Token | sel |
Token ID | 741 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
姫 | +1.071 |
� | +1.022 |
andr | +0.976 |
ennial | +0.971 |
sand | +0.960 |
Neg logit contributions | |
ascript | -1.046 |
lect | -0.896 |
spir | -0.869 |
NK | -0.855 |
Roll | -0.828 |
Token | ov |
Token ID | 709 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
姫 | +0.147 |
� | +0.140 |
ennial | +0.132 |
andr | +0.132 |
andestine | +0.131 |
Neg logit contributions | |
ascript | -0.161 |
lect | -0.141 |
spir | -0.137 |
NK | -0.136 |
Roll | -0.129 |
Token | from |
Token ID | 422 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
姫 | +1.180 |
� | +1.136 |
andr | +1.089 |
ennial | +1.078 |
oling | +1.062 |
Neg logit contributions | |
ascript | -1.371 |
lect | -1.196 |
spir | -1.171 |
NK | -1.155 |
Roll | -1.100 |
Token | the |
Token ID | 262 |
Feature activation | +2.57e-02 |
Pos logit contributions | |
姫 | +1.222 |
� | +1.179 |
andr | +1.129 |
ennial | +1.111 |
sand | +1.107 |
Neg logit contributions | |
ascript | -1.250 |
lect | -1.101 |
NK | -1.076 |
spir | -1.070 |
Roll | -1.024 |
Token | Harvard |
Token ID | 11131 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
姫 | +0.115 |
� | +0.109 |
andr | +0.103 |
ennial | +0.102 |
oling | +0.101 |
Neg logit contributions | |
ascript | -0.136 |
lect | -0.124 |
NK | -0.123 |
spir | -0.120 |
Roll | -0.116 |
Token | - |
Token ID | 12 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
姫 | +2.249 |
� | +2.172 |
andr | +2.080 |
ennial | +2.014 |
sand | +2.013 |
Neg logit contributions | |
ascript | -3.691 |
lect | -3.239 |
spir | -3.199 |
NK | -3.144 |
Roll | -3.088 |
Token | Smith |
Token ID | 17919 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
姫 | +6.050 |
� | +5.825 |
sand | +5.616 |
ennial | +5.573 |
jan | +5.525 |
Neg logit contributions | |
ascript | -11.107 |
hemy | -9.160 |
lect | -9.071 |
glers | -9.002 |
Citiz | -8.916 |
Token | sonian |
Token ID | 35202 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
姫 | +1.875 |
� | +1.805 |
andr | +1.737 |
andestine | +1.690 |
oling | +1.684 |
Neg logit contributions | |
ascript | -2.054 |
lect | -1.792 |
spir | -1.770 |
NK | -1.721 |
Roll | -1.683 |
Token | Center |
Token ID | 3337 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
姫 | +1.189 |
� | +1.136 |
andr | +1.105 |
sand | +1.077 |
andestine | +1.069 |
Neg logit contributions | |
ascript | -1.246 |
lect | -1.106 |
spir | -1.077 |
NK | -1.066 |
Roll | -1.033 |
Token | for |
Token ID | 329 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
姫 | +2.218 |
� | +2.159 |
andr | +2.095 |
ennial | +2.023 |
sand | +2.018 |
Neg logit contributions | |
ascript | -3.549 |
lect | -3.059 |
spir | -3.049 |
glers | -2.974 |
NK | -2.969 |
Token | Ast |
Token ID | 8304 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
ascript | +1.004 |
lect | +0.924 |
NK | +0.921 |
spir | +0.875 |
Roll | +0.821 |
Neg logit contributions | |
姫 | -1.400 |
� | -1.331 |
andr | -1.277 |
oling | -1.268 |
ennial | -1.247 |
Token | Imp |
Token ID | 9855 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
ascript | +1.797 |
lect | +1.610 |
spir | +1.553 |
NK | +1.521 |
Roll | +1.490 |
Neg logit contributions | |
姫 | -1.419 |
� | -1.357 |
andr | -1.300 |
ennial | -1.274 |
sand | -1.258 |
Token | é |
Token ID | 2634 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
姫 | +2.173 |
� | +2.143 |
andr | +2.077 |
ennial | +2.038 |
sand | +2.015 |
Neg logit contributions | |
ascript | -2.441 |
lect | -2.078 |
spir | -2.059 |
NK | -2.029 |
Roll | -1.977 |
Token | rial |
Token ID | 4454 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
姫 | +3.300 |
� | +3.246 |
andr | +3.137 |
ennial | +3.070 |
sand | +3.068 |
Neg logit contributions | |
ascript | -3.072 |
lect | -2.575 |
spir | -2.564 |
NK | -2.530 |
Roll | -2.452 |
Token | . |
Token ID | 13 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
ascript | +2.155 |
lect | +1.890 |
spir | +1.839 |
NK | +1.835 |
Roll | +1.773 |
Neg logit contributions | |
姫 | -1.743 |
� | -1.650 |
andr | -1.583 |
sand | -1.551 |
oling | -1.541 |
Token | Pr |
Token ID | 1736 |
Feature activation | +8.60e-02 |
Pos logit contributions | |
姫 | +0.625 |
� | +0.606 |
andr | +0.587 |
ennial | +0.574 |
sand | +0.570 |
Neg logit contributions | |
ascript | -0.717 |
lect | -0.624 |
spir | -0.612 |
NK | -0.604 |
Roll | -0.593 |
Token | at |
Token ID | 265 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
姫 | +0.390 |
� | +0.374 |
andr | +0.356 |
ennial | +0.348 |
sand | +0.347 |
Neg logit contributions | |
ascript | -0.464 |
lect | -0.405 |
spir | -0.393 |
NK | -0.380 |
Roll | -0.376 |
Token | iques |
Token ID | 6368 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
� | +6.911 |
姫 | +6.699 |
andr | +6.544 |
jan | +6.534 |
sand | +6.466 |
Neg logit contributions | |
ascript | -8.313 |
spir | -6.988 |
NK | -6.816 |
glers | -6.770 |
lect | -6.743 |
Token | qui |
Token ID | 45567 |
Feature activation | +8.51e-02 |
Pos logit contributions | |
ascript | +0.915 |
lect | +0.813 |
spir | +0.787 |
NK | +0.773 |
Roll | +0.756 |
Neg logit contributions | |
姫 | -0.707 |
� | -0.678 |
andr | -0.652 |
ennial | -0.640 |
sand | -0.633 |
Token | de |
Token ID | 390 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
姫 | +0.383 |
� | +0.370 |
andr | +0.356 |
ennial | +0.348 |
sand | +0.347 |
Neg logit contributions | |
ascript | -0.456 |
lect | -0.396 |
spir | -0.386 |
NK | -0.381 |
Roll | -0.371 |
Token | ven |
Token ID | 574 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
姫 | +3.287 |
� | +3.243 |
andr | +3.142 |
sand | +3.054 |
jan | +3.052 |
Neg logit contributions | |
ascript | -3.379 |
spir | -2.807 |
lect | -2.790 |
NK | -2.784 |
Roll | -2.712 |
Token | a |
Token ID | 64 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
姫 | +0.074 |
� | +0.071 |
andr | +0.069 |
ennial | +0.067 |
sand | +0.067 |
Neg logit contributions | |
ascript | -0.085 |
lect | -0.074 |
spir | -0.072 |
NK | -0.070 |
Roll | -0.068 |
Token | ness |
Token ID | 1108 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
姫 | +0.790 |
� | +0.757 |
andr | +0.734 |
ennial | +0.713 |
sand | +0.707 |
Neg logit contributions | |
ascript | -0.955 |
lect | -0.838 |
spir | -0.822 |
NK | -0.811 |
Roll | -0.789 |
Token | . |
Token ID | 13 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
姫 | +0.217 |
� | +0.208 |
andr | +0.201 |
ennial | +0.194 |
andestine | +0.194 |
Neg logit contributions | |
ascript | -0.244 |
lect | -0.209 |
spir | -0.206 |
NK | -0.202 |
Roll | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
ascript | +1.213 |
NK | +1.079 |
lect | +1.079 |
spir | +1.055 |
Roll | +1.046 |
Neg logit contributions | |
姫 | -1.218 |
� | -1.150 |
andr | -1.124 |
ennial | -1.108 |
oling | -1.108 |
Token |
|
Token ID | 198 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
ascript | +0.708 |
lect | +0.642 |
spir | +0.642 |
Roll | +0.641 |
NK | +0.625 |
Neg logit contributions | |
姫 | -0.620 |
� | -0.604 |
andr | -0.583 |
oling | -0.580 |
andestine | -0.574 |
Token | Continue |
Token ID | 29453 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
姫 | +0.025 |
� | +0.024 |
andr | +0.023 |
ennial | +0.023 |
oling | +0.023 |
Neg logit contributions | |
ascript | -0.029 |
lect | -0.026 |
spir | -0.025 |
Roll | -0.025 |
NK | -0.024 |
Token | Reading |
Token ID | 11725 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ascript | +0.361 |
lect | +0.298 |
spir | +0.283 |
NK | +0.280 |
Roll | +0.267 |
Neg logit contributions | |
姫 | -0.630 |
� | -0.609 |
andr | -0.599 |
ennial | -0.583 |
sand | -0.581 |
Token | Below |
Token ID | 10383 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
� | +5.210 |
姫 | +5.068 |
sand | +4.911 |
Ernest | +4.860 |
ennial | +4.859 |
Neg logit contributions | |
ascript | -9.844 |
spir | -8.524 |
lect | -8.392 |
glers | -8.243 |
Citiz | -8.213 |
Token | Advertisement |
Token ID | 39711 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
andr | +2.876 |
� | +2.868 |
姫 | +2.760 |
sand | +2.680 |
jan | +2.660 |
Neg logit contributions | |
ascript | -11.004 |
spir | -9.872 |
Citiz | -9.662 |
lect | -9.652 |
glers | -9.550 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
姫 | +2.377 |
� | +2.321 |
andr | +2.188 |
ennial | +2.141 |
sand | +2.138 |
Neg logit contributions | |
ascript | -5.452 |
lect | -4.791 |
Citiz | -4.764 |
spir | -4.738 |
glers | -4.636 |
Token | There |
Token ID | 1858 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
姫 | +0.261 |
� | +0.251 |
andr | +0.242 |
ennial | +0.237 |
oling | +0.235 |
Neg logit contributions | |
ascript | -0.298 |
lect | -0.262 |
spir | -0.258 |
Roll | -0.252 |
NK | -0.248 |
Token | are |
Token ID | 389 |
Feature activation | +1.17e-03 |
Pos logit contributions | |
姫 | +0.618 |
� | +0.599 |
andr | +0.581 |
ennial | +0.566 |
sand | +0.566 |
Neg logit contributions | |
ascript | -0.675 |
lect | -0.580 |
spir | -0.570 |
NK | -0.559 |
Roll | -0.546 |
Token | for |
Token ID | 329 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
姫 | +0.153 |
� | +0.147 |
andr | +0.143 |
andestine | +0.137 |
oling | +0.136 |
Neg logit contributions | |
ascript | -0.150 |
lect | -0.134 |
NK | -0.133 |
spir | -0.130 |
Roll | -0.130 |
Token | consideration |
Token ID | 9110 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
姫 | +3.062 |
� | +2.988 |
andr | +2.909 |
ennial | +2.844 |
sand | +2.843 |
Neg logit contributions | |
ascript | -2.401 |
lect | -1.991 |
spir | -1.951 |
NK | -1.905 |
Roll | -1.849 |
Token | for |
Token ID | 329 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
姫 | +3.999 |
� | +3.863 |
sand | +3.707 |
andr | +3.697 |
ennial | +3.675 |
Neg logit contributions | |
ascript | -6.668 |
lect | -5.719 |
Citiz | -5.676 |
spir | -5.675 |
NK | -5.484 |
Token | the |
Token ID | 262 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
姫 | +3.553 |
� | +3.462 |
andr | +3.355 |
ennial | +3.279 |
sand | +3.272 |
Neg logit contributions | |
ascript | -3.864 |
lect | -3.285 |
spir | -3.239 |
NK | -3.164 |
Roll | -3.088 |
Token | 2019 |
Token ID | 13130 |
Feature activation | +6.39e-01 |
Pos logit contributions | |
姫 | +1.418 |
� | +1.376 |
andr | +1.307 |
ennial | +1.279 |
sand | +1.267 |
Neg logit contributions | |
ascript | -2.473 |
lect | -2.172 |
spir | -2.155 |
NK | -2.106 |
Roll | -2.070 |
Token | award |
Token ID | 5764 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
姫 | +2.908 |
� | +2.792 |
andr | +2.695 |
ennial | +2.620 |
sand | +2.620 |
Neg logit contributions | |
ascript | -3.389 |
lect | -2.950 |
spir | -2.880 |
NK | -2.848 |
Roll | -2.783 |
Token | is |
Token ID | 318 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
andr | +4.126 |
� | +4.072 |
姫 | +4.008 |
ennial | +3.956 |
jan | +3.896 |
Neg logit contributions | |
ascript | -10.248 |
lect | -9.102 |
Citiz | -9.099 |
spir | -8.852 |
hemy | -8.741 |
Token | February |
Token ID | 3945 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
姫 | +2.602 |
� | +2.547 |
andr | +2.454 |
ennial | +2.413 |
jan | +2.401 |
Neg logit contributions | |
ascript | -3.462 |
spir | -2.983 |
lect | -2.965 |
NK | -2.877 |
Citiz | -2.840 |
Token | 28 |
Token ID | 2579 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
姫 | +1.442 |
� | +1.392 |
andr | +1.345 |
oling | +1.295 |
ennial | +1.290 |
Neg logit contributions | |
ascript | -1.705 |
lect | -1.510 |
spir | -1.474 |
NK | -1.471 |
Roll | -1.431 |
Token | , |
Token ID | 11 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
姫 | +2.116 |
� | +2.038 |
andr | +1.968 |
sand | +1.890 |
ennial | +1.882 |
Neg logit contributions | |
ascript | -2.952 |
lect | -2.605 |
spir | -2.554 |
NK | -2.529 |
Roll | -2.475 |
Token | 2019 |
Token ID | 13130 |
Feature activation | +6.39e-01 |
Pos logit contributions | |
ascript | +1.952 |
lect | +1.736 |
spir | +1.675 |
NK | +1.661 |
Roll | +1.639 |
Neg logit contributions | |
姫 | -1.486 |
� | -1.411 |
andr | -1.383 |
sand | -1.323 |
andestine | -1.321 |
Token | I |
Token ID | 314 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
姫 | +1.445 |
� | +1.384 |
andr | +1.313 |
ennial | +1.298 |
andestine | +1.280 |
Neg logit contributions | |
ascript | -1.544 |
lect | -1.373 |
spir | -1.332 |
NK | -1.325 |
Roll | -1.277 |
Token | think |
Token ID | 892 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
姫 | +1.236 |
� | +1.199 |
andr | +1.140 |
ennial | +1.126 |
andestine | +1.109 |
Neg logit contributions | |
ascript | -1.161 |
lect | -1.005 |
spir | -0.975 |
NK | -0.952 |
Roll | -0.922 |
Token | reduces |
Token ID | 12850 |
Feature activation | +6.04e-01 |
Pos logit contributions | |
姫 | +2.266 |
� | +2.234 |
andr | +2.129 |
ennial | +2.078 |
sand | +2.069 |
Neg logit contributions | |
ascript | -2.706 |
lect | -2.348 |
spir | -2.307 |
NK | -2.257 |
Roll | -2.197 |
Token | immersion |
Token ID | 36375 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
姫 | +3.379 |
� | +3.294 |
andr | +3.203 |
ennial | +3.135 |
sand | +3.132 |
Neg logit contributions | |
ascript | -2.560 |
lect | -2.120 |
spir | -2.075 |
NK | -2.027 |
Roll | -1.961 |
Token | more |
Token ID | 517 |
Feature activation | +8.00e-02 |
Pos logit contributions | |
姫 | +3.281 |
� | +3.210 |
andr | +3.123 |
sand | +3.065 |
ennial | +3.049 |
Neg logit contributions | |
ascript | -2.860 |
lect | -2.370 |
spir | -2.325 |
NK | -2.279 |
Roll | -2.215 |
Token | on |
Token ID | 319 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
姫 | +0.497 |
� | +0.480 |
andr | +0.466 |
ennial | +0.455 |
andestine | +0.455 |
Neg logit contributions | |
ascript | -0.294 |
lect | -0.249 |
spir | -0.237 |
NK | -0.237 |
Roll | -0.228 |
Token | H |
Token ID | 367 |
Feature activation | +5.71e-01 |
Pos logit contributions | |
� | +4.831 |
姫 | +4.577 |
andr | +4.564 |
sand | +4.432 |
jan | +4.297 |
Neg logit contributions | |
ascript | -9.962 |
lect | -8.403 |
spir | -8.395 |
Citiz | -8.394 |
hemy | -8.177 |
Token | MD |
Token ID | 12740 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
姫 | +3.226 |
� | +3.146 |
andr | +3.061 |
ennial | +2.998 |
sand | +2.994 |
Neg logit contributions | |
ascript | -2.384 |
lect | -1.976 |
spir | -1.927 |
NK | -1.882 |
Roll | -1.831 |
Token | s |
Token ID | 82 |
Feature activation | +9.37e-01 |
Pos logit contributions | |
姫 | +2.880 |
� | +2.757 |
andr | +2.656 |
sand | +2.600 |
ennial | +2.552 |
Neg logit contributions | |
ascript | -5.108 |
lect | -4.482 |
spir | -4.412 |
NK | -4.361 |
Roll | -4.277 |
Token | but |
Token ID | 475 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
姫 | +3.560 |
� | +3.452 |
andr | +3.363 |
sand | +3.272 |
ennial | +3.196 |
Neg logit contributions | |
ascript | -5.598 |
lect | -4.859 |
spir | -4.807 |
NK | -4.703 |
Citiz | -4.632 |
Token | it |
Token ID | 340 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
姫 | +1.800 |
� | +1.729 |
andr | +1.671 |
ennial | +1.643 |
oling | +1.618 |
Neg logit contributions | |
ascript | -1.688 |
lect | -1.455 |
spir | -1.433 |
NK | -1.409 |
Roll | -1.370 |
Token | . |
Token ID | 13 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
姫 | +0.217 |
� | +0.208 |
andr | +0.201 |
ennial | +0.194 |
andestine | +0.194 |
Neg logit contributions | |
ascript | -0.244 |
lect | -0.209 |
spir | -0.206 |
NK | -0.202 |
Roll | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
ascript | +1.213 |
NK | +1.079 |
lect | +1.079 |
spir | +1.055 |
Roll | +1.046 |
Neg logit contributions | |
姫 | -1.218 |
� | -1.150 |
andr | -1.124 |
ennial | -1.108 |
oling | -1.108 |
Token |
|
Token ID | 198 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
ascript | +0.708 |
lect | +0.642 |
spir | +0.642 |
Roll | +0.641 |
NK | +0.625 |
Neg logit contributions | |
姫 | -0.620 |
� | -0.604 |
andr | -0.583 |
oling | -0.580 |
andestine | -0.574 |
Token | Continue |
Token ID | 29453 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
姫 | +0.025 |
� | +0.024 |
andr | +0.023 |
ennial | +0.023 |
oling | +0.023 |
Neg logit contributions | |
ascript | -0.029 |
lect | -0.026 |
spir | -0.025 |
Roll | -0.025 |
NK | -0.024 |
Token | Reading |
Token ID | 11725 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ascript | +0.361 |
lect | +0.298 |
spir | +0.283 |
NK | +0.280 |
Roll | +0.267 |
Neg logit contributions | |
姫 | -0.630 |
� | -0.609 |
andr | -0.599 |
ennial | -0.583 |
sand | -0.581 |
Token | Below |
Token ID | 10383 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
� | +5.210 |
姫 | +5.068 |
sand | +4.911 |
Ernest | +4.860 |
ennial | +4.859 |
Neg logit contributions | |
ascript | -9.844 |
spir | -8.524 |
lect | -8.392 |
glers | -8.243 |
Citiz | -8.213 |
Token | Advertisement |
Token ID | 39711 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
andr | +2.876 |
� | +2.868 |
姫 | +2.760 |
sand | +2.680 |
jan | +2.660 |
Neg logit contributions | |
ascript | -11.004 |
spir | -9.872 |
Citiz | -9.662 |
lect | -9.652 |
glers | -9.550 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
姫 | +2.377 |
� | +2.321 |
andr | +2.188 |
ennial | +2.141 |
sand | +2.138 |
Neg logit contributions | |
ascript | -5.452 |
lect | -4.791 |
Citiz | -4.764 |
spir | -4.738 |
glers | -4.636 |
Token | There |
Token ID | 1858 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
姫 | +0.261 |
� | +0.251 |
andr | +0.242 |
ennial | +0.237 |
oling | +0.235 |
Neg logit contributions | |
ascript | -0.298 |
lect | -0.262 |
spir | -0.258 |
Roll | -0.252 |
NK | -0.248 |
Token | are |
Token ID | 389 |
Feature activation | +1.17e-03 |
Pos logit contributions | |
姫 | +0.618 |
� | +0.599 |
andr | +0.581 |
ennial | +0.566 |
sand | +0.566 |
Neg logit contributions | |
ascript | -0.675 |
lect | -0.580 |
spir | -0.570 |
NK | -0.559 |
Roll | -0.546 |
Token | many |
Token ID | 867 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
姫 | +0.006 |
� | +0.006 |
andr | +0.006 |
ennial | +0.006 |
sand | +0.006 |
Neg logit contributions | |
ascript | -0.005 |
lect | -0.005 |
spir | -0.005 |
NK | -0.005 |
Roll | -0.004 |
Token | games |
Token ID | 1830 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
姫 | +2.079 |
� | +1.997 |
andr | +1.923 |
ennial | +1.870 |
sand | +1.844 |
Neg logit contributions | |
ascript | -3.050 |
lect | -2.685 |
spir | -2.636 |
NK | -2.609 |
Roll | -2.544 |
Token | with |
Token ID | 351 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
姫 | +1.961 |
� | +1.895 |
andr | +1.837 |
ennial | +1.793 |
sand | +1.771 |
Neg logit contributions | |
ascript | -1.857 |
lect | -1.596 |
spir | -1.574 |
NK | -1.544 |
Roll | -1.499 |
Token | AA |
Token ID | 15923 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
姫 | +1.364 |
� | +1.324 |
andr | +1.274 |
ennial | +1.264 |
sand | +1.254 |
Neg logit contributions | |
ascript | -0.977 |
lect | -0.814 |
NK | -0.802 |
spir | -0.795 |
Roll | -0.765 |
Token | on |
Token ID | 319 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
ascript | +0.869 |
lect | +0.740 |
NK | +0.700 |
spir | +0.696 |
Roll | +0.682 |
Neg logit contributions | |
姫 | -1.561 |
� | -1.521 |
andr | -1.484 |
ennial | -1.452 |
andestine | -1.444 |
Token | ( |
Token ID | 357 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
姫 | +3.289 |
� | +3.192 |
andr | +3.067 |
sand | +2.991 |
ennial | +2.939 |
Neg logit contributions | |
ascript | -5.268 |
lect | -4.573 |
spir | -4.514 |
NK | -4.421 |
Citiz | -4.348 |
Token | Anti |
Token ID | 28795 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
姫 | +2.819 |
� | +2.725 |
andr | +2.616 |
ennial | +2.553 |
sand | +2.539 |
Neg logit contributions | |
ascript | -3.726 |
lect | -3.257 |
spir | -3.201 |
NK | -3.145 |
Roll | -3.093 |
Token | Ali |
Token ID | 37893 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
姫 | +4.776 |
� | +4.692 |
andr | +4.683 |
ennial | +4.419 |
jan | +4.410 |
Neg logit contributions | |
ascript | -9.352 |
Citiz | -8.117 |
NK | -8.081 |
spir | -8.040 |
lect | -8.011 |
Token | asing |
Token ID | 2313 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
ascript | +0.475 |
lect | +0.395 |
spir | +0.357 |
hemy | +0.352 |
pen | +0.348 |
Neg logit contributions | |
姫 | -0.630 |
� | -0.575 |
andr | -0.551 |
andestine | -0.548 |
sandy | -0.547 |
Token | is |
Token ID | 318 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
姫 | +1.376 |
� | +1.319 |
andr | +1.271 |
ennial | +1.241 |
oling | +1.228 |
Neg logit contributions | |
ascript | -1.361 |
lect | -1.188 |
spir | -1.151 |
NK | -1.145 |
Roll | -1.114 |
Token | a |
Token ID | 257 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
姫 | +1.300 |
� | +1.248 |
andr | +1.222 |
ennial | +1.197 |
andestine | +1.186 |
Neg logit contributions | |
ascript | -0.940 |
lect | -0.779 |
spir | -0.753 |
NK | -0.748 |
Roll | -0.719 |
Token | technique |
Token ID | 8173 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
姫 | +2.210 |
� | +2.115 |
andr | +2.041 |
ennial | +1.996 |
oling | +1.971 |
Neg logit contributions | |
ascript | -2.579 |
lect | -2.254 |
spir | -2.196 |
NK | -2.165 |
Roll | -2.097 |
Token | from |
Token ID | 422 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
姫 | +0.366 |
� | +0.353 |
andr | +0.344 |
sand | +0.335 |
ennial | +0.334 |
Neg logit contributions | |
ascript | -0.367 |
lect | -0.314 |
spir | -0.308 |
NK | -0.302 |
Roll | -0.292 |
Token | the |
Token ID | 262 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
姫 | +0.596 |
� | +0.560 |
andr | +0.548 |
sand | +0.529 |
ennial | +0.525 |
Neg logit contributions | |
ascript | -0.817 |
lect | -0.723 |
spir | -0.703 |
NK | -0.703 |
Roll | -0.681 |
Token | willful |
Token ID | 47860 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
姫 | +0.272 |
� | +0.255 |
andr | +0.248 |
sand | +0.243 |
ennial | +0.238 |
Neg logit contributions | |
ascript | -0.358 |
lect | -0.315 |
spir | -0.311 |
NK | -0.307 |
Roll | -0.297 |
Token | and |
Token ID | 290 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ascript | +0.315 |
lect | +0.282 |
spir | +0.275 |
NK | +0.270 |
glers | +0.264 |
Neg logit contributions | |
姫 | -0.265 |
� | -0.252 |
andr | -0.247 |
sand | -0.241 |
ennial | -0.238 |
Token | / |
Token ID | 14 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
� | +3.699 |
姫 | +3.575 |
andr | +3.353 |
ennial | +3.353 |
extinction | +3.258 |
Neg logit contributions | |
ascript | -7.334 |
lect | -6.263 |
NK | -6.224 |
spir | -6.155 |
hemy | -6.079 |
Token | or |
Token ID | 273 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ascript | +1.117 |
lect | +1.020 |
spir | +0.998 |
