Token | and |
Token ID | 290 |
Feature activation | +7.08e-01 |
Pos logit contributions | |
BELOW | +2.803 |
idated | +2.676 |
productive | +2.620 |
arrass | +2.436 |
ridge | +2.421 |
Neg logit contributions | |
tti | -2.857 |
neb | -2.507 |
etz | -2.500 |
Probe | -2.477 |
mysteries | -2.392 |
Token | click |
Token ID | 3904 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
BELOW | +4.493 |
idated | +4.179 |
productive | +4.072 |
arrass | +3.842 |
ridge | +3.813 |
Neg logit contributions | |
tti | -4.686 |
etz | -4.119 |
neb | -4.107 |
Probe | -4.021 |
earchers | -3.944 |
Token | like |
Token ID | 588 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
BELOW | +7.403 |
idated | +6.659 |
productive | +6.465 |
ridge | +6.105 |
arrass | +6.090 |
Neg logit contributions | |
tti | -7.504 |
neb | -6.624 |
etz | -6.447 |
Probe | -6.413 |
mysteries | -6.338 |
Token | on |
Token ID | 319 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
BELOW | +9.249 |
idated | +8.322 |
productive | +8.134 |
below | +7.715 |
ridge | +7.525 |
Neg logit contributions | |
tti | -9.403 |
neb | -8.211 |
etz | -7.922 |
Probe | -7.910 |
earchers | -7.856 |
Token | one |
Token ID | 530 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
BELOW | +8.002 |
idated | +7.320 |
productive | +7.205 |
ridge | +6.734 |
arrass | +6.650 |
Neg logit contributions | |
tti | -8.032 |
neb | -7.083 |
etz | -6.899 |
earchers | -6.833 |
Probe | -6.828 |
Token | of |
Token ID | 286 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
BELOW | +0.683 |
idated | +0.667 |
productive | +0.661 |
arrass | +0.611 |
ridge | +0.596 |
Neg logit contributions | |
tti | -0.699 |
neb | -0.614 |
etz | -0.612 |
Probe | -0.605 |
mysteries | -0.588 |
Token | many |
Token ID | 867 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
BELOW | +12.718 |
idated | +11.167 |
productive | +10.451 |
ridge | +10.333 |
below | +10.065 |
Neg logit contributions | |
tti | -17.219 |
earchers | -15.514 |
neb | -15.261 |
etz | -14.609 |
emis | -14.525 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
BELOW | +8.948 |
idated | +8.118 |
productive | +7.969 |
below | +7.396 |
arrass | +7.274 |
Neg logit contributions | |
tti | -7.972 |
etz | -7.026 |
Probe | -6.855 |
neb | -6.785 |
earchers | -6.775 |
Token | pages |
Token ID | 5468 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
BELOW | +3.097 |
idated | +2.990 |
productive | +2.932 |
arrass | +2.728 |
ridge | +2.709 |
Neg logit contributions | |
tti | -2.700 |
etz | -2.323 |
neb | -2.290 |
Probe | -2.284 |
earchers | -2.166 |
Token | . |
Token ID | 13 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
BELOW | +6.436 |
idated | +5.912 |
productive | +5.868 |
ridge | +5.335 |
arrass | +5.309 |
Neg logit contributions | |
tti | -6.087 |
neb | -5.340 |
Probe | -5.313 |
etz | -5.205 |
earchers | -5.149 |
Token | Join |
Token ID | 15251 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
BELOW | +3.993 |
idated | +3.796 |
productive | +3.754 |
arrass | +3.464 |
ridge | +3.447 |
Neg logit contributions | |
tti | -4.220 |
neb | -3.800 |
etz | -3.727 |
earchers | -3.616 |
Probe | -3.608 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
BELOW | +3.727 |
idated | +3.589 |
productive | +3.518 |
arrass | +3.204 |
ridge | +3.167 |
Neg logit contributions | |
tti | -4.838 |
etz | -4.325 |
neb | -4.299 |
Probe | -4.256 |
mysteries | -4.151 |
Token |
|
Token ID | 198 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
BELOW | +0.790 |
idated | +0.750 |
productive | +0.746 |
ridge | +0.667 |
arrass | +0.658 |
Neg logit contributions | |
tti | -1.108 |
neb | -1.010 |
etz | -0.998 |
Probe | -0.990 |
earchers | -0.989 |
Token | Video |
Token ID | 10798 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
tti | +3.466 |
etz | +2.987 |
neb | +2.913 |
Probe | +2.887 |
Television | +2.845 |
Neg logit contributions | |
BELOW | -3.808 |
idated | -3.737 |
productive | -3.663 |
arrass | -3.436 |
ridge | -3.326 |
Token | games |
Token ID | 1830 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
BELOW | +9.001 |
productive | +8.134 |
idated | +7.999 |
ridge | +7.771 |
arrass | +7.515 |
Neg logit contributions | |
tti | -8.750 |
etz | -7.601 |
earchers | -7.412 |
Probe | -7.368 |
neb | -7.286 |
Token | [ |
Token ID | 685 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
BELOW | +2.809 |
idated | +2.586 |
productive | +2.545 |
ridge | +2.291 |
arrass | +2.272 |
Neg logit contributions | |
tti | -4.547 |
etz | -4.095 |
neb | -4.056 |
Probe | -4.031 |
earchers | -3.977 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
BELOW | +4.484 |
productive | +4.123 |
idated | +3.951 |
ridge | +3.733 |
below | +3.524 |
Neg logit contributions | |
tti | -8.571 |
earchers | -8.392 |
neb | -7.989 |
Probe | -7.962 |
etz | -7.837 |
Token | ] |
Token ID | 2361 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
BELOW | +5.041 |
idated | +2.399 |
ridge | +2.336 |
productive | +2.222 |
below | +2.192 |
Neg logit contributions | |
mathemat | -26.542 |
cumbers | -26.522 |
earchers | -26.053 |
unbeliev | -25.043 |
neb | -24.379 |
Token |
|
Token ID | 198 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
BELOW | +4.447 |
idated | +4.288 |
productive | +4.189 |
arrass | +3.775 |
ridge | +3.760 |
Neg logit contributions | |
tti | -6.189 |
etz | -5.557 |
neb | -5.518 |
Probe | -5.465 |
mysteries | -5.349 |
Token |
|
Token ID | 198 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
BELOW | +1.721 |
productive | +1.612 |
idated | +1.612 |
ridge | +1.437 |
arrass | +1.386 |
Neg logit contributions | |
tti | -2.600 |
neb | -2.399 |
earchers | -2.391 |
etz | -2.360 |
Probe | -2.338 |
Token | In |
Token ID | 818 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
tti | +3.156 |
etz | +2.715 |
neb | +2.664 |
Probe | +2.643 |
Television | +2.592 |
Neg logit contributions | |
BELOW | -3.611 |
idated | -3.544 |
productive | -3.484 |
arrass | -3.261 |
ridge | -3.169 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
tti | +1.994 |
etz | +1.730 |
neb | +1.728 |
Probe | +1.716 |
mysteries | +1.661 |
Neg logit contributions | |
BELOW | -2.197 |
idated | -2.145 |
productive | -2.121 |
arrass | -1.974 |
ridge | -1.919 |
Token |
|
Token ID | 198 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
tti | +0.560 |
etz | +0.484 |
neb | +0.484 |
Probe | +0.476 |
mysteries | +0.461 |
Neg logit contributions | |
BELOW | -0.661 |
idated | -0.642 |
productive | -0.637 |
arrass | -0.591 |
ridge | -0.583 |
Token |
|
Token ID | 198 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
tti | +1.173 |
neb | +1.041 |
etz | +1.040 |
Probe | +1.030 |
mysteries | +1.002 |
Neg logit contributions | |
BELOW | -0.952 |
idated | -0.923 |
productive | -0.914 |
arrass | -0.834 |
ridge | -0.817 |
Token | Other |
Token ID | 6395 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
tti | +1.523 |
etz | +1.324 |
neb | +1.310 |
Probe | +1.297 |
mysteries | +1.260 |
Neg logit contributions | |
BELOW | -1.628 |
idated | -1.593 |
productive | -1.573 |
arrass | -1.464 |
ridge | -1.427 |
Token | media |
Token ID | 2056 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
BELOW | +0.219 |
idated | +0.212 |
productive | +0.211 |
arrass | +0.195 |
ridge | +0.193 |
Neg logit contributions | |
tti | -0.169 |
etz | -0.146 |
neb | -0.144 |
Probe | -0.144 |
earchers | -0.137 |
Token | [ |
Token ID | 685 |
Feature activation | +9.73e-01 |
Pos logit contributions | |
tti | +1.256 |
etz | +1.096 |
neb | +1.093 |
Probe | +1.083 |
mysteries | +1.048 |
Neg logit contributions | |
BELOW | -1.289 |
idated | -1.255 |
productive | -1.243 |
arrass | -1.150 |
ridge | -1.128 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
BELOW | +4.621 |
productive | +4.366 |
idated | +4.333 |
ridge | +3.818 |
arrass | +3.652 |
Neg logit contributions | |
tti | -7.606 |
etz | -7.058 |
neb | -7.001 |
Probe | -6.924 |
earchers | -6.879 |
Token | ] |
Token ID | 2361 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
BELOW | +7.171 |
idated | +5.285 |
ridge | +5.142 |
productive | +5.100 |
rik | +4.417 |
Neg logit contributions | |
tti | -21.408 |
earchers | -21.195 |
Probe | -20.506 |
neb | -20.414 |
ignt | -20.378 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
BELOW | +3.727 |
idated | +3.589 |
productive | +3.518 |
arrass | +3.204 |
ridge | +3.167 |
Neg logit contributions | |
tti | -4.838 |
etz | -4.325 |
neb | -4.299 |
Probe | -4.256 |
mysteries | -4.151 |
Token |
|
Token ID | 198 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
BELOW | +0.790 |
idated | +0.750 |
productive | +0.746 |
ridge | +0.667 |
arrass | +0.658 |
Neg logit contributions | |
tti | -1.108 |
neb | -1.010 |
etz | -0.998 |
Probe | -0.990 |
earchers | -0.989 |
Token | Video |
Token ID | 10798 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
tti | +3.466 |
etz | +2.987 |
neb | +2.913 |
Probe | +2.887 |
Television | +2.845 |
Neg logit contributions | |
BELOW | -3.808 |
idated | -3.737 |
productive | -3.663 |
arrass | -3.436 |
ridge | -3.326 |
Token | games |
Token ID | 1830 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
BELOW | +9.001 |
productive | +8.134 |
idated | +7.999 |
ridge | +7.771 |
arrass | +7.515 |
Neg logit contributions | |
tti | -8.750 |
etz | -7.601 |
earchers | -7.412 |
Probe | -7.368 |
neb | -7.286 |
Token |
|
Token ID | 198 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
tti | +1.128 |
etz | +1.001 |
neb | +1.001 |
Probe | +0.990 |
mysteries | +0.964 |
Neg logit contributions | |
BELOW | -0.913 |
idated | -0.885 |
productive | -0.876 |
arrass | -0.799 |
ridge | -0.784 |
Token | Enc |
Token ID | 27195 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
tti | +2.304 |
etz | +2.010 |
neb | +1.983 |
Probe | +1.965 |
mysteries | +1.911 |
Neg logit contributions | |
BELOW | -2.365 |
idated | -2.305 |
productive | -2.281 |
arrass | -2.122 |
ridge | -2.066 |
Token | ryption |
Token ID | 13168 |
Feature activation | -8.13e-01 |
Pos logit contributions | |
tti | +2.565 |
etz | +2.211 |
neb | +2.143 |
Probe | +2.125 |
recognition | +2.050 |
Neg logit contributions | |
BELOW | -3.745 |
productive | -3.631 |
idated | -3.606 |
arrass | -3.346 |
ridge | -3.300 |
Token | details |
Token ID | 3307 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
tti | +5.325 |
Probe | +4.730 |
etz | +4.722 |
neb | +4.714 |
recognition | +4.612 |
Neg logit contributions | |
BELOW | -5.038 |
productive | -4.958 |
idated | -4.890 |
arrass | -4.595 |
ridge | -4.369 |
Token | [ |
Token ID | 685 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
tti | +0.272 |
etz | +0.244 |
neb | +0.243 |
Probe | +0.241 |
mysteries | +0.235 |
Neg logit contributions | |
BELOW | -0.175 |
idated | -0.169 |
productive | -0.166 |
arrass | -0.150 |
ridge | -0.147 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
BELOW | +4.808 |
productive | +4.462 |
idated | +4.281 |
ridge | +4.077 |
below | +3.848 |
Neg logit contributions | |
tti | -9.788 |
earchers | -9.575 |
Probe | -9.125 |
neb | -9.106 |
etz | -8.940 |
Token | ] |
Token ID | 2361 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
BELOW | +6.527 |
idated | +4.052 |
productive | +3.955 |
below | +3.814 |
� | +3.809 |
Neg logit contributions | |
cumbers | -24.066 |
earchers | -23.949 |
mathemat | -23.872 |
unbeliev | -22.872 |
neb | -22.243 |
Token |
|
Token ID | 198 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
BELOW | +3.489 |
idated | +3.154 |
productive | +3.138 |
ridge | +2.810 |
arrass | +2.646 |
Neg logit contributions | |
tti | -5.667 |
earchers | -5.294 |
neb | -5.282 |
etz | -5.174 |
Probe | -5.087 |
Token |
|
Token ID | 198 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
tti | +1.369 |
etz | +1.213 |
neb | +1.211 |
Probe | +1.201 |
mysteries | +1.167 |
Neg logit contributions | |
BELOW | -1.123 |
idated | -1.091 |
productive | -1.079 |
arrass | -0.988 |
ridge | -0.966 |
Token | WE |
Token ID | 8845 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
tti | +5.400 |
etz | +4.713 |
neb | +4.522 |
Probe | +4.446 |
recognition | +4.425 |
Neg logit contributions | |
BELOW | -5.660 |
idated | -5.529 |
productive | -5.459 |
arrass | -5.156 |
ridge | -4.937 |
Token | P |
Token ID | 47 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
tti | +5.048 |
Probe | +4.253 |
etz | +4.208 |
recognition | +3.929 |
neb | +3.920 |
Neg logit contributions | |
idated | -8.589 |
productive | -8.465 |
BELOW | -8.457 |
arrass | -8.033 |
ridge | -7.880 |
Token |
|
Token ID | 198 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
BELOW | +1.701 |
idated | +1.645 |
productive | +1.621 |
arrass | +1.478 |
ridge | +1.463 |
Neg logit contributions | |
tti | -1.696 |
etz | -1.493 |
neb | -1.488 |
Probe | -1.434 |
mysteries | -1.432 |
Token |
|
Token ID | 198 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
tti | +1.240 |
etz | +1.100 |
neb | +1.099 |
Probe | +1.088 |
mysteries | +1.059 |
Neg logit contributions | |
BELOW | -1.010 |
idated | -0.979 |
productive | -0.969 |
arrass | -0.886 |
ridge | -0.867 |
Token | Other |
Token ID | 6395 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
tti | +2.590 |
etz | +2.274 |
neb | +2.242 |
Probe | +2.223 |
mysteries | +2.157 |
Neg logit contributions | |
BELOW | -2.601 |
idated | -2.531 |
productive | -2.492 |
arrass | -2.320 |
ridge | -2.266 |
Token | animals |
Token ID | 4695 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
BELOW | +1.344 |
productive | +1.306 |
idated | +1.302 |
arrass | +1.190 |
ridge | +1.175 |
Neg logit contributions | |
tti | -1.246 |
etz | -1.069 |
Probe | -1.050 |
neb | -1.049 |
earchers | -1.016 |
Token | [ |
Token ID | 685 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
BELOW | +3.818 |
idated | +3.714 |
productive | +3.629 |
arrass | +3.314 |
ridge | +3.283 |
Neg logit contributions | |
tti | -4.269 |
neb | -3.685 |
Probe | -3.662 |
etz | -3.659 |
earchers | -3.582 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
BELOW | +4.693 |
productive | +4.343 |
idated | +4.154 |
ridge | +3.946 |
below | +3.756 |
Neg logit contributions | |
tti | -9.477 |
earchers | -9.375 |
neb | -8.946 |
Probe | -8.859 |
etz | -8.679 |
Token | ] |
Token ID | 2361 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
BELOW | +5.454 |
idated | +3.190 |
productive | +3.056 |
below | +2.860 |
ridge | +2.854 |
Neg logit contributions | |
cumbers | -25.014 |
mathemat | -24.655 |
earchers | -24.420 |
unbeliev | -23.709 |
neb | -22.875 |
Token |
|
Token ID | 198 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
BELOW | +4.330 |
idated | +4.188 |
productive | +4.125 |
arrass | +3.705 |
ridge | +3.658 |
Neg logit contributions | |
tti | -6.005 |
etz | -5.350 |
neb | -5.338 |
Probe | -5.284 |
mysteries | -5.170 |
Token |
|
Token ID | 198 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
BELOW | +2.563 |
productive | +2.374 |
idated | +2.372 |
ridge | +2.125 |
arrass | +2.003 |
Neg logit contributions | |
tti | -4.149 |
earchers | -3.898 |
neb | -3.869 |
etz | -3.784 |
Probe | -3.756 |
Token | Wild |
Token ID | 25946 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
tti | +1.914 |
etz | +1.680 |
neb | +1.670 |
Probe | +1.655 |
mysteries | +1.606 |
Neg logit contributions | |
BELOW | -1.843 |
idated | -1.794 |
productive | -1.772 |
arrass | -1.640 |
ridge | -1.603 |
Token | nine |
Token ID | 5193 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
productive | +6.633 |
BELOW | +6.523 |
idated | +6.275 |
ridge | +6.068 |
arrass | +5.752 |
Neg logit contributions | |
tti | -6.856 |
earchers | -6.019 |
etz | -5.971 |
Probe | -5.967 |
emis | -5.963 |
Token | what |
Token ID | 644 |
Feature activation | +7.74e-01 |
Pos logit contributions | |
BELOW | +1.691 |
idated | +1.640 |
productive | +1.622 |
arrass | +1.519 |
ridge | +1.483 |
Neg logit contributions | |
tti | -1.518 |
etz | -1.334 |
Probe | -1.303 |
neb | -1.300 |
earchers | -1.260 |
Token | I |
Token ID | 314 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
BELOW | +5.161 |
idated | +4.977 |
productive | +4.945 |
arrass | +4.613 |
ridge | +4.459 |
Neg logit contributions | |
tti | -4.635 |
etz | -4.160 |
Probe | -4.033 |
neb | -3.957 |
earchers | -3.931 |
Token | saw |
Token ID | 2497 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
tti | +0.095 |
etz | +0.082 |
neb | +0.082 |
Probe | +0.082 |
mysteries | +0.078 |
Neg logit contributions | |
BELOW | -0.118 |
idated | -0.115 |
productive | -0.114 |
arrass | -0.106 |
ridge | -0.104 |
Token | was |
Token ID | 373 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
BELOW | +4.193 |
idated | +4.055 |
productive | +4.023 |
arrass | +3.738 |
ridge | +3.702 |
Neg logit contributions | |
tti | -3.776 |
etz | -3.333 |
Probe | -3.279 |
neb | -3.215 |
earchers | -3.156 |
Token | a |
Token ID | 257 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
BELOW | +3.845 |
idated | +3.732 |
productive | +3.671 |
arrass | +3.436 |
ridge | +3.388 |
Neg logit contributions | |
tti | -3.211 |
etz | -2.817 |
Probe | -2.741 |
neb | -2.683 |
earchers | -2.619 |
Token | level |
Token ID | 1241 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
BELOW | +3.799 |
idated | +3.668 |
productive | +3.639 |
arrass | +3.377 |
ridge | +3.305 |
Neg logit contributions | |
tti | -3.530 |
etz | -3.073 |
Probe | -2.998 |
earchers | -2.911 |
neb | -2.890 |
Token | of |
Token ID | 286 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
BELOW | +14.093 |
productive | +12.984 |
idated | +12.821 |
ridge | +12.204 |
deter | +11.052 |
Neg logit contributions | |
tti | -14.599 |
Probe | -13.072 |
earchers | -12.826 |
emis | -12.264 |
Satellite | -12.197 |
Token | dishon |
Token ID | 32045 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
BELOW | +5.055 |
idated | +4.867 |
productive | +4.831 |
arrass | +4.505 |
ridge | +4.395 |
Neg logit contributions | |
tti | -4.798 |
etz | -4.213 |
Probe | -4.135 |
earchers | -4.058 |
emis | -3.878 |
Token | esty |
Token ID | 9673 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
tti | +3.588 |
etz | +3.060 |
Probe | +2.850 |
neb | +2.847 |
recognition | +2.751 |
Neg logit contributions | |
BELOW | -7.787 |
productive | -7.583 |
idated | -7.500 |
覚醒 | -7.257 |
arrass | -7.095 |
Token | and |
Token ID | 290 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
tti | +4.087 |
neb | +3.566 |
etz | +3.445 |
Probe | +3.372 |
recognition | +3.346 |
Neg logit contributions | |
BELOW | -5.553 |
idated | -5.550 |
productive | -5.495 |
arrass | -5.141 |
ridge | -5.024 |
Token | ir |
Token ID | 4173 |
Feature activation | +6.43e-01 |
Pos logit contributions | |
BELOW | +1.252 |
idated | +1.207 |
