Token | / |
Token ID | 14 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Sov | +0.546 |
� | +0.512 |
Zhu | +0.507 |
vacc | +0.499 |
Qin | +0.485 |
Neg logit contributions | |
commute | -0.396 |
commuting | -0.324 |
notation | -0.313 |
flakes | -0.297 |
aeros | -0.297 |
Token | 3 |
Token ID | 18 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
Sov | +0.849 |
� | +0.778 |
Zhu | +0.765 |
vacc | +0.753 |
Qin | +0.721 |
Neg logit contributions | |
commute | -1.117 |
commuting | -0.968 |
notation | -0.945 |
aeros | -0.910 |
flakes | -0.907 |
Token | minutes |
Token ID | 2431 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
Sov | +0.370 |
� | +0.343 |
Zhu | +0.340 |
vacc | +0.334 |
Qin | +0.323 |
Neg logit contributions | |
commute | -0.333 |
commuting | -0.279 |
notation | -0.270 |
flakes | -0.260 |
aeros | -0.259 |
Token | until |
Token ID | 1566 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Sov | +0.315 |
� | +0.294 |
Zhu | +0.291 |
vacc | +0.284 |
Qin | +0.275 |
Neg logit contributions | |
commute | -0.348 |
commuting | -0.297 |
notation | -0.291 |
flakes | -0.279 |
aeros | -0.279 |
Token | golden |
Token ID | 10861 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
Sov | +2.404 |
Zhu | +2.276 |
� | +2.269 |
vacc | +2.199 |
Qin | +2.189 |
Neg logit contributions | |
commute | -2.237 |
commuting | -1.877 |
notation | -1.813 |
minster | -1.697 |
aeros | -1.691 |
Token | brown |
Token ID | 7586 |
Feature activation | +5.47e+00 |
Pos logit contributions | |
Sov | +3.632 |
� | +3.571 |
Zhu | +3.492 |
vacc | +3.412 |
Qin | +3.315 |
Neg logit contributions | |
commute | -3.032 |
commuting | -2.508 |
notation | -2.379 |
aeros | -2.269 |
minster | -2.195 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
Sov | +3.866 |
� | +3.784 |
Zhu | +3.677 |
vacc | +3.577 |
Qin | +3.510 |
Neg logit contributions | |
commute | -4.392 |
commuting | -3.754 |
notation | -3.533 |
aeros | -3.457 |
minster | -3.406 |
Token | These |
Token ID | 2312 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
commute | +0.194 |
commuting | +0.164 |
notation | +0.158 |
aeros | +0.152 |
flakes | +0.152 |
Neg logit contributions | |
Sov | -0.199 |
� | -0.186 |
Zhu | -0.185 |
vacc | -0.181 |
Qin | -0.175 |
Token | are |
Token ID | 389 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
commute | +0.518 |
commuting | +0.431 |
notation | +0.418 |
flakes | +0.404 |
aeros | +0.401 |
Neg logit contributions | |
Sov | -0.606 |
� | -0.561 |
Zhu | -0.558 |
vacc | -0.551 |
Qin | -0.529 |
Token | lightly |
Token ID | 15376 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
Sov | +0.209 |
� | +0.193 |
Zhu | +0.192 |
vacc | +0.189 |
Qin | +0.182 |
Neg logit contributions | |
commute | -0.204 |
commuting | -0.172 |
notation | -0.167 |
flakes | -0.161 |
aeros | -0.161 |
Token | fried |
Token ID | 23018 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Sov | +1.505 |
� | +1.425 |
Zhu | +1.420 |
vacc | +1.384 |
Qin | +1.345 |
Neg logit contributions | |
commute | -1.450 |
commuting | -1.222 |
notation | -1.187 |
minster | -1.123 |
aeros | -1.109 |
Token | Natural |
Token ID | 12068 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
commute | +0.141 |
commuting | +0.120 |
notation | +0.117 |
flakes | +0.112 |
aeros | +0.112 |
Neg logit contributions | |
Sov | -0.139 |
� | -0.128 |
Zhu | -0.127 |
vacc | -0.125 |
Qin | -0.121 |
Token | World |
Token ID | 2159 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
Sov | +0.880 |
� | +0.804 |
Zhu | +0.799 |
vacc | +0.779 |
Qin | +0.744 |
Neg logit contributions | |
commute | -1.191 |
commuting | -1.031 |
notation | -1.005 |
flakes | -0.973 |
aeros | -0.967 |
Token | episode |
Token ID | 4471 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
Sov | +1.794 |
� | +1.663 |
Zhu | +1.644 |
vacc | +1.605 |
Qin | +1.547 |
Neg logit contributions | |
commute | -1.953 |
commuting | -1.660 |
notation | -1.630 |
flakes | -1.561 |
aeros | -1.554 |
Token | " |
Token ID | 366 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
Sov | +0.482 |
� | +0.444 |
Zhu | +0.442 |
vacc | +0.434 |
Qin | +0.418 |
Neg logit contributions | |
commute | -0.504 |
commuting | -0.428 |
notation | -0.417 |
flakes | -0.401 |
aeros | -0.400 |
Token | Snow |
Token ID | 28974 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
Sov | +0.216 |
� | +0.196 |
Zhu | +0.194 |
vacc | +0.191 |
Qin | +0.182 |
Neg logit contributions | |
commute | -0.285 |
commuting | -0.247 |
notation | -0.241 |
flakes | -0.233 |
aeros | -0.232 |
Token | Leopard |
Token ID | 49072 |
Feature activation | +5.38e+00 |
Pos logit contributions | |
Sov | +2.012 |
� | +1.858 |
Zhu | +1.836 |
vacc | +1.788 |
Qin | +1.716 |
Neg logit contributions | |
commute | -2.544 |
commuting | -2.179 |
notation | -2.149 |
aeros | -2.061 |
flakes | -2.033 |
Token | – |
Token ID | 784 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Sov | +3.664 |
� | +3.446 |
Zhu | +3.426 |
vacc | +3.294 |
Qin | +3.202 |
Neg logit contributions | |
commute | -4.591 |
commuting | -3.917 |
notation | -3.869 |
aeros | -3.705 |
flakes | -3.640 |
Token | Beyond |
Token ID | 12197 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Sov | +0.010 |
� | +0.009 |
Zhu | +0.009 |
vacc | +0.009 |
Qin | +0.009 |
Neg logit contributions | |
commute | -0.011 |
commuting | -0.009 |
notation | -0.009 |
flakes | -0.009 |
aeros | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -9.06e-02 |
Pos logit contributions | |
Sov | +0.920 |
� | +0.844 |
Zhu | +0.842 |
vacc | +0.824 |
Qin | +0.792 |
Neg logit contributions | |
commute | -1.051 |
commuting | -0.896 |
notation | -0.873 |
flakes | -0.844 |
aeros | -0.841 |
Token | Myth |
Token ID | 18900 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
commute | +0.072 |
commuting | +0.061 |
notation | +0.060 |
flakes | +0.058 |
aeros | +0.057 |
Neg logit contributions | |
Sov | -0.070 |
� | -0.064 |
Zhu | -0.064 |
vacc | -0.063 |
Qin | -0.060 |
Token | ".[ |
Token ID | 42924 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Sov | +1.755 |
� | +1.607 |
Zhu | +1.579 |
vacc | +1.545 |
Qin | +1.469 |
Neg logit contributions | |
commute | -2.672 |
commuting | -2.336 |
notation | -2.273 |
aeros | -2.204 |
flakes | -2.200 |
Token | / |
Token ID | 14 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Sov | +0.010 |
Zhu | +0.009 |
� | +0.009 |
vacc | +0.009 |
�� | +0.009 |
Neg logit contributions | |
commute | -0.018 |
commuting | -0.015 |
notation | -0.015 |
flakes | -0.015 |
aeros | -0.015 |
Token | 3 |
Token ID | 18 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Sov | +0.809 |
� | +0.738 |
Zhu | +0.732 |
vacc | +0.720 |
Qin | +0.688 |
Neg logit contributions | |
commute | -0.986 |
commuting | -0.848 |
notation | -0.827 |
flakes | -0.800 |
aeros | -0.798 |
Token | minutes |
Token ID | 2431 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
Sov | +0.135 |
� | +0.125 |
Zhu | +0.125 |
vacc | +0.124 |
Qin | +0.119 |
Neg logit contributions | |
commute | -0.086 |
commuting | -0.069 |
notation | -0.067 |
flakes | -0.063 |
aeros | -0.063 |
Token | until |
Token ID | 1566 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
Sov | +0.226 |
� | +0.201 |
Zhu | +0.198 |
vacc | +0.196 |
�� | +0.188 |
Neg logit contributions | |
commute | -0.296 |
commuting | -0.255 |
notation | -0.247 |
flakes | -0.241 |
aeros | -0.239 |
Token | golden |
Token ID | 10861 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
Sov | +2.693 |
� | +2.604 |
Zhu | +2.568 |
vacc | +2.524 |
Qin | +2.467 |
Neg logit contributions | |
commute | -1.701 |
commuting | -1.350 |
notation | -1.290 |
aeros | -1.211 |
flakes | -1.169 |
Token | brown |
Token ID | 7586 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
Sov | +3.736 |
� | +3.589 |
Zhu | +3.559 |
vacc | +3.452 |
Qin | +3.409 |
Neg logit contributions | |
commute | -2.732 |
commuting | -2.217 |
notation | -2.132 |
aeros | -2.032 |
flakes | -1.997 |
Token | . |
Token ID | 13 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
Sov | +3.274 |
� | +3.091 |
Zhu | +3.068 |
vacc | +2.907 |
Qin | +2.858 |
Neg logit contributions | |
commute | -4.907 |
commuting | -4.277 |
notation | -4.129 |
aeros | -4.010 |
flakes | -3.932 |
Token | These |
Token ID | 2312 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
Sov | +0.024 |
� | +0.022 |
Zhu | +0.022 |
vacc | +0.022 |
Qin | +0.021 |
Neg logit contributions | |
commute | -0.023 |
commuting | -0.019 |
notation | -0.019 |
flakes | -0.018 |
aeros | -0.018 |
Token | are |
Token ID | 389 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
commute | +0.270 |
commuting | +0.219 |
notation | +0.211 |
flakes | +0.202 |
aeros | +0.201 |
Neg logit contributions | |
Sov | -0.389 |
� | -0.363 |
Zhu | -0.361 |
vacc | -0.357 |
Qin | -0.345 |
Token | lightly |
Token ID | 15376 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
Sov | +0.435 |
� | +0.401 |
Zhu | +0.399 |
vacc | +0.392 |
Qin | +0.377 |
Neg logit contributions | |
commute | -0.435 |
commuting | -0.367 |
notation | -0.357 |
flakes | -0.344 |
aeros | -0.343 |
Token | fried |
Token ID | 23018 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
Sov | +1.361 |
� | +1.268 |
Zhu | +1.261 |
vacc | +1.219 |
Qin | +1.195 |
Neg logit contributions | |
commute | -1.704 |
commuting | -1.457 |
notation | -1.428 |
aeros | -1.372 |
flakes | -1.359 |
Token | energy |
Token ID | 2568 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
commute | +0.026 |
commuting | +0.021 |
notation | +0.021 |
flakes | +0.020 |
aeros | +0.020 |
Neg logit contributions | |
Sov | -0.034 |
� | -0.031 |
Zhu | -0.031 |
vacc | -0.031 |
Qin | -0.030 |
Token | goals |
Token ID | 4661 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
commute | +0.677 |
commuting | +0.578 |
notation | +0.562 |
flakes | +0.543 |
aeros | +0.542 |
Neg logit contributions | |
Sov | -0.602 |
� | -0.551 |
Zhu | -0.547 |
vacc | -0.540 |
Qin | -0.516 |
Token | . |
Token ID | 13 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
Sov | +0.233 |
� | +0.219 |
Zhu | +0.215 |
vacc | +0.209 |
Qin | +0.202 |
Neg logit contributions | |
commute | -0.270 |
commuting | -0.231 |
notation | -0.226 |
flakes | -0.219 |
aeros | -0.218 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
Sov | +0.410 |
� | +0.383 |
Zhu | +0.381 |
vacc | +0.370 |
Qin | +0.361 |
Neg logit contributions | |
commute | -0.366 |
commuting | -0.306 |
notation | -0.300 |
flakes | -0.290 |
aeros | -0.284 |
Token | Media |
Token ID | 13152 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
commute | +0.307 |
commuting | +0.265 |
notation | +0.259 |
flakes | +0.250 |
aeros | +0.249 |
Neg logit contributions | |
Sov | -0.241 |
� | -0.220 |
Zhu | -0.219 |
vacc | -0.214 |
Qin | -0.205 |
Token | playback |
Token ID | 16388 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
Sov | +2.255 |
� | +2.113 |
Zhu | +2.109 |
vacc | +2.089 |
Qin | +2.030 |
Neg logit contributions | |
commute | -1.320 |
commuting | -1.040 |
notation | -0.992 |
aeros | -0.937 |
flakes | -0.933 |
Token | is |
Token ID | 318 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
Sov | +3.720 |
� | +3.546 |
Zhu | +3.545 |
vacc | +3.405 |
Qin | +3.368 |
Neg logit contributions | |
commute | -3.971 |
commuting | -3.350 |
notation | -3.258 |
aeros | -3.196 |
flakes | -3.146 |
Token | unsupported |
Token ID | 24222 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
Sov | +0.174 |
� | +0.158 |
Zhu | +0.157 |
vacc | +0.154 |
Qin | +0.147 |
Neg logit contributions | |
commute | -0.228 |
commuting | -0.197 |
notation | -0.192 |
flakes | -0.187 |
aeros | -0.186 |
Token | on |
Token ID | 319 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
Sov | +0.689 |
� | +0.634 |
Zhu | +0.631 |
vacc | +0.625 |
Qin | +0.601 |
Neg logit contributions | |
commute | -0.589 |
commuting | -0.490 |
notation | -0.477 |
flakes | -0.457 |
aeros | -0.455 |
Token | your |
Token ID | 534 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
Sov | +0.345 |
� | +0.308 |
Zhu | +0.305 |
vacc | +0.300 |
Qin | +0.286 |
Neg logit contributions | |
commute | -0.493 |
commuting | -0.428 |
notation | -0.420 |
flakes | -0.407 |
aeros | -0.406 |
Token | device |
Token ID | 3335 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
commute | +0.396 |
commuting | +0.324 |
notation | +0.316 |
aeros | +0.301 |
flakes | +0.301 |
Neg logit contributions | |
Sov | -0.520 |
� | -0.480 |
Zhu | -0.478 |
vacc | -0.473 |
Qin | -0.457 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
commute | +0.014 |
commuting | +0.011 |
notation | +0.011 |
flakes | +0.011 |
aeros | +0.011 |
Neg logit contributions | |
Sov | -0.014 |
� | -0.013 |
Zhu | -0.013 |
vacc | -0.013 |
Qin | -0.013 |
Token | said |
Token ID | 531 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
Sov | +0.797 |
� | +0.749 |
Zhu | +0.743 |
vacc | +0.725 |
Qin | +0.705 |
Neg logit contributions | |
commute | -0.780 |
commuting | -0.656 |
notation | -0.640 |
flakes | -0.617 |
aeros | -0.615 |
Token | Ross |
Token ID | 9847 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
commute | +0.124 |
commuting | +0.104 |
notation | +0.102 |
flakes | +0.098 |
aeros | +0.097 |
Neg logit contributions | |
Sov | -0.133 |
Zhu | -0.122 |
� | -0.122 |
vacc | -0.120 |
Qin | -0.116 |
Token | must |
Token ID | 1276 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
Sov | +0.868 |
� | +0.802 |
Zhu | +0.795 |
vacc | +0.783 |
Qin | +0.752 |
Neg logit contributions | |
commute | -0.863 |
commuting | -0.730 |
notation | -0.710 |
flakes | -0.685 |
aeros | -0.684 |
Token | rec |
Token ID | 664 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Sov | +0.429 |
� | +0.399 |
Zhu | +0.395 |
vacc | +0.390 |
Qin | +0.376 |
Neg logit contributions | |
commute | -0.395 |
commuting | -0.337 |
notation | -0.327 |
flakes | -0.315 |
aeros | -0.314 |
Token | use |
Token ID | 1904 |
Feature activation | +5.03e+00 |
Pos logit contributions | |
Sov | +1.075 |
� | +0.984 |
Zhu | +0.973 |
vacc | +0.960 |
Qin | +0.915 |
Neg logit contributions | |
commute | -1.463 |
commuting | -1.271 |
notation | -1.239 |
flakes | -1.205 |
aeros | -1.198 |
Token | himself |
Token ID | 2241 |
Feature activation | -5.74e-01 |
Pos logit contributions | |
Sov | +3.925 |
Zhu | +3.736 |
� | +3.683 |
Qin | +3.550 |
vacc | +3.547 |
Neg logit contributions | |
commute | -3.713 |
commuting | -3.086 |
notation | -3.016 |
aeros | -2.944 |
flakes | -2.920 |
Token | from |
Token ID | 422 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
commute | +0.441 |
commuting | +0.371 |
notation | +0.362 |
flakes | +0.348 |
aeros | +0.348 |
Neg logit contributions | |
Sov | -0.454 |
� | -0.423 |
Zhu | -0.419 |
vacc | -0.411 |
Qin | -0.397 |
Token | all |
Token ID | 477 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
commute | +1.064 |
commuting | +0.908 |
notation | +0.882 |
flakes | +0.852 |
aeros | +0.849 |
Neg logit contributions | |
Sov | -0.986 |
� | -0.904 |
Zhu | -0.896 |
vacc | -0.884 |
Qin | -0.846 |
Token | Russia |
Token ID | 3284 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
Sov | +0.178 |
� | +0.165 |
Zhu | +0.165 |
vacc | +0.161 |
Qin | +0.156 |
Neg logit contributions | |
commute | -0.151 |
commuting | -0.125 |
notation | -0.122 |
flakes | -0.117 |
aeros | -0.117 |
Token | - |
Token ID | 12 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
Sov | +0.391 |
� | +0.362 |
Zhu | +0.358 |
vacc | +0.352 |
Qin | +0.339 |
Neg logit contributions | |
commute | -0.383 |
commuting | -0.322 |
notation | -0.314 |
flakes | -0.302 |
aeros | -0.301 |
Token | to |
Token ID | 284 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
commute | +0.265 |
commuting | +0.221 |
notation | +0.215 |
flakes | +0.208 |
aeros | +0.206 |
Neg logit contributions | |
Sov | -0.299 |
� | -0.281 |
Zhu | -0.277 |
vacc | -0.272 |
Qin | -0.263 |
Token | every |
Token ID | 790 |
Feature activation | -8.57e-02 |
Pos logit contributions | |
Sov | +0.050 |
� | +0.046 |
Zhu | +0.046 |
vacc | +0.045 |
Qin | +0.043 |
Neg logit contributions | |
commute | -0.048 |
commuting | -0.041 |
notation | -0.040 |
flakes | -0.039 |
aeros | -0.038 |
Token | citizen |
Token ID | 9511 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
commute | +0.064 |
commuting | +0.054 |
notation | +0.052 |
flakes | +0.050 |
aeros | +0.050 |
Neg logit contributions | |
Sov | -0.070 |
� | -0.065 |
Zhu | -0.064 |
vacc | -0.063 |
Qin | -0.061 |
Token | over |
Token ID | 625 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
Sov | +1.156 |
� | +1.090 |
Zhu | +1.075 |
vacc | +1.053 |
Qin | +1.022 |
Neg logit contributions | |
commute | -0.958 |
commuting | -0.782 |
notation | -0.776 |
flakes | -0.745 |
aeros | -0.744 |
Token | the |
Token ID | 262 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
commute | +0.338 |
commuting | +0.288 |
notation | +0.283 |
flakes | +0.275 |
aeros | +0.273 |
Neg logit contributions | |
Sov | -0.296 |
� | -0.272 |
Zhu | -0.270 |
vacc | -0.266 |
Qin | -0.255 |
Token | age |
Token ID | 2479 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
Sov | +1.085 |
� | +1.020 |
Zhu | +1.015 |
vacc | +0.993 |
Qin | +0.969 |
Neg logit contributions | |
commute | -1.002 |
commuting | -0.843 |
notation | -0.828 |
flakes | -0.814 |
aeros | -0.800 |
Token | of |
Token ID | 286 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Sov | +3.897 |
� | +3.692 |
Zhu | +3.649 |
vacc | +3.557 |
Qin | +3.484 |
Neg logit contributions | |
commute | -3.672 |
commuting | -3.072 |
notation | -3.003 |
flakes | -2.943 |
aeros | -2.910 |
Token | 18 |
Token ID | 1248 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
Sov | +0.998 |
� | +0.918 |
Zhu | +0.912 |
vacc | +0.894 |
Qin | +0.862 |
Neg logit contributions | |
commute | -1.111 |
commuting | -0.946 |
notation | -0.924 |
flakes | -0.895 |
aeros | -0.890 |
Token | at |
Token ID | 379 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
Sov | +1.629 |
� | +1.548 |
Zhu | +1.518 |
vacc | +1.497 |
Qin | +1.445 |
Neg logit contributions | |
commute | -1.404 |
commuting | -1.149 |
notation | -1.135 |
flakes | -1.098 |
aeros | -1.097 |
Token | a |
Token ID | 257 |
Feature activation | +8.92e-01 |
Pos logit contributions | |
Sov | +0.344 |
� | +0.320 |
Zhu | +0.319 |
vacc | +0.309 |
Qin | +0.301 |
Neg logit contributions | |
commute | -0.379 |
commuting | -0.322 |
notation | -0.317 |
flakes | -0.310 |
aeros | -0.309 |
Token | federal |
Token ID | 2717 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
Sov | +0.732 |
� | +0.686 |
Zhu | +0.683 |
vacc | +0.665 |
Qin | +0.648 |
Neg logit contributions | |
commute | -0.644 |
commuting | -0.537 |
notation | -0.524 |
flakes | -0.511 |
aeros | -0.509 |
Token | � |
Token ID | 447 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
commute | +0.431 |
commuting | +0.381 |
notation | +0.375 |
flakes | +0.364 |
aeros | +0.363 |
Neg logit contributions | |
Sov | -0.211 |
Zhu | -0.186 |
� | -0.182 |
vacc | -0.179 |
Qin | -0.169 |
Token | � |
Token ID | 250 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
Sov | +0.046 |
� | +0.041 |
Zhu | +0.040 |
vacc | +0.039 |
Qin | +0.037 |
Neg logit contributions | |
commute | -0.091 |
commuting | -0.081 |
notation | -0.079 |
flakes | -0.077 |
aeros | -0.077 |
Token | At |
Token ID | 2953 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
commute | +0.264 |
commuting | +0.227 |
notation | +0.222 |
flakes | +0.214 |
aeros | +0.214 |
Neg logit contributions | |
Sov | -0.215 |
� | -0.195 |
Zhu | -0.194 |
vacc | -0.191 |
Qin | -0.182 |
Token | the |
Token ID | 262 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
commute | +0.147 |
commuting | +0.125 |
notation | +0.122 |
flakes | +0.118 |
aeros | +0.117 |
Neg logit contributions | |
Sov | -0.137 |
� | -0.125 |
Zhu | -0.124 |
vacc | -0.122 |
Qin | -0.118 |
Token | tender |
Token ID | 15403 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
Sov | +0.327 |
� | +0.303 |
Zhu | +0.300 |
vacc | +0.295 |
Qin | +0.284 |
Neg logit contributions | |
commute | -0.356 |
commuting | -0.304 |
notation | -0.296 |
flakes | -0.287 |
aeros | -0.285 |
Token | age |
Token ID | 2479 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
commute | +0.218 |
commuting | +0.180 |
notation | +0.175 |
flakes | +0.168 |
aeros | +0.167 |
Neg logit contributions | |
Sov | -0.271 |
� | -0.249 |
Zhu | -0.248 |
vacc | -0.245 |
Qin | -0.236 |
Token | of |
Token ID | 286 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
Sov | +3.736 |
� | +3.629 |
Zhu | +3.526 |
vacc | +3.483 |
Qin | +3.346 |
Neg logit contributions | |
commute | -3.732 |
commuting | -3.152 |
notation | -3.057 |
flakes | -2.989 |
aeros | -2.974 |
Token | 69 |
Token ID | 8644 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
Sov | +1.022 |
� | +0.939 |
Zhu | +0.932 |
vacc | +0.916 |
Qin | +0.879 |
Neg logit contributions | |
commute | -1.127 |
commuting | -0.960 |
notation | -0.936 |
flakes | -0.905 |
aeros | -0.901 |
Token | , |
Token ID | 11 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
Sov | +1.540 |
� | +1.453 |
Zhu | +1.432 |
vacc | +1.417 |
Qin | +1.365 |
Neg logit contributions | |
commute | -1.292 |
commuting | -1.069 |
notation | -1.039 |
flakes | -0.999 |
aeros | -0.995 |
Token | my |
Token ID | 616 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
Sov | +0.453 |
� | +0.414 |
Zhu | +0.409 |
vacc | +0.402 |
Qin | +0.384 |
Neg logit contributions | |
commute | -0.620 |
commuting | -0.537 |
notation | -0.525 |
flakes | -0.509 |
aeros | -0.507 |
Token | wife |
Token ID | 3656 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
commute | +0.905 |
commuting | +0.764 |
notation | +0.744 |
flakes | +0.715 |
aeros | +0.712 |
Neg logit contributions | |
Sov | -0.904 |
� | -0.824 |
Zhu | -0.820 |
vacc | -0.810 |
Qin | -0.776 |
Token | � |
Token ID | 247 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
commute | +0.042 |
commuting | +0.037 |
notation | +0.036 |
flakes | +0.035 |
aeros | +0.035 |
Neg logit contributions | |
Sov | -0.022 |
� | -0.019 |
Zhu | -0.019 |
vacc | -0.019 |
Qin | -0.018 |
Token | s |
Token ID | 82 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
commute | +0.387 |
commuting | +0.325 |
notation | +0.315 |
flakes | +0.304 |
aeros | +0.303 |
Neg logit contributions | |
Sov | -0.416 |
� | -0.387 |
Zhu | -0.383 |
vacc | -0.378 |
Qin | -0.364 |
Token | best |
Token ID | 1266 |
Feature activation | -9.01e-01 |
Pos logit contributions | |
commute | +0.108 |
commuting | +0.092 |
notation | +0.089 |
flakes | +0.086 |
aeros | +0.086 |
Neg logit contributions | |
Sov | -0.106 |
� | -0.098 |
Zhu | -0.097 |
vacc | -0.095 |
Qin | -0.092 |
Token | wel |
Token ID | 5029 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
commute | +0.753 |
commuting | +0.644 |
notation | +0.628 |
flakes | +0.607 |
aeros | +0.605 |
Neg logit contributions | |
Sov | -0.659 |
� | -0.601 |
Zhu | -0.597 |
vacc | -0.588 |
Qin | -0.563 |
Token | ter |
Token ID | 353 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Sov | +1.977 |
� | +1.830 |
Zhu | +1.817 |
vacc | +1.788 |
Qin | +1.728 |
Neg logit contributions | |
commute | -1.584 |
commuting | -1.314 |
notation | -1.272 |
flakes | -1.220 |
aeros | -1.215 |
Token | weight |
Token ID | 6551 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