Roll | +0.945 |
NK | +0.944 |
Neg logit contributions | |
姫 | -0.942 |
� | -0.896 |
extinction | -0.835 |
andestine | -0.834 |
andr | -0.833 |
Token | negligent |
Token ID | 42837 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
� | +4.612 |
姫 | +4.283 |
ennial | +4.151 |
andr | +4.095 |
oling | +4.036 |
Neg logit contributions | |
ascript | -9.554 |
NK | -8.185 |
lect | -8.152 |
Roll | -8.080 |
hemy | -8.002 |
Token | inclusion |
Token ID | 14900 |
Feature activation | +9.23e-02 |
Pos logit contributions | |
姫 | +0.500 |
� | +0.480 |
andr | +0.467 |
sand | +0.452 |
ennial | +0.448 |
Neg logit contributions | |
ascript | -0.667 |
lect | -0.591 |
spir | -0.578 |
NK | -0.569 |
Roll | -0.556 |
Token | of |
Token ID | 286 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
姫 | +0.387 |
� | +0.377 |
andr | +0.363 |
ennial | +0.356 |
sand | +0.350 |
Neg logit contributions | |
ascript | -0.516 |
lect | -0.451 |
spir | -0.445 |
NK | -0.438 |
Roll | -0.429 |
Token | inaccurate |
Token ID | 21873 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
姫 | +0.255 |
� | +0.239 |
andr | +0.235 |
sand | +0.228 |
ennial | +0.226 |
Neg logit contributions | |
ascript | -0.329 |
lect | -0.290 |
NK | -0.283 |
spir | -0.283 |
Roll | -0.272 |
Token | information |
Token ID | 1321 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
ascript | +0.201 |
lect | +0.179 |
NK | +0.174 |
spir | +0.172 |
Roll | +0.167 |
Neg logit contributions | |
姫 | -0.161 |
� | -0.151 |
andr | -0.148 |
sand | -0.143 |
oling | -0.142 |
Token | read |
Token ID | 1100 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
ascript | +0.391 |
lect | +0.347 |
NK | +0.345 |
spir | +0.329 |
Roll | +0.325 |
Neg logit contributions | |
姫 | -0.336 |
� | -0.314 |
andr | -0.310 |
ennial | -0.306 |
oling | -0.301 |
Token | our |
Token ID | 674 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
姫 | +2.905 |
� | +2.837 |
andr | +2.745 |
ennial | +2.675 |
sand | +2.667 |
Neg logit contributions | |
ascript | -3.250 |
lect | -2.779 |
spir | -2.750 |
NK | -2.676 |
Roll | -2.633 |
Token | comments |
Token ID | 3651 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
姫 | +3.898 |
� | +3.838 |
andr | +3.635 |
sand | +3.584 |
ennial | +3.545 |
Neg logit contributions | |
ascript | -5.756 |
spir | -4.922 |
lect | -4.892 |
NK | -4.871 |
hemy | -4.735 |
Token | policy |
Token ID | 2450 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
姫 | +5.654 |
� | +5.511 |
sand | +5.433 |
andr | +5.416 |
jan | +5.387 |
Neg logit contributions | |
ascript | -6.944 |
Citiz | -5.770 |
NK | -5.756 |
lect | -5.747 |
glers | -5.579 |
Token | before |
Token ID | 878 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
姫 | +2.195 |
� | +2.145 |
andr | +2.092 |
sand | +2.053 |
ennial | +2.041 |
Neg logit contributions | |
ascript | -1.667 |
lect | -1.369 |
spir | -1.345 |
NK | -1.308 |
Roll | -1.270 |
Token | commenting |
Token ID | 26387 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
姫 | +1.183 |
� | +1.140 |
andr | +1.096 |
ennial | +1.061 |
sand | +1.060 |
Neg logit contributions | |
ascript | -1.822 |
lect | -1.604 |
spir | -1.579 |
NK | -1.556 |
Roll | -1.524 |
Token | . |
Token ID | 13 |
Feature activation | +8.84e-01 |
Pos logit contributions | |
sand | +4.231 |
jan | +4.202 |
ennial | +4.201 |
andr | +4.009 |
� | +4.001 |
Neg logit contributions | |
ascript | -9.228 |
glers | -8.827 |
Citiz | -8.634 |
Palestin | -8.101 |
FTA | -7.795 |
Token |
|
Token ID | 198 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
� | +2.547 |
andr | +2.422 |
姫 | +2.406 |
sand | +2.389 |
jan | +2.373 |
Neg logit contributions | |
ascript | -6.086 |
spir | -5.194 |
Citiz | -5.125 |
lect | -5.105 |
glers | -5.019 |
Token |
|
Token ID | 198 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
姫 | +2.581 |
� | +2.493 |
andr | +2.389 |
ennial | +2.314 |
sand | +2.300 |
Neg logit contributions | |
ascript | -3.870 |
lect | -3.416 |
spir | -3.365 |
NK | -3.311 |
Roll | -3.252 |
Token | Note |
Token ID | 6425 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
姫 | +2.598 |
� | +2.533 |
andr | +2.440 |
sand | +2.386 |
ennial | +2.367 |
Neg logit contributions | |
ascript | -3.385 |
lect | -2.900 |
spir | -2.869 |
NK | -2.829 |
glers | -2.730 |
Token | : |
Token ID | 25 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
姫 | +2.691 |
� | +2.589 |
andr | +2.506 |
ennial | +2.434 |
sand | +2.432 |
Neg logit contributions | |
ascript | -3.183 |
lect | -2.777 |
spir | -2.714 |
NK | -2.682 |
Roll | -2.610 |
Token |
|
Token ID | 198 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
姫 | +2.097 |
� | +2.033 |
andr | +1.961 |
ennial | +1.906 |
sand | +1.904 |
Neg logit contributions | |
ascript | -2.695 |
lect | -2.335 |
spir | -2.299 |
NK | -2.259 |
Roll | -2.210 |
Token |
|
Token ID | 198 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
姫 | +0.637 |
� | +0.620 |
andr | +0.586 |
andestine | +0.577 |
oling | +0.570 |
Neg logit contributions | |
ascript | -0.733 |
lect | -0.667 |
spir | -0.656 |
NK | -0.644 |
Roll | -0.637 |
Token | The |
Token ID | 464 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
姫 | +0.366 |
� | +0.356 |
andr | +0.345 |
sand | +0.335 |
ennial | +0.335 |
Neg logit contributions | |
ascript | -0.399 |
lect | -0.345 |
spir | -0.339 |
NK | -0.331 |
Roll | -0.327 |
Token | annual |
Token ID | 5079 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
姫 | +1.887 |
� | +1.838 |
andr | +1.776 |
ennial | +1.753 |
oling | +1.735 |
Neg logit contributions | |
ascript | -1.956 |
lect | -1.642 |
spir | -1.630 |
NK | -1.575 |
Roll | -1.553 |
Token | L |
Token ID | 406 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
姫 | +0.867 |
� | +0.837 |
andr | +0.810 |
sand | +0.790 |
andestine | +0.789 |
Neg logit contributions | |
ascript | -0.833 |
lect | -0.727 |
spir | -0.705 |
NK | -0.704 |
Roll | -0.676 |
Token | oret |
Token ID | 9997 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
姫 | +1.531 |
� | +1.484 |
andr | +1.399 |
sand | +1.337 |
ennial | +1.320 |
Neg logit contributions | |
ascript | -2.596 |
lect | -2.330 |
spir | -2.277 |
NK | -2.229 |
Roll | -2.197 |
Token | Miller |
Token ID | 7920 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
姫 | +3.949 |
andr | +3.845 |
� | +3.792 |
ennial | +3.698 |
jan | +3.516 |
Neg logit contributions | |
ascript | -10.065 |
spir | -8.677 |
lect | -8.626 |
Citiz | -8.492 |
NK | -8.419 |
Token | R |
Token ID | 371 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
姫 | +3.216 |
� | +3.214 |
ennial | +3.068 |
andr | +3.047 |
extinction | +2.972 |
Neg logit contributions | |
ascript | -4.965 |
spir | -4.184 |
lect | -4.176 |
glers | -4.059 |
hemy | -4.036 |
Token | upp |
Token ID | 7211 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
姫 | +3.141 |
� | +3.053 |
andr | +2.957 |
ennial | +2.885 |
sand | +2.881 |
Neg logit contributions | |
ascript | -3.404 |
lect | -2.910 |
spir | -2.861 |
NK | -2.812 |
Roll | -2.738 |
Token | e |
Token ID | 68 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
姫 | +1.841 |
� | +1.758 |
andr | +1.676 |
sand | +1.620 |
ennial | +1.615 |
Neg logit contributions | |
ascript | -3.114 |
lect | -2.763 |
spir | -2.714 |
NK | -2.667 |
Roll | -2.627 |
Token | Award |
Token ID | 11289 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
� | +3.119 |
姫 | +3.114 |
andr | +3.011 |
ennial | +2.975 |
oling | +2.896 |
Neg logit contributions | |
ascript | -4.058 |
spir | -3.410 |
lect | -3.393 |
glers | -3.290 |
NK | -3.279 |
Token |
|
Token ID | 198 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
ascript | +0.744 |
lect | +0.656 |
NK | +0.648 |
spir | +0.648 |
Roll | +0.626 |
Neg logit contributions | |
姫 | -0.604 |
� | -0.582 |
andr | -0.563 |
oling | -0.544 |
andestine | -0.542 |
Token |
|
Token ID | 198 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
姫 | +0.834 |
� | +0.810 |
andr | +0.769 |
andestine | +0.754 |
oling | +0.749 |
Neg logit contributions | |
ascript | -0.963 |
lect | -0.878 |
spir | -0.866 |
NK | -0.846 |
Roll | -0.839 |
Token | A |
Token ID | 32 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
姫 | +0.728 |
� | +0.703 |
andr | +0.680 |
ennial | +0.658 |
andestine | +0.657 |
Neg logit contributions | |
ascript | -0.794 |
lect | -0.697 |
spir | -0.686 |
Roll | -0.665 |
NK | -0.661 |
Token | request |
Token ID | 2581 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
ascript | +0.553 |
lect | +0.487 |
spir | +0.474 |
NK | +0.467 |
Roll | +0.448 |
Neg logit contributions | |
姫 | -0.482 |
� | -0.462 |
andr | -0.450 |
andestine | -0.434 |
ennial | -0.431 |
Token | for |
Token ID | 329 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
姫 | +1.293 |
� | +1.250 |
andr | +1.206 |
sand | +1.171 |
ennial | +1.170 |
Neg logit contributions | |
ascript | -1.626 |
lect | -1.411 |
spir | -1.387 |
NK | -1.367 |
Roll | -1.334 |
Token | comment |
Token ID | 2912 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
姫 | +2.676 |
� | +2.616 |
andr | +2.529 |
ennial | +2.472 |
sand | +2.464 |
Neg logit contributions | |
ascript | -3.379 |
lect | -2.884 |
spir | -2.879 |
NK | -2.773 |
Roll | -2.749 |
Token | from |
Token ID | 422 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
� | +5.645 |
姫 | +5.570 |
ennial | +5.440 |
andr | +5.415 |
sand | +5.396 |
Neg logit contributions | |
ascript | -8.056 |
spir | -6.862 |
lect | -6.835 |
Citiz | -6.769 |
hemy | -6.665 |
Token | H |
Token ID | 367 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
姫 | +1.353 |
� | +1.314 |
andr | +1.271 |
ennial | +1.239 |
sand | +1.239 |
Neg logit contributions | |
ascript | -1.500 |
lect | -1.286 |
spir | -1.265 |
NK | -1.239 |
Roll | -1.212 |
Token | CCC |
Token ID | 46361 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
姫 | +1.148 |
� | +1.118 |
andr | +1.083 |
sand | +1.054 |
ennial | +1.052 |
Neg logit contributions | |
ascript | -0.891 |
lect | -0.750 |
spir | -0.737 |
NK | -0.711 |
Roll | -0.699 |
Token | President |
Token ID | 1992 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
姫 | +2.526 |
� | +2.445 |
andr | +2.365 |
ennial | +2.295 |
sand | +2.294 |
Neg logit contributions | |
ascript | -2.943 |
lect | -2.544 |
spir | -2.500 |
NK | -2.464 |
Roll | -2.398 |
Token | Glen |
Token ID | 20716 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
ascript | +2.747 |
NK | +2.521 |
lect | +2.516 |
spir | +2.415 |
Roll | +2.363 |
Neg logit contributions | |
姫 | -2.305 |
� | -2.157 |
andr | -2.057 |
oling | -2.054 |
ennial | -2.029 |
Token | delivering |
Token ID | 13630 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
姫 | +1.417 |
� | +1.370 |
andr | +1.328 |
ennial | +1.293 |
sand | +1.292 |
Neg logit contributions | |
ascript | -1.510 |
lect | -1.309 |
spir | -1.276 |
NK | -1.262 |
Roll | -1.228 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ascript | +1.771 |
lect | +1.556 |
NK | +1.484 |
spir | +1.457 |
Roll | +1.426 |
Neg logit contributions | |
姫 | -2.028 |
� | -1.913 |
andr | -1.865 |
ennial | -1.832 |
sand | -1.820 |
Token | 6 |
Token ID | 718 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
姫 | +0.521 |
� | +0.491 |
andr | +0.473 |
sand | +0.464 |
andestine | +0.461 |
Neg logit contributions | |
ascript | -0.574 |
lect | -0.516 |
NK | -0.496 |
spir | -0.489 |
Roll | -0.477 |
Token | th |
Token ID | 400 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
姫 | +0.386 |
� | +0.371 |
andr | +0.362 |
ennial | +0.352 |
sand | +0.351 |
Neg logit contributions | |
ascript | -0.367 |
lect | -0.316 |
spir | -0.307 |
NK | -0.303 |
Roll | -0.295 |
Token | Dil |
Token ID | 29328 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
ascript | +1.360 |
lect | +1.238 |
NK | +1.204 |
spir | +1.167 |
Roll | +1.159 |
Neg logit contributions | |
姫 | -1.399 |
� | -1.317 |
andr | -1.258 |
sand | -1.256 |
andestine | -1.244 |
Token | ip |
Token ID | 541 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
姫 | +1.872 |
� | +1.779 |
andr | +1.669 |
sand | +1.593 |
ennial | +1.589 |
Neg logit contributions | |
ascript | -4.751 |
lect | -4.270 |
spir | -4.219 |
NK | -4.157 |
Roll | -4.089 |
Token | Sard |
Token ID | 46997 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
姫 | +6.146 |
� | +5.982 |
andr | +5.889 |
sand | +5.741 |
ennial | +5.725 |
Neg logit contributions | |
ascript | -7.720 |
glers | -6.515 |
spir | -6.421 |
lect | -6.288 |
Roll | -6.204 |
Token | es |
Token ID | 274 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
姫 | +1.081 |
� | +1.038 |
andr | +0.987 |
ennial | +0.966 |
sand | +0.952 |
Neg logit contributions | |
ascript | -1.186 |
lect | -1.037 |
spir | -1.009 |
NK | -0.986 |
Roll | -0.956 |
Token | ai |
Token ID | 1872 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
姫 | +1.634 |
� | +1.584 |
andr | +1.492 |
sand | +1.487 |
ennial | +1.486 |
Neg logit contributions | |
ascript | -1.194 |
lect | -1.011 |
spir | -0.968 |
NK | -0.952 |
Roll | -0.922 |
Token | Memorial |
Token ID | 14861 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
姫 | +4.414 |
� | +4.325 |
andr | +4.221 |
ennial | +4.157 |
sand | +4.121 |
Neg logit contributions | |
ascript | -2.898 |
spir | -2.274 |
lect | -2.255 |
NK | -2.185 |
Roll | -2.120 |
Token | Lect |
Token ID | 31209 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
姫 | +1.509 |
� | +1.473 |
andr | +1.439 |
sand | +1.412 |
ennial | +1.411 |
Neg logit contributions | |
ascript | -0.773 |
lect | -0.627 |
spir | -0.596 |
NK | -0.582 |
Roll | -0.555 |
Token | . |
Token ID | 13 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
姫 | +2.516 |
� | +2.450 |
andr | +2.356 |
ennial | +2.309 |
sand | +2.308 |
Neg logit contributions | |
ascript | -2.911 |
lect | -2.500 |
spir | -2.469 |
NK | -2.415 |
Roll | -2.364 |
Token |
|
Token ID | 198 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
姫 | +1.066 |
� | +1.027 |
andr | +0.990 |
ennial | +0.963 |
sand | +0.959 |
Neg logit contributions | |
ascript | -1.377 |
lect | -1.216 |
spir | -1.198 |
NK | -1.184 |
Roll | -1.160 |
Token |
|
Token ID | 198 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
姫 | +0.552 |
� | +0.535 |
andr | +0.506 |
andestine | +0.499 |
oling | +0.493 |
Neg logit contributions | |
ascript | -0.601 |
lect | -0.555 |
spir | -0.546 |
NK | -0.533 |
Roll | -0.531 |
Token | Story |
Token ID | 11605 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
姫 | +0.709 |
� | +0.683 |
andr | +0.659 |
ennial | +0.643 |
andestine | +0.641 |
Neg logit contributions | |
ascript | -0.740 |
lect | -0.653 |
spir | -0.639 |
Roll | -0.625 |
NK | -0.608 |
Token | continues |
Token ID | 4477 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
ascript | +1.283 |
lect | +1.039 |
spir | +1.015 |
NK | +0.950 |
Roll | +0.941 |
Neg logit contributions | |
姫 | -1.531 |
� | -1.493 |
encia | -1.387 |
andestine | -1.383 |
andr | -1.381 |
Token | below |
Token ID | 2174 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
姫 | +1.103 |
� | +1.057 |
andr | +1.012 |
encia | +0.976 |
andestine | +0.973 |
Neg logit contributions | |
ascript | -1.422 |
lect | -1.271 |
NK | -1.233 |
spir | -1.229 |
Roll | -1.164 |
Token | advertisement |
Token ID | 15422 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
� | +3.095 |
姫 | +3.053 |
andr | +2.988 |
sand | +2.884 |
ennial | +2.789 |
Neg logit contributions | |
ascript | -10.633 |
spir | -9.323 |
lect | -9.312 |
Citiz | -9.215 |
NK | -9.167 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
姫 | +2.095 |
� | +2.022 |
andr | +1.936 |
ennial | +1.874 |
oling | +1.858 |
Neg logit contributions | |
ascript | -3.051 |
lect | -2.700 |
spir | -2.661 |
NK | -2.617 |
Roll | -2.570 |
Token |
|
Token ID | 198 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
姫 | +0.823 |
� | +0.800 |
andr | +0.759 |
andestine | +0.745 |
oling | +0.742 |
Neg logit contributions | |
ascript | -0.920 |
lect | -0.846 |
spir | -0.834 |
NK | -0.819 |
Roll | -0.810 |
Token | Story |
Token ID | 11605 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
姫 | +0.827 |
� | +0.805 |
andr | +0.775 |
ennial | +0.763 |
andestine | +0.760 |
Neg logit contributions | |
ascript | -0.676 |
lect | -0.587 |
spir | -0.571 |
Roll | -0.557 |
NK | -0.540 |
Token | continues |
Token ID | 4477 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
ascript | +0.084 |
lect | +0.071 |
spir | +0.070 |
NK | +0.068 |
Roll | +0.066 |
Neg logit contributions | |
姫 | -0.093 |
� | -0.090 |
andr | -0.084 |
andestine | -0.082 |
oling | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
ascript | +0.224 |
lect | +0.209 |
spir | +0.206 |
NK | +0.203 |
Roll | +0.201 |
Neg logit contributions | |
姫 | -0.222 |
� | -0.214 |
andr | -0.203 |
andestine | -0.201 |
oling | -0.200 |
Token | R |
Token ID | 49 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
姫 | +0.755 |
� | +0.717 |
andr | +0.693 |
ennial | +0.685 |
andestine | +0.683 |
Neg logit contributions | |
ascript | -0.803 |
lect | -0.719 |
spir | -0.708 |
Roll | -0.696 |
NK | -0.672 |
Token | OB |
Token ID | 9864 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
姫 | +1.867 |
� | +1.809 |
andr | +1.748 |
sand | +1.705 |
ennial | +1.704 |
Neg logit contributions | |
ascript | -2.019 |
lect | -1.738 |
spir | -1.706 |
NK | -1.669 |
Roll | -1.634 |
Token | ERT |
Token ID | 17395 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
姫 | +1.148 |
� | +1.092 |
andr | +1.051 |
ennial | +1.012 |
sand | +1.009 |
Neg logit contributions | |
ascript | -2.248 |
lect | -2.010 |
spir | -1.975 |
NK | -1.959 |
Roll | -1.921 |
Token | P |
Token ID | 350 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
姫 | +1.042 |
� | +1.009 |
andr | +0.976 |
ennial | +0.948 |
sand | +0.942 |
Neg logit contributions | |
ascript | -1.269 |
lect | -1.111 |
NK | -1.090 |
spir | -1.083 |
Roll | -1.053 |
Token | OR |
Token ID | 1581 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
姫 | +1.233 |
� | +1.191 |
andr | +1.156 |
ennial | +1.121 |
sand | +1.121 |
Neg logit contributions | |
ascript | -1.402 |
lect | -1.216 |
spir | -1.190 |
NK | -1.173 |
Roll | -1.147 |
Token | ATH |
Token ID | 12599 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
� | +3.355 |
姫 | +3.353 |
sand | +2.942 |
ennial | +2.886 |
PLE | +2.885 |
Neg logit contributions | |
ascript | -10.289 |
Citiz | -9.011 |
spir | -8.909 |
lect | -8.890 |
glers | -8.752 |
Token |
|
Token ID | 198 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
姫 | +1.608 |
� | +1.548 |
andr | +1.505 |
ennial | +1.463 |
sand | +1.459 |
Neg logit contributions | |
ascript | -1.909 |
lect | -1.661 |
spir | -1.628 |
NK | -1.610 |
Roll | -1.572 |
Token |
|
Token ID | 198 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
姫 | +0.859 |
� | +0.833 |
andr | +0.794 |
andestine | +0.775 |
oling | +0.773 |
Neg logit contributions | |
ascript | -1.012 |
lect | -0.923 |
spir | -0.910 |
NK | -0.896 |
Roll | -0.882 |
Token | B |
Token ID | 33 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
姫 | +0.136 |
� | +0.130 |
andr | +0.126 |
andestine | +0.124 |
extinction | +0.124 |
Neg logit contributions | |
ascript | -0.147 |
lect | -0.132 |
spir | -0.130 |
Roll | -0.128 |
NK | -0.123 |
Token | oulder |
Token ID | 17601 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
姫 | +1.308 |
� | +1.261 |
andr | +1.216 |
sand | +1.189 |
ennial | +1.186 |
Neg logit contributions | |
ascript | -1.431 |
lect | -1.238 |
spir | -1.209 |
NK | -1.187 |
Roll | -1.165 |
Token | was |
Token ID | 373 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ascript | +1.049 |
lect | +0.927 |
spir | +0.887 |
NK | +0.885 |
Roll | +0.853 |
Neg logit contributions | |
姫 | -1.046 |
� | -1.010 |
andr | -0.977 |
ennial | -0.959 |
sand | -0.955 |
Token | responsible |
Token ID | 4497 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ascript | +0.887 |
lect | +0.783 |
NK | +0.754 |
spir | +0.751 |
Roll | +0.707 |
Neg logit contributions | |
姫 | -0.961 |
� | -0.912 |
andr | -0.894 |
ennial | -0.881 |
sand | -0.879 |
Token | . |
Token ID | 13 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
姫 | +0.051 |
� | +0.049 |
andr | +0.047 |
sand | +0.046 |
ennial | +0.046 |
Neg logit contributions | |
ascript | -0.061 |
lect | -0.053 |
spir | -0.052 |
NK | -0.051 |
Roll | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
ascript | +0.100 |
lect | +0.088 |
NK | +0.087 |
spir | +0.086 |
Roll | +0.084 |
Neg logit contributions | |
姫 | -0.098 |
� | -0.094 |
andr | -0.091 |
oling | -0.089 |
ennial | -0.089 |
Token |
|
Token ID | 198 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
姫 | +0.162 |
� | +0.157 |
andr | +0.149 |
andestine | +0.146 |
oling | +0.145 |
Neg logit contributions | |
ascript | -0.174 |
lect | -0.161 |
spir | -0.159 |
NK | -0.156 |
Roll | -0.154 |
Token | TABLE |
Token ID | 38148 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
姫 | +0.532 |
� | +0.515 |
andr | +0.498 |
ennial | +0.488 |
oling | +0.485 |
Neg logit contributions | |
ascript | -0.556 |
lect | -0.487 |
spir | -0.477 |
Roll | -0.465 |
NK | -0.460 |
Token | 5 |
Token ID | 642 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
姫 | +5.003 |
� | +4.844 |
sand | +4.471 |
PLE | +4.470 |
jan | +4.419 |
Neg logit contributions | |
ascript | -8.603 |
spir | -7.261 |
lect | -7.049 |
hemy | -6.995 |
glers | -6.987 |
Token |
|
Token ID | 198 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
姫 | +2.117 |
� | +2.056 |
andr | +1.982 |
ennial | +1.929 |
sand | +1.927 |
Neg logit contributions | |
ascript | -2.603 |
lect | -2.242 |
spir | -2.212 |
NK | -2.165 |
Roll | -2.121 |
Token |
|
Token ID | 198 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
姫 | +1.637 |
� | +1.586 |
andr | +1.510 |
ennial | +1.467 |
andestine | +1.466 |
Neg logit contributions | |
ascript | -2.117 |
lect | -1.907 |
spir | -1.876 |
NK | -1.848 |
Roll | -1.818 |
Token | Table |
Token ID | 10962 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
姫 | +1.482 |
� | +1.440 |
andr | +1.394 |
ennial | +1.358 |
sand | +1.357 |
Neg logit contributions | |
ascript | -1.574 |
lect | -1.351 |
spir | -1.327 |
NK | -1.298 |
Roll | -1.273 |
Token | 5 |
Token ID | 642 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
姫 | +2.896 |
� | +2.832 |
ennial | +2.636 |
encia | +2.626 |
sand | +2.612 |
Neg logit contributions | |
ascript | -4.653 |
spir | -4.025 |
NK | -3.875 |
lect | -3.875 |
Citiz | -3.860 |
Token | aff |
Token ID | 2001 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