productive | +1.196 |
arrass | +1.111 |
ridge | +1.083 |
Neg logit contributions | |
tti | -1.353 |
etz | -1.193 |
Probe | -1.180 |
neb | -1.159 |
earchers | -1.136 |
Token |
|
Token ID | 198 |
Feature activation | +6.09e-02 |
Pos logit contributions | |
BELOW | +1.133 |
idated | +1.095 |
productive | +1.076 |
arrass | +0.994 |
ridge | +0.983 |
Neg logit contributions | |
tti | -1.111 |
neb | -0.983 |
etz | -0.980 |
Probe | -0.955 |
mysteries | -0.935 |
Token |
|
Token ID | 198 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
BELOW | +0.339 |
idated | +0.325 |
productive | +0.323 |
ridge | +0.288 |
arrass | +0.288 |
Neg logit contributions | |
tti | -0.457 |
neb | -0.413 |
etz | -0.409 |
Probe | -0.406 |
earchers | -0.401 |
Token | See |
Token ID | 6214 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
tti | +3.019 |
etz | +2.612 |
neb | +2.565 |
Probe | +2.536 |
mysteries | +2.485 |
Neg logit contributions | |
BELOW | -3.356 |
idated | -3.284 |
productive | -3.225 |
arrass | -3.008 |
ridge | -2.930 |
Token | also |
Token ID | 635 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
BELOW | +1.302 |
idated | +1.244 |
productive | +1.237 |
arrass | +1.154 |
ridge | +1.141 |
Neg logit contributions | |
tti | -1.199 |
etz | -1.056 |
neb | -1.042 |
Probe | -1.021 |
mysteries | -0.989 |
Token | [ |
Token ID | 685 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
BELOW | +3.708 |
productive | +3.298 |
idated | +3.280 |
ridge | +2.988 |
arrass | +2.960 |
Neg logit contributions | |
tti | -6.980 |
etz | -6.397 |
neb | -6.291 |
earchers | -6.199 |
Probe | -6.194 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
BELOW | +5.284 |
productive | +4.861 |
idated | +4.662 |
ridge | +4.476 |
below | +4.218 |
Neg logit contributions | |
earchers | -10.378 |
tti | -10.305 |
Probe | -9.832 |
neb | -9.826 |
etz | -9.729 |
Token | ] |
Token ID | 2361 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
BELOW | +5.004 |
productive | +2.603 |
idated | +2.568 |
ridge | +2.469 |
below | +2.198 |
Neg logit contributions | |
earchers | -23.833 |
mathemat | -23.286 |
cumbers | -23.257 |
ignt | -22.400 |
tti | -22.344 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.64e-02 |
Pos logit contributions | |
BELOW | +3.223 |
idated | +3.118 |
productive | +3.078 |
arrass | +2.842 |
ridge | +2.779 |
Neg logit contributions | |
tti | -3.672 |
etz | -3.250 |
neb | -3.240 |
Probe | -3.202 |
mysteries | -3.108 |
Token | I |
Token ID | 40 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
tti | +0.619 |
etz | +0.541 |
neb | +0.540 |
Probe | +0.535 |
mysteries | +0.518 |
Neg logit contributions | |
BELOW | -0.638 |
idated | -0.622 |
productive | -0.616 |
arrass | -0.570 |
ridge | -0.559 |
Token | ke |
Token ID | 365 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
tti | +0.886 |
etz | +0.759 |
neb | +0.758 |
Probe | +0.749 |
mysteries | +0.723 |
Neg logit contributions | |
BELOW | -1.150 |
idated | -1.124 |
productive | -1.114 |
arrass | -1.041 |
ridge | -1.022 |
Token | a |
Token ID | 64 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
BELOW | +2.499 |
idated | +2.433 |
productive | +2.418 |
arrass | +2.257 |
ridge | +2.220 |
Neg logit contributions | |
tti | -2.274 |
etz | -1.999 |
neb | -1.988 |
Probe | -1.972 |
mysteries | -1.905 |
Token |
|
Token ID | 198 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
BELOW | +1.776 |
idated | +1.704 |
productive | +1.683 |
ridge | +1.551 |
arrass | +1.550 |
Neg logit contributions | |
tti | -1.547 |
etz | -1.358 |
neb | -1.351 |
Probe | -1.296 |
earchers | -1.294 |
Token |
|
Token ID | 198 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
BELOW | +0.192 |
idated | +0.184 |
productive | +0.183 |
arrass | +0.163 |
ridge | +0.163 |
Neg logit contributions | |
tti | -0.257 |
neb | -0.232 |
etz | -0.230 |
Probe | -0.228 |
earchers | -0.225 |
Token | Sound |
Token ID | 21369 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
tti | +2.605 |
etz | +2.257 |
neb | +2.229 |
Probe | +2.207 |
mysteries | +2.152 |
Neg logit contributions | |
BELOW | -2.862 |
idated | -2.801 |
productive | -2.753 |
arrass | -2.567 |
ridge | -2.504 |
Token | tracks |
Token ID | 46074 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
tti | +0.698 |
etz | +0.559 |
neb | +0.559 |
Probe | +0.549 |
mysteries | +0.516 |
Neg logit contributions | |
BELOW | -1.494 |
idated | -1.463 |
productive | -1.453 |
arrass | -1.373 |
ridge | -1.356 |
Token | [ |
Token ID | 685 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
BELOW | +0.832 |
idated | +0.793 |
productive | +0.782 |
arrass | +0.702 |
ridge | +0.693 |
Neg logit contributions | |
tti | -1.296 |
neb | -1.168 |
etz | -1.168 |
Probe | -1.156 |
mysteries | -1.121 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
BELOW | +5.352 |
productive | +4.876 |
idated | +4.674 |
ridge | +4.577 |
below | +4.136 |
Neg logit contributions | |
earchers | -11.216 |
tti | -11.121 |
Probe | -10.631 |
neb | -10.620 |
etz | -10.418 |
Token | ] |
Token ID | 2361 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
BELOW | +4.593 |
idated | +2.191 |
productive | +2.143 |
ridge | +1.933 |
below | +1.818 |
Neg logit contributions | |
cumbers | -24.876 |
mathemat | -24.873 |
earchers | -24.582 |
unbeliev | -23.411 |
neb | -23.343 |
Token |
|
Token ID | 198 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
BELOW | +3.959 |
idated | +3.717 |
productive | +3.639 |
ridge | +3.258 |
arrass | +3.152 |
Neg logit contributions | |
tti | -6.389 |
neb | -5.838 |
etz | -5.819 |
earchers | -5.772 |
Probe | -5.730 |
Token |
|
Token ID | 198 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
BELOW | +1.658 |
idated | +1.558 |
productive | +1.551 |
ridge | +1.388 |
arrass | +1.345 |
Neg logit contributions | |
tti | -2.492 |
neb | -2.298 |
earchers | -2.284 |
etz | -2.264 |
Probe | -2.244 |
Token | Various |
Token ID | 40009 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
tti | +2.739 |
etz | +2.360 |
neb | +2.323 |
Probe | +2.300 |
mysteries | +2.246 |
Neg logit contributions | |
BELOW | -3.101 |
idated | -3.036 |
productive | -2.988 |
arrass | -2.795 |
ridge | -2.718 |
Token | sound |
Token ID | 2128 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
tti | +0.729 |
etz | +0.624 |
neb | +0.621 |
Probe | +0.615 |
mysteries | +0.591 |
Neg logit contributions | |
BELOW | -0.953 |
idated | -0.930 |
productive | -0.922 |
arrass | -0.861 |
ridge | -0.846 |
Token | Theme |
Token ID | 47863 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
tti | +2.143 |
etz | +1.860 |
neb | +1.843 |
Probe | +1.828 |
mysteries | +1.777 |
Neg logit contributions | |
BELOW | -2.264 |
idated | -2.207 |
productive | -2.181 |
arrass | -2.027 |
ridge | -1.982 |
Token | parks |
Token ID | 14860 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
tti | +1.361 |
etz | +1.166 |
neb | +1.161 |
Probe | +1.150 |
mysteries | +1.103 |
Neg logit contributions | |
BELOW | -1.809 |
idated | -1.769 |
productive | -1.753 |
arrass | -1.637 |
ridge | -1.610 |
Token | and |
Token ID | 290 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
BELOW | +3.446 |
idated | +3.288 |
productive | +3.225 |
ridge | +2.964 |
arrass | +2.956 |
Neg logit contributions | |
tti | -4.159 |
neb | -3.733 |
Probe | -3.699 |
etz | -3.696 |
earchers | -3.625 |
Token | displays |
Token ID | 11298 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
BELOW | +1.861 |
idated | +1.813 |
productive | +1.791 |
arrass | +1.659 |
ridge | +1.639 |
Neg logit contributions | |
tti | -1.608 |
etz | -1.401 |
Probe | -1.382 |
neb | -1.372 |
earchers | -1.336 |
Token | [ |
Token ID | 685 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
BELOW | +1.534 |
idated | +1.459 |
productive | +1.444 |
arrass | +1.301 |
ridge | +1.290 |
Neg logit contributions | |
tti | -2.116 |
etz | -1.892 |
neb | -1.883 |
Probe | -1.877 |
earchers | -1.831 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
BELOW | +5.831 |
productive | +5.131 |
idated | +4.951 |
ridge | +4.806 |
below | +4.494 |
Neg logit contributions | |
earchers | -13.941 |
tti | -13.561 |
neb | -12.913 |
Probe | -12.749 |
mathemat | -12.704 |
Token | ] |
Token ID | 2361 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
BELOW | +4.750 |
idated | +2.740 |
productive | +2.571 |
ridge | +2.338 |
below | +2.241 |
Neg logit contributions | |
cumbers | -24.561 |
mathemat | -24.451 |
earchers | -24.078 |
unbeliev | -23.249 |
neb | -22.778 |
Token |
|
Token ID | 198 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
BELOW | +4.742 |
idated | +4.170 |
productive | +4.122 |
ridge | +3.661 |
below | +3.378 |
Neg logit contributions | |
tti | -8.493 |
earchers | -8.293 |
neb | -8.101 |
etz | -7.907 |
Probe | -7.752 |
Token |
|
Token ID | 198 |
Feature activation | -7.05e-02 |
Pos logit contributions | |
BELOW | +3.491 |
idated | +3.200 |
productive | +3.183 |
ridge | +2.846 |
arrass | +2.625 |
Neg logit contributions | |
tti | -6.160 |
earchers | -5.977 |
neb | -5.835 |
etz | -5.649 |
Probe | -5.642 |
Token | M |
Token ID | 44 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
tti | +0.441 |
neb | +0.385 |
etz | +0.384 |
Probe | +0.381 |
mysteries | +0.368 |
Neg logit contributions | |
BELOW | -0.478 |
idated | -0.467 |
productive | -0.463 |
arrass | -0.428 |
ridge | -0.420 |
Token | oom |
Token ID | 4207 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
tti | +3.685 |
etz | +3.345 |
neb | +3.279 |
Probe | +3.258 |
mysteries | +3.188 |
Neg logit contributions | |
BELOW | -2.451 |
productive | -2.346 |
idated | -2.317 |
arrass | -2.053 |
ridge | -2.049 |
Token |
|
Token ID | 198 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
BELOW | +2.611 |
idated | +2.510 |
productive | +2.472 |
arrass | +2.248 |
ridge | +2.229 |
Neg logit contributions | |
tti | -2.572 |
neb | -2.273 |
etz | -2.261 |
earchers | -2.194 |
mysteries | -2.192 |
Token |
|
Token ID | 198 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
tti | +1.765 |
etz | +1.559 |
neb | +1.551 |
Probe | +1.542 |
mysteries | +1.498 |
Neg logit contributions | |
BELOW | -1.488 |
idated | -1.449 |
productive | -1.433 |
arrass | -1.320 |
ridge | -1.283 |
Token | Notable |
Token ID | 45533 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
tti | +1.223 |
etz | +1.070 |
neb | +1.062 |
Probe | +1.053 |
mysteries | +1.020 |
Neg logit contributions | |
BELOW | -1.252 |
idated | -1.220 |
productive | -1.207 |
arrass | -1.119 |
ridge | -1.095 |
Token | cases |
Token ID | 2663 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
BELOW | +1.524 |
idated | +1.467 |
productive | +1.461 |
arrass | +1.341 |
ridge | +1.312 |
Neg logit contributions | |
tti | -1.456 |
etz | -1.272 |
neb | -1.249 |
Probe | -1.248 |
earchers | -1.211 |
Token | [ |
Token ID | 685 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
BELOW | +6.804 |
idated | +6.448 |
productive | +6.366 |
arrass | +5.836 |
ridge | +5.818 |
Neg logit contributions | |
tti | -8.398 |
neb | -7.364 |
etz | -7.359 |
Probe | -7.354 |
earchers | -7.258 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
BELOW | +4.820 |
productive | +4.480 |
idated | +4.249 |
ridge | +4.019 |
below | +3.847 |
Neg logit contributions | |
tti | -9.981 |
earchers | -9.975 |
Probe | -9.441 |
neb | -9.387 |
etz | -9.213 |
Token | ] |
Token ID | 2361 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
BELOW | +5.347 |
idated | +3.187 |
productive | +3.142 |
below | +3.019 |
ridge | +2.922 |
Neg logit contributions | |
cumbers | -23.820 |
mathemat | -23.628 |
earchers | -23.500 |
unbeliev | -22.658 |
neb | -21.890 |
Token |
|
Token ID | 198 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
BELOW | +3.954 |
idated | +3.711 |
productive | +3.638 |
ridge | +3.206 |
arrass | +3.071 |
Neg logit contributions | |
tti | -6.441 |
neb | -6.008 |
earchers | -5.909 |
etz | -5.823 |
Probe | -5.762 |
Token |
|
Token ID | 198 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
BELOW | +2.756 |
idated | +2.533 |
productive | +2.522 |
ridge | +2.268 |
arrass | +2.115 |
Neg logit contributions | |
tti | -4.551 |
earchers | -4.313 |
neb | -4.265 |
etz | -4.165 |
Probe | -4.137 |
Token | Research |
Token ID | 25104 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
BELOW | +0.484 |
idated | +0.472 |
productive | +0.469 |
arrass | +0.432 |
ridge | +0.425 |
Neg logit contributions | |
tti | -0.484 |
neb | -0.425 |
etz | -0.422 |
Probe | -0.422 |
mysteries | -0.408 |
Token | directions |
Token ID | 11678 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
BELOW | +2.047 |
idated | +1.980 |
productive | +1.968 |
arrass | +1.794 |
ridge | +1.763 |
Neg logit contributions | |
tti | -2.155 |
etz | -1.886 |
neb | -1.877 |
Probe | -1.844 |
mysteries | -1.790 |
Token | ] |
Token ID | 2361 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
BELOW | +5.370 |
productive | +3.077 |
idated | +3.016 |
ridge | +2.950 |
Loading | +2.723 |
Neg logit contributions | |
earchers | -22.305 |
mathemat | -21.075 |
tti | -21.065 |
cumbers | -20.855 |
ignt | -20.845 |
Token |
|
Token ID | 198 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
BELOW | +3.309 |
idated | +3.201 |
productive | +3.161 |
arrass | +2.909 |
ridge | +2.850 |
Neg logit contributions | |
tti | -3.892 |
etz | -3.451 |
neb | -3.441 |
Probe | -3.410 |
mysteries | -3.313 |
Token |
|
Token ID | 198 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
BELOW | +0.384 |
idated | +0.367 |
productive | +0.364 |
arrass | +0.327 |
ridge | +0.326 |
Neg logit contributions | |
tti | -0.508 |
neb | -0.458 |
etz | -0.456 |
Probe | -0.451 |
earchers | -0.444 |
Token | References |
Token ID | 19927 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
tti | +2.351 |
etz | +2.045 |
neb | +2.026 |
Probe | +2.004 |
recognition | +1.954 |
Neg logit contributions | |
BELOW | -2.227 |
idated | -2.183 |
productive | -2.129 |
arrass | -1.969 |
ridge | -1.919 |
Token | [ |
Token ID | 685 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
BELOW | +4.290 |
productive | +3.877 |
idated | +3.821 |
arrass | +3.350 |
ridge | +3.316 |
Neg logit contributions | |
tti | -9.055 |
etz | -8.189 |
Probe | -8.094 |
neb | -8.075 |
earchers | -8.022 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
BELOW | +4.563 |
productive | +4.352 |
idated | +3.986 |
ridge | +3.813 |
below | +3.672 |
Neg logit contributions | |
tti | -8.428 |
earchers | -8.175 |
Probe | -7.948 |
etz | -7.895 |
neb | -7.888 |
Token | ] |
Token ID | 2361 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
BELOW | +5.187 |
productive | +2.744 |
idated | +2.623 |
ridge | +2.517 |
below | +2.429 |
Neg logit contributions | |
earchers | -23.307 |
cumbers | -22.926 |
mathemat | -22.728 |
ignt | -22.030 |
tti | -21.914 |
Token |
|
Token ID | 198 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
BELOW | +3.569 |
idated | +3.450 |
productive | +3.417 |
arrass | +3.143 |
ridge | +3.075 |
Neg logit contributions | |
tti | -3.991 |
etz | -3.521 |
neb | -3.518 |
Probe | -3.484 |
mysteries | -3.390 |
Token |
|
Token ID | 198 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
BELOW | +1.191 |
idated | +1.127 |
productive | +1.124 |
ridge | +1.004 |
arrass | +0.995 |
Neg logit contributions | |
tti | -1.660 |
neb | -1.512 |
etz | -1.496 |
Probe | -1.480 |
earchers | -1.478 |
Token | the |
Token ID | 1169 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
tti | +0.440 |
etz | +0.392 |
neb | +0.390 |
Probe | +0.386 |
mysteries | +0.377 |
Neg logit contributions | |
BELOW | -0.318 |
idated | -0.309 |
productive | -0.304 |
arrass | -0.277 |
ridge | -0.270 |
Token | ten |
Token ID | 3478 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
tti | +2.233 |
etz | +1.950 |
neb | +1.949 |
Probe | +1.923 |
mysteries | +1.866 |
Neg logit contributions | |
BELOW | -2.231 |
idated | -2.181 |
productive | -2.158 |
arrass | -2.000 |
ridge | -1.949 |
Token |
|
Token ID | 198 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
BELOW | +0.933 |
idated | +0.899 |
productive | +0.892 |
arrass | +0.820 |
ridge | +0.801 |
Neg logit contributions | |
tti | -0.960 |
etz | -0.846 |
neb | -0.842 |
Probe | -0.818 |
earchers | -0.809 |
Token |
|
Token ID | 198 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
tti | +3.551 |
etz | +3.093 |
Probe | +3.053 |
neb | +3.034 |
recognition | +2.965 |
Neg logit contributions | |
BELOW | -3.369 |
idated | -3.328 |
productive | -3.282 |
arrass | -3.095 |
覚醒 | -2.952 |
Token | See |
Token ID | 6214 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
tti | +4.147 |
etz | +3.632 |
neb | +3.551 |
Probe | +3.515 |
mysteries | +3.430 |
Neg logit contributions | |
BELOW | -4.310 |
idated | -4.193 |
productive | -4.098 |
arrass | -3.830 |
ridge | -3.744 |
Token | also |
Token ID | 635 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
BELOW | +2.099 |
idated | +1.994 |
productive | +1.994 |
arrass | +1.846 |
ridge | +1.826 |
Neg logit contributions | |
tti | -2.063 |
etz | -1.802 |
neb | -1.771 |
Probe | -1.751 |
mysteries | -1.701 |
Token | [ |
Token ID | 685 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
BELOW | +3.378 |
productive | +3.080 |
idated | +3.040 |
arrass | +2.727 |
ridge | +2.705 |
Neg logit contributions | |
tti | -6.186 |
etz | -5.587 |
neb | -5.488 |
Probe | -5.474 |
earchers | -5.427 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
BELOW | +4.602 |
productive | +4.265 |
idated | +4.044 |
ridge | +3.882 |
below | +3.720 |
Neg logit contributions | |
tti | -8.580 |
earchers | -8.458 |
neb | -8.115 |
Probe | -8.069 |
etz | -7.991 |
Token | ] |
Token ID | 2361 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
BELOW | +4.902 |
productive | +2.620 |
idated | +2.541 |
below | +2.392 |
ridge | +2.343 |
Neg logit contributions | |
earchers | -23.752 |
cumbers | -23.601 |
mathemat | -23.430 |
unbeliev | -22.400 |
ignt | -22.179 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
BELOW | +3.393 |
idated | +3.283 |
productive | +3.248 |
arrass | +2.987 |
ridge | +2.923 |
Neg logit contributions | |
tti | -3.913 |
etz | -3.458 |
neb | -3.446 |
Probe | -3.411 |
mysteries | -3.310 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
BELOW | +0.961 |
idated | +0.909 |
productive | +0.906 |
ridge | +0.809 |
arrass | +0.801 |
Neg logit contributions | |
tti | -1.340 |
neb | -1.221 |
etz | -1.209 |
earchers | -1.195 |
Probe | -1.195 |
Token | References |
Token ID | 19927 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
tti | +1.426 |
etz | +1.260 |
neb | +1.250 |
Probe | +1.238 |
mysteries | +1.206 |
Neg logit contributions | |
BELOW | -1.228 |
idated | -1.196 |
productive | -1.175 |