Sov | +1.150 |
� | +1.063 |
Zhu | +1.059 |
vacc | +1.043 |
Qin | +1.005 |
Neg logit contributions | |
commute | -0.980 |
commuting | -0.815 |
notation | -0.791 |
flakes | -0.760 |
aeros | -0.756 |
Token | while |
Token ID | 981 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
Sov | +3.475 |
� | +3.310 |
Zhu | +3.256 |
vacc | +3.116 |
Qin | +3.098 |
Neg logit contributions | |
commute | -3.866 |
commuting | -3.319 |
notation | -3.195 |
flakes | -3.127 |
aeros | -3.122 |
Token | suffering |
Token ID | 7195 |
Feature activation | -7.83e-01 |
Pos logit contributions | |
Sov | +0.464 |
� | +0.433 |
Zhu | +0.431 |
vacc | +0.421 |
Qin | +0.409 |
Neg logit contributions | |
commute | -0.415 |
commuting | -0.344 |
notation | -0.337 |
flakes | -0.323 |
aeros | -0.321 |
Token | double |
Token ID | 4274 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
commute | +0.614 |
commuting | +0.520 |
notation | +0.506 |
aeros | +0.488 |
flakes | +0.487 |
Neg logit contributions | |
Sov | -0.607 |
� | -0.562 |
Zhu | -0.560 |
vacc | -0.547 |
Qin | -0.529 |
Token | vision |
Token ID | 5761 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
Sov | +0.483 |
� | +0.447 |
Zhu | +0.445 |
vacc | +0.437 |
Qin | +0.421 |
Neg logit contributions | |
commute | -0.490 |
commuting | -0.413 |
notation | -0.403 |
flakes | -0.389 |
aeros | -0.388 |
Token | and |
Token ID | 290 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
Sov | +0.189 |
� | +0.176 |
Zhu | +0.173 |
vacc | +0.170 |
Qin | +0.164 |
Neg logit contributions | |
commute | -0.200 |
commuting | -0.169 |
notation | -0.166 |
aeros | -0.160 |
flakes | -0.159 |
Token | for |
Token ID | 329 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
Sov | +0.620 |
� | +0.584 |
Zhu | +0.575 |
vacc | +0.564 |
Qin | +0.546 |
Neg logit contributions | |
commute | -0.560 |
commuting | -0.468 |
notation | -0.452 |
flakes | -0.438 |
aeros | -0.437 |
Token | the |
Token ID | 262 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
Sov | +0.385 |
� | +0.364 |
Zhu | +0.362 |
vacc | +0.356 |
Qin | +0.348 |
Neg logit contributions | |
commute | -0.208 |
commuting | -0.162 |
notation | -0.154 |
flakes | -0.147 |
aeros | -0.145 |
Token | former |
Token ID | 1966 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
Sov | +0.109 |
� | +0.100 |
Zhu | +0.099 |
vacc | +0.097 |
Qin | +0.094 |
Neg logit contributions | |
commute | -0.118 |
commuting | -0.101 |
notation | -0.098 |
flakes | -0.095 |
aeros | -0.094 |
Token | UFC |
Token ID | 11448 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
commute | +0.318 |
commuting | +0.271 |
notation | +0.263 |
flakes | +0.255 |
aeros | +0.254 |
Neg logit contributions | |
Sov | -0.293 |
� | -0.272 |
Zhu | -0.269 |
vacc | -0.263 |
Qin | -0.255 |
Token | Heavy |
Token ID | 14089 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Sov | +1.205 |
� | +1.143 |
Zhu | +1.124 |
vacc | +1.093 |
Qin | +1.069 |
Neg logit contributions | |
commute | -1.145 |
commuting | -0.961 |
notation | -0.931 |
flakes | -0.903 |
aeros | -0.901 |
Token | weight |
Token ID | 6551 |
Feature activation | +4.80e+00 |
Pos logit contributions | |
Sov | +1.180 |
� | +1.126 |
Zhu | +1.112 |
vacc | +1.072 |
Qin | +1.055 |
Neg logit contributions | |
commute | -1.214 |
commuting | -1.027 |
notation | -1.000 |
flakes | -0.966 |
aeros | -0.962 |
Token | champion |
Token ID | 8783 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Sov | +3.177 |
� | +3.040 |
Zhu | +2.981 |
Qin | +2.821 |
vacc | +2.809 |
Neg logit contributions | |
commute | -4.125 |
commuting | -3.552 |
notation | -3.439 |
flakes | -3.388 |
aeros | -3.367 |
Token | who |
Token ID | 508 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
Sov | +1.259 |
� | +1.203 |
Zhu | +1.175 |
vacc | +1.152 |
Qin | +1.117 |
Neg logit contributions | |
commute | -1.142 |
commuting | -0.949 |
notation | -0.929 |
flakes | -0.892 |
aeros | -0.887 |
Token | vowed |
Token ID | 19982 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
commute | +0.699 |
commuting | +0.587 |
notation | +0.570 |
flakes | +0.549 |
aeros | +0.546 |
Neg logit contributions | |
Sov | -0.761 |
� | -0.701 |
Zhu | -0.698 |
vacc | -0.688 |
Qin | -0.662 |
Token | to |
Token ID | 284 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
commute | +1.554 |
commuting | +1.315 |
notation | +1.279 |
flakes | +1.233 |
aeros | +1.228 |
Neg logit contributions | |
Sov | -1.551 |
� | -1.427 |
Zhu | -1.418 |
vacc | -1.397 |
Qin | -1.342 |
Token | � |
Token ID | 564 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
commute | +0.608 |
commuting | +0.507 |
notation | +0.494 |
flakes | +0.474 |
aeros | +0.473 |
Neg logit contributions | |
Sov | -0.668 |
� | -0.614 |
Zhu | -0.611 |
vacc | -0.602 |
Qin | -0.579 |
Token | on |
Token ID | 319 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
commute | +0.157 |
commuting | +0.132 |
notation | +0.127 |
flakes | +0.123 |
aeros | +0.122 |
Neg logit contributions | |
Sov | -0.177 |
� | -0.163 |
Zhu | -0.162 |
vacc | -0.160 |
Qin | -0.154 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
Sov | +0.435 |
� | +0.401 |
Zhu | +0.398 |
vacc | +0.391 |
Qin | +0.376 |
Neg logit contributions | |
commute | -0.453 |
commuting | -0.385 |
notation | -0.376 |
flakes | -0.362 |
aeros | -0.361 |
Token | . |
Token ID | 13 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
Sov | +0.205 |
� | +0.189 |
Zhu | +0.188 |
vacc | +0.184 |
Qin | +0.178 |
Neg logit contributions | |
commute | -0.202 |
commuting | -0.171 |
notation | -0.166 |
flakes | -0.160 |
aeros | -0.160 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.49e-01 |
Pos logit contributions | |
commute | +0.378 |
commuting | +0.313 |
notation | +0.304 |
flakes | +0.292 |
aeros | +0.290 |
Neg logit contributions | |
Sov | -0.458 |
� | -0.424 |
Zhu | -0.418 |
vacc | -0.416 |
Qin | -0.398 |
Token | Media |
Token ID | 13152 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
commute | +0.740 |
commuting | +0.638 |
notation | +0.624 |
flakes | +0.603 |
aeros | +0.601 |
Neg logit contributions | |
Sov | -0.583 |
Zhu | -0.531 |
� | -0.531 |
vacc | -0.518 |
Qin | -0.499 |
Token | playback |
Token ID | 16388 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
Sov | +1.938 |
� | +1.818 |
Zhu | +1.813 |
vacc | +1.796 |
Qin | +1.745 |
Neg logit contributions | |
commute | -1.122 |
commuting | -0.883 |
notation | -0.842 |
aeros | -0.797 |
flakes | -0.793 |
Token | is |
Token ID | 318 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
Sov | +3.558 |
Zhu | +3.384 |
� | +3.381 |
vacc | +3.258 |
Qin | +3.218 |
Neg logit contributions | |
commute | -3.716 |
commuting | -3.125 |
notation | -3.045 |
aeros | -2.978 |
flakes | -2.940 |
Token | unsupported |
Token ID | 24222 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
Sov | +0.158 |
� | +0.143 |
Zhu | +0.142 |
vacc | +0.139 |
Qin | +0.133 |
Neg logit contributions | |
commute | -0.212 |
commuting | -0.184 |
notation | -0.179 |
flakes | -0.174 |
aeros | -0.173 |
Token | on |
Token ID | 319 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
Sov | +0.188 |
� | +0.173 |
Zhu | +0.172 |
vacc | +0.171 |
�� | +0.165 |
Neg logit contributions | |
commute | -0.156 |
commuting | -0.129 |
notation | -0.126 |
flakes | -0.120 |
aeros | -0.120 |
Token | your |
Token ID | 534 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
Sov | +0.329 |
� | +0.293 |
Zhu | +0.290 |
vacc | +0.286 |
�� | +0.273 |
Neg logit contributions | |
commute | -0.463 |
commuting | -0.401 |
notation | -0.393 |
flakes | -0.381 |
aeros | -0.380 |
Token | device |
Token ID | 3335 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
commute | +0.263 |
commuting | +0.214 |
notation | +0.209 |
aeros | +0.199 |
flakes | +0.199 |
Neg logit contributions | |
Sov | -0.348 |
� | -0.321 |
Zhu | -0.319 |
vacc | -0.315 |
Qin | -0.305 |
Token | b |
Token ID | 65 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
Sov | +0.548 |
� | +0.506 |
Zhu | +0.495 |
vacc | +0.494 |
Qin | +0.463 |
Neg logit contributions | |
commute | -0.766 |
commuting | -0.669 |
notation | -0.644 |
aeros | -0.631 |
flakes | -0.630 |
Token | ] |
Token ID | 60 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
Sov | +0.842 |
� | +0.790 |
Zhu | +0.784 |
vacc | +0.773 |
Qin | +0.750 |
Neg logit contributions | |
commute | -0.541 |
commuting | -0.435 |
notation | -0.417 |
flakes | -0.397 |
aeros | -0.395 |
Token | Personal |
Token ID | 30228 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
commute | +0.389 |
commuting | +0.332 |
notation | +0.323 |
flakes | +0.313 |
aeros | +0.311 |
Neg logit contributions | |
Sov | -0.355 |
� | -0.326 |
Zhu | -0.323 |
vacc | -0.318 |
Qin | -0.305 |
Token | best |
Token ID | 1266 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
Sov | +0.347 |
� | +0.327 |
Zhu | +0.321 |
vacc | +0.314 |
Qin | +0.304 |
Neg logit contributions | |
commute | -0.327 |
commuting | -0.277 |
notation | -0.266 |
flakes | -0.260 |
aeros | -0.258 |
Token | : |
Token ID | 25 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
Sov | +0.296 |
� | +0.274 |
Zhu | +0.273 |
vacc | +0.270 |
Qin | +0.260 |
Neg logit contributions | |
commute | -0.232 |
commuting | -0.192 |
notation | -0.186 |
flakes | -0.178 |
aeros | -0.177 |
Token | [/ |
Token ID | 13412 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
Sov | +0.320 |
� | +0.277 |
Zhu | +0.276 |
vacc | +0.271 |
Qin | +0.251 |
Neg logit contributions | |
commute | -0.673 |
commuting | -0.596 |
notation | -0.586 |
flakes | -0.572 |
aeros | -0.570 |
Token | b |
Token ID | 65 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
Sov | +2.880 |
� | +2.684 |
vacc | +2.621 |
Zhu | +2.604 |
Qin | +2.429 |
Neg logit contributions | |
commute | -4.425 |
commuting | -3.875 |
notation | -3.717 |
aeros | -3.671 |
flakes | -3.643 |
Token | ] |
Token ID | 60 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
Sov | +0.823 |
� | +0.769 |
Zhu | +0.766 |
vacc | +0.757 |
Qin | +0.733 |
Neg logit contributions | |
commute | -0.512 |
commuting | -0.409 |
notation | -0.394 |
flakes | -0.374 |
aeros | -0.372 |
Token | 00 |
Token ID | 3571 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
Sov | +0.486 |
� | +0.449 |
Zhu | +0.446 |
vacc | +0.436 |
Qin | +0.423 |
Neg logit contributions | |
commute | -0.527 |
commuting | -0.449 |
notation | -0.436 |
flakes | -0.421 |
aeros | -0.420 |
Token | : |
Token ID | 25 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
Sov | +0.234 |
� | +0.217 |
Zhu | +0.216 |
vacc | +0.213 |
Qin | +0.206 |
Neg logit contributions | |
commute | -0.181 |
commuting | -0.149 |
notation | -0.145 |
flakes | -0.139 |
aeros | -0.138 |
Token | 47 |
Token ID | 2857 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
commute | +0.284 |
commuting | +0.246 |
notation | +0.242 |
flakes | +0.235 |
aeros | +0.234 |
Neg logit contributions | |
Sov | -0.199 |
Zhu | -0.179 |
� | -0.178 |
vacc | -0.176 |
�� | -0.171 |
Token | b |
Token ID | 65 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
Sov | +0.428 |
� | +0.399 |
Zhu | +0.395 |
vacc | +0.392 |
Qin | +0.375 |
Neg logit contributions | |
commute | -0.391 |
commuting | -0.329 |
notation | -0.316 |
flakes | -0.306 |
aeros | -0.306 |
Token | ] |
Token ID | 60 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
Sov | +0.803 |
� | +0.758 |
Zhu | +0.744 |
vacc | +0.733 |
Qin | +0.714 |
Neg logit contributions | |
commute | -0.608 |
commuting | -0.505 |
notation | -0.475 |
flakes | -0.462 |
aeros | -0.460 |
Token | Personal |
Token ID | 30228 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
commute | +0.541 |
commuting | +0.469 |
notation | +0.458 |
flakes | +0.444 |
aeros | +0.443 |
Neg logit contributions | |
Sov | -0.397 |
� | -0.360 |
Zhu | -0.357 |
vacc | -0.351 |
Qin | -0.334 |
Token | best |
Token ID | 1266 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
commute | +0.247 |
commuting | +0.210 |
notation | +0.204 |
flakes | +0.198 |
aeros | +0.197 |
Neg logit contributions | |
Sov | -0.230 |
� | -0.211 |
Zhu | -0.210 |
vacc | -0.206 |
Qin | -0.198 |
Token | : |
Token ID | 25 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
commute | +0.135 |
commuting | +0.113 |
notation | +0.109 |
flakes | +0.106 |
aeros | +0.106 |
Neg logit contributions | |
Sov | -0.151 |
� | -0.140 |
Zhu | -0.139 |
vacc | -0.137 |
Qin | -0.132 |
Token | [/ |
Token ID | 13412 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
Sov | +0.054 |
� | +0.048 |
Zhu | +0.047 |
vacc | +0.046 |
Qin | +0.043 |
Neg logit contributions | |
commute | -0.119 |
commuting | -0.106 |
notation | -0.104 |
flakes | -0.102 |
aeros | -0.101 |
Token | b |
Token ID | 65 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
Sov | +2.855 |
� | +2.683 |
vacc | +2.622 |
Zhu | +2.601 |
Qin | +2.428 |
Neg logit contributions | |
commute | -4.388 |
commuting | -3.844 |
notation | -3.676 |
aeros | -3.646 |
flakes | -3.627 |
Token | ] |
Token ID | 60 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
Sov | +0.810 |
� | +0.763 |
Zhu | +0.751 |
vacc | +0.740 |
Qin | +0.718 |
Neg logit contributions | |
commute | -0.615 |
commuting | -0.509 |
notation | -0.486 |
aeros | -0.466 |
flakes | -0.466 |
Token | 1 |
Token ID | 352 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
Sov | +0.376 |
� | +0.346 |
Zhu | +0.344 |
vacc | +0.337 |
Qin | +0.325 |
Neg logit contributions | |
commute | -0.410 |
commuting | -0.349 |
notation | -0.340 |
flakes | -0.329 |
aeros | -0.328 |
Token | : |
Token ID | 25 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
Sov | +0.732 |
� | +0.680 |
Zhu | +0.673 |
vacc | +0.661 |
Qin | +0.642 |
Neg logit contributions | |
commute | -0.636 |
commuting | -0.532 |
notation | -0.513 |
flakes | -0.494 |
aeros | -0.492 |
Token | 07 |
Token ID | 2998 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
Sov | +0.018 |
� | +0.016 |
Zhu | +0.016 |
vacc | +0.016 |
Qin | +0.015 |
Neg logit contributions | |
commute | -0.024 |
commuting | -0.021 |
notation | -0.020 |
flakes | -0.020 |
aeros | -0.020 |
Token | b |
Token ID | 65 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
commute | +0.298 |
commuting | +0.250 |
notation | +0.242 |
flakes | +0.233 |
aeros | +0.232 |
Neg logit contributions | |
Sov | -0.330 |
� | -0.305 |
Zhu | -0.303 |
vacc | -0.299 |
Qin | -0.287 |
Token | ] |
Token ID | 60 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
Sov | +0.548 |
� | +0.516 |
Zhu | +0.506 |
vacc | +0.500 |
Qin | +0.486 |
Neg logit contributions | |
commute | -0.440 |
commuting | -0.367 |
notation | -0.347 |
flakes | -0.337 |
aeros | -0.336 |
Token | Est |
Token ID | 22362 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
commute | +0.881 |
commuting | +0.764 |
notation | +0.747 |
flakes | +0.724 |
aeros | +0.722 |
Neg logit contributions | |
Sov | -0.639 |
� | -0.579 |
Zhu | -0.576 |
vacc | -0.564 |
Qin | -0.538 |
Token | imate |
Token ID | 1920 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
commute | +0.035 |
commuting | +0.030 |
notation | +0.029 |
flakes | +0.028 |
aeros | +0.028 |
Neg logit contributions | |
Sov | -0.029 |
� | -0.026 |
Zhu | -0.026 |
vacc | -0.026 |
Qin | -0.025 |
Token | : |
Token ID | 25 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
commute | +0.021 |
commuting | +0.018 |
notation | +0.018 |
flakes | +0.017 |
aeros | +0.017 |
Neg logit contributions | |
Sov | -0.015 |
� | -0.014 |
Zhu | -0.013 |
vacc | -0.013 |
Qin | -0.013 |
Token | [/ |
Token ID | 13412 |
Feature activation | +4.64e+00 |
Pos logit contributions | |
commute | +0.967 |
commuting | +0.840 |
notation | +0.821 |
flakes | +0.796 |
aeros | +0.793 |
Neg logit contributions | |
Sov | -0.688 |
� | -0.618 |
Zhu | -0.615 |
vacc | -0.605 |
Qin | -0.574 |
Token | b |
Token ID | 65 |
Feature activation | +8.48e-01 |
Pos logit contributions | |
Sov | +2.741 |
� | +2.599 |
vacc | +2.554 |
Zhu | +2.503 |
Qin | +2.334 |
Neg logit contributions | |
commute | -4.380 |
commuting | -3.854 |
notation | -3.652 |
aeros | -3.646 |
flakes | -3.632 |
Token | ] |
Token ID | 60 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
Sov | +0.610 |
� | +0.569 |
Zhu | +0.557 |
vacc | +0.548 |
Qin | +0.528 |
Neg logit contributions | |
commute | -0.711 |
commuting | -0.613 |
notation | -0.588 |
flakes | -0.573 |
aeros | -0.573 |
Token | 1 |
Token ID | 352 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
Sov | +0.349 |
� | +0.321 |
Zhu | +0.319 |
vacc | +0.313 |
Qin | +0.302 |
Neg logit contributions | |
commute | -0.366 |
commuting | -0.311 |
notation | -0.302 |
flakes | -0.292 |
aeros | -0.292 |
Token | : |
Token ID | 25 |
Feature activation | -8.43e-02 |
Pos logit contributions | |
Sov | +0.619 |
� | +0.571 |
Zhu | +0.567 |
vacc | +0.556 |
Qin | +0.536 |
Neg logit contributions | |
commute | -0.647 |
commuting | -0.550 |
notation | -0.534 |
flakes | -0.517 |
aeros | -0.515 |
Token | 20 |
Token ID | 1238 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
commute | +0.075 |
commuting | +0.065 |
notation | +0.063 |
flakes | +0.061 |
aeros | +0.061 |
Neg logit contributions | |
Sov | -0.057 |
� | -0.052 |
Zhu | -0.051 |
vacc | -0.050 |
Qin | -0.048 |
Token | varieties |
Token ID | 20328 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
Sov | +0.215 |
� | +0.201 |
Zhu | +0.201 |
vacc | +0.197 |
Qin | +0.192 |
Neg logit contributions | |
commute | -0.154 |
commuting | -0.124 |
notation | -0.121 |
flakes | -0.114 |
aeros | -0.114 |
Token | for |
Token ID | 329 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Sov | +0.322 |
� | +0.302 |
Zhu | +0.299 |
vacc | +0.292 |
Qin | +0.282 |
Neg logit contributions | |
commute | -0.344 |
commuting | -0.291 |
notation | -0.285 |
aeros | -0.274 |
flakes | -0.273 |
Token | children |
Token ID | 1751 |
Feature activation | -9.06e-01 |
Pos logit contributions | |
Sov | +0.088 |
� | +0.081 |
Zhu | +0.081 |
vacc | +0.079 |
Qin | +0.077 |
Neg logit contributions | |
commute | -0.083 |
commuting | -0.069 |
notation | -0.068 |
flakes | -0.065 |
aeros | -0.065 |
Token | of |
Token ID | 286 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
commute | +0.698 |
commuting | +0.586 |
notation | +0.573 |
flakes | +0.551 |
aeros | +0.551 |
Neg logit contributions | |
Sov | -0.715 |
� | -0.665 |
Zhu | -0.660 |
vacc | -0.649 |
Qin | -0.625 |
Token | all |
Token ID | 477 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
Sov | +0.798 |
� | +0.738 |
Zhu | +0.735 |
vacc | +0.719 |
Qin | +0.692 |
Neg logit contributions | |
commute | -0.903 |
commuting | -0.769 |
notation | -0.755 |
flakes | -0.732 |
aeros | -0.727 |
Token | age |
Token ID | 2479 |
Feature activation | +4.61e+00 |
Pos logit contributions | |
commute | +0.149 |
commuting | +0.123 |
notation | +0.120 |
flakes | +0.115 |
aeros | +0.115 |
Neg logit contributions | |
Sov | -0.183 |
� | -0.171 |
Zhu | -0.169 |
vacc | -0.167 |
Qin | -0.161 |
Token | groups |
Token ID | 2628 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
Sov | +3.293 |
� | +3.107 |
Zhu | +3.048 |
vacc | +3.008 |
Qin | +2.851 |
Neg logit contributions | |
commute | -3.710 |
commuting | -3.121 |
notation | -3.070 |
flakes | -2.999 |
aeros | -2.999 |
Token | . |
Token ID | 13 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
commute | +0.324 |
commuting | +0.276 |
notation | +0.272 |
flakes | +0.262 |
aeros | +0.262 |
Neg logit contributions | |
Sov | -0.282 |
� | -0.262 |
Zhu | -0.259 |
vacc | -0.255 |
Qin | -0.244 |
Token | Every |
Token ID | 6109 |
Feature activation | -9.70e-01 |
Pos logit contributions | |
commute | +0.498 |
commuting | +0.421 |
notation | +0.409 |
flakes | +0.395 |
aeros | +0.393 |
Neg logit contributions | |
Sov | -0.499 |
� | -0.458 |
Zhu | -0.455 |
vacc | -0.449 |
Qin | -0.431 |
Token | parent |
Token ID | 2560 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
commute | +0.711 |
commuting | +0.594 |
notation | +0.576 |
flakes | +0.554 |
aeros | +0.551 |
Neg logit contributions | |
Sov | -0.809 |
� | -0.749 |
Zhu | -0.744 |
vacc | -0.734 |
Qin | -0.707 |
Token | loves |
Token ID | 10408 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
commute | +0.877 |
commuting | +0.734 |
notation | +0.714 |
flakes | +0.686 |
aeros | +0.684 |
Neg logit contributions | |
Sov | -0.969 |
� | -0.897 |
Zhu | -0.891 |
vacc | -0.878 |
Qin | -0.846 |
Token | younger |
Token ID | 7099 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Sov | +0.835 |
� | +0.783 |
Zhu | +0.775 |
vacc | +0.766 |
Qin | +0.737 |
Neg logit contributions | |
commute | -0.751 |
commuting | -0.624 |
notation | -0.609 |
flakes | -0.586 |
aeros | -0.584 |
Token | than |
Token ID | 621 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Sov | +1.136 |
� | +1.061 |
Zhu | +1.056 |
vacc | +1.042 |
Qin | +1.010 |
Neg logit contributions | |
commute | -0.753 |
commuting | -0.608 |
notation | -0.586 |
flakes | -0.558 |
aeros | -0.555 |
Token | 15 |
Token ID | 1315 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Sov | +0.108 |
� | +0.098 |
Zhu | +0.098 |
vacc | +0.096 |
Qin | +0.092 |
Neg logit contributions | |
commute | -0.114 |
commuting | -0.097 |
notation | -0.095 |
flakes | -0.091 |
aeros | -0.091 |
Token | years |
Token ID | 812 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
Sov | +0.998 |
� | +0.932 |
Zhu | +0.925 |
vacc | +0.912 |
Qin | +0.877 |
Neg logit contributions | |
commute | -1.007 |
commuting | -0.843 |
notation | -0.827 |
flakes | -0.797 |
aeros | -0.792 |
Token | of |
Token ID | 286 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
Sov | +0.266 |
� | +0.244 |
Zhu | +0.243 |
vacc | +0.236 |
Qin | +0.226 |
Neg logit contributions | |
commute | -0.396 |
commuting | -0.343 |
notation | -0.338 |
flakes | -0.329 |
aeros | -0.328 |
Token | age |
Token ID | 2479 |
Feature activation | +4.56e+00 |
Pos logit contributions | |
Sov | +0.097 |
� | +0.089 |
Zhu | +0.088 |
vacc | +0.087 |
�� | +0.084 |
Neg logit contributions | |
commute | -0.077 |
commuting | -0.064 |
notation | -0.062 |
aeros | -0.059 |
flakes | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
Sov | +3.253 |
� | +3.105 |
Zhu | +3.064 |
vacc | +2.992 |
Qin | +2.902 |
Neg logit contributions | |
commute | -3.649 |
commuting | -3.057 |
notation | -3.018 |
flakes | -2.939 |
aeros | -2.933 |
Token | according |
Token ID | 1864 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
Sov | +0.623 |
� | +0.577 |
Zhu | +0.575 |
vacc | +0.565 |
Qin | +0.545 |
Neg logit contributions | |
commute | -0.563 |
commuting | -0.471 |
notation | -0.458 |
flakes | -0.442 |
aeros | -0.439 |
Token | to |
Token ID | 284 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
Sov | +0.419 |
� | +0.383 |
Zhu | +0.381 |
vacc | +0.378 |
�� | +0.364 |
Neg logit contributions | |
commute | -0.429 |
commuting | -0.363 |
notation | -0.355 |
flakes | -0.342 |
aeros | -0.340 |
Token | the |
Token ID | 262 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
Sov | +0.416 |
� | +0.382 |
Zhu | +0.380 |
vacc | +0.372 |
Qin | +0.358 |
Neg logit contributions | |
commute | -0.458 |
commuting | -0.391 |
notation | -0.381 |
flakes | -0.368 |
aeros | -0.366 |
Token | online |
Token ID | 2691 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
Sov | +0.028 |
� | +0.026 |
Zhu | +0.026 |
vacc | +0.025 |