姫 | +1.943 |
� | +1.880 |
andr | +1.819 |
sand | +1.763 |
ennial | +1.760 |
Neg logit contributions | |
ascript | -1.967 |
lect | -1.694 |
spir | -1.660 |
NK | -1.625 |
Roll | -1.587 |
Token | in |
Token ID | 259 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
姫 | +0.551 |
� | +0.533 |
andr | +0.516 |
sand | +0.498 |
andestine | +0.490 |
Neg logit contributions | |
ascript | -0.602 |
lect | -0.531 |
spir | -0.516 |
NK | -0.507 |
Roll | -0.503 |
Token | Bay |
Token ID | 4696 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
姫 | +2.964 |
� | +2.874 |
andr | +2.786 |
sand | +2.692 |
ennial | +2.681 |
Neg logit contributions | |
ascript | -3.364 |
lect | -2.901 |
spir | -2.846 |
NK | -2.810 |
Roll | -2.740 |
Token | and |
Token ID | 290 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
姫 | +2.877 |
� | +2.783 |
andr | +2.694 |
ennial | +2.603 |
sand | +2.602 |
Neg logit contributions | |
ascript | -3.699 |
lect | -3.216 |
spir | -3.164 |
NK | -3.114 |
Roll | -3.042 |
Token | Davis |
Token ID | 7802 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
姫 | +2.470 |
� | +2.397 |
andr | +2.321 |
ennial | +2.257 |
sand | +2.253 |
Neg logit contributions | |
ascript | -2.723 |
lect | -2.342 |
spir | -2.296 |
NK | -2.267 |
Roll | -2.208 |
Token | Strait |
Token ID | 41407 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
ascript | +0.605 |
lect | +0.520 |
spir | +0.507 |
NK | +0.502 |
Roll | +0.490 |
Neg logit contributions | |
姫 | -0.655 |
� | -0.629 |
andr | -0.615 |
andestine | -0.590 |
sand | -0.590 |
Token | , |
Token ID | 11 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
姫 | +5.477 |
� | +5.325 |
sand | +5.049 |
andr | +5.025 |
ennial | +5.011 |
Neg logit contributions | |
ascript | -7.893 |
lect | -6.809 |
spir | -6.765 |
NK | -6.545 |
glers | -6.533 |
Token | a |
Token ID | 257 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
姫 | +2.296 |
� | +2.225 |
andr | +2.161 |
sand | +2.100 |
ennial | +2.100 |
Neg logit contributions | |
ascript | -2.417 |
lect | -2.067 |
spir | -2.030 |
NK | -2.004 |
Roll | -1.945 |
Token | prepar |
Token ID | 9198 |
Feature activation | +9.15e-01 |
Pos logit contributions | |
姫 | +1.120 |
� | +1.074 |
andr | +1.051 |
andestine | +1.010 |
ennial | +1.009 |
Neg logit contributions | |
ascript | -1.174 |
lect | -1.022 |
NK | -0.996 |
spir | -0.985 |
Roll | -0.948 |
Token | atory |
Token ID | 2870 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
姫 | +1.356 |
� | +1.293 |
andr | +1.199 |
ennial | +1.089 |
sand | +1.072 |
Neg logit contributions | |
ascript | -7.481 |
lect | -6.761 |
spir | -6.751 |
NK | -6.688 |
Roll | -6.586 |
Token | step |
Token ID | 2239 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
姫 | +2.540 |
� | +2.441 |
andr | +2.356 |
ennial | +2.281 |
sand | +2.274 |
Neg logit contributions | |
ascript | -2.742 |
lect | -2.395 |
NK | -2.335 |
spir | -2.318 |
Roll | -2.236 |
Token | We |
Token ID | 775 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
ascript | +0.077 |
lect | +0.068 |
NK | +0.067 |
spir | +0.067 |
Roll | +0.066 |
Neg logit contributions | |
姫 | -0.081 |
� | -0.078 |
andr | -0.076 |
oling | -0.075 |
ennial | -0.075 |
Token | want |
Token ID | 765 |
Feature activation | +4.94e-02 |
Pos logit contributions | |
姫 | +0.973 |
� | +0.937 |
andr | +0.909 |
ennial | +0.889 |
sand | +0.885 |
Neg logit contributions | |
ascript | -1.119 |
lect | -0.968 |
spir | -0.959 |
NK | -0.934 |
Roll | -0.924 |
Token | to |
Token ID | 284 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
姫 | +0.210 |
� | +0.202 |
andr | +0.195 |
ennial | +0.189 |
sand | +0.189 |
Neg logit contributions | |
ascript | -0.276 |
lect | -0.241 |
spir | -0.237 |
NK | -0.233 |
Roll | -0.228 |
Token | thank |
Token ID | 5875 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
姫 | +1.147 |
� | +1.076 |
andr | +1.024 |
ennial | +1.002 |
andestine | +0.998 |
Neg logit contributions | |
ascript | -2.063 |
lect | -1.865 |
spir | -1.829 |
NK | -1.812 |
Roll | -1.782 |
Token | each |
Token ID | 1123 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
姫 | +1.471 |
� | +1.417 |
andr | +1.374 |
ennial | +1.337 |
sand | +1.331 |
Neg logit contributions | |
ascript | -1.601 |
lect | -1.384 |
spir | -1.354 |
NK | -1.337 |
Roll | -1.306 |
Token | and |
Token ID | 290 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
姫 | +1.304 |
� | +1.273 |
andr | +1.232 |
ennial | +1.197 |
sand | +1.189 |
Neg logit contributions | |
ascript | -1.788 |
lect | -1.545 |
spir | -1.525 |
NK | -1.514 |
Roll | -1.467 |
Token | every |
Token ID | 790 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
� | +3.460 |
姫 | +3.414 |
ennial | +3.090 |
andr | +2.983 |
sand | +2.946 |
Neg logit contributions | |
ascript | -9.761 |
spir | -8.482 |
NK | -8.385 |
lect | -8.368 |
glers | -8.360 |
Token | Unt |
Token ID | 26970 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
姫 | +3.028 |
� | +2.965 |
andr | +2.865 |
sand | +2.807 |
ennial | +2.795 |
Neg logit contributions | |
ascript | -3.114 |
lect | -2.623 |
spir | -2.594 |
NK | -2.545 |
Roll | -2.466 |
Token | app |
Token ID | 1324 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
姫 | +2.582 |
� | +2.519 |
andr | +2.439 |
ennial | +2.387 |
oling | +2.379 |
Neg logit contributions | |
ascript | -1.149 |
lect | -0.892 |
spir | -0.856 |
NK | -0.809 |
Roll | -0.787 |
Token | d |
Token ID | 67 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
姫 | +0.251 |
� | +0.242 |
andr | +0.230 |
oling | +0.229 |
andestine | +0.225 |
Neg logit contributions | |
ascript | -0.286 |
lect | -0.256 |
spir | -0.253 |
Roll | -0.242 |
NK | -0.236 |
Token | user |
Token ID | 2836 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
姫 | +0.950 |
� | +0.919 |
andr | +0.889 |
sand | +0.862 |
ennial | +0.862 |
Neg logit contributions | |
ascript | -1.148 |
lect | -1.005 |
spir | -0.989 |
NK | -0.963 |
Roll | -0.952 |
Token | Minister |
Token ID | 4139 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
姫 | +1.498 |
� | +1.454 |
andr | +1.407 |
ennial | +1.375 |
sand | +1.373 |
Neg logit contributions | |
ascript | -1.693 |
lect | -1.464 |
spir | -1.438 |
NK | -1.418 |
Roll | -1.385 |
Token | S |
Token ID | 311 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
ascript | +3.112 |
NK | +3.107 |
lect | +2.954 |
spir | +2.850 |
Roll | +2.776 |
Neg logit contributions | |
姫 | -3.191 |
� | -2.987 |
ennial | -2.917 |
andestine | -2.893 |
encia | -2.866 |
Token | ures |
Token ID | 942 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
ascript | +1.103 |
lect | +0.951 |
NK | +0.936 |
spir | +0.927 |
Roll | +0.877 |
Neg logit contributions | |
姫 | -1.415 |
� | -1.332 |
andr | -1.310 |
ennial | -1.288 |
andestine | -1.273 |
Token | h |
Token ID | 71 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
姫 | +0.679 |
� | +0.620 |
andestine | +0.605 |
ennial | +0.605 |
andr | +0.591 |
Neg logit contributions | |
ascript | -0.497 |
lect | -0.454 |
NK | -0.437 |
spir | -0.437 |
pen | -0.407 |
Token | Pr |
Token ID | 1736 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
ascript | +2.278 |
NK | +2.209 |
lect | +2.090 |
Roll | +2.060 |
spir | +2.058 |
Neg logit contributions | |
姫 | -2.174 |
ennial | -2.004 |
� | -1.978 |
andr | -1.913 |
andestine | -1.900 |
Token | ab |
Token ID | 397 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ascript | +0.043 |
lect | +0.037 |
spir | +0.036 |
NK | +0.035 |
Roll | +0.034 |
Neg logit contributions | |
姫 | -0.043 |
� | -0.040 |
andr | -0.038 |
ennial | -0.038 |
oling | -0.037 |
Token | hu |
Token ID | 13415 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
� | +1.940 |
姫 | +1.876 |
andr | +1.812 |
oling | +1.635 |
sand | +1.633 |
Neg logit contributions | |
ascript | -11.346 |
lect | -10.183 |
spir | -10.093 |
Citiz | -9.995 |
Roll | -9.929 |
Token | estimates |
Token ID | 7746 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
ascript | +0.647 |
lect | +0.572 |
spir | +0.561 |
NK | +0.555 |
Roll | +0.542 |
Neg logit contributions | |
姫 | -0.453 |
� | -0.432 |
andr | -0.418 |
ennial | -0.412 |
sand | -0.408 |
Token | the |
Token ID | 262 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
ascript | +1.111 |
lect | +0.960 |
spir | +0.941 |
NK | +0.939 |
Roll | +0.903 |
Neg logit contributions | |
姫 | -1.202 |
� | -1.154 |
andr | -1.125 |
sand | -1.100 |
andestine | -1.091 |
Token | return |
Token ID | 1441 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
ascript | +1.208 |
lect | +1.066 |
NK | +1.065 |
spir | +1.030 |
FTA | +1.015 |
Neg logit contributions | |
姫 | -1.241 |
� | -1.172 |
andr | -1.142 |
sand | -1.113 |
andestine | -1.111 |
Token | to |
Token ID | 284 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
ascript | +1.752 |
lect | +1.549 |
NK | +1.515 |
spir | +1.514 |
Roll | +1.472 |
Neg logit contributions | |
姫 | -2.022 |
� | -1.916 |
andr | -1.885 |
andestine | -1.832 |
sand | -1.795 |
Token | graph |
Token ID | 4823 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
ascript | +1.102 |
lect | +0.981 |
NK | +0.962 |
spir | +0.925 |
Roll | +0.902 |
Neg logit contributions | |
姫 | -1.114 |
� | -1.030 |
andr | -1.007 |
ennial | -1.001 |
oling | -0.984 |
Token | = |
Token ID | 796 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
姫 | +1.092 |
� | +1.057 |
andr | +1.018 |
ennial | +0.988 |
sand | +0.988 |
Neg logit contributions | |
ascript | -1.490 |
lect | -1.297 |
spir | -1.278 |
NK | -1.257 |
Roll | -1.230 |
Token | Fl |
Token ID | 1610 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
ascript | +1.965 |
lect | +1.743 |
NK | +1.700 |
spir | +1.605 |
Roll | +1.581 |
Neg logit contributions | |
姫 | -2.477 |
� | -2.299 |
andr | -2.269 |
ennial | -2.258 |
andestine | -2.216 |
Token | atten |
Token ID | 41769 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
ascript | +0.785 |
lect | +0.678 |
spir | +0.642 |
NK | +0.623 |
Roll | +0.620 |
Neg logit contributions | |
姫 | -1.022 |
� | -0.979 |
andr | -0.948 |
ennial | -0.923 |
sand | -0.923 |
Token | [ |
Token ID | 58 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
姫 | +2.395 |
� | +2.366 |
andr | +2.255 |
sand | +2.210 |
ennial | +2.187 |
Neg logit contributions | |
ascript | -3.624 |
lect | -3.150 |
spir | -3.117 |
NK | -3.062 |
glers | -2.994 |
Token | Table |
Token ID | 8655 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
姫 | +2.210 |
� | +2.157 |
andr | +2.084 |
sand | +2.036 |
ennial | +2.028 |
Neg logit contributions | |
ascript | -2.410 |
lect | -2.060 |
spir | -2.030 |
NK | -1.987 |
Roll | -1.941 |
Token | [ |
Token ID | 58 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
� | +3.919 |
姫 | +3.765 |
sand | +3.557 |
andr | +3.489 |
ennial | +3.438 |
Neg logit contributions | |
ascript | -8.592 |
spir | -7.605 |
glers | -7.576 |
NK | -7.515 |
lect | -7.457 |
Token | Table |
Token ID | 10962 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
姫 | +2.366 |
� | +2.323 |
andr | +2.229 |
sand | +2.193 |
ennial | +2.167 |
Neg logit contributions | |
ascript | -2.866 |
lect | -2.459 |
spir | -2.430 |
NK | -2.371 |
Roll | -2.329 |
Token | [ |
Token ID | 58 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
� | +3.345 |
姫 | +3.314 |
sand | +3.091 |
andr | +3.089 |
jan | +3.050 |
Neg logit contributions | |
ascript | -5.935 |
spir | -5.138 |
lect | -5.111 |
NK | -5.023 |
glers | -5.017 |
Token | graph |
Token ID | 4823 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
姫 | +1.507 |
� | +1.462 |
andr | +1.414 |
ennial | +1.376 |
sand | +1.375 |
Neg logit contributions | |
ascript | -1.739 |
lect | -1.497 |
spir | -1.472 |
NK | -1.446 |
Roll | -1.412 |
Token | List |
Token ID | 8053 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
姫 | +1.463 |
� | +1.419 |
andr | +1.371 |
ennial | +1.334 |
sand | +1.333 |
Neg logit contributions | |
ascript | -1.741 |
lect | -1.502 |
spir | -1.478 |
NK | -1.452 |
Roll | -1.419 |
Token | EL |
Token ID | 17852 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
姫 | +0.430 |
� | +0.414 |
andr | +0.402 |
ennial | +0.394 |
sand | +0.392 |
Neg logit contributions | |
ascript | -0.394 |
lect | -0.333 |
spir | -0.326 |
NK | -0.325 |
Roll | -0.311 |
Token | F |
Token ID | 37 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
姫 | +0.754 |
� | +0.728 |
andr | +0.711 |
oling | +0.694 |
andestine | +0.690 |
Neg logit contributions | |
ascript | -0.818 |
lect | -0.725 |
spir | -0.713 |
NK | -0.699 |
Roll | -0.690 |
Token | _ |
Token ID | 62 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
姫 | +1.238 |
� | +1.205 |
andr | +1.159 |
oling | +1.132 |
andestine | +1.127 |
Neg logit contributions | |
ascript | -1.470 |
lect | -1.301 |
spir | -1.268 |
NK | -1.254 |
Roll | -1.236 |
Token | UM |
Token ID | 5883 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
姫 | +3.072 |
� | +3.008 |
andr | +2.909 |
sand | +2.866 |
ennial | +2.833 |
Neg logit contributions | |
ascript | -3.432 |
lect | -2.893 |
spir | -2.857 |
NK | -2.836 |
Citiz | -2.751 |
Token | RE |
Token ID | 2200 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
ascript | +1.372 |
lect | +1.228 |
spir | +1.170 |
Roll | +1.170 |
NK | +1.167 |
Neg logit contributions | |
姫 | -1.460 |
� | -1.400 |
andr | -1.381 |
oling | -1.366 |
andestine | -1.346 |
Token | ON |
Token ID | 1340 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
ascript | +2.537 |
lect | +2.291 |
Roll | +2.186 |
NK | +2.184 |
spir | +2.173 |
Neg logit contributions | |
姫 | -2.164 |
� | -2.043 |
andr | -2.005 |
andestine | -1.984 |
sand | -1.958 |
Token | . |
Token ID | 13 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
� | +4.956 |
姫 | +4.899 |
andr | +4.654 |
sand | +4.631 |
jan | +4.586 |
Neg logit contributions | |
ascript | -7.646 |
spir | -6.640 |
Citiz | -6.595 |
lect | -6.470 |
glers | -6.392 |
Token | B |
Token ID | 33 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
姫 | +1.915 |
� | +1.867 |
andr | +1.802 |
sand | +1.769 |
ennial | +1.746 |
Neg logit contributions | |
ascript | -2.230 |
lect | -1.908 |
spir | -1.881 |
NK | -1.846 |
Roll | -1.795 |
Token | ) |
Token ID | 8 |
Feature activation | +7.90e-02 |
Pos logit contributions | |
姫 | +1.634 |
� | +1.592 |
andr | +1.531 |
sand | +1.491 |
jan | +1.477 |
Neg logit contributions | |
ascript | -2.119 |
lect | -1.849 |
spir | -1.829 |
NK | -1.773 |
Citiz | -1.759 |
Token |
|
Token ID | 198 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
姫 | +0.287 |
� | +0.277 |
andr | +0.265 |
sand | +0.256 |
ennial | +0.254 |
Neg logit contributions | |
ascript | -0.486 |
lect | -0.430 |
spir | -0.425 |
NK | -0.415 |
Roll | -0.409 |
Token |
|
Token ID | 198 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
姫 | +1.404 |
� | +1.354 |
andr | +1.290 |
ennial | +1.250 |
andestine | +1.249 |
Neg logit contributions | |
ascript | -1.806 |
lect | -1.625 |
spir | -1.597 |
NK | -1.571 |
Roll | -1.544 |
Token | In |
Token ID | 554 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
ascript | +0.489 |
lect | +0.437 |
NK | +0.431 |
spir | +0.428 |
Roll | +0.423 |
Neg logit contributions | |
姫 | -0.423 |
� | -0.404 |
andr | -0.395 |
andestine | -0.388 |
oling | -0.388 |
Token | addition |
Token ID | 3090 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
姫 | +0.808 |
� | +0.775 |
andr | +0.756 |
sand | +0.732 |
ennial | +0.727 |
Neg logit contributions | |
ascript | -0.865 |
lect | -0.748 |
spir | -0.731 |
NK | -0.719 |
Roll | -0.702 |
Token | , |
Token ID | 11 |
Feature activation | +5.57e-02 |
Pos logit contributions | |
ascript | +1.956 |
lect | +1.713 |
spir | +1.683 |
NK | +1.660 |
Roll | +1.626 |
Neg logit contributions | |
姫 | -1.342 |
� | -1.293 |
andr | -1.248 |
sand | -1.210 |
ennial | -1.206 |
Token | a |
Token ID | 257 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
姫 | +0.261 |
� | +0.251 |
andr | +0.245 |
sand | +0.239 |
ennial | +0.238 |
Neg logit contributions | |
ascript | -0.284 |
lect | -0.247 |
spir | -0.240 |
NK | -0.239 |
Roll | -0.233 |
Token | Quad |
Token ID | 20648 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
ascript | +0.254 |
lect | +0.223 |
spir | +0.215 |
NK | +0.213 |
Roll | +0.208 |
Neg logit contributions | |
姫 | -0.241 |
� | -0.232 |
andr | -0.227 |
andestine | -0.222 |
sand | -0.220 |
Token | ren |
Token ID | 918 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ascript | +0.050 |
lect | +0.042 |
spir | +0.041 |
NK | +0.041 |
Roll | +0.040 |
Neg logit contributions | |
姫 | -0.064 |
� | -0.061 |
andr | -0.059 |
sand | -0.058 |
oling | -0.057 |
Token | n |
Token ID | 77 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
ascript | +6.575 |
lect | +6.281 |
spir | +6.050 |
NK | +5.642 |
forth | +5.495 |
Neg logit contributions | |
姫 | -8.705 |
� | -8.050 |
andestine | -7.675 |
ゼウス | -7.638 |
andr | -7.637 |
Token | ial |
Token ID | 498 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
ascript | +1.045 |
lect | +0.877 |
spir | +0.863 |
NK | +0.806 |
Roll | +0.776 |
Neg logit contributions | |
姫 | -2.161 |
� | -2.060 |
andestine | -1.957 |
encia | -1.957 |
oling | -1.953 |
Token | Energy |
Token ID | 6682 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
ascript | +1.021 |
lect | +0.902 |
spir | +0.893 |
NK | +0.884 |
Roll | +0.856 |
Neg logit contributions | |
姫 | -1.201 |
� | -1.165 |
andr | -1.120 |
andestine | -1.119 |
oling | -1.107 |
Token | Review |
Token ID | 6602 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
ascript | +0.885 |
lect | +0.761 |
spir | +0.740 |
NK | +0.740 |
Roll | +0.722 |
Neg logit contributions | |
姫 | -1.018 |
� | -0.996 |
andr | -0.956 |
andestine | -0.942 |
oling | -0.936 |
Token | — |
Token ID | 960 |
Feature activation | +2.40e-03 |
Pos logit contributions | |
ascript | +2.518 |
lect | +2.313 |
NK | +2.238 |
FTA | +2.178 |
Roll | +2.150 |
Neg logit contributions | |
姫 | -2.814 |
� | -2.672 |
andr | -2.605 |
andestine | -2.556 |
sand | -2.490 |
Token | Smart |
Token ID | 25610 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
ascript | +0.365 |
lect | +0.317 |
spir | +0.309 |
Roll | +0.300 |
NK | +0.300 |
Neg logit contributions | |
姫 | -0.347 |
� | -0.335 |
andr | -0.324 |
ennial | -0.315 |
oling | -0.313 |
Token | " |
Token ID | 1 |
Feature activation | +1.16e-04 |
Pos logit contributions | |
ascript | +0.755 |
lect | +0.652 |
spir | +0.627 |
NK | +0.626 |
Roll | +0.617 |
Neg logit contributions | |
姫 | -0.778 |
� | -0.751 |
andr | -0.725 |
ennial | -0.710 |
sand | -0.706 |
Token | tool |
Token ID | 2891 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
姫 | +0.001 |
� | +0.001 |
andr | +0.001 |
ennial | +0.001 |
sand | +0.001 |
Neg logit contributions | |
ascript | -0.001 |
lect | -0.000 |
NK | -0.000 |
spir | -0.000 |
Roll | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
姫 | +0.854 |
� | +0.827 |
andr | +0.799 |
ennial | +0.776 |
sand | +0.768 |
Neg logit contributions | |
ascript | -0.991 |
lect | -0.861 |
spir | -0.837 |
NK | -0.834 |
Roll | -0.811 |
Token | cross |
Token ID | 3272 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
ascript | +0.641 |
lect | +0.555 |
NK | +0.542 |
spir | +0.536 |
Roll | +0.527 |
Neg logit contributions | |
姫 | -0.598 |
� | -0.574 |
andr | -0.556 |
ennial | -0.545 |
sand | -0.541 |
Token | f |
Token ID | 69 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ascript | +2.102 |
lect | +1.881 |
spir | +1.790 |
NK | +1.783 |
Roll | +1.754 |
Neg logit contributions | |
姫 | -1.555 |
� | -1.430 |
andr | -1.371 |
ennial | -1.332 |
andestine | -1.318 |
Token | ades |
Token ID | 2367 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
lect | +6.074 |
ascript | +6.017 |
NK | +5.032 |
spir | +4.910 |
Roll | +4.647 |
Neg logit contributions | |
姫 | -9.630 |
� | -8.879 |
PLE | -8.220 |
sandy | -8.151 |
ギ | -8.113 |
Token | , |
Token ID | 11 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
ascript | +2.929 |
lect | +2.577 |
NK | +2.491 |
Roll | +2.429 |
spir | +2.414 |
Neg logit contributions | |
姫 | -2.595 |
� | -2.453 |
andr | -2.369 |
sand | -2.332 |
andestine | -2.317 |
Token | tim |
Token ID | 4628 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
ascript | +2.397 |
lect | +2.102 |
NK | +2.070 |
spir | +2.017 |
Roll | +2.002 |
Neg logit contributions | |
姫 | -2.223 |
� | -2.108 |
andr | -2.026 |
ennial | -2.004 |
sand | -1.982 |
Token | est |
Token ID | 395 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
ascript | +0.100 |
lect | +0.084 |
spir | +0.082 |
NK | +0.080 |
Roll | +0.078 |
Neg logit contributions | |
姫 | -0.121 |
� | -0.116 |
andr | -0.112 |
ennial | -0.110 |
sand | -0.109 |
Token | retch |
Token ID | 22592 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
ascript | +0.866 |
lect | +0.717 |
spir | +0.689 |
NK | +0.668 |
Roll | +0.654 |
Neg logit contributions | |
姫 | -1.462 |
� | -1.423 |
andr | -1.367 |
sand | -1.339 |
encia | -1.334 |
Token | campuses |
Token ID | 23003 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
姫 | +1.507 |
� | +1.444 |
andr | +1.381 |
ennial | +1.329 |
sand | +1.328 |
Neg logit contributions | |
ascript | -2.884 |
lect | -2.560 |
spir | -2.526 |
NK | -2.492 |
Roll | -2.444 |
Token | on |
Token ID | 319 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
姫 | +0.556 |
� | +0.535 |
andr | +0.520 |
sand | +0.505 |
ennial | +0.504 |
Neg logit contributions | |
ascript | -0.627 |
lect | -0.543 |
spir | -0.534 |
NK | -0.525 |
Roll | -0.512 |
Token | both |
Token ID | 1111 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
姫 | +0.558 |
� | +0.526 |
andr | +0.517 |
ennial | +0.500 |
sand | +0.499 |
Neg logit contributions | |
ascript | -0.610 |
lect | -0.532 |
spir | -0.520 |
NK | -0.515 |
Roll | -0.498 |
Token | sides |
Token ID | 5389 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
ascript | +1.308 |
lect | +1.170 |
NK | +1.149 |
spir | +1.113 |
Roll | +1.066 |
Neg logit contributions | |
姫 | -1.753 |
andr | -1.654 |
� | -1.641 |
ennial | -1.598 |
sand | -1.593 |
Token | of |
Token ID | 286 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ascript | +1.645 |
lect | +1.422 |
spir | +1.399 |
NK | +1.375 |
Roll | +1.343 |
Neg logit contributions | |
姫 | -1.349 |
� | -1.308 |
andr | -1.263 |
ennial | -1.229 |
sand | -1.228 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ascript | +5.563 |
lect | +4.614 |
spir | +4.429 |
NK | +4.369 |
Roll | +4.243 |
Neg logit contributions | |
姫 | -7.184 |
andr | -6.912 |
� | -6.769 |
encia | -6.596 |
sand | -6.565 |
Token | Atlantic |