arrass | -1.081 |
ridge | -1.057 |
Token | [ |
Token ID | 685 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
BELOW | +4.452 |
productive | +4.075 |
idated | +3.984 |
arrass | +3.508 |
ridge | +3.435 |
Neg logit contributions | |
tti | -8.971 |
etz | -8.070 |
neb | -7.939 |
Probe | -7.908 |
earchers | -7.897 |
Token | of |
Token ID | 286 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
tti | +1.313 |
etz | +1.171 |
neb | +1.168 |
Probe | +1.158 |
mysteries | +1.130 |
Neg logit contributions | |
BELOW | -0.968 |
idated | -0.939 |
productive | -0.927 |
arrass | -0.845 |
ridge | -0.822 |
Token | World |
Token ID | 2159 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
BELOW | +3.150 |
idated | +3.088 |
productive | +3.030 |
arrass | +2.821 |
ridge | +2.797 |
Neg logit contributions | |
tti | -2.996 |
etz | -2.592 |
neb | -2.569 |
Probe | -2.562 |
earchers | -2.547 |
Token | War |
Token ID | 1810 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
BELOW | +4.729 |
idated | +4.422 |
productive | +4.348 |
arrass | +3.934 |
ridge | +3.894 |
Neg logit contributions | |
tti | -9.205 |
earchers | -8.473 |
etz | -8.433 |
neb | -8.362 |
Probe | -8.276 |
Token | II |
Token ID | 2873 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
BELOW | +6.729 |
idated | +6.166 |
productive | +6.014 |
arrass | +5.730 |
ridge | +5.523 |
Neg logit contributions | |
tti | -9.553 |
earchers | -8.969 |
Probe | -8.474 |
neb | -8.455 |
resemb | -8.384 |
Token | [ |
Token ID | 685 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
BELOW | +2.046 |
idated | +1.948 |
productive | +1.918 |
arrass | +1.709 |
ridge | +1.701 |
Neg logit contributions | |
tti | -3.445 |
etz | -3.114 |
neb | -3.095 |
Probe | -3.065 |
earchers | -3.007 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
BELOW | +4.896 |
productive | +4.488 |
idated | +4.379 |
ridge | +4.105 |
below | +3.781 |
Neg logit contributions | |
tti | -10.809 |
earchers | -10.741 |
neb | -10.164 |
Probe | -10.050 |
etz | -9.972 |
Token | ] |
Token ID | 2361 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
BELOW | +5.319 |
idated | +3.567 |
productive | +3.145 |
ridge | +2.951 |
below | +2.920 |
Neg logit contributions | |
cumbers | -23.330 |
mathemat | -23.277 |
earchers | -23.153 |
unbeliev | -22.283 |
neb | -21.778 |
Token |
|
Token ID | 198 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
BELOW | +4.660 |
productive | +4.101 |
idated | +4.076 |
ridge | +3.630 |
below | +3.367 |
Neg logit contributions | |
tti | -8.630 |
earchers | -8.552 |
neb | -8.306 |
etz | -8.023 |
Probe | -7.903 |
Token |
|
Token ID | 198 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
BELOW | +2.326 |
idated | +2.182 |
productive | +2.164 |
ridge | +1.934 |
arrass | +1.849 |
Neg logit contributions | |
tti | -3.652 |
earchers | -3.404 |
neb | -3.399 |
etz | -3.320 |
Probe | -3.317 |
Token | General |
Token ID | 12218 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
BELOW | +1.174 |
idated | +1.143 |
productive | +1.137 |
arrass | +1.041 |
ridge | +1.024 |
Neg logit contributions | |
tti | -1.259 |
neb | -1.122 |
Probe | -1.109 |
etz | -1.100 |
mysteries | -1.072 |
Token | Ge |
Token ID | 2269 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
BELOW | +1.241 |
idated | +1.214 |
productive | +1.200 |
arrass | +1.103 |
ridge | +1.088 |
Neg logit contributions | |
tti | -1.165 |
neb | -1.014 |
etz | -1.010 |
Probe | -1.004 |
mysteries | -0.972 |
Token | em |
Token ID | 368 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
tti | +5.097 |
etz | +4.451 |
neb | +4.303 |
Probe | +4.151 |
bourg | +4.133 |
Neg logit contributions | |
BELOW | -6.458 |
idated | -6.257 |
productive | -6.179 |
arrass | -5.739 |
覚醒 | -5.668 |
Token | orative |
Token ID | 36478 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
BELOW | +1.478 |
idated | +1.453 |
productive | +1.426 |
arrass | +1.305 |
ridge | +1.286 |
Neg logit contributions | |
tti | -2.047 |
Probe | -1.820 |
etz | -1.818 |
neb | -1.811 |
mysteries | -1.769 |
Token | bank |
Token ID | 3331 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
tti | +4.813 |
neb | +4.191 |
etz | +4.146 |
recognition | +4.128 |
Probe | +4.099 |
Neg logit contributions | |
BELOW | -5.167 |
idated | -5.116 |
productive | -5.078 |
arrass | -4.716 |
ridge | -4.495 |
Token | notes |
Token ID | 17815 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
tti | +0.399 |
etz | +0.360 |
neb | +0.359 |
Probe | +0.356 |
mysteries | +0.348 |
Neg logit contributions | |
BELOW | -0.236 |
idated | -0.228 |
productive | -0.224 |
arrass | -0.201 |
ridge | -0.194 |
Token | [ |
Token ID | 685 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
BELOW | +3.466 |
idated | +3.277 |
productive | +3.233 |
ridge | +2.984 |
arrass | +2.968 |
Neg logit contributions | |
tti | -4.012 |
Probe | -3.546 |
neb | -3.534 |
etz | -3.534 |
earchers | -3.502 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
BELOW | +5.485 |
productive | +5.090 |
idated | +4.747 |
ridge | +4.598 |
below | +4.375 |
Neg logit contributions | |
earchers | -11.891 |
tti | -11.709 |
Probe | -11.105 |
neb | -11.084 |
resemb | -10.844 |
Token | ] |
Token ID | 2361 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
BELOW | +4.900 |
idated | +2.618 |
productive | +2.463 |
ridge | +2.328 |
below | +2.082 |
Neg logit contributions | |
earchers | -23.868 |
cumbers | -23.715 |
mathemat | -23.708 |
unbeliev | -22.517 |
neb | -22.267 |
Token |
|
Token ID | 198 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
BELOW | +4.635 |
idated | +4.150 |
productive | +4.084 |
ridge | +3.627 |
below | +3.365 |
Neg logit contributions | |
tti | -8.033 |
earchers | -7.753 |
neb | -7.681 |
etz | -7.448 |
mysteries | -7.271 |
Token |
|
Token ID | 198 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
BELOW | +2.269 |
idated | +2.113 |
productive | +2.103 |
ridge | +1.885 |
arrass | +1.793 |
Neg logit contributions | |
tti | -3.572 |
earchers | -3.334 |
neb | -3.315 |
etz | -3.254 |
Probe | -3.235 |
Token | $ |
Token ID | 3 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
BELOW | +0.131 |
idated | +0.128 |
productive | +0.127 |
arrass | +0.118 |
ridge | +0.115 |
Neg logit contributions | |
tti | -0.120 |
neb | -0.105 |
etz | -0.105 |
Probe | -0.104 |
mysteries | -0.101 |
Token | 25 |
Token ID | 1495 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
tti | +2.518 |
etz | +2.218 |
Probe | +2.202 |
neb | +2.201 |
mysteries | +2.129 |
Neg logit contributions | |
BELOW | -2.115 |
idated | -2.081 |
productive | -2.053 |
arrass | -1.890 |
ridge | -1.834 |
Token | ] |
Token ID | 2361 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
BELOW | +4.902 |
productive | +2.620 |
idated | +2.541 |
below | +2.392 |
ridge | +2.343 |
Neg logit contributions | |
earchers | -23.752 |
cumbers | -23.601 |
mathemat | -23.430 |
unbeliev | -22.400 |
ignt | -22.179 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
BELOW | +3.393 |
idated | +3.283 |
productive | +3.248 |
arrass | +2.987 |
ridge | +2.923 |
Neg logit contributions | |
tti | -3.913 |
etz | -3.458 |
neb | -3.446 |
Probe | -3.411 |
mysteries | -3.310 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
BELOW | +0.961 |
idated | +0.909 |
productive | +0.906 |
ridge | +0.809 |
arrass | +0.801 |
Neg logit contributions | |
tti | -1.340 |
neb | -1.221 |
etz | -1.209 |
earchers | -1.195 |
Probe | -1.195 |
Token | References |
Token ID | 19927 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
tti | +1.426 |
etz | +1.260 |
neb | +1.250 |
Probe | +1.238 |
mysteries | +1.206 |
Neg logit contributions | |
BELOW | -1.228 |
idated | -1.196 |
productive | -1.175 |
arrass | -1.081 |
ridge | -1.057 |
Token | [ |
Token ID | 685 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
BELOW | +4.452 |
productive | +4.075 |
idated | +3.984 |
arrass | +3.508 |
ridge | +3.435 |
Neg logit contributions | |
tti | -8.971 |
etz | -8.070 |
neb | -7.939 |
Probe | -7.908 |
earchers | -7.897 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
BELOW | +4.216 |
productive | +3.951 |
idated | +3.653 |
ridge | +3.517 |
below | +3.342 |
Neg logit contributions | |
tti | -7.784 |
earchers | -7.512 |
Probe | -7.307 |
neb | -7.254 |
etz | -7.234 |
Token | ] |
Token ID | 2361 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
BELOW | +4.873 |
productive | +2.543 |
idated | +2.449 |
below | +2.415 |
ridge | +2.250 |
Neg logit contributions | |
cumbers | -23.385 |
earchers | -23.364 |
mathemat | -23.099 |
unbeliev | -22.199 |
ignt | -22.063 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
BELOW | +2.435 |
idated | +2.367 |
productive | +2.344 |
arrass | +2.161 |
ridge | +2.114 |
Neg logit contributions | |
tti | -2.589 |
etz | -2.274 |
neb | -2.269 |
Probe | -2.248 |
mysteries | -2.182 |
Token | Z |
Token ID | 57 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
tti | +1.066 |
etz | +0.931 |
neb | +0.926 |
Probe | +0.917 |
mysteries | +0.889 |
Neg logit contributions | |
BELOW | -1.095 |
idated | -1.068 |
productive | -1.056 |
arrass | -0.979 |
ridge | -0.958 |
Token | ombie |
Token ID | 9081 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
tti | +0.269 |
neb | +0.235 |
etz | +0.235 |
Probe | +0.233 |
mysteries | +0.226 |
Neg logit contributions | |
BELOW | -0.269 |
idated | -0.262 |
productive | -0.259 |
arrass | -0.240 |
ridge | -0.236 |
Token | Deep |
Token ID | 10766 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
tti | +1.394 |
etz | +1.204 |
neb | +1.202 |
Probe | +1.190 |
mysteries | +1.150 |
Neg logit contributions | |
BELOW | -1.644 |
idated | -1.605 |
productive | -1.591 |
arrass | -1.481 |
ridge | -1.451 |
Token |
|
Token ID | 198 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
BELOW | +3.925 |
idated | +3.596 |
productive | +3.562 |
ridge | +3.158 |
arrass | +2.959 |
Neg logit contributions | |
tti | -6.388 |
earchers | -6.010 |
neb | -6.007 |
etz | -5.866 |
Probe | -5.751 |
Token |
|
Token ID | 198 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
BELOW | +1.387 |
idated | +1.311 |
productive | +1.305 |
ridge | +1.165 |
arrass | +1.142 |
Neg logit contributions | |
tti | -2.013 |
neb | -1.841 |
earchers | -1.818 |
etz | -1.817 |
Probe | -1.798 |
Token | Al |
Token ID | 2348 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
tti | +2.047 |
etz | +1.759 |
neb | +1.731 |
Probe | +1.721 |
mysteries | +1.676 |
Neg logit contributions | |
BELOW | -2.356 |
idated | -2.303 |
productive | -2.267 |
arrass | -2.119 |
ridge | -2.064 |
Token | bum |
Token ID | 4435 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
tti | +4.610 |
Probe | +4.038 |
etz | +4.029 |
neb | +3.960 |
Television | +3.956 |
Neg logit contributions | |
BELOW | -3.577 |
idated | -3.395 |
productive | -3.316 |
覚醒 | -2.990 |
arrass | -2.981 |
Token | [ |
Token ID | 685 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
BELOW | +4.477 |
productive | +4.377 |
idated | +4.352 |
ridge | +4.100 |
arrass | +4.059 |
Neg logit contributions | |
tti | -3.901 |
etz | -3.466 |
neb | -3.404 |
Probe | -3.378 |
earchers | -3.256 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
BELOW | +4.415 |
productive | +4.150 |
idated | +3.949 |
ridge | +3.752 |
below | +3.439 |
Neg logit contributions | |
tti | -8.195 |
earchers | -7.918 |
Probe | -7.706 |
neb | -7.636 |
etz | -7.572 |
Token | ] |
Token ID | 2361 |
Feature activation | +7.36e-01 |
Pos logit contributions | |
BELOW | +4.954 |
idated | +2.793 |
productive | +2.727 |
ridge | +2.678 |
below | +2.431 |
Neg logit contributions | |
earchers | -23.641 |
mathemat | -23.405 |
cumbers | -23.359 |
neb | -22.222 |
unbeliev | -22.146 |
Token |
|
Token ID | 198 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
BELOW | +4.055 |
idated | +3.899 |
productive | +3.845 |
arrass | +3.459 |
ridge | +3.436 |
Neg logit contributions | |
tti | -5.488 |
etz | -4.912 |
neb | -4.872 |
Probe | -4.848 |
mysteries | -4.722 |
Token |
|
Token ID | 198 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
BELOW | +0.458 |
idated | +0.438 |
productive | +0.434 |
ridge | +0.389 |
arrass | +0.388 |
Neg logit contributions | |
tti | -0.617 |
neb | -0.556 |
etz | -0.553 |
Probe | -0.547 |
earchers | -0.540 |
Token | A |
Token ID | 32 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
tti | +2.123 |
etz | +1.836 |
neb | +1.814 |
Probe | +1.795 |
mysteries | +1.747 |
Neg logit contributions | |
BELOW | -2.354 |
idated | -2.303 |
productive | -2.272 |
arrass | -2.116 |
ridge | -2.060 |
Token | solo |
Token ID | 12199 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
tti | +4.839 |
neb | +4.196 |
Probe | +4.140 |
etz | +4.104 |
Television | +4.013 |
Neg logit contributions | |
BELOW | -5.555 |
idated | -5.449 |
productive | -5.363 |
arrass | -5.006 |
ridge | -4.820 |
Token |
|
Token ID | 198 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
tti | +1.937 |
etz | +1.686 |
neb | +1.676 |
Probe | +1.661 |
mysteries | +1.618 |
Neg logit contributions | |
BELOW | -1.990 |
idated | -1.952 |
productive | -1.916 |
arrass | -1.786 |
ridge | -1.728 |
Token |
|
Token ID | 198 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
BELOW | +1.405 |
idated | +1.328 |
productive | +1.325 |
ridge | +1.186 |
arrass | +1.166 |
Neg logit contributions | |
tti | -2.021 |
neb | -1.856 |
etz | -1.826 |
earchers | -1.825 |
Probe | -1.810 |
Token | Begin |
Token ID | 44140 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
BELOW | +2.938 |
productive | +2.890 |
idated | +2.866 |
arrass | +2.629 |
ridge | +2.626 |
Neg logit contributions | |
tti | -3.159 |
neb | -2.876 |
Probe | -2.846 |
etz | -2.817 |
mysteries | -2.719 |
Token | with |
Token ID | 351 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
BELOW | +6.216 |
productive | +5.981 |
idated | +5.930 |
ridge | +5.571 |
arrass | +5.352 |
Neg logit contributions | |
tti | -6.966 |
etz | -6.323 |
neb | -6.252 |
Probe | -6.136 |
earchers | -6.026 |
Token | 2 |
Token ID | 362 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
BELOW | +5.835 |
idated | +5.593 |
productive | +5.504 |
ridge | +5.210 |
arrass | +5.144 |
Neg logit contributions | |
tti | -5.722 |
etz | -5.118 |
Probe | -5.003 |
neb | -4.983 |
earchers | -4.940 |
Token | cups |
Token ID | 14180 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
BELOW | +5.190 |
productive | +4.969 |
idated | +4.944 |
arrass | +4.565 |
ridge | +4.552 |
Neg logit contributions | |
tti | -6.199 |
etz | -5.669 |
neb | -5.594 |
Probe | -5.572 |
earchers | -5.534 |
Token | of |
Token ID | 286 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
BELOW | +11.634 |
idated | +10.970 |
productive | +10.834 |
ridge | +10.748 |
surplus | +10.105 |
Neg logit contributions | |
tti | -14.432 |
earchers | -13.496 |
Probe | -13.411 |
etz | -13.205 |
bourg | -12.805 |
Token | frozen |
Token ID | 12912 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
BELOW | +8.216 |
idated | +7.946 |
productive | +7.787 |
ridge | +7.461 |
arrass | +7.031 |
Neg logit contributions | |
tti | -9.561 |
etz | -8.672 |
earchers | -8.596 |
Probe | -8.477 |
bourg | -8.331 |
Token | mixed |
Token ID | 7668 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
BELOW | +7.691 |
idated | +7.220 |
productive | +6.983 |
ridge | +6.739 |
arrass | +6.523 |
Neg logit contributions | |
tti | -9.466 |
earchers | -8.510 |
etz | -8.490 |
Probe | -8.373 |
bourg | -8.021 |
Token | berries |
Token ID | 36322 |
Feature activation | +9.94e-01 |
Pos logit contributions | |
BELOW | +6.492 |
idated | +6.108 |
productive | +5.960 |
ridge | +5.752 |
arrass | +5.331 |
Neg logit contributions | |
tti | -9.816 |
etz | -9.002 |
Probe | -8.967 |
earchers | -8.873 |
emis | -8.625 |
Token | : |
Token ID | 25 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
BELOW | +6.235 |
idated | +5.903 |
productive | +5.816 |
ridge | +5.514 |
arrass | +5.383 |
Neg logit contributions | |
tti | -6.238 |
neb | -5.583 |
etz | -5.519 |
Probe | -5.498 |
earchers | -5.465 |
Token |
|
Token ID | 198 |
Feature activation | -5.86e-04 |
Pos logit contributions | |
BELOW | +2.057 |
idated | +1.976 |
productive | +1.946 |
arrass | +1.808 |
ridge | +1.790 |
Neg logit contributions | |
tti | -1.894 |
neb | -1.672 |
etz | -1.665 |
Probe | -1.610 |
mysteries | -1.597 |
Token |
|
Token ID | 198 |
Feature activation | +6.76e-02 |
Pos logit contributions | |
tti | +0.004 |
neb | +0.004 |
etz | +0.004 |
Probe | +0.004 |
earchers | +0.004 |
Neg logit contributions | |
BELOW | -0.003 |
idated | -0.003 |
productive | -0.003 |
arrass | -0.003 |
ridge | -0.003 |
Token | College |
Token ID | 38951 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
BELOW | +0.441 |
idated | +0.429 |
productive | +0.427 |
arrass | +0.394 |
ridge | +0.387 |
Neg logit contributions | |
tti | -0.441 |
neb | -0.388 |
etz | -0.386 |
Probe | -0.384 |
mysteries | -0.372 |
Token | career |
Token ID | 3451 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
BELOW | +5.066 |
productive | +4.919 |
idated | +4.883 |
arrass | +4.610 |
ridge | +4.578 |
Neg logit contributions | |
tti | -4.992 |
etz | -4.432 |
neb | -4.238 |
Probe | -4.218 |
earchers | -4.168 |
Token | [ |
Token ID | 685 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
BELOW | +3.298 |
productive | +3.138 |
idated | +3.129 |
arrass | +2.837 |
ridge | +2.799 |
Neg logit contributions | |
tti | -5.051 |
etz | -4.525 |
Probe | -4.491 |
neb | -4.470 |
earchers | -4.344 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
BELOW | +5.739 |
productive | +5.449 |
idated | +5.339 |
ridge | +4.805 |
below | +4.294 |
Neg logit contributions | |
tti | -11.481 |
earchers | -11.257 |
etz | -10.929 |
neb | -10.863 |
Probe | -10.664 |
Token | ] |
Token ID | 2361 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
BELOW | +6.087 |
idated | +3.780 |
productive | +3.695 |
ridge | +3.539 |
Loading | +3.188 |
Neg logit contributions | |
earchers | -22.008 |
mathemat | -21.087 |
tti | -20.778 |
cumbers | -20.756 |
ignt | -20.737 |
Token |
|
Token ID | 198 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
BELOW | +5.291 |
productive | +4.683 |
idated | +4.654 |
ridge | +4.138 |
Loading | +3.746 |
Neg logit contributions | |
tti | -10.270 |
earchers | -10.253 |
neb | -9.924 |
etz | -9.597 |
ignt | -9.457 |
Token |
|
Token ID | 198 |
Feature activation | +6.41e-01 |
Pos logit contributions | |
BELOW | +3.537 |
productive | +3.256 |
idated | +3.242 |
ridge | +2.904 |
arrass | +2.643 |
Neg logit contributions | |
tti | -6.274 |
earchers | -6.076 |
neb | -5.937 |
etz | -5.767 |
Probe | -5.725 |
Token | O |
Token ID | 46 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
