354 |
ower | -5.967 |
Piercing | -5.967 |
Token | all |
Token ID | 477 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
Magical | +1.986 |
adesh | +1.901 |
ower | +1.850 |
deen | +1.847 |
shed | +1.779 |
Neg logit contributions | |
roc | -2.391 |
opoly | -2.368 |
Times | -2.247 |
Concord | -2.201 |
hello | -2.181 |
Token | above |
Token ID | 2029 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
Magical | +3.902 |
ower | +3.767 |
adesh | +3.760 |
deen | +3.689 |
shed | +3.568 |
Neg logit contributions | |
opoly | -3.930 |
roc | -3.850 |
Times | -3.596 |
Concord | -3.571 |
TEXT | -3.502 |
Token | , |
Token ID | 11 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
Magical | +4.343 |
ower | +4.166 |
adesh | +4.157 |
deen | +4.044 |
shed | +3.858 |
Neg logit contributions | |
opoly | -5.482 |
roc | -5.393 |
Times | -5.124 |
Concord | -5.089 |
TEXT | -4.983 |
Token | to |
Token ID | 284 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
Magical | +6.285 |
ower | +5.851 |
adesh | +5.811 |
shed | +5.688 |
deen | +5.600 |
Neg logit contributions | |
opoly | -6.861 |
roc | -6.678 |
Times | -6.347 |
imity | -6.185 |
ustain | -6.164 |
Token | make |
Token ID | 787 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
Magical | +4.743 |
ower | +4.466 |
deen | +4.368 |
shed | +4.335 |
adesh | +4.333 |
Neg logit contributions | |
opoly | -5.475 |
roc | -5.384 |
Times | -5.025 |
Concord | -4.993 |
hello | -4.916 |
Token | power |
Token ID | 1176 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Magical | +8.228 |
ower | +7.979 |
ing | +7.434 |
shed | +7.312 |
adesh | +7.265 |
Neg logit contributions | |
opoly | -11.398 |
roc | -10.825 |
Concord | -10.337 |
Times | -10.244 |
hello | -10.194 |
Token | management |
Token ID | 4542 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ing | +9.185 |
ower | +9.040 |
Magical | +8.856 |
shed | +8.654 |
deen | +8.541 |
Neg logit contributions | |
opoly | -10.111 |
roc | -10.095 |
Concord | -9.573 |
Bee | -9.462 |
Sagan | -9.241 |
Token | off |
Token ID | 572 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ower | +13.301 |
ing | +13.289 |
Magical | +13.167 |
shed | +13.143 |
illuminate | +12.514 |
Neg logit contributions | |
opoly | -15.015 |
roc | -14.289 |
Concord | -13.354 |
hello | -13.311 |
icipated | -13.046 |
Token | the |
Token ID | 262 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
ower | +10.518 |
Magical | +10.510 |
ing | +10.143 |
shed | +9.673 |
deen | +9.618 |
Neg logit contributions | |
opoly | -11.724 |
roc | -11.483 |
hello | -10.905 |
Times | -10.886 |
Concord | -10.859 |
Token | new |
Token ID | 649 |
Feature activation | +8.21e-01 |
Pos logit contributions | |
Magical | +3.217 |
adesh | +3.067 |
ower | +3.044 |
deen | +2.993 |
shed | +2.909 |
Neg logit contributions | |
opoly | -3.009 |
roc | -2.996 |
Times | -2.782 |
Concord | -2.744 |
ustain | -2.716 |
Token | default |
Token ID | 4277 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
Magical | +6.969 |
ower | +6.499 |
shed | +6.182 |
adesh | +6.179 |
deen | +6.105 |
Neg logit contributions | |
opoly | -9.709 |
roc | -9.569 |
Times | -9.031 |
ustain | -8.953 |
Concord | -8.879 |
Token | . |
Token ID | 13 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
Magical | +11.105 |
ower | +10.868 |
ing | +10.656 |
shed | +10.285 |
deen | +9.972 |
Neg logit contributions | |
opoly | -13.248 |
roc | -12.772 |
Concord | -11.987 |
ustain | -11.913 |
Times | -11.832 |
Token | relay |
Token ID | 24248 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
Magical | +6.235 |
adesh | +5.943 |
ower | +5.712 |
shed | +5.638 |
deen | +5.560 |
Neg logit contributions | |
roc | -5.808 |
opoly | -5.800 |
Times | -5.333 |
ouched | -5.234 |
TEXT | -5.220 |
Token | improvements |
Token ID | 8561 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
Magical | +2.947 |
adesh | +2.875 |
ower | +2.797 |
deen | +2.720 |
shed | +2.675 |
Neg logit contributions | |
opoly | -3.292 |
roc | -3.277 |
Concord | -3.047 |
Times | -3.034 |
ouched | -2.998 |
Token | make |
Token ID | 787 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
Magical | +3.633 |
adesh | +3.452 |
ower | +3.379 |
deen | +3.318 |
shed | +3.283 |
Neg logit contributions | |
opoly | -4.057 |
roc | -4.034 |
Times | -3.738 |
Concord | -3.718 |
TEXT | -3.656 |
Token | se |
Token ID | 384 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
Magical | +7.222 |
ower | +6.764 |
adesh | +6.599 |
ing | +6.407 |
shed | +6.370 |
Neg logit contributions | |
opoly | -7.773 |
roc | -7.771 |
Times | -7.115 |
hello | -7.090 |
ouched | -6.978 |
Token | g |
Token ID | 70 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
Magical | +3.558 |
adesh | +3.550 |
ower | +3.493 |
deen | +3.398 |
ing | +3.266 |
Neg logit contributions | |
roc | -3.086 |
opoly | -3.024 |
Concord | -2.879 |
Times | -2.875 |
TEXT | -2.782 |
Token | wit |
Token ID | 39289 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Magical | +1.534 |
adesh | +1.522 |
ower | +1.478 |
deen | +1.459 |
shed | +1.383 |
Neg logit contributions | |
roc | -1.555 |
opoly | -1.544 |
Concord | -1.461 |
Times | -1.444 |
upfront | -1.419 |
Token | a |
Token ID | 257 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
Magical | +13.951 |
ower | +13.294 |
ing | +13.125 |
adesh | +12.725 |
shed | +12.538 |
Neg logit contributions | |
roc | -13.896 |
opoly | -13.499 |
Concord | -13.286 |
viron | -13.090 |
Times | -13.052 |
Token | reasonable |
Token ID | 6397 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
Magical | +6.060 |
adesh | +5.659 |
ower | +5.536 |
deen | +5.420 |
shed | +5.407 |
Neg logit contributions | |
roc | -5.620 |
opoly | -5.581 |
Times | -5.170 |
Concord | -5.122 |
hello | -5.110 |
Token | risk |
Token ID | 2526 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
Magical | +5.555 |
adesh | +5.250 |
ower | +5.195 |
shed | +4.983 |
deen | +4.973 |
Neg logit contributions | |
roc | -5.614 |
opoly | -5.553 |
Times | -5.079 |
ouched | -5.050 |
hello | -5.034 |
Token | . |
Token ID | 13 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
Magical | +1.927 |
adesh | +1.866 |
ower | +1.834 |
deen | +1.802 |
shed | +1.737 |
Neg logit contributions | |
roc | -1.941 |
opoly | -1.928 |
Concord | -1.794 |
Times | -1.793 |
hello | -1.742 |
Token | B |
Token ID | 347 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
Magical | +4.980 |
adesh | +4.625 |
ower | +4.465 |
Piercing | +4.416 |
deen | +4.370 |
Neg logit contributions | |
roc | -5.259 |
opoly | -5.230 |
ouched | -4.815 |
Times | -4.754 |
hello | -4.669 |
Token | � |
Token ID | 251 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
Magical | +0.241 |
adesh | +0.233 |
deen | +0.225 |
ower | +0.225 |
Piercing | +0.211 |
Neg logit contributions | |
roc | -0.352 |
opoly | -0.351 |
Times | -0.329 |
Concord | -0.329 |
upfront | -0.324 |
Token |
|
Token ID | 198 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
Magical | +1.568 |
adesh | +1.506 |
ower | +1.475 |
deen | +1.471 |
Pavilion | +1.399 |
Neg logit contributions | |
roc | -1.662 |
opoly | -1.657 |
Times | -1.533 |
Concord | -1.529 |
upfront | -1.504 |
Token |
|
Token ID | 198 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
Magical | +0.411 |
adesh | +0.399 |
ower | +0.388 |
deen | +0.388 |
Piercing | +0.369 |
Neg logit contributions | |
roc | -0.425 |
opoly | -0.424 |
Concord | -0.394 |
Times | -0.392 |
upfront | -0.386 |
Token | Aside |
Token ID | 32602 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
Magical | +2.787 |
adesh | +2.731 |
ower | +2.698 |
deen | +2.675 |
Pavilion | +2.522 |
Neg logit contributions | |
opoly | -2.298 |
roc | -2.290 |
Concord | -2.159 |
upfront | -2.071 |
Sagan | -2.040 |
Token | from |
Token ID | 422 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
Magical | +4.186 |
adesh | +4.003 |
ower | +3.945 |
deen | +3.790 |
ing | +3.738 |
Neg logit contributions | |
roc | -5.683 |
opoly | -5.665 |
Times | -5.388 |
Concord | -5.311 |
TEXT | -5.309 |
Token | H |
Token ID | 367 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Magical | +0.870 |
adesh | +0.841 |
ower | +0.821 |
deen | +0.817 |
Piercing | +0.782 |
Neg logit contributions | |
roc | -0.788 |
opoly | -0.787 |
Times | -0.725 |
Concord | -0.722 |
upfront | -0.707 |
Token | ower |
Token ID | 789 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
ower | +13.504 |
owers | +9.450 |
adesh | +9.216 |
ing | +9.113 |
room | +8.659 |
Neg logit contributions | |
Sagan | -18.356 |
Concord | -18.168 |
Times | -17.803 |
ゴ | -17.793 |
Homs | -17.735 |
Token | ton |
Token ID | 1122 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
Magical | +0.466 |
adesh | +0.454 |
ower | +0.441 |
deen | +0.440 |
Piercing | +0.413 |
Neg logit contributions | |
roc | -0.578 |
opoly | -0.577 |
Concord | -0.540 |
Times | -0.539 |
upfront | -0.530 |
Token | , |
Token ID | 11 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
Magical | +0.762 |
adesh | +0.739 |
ower | +0.721 |
deen | +0.718 |
Piercing | +0.683 |
Neg logit contributions | |
roc | -0.807 |
opoly | -0.804 |
Times | -0.747 |
Concord | -0.746 |
upfront | -0.731 |
Token | partners |
Token ID | 4887 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
roc | +1.844 |
opoly | +1.838 |
Concord | +1.715 |
Times | +1.692 |
upfront | +1.681 |
Neg logit contributions | |
Magical | -2.145 |
adesh | -2.139 |
deen | -2.070 |
ower | -2.046 |
Piercing | -1.969 |
Token | include |
Token ID | 2291 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
roc | +0.148 |
opoly | +0.147 |
Concord | +0.137 |
Times | +0.137 |
upfront | +0.134 |
Neg logit contributions | |
Magical | -0.155 |
adesh | -0.151 |
deen | -0.147 |
ower | -0.146 |
Piercing | -0.139 |
Token | First |
Token ID | 3274 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Magical | +8.074 |
adesh | +7.263 |
Pavilion | +7.099 |
ower | +7.077 |
deen | +7.070 |
Neg logit contributions | |
opoly | -7.153 |
roc | -6.918 |
Msg | -6.493 |
TEXT | -6.405 |
imity | -6.393 |
Token | , |
Token ID | 11 |
Feature activation | +9.15e-01 |
Pos logit contributions | |
Magical | +10.759 |
Pavilion | +9.468 |
ower | +9.389 |
Piercing | +9.341 |
adesh | +9.071 |
Neg logit contributions | |
opoly | -11.365 |
roc | -11.014 |
Msg | -10.496 |
TEXT | -10.455 |
upfront | -10.390 |
Token | Second |
Token ID | 5498 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
Magical | +9.435 |
adesh | +8.645 |
ower | +8.491 |
Piercing | +8.445 |
Pavilion | +8.210 |
Neg logit contributions | |
opoly | -8.951 |
roc | -8.948 |
Times | -8.285 |
hello | -8.162 |
TEXT | -8.093 |
Token | , |
Token ID | 11 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Magical | +9.398 |
ower | +8.533 |
adesh | +8.424 |
ing | +8.339 |
Pavilion | +8.225 |
Neg logit contributions | |
opoly | -13.133 |
roc | -13.044 |
upfront | -12.324 |
hello | -12.247 |
TEXT | -12.243 |
Token | and |
Token ID | 290 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
Magical | +10.669 |
ower | +9.258 |
adesh | +9.198 |
Pavilion | +9.084 |
Piercing | +9.045 |
Neg logit contributions | |
opoly | -13.773 |
roc | -13.656 |
hello | -13.048 |
Msg | -13.018 |
upfront | -13.009 |
Token | Fourth |
Token ID | 15692 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Magical | +12.345 |
ower | +10.499 |
Piercing | +10.457 |
Pavilion | +10.398 |
Magic | +9.892 |
Neg logit contributions | |
opoly | -15.346 |
roc | -15.113 |
hello | -14.896 |
gets | -14.599 |
Msg | -14.333 |
Token | Div |
Token ID | 4777 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
Magical | +15.260 |
Piercing | +12.899 |
ower | +12.638 |
Pavilion | +12.492 |
ing | +12.435 |
Neg logit contributions | |
opoly | -15.041 |
roc | -13.971 |
hello | -13.411 |
TEXT | -13.386 |
imity | -13.369 |
Token | isions |
Token ID | 3279 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
Magical | +2.293 |
adesh | +2.284 |
ower | +2.190 |
deen | +2.135 |
ing | +2.013 |
Neg logit contributions | |
roc | -2.700 |
opoly | -2.664 |
Concord | -2.526 |
upfront | -2.511 |
Times | -2.489 |
Token | all |
Token ID | 477 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
Magical | +4.139 |
adesh | +3.913 |
ower | +3.859 |
deen | +3.763 |
Piercing | +3.674 |
Neg logit contributions | |
opoly | -4.749 |
roc | -4.695 |
Times | -4.420 |
hello | -4.336 |
Msg | -4.332 |
Token | at |
Token ID | 379 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
roc | +0.975 |
opoly | +0.967 |
Concord | +0.901 |
Times | +0.893 |
upfront | +0.888 |
Neg logit contributions | |
Magical | -1.014 |
adesh | -0.995 |
deen | -0.968 |
ower | -0.957 |
Piercing | -0.916 |
Token | once |
Token ID | 1752 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Magical | +2.946 |
adesh | +2.834 |
ower | +2.791 |
deen | +2.774 |
Piercing | +2.715 |
Neg logit contributions | |
opoly | -1.550 |
roc | -1.502 |
Times | -1.365 |
Concord | -1.334 |
TEXT | -1.325 |
Token |
|
Token ID | 198 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
Magical | +3.032 |
adesh | +2.891 |
ower | +2.802 |
deen | +2.784 |
Piercing | +2.678 |
Neg logit contributions | |
roc | -3.424 |
opoly | -3.394 |
Times | -3.182 |
Concord | -3.137 |
ouched | -3.114 |
Token |
|
Token ID | 198 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
Magical | +0.465 |
adesh | +0.452 |
ower | +0.439 |
deen | +0.438 |
Piercing | +0.416 |
Neg logit contributions | |
roc | -0.483 |
opoly | -0.482 |
Concord | -0.447 |
Times | -0.446 |
upfront | -0.438 |
Token | and |
Token ID | 392 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
Magical | +7.173 |
adesh | +7.151 |
ower | +6.924 |
deen | +6.909 |
ing | +6.706 |
Neg logit contributions | |
opoly | -6.275 |
roc | -6.256 |
Concord | -6.209 |
Sagan | -5.960 |
upfront | -5.768 |
Token | how |
Token ID | 703 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Magical | +3.699 |
adesh | +3.534 |
ower | +3.488 |
deen | +3.412 |
shed | +3.354 |
Neg logit contributions | |
opoly | -3.663 |
roc | -3.653 |
Times | -3.447 |
Concord | -3.367 |
ouched | -3.357 |
Token | does |
Token ID | 857 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
Magical | +1.873 |
adesh | +1.847 |
deen | +1.765 |
ower | +1.761 |
Piercing | +1.708 |
Neg logit contributions | |
roc | -1.778 |
opoly | -1.767 |
Times | -1.633 |
Concord | -1.628 |
ouched | -1.613 |
Token | it |
Token ID | 340 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Magical | +7.267 |
ing | +6.587 |
shed | +6.487 |
adesh | +6.273 |
ower | +6.145 |
Neg logit contributions | |
opoly | -12.456 |
roc | -11.750 |
ailability | -11.623 |
Times | -11.621 |
Bee | -11.505 |
Token | look |
Token ID | 804 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
ing | +3.394 |
shed | +2.988 |
Magical | +2.929 |
illuminate | +2.383 |
ower | +1.991 |
Neg logit contributions | |
ailability | -27.247 |
opoly | -25.216 |
roc | -25.051 |
hovah | -24.998 |
Times | -24.993 |
Token | from |
Token ID | 422 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
Magical | +5.428 |
adesh | +5.303 |
ing | +4.978 |
deen | +4.975 |
shed | +4.953 |
Neg logit contributions | |
roc | -10.112 |
opoly | -10.053 |
Times | -9.505 |
ouched | -9.473 |
ustain | -9.453 |
Token | the |
Token ID | 262 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
Magical | +5.721 |
ing | +5.184 |
ower | +5.109 |
adesh | +5.088 |
shed | +5.010 |
Neg logit contributions | |
opoly | -8.917 |
roc | -8.908 |
ailability | -8.619 |
Times | -8.359 |
ustain | -8.284 |
Token | perspective |
Token ID | 6650 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
Magical | +4.214 |
ower | +3.943 |
adesh | +3.833 |
shed | +3.801 |
Pavilion | +3.722 |
Neg logit contributions | |
roc | -7.439 |
opoly | -7.296 |
Msg | -6.975 |
Times | -6.941 |
Bee | -6.938 |
Token | of |
Token ID | 286 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
Magical | +4.750 |
ing | +4.305 |
ower | +4.261 |
adesh | +4.228 |
shed | +4.084 |
Neg logit contributions | |
opoly | -7.465 |
roc | -7.382 |
ailability | -7.109 |
ustain | -6.868 |
Times | -6.853 |
Token | what |
Token ID | 644 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
Magical | +6.265 |
ower | +5.909 |
adesh | +5.783 |
shed | +5.769 |
Piercing | +5.685 |
Neg logit contributions | |
opoly | -5.544 |
roc | -5.476 |
Times | -5.283 |
Concord | -5.055 |
TEXT | -4.986 |
Token | scope |
Token ID | 8354 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
Magical | +4.472 |
ower | +4.068 |
adesh | +4.049 |
shed | +4.022 |
Pavilion | +3.816 |
Neg logit contributions | |
roc | -7.506 |
opoly | -7.324 |
Times | -7.156 |
Bee | -6.911 |
ustain | -6.835 |
Token | do |
Token ID | 466 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
Magical | +3.875 |
adesh | +3.863 |
shed | +3.662 |
ower | +3.658 |
deen | +3.553 |
Neg logit contributions | |
opoly | -5.716 |
roc | -5.663 |
Times | -5.402 |
ouched | -5.326 |
TEXT | -5.254 |
Token | those |
Token ID | 883 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
Magical | +5.605 |
adesh | +5.333 |
ing | +5.039 |
ower | +5.005 |
shed | +4.941 |
Neg logit contributions | |
opoly | -6.895 |
roc | -6.892 |
ouched | -6.359 |
Bee | -6.315 |
Times | -6.312 |
Token | parameters |
Token ID | 10007 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
roc | +2.165 |
opoly | +2.122 |
upfront | +2.032 |
Concord | +1.980 |
Times | +1.928 |
Neg logit contributions | |
Magical | -3.986 |
adesh | -3.943 |
ower | -3.905 |
deen | -3.881 |
Piercing | -3.752 |
Token | have |
Token ID | 423 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
roc | +4.107 |
opoly | +4.009 |
upfront | +3.923 |
Concord | +3.719 |
ouched | +3.709 |
Neg logit contributions | |
Magical | -6.738 |
adesh | -6.549 |
deen | -6.441 |
Piercing | -6.418 |
ower | -6.312 |
Token | ? |
Token ID | 30 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
?) | +4.576 |
Magical | +3.448 |
ing | +3.172 |
?). | +3.001 |
(@ | +2.678 |
Neg logit contributions | |
ailability | -26.596 |
��� | -25.784 |
unbeliev | -25.224 |
neighb | -24.845 |
tremend | -24.840 |
Token | can |
Token ID | 460 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
Magical | +0.777 |
adesh | +0.751 |
ower | +0.728 |
deen | +0.727 |
Piercing | +0.697 |
Neg logit contributions | |
roc | -0.803 |
opoly | -0.799 |
Times | -0.742 |
Concord | -0.738 |
upfront | -0.724 |
Token | a |
Token ID | 257 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
Magical | +2.498 |
adesh | +2.395 |
ower | +2.346 |
deen | +2.314 |
Piercing | +2.227 |
Neg logit contributions | |
opoly | -2.380 |
roc | -2.370 |
Times | -2.200 |
Concord | -2.178 |
Bee | -2.130 |
Token | block |
Token ID | 2512 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
Magical | +2.122 |
adesh | +2.013 |
ower | +1.961 |
deen | +1.940 |
Piercing | +1.849 |
Neg logit contributions | |
opoly | -2.696 |
roc | -2.686 |
Times | -2.516 |
Concord | -2.489 |
Bee | -2.450 |
Token | be |
Token ID | 307 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
roc | +0.134 |
opoly | +0.134 |
Concord | +0.124 |
Times | +0.124 |
upfront | +0.122 |
Neg logit contributions | |
Magical | -0.133 |
adesh | -0.130 |
deen | -0.126 |
ower | -0.126 |
Piercing | -0.120 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
roc | +3.697 |
opoly | +3.544 |
Times | +3.404 |
upfront | +3.374 |
Concord | +3.304 |
Neg logit contributions | |
adesh | -5.153 |
Magical | -5.092 |
deen | -5.041 |
ower | -4.862 |
Piercing | -4.739 |
Token | But |
Token ID | 1537 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
roc | +1.068 |
opoly | +1.052 |
Times | +0.991 |
Bee | +0.947 |
Concord | +0.943 |
Neg logit contributions | |
Magical | -1.525 |
adesh | -1.496 |
deen | -1.455 |
ower | -1.442 |
Piercing | -1.396 |
Token | what |
Token ID | 644 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
roc | +0.590 |
opoly | +0.585 |
Concord | +0.543 |
Times | +0.542 |
upfront | +0.535 |
Neg logit contributions | |
Magical | -0.640 |
adesh | -0.628 |
deen | -0.610 |
ower | -0.606 |
Piercing | -0.578 |
Token | made |
Token ID | 925 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Magical | +0.639 |
adesh | +0.620 |
deen | +0.604 |
ower | +0.603 |
Piercing | +0.576 |
Neg logit contributions | |
roc | -0.613 |
opoly | -0.610 |
Times | -0.565 |
Concord | -0.564 |
upfront | -0.554 |
Token | the |
Token ID | 262 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
Magical | +1.447 |
adesh | +1.394 |
ower | +1.367 |
deen | +1.360 |
Piercing | +1.307 |
Neg logit contributions | |
opoly | -1.380 |
roc | -1.375 |
Times | -1.272 |
Concord | -1.268 |
upfront | -1.245 |
Token | reactions |
Token ID | 12737 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Magical | +3.238 |
adesh | +3.083 |
ower | +3.046 |
deen | +3.011 |
shed | +2.970 |
Neg logit contributions | |
opoly | -3.045 |
roc | -3.007 |
hello | -2.749 |
Times | -2.747 |
ustain | -2.743 |
Token | of |
Token ID | 286 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
Magical | +11.959 |