Token ID | 10596 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
ascript | +6.682 |
lect | +5.752 |
Roll | +5.390 |
Compass | +5.381 |
NK | +5.277 |
Neg logit contributions | |
姫 | -7.624 |
encia | -6.969 |
andr | -6.872 |
� | -6.830 |
andestine | -6.685 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ascript | +4.120 |
lect | +3.578 |
spir | +3.545 |
NK | +3.421 |
Roll | +3.395 |
Neg logit contributions | |
姫 | -4.549 |
� | -4.363 |
andr | -4.307 |
sand | -4.106 |
encia | -4.093 |
Token |
|
Token ID | 198 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
ascript | +0.574 |
lect | +0.507 |
spir | +0.500 |
NK | +0.493 |
Roll | +0.483 |
Neg logit contributions | |
姫 | -0.455 |
� | -0.435 |
andr | -0.425 |
oling | -0.412 |
ennial | -0.412 |
Token |
|
Token ID | 198 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
姫 | +0.638 |
� | +0.621 |
andr | +0.591 |
andestine | +0.578 |
oling | +0.577 |
Neg logit contributions | |
ascript | -0.734 |
lect | -0.671 |
spir | -0.662 |
NK | -0.650 |
Roll | -0.643 |
Token | After |
Token ID | 3260 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
姫 | +0.654 |
� | +0.630 |
andr | +0.613 |
ennial | +0.597 |
oling | +0.595 |
Neg logit contributions | |
ascript | -0.676 |
lect | -0.594 |
spir | -0.583 |
Roll | -0.566 |
NK | -0.561 |
Token | lawmakers |
Token ID | 10191 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
ascript | +0.227 |
lect | +0.196 |
spir | +0.195 |
NK | +0.189 |
Roll | +0.184 |
Neg logit contributions | |
姫 | -0.205 |
� | -0.196 |
andr | -0.192 |
ennial | -0.187 |
sand | -0.185 |
Token | from |
Token ID | 422 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
ascript | +0.591 |
lect | +0.512 |
spir | +0.503 |
NK | +0.494 |
Roll | +0.484 |
Neg logit contributions | |
姫 | -0.585 |
� | -0.563 |
andr | -0.547 |
sand | -0.535 |
ennial | -0.533 |
Token | both |
Token ID | 1111 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
ascript | +0.119 |
lect | +0.101 |
spir | +0.098 |
NK | +0.096 |
Roll | +0.094 |
Neg logit contributions | |
姫 | -0.156 |
� | -0.150 |
andr | -0.148 |
ennial | -0.144 |
sand | -0.143 |
Token | sides |
Token ID | 5389 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
ascript | +1.451 |
lect | +1.301 |
NK | +1.230 |
spir | +1.221 |
Roll | +1.194 |
Neg logit contributions | |
姫 | -1.853 |
� | -1.761 |
andr | -1.751 |
oling | -1.685 |
ennial | -1.681 |
Token | of |
Token ID | 286 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ascript | +1.328 |
lect | +1.140 |
spir | +1.131 |
NK | +1.078 |
Roll | +1.062 |
Neg logit contributions | |
姫 | -1.269 |
� | -1.210 |
andr | -1.190 |
sand | -1.143 |
ennial | -1.143 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ascript | +5.107 |
lect | +4.317 |
spir | +4.123 |
Roll | +4.033 |
NK | +3.992 |
Neg logit contributions | |
姫 | -6.985 |
andr | -6.795 |
� | -6.680 |
andestine | -6.504 |
encia | -6.430 |
Token | aisle |
Token ID | 28960 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
ascript | +4.340 |
lect | +3.648 |
Roll | +3.232 |
spir | +3.131 |
Compass | +3.048 |
Neg logit contributions | |
姫 | -9.542 |
andr | -9.113 |
encia | -9.016 |
andestine | -8.980 |
� | -8.961 |
Token | said |
Token ID | 531 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
ascript | +2.761 |
lect | +2.386 |
spir | +2.372 |
NK | +2.262 |
Roll | +2.254 |
Neg logit contributions | |
姫 | -2.906 |
� | -2.770 |
andr | -2.723 |
ennial | -2.630 |
andestine | -2.620 |
Token | they |
Token ID | 484 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
姫 | +0.455 |
� | +0.441 |
andr | +0.430 |
ennial | +0.418 |
sand | +0.418 |
Neg logit contributions | |
ascript | -0.409 |
lect | -0.347 |
spir | -0.341 |
NK | -0.332 |
Roll | -0.324 |
Token | are |
Token ID | 389 |
Feature activation | +4.11e-03 |
Pos logit contributions | |
姫 | +0.649 |
� | +0.631 |
andr | +0.610 |
ennial | +0.596 |
sand | +0.595 |
Neg logit contributions | |
ascript | -0.703 |
lect | -0.605 |
spir | -0.599 |
NK | -0.586 |
Roll | -0.569 |
Token | receiving |
Token ID | 6464 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
姫 | +0.021 |
� | +0.020 |
andr | +0.020 |
ennial | +0.020 |
sand | +0.019 |
Neg logit contributions | |
ascript | -0.019 |
lect | -0.016 |
spir | -0.016 |
NK | -0.016 |
Roll | -0.015 |
Token | Gram |
Token ID | 20159 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
ascript | +0.499 |
lect | +0.462 |
spir | +0.427 |
NK | +0.420 |
Roll | +0.415 |
Neg logit contributions | |
姫 | -1.080 |
� | -1.038 |
andestine | -1.015 |
oling | -0.995 |
andr | -0.994 |
Token | mer |
Token ID | 647 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
ascript | +1.726 |
spir | +1.597 |
lect | +1.595 |
pen | +1.474 |
Roll | +1.453 |
Neg logit contributions | |
姫 | -2.246 |
� | -2.173 |
andestine | -2.016 |
andr | -1.995 |
extinction | -1.985 |
Token |
|
Token ID | 198 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
姫 | +0.652 |
� | +0.636 |
andr | +0.608 |
oling | +0.590 |
andestine | +0.585 |
Neg logit contributions | |
ascript | -0.804 |
lect | -0.727 |
spir | -0.710 |
NK | -0.706 |
Roll | -0.692 |
Token |
|
Token ID | 198 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
ascript | +0.991 |
lect | +0.948 |
spir | +0.923 |
NK | +0.912 |
Roll | +0.901 |
Neg logit contributions | |
姫 | -1.072 |
� | -1.047 |
andr | -0.986 |
andestine | -0.977 |
oling | -0.965 |
Token | 1999 |
Token ID | 18946 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
lect | +2.640 |
ascript | +2.616 |
Roll | +2.425 |
spir | +2.388 |
Factor | +2.344 |
Neg logit contributions | |
姫 | -4.487 |
andestine | -4.431 |
andr | -4.381 |
� | -4.356 |
extinction | -4.291 |
Token | Paul |
Token ID | 3362 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ascript | +2.772 |
lect | +2.519 |
NK | +2.445 |
spir | +2.395 |
Roll | +2.369 |
Neg logit contributions | |
姫 | -2.510 |
� | -2.355 |
andr | -2.270 |
andestine | -2.213 |
sand | -2.210 |
Token | Re |
Token ID | 797 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
NK | +4.348 |
lect | +4.215 |
ascript | +4.089 |
Roll | +3.915 |
Roll | +3.897 |
Neg logit contributions | |
姫 | -8.940 |
oling | -8.086 |
andestine | -8.067 |
�� | -8.006 |
� | -7.862 |
Token | iser |
Token ID | 5847 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ascript | +2.578 |
lect | +2.333 |
spir | +2.303 |
Roll | +2.136 |
live | +1.969 |
Neg logit contributions | |
姫 | -5.026 |
� | -4.642 |
extinction | -4.432 |
andestine | -4.404 |
sandy | -4.404 |
Token |
|
Token ID | 198 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
姫 | +1.019 |
� | +0.989 |
andr | +0.953 |
oling | +0.919 |
ennial | +0.915 |
Neg logit contributions | |
ascript | -1.327 |
lect | -1.191 |
spir | -1.165 |
NK | -1.160 |
Roll | -1.135 |
Token |
|
Token ID | 198 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
ascript | +0.855 |
lect | +0.814 |
spir | +0.794 |
NK | +0.783 |
Roll | +0.774 |
Neg logit contributions | |
姫 | -0.901 |
� | -0.879 |
andr | -0.829 |
andestine | -0.818 |
oling | -0.808 |
Token | 1999 |
Token ID | 18946 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
lect | +2.665 |
ascript | +2.606 |
Roll | +2.467 |
spir | +2.426 |
Factor | +2.370 |
Neg logit contributions | |
姫 | -4.456 |
andestine | -4.390 |
andr | -4.328 |
� | -4.299 |
extinction | -4.282 |
Token | not |
Token ID | 407 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
姫 | +0.085 |
� | +0.081 |
andr | +0.079 |
ennial | +0.077 |
sand | +0.077 |
Neg logit contributions | |
ascript | -0.093 |
lect | -0.080 |
spir | -0.079 |
NK | -0.079 |
Roll | -0.076 |
Token | yet |
Token ID | 1865 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ascript | +1.187 |
lect | +1.031 |
spir | +1.017 |
NK | +1.000 |
Roll | +0.954 |
Neg logit contributions | |
姫 | -1.140 |
� | -1.108 |
andr | -1.082 |
oling | -1.046 |
andestine | -1.043 |
Token | been |
Token ID | 587 |
Feature activation | +8.79e-02 |
Pos logit contributions | |
ascript | +0.960 |
lect | +0.846 |
spir | +0.835 |
NK | +0.833 |
Roll | +0.797 |
Neg logit contributions | |
姫 | -0.794 |
� | -0.761 |
andr | -0.743 |
ennial | -0.718 |
sand | -0.709 |
Token | implemented |
Token ID | 9177 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
姫 | +0.410 |
� | +0.396 |
andr | +0.386 |
ennial | +0.377 |
oling | +0.372 |
Neg logit contributions | |
ascript | -0.454 |
lect | -0.393 |
spir | -0.391 |
NK | -0.385 |
Roll | -0.372 |
Token | , |
Token ID | 11 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
ascript | +0.842 |
lect | +0.720 |
NK | +0.702 |
spir | +0.700 |
Roll | +0.678 |
Neg logit contributions | |
姫 | -0.869 |
� | -0.848 |
andr | -0.813 |
ennial | -0.796 |
sand | -0.792 |
Token | but |
Token ID | 475 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ascript | +1.841 |
NK | +1.576 |
lect | +1.573 |
spir | +1.553 |
Roll | +1.501 |
Neg logit contributions | |
姫 | -1.866 |
� | -1.759 |
andr | -1.702 |
ennial | -1.669 |
oling | -1.654 |
Token | fortunately |
Token ID | 39955 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
ascript | +5.596 |
NK | +5.348 |
lect | +4.974 |
Roll | +4.936 |
FTA | +4.864 |
Neg logit contributions | |
姫 | -7.469 |
� | -7.004 |
ennial | -6.751 |
andr | -6.742 |
encia | -6.694 |
Token | the |
Token ID | 262 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
ascript | +3.500 |
NK | +3.230 |
lect | +3.138 |
spir | +3.055 |
Roll | +3.006 |
Neg logit contributions | |
姫 | -4.207 |
� | -4.010 |
andr | -3.900 |
ennial | -3.816 |
sand | -3.794 |
Token | path |
Token ID | 3108 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
ascript | +1.418 |
NK | +1.240 |
lect | +1.236 |
spir | +1.225 |
Roll | +1.186 |
Neg logit contributions | |
姫 | -1.345 |
� | -1.272 |
andr | -1.223 |
ennial | -1.195 |
sand | -1.192 |
Token | appears |
Token ID | 3568 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
ascript | +1.833 |
lect | +1.622 |
NK | +1.586 |
spir | +1.579 |
Roll | +1.517 |
Neg logit contributions | |
姫 | -1.705 |
� | -1.633 |
andr | -1.576 |
ennial | -1.511 |
oling | -1.511 |
Token | clear |
Token ID | 1598 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
ascript | +2.287 |
lect | +1.971 |
spir | +1.967 |
NK | +1.966 |
Roll | +1.853 |
Neg logit contributions | |
姫 | -2.108 |
� | -2.028 |
andr | -1.999 |
ennial | -1.946 |
andestine | -1.927 |
Token | 're |
Token ID | 821 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
姫 | +1.685 |
� | +1.631 |
andr | +1.581 |
ennial | +1.537 |
sand | +1.537 |
Neg logit contributions | |
ascript | -1.894 |
lect | -1.631 |
spir | -1.604 |
NK | -1.576 |
Roll | -1.538 |
Token | getting |
Token ID | 1972 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
姫 | +1.479 |
� | +1.427 |
andr | +1.383 |
ennial | +1.347 |
sand | +1.345 |
Neg logit contributions | |
ascript | -1.660 |
lect | -1.432 |
spir | -1.405 |
NK | -1.381 |
Roll | -1.346 |
Token | a |
Token ID | 257 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
姫 | +1.478 |
� | +1.424 |
andr | +1.381 |
ennial | +1.343 |
sand | +1.342 |
Neg logit contributions | |
ascript | -1.561 |
lect | -1.341 |
spir | -1.311 |
NK | -1.299 |
Roll | -1.264 |
Token | cross |
Token ID | 3272 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
姫 | +0.679 |
� | +0.651 |
andr | +0.630 |
ennial | +0.611 |
andestine | +0.609 |
Neg logit contributions | |
ascript | -0.767 |
lect | -0.678 |
spir | -0.658 |
NK | -0.655 |
Roll | -0.632 |
Token | - |
Token ID | 12 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
ascript | +1.606 |
lect | +1.437 |
spir | +1.395 |
NK | +1.361 |
Roll | +1.344 |
Neg logit contributions | |
姫 | -1.277 |
� | -1.201 |
andr | -1.148 |
andestine | -1.103 |
sand | -1.090 |
Token | f |
Token ID | 69 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ascript | +3.759 |
spir | +3.340 |
lect | +3.340 |
Roll | +3.104 |
pen | +3.084 |
Neg logit contributions | |
姫 | -3.730 |
� | -3.594 |
andr | -3.417 |
andestine | -3.308 |
encia | -3.249 |
Token | ert |
Token ID | 861 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
ascript | +7.898 |
lect | +7.663 |
spir | +7.023 |
NK | +6.966 |
roller | +6.720 |
Neg logit contributions | |
姫 | -5.742 |
� | -5.311 |
andr | -4.645 |
ギ | -4.601 |
sand | -4.566 |
Token | il |
Token ID | 346 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
ascript | +1.189 |
lect | +1.029 |
NK | +1.000 |
spir | +0.992 |
Roll | +0.955 |
Neg logit contributions | |
姫 | -1.156 |
� | -1.131 |
andr | -1.068 |
sand | -1.021 |
ennial | -1.018 |
Token | ization |
Token ID | 1634 |
Feature activation | +8.47e-02 |
Pos logit contributions | |
ascript | +2.747 |
lect | +2.429 |
spir | +2.296 |
NK | +2.261 |
Roll | +2.182 |
Neg logit contributions | |
姫 | -2.550 |
� | -2.469 |
andr | -2.334 |
sand | -2.309 |
andestine | -2.273 |
Token | of |
Token ID | 286 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
姫 | +0.400 |
� | +0.386 |
andr | +0.371 |
ennial | +0.360 |
sand | +0.359 |
Neg logit contributions | |
ascript | -0.435 |
lect | -0.378 |
spir | -0.371 |
NK | -0.366 |
Roll | -0.355 |
Token | forms |
Token ID | 5107 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
ascript | +0.889 |
lect | +0.777 |
NK | +0.763 |
spir | +0.753 |
Roll | +0.724 |
Neg logit contributions | |
姫 | -0.921 |
� | -0.882 |
andr | -0.847 |
oling | -0.829 |
ennial | -0.827 |
Token | , |
Token ID | 11 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
ascript | +0.759 |
lect | +0.712 |
NK | +0.711 |
spir | +0.698 |
Roll | +0.685 |
Neg logit contributions | |
姫 | -0.833 |
� | -0.797 |
andr | -0.777 |
andestine | -0.757 |
oling | -0.751 |
Token | The |
Token ID | 383 |
Feature activation | -6.47e-01 |
Pos logit contributions | |
NK | +1.360 |
ascript | +1.356 |
lect | +1.304 |
spir | +1.267 |
Roll | +1.244 |
Neg logit contributions | |
姫 | -3.409 |
� | -3.263 |
oling | -3.236 |
andr | -3.206 |
andestine | -3.205 |
Token | Chronicle |
Token ID | 14160 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
ascript | +1.207 |
lect | +1.122 |
NK | +1.003 |
spir | +0.993 |
Roll | +0.913 |
Neg logit contributions | |
姫 | -4.851 |
oling | -4.635 |
andestine | -4.608 |
andr | -4.574 |
� | -4.558 |
Token | 13 |
Token ID | 1511 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
ascript | +1.038 |
lect | +0.977 |
spir | +0.959 |
NK | +0.940 |
Roll | +0.907 |
Neg logit contributions | |
姫 | -1.115 |
� | -1.032 |
andr | -1.012 |
andestine | -1.010 |
oling | -1.009 |
Token | of |
Token ID | 286 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
ascript | +1.222 |
lect | +1.148 |
spir | +1.118 |
NK | +1.099 |
Roll | +1.068 |
Neg logit contributions | |
姫 | -1.695 |
� | -1.620 |
andr | -1.552 |
andestine | -1.534 |
encia | -1.516 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
ascript | +1.981 |
lect | +1.868 |
NK | +1.842 |
spir | +1.809 |
FTA | +1.666 |
Neg logit contributions | |
姫 | -3.159 |
� | -2.966 |
andr | -2.916 |
andestine | -2.877 |
oling | -2.865 |
Token | Photo |
Token ID | 5555 |
Feature activation | -5.53e-02 |
Pos logit contributions | |
NK | +3.268 |
lect | +3.213 |
FTA | +3.206 |
spir | +3.097 |
Live | +2.998 |
Neg logit contributions | |
姫 | -9.086 |
andestine | -8.815 |
oling | -8.759 |
� | -8.638 |
adian | -8.400 |
Token | : |
Token ID | 25 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
ascript | +0.249 |
lect | +0.227 |
spir | +0.224 |
NK | +0.219 |
Roll | +0.216 |
Neg logit contributions | |
姫 | -0.285 |
� | -0.280 |
andr | -0.269 |
oling | -0.262 |
andestine | -0.262 |
Token | Le |
Token ID | 1004 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
ascript | +3.268 |
lect | +2.929 |
NK | +2.862 |
spir | +2.848 |
Roll | +2.775 |
Neg logit contributions | |
姫 | -2.468 |
� | -2.345 |
andr | -2.277 |
ennial | -2.202 |
oling | -2.174 |
Token | a |
Token ID | 64 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
ascript | +1.682 |
lect | +1.467 |
spir | +1.387 |
Roll | +1.307 |
NK | +1.279 |
Neg logit contributions | |
姫 | -3.268 |
� | -3.063 |
andestine | -2.922 |
andr | -2.909 |
extinction | -2.909 |
Token | Suzuki |
Token ID | 35807 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
姫 | +0.502 |
� | +0.476 |
andr | +0.462 |
oling | +0.458 |
andestine | +0.458 |
Neg logit contributions | |
ascript | -0.515 |
NK | -0.455 |
lect | -0.455 |
spir | -0.454 |
Roll | -0.433 |
Token | right |
Token ID | 826 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
姫 | +0.460 |
� | +0.430 |
andr | +0.419 |
sand | +0.405 |
oling | +0.403 |
Neg logit contributions | |
ascript | -0.593 |
lect | -0.530 |
spir | -0.524 |
NK | -0.516 |
Roll | -0.501 |
Token | was |
Token ID | 373 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
姫 | +1.936 |
� | +1.876 |
andr | +1.812 |
ennial | +1.766 |
sand | +1.760 |
Neg logit contributions | |
ascript | -2.333 |
lect | -2.021 |
spir | -1.992 |
NK | -1.957 |
Roll | -1.909 |
Token | a |
Token ID | 257 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
姫 | +1.083 |
� | +1.049 |
andr | +1.024 |
ennial | +1.003 |
sand | +0.998 |
Neg logit contributions | |
ascript | -1.035 |
lect | -0.886 |
spir | -0.876 |
NK | -0.858 |
Roll | -0.831 |
Token | true |
Token ID | 2081 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
姫 | +1.711 |
� | +1.655 |
andr | +1.597 |
ennial | +1.559 |
sand | +1.548 |
Neg logit contributions | |
ascript | -2.322 |
lect | -2.084 |
spir | -2.063 |
NK | -2.019 |
Roll | -1.972 |
Token | rock |
Token ID | 3881 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
姫 | +1.390 |
� | +1.341 |
andr | +1.288 |
ennial | +1.255 |
andestine | +1.250 |
Neg logit contributions | |
ascript | -2.098 |
lect | -1.877 |
spir | -1.859 |
NK | -1.832 |
Roll | -1.784 |
Token | n |
Token ID | 299 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
姫 | +0.975 |
� | +0.950 |
andr | +0.903 |
ennial | +0.888 |
sand | +0.884 |
Neg logit contributions | |
ascript | -1.339 |
lect | -1.176 |
spir | -1.161 |
NK | -1.140 |
Roll | -1.122 |
Token | � |
Token ID | 447 |
Feature activation | -9.22e-02 |
Pos logit contributions | |
ascript | +4.615 |
Roll | +4.441 |
lect | +4.228 |
Roll | +3.934 |
roller | +3.911 |
Neg logit contributions | |
姫 | -8.123 |
� | -7.946 |
andestine | -7.586 |
ennial | -7.570 |
andr | -7.537 |
Token | � |
Token ID | 247 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
ascript | +0.533 |
lect | +0.469 |
spir | +0.461 |
Roll | +0.444 |
NK | +0.440 |
Neg logit contributions | |
姫 | -0.370 |
andr | -0.347 |
sand | -0.335 |
andestine | -0.332 |
jan | -0.331 |
Token | roll |
Token ID | 4836 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
姫 | +0.219 |
� | +0.210 |
andr | +0.206 |
ennial | +0.200 |
andestine | +0.199 |
Neg logit contributions | |
ascript | -0.257 |
lect | -0.230 |
spir | -0.225 |
Roll | -0.217 |
NK | -0.216 |
Token | pioneer |
Token ID | 29570 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
姫 | +0.581 |
� | +0.559 |
andr | +0.538 |
ennial | +0.523 |
sand | +0.521 |
Neg logit contributions | |
ascript | -0.834 |
lect | -0.743 |
spir | -0.732 |
NK | -0.722 |
Roll | -0.709 |
Token | . |
Token ID | 13 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
ascript | +0.221 |
lect | +0.191 |
spir | +0.188 |
NK | +0.183 |
Roll | +0.179 |
Neg logit contributions | |
姫 | -0.222 |
� | -0.215 |
andr | -0.210 |
andestine | -0.205 |
sand | -0.203 |
Token | Head |
Token ID | 7123 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
ascript | +0.504 |
lect | +0.438 |
spir | +0.430 |
NK | +0.424 |
Roll | +0.414 |
Neg logit contributions | |
姫 | -0.471 |
� | -0.445 |
andr | -0.431 |
ennial | -0.419 |
andestine | -0.419 |
Token | - |
Token ID | 12 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
ascript | +0.958 |
lect | +0.847 |
spir | +0.833 |
NK | +0.818 |
Roll | +0.808 |
Neg logit contributions | |
姫 | -0.856 |
� | -0.811 |
andr | -0.776 |
ennial | -0.759 |
andestine | -0.756 |
Token | Anti |
Token ID | 28795 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ascript | +0.189 |
lect | +0.167 |
spir | +0.165 |
Roll | +0.162 |
NK | +0.160 |
Neg logit contributions | |
姫 | -0.160 |
� | -0.154 |
andr | -0.149 |
ennial | -0.146 |
andestine | -0.145 |
Token | E |
Token ID | 412 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
姫 | +0.807 |
� | +0.773 |
andr | +0.751 |
ennial | +0.728 |
sand | +0.727 |
Neg logit contributions | |
ascript | -0.944 |
lect | -0.823 |
spir | -0.808 |
NK | -0.793 |
Roll | -0.778 |
Token | pt |
Token ID | 457 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
ascript | +4.659 |
lect | +4.289 |
spir | +4.080 |
Roll | +3.995 |
glers | +3.891 |
Neg logit contributions | |
姫 | -4.085 |
� | -3.670 |
andestine | -3.572 |
andr | -3.536 |
ennial | -3.464 |
Token | ch |
Token ID | 354 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
ascript | +0.773 |
lect | +0.669 |
spir | +0.645 |
NK | +0.634 |
Roll | +0.627 |
Neg logit contributions | |
姫 | -0.713 |
� | -0.675 |
andr | -0.655 |
andestine | -0.630 |
ennial | -0.627 |
Token | stein |
Token ID | 5714 |
Feature activation | -8.87e-01 |
Pos logit contributions | |
ascript | +7.305 |
lect | +6.862 |
spir | +6.571 |
NK | +6.386 |
Roll | +6.363 |
Neg logit contributions | |
姫 | -5.913 |
� | -5.221 |
andr | -4.968 |
andestine | -4.907 |
encia | -4.854 |
Token | Prom |
Token ID | 10335 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
ascript | +4.513 |
spir | +4.061 |
lect | +4.059 |
NK | +3.872 |
Roll | +3.841 |
Neg logit contributions | |
姫 | -4.061 |
� | -3.676 |
andr | -3.545 |
andestine | -3.543 |
ennial | -3.485 |
Token | o |
Token ID | 78 |
Feature activation | -8.71e-02 |
Pos logit contributions | |
ascript | +0.231 |
lect | +0.201 |
spir | +0.197 |
NK | +0.195 |
Roll | +0.191 |
Neg logit contributions | |
姫 | -0.195 |
� | -0.185 |
andr | -0.181 |
andestine | -0.175 |
ennial | -0.174 |
Token | 2008 |
Token ID | 3648 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
ascript | +0.522 |
lect | +0.459 |
spir | +0.453 |
NK | +0.446 |
Roll | +0.437 |
Neg logit contributions | |
姫 | -0.333 |
� | -0.320 |
andr | -0.308 |
ennial | -0.298 |
sand | -0.297 |
Token | 01 |
Token ID | 486 |
Feature activation | +2.18e-03 |
Pos logit contributions | |
ascript | +0.743 |
lect | +0.646 |
Roll | +0.624 |
spir | +0.618 |
NK | +0.592 |
Neg logit contributions | |
姫 | -1.270 |
� | -1.240 |
andr | -1.220 |
andestine | -1.188 |
oling | -1.171 |
Token | , |
Token ID | 11 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