Qin | +0.024 |
Neg logit contributions | |
commute | -0.027 |
commuting | -0.023 |
notation | -0.022 |
flakes | -0.021 |
aeros | -0.021 |
Token | b |
Token ID | 65 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
commute | +0.014 |
commuting | +0.012 |
notation | +0.012 |
flakes | +0.011 |
aeros | +0.011 |
Neg logit contributions | |
Sov | -0.010 |
� | -0.009 |
Zhu | -0.009 |
vacc | -0.009 |
Qin | -0.009 |
Token | ] |
Token ID | 60 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
Sov | +0.703 |
� | +0.659 |
Zhu | +0.654 |
vacc | +0.645 |
Qin | +0.626 |
Neg logit contributions | |
commute | -0.450 |
commuting | -0.362 |
notation | -0.347 |
flakes | -0.331 |
aeros | -0.329 |
Token | Est |
Token ID | 22362 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
commute | +0.568 |
commuting | +0.486 |
notation | +0.473 |
flakes | +0.458 |
aeros | +0.456 |
Neg logit contributions | |
Sov | -0.499 |
� | -0.456 |
Zhu | -0.453 |
vacc | -0.445 |
Qin | -0.427 |
Token | imate |
Token ID | 1920 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Sov | +0.593 |
� | +0.539 |
Zhu | +0.535 |
vacc | +0.526 |
Qin | +0.502 |
Neg logit contributions | |
commute | -0.736 |
commuting | -0.631 |
notation | -0.619 |
flakes | -0.596 |
aeros | -0.595 |
Token | : |
Token ID | 25 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
Sov | +0.155 |
� | +0.143 |
Zhu | +0.143 |
vacc | +0.141 |
Qin | +0.136 |
Neg logit contributions | |
commute | -0.121 |
commuting | -0.099 |
notation | -0.096 |
flakes | -0.092 |
aeros | -0.092 |
Token | [/ |
Token ID | 13412 |
Feature activation | +4.56e+00 |
Pos logit contributions | |
commute | +0.495 |
commuting | +0.439 |
notation | +0.431 |
flakes | +0.420 |
aeros | +0.419 |
Neg logit contributions | |
Sov | -0.231 |
� | -0.201 |
Zhu | -0.199 |
vacc | -0.195 |
Qin | -0.181 |
Token | b |
Token ID | 65 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
Sov | +2.788 |
� | +2.598 |
vacc | +2.539 |
Zhu | +2.515 |
Qin | +2.350 |
Neg logit contributions | |
commute | -4.285 |
commuting | -3.753 |
notation | -3.597 |
aeros | -3.553 |
flakes | -3.533 |
Token | ] |
Token ID | 60 |
Feature activation | +6.69e-01 |
Pos logit contributions | |
Sov | +0.671 |
� | +0.627 |
Zhu | +0.624 |
vacc | +0.617 |
Qin | +0.597 |
Neg logit contributions | |
commute | -0.415 |
commuting | -0.331 |
notation | -0.319 |
flakes | -0.303 |
aeros | -0.302 |
Token | 00 |
Token ID | 3571 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
Sov | +0.473 |
� | +0.435 |
Zhu | +0.433 |
vacc | +0.421 |
Qin | +0.407 |
Neg logit contributions | |
commute | -0.572 |
commuting | -0.491 |
notation | -0.478 |
flakes | -0.463 |
aeros | -0.463 |
Token | : |
Token ID | 25 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
Sov | +0.235 |
� | +0.218 |
Zhu | +0.216 |
vacc | +0.213 |
Qin | +0.206 |
Neg logit contributions | |
commute | -0.203 |
commuting | -0.170 |
notation | -0.164 |
flakes | -0.158 |
aeros | -0.157 |
Token | 55 |
Token ID | 2816 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
commute | +0.585 |
commuting | +0.504 |
notation | +0.496 |
flakes | +0.478 |
aeros | +0.477 |
Neg logit contributions | |
Sov | -0.458 |
Zhu | -0.413 |
� | -0.411 |
vacc | -0.408 |
�� | -0.396 |
Token | reality |
Token ID | 3950 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Sov | +0.389 |
� | +0.360 |
Zhu | +0.358 |
vacc | +0.351 |
Qin | +0.340 |
Neg logit contributions | |
commute | -0.395 |
commuting | -0.334 |
notation | -0.326 |
flakes | -0.314 |
aeros | -0.312 |
Token | this |
Token ID | 428 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
commute | +0.865 |
commuting | +0.722 |
notation | +0.701 |
flakes | +0.673 |
aeros | +0.670 |
Neg logit contributions | |
Sov | -0.991 |
� | -0.918 |
Zhu | -0.912 |
vacc | -0.898 |
Qin | -0.867 |
Token | is |
Token ID | 318 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
Sov | +0.453 |
� | +0.420 |
Zhu | +0.417 |
vacc | +0.410 |
Qin | +0.399 |
Neg logit contributions | |
commute | -0.402 |
commuting | -0.337 |
notation | -0.324 |
flakes | -0.313 |
aeros | -0.312 |
Token | seldom |
Token ID | 25129 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Sov | +0.436 |
� | +0.406 |
Zhu | +0.403 |
vacc | +0.397 |
Qin | +0.388 |
Neg logit contributions | |
commute | -0.304 |
commuting | -0.246 |
notation | -0.237 |
flakes | -0.227 |
aeros | -0.226 |
Token | the |
Token ID | 262 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Sov | +0.866 |
� | +0.793 |
Zhu | +0.782 |
vacc | +0.767 |
Qin | +0.739 |
Neg logit contributions | |
commute | -1.058 |
commuting | -0.910 |
notation | -0.888 |
flakes | -0.864 |
aeros | -0.857 |
Token | case |
Token ID | 1339 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
Sov | +0.971 |
� | +0.899 |
Zhu | +0.895 |
vacc | +0.875 |
Qin | +0.852 |
Neg logit contributions | |
commute | -0.925 |
commuting | -0.781 |
notation | -0.753 |
flakes | -0.736 |
aeros | -0.723 |
Token | . |
Token ID | 13 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Sov | +3.236 |
� | +3.159 |
Zhu | +3.027 |
vacc | +2.915 |
Qin | +2.863 |
Neg logit contributions | |
commute | -3.706 |
commuting | -3.162 |
notation | -3.072 |
flakes | -2.998 |
aeros | -2.985 |
Token |
|
Token ID | 198 |
Feature activation | +7.41e-01 |
Pos logit contributions | |
Sov | +0.926 |
� | +0.860 |
Zhu | +0.853 |
vacc | +0.830 |
Qin | +0.810 |
Neg logit contributions | |
commute | -0.980 |
commuting | -0.833 |
notation | -0.808 |
flakes | -0.778 |
aeros | -0.775 |
Token |
|
Token ID | 198 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
Sov | +0.656 |
� | +0.612 |
Zhu | +0.608 |
vacc | +0.599 |
Qin | +0.580 |
Neg logit contributions | |
commute | -0.505 |
commuting | -0.416 |
notation | -0.401 |
flakes | -0.384 |
aeros | -0.382 |
Token | To |
Token ID | 2514 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
Sov | +0.525 |
� | +0.478 |
Zhu | +0.471 |
vacc | +0.464 |
Qin | +0.443 |
Neg logit contributions | |
commute | -0.678 |
commuting | -0.586 |
notation | -0.570 |
flakes | -0.553 |
aeros | -0.551 |
Token | understand |
Token ID | 1833 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
Sov | +0.731 |
� | +0.673 |
Zhu | +0.668 |
vacc | +0.657 |
Qin | +0.632 |
Neg logit contributions | |
commute | -0.772 |
commuting | -0.658 |
notation | -0.640 |
flakes | -0.618 |
aeros | -0.615 |
Token | the |
Token ID | 262 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Sov | +1.521 |
� | +1.405 |
Zhu | +1.401 |
vacc | +1.353 |
Qin | +1.314 |
Neg logit contributions | |
commute | -1.756 |
commuting | -1.512 |
notation | -1.480 |
flakes | -1.444 |
aeros | -1.433 |
Token | White |
Token ID | 2635 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
Sov | +0.909 |
� | +0.852 |
Zhu | +0.845 |
vacc | +0.824 |
Qin | +0.801 |
Neg logit contributions | |
commute | -0.842 |
commuting | -0.705 |
notation | -0.686 |
flakes | -0.664 |
aeros | -0.658 |
Token | House |
Token ID | 2097 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
Sov | +0.508 |
� | +0.463 |
Zhu | +0.459 |
vacc | +0.457 |
�� | +0.437 |
Neg logit contributions | |
commute | -0.505 |
commuting | -0.428 |
notation | -0.416 |
flakes | -0.403 |
aeros | -0.400 |
Token | at |
Token ID | 379 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
commute | +0.260 |
commuting | +0.215 |
notation | +0.209 |
flakes | +0.201 |
aeros | +0.199 |
Neg logit contributions | |
Sov | -0.323 |
� | -0.301 |
Zhu | -0.300 |
vacc | -0.295 |
Qin | -0.285 |
Token | the |
Token ID | 262 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
Sov | +0.078 |
� | +0.072 |
Zhu | +0.072 |
vacc | +0.070 |
Qin | +0.068 |
Neg logit contributions | |
commute | -0.087 |
commuting | -0.074 |
notation | -0.072 |
flakes | -0.070 |
aeros | -0.070 |
Token | age |
Token ID | 2479 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
Sov | +0.577 |
� | +0.533 |
Zhu | +0.530 |
vacc | +0.516 |
Qin | +0.499 |
Neg logit contributions | |
commute | -0.652 |
commuting | -0.558 |
notation | -0.542 |
flakes | -0.527 |
aeros | -0.526 |
Token | of |
Token ID | 286 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Sov | +3.600 |
� | +3.410 |
Zhu | +3.387 |
vacc | +3.285 |
Qin | +3.207 |
Neg logit contributions | |
commute | -3.371 |
commuting | -2.822 |
notation | -2.738 |
flakes | -2.663 |
aeros | -2.637 |
Token | just |
Token ID | 655 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
Sov | +0.768 |
� | +0.708 |
Zhu | +0.703 |
vacc | +0.689 |
Qin | +0.663 |
Neg logit contributions | |
commute | -0.865 |
commuting | -0.738 |
notation | -0.719 |
flakes | -0.697 |
aeros | -0.693 |
Token | 25 |
Token ID | 1679 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
commute | +0.509 |
commuting | +0.434 |
notation | +0.422 |
flakes | +0.405 |
aeros | +0.405 |
Neg logit contributions | |
Sov | -0.489 |
� | -0.444 |
Zhu | -0.442 |
vacc | -0.439 |
�� | -0.419 |
Token | , |
Token ID | 11 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
Sov | +1.008 |
� | +0.957 |
Zhu | +0.947 |
vacc | +0.924 |
Qin | +0.893 |
Neg logit contributions | |
commute | -0.858 |
commuting | -0.710 |
notation | -0.691 |
flakes | -0.663 |
aeros | -0.663 |
Token | one |
Token ID | 530 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Sov | +0.539 |
� | +0.505 |
Zhu | +0.502 |
vacc | +0.488 |
Qin | +0.474 |
Neg logit contributions | |
commute | -0.528 |
commuting | -0.443 |
notation | -0.434 |
flakes | -0.417 |
aeros | -0.417 |
Token | at |
Token ID | 379 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
Sov | +0.369 |
� | +0.338 |
Zhu | +0.336 |
vacc | +0.331 |
Qin | +0.317 |
Neg logit contributions | |
commute | -0.395 |
commuting | -0.336 |
notation | -0.327 |
flakes | -0.316 |
aeros | -0.315 |
Token | medium |
Token ID | 7090 |
Feature activation | +4.46e+00 |
Pos logit contributions | |
Sov | +0.644 |
� | +0.597 |
Zhu | +0.594 |
vacc | +0.577 |
Qin | +0.558 |
Neg logit contributions | |
commute | -0.718 |
commuting | -0.613 |
notation | -0.594 |
flakes | -0.571 |
aeros | -0.566 |
Token | - |
Token ID | 12 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
Sov | +3.768 |
� | +3.609 |
Zhu | +3.557 |
vacc | +3.498 |
Qin | +3.407 |
Neg logit contributions | |
commute | -3.019 |
commuting | -2.522 |
notation | -2.356 |
flakes | -2.228 |
minster | -2.214 |
Token | high |
Token ID | 8929 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Sov | +0.261 |
� | +0.238 |
vacc | +0.233 |
Zhu | +0.232 |
Qin | +0.217 |
Neg logit contributions | |
commute | -0.410 |
commuting | -0.359 |
notation | -0.349 |
flakes | -0.339 |
aeros | -0.338 |
Token | heat |
Token ID | 4894 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
Sov | +1.291 |
� | +1.227 |
Zhu | +1.217 |
vacc | +1.195 |
Qin | +1.166 |
Neg logit contributions | |
commute | -0.954 |
commuting | -0.780 |
notation | -0.732 |
flakes | -0.701 |
aeros | -0.694 |
Token | until |
Token ID | 1566 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
Sov | +0.062 |
� | +0.057 |
Zhu | +0.056 |
vacc | +0.055 |
Qin | +0.053 |
Neg logit contributions | |
commute | -0.072 |
commuting | -0.062 |
notation | -0.060 |
flakes | -0.058 |
aeros | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Sov | +3.479 |
Zhu | +3.388 |
� | +3.336 |
vacc | +3.232 |
Qin | +3.214 |
Neg logit contributions | |
commute | -3.307 |
commuting | -2.771 |
notation | -2.641 |
minster | -2.499 |
aeros | -2.474 |
Token | onion |
Token ID | 21670 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
Sov | +0.758 |
Zhu | +0.702 |
� | +0.701 |
vacc | +0.680 |
Qin | +0.661 |
Neg logit contributions | |
commute | -0.826 |
commuting | -0.706 |
notation | -0.681 |
aeros | -0.650 |
minster | -0.641 |
Token | is |
Token ID | 318 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Sov | +1.732 |
� | +1.618 |
Zhu | +1.612 |
vacc | +1.574 |
Qin | +1.522 |
Neg logit contributions | |
commute | -1.615 |
commuting | -1.364 |
notation | -1.309 |
aeros | -1.234 |
minster | -1.208 |
Token | clear |
Token ID | 1598 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Sov | +0.922 |
� | +0.872 |
Zhu | +0.871 |
vacc | +0.852 |
Qin | +0.827 |
Neg logit contributions | |
commute | -0.683 |
commuting | -0.558 |
notation | -0.539 |
aeros | -0.504 |
flakes | -0.497 |
Token | or |
Token ID | 393 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
Sov | +1.136 |
� | +1.061 |
Zhu | +1.054 |
vacc | +1.033 |
Qin | +0.996 |
Neg logit contributions | |
commute | -1.152 |
commuting | -0.975 |
notation | -0.939 |
aeros | -0.901 |
flakes | -0.895 |
Token | oil |
Token ID | 3056 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
Sov | +1.170 |
� | +1.099 |
Zhu | +1.095 |
vacc | +1.075 |
Qin | +1.042 |
Neg logit contributions | |
commute | -0.811 |
commuting | -0.656 |
notation | -0.633 |
aeros | -0.597 |
minster | -0.583 |
Token | and |
Token ID | 290 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Sov | +0.904 |
� | +0.840 |
Zhu | +0.834 |
vacc | +0.817 |
Qin | +0.788 |
Neg logit contributions | |
commute | -0.916 |
commuting | -0.774 |
notation | -0.751 |
aeros | -0.715 |
flakes | -0.708 |
Token | fry |
Token ID | 32959 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
commute | +0.191 |
commuting | +0.162 |
notation | +0.158 |
flakes | +0.152 |
aeros | +0.151 |
Neg logit contributions | |
Sov | -0.188 |
� | -0.173 |
Zhu | -0.172 |
vacc | -0.169 |
Qin | -0.163 |
Token | it |
Token ID | 340 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
Sov | +1.402 |
� | +1.298 |
Zhu | +1.292 |
vacc | +1.263 |
Qin | +1.217 |
Neg logit contributions | |
commute | -1.557 |
commuting | -1.328 |
notation | -1.292 |
aeros | -1.242 |
flakes | -1.241 |
Token | at |
Token ID | 379 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
Sov | +0.369 |
� | +0.338 |
Zhu | +0.336 |
vacc | +0.331 |
Qin | +0.317 |
Neg logit contributions | |
commute | -0.395 |
commuting | -0.336 |
notation | -0.327 |
flakes | -0.316 |
aeros | -0.315 |
Token | medium |
Token ID | 7090 |
Feature activation | +4.46e+00 |
Pos logit contributions | |
Sov | +0.644 |
� | +0.597 |
Zhu | +0.594 |
vacc | +0.577 |
Qin | +0.558 |
Neg logit contributions | |
commute | -0.718 |
commuting | -0.613 |
notation | -0.594 |
flakes | -0.571 |
aeros | -0.566 |
Token | - |
Token ID | 12 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
Sov | +3.768 |
� | +3.609 |
Zhu | +3.557 |
vacc | +3.498 |
Qin | +3.407 |
Neg logit contributions | |
commute | -3.019 |
commuting | -2.522 |
notation | -2.356 |
flakes | -2.228 |
minster | -2.214 |
Token | high |
Token ID | 8929 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Sov | +0.261 |
� | +0.238 |
vacc | +0.233 |
Zhu | +0.232 |
Qin | +0.217 |
Neg logit contributions | |
commute | -0.410 |
commuting | -0.359 |
notation | -0.349 |
flakes | -0.339 |
aeros | -0.338 |
Token | heat |
Token ID | 4894 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
Sov | +1.291 |
� | +1.227 |
Zhu | +1.217 |
vacc | +1.195 |
Qin | +1.166 |
Neg logit contributions | |
commute | -0.954 |
commuting | -0.780 |
notation | -0.732 |
flakes | -0.701 |
aeros | -0.694 |
Token | until |
Token ID | 1566 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
Sov | +0.062 |
� | +0.057 |
Zhu | +0.056 |
vacc | +0.055 |
Qin | +0.053 |
Neg logit contributions | |
commute | -0.072 |
commuting | -0.062 |
notation | -0.060 |
flakes | -0.058 |
aeros | -0.058 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Sov | +3.479 |
Zhu | +3.388 |
� | +3.336 |
vacc | +3.232 |
Qin | +3.214 |
Neg logit contributions | |
commute | -3.307 |
commuting | -2.771 |
notation | -2.641 |
minster | -2.499 |
aeros | -2.474 |
Token | when |
Token ID | 618 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
commute | +0.198 |
commuting | +0.168 |
notation | +0.164 |
flakes | +0.158 |
aeros | +0.158 |
Neg logit contributions | |
Sov | -0.188 |
� | -0.173 |
Zhu | -0.172 |
vacc | -0.169 |
Qin | -0.162 |
Token | perform |
Token ID | 1620 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Sov | +1.216 |
� | +1.113 |
Zhu | +1.109 |
vacc | +1.079 |
Qin | +1.044 |
Neg logit contributions | |
commute | -1.418 |
commuting | -1.214 |
notation | -1.178 |
flakes | -1.148 |
aeros | -1.142 |
Token | a |
Token ID | 257 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
commute | +0.939 |
commuting | +0.796 |
notation | +0.775 |
flakes | +0.748 |
aeros | +0.745 |
Neg logit contributions | |
Sov | -0.914 |
� | -0.840 |
Zhu | -0.834 |
vacc | -0.821 |
Qin | -0.789 |
Token | R |
Token ID | 371 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
commute | +0.110 |
commuting | +0.091 |
notation | +0.088 |
flakes | +0.085 |
aeros | +0.085 |
Neg logit contributions | |
Sov | -0.133 |
� | -0.124 |
Zhu | -0.123 |
vacc | -0.121 |
Qin | -0.117 |
Token | OP |
Token ID | 3185 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
commute | +0.036 |
commuting | +0.031 |
notation | +0.030 |
flakes | +0.030 |
aeros | +0.029 |
Neg logit contributions | |
Sov | -0.030 |
� | -0.027 |
Zhu | -0.027 |
vacc | -0.026 |
Qin | -0.025 |
Token | exploit |
Token ID | 14561 |
Feature activation | -4.93e+00 |
Pos logit contributions | |
commute | +0.247 |
commuting | +0.207 |
notation | +0.200 |
flakes | +0.194 |
aeros | +0.192 |
Neg logit contributions | |
Sov | -0.287 |
� | -0.266 |
Zhu | -0.264 |
vacc | -0.260 |
Qin | -0.252 |
Token | , |
Token ID | 11 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
commute | +3.400 |
commuting | +2.789 |
notation | +2.753 |
flakes | +2.593 |
aeros | +2.577 |
Neg logit contributions | |
Sov | -4.284 |
� | -3.933 |
vacc | -3.915 |
Zhu | -3.906 |
�� | -3.730 |
Token | there |
Token ID | 612 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
commute | +0.433 |
commuting | +0.369 |
notation | +0.360 |
flakes | +0.347 |
aeros | +0.346 |
Neg logit contributions | |
Sov | -0.395 |
� | -0.362 |
Zhu | -0.359 |
vacc | -0.354 |
Qin | -0.339 |
Token | 's |
Token ID | 338 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
commute | +0.376 |
commuting | +0.311 |
notation | +0.300 |
flakes | +0.288 |
aeros | +0.287 |
Neg logit contributions | |
Sov | -0.470 |
� | -0.436 |
Zhu | -0.434 |
vacc | -0.427 |
Qin | -0.413 |
Token | a |
Token ID | 257 |
Feature activation | -9.06e-01 |
Pos logit contributions | |
Sov | +0.356 |
� | +0.328 |
Zhu | +0.327 |
vacc | +0.320 |
Qin | +0.309 |
Neg logit contributions | |
commute | -0.360 |
commuting | -0.305 |
notation | -0.294 |
flakes | -0.285 |
aeros | -0.284 |
Token | very |
Token ID | 845 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
commute | +0.711 |
commuting | +0.601 |
notation | +0.587 |
flakes | +0.564 |
aeros | +0.561 |
Neg logit contributions | |
Sov | -0.710 |
� | -0.653 |
Zhu | -0.647 |
vacc | -0.639 |
Qin | -0.613 |
Token | center |
Token ID | 3641 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Sov | +0.931 |
� | +0.848 |
Zhu | +0.843 |
vacc | +0.821 |
Qin | +0.787 |
Neg logit contributions | |
commute | -1.238 |
commuting | -1.071 |
notation | -1.050 |
flakes | -1.028 |
aeros | -1.017 |
Token | for |
Token ID | 329 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
Sov | +0.941 |
� | +0.873 |
Zhu | +0.865 |
vacc | +0.849 |
Qin | +0.820 |
Neg logit contributions | |
commute | -0.915 |
commuting | -0.771 |
notation | -0.753 |
flakes | -0.730 |
aeros | -0.726 |
Token | the |
Token ID | 262 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
Sov | +0.171 |
� | +0.157 |
Zhu | +0.156 |
vacc | +0.153 |
Qin | +0.147 |
Neg logit contributions | |
commute | -0.194 |
commuting | -0.165 |
notation | -0.162 |
flakes | -0.156 |
aeros | -0.156 |
Token | city |
Token ID | 1748 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
Sov | +0.156 |
� | +0.145 |
Zhu | +0.144 |
vacc | +0.142 |
Qin | +0.137 |
Neg logit contributions | |
commute | -0.139 |
commuting | -0.116 |
notation | -0.113 |
flakes | -0.109 |
aeros | -0.109 |
Token | ," |
Token ID | 553 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
commute | +0.323 |
commuting | +0.274 |
notation | +0.269 |
flakes | +0.259 |
aeros | +0.258 |
Neg logit contributions | |
Sov | -0.305 |
� | -0.281 |
Zhu | -0.279 |
vacc | -0.274 |
Qin | -0.263 |
Token | Ful |
Token ID | 28040 |
Feature activation | -4.88e+00 |
Pos logit contributions | |
commute | +0.835 |
commuting | +0.691 |
notation | +0.673 |
flakes | +0.644 |
aeros | +0.643 |
Neg logit contributions | |
Sov | -1.020 |
� | -0.948 |
Zhu | -0.948 |
vacc | -0.928 |
Qin | -0.900 |
Token | op |
Token ID | 404 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
commute | +4.793 |
commuting | +4.168 |
notation | +4.125 |
aster | +4.117 |
flakes | +4.032 |
Neg logit contributions | |
Sov | -2.746 |
Zhu | -2.447 |
� | -2.347 |
�� | -2.311 |
��極 | -2.285 |
Token | said |
Token ID | 531 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
commute | +0.258 |
commuting | +0.217 |
notation | +0.212 |
flakes | +0.205 |
aeros | +0.204 |
Neg logit contributions | |
Sov | -0.261 |
� | -0.244 |
Zhu | -0.241 |
vacc | -0.236 |
Qin | -0.229 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
commute | +0.907 |
commuting | +0.769 |
notation | +0.750 |
flakes | +0.724 |
aeros | +0.721 |
Neg logit contributions | |
Sov | -0.876 |
� | -0.807 |
Zhu | -0.801 |
vacc | -0.787 |
Qin | -0.757 |
Token |
|
Token ID | 198 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
commute | +0.214 |
commuting | +0.180 |
notation | +0.176 |
aeros | +0.169 |
flakes | +0.169 |
Neg logit contributions | |
Sov | -0.231 |
Zhu | -0.215 |
� | -0.215 |
vacc | -0.210 |
Qin | -0.204 |
Token |
|
Token ID | 198 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
commute | +0.362 |
commuting | +0.295 |
notation | +0.286 |
flakes | +0.273 |
aeros | +0.272 |
Neg logit contributions | |
Sov | -0.500 |
� | -0.464 |
Zhu | -0.463 |
vacc | -0.457 |
Qin | -0.441 |
Token | Monday |
Token ID | 3321 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
Sov | +0.050 |
Zhu | +0.046 |
� | +0.046 |
vacc | +0.045 |
Qin | +0.044 |
Neg logit contributions | |
commute | -0.044 |
commuting | -0.037 |
notation | -0.036 |
flakes | -0.034 |
aeros | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
Sov | +0.589 |
� | +0.553 |
Zhu | +0.552 |
vacc | +0.536 |
Qin | +0.520 |
Neg logit contributions | |
commute | -0.594 |
commuting | -0.504 |
notation | -0.496 |
flakes | -0.476 |
aeros | -0.472 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
commute | +0.257 |
commuting | +0.216 |
notation | +0.210 |
flakes | +0.198 |
aeros | +0.197 |
Neg logit contributions | |
Sov | -0.287 |
Zhu | -0.268 |
� | -0.268 |
vacc | -0.261 |
Qin | -0.256 |
Token |
|
Token ID | 198 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
commute | +0.091 |
commuting | +0.074 |
notation | +0.072 |
flakes | +0.069 |
aeros | +0.068 |
Neg logit contributions | |
Sov | -0.124 |
� | -0.115 |
Zhu | -0.115 |
vacc | -0.113 |
Qin | -0.110 |
Token | ATT |
Token ID | 17139 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Sov | +0.036 |
� | +0.033 |
Zhu | +0.032 |
vacc | +0.032 |
Qin | +0.030 |
Neg logit contributions | |
commute | -0.045 |
commuting | -0.038 |
notation | -0.037 |
flakes | -0.036 |
aeros | -0.036 |
Token | ENTION |
Token ID | 45589 |
Feature activation | -4.80e+00 |
Pos logit contributions | |
commute | +1.218 |