shed | +11.811 |
adesh | +11.565 |
ing | +11.247 |
ower | +10.792 |
Neg logit contributions | |
roc | -15.081 |
opoly | -14.933 |
hello | -14.211 |
ussian | -14.135 |
icipated | -14.052 |
Token | Belle |
Token ID | 31974 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
Magical | +3.894 |
ower | +3.661 |
adesh | +3.632 |
deen | +3.609 |
shed | +3.529 |
Neg logit contributions | |
opoly | -3.702 |
roc | -3.629 |
Times | -3.397 |
ustain | -3.338 |
Concord | -3.333 |
Token | w |
Token ID | 86 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
opoly | +3.996 |
roc | +3.972 |
hello | +3.577 |
Concord | +3.574 |
upfront | +3.553 |
Neg logit contributions | |
Magical | -4.905 |
adesh | -4.676 |
deen | -4.585 |
Piercing | -4.496 |
ower | -4.425 |
Token | particularly |
Token ID | 3573 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
Magical | +8.913 |
adesh | +8.787 |
ower | +8.630 |
shed | +8.580 |
deen | +8.389 |
Neg logit contributions | |
roc | -11.586 |
opoly | -11.513 |
Times | -11.310 |
ustain | -10.814 |
hello | -10.811 |
Token | cr |
Token ID | 1067 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
Magical | +8.480 |
shed | +8.080 |
adesh | +8.049 |
ing | +7.774 |
Piercing | +7.486 |
Neg logit contributions | |
opoly | -9.781 |
roc | -9.547 |
ustain | -8.895 |
Times | -8.852 |
ouched | -8.715 |
Token | , |
Token ID | 11 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
roc | +0.198 |
opoly | +0.197 |
Concord | +0.182 |
Times | +0.182 |
upfront | +0.179 |
Neg logit contributions | |
Magical | -0.208 |
adesh | -0.203 |
deen | -0.197 |
ower | -0.197 |
Piercing | -0.188 |
Token | was |
Token ID | 373 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
Magical | +4.247 |
adesh | +4.038 |
ower | +3.957 |
deen | +3.947 |
shed | +3.911 |
Neg logit contributions | |
opoly | -4.269 |
roc | -4.208 |
Times | -3.972 |
ustain | -3.872 |
hello | -3.854 |
Token | killed |
Token ID | 2923 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
Magical | +3.969 |
adesh | +3.774 |
deen | +3.689 |
ower | +3.681 |
shed | +3.673 |
Neg logit contributions | |
opoly | -4.925 |
roc | -4.823 |
Times | -4.566 |
hello | -4.497 |
ustain | -4.492 |
Token | trying |
Token ID | 2111 |
Feature activation | -6.04e-02 |
Pos logit contributions | |
roc | +1.983 |
opoly | +1.938 |
Concord | +1.818 |
upfront | +1.801 |
Times | +1.789 |
Neg logit contributions | |
Magical | -2.047 |
adesh | -2.016 |
deen | -1.934 |
ower | -1.912 |
Piercing | -1.842 |
Token | to |
Token ID | 284 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
roc | +0.636 |
opoly | +0.631 |
Concord | +0.585 |
Times | +0.583 |
upfront | +0.575 |
Neg logit contributions | |
Magical | -0.652 |
adesh | -0.636 |
deen | -0.619 |
ower | -0.616 |
Piercing | -0.588 |
Token | cross |
Token ID | 3272 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Magical | +2.343 |
adesh | +2.278 |
deen | +2.182 |
ower | +2.179 |
shed | +2.140 |
Neg logit contributions | |
opoly | -3.023 |
roc | -3.019 |
Times | -2.854 |
Concord | -2.791 |
hello | -2.784 |
Token | into |
Token ID | 656 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
Magical | +12.200 |
ing | +11.752 |
adesh | +11.685 |
ower | +11.297 |
shed | +10.832 |
Neg logit contributions | |
TEXT | -14.436 |
opoly | -14.174 |
Msg | -14.060 |
hello | -14.021 |
ouched | -13.931 |
Token | Turkey |
Token ID | 7137 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
Magical | +6.883 |
adesh | +6.229 |
deen | +6.181 |
ower | +6.126 |
shed | +5.967 |
Neg logit contributions | |
opoly | -9.045 |
roc | -8.698 |
TEXT | -8.567 |
Msg | -8.545 |
hello | -8.494 |
Token | . |
Token ID | 13 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
Magical | +5.058 |
adesh | +4.824 |
ower | +4.707 |
deen | +4.674 |
shed | +4.579 |
Neg logit contributions | |
opoly | -6.894 |
roc | -6.696 |
Times | -6.450 |
ustain | -6.412 |
TEXT | -6.350 |
Token | On |
Token ID | 1550 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
Magical | +3.066 |
adesh | +2.843 |
ower | +2.787 |
deen | +2.763 |
Piercing | +2.749 |
Neg logit contributions | |
opoly | -3.221 |
roc | -3.183 |
Times | -2.917 |
ouched | -2.915 |
upfront | -2.911 |
Token | March |
Token ID | 2805 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
roc | +0.958 |
opoly | +0.948 |
Concord | +0.887 |
Times | +0.881 |
upfront | +0.870 |
Neg logit contributions | |
Magical | -0.938 |
adesh | -0.925 |
deen | -0.894 |
ower | -0.887 |
Piercing | -0.846 |
Token | as |
Token ID | 355 |
Feature activation | +9.83e-01 |
Pos logit contributions | |
Magical | +6.436 |
deen | +6.307 |
ower | +6.151 |
ing | +6.123 |
adesh | +6.049 |
Neg logit contributions | |
opoly | -8.829 |
roc | -8.682 |
Msg | -8.250 |
TEXT | -8.188 |
Times | -8.133 |
Token | a |
Token ID | 257 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
Magical | +8.664 |
shed | +8.200 |
deen | +8.181 |
adesh | +8.140 |
ower | +7.933 |
Neg logit contributions | |
opoly | -10.701 |
roc | -10.335 |
TEXT | -9.943 |
Times | -9.470 |
imity | -9.433 |
Token | delegate |
Token ID | 23191 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
Magical | +6.535 |
deen | +6.029 |
adesh | +5.992 |
ower | +5.966 |
shed | +5.930 |
Neg logit contributions | |
opoly | -5.997 |
roc | -5.859 |
TEXT | -5.438 |
Times | -5.331 |
hello | -5.318 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Magical | +4.404 |
adesh | +4.286 |
deen | +4.212 |
ower | +4.202 |
shed | +4.001 |
Neg logit contributions | |
opoly | -5.444 |
roc | -5.396 |
Times | -5.112 |
TEXT | -4.995 |
Concord | -4.949 |
Token | Where |
Token ID | 6350 |
Feature activation | +6.23e-01 |
Pos logit contributions | |
Magical | +9.239 |
Pavilion | +8.082 |
deen | +7.923 |
Rated | +7.826 |
adesh | +7.803 |
Neg logit contributions | |
opoly | -11.186 |
roc | -11.040 |
ouched | -10.312 |
etition | -10.221 |
ussian | -10.131 |
Token | is |
Token ID | 318 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
Magical | +6.408 |
adesh | +6.266 |
deen | +6.123 |
ower | +6.023 |
shed | +5.837 |
Neg logit contributions | |
roc | -6.385 |
opoly | -6.261 |
Times | -5.768 |
TEXT | -5.764 |
Concord | -5.719 |
Token | she |
Token ID | 673 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
Magical | +13.205 |
Pavilion | +12.201 |
deen | +12.191 |
shed | +12.034 |
ing | +11.911 |
Neg logit contributions | |
opoly | -14.109 |
roc | -14.001 |
TEXT | -13.247 |
ouched | -13.115 |
Msg | -12.829 |
Token | ? |
Token ID | 30 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Magical | +8.688 |
deen | +8.499 |
ing | +8.317 |
shed | +8.244 |
ower | +8.243 |
Neg logit contributions | |
roc | -10.533 |
opoly | -10.408 |
Times | -9.800 |
TEXT | -9.647 |
ustain | -9.381 |
Token | Cathy |
Token ID | 40636 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
Magical | +9.225 |
Pavilion | +8.236 |
deen | +8.172 |
adesh | +7.991 |
ing | +7.893 |
Neg logit contributions | |
opoly | -11.330 |
roc | -11.246 |
ouched | -10.285 |
ussian | -10.210 |
TEXT | -10.192 |
Token | � |
Token ID | 447 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
Magical | +3.410 |
adesh | +3.278 |
deen | +3.255 |
ower | +3.212 |
shed | +3.122 |
Neg logit contributions | |
opoly | -3.721 |
roc | -3.696 |
Times | -3.439 |
ouched | -3.365 |
Concord | -3.360 |
Token | � |
Token ID | 247 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
roc | +2.150 |
opoly | +2.103 |
upfront | +1.937 |
Times | +1.927 |
Concord | +1.921 |
Neg logit contributions | |
Magical | -3.416 |
adesh | -3.365 |
deen | -3.256 |
ower | -3.247 |
Piercing | -3.183 |
Token | special |
Token ID | 2041 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
Magical | +2.208 |
adesh | +2.112 |
deen | +2.056 |
ower | +2.048 |
shed | +1.985 |
Neg logit contributions | |
opoly | -2.292 |
roc | -2.274 |
Times | -2.119 |
Concord | -2.081 |
hello | -2.076 |
Token | mantra |
Token ID | 32628 |
Feature activation | -7.23e-01 |
Pos logit contributions | |
Magical | +1.182 |
adesh | +1.140 |
deen | +1.106 |
ower | +1.103 |
Piercing | +1.060 |
Neg logit contributions | |
opoly | -1.245 |
roc | -1.241 |
Times | -1.153 |
Concord | -1.143 |
upfront | -1.128 |
Token | for |
Token ID | 329 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
roc | +6.829 |
Concord | +6.294 |
upfront | +6.281 |
opoly | +6.240 |
Times | +6.057 |
Neg logit contributions | |
adesh | -7.627 |
deen | -7.580 |
Magical | -7.515 |
ower | -7.249 |
Piercing | -6.858 |
Token | Kar |
Token ID | 9375 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
Magical | +1.365 |
adesh | +1.316 |
deen | +1.282 |
ower | +1.280 |
Piercing | +1.237 |
Neg logit contributions | |
opoly | -1.234 |
roc | -1.223 |
Times | -1.125 |
Concord | -1.115 |
upfront | -1.100 |
Token | th |
Token ID | 400 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
adesh | +3.042 |
Magical | +3.007 |
deen | +2.880 |
ower | +2.871 |
shed | +2.691 |
Neg logit contributions | |
opoly | -3.375 |
roc | -3.342 |
Times | -3.172 |
Concord | -3.152 |
Bee | -3.117 |
Token | ave |
Token ID | 1015 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
adesh | +6.845 |
ower | +6.763 |
Magical | +6.629 |
deen | +6.464 |
mage | +6.030 |
Neg logit contributions | |
Times | -14.091 |
opoly | -13.912 |
roc | -13.534 |
TEXT | -13.502 |
Albany | -13.374 |
Token | ery |
Token ID | 1924 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
ower | +9.876 |
Magical | +9.475 |
adesh | +9.212 |
ing | +8.890 |
mage | +8.857 |
Neg logit contributions | |
opoly | -16.836 |
Concord | -16.825 |
unbeliev | -16.614 |
constitu | -16.519 |
Times | -16.461 |
Token | ar |
Token ID | 283 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
adesh | +5.376 |
Magical | +5.363 |
ower | +5.270 |
deen | +5.194 |
ing | +5.015 |
Neg logit contributions | |
opoly | -4.640 |
roc | -4.503 |
Concord | -4.393 |
Times | -4.353 |
Msg | -4.272 |
Token | j |
Token ID | 73 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
roc | +3.199 |
opoly | +3.040 |
hello | +2.820 |
Times | +2.816 |
upfront | +2.787 |
Neg logit contributions | |
Magical | -4.516 |
adesh | -4.337 |
ower | -4.292 |
deen | -4.292 |
Piercing | -4.143 |
Token | una |
Token ID | 9613 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
adesh | +5.535 |
deen | +5.042 |
ower | +4.928 |
Magical | +4.637 |
ing | +4.397 |
Neg logit contributions | |
opoly | -15.038 |
roc | -14.971 |
Concord | -14.882 |
Homs | -14.595 |
Ambro | -14.590 |
Token | and |
Token ID | 290 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
roc | +4.071 |
opoly | +3.774 |
Concord | +3.772 |
upfront | +3.711 |
Times | +3.676 |
Neg logit contributions | |
adesh | -4.789 |
Magical | -4.788 |
deen | -4.721 |
ower | -4.582 |
Piercing | -4.365 |
Token | : |
Token ID | 25 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
Magical | +0.616 |
adesh | +0.600 |
ower | +0.584 |
deen | +0.583 |
Piercing | +0.557 |
Neg logit contributions | |
roc | -0.536 |
opoly | -0.534 |
Concord | -0.492 |
Times | -0.491 |
upfront | -0.483 |
Token | What |
Token ID | 1867 |
Feature activation | +6.52e-02 |
Pos logit contributions | |
Magical | +2.696 |
adesh | +2.476 |
ower | +2.409 |
deen | +2.401 |
Piercing | +2.397 |
Neg logit contributions | |
roc | -2.906 |
opoly | -2.888 |
ouched | -2.656 |
Times | -2.636 |
etition | -2.623 |
Token | made |
Token ID | 925 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
Magical | +0.700 |
adesh | +0.681 |
deen | +0.660 |
ower | +0.658 |
Piercing | +0.629 |
Neg logit contributions | |
roc | -0.691 |
opoly | -0.685 |
Times | -0.636 |
Concord | -0.635 |
upfront | -0.623 |
Token | Mid |
Token ID | 7215 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
Magical | +5.293 |
adesh | +4.879 |
deen | +4.824 |
ower | +4.756 |
Piercing | +4.725 |
Neg logit contributions | |
roc | -5.252 |
opoly | -5.236 |
Times | -4.909 |
hello | -4.898 |
ouched | -4.875 |
Token | n |
Token ID | 77 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Magical | +3.606 |
ower | +3.574 |
deen | +3.471 |
adesh | +3.461 |
shed | +3.191 |
Neg logit contributions | |
roc | -5.595 |
opoly | -5.521 |
Times | -5.386 |
Concord | -5.352 |
Sagan | -5.258 |
Token | ighter |
Token ID | 4799 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ower | +4.812 |
ighter | +4.334 |
adesh | +3.961 |
Magical | +3.809 |
deen | +3.479 |
Neg logit contributions | |
roc | -16.567 |
Times | -16.279 |
TEXT | -15.994 |
opoly | -15.909 |
Sagan | -15.804 |
Token | initially |
Token ID | 7317 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Magical | +13.603 |
shed | +12.385 |
ing | +12.272 |
deen | +11.943 |
ower | +11.822 |
Neg logit contributions | |
Times | -13.805 |
roc | -13.699 |
opoly | -13.697 |
ustain | -13.151 |
icipated | -13.079 |
Token | come |
Token ID | 1282 |
Feature activation | -7.84e-02 |
Pos logit contributions | |
Magical | +12.072 |
shed | +11.213 |
ing | +10.603 |
deen | +10.543 |
adesh | +10.533 |
Neg logit contributions | |
roc | -12.632 |
opoly | -12.450 |
Times | -12.372 |
TEXT | -11.697 |
ustain | -11.639 |
Token | to |
Token ID | 284 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
roc | +0.821 |
opoly | +0.821 |
Concord | +0.759 |
Times | +0.755 |
upfront | +0.751 |
Neg logit contributions | |
Magical | -0.848 |
adesh | -0.824 |
deen | -0.802 |
ower | -0.798 |
Piercing | -0.763 |
Token | mind |
Token ID | 2000 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
Magical | +2.526 |
adesh | +2.422 |
deen | +2.393 |
ower | +2.364 |
Piercing | +2.274 |
Neg logit contributions | |
roc | -2.232 |
opoly | -2.208 |
Times | -2.027 |
upfront | -1.997 |
hello | -1.997 |
Token | ? |
Token ID | 30 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
Magical | +5.417 |
adesh | +5.137 |
deen | +4.995 |
ower | +4.908 |
shed | +4.889 |
Neg logit contributions | |
roc | -7.497 |
opoly | -7.471 |
Times | -7.181 |
TEXT | -6.993 |
Msg | -6.921 |
Token | a |
Token ID | 257 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
Magical | +3.423 |
adesh | +3.237 |
ower | +3.153 |
deen | +3.150 |
shed | +3.109 |
Neg logit contributions | |
opoly | -3.404 |
roc | -3.309 |
Times | -3.077 |
hello | -3.019 |
Concord | -2.999 |
Token | shrine |
Token ID | 29981 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
Magical | +2.016 |
adesh | +1.934 |
deen | +1.886 |
ower | +1.876 |
shed | +1.812 |
Neg logit contributions | |
opoly | -2.013 |
roc | -1.996 |
Times | -1.854 |
Concord | -1.819 |
hello | -1.806 |
Token | for |
Token ID | 329 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
roc | +1.876 |
opoly | +1.834 |
Concord | +1.742 |
upfront | +1.726 |
Times | +1.714 |
Neg logit contributions | |
Magical | -2.004 |
adesh | -1.978 |
deen | -1.925 |
ower | -1.906 |
Piercing | -1.817 |
Token | Kar |
Token ID | 9375 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
Magical | +5.447 |
adesh | +5.113 |
deen | +5.030 |
shed | +4.933 |
ower | +4.915 |
Neg logit contributions | |
opoly | -5.475 |
roc | -5.180 |
hello | -4.861 |
Times | -4.857 |
Bee | -4.755 |
Token | th |
Token ID | 400 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
Magical | +1.945 |
adesh | +1.940 |
deen | +1.857 |
ower | +1.850 |
Piercing | +1.744 |
Neg logit contributions | |
opoly | -2.008 |
roc | -2.001 |
Concord | -1.876 |
Times | -1.867 |
upfront | -1.850 |
Token | ave |
Token ID | 1015 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
adesh | +5.929 |
Magical | +5.830 |
ower | +5.747 |
deen | +5.611 |
mage | +5.182 |
Neg logit contributions | |
opoly | -9.929 |
roc | -9.799 |
Times | -9.774 |
TEXT | -9.446 |
Concord | -9.441 |
Token | ery |
Token ID | 1924 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
ower | +10.601 |
Magical | +10.356 |
adesh | +9.945 |
mage | +9.536 |
ing | +9.497 |
Neg logit contributions | |
opoly | -15.630 |
Concord | -15.491 |
roc | -15.166 |
TEXT | -15.103 |
Times | -15.039 |
Token | ar |
Token ID | 283 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
Magical | +4.995 |
adesh | +4.988 |
ower | +4.872 |
deen | +4.826 |
ing | +4.586 |
Neg logit contributions | |
opoly | -4.109 |
roc | -4.014 |
Times | -3.863 |
Concord | -3.837 |
Msg | -3.791 |
Token | j |
Token ID | 73 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
roc | +3.690 |
opoly | +3.486 |
Times | +3.281 |
hello | +3.218 |
Concord | +3.203 |
Neg logit contributions | |
Magical | -5.267 |
adesh | -5.105 |
ower | -5.057 |
deen | -5.010 |
Piercing | -4.823 |
Token | una |
Token ID | 9613 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
adesh | +5.029 |
deen | +4.585 |
ower | +4.469 |
Magical | +4.312 |
ing | +3.976 |
Neg logit contributions | |
opoly | -12.439 |
roc | -12.398 |
Concord | -12.278 |
Msg | -11.892 |
Homs | -11.852 |
Token | . |
Token ID | 13 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
Magical | +1.430 |
adesh | +1.386 |
ower | +1.347 |
deen | +1.342 |
Piercing | +1.280 |
Neg logit contributions | |
opoly | -1.525 |
roc | -1.507 |
Times | -1.400 |
Concord | -1.390 |
upfront | -1.370 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
roc | +3.087 |
opoly | +3.022 |
upfront | +2.880 |
Concord | +2.828 |
Times | +2.811 |
Neg logit contributions | |
Magical | -3.564 |
adesh | -3.550 |
deen | -3.464 |
ower | -3.408 |
Piercing | -3.293 |
Token | In |
Token ID | 554 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
roc | +0.172 |
opoly | +0.171 |
Concord | +0.159 |
Times | +0.159 |
upfront | +0.157 |
Neg logit contributions | |
Magical | -0.162 |
adesh | -0.158 |
deen | -0.153 |
ower | -0.153 |
Piercing | -0.145 |
Token | this |
Token ID | 428 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
Magical | +0.482 |
adesh | +0.469 |
ower | +0.457 |
deen | +0.456 |
Piercing | +0.438 |
Neg logit contributions | |
roc | -0.396 |
opoly | -0.395 |
Times | -0.363 |
Concord | -0.362 |
upfront | -0.354 |
Token | post |
Token ID | 1281 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
Magical | +4.464 |
adesh | +4.198 |
ower | +4.183 |
shed | +4.147 |
deen | +4.091 |
Neg logit contributions | |
opoly | -3.600 |
roc | -3.597 |
Times | -3.280 |
Concord | -3.265 |
ustain | -3.259 |
Token | I |
Token ID | 314 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
Magical | +3.802 |
adesh | +3.697 |
ower | +3.633 |
shed | +3.526 |
deen | +3.495 |
Neg logit contributions | |
roc | -3.647 |
opoly | -3.640 |
Times | -3.392 |
Concord | -3.373 |
ustain | -3.305 |
Token | 'll |
Token ID | 1183 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ower | +6.220 |
shed | +5.630 |
Magical | +5.033 |
ing | +4.687 |
illuminate | +4.420 |
Neg logit contributions | |
ailability | -23.671 |
Ambro | -23.225 |
Bee | -23.113 |
Concord | -22.809 |
Times | -22.703 |
Token | round |
Token ID | 2835 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
shed | +8.583 |
Magical | +8.321 |
ower | +7.617 |
ing | +7.532 |
illuminate | +7.450 |
Neg logit contributions | |
ailability | -18.997 |
Times | -18.754 |
Bee | -18.638 |
opoly | -18.631 |
unbeliev | -18.153 |
Token | up |
Token ID | 510 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
Magical | +5.578 |
ower | +5.302 |
adesh | +5.249 |
ing | +5.184 |
shed | +5.157 |
Neg logit contributions | |
opoly | -5.976 |
roc | -5.873 |
Times | -5.567 |
Concord | -5.541 |
hello | -5.473 |
Token | some |
Token ID | 617 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
roc | +2.826 |
opoly | +2.588 |
upfront | +2.471 |
Times | +2.361 |
Concord | +2.342 |
Neg logit contributions | |
deen | -7.607 |
Magical | -7.535 |
adesh | -7.492 |
ower | -7.400 |
Piercing | -7.099 |
Token | got |
Token ID | 1392 |
Feature activation | -8.64e-01 |
Pos logit contributions | |
roc | +5.253 |
upfront | +5.033 |
opoly | +4.897 |
hello | +4.797 |
Times | +4.682 |
Neg logit contributions | |
deen | -9.399 |
Magical | -9.071 |
ower | -9.067 |
adesh | -8.931 |
Piercing | -8.624 |
Token | cha |
Token ID | 11693 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
roc | +1.283 |
opoly | +1.050 |
hello | +0.845 |
gets | +0.705 |
ouched | +0.546 |
Neg logit contributions | |
Magical | -15.891 |
deen | -15.772 |
ower | -15.568 |
adesh | -15.241 |
Pavilion | -15.172 |
Token | to |
Token ID | 284 |
Feature activation | -8.11e-03 |
Pos logit contributions | |
Magical | +1.263 |
adesh | +1.227 |
ower | +1.193 |
deen | +1.189 |
shed | +1.136 |
Neg logit contributions | |
opoly | -1.260 |
roc | -1.259 |
Times | -1.170 |
Concord | -1.154 |
upfront | -1.137 |
Token | prepare |
Token ID | 8335 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
roc | +0.091 |
opoly | +0.091 |
Concord | +0.085 |
Times | +0.085 |
upfront | +0.083 |
Neg logit contributions | |
Magical | -0.081 |
adesh | -0.079 |
deen | -0.077 |
ower | -0.077 |
Piercing | -0.073 |
Token | for |
Token ID | 329 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
roc | +4.179 |
Concord | +4.033 |
opoly | +4.019 |
upfront | +3.928 |
Times | +3.831 |
Neg logit contributions | |
Magical | -5.017 |
adesh | -4.927 |