ascript | +0.835 |
lect | +0.785 |
NK | +0.783 |
spir | +0.769 |
Roll | +0.755 |
Neg logit contributions | |
姫 | -0.926 |
� | -0.886 |
andr | -0.864 |
andestine | -0.842 |
oling | -0.835 |
Token | The |
Token ID | 383 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
NK | +1.415 |
ascript | +1.400 |
lect | +1.353 |
spir | +1.314 |
Roll | +1.293 |
Neg logit contributions | |
姫 | -3.589 |
� | -3.435 |
oling | -3.409 |
andestine | -3.379 |
andr | -3.376 |
Token | Chronicle |
Token ID | 14160 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
ascript | +1.200 |
lect | +1.114 |
NK | +0.997 |
spir | +0.983 |
Roll | +0.907 |
Neg logit contributions | |
姫 | -4.838 |
oling | -4.624 |
andestine | -4.599 |
andr | -4.563 |
� | -4.549 |
Token | 14 |
Token ID | 1478 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
ascript | +0.928 |
lect | +0.847 |
spir | +0.846 |
NK | +0.839 |
forth | +0.780 |
Neg logit contributions | |
姫 | -0.879 |
� | -0.835 |
andr | -0.813 |
andestine | -0.798 |
oling | -0.796 |
Token | of |
Token ID | 286 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
ascript | +1.349 |
lect | +1.264 |
spir | +1.227 |
NK | +1.207 |
Roll | +1.172 |
Neg logit contributions | |
姫 | -1.891 |
� | -1.809 |
andr | -1.725 |
andestine | -1.718 |
encia | -1.686 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ascript | +2.040 |
lect | +1.928 |
NK | +1.900 |
spir | +1.861 |
FTA | +1.726 |
Neg logit contributions | |
姫 | -3.258 |
� | -3.062 |
andr | -3.012 |
andestine | -2.979 |
oling | -2.967 |
Token | Photo |
Token ID | 5555 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
NK | +3.180 |
lect | +3.126 |
FTA | +3.117 |
spir | +3.006 |
Live | +2.938 |
Neg logit contributions | |
姫 | -8.941 |
andestine | -8.662 |
oling | -8.593 |
� | -8.498 |
adian | -8.236 |
Token | : |
Token ID | 25 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
姫 | +0.069 |
� | +0.068 |
andr | +0.065 |
oling | +0.063 |
andestine | +0.063 |
Neg logit contributions | |
ascript | -0.062 |
lect | -0.056 |
spir | -0.056 |
NK | -0.054 |
Roll | -0.054 |
Token | Le |
Token ID | 1004 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
ascript | +3.276 |
lect | +2.931 |
NK | +2.862 |
spir | +2.851 |
Roll | +2.774 |
Neg logit contributions | |
姫 | -2.522 |
� | -2.399 |
andr | -2.330 |
ennial | -2.255 |
oling | -2.227 |
Token | a |
Token ID | 64 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
ascript | +1.621 |
lect | +1.414 |
spir | +1.338 |
Roll | +1.262 |
NK | +1.234 |
Neg logit contributions | |
姫 | -3.129 |
� | -2.935 |
andestine | -2.801 |
andr | -2.792 |
extinction | -2.788 |
Token | Suzuki |
Token ID | 35807 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
姫 | +0.513 |
� | +0.487 |
andr | +0.472 |
oling | +0.468 |
andestine | +0.467 |
Neg logit contributions | |
ascript | -0.527 |
lect | -0.466 |
NK | -0.466 |
spir | -0.465 |
Roll | -0.443 |
Token | , |
Token ID | 11 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
ascript | +0.707 |
NK | +0.664 |
lect | +0.662 |
spir | +0.648 |
Roll | +0.639 |
Neg logit contributions | |
姫 | -0.771 |
� | -0.737 |
andr | -0.720 |
andestine | -0.701 |
oling | -0.696 |
Token | The |
Token ID | 383 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
NK | +1.328 |
ascript | +1.321 |
lect | +1.272 |
spir | +1.234 |
Roll | +1.214 |
Neg logit contributions | |
姫 | -3.324 |
� | -3.181 |
oling | -3.155 |
andr | -3.126 |
andestine | -3.123 |
Token | Chronicle |
Token ID | 14160 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
ascript | +1.198 |
lect | +1.114 |
NK | +0.997 |
spir | +0.984 |
Roll | +0.908 |
Neg logit contributions | |
姫 | -4.877 |
oling | -4.658 |
andestine | -4.634 |
andr | -4.599 |
� | -4.582 |
Token | 12 |
Token ID | 1105 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
ascript | +0.300 |
lect | +0.266 |
spir | +0.264 |
NK | +0.261 |
Roll | +0.252 |
Neg logit contributions | |
姫 | -0.348 |
� | -0.336 |
andr | -0.323 |
andestine | -0.322 |
oling | -0.319 |
Token | of |
Token ID | 286 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
ascript | +0.587 |
lect | +0.538 |
spir | +0.524 |
NK | +0.516 |
Roll | +0.501 |
Neg logit contributions | |
姫 | -0.742 |
� | -0.712 |
andr | -0.685 |
andestine | -0.675 |
oling | -0.665 |
Token | 15 |
Token ID | 1315 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ascript | +1.755 |
lect | +1.640 |
NK | +1.621 |
spir | +1.595 |
FTA | +1.465 |
Neg logit contributions | |
姫 | -2.753 |
� | -2.583 |
andr | -2.546 |
andestine | -2.498 |
oling | -2.492 |
Token | Photo |
Token ID | 5555 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
NK | +3.230 |
lect | +3.206 |
FTA | +3.155 |
spir | +3.043 |
Live | +2.956 |
Neg logit contributions | |
姫 | -8.943 |
andestine | -8.652 |
oling | -8.582 |
� | -8.435 |
adian | -8.243 |
Token | : |
Token ID | 25 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
ascript | +0.096 |
lect | +0.087 |
spir | +0.086 |
NK | +0.084 |
Roll | +0.083 |
Neg logit contributions | |
姫 | -0.108 |
� | -0.106 |
andr | -0.102 |
oling | -0.100 |
andestine | -0.099 |
Token | Le |
Token ID | 1004 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
ascript | +3.007 |
lect | +2.691 |
NK | +2.629 |
spir | +2.620 |
Roll | +2.553 |
Neg logit contributions | |
姫 | -2.229 |
� | -2.119 |
andr | -2.056 |
ennial | -1.989 |
oling | -1.964 |
Token | a |
Token ID | 64 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
ascript | +1.611 |
lect | +1.403 |
spir | +1.330 |
Roll | +1.251 |
NK | +1.228 |
Neg logit contributions | |
姫 | -3.110 |
� | -2.913 |
andestine | -2.778 |
andr | -2.767 |
extinction | -2.766 |
Token | Suzuki |
Token ID | 35807 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
姫 | +0.650 |
� | +0.616 |
andr | +0.598 |
oling | +0.592 |
andestine | +0.591 |
Neg logit contributions | |
ascript | -0.674 |
lect | -0.595 |
NK | -0.594 |
spir | -0.594 |
Roll | -0.566 |
Token | ess |
Token ID | 408 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
姫 | +0.759 |
� | +0.730 |
andr | +0.706 |
ennial | +0.688 |
sand | +0.683 |
Neg logit contributions | |
ascript | -0.795 |
lect | -0.685 |
spir | -0.674 |
NK | -0.660 |
Roll | -0.644 |
Token |
|
Token ID | 198 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
姫 | +0.799 |
� | +0.765 |
andr | +0.737 |
ennial | +0.719 |
sand | +0.718 |
Neg logit contributions | |
ascript | -0.946 |
lect | -0.823 |
spir | -0.804 |
NK | -0.797 |
Roll | -0.774 |
Token |
|
Token ID | 198 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
姫 | +1.249 |
� | +1.206 |
andr | +1.150 |
andestine | +1.115 |
ennial | +1.112 |
Neg logit contributions | |
ascript | -1.541 |
lect | -1.392 |
spir | -1.365 |
NK | -1.344 |
Roll | -1.329 |
Token | Cross |
Token ID | 21544 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
ascript | +1.257 |
lect | +1.137 |
spir | +1.128 |
Roll | +1.107 |
NK | +1.069 |
Neg logit contributions | |
姫 | -1.041 |
� | -0.995 |
andr | -0.972 |
ennial | -0.945 |
extinction | -0.941 |
Token | - |
Token ID | 12 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
ascript | +1.188 |
lect | +1.068 |
spir | +1.047 |
NK | +1.020 |
Roll | +1.008 |
Neg logit contributions | |
姫 | -0.670 |
� | -0.616 |
andr | -0.594 |
sand | -0.569 |
ennial | -0.566 |
Token | d |
Token ID | 67 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ascript | +5.104 |
lect | +4.538 |
spir | +4.535 |
Roll | +4.319 |
DX | +4.212 |
Neg logit contributions | |
姫 | -4.005 |
� | -3.790 |
andr | -3.681 |
ennial | -3.530 |
jan | -3.527 |
Token | ressing |
Token ID | 11697 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
ascript | +4.955 |
lect | +4.316 |
spir | +4.133 |
Roll | +3.700 |
ivid | +3.658 |
Neg logit contributions | |
姫 | -8.740 |
� | -7.602 |
sand | -7.454 |
KEN | -7.441 |
andr | -7.412 |
Token | : |
Token ID | 25 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
ascript | +2.635 |
lect | +2.277 |
spir | +2.238 |
NK | +2.218 |
glers | +2.157 |
Neg logit contributions | |
姫 | -2.548 |
� | -2.433 |
andr | -2.412 |
sand | -2.309 |
ennial | -2.300 |
Token | Cross |
Token ID | 6372 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
姫 | +0.781 |
� | +0.757 |
andr | +0.745 |
ennial | +0.728 |
sand | +0.726 |
Neg logit contributions | |
ascript | -0.506 |
lect | -0.416 |
spir | -0.408 |
NK | -0.401 |
Roll | -0.385 |
Token | - |
Token ID | 12 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
ascript | +1.833 |
lect | +1.634 |
spir | +1.612 |
NK | +1.588 |
Roll | +1.562 |
Neg logit contributions | |
姫 | -0.860 |
� | -0.821 |
andr | -0.782 |
ennial | -0.749 |
sand | -0.748 |
Token | d |
Token ID | 67 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ascript | +3.391 |
lect | +2.983 |
spir | +2.945 |
Roll | +2.815 |
NK | +2.806 |
Neg logit contributions | |
姫 | -2.424 |
� | -2.337 |
andr | -2.241 |
ennial | -2.170 |
oling | -2.143 |
Token | : |
Token ID | 25 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
姫 | +1.673 |
� | +1.619 |
andr | +1.561 |
ennial | +1.515 |
oling | +1.514 |
Neg logit contributions | |
ascript | -1.770 |
lect | -1.567 |
spir | -1.540 |
NK | -1.516 |
Roll | -1.487 |
Token | Just |
Token ID | 2329 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
ascript | +0.145 |
NK | +0.130 |
lect | +0.128 |
spir | +0.128 |
Roll | +0.126 |
Neg logit contributions | |
姫 | -0.149 |
� | -0.142 |
andr | -0.138 |
ennial | -0.136 |
oling | -0.136 |
Token | about |
Token ID | 546 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
姫 | +2.404 |
� | +2.320 |
andr | +2.243 |
sand | +2.179 |
ennial | +2.178 |
Neg logit contributions | |
ascript | -2.658 |
lect | -2.291 |
spir | -2.247 |
NK | -2.222 |
Roll | -2.166 |
Token | the |
Token ID | 262 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
姫 | +1.724 |
� | +1.671 |
andr | +1.620 |
ennial | +1.576 |
sand | +1.572 |
Neg logit contributions | |
ascript | -1.871 |
lect | -1.610 |
spir | -1.580 |
NK | -1.556 |
Roll | -1.517 |
Token | key |
Token ID | 1994 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
姫 | +0.537 |
� | +0.514 |
andr | +0.502 |
ennial | +0.486 |
sand | +0.481 |
Neg logit contributions | |
ascript | -0.573 |
lect | -0.497 |
spir | -0.490 |
NK | -0.483 |
Roll | -0.469 |
Token | w |
Token ID | 86 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
姫 | +0.650 |
� | +0.629 |
andr | +0.606 |
ennial | +0.585 |
oling | +0.582 |
Neg logit contributions | |
ascript | -0.813 |
lect | -0.706 |
spir | -0.699 |
NK | -0.682 |
Roll | -0.672 |
Token | ording |
Token ID | 1284 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ascript | +5.827 |
lect | +5.284 |
Roll | +4.887 |
spir | +4.847 |
roller | +4.812 |
Neg logit contributions | |
姫 | -7.066 |
� | -6.784 |
andr | -6.185 |
ゼウス | -6.079 |
encia | -6.079 |
Token | of |
Token ID | 286 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
姫 | +0.545 |
� | +0.524 |
andr | +0.505 |
ennial | +0.490 |
sand | +0.486 |
Neg logit contributions | |
ascript | -0.561 |
lect | -0.483 |
spir | -0.477 |
NK | -0.467 |
Roll | -0.458 |
Token | potions |
Token ID | 28074 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
姫 | +0.739 |
� | +0.716 |
andr | +0.695 |
ennial | +0.679 |
sand | +0.674 |
Neg logit contributions | |
ascript | -0.730 |
lect | -0.625 |
spir | -0.614 |
NK | -0.604 |
Roll | -0.590 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
ascript | +0.127 |
lect | +0.110 |
spir | +0.109 |
NK | +0.106 |
Roll | +0.105 |
Neg logit contributions | |
姫 | -0.120 |
� | -0.116 |
andr | -0.112 |
ennial | -0.108 |
sand | -0.108 |
Token |
|
Token ID | 198 |
Feature activation | -9.32e-04 |
Pos logit contributions | |
姫 | +2.265 |
� | +2.188 |
andr | +2.117 |
ennial | +2.060 |
sand | +2.051 |
Neg logit contributions | |
ascript | -2.775 |
lect | -2.410 |
spir | -2.378 |
NK | -2.344 |
Roll | -2.290 |
Token | campaign |
Token ID | 1923 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
ascript | +1.673 |
lect | +1.471 |
NK | +1.429 |
spir | +1.415 |
Roll | +1.388 |
Neg logit contributions | |
姫 | -1.628 |
� | -1.562 |
andr | -1.519 |
sand | -1.460 |
andestine | -1.450 |
Token | headquarters |
Token ID | 10043 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
姫 | +1.233 |
� | +1.181 |
andr | +1.128 |
ennial | +1.081 |
sand | +1.076 |
Neg logit contributions | |
ascript | -2.289 |
lect | -2.036 |
spir | -2.004 |
NK | -1.984 |
Roll | -1.943 |
Token | on |
Token ID | 319 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
ascript | +0.732 |
lect | +0.615 |
NK | +0.602 |
spir | +0.598 |
Roll | +0.571 |
Neg logit contributions | |
姫 | -0.933 |
� | -0.904 |
andr | -0.882 |
sand | -0.858 |
ennial | -0.857 |
Token | November |
Token ID | 3389 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
ascript | +1.191 |
NK | +1.072 |
lect | +1.071 |
spir | +1.038 |
FTA | +1.004 |
Neg logit contributions | |
姫 | -1.067 |
� | -1.012 |
andr | -0.984 |
oling | -0.955 |
ennial | -0.951 |
Token | 3 |
Token ID | 513 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
ascript | +0.173 |
lect | +0.156 |
NK | +0.154 |
spir | +0.152 |
Roll | +0.148 |
Neg logit contributions | |
姫 | -0.161 |
� | -0.154 |
andr | -0.149 |
oling | -0.144 |
andestine | -0.144 |
Token | , |
Token ID | 11 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ascript | +1.639 |
lect | +1.496 |
NK | +1.478 |
spir | +1.461 |
Roll | +1.402 |
Neg logit contributions | |
姫 | -1.664 |
� | -1.560 |
andr | -1.533 |
andestine | -1.495 |
extinction | -1.457 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
ascript | +4.612 |
NK | +3.908 |
lect | +3.852 |
spir | +3.778 |
pen | +3.640 |
Neg logit contributions | |
姫 | -8.035 |
andr | -7.613 |
� | -7.582 |
andestine | -7.397 |
ennial | -7.300 |
Token | , |
Token ID | 11 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
ascript | +3.363 |
lect | +2.900 |
spir | +2.853 |
NK | +2.802 |
Roll | +2.738 |
Neg logit contributions | |
姫 | -2.915 |
� | -2.824 |
andr | -2.733 |
sand | -2.659 |
ennial | -2.655 |
Token | in |
Token ID | 287 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
ascript | +2.147 |
lect | +1.814 |
NK | +1.795 |
spir | +1.765 |
Roll | +1.699 |
Neg logit contributions | |
姫 | -2.935 |
� | -2.817 |
andr | -2.757 |
ennial | -2.673 |
andestine | -2.671 |
Token | Irvine |
Token ID | 37277 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
ascript | +1.773 |
lect | +1.588 |
NK | +1.569 |
spir | +1.547 |
Roll | +1.519 |
Neg logit contributions | |
姫 | -1.395 |
� | -1.341 |
andr | -1.292 |
ennial | -1.231 |
andestine | -1.228 |
Token | , |
Token ID | 11 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
ascript | +0.408 |
lect | +0.357 |
spir | +0.353 |
NK | +0.347 |
Roll | +0.339 |
Neg logit contributions | |
姫 | -0.299 |
� | -0.288 |
andr | -0.279 |
sand | -0.269 |
ennial | -0.267 |
Token | Jim |
Token ID | 5395 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
姫 | +0.093 |
� | +0.090 |
andr | +0.087 |
ennial | +0.084 |
oling | +0.083 |
Neg logit contributions | |
ascript | -0.114 |
lect | -0.101 |
NK | -0.099 |
spir | -0.099 |
Roll | -0.096 |
Token | Parsons |
Token ID | 35925 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
ascript | +1.992 |
NK | +1.851 |
lect | +1.798 |
spir | +1.787 |
Roll | +1.743 |
Neg logit contributions | |
姫 | -2.735 |
� | -2.664 |
andr | -2.543 |
andestine | -2.488 |
oling | -2.462 |
Token |
|
Token ID | 198 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
ascript | +1.374 |
lect | +1.264 |
NK | +1.254 |
spir | +1.223 |
Roll | +1.201 |
Neg logit contributions | |
姫 | -1.372 |
� | -1.303 |
andr | -1.247 |
oling | -1.228 |
andestine | -1.226 |
Token |
|
Token ID | 198 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
ascript | +1.070 |
lect | +1.020 |
spir | +0.995 |
NK | +0.981 |
Roll | +0.969 |
Neg logit contributions | |
姫 | -1.153 |
� | -1.123 |
andr | -1.055 |
andestine | -1.041 |
oling | -1.029 |
Token | 2014 |
Token ID | 4967 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
ascript | +2.483 |
lect | +2.359 |
spir | +2.329 |
Roll | +2.244 |
forth | +2.103 |
Neg logit contributions | |
姫 | -3.708 |
� | -3.566 |
extinction | -3.500 |
sandy | -3.473 |
oling | -3.465 |
Token | Louis |
Token ID | 5593 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ascript | +2.191 |
lect | +1.991 |
NK | +1.982 |
spir | +1.922 |
Roll | +1.883 |
Neg logit contributions | |
姫 | -1.875 |
� | -1.779 |
andr | -1.704 |
ennial | -1.661 |
andestine | -1.649 |
Token | C |
Token ID | 327 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
NK | +3.407 |
lect | +3.208 |
ascript | +3.011 |
Ct | +2.977 |
FTA | +2.948 |
Neg logit contributions | |
姫 | -8.982 |
� | -8.430 |
andr | -8.115 |
ennial | -7.993 |
oling | -7.990 |
Token | . |
Token ID | 13 |
Feature activation | -8.51e-01 |
Pos logit contributions | |
姫 | +0.145 |
� | +0.139 |
andr | +0.133 |
andestine | +0.129 |
oling | +0.129 |
Neg logit contributions | |
ascript | -0.118 |
lect | -0.106 |
spir | -0.103 |
NK | -0.101 |
Roll | -0.099 |
Token | K |
Token ID | 42 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
Roll | +2.323 |
NK | +2.301 |
spir | +2.300 |
ascript | +2.288 |
lect | +2.191 |
Neg logit contributions | |
姫 | -5.375 |
andestine | -5.213 |
� | -5.210 |
andr | -5.206 |
�� | -5.135 |
Token | . |
Token ID | 13 |
Feature activation | -9.19e-02 |
Pos logit contributions | |
姫 | +0.083 |
� | +0.079 |
andr | +0.075 |
oling | +0.074 |
andestine | +0.074 |
Neg logit contributions | |
ascript | -0.064 |
lect | -0.057 |
spir | -0.055 |
NK | -0.055 |
Roll | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
ascript | +0.467 |
lect | +0.420 |
spir | +0.417 |
NK | +0.417 |
Roll | +0.404 |
Neg logit contributions | |
姫 | -0.427 |
� | -0.411 |
andr | -0.392 |
oling | -0.390 |
andestine | -0.379 |
Token | ency |
Token ID | 1387 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
姫 | +1.913 |
� | +1.856 |
andr | +1.796 |
ennial | +1.756 |
sand | +1.749 |
Neg logit contributions | |
ascript | -2.158 |
lect | -1.847 |
spir | -1.822 |
NK | -1.786 |
Roll | -1.744 |
Token | Irvine |
Token ID | 37277 |
Feature activation | -8.62e-02 |
Pos logit contributions | |
姫 | +2.899 |
� | +2.830 |
andr | +2.723 |
ennial | +2.684 |
sand | +2.663 |
Neg logit contributions | |
ascript | -3.149 |
lect | -2.650 |
spir | -2.632 |
NK | -2.542 |
hemy | -2.510 |
Token | , |
Token ID | 11 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
ascript | +0.411 |
lect | +0.353 |
spir | +0.346 |
NK | +0.341 |
Roll | +0.330 |
Neg logit contributions | |
姫 | -0.432 |
� | -0.420 |
andr | -0.409 |
sand | -0.396 |
ennial | -0.395 |
Token | November |
Token ID | 3389 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
ascript | +0.465 |
lect | +0.412 |
spir | +0.404 |
NK | +0.399 |
Roll | +0.390 |
Neg logit contributions | |
姫 | -0.302 |
� | -0.290 |
andr | -0.280 |
ennial | -0.270 |
sand | -0.270 |
Token | 2 |
Token ID | 362 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
ascript | +1.668 |
lect | +1.559 |
NK | +1.541 |
spir | +1.477 |
Roll | +1.473 |
Neg logit contributions | |
姫 | -2.040 |
� | -1.923 |
andr | -1.888 |
andestine | -1.852 |
oling | -1.831 |
Token | , |
Token ID | 11 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ascript | +2.828 |
lect | +2.476 |
spir | +2.443 |
NK | +2.404 |
Roll | +2.355 |
Neg logit contributions | |
姫 | -1.835 |
� | -1.776 |
andr | -1.700 |
ennial | -1.653 |
sand | -1.648 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
ascript | +3.472 |
lect | +2.906 |
spir | +2.711 |
NK | +2.687 |
Roll | +2.582 |
Neg logit contributions | |
姫 | -9.066 |
� | -8.642 |
andestine | -8.482 |
andr | -8.478 |
encia | -8.257 |
Token | , |
Token ID | 11 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
ascript | +3.353 |
lect | +2.901 |
spir | +2.853 |
NK | +2.804 |
Roll | +2.740 |
Neg logit contributions | |
姫 | -2.788 |
� | -2.700 |
andr | -2.612 |
sand | -2.538 |
ennial | -2.537 |
Token | in |
Token ID | 287 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
ascript | +1.967 |
lect | +1.683 |
NK | +1.626 |
spir | +1.619 |
Roll | +1.556 |
Neg logit contributions | |
姫 | -2.470 |
� | -2.371 |
andr | -2.331 |
ennial | -2.262 |
andestine | -2.242 |
Token | Irvine |
Token ID | 37277 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
ascript | +1.337 |
lect | +1.196 |
spir | +1.169 |
NK | +1.162 |
Roll | +1.130 |
Neg logit contributions | |
姫 | -0.939 |
� | -0.911 |
andr | -0.872 |
ennial | -0.834 |
sand | -0.832 |
Token | , |
Token ID | 11 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
ascript | +0.530 |
lect | +0.463 |
spir | +0.457 |
NK | +0.450 |
Roll | +0.439 |
Neg logit contributions | |
姫 | -0.448 |
� | -0.430 |
andr | -0.419 |
andestine | -0.398 |
sand | -0.398 |
Token | university |
Token ID | 6403 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
姫 | +1.727 |
� | +1.672 |
andr | +1.619 |
sand | +1.573 |
ennial | +1.571 |
Neg logit contributions | |
ascript | -2.044 |
lect | -1.774 |
spir | -1.740 |
NK | -1.721 |
Roll | -1.678 |
Token | campuses |
Token ID | 23003 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
姫 | +1.507 |
� | +1.444 |
andr | +1.381 |
ennial | +1.329 |
sand | +1.328 |
Neg logit contributions | |
ascript | -2.884 |
lect | -2.560 |
spir | -2.526 |
NK | -2.492 |
Roll | -2.444 |
Token | on |
Token ID | 319 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
姫 | +0.556 |
� | +0.535 |
andr | +0.520 |
sand | +0.505 |
ennial | +0.504 |
Neg logit contributions | |
ascript | -0.627 |
lect | -0.543 |
spir | -0.534 |
NK | -0.525 |
Roll | -0.512 |
Token | both |
Token ID | 1111 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
姫 | +0.558 |
� | +0.526 |
andr | +0.517 |
ennial | +0.500 |
sand | +0.499 |
Neg logit contributions | |
ascript | -0.610 |
lect | -0.532 |
spir | -0.520 |
NK | -0.515 |
Roll | -0.498 |
Token | sides |
Token ID | 5389 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
ascript | +1.308 |
lect | +1.170 |
NK | +1.149 |
spir | +1.113 |
Roll | +1.066 |
Neg logit contributions | |
姫 | -1.753 |
andr | -1.654 |
� | -1.641 |
ennial | -1.598 |
sand | -1.593 |
Token | of |
Token ID | 286 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ascript | +1.645 |
lect | +1.422 |
spir | +1.399 |
NK | +1.375 |
Roll | +1.343 |
Neg logit contributions | |
姫 | -1.349 |
� | -1.308 |
andr | -1.263 |
ennial | -1.229 |