BELOW | +3.989 |
productive | +3.927 |
idated | +3.887 |
arrass | +3.571 |
ridge | +3.560 |
Neg logit contributions | |
tti | -4.204 |
neb | -3.806 |
Probe | -3.792 |
etz | -3.716 |
earchers | -3.638 |
Token | ka |
Token ID | 4914 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
tti | +1.035 |
etz | +0.896 |
neb | +0.891 |
Probe | +0.880 |
mysteries | +0.852 |
Neg logit contributions | |
BELOW | -1.180 |
idated | -1.153 |
productive | -1.142 |
arrass | -1.060 |
ridge | -1.037 |
Token | done |
Token ID | 1760 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
tti | +3.426 |
neb | +3.027 |
etz | +2.986 |
Probe | +2.930 |
recognition | +2.864 |
Neg logit contributions | |
BELOW | -3.486 |
idated | -3.346 |
productive | -3.331 |
arrass | -3.147 |
ridge | -3.076 |
Token | , |
Token ID | 11 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
BELOW | +2.123 |
idated | +2.066 |
productive | +2.045 |
arrass | +1.881 |
ridge | +1.840 |
Neg logit contributions | |
tti | -2.335 |
etz | -2.061 |
neb | -2.057 |
Probe | -2.046 |
mysteries | -1.978 |
Token | would |
Token ID | 561 |
Feature activation | +9.58e-01 |
Pos logit contributions | |
BELOW | +3.683 |
idated | +3.602 |
productive | +3.579 |
arrass | +3.306 |
ridge | +3.240 |
Neg logit contributions | |
tti | -3.898 |
etz | -3.456 |
Probe | -3.423 |
neb | -3.410 |
earchers | -3.343 |
Token | likely |
Token ID | 1884 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
BELOW | +5.673 |
idated | +5.490 |
productive | +5.439 |
arrass | +5.021 |
ridge | +4.977 |
Neg logit contributions | |
tti | -6.374 |
etz | -5.709 |
earchers | -5.565 |
Probe | -5.559 |
neb | -5.489 |
Token | do |
Token ID | 466 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
BELOW | +6.694 |
idated | +6.427 |
productive | +6.356 |
deter | +5.995 |
arrass | +5.812 |
Neg logit contributions | |
tti | -8.453 |
etz | -7.655 |
earchers | -7.514 |
Probe | -7.432 |
mysteries | -7.264 |
Token | just |
Token ID | 655 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
BELOW | +8.565 |
productive | +8.416 |
idated | +8.258 |
arrass | +7.757 |
ridge | +7.587 |
Neg logit contributions | |
tti | -9.329 |
Probe | -8.341 |
neb | -8.085 |
etz | -8.081 |
bourg | -7.959 |
Token | that |
Token ID | 326 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
BELOW | +13.448 |
productive | +12.489 |
idated | +12.201 |
ridge | +11.262 |
enough | +11.131 |
Neg logit contributions | |
tti | -13.599 |
Probe | -12.772 |
earchers | -12.275 |
etz | -11.955 |
neb | -11.781 |
Token | . |
Token ID | 13 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
BELOW | +7.271 |
productive | +7.037 |
idated | +6.888 |
arrass | +6.393 |
ridge | +6.304 |
Neg logit contributions | |
tti | -7.280 |
Probe | -6.490 |
etz | -6.428 |
neb | -6.363 |
earchers | -6.339 |
Token |
|
Token ID | 198 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
BELOW | +3.149 |
idated | +3.046 |
productive | +2.996 |
arrass | +2.745 |
ridge | +2.727 |
Neg logit contributions | |
tti | -3.373 |
neb | -3.048 |
etz | -3.008 |
earchers | -2.923 |
mysteries | -2.898 |
Token |
|
Token ID | 198 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
tti | +1.293 |
etz | +1.144 |
neb | +1.141 |
Probe | +1.132 |
mysteries | +1.100 |
Neg logit contributions | |
BELOW | -1.073 |
idated | -1.044 |
productive | -1.033 |
arrass | -0.948 |
ridge | -0.924 |
Token | Large |
Token ID | 21968 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
BELOW | +1.296 |
idated | +1.264 |
productive | +1.255 |
arrass | +1.156 |
ridge | +1.133 |
Neg logit contributions | |
tti | -1.358 |
neb | -1.205 |
etz | -1.194 |
Probe | -1.187 |
mysteries | -1.152 |
Token |
|
Token ID | 198 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
tti | +0.872 |
neb | +0.774 |
etz | +0.774 |
Probe | +0.766 |
mysteries | +0.746 |
Neg logit contributions | |
BELOW | -0.702 |
idated | -0.681 |
productive | -0.674 |
arrass | -0.614 |
ridge | -0.603 |
Token | Results |
Token ID | 25468 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
tti | +1.581 |
etz | +1.380 |
neb | +1.376 |
Probe | +1.364 |
mysteries | +1.321 |
Neg logit contributions | |
BELOW | -1.588 |
idated | -1.547 |
productive | -1.530 |
arrass | -1.417 |
ridge | -1.385 |
Token | and |
Token ID | 290 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
BELOW | +0.314 |
idated | +0.304 |
productive | +0.300 |
arrass | +0.274 |
ridge | +0.268 |
Neg logit contributions | |
tti | -0.396 |
etz | -0.352 |
neb | -0.349 |
Probe | -0.346 |
mysteries | -0.338 |
Token | achievements |
Token ID | 16970 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
tti | +0.951 |
etz | +0.820 |
neb | +0.814 |
Probe | +0.806 |
mysteries | +0.777 |
Neg logit contributions | |
BELOW | -1.160 |
idated | -1.132 |
productive | -1.123 |
arrass | -1.046 |
ridge | -1.027 |
Token | [ |
Token ID | 685 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
tti | +0.672 |
etz | +0.598 |
neb | +0.597 |
Probe | +0.592 |
mysteries | +0.577 |
Neg logit contributions | |
BELOW | -0.499 |
idated | -0.484 |
productive | -0.478 |
arrass | -0.436 |
ridge | -0.425 |
Token | edit |
Token ID | 4370 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
BELOW | +6.325 |
productive | +5.963 |
idated | +5.673 |
ridge | +5.204 |
below | +4.995 |
Neg logit contributions | |
tti | -12.425 |
earchers | -12.025 |
etz | -11.915 |
Probe | -11.680 |
neb | -11.676 |
Token | ] |
Token ID | 2361 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
BELOW | +5.217 |
idated | +3.311 |
productive | +3.162 |
ridge | +3.102 |
Loading | +2.564 |
Neg logit contributions | |
earchers | -21.862 |
tti | -21.320 |
mathemat | -20.941 |
etz | -20.691 |
neb | -20.663 |
Token |
|
Token ID | 198 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
BELOW | +4.023 |
productive | +3.622 |
idated | +3.619 |
ridge | +3.177 |
arrass | +2.968 |
Neg logit contributions | |
tti | -6.754 |
earchers | -6.326 |
neb | -6.278 |
etz | -6.212 |
mysteries | -6.017 |
Token |
|
Token ID | 198 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
BELOW | +1.277 |
idated | +1.210 |
productive | +1.202 |
ridge | +1.074 |
arrass | +1.057 |
Neg logit contributions | |
tti | -1.829 |
neb | -1.666 |
etz | -1.649 |
earchers | -1.642 |
Probe | -1.635 |
Token | The |
Token ID | 464 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
tti | +1.898 |
etz | +1.649 |
neb | +1.646 |
Probe | +1.632 |
mysteries | +1.578 |
Neg logit contributions | |
BELOW | -1.954 |
idated | -1.902 |
productive | -1.875 |
arrass | -1.743 |
ridge | -1.702 |
Token | ion |
Token ID | 22088 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
tti | +3.251 |
neb | +2.859 |
Probe | +2.833 |
etz | +2.807 |
mysteries | +2.723 |
Neg logit contributions | |
BELOW | -3.543 |
idated | -3.449 |
productive | -3.427 |
arrass | -3.190 |
ridge | -3.117 |
Token | one |
Token ID | 530 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
BELOW | +4.794 |
idated | +4.576 |
productive | +4.505 |
arrass | +4.225 |
ridge | +4.193 |
Neg logit contributions | |
tti | -4.304 |
etz | -3.803 |
Probe | -3.718 |
neb | -3.663 |
earchers | -3.600 |
Token | , |
Token ID | 11 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
tti | +3.050 |
neb | +2.673 |
etz | +2.639 |
Probe | +2.613 |
mysteries | +2.540 |
Neg logit contributions | |
BELOW | -3.094 |
idated | -3.019 |
productive | -3.003 |
arrass | -2.775 |
ridge | -2.697 |
Token | I |
Token ID | 314 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
tti | +2.459 |
etz | +2.180 |
neb | +2.178 |
Probe | +2.157 |
mysteries | +2.103 |
Neg logit contributions | |
BELOW | -1.929 |
idated | -1.877 |
productive | -1.856 |
arrass | -1.694 |
ridge | -1.652 |
Token | cannot |
Token ID | 2314 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
tti | +0.894 |
etz | +0.777 |
neb | +0.774 |
Probe | +0.767 |
mysteries | +0.741 |
Neg logit contributions | |
BELOW | -1.000 |
idated | -0.975 |
productive | -0.966 |
arrass | -0.899 |
ridge | -0.881 |
Token | , |
Token ID | 11 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
tti | +0.604 |
etz | +0.519 |
neb | +0.516 |
Probe | +0.511 |
mysteries | +0.493 |
Neg logit contributions | |
BELOW | -0.777 |
idated | -0.757 |
productive | -0.751 |
arrass | -0.701 |
ridge | -0.690 |
Token | for |
Token ID | 329 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
tti | +3.390 |
neb | +3.036 |
etz | +3.015 |
Probe | +2.992 |
mysteries | +2.935 |
Neg logit contributions | |
BELOW | -2.420 |
idated | -2.361 |
productive | -2.324 |
arrass | -2.119 |
ridge | -2.043 |
Token | the |
Token ID | 262 |
Feature activation | -9.27e-01 |
Pos logit contributions | |
tti | +17.298 |
obvious | +15.984 |
recognition | +15.656 |
neb | +15.325 |
ked | +15.148 |
Neg logit contributions | |
idated | -16.236 |
覚醒 | -15.868 |
BELOW | -15.500 |
productive | -15.296 |
arrass | -14.545 |
Token | life |
Token ID | 1204 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
tti | +5.559 |
neb | +4.923 |
etz | +4.731 |
Probe | +4.713 |
mysteries | +4.709 |
Neg logit contributions | |
BELOW | -6.186 |
productive | -6.110 |
idated | -6.097 |
arrass | -5.728 |
ridge | -5.489 |
Token | of |
Token ID | 286 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
BELOW | +1.134 |
idated | +1.104 |
productive | +1.093 |
arrass | +1.009 |
ridge | +0.987 |
Neg logit contributions | |
tti | -1.180 |
etz | -1.035 |
neb | -1.034 |
Probe | -1.022 |
mysteries | -0.994 |
Token | me |
Token ID | 502 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
BELOW | +1.401 |
idated | +1.372 |
productive | +1.361 |
arrass | +1.280 |
ridge | +1.265 |
Neg logit contributions | |
tti | -0.785 |
etz | -0.648 |
neb | -0.645 |
Probe | -0.639 |
mysteries | -0.604 |
Token | , |
Token ID | 11 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
tti | +2.613 |
neb | +2.257 |
etz | +2.247 |
Probe | +2.194 |
recognition | +2.181 |
Neg logit contributions | |
BELOW | -2.707 |
idated | -2.666 |
productive | -2.607 |
arrass | -2.425 |
ridge | -2.362 |
Token | preserved |
Token ID | 17232 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
tti | +2.380 |
neb | +2.042 |
etz | +2.038 |
Probe | +1.991 |
mysteries | +1.953 |
Neg logit contributions | |
BELOW | -3.064 |
idated | -3.008 |
productive | -2.971 |
arrass | -2.790 |
ridge | -2.725 |
Token | that |
Token ID | 326 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
tti | +7.615 |
neb | +6.653 |
mysteries | +6.555 |
recognition | +6.446 |
etz | +6.368 |
Neg logit contributions | |
BELOW | -9.306 |
productive | -9.146 |
idated | -9.116 |
arrass | -8.717 |
覚醒 | -8.348 |
Token | researchers |
Token ID | 4837 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
tti | +5.697 |
neb | +4.973 |
Probe | +4.898 |
mysteries | +4.887 |
etz | +4.862 |
Neg logit contributions | |
BELOW | -7.333 |
productive | -7.174 |
idated | -7.120 |
arrass | -6.649 |
ridge | -6.471 |
Token | have |
Token ID | 423 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
tti | +4.698 |
etz | +4.025 |
neb | +4.014 |
Probe | +4.009 |
mysteries | +3.884 |
Neg logit contributions | |
BELOW | -6.632 |
idated | -6.515 |
productive | -6.453 |
arrass | -6.009 |
ridge | -5.933 |
Token | seen |
Token ID | 1775 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
tti | +3.572 |
neb | +3.108 |
etz | +3.065 |
Probe | +3.042 |
mysteries | +3.016 |
Neg logit contributions | |
BELOW | -4.108 |
idated | -4.002 |
productive | -3.968 |
arrass | -3.707 |
ridge | -3.652 |
Token | for |
Token ID | 329 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
BELOW | +0.607 |
idated | +0.586 |
productive | +0.581 |
arrass | +0.539 |
ridge | +0.533 |
Neg logit contributions | |
tti | -0.550 |
etz | -0.480 |
Probe | -0.473 |
neb | -0.468 |
Television | -0.450 |
Token | the |
Token ID | 262 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
tti | +16.523 |
ked | +14.867 |
recognition | +14.673 |
decades | +14.530 |
years | +14.422 |
Neg logit contributions | |
BELOW | -17.377 |
idated | -16.520 |
覚醒 | -16.258 |
productive | -16.164 |
bleacher | -15.596 |
Token | first |
Token ID | 717 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
tti | +7.732 |
neb | +7.079 |
mysteries | +6.876 |
recognition | +6.809 |
Probe | +6.789 |
Neg logit contributions | |
BELOW | -6.798 |
productive | -6.627 |
idated | -6.619 |
arrass | -6.121 |
覚醒 | -5.820 |
Token | time |
Token ID | 640 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
tti | +3.881 |
neb | +3.311 |
Probe | +3.128 |
recognition | +3.109 |
mysteries | +3.039 |
Neg logit contributions | |
BELOW | -8.316 |
idated | -8.146 |
productive | -8.095 |
arrass | -7.695 |
itably | -7.423 |
Token | two |
Token ID | 734 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
tti | +3.639 |
neb | +3.174 |
etz | +3.115 |
Probe | +3.094 |
mysteries | +3.054 |
Neg logit contributions | |
BELOW | -4.619 |
idated | -4.548 |
productive | -4.527 |
arrass | -4.231 |
ridge | -4.096 |
Token | fl |
Token ID | 781 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
tti | +3.347 |
neb | +2.936 |
etz | +2.869 |
Probe | +2.850 |
mysteries | +2.840 |
Neg logit contributions | |
BELOW | -3.646 |
idated | -3.592 |
productive | -3.540 |
arrass | -3.327 |
ridge | -3.218 |
Token | claims |
Token ID | 3667 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
tti | +2.515 |
neb | +2.152 |
etz | +2.138 |
Probe | +2.097 |
mysteries | +2.064 |
Neg logit contributions | |
BELOW | -3.633 |
idated | -3.576 |
productive | -3.539 |
arrass | -3.309 |
ridge | -3.248 |
Token | that |
Token ID | 326 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
tti | +2.814 |
etz | +2.436 |
neb | +2.424 |
Probe | +2.373 |
mysteries | +2.310 |
Neg logit contributions | |
BELOW | -3.477 |
idated | -3.417 |
productive | -3.383 |
arrass | -3.156 |
ridge | -3.087 |
Token | have |
Token ID | 423 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
BELOW | +1.170 |
idated | +1.142 |
productive | +1.125 |
arrass | +1.046 |
ridge | +1.031 |
Neg logit contributions | |
tti | -1.088 |
etz | -0.943 |
neb | -0.941 |
Probe | -0.935 |
mysteries | -0.902 |
Token | been |
Token ID | 587 |
Feature activation | -9.87e-01 |
Pos logit contributions | |
BELOW | +0.800 |
idated | +0.780 |
productive | +0.770 |
arrass | +0.713 |
ridge | +0.700 |
Neg logit contributions | |
tti | -0.781 |
etz | -0.686 |
Probe | -0.680 |
neb | -0.675 |
mysteries | -0.649 |
Token | pending |
Token ID | 13310 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
tti | +5.049 |
neb | +4.390 |
etz | +4.278 |
mysteries | +4.140 |
Probe | +4.110 |
Neg logit contributions | |
BELOW | -7.420 |
idated | -7.247 |
productive | -7.224 |
arrass | -6.747 |
ridge | -6.610 |
Token | for |
Token ID | 329 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
tti | +9.765 |
neb | +8.811 |
recognition | +8.609 |
mysteries | +8.556 |
etz | +8.476 |
Neg logit contributions | |
idated | -12.664 |
productive | -12.368 |
BELOW | -12.308 |
覚醒 | -11.720 |
arrass | -11.411 |
Token | at |
Token ID | 379 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
tti | +12.710 |
years | +12.304 |
decades | +12.303 |
neb | +11.669 |
recognition | +11.542 |
Neg logit contributions | |
idated | -18.968 |
覚醒 | -18.527 |
BELOW | -17.792 |
productive | -17.563 |
arrass | -17.266 |
Token | least |
Token ID | 1551 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
tti | +3.381 |
etz | +3.104 |
neb | +3.094 |
Probe | +3.069 |
mysteries | +2.999 |
Neg logit contributions | |
BELOW | -1.463 |
idated | -1.397 |
productive | -1.383 |
arrass | -1.211 |
ridge | -1.151 |
Token | a |
Token ID | 257 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
tti | +7.242 |
neb | +6.382 |
Probe | +6.140 |
recognition | +6.096 |
etz | +5.944 |
Neg logit contributions | |
BELOW | -10.041 |
idated | -10.021 |
productive | -9.762 |
覚醒 | -9.445 |
arrass | -9.399 |
Token | decade |
Token ID | 5707 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
tti | +4.218 |
neb | +3.792 |
recognition | +3.429 |
Probe | +3.264 |
etz | +3.124 |
Neg logit contributions | |
idated | -11.576 |
BELOW | -11.517 |
productive | -11.238 |
arrass | -11.063 |
覚醒 | -10.820 |
Token | — |
Token ID | 851 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
tti | +4.181 |
etz | +3.538 |
neb | +3.520 |
Probe | +3.420 |
mysteries | +3.376 |
Neg logit contributions | |
BELOW | -6.052 |
idated | -6.013 |
productive | -5.957 |
arrass | -5.574 |
ridge | -5.398 |
Token | nineteenth |
Token ID | 29666 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
tti | +1.549 |
etz | +1.385 |
neb | +1.378 |
Probe | +1.370 |
mysteries | +1.332 |
Neg logit contributions | |
BELOW | -1.077 |
idated | -1.043 |
productive | -1.030 |
arrass | -0.933 |
ridge | -0.911 |
Token | century |
Token ID | 4289 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
tti | +2.744 |
neb | +2.277 |
etz | +2.208 |
Probe | +2.151 |
mysteries | +2.147 |
Neg logit contributions | |
BELOW | -5.886 |
productive | -5.760 |
idated | -5.742 |
arrass | -5.459 |
itably | -5.242 |
Token | , |
Token ID | 11 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
BELOW | +0.545 |
idated | +0.530 |
productive | +0.524 |
arrass | +0.483 |
ridge | +0.474 |
Neg logit contributions | |
tti | -0.585 |
etz | -0.517 |
neb | -0.512 |
Probe | -0.510 |
mysteries | -0.490 |
Token | Europe |
Token ID | 2031 |
Feature activation | -4.70e-03 |
Pos logit contributions | |
tti | +2.540 |
neb | +2.222 |
etz | +2.200 |
Probe | +2.186 |
mysteries | +2.144 |
Neg logit contributions | |
BELOW | -2.601 |
idated | -2.533 |
productive | -2.515 |
arrass | -2.329 |
ridge | -2.274 |
Token | had |
Token ID | 550 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
tti | +0.029 |
etz | +0.025 |
neb | +0.025 |
Probe | +0.025 |
mysteries | +0.024 |
Neg logit contributions | |
BELOW | -0.032 |
idated | -0.031 |
productive | -0.031 |
arrass | -0.029 |
ridge | -0.028 |
Token | for |
Token ID | 329 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
BELOW | +0.962 |
idated | +0.940 |
productive | +0.927 |
arrass | +0.857 |
ridge | +0.842 |
Neg logit contributions | |
tti | -0.893 |
etz | -0.785 |
Probe | -0.778 |
neb | -0.769 |
earchers | -0.735 |
Token | the |
Token ID | 262 |
Feature activation | -7.25e-01 |
Pos logit contributions | |
tti | +16.731 |
ked | +15.583 |
recognition | +14.827 |
mysteries | +14.649 |
neb | +14.524 |
Neg logit contributions | |
BELOW | -16.204 |
覚醒 | -14.337 |
idated | -14.251 |
productive | -13.969 |
arrass | -13.945 |
Token | first |
Token ID | 717 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
tti | +5.183 |