deen | -4.874 |
Piercing | -4.736 |
ower | -4.734 |
Token | the |
Token ID | 262 |
Feature activation | -9.81e-01 |
Pos logit contributions | |
roc | +6.681 |
Concord | +6.526 |
upfront | +6.506 |
opoly | +6.366 |
ustain | +6.209 |
Neg logit contributions | |
Magical | -8.913 |
adesh | -8.826 |
deen | -8.644 |
Piercing | -8.392 |
ower | -8.364 |
Token | UN |
Token ID | 4725 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
upfront | +8.321 |
roc | +8.313 |
Concord | +8.265 |
opoly | +7.882 |
ustain | +7.779 |
Neg logit contributions | |
adesh | -10.335 |
Magical | -9.913 |
deen | -9.779 |
ower | -9.528 |
mage | -9.480 |
Token | climate |
Token ID | 4258 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
roc | +9.144 |
ustain | +8.846 |
Concord | +8.777 |
opoly | +8.766 |
Gl | +8.671 |
Neg logit contributions | |
Magical | -11.714 |
adesh | -11.554 |
deen | -11.319 |
Piercing | -11.219 |
ower | -11.159 |
Token | change |
Token ID | 1487 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
Concord | +12.539 |
roc | +12.340 |
ustain | +11.886 |
opoly | +11.693 |
upfront | +11.691 |
Neg logit contributions | |
Magical | -16.410 |
adesh | -16.310 |
Piercing | -15.958 |
田 | -15.781 |
Leilan | -15.009 |
Token | conference |
Token ID | 4495 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
Concord | +7.529 |
roc | +7.461 |
upfront | +7.254 |
ustain | +7.195 |
opoly | +7.128 |
Neg logit contributions | |
Magical | -12.577 |
adesh | -12.531 |
Piercing | -12.187 |
ower | -11.933 |
deen | -11.817 |
Token | in |
Token ID | 287 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
roc | +3.324 |
opoly | +3.228 |
upfront | +3.066 |
Concord | +3.061 |
Times | +2.988 |
Neg logit contributions | |
Magical | -3.784 |
adesh | -3.739 |
deen | -3.640 |
ower | -3.593 |
Piercing | -3.524 |
Token | Paris |
Token ID | 6342 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Concord | +6.856 |
Albany | +6.404 |
Sagan | +6.318 |
roc | +6.312 |
upfront | +6.070 |
Neg logit contributions | |
adesh | -16.149 |
Magical | -15.394 |
ower | -14.986 |
Piercing | -14.906 |
deen | -14.890 |
Token | in |
Token ID | 287 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
roc | +5.159 |
opoly | +4.961 |
Concord | +4.767 |
upfront | +4.760 |
Sagan | +4.559 |
Neg logit contributions | |
adesh | -6.981 |
Magical | -6.880 |
ower | -6.694 |
deen | -6.673 |
Piercing | -6.492 |
Token | ius |
Token ID | 3754 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
Magical | +1.799 |
adesh | +1.755 |
ower | +1.725 |
deen | +1.720 |
Piercing | +1.642 |
Neg logit contributions | |
opoly | -1.156 |
roc | -1.149 |
Times | -1.058 |
Concord | -1.048 |
upfront | -1.045 |
Token | says |
Token ID | 1139 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
roc | +2.623 |
opoly | +2.614 |
Concord | +2.460 |
Times | +2.444 |
upfront | +2.427 |
Neg logit contributions | |
Magical | -2.274 |
adesh | -2.253 |
deen | -2.169 |
ower | -2.132 |
Piercing | -2.048 |
Token | ahead |
Token ID | 4058 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
roc | +3.002 |
opoly | +3.002 |
Concord | +2.817 |
Times | +2.808 |
upfront | +2.798 |
Neg logit contributions | |
Magical | -3.254 |
adesh | -3.216 |
deen | -3.094 |
ower | -3.061 |
Piercing | -2.949 |
Token | of |
Token ID | 286 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
Magical | +0.623 |
adesh | +0.605 |
ower | +0.589 |
deen | +0.588 |
Piercing | +0.559 |
Neg logit contributions | |
roc | -0.651 |
opoly | -0.648 |
Times | -0.602 |
Concord | -0.601 |
upfront | -0.590 |
Token | UN |
Token ID | 4725 |
Feature activation | -5.88e-01 |
Pos logit contributions | |
roc | +3.048 |
opoly | +2.989 |
Concord | +2.847 |
upfront | +2.789 |
Times | +2.775 |
Neg logit contributions | |
adesh | -3.470 |
Magical | -3.434 |
deen | -3.355 |
ower | -3.355 |
Piercing | -3.197 |
Token | climate |
Token ID | 4258 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
roc | +5.802 |
opoly | +5.585 |
Concord | +5.338 |
Times | +5.248 |
Gl | +5.216 |
Neg logit contributions | |
Magical | -6.214 |
adesh | -6.167 |
deen | -6.060 |
ower | -5.940 |
Piercing | -5.834 |
Token | summit |
Token ID | 14237 |
Feature activation | -6.03e-01 |
Pos logit contributions | |
Concord | +10.909 |
roc | +10.632 |
opoly | +10.295 |
ustain | +10.120 |
Albany | +10.003 |
Neg logit contributions | |
adesh | -16.780 |
Magical | -16.679 |
Piercing | -16.201 |
田 | -16.114 |
ower | -15.291 |
Token | in |
Token ID | 287 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
roc | +5.605 |
opoly | +5.366 |
upfront | +5.250 |
Concord | +5.221 |
hello | +5.117 |
Neg logit contributions | |
Magical | -6.612 |
adesh | -6.491 |
deen | -6.392 |
ower | -6.256 |
Piercing | -6.157 |
Token | Paris |
Token ID | 6342 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
Concord | +10.357 |
Albany | +10.107 |
Sagan | +9.786 |
roc | +9.463 |
Homs | +9.172 |
Neg logit contributions | |
adesh | -14.915 |
Magical | -14.206 |
deen | -14.057 |
田 | -13.638 |
Piercing | -13.622 |
Token |
|
Token ID | 198 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
roc | +7.048 |
opoly | +6.778 |
Concord | +6.529 |
upfront | +6.512 |
Times | +6.335 |
Neg logit contributions | |
adesh | -8.655 |
Magical | -8.637 |
ower | -8.328 |
deen | -8.319 |
Piercing | -8.160 |
Token |
|
Token ID | 198 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
roc | +2.017 |
opoly | +1.966 |
Times | +1.877 |
Concord | +1.844 |
upfront | +1.842 |
Neg logit contributions | |
Magical | -2.377 |
adesh | -2.362 |
deen | -2.293 |
ower | -2.254 |
Piercing | -2.178 |
Token |
|
Token ID | 198 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
roc | +3.735 |
opoly | +3.605 |
Times | +3.542 |
Concord | +3.493 |
Msg | +3.477 |
Neg logit contributions | |
adesh | -5.113 |
deen | -5.019 |
Magical | -4.958 |
ower | -4.845 |
shed | -4.670 |
Token |
|
Token ID | 198 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
roc | +4.250 |
Times | +3.977 |
opoly | +3.957 |
upfront | +3.924 |
Bee | +3.881 |
Neg logit contributions | |
adesh | -6.597 |
deen | -6.506 |
Magical | -6.404 |
ower | -6.211 |
shed | -6.038 |
Token | On |
Token ID | 2202 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
roc | +4.407 |
Times | +4.292 |
Msg | +4.222 |
opoly | +4.156 |
Bee | +4.133 |
Neg logit contributions | |
Magical | -9.621 |
deen | -9.582 |
adesh | -9.316 |
ower | -9.161 |
Piercing | -9.074 |
Token | September |
Token ID | 2693 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
roc | +3.067 |
opoly | +2.719 |
upfront | +2.493 |
Msg | +2.424 |
Concord | +2.407 |
Neg logit contributions | |
deen | -8.439 |
adesh | -8.425 |
Magical | -8.199 |
ower | -8.136 |
Piercing | -7.906 |
Token | 20 |
Token ID | 1160 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Concord | +4.217 |
opoly | +4.176 |
roc | +4.146 |
Sagan | +3.551 |
Homs | +3.512 |
Neg logit contributions | |
adesh | -15.176 |
deen | -14.989 |
ower | -14.785 |
Pavilion | -14.494 |
mage | -14.360 |
Token | 2013 |
Token ID | 2211 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Magical | +1.129 |
adesh | +1.060 |
ower | +1.034 |
deen | +1.029 |
Piercing | +0.972 |
Neg logit contributions | |
opoly | -1.719 |
roc | -1.706 |
Times | -1.594 |
Concord | -1.585 |
hello | -1.572 |
Token | 07 |
Token ID | 8753 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
Concord | +8.390 |
roc | +7.501 |
Sagan | +7.304 |
upfront | +7.251 |
Schmidt | +6.976 |
Neg logit contributions | |
adesh | -20.052 |
deen | -19.653 |
ower | -19.652 |
田 | -18.442 |
bley | -18.399 |
Token | : |
Token ID | 25 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
roc | +2.315 |
opoly | +2.225 |
Times | +2.110 |
upfront | +2.083 |
Concord | +2.065 |
Neg logit contributions | |
Magical | -3.385 |
adesh | -3.371 |
deen | -3.313 |
ower | -3.241 |
Piercing | -3.149 |
Token | 44 |
Token ID | 2598 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
roc | +1.765 |
opoly | +1.731 |
Times | +1.649 |
Concord | +1.620 |
upfront | +1.599 |
Neg logit contributions | |
Magical | -1.901 |
adesh | -1.887 |
deen | -1.836 |
ower | -1.812 |
Piercing | -1.737 |
Token | Se |
Token ID | 1001 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
roc | +3.872 |
opoly | +3.844 |
Concord | +3.718 |
Times | +3.576 |
upfront | +3.575 |
Neg logit contributions | |
adesh | -3.753 |
Magical | -3.714 |
deen | -3.673 |
ower | -3.619 |
Pavilion | -3.370 |
Token | eker |
Token ID | 28233 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
adesh | +1.945 |
Magical | +1.894 |
ower | +1.884 |
deen | +1.813 |
Pavilion | +1.590 |
Neg logit contributions | |
opoly | -5.730 |
roc | -5.712 |
Concord | -5.520 |
Times | -5.512 |
TEXT | -5.447 |
Token |
|
Token ID | 198 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
Concord | +6.314 |
roc | +6.099 |
Albany | +6.053 |
opoly | +5.981 |
Times | +5.906 |
Neg logit contributions | |
deen | -8.423 |
adesh | -8.202 |
ower | -7.829 |
Magical | -7.652 |
shed | -7.604 |
Token |
|
Token ID | 198 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
roc | +2.989 |
opoly | +2.886 |
Times | +2.792 |
upfront | +2.726 |
Concord | +2.708 |
Neg logit contributions | |
adesh | -3.842 |
Magical | -3.834 |
deen | -3.749 |
ower | -3.658 |
shed | -3.531 |
Token | In |
Token ID | 818 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
roc | +4.501 |
Times | +4.470 |
opoly | +4.291 |
Gl | +4.218 |
Bee | +4.165 |
Neg logit contributions | |
Magical | -7.646 |
deen | -7.470 |
adesh | -7.437 |
shed | -7.189 |
Piercing | -7.129 |
Token | a |
Token ID | 257 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
roc | +5.149 |
opoly | +5.064 |
Concord | +4.953 |
Albany | +4.834 |
upfront | +4.672 |
Neg logit contributions | |
adesh | -7.091 |
deen | -7.022 |
Magical | -6.998 |
ower | -6.723 |
Piercing | -6.517 |
Token | New |
Token ID | 968 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
Concord | +9.146 |
roc | +9.099 |
opoly | +8.773 |
Albany | +8.593 |
upfront | +8.455 |
Neg logit contributions | |
deen | -10.570 |
adesh | -10.437 |
Magical | -10.062 |
ower | -10.031 |
田 | -9.934 |
Token | York |
Token ID | 1971 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
roc | +4.202 |
Albany | +4.031 |
Times | +3.932 |
opoly | +3.915 |
Concord | +3.756 |
Neg logit contributions | |
adesh | -13.538 |
Magical | -13.420 |
deen | -13.363 |
ower | -13.009 |
Piercing | -12.771 |
Token | State |
Token ID | 1812 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Times | +12.742 |
roc | +12.668 |
Concord | +12.051 |
opoly | +11.814 |
Albany | +11.737 |
Neg logit contributions | |
deen | -16.048 |
adesh | -15.375 |
田 | -14.859 |
Magical | -14.681 |
ower | -14.424 |
Token | poll |
Token ID | 3278 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
Concord | +14.146 |
roc | +13.945 |
opoly | +13.276 |
Albany | +13.209 |
Times | +13.136 |
Neg logit contributions | |
deen | -13.047 |
adesh | -12.558 |
田 | -11.927 |
Magical | -11.513 |
atche | -11.434 |
Token | of |
Token ID | 286 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
roc | +4.609 |
opoly | +4.544 |
Concord | +4.363 |
upfront | +4.257 |
Times | +4.227 |
Neg logit contributions | |
Magical | -5.567 |
deen | -5.501 |
adesh | -5.431 |
ower | -5.308 |
Piercing | -5.127 |
Token | 1 |
Token ID | 352 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
roc | +7.718 |
Concord | +7.498 |
Albany | +7.359 |
opoly | +7.158 |
upfront | +7.080 |
Neg logit contributions | |
adesh | -9.575 |
deen | -9.475 |
Magical | -9.374 |
ower | -9.177 |
田 | -9.017 |
Token | , |
Token ID | 11 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
roc | +3.768 |
opoly | +3.727 |
Concord | +3.518 |
Times | +3.454 |
upfront | +3.449 |
Neg logit contributions | |
Magical | -4.829 |
adesh | -4.740 |
deen | -4.723 |
ower | -4.598 |
Piercing | -4.469 |
Token | Fantasy |
Token ID | 10982 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
opoly | +7.972 |
upfront | +7.820 |
roc | +7.743 |
Concord | +7.420 |
Times | +6.916 |
Neg logit contributions | |
adesh | -15.915 |
deen | -15.679 |
ower | -14.808 |
Magical | -14.372 |
田 | -14.326 |
Token | points |
Token ID | 2173 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
opoly | +3.923 |
roc | +3.877 |
Concord | +3.670 |
upfront | +3.663 |
Times | +3.620 |
Neg logit contributions | |
adesh | -4.837 |
Magical | -4.772 |
deen | -4.764 |
ower | -4.688 |
Piercing | -4.427 |
Token | in |
Token ID | 287 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
roc | +5.222 |
upfront | +5.171 |
opoly | +5.141 |
Concord | +4.959 |
Albany | +4.735 |
Neg logit contributions | |
Magical | -6.920 |
adesh | -6.890 |
deen | -6.762 |
ower | -6.634 |
Piercing | -6.443 |
Token | his |
Token ID | 465 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
roc | +5.262 |
opoly | +5.041 |
Concord | +5.016 |
upfront | +4.985 |
Albany | +4.816 |
Neg logit contributions | |
adesh | -8.424 |
deen | -8.247 |
Magical | -8.102 |
ower | -8.032 |
田 | -7.704 |
Token | past |
Token ID | 1613 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
roc | +6.969 |
opoly | +6.771 |
upfront | +6.672 |
Concord | +6.515 |
Times | +6.330 |
Neg logit contributions | |
adesh | -6.930 |
deen | -6.921 |
Magical | -6.747 |
ower | -6.577 |
Piercing | -6.287 |
Token | three |
Token ID | 1115 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
roc | +5.159 |
upfront | +4.847 |
opoly | +4.773 |
Concord | +4.758 |
Albany | +4.591 |
Neg logit contributions | |
deen | -9.998 |
adesh | -9.961 |
Magical | -9.857 |
ower | -9.749 |
田 | -9.363 |
Token | home |
Token ID | 1363 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
upfront | +10.218 |
roc | +10.032 |
Concord | +9.643 |
Albany | +9.592 |
contest | +9.461 |
Neg logit contributions | |
adesh | -15.925 |
deen | -15.762 |
田 | -15.257 |
ower | -15.053 |
Piercing | -14.781 |
Token | games |
Token ID | 1830 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
roc | +7.960 |
opoly | +7.499 |
upfront | +7.011 |
Concord | +6.894 |
Times | +6.827 |
Neg logit contributions | |
deen | -12.066 |
adesh | -11.962 |
Magical | -11.853 |
田 | -11.626 |
Piercing | -11.611 |
Token | , |
Token ID | 11 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
roc | +5.376 |
opoly | +5.295 |
upfront | +5.153 |
Concord | +5.011 |
Albany | +4.857 |
Neg logit contributions | |
Magical | -6.604 |
adesh | -6.563 |
deen | -6.455 |
ower | -6.293 |
Piercing | -6.148 |
Token | and |
Token ID | 290 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
roc | +0.914 |
opoly | +0.909 |
Concord | +0.850 |
Times | +0.844 |
upfront | +0.838 |
Neg logit contributions | |
Magical | -0.899 |
adesh | -0.879 |
deen | -0.859 |
ower | -0.851 |
Piercing | -0.811 |
Token | he |
Token ID | 339 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
roc | +3.508 |
opoly | +3.459 |
Concord | +3.336 |
upfront | +3.259 |
Times | +3.193 |
Neg logit contributions | |
adesh | -4.247 |
Magical | -4.189 |
deen | -4.131 |
ower | -4.037 |
Piercing | -3.849 |
Token |
|
Token ID | 198 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
roc | +2.989 |
opoly | +2.886 |
Times | +2.792 |
upfront | +2.726 |
Concord | +2.708 |
Neg logit contributions | |
adesh | -3.842 |
Magical | -3.834 |
deen | -3.749 |
ower | -3.658 |
shed | -3.531 |
Token | In |
Token ID | 818 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
roc | +4.501 |
Times | +4.470 |
opoly | +4.291 |
Gl | +4.218 |
Bee | +4.165 |
Neg logit contributions | |
Magical | -7.646 |
deen | -7.470 |
adesh | -7.437 |
shed | -7.189 |
Piercing | -7.129 |
Token | a |
Token ID | 257 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
roc | +5.149 |
opoly | +5.064 |
Concord | +4.953 |
Albany | +4.834 |
upfront | +4.672 |
Neg logit contributions | |
adesh | -7.091 |
deen | -7.022 |
Magical | -6.998 |
ower | -6.723 |
Piercing | -6.517 |
Token | New |
Token ID | 968 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
Concord | +9.146 |
roc | +9.099 |
opoly | +8.773 |
Albany | +8.593 |
upfront | +8.455 |
Neg logit contributions | |
deen | -10.570 |
adesh | -10.437 |
Magical | -10.062 |
ower | -10.031 |
田 | -9.934 |
Token | York |
Token ID | 1971 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
roc | +4.202 |
Albany | +4.031 |
Times | +3.932 |
opoly | +3.915 |
Concord | +3.756 |
Neg logit contributions | |
adesh | -13.538 |
Magical | -13.420 |
deen | -13.363 |
ower | -13.009 |
Piercing | -12.771 |
Token | State |
Token ID | 1812 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Times | +12.742 |
roc | +12.668 |
Concord | +12.051 |
opoly | +11.814 |
Albany | +11.737 |
Neg logit contributions | |
deen | -16.048 |
adesh | -15.375 |
田 | -14.859 |
Magical | -14.681 |
ower | -14.424 |
Token | poll |
Token ID | 3278 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
Concord | +14.146 |
roc | +13.945 |
opoly | +13.276 |
Albany | +13.209 |
Times | +13.136 |
Neg logit contributions | |
deen | -13.047 |
adesh | -12.558 |
田 | -11.927 |
Magical | -11.513 |
atche | -11.434 |
Token | of |
Token ID | 286 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
roc | +4.609 |
opoly | +4.544 |
Concord | +4.363 |
upfront | +4.257 |
Times | +4.227 |
Neg logit contributions | |
Magical | -5.567 |
deen | -5.501 |
adesh | -5.431 |
ower | -5.308 |
Piercing | -5.127 |
Token | 1 |
Token ID | 352 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
roc | +7.718 |
Concord | +7.498 |
Albany | +7.359 |
opoly | +7.158 |
upfront | +7.080 |
Neg logit contributions | |
adesh | -9.575 |
deen | -9.475 |
Magical | -9.374 |
ower | -9.177 |
田 | -9.017 |
Token | , |
Token ID | 11 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
roc | +3.768 |
opoly | +3.727 |
Concord | +3.518 |
Times | +3.454 |
upfront | +3.449 |
Neg logit contributions | |
Magical | -4.829 |
adesh | -4.740 |
deen | -4.723 |
ower | -4.598 |
Piercing | -4.469 |
Token | 302 |
Token ID | 22709 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
roc | +3.348 |
opoly | +3.299 |
Times | +3.088 |
Concord | +3.079 |
upfront | +2.979 |
Neg logit contributions | |
Magical | -3.764 |
adesh | -3.684 |
deen | -3.614 |
ower | -3.537 |
Piercing | -3.418 |
Token | of |
Token ID | 286 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
Magical | +0.647 |
adesh | +0.630 |
ower | +0.613 |
deen | +0.612 |
Piercing | +0.580 |
Neg logit contributions | |
roc | -0.678 |
opoly | -0.674 |
Times | -0.628 |
Concord | -0.627 |
upfront | -0.616 |
Token | thousands |
Token ID | 4138 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
opoly | +4.780 |
roc | +4.748 |
upfront | +4.508 |
Concord | +4.415 |
Times | +4.384 |
Neg logit contributions | |
deen | -6.276 |
adesh | -6.231 |
Magical | -6.177 |
ower | -5.919 |
Piercing | -5.840 |
Token | of |
Token ID | 286 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
roc | +4.535 |
opoly | +4.504 |
upfront | +4.173 |
Concord | +4.129 |
Times | +4.069 |
Neg logit contributions | |
Magical | -5.891 |
adesh | -5.864 |
deen | -5.708 |
ower | -5.571 |
Piercing | -5.456 |
Token | duplicate |
Token ID | 23418 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
upfront | +10.494 |
Albany | +10.265 |
roc | +10.188 |
opoly | +10.118 |
Concord | +10.089 |
Neg logit contributions | |
deen | -14.024 |
adesh | -13.567 |
Magical | -12.636 |
ower | -12.520 |
Piercing | -12.471 |
Token | and |
Token ID | 290 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
roc | +9.364 |
opoly | +9.358 |
upfront | +9.156 |
Concord | +8.854 |
Times | +8.812 |
Neg logit contributions | |
deen | -11.783 |
adesh | -11.149 |
Magical | -10.885 |
ower | -10.692 |
Piercing | -10.561 |
Token | ineligible |
Token ID | 39538 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
opoly | +4.636 |
roc | +4.598 |
upfront | +4.335 |
Concord | +4.234 |
Times | +4.205 |
Neg logit contributions | |
Magical | -4.854 |
deen | -4.766 |
adesh | -4.766 |
ower | -4.535 |
Piercing | -4.426 |
Token | consumers |
Token ID | 7008 |
Feature activation | -8.07e-02 |
Pos logit contributions | |
upfront | +11.510 |
opoly | +11.206 |
roc | +11.155 |
Caller | +11.143 |
Albany | +10.976 |
Neg logit contributions | |
deen | -14.754 |
adesh | -13.849 |
Magical | -13.415 |
田 | -13.249 |
Piercing | -13.211 |
Token | in |
Token ID | 287 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
roc | +0.880 |
opoly | +0.871 |
Concord | +0.813 |
Times | +0.810 |
upfront | +0.807 |
Neg logit contributions | |
Magical | -0.839 |
adesh | -0.821 |
deen | -0.796 |
ower | -0.790 |
Piercing | -0.756 |
Token | the |
Token ID | 262 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
roc | +4.521 |
opoly | +4.493 |
upfront | +4.189 |
Concord | +4.188 |
Times | +4.108 |
Neg logit contributions | |
deen | -4.281 |
adesh | -4.222 |
Magical | -4.189 |
ower | -4.018 |
Piercing | -3.840 |
Token | Lif |
Token ID | 21073 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
opoly | +6.081 |
roc | +6.052 |
upfront | +5.795 |
Concord | +5.751 |
Albany | +5.636 |