sand | -1.228 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ascript | +5.563 |
lect | +4.614 |
spir | +4.429 |
NK | +4.369 |
Roll | +4.243 |
Neg logit contributions | |
姫 | -7.184 |
andr | -6.912 |
� | -6.769 |
encia | -6.596 |
sand | -6.565 |
Token | Atlantic |
Token ID | 10596 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
ascript | +6.682 |
lect | +5.752 |
Roll | +5.390 |
Compass | +5.381 |
NK | +5.277 |
Neg logit contributions | |
姫 | -7.624 |
encia | -6.969 |
andr | -6.872 |
� | -6.830 |
andestine | -6.685 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ascript | +4.120 |
lect | +3.578 |
spir | +3.545 |
NK | +3.421 |
Roll | +3.395 |
Neg logit contributions | |
姫 | -4.549 |
� | -4.363 |
andr | -4.307 |
sand | -4.106 |
encia | -4.093 |
Token |
|
Token ID | 198 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
ascript | +0.574 |
lect | +0.507 |
spir | +0.500 |
NK | +0.493 |
Roll | +0.483 |
Neg logit contributions | |
姫 | -0.455 |
� | -0.435 |
andr | -0.425 |
oling | -0.412 |
ennial | -0.412 |
Token |
|
Token ID | 198 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
姫 | +0.638 |
� | +0.621 |
andr | +0.591 |
andestine | +0.578 |
oling | +0.577 |
Neg logit contributions | |
ascript | -0.734 |
lect | -0.671 |
spir | -0.662 |
NK | -0.650 |
Roll | -0.643 |
Token | modes |
Token ID | 12881 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
ascript | +0.525 |
lect | +0.462 |
NK | +0.452 |
spir | +0.442 |
Roll | +0.437 |
Neg logit contributions | |
姫 | -0.532 |
� | -0.513 |
andr | -0.493 |
ennial | -0.490 |
sand | -0.485 |
Token | for |
Token ID | 329 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
ascript | +3.186 |
lect | +2.829 |
NK | +2.801 |
spir | +2.673 |
Roll | +2.664 |
Neg logit contributions | |
姫 | -3.611 |
� | -3.418 |
andr | -3.345 |
ennial | -3.277 |
encia | -3.242 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ascript | +1.941 |
lect | +1.740 |
NK | +1.703 |
Roll | +1.655 |
spir | +1.635 |
Neg logit contributions | |
姫 | -1.835 |
� | -1.760 |
ennial | -1.673 |
andr | -1.671 |
encia | -1.649 |
Token | truck |
Token ID | 7779 |
Feature activation | -8.11e-02 |
Pos logit contributions | |
ascript | +0.688 |
lect | +0.607 |
NK | +0.596 |
spir | +0.582 |
Roll | +0.577 |
Neg logit contributions | |
姫 | -0.609 |
� | -0.580 |
andr | -0.558 |
ennial | -0.550 |
andestine | -0.548 |
Token | depending |
Token ID | 6906 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
ascript | +0.412 |
lect | +0.359 |
spir | +0.350 |
NK | +0.346 |
Roll | +0.341 |
Neg logit contributions | |
姫 | -0.387 |
� | -0.372 |
andr | -0.361 |
sand | -0.347 |
ennial | -0.347 |
Token | on |
Token ID | 319 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
ascript | +2.597 |
lect | +2.149 |
spir | +2.087 |
NK | +2.051 |
Roll | +1.984 |
Neg logit contributions | |
姫 | -3.505 |
� | -3.409 |
andr | -3.323 |
sand | -3.257 |
ennial | -3.244 |
Token | terrain |
Token ID | 15510 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ascript | +5.302 |
NK | +4.789 |
lect | +4.789 |
FTA | +4.741 |
Roll | +4.637 |
Neg logit contributions | |
姫 | -7.043 |
� | -6.614 |
encia | -6.367 |
andestine | -6.287 |
andr | -6.248 |
Token | and |
Token ID | 290 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
ascript | +4.344 |
lect | +3.957 |
NK | +3.859 |
Roll | +3.709 |
spir | +3.617 |
Neg logit contributions | |
姫 | -5.260 |
� | -5.103 |
andr | -4.906 |
encia | -4.824 |
andestine | -4.757 |
Token | driving |
Token ID | 5059 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
ascript | +3.731 |
lect | +3.319 |
NK | +3.266 |
Roll | +3.222 |
FTA | +3.160 |
Neg logit contributions | |
姫 | -4.242 |
� | -4.019 |
andr | -3.850 |
andestine | -3.805 |
ennial | -3.801 |
Token | conditions |
Token ID | 3403 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
ascript | +1.454 |
lect | +1.285 |
NK | +1.258 |
spir | +1.233 |
Roll | +1.201 |
Neg logit contributions | |
姫 | -1.696 |
� | -1.619 |
andr | -1.562 |
ennial | -1.548 |
andestine | -1.543 |
Token | : |
Token ID | 25 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
ascript | +4.001 |
lect | +3.457 |
NK | +3.360 |
spir | +3.276 |
Roll | +3.270 |
Neg logit contributions | |
姫 | -4.754 |
� | -4.455 |
andr | -4.364 |
andestine | -4.277 |
encia | -4.263 |
Token | will |
Token ID | 481 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
ascript | +0.263 |
lect | +0.227 |
spir | +0.224 |
NK | +0.221 |
Roll | +0.214 |
Neg logit contributions | |
姫 | -0.252 |
� | -0.244 |
andr | -0.234 |
ennial | -0.229 |
sand | -0.229 |
Token | soon |
Token ID | 2582 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
ascript | +1.113 |
lect | +0.973 |
spir | +0.958 |
NK | +0.958 |
Roll | +0.927 |
Neg logit contributions | |
姫 | -1.091 |
� | -1.040 |
andr | -0.999 |
ennial | -0.994 |
sand | -0.985 |
Token | deliver |
Token ID | 5203 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
姫 | +0.516 |
� | +0.496 |
andr | +0.475 |
ennial | +0.468 |
sand | +0.466 |
Neg logit contributions | |
ascript | -0.607 |
lect | -0.536 |
spir | -0.525 |
NK | -0.521 |
Roll | -0.509 |
Token | " |
Token ID | 366 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
ascript | +0.632 |
lect | +0.541 |
spir | +0.528 |
NK | +0.526 |
Roll | +0.516 |
Neg logit contributions | |
姫 | -0.680 |
� | -0.648 |
andr | -0.625 |
andestine | -0.611 |
ennial | -0.610 |
Token | big |
Token ID | 14261 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
ascript | +1.597 |
lect | +1.386 |
spir | +1.365 |
Roll | +1.323 |
NK | +1.285 |
Neg logit contributions | |
姫 | -1.713 |
� | -1.651 |
andr | -1.592 |
andestine | -1.566 |
ennial | -1.562 |
Token | news |
Token ID | 1705 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ascript | +1.126 |
lect | +0.955 |
spir | +0.938 |
NK | +0.931 |
Roll | +0.904 |
Neg logit contributions | |
姫 | -1.226 |
� | -1.195 |
andr | -1.162 |
ennial | -1.121 |
andestine | -1.119 |
Token | ". |
Token ID | 1911 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
ascript | +5.231 |
lect | +4.694 |
spir | +4.399 |
Roll | +4.338 |
NK | +4.308 |
Neg logit contributions | |
姫 | -7.248 |
� | -6.860 |
andr | -6.613 |
andestine | -6.468 |
encia | -6.384 |
Token | According |
Token ID | 4821 |
Feature activation | -5.42e-03 |
Pos logit contributions | |
ascript | +0.237 |
lect | +0.209 |
spir | +0.207 |
NK | +0.202 |
Roll | +0.201 |
Neg logit contributions | |
姫 | -0.208 |
� | -0.200 |
andr | -0.192 |
ennial | -0.189 |
andestine | -0.189 |
Token | to |
Token ID | 284 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
ascript | +0.032 |
lect | +0.028 |
spir | +0.027 |
NK | +0.027 |
Roll | +0.026 |
Neg logit contributions | |
姫 | -0.021 |
� | -0.021 |
andr | -0.020 |
sand | -0.019 |
ennial | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
ascript | +3.730 |
lect | +3.414 |
spir | +3.292 |
NK | +3.276 |
Roll | +3.245 |
Neg logit contributions | |
姫 | -3.472 |
� | -3.257 |
ennial | -3.102 |
sand | -3.089 |
andestine | -3.064 |
Token | magazine |
Token ID | 7093 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
ascript | +3.353 |
lect | +3.018 |
spir | +2.978 |
NK | +2.947 |
Roll | +2.888 |
Neg logit contributions | |
姫 | -2.939 |
� | -2.733 |
ennial | -2.562 |
andr | -2.560 |
andestine | -2.544 |
Token | in |
Token ID | 287 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
姫 | +3.225 |
� | +3.146 |
andr | +3.038 |
ennial | +2.950 |
sand | +2.935 |
Neg logit contributions | |
ascript | -3.855 |
lect | -3.304 |
spir | -3.248 |
NK | -3.194 |
glers | -3.127 |
Token | conjunction |
Token ID | 17856 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
姫 | +1.463 |
� | +1.417 |
andr | +1.374 |
sand | +1.336 |
ennial | +1.328 |
Neg logit contributions | |
ascript | -1.610 |
lect | -1.393 |
spir | -1.372 |
NK | -1.353 |
Roll | -1.314 |
Token | with |
Token ID | 351 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
姫 | +1.355 |
� | +1.307 |
andr | +1.265 |
ennial | +1.234 |
andestine | +1.234 |
Neg logit contributions | |
ascript | -1.153 |
spir | -0.979 |
lect | -0.978 |
NK | -0.948 |
Roll | -0.917 |
Token | Edwin |
Token ID | 37016 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
姫 | +0.668 |
� | +0.643 |
andr | +0.624 |
sand | +0.611 |
ennial | +0.608 |
Neg logit contributions | |
ascript | -0.757 |
lect | -0.660 |
spir | -0.650 |
NK | -0.645 |
Roll | -0.626 |
Token | Earl |
Token ID | 21647 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
姫 | +0.544 |
� | +0.518 |
andr | +0.505 |
sand | +0.493 |
oling | +0.492 |
Neg logit contributions | |
ascript | -0.627 |
lect | -0.541 |
spir | -0.533 |
NK | -0.524 |
Roll | -0.512 |
Token | of |
Token ID | 286 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
姫 | +1.783 |
� | +1.727 |
andr | +1.670 |
ennial | +1.624 |
sand | +1.624 |
Neg logit contributions | |
ascript | -2.125 |
lect | -1.836 |
spir | -1.805 |
NK | -1.774 |
Roll | -1.734 |
Token | Merc |
Token ID | 12185 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
� | +4.705 |
姫 | +4.698 |
andr | +4.545 |
oling | +4.428 |
sand | +4.418 |
Neg logit contributions | |
ascript | -6.369 |
lect | -5.354 |
spir | -5.302 |
NK | -5.057 |
glers | -5.057 |
Token | ia |
Token ID | 544 |
Feature activation | +6.01e-02 |
Pos logit contributions | |
姫 | +0.168 |
� | +0.162 |
andr | +0.157 |
sand | +0.153 |
ennial | +0.152 |
Neg logit contributions | |
ascript | -0.151 |
lect | -0.129 |
spir | -0.127 |
NK | -0.124 |
Roll | -0.120 |
Token | , |
Token ID | 11 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
姫 | +0.280 |
� | +0.269 |
andr | +0.261 |
sand | +0.257 |
andestine | +0.254 |
Neg logit contributions | |
ascript | -0.308 |
lect | -0.269 |
spir | -0.262 |
NK | -0.257 |
Roll | -0.252 |
Token | Mor |
Token ID | 3461 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
姫 | +1.734 |
� | +1.683 |
andr | +1.628 |
sand | +1.586 |
ennial | +1.586 |
Neg logit contributions | |
ascript | -1.929 |
lect | -1.657 |
spir | -1.629 |
NK | -1.601 |
Roll | -1.562 |
Token | car |
Token ID | 7718 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
ascript | +0.044 |
lect | +0.039 |
spir | +0.038 |
NK | +0.037 |
Roll | +0.036 |
Neg logit contributions | |
姫 | -0.037 |
� | -0.036 |
andr | -0.034 |
sand | -0.033 |
ennial | -0.033 |
Token | their |
Token ID | 511 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
姫 | +2.159 |
� | +2.094 |
andr | +2.029 |
ennial | +1.984 |
sand | +1.981 |
Neg logit contributions | |
ascript | -2.191 |
lect | -1.878 |
spir | -1.848 |
NK | -1.812 |
Roll | -1.760 |
Token | lost |
Token ID | 2626 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
姫 | +2.427 |
� | +2.337 |
andr | +2.254 |
ennial | +2.193 |
sand | +2.193 |
Neg logit contributions | |
ascript | -3.288 |
lect | -2.876 |
spir | -2.838 |
NK | -2.788 |
Roll | -2.725 |
Token | leader |
Token ID | 3554 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
ascript | +1.945 |
lect | +1.809 |
spir | +1.803 |
NK | +1.773 |
Roll | +1.686 |
Neg logit contributions | |
姫 | -1.588 |
� | -1.466 |
ennial | -1.419 |
andestine | -1.418 |
andr | -1.414 |
Token | O |
Token ID | 440 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
姫 | +2.686 |
� | +2.608 |
andr | +2.525 |
ennial | +2.456 |
sand | +2.454 |
Neg logit contributions | |
ascript | -3.061 |
lect | -2.623 |
spir | -2.577 |
NK | -2.531 |
Roll | -2.472 |
Token | der |
Token ID | 1082 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
姫 | +0.570 |
� | +0.552 |
andr | +0.525 |
andestine | +0.521 |
oling | +0.520 |
Neg logit contributions | |
ascript | -0.361 |
lect | -0.303 |
spir | -0.292 |
NK | -0.272 |
glers | -0.267 |
Token | us |
Token ID | 385 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
姫 | +1.915 |
� | +1.852 |
andr | +1.739 |
ennial | +1.694 |
andestine | +1.681 |
Neg logit contributions | |
ascript | -2.807 |
lect | -2.500 |
spir | -2.462 |
NK | -2.394 |
Roll | -2.344 |
Token | and |
Token ID | 290 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
姫 | +5.224 |
� | +5.005 |
andr | +4.987 |
ennial | +4.717 |
extinction | +4.640 |
Neg logit contributions | |
ascript | -7.073 |
lect | -6.026 |
NK | -5.897 |
spir | -5.877 |
glers | -5.750 |
Token | they |
Token ID | 484 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
姫 | +2.860 |
� | +2.779 |
andr | +2.691 |
ennial | +2.625 |
sand | +2.623 |
Neg logit contributions | |
ascript | -2.976 |
lect | -2.546 |
spir | -2.502 |
NK | -2.456 |
Roll | -2.394 |
Token | pay |
Token ID | 1414 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
姫 | +1.648 |
� | +1.596 |
andr | +1.543 |
ennial | +1.510 |
sand | +1.502 |
Neg logit contributions | |
ascript | -1.718 |
lect | -1.474 |
spir | -1.458 |
NK | -1.423 |
Roll | -1.387 |
Token | ed |
Token ID | 276 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
姫 | +1.577 |
� | +1.520 |
andr | +1.468 |
sand | +1.437 |
ennial | +1.434 |
Neg logit contributions | |
ascript | -1.648 |
lect | -1.429 |
spir | -1.402 |
NK | -1.378 |
Roll | -1.341 |
Token | homage |
Token ID | 31044 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
姫 | +1.067 |
� | +1.025 |
andr | +0.990 |
ennial | +0.972 |
sand | +0.969 |
Neg logit contributions | |
ascript | -1.072 |
lect | -0.938 |
spir | -0.915 |
NK | -0.909 |
Roll | -0.879 |
Token | 10 |
Token ID | 838 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
姫 | +2.191 |
� | +2.103 |
andr | +2.086 |
ennial | +2.050 |
sand | +2.013 |
Neg logit contributions | |
ascript | -3.180 |
lect | -2.735 |
spir | -2.724 |
NK | -2.643 |
Citiz | -2.613 |
Token | 52 |
Token ID | 4309 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
ascript | +0.962 |
lect | +0.887 |
spir | +0.879 |
Roll | +0.856 |
NK | +0.852 |
Neg logit contributions | |
姫 | -1.088 |
� | -1.064 |
andr | -1.012 |
andestine | -1.010 |
ennial | -0.998 |
Token | . |
Token ID | 764 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
姫 | +0.149 |
� | +0.147 |
andr | +0.141 |
andestine | +0.139 |
oling | +0.138 |
Neg logit contributions | |
ascript | -0.097 |
lect | -0.084 |
NK | -0.082 |
spir | -0.082 |
Roll | -0.080 |
Token |
|
Token ID | 198 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
ascript | +0.720 |
lect | +0.661 |
spir | +0.653 |
NK | +0.650 |
Roll | +0.639 |
Neg logit contributions | |
姫 | -0.686 |
� | -0.670 |
andestine | -0.640 |
andr | -0.637 |
oling | -0.626 |
Token |
|
Token ID | 198 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
姫 | +0.422 |
� | +0.409 |
andr | +0.387 |
andestine | +0.380 |
oling | +0.376 |
Neg logit contributions | |
ascript | -0.470 |
lect | -0.432 |
spir | -0.425 |
NK | -0.419 |
Roll | -0.412 |
Token | Google |
Token ID | 11708 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
姫 | +2.218 |
� | +2.163 |
andr | +2.098 |
andestine | +2.038 |
ennial | +2.036 |
Neg logit contributions | |
ascript | -2.140 |
lect | -1.869 |
spir | -1.832 |
Roll | -1.783 |
NK | -1.774 |
Token | Scholar |
Token ID | 11713 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
姫 | +4.277 |
� | +4.171 |
andr | +3.993 |
sand | +3.933 |
ennial | +3.900 |
Neg logit contributions | |
ascript | -5.788 |
lect | -4.934 |
spir | -4.892 |
NK | -4.764 |
hemy | -4.690 |
Token | Crossref |
Token ID | 23613 |
Feature activation | +7.52e-01 |
Pos logit contributions | |
姫 | +3.518 |
� | +3.432 |
ennial | +3.180 |
sand | +3.164 |
andr | +3.144 |
Neg logit contributions | |
ascript | -5.335 |
lect | -4.593 |
spir | -4.570 |
glers | -4.444 |
hemy | -4.439 |
Token | | |
Token ID | 930 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
姫 | +2.575 |
� | +2.479 |
andr | +2.275 |
ennial | +2.243 |
sand | +2.232 |
Neg logit contributions | |
ascript | -4.701 |
lect | -4.106 |
spir | -4.104 |
NK | -4.026 |
glers | -3.942 |
Token | Medline |
Token ID | 34582 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
姫 | +1.286 |
� | +1.250 |
andr | +1.211 |
ennial | +1.181 |
sand | +1.181 |
Neg logit contributions | |
ascript | -1.313 |
lect | -1.119 |
spir | -1.100 |
NK | -1.078 |
Roll | -1.051 |
Token | | |
Token ID | 930 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
姫 | +1.618 |
� | +1.543 |
andr | +1.441 |
ennial | +1.411 |
sand | +1.396 |
Neg logit contributions | |
ascript | -2.929 |
lect | -2.581 |
spir | -2.576 |
NK | -2.510 |
Citiz | -2.481 |
Token | and |
Token ID | 290 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
姫 | +0.320 |
� | +0.310 |
andr | +0.300 |
sand | +0.292 |
ennial | +0.291 |
Neg logit contributions | |
ascript | -0.350 |
lect | -0.302 |
spir | -0.295 |
NK | -0.291 |
Roll | -0.283 |
Token | playing |
Token ID | 2712 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
ascript | +1.460 |
lect | +1.315 |
NK | +1.285 |
spir | +1.241 |
glers | +1.195 |
Neg logit contributions | |
姫 | -1.474 |
� | -1.395 |
andr | -1.390 |
ennial | -1.353 |
sand | -1.345 |
Token | fields |
Token ID | 7032 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
ascript | +1.885 |
lect | +1.678 |
spir | +1.640 |
NK | +1.614 |
pen | +1.590 |
Neg logit contributions | |
姫 | -2.046 |
� | -1.955 |
andr | -1.925 |
ennial | -1.870 |
sand | -1.850 |
Token | ," |
Token ID | 553 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
姫 | +1.577 |
� | +1.533 |
andr | +1.485 |
ennial | +1.448 |
sand | +1.447 |
Neg logit contributions | |
ascript | -1.602 |
lect | -1.367 |
spir | -1.342 |
NK | -1.318 |
Roll | -1.282 |
Token | HR |
Token ID | 15172 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
姫 | +2.314 |
� | +2.270 |
andr | +2.194 |
ennial | +2.136 |
sand | +2.131 |
Neg logit contributions | |
ascript | -2.636 |
lect | -2.247 |
spir | -2.224 |
NK | -2.151 |
Roll | -2.114 |
Token | W |
Token ID | 54 |
Feature activation | +9.63e-01 |
Pos logit contributions | |
姫 | +2.909 |
� | +2.810 |
andr | +2.702 |
ennial | +2.620 |
sand | +2.619 |
Neg logit contributions | |
ascript | -4.310 |
lect | -3.769 |
spir | -3.715 |
NK | -3.655 |
Roll | -3.578 |
Token | researcher |
Token ID | 13453 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
姫 | +3.309 |
� | +3.250 |
andr | +3.109 |
jan | +2.989 |
ennial | +2.989 |
Neg logit contributions | |
ascript | -6.041 |
spir | -5.271 |
lect | -5.253 |
Roll | -5.030 |
NK | -5.027 |
Token | La |
Token ID | 4689 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
ascript | +0.932 |
lect | +0.829 |
NK | +0.825 |
spir | +0.797 |
Roll | +0.782 |
Neg logit contributions | |
姫 | -0.788 |
� | -0.738 |
andr | -0.707 |
ennial | -0.697 |
oling | -0.695 |
Token | et |
Token ID | 316 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
ascript | +3.229 |
lect | +2.856 |
spir | +2.759 |
Roll | +2.693 |
NK | +2.682 |
Neg logit contributions | |
姫 | -3.048 |
� | -2.765 |
sand | -2.671 |
andr | -2.665 |
andestine | -2.655 |
Token | itia |
Token ID | 36723 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
姫 | +1.152 |
� | +1.112 |
andr | +1.065 |
ennial | +1.043 |
sand | +1.042 |
Neg logit contributions | |
ascript | -1.232 |
lect | -1.072 |
spir | -1.052 |
NK | -1.029 |
Roll | -1.007 |
Token | B |
Token ID | 347 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
ascript | +1.725 |
lect | +1.527 |
NK | +1.514 |
spir | +1.496 |
Roll | +1.460 |
Neg logit contributions | |
姫 | -1.486 |
� | -1.398 |
andr | -1.348 |
ennial | -1.325 |
oling | -1.323 |
Token |
|
Token ID | 198 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
ascript | +0.875 |
lect | +0.817 |
spir | +0.809 |
NK | +0.787 |
Roll | +0.779 |
Neg logit contributions | |
姫 | -0.916 |
� | -0.890 |
andr | -0.842 |
andestine | -0.830 |
oling | -0.824 |
Token | Early |
Token ID | 20457 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ascript | +0.964 |
lect | +0.900 |
Roll | +0.890 |
spir | +0.880 |
NK | +0.834 |
Neg logit contributions | |
姫 | -0.909 |
� | -0.872 |
oling | -0.837 |
andr | -0.834 |
andestine | -0.832 |
Token | in |
Token ID | 287 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
姫 | +0.801 |
� | +0.775 |
andr | +0.756 |
sand | +0.735 |
oling | +0.732 |
Neg logit contributions | |
ascript | -0.738 |
lect | -0.641 |
spir | -0.621 |
NK | -0.617 |
Roll | -0.603 |
Token | the |
Token ID | 262 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
ascript | +0.670 |
lect | +0.591 |
spir | +0.578 |
NK | +0.569 |
Roll | +0.560 |
Neg logit contributions | |
姫 | -0.806 |
� | -0.773 |
andr | -0.753 |
ennial | -0.733 |
sand | -0.732 |
Token | first |
Token ID | 717 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
ascript | +0.251 |
lect | +0.220 |
spir | +0.216 |
NK | +0.214 |
Roll | +0.210 |
Neg logit contributions | |
姫 | -0.245 |
� | -0.228 |
andr | -0.223 |
andestine | -0.216 |
ennial | -0.215 |
Token | quarter |
Token ID | 3860 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
ascript | +1.556 |
lect | +1.330 |
spir | +1.309 |
NK | +1.301 |
Roll | +1.257 |
Neg logit contributions | |
姫 | -1.899 |
� | -1.775 |
andr | -1.740 |
sand | -1.691 |
andestine | -1.690 |
Token | Carlton |
Token ID | 44348 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
姫 | +3.979 |
� | +3.927 |
andr | +3.723 |
sand | +3.652 |
ennial | +3.635 |
Neg logit contributions | |
ascript | -5.712 |
lect | -4.886 |
spir | -4.819 |
NK | -4.691 |
Citiz | -4.618 |
Token | � |
Token ID | 447 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
姫 | +0.132 |
� | +0.128 |
andr | +0.124 |
ennial | +0.122 |
oling | +0.122 |
Neg logit contributions | |
ascript | -0.147 |
lect | -0.129 |
spir | -0.128 |
NK | -0.124 |
Roll | -0.123 |
Token | � |
Token ID | 247 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
ascript | +0.186 |
lect | +0.153 |
spir | +0.152 |
Roll | +0.143 |
NK | +0.139 |
Neg logit contributions | |
姫 | -0.336 |
andr | -0.317 |
andestine | -0.316 |
oling | -0.314 |
extinction | -0.312 |
Token | s |
Token ID | 82 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
姫 | +0.291 |
� | +0.281 |
andr | +0.270 |
ennial | +0.265 |
andestine | +0.264 |
Neg logit contributions | |
ascript | -0.385 |
spir | -0.350 |
lect | -0.350 |
Roll | -0.332 |
NK | -0.332 |
Token | Mitch |
Token ID | 20472 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
姫 | +1.560 |
� | +1.494 |
andr | +1.443 |
ennial | +1.409 |
sand | +1.405 |
Neg logit contributions | |
ascript | -1.964 |
lect | -1.728 |
spir | -1.694 |
NK | -1.686 |
Roll | -1.640 |
Token | debt |
Token ID | 5057 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
姫 | +0.387 |
� | +0.372 |
andr | +0.361 |
sand | +0.352 |
ennial | +0.350 |