neb | +4.705 |
etz | +4.578 |
mysteries | +4.545 |
Probe | +4.533 |
Neg logit contributions | |
BELOW | -4.156 |
idated | -4.005 |
productive | -3.982 |
arrass | -3.731 |
ridge | -3.523 |
Token | time |
Token ID | 640 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
tti | +3.757 |
neb | +3.181 |
recognition | +3.001 |
etz | +2.978 |
Probe | +2.971 |
Neg logit contributions | |
BELOW | -7.452 |
idated | -7.300 |
productive | -7.247 |
arrass | -6.886 |
ridge | -6.637 |
Token | come |
Token ID | 1282 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
tti | +3.820 |
neb | +3.379 |
etz | +3.317 |
mysteries | +3.243 |
recognition | +3.242 |
Neg logit contributions | |
BELOW | -4.162 |
idated | -4.065 |
productive | -4.055 |
arrass | -3.774 |
ridge | -3.678 |
Token | to |
Token ID | 284 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
tti | +4.245 |
neb | +3.656 |
etz | +3.609 |
recognition | +3.541 |
Probe | +3.500 |
Neg logit contributions | |
BELOW | -5.354 |
idated | -5.335 |
productive | -5.241 |
arrass | -4.939 |
ridge | -4.769 |
Token | I |
Token ID | 40 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
tti | +0.893 |
etz | +0.776 |
neb | +0.776 |
Probe | +0.767 |
mysteries | +0.741 |
Neg logit contributions | |
BELOW | -0.997 |
idated | -0.971 |
productive | -0.964 |
arrass | -0.894 |
ridge | -0.878 |
Token | � |
Token ID | 447 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
tti | +1.877 |
etz | +1.598 |
neb | +1.592 |
Probe | +1.576 |
mysteries | +1.517 |
Neg logit contributions | |
BELOW | -2.443 |
idated | -2.383 |
productive | -2.369 |
arrass | -2.207 |
ridge | -2.167 |
Token | � |
Token ID | 247 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
tti | +3.333 |
neb | +2.980 |
etz | +2.954 |
Probe | +2.953 |
mysteries | +2.877 |
Neg logit contributions | |
BELOW | -2.342 |
idated | -2.264 |
productive | -2.239 |
arrass | -2.017 |
ridge | -1.984 |
Token | ve |
Token ID | 303 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
tti | +2.424 |
etz | +2.062 |
neb | +2.044 |
Probe | +2.029 |
mysteries | +1.946 |
Neg logit contributions | |
BELOW | -3.140 |
idated | -3.063 |
productive | -3.033 |
arrass | -2.838 |
ridge | -2.777 |
Token | said |
Token ID | 531 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
BELOW | +0.707 |
idated | +0.691 |
productive | +0.681 |
arrass | +0.632 |
ridge | +0.622 |
Neg logit contributions | |
tti | -0.598 |
etz | -0.527 |
neb | -0.521 |
Probe | -0.520 |
mysteries | -0.495 |
Token | for |
Token ID | 329 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
BELOW | +4.549 |
idated | +4.383 |
productive | +4.337 |
arrass | +4.106 |
ridge | +3.951 |
Neg logit contributions | |
tti | -4.000 |
etz | -3.584 |
Probe | -3.503 |
neb | -3.443 |
earchers | -3.350 |
Token | a |
Token ID | 257 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
tti | +15.081 |
years | +13.403 |
neb | +13.278 |
recognition | +13.133 |
ked | +12.707 |
Neg logit contributions | |
idated | -17.113 |
BELOW | -16.225 |
productive | -15.953 |
覚醒 | -15.698 |
arrass | -14.936 |
Token | couple |
Token ID | 3155 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
tti | +5.387 |
neb | +4.718 |
Probe | +4.517 |
recognition | +4.504 |
etz | +4.488 |
Neg logit contributions | |
BELOW | -6.245 |
idated | -6.106 |
productive | -5.983 |
arrass | -5.683 |
ridge | -5.490 |
Token | of |
Token ID | 286 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
tti | +6.446 |
neb | +5.496 |
mysteries | +5.320 |
Television | +5.082 |
recognition | +5.009 |
Neg logit contributions | |
BELOW | -8.919 |
idated | -8.718 |
productive | -8.542 |
arrass | -8.199 |
ridge | -8.116 |
Token | years |
Token ID | 812 |
Feature activation | +9.15e-03 |
Pos logit contributions | |
tti | +5.183 |
neb | +4.391 |
mysteries | +4.126 |
recognition | +3.914 |
Probe | +3.875 |
Neg logit contributions | |
BELOW | -10.496 |
idated | -10.344 |
productive | -10.152 |
arrass | -9.739 |
ridge | -9.707 |
Token | I |
Token ID | 314 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
BELOW | +0.064 |
idated | +0.062 |
productive | +0.062 |
arrass | +0.058 |
ridge | +0.057 |
Neg logit contributions | |
tti | -0.055 |
etz | -0.048 |
neb | -0.048 |
Probe | -0.047 |
mysteries | -0.046 |
Token | � |
Token ID | 564 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
BELOW | +3.341 |
idated | +3.223 |
productive | +3.205 |
arrass | +2.955 |
ridge | +2.900 |
Neg logit contributions | |
tti | -3.530 |
etz | -3.108 |
neb | -3.057 |
Probe | -3.056 |
earchers | -2.971 |
Token | � |
Token ID | 250 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
BELOW | +1.783 |
idated | +1.731 |
productive | +1.717 |
arrass | +1.555 |
ridge | +1.516 |
Neg logit contributions | |
tti | -2.409 |
etz | -2.159 |
neb | -2.139 |
Probe | -2.130 |
earchers | -2.080 |
Token | I |
Token ID | 40 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
BELOW | +0.512 |
productive | +0.499 |
idated | +0.498 |
arrass | +0.461 |
ridge | +0.452 |
Neg logit contributions | |
tti | -0.470 |
etz | -0.409 |
neb | -0.409 |
Probe | -0.405 |
mysteries | -0.392 |
Token | have |
Token ID | 423 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
BELOW | +1.559 |
idated | +1.526 |
productive | +1.504 |
arrass | +1.414 |
ridge | +1.386 |
Neg logit contributions | |
tti | -1.226 |
neb | -1.071 |
etz | -1.062 |
Probe | -1.049 |
earchers | -1.026 |
Token | said |
Token ID | 531 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
BELOW | +5.847 |
idated | +5.640 |
productive | +5.608 |
arrass | +5.076 |
ridge | +4.922 |
Neg logit contributions | |
tti | -5.321 |
etz | -4.825 |
Probe | -4.731 |
neb | -4.706 |
earchers | -4.603 |
Token | for |
Token ID | 329 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
BELOW | +2.960 |
idated | +2.837 |
productive | +2.832 |
arrass | +2.663 |
ridge | +2.562 |
Neg logit contributions | |
tti | -2.690 |
etz | -2.353 |
neb | -2.304 |
Probe | -2.297 |
mysteries | -2.224 |
Token | more |
Token ID | 517 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
tti | +13.301 |
years | +12.655 |
recognition | +12.517 |
neb | +12.229 |
decades | +12.154 |
Neg logit contributions | |
idated | -17.761 |
BELOW | -16.329 |
productive | -16.239 |
覚醒 | -16.195 |
arrass | -15.589 |
Token | than |
Token ID | 621 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
tti | +6.832 |
neb | +5.672 |
Than | +5.209 |
recognition | +5.193 |
mysteries | +4.781 |
Neg logit contributions | |
idated | -20.130 |
BELOW | -18.970 |
覚醒 | -18.921 |
arrass | -18.557 |
productive | -18.553 |
Token | 22 |
Token ID | 2534 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
tti | +5.307 |
neb | +4.696 |
Probe | +4.510 |
recognition | +4.428 |
etz | +4.416 |
Neg logit contributions | |
idated | -7.345 |
BELOW | -7.278 |
productive | -7.098 |
arrass | -6.631 |
ridge | -6.575 |
Token | years |
Token ID | 812 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
tti | +2.196 |
neb | +1.579 |
Probe | +1.441 |
Television | +1.394 |
mysteries | +1.390 |
Neg logit contributions | |
idated | -9.548 |
BELOW | -9.459 |
productive | -9.212 |
ridge | -8.981 |
arrass | -8.928 |
Token | that |
Token ID | 326 |
Feature activation | +5.32e-02 |
Pos logit contributions | |
tti | +0.071 |
etz | +0.062 |
neb | +0.062 |
Probe | +0.061 |
mysteries | +0.059 |
Neg logit contributions | |
BELOW | -0.073 |
idated | -0.071 |
productive | -0.071 |
arrass | -0.066 |
ridge | -0.064 |
Token | very |
Token ID | 845 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
BELOW | +2.563 |
productive | +2.429 |
idated | +2.425 |
arrass | +2.258 |
ridge | +2.221 |
Neg logit contributions | |
tti | -2.452 |
etz | -2.143 |
Probe | -2.123 |
neb | -2.096 |
mysteries | -2.021 |
Token | same |
Token ID | 976 |
Feature activation | +7.22e-01 |
Pos logit contributions | |
BELOW | +3.197 |
productive | +3.075 |
idated | +3.048 |
arrass | +2.824 |
ridge | +2.786 |
Neg logit contributions | |
tti | -3.246 |
etz | -2.856 |
Probe | -2.818 |
neb | -2.742 |
mysteries | -2.697 |
Token | thing |
Token ID | 1517 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
BELOW | +4.731 |
productive | +4.508 |
idated | +4.481 |
arrass | +4.197 |
ridge | +4.104 |
Neg logit contributions | |
tti | -4.513 |
etz | -3.924 |
Probe | -3.882 |
neb | -3.726 |
earchers | -3.719 |
Token | , |
Token ID | 11 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
tti | +0.874 |
etz | +0.759 |
neb | +0.757 |
Probe | +0.750 |
mysteries | +0.725 |
Neg logit contributions | |
BELOW | -0.959 |
idated | -0.934 |
productive | -0.926 |
arrass | -0.860 |
ridge | -0.843 |
Token | without |
Token ID | 1231 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
tti | +1.481 |
etz | +1.288 |
neb | +1.282 |
Probe | +1.272 |
mysteries | +1.230 |
Neg logit contributions | |
BELOW | -1.599 |
idated | -1.556 |
productive | -1.543 |
arrass | -1.432 |
ridge | -1.404 |
Token | for |
Token ID | 329 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
BELOW | +1.495 |
idated | +1.439 |
productive | +1.428 |
arrass | +1.328 |
ridge | +1.293 |
Neg logit contributions | |
tti | -1.588 |
etz | -1.393 |
Probe | -1.362 |
neb | -1.354 |
earchers | -1.313 |
Token | king |
Token ID | 3364 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
tti | +15.309 |
ked | +14.073 |
bes | +13.184 |
ft | +13.083 |
etz | +13.042 |
Neg logit contributions | |
idated | -16.574 |
覚醒 | -16.156 |
BELOW | -15.826 |
productive | -15.101 |
arrass | -13.982 |
Token | over |
Token ID | 625 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
tti | +1.609 |
etz | +1.388 |
neb | +1.386 |
Probe | +1.370 |
mysteries | +1.324 |
Neg logit contributions | |
BELOW | -1.935 |
idated | -1.893 |
productive | -1.875 |
arrass | -1.747 |
ridge | -1.713 |
Token | $ |
Token ID | 720 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
tti | +6.725 |
neb | +6.077 |
mysteries | +5.791 |
recognition | +5.777 |
Probe | +5.726 |
Neg logit contributions | |
idated | -8.420 |
BELOW | -8.266 |
productive | -8.172 |
覚醒 | -7.705 |
arrass | -7.668 |
Token | 75 |
Token ID | 2425 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
tti | +2.506 |
etz | +2.180 |
Probe | +2.159 |
neb | +2.149 |
mysteries | +2.080 |
Neg logit contributions | |
BELOW | -2.530 |
idated | -2.494 |
productive | -2.455 |
arrass | -2.287 |
ridge | -2.218 |
Token | a |
Token ID | 257 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
BELOW | +0.291 |
idated | +0.281 |
productive | +0.278 |
arrass | +0.259 |
ridge | +0.255 |
Neg logit contributions | |
tti | -0.257 |
etz | -0.223 |
neb | -0.222 |
Probe | -0.220 |
mysteries | -0.212 |
Token | have |
Token ID | 423 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
tti | +0.659 |
etz | +0.568 |
neb | +0.567 |
Probe | +0.561 |
mysteries | +0.542 |
Neg logit contributions | |
BELOW | -0.781 |
idated | -0.762 |
productive | -0.755 |
arrass | -0.703 |
ridge | -0.690 |
Token | steadfast |
Token ID | 35690 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
BELOW | +1.548 |
idated | +1.503 |
productive | +1.488 |
arrass | +1.373 |
ridge | +1.339 |
Neg logit contributions | |
tti | -1.468 |
etz | -1.288 |
Probe | -1.286 |
neb | -1.265 |
earchers | -1.234 |
Token | ly |
Token ID | 306 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
BELOW | +2.624 |
idated | +2.590 |
productive | +2.531 |
arrass | +2.284 |
ridge | +2.218 |
Neg logit contributions | |
tti | -3.830 |
Probe | -3.475 |
etz | -3.469 |
neb | -3.429 |
earchers | -3.348 |
Token | avoided |
Token ID | 13941 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
tti | +1.044 |
etz | +0.919 |
neb | +0.918 |
Probe | +0.907 |
mysteries | +0.883 |
Neg logit contributions | |
BELOW | -0.943 |
idated | -0.917 |
productive | -0.909 |
arrass | -0.836 |
ridge | -0.818 |
Token | it |
Token ID | 340 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
tti | +3.370 |
neb | +3.025 |
etz | +2.968 |
Probe | +2.921 |
recognition | +2.904 |
Neg logit contributions | |
BELOW | -3.187 |
idated | -3.151 |
productive | -3.089 |
arrass | -2.853 |
ridge | -2.799 |
Token | for |
Token ID | 329 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
tti | +2.293 |
neb | +1.976 |
etz | +1.974 |
Probe | +1.914 |
recognition | +1.881 |
Neg logit contributions | |
BELOW | -2.997 |
idated | -2.942 |
productive | -2.900 |
arrass | -2.719 |
ridge | -2.669 |
Token | as |
Token ID | 355 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
tti | +13.881 |
neb | +12.920 |
recognition | +12.838 |
years | +12.513 |
obvious | +12.509 |
Neg logit contributions | |
idated | -16.892 |
BELOW | -16.310 |
覚醒 | -15.932 |
productive | -15.709 |
arrass | -15.217 |
Token | long |
Token ID | 890 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
tti | +1.750 |
neb | +1.488 |
etz | +1.470 |
Probe | +1.444 |
mysteries | +1.400 |
Neg logit contributions | |
BELOW | -2.582 |
idated | -2.518 |
productive | -2.482 |
arrass | -2.339 |
ridge | -2.300 |
Token | as |
Token ID | 355 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
tti | +3.611 |
neb | +3.124 |
etz | +3.011 |
mysteries | +2.967 |
recognition | +2.944 |
Neg logit contributions | |
BELOW | -6.395 |
idated | -6.248 |
productive | -6.158 |
arrass | -5.821 |
ridge | -5.776 |
Token | they |
Token ID | 484 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
tti | +2.391 |
neb | +1.999 |
etz | +1.994 |
Probe | +1.978 |
mysteries | +1.900 |
Neg logit contributions | |
BELOW | -4.062 |
idated | -3.996 |
productive | -3.951 |
arrass | -3.713 |
ridge | -3.670 |
Token | can |
Token ID | 460 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
tti | +1.075 |
etz | +0.930 |
neb | +0.926 |
Probe | +0.917 |
mysteries | +0.886 |
Neg logit contributions | |
BELOW | -1.251 |
idated | -1.222 |
productive | -1.211 |
arrass | -1.126 |
ridge | -1.105 |
Token | lost |
Token ID | 2626 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
tti | +0.402 |
neb | +0.350 |
etz | +0.350 |
Probe | +0.346 |
mysteries | +0.335 |
Neg logit contributions | |
BELOW | -0.443 |
idated | -0.433 |
productive | -0.427 |
arrass | -0.397 |
ridge | -0.390 |
Token | at |
Token ID | 379 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
tti | +4.326 |
neb | +3.715 |
Probe | +3.700 |
recognition | +3.669 |
etz | +3.651 |
Neg logit contributions | |
BELOW | -5.050 |
productive | -4.937 |
idated | -4.927 |
arrass | -4.605 |
ridge | -4.447 |
Token | Crystal |
Token ID | 12969 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
tti | +0.914 |
etz | +0.805 |
neb | +0.804 |
Probe | +0.799 |
mysteries | +0.774 |
Neg logit contributions | |
BELOW | -0.810 |
idated | -0.786 |
productive | -0.782 |
arrass | -0.719 |
ridge | -0.702 |
Token | Palace |
Token ID | 15301 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
tti | +0.061 |
etz | +0.052 |
neb | +0.052 |
Probe | +0.051 |
mysteries | +0.050 |
Neg logit contributions | |
BELOW | -0.080 |
idated | -0.078 |
productive | -0.077 |
arrass | -0.072 |
ridge | -0.071 |
Token | , |
Token ID | 11 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
tti | +0.289 |
etz | +0.252 |
neb | +0.252 |
Probe | +0.249 |
mysteries | +0.240 |
Neg logit contributions | |
BELOW | -0.322 |
idated | -0.314 |
productive | -0.311 |
arrass | -0.289 |
ridge | -0.284 |
Token | for |
Token ID | 329 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
tti | +0.210 |
etz | +0.182 |
neb | +0.181 |
Probe | +0.180 |
mysteries | +0.173 |
Neg logit contributions | |
BELOW | -0.251 |
idated | -0.246 |
productive | -0.243 |
arrass | -0.227 |
ridge | -0.223 |
Token | example |
Token ID | 1672 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
tti | +13.717 |
obvious | +12.016 |
recognition | +11.907 |
ked | +11.611 |
bes | +11.318 |
Neg logit contributions | |
BELOW | -17.301 |
覚醒 | -16.986 |
idated | -16.666 |
productive | -16.270 |
arrass | -15.653 |
Token | , |
Token ID | 11 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
BELOW | +1.014 |
idated | +0.986 |
productive | +0.975 |
arrass | +0.899 |
ridge | +0.881 |
Neg logit contributions | |
tti | -1.132 |
etz | -1.003 |
neb | -0.999 |
Probe | -0.992 |
mysteries | -0.960 |
Token | the |
Token ID | 262 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
tti | +0.569 |
etz | +0.491 |
neb | +0.489 |
Probe | +0.484 |
mysteries | +0.466 |
Neg logit contributions | |
BELOW | -0.701 |
idated | -0.685 |
productive | -0.679 |
arrass | -0.633 |
ridge | -0.622 |
Token | media |
Token ID | 2056 |
Feature activation | -5.50e-01 |
Pos logit contributions | |
tti | +0.534 |
etz | +0.468 |
neb | +0.466 |
Probe | +0.462 |
mysteries | +0.447 |
Neg logit contributions | |
BELOW | -0.529 |
idated | -0.516 |
productive | -0.511 |
arrass | -0.472 |
ridge | -0.462 |
Token | was |
Token ID | 373 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
tti | +3.308 |
neb | +2.872 |
etz | +2.859 |
Probe | +2.846 |
recognition | +2.770 |
Neg logit contributions | |
BELOW | -3.818 |
idated | -3.738 |
productive | -3.725 |
arrass | -3.450 |
ridge | -3.372 |
Token | eye |
Token ID | 4151 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
tti | +0.992 |
neb | +0.819 |
etz | +0.819 |
Probe | +0.805 |
mysteries | +0.772 |
Neg logit contributions | |
BELOW | -1.775 |
idated | -1.747 |
productive | -1.732 |
arrass | -1.632 |
ridge | -1.606 |
Token | again |
Token ID | 757 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
BELOW | +1.067 |
idated | +1.040 |
productive | +1.032 |
arrass | +0.960 |
ridge | +0.941 |
Neg logit contributions | |
tti | -0.962 |
etz | -0.837 |
neb | -0.832 |
Probe | -0.825 |
mysteries | -0.795 |
Token | ? |
Token ID | 30 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
BELOW | +0.087 |
idated | +0.085 |
productive | +0.084 |
arrass | +0.079 |
ridge | +0.077 |
Neg logit contributions | |
tti | -0.075 |
etz | -0.065 |
neb | -0.065 |
Probe | -0.064 |
mysteries | -0.062 |
Token | She |
Token ID | 1375 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
BELOW | +0.902 |
idated | +0.873 |
productive | +0.865 |
arrass | +0.803 |
ridge | +0.796 |
Neg logit contributions | |
tti | -0.760 |
etz | -0.663 |
neb | -0.662 |
Probe | -0.641 |
earchers | -0.628 |
Token | thinks |
Token ID | 6834 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
tti | +1.626 |
etz | +1.399 |
neb | +1.398 |
Probe | +1.370 |
mysteries | +1.337 |
Neg logit contributions | |
BELOW | -1.991 |
idated | -1.947 |
productive | -1.929 |
arrass | -1.796 |
ridge | -1.755 |
Token | for |
Token ID | 329 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
tti | +4.554 |
neb | +3.904 |
etz | +3.810 |
Probe | +3.781 |
mysteries | +3.733 |