commuting | +1.058 |
notation | +1.033 |
flakes | +1.004 |
aeros | +1.000 |
Neg logit contributions | |
Sov | -0.853 |
� | -0.773 |
Zhu | -0.765 |
vacc | -0.752 |
Qin | -0.715 |
Token | , |
Token ID | 837 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
commute | +3.578 |
commuting | +3.003 |
notation | +2.913 |
flakes | +2.826 |
aeros | +2.824 |
Neg logit contributions | |
Sov | -3.933 |
Zhu | -3.569 |
� | -3.557 |
vacc | -3.511 |
Qin | -3.370 |
Token | en |
Token ID | 551 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
commute | +1.325 |
commuting | +1.122 |
notation | +1.093 |
flakes | +1.051 |
aeros | +1.046 |
Neg logit contributions | |
Sov | -1.309 |
� | -1.209 |
Zhu | -1.200 |
vacc | -1.182 |
Qin | -1.136 |
Token | ra |
Token ID | 2179 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
commute | +1.118 |
commuting | +0.948 |
notation | +0.923 |
flakes | +0.891 |
aeros | +0.887 |
Neg logit contributions | |
Sov | -1.085 |
� | -0.996 |
Zhu | -0.990 |
vacc | -0.974 |
Qin | -0.936 |
Token | ison |
Token ID | 1653 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
commute | +0.647 |
commuting | +0.556 |
notation | +0.540 |
flakes | +0.522 |
aeros | +0.520 |
Neg logit contributions | |
Sov | -0.549 |
� | -0.505 |
Zhu | -0.501 |
vacc | -0.493 |
Qin | -0.472 |
Token | des |
Token ID | 748 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
Sov | +0.180 |
� | +0.173 |
Zhu | +0.169 |
vacc | +0.165 |
Qin | +0.160 |
Neg logit contributions | |
commute | -0.169 |
commuting | -0.143 |
notation | -0.138 |
flakes | -0.134 |
aeros | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
commute | +0.533 |
commuting | +0.444 |
notation | +0.432 |
flakes | +0.413 |
aeros | +0.412 |
Neg logit contributions | |
Sov | -0.612 |
� | -0.570 |
Zhu | -0.566 |
vacc | -0.556 |
Qin | -0.538 |
Token | his |
Token ID | 465 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
Sov | +0.303 |
Zhu | +0.277 |
� | +0.276 |
vacc | +0.269 |
Qin | +0.260 |
Neg logit contributions | |
commute | -0.383 |
commuting | -0.329 |
notation | -0.322 |
flakes | -0.311 |
aeros | -0.310 |
Token | co |
Token ID | 763 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
commute | +0.175 |
commuting | +0.146 |
notation | +0.142 |
flakes | +0.136 |
aeros | +0.135 |
Neg logit contributions | |
Sov | -0.210 |
� | -0.195 |
Zhu | -0.194 |
vacc | -0.191 |
Qin | -0.185 |
Token | - |
Token ID | 12 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
commute | +0.587 |
commuting | +0.494 |
notation | +0.481 |
flakes | +0.464 |
aeros | +0.461 |
Neg logit contributions | |
Sov | -0.621 |
� | -0.574 |
Zhu | -0.569 |
vacc | -0.563 |
Qin | -0.542 |
Token | p |
Token ID | 79 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
commute | +0.313 |
commuting | +0.265 |
notation | +0.258 |
flakes | +0.249 |
aeros | +0.248 |
Neg logit contributions | |
Sov | -0.312 |
� | -0.287 |
Zhu | -0.285 |
vacc | -0.281 |
Qin | -0.270 |
Token | il |
Token ID | 346 |
Feature activation | -4.80e+00 |
Pos logit contributions | |
commute | +0.299 |
commuting | +0.257 |
notation | +0.249 |
aeros | +0.240 |
flakes | +0.240 |
Neg logit contributions | |
Sov | -0.262 |
� | -0.241 |
Zhu | -0.239 |
vacc | -0.238 |
Qin | -0.226 |
Token | ots |
Token ID | 1747 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
commute | +2.580 |
commuting | +1.999 |
notation | +1.900 |
flakes | +1.811 |
aeros | +1.798 |
Neg logit contributions | |
Sov | -4.937 |
� | -4.640 |
Zhu | -4.606 |
vacc | -4.551 |
Qin | -4.423 |
Token | tried |
Token ID | 3088 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
commute | +0.972 |
commuting | +0.807 |
notation | +0.783 |
flakes | +0.751 |
aeros | +0.747 |
Neg logit contributions | |
Sov | -1.180 |
� | -1.096 |
Zhu | -1.089 |
vacc | -1.074 |
Qin | -1.037 |
Token | to |
Token ID | 284 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
commute | +0.552 |
commuting | +0.470 |
notation | +0.457 |
flakes | +0.442 |
aeros | +0.440 |
Neg logit contributions | |
Sov | -0.521 |
� | -0.478 |
Zhu | -0.475 |
vacc | -0.468 |
Qin | -0.449 |
Token | stop |
Token ID | 2245 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
commute | +0.583 |
commuting | +0.485 |
notation | +0.470 |
flakes | +0.452 |
aeros | +0.450 |
Neg logit contributions | |
Sov | -0.684 |
� | -0.634 |
Zhu | -0.630 |
vacc | -0.621 |
Qin | -0.599 |
Token | them |
Token ID | 606 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
Sov | +0.809 |
Zhu | +0.749 |
� | +0.741 |
vacc | +0.726 |
Qin | +0.711 |
Neg logit contributions | |
commute | -0.981 |
commuting | -0.837 |
notation | -0.824 |
flakes | -0.793 |
aeros | -0.790 |
Token | , |
Token ID | 11 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
commute | +0.029 |
commuting | +0.024 |
notation | +0.024 |
flakes | +0.023 |
aeros | +0.023 |
Neg logit contributions | |
Sov | -0.034 |
� | -0.032 |
Zhu | -0.032 |
vacc | -0.031 |
Qin | -0.030 |
Token | change |
Token ID | 1487 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
commute | +0.153 |
commuting | +0.128 |
notation | +0.124 |
flakes | +0.119 |
aeros | +0.118 |
Neg logit contributions | |
Sov | -0.179 |
� | -0.166 |
Zhu | -0.165 |
vacc | -0.162 |
Qin | -0.157 |
Token | the |
Token ID | 262 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
commute | +1.746 |
commuting | +1.486 |
notation | +1.456 |
flakes | +1.399 |
aeros | +1.392 |
Neg logit contributions | |
Sov | -1.590 |
� | -1.441 |
Zhu | -1.428 |
vacc | -1.423 |
Qin | -1.343 |
Token | Fire |
Token ID | 3764 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
commute | +0.378 |
commuting | +0.319 |
notation | +0.310 |
flakes | +0.299 |
aeros | +0.298 |
Neg logit contributions | |
Sov | -0.388 |
� | -0.357 |
Zhu | -0.355 |
vacc | -0.350 |
Qin | -0.336 |
Token | wall |
Token ID | 11930 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Sov | +0.615 |
� | +0.554 |
Zhu | +0.548 |
vacc | +0.535 |
Qin | +0.508 |
Neg logit contributions | |
commute | -1.073 |
commuting | -0.944 |
notation | -0.925 |
flakes | -0.900 |
aeros | -0.898 |
Token | Settings |
Token ID | 16163 |
Feature activation | -4.75e+00 |
Pos logit contributions | |
commute | +1.076 |
commuting | +0.900 |
notation | +0.874 |
flakes | +0.840 |
aeros | +0.836 |
Neg logit contributions | |
Sov | -1.212 |
� | -1.121 |
Zhu | -1.114 |
vacc | -1.098 |
Qin | -1.058 |
Token | to |
Token ID | 284 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
commute | +3.482 |
commuting | +2.883 |
notation | +2.876 |
flakes | +2.723 |
aeros | +2.698 |
Neg logit contributions | |
Sov | -3.979 |
vacc | -3.608 |
� | -3.601 |
Zhu | -3.593 |
�� | -3.416 |
Token | the |
Token ID | 262 |
Feature activation | -8.72e-01 |
Pos logit contributions | |
commute | +0.586 |
commuting | +0.492 |
notation | +0.478 |
flakes | +0.461 |
aeros | +0.458 |
Neg logit contributions | |
Sov | -0.632 |
� | -0.583 |
Zhu | -0.580 |
vacc | -0.571 |
Qin | -0.550 |
Token | minimum |
Token ID | 5288 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
commute | +0.691 |
commuting | +0.585 |
notation | +0.570 |
flakes | +0.550 |
aeros | +0.548 |
Neg logit contributions | |
Sov | -0.676 |
� | -0.621 |
Zhu | -0.617 |
vacc | -0.608 |
Qin | -0.583 |
Token | setting |
Token ID | 4634 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
commute | +0.364 |
commuting | +0.306 |
notation | +0.296 |
flakes | +0.287 |
aeros | +0.285 |
Neg logit contributions | |
Sov | -0.389 |
� | -0.361 |
Zhu | -0.358 |
vacc | -0.351 |
Qin | -0.340 |
Token | . |
Token ID | 13 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
commute | +2.856 |
commuting | +2.397 |
notation | +2.341 |
flakes | +2.233 |
aeros | +2.228 |
Neg logit contributions | |
Sov | -3.157 |
� | -2.876 |
Zhu | -2.859 |
vacc | -2.845 |
�� | -2.709 |
Token | ITT |
Token ID | 22470 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Sov | +0.329 |
� | +0.304 |
Zhu | +0.302 |
vacc | +0.296 |
Qin | +0.285 |
Neg logit contributions | |
commute | -0.346 |
commuting | -0.294 |
notation | -0.286 |
flakes | -0.276 |
aeros | -0.276 |
Token | SM |
Token ID | 9447 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
commute | +1.707 |
commuting | +1.442 |
notation | +1.408 |
flakes | +1.365 |
aeros | +1.359 |
Neg logit contributions | |
Sov | -1.683 |
� | -1.532 |
Zhu | -1.519 |
vacc | -1.519 |
�� | -1.448 |
Token | ITH |
Token ID | 10554 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
commute | +0.217 |
commuting | +0.194 |
notation | +0.190 |
flakes | +0.185 |
aeros | +0.185 |
Neg logit contributions | |
Sov | -0.089 |
� | -0.078 |
Zhu | -0.076 |
vacc | -0.074 |
Qin | -0.069 |
Token | , |
Token ID | 11 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
Sov | +1.976 |
� | +1.875 |
Zhu | +1.824 |
vacc | +1.807 |
Qin | +1.728 |
Neg logit contributions | |
commute | -1.784 |
commuting | -1.503 |
notation | -1.442 |
flakes | -1.384 |
aeros | -1.376 |
Token | FL |
Token ID | 9977 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
Sov | +0.230 |
� | +0.213 |
Zhu | +0.212 |
vacc | +0.207 |
Qin | +0.200 |
Neg logit contributions | |
commute | -0.239 |
commuting | -0.202 |
notation | -0.197 |
flakes | -0.190 |
aeros | -0.189 |
Token | OR |
Token ID | 1581 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
Sov | +0.747 |
� | +0.693 |
Zhu | +0.686 |
vacc | +0.676 |
Qin | +0.651 |
Neg logit contributions | |
commute | -0.751 |
commuting | -0.636 |
notation | -0.616 |
aeros | -0.591 |
flakes | -0.590 |
Token | IDA |
Token ID | 41957 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
commute | +3.895 |
commuting | +3.342 |
notation | +3.288 |
aeros | +3.199 |
flakes | +3.180 |
Neg logit contributions | |
Sov | -3.323 |
vacc | -3.038 |
Zhu | -3.029 |
� | -3.029 |
�� | -2.887 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
commute | +1.367 |
commuting | +1.124 |
notation | +1.087 |
flakes | +1.042 |
aeros | +1.037 |
Neg logit contributions | |
Sov | -1.794 |
� | -1.663 |
Zhu | -1.657 |
vacc | -1.636 |
Qin | -1.578 |
Token |
|
Token ID | 198 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
commute | +0.671 |
commuting | +0.542 |
notation | +0.528 |
flakes | +0.502 |
aeros | +0.501 |
Neg logit contributions | |
Sov | -0.979 |
Zhu | -0.904 |
� | -0.904 |
vacc | -0.896 |
�� | -0.876 |
Token | Y |
Token ID | 56 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
commute | +0.447 |
commuting | +0.394 |
notation | +0.386 |
flakes | +0.376 |
aeros | +0.375 |
Neg logit contributions | |
Sov | -0.245 |
� | -0.217 |
Zhu | -0.215 |
vacc | -0.210 |
Qin | -0.198 |
Token | ards |
Token ID | 1371 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
commute | +1.139 |
commuting | +0.915 |
notation | +0.901 |
flakes | +0.854 |
aeros | +0.854 |
Neg logit contributions | |
Sov | -1.776 |
Zhu | -1.646 |
� | -1.639 |
vacc | -1.621 |
�� | -1.608 |
Token | of |
Token ID | 286 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
commute | +0.572 |
commuting | +0.478 |
notation | +0.461 |
flakes | +0.442 |
aeros | +0.441 |
Neg logit contributions | |
Sov | -0.647 |
� | -0.604 |
Zhu | -0.599 |
vacc | -0.587 |
Qin | -0.570 |
Token | Pre |
Token ID | 3771 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
Sov | +0.193 |
� | +0.181 |
Zhu | +0.181 |
vacc | +0.179 |
Qin | +0.174 |
Neg logit contributions | |
commute | -0.090 |
commuting | -0.068 |
notation | -0.065 |
flakes | -0.061 |
aeros | -0.061 |
Token | y |
Token ID | 88 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
commute | +0.650 |
commuting | +0.540 |
notation | +0.531 |
flakes | +0.507 |
aeros | +0.506 |
Neg logit contributions | |
Sov | -0.746 |
Zhu | -0.683 |
� | -0.683 |
vacc | -0.672 |
�� | -0.650 |
Token |
|
Token ID | 198 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
commute | +0.805 |
commuting | +0.671 |
notation | +0.651 |
flakes | +0.625 |
aeros | +0.622 |
Neg logit contributions | |
Sov | -0.939 |
� | -0.869 |
Zhu | -0.864 |
vacc | -0.852 |
Qin | -0.821 |
Token |
|
Token ID | 198 |
Feature activation | -9.92e-01 |
Pos logit contributions | |
commute | +0.730 |
commuting | +0.588 |
notation | +0.574 |
flakes | +0.544 |
aeros | +0.543 |
Neg logit contributions | |
Sov | -1.083 |
Zhu | -0.999 |
� | -0.997 |
vacc | -0.992 |
�� | -0.974 |
Token | https |
Token ID | 5450 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
commute | +0.873 |
commuting | +0.753 |
notation | +0.738 |
flakes | +0.714 |
aeros | +0.711 |
Neg logit contributions | |
Sov | -0.676 |
� | -0.613 |
Zhu | -0.613 |
vacc | -0.600 |
Qin | -0.575 |
Token | :// |
Token ID | 1378 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
commute | +3.000 |
commuting | +2.419 |
notation | +2.407 |
aster | +2.351 |
flakes | +2.318 |
Neg logit contributions | |
Sov | -3.807 |
�� | -3.573 |
Zhu | -3.506 |
vacc | -3.466 |
��極 | -3.420 |
Token | kill |
Token ID | 12728 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
commute | +1.790 |
commuting | +1.544 |
notation | +1.509 |
flakes | +1.461 |
aeros | +1.456 |
Neg logit contributions | |
Sov | -1.393 |
� | -1.263 |
Zhu | -1.258 |
vacc | -1.230 |
Qin | -1.181 |
Token | ert |
Token ID | 861 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
commute | +1.030 |
commuting | +0.883 |
notation | +0.860 |
flakes | +0.833 |
aeros | +0.830 |
Neg logit contributions | |
Sov | -0.873 |
� | -0.799 |
Zhu | -0.792 |
vacc | -0.782 |
Qin | -0.746 |
Token | of |
Token ID | 1659 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
Sov | +0.095 |
� | +0.087 |
Zhu | +0.086 |
vacc | +0.085 |
Qin | +0.081 |
Neg logit contributions | |
commute | -0.114 |
commuting | -0.099 |
notation | -0.096 |
flakes | -0.093 |
aeros | -0.092 |
Token | ure |
Token ID | 495 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
commute | +0.313 |
commuting | +0.266 |
notation | +0.258 |
flakes | +0.249 |
aeros | +0.248 |
Neg logit contributions | |
Sov | -0.299 |
� | -0.276 |
vacc | -0.273 |
Zhu | -0.271 |
Qin | -0.258 |
Token | the |
Token ID | 262 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
Sov | +0.939 |
Zhu | +0.884 |
� | +0.882 |
vacc | +0.855 |
Qin | +0.837 |
Neg logit contributions | |
commute | -1.066 |
commuting | -0.906 |
notation | -0.890 |
flakes | -0.855 |
aeros | -0.850 |
Token | Club |
Token ID | 6289 |
Feature activation | -8.89e-01 |
Pos logit contributions | |
commute | +0.349 |
commuting | +0.292 |
notation | +0.284 |
flakes | +0.273 |
aeros | +0.272 |
Neg logit contributions | |
Sov | -0.386 |
� | -0.357 |
Zhu | -0.355 |
vacc | -0.349 |
Qin | -0.337 |
Token | and |
Token ID | 290 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
commute | +0.645 |
commuting | +0.537 |
notation | +0.522 |
flakes | +0.501 |
aeros | +0.499 |
Neg logit contributions | |
Sov | -0.744 |
� | -0.692 |
Zhu | -0.688 |
vacc | -0.676 |
Qin | -0.655 |
Token | a |
Token ID | 257 |
Feature activation | -7.25e-01 |
Pos logit contributions | |
commute | +0.132 |
commuting | +0.112 |
notation | +0.109 |
flakes | +0.105 |
aeros | +0.104 |
Neg logit contributions | |
Sov | -0.130 |
� | -0.121 |
Zhu | -0.120 |
vacc | -0.118 |
Qin | -0.114 |
Token | new |
Token ID | 649 |
Feature activation | +5.33e-02 |
Pos logit contributions | |
commute | +0.566 |
commuting | +0.479 |
notation | +0.466 |
flakes | +0.449 |
aeros | +0.448 |
Neg logit contributions | |
Sov | -0.573 |
� | -0.525 |
Zhu | -0.521 |
vacc | -0.514 |
Qin | -0.493 |
Token | sponsor |
Token ID | 17487 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
Sov | +0.043 |
� | +0.040 |
Zhu | +0.040 |
vacc | +0.039 |
Qin | +0.038 |
Neg logit contributions | |
commute | -0.040 |
commuting | -0.033 |
notation | -0.032 |
flakes | -0.031 |
aeros | -0.031 |
Token | is |
Token ID | 318 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
commute | +3.066 |
commuting | +2.532 |
notation | +2.463 |
flakes | +2.359 |
aeros | +2.356 |
Neg logit contributions | |
Sov | -3.899 |
� | -3.565 |
vacc | -3.541 |
Zhu | -3.534 |
�� | -3.400 |
Token | for |
Token ID | 329 |
Feature activation | -6.86e-02 |
Pos logit contributions | |
commute | +0.247 |
commuting | +0.208 |
notation | +0.202 |
flakes | +0.195 |
aeros | +0.194 |
Neg logit contributions | |
Sov | -0.262 |
� | -0.242 |
Zhu | -0.241 |
vacc | -0.236 |
Qin | -0.228 |
Token | the |
Token ID | 262 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
commute | +0.056 |
commuting | +0.047 |
notation | +0.046 |
flakes | +0.045 |
aeros | +0.044 |
Neg logit contributions | |
Sov | -0.051 |
� | -0.047 |
Zhu | -0.047 |
vacc | -0.046 |
Qin | -0.045 |
Token | Premier |
Token ID | 9952 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
commute | +0.105 |
commuting | +0.088 |
notation | +0.086 |
flakes | +0.082 |
aeros | +0.082 |
Neg logit contributions | |
Sov | -0.111 |
� | -0.102 |
Zhu | -0.102 |
vacc | -0.100 |
Qin | -0.096 |
Token | League |
Token ID | 4041 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
Sov | +0.410 |
� | +0.386 |
Zhu | +0.384 |
vacc | +0.375 |
Qin | +0.366 |
Neg logit contributions | |
commute | -0.360 |
commuting | -0.300 |
notation | -0.292 |
flakes | -0.281 |
aeros | -0.280 |
Token | when |
Token ID | 618 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
commute | +0.476 |
commuting | +0.396 |
notation | +0.389 |
flakes | +0.373 |
aeros | +0.372 |
Neg logit contributions | |
Sov | -0.517 |
� | -0.481 |
Zhu | -0.481 |
vacc | -0.469 |
Qin | -0.456 |
Token | the |
Token ID | 262 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
Sov | +0.380 |
Zhu | +0.356 |
� | +0.352 |
vacc | +0.342 |
Qin | +0.339 |
Neg logit contributions | |
commute | -0.374 |
commuting | -0.312 |
notation | -0.306 |
aeros | -0.293 |
flakes | -0.292 |
Token | rubber |
Token ID | 14239 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
commute | +0.220 |
commuting | +0.186 |
notation | +0.180 |
flakes | +0.174 |
aeros | +0.173 |
Neg logit contributions | |
Sov | -0.217 |
� | -0.200 |
Zhu | -0.199 |
vacc | -0.195 |
Qin | -0.189 |
Token | ' |
Token ID | 705 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
Sov | +0.555 |
Zhu | +0.514 |
� | +0.514 |
vacc | +0.505 |
Qin | +0.489 |
Neg logit contributions | |
commute | -0.501 |
commuting | -0.419 |
notation | -0.405 |
aeros | -0.386 |
flakes | -0.385 |
Token | t |
Token ID | 83 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
commute | +0.513 |
commuting | +0.440 |
notation | +0.429 |
flakes | +0.416 |
aeros | +0.414 |
Neg logit contributions | |
Sov | -0.427 |
� | -0.390 |
Zhu | -0.386 |
vacc | -0.381 |
Qin | -0.364 |
Token | ent |
Token ID | 298 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
Sov | +0.174 |
� | +0.161 |
Zhu | +0.158 |
vacc | +0.157 |
Qin | +0.150 |
Neg logit contributions | |
commute | -0.211 |
commuting | -0.182 |
notation | -0.177 |
flakes | -0.171 |
aeros | -0.171 |
Token | acles |
Token ID | 9928 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
commute | +3.496 |
commuting | +2.975 |
notation | +2.916 |
flakes | +2.833 |
aeros | +2.787 |
Neg logit contributions | |
Sov | -3.377 |
� | -3.099 |
Zhu | -3.079 |
vacc | -3.035 |
�� | -2.912 |
Token | ' |
Token ID | 6 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
commute | +0.116 |
commuting | +0.101 |
notation | +0.098 |
aeros | +0.095 |
flakes | +0.095 |
Neg logit contributions | |
Sov | -0.086 |
� | -0.078 |
Zhu | -0.077 |
vacc | -0.076 |
Qin | -0.072 |
Token | started |
Token ID | 2067 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
commute | +0.780 |
commuting | +0.649 |
notation | +0.629 |
flakes | +0.603 |
aeros | +0.601 |
Neg logit contributions | |
Sov | -0.905 |
� | -0.841 |
Zhu | -0.837 |
vacc | -0.823 |
Qin | -0.796 |
Token | interfering |
Token ID | 32874 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
commute | +1.372 |
commuting | +1.163 |
notation | +1.130 |
flakes | +1.092 |
aeros | +1.087 |
Neg logit contributions | |
Sov | -1.367 |
� | -1.256 |
Zhu | -1.247 |
vacc | -1.229 |
Qin | -1.181 |
Token | with |
Token ID | 351 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
Sov | +1.559 |
� | +1.460 |
Zhu | +1.455 |
vacc | +1.405 |
Qin | +1.380 |
Neg logit contributions | |
commute | -1.877 |
commuting | -1.597 |
notation | -1.563 |
aeros | -1.504 |
flakes | -1.494 |
Token | put |
Token ID | 1234 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
commute | +0.365 |
commuting | +0.304 |
notation | +0.292 |
flakes | +0.281 |
aeros | +0.281 |
Neg logit contributions | |
Sov | -0.439 |
� | -0.408 |
Zhu | -0.405 |
vacc | -0.398 |
Qin | -0.386 |
Token | ts |
Token ID | 912 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
commute | +0.134 |
commuting | +0.118 |
notation | +0.115 |
flakes | +0.112 |
aeros | +0.112 |
Neg logit contributions | |
Sov | -0.071 |
� | -0.064 |
Zhu | -0.063 |
vacc | -0.061 |
Qin | -0.058 |
Token | easily |
Token ID | 3538 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
commute | +0.115 |
commuting | +0.096 |
notation | +0.093 |
flakes | +0.090 |
aeros | +0.090 |
Neg logit contributions | |
Sov | -0.131 |
� | -0.122 |
Zhu | -0.121 |
vacc | -0.119 |
Qin | -0.115 |
Token | , |
Token ID | 11 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
Sov | +0.718 |
� | +0.677 |
Zhu | +0.671 |
vacc | +0.655 |
Qin | +0.637 |
Neg logit contributions | |
commute | -0.634 |
commuting | -0.528 |
notation | -0.511 |
aeros | -0.495 |
flakes | -0.493 |
Token | feel |
Token ID | 1254 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
commute | +0.230 |
commuting | +0.194 |
notation | +0.188 |
flakes | +0.182 |
aeros | +0.181 |
Neg logit contributions | |
Sov | -0.236 |
� | -0.220 |
Zhu | -0.218 |
vacc | -0.214 |
Qin | -0.207 |
Token | free |
Token ID | 1479 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
commute | +0.799 |
commuting | +0.663 |
notation | +0.644 |
flakes | +0.616 |
aeros | +0.612 |
Neg logit contributions | |
Sov | -0.982 |
� | -0.908 |
Zhu | -0.899 |
vacc | -0.891 |
Qin | -0.857 |
Token | to |
Token ID | 284 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
commute | +3.505 |
commuting | +2.974 |
notation | +2.894 |
flakes | +2.782 |
aeros | +2.761 |
Neg logit contributions | |
Sov | -3.441 |
� | -3.135 |
Zhu | -3.129 |
vacc | -3.092 |
Qin | -2.957 |
Token | make |
Token ID | 787 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
commute | +1.027 |
commuting | +0.850 |
notation | +0.823 |
flakes | +0.790 |
aeros | +0.786 |
Neg logit contributions | |
Sov | -1.275 |
� | -1.184 |
Zhu | -1.177 |
vacc | -1.161 |
Qin | -1.121 |
Token | more |
Token ID | 517 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
commute | +0.607 |
commuting | +0.518 |
notation | +0.504 |
flakes | +0.488 |
aeros | +0.487 |
Neg logit contributions | |
Sov | -0.536 |
� | -0.492 |
Zhu | -0.489 |
vacc | -0.479 |
Qin | -0.461 |
Token | than |
Token ID | 621 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
Sov | +0.575 |
� | +0.536 |
Zhu | +0.532 |
vacc | +0.521 |
Qin | +0.507 |
Neg logit contributions | |
commute | -0.520 |
commuting | -0.435 |
notation | -0.419 |
aeros | -0.404 |
flakes | -0.403 |
Token | one |
Token ID | 530 |
Feature activation | -8.70e-02 |
Pos logit contributions | |
Sov | +0.210 |
� | +0.191 |
Zhu | +0.190 |
vacc | +0.187 |
Qin | +0.179 |
Neg logit contributions | |
commute | -0.244 |