Neg logit contributions | |
ascript | -0.406 |
lect | -0.347 |
spir | -0.342 |
NK | -0.338 |
Roll | -0.327 |
Token | sustainability |
Token ID | 26809 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
ascript | +0.863 |
lect | +0.748 |
spir | +0.734 |
NK | +0.714 |
Roll | +0.701 |
Neg logit contributions | |
姫 | -0.822 |
� | -0.787 |
andr | -0.763 |
sand | -0.742 |
ennial | -0.741 |
Token | and |
Token ID | 290 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
ascript | +1.141 |
lect | +0.983 |
spir | +0.965 |
NK | +0.943 |
Roll | +0.925 |
Neg logit contributions | |
姫 | -1.072 |
� | -1.033 |
andr | -1.005 |
andestine | -0.972 |
ennial | -0.971 |
Token | he |
Token ID | 339 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
姫 | +0.742 |
� | +0.709 |
andr | +0.688 |
sand | +0.675 |
ennial | +0.671 |
Neg logit contributions | |
ascript | -0.798 |
lect | -0.688 |
spir | -0.680 |
NK | -0.671 |
Roll | -0.652 |
Token | would |
Token ID | 561 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
姫 | +0.447 |
� | +0.430 |
andr | +0.413 |
sand | +0.402 |
ennial | +0.401 |
Neg logit contributions | |
ascript | -0.675 |
lect | -0.594 |
spir | -0.585 |
NK | -0.569 |
Roll | -0.560 |
Token | instead |
Token ID | 2427 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
姫 | +0.918 |
� | +0.887 |
andr | +0.859 |
ennial | +0.838 |
sand | +0.837 |
Neg logit contributions | |
ascript | -1.009 |
lect | -0.865 |
spir | -0.849 |
NK | -0.839 |
Roll | -0.818 |
Token | rely |
Token ID | 8814 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
姫 | +1.306 |
� | +1.254 |
andr | +1.207 |
sand | +1.177 |
ennial | +1.173 |
Neg logit contributions | |
ascript | -1.772 |
lect | -1.551 |
spir | -1.521 |
NK | -1.503 |
Roll | -1.466 |
Token | on |
Token ID | 319 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
姫 | +0.514 |
� | +0.505 |
andr | +0.487 |
ennial | +0.479 |
sand | +0.473 |
Neg logit contributions | |
ascript | -0.438 |
lect | -0.364 |
spir | -0.364 |
NK | -0.353 |
Roll | -0.343 |
Token | a |
Token ID | 257 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
ascript | +0.161 |
lect | +0.139 |
spir | +0.136 |
NK | +0.135 |
Roll | +0.130 |
Neg logit contributions | |
姫 | -0.159 |
� | -0.151 |
andr | -0.148 |
sand | -0.145 |
ennial | -0.143 |
Token | 2016 |
Token ID | 1584 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
ascript | +0.871 |
lect | +0.760 |
spir | +0.741 |
NK | +0.733 |
Roll | +0.708 |
Neg logit contributions | |
姫 | -0.850 |
� | -0.811 |
andr | -0.795 |
sand | -0.776 |
andestine | -0.773 |
Token | forecast |
Token ID | 11092 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
ascript | +1.261 |
lect | +1.109 |
spir | +1.085 |
NK | +1.078 |
Roll | +1.042 |
Neg logit contributions | |
姫 | -1.179 |
� | -1.117 |
andr | -1.097 |
sand | -1.072 |
andestine | -1.060 |
Token | to |
Token ID | 284 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
姫 | +0.386 |
� | +0.368 |
andr | +0.357 |
ennial | +0.344 |
sand | +0.342 |
Neg logit contributions | |
ascript | -0.424 |
lect | -0.367 |
NK | -0.359 |
spir | -0.359 |
Roll | -0.347 |
Token | help |
Token ID | 1037 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
姫 | +0.497 |
� | +0.480 |
andr | +0.466 |
andestine | +0.458 |
sand | +0.457 |
Neg logit contributions | |
ascript | -0.532 |
lect | -0.475 |
NK | -0.457 |
spir | -0.454 |
Roll | -0.446 |
Token | protect |
Token ID | 1805 |
Feature activation | +3.08e-03 |
Pos logit contributions | |
姫 | +1.251 |
� | +1.195 |
andr | +1.173 |
sand | +1.140 |
andestine | +1.140 |
Neg logit contributions | |
ascript | -1.338 |
lect | -1.179 |
spir | -1.151 |
NK | -1.137 |
Roll | -1.110 |
Token | the |
Token ID | 262 |
Feature activation | -8.65e-02 |
Pos logit contributions | |
姫 | +0.016 |
� | +0.015 |
andr | +0.014 |
sand | +0.014 |
andestine | +0.014 |
Neg logit contributions | |
ascript | -0.015 |
lect | -0.013 |
spir | -0.013 |
NK | -0.012 |
Roll | -0.012 |
Token | insurance |
Token ID | 5096 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
ascript | +0.443 |
lect | +0.395 |
spir | +0.392 |
NK | +0.383 |
Roll | +0.371 |
Neg logit contributions | |
姫 | -0.401 |
� | -0.379 |
andr | -0.362 |
andestine | -0.360 |
sand | -0.355 |
Token | companies |
Token ID | 2706 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
姫 | +0.876 |
� | +0.849 |
andr | +0.818 |
ennial | +0.798 |
sand | +0.794 |
Neg logit contributions | |
ascript | -0.845 |
lect | -0.729 |
spir | -0.713 |
NK | -0.699 |
Roll | -0.683 |
Token | from |
Token ID | 422 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
姫 | +1.419 |
� | +1.371 |
andr | +1.328 |
ennial | +1.295 |
sand | +1.294 |
Neg logit contributions | |
ascript | -1.438 |
lect | -1.242 |
spir | -1.215 |
NK | -1.194 |
Roll | -1.164 |
Token | extreme |
Token ID | 3257 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
姫 | +1.258 |
� | +1.205 |
andr | +1.175 |
sand | +1.139 |
ennial | +1.138 |
Neg logit contributions | |
ascript | -1.352 |
lect | -1.176 |
spir | -1.146 |
NK | -1.138 |
Roll | -1.107 |
Token | enrollment |
Token ID | 20753 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
ascript | +0.310 |
lect | +0.274 |
NK | +0.270 |
spir | +0.269 |
Roll | +0.262 |
Neg logit contributions | |
姫 | -0.283 |
� | -0.266 |
andr | -0.259 |
ennial | -0.249 |
jan | -0.249 |
Token | of |
Token ID | 286 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
姫 | +1.649 |
� | +1.596 |
andr | +1.543 |
ennial | +1.499 |
sand | +1.495 |
Neg logit contributions | |
ascript | -1.874 |
lect | -1.616 |
spir | -1.590 |
NK | -1.566 |
Roll | -1.531 |
Token | high |
Token ID | 1029 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
姫 | +0.639 |
� | +0.609 |
andr | +0.591 |
ennial | +0.575 |
sand | +0.573 |
Neg logit contributions | |
ascript | -0.645 |
lect | -0.564 |
spir | -0.550 |
NK | -0.550 |
Roll | -0.531 |
Token | the |
Token ID | 262 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
姫 | +0.398 |
� | +0.382 |
andr | +0.370 |
ennial | +0.363 |
sand | +0.362 |
Neg logit contributions | |
ascript | -0.382 |
lect | -0.328 |
spir | -0.319 |
NK | -0.319 |
Roll | -0.305 |
Token | other |
Token ID | 584 |
Feature activation | -3.76e-02 |
Pos logit contributions | |
姫 | +0.679 |
� | +0.645 |
andr | +0.627 |
ennial | +0.610 |
sand | +0.604 |
Neg logit contributions | |
ascript | -0.714 |
lect | -0.618 |
NK | -0.603 |
spir | -0.601 |
Roll | -0.578 |
Token | languages |
Token ID | 8950 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
ascript | +0.200 |
lect | +0.176 |
NK | +0.169 |
spir | +0.168 |
Roll | +0.163 |
Neg logit contributions | |
姫 | -0.175 |
� | -0.164 |
andr | -0.160 |
ennial | -0.156 |
sand | -0.154 |
Token | � |
Token ID | 564 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
姫 | +2.287 |
� | +2.214 |
andr | +2.140 |
ennial | +2.083 |
sand | +2.081 |
Neg logit contributions | |
ascript | -2.729 |
lect | -2.358 |
spir | -2.318 |
NK | -2.280 |
Roll | -2.225 |
Token | � |
Token ID | 250 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
ascript | +0.199 |
lect | +0.175 |
spir | +0.172 |
NK | +0.169 |
Roll | +0.167 |
Neg logit contributions | |
姫 | -0.124 |
andr | -0.116 |
� | -0.114 |
sand | -0.112 |
ennial | -0.111 |
Token | warn |
Token ID | 40539 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
姫 | +1.157 |
� | +1.114 |
andr | +1.079 |
ennial | +1.060 |
andestine | +1.047 |
Neg logit contributions | |
ascript | -1.196 |
lect | -1.036 |
spir | -1.016 |
NK | -0.985 |
Roll | -0.973 |
Token | ed |
Token ID | 276 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
姫 | +0.696 |
� | +0.674 |
andr | +0.651 |
sand | +0.632 |
ennial | +0.632 |
Neg logit contributions | |
ascript | -0.703 |
lect | -0.609 |
spir | -0.596 |
NK | -0.584 |
Roll | -0.567 |
Token | � |
Token ID | 447 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
姫 | +1.433 |
� | +1.380 |
andr | +1.335 |
sand | +1.301 |
ennial | +1.300 |
Neg logit contributions | |
ascript | -1.574 |
lect | -1.368 |
spir | -1.333 |
NK | -1.317 |
Roll | -1.277 |
Token | � |
Token ID | 251 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
ascript | +0.526 |
lect | +0.450 |
spir | +0.444 |
NK | +0.430 |
Roll | +0.428 |
Neg logit contributions | |
姫 | -0.484 |
andr | -0.460 |
andestine | -0.457 |
sand | -0.446 |
oling | -0.444 |
Token | may |
Token ID | 743 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
姫 | +1.693 |
� | +1.634 |
andr | +1.578 |
ennial | +1.541 |
sand | +1.531 |
Neg logit contributions | |
ascript | -1.941 |
lect | -1.683 |
spir | -1.652 |
NK | -1.627 |
Roll | -1.585 |
Token | be |
Token ID | 307 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
ascript | +2.531 |
lect | +2.250 |
spir | +2.171 |
NK | +2.146 |
Roll | +2.071 |
Neg logit contributions | |
姫 | -2.111 |
� | -2.015 |
andr | -1.924 |
andestine | -1.898 |
sand | -1.870 |
Token | . |
Token ID | 13 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
姫 | +3.889 |
� | +3.792 |
andr | +3.693 |
sand | +3.626 |
ennial | +3.603 |
Neg logit contributions | |
ascript | -4.732 |
lect | -4.026 |
spir | -3.974 |
NK | -3.898 |
Citiz | -3.845 |
Token | Mor |
Token ID | 3461 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
姫 | +0.843 |
� | +0.810 |
andr | +0.782 |
ennial | +0.756 |
sand | +0.756 |
Neg logit contributions | |
ascript | -1.190 |
lect | -1.042 |
spir | -1.026 |
NK | -1.011 |
Roll | -0.990 |
Token | neau |
Token ID | 46533 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
姫 | +1.160 |
� | +1.114 |
andr | +1.043 |
andestine | +1.037 |
extinction | +1.026 |
Neg logit contributions | |
ascript | -1.154 |
lect | -1.016 |
spir | -0.977 |
Roll | -0.930 |
NK | -0.922 |
Token | was |
Token ID | 373 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
姫 | +1.244 |
� | +1.200 |
andr | +1.157 |
ennial | +1.129 |
sand | +1.124 |
Neg logit contributions | |
ascript | -1.540 |
lect | -1.335 |
spir | -1.315 |
NK | -1.291 |
Roll | -1.260 |
Token | repeatedly |
Token ID | 7830 |
Feature activation | +4.41e-01 |
Pos logit contributions | |
姫 | +0.094 |
� | +0.088 |
andr | +0.088 |
ennial | +0.085 |
sand | +0.085 |
Neg logit contributions | |
ascript | -0.090 |
lect | -0.077 |
spir | -0.074 |
NK | -0.074 |
Roll | -0.072 |
Token | on |
Token ID | 319 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
姫 | +2.099 |
� | +2.010 |
andr | +1.969 |
ennial | +1.916 |
sand | +1.910 |
Neg logit contributions | |
ascript | -2.249 |
lect | -1.939 |
spir | -1.894 |
NK | -1.872 |
Roll | -1.818 |
Token | the |
Token ID | 262 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
姫 | +2.002 |
� | +1.939 |
andr | +1.880 |
sand | +1.831 |
ennial | +1.831 |
Neg logit contributions | |
ascript | -2.056 |
lect | -1.765 |
spir | -1.729 |
NK | -1.701 |
Roll | -1.655 |
Token | firing |
Token ID | 9645 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
姫 | +1.848 |
� | +1.779 |
andr | +1.720 |
ennial | +1.668 |
sand | +1.662 |
Neg logit contributions | |
ascript | -2.183 |
lect | -1.905 |
spir | -1.855 |
NK | -1.837 |
Roll | -1.790 |
Token | line |
Token ID | 1627 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
姫 | +0.049 |
� | +0.047 |
andr | +0.045 |
andestine | +0.044 |
ennial | +0.044 |
Neg logit contributions | |
ascript | -0.052 |
lect | -0.046 |
NK | -0.044 |
spir | -0.044 |
Roll | -0.043 |
Token | during |
Token ID | 1141 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
姫 | +0.631 |
� | +0.607 |
andr | +0.587 |
ennial | +0.570 |
sand | +0.569 |
Neg logit contributions | |
ascript | -0.620 |
lect | -0.532 |
spir | -0.522 |
NK | -0.514 |
Roll | -0.496 |
Token | Commons |
Token ID | 13815 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
姫 | +0.704 |
� | +0.679 |
andr | +0.660 |
sand | +0.641 |
ennial | +0.641 |
Neg logit contributions | |
ascript | -0.761 |
lect | -0.656 |
spir | -0.639 |
NK | -0.630 |
Roll | -0.616 |
Token | story |
Token ID | 1621 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
ascript | +2.075 |
lect | +1.825 |
NK | +1.824 |
spir | +1.780 |
Roll | +1.720 |
Neg logit contributions | |
姫 | -1.806 |
� | -1.700 |
andr | -1.645 |
andestine | -1.597 |
sand | -1.594 |
Token | as |
Token ID | 355 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ascript | +0.619 |
lect | +0.530 |
spir | +0.516 |
NK | +0.505 |
Roll | +0.487 |
Neg logit contributions | |
姫 | -0.703 |
� | -0.674 |
andr | -0.662 |
ennial | -0.641 |
sand | -0.637 |
Token | possible |
Token ID | 1744 |
Feature activation | -8.00e-02 |
Pos logit contributions | |
ascript | +0.515 |
lect | +0.444 |
spir | +0.434 |
NK | +0.429 |
Roll | +0.415 |
Neg logit contributions | |
姫 | -0.516 |
� | -0.495 |
andr | -0.483 |
ennial | -0.472 |
sand | -0.469 |
Token | . |
Token ID | 13 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
ascript | +0.395 |
lect | +0.340 |
spir | +0.330 |
NK | +0.329 |
Roll | +0.317 |
Neg logit contributions | |
姫 | -0.393 |
� | -0.379 |
andr | -0.367 |
sand | -0.356 |
ennial | -0.355 |
Token | I |
Token ID | 314 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
ascript | +0.250 |
lect | +0.220 |
NK | +0.218 |
spir | +0.217 |
Roll | +0.212 |
Neg logit contributions | |
姫 | -0.242 |
� | -0.232 |
andr | -0.226 |
ennial | -0.223 |
oling | -0.223 |
Token | didn |
Token ID | 1422 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
姫 | +0.123 |
� | +0.118 |
andr | +0.115 |
ennial | +0.111 |
sand | +0.111 |
Neg logit contributions | |
ascript | -0.139 |
lect | -0.121 |
spir | -0.119 |
NK | -0.118 |
Roll | -0.113 |
Token | � |
Token ID | 447 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
姫 | +0.916 |
� | +0.875 |
andr | +0.834 |
andestine | +0.808 |
ennial | +0.807 |
Neg logit contributions | |
ascript | -1.340 |
lect | -1.216 |
spir | -1.196 |
NK | -1.174 |
Roll | -1.158 |
Token | � |
Token ID | 247 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
姫 | +0.236 |
andr | +0.219 |
sand | +0.215 |
ennial | +0.212 |
andestine | +0.211 |
Neg logit contributions | |
ascript | -0.321 |
lect | -0.283 |
spir | -0.280 |
NK | -0.268 |
Roll | -0.266 |
Token | t |
Token ID | 83 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
姫 | +0.786 |
� | +0.752 |
andr | +0.734 |
ennial | +0.721 |
sand | +0.717 |
Neg logit contributions | |
ascript | -0.736 |
lect | -0.643 |
spir | -0.620 |
NK | -0.600 |
Roll | -0.586 |
Token | really |
Token ID | 1107 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
姫 | +1.397 |
� | +1.347 |
andr | +1.300 |
ennial | +1.267 |
sand | +1.261 |
Neg logit contributions | |
ascript | -1.656 |
lect | -1.438 |
spir | -1.411 |
NK | -1.387 |
Roll | -1.347 |
Token | have |
Token ID | 423 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
姫 | +1.338 |
� | +1.286 |
andr | +1.242 |
ennial | +1.210 |
sand | +1.206 |
Neg logit contributions | |
ascript | -1.612 |
lect | -1.403 |
spir | -1.376 |
NK | -1.356 |
Roll | -1.316 |
Token | Reuters |
Token ID | 8428 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
ascript | +1.968 |
NK | +1.771 |
lect | +1.739 |
Roll | +1.632 |
spir | +1.631 |
Neg logit contributions | |
姫 | -2.546 |
� | -2.407 |
andr | -2.355 |
ennial | -2.317 |
oling | -2.288 |
Token | the |
Token ID | 262 |
Feature activation | +9.82e-03 |
Pos logit contributions | |
ascript | +0.440 |
lect | +0.377 |
spir | +0.368 |
NK | +0.368 |
Roll | +0.354 |
Neg logit contributions | |
姫 | -0.440 |
� | -0.425 |
andr | -0.414 |
sand | -0.404 |
ennial | -0.403 |
Token | decision |
Token ID | 2551 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
姫 | +0.047 |
� | +0.044 |
andr | +0.043 |
ennial | +0.042 |
sand | +0.042 |
Neg logit contributions | |
ascript | -0.050 |
lect | -0.044 |
NK | -0.043 |
spir | -0.043 |
Roll | -0.041 |
Token | came |
Token ID | 1625 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
ascript | +0.592 |
lect | +0.520 |
spir | +0.506 |
NK | +0.505 |
glers | +0.480 |
Neg logit contributions | |
姫 | -0.527 |
� | -0.503 |
andr | -0.480 |
ennial | -0.469 |
sand | -0.465 |
Token | as |
Token ID | 355 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
ascript | +1.035 |
lect | +0.856 |
NK | +0.845 |
spir | +0.837 |
glers | +0.796 |
Neg logit contributions | |
姫 | -1.191 |
� | -1.142 |
andr | -1.107 |
ennial | -1.084 |
sand | -1.077 |
Token | a |
Token ID | 257 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
ascript | +2.314 |
NK | +2.053 |
lect | +2.006 |
spir | +1.970 |
FTA | +1.938 |
Neg logit contributions | |
姫 | -2.403 |
� | -2.308 |
andr | -2.243 |
ennial | -2.198 |
sand | -2.168 |
Token | " |
Token ID | 366 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
ascript | +2.251 |
NK | +2.008 |
lect | +2.007 |
spir | +1.963 |
FTA | +1.884 |
Neg logit contributions | |
姫 | -2.437 |
� | -2.306 |
andr | -2.214 |
andestine | -2.186 |
ennial | -2.155 |
Token | big |
Token ID | 14261 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
ascript | +1.521 |
lect | +1.368 |
spir | +1.360 |
Roll | +1.284 |
pen | +1.257 |
Neg logit contributions | |
姫 | -1.351 |
� | -1.283 |
andr | -1.237 |
andestine | -1.215 |
extinction | -1.204 |
Token | shock |
Token ID | 6380 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
ascript | +1.607 |
lect | +1.396 |
spir | +1.386 |
NK | +1.367 |
Roll | +1.322 |
Neg logit contributions | |
姫 | -1.561 |
� | -1.505 |
andr | -1.463 |
andestine | -1.422 |
ennial | -1.406 |
Token | " |
Token ID | 1 |
Feature activation | +4.89e-02 |
Pos logit contributions | |
ascript | +2.066 |
lect | +1.789 |
spir | +1.739 |
NK | +1.695 |
Roll | +1.652 |
Neg logit contributions | |
姫 | -2.279 |
� | -2.193 |
andr | -2.093 |
ennial | -2.046 |
andestine | -2.034 |
Token | and |
Token ID | 290 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
姫 | +0.221 |
� | +0.212 |
andr | +0.203 |
ennial | +0.199 |
sand | +0.198 |
Neg logit contributions | |
ascript | -0.261 |
lect | -0.227 |
spir | -0.223 |
NK | -0.221 |
Roll | -0.214 |
Token | � |
Token ID | 447 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
姫 | +1.222 |
� | +1.186 |
andr | +1.151 |
sand | +1.121 |
ennial | +1.119 |
Neg logit contributions | |
ascript | -1.273 |
lect | -1.090 |
spir | -1.071 |
NK | -1.048 |
Roll | -1.025 |
Token | � |
Token ID | 251 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
姫 | +0.949 |
� | +0.924 |
andr | +0.887 |
ennial | +0.864 |
sand | +0.862 |
Neg logit contributions | |
ascript | -1.192 |
lect | -1.032 |
spir | -1.016 |
NK | -0.999 |
Roll | -0.976 |
Token | from |
Token ID | 422 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
姫 | +0.921 |
� | +0.894 |
andr | +0.866 |
ennial | +0.847 |
sand | +0.842 |
Neg logit contributions | |
ascript | -0.972 |
lect | -0.837 |
spir | -0.824 |
NK | -0.805 |
Roll | -0.787 |
Token | which |
Token ID | 543 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
姫 | +0.220 |
� | +0.213 |
andr | +0.207 |
sand | +0.202 |
ennial | +0.202 |
Neg logit contributions | |
ascript | -0.225 |
lect | -0.194 |
spir | -0.190 |
NK | -0.187 |
Roll | -0.182 |
Token | the |
Token ID | 262 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
ascript | +0.969 |
lect | +0.856 |
spir | +0.840 |
NK | +0.829 |
Roll | +0.804 |
Neg logit contributions | |
姫 | -0.809 |
� | -0.779 |
andr | -0.764 |
ennial | -0.744 |
sand | -0.742 |
Token | game |
Token ID | 983 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
ascript | +0.069 |
lect | +0.061 |
spir | +0.060 |
NK | +0.059 |
Roll | +0.057 |
Neg logit contributions | |
姫 | -0.057 |
� | -0.055 |
andr | -0.053 |
sand | -0.052 |
ennial | -0.052 |
Token | � |
Token ID | 447 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
ascript | +0.855 |
lect | +0.744 |
spir | +0.736 |
NK | +0.717 |
Roll | +0.701 |
Neg logit contributions | |
姫 | -0.747 |
� | -0.716 |
andr | -0.691 |
ennial | -0.682 |
sand | -0.681 |
Token | � |
Token ID | 247 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
ascript | +1.145 |
lect | +1.004 |
spir | +0.987 |
Roll | +0.962 |
NK | +0.960 |
Neg logit contributions | |
姫 | -0.930 |
andr | -0.862 |
andestine | -0.860 |
oling | -0.831 |
sand | -0.827 |
Token | s |
Token ID | 82 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
ascript | +0.597 |
lect | +0.526 |
spir | +0.522 |
NK | +0.502 |
Roll | +0.496 |
Neg logit contributions | |
姫 | -0.428 |
� | -0.411 |
andr | -0.397 |
ennial | -0.389 |
oling | -0.384 |
Token | story |
Token ID | 1621 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
ascript | +1.877 |
lect | +1.647 |
spir | +1.607 |
NK | +1.602 |
Roll | +1.558 |
Neg logit contributions | |
姫 | -1.734 |
� | -1.643 |
andr | -1.605 |
andestine | -1.587 |
sand | -1.586 |
Token | unfolds |
Token ID | 45995 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
ascript | +2.227 |
lect | +1.942 |
spir | +1.930 |
NK | +1.852 |
Roll | +1.838 |
Neg logit contributions | |
姫 | -1.950 |
� | -1.843 |
andr | -1.784 |
andestine | -1.750 |
ennial | -1.746 |
Token | It |
Token ID | 632 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
姫 | +1.056 |
� | +1.013 |
andr | +0.985 |
ennial | +0.970 |
andestine | +0.958 |
Neg logit contributions | |
ascript | -1.160 |
lect | -1.019 |
spir | -1.000 |
NK | -0.998 |
Roll | -0.977 |
Token | helps |
Token ID | 5419 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
姫 | +0.058 |
� | +0.056 |
andr | +0.054 |
ennial | +0.053 |
andestine | +0.053 |
Neg logit contributions | |
ascript | -0.057 |
lect | -0.049 |
spir | -0.048 |
NK | -0.046 |
Roll | -0.046 |
Token | immensely |
Token ID | 29731 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
ascript | +0.060 |
lect | +0.053 |
NK | +0.052 |
spir | +0.052 |
Roll | +0.050 |
Neg logit contributions | |
姫 | -0.048 |
� | -0.046 |
andr | -0.045 |
ennial | -0.044 |
sand | -0.043 |
Token | with |
Token ID | 351 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
ascript | +0.995 |
lect | +0.859 |
spir | +0.839 |
NK | +0.836 |
Roll | +0.796 |
Neg logit contributions | |
姫 | -1.056 |
� | -1.026 |
andr | -0.989 |
sand | -0.970 |
andestine | -0.968 |
Token | iTunes |
Token ID | 4830 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
ascript | +1.110 |
lect | +0.955 |
NK | +0.931 |
spir | +0.924 |
Roll | +0.886 |
Neg logit contributions | |
姫 | -1.173 |
� | -1.131 |
andr | -1.098 |
andestine | -1.067 |
ennial | -1.067 |
Token | search |
Token ID | 2989 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
ascript | +1.115 |
lect | +0.982 |
NK | +0.944 |
spir | +0.935 |
Roll | +0.915 |