Neg logit contributions | |
BELOW | -6.073 |
idated | -6.056 |
productive | -5.968 |
arrass | -5.542 |
ridge | -5.434 |
Token | a |
Token ID | 257 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
tti | +16.777 |
recognition | +14.778 |
neb | +14.711 |
bes | +14.699 |
etz | +14.430 |
Neg logit contributions | |
idated | -14.378 |
productive | -14.117 |
BELOW | -13.904 |
覚醒 | -13.083 |
arrass | -12.705 |
Token | moment |
Token ID | 2589 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
tti | +3.527 |
neb | +3.050 |
etz | +2.963 |
Probe | +2.936 |
recognition | +2.889 |
Neg logit contributions | |
BELOW | -4.899 |
idated | -4.804 |
productive | -4.756 |
arrass | -4.460 |
ridge | -4.358 |
Token | . |
Token ID | 13 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
tti | +2.014 |
neb | +1.741 |
etz | +1.733 |
Probe | +1.712 |
mysteries | +1.666 |
Neg logit contributions | |
BELOW | -2.427 |
idated | -2.386 |
productive | -2.357 |
arrass | -2.195 |
ridge | -2.144 |
Token |
|
Token ID | 198 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
BELOW | +0.298 |
idated | +0.288 |
productive | +0.285 |
arrass | +0.264 |
ridge | +0.262 |
Neg logit contributions | |
tti | -0.260 |
neb | -0.227 |
etz | -0.227 |
Probe | -0.220 |
mysteries | -0.216 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
BELOW | +1.971 |
idated | +1.861 |
productive | +1.845 |
ridge | +1.648 |
arrass | +1.633 |
Neg logit contributions | |
tti | -2.842 |
neb | -2.624 |
etz | -2.556 |
earchers | -2.545 |
Probe | -2.512 |
Token | City |
Token ID | 2254 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
BELOW | +2.116 |
idated | +2.052 |
productive | +2.025 |
arrass | +1.859 |
ridge | +1.838 |
Neg logit contributions | |
tti | -2.230 |
etz | -1.981 |
neb | -1.969 |
Probe | -1.938 |
mysteries | -1.882 |
Token | Council |
Token ID | 4281 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
tti | +2.051 |
etz | +1.741 |
neb | +1.736 |
Probe | +1.721 |
Television | +1.670 |
Neg logit contributions | |
BELOW | -2.567 |
idated | -2.499 |
productive | -2.485 |
arrass | -2.326 |
ridge | -2.273 |
Token | has |
Token ID | 468 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
BELOW | +0.870 |
idated | +0.854 |
productive | +0.846 |
arrass | +0.778 |
ridge | +0.766 |
Neg logit contributions | |
tti | -0.835 |
neb | -0.723 |
etz | -0.720 |
Probe | -0.711 |
mysteries | -0.694 |
Token | been |
Token ID | 587 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
tti | +0.960 |
etz | +0.835 |
neb | +0.834 |
Probe | +0.825 |
mysteries | +0.799 |
Neg logit contributions | |
BELOW | -1.034 |
idated | -1.007 |
productive | -0.998 |
arrass | -0.926 |
ridge | -0.908 |
Token | trying |
Token ID | 2111 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
tti | +3.765 |
neb | +3.362 |
etz | +3.336 |
Probe | +3.296 |
mysteries | +3.215 |
Neg logit contributions | |
BELOW | -3.969 |
idated | -3.858 |
productive | -3.813 |
arrass | -3.544 |
ridge | -3.464 |
Token | for |
Token ID | 329 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
tti | +6.254 |
etz | +5.503 |
neb | +5.480 |
recognition | +5.262 |
Probe | +5.249 |
Neg logit contributions | |
BELOW | -9.839 |
idated | -9.789 |
productive | -9.595 |
arrass | -8.979 |
ridge | -8.936 |
Token | six |
Token ID | 2237 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
tti | +12.384 |
recognition | +11.636 |
years | +11.134 |
neb | +11.119 |
Probe | +10.969 |
Neg logit contributions | |
覚醒 | -17.464 |
BELOW | -17.228 |
idated | -17.170 |
productive | -16.797 |
itably | -15.828 |
Token | years |
Token ID | 812 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
tti | +1.466 |
neb | +0.917 |
mysteries | +0.778 |
etz | +0.687 |
recognition | +0.622 |
Neg logit contributions | |
BELOW | -11.094 |
idated | -10.899 |
productive | -10.679 |
arrass | -10.518 |
ridge | -10.396 |
Token | to |
Token ID | 284 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
tti | +2.291 |
etz | +1.981 |
neb | +1.976 |
Probe | +1.939 |
mysteries | +1.891 |
Neg logit contributions | |
BELOW | -2.816 |
idated | -2.759 |
productive | -2.731 |
arrass | -2.552 |
ridge | -2.498 |
Token | figure |
Token ID | 3785 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
tti | +4.352 |
neb | +3.799 |
etz | +3.762 |
Probe | +3.728 |
recognition | +3.593 |
Neg logit contributions | |
BELOW | -5.344 |
idated | -5.228 |
productive | -5.213 |
arrass | -4.857 |
ridge | -4.781 |
Token | out |
Token ID | 503 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
tti | +1.130 |
neb | +0.975 |
etz | +0.967 |
Probe | +0.953 |
mysteries | +0.933 |
Neg logit contributions | |
BELOW | -1.469 |
idated | -1.435 |
productive | -1.432 |
arrass | -1.333 |
ridge | -1.306 |
Token | release |
Token ID | 2650 |
Feature activation | -4.70e-03 |
Pos logit contributions | |
tti | +1.813 |
neb | +1.569 |
etz | +1.564 |
Probe | +1.546 |
mysteries | +1.499 |
Neg logit contributions | |
BELOW | -2.235 |
idated | -2.193 |
productive | -2.172 |
arrass | -2.022 |
ridge | -1.981 |
Token | as |
Token ID | 355 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
tti | +0.027 |
etz | +0.023 |
neb | +0.023 |
Probe | +0.023 |
mysteries | +0.022 |
Neg logit contributions | |
BELOW | -0.034 |
idated | -0.033 |
productive | -0.033 |
arrass | -0.031 |
ridge | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
BELOW | +2.257 |
idated | +2.203 |
productive | +2.197 |
arrass | +2.029 |
ridge | +1.988 |
Neg logit contributions | |
tti | -2.025 |
etz | -1.756 |
Probe | -1.705 |
neb | -1.692 |
earchers | -1.640 |
Token | grand |
Token ID | 4490 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
BELOW | +1.496 |
idated | +1.455 |
productive | +1.451 |
arrass | +1.333 |
ridge | +1.306 |
Neg logit contributions | |
tti | -1.475 |
etz | -1.293 |
Probe | -1.259 |
neb | -1.257 |
mysteries | -1.219 |
Token | gesture |
Token ID | 18342 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
tti | +1.347 |
neb | +1.185 |
etz | +1.177 |
Probe | +1.163 |
recognition | +1.134 |
Neg logit contributions | |
BELOW | -1.555 |
idated | -1.520 |
productive | -1.496 |
arrass | -1.393 |
ridge | -1.362 |
Token | remains |
Token ID | 3793 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
BELOW | +0.147 |
idated | +0.143 |
productive | +0.142 |
arrass | +0.131 |
ridge | +0.129 |
Neg logit contributions | |
tti | -0.138 |
etz | -0.120 |
neb | -0.120 |
Probe | -0.118 |
mysteries | -0.115 |
Token | a |
Token ID | 257 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
neb | +13.745 |
mysteries | +13.338 |
tti | +12.069 |
mystery | +12.043 |
secrets | +11.409 |
Neg logit contributions | |
idated | -17.887 |
覚醒 | -17.482 |
productive | -16.778 |
isode | -16.718 |
BELOW | -16.702 |
Token | mystery |
Token ID | 10715 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
neb | +7.509 |
tti | +7.336 |
mysteries | +7.273 |
mystery | +7.233 |
recognition | +6.363 |
Neg logit contributions | |
idated | -13.433 |
BELOW | -13.422 |
productive | -12.738 |
覚醒 | -12.522 |
accompanied | -12.039 |
Token | . |
Token ID | 13 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
tti | +6.828 |
neb | +5.873 |
mysteries | +5.750 |
etz | +5.710 |
Probe | +5.619 |
Neg logit contributions | |
idated | -9.045 |
BELOW | -8.954 |
productive | -8.733 |
arrass | -8.269 |
ridge | -8.176 |
Token | Wikileaks |
Token ID | 35828 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
tti | +2.213 |
Probe | +1.923 |
etz | +1.916 |
neb | +1.898 |
Television | +1.840 |
Neg logit contributions | |
BELOW | -2.359 |
idated | -2.324 |
productive | -2.302 |
arrass | -2.137 |
ridge | -2.083 |
Token | provides |
Token ID | 3769 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
BELOW | +3.645 |
idated | +3.442 |
productive | +3.431 |
arrass | +3.197 |
ridge | +3.176 |
Neg logit contributions | |
tti | -3.500 |
neb | -3.052 |
etz | -3.008 |
Probe | -2.954 |
earchers | -2.905 |
Token | involving |
Token ID | 7411 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
tti | +5.853 |
neb | +5.200 |
etz | +5.063 |
Probe | +5.045 |
recognition | +4.992 |
Neg logit contributions | |
BELOW | -7.267 |
idated | -7.220 |
productive | -7.205 |
arrass | -6.678 |
覚醒 | -6.486 |
Token | the |
Token ID | 262 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
tti | +0.621 |
etz | +0.537 |
neb | +0.536 |
Probe | +0.530 |
mysteries | +0.512 |
Neg logit contributions | |
BELOW | -0.704 |
idated | -0.686 |
productive | -0.680 |
arrass | -0.632 |
ridge | -0.620 |
Token | dead |
Token ID | 2636 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
tti | +0.647 |
etz | +0.562 |
neb | +0.559 |
Probe | +0.555 |
mysteries | +0.536 |
Neg logit contributions | |
BELOW | -0.699 |
idated | -0.681 |
productive | -0.673 |
arrass | -0.625 |
ridge | -0.614 |
Token | driver |
Token ID | 4639 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
tti | +1.960 |
etz | +1.703 |
neb | +1.703 |
Probe | +1.682 |
mysteries | +1.631 |
Neg logit contributions | |
BELOW | -2.087 |
idated | -2.040 |
productive | -2.020 |
arrass | -1.875 |
ridge | -1.831 |
Token | are |
Token ID | 389 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
tti | +3.624 |
etz | +3.140 |
neb | +3.126 |
Probe | +3.072 |
mysteries | +3.030 |
Neg logit contributions | |
BELOW | -4.010 |
idated | -3.917 |
productive | -3.900 |
arrass | -3.639 |
ridge | -3.528 |
Token | under |
Token ID | 739 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
tti | +6.379 |
neb | +5.791 |
mysteries | +5.701 |
Probe | +5.566 |
recognition | +5.494 |
Neg logit contributions | |
BELOW | -8.435 |
idated | -8.433 |
productive | -8.392 |
覚醒 | -7.857 |
arrass | -7.727 |
Token | investigation |
Token ID | 3645 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
investigation | +7.831 |
tti | +7.338 |
Probe | +7.269 |
recognition | +6.956 |
neb | +6.633 |
Neg logit contributions | |
BELOW | -22.843 |
覚醒 | -21.991 |
idated | -21.902 |
productive | -21.434 |
bleacher | -21.068 |
Token | . |
Token ID | 13 |
Feature activation | -9.21e-01 |
Pos logit contributions | |
tti | +5.273 |
etz | +4.447 |
Probe | +4.444 |
neb | +4.424 |
mysteries | +4.389 |
Neg logit contributions | |
BELOW | -8.778 |
idated | -8.722 |
productive | -8.685 |
arrass | -8.205 |
ridge | -8.063 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
tti | +5.548 |
Probe | +4.885 |
etz | +4.787 |
Television | +4.739 |
Satellite | +4.632 |
Neg logit contributions | |
idated | -6.169 |
BELOW | -6.103 |
productive | -6.093 |
arrass | -5.661 |
ridge | -5.470 |
Token | Ready |
Token ID | 35474 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
tti | +1.170 |
etz | +1.021 |
neb | +1.015 |
Probe | +1.005 |
mysteries | +0.975 |
Neg logit contributions | |
BELOW | -1.201 |
idated | -1.171 |
productive | -1.158 |
arrass | -1.074 |
ridge | -1.050 |
Token | to |
Token ID | 284 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
BELOW | +1.803 |
idated | +1.749 |
productive | +1.738 |
arrass | +1.602 |
ridge | +1.588 |
Neg logit contributions | |
tti | -1.629 |
neb | -1.431 |
etz | -1.428 |
Probe | -1.404 |
mysteries | -1.353 |
Token | of |
Token ID | 286 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
tti | +3.970 |
neb | +3.460 |
etz | +3.452 |
recognition | +3.341 |
Probe | +3.260 |
Neg logit contributions | |
BELOW | -4.744 |
idated | -4.647 |
productive | -4.600 |
arrass | -4.320 |
ridge | -4.155 |
Token | the |
Token ID | 262 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
tti | +1.334 |
neb | +1.174 |
etz | +1.168 |
Probe | +1.157 |
mysteries | +1.121 |
Neg logit contributions | |
BELOW | -1.375 |
idated | -1.343 |
productive | -1.329 |
arrass | -1.230 |
ridge | -1.203 |
Token | Iran |
Token ID | 4068 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
tti | +2.879 |
neb | +2.568 |
Probe | +2.522 |
etz | +2.515 |
recognition | +2.464 |
Neg logit contributions | |
BELOW | -3.218 |
idated | -3.161 |
productive | -3.107 |
arrass | -2.906 |
ridge | -2.832 |
Token | deal |
Token ID | 1730 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
tti | +3.317 |
Probe | +2.897 |
neb | +2.891 |
recognition | +2.803 |
etz | +2.789 |
Neg logit contributions | |
BELOW | -5.041 |
idated | -4.930 |
productive | -4.849 |
arrass | -4.580 |
ridge | -4.490 |
Token | , |
Token ID | 11 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
tti | +1.429 |
etz | +1.246 |
neb | +1.245 |
Probe | +1.232 |
mysteries | +1.191 |
Neg logit contributions | |
BELOW | -1.492 |
idated | -1.458 |
productive | -1.441 |
arrass | -1.336 |
ridge | -1.307 |
Token | for |
Token ID | 329 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
tti | +3.025 |
neb | +2.611 |
etz | +2.601 |
Probe | +2.560 |
recognition | +2.487 |
Neg logit contributions | |
BELOW | -3.772 |
idated | -3.670 |
productive | -3.647 |
arrass | -3.392 |
ridge | -3.329 |
Token | example |
Token ID | 1672 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
tti | +11.987 |
recognition | +11.079 |
neb | +11.068 |
obvious | +10.632 |
ked | +10.345 |
Neg logit contributions | |
BELOW | -18.163 |
idated | -17.991 |
覚醒 | -17.493 |
productive | -17.092 |
arrass | -16.531 |
Token | , |
Token ID | 11 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
BELOW | +0.566 |
idated | +0.552 |
productive | +0.547 |
arrass | +0.507 |
ridge | +0.497 |
Neg logit contributions | |
tti | -0.542 |
etz | -0.472 |
neb | -0.471 |
Probe | -0.467 |
mysteries | -0.452 |
Token | despite |
Token ID | 3805 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
tti | +0.159 |
etz | +0.138 |
neb | +0.137 |
Probe | +0.136 |
mysteries | +0.132 |
Neg logit contributions | |
BELOW | -0.181 |
idated | -0.177 |
productive | -0.175 |
arrass | -0.163 |
ridge | -0.160 |
Token | the |
Token ID | 262 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
tti | +2.979 |
neb | +2.608 |
etz | +2.563 |
Probe | +2.552 |
mysteries | +2.490 |
Neg logit contributions | |
BELOW | -3.482 |
idated | -3.402 |
productive | -3.363 |
arrass | -3.118 |
ridge | -3.064 |
Token | fact |
Token ID | 1109 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
tti | +2.181 |
neb | +1.854 |
Probe | +1.791 |
etz | +1.791 |
recognition | +1.734 |
Neg logit contributions | |
BELOW | -3.888 |
idated | -3.826 |
productive | -3.776 |
arrass | -3.557 |
ridge | -3.506 |
Token | have |
Token ID | 423 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
tti | +2.749 |
neb | +2.355 |
etz | +2.347 |
Probe | +2.309 |
recognition | +2.248 |
Neg logit contributions | |
BELOW | -3.598 |
idated | -3.543 |
productive | -3.495 |
arrass | -3.270 |
ridge | -3.206 |
Token | been |
Token ID | 587 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
tti | +0.785 |
etz | +0.678 |
neb | +0.674 |
Probe | +0.669 |
mysteries | +0.644 |
Neg logit contributions | |
BELOW | -0.951 |
idated | -0.928 |
productive | -0.920 |
arrass | -0.857 |
ridge | -0.840 |
Token | flag |
Token ID | 6056 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
tti | +0.424 |
etz | +0.372 |
neb | +0.369 |
Probe | +0.368 |
mysteries | +0.355 |
Neg logit contributions | |
BELOW | -0.410 |
idated | -0.397 |
productive | -0.395 |
arrass | -0.364 |
ridge | -0.356 |
Token | ging |
Token ID | 2667 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
BELOW | +4.805 |
productive | +4.671 |
idated | +4.636 |
arrass | +4.432 |
ridge | +4.301 |
Neg logit contributions | |
tti | -4.075 |
etz | -3.593 |
Probe | -3.499 |
neb | -3.441 |
mysteries | -3.338 |
Token | it |
Token ID | 340 |
Feature activation | -6.05e-01 |
Pos logit contributions | |
tti | +3.134 |
neb | +2.726 |
etz | +2.693 |
Probe | +2.668 |
mysteries | +2.642 |
Neg logit contributions | |
BELOW | -3.431 |
idated | -3.368 |
productive | -3.329 |
arrass | -3.071 |
ridge | -3.022 |
Token | for |
Token ID | 329 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
tti | +3.212 |
neb | +2.766 |
etz | +2.750 |
Probe | +2.677 |
mysteries | +2.635 |
Neg logit contributions | |
BELOW | -4.594 |
idated | -4.550 |
productive | -4.469 |
arrass | -4.182 |
ridge | -4.092 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
tti | +11.995 |
recognition | +11.184 |
decades | +11.053 |
years | +10.995 |
neb | +10.951 |
Neg logit contributions | |
idated | -18.432 |
BELOW | -17.625 |
覚醒 | -17.620 |
productive | -17.529 |
arrass | -16.796 |
Token | last |
Token ID | 938 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
tti | +6.433 |
neb | +5.791 |
recognition | +5.636 |
mysteries | +5.590 |
Probe | +5.455 |
Neg logit contributions | |
BELOW | -8.492 |
idated | -8.464 |
productive | -8.369 |
arrass | -7.987 |
覚醒 | -7.722 |
Token | 50 |
Token ID | 2026 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
tti | +8.575 |
neb | +7.868 |
recognition | +7.625 |
mysteries | +7.416 |
Probe | +7.389 |
Neg logit contributions | |
idated | -10.853 |
productive | -10.696 |
BELOW | -10.565 |
arrass | -10.041 |
覚醒 | -9.748 |
Token | years |
Token ID | 812 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
tti | +4.529 |
neb | +3.484 |
etz | +3.390 |
mysteries | +3.379 |
Probe | +3.378 |
Neg logit contributions | |
BELOW | -10.876 |
idated | -10.676 |
productive | -10.563 |
arrass | -10.138 |
ridge | -9.989 |
Token | , |
Token ID | 11 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
tti | +1.168 |
etz | +1.013 |
neb | +1.011 |
Probe | +0.999 |
mysteries | +0.970 |
Neg logit contributions | |
BELOW | -1.282 |
idated | -1.255 |
productive | -1.241 |
arrass | -1.152 |
ridge | -1.127 |
Token | been |
Token ID | 587 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
tti | +3.572 |
neb | +3.074 |
etz | +3.045 |
Probe | +2.998 |
mysteries | +2.933 |
Neg logit contributions | |
BELOW | -4.490 |
productive | -4.387 |
idated | -4.372 |
arrass | -4.091 |
ridge | -4.040 |
Token | trying |
Token ID | 2111 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
tti | +2.880 |
neb | +2.528 |
etz | +2.513 |
Probe | +2.489 |
mysteries | +2.426 |
Neg logit contributions | |
BELOW | -2.995 |
idated | -2.929 |
productive | -2.896 |
arrass | -2.686 |
ridge | -2.632 |
Token | to |
Token ID | 284 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
tti | +4.562 |
neb | +4.008 |
etz | +3.959 |
Probe | +3.931 |
mysteries | +3.855 |
Neg logit contributions | |
BELOW | -5.958 |
idated | -5.919 |
productive | -5.806 |
arrass | -5.440 |
ridge | -5.373 |
Token | do |
Token ID | 466 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
tti | +6.579 |
neb | +5.789 |
Probe | +5.645 |
recognition | +5.536 |
etz | +5.499 |
Neg logit contributions | |
BELOW | -8.607 |
idated | -8.573 |
productive | -8.555 |
arrass | -8.007 |
ridge | -7.948 |