commuting | -0.209 |
notation | -0.204 |
flakes | -0.197 |
aeros | -0.196 |
Token | category |
Token ID | 6536 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
commute | +0.630 |
commuting | +0.527 |
notation | +0.510 |
flakes | +0.490 |
aeros | +0.489 |
Neg logit contributions | |
Sov | -0.717 |
� | -0.665 |
Zhu | -0.660 |
vacc | -0.650 |
Qin | -0.627 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
commute | +1.251 |
commuting | +1.051 |
notation | +1.020 |
flakes | +0.982 |
aeros | +0.978 |
Neg logit contributions | |
Sov | -1.349 |
� | -1.248 |
Zhu | -1.240 |
vacc | -1.220 |
Qin | -1.175 |
Token | where |
Token ID | 810 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
commute | +0.112 |
commuting | +0.094 |
notation | +0.091 |
flakes | +0.088 |
aeros | +0.088 |
Neg logit contributions | |
Sov | -0.121 |
� | -0.112 |
Zhu | -0.111 |
vacc | -0.109 |
Qin | -0.105 |
Token | Emmy |
Token ID | 33953 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
commute | +0.161 |
commuting | +0.137 |
notation | +0.133 |
flakes | +0.128 |
aeros | +0.128 |
Neg logit contributions | |
Sov | -0.155 |
� | -0.144 |
Zhu | -0.143 |
vacc | -0.140 |
Qin | -0.135 |
Token | - |
Token ID | 12 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
commute | +1.090 |
commuting | +0.905 |
notation | +0.876 |
flakes | +0.840 |
aeros | +0.838 |
Neg logit contributions | |
Sov | -1.305 |
� | -1.210 |
Zhu | -1.204 |
vacc | -1.186 |
Qin | -1.144 |
Token | neg |
Token ID | 12480 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
commute | +0.964 |
commuting | +0.802 |
notation | +0.777 |
aeros | +0.748 |
flakes | +0.746 |
Neg logit contributions | |
Sov | -1.128 |
� | -1.047 |
Zhu | -1.037 |
vacc | -1.029 |
Qin | -0.984 |
Token | lected |
Token ID | 12609 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
commute | +3.052 |
commuting | +2.523 |
notation | +2.434 |
flakes | +2.311 |
aeros | +2.287 |
Neg logit contributions | |
Sov | -3.811 |
� | -3.541 |
Zhu | -3.540 |
vacc | -3.495 |
Qin | -3.374 |
Token | The |
Token ID | 383 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
commute | +0.593 |
commuting | +0.496 |
notation | +0.481 |
flakes | +0.462 |
aeros | +0.461 |
Neg logit contributions | |
Sov | -0.672 |
� | -0.624 |
Zhu | -0.621 |
vacc | -0.610 |
Qin | -0.590 |
Token | Good |
Token ID | 4599 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
commute | +0.497 |
commuting | +0.426 |
notation | +0.416 |
aeros | +0.402 |
flakes | +0.399 |
Neg logit contributions | |
Sov | -0.427 |
� | -0.394 |
Zhu | -0.394 |
vacc | -0.384 |
Qin | -0.371 |
Token | Wife |
Token ID | 34188 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Sov | +0.050 |
� | +0.045 |
Zhu | +0.044 |
vacc | +0.043 |
Qin | +0.041 |
Neg logit contributions | |
commute | -0.071 |
commuting | -0.062 |
notation | -0.060 |
flakes | -0.059 |
aeros | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | -6.15e-01 |
Pos logit contributions | |
commute | +1.132 |
commuting | +0.956 |
notation | +0.929 |
flakes | +0.899 |
aeros | +0.894 |
Neg logit contributions | |
Sov | -1.143 |
� | -1.049 |
Zhu | -1.043 |
vacc | -1.029 |
Qin | -0.988 |
Token | SAS |
Token ID | 35516 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
commute | +0.644 |
commuting | +0.543 |
notation | +0.528 |
flakes | +0.508 |
aeros | +0.506 |
Neg logit contributions | |
Sov | -0.662 |
� | -0.612 |
Zhu | -0.610 |
vacc | -0.598 |
Qin | -0.577 |
Token | K |
Token ID | 42 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
commute | +0.668 |
commuting | +0.575 |
notation | +0.560 |
flakes | +0.543 |
aeros | +0.541 |
Neg logit contributions | |
Sov | -0.537 |
� | -0.489 |
Zhu | -0.486 |
vacc | -0.478 |
Qin | -0.456 |
Token | ATCH |
Token ID | 11417 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
commute | +0.261 |
commuting | +0.224 |
notation | +0.218 |
flakes | +0.211 |
aeros | +0.210 |
Neg logit contributions | |
Sov | -0.223 |
� | -0.203 |
Zhu | -0.202 |
vacc | -0.199 |
Qin | -0.190 |
Token | EW |
Token ID | 6217 |
Feature activation | -7.34e-01 |
Pos logit contributions | |
commute | +1.604 |
commuting | +1.365 |
notation | +1.333 |
flakes | +1.287 |
aeros | +1.283 |
Neg logit contributions | |
Sov | -1.491 |
� | -1.367 |
Zhu | -1.357 |
vacc | -1.335 |
Qin | -1.281 |
Token | AN |
Token ID | 1565 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
commute | +0.597 |
commuting | +0.508 |
notation | +0.494 |
flakes | +0.477 |
aeros | +0.475 |
Neg logit contributions | |
Sov | -0.554 |
� | -0.509 |
Zhu | -0.505 |
vacc | -0.497 |
Qin | -0.477 |
Token | - |
Token ID | 532 |
Feature activation | -4.37e+00 |
Pos logit contributions | |
commute | +1.454 |
commuting | +1.223 |
notation | +1.187 |
flakes | +1.143 |
aeros | +1.138 |
Neg logit contributions | |
Sov | -1.552 |
� | -1.432 |
Zhu | -1.425 |
vacc | -1.403 |
Qin | -1.350 |
Token | favorite |
Token ID | 4004 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
commute | +3.024 |
commuting | +2.473 |
notation | +2.389 |
flakes | +2.294 |
aeros | +2.271 |
Neg logit contributions | |
Sov | -3.944 |
� | -3.569 |
vacc | -3.531 |
Zhu | -3.523 |
�� | -3.514 |
Token | favorite |
Token ID | 4004 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
commute | +1.996 |
commuting | +1.655 |
notation | +1.605 |
flakes | +1.524 |
aeros | +1.519 |
Neg logit contributions | |
Sov | -2.405 |
� | -2.190 |
Zhu | -2.166 |
vacc | -2.156 |
�� | -2.113 |
Token | favorite |
Token ID | 4004 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
commute | +1.855 |
commuting | +1.540 |
notation | +1.492 |
flakes | +1.420 |
aeros | +1.415 |
Neg logit contributions | |
Sov | -2.212 |
� | -2.017 |
Zhu | -1.995 |
vacc | -1.985 |
�� | -1.941 |
Token | favorite |
Token ID | 4004 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
commute | +1.671 |
commuting | +1.389 |
notation | +1.345 |
flakes | +1.284 |
aeros | +1.277 |
Neg logit contributions | |
Sov | -1.973 |
� | -1.804 |
Zhu | -1.784 |
vacc | -1.772 |
�� | -1.725 |
Token | favorite |
Token ID | 4004 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
commute | +1.452 |
commuting | +1.218 |
notation | +1.184 |
flakes | +1.140 |
aeros | +1.136 |
Neg logit contributions | |
Sov | -1.583 |
� | -1.465 |
Zhu | -1.457 |
vacc | -1.435 |
Qin | -1.382 |
Token | � |
Token ID | 250 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
commute | +0.078 |
commuting | +0.069 |
notation | +0.067 |
flakes | +0.066 |
aeros | +0.066 |
Neg logit contributions | |
Sov | -0.040 |
� | -0.035 |
Zhu | -0.035 |
vacc | -0.034 |
Qin | -0.032 |
Token | shop |
Token ID | 24643 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Sov | +0.556 |
� | +0.513 |
Zhu | +0.501 |
vacc | +0.499 |
Qin | +0.472 |
Neg logit contributions | |
commute | -0.641 |
commuting | -0.552 |
notation | -0.535 |
flakes | -0.521 |
aeros | -0.518 |
Token | � |
Token ID | 447 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
commute | +0.947 |
commuting | +0.800 |
notation | +0.777 |
flakes | +0.749 |
aeros | +0.746 |
Neg logit contributions | |
Sov | -0.967 |
� | -0.886 |
Zhu | -0.883 |
vacc | -0.871 |
Qin | -0.837 |
Token | � |
Token ID | 251 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
commute | +1.694 |
commuting | +1.492 |
notation | +1.463 |
flakes | +1.429 |
aeros | +1.417 |
Neg logit contributions | |
Sov | -0.952 |
Zhu | -0.832 |
� | -0.830 |
vacc | -0.813 |
Qin | -0.769 |
Token | for |
Token ID | 329 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
commute | +0.241 |
commuting | +0.205 |
notation | +0.199 |
flakes | +0.193 |
aeros | +0.192 |
Neg logit contributions | |
Sov | -0.223 |
� | -0.207 |
Zhu | -0.204 |
vacc | -0.201 |
Qin | -0.193 |
Token | free |
Token ID | 1479 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
commute | +0.573 |
commuting | +0.488 |
notation | +0.476 |
flakes | +0.460 |
aeros | +0.458 |
Neg logit contributions | |
Sov | -0.516 |
� | -0.475 |
Zhu | -0.471 |
vacc | -0.463 |
Qin | -0.444 |
Token | , |
Token ID | 11 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
commute | +3.246 |
commuting | +2.731 |
notation | +2.660 |
flakes | +2.577 |
aeros | +2.551 |
Neg logit contributions | |
Sov | -3.576 |
Zhu | -3.203 |
vacc | -3.198 |
� | -3.186 |
�� | -3.074 |
Token | legal |
Token ID | 2742 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
commute | +0.200 |
commuting | +0.168 |
notation | +0.163 |
flakes | +0.157 |
aeros | +0.156 |
Neg logit contributions | |
Sov | -0.219 |
� | -0.205 |
Zhu | -0.203 |
vacc | -0.199 |
Qin | -0.192 |
Token | copies |
Token ID | 9088 |
Feature activation | +7.98e-02 |
Pos logit contributions | |
commute | +0.074 |
commuting | +0.061 |
notation | +0.059 |
flakes | +0.057 |
aeros | +0.057 |
Neg logit contributions | |
Sov | -0.084 |
� | -0.079 |
Zhu | -0.078 |
vacc | -0.076 |
Qin | -0.074 |
Token | of |
Token ID | 286 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
Sov | +0.066 |
� | +0.062 |
Zhu | +0.061 |
vacc | +0.060 |
Qin | +0.058 |
Neg logit contributions | |
commute | -0.059 |
commuting | -0.049 |
notation | -0.048 |
flakes | -0.046 |
aeros | -0.046 |
Token | e |
Token ID | 304 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Sov | +0.432 |
� | +0.396 |
Zhu | +0.394 |
vacc | +0.388 |
Qin | +0.372 |
Neg logit contributions | |
commute | -0.446 |
commuting | -0.379 |
notation | -0.368 |
flakes | -0.356 |
aeros | -0.354 |
Token | : |
Token ID | 25 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
Sov | +1.043 |
� | +0.977 |
Zhu | +0.974 |
vacc | +0.944 |
Qin | +0.927 |
Neg logit contributions | |
commute | -0.947 |
commuting | -0.791 |
notation | -0.773 |
aeros | -0.742 |
flakes | -0.741 |
Token | Sylvia |
Token ID | 45425 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
Sov | +0.285 |
� | +0.264 |
Zhu | +0.263 |
vacc | +0.256 |
Qin | +0.249 |
Neg logit contributions | |
commute | -0.290 |
commuting | -0.245 |
notation | -0.239 |
flakes | -0.230 |
aeros | -0.229 |
Token | 's |
Token ID | 338 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
commute | +0.114 |
commuting | +0.097 |
notation | +0.094 |
flakes | +0.090 |
aeros | +0.090 |
Neg logit contributions | |
Sov | -0.116 |
� | -0.107 |
Zhu | -0.107 |
vacc | -0.105 |
Qin | -0.101 |
Token | En |
Token ID | 2039 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Sov | +0.539 |
� | +0.500 |
Zhu | +0.498 |
vacc | +0.486 |
Qin | +0.472 |
Neg logit contributions | |
commute | -0.535 |
commuting | -0.452 |
notation | -0.439 |
flakes | -0.422 |
aeros | -0.422 |
Token | ch |
Token ID | 354 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
commute | +0.899 |
commuting | +0.763 |
notation | +0.744 |
flakes | +0.718 |
aeros | +0.715 |
Neg logit contributions | |
Sov | -0.855 |
� | -0.783 |
Zhu | -0.775 |
vacc | -0.765 |
Qin | -0.733 |
Token | il |
Token ID | 346 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
commute | +0.957 |
commuting | +0.798 |
notation | +0.776 |
flakes | +0.749 |
aeros | +0.745 |
Neg logit contributions | |
Sov | -1.066 |
� | -0.980 |
Zhu | -0.976 |
vacc | -0.961 |
Qin | -0.928 |
Token | ada |
Token ID | 4763 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
commute | +3.014 |
commuting | +2.500 |
notation | +2.457 |
flakes | +2.384 |
aeros | +2.358 |
Neg logit contributions | |
Sov | -3.667 |
� | -3.362 |
vacc | -3.306 |
Zhu | -3.296 |
�� | -3.185 |
Token | Kitchen |
Token ID | 22541 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
Sov | +0.208 |
� | +0.194 |
Zhu | +0.193 |
vacc | +0.189 |
Qin | +0.183 |
Neg logit contributions | |
commute | -0.191 |
commuting | -0.160 |
notation | -0.155 |
flakes | -0.149 |
aeros | -0.149 |
Token | Previously |
Token ID | 29875 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
Sov | +0.431 |
� | +0.401 |
Zhu | +0.399 |
vacc | +0.391 |
Qin | +0.379 |
Neg logit contributions | |
commute | -0.383 |
commuting | -0.320 |
notation | -0.310 |
flakes | -0.298 |
aeros | -0.297 |
Token | : |
Token ID | 25 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
commute | +0.616 |
commuting | +0.517 |
notation | +0.502 |
flakes | +0.484 |
aeros | +0.482 |
Neg logit contributions | |
Sov | -0.665 |
� | -0.615 |
Zhu | -0.611 |
vacc | -0.601 |
Qin | -0.580 |
Token | Gro |
Token ID | 10299 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
Sov | +0.231 |
� | +0.213 |
Zhu | +0.212 |
vacc | +0.207 |
Qin | +0.201 |
Neg logit contributions | |
commute | -0.240 |
commuting | -0.204 |
notation | -0.198 |
flakes | -0.191 |
aeros | -0.190 |
Token | ) |
Token ID | 8 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
commute | +0.989 |
commuting | +0.838 |
notation | +0.814 |
flakes | +0.785 |
aeros | +0.782 |
Neg logit contributions | |
Sov | -0.969 |
� | -0.890 |
Zhu | -0.884 |
vacc | -0.870 |
Qin | -0.837 |
Token |
|
Token ID | 198 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Sov | +0.135 |
� | +0.127 |
Zhu | +0.125 |
vacc | +0.123 |
Qin | +0.119 |
Neg logit contributions | |
commute | -0.119 |
commuting | -0.100 |
notation | -0.097 |
flakes | -0.094 |
aeros | -0.093 |
Token |
|
Token ID | 198 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
commute | +0.982 |
commuting | +0.789 |
notation | +0.771 |
flakes | +0.732 |
aeros | +0.730 |
Neg logit contributions | |
Sov | -1.458 |
Zhu | -1.351 |
� | -1.344 |
vacc | -1.338 |
�� | -1.314 |
Token | ( |
Token ID | 7 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
commute | +0.683 |
commuting | +0.592 |
notation | +0.578 |
flakes | +0.560 |
aeros | +0.558 |
Neg logit contributions | |
Sov | -0.507 |
� | -0.459 |
Zhu | -0.456 |
vacc | -0.448 |
Qin | -0.427 |
Token | K |
Token ID | 42 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
Sov | +0.516 |
� | +0.470 |
Zhu | +0.462 |
vacc | +0.456 |
Qin | +0.434 |
Neg logit contributions | |
commute | -0.672 |
commuting | -0.581 |
notation | -0.567 |
flakes | -0.551 |
aeros | -0.549 |
Token | ath |
Token ID | 776 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
Sov | +0.490 |
� | +0.446 |
Zhu | +0.441 |
vacc | +0.436 |
Qin | +0.416 |
Neg logit contributions | |
commute | -0.587 |
commuting | -0.505 |
notation | -0.490 |
flakes | -0.478 |
aeros | -0.477 |
Token | leen |
Token ID | 20042 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
commute | +3.147 |
commuting | +2.642 |
notation | +2.555 |
flakes | +2.473 |
aeros | +2.438 |
Neg logit contributions | |
Sov | -3.557 |
� | -3.291 |
Zhu | -3.241 |
vacc | -3.207 |
Qin | -3.063 |
Token | Flynn |
Token ID | 14654 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
commute | +1.057 |
commuting | +0.911 |
notation | +0.889 |
flakes | +0.861 |
aeros | +0.859 |
Neg logit contributions | |
Sov | -0.837 |
� | -0.763 |
Zhu | -0.758 |
vacc | -0.743 |
Qin | -0.711 |
Token | ) |
Token ID | 8 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
Sov | +0.111 |
� | +0.103 |
Zhu | +0.102 |
vacc | +0.100 |
Qin | +0.096 |
Neg logit contributions | |
commute | -0.124 |
commuting | -0.105 |
notation | -0.103 |
flakes | -0.099 |
aeros | -0.099 |
Token |
|
Token ID | 198 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
Sov | +0.530 |
� | +0.495 |
Zhu | +0.490 |
vacc | +0.481 |
Qin | +0.466 |
Neg logit contributions | |
commute | -0.470 |
commuting | -0.394 |
notation | -0.384 |
flakes | -0.371 |
aeros | -0.368 |
Token |
|
Token ID | 198 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
commute | +0.258 |
commuting | +0.210 |
notation | +0.204 |
flakes | +0.195 |
aeros | +0.194 |
Neg logit contributions | |
Sov | -0.357 |
� | -0.331 |
Zhu | -0.330 |
vacc | -0.327 |
Qin | -0.315 |
Token | www |
Token ID | 2503 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
commute | +1.932 |
commuting | +1.629 |
notation | +1.565 |
aster | +1.514 |
minster | +1.511 |
Neg logit contributions | |
Sov | -1.830 |
Zhu | -1.668 |
� | -1.624 |
�� | -1.590 |
vacc | -1.567 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
commute | +2.615 |
commuting | +2.159 |
notation | +2.120 |
flakes | +2.081 |
aster | +2.049 |
Neg logit contributions | |
Sov | -2.890 |
�� | -2.666 |
Zhu | -2.622 |
vacc | -2.598 |
� | -2.563 |
Token | bel |
Token ID | 6667 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
commute | +0.729 |
commuting | +0.606 |
notation | +0.591 |
flakes | +0.566 |
aeros | +0.565 |
Neg logit contributions | |
Sov | -0.856 |
� | -0.796 |
Zhu | -0.788 |
vacc | -0.784 |
Qin | -0.749 |
Token | fast |
Token ID | 7217 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
Sov | +0.432 |
� | +0.398 |
vacc | +0.392 |
Zhu | +0.391 |
Qin | +0.369 |
Neg logit contributions | |
commute | -0.470 |
commuting | -0.400 |
notation | -0.391 |
aeros | -0.376 |
flakes | -0.376 |
Token | te |
Token ID | 660 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
commute | +0.638 |
commuting | +0.545 |
notation | +0.534 |
flakes | +0.515 |
aeros | +0.513 |
Neg logit contributions | |
Sov | -0.580 |
� | -0.543 |
Zhu | -0.532 |
vacc | -0.528 |
Qin | -0.502 |
Token | legraph |
Token ID | 16606 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
commute | +0.884 |
commuting | +0.728 |
notation | +0.706 |
flakes | +0.677 |
aeros | +0.670 |
Neg logit contributions | |
Sov | -1.161 |
� | -1.089 |
Zhu | -1.078 |
vacc | -1.065 |
Qin | -1.029 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
commute | +3.088 |
commuting | +2.546 |
notation | +2.469 |
aster | +2.462 |
flakes | +2.434 |
Neg logit contributions | |
Sov | -3.528 |
�� | -3.260 |
Zhu | -3.196 |
��極 | -3.114 |
� | -3.102 |
Token | co |
Token ID | 1073 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
commute | +0.850 |
commuting | +0.683 |
notation | +0.652 |
flakes | +0.628 |
aeros | +0.620 |
Neg logit contributions | |
Sov | -1.314 |
� | -1.225 |
Zhu | -1.218 |
vacc | -1.194 |
�� | -1.165 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Sov | +0.883 |
� | +0.804 |
Zhu | +0.800 |
vacc | +0.786 |
Qin | +0.753 |
Neg logit contributions | |
commute | -1.028 |
commuting | -0.880 |
notation | -0.859 |
flakes | -0.831 |
aeros | -0.828 |
Token | uk |
Token ID | 2724 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
commute | +0.925 |
commuting | +0.801 |
notation | +0.776 |
flakes | +0.756 |
aeros | +0.752 |
Neg logit contributions | |
Sov | -0.678 |
Zhu | -0.603 |
� | -0.600 |
vacc | -0.583 |
�� | -0.571 |
Token | / |
Token ID | 14 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
Sov | +1.154 |
� | +1.029 |
Zhu | +1.015 |
vacc | +0.989 |
Qin | +0.931 |
Neg logit contributions | |
commute | -2.204 |
commuting | -1.947 |
notation | -1.906 |
flakes | -1.857 |
aeros | -1.851 |
Token | ) |
Token ID | 8 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
commute | +2.003 |
commuting | +1.682 |
notation | +1.637 |
flakes | +1.572 |
aeros | +1.565 |
Neg logit contributions | |
Sov | -2.179 |
� | -1.995 |
Zhu | -1.994 |
vacc | -1.964 |
Qin | -1.894 |
Token | and |
Token ID | 290 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
commute | +0.097 |
commuting | +0.080 |
notation | +0.078 |
flakes | +0.075 |
aeros | +0.075 |
Neg logit contributions | |
Sov | -0.109 |
� | -0.102 |
Zhu | -0.101 |
vacc | -0.099 |
Qin | -0.096 |
Token | Henry |
Token ID | 8616 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
commute | +1.039 |
commuting | +0.883 |
notation | +0.857 |
flakes | +0.827 |
aeros | +0.822 |
Neg logit contributions | |
Sov | -1.026 |
� | -0.936 |
Zhu | -0.930 |
vacc | -0.920 |
Qin | -0.879 |
Token | ( |
Token ID | 357 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
commute | +0.188 |
commuting | +0.155 |
notation | +0.151 |
flakes | +0.144 |
aeros | +0.144 |
Neg logit contributions | |
Sov | -0.233 |
� | -0.217 |
Zhu | -0.215 |
vacc | -0.212 |
Qin | -0.205 |
Token | P |
Token ID | 47 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
commute | +0.399 |
commuting | +0.348 |
notation | +0.341 |
flakes | +0.331 |
aeros | +0.329 |
Neg logit contributions | |
Sov | -0.260 |
� | -0.236 |
Zhu | -0.235 |
vacc | -0.228 |
Qin | -0.218 |
Token | yr |
Token ID | 2417 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
commute | +0.249 |
commuting | +0.213 |
notation | +0.209 |
flakes | +0.202 |
aeros | +0.201 |
Neg logit contributions | |
Sov | -0.209 |
� | -0.191 |
Zhu | -0.190 |
vacc | -0.186 |
Qin | -0.178 |
Token | g |
Token ID | 70 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
commute | +2.620 |
commuting | +2.108 |
notation | +2.054 |
flakes | +1.968 |
aeros | +1.942 |
Neg logit contributions | |
Sov | -3.995 |
� | -3.688 |
Zhu | -3.685 |
vacc | -3.564 |
�� | -3.533 |
Token | os |
Token ID | 418 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
commute | +0.134 |
commuting | +0.108 |
notation | +0.105 |
aeros | +0.100 |
flakes | +0.100 |
Neg logit contributions | |
Sov | -0.196 |
� | -0.182 |
Zhu | -0.182 |
vacc | -0.179 |
Qin | -0.173 |
Token | ) |
Token ID | 8 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
commute | +0.816 |
commuting | +0.703 |
notation | +0.686 |
flakes | +0.665 |
aeros | +0.662 |
Neg logit contributions | |
Sov | -0.652 |
� | -0.594 |
Zhu | -0.589 |
vacc | -0.579 |
Qin | -0.553 |
Token | on |
Token ID | 319 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
Sov | +0.303 |
� | +0.282 |
Zhu | +0.279 |
vacc | +0.275 |
Qin | +0.265 |
Neg logit contributions | |
commute | -0.282 |
commuting | -0.235 |
notation | -0.230 |
flakes | -0.221 |
aeros | -0.220 |
Token | the |
Token ID | 262 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
commute | +0.039 |
commuting | +0.033 |
notation | +0.032 |
flakes | +0.031 |
aeros | +0.031 |
Neg logit contributions | |
Sov | -0.033 |
� | -0.030 |
Zhu | -0.030 |
vacc | -0.030 |
Qin | -0.028 |
Token | ) |
Token ID | 8 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
commute | +1.378 |
commuting | +1.158 |
notation | +1.128 |
flakes | +1.085 |
aeros | +1.081 |
Neg logit contributions | |
Sov | -1.466 |
� | -1.347 |
Zhu | -1.342 |
vacc | -1.322 |
Qin | -1.272 |
Token | shows |
Token ID | 2523 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
commute | +0.505 |
commuting | +0.416 |
notation | +0.399 |
flakes | +0.383 |
aeros | +0.383 |
Neg logit contributions | |
Sov | -0.663 |
� | -0.622 |
Zhu | -0.616 |
vacc | -0.607 |
Qin | -0.589 |
Token | the |
Token ID | 262 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
commute | +1.050 |
commuting | +0.884 |
notation | +0.859 |
flakes | +0.827 |
aeros | +0.824 |
Neg logit contributions | |
Sov | -1.101 |
� | -1.017 |
Zhu | -1.010 |
vacc | -0.994 |
Qin | -0.957 |
Token | three |
Token ID | 1115 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
commute | +0.822 |
commuting | +0.693 |
notation | +0.673 |
flakes | +0.649 |
aeros | +0.646 |
Neg logit contributions | |
Sov | -0.851 |
� | -0.785 |
Zhu | -0.780 |
vacc | -0.768 |
Qin | -0.739 |
Token | available |
Token ID | 1695 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
commute | +1.166 |
commuting | +0.985 |
notation | +0.959 |
flakes | +0.925 |
aeros | +0.920 |
Neg logit contributions | |
Sov | -1.191 |
� | -1.092 |
Zhu | -1.087 |
vacc | -1.074 |
Qin | -1.028 |
Token | settings |
Token ID | 6460 |
Feature activation | -4.23e+00 |
Pos logit contributions | |
commute | +0.839 |
commuting | +0.713 |
notation | +0.694 |