Neg logit contributions | |
deen | -4.967 |
adesh | -4.953 |
Magical | -4.864 |
ower | -4.688 |
Piercing | -4.405 |
Token | eline |
Token ID | 4470 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
roc | +1.347 |
opoly | +1.337 |
Times | +1.257 |
Concord | +1.253 |
upfront | +1.242 |
Neg logit contributions | |
Magical | -0.766 |
adesh | -0.734 |
deen | -0.703 |
ower | -0.695 |
Piercing | -0.663 |
Token | , |
Token ID | 11 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
roc | +4.939 |
opoly | +4.891 |
upfront | +4.638 |
Concord | +4.514 |
Bee | +4.370 |
Neg logit contributions | |
Magical | -6.560 |
adesh | -6.416 |
deen | -6.316 |
ower | -6.138 |
Piercing | -5.986 |
Token | two |
Token ID | 734 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
roc | +3.189 |
opoly | +3.158 |
Concord | +2.982 |
upfront | +2.915 |
Times | +2.902 |
Neg logit contributions | |
Magical | -3.835 |
adesh | -3.755 |
deen | -3.731 |
ower | -3.671 |
Piercing | -3.464 |
Token | facts |
Token ID | 6419 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
opoly | +4.563 |
roc | +4.556 |
Concord | +4.362 |
upfront | +4.313 |
Times | +4.302 |
Neg logit contributions | |
Magical | -5.090 |
deen | -5.028 |
adesh | -4.954 |
ower | -4.828 |
Piercing | -4.649 |
Token | need |
Token ID | 761 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
upfront | +8.558 |
roc | +8.183 |
opoly | +8.168 |
Concord | +7.827 |
Times | +7.510 |
Neg logit contributions | |
adesh | -12.651 |
Magical | -12.604 |
deen | -12.441 |
ower | -12.327 |
Pavilion | -12.052 |
Token | to |
Token ID | 284 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
roc | +7.820 |
upfront | +7.224 |
Concord | +6.984 |
opoly | +6.810 |
Times | +6.773 |
Neg logit contributions | |
adesh | -13.153 |
Magical | -13.101 |
deen | -12.662 |
ower | -12.115 |
Pavilion | -12.115 |
Token | be |
Token ID | 307 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
roc | +7.248 |
upfront | +7.200 |
Concord | +6.869 |
opoly | +6.845 |
ouched | +6.570 |
Neg logit contributions | |
Magical | -12.496 |
adesh | -12.399 |
Pavilion | -11.612 |
Piercing | -11.608 |
deen | -11.395 |
Token | determined |
Token ID | 5295 |
Feature activation | -9.36e-01 |
Pos logit contributions | |
upfront | +12.102 |
Concord | +11.357 |
roc | +10.999 |
Albany | +10.848 |
ouched | +10.496 |
Neg logit contributions | |
adesh | -14.451 |
Magical | -14.330 |
deen | -14.223 |
Piercing | -13.625 |
ower | -13.616 |
Token | . |
Token ID | 13 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
upfront | +7.437 |
roc | +7.065 |
opoly | +6.673 |
Concord | +6.452 |
Albany | +6.164 |
Neg logit contributions | |
Magical | -10.838 |
adesh | -10.838 |
deen | -10.740 |
ower | -10.436 |
Piercing | -10.163 |
Token | If |
Token ID | 1002 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
roc | +4.091 |
opoly | +4.049 |
Concord | +3.960 |
Times | +3.862 |
upfront | +3.771 |
Neg logit contributions | |
adesh | -5.011 |
deen | -5.006 |
ower | -4.933 |
Magical | -4.907 |
shed | -4.621 |
Token | tree |
Token ID | 5509 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
roc | +1.711 |
opoly | +1.695 |
Concord | +1.581 |
Times | +1.564 |
upfront | +1.557 |
Neg logit contributions | |
Magical | -2.043 |
adesh | -2.001 |
deen | -1.969 |
ower | -1.945 |
Piercing | -1.853 |
Token | roots |
Token ID | 11135 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
roc | +1.465 |
opoly | +1.445 |
Concord | +1.303 |
Times | +1.292 |
upfront | +1.282 |
Neg logit contributions | |
Magical | -2.602 |
adesh | -2.553 |
deen | -2.516 |
ower | -2.460 |
Piercing | -2.393 |
Token | of |
Token ID | 286 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
roc | +2.606 |
opoly | +2.503 |
Concord | +2.425 |
upfront | +2.360 |
Times | +2.347 |
Neg logit contributions | |
adesh | -3.524 |
Magical | -3.487 |
deen | -3.422 |
ower | -3.342 |
Piercing | -3.227 |
Token | a |
Token ID | 257 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
Magical | +0.955 |
adesh | +0.922 |
ower | +0.901 |
deen | +0.899 |
Piercing | +0.860 |
Neg logit contributions | |
opoly | -0.902 |
roc | -0.897 |
Times | -0.833 |
Concord | -0.824 |
upfront | -0.816 |
Token | missing |
Token ID | 4814 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
roc | +4.089 |
opoly | +3.943 |
Concord | +3.891 |
Albany | +3.666 |
upfront | +3.665 |
Neg logit contributions | |
adesh | -4.923 |
Magical | -4.783 |
deen | -4.765 |
ower | -4.649 |
Piercing | -4.400 |
Token | 1982 |
Token ID | 14489 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
roc | +6.450 |
Concord | +6.008 |
opoly | +5.977 |
Albany | +5.862 |
Sagan | +5.509 |
Neg logit contributions | |
adesh | -9.653 |
deen | -9.382 |
Magical | -9.140 |
ower | -9.097 |
Piercing | -8.736 |
Token | V |
Token ID | 569 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
roc | +7.044 |
Concord | +6.616 |
opoly | +6.337 |
Albany | +6.252 |
Sagan | +6.061 |
Neg logit contributions | |
adesh | -9.861 |
deen | -9.685 |
Magical | -9.431 |
ower | -9.344 |
Piercing | -8.974 |
Token | itus |
Token ID | 17506 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
roc | +0.690 |
opoly | +0.674 |
Times | +0.584 |
Concord | +0.560 |
upfront | +0.530 |
Neg logit contributions | |
Magical | -2.507 |
adesh | -2.438 |
deen | -2.404 |
ower | -2.398 |
Piercing | -2.353 |
Token | road |
Token ID | 2975 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
roc | +12.199 |
Concord | +11.241 |
Albany | +10.610 |
opoly | +10.144 |
Schmidt | +10.084 |
Neg logit contributions | |
adesh | -15.510 |
deen | -14.243 |
ower | -13.664 |
Magical | -13.581 |
田 | -13.098 |
Token | bike |
Token ID | 7161 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
roc | +1.733 |
opoly | +1.685 |
Concord | +1.512 |
Times | +1.468 |
hello | +1.437 |
Neg logit contributions | |
Magical | -3.290 |
adesh | -3.238 |
deen | -3.172 |
ower | -3.133 |
Piercing | -3.052 |
Token | . |
Token ID | 13 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
roc | +5.564 |
opoly | +5.292 |
Concord | +5.291 |
Albany | +5.066 |
upfront | +5.030 |
Neg logit contributions | |
adesh | -7.025 |
Magical | -6.841 |
deen | -6.770 |
ower | -6.567 |
Piercing | -6.365 |
Token |
|
Token ID | 198 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
Magical | +3.539 |
adesh | +3.252 |
ower | +3.188 |
deen | +3.138 |
Piercing | +3.115 |
Neg logit contributions | |
opoly | -4.178 |
roc | -4.115 |
Times | -3.816 |
ouched | -3.813 |
upfront | -3.812 |
Token |
|
Token ID | 198 |
Feature activation | +4.54e-01 |
Pos logit contributions | |
Magical | +3.936 |
ower | +3.715 |
adesh | +3.658 |
ing | +3.627 |
deen | +3.533 |
Neg logit contributions | |
opoly | -5.940 |
roc | -5.781 |
ailability | -5.592 |
Concord | -5.491 |
ustain | -5.402 |
Token | in |
Token ID | 287 |
Feature activation | -6.98e-01 |
Pos logit contributions | |
roc | +3.931 |
opoly | +3.876 |
upfront | +3.740 |
Concord | +3.603 |
Times | +3.548 |
Neg logit contributions | |
Magical | -4.504 |
adesh | -4.366 |
deen | -4.315 |
ower | -4.249 |
Piercing | -4.137 |
Token | each |
Token ID | 1123 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
roc | +5.795 |
opoly | +5.580 |
Concord | +5.558 |
Albany | +5.454 |
upfront | +5.430 |
Neg logit contributions | |
adesh | -8.336 |
deen | -8.133 |
Magical | -8.088 |
ower | -8.007 |
田 | -7.678 |
Token | of |
Token ID | 286 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
opoly | +6.337 |
roc | +6.220 |
upfront | +5.963 |
Concord | +5.955 |
etition | +5.935 |
Neg logit contributions | |
adesh | -10.066 |
Magical | -9.730 |
deen | -9.695 |
ower | -9.512 |
Piercing | -9.262 |
Token | his |
Token ID | 465 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
roc | +5.740 |
Concord | +5.441 |
opoly | +5.389 |
Albany | +5.347 |
upfront | +5.178 |
Neg logit contributions | |
adesh | -9.300 |
deen | -8.979 |
Magical | -8.826 |
ower | -8.715 |
Piercing | -8.370 |
Token | past |
Token ID | 1613 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
roc | +6.478 |
upfront | +6.306 |
Concord | +6.109 |
opoly | +6.094 |
Albany | +6.052 |
Neg logit contributions | |
deen | -8.911 |
adesh | -8.863 |
Magical | -8.647 |
ower | -8.470 |
Piercing | -8.125 |
Token | three |
Token ID | 1115 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
roc | +4.619 |
upfront | +4.314 |
Concord | +4.281 |
opoly | +4.211 |
Albany | +4.167 |
Neg logit contributions | |
adesh | -8.611 |
Magical | -8.535 |
deen | -8.509 |
ower | -8.420 |
田 | -8.074 |
Token | games |
Token ID | 1830 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
upfront | +9.675 |
roc | +9.543 |
Albany | +9.313 |
Concord | +9.146 |
contest | +9.028 |
Neg logit contributions | |
adesh | -16.136 |
deen | -15.755 |
田 | -15.466 |
ower | -15.116 |
Piercing | -14.920 |
Token | . |
Token ID | 13 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
roc | +5.866 |
opoly | +5.653 |
upfront | +5.646 |
Concord | +5.400 |
Albany | +5.289 |
Neg logit contributions | |
Magical | -7.722 |
adesh | -7.714 |
deen | -7.566 |
ower | -7.350 |
Piercing | -7.273 |
Token | We |
Token ID | 775 |
Feature activation | +5.27e-01 |
Pos logit contributions | |
roc | +5.118 |
Concord | +5.043 |
opoly | +5.010 |
Albany | +4.953 |
Times | +4.938 |
Neg logit contributions | |
adesh | -6.578 |
deen | -6.427 |
ower | -6.126 |
Magical | -6.088 |
shed | -5.916 |
Token | hope |
Token ID | 2911 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
Magical | +4.975 |
ower | +4.801 |
shed | +4.691 |
adesh | +4.667 |
deen | +4.641 |
Neg logit contributions | |
opoly | -6.042 |
roc | -5.896 |
Times | -5.561 |
TEXT | -5.521 |
Concord | -5.478 |
Token | that |
Token ID | 326 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
roc | +4.954 |
opoly | +4.695 |
Concord | +4.617 |
upfront | +4.612 |
Albany | +4.559 |
Neg logit contributions | |
adesh | -7.157 |
deen | -7.018 |
Magical | -6.861 |
ower | -6.685 |
Piercing | -6.450 |
Token | quarterbacks |
Token ID | 23643 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
roc | +7.991 |
opoly | +7.825 |
upfront | +7.824 |
Concord | +7.739 |
Albany | +7.573 |
Neg logit contributions | |
adesh | -9.403 |
deen | -9.198 |
Magical | -9.012 |
ower | -8.554 |
Piercing | -8.438 |
Token | to |
Token ID | 284 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
roc | +4.635 |
opoly | +4.575 |
upfront | +4.467 |
Concord | +4.393 |
Albany | +4.264 |
Neg logit contributions | |
adesh | -5.941 |
Magical | -5.814 |
deen | -5.698 |
ower | -5.560 |
Piercing | -5.367 |
Token | score |
Token ID | 4776 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
roc | +3.927 |
opoly | +3.910 |
Concord | +3.639 |
upfront | +3.612 |
Times | +3.595 |
Neg logit contributions | |
Magical | -4.126 |
adesh | -4.119 |
deen | -3.874 |
ower | -3.872 |
Piercing | -3.765 |
Token | at |
Token ID | 379 |
Feature activation | -9.81e-01 |
Pos logit contributions | |
roc | +6.821 |
opoly | +6.742 |
upfront | +6.712 |
Concord | +6.525 |
Albany | +6.301 |
Neg logit contributions | |
adesh | -9.300 |
deen | -8.950 |
ower | -8.901 |
Magical | -8.863 |
Piercing | -8.323 |
Token | least |
Token ID | 1551 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
roc | +7.502 |
Concord | +7.059 |
opoly | +6.961 |
Albany | +6.813 |
upfront | +6.715 |
Neg logit contributions | |
adesh | -12.077 |
deen | -11.413 |
ower | -11.193 |
Magical | -11.142 |
田 | -10.591 |
Token | 20 |
Token ID | 1160 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
upfront | +7.020 |
roc | +6.830 |
opoly | +6.485 |
Concord | +6.475 |
Albany | +6.226 |
Neg logit contributions | |
adesh | -13.823 |
deen | -13.097 |
ower | -13.082 |
Magical | -12.513 |
田 | -12.277 |
Token | Fantasy |
Token ID | 10982 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
upfront | +9.402 |
opoly | +9.299 |
roc | +9.087 |
Concord | +8.874 |
Gl | +8.245 |
Neg logit contributions | |
adesh | -16.313 |
deen | -16.069 |
ower | -15.225 |
田 | -15.107 |
Magical | -14.684 |
Token | points |
Token ID | 2173 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
opoly | +5.533 |
roc | +5.423 |
Concord | +5.185 |
upfront | +5.140 |
Times | +5.057 |
Neg logit contributions | |
adesh | -7.561 |
deen | -7.514 |
Magical | -7.504 |
ower | -7.408 |
Piercing | -6.998 |
Token | against |
Token ID | 1028 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
upfront | +6.460 |
roc | +6.444 |
opoly | +6.376 |
Concord | +6.143 |
Albany | +5.926 |
Neg logit contributions | |
adesh | -8.303 |
Magical | -8.287 |
deen | -8.090 |
ower | -7.902 |
Piercing | -7.727 |
Token | Denver |
Token ID | 10656 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
roc | +7.647 |
Concord | +7.416 |
upfront | +7.307 |
opoly | +7.300 |
Albany | +7.240 |
Neg logit contributions | |
adesh | -11.440 |
deen | -10.810 |
Magical | -10.686 |
ower | -10.426 |
Piercing | -10.251 |
Token | this |
Token ID | 428 |
Feature activation | -8.96e-01 |
Pos logit contributions | |
roc | +5.189 |
opoly | +5.173 |
upfront | +5.067 |
Concord | +4.962 |
Times | +4.793 |
Neg logit contributions | |
Magical | -6.479 |
adesh | -6.435 |
deen | -6.294 |
ower | -6.217 |
Piercing | -6.047 |
Token | an |
Token ID | 281 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
roc | +5.922 |
opoly | +5.804 |
upfront | +5.629 |
Concord | +5.611 |
Times | +5.423 |
Neg logit contributions | |
Magical | -4.646 |
adesh | -4.537 |
deen | -4.413 |
ower | -4.279 |
Piercing | -4.189 |
Token | ambitious |
Token ID | 14742 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
upfront | +5.030 |
opoly | +4.798 |
roc | +4.693 |
Concord | +4.627 |
Albany | +4.576 |
Neg logit contributions | |
adesh | -9.780 |
Magical | -9.666 |
deen | -9.490 |
Piercing | -9.435 |
ower | -9.247 |
Token | goal |
Token ID | 3061 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
upfront | +8.952 |
Concord | +8.900 |
roc | +8.664 |
opoly | +8.514 |
Times | +8.396 |
Neg logit contributions | |
adesh | -10.619 |
deen | -10.278 |
Magical | -10.215 |
Piercing | -10.114 |
mage | -9.975 |
Token | for |
Token ID | 329 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
roc | +6.798 |
opoly | +6.616 |
upfront | +6.505 |
Concord | +6.311 |
Times | +6.095 |
Neg logit contributions | |
Magical | -10.249 |
adesh | -10.059 |
deen | -10.042 |
Piercing | -9.739 |
ower | -9.656 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
roc | +5.709 |
Concord | +5.609 |
upfront | +5.542 |
opoly | +5.393 |
Albany | +5.208 |
Neg logit contributions | |
adesh | -8.337 |
Magical | -8.246 |
deen | -8.130 |
ower | -7.837 |
Piercing | -7.662 |
Token | Paris |
Token ID | 6342 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Concord | +9.113 |
roc | +8.925 |
upfront | +8.916 |
Albany | +8.275 |
ustain | +8.274 |
Neg logit contributions | |
adesh | -12.212 |
Magical | -11.433 |
deen | -11.416 |
田 | -11.162 |
Piercing | -11.122 |
Token | meeting |
Token ID | 3249 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
roc | +10.825 |
Concord | +10.445 |
upfront | +9.955 |
opoly | +9.829 |
ustain | +9.809 |
Neg logit contributions | |
adesh | -14.476 |
Piercing | -13.989 |
ower | -13.783 |
Magical | -13.607 |
田 | -13.463 |
Token | – |
Token ID | 784 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
roc | +3.495 |
opoly | +3.371 |
upfront | +3.236 |
Concord | +3.202 |
Times | +3.176 |
Neg logit contributions | |
Magical | -3.947 |
adesh | -3.864 |
deen | -3.808 |
ower | -3.722 |
Piercing | -3.646 |
Token | an |
Token ID | 281 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
roc | +4.457 |
opoly | +4.309 |
Concord | +4.176 |
upfront | +4.155 |
Times | +4.021 |
Neg logit contributions | |
Magical | -5.235 |
adesh | -5.088 |
deen | -5.049 |
ower | -4.902 |
Piercing | -4.764 |
Token | agreement |
Token ID | 4381 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
upfront | +6.262 |
roc | +6.006 |
Concord | +5.748 |
Albany | +5.657 |
opoly | +5.648 |
Neg logit contributions | |
adesh | -9.055 |
Magical | -9.017 |
deen | -8.963 |
Piercing | -8.779 |
ower | -8.440 |
Token | to |
Token ID | 284 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
roc | +2.216 |
opoly | +2.185 |
Concord | +2.069 |
upfront | +2.066 |
Times | +2.025 |
Neg logit contributions | |
Magical | -2.462 |
adesh | -2.402 |
deen | -2.364 |
ower | -2.336 |
Piercing | -2.253 |
Token | arm |
Token ID | 3211 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
roc | +2.609 |
opoly | +2.581 |
Concord | +2.403 |
upfront | +2.383 |
Times | +2.336 |
Neg logit contributions | |
adesh | -3.367 |
Magical | -3.356 |
deen | -3.257 |
ower | -3.180 |
Piercing | -3.137 |
Token | rest |
Token ID | 2118 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
roc | +3.976 |
opoly | +3.911 |
upfront | +3.604 |
Concord | +3.592 |
Times | +3.585 |
Neg logit contributions | |
adesh | -4.578 |
Magical | -4.541 |
deen | -4.389 |
ower | -4.263 |
Piercing | -4.177 |
Token | . |
Token ID | 13 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
roc | +4.435 |
opoly | +4.308 |
Concord | +4.003 |
upfront | +3.997 |
Times | +3.957 |
Neg logit contributions | |
adesh | -5.229 |
Magical | -5.196 |
deen | -5.004 |
ower | -4.866 |
Piercing | -4.764 |
Token | A |
Token ID | 317 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
Magical | +0.034 |
adesh | +0.034 |
deen | +0.033 |
ower | +0.033 |
Piercing | +0.031 |
Neg logit contributions | |
roc | -0.033 |
opoly | -0.033 |
Times | -0.031 |
Concord | -0.031 |
upfront | -0.030 |
Token | restraining |
Token ID | 35078 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
roc | +3.001 |
opoly | +2.956 |
Concord | +2.763 |
Times | +2.718 |
upfront | +2.716 |
Neg logit contributions | |
Magical | -3.179 |
adesh | -3.150 |
deen | -3.052 |
ower | -3.036 |
Piercing | -2.894 |
Token | seat |
Token ID | 5852 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
roc | +7.692 |
opoly | +7.605 |
Concord | +7.470 |
upfront | +7.242 |
actu | +7.087 |
Neg logit contributions | |
adesh | -9.137 |
deen | -8.985 |
Magical | -8.664 |
ower | -8.511 |
Piercing | -8.243 |
Token | kept |
Token ID | 4030 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
roc | +10.873 |
upfront | +10.260 |
opoly | +10.104 |
Concord | +9.918 |
actu | +9.759 |
Neg logit contributions | |
adesh | -14.482 |
Magical | -14.151 |
deen | -14.141 |
田 | -13.649 |
Piercing | -13.507 |
Token | the |
Token ID | 262 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
roc | +5.413 |
opoly | +5.174 |
Concord | +4.913 |
upfront | +4.818 |
ouched | +4.653 |
Neg logit contributions | |
adesh | -7.208 |
Magical | -7.129 |
deen | -7.038 |
ower | -6.896 |
Piercing | -6.690 |
Token | monkey |
Token ID | 21657 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
roc | +3.065 |
opoly | +2.970 |
Concord | +2.812 |
upfront | +2.783 |
Times | +2.753 |
Neg logit contributions | |
Magical | -3.391 |
adesh | -3.371 |
deen | -3.309 |
ower | -3.246 |
Piercing | -3.154 |
Token | from |
Token ID | 422 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
roc | +9.198 |
upfront | +9.166 |
opoly | +8.532 |
ouched | +8.330 |
Concord | +8.274 |
Neg logit contributions | |
Magical | -13.330 |
adesh | -13.184 |
Piercing | -12.825 |
deen | -12.796 |
田 | -12.233 |
Token | leaving |
Token ID | 4305 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
roc | +5.326 |
opoly | +5.085 |
upfront | +4.928 |
Concord | +4.820 |
Albany | +4.665 |
Neg logit contributions | |
Magical | -6.086 |
adesh | -5.964 |
deen | -5.854 |
ower | -5.736 |
Piercing | -5.520 |
Token | finding |
Token ID | 4917 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
roc | +2.880 |
opoly | +2.850 |
Concord | +2.660 |
upfront | +2.612 |
Albany | +2.610 |
Neg logit contributions | |
Magical | -4.216 |
adesh | -4.196 |
deen | -4.138 |
ower | -4.064 |
Piercing | -3.854 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
roc | +8.367 |
Albany | +8.052 |
Concord | +7.972 |
upfront | +7.824 |
opoly | +7.782 |
Neg logit contributions | |
adesh | -10.722 |
deen | -10.294 |
Magical | -9.984 |
ower | -9.823 |
Piercing | -9.277 |
Token | back |
Token ID | 736 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
roc | +1.516 |
opoly | +1.497 |
Concord | +1.404 |
Times | +1.390 |
upfront | +1.381 |
Neg logit contributions | |
Magical | -1.656 |
adesh | -1.650 |
deen | -1.593 |
ower | -1.579 |
Piercing | -1.505 |
Token | of |
Token ID | 286 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
roc | +4.911 |
opoly | +4.843 |
ouched | +4.480 |
upfront | +4.443 |
Times | +4.412 |
Neg logit contributions | |
Magical | -6.547 |
adesh | -6.514 |
deen | -6.229 |
ower | -6.056 |
Piercing | -6.011 |
Token | the |
Token ID | 262 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
roc | +5.739 |
opoly | +5.466 |
Concord | +5.432 |
Albany | +5.361 |
upfront | +5.124 |
Neg logit contributions | |
adesh | -8.715 |
Magical | -8.433 |
deen | -8.264 |
ower | -8.257 |
Piercing | -7.709 |
Token | net |