Token | this |
Token ID | 428 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
BELOW | +1.184 |
idated | +1.148 |
productive | +1.140 |
arrass | +1.048 |
ridge | +1.034 |
Neg logit contributions | |
tti | -1.182 |
etz | -1.042 |
neb | -1.023 |
Probe | -1.023 |
earchers | -0.985 |
Token | for |
Token ID | 329 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
tti | +2.569 |
neb | +2.212 |
etz | +2.195 |
Probe | +2.172 |
mysteries | +2.106 |
Neg logit contributions | |
BELOW | -3.252 |
idated | -3.184 |
productive | -3.152 |
arrass | -2.944 |
ridge | -2.885 |
Token | six |
Token ID | 2237 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
tti | +12.676 |
neb | +12.165 |
recognition | +12.011 |
years | +11.531 |
mysteries | +11.474 |
Neg logit contributions | |
idated | -17.449 |
productive | -16.603 |
覚醒 | -16.426 |
BELOW | -16.284 |
arrass | -15.989 |
Token | years |
Token ID | 812 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
tti | +5.231 |
neb | +4.333 |
mysteries | +4.163 |
etz | +4.107 |
Probe | +4.043 |
Neg logit contributions | |
BELOW | -9.180 |
idated | -9.076 |
productive | -8.858 |
arrass | -8.620 |
ridge | -8.441 |
Token | now |
Token ID | 783 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
tti | +0.759 |
etz | +0.656 |
neb | +0.655 |
Probe | +0.648 |
mysteries | +0.626 |
Neg logit contributions | |
BELOW | -0.881 |
idated | -0.860 |
productive | -0.852 |
arrass | -0.793 |
ridge | -0.777 |
Token | , |
Token ID | 11 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
tti | +2.213 |
neb | +1.911 |
etz | +1.904 |
Probe | +1.877 |
mysteries | +1.828 |
Neg logit contributions | |
BELOW | -2.660 |
idated | -2.608 |
productive | -2.578 |
arrass | -2.405 |
ridge | -2.349 |
Token | and |
Token ID | 290 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
tti | +4.477 |
neb | +3.864 |
etz | +3.839 |
Probe | +3.799 |
mysteries | +3.706 |
Neg logit contributions | |
BELOW | -5.541 |
idated | -5.400 |
productive | -5.387 |
arrass | -5.045 |
ridge | -4.928 |
Token | other |
Token ID | 584 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
tti | +4.138 |
neb | +3.593 |
Probe | +3.571 |
etz | +3.554 |
mysteries | +3.454 |
Neg logit contributions | |
BELOW | -4.543 |
idated | -4.512 |
productive | -4.434 |
arrass | -4.147 |
ridge | -4.021 |
Token | formats |
Token ID | 17519 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
BELOW | +2.070 |
productive | +1.984 |
idated | +1.983 |
ridge | +1.810 |
arrass | +1.795 |
Neg logit contributions | |
tti | -2.008 |
etz | -1.789 |
Probe | -1.758 |
neb | -1.738 |
earchers | -1.671 |
Token | — |
Token ID | 960 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
BELOW | +4.510 |
idated | +4.226 |
productive | +4.210 |
ridge | +3.953 |
arrass | +3.921 |
Neg logit contributions | |
tti | -4.123 |
Probe | -3.640 |
etz | -3.639 |
neb | -3.620 |
earchers | -3.531 |
Token | HTML |
Token ID | 28656 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
tti | +4.803 |
etz | +4.245 |
neb | +4.192 |
Probe | +4.160 |
mysteries | +4.069 |
Neg logit contributions | |
BELOW | -3.937 |
idated | -3.853 |
productive | -3.721 |
arrass | -3.483 |
ridge | -3.323 |
Token | , |
Token ID | 11 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
tti | +0.857 |
etz | +0.748 |
neb | +0.746 |
Probe | +0.743 |
mysteries | +0.721 |
Neg logit contributions | |
BELOW | -0.849 |
idated | -0.832 |
productive | -0.824 |
arrass | -0.759 |
ridge | -0.743 |
Token | for |
Token ID | 329 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
tti | +2.584 |
neb | +2.278 |
etz | +2.272 |
Probe | +2.266 |
mysteries | +2.213 |
Neg logit contributions | |
BELOW | -2.126 |
idated | -2.098 |
productive | -2.066 |
arrass | -1.885 |
ridge | -1.837 |
Token | example |
Token ID | 1672 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
tti | +12.760 |
ked | +11.434 |
bes | +10.912 |
etz | +10.698 |
recognition | +10.691 |
Neg logit contributions | |
idated | -17.073 |
BELOW | -16.841 |
productive | -16.204 |
覚醒 | -16.007 |
arrass | -15.646 |
Token | — |
Token ID | 960 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
tti | +1.066 |
neb | +0.930 |
etz | +0.929 |
Probe | +0.917 |
mysteries | +0.893 |
Neg logit contributions | |
BELOW | -1.054 |
idated | -1.036 |
productive | -1.028 |
arrass | -0.949 |
ridge | -0.920 |
Token | will |
Token ID | 10594 |
Feature activation | +8.76e-02 |
Pos logit contributions | |
tti | +2.463 |
etz | +2.158 |
neb | +2.115 |
Probe | +2.104 |
mysteries | +2.046 |
Neg logit contributions | |
BELOW | -2.231 |
idated | -2.173 |
productive | -2.109 |
arrass | -1.966 |
ridge | -1.889 |
Token | require |
Token ID | 2421 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
BELOW | +0.605 |
idated | +0.586 |
productive | +0.581 |
arrass | +0.541 |
ridge | +0.533 |
Neg logit contributions | |
tti | -0.537 |
etz | -0.468 |
neb | -0.465 |
Probe | -0.462 |
mysteries | -0.447 |
Token | conversion |
Token ID | 11315 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
tti | +4.058 |
neb | +3.535 |
Probe | +3.498 |
etz | +3.476 |
recognition | +3.410 |
Neg logit contributions | |
BELOW | -4.586 |
idated | -4.571 |
productive | -4.509 |
arrass | -4.201 |
ridge | -4.054 |
Token | it |
Token ID | 270 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
tti | +4.721 |
etz | +4.042 |
neb | +4.037 |
Probe | +3.951 |
mysteries | +3.874 |
Neg logit contributions | |
BELOW | -5.550 |
idated | -5.404 |
productive | -5.194 |
arrass | -4.987 |
ridge | -4.715 |
Token | 's |
Token ID | 338 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
tti | +4.041 |
neb | +3.460 |
etz | +3.455 |
mysteries | +3.294 |
Probe | +3.266 |
Neg logit contributions | |
BELOW | -6.516 |
idated | -6.312 |
productive | -6.280 |
arrass | -5.919 |
覚醒 | -5.756 |
Token | not |
Token ID | 407 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
neb | +10.528 |
mysteries | +10.189 |
tti | +10.147 |
recognition | +9.328 |
obvious | +8.927 |
Neg logit contributions | |
idated | -15.099 |
productive | -14.768 |
覚醒 | -14.712 |
BELOW | -14.527 |
isode | -14.046 |
Token | clear |
Token ID | 1598 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
tti | +7.594 |
neb | +6.935 |
mysteries | +6.657 |
recognition | +6.580 |
etz | +6.526 |
Neg logit contributions | |
BELOW | -9.110 |
idated | -9.097 |
productive | -8.973 |
覚醒 | -8.535 |
arrass | -8.487 |
Token | , |
Token ID | 11 |
Feature activation | -8.82e-01 |
Pos logit contributions | |
BELOW | +0.605 |
idated | +0.583 |
productive | +0.579 |
arrass | +0.534 |
ridge | +0.528 |
Neg logit contributions | |
tti | -0.598 |
etz | -0.526 |
neb | -0.517 |
Probe | -0.516 |
earchers | -0.497 |
Token | for |
Token ID | 329 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
tti | +5.189 |
neb | +4.502 |
Probe | +4.480 |
etz | +4.399 |
mysteries | +4.346 |
Neg logit contributions | |
BELOW | -6.115 |
idated | -6.025 |
productive | -5.997 |
arrass | -5.540 |
覚醒 | -5.422 |
Token | instance |
Token ID | 4554 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
tti | +10.357 |
obvious | +8.877 |
recognition | +8.546 |
mysteries | +8.359 |
neb | +8.315 |
Neg logit contributions | |
覚醒 | -19.520 |
idated | -19.304 |
BELOW | -19.297 |
productive | -19.016 |
arrass | -18.511 |
Token | , |
Token ID | 11 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
BELOW | +1.039 |
idated | +0.986 |
productive | +0.982 |
ridge | +0.892 |
arrass | +0.890 |
Neg logit contributions | |
tti | -1.280 |
etz | -1.145 |
neb | -1.139 |
Probe | -1.132 |
earchers | -1.119 |
Token | whether |
Token ID | 1771 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
tti | +0.298 |
etz | +0.261 |
neb | +0.258 |
Probe | +0.257 |
mysteries | +0.248 |
Neg logit contributions | |
BELOW | -0.298 |
idated | -0.290 |
productive | -0.287 |
arrass | -0.264 |
ridge | -0.261 |
Token | those |
Token ID | 883 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
tti | +0.527 |
etz | +0.459 |
neb | +0.455 |
Probe | +0.453 |
mysteries | +0.437 |
Neg logit contributions | |
BELOW | -0.557 |
idated | -0.541 |
productive | -0.537 |
arrass | -0.497 |
ridge | -0.488 |
Token | levels |
Token ID | 2974 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
tti | +4.467 |
neb | +4.005 |
etz | +3.882 |
Probe | +3.873 |
mysteries | +3.865 |
Neg logit contributions | |
BELOW | -4.804 |
idated | -4.786 |
productive | -4.733 |
arrass | -4.419 |
ridge | -4.288 |
Token | a |
Token ID | 257 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
BELOW | +0.869 |
idated | +0.844 |
productive | +0.836 |
arrass | +0.775 |
ridge | +0.767 |
Neg logit contributions | |
tti | -0.759 |
etz | -0.656 |
neb | -0.655 |
Probe | -0.648 |
mysteries | -0.628 |
Token | UCS |
Token ID | 49553 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
tti | +0.125 |
etz | +0.109 |
neb | +0.109 |
Probe | +0.108 |
mysteries | +0.104 |
Neg logit contributions | |
BELOW | -0.123 |
idated | -0.120 |
productive | -0.119 |
arrass | -0.109 |
ridge | -0.108 |
Token | F |
Token ID | 37 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
BELOW | +4.094 |
idated | +3.972 |
productive | +3.848 |
ridge | +3.637 |
arrass | +3.568 |
Neg logit contributions | |
tti | -4.247 |
etz | -3.656 |
Probe | -3.619 |
neb | -3.617 |
mysteries | -3.560 |
Token | professor |
Token ID | 6240 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
tti | +0.275 |
etz | +0.237 |
neb | +0.237 |
Probe | +0.235 |
mysteries | +0.227 |
Neg logit contributions | |
BELOW | -0.325 |
idated | -0.317 |
productive | -0.314 |
arrass | -0.292 |
ridge | -0.288 |
Token | who |
Token ID | 508 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
tti | +2.775 |
etz | +2.391 |
neb | +2.368 |
Probe | +2.348 |
mysteries | +2.265 |
Neg logit contributions | |
BELOW | -3.408 |
idated | -3.318 |
productive | -3.277 |
arrass | -3.071 |
ridge | -2.975 |
Token | for |
Token ID | 329 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
tti | +4.777 |
etz | +4.169 |
Probe | +4.014 |
neb | +3.995 |
mysteries | +3.908 |
Neg logit contributions | |
BELOW | -5.114 |
idated | -4.898 |
productive | -4.884 |
arrass | -4.505 |
ridge | -4.415 |
Token | many |
Token ID | 867 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
tti | +13.090 |
ked | +12.056 |
etz | +11.786 |
Probe | +11.736 |
bes | +11.729 |
Neg logit contributions | |
BELOW | -16.430 |
idated | -15.784 |
覚醒 | -15.473 |
productive | -15.094 |
arrass | -14.518 |
Token | years |
Token ID | 812 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
tti | +6.894 |
neb | +5.830 |
mysteries | +5.758 |
recognition | +5.464 |
Probe | +5.292 |
Neg logit contributions | |
BELOW | -15.458 |
idated | -15.189 |
productive | -14.926 |
覚醒 | -14.551 |
arrass | -14.500 |
Token | has |
Token ID | 468 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
tti | +1.086 |
etz | +0.944 |
neb | +0.939 |
Probe | +0.930 |
mysteries | +0.903 |
Neg logit contributions | |
BELOW | -1.191 |
idated | -1.158 |
productive | -1.151 |
arrass | -1.066 |
ridge | -1.046 |
Token | helped |
Token ID | 4193 |
Feature activation | -5.80e-01 |
Pos logit contributions | |
tti | +4.466 |
etz | +3.872 |
neb | +3.822 |
Probe | +3.784 |
mysteries | +3.696 |
Neg logit contributions | |
BELOW | -5.037 |
productive | -4.835 |
idated | -4.789 |
arrass | -4.483 |
ridge | -4.426 |
Token | craft |
Token ID | 5977 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
tti | +4.371 |
neb | +3.932 |
etz | +3.910 |
Probe | +3.881 |
mysteries | +3.801 |
Neg logit contributions | |
BELOW | -3.088 |
idated | -3.024 |
productive | -3.011 |
arrass | -2.749 |
ridge | -2.658 |
Token | adults |
Token ID | 6490 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
tti | +8.122 |
neb | +7.314 |
recognition | +7.273 |
mysteries | +7.086 |
Television | +7.042 |
Neg logit contributions | |
BELOW | -9.633 |
idated | -9.512 |
productive | -9.275 |
arrass | -8.708 |
覚醒 | -8.565 |
Token | that |
Token ID | 326 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
tti | +2.992 |
neb | +2.537 |
etz | +2.511 |
Probe | +2.449 |
mysteries | +2.428 |
Neg logit contributions | |
BELOW | -4.233 |
idated | -4.149 |
productive | -4.099 |
arrass | -3.880 |
ridge | -3.751 |
Token | had |
Token ID | 550 |
Feature activation | +6.11e-02 |
Pos logit contributions | |
tti | +0.764 |
etz | +0.662 |
neb | +0.660 |
Probe | +0.654 |
mysteries | +0.631 |
Neg logit contributions | |
BELOW | -0.862 |
idated | -0.841 |
productive | -0.833 |
arrass | -0.774 |
ridge | -0.760 |
Token | been |
Token ID | 587 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
BELOW | +0.390 |
idated | +0.379 |
productive | +0.376 |
arrass | +0.347 |
ridge | +0.340 |
Neg logit contributions | |
tti | -0.405 |
etz | -0.359 |
Probe | -0.355 |
neb | -0.353 |
mysteries | -0.338 |
Token | ignored |
Token ID | 9514 |
Feature activation | -9.27e-01 |
Pos logit contributions | |
tti | +4.737 |
neb | +4.227 |
etz | +4.081 |
mysteries | +4.064 |
Probe | +4.020 |
Neg logit contributions | |
BELOW | -4.829 |
idated | -4.783 |
productive | -4.630 |
arrass | -4.350 |
ridge | -4.217 |
Token | for |
Token ID | 329 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
tti | +5.171 |
neb | +4.590 |
etz | +4.478 |
mysteries | +4.433 |
Probe | +4.387 |
Neg logit contributions | |
BELOW | -6.513 |
idated | -6.437 |
productive | -6.310 |
arrass | -5.962 |
ridge | -5.773 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
tti | +11.789 |
recognition | +11.174 |
neb | +11.017 |
decades | +10.928 |
mysteries | +10.668 |
Neg logit contributions | |
idated | -17.386 |
覚醒 | -16.947 |
BELOW | -16.740 |
productive | -16.608 |
arrass | -16.039 |
Token | past |
Token ID | 1613 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
tti | +6.366 |
neb | +5.831 |
recognition | +5.630 |
mysteries | +5.572 |
Television | +5.530 |
Neg logit contributions | |
BELOW | -7.642 |
idated | -7.547 |
productive | -7.484 |
arrass | -7.173 |
覚醒 | -6.818 |
Token | decade |
Token ID | 5707 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
tti | +8.834 |
neb | +8.061 |
recognition | +7.933 |
mysteries | +7.691 |
Probe | +7.483 |
Neg logit contributions | |
idated | -12.945 |
BELOW | -12.670 |
productive | -12.594 |
覚醒 | -12.132 |
arrass | -11.891 |
Token | . |
Token ID | 13 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
tti | +2.712 |
neb | +2.317 |
etz | +2.298 |
Probe | +2.253 |
mysteries | +2.234 |
Neg logit contributions | |
BELOW | -3.644 |
idated | -3.567 |
productive | -3.532 |
arrass | -3.316 |
ridge | -3.222 |
Token |
|
Token ID | 198 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
BELOW | +1.648 |
idated | +1.582 |
productive | +1.571 |
arrass | +1.447 |
ridge | +1.447 |
Neg logit contributions | |
tti | -1.573 |
neb | -1.401 |
etz | -1.393 |
Probe | -1.338 |
earchers | -1.332 |
Token | up |
Token ID | 510 |
Feature activation | +7.55e-01 |
Pos logit contributions | |
BELOW | +3.865 |
productive | +3.675 |
idated | +3.631 |
arrass | +3.441 |
ridge | +3.397 |
Neg logit contributions | |
tti | -3.915 |
etz | -3.436 |
Probe | -3.415 |
earchers | -3.334 |
neb | -3.330 |
Token | and |
Token ID | 290 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
BELOW | +4.887 |
productive | +4.657 |
idated | +4.604 |
arrass | +4.323 |
ridge | +4.293 |
Neg logit contributions | |
tti | -4.744 |
etz | -4.152 |
Probe | -4.136 |
neb | -4.060 |
earchers | -3.970 |
Token | wash |
Token ID | 13502 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
BELOW | +0.269 |
idated | +0.259 |
productive | +0.258 |
arrass | +0.242 |
ridge | +0.236 |
Neg logit contributions | |
tti | -0.243 |
etz | -0.212 |
neb | -0.209 |
Probe | -0.209 |
earchers | -0.201 |
Token | them |
Token ID | 606 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
BELOW | +1.857 |
idated | +1.792 |
productive | +1.780 |
arrass | +1.629 |
ridge | +1.593 |
Neg logit contributions | |
tti | -2.349 |
etz | -2.083 |
Probe | -2.081 |
neb | -2.073 |
mysteries | -1.993 |
Token | down |
Token ID | 866 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
BELOW | +3.592 |
idated | +3.431 |
productive | +3.431 |
arrass | +3.154 |
ridge | +3.140 |
Neg logit contributions | |
tti | -3.677 |
etz | -3.257 |
Probe | -3.255 |
neb | -3.191 |
earchers | -3.126 |
Token | with |
Token ID | 351 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
BELOW | +4.959 |
productive | +4.691 |
idated | +4.685 |
ridge | +4.309 |
arrass | +4.300 |
Neg logit contributions | |
tti | -5.201 |
Probe | -4.596 |
etz | -4.561 |
neb | -4.546 |
earchers | -4.410 |
Token | orange |
Token ID | 10912 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
BELOW | +7.521 |
idated | +7.376 |
productive | +6.989 |
deter | +6.900 |
arrass | +6.601 |
Neg logit contributions | |
tti | -9.086 |
bourg | -7.975 |
etz | -7.948 |
Probe | -7.878 |
earchers | -7.730 |
Token | juice |
Token ID | 13135 |
Feature activation | +8.30e-01 |
Pos logit contributions | |
tti | +0.261 |
etz | +0.225 |
neb | +0.224 |
Probe | +0.222 |
mysteries | +0.214 |
Neg logit contributions | |
BELOW | -0.316 |
idated | -0.309 |
productive | -0.306 |
arrass | -0.285 |
ridge | -0.280 |
Token | . |
Token ID | 13 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
BELOW | +5.270 |
idated | +5.051 |
productive | +4.969 |
arrass | +4.662 |
ridge | +4.634 |
Neg logit contributions | |
tti | -5.204 |
Probe | -4.571 |
neb | -4.553 |
etz | -4.540 |
mysteries | -4.386 |
Token | Some |
Token ID | 4366 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
BELOW | +1.197 |
idated | +1.154 |
productive | +1.145 |
arrass | +1.065 |
ridge | +1.048 |
Neg logit contributions | |
tti | -1.031 |
neb | -0.900 |
etz | -0.895 |
Probe | -0.871 |
mysteries | -0.852 |
Token | people |
Token ID | 661 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
tti | +1.976 |
neb | +1.668 |
etz | +1.652 |
Probe | +1.625 |
mysteries | +1.589 |
Neg logit contributions | |
BELOW | -2.977 |
idated | -2.919 |
productive | -2.887 |
arrass | -2.714 |
ridge | -2.661 |
Token | all |
Token ID | 477 |
Feature activation | -7.33e-01 |
Pos logit contributions | |
tti | +2.196 |
neb | +1.929 |
etz | +1.904 |
Probe | +1.889 |
mysteries | +1.859 |
Neg logit contributions | |
BELOW | -2.320 |
idated | -2.276 |
productive | -2.244 |
arrass | -2.086 |
ridge | -2.034 |
Token | elbows |
Token ID | 37076 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
tti | +4.289 |
neb | +3.792 |
mysteries | +3.673 |
etz | +3.663 |
Probe | +3.630 |
Neg logit contributions | |
BELOW | -5.112 |
idated | -5.073 |
productive | -4.951 |
arrass | -4.637 |
ridge | -4.501 |
Token | and |
Token ID | 290 |