flakes | +0.670 |
aeros | +0.667 |
Neg logit contributions | |
Sov | -0.802 |
� | -0.736 |
Zhu | -0.731 |
vacc | -0.720 |
Qin | -0.691 |
Token | , |
Token ID | 11 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
commute | +2.999 |
commuting | +2.491 |
notation | +2.421 |
flakes | +2.316 |
aeros | +2.309 |
Neg logit contributions | |
Sov | -3.630 |
Zhu | -3.304 |
� | -3.301 |
vacc | -3.289 |
�� | -3.164 |
Token | including |
Token ID | 1390 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
commute | +0.232 |
commuting | +0.195 |
notation | +0.189 |
flakes | +0.182 |
aeros | +0.181 |
Neg logit contributions | |
Sov | -0.248 |
� | -0.230 |
Zhu | -0.228 |
vacc | -0.225 |
Qin | -0.217 |
Token | � |
Token ID | 564 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
Sov | +1.507 |
� | +1.416 |
Zhu | +1.390 |
vacc | +1.343 |
Qin | +1.316 |
Neg logit contributions | |
commute | -1.771 |
commuting | -1.518 |
notation | -1.484 |
flakes | -1.437 |
aeros | -1.426 |
Token | � |
Token ID | 246 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
commute | +1.566 |
commuting | +1.336 |
notation | +1.299 |
flakes | +1.254 |
aeros | +1.252 |
Neg logit contributions | |
Sov | -1.431 |
Zhu | -1.301 |
� | -1.294 |
vacc | -1.281 |
Qin | -1.223 |
Token | sil |
Token ID | 18217 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
commute | +0.526 |
commuting | +0.445 |
notation | +0.432 |
flakes | +0.417 |
aeros | +0.415 |
Neg logit contributions | |
Sov | -0.532 |
� | -0.491 |
Zhu | -0.487 |
vacc | -0.479 |
Qin | -0.461 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
commute | +0.160 |
commuting | +0.136 |
notation | +0.132 |
flakes | +0.127 |
aeros | +0.127 |
Neg logit contributions | |
Sov | -0.155 |
� | -0.144 |
Zhu | -0.143 |
vacc | -0.140 |
Qin | -0.135 |
Token | brain |
Token ID | 3632 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
commute | +0.100 |
commuting | +0.084 |
notation | +0.082 |
flakes | +0.078 |
aeros | +0.078 |
Neg logit contributions | |
Sov | -0.103 |
� | -0.096 |
Zhu | -0.095 |
vacc | -0.093 |
Qin | -0.090 |
Token | areas |
Token ID | 3006 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
commute | +0.189 |
commuting | +0.158 |
notation | +0.154 |
flakes | +0.147 |
aeros | +0.147 |
Neg logit contributions | |
Sov | -0.207 |
� | -0.193 |
Zhu | -0.191 |
vacc | -0.188 |
Qin | -0.181 |
Token | controlling |
Token ID | 12755 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
commute | +0.027 |
commuting | +0.022 |
notation | +0.022 |
flakes | +0.021 |
aeros | +0.021 |
Neg logit contributions | |
Sov | -0.029 |
� | -0.027 |
Zhu | -0.027 |
vacc | -0.026 |
Qin | -0.025 |
Token | these |
Token ID | 777 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Sov | +0.152 |
� | +0.142 |
Zhu | +0.141 |
vacc | +0.138 |
Qin | +0.133 |
Neg logit contributions | |
commute | -0.157 |
commuting | -0.132 |
notation | -0.129 |
flakes | -0.124 |
aeros | -0.124 |
Token | movements |
Token ID | 8650 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
commute | +0.899 |
commuting | +0.747 |
notation | +0.724 |
flakes | +0.696 |
aeros | +0.693 |
Neg logit contributions | |
Sov | -1.064 |
� | -0.988 |
Zhu | -0.982 |
vacc | -0.967 |
Qin | -0.933 |
Token | for |
Token ID | 329 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
commute | +2.823 |
commuting | +2.319 |
notation | +2.253 |
flakes | +2.152 |
aeros | +2.135 |
Neg logit contributions | |
Sov | -3.729 |
Zhu | -3.416 |
� | -3.397 |
vacc | -3.382 |
Qin | -3.261 |
Token | the |
Token ID | 262 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
commute | +0.513 |
commuting | +0.433 |
notation | +0.421 |
flakes | +0.407 |
aeros | +0.405 |
Neg logit contributions | |
Sov | -0.505 |
� | -0.467 |
Zhu | -0.464 |
vacc | -0.455 |
Qin | -0.439 |
Token | first |
Token ID | 717 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
commute | +0.131 |
commuting | +0.111 |
notation | +0.108 |
flakes | +0.105 |
aeros | +0.104 |
Neg logit contributions | |
Sov | -0.126 |
� | -0.117 |
Zhu | -0.116 |
vacc | -0.114 |
Qin | -0.109 |
Token | time |
Token ID | 640 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
commute | +0.995 |
commuting | +0.810 |
notation | +0.782 |
flakes | +0.747 |
aeros | +0.744 |
Neg logit contributions | |
Sov | -1.402 |
� | -1.308 |
Zhu | -1.300 |
vacc | -1.283 |
Qin | -1.242 |
Token | , |
Token ID | 11 |
Feature activation | -9.34e-02 |
Pos logit contributions | |
commute | +0.687 |
commuting | +0.570 |
notation | +0.550 |
flakes | +0.529 |
aeros | +0.526 |
Neg logit contributions | |
Sov | -0.839 |
� | -0.786 |
Zhu | -0.776 |
vacc | -0.765 |
Qin | -0.739 |
Token | oppose |
Token ID | 12856 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
Sov | +0.443 |
� | +0.416 |
Zhu | +0.410 |
vacc | +0.405 |
Qin | +0.390 |
Neg logit contributions | |
commute | -0.386 |
commuting | -0.323 |
notation | -0.315 |
flakes | -0.303 |
aeros | -0.302 |
Token | what |
Token ID | 644 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
commute | +0.571 |
commuting | +0.485 |
notation | +0.472 |
flakes | +0.456 |
aeros | +0.454 |
Neg logit contributions | |
Sov | -0.546 |
� | -0.501 |
Zhu | -0.498 |
vacc | -0.490 |
Qin | -0.471 |
Token | my |
Token ID | 616 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
commute | +0.746 |
commuting | +0.626 |
notation | +0.608 |
flakes | +0.585 |
aeros | +0.583 |
Neg logit contributions | |
Sov | -0.817 |
� | -0.756 |
Zhu | -0.750 |
vacc | -0.739 |
Qin | -0.713 |
Token | president |
Token ID | 1893 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
commute | +1.289 |
commuting | +1.070 |
notation | +1.038 |
flakes | +0.996 |
aeros | +0.992 |
Neg logit contributions | |
Sov | -1.551 |
� | -1.439 |
Zhu | -1.430 |
vacc | -1.409 |
Qin | -1.360 |
Token | stands |
Token ID | 6296 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
commute | +1.250 |
commuting | +1.041 |
notation | +1.008 |
flakes | +0.969 |
aeros | +0.965 |
Neg logit contributions | |
Sov | -1.467 |
� | -1.356 |
Zhu | -1.349 |
vacc | -1.330 |
Qin | -1.282 |
Token | for |
Token ID | 329 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
commute | +1.097 |
commuting | +0.914 |
notation | +0.887 |
flakes | +0.852 |
aeros | +0.848 |
Neg logit contributions | |
Sov | -1.282 |
� | -1.186 |
Zhu | -1.179 |
vacc | -1.163 |
Qin | -1.121 |
Token | , |
Token ID | 11 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
commute | +3.150 |
commuting | +2.658 |
notation | +2.556 |
flakes | +2.462 |
aeros | +2.453 |
Neg logit contributions | |
Sov | -3.359 |
Zhu | -3.057 |
vacc | -2.996 |
� | -2.996 |
Qin | -2.880 |
Token | his |
Token ID | 465 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
commute | +0.915 |
commuting | +0.783 |
notation | +0.763 |
flakes | +0.738 |
aeros | +0.735 |
Neg logit contributions | |
Sov | -0.806 |
� | -0.737 |
Zhu | -0.732 |
vacc | -0.720 |
Qin | -0.690 |
Token | incompetence |
Token ID | 39674 |
Feature activation | -8.89e-01 |
Pos logit contributions | |
commute | +0.409 |
commuting | +0.343 |
notation | +0.334 |
flakes | +0.322 |
aeros | +0.320 |
Neg logit contributions | |
Sov | -0.452 |
� | -0.419 |
Zhu | -0.416 |
vacc | -0.410 |
Qin | -0.395 |
Token | , |
Token ID | 11 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
commute | +0.645 |
commuting | +0.538 |
notation | +0.522 |
flakes | +0.501 |
aeros | +0.500 |
Neg logit contributions | |
Sov | -0.747 |
� | -0.694 |
Zhu | -0.688 |
vacc | -0.678 |
Qin | -0.655 |
Token | potential |
Token ID | 2785 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
Sov | +0.141 |
� | +0.131 |
Zhu | +0.129 |
vacc | +0.127 |
Qin | +0.122 |
Neg logit contributions | |
commute | -0.160 |
commuting | -0.136 |
notation | -0.133 |
flakes | -0.129 |
aeros | -0.128 |
Token | decades |
Token ID | 4647 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Sov | +0.211 |
Zhu | +0.197 |
� | +0.196 |
vacc | +0.191 |
Qin | +0.186 |
Neg logit contributions | |
commute | -0.210 |
commuting | -0.177 |
notation | -0.173 |
flakes | -0.167 |
aeros | -0.164 |
Token | to |
Token ID | 284 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
Sov | +0.847 |
� | +0.793 |
Zhu | +0.788 |
vacc | +0.767 |
Qin | +0.744 |
Neg logit contributions | |
commute | -0.912 |
commuting | -0.771 |
notation | -0.760 |
flakes | -0.736 |
aeros | -0.731 |
Token | come |
Token ID | 1282 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
Sov | +0.360 |
� | +0.335 |
Zhu | +0.332 |
vacc | +0.323 |
Qin | +0.314 |
Neg logit contributions | |
commute | -0.391 |
commuting | -0.339 |
notation | -0.331 |
flakes | -0.317 |
aeros | -0.313 |
Token | back |
Token ID | 736 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
Sov | +0.645 |
� | +0.600 |
Zhu | +0.594 |
vacc | +0.578 |
Qin | +0.563 |
Neg logit contributions | |
commute | -0.645 |
commuting | -0.544 |
notation | -0.533 |
flakes | -0.511 |
aeros | -0.508 |
Token | after |
Token ID | 706 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
Sov | +0.463 |
� | +0.431 |
Zhu | +0.427 |
vacc | +0.416 |
Qin | +0.404 |
Neg logit contributions | |
commute | -0.467 |
commuting | -0.394 |
notation | -0.386 |
flakes | -0.371 |
aeros | -0.368 |
Token | finishing |
Token ID | 12848 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
Sov | +0.109 |
� | +0.100 |
Zhu | +0.100 |
vacc | +0.098 |
Qin | +0.095 |
Neg logit contributions | |
commute | -0.106 |
commuting | -0.089 |
notation | -0.087 |
flakes | -0.084 |
aeros | -0.083 |
Token | operations |
Token ID | 4560 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
Sov | +2.070 |
� | +1.914 |
Zhu | +1.910 |
vacc | +1.836 |
Qin | +1.804 |
Neg logit contributions | |
commute | -2.472 |
commuting | -2.104 |
notation | -2.084 |
flakes | -2.015 |
minster | -1.995 |
Token | ; |
Token ID | 26 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
Sov | +1.386 |
� | +1.299 |
Zhu | +1.287 |
vacc | +1.251 |
Qin | +1.222 |
Neg logit contributions | |
commute | -1.426 |
commuting | -1.201 |
notation | -1.184 |
flakes | -1.133 |
aeros | -1.123 |
Token | we |
Token ID | 356 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
Sov | +1.419 |
Zhu | +1.327 |
� | +1.316 |
vacc | +1.278 |
Qin | +1.251 |
Neg logit contributions | |
commute | -1.499 |
commuting | -1.260 |
notation | -1.240 |
flakes | -1.193 |
aeros | -1.175 |
Token | will |
Token ID | 481 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
Sov | +1.131 |
� | +1.060 |
Zhu | +1.048 |
vacc | +1.029 |
Qin | +1.000 |
Neg logit contributions | |
commute | -1.034 |
commuting | -0.878 |
notation | -0.857 |
flakes | -0.818 |
aeros | -0.810 |
Token | re |
Token ID | 302 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
Sov | +1.120 |
� | +1.048 |
Zhu | +1.036 |
vacc | +1.016 |
Qin | +0.986 |
Neg logit contributions | |
commute | -1.027 |
commuting | -0.877 |
notation | -0.849 |
flakes | -0.815 |
aeros | -0.808 |
Token | ma |
Token ID | 2611 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
commute | +1.476 |
commuting | +1.292 |
notation | +1.271 |
flakes | +1.238 |
aeros | +1.236 |
Neg logit contributions | |
Sov | -0.841 |
Zhu | -0.738 |
� | -0.737 |
vacc | -0.711 |
�� | -0.696 |
Token | v |
Token ID | 410 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
Sov | +0.032 |
� | +0.029 |
Zhu | +0.029 |
vacc | +0.028 |
Qin | +0.027 |
Neg logit contributions | |
commute | -0.033 |
commuting | -0.028 |
notation | -0.027 |
flakes | -0.026 |
aeros | -0.026 |
Token | Goodell |
Token ID | 41648 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
commute | +0.686 |
commuting | +0.588 |
notation | +0.573 |
flakes | +0.555 |
aeros | +0.552 |
Neg logit contributions | |
Sov | -0.593 |
� | -0.540 |
Zhu | -0.535 |
vacc | -0.527 |
Qin | -0.503 |
Token | Original |
Token ID | 13745 |
Feature activation | +5.43e-01 |
Pos logit contributions | |
Sov | +1.593 |
Zhu | +1.485 |
� | +1.478 |
vacc | +1.456 |
Qin | +1.416 |
Neg logit contributions | |
commute | -1.558 |
commuting | -1.311 |
notation | -1.263 |
flakes | -1.228 |
aeros | -1.222 |
Token | Compl |
Token ID | 20011 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
Sov | +0.423 |
� | +0.391 |
Zhu | +0.389 |
vacc | +0.382 |
Qin | +0.369 |
Neg logit contributions | |
commute | -0.425 |
commuting | -0.360 |
notation | -0.348 |
flakes | -0.338 |
aeros | -0.336 |
Token | aint |
Token ID | 2913 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
Sov | +1.965 |
� | +1.757 |
Zhu | +1.741 |
vacc | +1.688 |
Qin | +1.615 |
Neg logit contributions | |
commute | -3.727 |
commuting | -3.297 |
notation | -3.196 |
flakes | -3.151 |
aeros | -3.149 |
Token |
|
Token ID | 198 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Sov | +3.479 |
Zhu | +3.255 |
� | +3.255 |
vacc | +3.154 |
Qin | +3.102 |
Neg logit contributions | |
commute | -3.384 |
commuting | -2.845 |
notation | -2.693 |
flakes | -2.641 |
aeros | -2.636 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
commute | +0.157 |
commuting | +0.128 |
notation | +0.124 |
flakes | +0.118 |
aeros | +0.118 |
Neg logit contributions | |
Sov | -0.221 |
� | -0.205 |
Zhu | -0.204 |
vacc | -0.202 |
�� | -0.196 |
Token | V |
Token ID | 53 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
commute | +0.125 |
commuting | +0.109 |
notation | +0.107 |
flakes | +0.104 |
aeros | +0.104 |
Neg logit contributions | |
Sov | -0.080 |
� | -0.072 |
Zhu | -0.072 |
vacc | -0.070 |
Qin | -0.067 |
Token | il |
Token ID | 346 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Sov | +0.310 |
� | +0.286 |
Zhu | +0.283 |
vacc | +0.281 |
Qin | +0.269 |
Neg logit contributions | |
commute | -0.295 |
commuting | -0.249 |
notation | -0.240 |
aeros | -0.232 |
flakes | -0.231 |
Token | ma |
Token ID | 2611 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
commute | +1.317 |
commuting | +1.154 |
notation | +1.135 |
flakes | +1.106 |
aeros | +1.106 |
Neg logit contributions | |
Sov | -0.742 |
Zhu | -0.654 |
� | -0.651 |
vacc | -0.628 |
�� | -0.611 |
Token | ume |
Token ID | 2454 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
Sov | +0.514 |
� | +0.473 |
Zhu | +0.468 |
vacc | +0.461 |
Qin | +0.441 |
Neg logit contributions | |
commute | -0.576 |
commuting | -0.493 |
notation | -0.480 |
flakes | -0.463 |
aeros | -0.462 |
Token | is |
Token ID | 318 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
commute | +0.020 |
commuting | +0.017 |
notation | +0.016 |
aeros | +0.016 |
flakes | +0.015 |
Neg logit contributions | |
Sov | -0.023 |
� | -0.021 |
Zhu | -0.021 |
vacc | -0.021 |
Qin | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
Sov | +0.180 |
� | +0.167 |
Zhu | +0.166 |
vacc | +0.163 |
Qin | +0.158 |
Neg logit contributions | |
commute | -0.168 |
commuting | -0.140 |
notation | -0.137 |
flakes | -0.131 |
aeros | -0.131 |
Token | finally |
Token ID | 3443 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Sov | +0.266 |
� | +0.249 |
Zhu | +0.249 |
vacc | +0.242 |
Qin | +0.236 |
Neg logit contributions | |
commute | -0.234 |
commuting | -0.194 |
notation | -0.190 |
flakes | -0.182 |
aeros | -0.182 |
Token | , |
Token ID | 11 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
Sov | +1.666 |
� | +1.585 |
Zhu | +1.565 |
vacc | +1.518 |
Qin | +1.474 |
Neg logit contributions | |
commute | -1.487 |
commuting | -1.241 |
notation | -1.211 |
aeros | -1.160 |
flakes | -1.142 |
Token | very |
Token ID | 845 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
commute | +0.278 |
commuting | +0.234 |
notation | +0.228 |
flakes | +0.220 |
aeros | +0.219 |
Neg logit contributions | |
Sov | -0.282 |
� | -0.260 |
Zhu | -0.258 |
vacc | -0.254 |
Qin | -0.245 |
Token | much |
Token ID | 881 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Sov | +2.365 |
Zhu | +2.216 |
� | +2.203 |
vacc | +2.133 |
Qin | +2.081 |
Neg logit contributions | |
commute | -2.321 |
commuting | -1.936 |
notation | -1.884 |
aeros | -1.821 |
minster | -1.808 |
Token | part |
Token ID | 636 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
commute | +0.781 |
commuting | +0.657 |
notation | +0.638 |
flakes | +0.615 |
aeros | +0.612 |
Neg logit contributions | |
Sov | -0.830 |
� | -0.766 |
Zhu | -0.761 |
vacc | -0.750 |
Qin | -0.721 |
Token | of |
Token ID | 286 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
Sov | +0.247 |
� | +0.230 |
Zhu | +0.229 |
vacc | +0.225 |
Qin | +0.218 |
Neg logit contributions | |
commute | -0.216 |
commuting | -0.179 |
notation | -0.176 |
aeros | -0.168 |
flakes | -0.168 |
Token | the |
Token ID | 262 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
Sov | +0.272 |
� | +0.250 |
Zhu | +0.248 |
vacc | +0.243 |
Qin | +0.233 |
Neg logit contributions | |
commute | -0.317 |
commuting | -0.270 |
notation | -0.264 |
flakes | -0.256 |
aeros | -0.255 |
Token | women |
Token ID | 1466 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
commute | +0.029 |
commuting | +0.024 |
notation | +0.024 |
flakes | +0.023 |
aeros | +0.023 |
Neg logit contributions | |
Sov | -0.030 |
� | -0.027 |
Zhu | -0.027 |
vacc | -0.027 |
Qin | -0.026 |
Token | August |
Token ID | 2932 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
commute | +0.511 |
commuting | +0.435 |
notation | +0.427 |
flakes | +0.410 |
aeros | +0.409 |
Neg logit contributions | |
Sov | -0.451 |
Zhu | -0.414 |
� | -0.413 |
vacc | -0.402 |
Qin | -0.390 |
Token | A |
Token ID | 317 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
commute | +0.533 |
commuting | +0.452 |
notation | +0.439 |
aeros | +0.423 |
flakes | +0.423 |
Neg logit contributions | |
Sov | -0.521 |
� | -0.481 |
Zhu | -0.479 |
vacc | -0.469 |
Qin | -0.454 |
Token | . |
Token ID | 13 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
commute | +0.989 |
commuting | +0.845 |
notation | +0.823 |
flakes | +0.796 |
aeros | +0.793 |
Neg logit contributions | |
Sov | -0.877 |
� | -0.803 |
Zhu | -0.798 |
vacc | -0.785 |
Qin | -0.752 |
Token | Bus |
Token ID | 5869 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
Sov | +0.418 |
� | +0.384 |
Zhu | +0.383 |
vacc | +0.375 |
Qin | +0.362 |
Neg logit contributions | |
commute | -0.458 |
commuting | -0.390 |
notation | -0.380 |
flakes | -0.366 |
aeros | -0.365 |
Token | ch |
Token ID | 354 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
Sov | +0.066 |
� | +0.060 |
Zhu | +0.060 |
vacc | +0.059 |
Qin | +0.056 |
Neg logit contributions | |
commute | -0.088 |
commuting | -0.076 |
notation | -0.074 |
flakes | -0.072 |
aeros | -0.072 |
Token | Jr |
Token ID | 7504 |
Feature activation | +3.05e+00 |
Pos logit contributions | |
Sov | +0.251 |
� | +0.233 |
Zhu | +0.231 |
vacc | +0.228 |
Qin | +0.219 |
Neg logit contributions | |
commute | -0.236 |
commuting | -0.198 |
notation | -0.192 |
flakes | -0.185 |
aeros | -0.184 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
Sov | +1.877 |
� | +1.723 |
Zhu | +1.687 |
vacc | +1.671 |
Qin | +1.572 |
Neg logit contributions | |
commute | -2.824 |
commuting | -2.465 |
notation | -2.409 |
aeros | -2.336 |
flakes | -2.332 |
Token | and |
Token ID | 290 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
Sov | +0.804 |
� | +0.748 |
Zhu | +0.743 |
vacc | +0.728 |
Qin | +0.705 |
Neg logit contributions | |
commute | -0.785 |
commuting | -0.662 |
notation | -0.644 |
flakes | -0.622 |
aeros | -0.617 |
Token | Ad |
Token ID | 1215 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
Sov | +0.183 |
Zhu | +0.169 |
� | +0.169 |
vacc | +0.164 |
Qin | +0.160 |
Neg logit contributions | |
commute | -0.195 |
commuting | -0.166 |
notation | -0.162 |
flakes | -0.156 |
aeros | -0.155 |
Token | olph |
Token ID | 10196 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
commute | +0.393 |
commuting | +0.333 |
notation | +0.323 |
flakes | +0.311 |
aeros | +0.311 |
Neg logit contributions | |
Sov | -0.387 |
� | -0.357 |
Zhu | -0.355 |
vacc | -0.350 |
Qin | -0.336 |
Token | us |
Token ID | 385 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
commute | +1.442 |
commuting | +1.234 |
notation | +1.202 |
flakes | +1.162 |
aeros | +1.156 |
Neg logit contributions | |
Sov | -1.293 |
� | -1.177 |
Zhu | -1.165 |
vacc | -1.154 |
Qin | -1.095 |
Token | ention |
Token ID | 1463 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
Sov | +0.177 |
� | +0.167 |
Zhu | +0.164 |
vacc | +0.163 |
Qin | +0.157 |
Neg logit contributions | |
commute | -0.116 |
commuting | -0.093 |
notation | -0.089 |
aeros | -0.085 |
flakes | -0.084 |
Token | : |
Token ID | 25 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
Sov | +0.324 |
� | +0.306 |
Zhu | +0.304 |
vacc | +0.294 |
Qin | +0.288 |
Neg logit contributions | |
commute | -0.285 |
commuting | -0.238 |
notation | -0.228 |
aeros | -0.222 |
flakes | -0.220 |
Token |
|
Token ID | 198 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
Sov | +0.143 |
� | +0.132 |
Zhu | +0.132 |
vacc | +0.129 |
Qin | +0.125 |
Neg logit contributions | |
commute | -0.150 |
commuting | -0.127 |
notation | -0.124 |
flakes | -0.119 |
aeros | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
Sov | +0.490 |
� | +0.458 |
Zhu | +0.454 |
vacc | +0.447 |
Qin | +0.433 |
Neg logit contributions | |
commute | -0.381 |
commuting | -0.315 |
notation | -0.304 |
flakes | -0.291 |
aeros | -0.290 |
Token | 706 |
Token ID | 35402 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
Sov | +0.481 |
� | +0.436 |
Zhu | +0.430 |
vacc | +0.423 |
Qin | +0.403 |
Neg logit contributions | |
commute | -0.638 |
commuting | -0.553 |
notation | -0.539 |
flakes | -0.523 |
aeros | -0.521 |
Token | Main |
Token ID | 8774 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
commute | +0.273 |
commuting | +0.232 |
notation | +0.226 |
flakes | +0.219 |
aeros | +0.218 |
Neg logit contributions | |
Sov | -0.256 |
� | -0.234 |
Zhu | -0.232 |
vacc | -0.230 |
Qin | -0.219 |
Token | Today |
Token ID | 6288 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
Sov | +2.523 |
� | +2.376 |
Zhu | +2.361 |
vacc | +2.291 |
Qin | +2.224 |
Neg logit contributions | |
commute | -2.631 |
commuting | -2.227 |
notation | -2.181 |
flakes | -2.103 |
aeros | -2.090 |
Token | : |
Token ID | 25 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
Sov | +1.592 |
� | +1.496 |
Zhu | +1.484 |
vacc | +1.437 |
Qin | +1.408 |
Neg logit contributions | |
commute | -1.463 |
commuting | -1.233 |
notation | -1.197 |
flakes | -1.144 |
aeros | -1.143 |
Token | The |
Token ID | 383 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Sov | +0.748 |
Zhu | +0.693 |
� | +0.691 |
vacc | +0.669 |
Qin | +0.655 |
Neg logit contributions | |
commute | -0.791 |
commuting | -0.673 |
notation | -0.655 |
flakes | -0.629 |
aeros | -0.628 |
Token | Burger |
Token ID | 28983 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
Sov | +1.132 |
Zhu | +1.052 |
� | +1.049 |
vacc | +1.014 |
Qin | +0.995 |
Neg logit contributions | |
commute | -1.203 |
commuting | -1.026 |
notation | -0.994 |
flakes | -0.956 |
aeros | -0.956 |
Token | Guys |
Token ID | 29899 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
Sov | +1.270 |
� | +1.171 |
Zhu | +1.164 |
vacc | +1.135 |
Qin | +1.099 |
Neg logit contributions | |
commute | -1.377 |
commuting | -1.174 |
notation | -1.141 |
flakes | -1.100 |
aeros | -1.098 |
Token |
|
Token ID | 198 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