Token ID | 2010 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
roc | +3.583 |
opoly | +3.491 |
Concord | +3.364 |
Albany | +3.262 |
upfront | +3.259 |
Neg logit contributions | |
adesh | -4.625 |
Magical | -4.607 |
deen | -4.424 |
ower | -4.354 |
Piercing | -4.178 |
Token | . |
Token ID | 13 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
upfront | +11.407 |
roc | +11.155 |
Albany | +10.539 |
Concord | +10.245 |
ouched | +9.969 |
Neg logit contributions | |
adesh | -14.463 |
Magical | -13.703 |
deen | -13.552 |
田 | -13.166 |
ower | -12.987 |
Token | In |
Token ID | 554 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
Concord | +5.896 |
Albany | +5.895 |
opoly | +5.882 |
roc | +5.882 |
Sagan | +5.638 |
Neg logit contributions | |
adesh | -7.301 |
deen | -7.209 |
ower | -6.823 |
Magical | -6.717 |
shed | -6.582 |
Token | addition |
Token ID | 3090 |
Feature activation | -9.28e-01 |
Pos logit contributions | |
roc | +6.212 |
opoly | +6.143 |
Concord | +5.837 |
Albany | +5.739 |
upfront | +5.709 |
Neg logit contributions | |
adesh | -7.842 |
Magical | -7.684 |
deen | -7.684 |
ower | -7.379 |
Piercing | -7.120 |
Token | to |
Token ID | 284 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
roc | +7.296 |
opoly | +7.161 |
Concord | +6.969 |
Albany | +6.926 |
upfront | +6.850 |
Neg logit contributions | |
adesh | -10.781 |
deen | -10.701 |
Magical | -10.432 |
ower | -9.938 |
Piercing | -9.729 |
Token | the |
Token ID | 262 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
Albany | +9.268 |
roc | +9.121 |
Concord | +8.924 |
upfront | +8.871 |
Sagan | +8.732 |
Neg logit contributions | |
adesh | -11.037 |
deen | -10.833 |
Magical | -9.958 |
ower | -9.933 |
Pavilion | -9.314 |
Token | the |
Token ID | 262 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
roc | +7.005 |
Concord | +6.753 |
upfront | +6.475 |
opoly | +6.423 |
ustain | +6.368 |
Neg logit contributions | |
adesh | -10.147 |
deen | -9.865 |
Magical | -9.848 |
ower | -9.585 |
Piercing | -9.477 |
Token | target |
Token ID | 2496 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Concord | +8.916 |
roc | +8.807 |
upfront | +8.538 |
Sagan | +8.271 |
opoly | +8.267 |
Neg logit contributions | |
adesh | -11.898 |
田 | -11.310 |
deen | -11.310 |
ower | -11.294 |
Magical | -11.275 |
Token | eyed |
Token ID | 45320 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
roc | +9.053 |
upfront | +8.707 |
Concord | +8.364 |
opoly | +8.291 |
Sagan | +8.133 |
Neg logit contributions | |
Magical | -12.285 |
adesh | -12.100 |
Piercing | -12.035 |
ower | -11.895 |
deen | -11.787 |
Token | at |
Token ID | 379 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
roc | +5.781 |
opoly | +5.603 |
Concord | +5.327 |
upfront | +5.268 |
Times | +5.159 |
Neg logit contributions | |
Magical | -6.475 |
adesh | -6.407 |
deen | -6.218 |
Piercing | -6.026 |
ower | -6.005 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
roc | +7.322 |
Concord | +6.793 |
opoly | +6.616 |
upfront | +6.462 |
Sagan | +6.412 |
Neg logit contributions | |
adesh | -9.071 |
Magical | -9.067 |
deen | -8.735 |
ower | -8.568 |
Piercing | -8.381 |
Token | Paris |
Token ID | 6342 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
roc | +9.969 |
Concord | +9.947 |
upfront | +9.908 |
Albany | +9.423 |
Sagan | +9.366 |
Neg logit contributions | |
adesh | -12.767 |
Magical | -12.286 |
田 | -11.898 |
deen | -11.825 |
Piercing | -11.815 |
Token | meeting |
Token ID | 3249 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
roc | +10.730 |
Concord | +10.122 |
upfront | +10.036 |
Times | +9.861 |
Gl | +9.804 |
Neg logit contributions | |
adesh | -14.430 |
Magical | -13.848 |
Piercing | -13.713 |
ower | -13.428 |
deen | -13.351 |
Token | , |
Token ID | 11 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
roc | +4.264 |
opoly | +4.102 |
upfront | +3.958 |
Concord | +3.901 |
Times | +3.830 |
Neg logit contributions | |
Magical | -4.927 |
adesh | -4.804 |
deen | -4.716 |
ower | -4.644 |
Piercing | -4.550 |
Token | Fab |
Token ID | 14236 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
roc | +5.778 |
opoly | +5.627 |
Concord | +5.524 |
upfront | +5.316 |
Sagan | +5.222 |
Neg logit contributions | |
adesh | -6.925 |
Magical | -6.919 |
deen | -6.862 |
ower | -6.566 |
Piercing | -6.374 |
Token | ius |
Token ID | 3754 |
Feature activation | -5.74e-01 |
Pos logit contributions | |
opoly | +4.123 |
roc | +4.116 |
Times | +3.281 |
hello | +3.203 |
gets | +3.169 |
Neg logit contributions | |
Magical | -10.754 |
deen | -10.317 |
adesh | -10.312 |
Piercing | -10.242 |
Pavilion | -10.077 |
Token | said |
Token ID | 531 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
roc | +3.887 |
opoly | +3.835 |
upfront | +3.560 |
Concord | +3.531 |
Times | +3.376 |
Neg logit contributions | |
Magical | -7.867 |
adesh | -7.746 |
deen | -7.562 |
Piercing | -7.346 |
ower | -7.246 |
Token | While |
Token ID | 3633 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
roc | +2.594 |
opoly | +2.525 |
Times | +2.466 |
Bee | +2.353 |
TEXT | +2.352 |
Neg logit contributions | |
Magical | -3.747 |
adesh | -3.649 |
deen | -3.558 |
ower | -3.485 |
Pavilion | -3.457 |
Token | Russia |
Token ID | 3284 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
roc | +4.922 |
opoly | +4.739 |
Concord | +4.636 |
upfront | +4.553 |
Albany | +4.494 |
Neg logit contributions | |
adesh | -6.362 |
Magical | -6.320 |
deen | -6.187 |
ower | -5.951 |
Piercing | -5.879 |
Token | retains |
Token ID | 27452 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
roc | +2.009 |
opoly | +2.006 |
Concord | +1.844 |
Times | +1.841 |
upfront | +1.834 |
Neg logit contributions | |
Magical | -2.270 |
adesh | -2.198 |
deen | -2.136 |
ower | -2.127 |
Piercing | -2.054 |
Token | substantial |
Token ID | 8904 |
Feature activation | -6.94e-02 |
Pos logit contributions | |
roc | +4.369 |
opoly | +4.283 |
Concord | +4.066 |
upfront | +4.011 |
Albany | +3.875 |
Neg logit contributions | |
adesh | -5.383 |
Magical | -5.339 |
deen | -5.247 |
ower | -5.109 |
Piercing | -4.972 |
Token | forces |
Token ID | 3386 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
roc | +0.664 |
opoly | +0.661 |
Concord | +0.611 |
Times | +0.606 |
upfront | +0.602 |
Neg logit contributions | |
Magical | -0.810 |
adesh | -0.796 |
deen | -0.776 |
ower | -0.771 |
Piercing | -0.740 |
Token | on |
Token ID | 319 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
roc | +3.392 |
opoly | +3.357 |
upfront | +3.147 |
Concord | +3.105 |
ouched | +3.030 |
Neg logit contributions | |
Magical | -3.974 |
adesh | -3.909 |
deen | -3.776 |
ower | -3.722 |
Piercing | -3.644 |
Token | Ukraine |
Token ID | 7049 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
roc | +10.000 |
opoly | +9.476 |
Concord | +9.295 |
ground | +9.198 |
Homs | +8.785 |
Neg logit contributions | |
田 | -16.002 |
adesh | -15.831 |
deen | -15.449 |
Magical | -15.219 |
Piercing | -15.055 |
Token | � |
Token ID | 447 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
roc | +5.485 |
opoly | +5.252 |
upfront | +4.967 |
Concord | +4.830 |
Msg | +4.759 |
Neg logit contributions | |
Magical | -6.659 |
adesh | -6.608 |
deen | -6.399 |
Piercing | -6.229 |
ower | -6.198 |
Token | � |
Token ID | 247 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
roc | +1.328 |
opoly | +1.319 |
Times | +1.206 |
upfront | +1.194 |
Concord | +1.192 |
Neg logit contributions | |
Magical | -1.952 |
adesh | -1.918 |
deen | -1.867 |
ower | -1.854 |
Piercing | -1.798 |
Token | s |
Token ID | 82 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
roc | +3.494 |
opoly | +3.480 |
Times | +3.107 |
Concord | +3.072 |
upfront | +3.067 |
Neg logit contributions | |
Magical | -6.093 |
adesh | -5.939 |
deen | -5.825 |
ower | -5.774 |
Piercing | -5.655 |
Token | eastern |
Token ID | 10183 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
roc | +6.503 |
opoly | +6.090 |
Concord | +6.061 |
upfront | +5.970 |
Times | +5.685 |
Neg logit contributions | |
adesh | -7.835 |
Magical | -7.672 |
deen | -7.583 |
Piercing | -7.325 |
ower | -7.313 |
Token | interceptions |
Token ID | 28891 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
roc | +3.149 |
opoly | +3.131 |
Concord | +2.941 |
upfront | +2.923 |
Times | +2.891 |
Neg logit contributions | |
Magical | -3.257 |
adesh | -3.236 |
deen | -3.199 |
ower | -3.124 |
Pavilion | -2.948 |
Token | . |
Token ID | 13 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
roc | +2.389 |
opoly | +2.377 |
upfront | +2.227 |
Concord | +2.212 |
Times | +2.175 |
Neg logit contributions | |
Magical | -2.555 |
adesh | -2.532 |
deen | -2.462 |
ower | -2.422 |
Piercing | -2.343 |
Token | In |
Token ID | 554 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
roc | +4.538 |
opoly | +4.533 |
Concord | +4.501 |
Times | +4.364 |
Albany | +4.331 |
Neg logit contributions | |
adesh | -5.846 |
deen | -5.682 |
Magical | -5.543 |
ower | -5.473 |
shed | -5.298 |
Token | his |
Token ID | 465 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
opoly | +4.635 |
roc | +4.623 |
Concord | +4.375 |
upfront | +4.260 |
Albany | +4.217 |
Neg logit contributions | |
adesh | -5.856 |
Magical | -5.796 |
deen | -5.671 |
ower | -5.509 |
Piercing | -5.295 |
Token | past |
Token ID | 1613 |
Feature activation | -9.92e-01 |
Pos logit contributions | |
roc | +6.459 |
opoly | +6.377 |
upfront | +6.239 |
Concord | +6.121 |
Times | +6.032 |
Neg logit contributions | |
deen | -8.358 |
adesh | -8.308 |
Magical | -8.106 |
ower | -7.973 |
田 | -7.559 |
Token | two |
Token ID | 734 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
roc | +5.944 |
upfront | +5.741 |
Concord | +5.586 |
opoly | +5.482 |
Albany | +5.387 |
Neg logit contributions | |
adesh | -13.101 |
deen | -12.972 |
ower | -12.873 |
Magical | -12.704 |
田 | -12.398 |
Token | meetings |
Token ID | 8292 |
Feature activation | -8.56e-01 |
Pos logit contributions | |
upfront | +9.887 |
roc | +9.465 |
Concord | +9.388 |
Albany | +9.386 |
opoly | +9.116 |
Neg logit contributions | |
adesh | -15.243 |
deen | -15.086 |
ower | -14.590 |
田 | -14.527 |
Magical | -14.090 |
Token | with |
Token ID | 351 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
roc | +6.980 |
opoly | +6.945 |
upfront | +6.804 |
Concord | +6.751 |
Albany | +6.648 |
Neg logit contributions | |
adesh | -9.702 |
Magical | -9.559 |
deen | -9.434 |
ower | -9.305 |
Piercing | -9.047 |
Token | Ron |
Token ID | 6575 |
Feature activation | -6.35e-02 |
Pos logit contributions | |
roc | +5.300 |
Concord | +5.154 |
opoly | +5.145 |
Albany | +4.984 |
upfront | +4.833 |
Neg logit contributions | |
adesh | -9.025 |
deen | -8.568 |
Magical | -8.546 |
ower | -8.478 |
Piercing | -8.095 |
Token | Rivera |
Token ID | 31325 |
Feature activation | -6.33e-01 |
Pos logit contributions | |
roc | +0.477 |
opoly | +0.475 |
Concord | +0.422 |
Times | +0.420 |
upfront | +0.410 |
Neg logit contributions | |
Magical | -0.875 |
adesh | -0.865 |
deen | -0.839 |
ower | -0.835 |
Piercing | -0.809 |
Token | 's |
Token ID | 338 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
opoly | +5.515 |
roc | +5.401 |
Concord | +5.268 |
upfront | +5.251 |
Albany | +5.057 |
Neg logit contributions | |
adesh | -7.366 |
Magical | -7.331 |
deen | -7.182 |
ower | -6.926 |
Piercing | -6.790 |
Token | a |
Token ID | 257 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
roc | +5.836 |
opoly | +5.815 |
upfront | +5.700 |
Concord | +5.405 |
Times | +5.377 |
Neg logit contributions | |
deen | -7.938 |
adesh | -7.919 |
ower | -7.807 |
Magical | -7.707 |
Piercing | -7.432 |
Token | requirement |
Token ID | 9079 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
opoly | +4.527 |
roc | +4.480 |
upfront | +4.224 |
Concord | +4.184 |
Times | +4.161 |
Neg logit contributions | |
adesh | -5.118 |
deen | -5.070 |
Magical | -5.063 |
ower | -4.994 |
Piercing | -4.756 |
Token | to |
Token ID | 284 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
upfront | +6.638 |
roc | +6.573 |
opoly | +6.376 |
Concord | +5.999 |
Times | +5.920 |
Neg logit contributions | |
adesh | -8.810 |
Magical | -8.529 |
deen | -8.452 |
ower | -8.111 |
Piercing | -7.946 |
Token | solve |
Token ID | 8494 |
Feature activation | -9.01e-01 |
Pos logit contributions | |
roc | +6.610 |
opoly | +6.443 |
Concord | +6.368 |
upfront | +6.291 |
Albany | +6.144 |
Neg logit contributions | |
adesh | -7.166 |
Magical | -6.980 |
ower | -6.814 |
deen | -6.786 |
Piercing | -6.493 |
Token | the |
Token ID | 262 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
roc | +7.108 |
opoly | +6.985 |
Concord | +6.914 |
upfront | +6.880 |
Sagan | +6.621 |
Neg logit contributions | |
deen | -10.407 |
adesh | -10.368 |
Magical | -10.013 |
ower | -9.838 |
Piercing | -9.561 |
Token | tough |
Token ID | 5802 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
roc | +7.332 |
opoly | +7.257 |
Concord | +7.177 |
upfront | +7.090 |
Albany | +6.813 |
Neg logit contributions | |
deen | -9.639 |
adesh | -9.632 |
ower | -9.253 |
Magical | -9.245 |
mage | -8.868 |
Token | questions |
Token ID | 2683 |
Feature activation | -6.43e-01 |
Pos logit contributions | |
upfront | +11.023 |
opoly | +10.816 |
roc | +10.659 |
Concord | +10.385 |
macro | +9.975 |
Neg logit contributions | |
deen | -14.586 |
adesh | -13.599 |
Magical | -13.578 |
ower | -13.251 |
Piercing | -12.934 |
Token | , |
Token ID | 11 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
opoly | +5.492 |
roc | +5.451 |
upfront | +5.443 |
Concord | +5.154 |
Sagan | +5.087 |
Neg logit contributions | |
adesh | -7.455 |
Magical | -7.286 |
deen | -7.282 |
ower | -7.025 |
Piercing | -6.797 |
Token | � |
Token ID | 447 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
roc | +8.720 |
opoly | +8.518 |
Sagan | +8.110 |
Concord | +8.088 |
Gl | +7.997 |
Neg logit contributions | |
adesh | -6.755 |
Magical | -6.690 |
deen | -6.659 |
ower | -6.547 |
Piercing | -6.189 |
Token | � |
Token ID | 251 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
roc | +3.462 |
opoly | +3.439 |
Concord | +3.233 |
Times | +3.227 |
upfront | +3.199 |
Neg logit contributions | |
Magical | -3.045 |
adesh | -2.982 |
deen | -2.923 |
ower | -2.876 |
Piercing | -2.745 |
Token | said |
Token ID | 531 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
roc | +4.147 |
opoly | +4.063 |
Concord | +3.857 |
Sagan | +3.765 |
upfront | +3.739 |
Neg logit contributions | |
adesh | -5.669 |
Magical | -5.594 |
deen | -5.510 |
ower | -5.467 |
Piercing | -5.269 |
Token | , |
Token ID | 11 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
Magical | +1.073 |
adesh | +1.033 |
ower | +1.002 |
deen | +0.999 |
Piercing | +0.936 |
Neg logit contributions | |
roc | -1.547 |
opoly | -1.536 |
Times | -1.452 |
Concord | -1.444 |
upfront | -1.422 |
Token | lightly |
Token ID | 15376 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
roc | +0.194 |
opoly | +0.193 |
Concord | +0.180 |
Times | +0.180 |
upfront | +0.177 |
Neg logit contributions | |
Magical | -0.184 |
adesh | -0.179 |
deen | -0.174 |
ower | -0.174 |
Piercing | -0.165 |
Token | a |
Token ID | 257 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
roc | +0.277 |
opoly | +0.275 |
Concord | +0.256 |
Times | +0.255 |
upfront | +0.252 |
Neg logit contributions | |
Magical | -0.278 |
adesh | -0.272 |
deen | -0.264 |
ower | -0.263 |
Piercing | -0.251 |
Token | lder |
Token ID | 6499 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
roc | +4.054 |
opoly | +3.984 |
Concord | +3.642 |
hello | +3.622 |
ouched | +3.572 |
Neg logit contributions | |
Magical | -5.214 |
adesh | -5.117 |
deen | -5.023 |
ower | -4.876 |
Piercing | -4.842 |
Token | smoked |
Token ID | 21603 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
Magical | +3.680 |
adesh | +3.454 |
ower | +3.436 |
deen | +3.410 |
shed | +3.309 |
Neg logit contributions | |
opoly | -4.369 |
roc | -4.293 |
Times | -4.110 |
Concord | -3.954 |
TEXT | -3.951 |
Token | mal |
Token ID | 6428 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
roc | +3.654 |
opoly | +3.578 |
upfront | +3.416 |
Concord | +3.411 |
Times | +3.351 |
Neg logit contributions | |
Magical | -3.663 |
adesh | -3.638 |
ower | -3.500 |
deen | -3.492 |
Piercing | -3.345 |
Token | ts |
Token ID | 912 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
roc | +6.319 |
opoly | +6.273 |
hello | +5.282 |
gets | +5.232 |
ouched | +5.222 |
Neg logit contributions | |
Magical | -19.640 |
Piercing | -19.255 |
Pavilion | -18.679 |
田 | -18.322 |
adesh | -18.004 |
Token | , |
Token ID | 11 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
Magical | +0.559 |
adesh | +0.543 |
deen | +0.528 |
ower | +0.527 |
Piercing | +0.500 |
Neg logit contributions | |
roc | -0.619 |
opoly | -0.617 |
Times | -0.575 |
Concord | -0.573 |
upfront | -0.563 |
Token | probably |
Token ID | 2192 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
roc | +0.285 |
opoly | +0.284 |
Concord | +0.264 |
Times | +0.263 |
upfront | +0.259 |
Neg logit contributions | |
Magical | -0.294 |
adesh | -0.288 |
deen | -0.279 |
ower | -0.279 |
Piercing | -0.265 |
Token | of |
Token ID | 286 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
Magical | +0.624 |
adesh | +0.605 |
ower | +0.589 |
deen | +0.588 |
Piercing | +0.561 |
Neg logit contributions | |
roc | -0.615 |
opoly | -0.613 |
Times | -0.568 |
Concord | -0.565 |
upfront | -0.556 |
Token | relatively |
Token ID | 5365 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
Magical | +3.374 |
adesh | +3.197 |
ower | +3.163 |
deen | +3.127 |
shed | +3.053 |
Neg logit contributions | |
opoly | -3.232 |
roc | -3.206 |
Times | -3.006 |
TEXT | -2.930 |
Msg | -2.913 |
Token | coordinator |
Token ID | 16052 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
Magical | +2.363 |
adesh | +2.252 |
deen | +2.216 |
ower | +2.216 |
shed | +2.121 |
Neg logit contributions | |
opoly | -2.187 |
roc | -2.175 |
Times | -2.012 |
Concord | -1.993 |
TEXT | -1.973 |
Token | Dennis |
Token ID | 16902 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
roc | +1.853 |
opoly | +1.849 |
Concord | +1.736 |
Times | +1.688 |
upfront | +1.686 |
Neg logit contributions | |
adesh | -2.123 |
Magical | -2.123 |
deen | -2.043 |
ower | -2.031 |
Piercing | -1.942 |
Token | Allen |
Token ID | 9659 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
roc | +5.478 |
Concord | +5.377 |
opoly | +5.302 |
Sagan | +5.248 |
Schmidt | +5.170 |
Neg logit contributions | |
adesh | -5.111 |
Magical | -4.772 |
ower | -4.753 |
deen | -4.735 |
Pavilion | -4.452 |
Token | the |
Token ID | 262 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
roc | +2.321 |
opoly | +2.321 |
upfront | +2.175 |
Concord | +2.166 |
Times | +2.126 |
Neg logit contributions | |
Magical | -2.556 |
adesh | -2.516 |
deen | -2.452 |
ower | -2.405 |
Piercing | -2.352 |
Token | past |
Token ID | 1613 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
roc | +7.147 |
opoly | +6.984 |
upfront | +6.977 |
Concord | +6.860 |
Times | +6.748 |
Neg logit contributions | |
adesh | -6.999 |
Magical | -6.734 |
deen | -6.682 |
ower | -6.408 |
Piercing | -6.302 |
Token | three |
Token ID | 1115 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
roc | +5.578 |
upfront | +5.227 |
Concord | +5.214 |
opoly | +5.172 |
Times | +4.994 |
Neg logit contributions | |
adesh | -12.856 |
deen | -12.838 |
Magical | -12.740 |
ower | -12.405 |
Piercing | -12.101 |
Token | games |
Token ID | 1830 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
upfront | +10.711 |
roc | +10.598 |
Concord | +10.553 |
Times | +10.337 |
opoly | +10.272 |
Neg logit contributions | |
deen | -14.330 |
adesh | -14.069 |
Magical | -13.540 |
Piercing | -13.207 |
田 | -13.082 |
Token | than |
Token ID | 621 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
roc | +3.534 |
opoly | +3.514 |
upfront | +3.339 |
Concord | +3.289 |
Times | +3.204 |
Neg logit contributions | |
Magical | -4.018 |
adesh | -3.956 |
deen | -3.867 |
ower | -3.798 |
Piercing | -3.705 |
Token | at |
Token ID | 379 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
Magical | +0.832 |
adesh | +0.807 |
deen | +0.785 |
ower | +0.785 |
Piercing | +0.747 |
Neg logit contributions | |
roc | -0.776 |
opoly | -0.773 |
Times | -0.714 |
Concord | -0.707 |
upfront | -0.697 |
Token | any |
Token ID | 597 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
roc | +7.355 |
Concord | +7.004 |
opoly | +6.888 |
Albany | +6.713 |
upfront | +6.585 |
Neg logit contributions | |
adesh | -7.807 |
deen | -7.459 |
Magical | -7.431 |
ower | -7.168 |
Piercing | -6.890 |
Token | point |
Token ID | 966 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
roc | +3.843 |
opoly | +3.794 |
Concord | +3.650 |
Times | +3.648 |
upfront | +3.640 |
Neg logit contributions | |
adesh | -3.658 |
Magical | -3.596 |
deen | -3.513 |
ower | -3.421 |
Piercing | -3.256 |
Token | Congress |
Token ID | 3162 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
opoly | +3.475 |
roc | +3.442 |
Concord | +3.228 |
Times | +3.178 |