Neg logit contributions | |
姫 | -1.067 |
� | -1.035 |
andr | -0.987 |
andestine | -0.975 |
ennial | -0.972 |
Token | and |
Token ID | 290 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
ascript | +0.787 |
lect | +0.682 |
NK | +0.660 |
spir | +0.656 |
Roll | +0.639 |
Neg logit contributions | |
姫 | -0.774 |
� | -0.756 |
andr | -0.722 |
ennial | -0.708 |
andestine | -0.706 |
Token | ranking |
Token ID | 12759 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ascript | +1.202 |
lect | +1.058 |
NK | +1.050 |
spir | +1.011 |
Roll | +0.999 |
Neg logit contributions | |
姫 | -1.115 |
� | -1.064 |
andr | -1.016 |
ennial | -1.006 |
andestine | -0.991 |
Token | , |
Token ID | 11 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
ascript | +0.665 |
lect | +0.568 |
NK | +0.559 |
spir | +0.549 |
Roll | +0.538 |
Neg logit contributions | |
姫 | -0.687 |
� | -0.667 |
andr | -0.643 |
ennial | -0.631 |
andestine | -0.630 |
Token | and |
Token ID | 290 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
ascript | +0.673 |
lect | +0.582 |
NK | +0.570 |
spir | +0.564 |
Roll | +0.549 |
Neg logit contributions | |
姫 | -0.645 |
� | -0.619 |
andr | -0.597 |
ennial | -0.587 |
andestine | -0.581 |
Token | helps |
Token ID | 5419 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ascript | +1.233 |
lect | +1.092 |
NK | +1.079 |
spir | +1.040 |
Roll | +1.030 |
Neg logit contributions | |
姫 | -1.186 |
� | -1.139 |
andr | -1.094 |
ennial | -1.073 |
andestine | -1.064 |
Token | obtain |
Token ID | 7330 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
姫 | +1.067 |
� | +1.021 |
andr | +1.001 |
ennial | +0.974 |
andestine | +0.973 |
Neg logit contributions | |
ascript | -1.186 |
lect | -1.020 |
spir | -0.995 |
NK | -0.988 |
Roll | -0.956 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ascript | +2.032 |
lect | +1.734 |
spir | +1.705 |
NK | +1.662 |
Roll | +1.604 |
Neg logit contributions | |
姫 | -2.139 |
� | -2.044 |
andr | -2.019 |
andestine | -1.971 |
jan | -1.965 |
Token | Ms |
Token ID | 6997 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
姫 | +0.582 |
� | +0.556 |
andr | +0.544 |
ennial | +0.537 |
andestine | +0.532 |
Neg logit contributions | |
ascript | -0.562 |
lect | -0.492 |
spir | -0.490 |
NK | -0.489 |
Roll | -0.473 |
Token | . |
Token ID | 13 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
姫 | +0.591 |
� | +0.574 |
andr | +0.555 |
ennial | +0.541 |
sand | +0.541 |
Neg logit contributions | |
ascript | -0.675 |
lect | -0.580 |
spir | -0.571 |
NK | -0.561 |
Roll | -0.547 |
Token | Gab |
Token ID | 12300 |
Feature activation | -7.62e-02 |
Pos logit contributions | |
姫 | +1.713 |
� | +1.662 |
andr | +1.608 |
ennial | +1.565 |
sand | +1.565 |
Neg logit contributions | |
ascript | -1.949 |
lect | -1.676 |
spir | -1.648 |
NK | -1.618 |
Roll | -1.579 |
Token | ald |
Token ID | 1940 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
ascript | +0.395 |
lect | +0.342 |
spir | +0.334 |
NK | +0.327 |
Roll | +0.320 |
Neg logit contributions | |
姫 | -0.357 |
� | -0.343 |
andr | -0.330 |
ennial | -0.321 |
andestine | -0.320 |
Token | on |
Token ID | 261 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
ascript | +2.190 |
lect | +1.897 |
spir | +1.825 |
NK | +1.757 |
Roll | +1.692 |
Neg logit contributions | |
姫 | -2.913 |
� | -2.823 |
andestine | -2.719 |
andr | -2.670 |
sand | -2.648 |
Token | � |
Token ID | 447 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
ascript | +0.263 |
lect | +0.229 |
spir | +0.223 |
NK | +0.218 |
Roll | +0.213 |
Neg logit contributions | |
姫 | -0.260 |
� | -0.250 |
andr | -0.243 |
ennial | -0.237 |
andestine | -0.237 |
Token | � |
Token ID | 247 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
姫 | +0.666 |
andr | +0.630 |
andestine | +0.623 |
sand | +0.613 |
oling | +0.613 |
Neg logit contributions | |
ascript | -0.414 |
lect | -0.343 |
spir | -0.336 |
Roll | -0.318 |
NK | -0.313 |
Token | s |
Token ID | 82 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
姫 | +0.223 |
� | +0.215 |
andr | +0.209 |
ennial | +0.201 |
andestine | +0.199 |
Neg logit contributions | |
ascript | -0.329 |
lect | -0.291 |
spir | -0.287 |
NK | -0.278 |
Roll | -0.274 |
Token | books |
Token ID | 3835 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
姫 | +0.670 |
� | +0.637 |
andr | +0.620 |
sand | +0.602 |
ennial | +0.602 |
Neg logit contributions | |
ascript | -0.782 |
lect | -0.681 |
spir | -0.674 |
NK | -0.661 |
Roll | -0.644 |
Token | we |
Token ID | 356 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
姫 | +0.740 |
� | +0.716 |
andr | +0.692 |
ennial | +0.673 |
sand | +0.672 |
Neg logit contributions | |
ascript | -0.797 |
lect | -0.684 |
spir | -0.670 |
NK | -0.663 |
Roll | -0.648 |
Token | � |
Token ID | 447 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
姫 | +1.854 |
� | +1.796 |
andr | +1.738 |
sand | +1.697 |
ennial | +1.694 |
Neg logit contributions | |
ascript | -1.984 |
lect | -1.702 |
spir | -1.674 |
NK | -1.636 |
Roll | -1.608 |
Token | � |
Token ID | 247 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
ascript | +0.270 |
lect | +0.228 |
spir | +0.221 |
Roll | +0.214 |
NK | +0.204 |
Neg logit contributions | |
姫 | -0.427 |
andr | -0.404 |
andestine | -0.400 |
oling | -0.398 |
extinction | -0.396 |
Token | ve |
Token ID | 303 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ascript | +0.282 |
lect | +0.248 |
spir | +0.240 |
NK | +0.232 |
Roll | +0.229 |
Neg logit contributions | |
姫 | -0.267 |
� | -0.251 |
andr | -0.246 |
ennial | -0.242 |
oling | -0.241 |
Token | created |
Token ID | 2727 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
ascript | +0.527 |
lect | +0.461 |
spir | +0.458 |
NK | +0.441 |
Roll | +0.434 |
Neg logit contributions | |
姫 | -0.543 |
� | -0.510 |
andr | -0.495 |
sand | -0.489 |
andestine | -0.487 |
Token | a |
Token ID | 257 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
ascript | +1.687 |
lect | +1.400 |
spir | +1.373 |
NK | +1.366 |
Roll | +1.335 |
Neg logit contributions | |
姫 | -2.077 |
� | -1.977 |
andr | -1.925 |
oling | -1.898 |
andestine | -1.880 |
Token | special |
Token ID | 2041 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
ascript | +1.313 |
lect | +1.150 |
spir | +1.120 |
NK | +1.100 |
Roll | +1.088 |
Neg logit contributions | |
姫 | -1.210 |
� | -1.151 |
andr | -1.112 |
andestine | -1.102 |
ennial | -1.070 |
Token | commemor |
Token ID | 28723 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
ascript | +1.397 |
lect | +1.218 |
NK | +1.207 |
spir | +1.193 |
Roll | +1.167 |
Neg logit contributions | |
姫 | -1.223 |
� | -1.165 |
andr | -1.115 |
andestine | -1.095 |
ennial | -1.084 |
Token | ative |
Token ID | 876 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
姫 | +1.636 |
� | +1.577 |
andr | +1.503 |
ennial | +1.446 |
sand | +1.442 |
Neg logit contributions | |
ascript | -3.363 |
lect | -2.985 |
spir | -2.950 |
NK | -2.911 |
Roll | -2.858 |
Token | badge |
Token ID | 23009 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
姫 | +0.413 |
� | +0.394 |
andr | +0.380 |
ennial | +0.369 |
andestine | +0.368 |
Neg logit contributions | |
ascript | -0.464 |
lect | -0.412 |
NK | -0.401 |
spir | -0.400 |
Roll | -0.391 |
Token | for |
Token ID | 329 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
姫 | +2.214 |
� | +2.149 |
andr | +2.078 |
ennial | +2.023 |
sand | +2.022 |
Neg logit contributions | |
ascript | -2.533 |
lect | -2.179 |
spir | -2.143 |
NK | -2.104 |
Roll | -2.055 |
Token | is |
Token ID | 318 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
姫 | +0.030 |
� | +0.028 |
andr | +0.027 |
ennial | +0.027 |
andestine | +0.026 |
Neg logit contributions | |
ascript | -0.037 |
lect | -0.034 |
NK | -0.034 |
spir | -0.033 |
Roll | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
姫 | +0.232 |
� | +0.219 |
andr | +0.217 |
ennial | +0.213 |
sand | +0.211 |
Neg logit contributions | |
ascript | -0.233 |
lect | -0.202 |
spir | -0.199 |
NK | -0.198 |
Roll | -0.190 |
Token | case |
Token ID | 1339 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
姫 | +0.235 |
� | +0.216 |
andr | +0.210 |
ennial | +0.204 |
sand | +0.204 |
Neg logit contributions | |
ascript | -0.269 |
lect | -0.241 |
NK | -0.238 |
spir | -0.234 |
Roll | -0.223 |
Token | irrespective |
Token ID | 40611 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
ascript | +1.837 |
lect | +1.576 |
spir | +1.548 |
NK | +1.531 |
forth | +1.456 |
Neg logit contributions | |
姫 | -2.013 |
� | -1.940 |
andr | -1.897 |
sand | -1.838 |
andestine | -1.835 |
Token | of |
Token ID | 286 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
ascript | +1.245 |
lect | +1.104 |
NK | +1.075 |
spir | +1.064 |
Roll | +1.014 |
Neg logit contributions | |
姫 | -1.588 |
� | -1.512 |
andr | -1.432 |
andestine | -1.430 |
oling | -1.412 |
Token | whether |
Token ID | 1771 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
ascript | +3.773 |
NK | +3.324 |
lect | +3.294 |
spir | +3.146 |
FTA | +3.124 |
Neg logit contributions | |
姫 | -4.462 |
andr | -4.149 |
� | -4.112 |
jan | -4.036 |
sand | -3.989 |
Token | the |
Token ID | 262 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
ascript | +4.240 |
NK | +3.705 |
lect | +3.699 |
spir | +3.646 |
FTA | +3.452 |
Neg logit contributions | |
姫 | -5.118 |
andr | -4.893 |
� | -4.858 |
oling | -4.706 |
sand | -4.681 |
Token | woman |
Token ID | 2415 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
ascript | +1.321 |
lect | +1.169 |
NK | +1.168 |
spir | +1.161 |
FTA | +1.112 |
Neg logit contributions | |
姫 | -1.003 |
� | -0.928 |
andr | -0.917 |
encia | -0.884 |
oling | -0.884 |
Token | is |
Token ID | 318 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
ascript | +0.667 |
lect | +0.585 |
spir | +0.575 |
NK | +0.569 |
glers | +0.541 |
Neg logit contributions | |
姫 | -0.673 |
� | -0.645 |
andr | -0.634 |
sand | -0.609 |
oling | -0.608 |
Token | younger |
Token ID | 7099 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ascript | +1.459 |
lect | +1.281 |
NK | +1.267 |
spir | +1.253 |
FTA | +1.180 |
Neg logit contributions | |
姫 | -1.492 |
� | -1.389 |
andr | -1.378 |
andestine | -1.341 |
sand | -1.337 |
Token | or |
Token ID | 393 |
Feature activation | -7.35e-02 |
Pos logit contributions | |
姫 | +0.529 |
� | +0.509 |
andr | +0.497 |
sand | +0.479 |
ennial | +0.478 |
Neg logit contributions | |
ascript | -0.619 |
lect | -0.538 |
spir | -0.527 |
NK | -0.519 |
Roll | -0.503 |
Token | . |
Token ID | 13 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
姫 | +0.338 |
� | +0.324 |
andr | +0.315 |
ennial | +0.307 |
sand | +0.306 |
Neg logit contributions | |
ascript | -0.359 |
lect | -0.311 |
spir | -0.303 |
NK | -0.298 |
Roll | -0.292 |
Token |
|
Token ID | 198 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
ascript | +0.139 |
lect | +0.122 |
spir | +0.120 |
NK | +0.119 |
Roll | +0.116 |
Neg logit contributions | |
姫 | -0.113 |
� | -0.109 |
andr | -0.105 |
oling | -0.103 |
ennial | -0.103 |
Token |
|
Token ID | 198 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
姫 | +0.091 |
� | +0.088 |
andr | +0.084 |
andestine | +0.083 |
oling | +0.082 |
Neg logit contributions | |
ascript | -0.098 |
lect | -0.091 |
spir | -0.089 |
NK | -0.088 |
Roll | -0.087 |
Token | More |
Token ID | 5167 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
姫 | +0.247 |
� | +0.238 |
andr | +0.229 |
ennial | +0.225 |
oling | +0.224 |
Neg logit contributions | |
ascript | -0.290 |
lect | -0.258 |
spir | -0.254 |
Roll | -0.246 |
NK | -0.243 |
Token | than |
Token ID | 621 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
ascript | +1.238 |
lect | +1.087 |
spir | +1.050 |
NK | +1.038 |
Roll | +1.002 |
Neg logit contributions | |
姫 | -1.153 |
� | -1.099 |
andr | -1.066 |
ennial | -1.055 |
oling | -1.033 |
Token | 1 |
Token ID | 352 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
ascript | +3.276 |
NK | +2.852 |
lect | +2.828 |
spir | +2.755 |
FTA | +2.690 |
Neg logit contributions | |
姫 | -3.639 |
� | -3.368 |
andr | -3.305 |
ennial | -3.288 |
encia | -3.212 |
Token | , |
Token ID | 11 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
ascript | +4.584 |
lect | +3.999 |
spir | +3.957 |
NK | +3.876 |
Roll | +3.696 |
Neg logit contributions | |
姫 | -4.337 |
andr | -4.074 |
� | -4.063 |
ennial | -3.948 |
oling | -3.899 |
Token | 000 |
Token ID | 830 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
ascript | +1.923 |
lect | +1.600 |
spir | +1.571 |
NK | +1.515 |
Roll | +1.480 |
Neg logit contributions | |
姫 | -2.729 |
� | -2.665 |
andr | -2.604 |
ennial | -2.553 |
andestine | -2.523 |
Token | flag |
Token ID | 6056 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
ascript | +1.750 |
lect | +1.544 |
spir | +1.493 |
NK | +1.479 |
Roll | +1.440 |
Neg logit contributions | |
姫 | -1.483 |
� | -1.405 |
andr | -1.393 |
ennial | -1.365 |
oling | -1.334 |
Token | - |
Token ID | 12 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
姫 | +1.082 |
� | +1.040 |
andr | +1.007 |
sand | +0.963 |
ennial | +0.960 |
Neg logit contributions | |
ascript | -1.724 |
lect | -1.528 |
spir | -1.499 |
NK | -1.467 |
Roll | -1.455 |
Token | w |
Token ID | 86 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
姫 | +0.632 |
� | +0.608 |
andr | +0.585 |
andestine | +0.566 |
sand | +0.563 |
Neg logit contributions | |
ascript | -0.731 |
lect | -0.643 |
spir | -0.632 |
Roll | -0.611 |
NK | -0.601 |
Token | jumping |
Token ID | 14284 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
ascript | +1.336 |
lect | +1.146 |
NK | +1.131 |
spir | +1.124 |
Roll | +1.091 |
Neg logit contributions | |
姫 | -1.698 |
� | -1.609 |
andr | -1.592 |
ennial | -1.548 |
sand | -1.545 |
Token | out |
Token ID | 503 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
ascript | +0.107 |
lect | +0.093 |
NK | +0.091 |
spir | +0.090 |
Roll | +0.087 |
Neg logit contributions | |
姫 | -0.107 |
� | -0.103 |
andr | -0.101 |
ennial | -0.098 |
oling | -0.097 |
Token | on |
Token ID | 319 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
ascript | +0.338 |
lect | +0.292 |
spir | +0.284 |
NK | +0.282 |
Roll | +0.275 |
Neg logit contributions | |
姫 | -0.334 |
� | -0.322 |
andr | -0.313 |
andestine | -0.305 |
ennial | -0.305 |
Token | top |
Token ID | 1353 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
ascript | +0.458 |
lect | +0.401 |
NK | +0.390 |
spir | +0.389 |
Roll | +0.378 |
Neg logit contributions | |
姫 | -0.452 |
� | -0.428 |
andr | -0.418 |
ennial | -0.409 |
sand | -0.404 |
Token | of |
Token ID | 286 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
ascript | +0.524 |
lect | +0.452 |
spir | +0.443 |
NK | +0.437 |
Roll | +0.428 |
Neg logit contributions | |
姫 | -0.460 |
� | -0.447 |
andr | -0.432 |
sand | -0.420 |
ennial | -0.420 |
Token | Ann |
Token ID | 5506 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
ascript | +0.729 |
lect | +0.631 |
spir | +0.617 |
NK | +0.614 |
Roll | +0.600 |
Neg logit contributions | |
姫 | -0.736 |
� | -0.708 |
andr | -0.690 |
ennial | -0.676 |
sand | -0.671 |
Token | al |
Token ID | 282 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
ascript | +2.558 |
lect | +2.328 |
spir | +2.267 |
NK | +2.070 |
til | +2.059 |
Neg logit contributions | |
姫 | -5.404 |
� | -5.169 |
KEN | -4.880 |
andestine | -4.871 |
andr | -4.862 |
Token | ise |
Token ID | 786 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
ascript | +0.821 |
lect | +0.743 |
spir | +0.707 |
NK | +0.682 |
Roll | +0.672 |
Neg logit contributions | |
姫 | -0.790 |
� | -0.758 |
andr | -0.725 |
andestine | -0.706 |
oling | -0.704 |
Token | . |
Token ID | 13 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
ascript | +0.951 |
lect | +0.829 |
spir | +0.811 |
NK | +0.790 |
Roll | +0.776 |
Neg logit contributions | |
姫 | -1.001 |
� | -0.968 |
andr | -0.939 |
ennial | -0.918 |
andestine | -0.916 |
Token | There |
Token ID | 1318 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
姫 | +0.171 |
� | +0.163 |
andr | +0.160 |
oling | +0.156 |
ennial | +0.156 |
Neg logit contributions | |
ascript | -0.166 |
lect | -0.146 |
NK | -0.146 |
spir | -0.146 |
Roll | -0.142 |
Token | � |
Token ID | 447 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
姫 | +0.278 |
� | +0.267 |
andr | +0.259 |
ennial | +0.254 |
sand | +0.254 |
Neg logit contributions | |
ascript | -0.283 |
lect | -0.244 |
spir | -0.240 |
NK | -0.233 |
Roll | -0.230 |
Token | as |
Token ID | 355 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
姫 | +0.913 |
� | +0.882 |
andr | +0.854 |
sand | +0.835 |
ennial | +0.832 |
Neg logit contributions | |
ascript | -0.980 |
lect | -0.850 |
spir | -0.828 |
NK | -0.823 |
Roll | -0.796 |
Token | local |
Token ID | 1957 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
ascript | +0.742 |
lect | +0.648 |
NK | +0.641 |
spir | +0.629 |
Roll | +0.605 |
Neg logit contributions | |
姫 | -0.753 |
� | -0.719 |
andr | -0.699 |
ennial | -0.685 |
sand | -0.681 |
Token | deputies |
Token ID | 21861 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
ascript | +1.937 |
lect | +1.720 |
NK | +1.692 |
spir | +1.631 |
Roll | +1.577 |
Neg logit contributions | |
姫 | -2.196 |
� | -2.102 |
andr | -2.033 |
ennial | -2.016 |
sand | -2.001 |
Token | , |
Token ID | 11 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
ascript | +0.501 |
lect | +0.436 |
spir | +0.425 |
NK | +0.422 |
Roll | +0.406 |
Neg logit contributions | |
姫 | -0.484 |
� | -0.464 |
andr | -0.449 |
sand | -0.438 |
ennial | -0.437 |
Token | representing |
Token ID | 10200 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
ascript | +1.479 |
NK | +1.299 |
lect | +1.290 |
spir | +1.270 |
Roll | +1.213 |
Neg logit contributions | |
姫 | -1.542 |
� | -1.464 |
andr | -1.408 |
ennial | -1.380 |
andestine | -1.378 |
Token | numerous |
Token ID | 6409 |
Feature activation | -8.38e-01 |
Pos logit contributions | |
ascript | +0.916 |
NK | +0.813 |
lect | +0.812 |
spir | +0.788 |
Roll | +0.754 |
Neg logit contributions | |
姫 | -0.853 |
� | -0.805 |
andr | -0.778 |
ennial | -0.760 |
sand | -0.758 |
Token | parts |
Token ID | 3354 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
ascript | +4.389 |
NK | +4.119 |
lect | +4.037 |
spir | +3.841 |
FTA | +3.710 |
Neg logit contributions | |
姫 | -3.576 |
� | -3.332 |
andr | -3.256 |
ennial | -3.163 |
oling | -3.138 |
Token | of |
Token ID | 286 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
姫 | +0.911 |
� | +0.874 |
andr | +0.836 |
ennial | +0.815 |
oling | +0.815 |
Neg logit contributions | |
ascript | -1.336 |
spir | -1.157 |
lect | -1.153 |
glers | -1.119 |
Roll | -1.114 |
Token | Korean |
Token ID | 6983 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
ascript | +1.447 |
NK | +1.290 |
lect | +1.271 |
spir | +1.214 |
Roll | +1.166 |
Neg logit contributions | |
姫 | -1.641 |
� | -1.568 |
andr | -1.518 |
sand | -1.490 |
ennial | -1.490 |
Token | society |
Token ID | 3592 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
ascript | +1.342 |
lect | +1.201 |
NK | +1.201 |
spir | +1.136 |
Roll | +1.109 |
Neg logit contributions | |
姫 | -1.302 |
� | -1.233 |
andr | -1.213 |
sand | -1.186 |
ennial | -1.182 |
Token | . |
Token ID | 13 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
ascript | +1.384 |
lect | +1.196 |
spir | +1.177 |
NK | +1.161 |
Roll | +1.119 |
Neg logit contributions | |
姫 | -1.366 |
� | -1.317 |
andr | -1.278 |
ennial | -1.238 |
sand | -1.235 |
Token | releases |
Token ID | 10050 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
ascript | +0.219 |
lect | +0.188 |
NK | +0.185 |
spir | +0.185 |
Roll | +0.178 |
Neg logit contributions | |
姫 | -0.225 |
� | -0.215 |
andr | -0.209 |
ennial | -0.205 |
andestine | -0.202 |
Token | , |
Token ID | 11 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
姫 | +0.548 |
� | +0.529 |
andr | +0.514 |
ennial | +0.498 |
oling | +0.496 |
Neg logit contributions | |
ascript | -0.519 |
lect | -0.442 |
NK | -0.430 |
spir | -0.430 |
Roll | -0.414 |
Token | [ |
Token ID | 685 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
ascript | +1.420 |
NK | +1.221 |
lect | +1.215 |
spir | +1.180 |
Roll | +1.154 |
Neg logit contributions | |
姫 | -1.486 |
� | -1.391 |
andr | -1.346 |
ennial | -1.333 |
andestine | -1.317 |
Token | an |
Token ID | 272 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
ascript | +1.032 |
lect | +0.918 |
spir | +0.890 |
Roll | +0.868 |
NK | +0.853 |
Neg logit contributions | |
姫 | -0.998 |
� | -0.949 |
andr | -0.912 |
andestine | -0.904 |
ennial | -0.903 |
Token | ] |
Token ID | 60 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
ascript | +0.788 |
lect | +0.682 |
spir | +0.667 |
NK | +0.666 |
Roll | +0.646 |
Neg logit contributions | |
姫 | -0.616 |
� | -0.590 |
andr | -0.570 |
sand | -0.553 |
ennial | -0.553 |
Token | estimate |
Token ID | 8636 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
ascript | +2.582 |
NK | +2.295 |
lect | +2.253 |
FTA | +2.176 |
spir | +2.168 |
Neg logit contributions | |
姫 | -2.342 |
� | -2.180 |
andr | -2.096 |
andestine | -2.028 |
ennial | -2.021 |
Token | ( |
Token ID | 357 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
ascript | +3.759 |
NK | +3.290 |
lect | +3.228 |
spir | +3.059 |
Roll | +3.029 |
Neg logit contributions | |
姫 | -4.857 |
� | -4.552 |
andr | -4.436 |
andestine | -4.324 |
ennial | -4.314 |
Token | within |
Token ID | 33479 |
Feature activation | +4.17e-03 |
Pos logit contributions | |
ascript | +2.552 |
lect | +2.282 |
spir | +2.234 |
Roll | +2.159 |
yet | +2.100 |
Neg logit contributions | |
姫 | -2.592 |
� | -2.466 |
andr | -2.343 |
extinction | -2.335 |
andestine | -2.328 |
Token | 4 |
Token ID | 604 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
姫 | +0.020 |
� | +0.019 |
andr | +0.018 |
ennial | +0.018 |
sand | +0.018 |
Neg logit contributions | |
ascript | -0.022 |
lect | -0.019 |
spir | -0.018 |
NK | -0.018 |
Roll | -0.017 |
Token | %) |
Token ID | 4407 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
ascript | +1.134 |
lect | +0.966 |
spir | +0.955 |
NK | +0.953 |
Roll | +0.917 |
Neg logit contributions | |
姫 | -0.997 |
� | -0.967 |
andr | -0.942 |
ennial | -0.908 |
oling | -0.906 |
Token | of |
Token ID | 286 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
ascript | +0.850 |
lect | +0.732 |
NK | +0.715 |
spir | +0.708 |
Roll | +0.685 |
Neg logit contributions | |
姫 | -0.851 |
� | -0.814 |
andr | -0.796 |
ennial | -0.770 |
andestine | -0.763 |