Sov | +0.314 |
� | +0.292 |
Zhu | +0.290 |
vacc | +0.286 |
Qin | +0.277 |
Neg logit contributions | |
commute | -0.238 |
commuting | -0.196 |
notation | -0.189 |
flakes | -0.181 |
aeros | -0.180 |
Token | " |
Token ID | 1 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
Sov | +0.427 |
� | +0.386 |
Zhu | +0.382 |
vacc | +0.376 |
Qin | +0.357 |
Neg logit contributions | |
commute | -0.576 |
commuting | -0.499 |
notation | -0.486 |
flakes | -0.472 |
aeros | -0.471 |
Token | Working |
Token ID | 28516 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
Sov | +0.198 |
� | +0.180 |
Zhu | +0.178 |
vacc | +0.175 |
Qin | +0.166 |
Neg logit contributions | |
commute | -0.271 |
commuting | -0.235 |
notation | -0.229 |
flakes | -0.223 |
aeros | -0.222 |
Token | on |
Token ID | 319 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
commute | +0.977 |
commuting | +0.822 |
notation | +0.799 |
flakes | +0.770 |
aeros | +0.767 |
Neg logit contributions | |
Sov | -1.031 |
� | -0.952 |
Zhu | -0.946 |
vacc | -0.931 |
Qin | -0.897 |
Token | Yog |
Token ID | 31604 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
commute | +0.374 |
commuting | +0.322 |
notation | +0.314 |
flakes | +0.304 |
aeros | +0.303 |
Neg logit contributions | |
Sov | -0.310 |
� | -0.282 |
Zhu | -0.279 |
vacc | -0.276 |
Qin | -0.263 |
Token | ventures |
Token ID | 10065 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
Sov | +0.220 |
� | +0.197 |
Zhu | +0.195 |
vacc | +0.191 |
Qin | +0.181 |
Neg logit contributions | |
commute | -0.350 |
commuting | -0.306 |
notation | -0.300 |
flakes | -0.291 |
aeros | -0.290 |
Token | was |
Token ID | 373 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
Sov | +0.984 |
� | +0.913 |
Zhu | +0.906 |
vacc | +0.888 |
Qin | +0.861 |
Neg logit contributions | |
commute | -0.881 |
commuting | -0.736 |
notation | -0.716 |
flakes | -0.688 |
aeros | -0.686 |
Token | an |
Token ID | 281 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Sov | +0.293 |
� | +0.271 |
Zhu | +0.270 |
vacc | +0.264 |
Qin | +0.256 |
Neg logit contributions | |
commute | -0.286 |
commuting | -0.241 |
notation | -0.236 |
flakes | -0.228 |
aeros | -0.226 |
Token | amazing |
Token ID | 4998 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
Sov | +0.906 |
� | +0.848 |
Zhu | +0.843 |
vacc | +0.821 |
Qin | +0.802 |
Neg logit contributions | |
commute | -0.776 |
commuting | -0.650 |
notation | -0.635 |
flakes | -0.610 |
aeros | -0.600 |
Token | experience |
Token ID | 1998 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
commute | +0.387 |
commuting | +0.324 |
notation | +0.315 |
flakes | +0.304 |
aeros | +0.301 |
Neg logit contributions | |
Sov | -0.446 |
� | -0.415 |
Zhu | -0.413 |
vacc | -0.405 |
Qin | -0.392 |
Token | and |
Token ID | 290 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
Sov | +0.449 |
� | +0.421 |
Zhu | +0.417 |
vacc | +0.406 |
Qin | +0.395 |
Neg logit contributions | |
commute | -0.430 |
commuting | -0.360 |
notation | -0.355 |
flakes | -0.343 |
aeros | -0.340 |
Token | uses |
Token ID | 3544 |
Feature activation | +7.38e-01 |
Pos logit contributions | |
commute | +0.314 |
commuting | +0.258 |
notation | +0.250 |
flakes | +0.239 |
aeros | +0.238 |
Neg logit contributions | |
Sov | -0.412 |
� | -0.384 |
Zhu | -0.381 |
vacc | -0.376 |
Qin | -0.364 |
Token | ? |
Token ID | 30 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Sov | +0.554 |
� | +0.517 |
Zhu | +0.507 |
vacc | +0.499 |
Qin | +0.481 |
Neg logit contributions | |
commute | -0.595 |
commuting | -0.507 |
notation | -0.493 |
flakes | -0.476 |
aeros | -0.475 |
Token | How |
Token ID | 1374 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
Sov | +0.811 |
� | +0.750 |
Zhu | +0.743 |
vacc | +0.725 |
Qin | +0.701 |
Neg logit contributions | |
commute | -0.934 |
commuting | -0.800 |
notation | -0.777 |
flakes | -0.752 |
aeros | -0.750 |
Token | about |
Token ID | 546 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
Sov | +0.225 |
� | +0.208 |
Zhu | +0.206 |
vacc | +0.203 |
Qin | +0.195 |
Neg logit contributions | |
commute | -0.227 |
commuting | -0.193 |
notation | -0.187 |
flakes | -0.180 |
aeros | -0.180 |
Token | after |
Token ID | 706 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
commute | +0.409 |
commuting | +0.339 |
notation | +0.329 |
flakes | +0.315 |
aeros | +0.314 |
Neg logit contributions | |
Sov | -0.500 |
� | -0.464 |
Zhu | -0.461 |
vacc | -0.455 |
Qin | -0.439 |
Token | a |
Token ID | 257 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
Sov | +0.225 |
� | +0.207 |
Zhu | +0.206 |
vacc | +0.202 |
Qin | +0.195 |
Neg logit contributions | |
commute | -0.232 |
commuting | -0.197 |
notation | -0.192 |
flakes | -0.185 |
aeros | -0.184 |
Token | year |
Token ID | 614 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
Sov | +0.258 |
� | +0.237 |
Zhu | +0.236 |
vacc | +0.232 |
Qin | +0.223 |
Neg logit contributions | |
commute | -0.258 |
commuting | -0.218 |
notation | -0.212 |
flakes | -0.204 |
aeros | -0.204 |
Token | ? |
Token ID | 30 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
commute | +0.020 |
commuting | +0.017 |
notation | +0.017 |
flakes | +0.016 |
aeros | +0.016 |
Neg logit contributions | |
Sov | -0.019 |
� | -0.017 |
Zhu | -0.017 |
vacc | -0.017 |
Qin | -0.016 |
Token | How |
Token ID | 1374 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Sov | +0.217 |
� | +0.199 |
Zhu | +0.197 |
vacc | +0.193 |
Qin | +0.186 |
Neg logit contributions | |
commute | -0.265 |
commuting | -0.228 |
notation | -0.222 |
flakes | -0.215 |
aeros | -0.215 |
Token | about |
Token ID | 546 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
Sov | +0.097 |
� | +0.090 |
Zhu | +0.089 |
vacc | +0.087 |
Qin | +0.084 |
Neg logit contributions | |
commute | -0.101 |
commuting | -0.086 |
notation | -0.083 |
flakes | -0.080 |
aeros | -0.080 |
Token | when |
Token ID | 618 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
commute | +0.440 |
commuting | +0.365 |
notation | +0.353 |
flakes | +0.339 |
aeros | +0.338 |
Neg logit contributions | |
Sov | -0.535 |
� | -0.496 |
Zhu | -0.493 |
vacc | -0.486 |
Qin | -0.469 |
Token | in |
Token ID | 287 |
Feature activation | +7.07e-01 |
Pos logit contributions | |
Sov | +0.382 |
� | +0.353 |
Zhu | +0.351 |
vacc | +0.345 |
Qin | +0.334 |
Neg logit contributions | |
commute | -0.352 |
commuting | -0.293 |
notation | -0.289 |
flakes | -0.277 |
aeros | -0.276 |
Token | the |
Token ID | 262 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
Sov | +0.495 |
� | +0.453 |
Zhu | +0.449 |
vacc | +0.441 |
Qin | +0.425 |
Neg logit contributions | |
commute | -0.603 |
commuting | -0.516 |
notation | -0.509 |
flakes | -0.493 |
aeros | -0.490 |
Token | Medicare |
Token ID | 13771 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
Sov | +0.387 |
� | +0.358 |
Zhu | +0.355 |
vacc | +0.348 |
Qin | +0.337 |
Neg logit contributions | |
commute | -0.416 |
commuting | -0.354 |
notation | -0.347 |
flakes | -0.335 |
aeros | -0.333 |
Token | Advantage |
Token ID | 45318 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
Sov | +0.055 |
� | +0.051 |
Zhu | +0.051 |
vacc | +0.050 |
Qin | +0.048 |
Neg logit contributions | |
commute | -0.059 |
commuting | -0.051 |
notation | -0.049 |
flakes | -0.048 |
aeros | -0.047 |
Token | business |
Token ID | 1597 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Sov | +0.433 |
� | +0.398 |
Zhu | +0.394 |
vacc | +0.387 |
Qin | +0.374 |
Neg logit contributions | |
commute | -0.520 |
commuting | -0.447 |
notation | -0.436 |
flakes | -0.423 |
aeros | -0.420 |
Token | ; |
Token ID | 26 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
Sov | +1.252 |
� | +1.149 |
Zhu | +1.145 |
vacc | +1.124 |
Qin | +1.088 |
Neg logit contributions | |
commute | -1.287 |
commuting | -1.091 |
notation | -1.064 |
flakes | -1.036 |
aeros | -1.026 |
Token | they |
Token ID | 484 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
Sov | +0.662 |
� | +0.614 |
Zhu | +0.613 |
vacc | +0.603 |
Qin | +0.586 |
Neg logit contributions | |
commute | -0.528 |
commuting | -0.434 |
notation | -0.424 |
flakes | -0.407 |
aeros | -0.403 |
Token | 're |
Token ID | 821 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Sov | +0.699 |
� | +0.650 |
Zhu | +0.641 |
vacc | +0.634 |
Qin | +0.612 |
Neg logit contributions | |
commute | -0.660 |
commuting | -0.566 |
notation | -0.554 |
flakes | -0.529 |
aeros | -0.525 |
Token | not |
Token ID | 407 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
Sov | +0.885 |
� | +0.820 |
Zhu | +0.817 |
vacc | +0.803 |
Qin | +0.782 |
Neg logit contributions | |
commute | -0.687 |
commuting | -0.559 |
notation | -0.549 |
flakes | -0.525 |
aeros | -0.523 |
Token | in |
Token ID | 287 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
Sov | +0.579 |
� | +0.535 |
Zhu | +0.532 |
vacc | +0.523 |
Qin | +0.507 |
Neg logit contributions | |
commute | -0.522 |
commuting | -0.434 |
notation | -0.426 |
flakes | -0.410 |
aeros | -0.407 |
Token | the |
Token ID | 262 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Sov | +0.541 |
� | +0.494 |
Zhu | +0.491 |
vacc | +0.482 |
Qin | +0.464 |
Neg logit contributions | |
commute | -0.651 |
commuting | -0.557 |
notation | -0.546 |
flakes | -0.531 |
aeros | -0.526 |
Token | ci |
Token ID | 979 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
Sov | +0.480 |
� | +0.436 |
Zhu | +0.433 |
vacc | +0.423 |
Qin | +0.409 |
Neg logit contributions | |
commute | -0.532 |
commuting | -0.453 |
notation | -0.442 |
flakes | -0.427 |
aeros | -0.426 |
Token | / |
Token ID | 14 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
Sov | +0.144 |
Zhu | +0.128 |
� | +0.127 |
vacc | +0.124 |
�� | +0.122 |
Neg logit contributions | |
commute | -0.223 |
commuting | -0.194 |
notation | -0.190 |
aeros | -0.185 |
flakes | -0.183 |
Token | 2 |
Token ID | 17 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
Sov | +1.954 |
� | +1.747 |
Zhu | +1.732 |
vacc | +1.711 |
Qin | +1.616 |
Neg logit contributions | |
commute | -3.066 |
commuting | -2.686 |
notation | -2.621 |
flakes | -2.563 |
aeros | -2.541 |
Token | yp |
Token ID | 4464 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Sov | +0.643 |
� | +0.582 |
Zhu | +0.577 |
vacc | +0.568 |
Qin | +0.540 |
Neg logit contributions | |
commute | -0.847 |
commuting | -0.734 |
notation | -0.713 |
flakes | -0.693 |
aeros | -0.690 |
Token | d |
Token ID | 67 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
commute | +1.393 |
commuting | +1.209 |
notation | +1.192 |
flakes | +1.152 |
aeros | +1.152 |
Neg logit contributions | |
Sov | -0.937 |
� | -0.851 |
Zhu | -0.843 |
vacc | -0.824 |
Qin | -0.787 |
Token | G |
Token ID | 38 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Sov | +0.786 |
� | +0.706 |
Zhu | +0.700 |
vacc | +0.686 |
Qin | +0.651 |
Neg logit contributions | |
commute | -1.190 |
commuting | -1.040 |
notation | -1.014 |
flakes | -0.987 |
aeros | -0.983 |
Token | uf |
Token ID | 3046 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
Sov | +0.831 |
� | +0.747 |
Zhu | +0.741 |
vacc | +0.727 |
Qin | +0.690 |
Neg logit contributions | |
commute | -1.229 |
commuting | -1.072 |
notation | -1.044 |
flakes | -1.016 |
aeros | -1.012 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
commute | +0.018 |
commuting | +0.015 |
notation | +0.015 |
flakes | +0.014 |
aeros | +0.014 |
Neg logit contributions | |
Sov | -0.022 |
� | -0.020 |
Zhu | -0.020 |
vacc | -0.020 |
Qin | -0.019 |
Token | A |
Token ID | 32 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
commute | +0.254 |
commuting | +0.219 |
notation | +0.214 |
flakes | +0.207 |
aeros | +0.206 |
Neg logit contributions | |
Sov | -0.200 |
� | -0.182 |
Zhu | -0.181 |
vacc | -0.177 |
Qin | -0.170 |
Token | statistical |
Token ID | 13905 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
Sov | +0.959 |
� | +0.888 |
Zhu | +0.883 |
vacc | +0.871 |
Qin | +0.837 |
Neg logit contributions | |
commute | -0.952 |
commuting | -0.804 |
notation | -0.782 |
flakes | -0.758 |
aeros | -0.753 |
Token | modeling |
Token ID | 21128 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
Sov | +0.246 |
� | +0.228 |
Zhu | +0.226 |
vacc | +0.222 |
Qin | +0.212 |
Neg logit contributions | |
commute | -0.311 |
commuting | -0.268 |
notation | -0.259 |
flakes | -0.254 |
aeros | -0.252 |
Token |
|
Token ID | 198 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
commute | +1.416 |
commuting | +1.122 |
notation | +1.099 |
flakes | +1.043 |
aeros | +1.035 |
Neg logit contributions | |
Sov | -2.268 |
�� | -2.095 |
Zhu | -2.095 |
vacc | -2.088 |
� | -2.086 |
Token | This |
Token ID | 1212 |
Feature activation | -1.61e-02 |
Pos logit contributions | |
commute | +1.603 |
commuting | +1.371 |
notation | +1.353 |
flakes | +1.299 |
aeros | +1.290 |
Neg logit contributions | |
Sov | -1.298 |
Zhu | -1.205 |
� | -1.188 |
vacc | -1.168 |
Qin | -1.130 |
Token | teen |
Token ID | 6036 |
Feature activation | +7.47e-01 |
Pos logit contributions | |
commute | +0.012 |
commuting | +0.010 |
notation | +0.010 |
flakes | +0.009 |
aeros | +0.009 |
Neg logit contributions | |
Sov | -0.013 |
� | -0.012 |
Zhu | -0.012 |
vacc | -0.012 |
Qin | -0.012 |
Token | is |
Token ID | 318 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
Sov | +0.604 |
� | +0.559 |
Zhu | +0.559 |
vacc | +0.549 |
Qin | +0.529 |
Neg logit contributions | |
commute | -0.556 |
commuting | -0.467 |
notation | -0.454 |
flakes | -0.438 |
aeros | -0.435 |
Token | being |
Token ID | 852 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
Sov | +0.265 |
� | +0.246 |
Zhu | +0.244 |
vacc | +0.241 |
Qin | +0.232 |
Neg logit contributions | |
commute | -0.236 |
commuting | -0.197 |
notation | -0.191 |
flakes | -0.184 |
aeros | -0.183 |
Token | ID |
Token ID | 4522 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
Sov | +0.069 |
� | +0.064 |
Zhu | +0.064 |
vacc | +0.063 |
Qin | +0.060 |
Neg logit contributions | |
commute | -0.065 |
commuting | -0.054 |
notation | -0.053 |
flakes | -0.051 |
aeros | -0.051 |
Token | 'd |
Token ID | 1549 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
Sov | +0.540 |
� | +0.491 |
Zhu | +0.487 |
vacc | +0.478 |
Qin | +0.457 |
Neg logit contributions | |
commute | -0.692 |
commuting | -0.597 |
notation | -0.583 |
flakes | -0.564 |
aeros | -0.563 |
Token | because |
Token ID | 780 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
Sov | +0.913 |
� | +0.861 |
Zhu | +0.859 |
vacc | +0.840 |
Qin | +0.817 |
Neg logit contributions | |
commute | -0.726 |
commuting | -0.597 |
notation | -0.579 |
flakes | -0.558 |
aeros | -0.556 |
Token | he |
Token ID | 339 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
commute | +0.196 |
commuting | +0.164 |
notation | +0.160 |
flakes | +0.153 |
aeros | +0.153 |
Neg logit contributions | |
Sov | -0.221 |
� | -0.205 |
Zhu | -0.205 |
vacc | -0.201 |
Qin | -0.195 |
Token | 's |
Token ID | 338 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
commute | +0.168 |
commuting | +0.141 |
notation | +0.137 |
flakes | +0.131 |
aeros | +0.131 |
Neg logit contributions | |
Sov | -0.187 |
� | -0.174 |
Zhu | -0.173 |
vacc | -0.170 |
Qin | -0.164 |
Token | a |
Token ID | 257 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
Sov | +0.106 |
� | +0.098 |
Zhu | +0.097 |
vacc | +0.096 |
Qin | +0.092 |
Neg logit contributions | |
commute | -0.097 |
commuting | -0.082 |
notation | -0.079 |
flakes | -0.076 |
aeros | -0.076 |
Token | men |
Token ID | 3653 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
commute | +0.815 |
commuting | +0.686 |
notation | +0.666 |
flakes | +0.642 |
aeros | +0.640 |
Neg logit contributions | |
Sov | -0.864 |
� | -0.795 |
Zhu | -0.791 |
vacc | -0.779 |
Qin | -0.749 |
Token | , |
Token ID | 11 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
Sov | +0.180 |
� | +0.171 |
Zhu | +0.167 |
vacc | +0.164 |
Qin | +0.160 |
Neg logit contributions | |
commute | -0.154 |
commuting | -0.128 |
notation | -0.123 |
flakes | -0.119 |
aeros | -0.118 |
Token | 18 |
Token ID | 1248 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
commute | +0.615 |
commuting | +0.520 |
notation | +0.506 |
flakes | +0.488 |
aeros | +0.486 |
Neg logit contributions | |
Sov | -0.614 |
� | -0.564 |
Zhu | -0.561 |
vacc | -0.553 |
Qin | -0.530 |
Token | 59 |
Token ID | 3270 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
commute | +1.184 |
commuting | +1.022 |
notation | +1.005 |
flakes | +0.976 |
aeros | +0.972 |
Neg logit contributions | |
Sov | -0.885 |
Zhu | -0.793 |
� | -0.791 |
vacc | -0.788 |
�� | -0.757 |
Token | . |
Token ID | 13 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
commute | +1.055 |
commuting | +0.906 |
notation | +0.882 |
flakes | +0.853 |
aeros | +0.850 |
Neg logit contributions | |
Sov | -0.888 |
� | -0.816 |
Zhu | -0.807 |
vacc | -0.793 |
Qin | -0.759 |
Token |
|
Token ID | 198 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
commute | +0.809 |
commuting | +0.669 |
notation | +0.645 |
flakes | +0.619 |
aeros | +0.616 |
Neg logit contributions | |
Sov | -1.018 |
� | -0.955 |
Zhu | -0.944 |
vacc | -0.931 |
Qin | -0.901 |
Token |
|
Token ID | 198 |
Feature activation | -6.45e-02 |
Pos logit contributions | |
commute | +0.115 |
commuting | +0.094 |
notation | +0.091 |
flakes | +0.087 |
aeros | +0.086 |
Neg logit contributions | |
Sov | -0.159 |
� | -0.147 |
Zhu | -0.147 |
vacc | -0.145 |
Qin | -0.140 |
Token | 11 |
Token ID | 1157 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
commute | +0.057 |
commuting | +0.050 |
notation | +0.048 |
flakes | +0.047 |
aeros | +0.047 |
Neg logit contributions | |
Sov | -0.044 |
� | -0.040 |
Zhu | -0.039 |
vacc | -0.039 |
Qin | -0.037 |
Token | ) |
Token ID | 8 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
commute | +0.543 |
commuting | +0.450 |
notation | +0.442 |
flakes | +0.423 |
aeros | +0.421 |
Neg logit contributions | |
Sov | -0.649 |
Zhu | -0.592 |
� | -0.590 |
vacc | -0.588 |
�� | -0.573 |
Token | When |
Token ID | 1649 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
commute | +0.508 |
commuting | +0.415 |
notation | +0.400 |
flakes | +0.382 |
aeros | +0.381 |
Neg logit contributions | |
Sov | -0.709 |
� | -0.661 |
Zhu | -0.658 |
vacc | -0.648 |
Qin | -0.628 |
Token | Dancing |
Token ID | 31799 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
Sov | +0.486 |
� | +0.450 |
Zhu | +0.447 |
vacc | +0.437 |
Qin | +0.424 |
Neg logit contributions | |
commute | -0.472 |
commuting | -0.397 |
notation | -0.386 |
flakes | -0.372 |
aeros | -0.371 |
Token | � |
Token ID | 447 |
Feature activation | +5.72e-02 |
Pos logit contributions | |
commute | +0.143 |
commuting | +0.120 |
notation | +0.117 |
flakes | +0.112 |
aeros | +0.112 |
Neg logit contributions | |
Sov | -0.161 |
� | -0.150 |
Zhu | -0.148 |
vacc | -0.146 |
Qin | -0.141 |
Token | � |
Token ID | 247 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
Sov | +0.031 |
� | +0.027 |
Zhu | +0.027 |
vacc | +0.026 |
Qin | +0.025 |
Neg logit contributions | |
commute | -0.059 |
commuting | -0.052 |
notation | -0.051 |
flakes | -0.050 |
aeros | -0.049 |
Token | s |
Token ID | 82 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
commute | +0.176 |
commuting | +0.147 |
notation | +0.143 |
flakes | +0.137 |
aeros | +0.137 |
Neg logit contributions | |
Sov | -0.201 |
� | -0.188 |
Zhu | -0.185 |
vacc | -0.183 |
Qin | -0.176 |
Token | important |
Token ID | 1593 |
Feature activation | -9.44e-01 |
Pos logit contributions | |
commute | +0.318 |
commuting | +0.266 |
notation | +0.258 |
flakes | +0.248 |
aeros | +0.247 |
Neg logit contributions | |
Sov | -0.362 |
� | -0.336 |
Zhu | -0.333 |
vacc | -0.328 |
Qin | -0.316 |
Token | to |
Token ID | 284 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
commute | +0.737 |
commuting | +0.623 |
notation | +0.605 |
flakes | +0.585 |
aeros | +0.582 |
Neg logit contributions | |
Sov | -0.742 |
� | -0.685 |
Zhu | -0.679 |
vacc | -0.668 |
Qin | -0.642 |
Token | enable |
Token ID | 7139 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Sov | +0.288 |
� | +0.267 |
Zhu | +0.265 |
vacc | +0.260 |
Qin | +0.251 |
Neg logit contributions | |
commute | -0.265 |
commuting | -0.222 |
notation | -0.216 |
flakes | -0.208 |
aeros | -0.207 |
Token | them |
Token ID | 606 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
commute | +0.095 |
commuting | +0.080 |
notation | +0.078 |
flakes | +0.075 |
aeros | +0.075 |
Neg logit contributions | |
Sov | -0.092 |
� | -0.085 |
Zhu | -0.084 |
vacc | -0.083 |
Qin | -0.080 |
Token | for |
Token ID | 329 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Sov | +0.450 |
� | +0.421 |
Zhu | +0.413 |
vacc | +0.407 |
Qin | +0.393 |
Neg logit contributions | |
commute | -0.415 |
commuting | -0.348 |
notation | -0.338 |
flakes | -0.326 |
aeros | -0.325 |
Token | the |
Token ID | 262 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Sov | +0.129 |
� | +0.119 |
Zhu | +0.118 |
vacc | +0.116 |
Qin | +0.111 |
Neg logit contributions | |
commute | -0.134 |
commuting | -0.114 |
notation | -0.111 |
flakes | -0.108 |
aeros | -0.107 |
Token | future |
Token ID | 2003 |
Feature activation | +9.07e-01 |
Pos logit contributions | |
Sov | +0.183 |
� | +0.170 |
Zhu | +0.169 |
vacc | +0.165 |
Qin | +0.160 |
Neg logit contributions | |
commute | -0.168 |
commuting | -0.141 |
notation | -0.137 |
flakes | -0.133 |
aeros | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | +9.87e-02 |
Pos logit contributions | |
Sov | +0.693 |
� | +0.650 |
Zhu | +0.635 |
vacc | +0.623 |
Qin | +0.602 |
Neg logit contributions | |
commute | -0.714 |
commuting | -0.606 |
notation | -0.592 |
flakes | -0.572 |
aeros | -0.571 |
Token | House |
Token ID | 2097 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
Sov | +0.508 |
� | +0.463 |
Zhu | +0.459 |
vacc | +0.457 |
�� | +0.437 |
Neg logit contributions | |
commute | -0.505 |
commuting | -0.428 |
notation | -0.416 |
flakes | -0.403 |
aeros | -0.400 |
Token | at |
Token ID | 379 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
commute | +0.260 |
commuting | +0.215 |
notation | +0.209 |
flakes | +0.201 |
aeros | +0.199 |
Neg logit contributions | |
Sov | -0.323 |
� | -0.301 |
Zhu | -0.300 |
vacc | -0.295 |
Qin | -0.285 |
Token | the |
Token ID | 262 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
Sov | +0.078 |
� | +0.072 |
Zhu | +0.072 |
vacc | +0.070 |
Qin | +0.068 |
Neg logit contributions | |
commute | -0.087 |
commuting | -0.074 |
notation | -0.072 |
flakes | -0.070 |
aeros | -0.070 |
Token | age |
Token ID | 2479 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
Sov | +0.577 |
� | +0.533 |
Zhu | +0.530 |
vacc | +0.516 |
Qin | +0.499 |
Neg logit contributions | |
commute | -0.652 |
commuting | -0.558 |
notation | -0.542 |
flakes | -0.527 |
aeros | -0.526 |
Token | of |
Token ID | 286 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Sov | +3.600 |
� | +3.410 |
Zhu | +3.387 |
vacc | +3.285 |
Qin | +3.207 |
Neg logit contributions | |
commute | -3.371 |
commuting | -2.822 |
notation | -2.738 |
flakes | -2.663 |
aeros | -2.637 |
Token | just |
Token ID | 655 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
Sov | +0.768 |
� | +0.708 |
Zhu | +0.703 |