upfront | +3.141 |
Neg logit contributions | |
Magical | -4.460 |
adesh | -4.438 |
deen | -4.369 |
ower | -4.257 |
Piercing | -4.099 |
Token | 2019 |
Token ID | 13130 |
Feature activation | +9.55e-02 |
Pos logit contributions | |
Magical | +3.532 |
adesh | +3.375 |
ower | +3.330 |
deen | +3.257 |
Pavilion | +3.163 |
Neg logit contributions | |
roc | -3.509 |
opoly | -3.506 |
Times | -3.254 |
Concord | -3.219 |
ouched | -3.189 |
Token | with |
Token ID | 351 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Magical | +0.999 |
adesh | +0.969 |
ower | +0.945 |
deen | +0.939 |
Piercing | +0.894 |
Neg logit contributions | |
roc | -1.035 |
opoly | -1.031 |
Times | -0.958 |
Concord | -0.953 |
upfront | -0.933 |
Token | a |
Token ID | 257 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Magical | +2.219 |
adesh | +2.139 |
ower | +2.086 |
deen | +2.053 |
shed | +2.010 |
Neg logit contributions | |
roc | -2.117 |
opoly | -2.115 |
Times | -1.959 |
Concord | -1.939 |
Bee | -1.899 |
Token | two |
Token ID | 734 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
Magical | +1.797 |
adesh | +1.728 |
ower | +1.679 |
deen | +1.669 |
shed | +1.612 |
Neg logit contributions | |
roc | -1.901 |
opoly | -1.888 |
Times | -1.751 |
Concord | -1.742 |
hello | -1.710 |
Token | - |
Token ID | 12 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Magical | +5.921 |
ower | +5.720 |
adesh | +5.652 |
Pavilion | +5.422 |
shed | +5.320 |
Neg logit contributions | |
roc | -9.917 |
opoly | -9.831 |
Concord | -9.341 |
hello | -9.262 |
ussian | -9.216 |
Token | day |
Token ID | 820 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
ower | +5.963 |
Magical | +5.717 |
adesh | +5.696 |
deen | +5.651 |
ing | +5.567 |
Neg logit contributions | |
opoly | -13.377 |
roc | -13.168 |
Concord | -13.079 |
Homs | -12.704 |
Msg | -12.555 |
Token | Executive |
Token ID | 10390 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
Magical | +1.256 |
adesh | +1.215 |
ower | +1.183 |
deen | +1.182 |
Piercing | +1.131 |
Neg logit contributions | |
roc | -1.172 |
opoly | -1.167 |
Times | -1.077 |
Concord | -1.074 |
hello | -1.047 |
Token | 5 |
Token ID | 642 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
Magical | +2.471 |
adesh | +2.321 |
ower | +2.303 |
deen | +2.277 |
Piercing | +2.229 |
Neg logit contributions | |
opoly | -2.676 |
roc | -2.671 |
Times | -2.468 |
Concord | -2.462 |
hello | -2.428 |
Token | G |
Token ID | 38 |
Feature activation | +4.59e-01 |
Pos logit contributions | |
roc | +2.030 |
opoly | +2.009 |
Times | +1.840 |
Concord | +1.796 |
upfront | +1.761 |
Neg logit contributions | |
Magical | -3.904 |
adesh | -3.880 |
deen | -3.795 |
ower | -3.722 |
Piercing | -3.637 |
Token | Panel |
Token ID | 18810 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
Magical | +4.637 |
adesh | +4.358 |
ower | +4.357 |
Pavilion | +4.243 |
deen | +4.219 |
Neg logit contributions | |
roc | -4.956 |
opoly | -4.941 |
Times | -4.572 |
ouched | -4.511 |
hello | -4.480 |
Token | _ |
Token ID | 62 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
roc | +2.130 |
opoly | +2.101 |
Times | +1.955 |
Concord | +1.932 |
hello | +1.922 |
Neg logit contributions | |
Magical | -2.095 |
adesh | -2.034 |
deen | -1.959 |
ower | -1.953 |
Piercing | -1.900 |
Token | ler |
Token ID | 1754 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
roc | +2.316 |
opoly | +2.281 |
Times | +2.117 |
hello | +2.092 |
Concord | +2.070 |
Neg logit contributions | |
Magical | -2.294 |
adesh | -2.216 |
deen | -2.127 |
ower | -2.119 |
Piercing | -2.081 |
Token | oy |
Token ID | 726 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
roc | +0.712 |
opoly | +0.710 |
Concord | +0.655 |
Times | +0.654 |
upfront | +0.643 |
Neg logit contributions | |
Magical | -0.715 |
adesh | -0.695 |
deen | -0.673 |
ower | -0.672 |
Piercing | -0.646 |
Token | - |
Token ID | 532 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
Magical | +4.595 |
ower | +4.449 |
adesh | +4.387 |
deen | +4.367 |
ing | +4.149 |
Neg logit contributions | |
opoly | -5.753 |
roc | -5.704 |
Concord | -5.509 |
Times | -5.457 |
upfront | -5.408 |
Token | favorite |
Token ID | 4004 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Magical | +2.564 |
shed | +1.936 |
Pavilion | +1.929 |
Rated | +1.898 |
ower | +1.842 |
Neg logit contributions | |
opoly | -13.478 |
roc | -13.100 |
TEXT | -12.711 |
Times | -12.582 |
ustain | -12.574 |
Token | favorite |
Token ID | 4004 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Magical | +3.715 |
shed | +3.093 |
Pavilion | +2.919 |
ing | +2.839 |
ower | +2.794 |
Neg logit contributions | |
opoly | -17.166 |
ailability | -16.709 |
roc | -16.368 |
ustain | -15.969 |
imity | -15.921 |
Token | favorite |
Token ID | 4004 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Magical | +3.880 |
shed | +3.227 |
Pavilion | +3.057 |
ing | +2.998 |
ower | +2.914 |
Neg logit contributions | |
opoly | -17.391 |
ailability | -16.893 |
roc | -16.580 |
ustain | -16.184 |
imity | -16.106 |
Token | favorite |
Token ID | 4004 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
Magical | +3.779 |
shed | +3.023 |
Pavilion | +2.961 |
ower | +2.913 |
ing | +2.846 |
Neg logit contributions | |
opoly | -16.485 |
roc | -15.836 |
ailability | -15.654 |
ustain | -15.373 |
imity | -15.339 |
Token | - |
Token ID | 532 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
Magical | +3.008 |
ower | +2.666 |
adesh | +2.409 |
shed | +2.390 |
ing | +2.334 |
Neg logit contributions | |
opoly | -12.539 |
roc | -12.318 |
Times | -11.880 |
ustain | -11.757 |
Concord | -11.740 |
Token | January |
Token ID | 3269 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
roc | +0.388 |
opoly | +0.385 |
Concord | +0.358 |
Times | +0.357 |
upfront | +0.351 |
Neg logit contributions | |
Magical | -0.453 |
adesh | -0.446 |
deen | -0.433 |
ower | -0.432 |
Piercing | -0.411 |
Token | 4 |
Token ID | 604 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
roc | +2.355 |
opoly | +2.302 |
Concord | +2.212 |
Times | +2.162 |
upfront | +2.135 |
Neg logit contributions | |
adesh | -3.426 |
Magical | -3.375 |
deen | -3.307 |
ower | -3.307 |
shed | -3.134 |
Token | and |
Token ID | 290 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
Magical | +4.214 |
adesh | +3.979 |
ower | +3.960 |
deen | +3.853 |
shed | +3.745 |
Neg logit contributions | |
roc | -4.647 |
opoly | -4.602 |
Times | -4.199 |
Msg | -4.192 |
Concord | -4.191 |
Token | Labor |
Token ID | 7882 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
Magical | +2.334 |
adesh | +2.236 |
ower | +2.180 |
deen | +2.143 |
Piercing | +2.092 |
Neg logit contributions | |
roc | -2.420 |
opoly | -2.400 |
Times | -2.202 |
hello | -2.181 |
ouched | -2.173 |
Token | Turn |
Token ID | 6756 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
Magical | +4.345 |
adesh | +4.214 |
ower | +4.118 |
deen | +4.062 |
Piercing | +3.959 |
Neg logit contributions | |
roc | -3.800 |
opoly | -3.765 |
hello | -3.465 |
upfront | -3.444 |
Concord | -3.400 |
Token | over |
Token ID | 2502 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Magical | +3.193 |
ower | +3.085 |
adesh | +3.024 |
ing | +2.826 |
deen | +2.652 |
Neg logit contributions | |
opoly | -7.939 |
roc | -7.919 |
upfront | -7.658 |
Concord | -7.607 |
hello | -7.569 |
Token | Survey |
Token ID | 13084 |
Feature activation | +8.94e-01 |
Pos logit contributions | |
Magical | +9.294 |
adesh | +8.898 |
ower | +8.737 |
ing | +8.530 |
shed | +8.303 |
Neg logit contributions | |
roc | -10.735 |
opoly | -10.692 |
hello | -10.160 |
ouched | -9.901 |
etition | -9.747 |
Token | data |
Token ID | 1366 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Magical | +7.915 |
ower | +7.309 |
adesh | +7.299 |
ing | +7.143 |
shed | +6.980 |
Neg logit contributions | |
opoly | -9.899 |
roc | -9.773 |
hello | -9.101 |
ouched | -9.042 |
Bee | -9.031 |
Token | , |
Token ID | 11 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
Magical | +9.933 |
shed | +9.427 |
adesh | +9.160 |
ower | +9.109 |
deen | +9.093 |
Neg logit contributions | |
roc | -13.118 |
opoly | -12.941 |
Msg | -12.518 |
hello | -12.250 |
TEXT | -12.182 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
Magical | +3.816 |
adesh | +3.531 |
ower | +3.469 |
deen | +3.398 |
shed | +3.375 |
Neg logit contributions | |
roc | -4.119 |
opoly | -4.102 |
Times | -3.758 |
Msg | -3.752 |
hello | -3.732 |
Token | – |
Token ID | 1906 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
roc | +0.712 |
opoly | +0.708 |
Concord | +0.663 |
Times | +0.662 |
upfront | +0.654 |
Neg logit contributions | |
Magical | -0.591 |
adesh | -0.576 |
deen | -0.558 |
ower | -0.556 |
Piercing | -0.527 |
Token | 2016 |
Token ID | 5304 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
roc | +0.565 |
opoly | +0.558 |
Times | +0.517 |
Concord | +0.512 |
upfront | +0.503 |
Neg logit contributions | |
Magical | -0.791 |
adesh | -0.777 |
deen | -0.757 |
ower | -0.753 |
Piercing | -0.724 |
Token | ). |
Token ID | 737 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
Magical | +4.404 |
adesh | +4.180 |
ower | +4.042 |
deen | +4.013 |
shed | +3.850 |
Neg logit contributions | |
opoly | -7.409 |
roc | -7.337 |
Msg | -6.948 |
hello | -6.903 |
Times | -6.858 |
Token | cycling |
Token ID | 16259 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
Magical | +3.447 |
adesh | +3.245 |
ower | +3.138 |
deen | +3.133 |
shed | +3.074 |
Neg logit contributions | |
opoly | -3.722 |
roc | -3.702 |
Times | -3.481 |
Msg | -3.424 |
hello | -3.414 |
Token | specialist |
Token ID | 15670 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
Magical | +5.518 |
adesh | +5.184 |
ower | +5.077 |
shed | +5.057 |
Pavilion | +4.953 |
Neg logit contributions | |
opoly | -5.602 |
roc | -5.560 |
TEXT | -5.267 |
Times | -5.206 |
hello | -5.197 |
Token | , |
Token ID | 11 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
Magical | +3.661 |
adesh | +3.410 |
ower | +3.406 |
deen | +3.307 |
shed | +3.267 |
Neg logit contributions | |
opoly | -4.062 |
roc | -4.054 |
Times | -3.752 |
TEXT | -3.738 |
ouched | -3.730 |
Token | offers |
Token ID | 4394 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
Magical | +4.842 |
adesh | +4.591 |
ower | +4.528 |
shed | +4.399 |
deen | +4.372 |
Neg logit contributions | |
roc | -5.306 |
opoly | -5.268 |
TEXT | -4.923 |
Times | -4.900 |
ouched | -4.894 |
Token | a |
Token ID | 257 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
Magical | +5.347 |
adesh | +4.963 |
ower | +4.786 |
shed | +4.735 |
deen | +4.731 |
Neg logit contributions | |
opoly | -5.141 |
roc | -5.074 |
Times | -4.684 |
TEXT | -4.626 |
hello | -4.608 |
Token | Highlights |
Token ID | 30776 |
Feature activation | +9.72e-01 |
Pos logit contributions | |
Magical | +2.586 |
adesh | +2.467 |
deen | +2.388 |
ower | +2.382 |
shed | +2.321 |
Neg logit contributions | |
roc | -2.341 |
opoly | -2.338 |
Times | -2.151 |
Concord | -2.114 |
Msg | -2.101 |
Token | of |
Token ID | 286 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
Magical | +10.218 |
Pavilion | +9.311 |
ower | +9.250 |
adesh | +9.098 |
ing | +9.010 |
Neg logit contributions | |
roc | -8.994 |
opoly | -8.920 |
TEXT | -8.403 |
Msg | -8.241 |
Times | -8.225 |
Token | T |
Token ID | 309 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
Magical | +7.041 |
adesh | +6.474 |
deen | +6.184 |
ower | +6.181 |
Pavilion | +6.083 |
Neg logit contributions | |
opoly | -6.740 |
roc | -6.707 |
TEXT | -6.310 |
Msg | -6.226 |
Times | -6.163 |
Token | ener |
Token ID | 877 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
roc | +0.480 |
opoly | +0.476 |
Concord | +0.441 |
Times | +0.440 |
upfront | +0.432 |
Neg logit contributions | |
Magical | -0.513 |
adesh | -0.500 |
deen | -0.487 |
ower | -0.483 |
Piercing | -0.464 |
Token | ife |
Token ID | 901 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
opoly | +1.903 |
roc | +1.902 |
Concord | +1.816 |
Times | +1.808 |
upfront | +1.792 |
Neg logit contributions | |
Magical | -0.440 |
adesh | -0.407 |
deen | -0.371 |
ower | -0.366 |
Piercing | -0.329 |
Token | walking |
Token ID | 6155 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
Magical | +7.066 |
adesh | +6.430 |
Pavilion | +6.375 |
ower | +6.263 |
deen | +6.206 |
Neg logit contributions | |
roc | -6.951 |
opoly | -6.901 |
TEXT | -6.348 |
ouched | -6.318 |
Times | -6.316 |
Token | ipp |
Token ID | 3974 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
Magical | +0.373 |
adesh | +0.364 |
ower | +0.351 |
deen | +0.351 |
Piercing | +0.330 |
Neg logit contributions | |
roc | -0.496 |
opoly | -0.494 |
Concord | -0.464 |
Times | -0.463 |
upfront | -0.457 |
Token | ew |
Token ID | 413 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
roc | +1.005 |
opoly | +0.998 |
Times | +0.902 |
Concord | +0.900 |
upfront | +0.881 |
Neg logit contributions | |
Magical | -1.523 |
adesh | -1.472 |
deen | -1.441 |
ower | -1.428 |
Piercing | -1.401 |
Token | as |
Token ID | 292 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
Magical | +2.398 |
adesh | +2.392 |
deen | +2.302 |
ower | +2.297 |
shed | +2.164 |
Neg logit contributions | |
roc | -2.022 |
opoly | -2.005 |
Concord | -1.867 |
Times | -1.859 |
upfront | -1.837 |
Token | regard |
Token ID | 2754 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
Magical | +0.482 |
adesh | +0.471 |
deen | +0.459 |
ower | +0.459 |
Piercing | +0.439 |
Neg logit contributions | |
roc | -0.360 |
opoly | -0.358 |
Times | -0.328 |
Concord | -0.327 |
upfront | -0.321 |
Token | as |
Token ID | 355 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
roc | +1.069 |
opoly | +1.056 |
Concord | +0.977 |
Times | +0.971 |
upfront | +0.961 |
Neg logit contributions | |
Magical | -1.232 |
adesh | -1.212 |
ower | -1.171 |
deen | -1.169 |
Piercing | -1.118 |
Token | their |
Token ID | 511 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
Magical | +3.113 |
adesh | +2.954 |
ower | +2.905 |
deen | +2.905 |
shed | +2.790 |
Neg logit contributions | |
roc | -3.196 |
opoly | -3.175 |
Times | -3.022 |
hello | -2.934 |
TEXT | -2.929 |
Token | traditional |
Token ID | 4569 |
Feature activation | +9.70e-01 |
Pos logit contributions | |
Magical | +9.224 |
shed | +8.307 |
ing | +8.099 |
adesh | +8.060 |
ower | +8.045 |
Neg logit contributions | |
opoly | -10.200 |
roc | -10.038 |
Times | -9.907 |
hello | -9.624 |
Msg | -9.365 |
Token | territory |
Token ID | 7674 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
Magical | +9.699 |
shed | +8.931 |
ing | +8.788 |
adesh | +8.680 |
ower | +8.583 |
Neg logit contributions | |
opoly | -9.693 |
roc | -9.310 |
Times | -9.009 |
Msg | -8.795 |
ouched | -8.770 |
Token | . |
Token ID | 13 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
roc | +0.975 |
opoly | +0.968 |
Concord | +0.901 |
Times | +0.893 |
upfront | +0.890 |
Neg logit contributions | |
Magical | -0.975 |
adesh | -0.955 |
deen | -0.924 |
ower | -0.920 |
Piercing | -0.877 |
Token | They |
Token ID | 1119 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
Magical | +2.764 |
adesh | +2.610 |
ower | +2.561 |
deen | +2.509 |
Piercing | +2.489 |
Neg logit contributions | |
roc | -2.999 |
opoly | -2.971 |
Times | -2.752 |
ouched | -2.716 |
upfront | -2.708 |
Token | had |
Token ID | 550 |
Feature activation | +6.36e-01 |
Pos logit contributions | |
Magical | +3.165 |
adesh | +3.113 |
ower | +3.056 |
deen | +3.020 |
shed | +2.969 |
Neg logit contributions | |
roc | -3.258 |
opoly | -3.240 |
Times | -3.055 |
TEXT | -2.982 |
Concord | -2.967 |
Token | that |
Token ID | 326 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
opoly | +1.559 |
roc | +1.556 |
Concord | +1.447 |
Times | +1.440 |
upfront | +1.421 |
Neg logit contributions | |
Magical | -1.592 |
adesh | -1.568 |
deen | -1.508 |
ower | -1.505 |
Piercing | -1.440 |
Token | included |
Token ID | 3017 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
Magical | +2.912 |
adesh | +2.813 |
ower | +2.757 |
deen | +2.728 |
shed | +2.710 |
Neg logit contributions | |
opoly | -2.684 |
roc | -2.680 |
Times | -2.459 |
Concord | -2.436 |
ouched | -2.407 |
Token | the |
Token ID | 262 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
roc | +0.735 |
opoly | +0.727 |
Concord | +0.678 |
Times | +0.672 |
upfront | +0.663 |
Neg logit contributions | |
Magical | -0.855 |
adesh | -0.839 |
deen | -0.818 |
ower | -0.814 |
Piercing | -0.778 |
Token | club |
Token ID | 3430 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Magical | +3.312 |
adesh | +3.094 |
ower | +3.011 |
deen | +2.993 |
shed | +2.953 |
Neg logit contributions | |
opoly | -3.593 |
roc | -3.564 |
Times | -3.323 |
TEXT | -3.295 |
Msg | -3.291 |
Token | � |
Token ID | 447 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
roc | +1.479 |
opoly | +1.469 |
Concord | +1.364 |
Times | +1.357 |
upfront | +1.345 |
Neg logit contributions | |
Magical | -1.539 |
adesh | -1.493 |
deen | -1.468 |
ower | -1.450 |
Piercing | -1.387 |
Token | � |
Token ID | 247 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
Magical | +1.344 |
adesh | +1.306 |
deen | +1.275 |
ower | +1.275 |
Piercing | +1.210 |
Neg logit contributions | |
roc | -1.283 |
opoly | -1.279 |
Concord | -1.182 |
Times | -1.177 |
upfront | -1.151 |
Token | s |
Token ID | 82 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
Magical | +4.147 |
adesh | +4.076 |
ower | +4.005 |
deen | +3.962 |
ing | +3.876 |
Neg logit contributions | |
opoly | -3.551 |
roc | -3.548 |
Concord | -3.235 |
Times | -3.201 |
upfront | -3.197 |
Token | third |
Token ID | 2368 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
Magical | +6.059 |
adesh | +5.692 |
ower | +5.500 |
shed | +5.499 |
deen | +5.350 |
Neg logit contributions | |
opoly | -6.568 |
roc | -6.443 |
TEXT | -6.162 |
Times | -6.074 |
hello | -6.017 |
Token | Stanley |
Token ID | 14167 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
Magical | +4.206 |
adesh | +4.041 |
ower | +3.908 |
deen | +3.826 |
Piercing | +3.769 |
Neg logit contributions | |
opoly | -4.799 |
roc | -4.737 |
Times | -4.509 |
TEXT | -4.411 |
ouched | -4.408 |
Token | Cup |
Token ID | 5454 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
roc | +1.355 |
opoly | +1.340 |
Concord | +1.270 |
Times | +1.252 |
upfront | +1.244 |
Neg logit contributions | |
Magical | -1.006 |
adesh | -0.984 |
deen | -0.958 |
ower | -0.945 |
Piercing | -0.888 |
Token | championship |
Token ID | 12184 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
roc | +1.855 |
opoly | +1.851 |
Concord | +1.720 |
upfront | +1.668 |
Times | +1.661 |
Neg logit contributions | |
Magical | -2.383 |
adesh | -2.337 |
deen | -2.304 |
ower | -2.267 |
Piercing | -2.173 |
Token | Dan |
Token ID | 6035 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
roc | +2.111 |
opoly | +2.107 |
Concord | +1.981 |
Times | +1.938 |
upfront | +1.937 |
Neg logit contributions | |
Magical | -2.305 |
adesh | -2.294 |
deen | -2.230 |
ower | -2.216 |
Piercing | -2.111 |
Token | Fry |
Token ID | 25305 |
Feature activation | +7.74e-01 |
Pos logit contributions | |
Magical | +1.926 |
adesh | +1.875 |
deen | +1.826 |
ower | +1.826 |
Piercing | +1.742 |
Neg logit contributions | |
roc | -1.696 |
opoly | -1.682 |
Times | -1.573 |
Concord | -1.556 |
upfront | -1.539 |
Token | d |
Token ID | 67 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
Magical | +7.330 |
deen | +7.172 |
adesh | +7.148 |
ower | +6.862 |
ing | +6.772 |
Neg logit contributions | |
roc | -8.101 |
opoly | -8.064 |
Times | -7.671 |
TEXT | -7.629 |
ouched | -7.627 |
Token | man |
Token ID | 805 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
Magical | +4.287 |
ower | +4.208 |
adesh | +4.166 |
deen | +4.110 |
shed | +3.897 |
Neg logit contributions | |
roc | -4.466 |
opoly | -4.412 |
Times | -4.241 |
Concord | -4.210 |
TEXT | -4.205 |
Token | of |
Token ID | 286 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
Magical | +5.997 |
ower | +5.753 |
adesh | +5.638 |
deen | +5.573 |
ing | +5.378 |
Neg logit contributions | |
roc | -7.931 |
opoly | -7.661 |
ouched | -7.330 |
Times | -7.329 |
TEXT | -7.313 |
Token | web |
Token ID | 3992 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
opoly | +4.758 |
roc | +4.735 |
Concord | +4.579 |
upfront | +4.379 |
Times | +4.376 |
Neg logit contributions | |
adesh | -4.859 |
Magical | -4.706 |
deen | -4.705 |
ower | -4.627 |
Piercing | -4.295 |
Token | firm |
Token ID | 4081 |
Feature activation | -6.33e-01 |
Pos logit contributions | |
Magical | +3.741 |
adesh | +3.651 |
ower | +3.616 |
deen | +3.564 |
shed | +3.396 |
Neg logit contributions | |
roc | -3.793 |
opoly | -3.747 |
Concord | -3.566 |
Times | -3.509 |
TEXT | -3.449 |
Token | In |
Token ID | 554 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
roc | +6.560 |
opoly | +6.515 |
Concord | +6.393 |
upfront | +6.118 |
Albany | +6.088 |
Neg logit contributions | |
adesh | -6.279 |
deen | -6.088 |
Magical | -5.909 |
ower | -5.856 |