vacc | +0.689 |
Qin | +0.663 |
Neg logit contributions | |
commute | -0.865 |
commuting | -0.738 |
notation | -0.719 |
flakes | -0.697 |
aeros | -0.693 |
Token | 25 |
Token ID | 1679 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
commute | +0.509 |
commuting | +0.434 |
notation | +0.422 |
flakes | +0.405 |
aeros | +0.405 |
Neg logit contributions | |
Sov | -0.489 |
� | -0.444 |
Zhu | -0.442 |
vacc | -0.439 |
�� | -0.419 |
Token | , |
Token ID | 11 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
Sov | +1.008 |
� | +0.957 |
Zhu | +0.947 |
vacc | +0.924 |
Qin | +0.893 |
Neg logit contributions | |
commute | -0.858 |
commuting | -0.710 |
notation | -0.691 |
flakes | -0.663 |
aeros | -0.663 |
Token | one |
Token ID | 530 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
Sov | +0.539 |
� | +0.505 |
Zhu | +0.502 |
vacc | +0.488 |
Qin | +0.474 |
Neg logit contributions | |
commute | -0.528 |
commuting | -0.443 |
notation | -0.434 |
flakes | -0.417 |
aeros | -0.417 |
Token | of |
Token ID | 286 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Sov | +0.229 |
� | +0.215 |
Zhu | +0.214 |
vacc | +0.209 |
Qin | +0.203 |
Neg logit contributions | |
commute | -0.208 |
commuting | -0.174 |
notation | -0.169 |
flakes | -0.163 |
aeros | -0.162 |
Token | her |
Token ID | 607 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
Sov | +0.825 |
� | +0.762 |
Zhu | +0.758 |
vacc | +0.734 |
Qin | +0.708 |
Neg logit contributions | |
commute | -1.077 |
commuting | -0.926 |
notation | -0.909 |
flakes | -0.878 |
aeros | -0.877 |
Token | accommodate |
Token ID | 15550 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
commute | +0.284 |
commuting | +0.239 |
notation | +0.232 |
flakes | +0.223 |
aeros | +0.222 |
Neg logit contributions | |
Sov | -0.306 |
� | -0.283 |
Zhu | -0.281 |
vacc | -0.277 |
Qin | -0.267 |
Token | a |
Token ID | 257 |
Feature activation | -8.17e-03 |
Pos logit contributions | |
Sov | +0.306 |
� | +0.283 |
Zhu | +0.281 |
vacc | +0.276 |
Qin | +0.266 |
Neg logit contributions | |
commute | -0.313 |
commuting | -0.265 |
notation | -0.259 |
flakes | -0.250 |
aeros | -0.247 |
Token | brief |
Token ID | 4506 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
commute | +0.006 |
commuting | +0.005 |
notation | +0.005 |
flakes | +0.005 |
aeros | +0.005 |
Neg logit contributions | |
Sov | -0.006 |
� | -0.006 |
Zhu | -0.006 |
vacc | -0.006 |
Qin | -0.006 |
Token | work |
Token ID | 670 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
Sov | +0.094 |
� | +0.087 |
Zhu | +0.086 |
vacc | +0.085 |
Qin | +0.082 |
Neg logit contributions | |
commute | -0.088 |
commuting | -0.074 |
notation | -0.072 |
flakes | -0.070 |
aeros | -0.069 |
Token | stopp |
Token ID | 43804 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
Sov | +0.072 |
� | +0.067 |
Zhu | +0.067 |
vacc | +0.066 |
Qin | +0.063 |
Neg logit contributions | |
commute | -0.055 |
commuting | -0.046 |
notation | -0.044 |
flakes | -0.043 |
aeros | -0.042 |
Token | age |
Token ID | 496 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
commute | +0.040 |
commuting | +0.034 |
notation | +0.033 |
aeros | +0.031 |
flakes | +0.031 |
Neg logit contributions | |
Sov | -0.038 |
� | -0.034 |
Zhu | -0.034 |
vacc | -0.034 |
�� | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
commute | +0.572 |
commuting | +0.478 |
notation | +0.464 |
flakes | +0.446 |
aeros | +0.444 |
Neg logit contributions | |
Sov | -0.647 |
� | -0.599 |
Zhu | -0.595 |
vacc | -0.586 |
Qin | -0.565 |
Token | But |
Token ID | 887 |
Feature activation | -5.19e-01 |
Pos logit contributions | |
commute | +0.114 |
commuting | +0.096 |
notation | +0.093 |
flakes | +0.089 |
aeros | +0.089 |
Neg logit contributions | |
Sov | -0.118 |
� | -0.110 |
Zhu | -0.109 |
vacc | -0.106 |
Qin | -0.103 |
Token | if |
Token ID | 611 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
commute | +0.407 |
commuting | +0.344 |
notation | +0.335 |
flakes | +0.323 |
aeros | +0.321 |
Neg logit contributions | |
Sov | -0.405 |
� | -0.374 |
Zhu | -0.371 |
vacc | -0.365 |
Qin | -0.351 |
Token | MA |
Token ID | 8779 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
Sov | +0.674 |
� | +0.622 |
Zhu | +0.617 |
vacc | +0.604 |
Qin | +0.585 |
Neg logit contributions | |
commute | -0.682 |
commuting | -0.576 |
notation | -0.563 |
flakes | -0.541 |
aeros | -0.533 |
Token | VEN |
Token ID | 28290 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
Sov | +1.079 |
� | +0.967 |
Zhu | +0.960 |
vacc | +0.940 |
Qin | +0.891 |
Neg logit contributions | |
commute | -1.741 |
commuting | -1.523 |
notation | -1.491 |
flakes | -1.449 |
aeros | -1.445 |
Token | , |
Token ID | 11 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
commute | +1.009 |
commuting | +0.836 |
notation | +0.807 |
flakes | +0.777 |
aeros | +0.775 |
Neg logit contributions | |
Sov | -1.230 |
� | -1.144 |
Zhu | -1.136 |
vacc | -1.118 |
Qin | -1.083 |
Token | an |
Token ID | 281 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
Sov | +0.136 |
� | +0.126 |
Zhu | +0.125 |
vacc | +0.123 |
Qin | +0.119 |
Neg logit contributions | |
commute | -0.139 |
commuting | -0.118 |
notation | -0.114 |
flakes | -0.110 |
aeros | -0.110 |
Token | admin |
Token ID | 13169 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Sov | +0.305 |
� | +0.282 |
Zhu | +0.280 |
vacc | +0.275 |
Qin | +0.265 |
Neg logit contributions | |
commute | -0.296 |
commuting | -0.250 |
notation | -0.242 |
flakes | -0.234 |
aeros | -0.232 |
Token | must |
Token ID | 1276 |
Feature activation | -4.27e-03 |
Pos logit contributions | |
commute | +0.870 |
commuting | +0.722 |
notation | +0.699 |
flakes | +0.672 |
aeros | +0.669 |
Neg logit contributions | |
Sov | -1.046 |
� | -0.970 |
Zhu | -0.964 |
vacc | -0.950 |
Qin | -0.918 |
Token | be |
Token ID | 307 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
commute | +0.003 |
commuting | +0.003 |
notation | +0.002 |
flakes | +0.002 |
aeros | +0.002 |
Neg logit contributions | |
Sov | -0.004 |
� | -0.003 |
Zhu | -0.003 |
vacc | -0.003 |
Qin | -0.003 |
Token | placed |
Token ID | 4624 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
commute | +0.051 |
commuting | +0.043 |
notation | +0.042 |
flakes | +0.040 |
aeros | +0.040 |
Neg logit contributions | |
Sov | -0.058 |
� | -0.054 |
Zhu | -0.054 |
vacc | -0.053 |
Qin | -0.051 |
Token | behind |
Token ID | 2157 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
commute | +0.223 |
commuting | +0.186 |
notation | +0.181 |
flakes | +0.175 |
aeros | +0.174 |
Neg logit contributions | |
Sov | -0.251 |
� | -0.235 |
Zhu | -0.232 |
vacc | -0.228 |
Qin | -0.221 |
Token | every |
Token ID | 790 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
Sov | +0.413 |
� | +0.379 |
Zhu | +0.375 |
vacc | +0.367 |
Qin | +0.354 |
Neg logit contributions | |
commute | -0.494 |
commuting | -0.424 |
notation | -0.411 |
flakes | -0.399 |
aeros | -0.397 |
Token | team |
Token ID | 1074 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
commute | +0.393 |
commuting | +0.332 |
notation | +0.323 |
flakes | +0.312 |
aeros | +0.310 |
Neg logit contributions | |
Sov | -0.399 |
� | -0.367 |
Zhu | -0.365 |
vacc | -0.359 |
Qin | -0.345 |
Token | for |
Token ID | 329 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
commute | +1.056 |
commuting | +0.885 |
notation | +0.860 |
flakes | +0.826 |
aeros | +0.823 |
Neg logit contributions | |
Sov | -1.169 |
� | -1.079 |
Zhu | -1.073 |
vacc | -1.057 |
Qin | -1.018 |
Token | the |
Token ID | 262 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Sov | +0.457 |
� | +0.424 |
Zhu | +0.422 |
vacc | +0.412 |
Qin | +0.399 |
Neg logit contributions | |
commute | -0.493 |
commuting | -0.417 |
notation | -0.407 |
flakes | -0.396 |
aeros | -0.392 |
Token | as |
Token ID | 355 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Sov | +0.594 |
� | +0.549 |
Zhu | +0.547 |
vacc | +0.536 |
Qin | +0.518 |
Neg logit contributions | |
commute | -0.566 |
commuting | -0.477 |
notation | -0.461 |
flakes | -0.446 |
aeros | -0.445 |
Token | it |
Token ID | 340 |
Feature activation | +7.99e-01 |
Pos logit contributions | |
Sov | +0.880 |
� | +0.814 |
Zhu | +0.809 |
vacc | +0.792 |
Qin | +0.766 |
Neg logit contributions | |
commute | -0.878 |
commuting | -0.743 |
notation | -0.719 |
flakes | -0.695 |
aeros | -0.694 |
Token | can |
Token ID | 460 |
Feature activation | +4.65e-03 |
Pos logit contributions | |
Sov | +0.670 |
� | +0.624 |
Zhu | +0.619 |
vacc | +0.608 |
Qin | +0.590 |
Neg logit contributions | |
commute | -0.575 |
commuting | -0.482 |
notation | -0.465 |
aeros | -0.447 |
flakes | -0.446 |
Token | be |
Token ID | 307 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
Sov | +0.004 |
� | +0.004 |
Zhu | +0.004 |
vacc | +0.004 |
Qin | +0.004 |
Neg logit contributions | |
commute | -0.003 |
commuting | -0.003 |
notation | -0.003 |
flakes | -0.002 |
aeros | -0.002 |
Token | mult |
Token ID | 1963 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
commute | +0.178 |
commuting | +0.149 |
notation | +0.145 |
flakes | +0.140 |
aeros | +0.139 |
Neg logit contributions | |
Sov | -0.191 |
� | -0.176 |
Zhu | -0.175 |
vacc | -0.173 |
Qin | -0.166 |
Token | il |
Token ID | 346 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
Sov | +0.633 |
� | +0.586 |
Zhu | +0.581 |
vacc | +0.573 |
Qin | +0.550 |
Neg logit contributions | |
commute | -0.620 |
commuting | -0.525 |
notation | -0.508 |
flakes | -0.490 |
aeros | -0.490 |
Token | ine |
Token ID | 500 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
commute | +1.948 |
commuting | +1.586 |
notation | +1.552 |
flakes | +1.472 |
aeros | +1.453 |
Neg logit contributions | |
Sov | -2.645 |
� | -2.449 |
Zhu | -2.439 |
vacc | -2.406 |
Qin | -2.333 |
Token | $ |
Token ID | 720 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
commute | +1.343 |
commuting | +1.127 |
notation | +1.106 |
flakes | +1.061 |
aeros | +1.051 |
Neg logit contributions | |
Sov | -1.412 |
� | -1.293 |
Zhu | -1.280 |
vacc | -1.273 |
Qin | -1.214 |
Token | art |
Token ID | 433 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
Sov | +1.004 |
� | +0.913 |
Zhu | +0.909 |
vacc | +0.894 |
Qin | +0.851 |
Neg logit contributions | |
commute | -1.231 |
commuting | -1.061 |
notation | -1.022 |
aeros | -0.999 |
flakes | -0.995 |
Token | ext |
Token ID | 2302 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
commute | +0.213 |
commuting | +0.181 |
notation | +0.174 |
flakes | +0.170 |
aeros | +0.169 |
Neg logit contributions | |
Sov | -0.210 |
� | -0.193 |
Zhu | -0.192 |
vacc | -0.189 |
Qin | -0.182 |
Token | = |
Token ID | 796 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
commute | +0.891 |
commuting | +0.750 |
notation | +0.728 |
flakes | +0.702 |
aeros | +0.699 |
Neg logit contributions | |
Sov | -0.940 |
� | -0.867 |
Zhu | -0.862 |
vacc | -0.849 |
Qin | -0.817 |
Token | who |
Token ID | 508 |
Feature activation | -9.00e-02 |
Pos logit contributions | |
commute | +0.473 |
commuting | +0.392 |
notation | +0.381 |
flakes | +0.366 |
aeros | +0.365 |
Neg logit contributions | |
Sov | -0.551 |
� | -0.511 |
Zhu | -0.509 |
vacc | -0.501 |
Qin | -0.484 |
Token | paid |
Token ID | 3432 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
commute | +0.065 |
commuting | +0.055 |
notation | +0.053 |
flakes | +0.051 |
aeros | +0.051 |
Neg logit contributions | |
Sov | -0.075 |
� | -0.070 |
Zhu | -0.069 |
vacc | -0.068 |
Qin | -0.066 |
Token | extra |
Token ID | 3131 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Sov | +0.178 |
Zhu | +0.164 |
� | +0.164 |
vacc | +0.160 |
Qin | +0.155 |
Neg logit contributions | |
commute | -0.174 |
commuting | -0.146 |
notation | -0.142 |
flakes | -0.137 |
aeros | -0.137 |
Token | for |
Token ID | 329 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
commute | +1.132 |
commuting | +0.950 |
notation | +0.921 |
flakes | +0.889 |
aeros | +0.887 |
Neg logit contributions | |
Sov | -1.203 |
� | -1.111 |
Zhu | -1.110 |
vacc | -1.087 |
Qin | -1.052 |
Token | other |
Token ID | 584 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
commute | +0.498 |
commuting | +0.428 |
notation | +0.418 |
flakes | +0.405 |
aeros | +0.403 |
Neg logit contributions | |
Sov | -0.402 |
� | -0.366 |
Zhu | -0.364 |
vacc | -0.357 |
Qin | -0.342 |
Token | rewards |
Token ID | 11530 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
Sov | +0.839 |
� | +0.774 |
Zhu | +0.771 |
vacc | +0.755 |
Qin | +0.729 |
Neg logit contributions | |
commute | -0.854 |
commuting | -0.722 |
notation | -0.705 |
flakes | -0.679 |
aeros | -0.677 |
Token | on |
Token ID | 319 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
commute | +1.995 |
commuting | +1.667 |
notation | +1.617 |
flakes | +1.555 |
aeros | +1.546 |
Neg logit contributions | |
Sov | -2.296 |
� | -2.123 |
Zhu | -2.104 |
vacc | -2.083 |
Qin | -2.001 |
Token | top |
Token ID | 1353 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
commute | +0.175 |
commuting | +0.152 |
notation | +0.149 |
flakes | +0.144 |
aeros | +0.144 |
Neg logit contributions | |
Sov | -0.125 |
� | -0.113 |
Zhu | -0.112 |
vacc | -0.110 |
Qin | -0.105 |
Token | of |
Token ID | 286 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
Sov | +0.505 |
� | +0.469 |
Zhu | +0.468 |
vacc | +0.458 |
Qin | +0.443 |
Neg logit contributions | |
commute | -0.460 |
commuting | -0.386 |
notation | -0.373 |
flakes | -0.359 |
aeros | -0.358 |
Token | a |
Token ID | 257 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
Sov | +0.286 |
� | +0.259 |
Zhu | +0.258 |
vacc | +0.252 |
Qin | +0.241 |
Neg logit contributions | |
commute | -0.399 |
commuting | -0.345 |
notation | -0.338 |
flakes | -0.328 |
aeros | -0.327 |
Token | copy |
Token ID | 4866 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
Sov | +0.636 |
Zhu | +0.592 |
� | +0.589 |
vacc | +0.576 |
Qin | +0.559 |
Neg logit contributions | |
commute | -0.600 |
commuting | -0.506 |
notation | -0.489 |
flakes | -0.477 |
aeros | -0.472 |
Token | There |
Token ID | 1318 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
commute | +0.257 |
commuting | +0.218 |
notation | +0.212 |
flakes | +0.204 |
aeros | +0.204 |
Neg logit contributions | |
Sov | -0.248 |
� | -0.229 |
Zhu | -0.227 |
vacc | -0.222 |
Qin | -0.215 |
Token | are |
Token ID | 389 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
commute | +0.112 |
commuting | +0.092 |
notation | +0.089 |
flakes | +0.086 |
aeros | +0.085 |
Neg logit contributions | |
Sov | -0.139 |
� | -0.130 |
Zhu | -0.129 |
vacc | -0.127 |
Qin | -0.122 |
Token | knit |
Token ID | 30495 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
Sov | +0.275 |
� | +0.256 |
Zhu | +0.254 |
vacc | +0.249 |
Qin | +0.241 |
Neg logit contributions | |
commute | -0.245 |
commuting | -0.204 |
notation | -0.198 |
flakes | -0.191 |
aeros | -0.190 |
Token | - |
Token ID | 12 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Sov | +0.551 |
� | +0.504 |
Zhu | +0.495 |
vacc | +0.489 |
Qin | +0.464 |
Neg logit contributions | |
commute | -0.714 |
commuting | -0.614 |
notation | -0.597 |
flakes | -0.581 |
aeros | -0.581 |
Token | a |
Token ID | 64 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
Sov | +0.916 |
� | +0.843 |
vacc | +0.829 |
Zhu | +0.829 |
Qin | +0.781 |
Neg logit contributions | |
commute | -1.073 |
commuting | -0.916 |
notation | -0.886 |
flakes | -0.866 |
aeros | -0.865 |
Token | ... |
Token ID | 986 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
Sov | +0.228 |
� | +0.213 |
Zhu | +0.209 |
vacc | +0.206 |
Qin | +0.198 |
Neg logit contributions | |
commute | -0.225 |
commuting | -0.190 |
notation | -0.184 |
aeros | -0.179 |
flakes | -0.178 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.93e-02 |
Pos logit contributions | |
commute | +1.761 |
commuting | +1.451 |
notation | +1.405 |
flakes | +1.345 |
aeros | +1.338 |
Neg logit contributions | |
Sov | -2.271 |
� | -2.113 |
Zhu | -2.100 |
vacc | -2.072 |
Qin | -2.001 |
Token | Buy |
Token ID | 14518 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
commute | +0.078 |
commuting | +0.067 |
notation | +0.066 |
flakes | +0.064 |
aeros | +0.064 |
Neg logit contributions | |
Sov | -0.062 |
� | -0.056 |
Zhu | -0.056 |
vacc | -0.055 |
Qin | -0.052 |
Token | Photo |
Token ID | 5555 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
commute | +0.630 |
commuting | +0.528 |
notation | +0.512 |
flakes | +0.493 |
aeros | +0.491 |
Neg logit contributions | |
Sov | -0.699 |
� | -0.645 |
Zhu | -0.640 |
vacc | -0.632 |
Qin | -0.608 |
Token | F |
Token ID | 376 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
Sov | +0.917 |
� | +0.859 |
Zhu | +0.857 |
vacc | +0.829 |
Qin | +0.811 |
Neg logit contributions | |
commute | -0.934 |
commuting | -0.792 |
notation | -0.770 |
flakes | -0.740 |
aeros | -0.739 |
Token | SU |
Token ID | 12564 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
commute | +0.040 |
commuting | +0.034 |
notation | +0.033 |
flakes | +0.032 |
aeros | +0.032 |
Neg logit contributions | |
Sov | -0.040 |
� | -0.037 |
Zhu | -0.036 |
vacc | -0.036 |
Qin | -0.034 |
Token | this |
Token ID | 428 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
Sov | +0.078 |
� | +0.071 |
Zhu | +0.071 |
vacc | +0.070 |
Qin | +0.067 |
Neg logit contributions | |
commute | -0.086 |
commuting | -0.073 |
notation | -0.071 |
flakes | -0.069 |
aeros | -0.069 |
Token | visualization |
Token ID | 32704 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
Sov | +0.142 |
� | +0.132 |
Zhu | +0.131 |
vacc | +0.129 |
Qin | +0.124 |
Neg logit contributions | |
commute | -0.144 |
commuting | -0.122 |
notation | -0.119 |
flakes | -0.115 |
aeros | -0.114 |
Token | , |
Token ID | 11 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
commute | +0.078 |
commuting | +0.065 |
notation | +0.063 |
flakes | +0.061 |
aeros | +0.061 |
Neg logit contributions | |
Sov | -0.092 |
� | -0.086 |
Zhu | -0.085 |
vacc | -0.084 |
Qin | -0.081 |
Token | you |
Token ID | 345 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
commute | +0.942 |
commuting | +0.802 |
notation | +0.781 |
flakes | +0.754 |
aeros | +0.751 |
Neg logit contributions | |
Sov | -0.873 |
� | -0.802 |
Zhu | -0.796 |
vacc | -0.783 |
Qin | -0.752 |
Token | can |
Token ID | 460 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
commute | +0.387 |
commuting | +0.321 |
notation | +0.311 |
flakes | +0.298 |
aeros | +0.296 |
Neg logit contributions | |
Sov | -0.486 |
� | -0.453 |
Zhu | -0.450 |
vacc | -0.444 |
Qin | -0.428 |
Token | find |
Token ID | 1064 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
commute | +0.797 |
commuting | +0.662 |
notation | +0.639 |
flakes | +0.615 |
aeros | +0.611 |
Neg logit contributions | |
Sov | -1.003 |
� | -0.938 |
Zhu | -0.929 |
vacc | -0.917 |
Qin | -0.886 |
Token | a |
Token ID | 257 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
commute | +2.044 |
commuting | +1.738 |
notation | +1.696 |
flakes | +1.629 |
aeros | +1.622 |
Neg logit contributions | |
Sov | -1.955 |
� | -1.777 |
Zhu | -1.765 |
vacc | -1.749 |
Qin | -1.666 |
Token | static |
Token ID | 9037 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
commute | +1.598 |
commuting | +1.354 |
notation | +1.323 |
flakes | +1.267 |
aeros | +1.265 |
Neg logit contributions | |
Sov | -1.545 |
� | -1.412 |
Zhu | -1.404 |
vacc | -1.384 |
Qin | -1.327 |
Token | version |
Token ID | 2196 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
commute | +0.017 |
commuting | +0.014 |
notation | +0.014 |
flakes | +0.013 |
aeros | +0.013 |
Neg logit contributions | |
Sov | -0.016 |
� | -0.015 |
Zhu | -0.015 |
vacc | -0.015 |
Qin | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
commute | +0.843 |
commuting | +0.718 |
notation | +0.700 |
flakes | +0.675 |
aeros | +0.673 |
Neg logit contributions | |
Sov | -0.780 |
� | -0.712 |
Zhu | -0.709 |
vacc | -0.698 |
Qin | -0.669 |
Token | it |
Token ID | 340 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
Sov | +0.266 |
� | +0.245 |
Zhu | +0.243 |
vacc | +0.238 |
Qin | +0.229 |
Neg logit contributions | |
commute | -0.298 |
commuting | -0.254 |
notation | -0.248 |
flakes | -0.240 |
aeros | -0.239 |
Token | make |
Token ID | 787 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
Sov | +0.712 |
� | +0.664 |
Zhu | +0.660 |
vacc | +0.648 |
Qin | +0.628 |
Neg logit contributions | |
commute | -0.581 |
commuting | -0.483 |
notation | -0.467 |
flakes | -0.449 |
aeros | -0.446 |
Token | sure |
Token ID | 1654 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
Sov | +0.578 |
� | +0.538 |
Zhu | +0.534 |
vacc | +0.521 |
Qin | +0.507 |
Neg logit contributions | |
commute | -0.559 |
commuting | -0.470 |
notation | -0.457 |
flakes | -0.442 |
aeros | -0.440 |
Token | he |
Token ID | 339 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
commute | +0.744 |
commuting | +0.632 |
notation | +0.615 |
flakes | +0.594 |
aeros | +0.592 |
Neg logit contributions | |
Sov | -0.713 |
� | -0.652 |
Zhu | -0.647 |
vacc | -0.641 |
Qin | -0.611 |
Token | knew |
Token ID | 2993 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Sov | +0.446 |
� | +0.417 |
Zhu | +0.410 |
vacc | +0.404 |
Qin | +0.389 |
Neg logit contributions | |
commute | -0.418 |
commuting | -0.352 |
notation | -0.341 |
flakes | -0.328 |
aeros | -0.327 |
Token | the |
Token ID | 262 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
Sov | +1.006 |
� | +0.935 |
Zhu | +0.929 |
vacc | +0.903 |
Qin | +0.877 |
Neg logit contributions | |
commute | -1.075 |
commuting | -0.913 |
notation | -0.886 |
flakes | -0.857 |
aeros | -0.854 |
Token | rules |
Token ID | 3173 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
commute | +0.197 |
commuting | +0.166 |
notation | +0.160 |
flakes | +0.155 |
aeros | +0.154 |
Neg logit contributions | |
Sov | -0.214 |
� | -0.199 |
Zhu | -0.198 |
vacc | -0.193 |
Qin | -0.187 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
commute | +2.394 |
commuting | +2.012 |
notation | +1.970 |
flakes | +1.882 |
aeros | +1.869 |
Neg logit contributions | |
Sov | -2.581 |
Zhu | -2.354 |
� | -2.342 |
vacc | -2.331 |
�� | -2.225 |
Token | He |
Token ID | 679 |
Feature activation | +8.97e-02 |
Pos logit contributions | |
commute | +0.121 |
commuting | +0.103 |
notation | +0.101 |
flakes | +0.097 |
aeros | +0.097 |
Neg logit contributions | |
Sov | -0.112 |
� | -0.103 |
Zhu | -0.102 |
vacc | -0.101 |
Qin | -0.096 |
Token | hadn |
Token ID | 8020 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Sov | +0.071 |
� | +0.066 |
Zhu | +0.065 |
vacc | +0.064 |
Qin | +0.062 |
Neg logit contributions | |
commute | -0.069 |
commuting | -0.058 |
notation | -0.057 |
flakes | -0.055 |
aeros | -0.054 |
Token | � |
Token ID | 447 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
Sov | +0.746 |
� | +0.682 |
Zhu | +0.678 |
vacc | +0.667 |
Qin | +0.639 |
Neg logit contributions | |
commute | -0.839 |
commuting | -0.717 |
notation | -0.699 |
flakes | -0.675 |
aeros | -0.673 |
Token | � |
Token ID | 247 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
Sov | +0.181 |
Zhu | +0.158 |
� | +0.156 |
vacc | +0.154 |
Qin | +0.145 |
Neg logit contributions | |
commute | -0.327 |
commuting | -0.287 |
notation | -0.283 |
flakes | -0.276 |
aeros | -0.273 |