田 | -5.442 |
Token | igo |
Token ID | 14031 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
roc | +5.569 |
opoly | +5.564 |
Times | +5.176 |
Bee | +5.137 |
Concord | +5.085 |
Neg logit contributions | |
Magical | -5.031 |
adesh | -4.934 |
deen | -4.748 |
Piercing | -4.610 |
ower | -4.582 |
Token | . |
Token ID | 13 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
Magical | +0.629 |
adesh | +0.609 |
ower | +0.594 |
deen | +0.592 |
Piercing | +0.563 |
Neg logit contributions | |
roc | -0.657 |
opoly | -0.653 |
Concord | -0.607 |
Times | -0.606 |
upfront | -0.597 |
Token | A |
Token ID | 317 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
roc | +0.178 |
opoly | +0.178 |
Times | +0.164 |
Concord | +0.164 |
upfront | +0.161 |
Neg logit contributions | |
Magical | -0.189 |
adesh | -0.184 |
deen | -0.179 |
ower | -0.179 |
Piercing | -0.171 |
Token | US |
Token ID | 1294 |
Feature activation | +8.22e-01 |
Pos logit contributions | |
Magical | +1.586 |
adesh | +1.524 |
ower | +1.487 |
deen | +1.473 |
shed | +1.415 |
Neg logit contributions | |
opoly | -1.574 |
roc | -1.561 |
Times | -1.448 |
Concord | -1.438 |
hello | -1.424 |
Token | security |
Token ID | 2324 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
Magical | +7.300 |
ower | +6.736 |
adesh | +6.675 |
shed | +6.383 |
deen | +6.367 |
Neg logit contributions | |
opoly | -9.165 |
roc | -8.983 |
hello | -8.795 |
Msg | -8.525 |
ouched | -8.524 |
Token | services |
Token ID | 2594 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
Magical | +2.819 |
adesh | +2.705 |
ower | +2.661 |
deen | +2.603 |
shed | +2.543 |
Neg logit contributions | |
opoly | -3.186 |
roc | -3.181 |
Times | -2.942 |
Concord | -2.930 |
TEXT | -2.900 |
Token | . |
Token ID | 13 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
Magical | +3.904 |
ower | +3.662 |
adesh | +3.656 |
deen | +3.623 |
shed | +3.526 |
Neg logit contributions | |
opoly | -4.508 |
roc | -4.396 |
Times | -4.134 |
Concord | -4.090 |
hello | -4.078 |
Token |
|
Token ID | 198 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
Magical | +0.850 |
adesh | +0.820 |
ower | +0.797 |
deen | +0.795 |
Piercing | +0.762 |
Neg logit contributions | |
roc | -0.864 |
opoly | -0.862 |
Times | -0.795 |
Concord | -0.793 |
upfront | -0.785 |
Token |
|
Token ID | 198 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
roc | +2.673 |
opoly | +2.576 |
Times | +2.461 |
upfront | +2.427 |
Concord | +2.406 |
Neg logit contributions | |
adesh | -3.345 |
Magical | -3.343 |
deen | -3.254 |
ower | -3.169 |
Piercing | -3.077 |
Token | They |
Token ID | 2990 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
Magical | +4.196 |
adesh | +4.128 |
ower | +4.091 |
deen | +4.024 |
ing | +3.851 |
Neg logit contributions | |
opoly | -3.406 |
roc | -3.350 |
Concord | -3.225 |
upfront | -3.083 |
Sagan | -3.058 |
Token | also |
Token ID | 635 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
Magical | +4.011 |
adesh | +3.925 |
ower | +3.837 |
deen | +3.832 |
shed | +3.824 |
Neg logit contributions | |
roc | -3.924 |
opoly | -3.918 |
Times | -3.618 |
hello | -3.602 |
Concord | -3.587 |
Token | indicated |
Token ID | 8203 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
Magical | +3.954 |
adesh | +3.862 |
shed | +3.861 |
ower | +3.741 |
deen | +3.698 |
Neg logit contributions | |
opoly | -4.793 |
roc | -4.739 |
Times | -4.417 |
hello | -4.409 |
Bee | -4.384 |
Token | that |
Token ID | 326 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
Magical | +2.260 |
adesh | +2.170 |
ower | +2.136 |
deen | +2.105 |
shed | +2.040 |
Neg logit contributions | |
opoly | -2.277 |
roc | -2.237 |
Times | -2.079 |
Concord | -2.070 |
hello | -2.053 |
Token | the |
Token ID | 262 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
Magical | +2.327 |
adesh | +2.234 |
ower | +2.191 |
deen | +2.162 |
shed | +2.088 |
Neg logit contributions | |
opoly | -2.302 |
roc | -2.283 |
Times | -2.121 |
Concord | -2.090 |
hello | -2.080 |
Token | � |
Token ID | 447 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
Magical | +5.145 |
adesh | +4.915 |
ower | +4.856 |
deen | +4.833 |
ing | +4.660 |
Neg logit contributions | |
opoly | -4.353 |
roc | -4.258 |
Concord | -4.094 |
upfront | -3.892 |
Albany | -3.859 |
Token | � |
Token ID | 250 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
roc | +0.261 |
opoly | +0.260 |
Times | +0.243 |
Concord | +0.242 |
upfront | +0.239 |
Neg logit contributions | |
Magical | -0.231 |
adesh | -0.225 |
deen | -0.218 |
ower | -0.217 |
Piercing | -0.206 |
Token | I |
Token ID | 40 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
roc | +0.462 |
opoly | +0.458 |
Times | +0.423 |
Concord | +0.419 |
upfront | +0.412 |
Neg logit contributions | |
Magical | -0.620 |
adesh | -0.609 |
deen | -0.592 |
ower | -0.590 |
Piercing | -0.566 |
Token | knew |
Token ID | 2993 |
Feature activation | -8.50e-03 |
Pos logit contributions | |
Magical | +6.685 |
ower | +6.441 |
adesh | +6.386 |
deen | +6.375 |
shed | +6.271 |
Neg logit contributions | |
opoly | -7.479 |
roc | -7.127 |
Times | -7.042 |
Concord | -6.871 |
Bee | -6.845 |
Token | from |
Token ID | 422 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
roc | +0.085 |
opoly | +0.084 |
Concord | +0.078 |
Times | +0.078 |
upfront | +0.077 |
Neg logit contributions | |
Magical | -0.096 |
adesh | -0.094 |
deen | -0.091 |
ower | -0.091 |
Piercing | -0.087 |
Token | the |
Token ID | 262 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
roc | +1.700 |
opoly | +1.664 |
Concord | +1.570 |
Times | +1.558 |
upfront | +1.544 |
Neg logit contributions | |
Magical | -2.000 |
adesh | -1.975 |
deen | -1.912 |
ower | -1.912 |
Piercing | -1.820 |
Token | start |
Token ID | 923 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
Magical | +2.070 |
adesh | +2.002 |
ower | +1.967 |
deen | +1.955 |
Piercing | +1.885 |
Neg logit contributions | |
opoly | -1.576 |
roc | -1.567 |
Times | -1.419 |
Concord | -1.417 |
ustain | -1.396 |
Token | I |
Token ID | 314 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
roc | +0.328 |
opoly | +0.325 |
Concord | +0.301 |
Times | +0.301 |
upfront | +0.297 |
Neg logit contributions | |
Magical | -0.365 |
adesh | -0.357 |
deen | -0.346 |
ower | -0.345 |
Piercing | -0.330 |
Token | could |
Token ID | 714 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
Magical | +0.265 |
adesh | +0.258 |
deen | +0.251 |
ower | +0.251 |
Piercing | +0.239 |
Neg logit contributions | |
roc | -0.235 |
opoly | -0.234 |
Times | -0.216 |
Concord | -0.216 |
upfront | -0.212 |
Token | never |
Token ID | 1239 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
Magical | +5.126 |
ower | +4.801 |
adesh | +4.757 |
deen | +4.746 |
shed | +4.733 |
Neg logit contributions | |
opoly | -5.880 |
roc | -5.718 |
Times | -5.436 |
Concord | -5.388 |
upfront | -5.248 |
Token | become |
Token ID | 1716 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
Magical | +1.772 |
adesh | +1.693 |
ower | +1.666 |
deen | +1.663 |
shed | +1.606 |
Neg logit contributions | |
opoly | -1.913 |
roc | -1.900 |
Times | -1.778 |
Concord | -1.761 |
upfront | -1.720 |
Token | а |
Token ID | 16142 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
Magical | +3.358 |
adesh | +3.253 |
ower | +3.163 |
deen | +3.132 |
Piercing | +2.990 |
Neg logit contributions | |
opoly | -4.040 |
roc | -4.004 |
Times | -3.744 |
Concord | -3.741 |
upfront | -3.690 |
Token | с |
Token ID | 21727 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
Magical | +2.947 |
adesh | +2.784 |
ower | +2.718 |
deen | +2.694 |
Piercing | +2.559 |
Neg logit contributions | |
opoly | -4.194 |
roc | -4.156 |
Concord | -3.922 |
Times | -3.922 |
ouched | -3.867 |
Token | с |
Token ID | 21727 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
Magical | +2.827 |
adesh | +2.662 |
deen | +2.631 |
ower | +2.614 |
shed | +2.479 |
Neg logit contributions | |
opoly | -3.780 |
roc | -3.780 |
Times | -3.537 |
Concord | -3.530 |
upfront | -3.511 |
Token | е |
Token ID | 16843 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
Magical | +2.802 |
adesh | +2.653 |
deen | +2.606 |
ower | +2.602 |
shed | +2.468 |
Neg logit contributions | |
roc | -3.255 |
opoly | -3.248 |
Times | -3.030 |
Concord | -3.013 |
upfront | -3.005 |
Token | , |
Token ID | 11 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
Magical | +4.576 |
adesh | +4.291 |
deen | +4.274 |
ower | +4.158 |
ing | +4.014 |
Neg logit contributions | |
opoly | -6.638 |
roc | -6.495 |
Concord | -6.174 |
Times | -6.169 |
ouched | -6.153 |
Token | � |
Token ID | 12466 |
Feature activation | +6.19e-02 |
Pos logit contributions | |
roc | +0.674 |
opoly | +0.671 |
Concord | +0.624 |
Times | +0.623 |
upfront | +0.614 |
Neg logit contributions | |
Magical | -0.648 |
adesh | -0.633 |
deen | -0.615 |
ower | -0.613 |
Piercing | -0.582 |
Token | � |
Token ID | 121 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
Magical | +0.613 |
adesh | +0.596 |
deen | +0.578 |
ower | +0.577 |
Piercing | +0.547 |
Neg logit contributions | |
roc | -0.702 |
opoly | -0.700 |
Concord | -0.653 |
Times | -0.653 |
upfront | -0.642 |
Token | е |
Token ID | 16843 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
Magical | +0.956 |
adesh | +0.930 |
ower | +0.903 |
deen | +0.896 |
Piercing | +0.851 |
Neg logit contributions | |
roc | -1.178 |
opoly | -1.174 |
Concord | -1.098 |
Times | -1.098 |
upfront | -1.081 |
Token | к |
Token ID | 31583 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
Magical | +2.023 |
adesh | +1.940 |
deen | +1.888 |
ower | +1.884 |
Piercing | +1.802 |
Neg logit contributions | |
opoly | -2.457 |
roc | -2.451 |
Times | -2.288 |
Concord | -2.279 |
upfront | -2.242 |
Token | о |
Token ID | 15166 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
roc | +1.876 |
opoly | +1.858 |
Times | +1.742 |
Concord | +1.731 |
upfront | +1.708 |
Neg logit contributions | |
Magical | -1.704 |
adesh | -1.672 |
deen | -1.616 |
ower | -1.603 |
Piercing | -1.523 |
Token | т |
Token ID | 20375 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Magical | +3.295 |
adesh | +3.210 |
deen | +3.145 |
ower | +3.079 |
shed | +2.894 |
Neg logit contributions | |
opoly | -5.152 |
roc | -5.120 |
Concord | -4.807 |
Times | -4.790 |
hello | -4.730 |
Token | identify |
Token ID | 5911 |
Feature activation | -8.50e-03 |
Pos logit contributions | |
Magical | +0.730 |
adesh | +0.709 |
ower | +0.686 |
deen | +0.686 |
shed | +0.648 |
Neg logit contributions | |
roc | -0.890 |
opoly | -0.887 |
Concord | -0.828 |
Times | -0.828 |
upfront | -0.811 |
Token | in |
Token ID | 287 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
roc | +0.093 |
opoly | +0.093 |
Concord | +0.086 |
Times | +0.086 |
upfront | +0.085 |
Neg logit contributions | |
Magical | -0.088 |
adesh | -0.085 |
deen | -0.083 |
ower | -0.083 |
Piercing | -0.079 |
Token | all |
Token ID | 477 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
Magical | +3.571 |
adesh | +3.386 |
ower | +3.305 |
deen | +3.265 |
shed | +3.175 |
Neg logit contributions | |
opoly | -3.407 |
roc | -3.385 |
Times | -3.141 |
Concord | -3.130 |
Bee | -3.056 |
Token | of |
Token ID | 286 |
Feature activation | +8.72e-02 |
Pos logit contributions | |
Magical | +5.167 |
ower | +4.878 |
adesh | +4.866 |
shed | +4.687 |
deen | +4.672 |
Neg logit contributions | |
roc | -5.451 |
opoly | -5.406 |
Concord | -5.036 |
hello | -4.952 |
Msg | -4.927 |
Token | comics |
Token ID | 12770 |
Feature activation | +3.48e-03 |
Pos logit contributions | |
Magical | +0.928 |
adesh | +0.895 |
ower | +0.870 |
deen | +0.867 |
Piercing | +0.830 |
Neg logit contributions | |
roc | -0.933 |
opoly | -0.928 |
Concord | -0.860 |
Times | -0.860 |
upfront | -0.844 |
Token | . |
Token ID | 13 |
Feature activation | -6.28e-03 |
Pos logit contributions | |
Magical | +0.036 |
adesh | +0.035 |
deen | +0.034 |
ower | +0.034 |
Piercing | +0.032 |
Neg logit contributions | |
roc | -0.038 |
opoly | -0.038 |
Times | -0.035 |
Concord | -0.035 |
upfront | -0.035 |
Token | I |
Token ID | 314 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
roc | +0.066 |
opoly | +0.065 |
Concord | +0.061 |
Times | +0.061 |
upfront | +0.060 |
Neg logit contributions | |
Magical | -0.068 |
adesh | -0.066 |
deen | -0.064 |
ower | -0.064 |
Piercing | -0.061 |
Token | dropped |
Token ID | 5710 |
Feature activation | -6.35e-02 |
Pos logit contributions | |
roc | +0.342 |
opoly | +0.339 |
Concord | +0.316 |
Times | +0.316 |
upfront | +0.312 |
Neg logit contributions | |
Magical | -0.300 |
adesh | -0.293 |
deen | -0.283 |
ower | -0.282 |
Piercing | -0.268 |
Token | the |
Token ID | 262 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
roc | +0.623 |
opoly | +0.619 |
Concord | +0.572 |
Times | +0.569 |
upfront | +0.563 |
Neg logit contributions | |
Magical | -0.726 |
adesh | -0.713 |
deen | -0.696 |
ower | -0.689 |
Piercing | -0.659 |
Token | series |
Token ID | 2168 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
roc | +0.919 |
opoly | +0.911 |
Concord | +0.843 |
Times | +0.838 |
upfront | +0.831 |
Neg logit contributions | |
Magical | -1.070 |
adesh | -1.053 |
deen | -1.030 |
ower | -1.020 |
Piercing | -0.973 |
Token | within |
Token ID | 1626 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
roc | +3.362 |
opoly | +3.324 |
upfront | +3.169 |
Concord | +3.108 |
Times | +3.051 |
Neg logit contributions | |
Magical | -3.640 |
adesh | -3.636 |
deen | -3.562 |
ower | -3.474 |
Piercing | -3.299 |
Token | groups |
Token ID | 2628 |
Feature activation | -4.68e-03 |
Pos logit contributions | |
Magical | +0.019 |
adesh | +0.018 |
deen | +0.018 |
ower | +0.018 |
Piercing | +0.017 |
Neg logit contributions | |
roc | -0.017 |
opoly | -0.017 |
Times | -0.016 |
Concord | -0.016 |
upfront | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
roc | +0.049 |
opoly | +0.049 |
Times | +0.045 |
Concord | +0.045 |
upfront | +0.044 |
Neg logit contributions | |
Magical | -0.051 |
adesh | -0.050 |
deen | -0.048 |
ower | -0.048 |
Piercing | -0.046 |
Token | certain |
Token ID | 1728 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
roc | +1.042 |
opoly | +1.033 |
Concord | +0.956 |
Times | +0.952 |
upfront | +0.937 |
Neg logit contributions | |
Magical | -1.220 |
adesh | -1.202 |
deen | -1.174 |
ower | -1.158 |
Piercing | -1.116 |
Token | countries |
Token ID | 2678 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
Magical | +3.970 |
adesh | +3.756 |
ower | +3.708 |
shed | +3.535 |
deen | +3.506 |
Neg logit contributions | |
roc | -4.970 |
opoly | -4.914 |
hello | -4.599 |
Concord | -4.580 |
Msg | -4.542 |
Token | . |
Token ID | 13 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
roc | +0.748 |
opoly | +0.743 |
Concord | +0.689 |
Times | +0.688 |
upfront | +0.683 |
Neg logit contributions | |
Magical | -0.733 |
adesh | -0.716 |
deen | -0.695 |
ower | -0.691 |
Piercing | -0.661 |
Token | � |
Token ID | 447 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
Magical | +1.718 |
adesh | +1.643 |
ower | +1.595 |
deen | +1.586 |
Piercing | +1.538 |
Neg logit contributions | |
roc | -1.964 |
opoly | -1.961 |
Times | -1.805 |
Concord | -1.792 |
upfront | -1.782 |
Token | � |
Token ID | 251 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
roc | +1.032 |
opoly | +1.027 |
Times | +0.965 |
Concord | +0.963 |
upfront | +0.950 |
Neg logit contributions | |
Magical | -0.777 |
adesh | -0.755 |
deen | -0.732 |
ower | -0.728 |
Piercing | -0.687 |
Token | While |
Token ID | 2893 |
Feature activation | -2.41e-03 |
Pos logit contributions | |
Magical | +2.114 |
adesh | +2.012 |
ower | +1.956 |
deen | +1.945 |
Piercing | +1.886 |
Neg logit contributions | |
opoly | -2.437 |
roc | -2.433 |
Times | -2.230 |
ouched | -2.218 |
Concord | -2.213 |
Token | she |
Token ID | 673 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
roc | +0.024 |
opoly | +0.024 |
Concord | +0.022 |
Times | +0.022 |
upfront | +0.022 |
Neg logit contributions | |
Magical | -0.027 |
adesh | -0.026 |
deen | -0.026 |
ower | -0.026 |
Piercing | -0.025 |
Token | does |
Token ID | 857 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
roc | +1.739 |
opoly | +1.727 |
Concord | +1.606 |
Times | +1.602 |
upfront | +1.600 |
Neg logit contributions | |
Magical | -1.569 |
adesh | -1.499 |
deen | -1.469 |
ower | -1.451 |
Piercing | -1.401 |
Token | believe |
Token ID | 1975 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
roc | +1.740 |
opoly | +1.721 |
Concord | +1.609 |
upfront | +1.596 |
Times | +1.589 |
Neg logit contributions | |
Magical | -1.882 |
adesh | -1.832 |
deen | -1.780 |
ower | -1.763 |
Piercing | -1.711 |
Token | this |
Token ID | 428 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
Magical | +2.669 |
adesh | +2.561 |
ower | +2.524 |
deen | +2.479 |
Piercing | +2.377 |
Neg logit contributions | |
roc | -2.673 |
opoly | -2.653 |
Times | -2.470 |
Concord | -2.430 |
hello | -2.418 |
Token | far |
Token ID | 1290 |
Feature activation | -9.78e-02 |
Pos logit contributions | |
roc | +5.119 |
opoly | +5.062 |
upfront | +4.934 |
Concord | +4.928 |
Times | +4.736 |
Neg logit contributions | |
adesh | -6.214 |
Magical | -6.148 |
deen | -6.115 |
ower | -6.031 |
Piercing | -5.723 |
Token | , |
Token ID | 11 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
roc | +1.004 |
opoly | +0.999 |
Concord | +0.929 |
Times | +0.921 |
upfront | +0.918 |
Neg logit contributions | |
Magical | -1.078 |
adesh | -1.048 |
deen | -1.023 |
ower | -1.015 |
Piercing | -0.971 |
Token | I |
Token ID | 314 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
roc | +4.495 |
opoly | +4.494 |
upfront | +4.225 |
Concord | +4.205 |
hello | +4.052 |
Neg logit contributions | |
Magical | -5.966 |
adesh | -5.878 |
deen | -5.801 |
ower | -5.703 |
Pavilion | -5.454 |
Token | � |
Token ID | 447 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
roc | +1.446 |
opoly | +1.427 |
Concord | +1.337 |
Times | +1.334 |
upfront | +1.323 |
Neg logit contributions | |
Magical | -1.199 |
adesh | -1.167 |
deen | -1.124 |
ower | -1.113 |
Piercing | -1.069 |
Token | � |
Token ID | 247 |
Feature activation | -7.57e-02 |
Pos logit contributions | |
roc | +0.350 |
opoly | +0.346 |
Times | +0.318 |
Concord | +0.318 |
upfront | +0.312 |
Neg logit contributions | |
Magical | -0.466 |
adesh | -0.456 |
deen | -0.444 |
ower | -0.442 |
Piercing | -0.426 |
Token | d |
Token ID | 67 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
roc | +0.893 |
opoly | +0.884 |
Times | +0.829 |
Concord | +0.828 |
upfront | +0.818 |
Neg logit contributions | |
Magical | -0.719 |
adesh | -0.701 |
deen | -0.674 |
ower | -0.671 |
Piercing | -0.640 |
Token | love |
Token ID | 1842 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
roc | +1.440 |
opoly | +1.410 |
Concord | +1.309 |
upfront | +1.302 |
Times | +1.293 |
Neg logit contributions | |
Magical | -1.745 |
adesh | -1.718 |
deen | -1.652 |
ower | -1.641 |
Piercing | -1.592 |
Token | it |
Token ID | 340 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Magical | +0.231 |
adesh | +0.225 |
deen | +0.219 |
ower | +0.218 |
Piercing | +0.208 |
Neg logit contributions | |
roc | -0.209 |
opoly | -0.208 |
Times | -0.192 |
Concord | -0.192 |
upfront | -0.189 |
Token | if |
Token ID | 611 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
Magical | +1.032 |
adesh | +1.003 |
deen | +0.975 |
ower | +0.973 |
Piercing | +0.921 |
Neg logit contributions | |
roc | -1.169 |
opoly | -1.168 |
Times | -1.088 |
Concord | -1.088 |
upfront | -1.062 |
Token | you |
Token ID | 345 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
Magical | +0.314 |
adesh | +0.307 |
deen | +0.298 |
ower | +0.298 |
Piercing | +0.285 |
Neg logit contributions | |
roc | -0.263 |
opoly | -0.262 |
Times | -0.241 |
Concord | -0.241 |
upfront | -0.237 |
Token | � |
Token ID | 447 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
roc | +2.201 |
opoly | +2.186 |
upfront | +2.048 |
Concord | +2.040 |
Times | +2.028 |
Neg logit contributions | |
Magical | -2.149 |
adesh | -2.089 |
deen | -2.031 |
ower | -2.009 |
Piercing | -1.945 |
Token | � |
Token ID | 247 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
roc | +0.088 |
opoly | +0.087 |
Times | +0.081 |
Concord | +0.081 |
upfront | +0.080 |
Neg logit contributions | |
Magical | -0.087 |
adesh | -0.085 |
deen | -0.083 |
ower | -0.083 |
Piercing | -0.079 |
Token | self |
Token ID | 2116 |
Feature activation | +7.29e-02 |
Pos logit contributions | |
roc | +1.379 |
opoly | +1.377 |
Times | +1.276 |
Concord | +1.271 |
upfront | +1.262 |
Neg logit contributions | |
Magical | -1.451 |
adesh | -1.404 |
deen | -1.367 |
ower | -1.360 |
Piercing | -1.312 |
Token | - |
Token ID | 12 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Magical | +0.736 |
adesh | +0.716 |
ower | +0.696 |
deen | +0.694 |
Piercing | +0.657 |
Neg logit contributions | |
roc | -0.816 |
opoly | -0.811 |
Times | -0.755 |
Concord | -0.755 |
upfront | -0.740 |
Token | per |
Token ID | 525 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
roc | +0.141 |
opoly | +0.140 |
Times | +0.131 |
Concord | +0.131 |
upfront | +0.129 |
Neg logit contributions | |
Magical | -0.118 |
adesh | -0.115 |
deen | -0.111 |
ower | -0.111 |
Piercing | -0.105 |
Token | ception |
Token ID | 4516 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
roc | +2.292 |
opoly | +2.280 |
Concord | +2.069 |
Times | +2.052 |
upfront | +2.023 |
Neg logit contributions | |
Magical | -3.517 |
adesh | -3.475 |
ower | -3.351 |
deen | -3.344 |
Piercing | -3.263 |
Token | and |
Token ID | 290 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
roc | +1.586 |
opoly | +1.581 |
Times | +1.470 |
Concord | +1.465 |
upfront | +1.463 |
Neg logit contributions | |
Magical | -1.553 |
adesh | -1.513 |
deen | -1.471 |
ower | -1.456 |
Piercing | -1.401 |
Token | objective |
Token ID | 9432 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
upfront | +6.607 |
roc | +6.597 |
opoly | +6.547 |
Concord | +6.410 |
Times | +6.292 |
Neg logit contributions | |
deen | -8.125 |
adesh | -8.117 |
Magical | -8.105 |
ower | -7.798 |
Piercing | -7.685 |
Token | reality |
Token ID | 3950 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
roc | +6.191 |
opoly | +6.157 |
upfront | +6.076 |
Concord | +5.977 |
Times | +5.938 |
Neg logit contributions | |
deen | -7.688 |
Magical | -7.651 |
adesh | -7.597 |
ower | -7.329 |
Piercing | -7.249 |
Token | , |
Token ID | 11 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
opoly | +5.422 |
roc | +5.411 |
upfront | +5.194 |
Times | +5.030 |
Concord | +5.001 |
Neg logit contributions | |
Magical | -6.408 |
adesh | -6.271 |
deen | -6.196 |
ower | -6.068 |
Piercing | -5.867 |
Token | the |
Token ID | 262 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
roc | +5.856 |
opoly | +5.729 |
upfront | +5.569 |
Concord | +5.453 |
Sagan | +5.386 |
Neg logit contributions | |
Magical | -7.237 |
deen | -7.191 |
adesh | -7.099 |
ower | -6.880 |
mage | -6.693 |
Token | the |
Token ID | 262 |
Feature activation | +2.30e-03 |
Pos logit contributions | |
roc | +3.210 |
opoly | +3.178 |
upfront | +2.975 |
Concord | +2.969 |
Times | +2.950 |
Neg logit contributions | |
Magical | -3.626 |
adesh | -3.573 |
deen | -3.464 |
ower | -3.439 |
Piercing | -3.359 |
Token | age |
Token ID | 2479 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
Magical | +0.019 |
adesh | +0.018 |
deen | +0.018 |
ower | +0.017 |
Piercing | +0.016 |
Neg logit contributions | |
roc | -0.030 |
opoly | -0.030 |
Concord | -0.028 |
Times | -0.028 |
upfront | -0.028 |
Token | of |
Token ID | 286 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
roc | +2.934 |
opoly | +2.889 |
upfront | +2.751 |
Times | +2.715 |
Concord | +2.695 |
Neg logit contributions | |
Magical | -3.574 |
adesh | -3.565 |
deen | -3.481 |
ower | -3.397 |
Piercing | -3.303 |
Token | 18 |
Token ID | 1248 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
roc | +5.586 |
upfront | +5.439 |
opoly | +5.384 |
Albany | +5.143 |
Concord | +5.066 |
Neg logit contributions | |
adesh | -9.243 |
deen | -9.236 |
ower | -8.965 |
Magical | -8.911 |
Piercing | -8.724 |
Token | at |
Token ID | 379 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
roc | +3.648 |
opoly | +3.595 |
upfront | +3.461 |
Concord | +3.314 |
Times | +3.279 |
Neg logit contributions | |
Magical | -4.254 |
adesh | -4.179 |
deen | -4.074 |
ower | -4.011 |
Piercing | -3.907 |
Token | a |
Token ID | 257 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
roc | +4.426 |
opoly | +4.289 |
upfront | +4.124 |
Concord | +4.070 |
Times | +3.938 |
Neg logit contributions | |
adesh | -5.202 |
Magical | -5.149 |
deen | -5.083 |
ower | -4.992 |
Piercing | -4.789 |
Token | federal |
Token ID | 2717 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
roc | +2.652 |
opoly | +2.627 |
upfront | +2.464 |
Concord | +2.434 |
Times | +2.416 |
Neg logit contributions | |
Magical | -3.151 |
adesh | -3.128 |
deen | -3.047 |
ower | -3.034 |
Piercing | -2.905 |
Token | level |
Token ID | 1241 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
roc | +6.433 |
opoly | +6.369 |
upfront | +6.188 |
Concord | +5.892 |
Albany | +5.770 |
Neg logit contributions | |
Magical | -7.295 |
adesh | -7.288 |
deen | -7.110 |
ower | -6.934 |
Piercing | -6.807 |
Token | ( |
Token ID | 357 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
roc | +5.618 |
opoly | +5.583 |
upfront | +5.417 |
Concord | +5.063 |
Albany | +4.959 |
Neg logit contributions | |
adesh | -7.072 |
Magical | -6.993 |
deen | -6.814 |
ower | -6.651 |
Piercing | -6.580 |
Token | the |
Token ID | 1169 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
Magical | +1.189 |
adesh | +1.164 |
ower | +1.132 |
deen | +1.130 |
Piercing | +1.069 |
Neg logit contributions | |
roc | -1.096 |
opoly | -1.094 |
Concord | -1.015 |
Times | -1.005 |
upfront | -0.994 |
Token | proposal |
Token ID | 6961 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
roc | +0.274 |
opoly | +0.273 |
Concord | +0.252 |
Times | +0.252 |
upfront | +0.248 |
Neg logit contributions | |
Magical | -0.309 |
adesh | -0.303 |
deen | -0.294 |
ower | -0.294 |
Piercing | -0.281 |
Token |
|
Token ID | 198 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
roc | +2.866 |
opoly | +2.864 |
Concord | +2.716 |
Times | +2.699 |
upfront | +2.609 |
Neg logit contributions | |
adesh | -3.215 |
Magical | -3.183 |
deen | -3.170 |
ower | -3.098 |
shed | -2.952 |
Token |
|
Token ID | 198 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
Magical | +4.291 |
ower | +4.042 |
ing | +3.985 |
adesh | +3.942 |
deen | +3.802 |
Neg logit contributions | |
opoly | -6.788 |
ailability | -6.606 |
roc | -6.567 |
Concord | -6.301 |
unbeliev | -6.251 |
Token | In |
Token ID | 818 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
roc | +0.952 |
opoly | +0.946 |
Times | +0.883 |
Concord | +0.857 |
upfront | +0.848 |
Neg logit contributions | |
Magical | -1.232 |
adesh | -1.204 |
deen | -1.175 |
ower | -1.165 |
Piercing | -1.126 |
Token | order |
Token ID | 1502 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
roc | +2.621 |
opoly | +2.616 |
Concord | +2.463 |
Times | +2.398 |
upfront | +2.387 |
Neg logit contributions | |
Magical | -3.019 |
adesh | -2.994 |
deen | -2.924 |
ower | -2.857 |
Piercing | -2.763 |
Token | to |
Token ID | 284 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
roc | +0.194 |
opoly | +0.193 |
Times | +0.179 |
Concord | +0.179 |
upfront | +0.177 |
Neg logit contributions | |
Magical | -0.194 |
adesh | -0.189 |
deen | -0.184 |
ower | -0.183 |
Piercing | -0.175 |
Token | win |
Token ID | 1592 |
Feature activation | -9.70e-01 |
Pos logit contributions | |
opoly | +4.104 |
roc | +4.060 |
Concord | +3.806 |
upfront | +3.789 |
Times | +3.760 |
Neg logit contributions | |
Magical | -4.386 |
adesh | -4.333 |
deen | -4.176 |
ower | -4.129 |
Piercing | -3.967 |
Token | the |
Token ID | 262 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
opoly | +8.132 |
roc | +7.828 |
Concord | +7.534 |
Albany | +7.369 |
upfront | +7.321 |
Neg logit contributions | |
deen | -10.779 |
adesh | -10.667 |
Magical | -10.598 |
Piercing | -10.318 |
ower | -10.039 |
Token | White |
Token ID | 2635 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
roc | +3.220 |
opoly | +3.207 |
Concord | +3.030 |
upfront | +2.959 |
Times | +2.910 |
Neg logit contributions | |
adesh | -4.106 |
Magical | -4.060 |
deen | -4.009 |
ower | -3.939 |
Piercing | -3.805 |
Token | House |
Token ID | 2097 |
Feature activation | -7.01e-01 |
Pos logit contributions | |
opoly | +3.396 |
roc | +3.348 |
Concord | +3.187 |
Times | +3.140 |
upfront | +3.129 |
Neg logit contributions | |
adesh | -3.123 |
Magical | -3.038 |
deen | -2.992 |
ower | -2.928 |
shed | -2.774 |
Token | in |
Token ID | 287 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
opoly | +6.114 |
roc | +5.957 |
upfront | +5.743 |
Concord | +5.581 |
Albany | +5.348 |
Neg logit contributions | |
Magical | -8.070 |
adesh | -7.987 |
deen | -7.900 |
Piercing | -7.558 |
ower | -7.515 |
Token | November |
Token ID | 3389 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
opoly | +3.381 |
roc | +3.336 |
Concord | +3.182 |
Albany | +3.045 |
Times | +3.034 |
Neg logit contributions | |
adesh | -5.030 |
Magical | -5.001 |
deen | -4.941 |
ower | -4.798 |
Piercing | -4.666 |
Token | and |
Token ID | 290 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
roc | +3.106 |
opoly | +3.069 |
Concord | +2.869 |
upfront | +2.849 |
Times | +2.800 |
Neg logit contributions | |
Magical | -3.503 |
adesh | -3.467 |
deen | -3.387 |
ower | -3.312 |
Piercing | -3.201 |
Token | human |
Token ID | 1692 |
Feature activation | -5.60e-01 |
Pos logit contributions | |
roc | +3.531 |
opoly | +3.504 |
upfront | +3.267 |
Concord | +3.263 |
Times | +3.169 |
Neg logit contributions | |
Magical | -4.217 |
adesh | -4.152 |
deen | -4.060 |
ower | -3.954 |
Piercing | -3.867 |
Token | health |
Token ID | 1535 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
roc | +5.310 |
opoly | +5.257 |
upfront | +4.977 |
Concord | +4.834 |
ustain | +4.775 |
Neg logit contributions | |
Magical | -6.228 |
deen | -6.136 |
adesh | -6.081 |
Piercing | -5.966 |
ower | -5.918 |
Token | impacts |
Token ID | 12751 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
roc | +4.349 |
opoly | +4.300 |
upfront | +4.004 |
Concord | +3.954 |
Times | +3.829 |
Neg logit contributions | |
Magical | -5.981 |
adesh | -5.969 |
deen | -5.860 |
ower | -5.656 |
Piercing | -5.597 |
Token | that |
Token ID | 326 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
opoly | +5.140 |
roc | +5.105 |
upfront | +4.824 |
Concord | +4.640 |
Times | +4.629 |
Neg logit contributions | |
Magical | -6.502 |
adesh | -6.502 |
deen | -6.272 |
ower | -6.177 |
Piercing | -6.081 |
Token | may |
Token ID | 743 |
Feature activation | -9.48e-01 |
Pos logit contributions | |
roc | +3.233 |
opoly | +3.231 |
Concord | +3.037 |
upfront | +2.993 |
Times | +2.989 |
Neg logit contributions | |
Magical | -3.656 |
adesh | -3.602 |
deen | -3.526 |
ower | -3.486 |
Piercing | -3.342 |
Token | extend |
Token ID | 9117 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
opoly | +8.284 |
roc | +8.223 |
upfront | +8.185 |
ouched | +7.877 |
ustain | +7.624 |
Neg logit contributions | |
Magical | -9.955 |
adesh | -9.556 |
deen | -9.262 |
Piercing | -9.174 |
Pavilion | -8.992 |
Token | beyond |
Token ID | 3675 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
Magical | +1.736 |
adesh | +1.697 |
ower | +1.656 |
deen | +1.628 |
shed | +1.563 |
Neg logit contributions | |
opoly | -1.776 |
roc | -1.771 |
Times | -1.639 |
Concord | -1.632 |
hello | -1.609 |
Token | Iran |
Token ID | 4068 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
Magical | +4.165 |
adesh | +3.994 |
ower | +3.926 |
deen | +3.833 |
shed | +3.715 |
Neg logit contributions | |
opoly | -4.113 |
roc | -4.063 |
Times | -3.768 |
Concord | -3.737 |
hello | -3.731 |
Token | � |
Token ID | 447 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
roc | +0.440 |
opoly | +0.438 |
Times | +0.406 |
Concord | +0.406 |
upfront | +0.401 |
Neg logit contributions | |
Magical | -0.423 |
adesh | -0.412 |
deen | -0.401 |
ower | -0.399 |
Piercing | -0.381 |
Token | � |
Token ID | 247 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
Magical | +1.401 |
adesh | +1.367 |
ower | +1.327 |
deen | +1.326 |
Piercing | +1.259 |
Neg logit contributions | |
roc | -1.273 |
opoly | -1.270 |
Concord | -1.172 |
Times | -1.168 |
upfront | -1.138 |
Token | time |
Token ID | 640 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
roc | +3.858 |
opoly | +3.852 |
Times | +3.512 |
Concord | +3.510 |
upfront | +3.482 |
Neg logit contributions | |
Magical | -4.291 |
adesh | -4.171 |
deen | -4.110 |
ower | -4.054 |
Piercing | -3.851 |
Token | "). |
Token ID | 11074 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
opoly | +3.200 |
roc | +3.188 |
Times | +2.937 |
Concord | +2.933 |
upfront | +2.915 |
Neg logit contributions | |
Magical | -3.615 |
adesh | -3.488 |
deen | -3.450 |
ower | -3.406 |
Piercing | -3.271 |
Token | So |
Token ID | 1406 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
opoly | +3.154 |
roc | +3.152 |
Concord | +2.994 |
Times | +2.976 |
upfront | +2.886 |
Neg logit contributions | |
adesh | -3.613 |
Magical | -3.574 |
deen | -3.569 |
ower | -3.506 |
shed | -3.329 |
Token | regardless |
Token ID | 7692 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
roc | +2.865 |
opoly | +2.847 |
Concord | +2.625 |
Times | +2.606 |
upfront | +2.597 |
Neg logit contributions | |
Magical | -3.195 |
adesh | -3.137 |
deen | -3.105 |
ower | -3.004 |
Piercing | -2.911 |
Token | of |
Token ID | 286 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
roc | +0.857 |
opoly | +0.855 |
Concord | +0.792 |
Times | +0.790 |
upfront | +0.784 |
Neg logit contributions | |
Magical | -0.894 |
adesh | -0.880 |
deen | -0.855 |
ower | -0.848 |
Piercing | -0.811 |
Token | social |
Token ID | 1919 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
roc | +2.058 |
opoly | +2.036 |
Concord | +1.900 |
Times | +1.895 |
upfront | +1.891 |
Neg logit contributions | |
Magical | -2.420 |
adesh | -2.394 |
deen | -2.354 |
ower | -2.318 |
Piercing | -2.218 |
Token | time |
Token ID | 640 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
opoly | +8.853 |
roc | +8.639 |
Times | +8.569 |
Concord | +8.236 |
upfront | +8.175 |
Neg logit contributions | |
deen | -9.853 |
Magical | -9.648 |
adesh | -9.457 |
ower | -9.078 |
田 | -8.942 |
Token | mandates |
Token ID | 33568 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
Magical | +1.031 |
adesh | +1.005 |
ower | +0.975 |
deen | +0.973 |
shed | +0.928 |
Neg logit contributions | |
roc | -1.036 |
opoly | -1.030 |
Concord | -0.954 |
Times | -0.954 |
upfront | -0.934 |
Token | , |
Token ID | 11 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
opoly | +2.379 |
roc | +2.378 |
Concord | +2.204 |
upfront | +2.201 |
Times | +2.196 |
Neg logit contributions | |
Magical | -2.667 |
adesh | -2.627 |
deen | -2.569 |
ower | -2.538 |
Piercing | -2.429 |
Token | if |
Token ID | 611 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
opoly | +3.132 |
roc | +3.130 |
Concord | +2.928 |
upfront | +2.894 |
Times | +2.892 |
Neg logit contributions | |
Magical | -3.717 |
adesh | -3.691 |
deen | -3.660 |
ower | -3.583 |
Piercing | -3.418 |
Token | the |
Token ID | 262 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
roc | +3.666 |
opoly | +3.649 |
Concord | +3.413 |
Times | +3.373 |
upfront | +3.368 |
Neg logit contributions | |
Magical | -4.403 |
adesh | -4.342 |
deen | -4.318 |
ower | -4.226 |
Piercing | -4.081 |
Token | to |
Token ID | 1462 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
roc | +3.429 |
opoly | +3.369 |
Times | +3.189 |
hello | +3.158 |
Bee | +3.111 |
Neg logit contributions | |
Magical | -4.068 |
adesh | -3.922 |
deen | -3.865 |
ower | -3.764 |
Piercing | -3.697 |
Token | just |
Token ID | 655 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
roc | +4.541 |
opoly | +4.433 |
Concord | +4.263 |
Albany | +4.203 |
upfront | +4.123 |
Neg logit contributions | |
Magical | -5.079 |
adesh | -5.063 |
deen | -4.936 |
ower | -4.860 |
Piercing | -4.585 |
Token | six |
Token ID | 2237 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
roc | +1.594 |
opoly | +1.563 |
Concord | +1.463 |
upfront | +1.443 |
Times | +1.435 |
Neg logit contributions | |
Magical | -2.087 |
adesh | -2.051 |
deen | -2.003 |
ower | -1.994 |
Piercing | -1.901 |
Token | from |
Token ID | 422 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
roc | +4.806 |
opoly | +4.692 |
upfront | +4.467 |
Concord | +4.461 |
Times | +4.428 |
Neg logit contributions | |
Magical | -5.037 |
adesh | -5.000 |
deen | -4.917 |
ower | -4.773 |
Pavilion | -4.593 |
Token | Columbus |
Token ID | 14939 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
roc | +6.216 |
Albany | +6.000 |
Concord | +5.989 |
opoly | +5.967 |
upfront | +5.688 |
Neg logit contributions | |
adesh | -7.412 |
deen | -7.287 |
Magical | -7.206 |
ower | -7.054 |
Piercing | -6.617 |
Token | ), |
Token ID | 828 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
roc | +1.724 |
opoly | +1.710 |
Concord | +1.598 |
Times | +1.594 |
upfront | +1.573 |
Neg logit contributions | |
Magical | -1.656 |
adesh | -1.639 |
deen | -1.586 |
ower | -1.564 |
Piercing | -1.496 |
Token | three |
Token ID | 1115 |
Feature activation | -5.74e-01 |
Pos logit contributions | |
roc | +7.870 |
Albany | +7.749 |
upfront | +7.698 |
Concord | +7.691 |
opoly | +7.622 |
Neg logit contributions | |
adesh | -9.769 |
deen | -9.591 |
Magical | -9.368 |
ower | -9.142 |
田 | -8.734 |
Token | of |
Token ID | 286 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
roc | +5.435 |
opoly | +5.276 |
upfront | +5.136 |
Concord | +5.044 |
Albany | +4.987 |
Neg logit contributions | |
Magical | -6.378 |
adesh | -6.356 |
deen | -6.235 |
ower | -6.032 |
Pavilion | -5.821 |
Token | which |
Token ID | 543 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
roc | +5.545 |
opoly | +5.404 |
Concord | +5.376 |
Albany | +5.341 |
Sagan | +5.079 |
Neg logit contributions | |
adesh | -10.129 |
deen | -9.972 |
Magical | -9.782 |
ower | -9.733 |
Pavilion | -9.103 |
Token | finding |
Token ID | 4917 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
roc | +2.880 |
opoly | +2.850 |
Concord | +2.660 |
upfront | +2.612 |
Albany | +2.610 |
Neg logit contributions | |
Magical | -4.216 |
adesh | -4.196 |
deen | -4.138 |
ower | -4.064 |
Piercing | -3.854 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
roc | +8.367 |
Albany | +8.052 |
Concord | +7.972 |
upfront | +7.824 |
opoly | +7.782 |
Neg logit contributions | |
adesh | -10.722 |
deen | -10.294 |
Magical | -9.984 |
ower | -9.823 |
Piercing | -9.277 |
Token | in |
Token ID | 287 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
roc | +0.787 |
opoly | +0.783 |
Concord | +0.727 |
Times | +0.726 |
upfront | +0.717 |
Neg logit contributions | |
Magical | -0.788 |
adesh | -0.772 |
deen | -0.752 |
ower | -0.747 |
Piercing | -0.711 |
Token | Film |
Token ID | 13741 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
roc | +3.144 |
opoly | +3.114 |
Concord | +2.968 |
upfront | +2.893 |
Times | +2.892 |
Neg logit contributions | |
Magical | -3.322 |
adesh | -3.307 |
deen | -3.233 |
ower | -3.216 |
Piercing | -3.035 |
Token | city |
Token ID | 19205 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
Magical | +1.259 |
adesh | +1.227 |
ower | +1.191 |
deen | +1.189 |
Piercing | +1.134 |
Neg logit contributions | |
roc | -1.180 |
opoly | -1.173 |
Times | -1.089 |
Concord | -1.087 |
upfront | -1.074 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
opoly | +3.263 |
roc | +3.244 |
Times | +3.005 |
Concord | +2.986 |
upfront | +2.964 |
Neg logit contributions | |
Magical | -3.338 |
adesh | -3.280 |
deen | -3.215 |
ower | -3.162 |
Piercing | -3.031 |
Token | on |
Token ID | 319 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
roc | +1.487 |
opoly | +1.478 |
Concord | +1.375 |
Times | +1.365 |
upfront | +1.346 |
Neg logit contributions | |
Magical | -1.646 |
adesh | -1.611 |
deen | -1.580 |
ower | -1.567 |
Piercing | -1.491 |
Token | the |
Token ID | 262 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
opoly | +6.833 |
roc | +6.749 |
upfront | +6.271 |
Concord | +6.249 |
Albany | +6.186 |
Neg logit contributions | |
Magical | -8.224 |
adesh | -8.200 |
deen | -8.103 |
ower | -7.778 |
Piercing | -7.575 |
Token | outskirts |
Token ID | 30140 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
opoly | +7.408 |
roc | +7.223 |
upfront | +7.208 |
ground | +7.098 |
Concord | +7.040 |
Neg logit contributions | |
adesh | -10.807 |
deen | -10.652 |
Magical | -10.568 |
ower | -10.225 |
田 | -9.949 |
Token | of |
Token ID | 286 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
roc | +1.422 |
opoly | +1.410 |
Concord | +1.318 |
Times | +1.310 |
upfront | +1.298 |
Neg logit contributions | |
Magical | -1.526 |
adesh | -1.512 |
deen | -1.467 |
ower | -1.451 |
Piercing | -1.387 |
Token | Mumbai |
Token ID | 22917 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
roc | +6.167 |
opoly | +6.035 |
Concord | +5.881 |
Albany | +5.744 |
Times | +5.664 |
Neg logit contributions | |
adesh | -7.976 |
ower | -7.803 |
deen | -7.737 |
Magical | -7.717 |
mage | -7.240 |
Token | . |
Token ID | 13 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
roc | +5.068 |
opoly | +5.060 |
Times | +4.654 |
upfront | +4.592 |
Concord | +4.547 |
Neg logit contributions | |
Magical | -5.828 |
adesh | -5.812 |
deen | -5.678 |
ower | -5.599 |
Piercing | -5.398 |
Token |
|
Token ID | 198 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
opoly | +6.271 |
Times | +6.199 |
roc | +6.158 |
Gl | +5.967 |
Concord | +5.841 |
Neg logit contributions | |
deen | -8.687 |
adesh | -8.675 |
Magical | -8.503 |
ower | -8.402 |
shed | -7.975 |