Feature activation | +7.75e-01 |
Pos logit contributions | |
BELOW | +1.091 |
idated | +1.055 |
productive | +1.053 |
arrass | +0.979 |
ridge | +0.960 |
Neg logit contributions | |
tti | -0.886 |
etz | -0.759 |
neb | -0.758 |
Probe | -0.749 |
mysteries | -0.722 |
Token | knees |
Token ID | 14475 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
BELOW | +4.618 |
productive | +4.406 |
idated | +4.405 |
arrass | +4.045 |
ridge | +3.902 |
Neg logit contributions | |
tti | -5.104 |
etz | -4.638 |
Probe | -4.579 |
neb | -4.391 |
earchers | -4.387 |
Token | , |
Token ID | 11 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
BELOW | +1.501 |
idated | +1.445 |
productive | +1.443 |
arrass | +1.336 |
ridge | +1.313 |
Neg logit contributions | |
tti | -1.404 |
etz | -1.218 |
neb | -1.214 |
Probe | -1.213 |
mysteries | -1.155 |
Token | Pl |
Token ID | 1345 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
BELOW | +0.666 |
idated | +0.645 |
productive | +0.642 |
arrass | +0.594 |
ridge | +0.586 |
Neg logit contributions | |
tti | -0.618 |
etz | -0.542 |
Probe | -0.533 |
neb | -0.531 |
Television | -0.509 |
Token | ath |
Token ID | 776 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
idated | +7.346 |
BELOW | +7.226 |
productive | +7.043 |
arrass | +6.825 |
ridge | +6.620 |
Neg logit contributions | |
tti | -8.059 |
Probe | -7.598 |
neb | -7.579 |
mysteries | -7.378 |
Television | -7.182 |
Token | ian |
Token ID | 666 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
BELOW | +2.114 |
idated | +2.086 |
productive | +2.052 |
arrass | +1.916 |
ridge | +1.902 |
Neg logit contributions | |
tti | -1.824 |
Probe | -1.569 |
neb | -1.565 |
etz | -1.558 |
mysteries | -1.488 |
Token | with |
Token ID | 351 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
tti | +2.951 |
etz | +2.561 |
neb | +2.553 |
Probe | +2.523 |
mysteries | +2.466 |
Neg logit contributions | |
BELOW | -3.356 |
idated | -3.261 |
productive | -3.248 |
arrass | -3.018 |
ridge | -2.940 |
Token | a |
Token ID | 257 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
tti | +1.638 |
neb | +1.419 |
etz | +1.414 |
Probe | +1.396 |
mysteries | +1.361 |
Neg logit contributions | |
BELOW | -1.903 |
idated | -1.855 |
productive | -1.840 |
arrass | -1.710 |
ridge | -1.674 |
Token | r |
Token ID | 374 |
Feature activation | -9.77e-02 |
Pos logit contributions | |
tti | +0.991 |
etz | +0.862 |
neb | +0.859 |
Probe | +0.851 |
mysteries | +0.824 |
Neg logit contributions | |
BELOW | -1.067 |
idated | -1.041 |
productive | -1.031 |
arrass | -0.956 |
ridge | -0.938 |
Token | protecting |
Token ID | 10192 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
tti | +0.204 |
etz | +0.178 |
neb | +0.176 |
Probe | +0.175 |
mysteries | +0.169 |
Neg logit contributions | |
BELOW | -0.210 |
idated | -0.204 |
productive | -0.202 |
arrass | -0.188 |
ridge | -0.184 |
Token | manufacturing |
Token ID | 9138 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
tti | +3.423 |
neb | +3.014 |
etz | +2.988 |
Probe | +2.941 |
mysteries | +2.897 |
Neg logit contributions | |
BELOW | -3.745 |
idated | -3.668 |
productive | -3.610 |
arrass | -3.389 |
ridge | -3.303 |
Token | jobs |
Token ID | 3946 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
BELOW | +0.471 |
idated | +0.458 |
productive | +0.457 |
arrass | +0.419 |
ridge | +0.413 |
Neg logit contributions | |
tti | -0.448 |
etz | -0.393 |
neb | -0.390 |
Probe | -0.386 |
mysteries | -0.371 |
Token | through |
Token ID | 832 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
tti | +2.632 |
etz | +2.266 |
neb | +2.265 |
Probe | +2.224 |
mysteries | +2.159 |
Neg logit contributions | |
BELOW | -3.235 |
idated | -3.169 |
productive | -3.126 |
arrass | -2.921 |
ridge | -2.862 |
Token | a |
Token ID | 257 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
tti | +0.278 |
etz | +0.242 |
neb | +0.241 |
Probe | +0.239 |
mysteries | +0.230 |
Neg logit contributions | |
BELOW | -0.311 |
idated | -0.303 |
productive | -0.301 |
arrass | -0.279 |
ridge | -0.274 |
Token | range |
Token ID | 2837 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
tti | +0.194 |
etz | +0.169 |
Probe | +0.167 |
neb | +0.167 |
mysteries | +0.161 |
Neg logit contributions | |
BELOW | -0.204 |
idated | -0.199 |
productive | -0.197 |
arrass | -0.183 |
ridge | -0.179 |
Token | of |
Token ID | 286 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
BELOW | +6.569 |
productive | +6.131 |
idated | +5.748 |
ridge | +5.391 |
surplus | +4.897 |
Neg logit contributions | |
tti | -12.397 |
earchers | -12.259 |
Probe | -11.585 |
etz | -11.484 |
neb | -11.240 |
Token | means |
Token ID | 1724 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
BELOW | +6.901 |
productive | +6.732 |
idated | +6.634 |
arrass | +6.330 |
ridge | +5.891 |
Neg logit contributions | |
tti | -7.168 |
etz | -6.334 |
earchers | -6.264 |
Probe | -6.191 |
bourg | -5.945 |
Token | such |
Token ID | 884 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
BELOW | +2.999 |
productive | +2.911 |
idated | +2.898 |
arrass | +2.657 |
ridge | +2.602 |
Neg logit contributions | |
tti | -3.126 |
Probe | -2.738 |
neb | -2.728 |
etz | -2.725 |
mysteries | -2.644 |
Token | as |
Token ID | 355 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
tti | +2.713 |
neb | +2.306 |
etz | +2.295 |
Probe | +2.264 |
recognition | +2.210 |
Neg logit contributions | |
BELOW | -3.850 |
idated | -3.783 |
productive | -3.735 |
arrass | -3.523 |
ridge | -3.432 |
Token | cur |
Token ID | 1090 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
BELOW | +3.155 |
idated | +3.082 |
productive | +3.066 |
arrass | +2.868 |
ridge | +2.768 |
Neg logit contributions | |
tti | -3.077 |
etz | -2.695 |
Probe | -2.642 |
neb | -2.630 |
earchers | -2.561 |
Token | . |
Token ID | 13 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
BELOW | +0.250 |
idated | +0.244 |
productive | +0.242 |
arrass | +0.225 |
ridge | +0.221 |
Neg logit contributions | |
tti | -0.209 |
etz | -0.180 |
neb | -0.179 |
Probe | -0.177 |
mysteries | -0.171 |
Token | ( |
Token ID | 357 |
Feature activation | -2.91e-01 |
Pos logit contributions | |
BELOW | +1.005 |
idated | +0.975 |
productive | +0.966 |
arrass | +0.898 |
ridge | +0.886 |
Neg logit contributions | |
tti | -0.844 |
neb | -0.736 |
etz | -0.735 |
Probe | -0.716 |
mysteries | -0.698 |
Token | It |
Token ID | 1026 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
tti | +1.888 |
etz | +1.639 |
neb | +1.631 |
Probe | +1.629 |
mysteries | +1.573 |
Neg logit contributions | |
BELOW | -1.909 |
idated | -1.865 |
productive | -1.842 |
arrass | -1.711 |
ridge | -1.673 |
Token | was |
Token ID | 373 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
tti | +1.049 |
etz | +0.905 |
neb | +0.899 |
Probe | +0.891 |
mysteries | +0.862 |
Neg logit contributions | |
BELOW | -1.209 |
idated | -1.177 |
productive | -1.166 |
arrass | -1.084 |
ridge | -1.064 |
Token | an |
Token ID | 281 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
tti | +1.537 |
neb | +1.334 |
etz | +1.334 |
Probe | +1.321 |
mysteries | +1.279 |
Neg logit contributions | |
BELOW | -1.654 |
idated | -1.616 |
productive | -1.599 |
arrass | -1.484 |
ridge | -1.454 |
Token | explosive |
Token ID | 13835 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
BELOW | +4.263 |
idated | +4.059 |
productive | +4.023 |
arrass | +3.785 |
ridge | +3.590 |
Neg logit contributions | |
tti | -3.773 |
etz | -3.296 |
neb | -3.221 |
earchers | -3.180 |
Probe | -3.179 |
Token | 80 |
Token ID | 4019 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
BELOW | +7.901 |
productive | +7.772 |
idated | +7.513 |
arrass | +7.072 |
ridge | +6.813 |
Neg logit contributions | |
tti | -8.297 |
earchers | -7.209 |
etz | -7.192 |
emis | -7.037 |
bourg | -6.816 |
Token | - |
Token ID | 12 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
BELOW | +1.376 |
idated | +1.331 |
productive | +1.330 |
arrass | +1.241 |
ridge | +1.217 |
Neg logit contributions | |
tti | -1.106 |
etz | -0.951 |
neb | -0.947 |
Probe | -0.926 |
mysteries | -0.896 |
Token | yard |
Token ID | 9413 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
productive | +3.045 |
BELOW | +3.032 |
idated | +2.947 |
arrass | +2.684 |
ridge | +2.667 |
Neg logit contributions | |
tti | -4.465 |
neb | -4.076 |
etz | -4.003 |
Probe | -3.961 |
mysteries | -3.890 |
Token | sprint |
Token ID | 18553 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
BELOW | +3.608 |
productive | +3.541 |
idated | +3.504 |
arrass | +3.249 |
ridge | +3.151 |
Neg logit contributions | |
tti | -3.784 |
etz | -3.262 |
neb | -3.232 |
Probe | -3.177 |
mysteries | -3.111 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
tti | +0.405 |
etz | +0.350 |
neb | +0.349 |
Probe | +0.345 |
mysteries | +0.334 |
Neg logit contributions | |
BELOW | -0.470 |
idated | -0.459 |
productive | -0.455 |
arrass | -0.423 |
ridge | -0.415 |
Token | are |
Token ID | 389 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
BELOW | +2.413 |
idated | +2.345 |
productive | +2.316 |
arrass | +2.148 |
ridge | +2.127 |
Neg logit contributions | |
tti | -2.304 |
neb | -2.019 |
etz | -2.000 |
Probe | -1.969 |
mysteries | -1.919 |
Token | looking |
Token ID | 2045 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
BELOW | +1.787 |
idated | +1.719 |
productive | +1.700 |
arrass | +1.576 |
ridge | +1.554 |
Neg logit contributions | |
tti | -1.642 |
etz | -1.438 |
Probe | -1.404 |
neb | -1.400 |
earchers | -1.352 |
Token | to |
Token ID | 284 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
BELOW | +0.967 |
idated | +0.931 |
productive | +0.925 |
arrass | +0.855 |
ridge | +0.838 |
Neg logit contributions | |
tti | -0.872 |
etz | -0.761 |
neb | -0.748 |
Probe | -0.745 |
mysteries | -0.718 |
Token | be |
Token ID | 307 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
tti | +2.702 |
neb | +2.355 |
etz | +2.346 |
Probe | +2.333 |
mysteries | +2.253 |
Neg logit contributions | |
BELOW | -2.919 |
idated | -2.860 |
productive | -2.848 |
arrass | -2.631 |
ridge | -2.581 |
Token | the |
Token ID | 262 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
BELOW | +2.263 |
idated | +2.187 |
productive | +2.173 |
arrass | +2.012 |
ridge | +1.965 |
Neg logit contributions | |
tti | -2.113 |
etz | -1.833 |
Probe | -1.798 |
neb | -1.795 |
mysteries | -1.737 |
Token | little |
Token ID | 1310 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
BELOW | +6.612 |
idated | +6.329 |
productive | +6.151 |
arrass | +5.735 |
ridge | +5.566 |
Neg logit contributions | |
tti | -6.255 |
etz | -5.635 |
earchers | -5.309 |
emis | -5.290 |
neb | -5.263 |
Token | team |
Token ID | 1074 |
Feature activation | +7.73e-01 |
Pos logit contributions | |
BELOW | +8.819 |
idated | +8.242 |
productive | +8.102 |
arrass | +7.671 |
ridge | +7.652 |
Neg logit contributions | |
tti | -8.344 |
etz | -7.856 |
Probe | -7.205 |
emis | -7.148 |
earchers | -6.956 |
Token | that |
Token ID | 326 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
BELOW | +5.525 |
idated | +5.306 |
productive | +5.260 |
ridge | +4.920 |
arrass | +4.891 |
Neg logit contributions | |
tti | -4.381 |
etz | -3.794 |
neb | -3.658 |
Probe | -3.604 |
earchers | -3.526 |
Token | could |
Token ID | 714 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
BELOW | +3.991 |
idated | +3.793 |
productive | +3.712 |
arrass | +3.495 |
ridge | +3.403 |
Neg logit contributions | |
tti | -3.951 |
etz | -3.473 |
neb | -3.438 |
earchers | -3.355 |
Probe | -3.353 |
Token |
|
Token ID | 198 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
BELOW | +2.070 |
idated | +1.979 |
productive | +1.940 |
arrass | +1.837 |
ridge | +1.796 |
Neg logit contributions | |
tti | -1.886 |
etz | -1.656 |
neb | -1.632 |
Probe | -1.607 |
mysteries | -1.567 |
Token |
|
Token ID | 198 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
BELOW | +0.949 |
idated | +0.903 |
productive | +0.898 |
ridge | +0.801 |
arrass | +0.794 |
Neg logit contributions | |
tti | -1.333 |
neb | -1.214 |
etz | -1.201 |
Probe | -1.190 |
earchers | -1.186 |
Token | ed |
Token ID | 276 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
tti | +3.038 |
etz | +2.732 |
neb | +2.629 |
Probe | +2.588 |
mysteries | +2.548 |
Neg logit contributions | |
BELOW | -3.161 |
idated | -2.967 |
productive | -2.961 |
arrass | -2.752 |
ridge | -2.700 |
Token | by |
Token ID | 416 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
tti | +1.103 |
etz | +0.957 |
neb | +0.954 |
Probe | +0.944 |
mysteries | +0.914 |
Neg logit contributions | |
BELOW | -1.243 |
idated | -1.214 |
productive | -1.202 |
arrass | -1.117 |
ridge | -1.095 |
Token | German |
Token ID | 2679 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
BELOW | +0.381 |
idated | +0.373 |
productive | +0.369 |
arrass | +0.343 |
ridge | +0.337 |
Neg logit contributions | |
tti | -0.330 |
etz | -0.285 |
neb | -0.284 |
Probe | -0.282 |
mysteries | -0.272 |
Token | forces |
Token ID | 3386 |
Feature activation | -9.19e-02 |
Pos logit contributions | |
tti | +1.416 |
etz | +1.239 |
neb | +1.236 |
Probe | +1.224 |
mysteries | +1.187 |
Neg logit contributions | |
BELOW | -1.441 |
idated | -1.405 |
productive | -1.391 |
arrass | -1.286 |
ridge | -1.259 |
Token | in |
Token ID | 287 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
tti | +0.581 |
etz | +0.506 |
neb | +0.505 |
Probe | +0.501 |
mysteries | +0.484 |
Neg logit contributions | |
BELOW | -0.617 |
idated | -0.601 |
productive | -0.596 |
arrass | -0.552 |
ridge | -0.542 |
Token | 1943 |
Token ID | 21577 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
tti | +0.349 |
etz | +0.302 |
neb | +0.302 |
Probe | +0.299 |
mysteries | +0.289 |
Neg logit contributions | |
BELOW | -0.398 |
idated | -0.389 |
productive | -0.385 |
arrass | -0.358 |
ridge | -0.351 |
Token | , |
Token ID | 11 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
BELOW | +0.644 |
idated | +0.627 |
productive | +0.621 |
arrass | +0.578 |
ridge | +0.570 |
Neg logit contributions | |
tti | -0.557 |
etz | -0.482 |
neb | -0.481 |
Probe | -0.477 |
mysteries | -0.458 |
Token | many |
Token ID | 867 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
BELOW | +2.679 |
idated | +2.646 |
productive | +2.584 |
arrass | +2.418 |
ridge | +2.362 |
Neg logit contributions | |
tti | -2.717 |
neb | -2.372 |
etz | -2.371 |
Probe | -2.365 |
earchers | -2.301 |
Token | military |
Token ID | 2422 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
BELOW | +1.661 |
idated | +1.612 |
productive | +1.593 |
arrass | +1.468 |
ridge | +1.436 |
Neg logit contributions | |
tti | -1.778 |
etz | -1.572 |
Probe | -1.562 |
neb | -1.555 |
earchers | -1.487 |
Token | - |
Token ID | 12 |
Feature activation | +6.64e-02 |
Pos logit contributions | |
BELOW | +1.029 |
idated | +1.003 |
productive | +0.990 |
arrass | +0.917 |
ridge | +0.899 |
Neg logit contributions | |
tti | -1.005 |
etz | -0.877 |
Probe | -0.869 |
neb | -0.868 |
Television | -0.828 |
Token | aged |
Token ID | 1886 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
BELOW | +0.417 |
idated | +0.406 |
productive | +0.403 |
arrass | +0.371 |
ridge | +0.363 |
Neg logit contributions | |
tti | -0.450 |
etz | -0.395 |
neb | -0.395 |
Probe | -0.392 |
mysteries | -0.379 |
Token | : |
Token ID | 25 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
BELOW | +3.952 |
idated | +3.662 |
productive | +3.630 |
ridge | +3.347 |
arrass | +3.327 |
Neg logit contributions | |
tti | -3.851 |
neb | -3.425 |
etz | -3.410 |
Probe | -3.338 |
earchers | -3.256 |
Token | Fate |
Token ID | 20341 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
tti | +1.300 |
etz | +1.130 |
neb | +1.127 |
Probe | +1.122 |
mysteries | +1.084 |
Neg logit contributions | |
BELOW | -1.358 |
idated | -1.330 |
productive | -1.319 |
arrass | -1.223 |
ridge | -1.194 |
Token | Makes |
Token ID | 27433 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
tti | +6.803 |
etz | +5.811 |
neb | +5.802 |
Probe | +5.776 |
mysteries | +5.616 |
Neg logit contributions | |
BELOW | -7.373 |
idated | -7.351 |
productive | -7.272 |
arrass | -6.764 |
ridge | -6.404 |
Token | a |
Token ID | 257 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
tti | +3.776 |
Probe | +3.205 |
etz | +3.201 |
neb | +3.175 |
Television | +3.095 |
Neg logit contributions | |
BELOW | -5.059 |
idated | -4.994 |
productive | -4.917 |
arrass | -4.599 |
ridge | -4.487 |
Token | Match |
Token ID | 13225 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
tti | +1.520 |
neb | +1.320 |
Probe | +1.320 |
etz | +1.314 |
Television | +1.261 |
Neg logit contributions | |
BELOW | -1.705 |
idated | -1.671 |
productive | -1.648 |
arrass | -1.537 |
ridge | -1.498 |
Token | - |
Token ID | 12 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
tti | +2.680 |
etz | +2.334 |
Probe | +2.327 |
neb | +2.322 |
Television | +2.245 |
Neg logit contributions | |
BELOW | -2.620 |
idated | -2.596 |
productive | -2.549 |
arrass | -2.372 |
ridge | -2.280 |
Token | Up |
Token ID | 4933 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
BELOW | +4.340 |
productive | +4.215 |
idated | +4.209 |
ridge | +3.820 |
arrass | +3.748 |
Neg logit contributions | |
tti | -5.800 |
etz | -5.259 |
neb | -5.211 |
Probe | -5.019 |
mysteries | -5.018 |
Token |
|
Token ID | 198 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
tti | +0.306 |
etz | +0.269 |
neb | +0.268 |
Probe | +0.265 |
mysteries | +0.258 |
Neg logit contributions | |
BELOW | -0.287 |
idated | -0.279 |
productive | -0.276 |
arrass | -0.255 |
ridge | -0.250 |
Token |
|
Token ID | 198 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
BELOW | +1.476 |
idated | +1.390 |
productive | +1.379 |
ridge | +1.232 |
arrass | +1.203 |
Neg logit contributions | |
tti | -2.165 |
neb | -1.995 |
earchers | -1.967 |
etz | -1.963 |
Probe | -1.940 |
Token | On |
Token ID | 2202 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
BELOW | +2.113 |
idated | +2.041 |
productive | +2.031 |
arrass | +1.852 |
ridge | +1.845 |
Neg logit contributions | |
tti | -2.228 |
neb | -2.022 |
Probe | -1.984 |
etz | -1.979 |
mysteries | -1.911 |
Token | December |
Token ID | 3426 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
tti | +2.215 |
etz | +1.916 |
neb | +1.916 |
Probe | +1.906 |
mysteries | +1.838 |
Neg logit contributions | |
BELOW | -2.400 |
idated | -2.348 |
productive | -2.322 |
arrass | -2.169 |
ridge | -2.103 |
Token | put |
Token ID | 1234 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
BELOW | +4.702 |
productive | +4.397 |
idated | +4.319 |
arrass | +4.050 |
ridge | +4.040 |
Neg logit contributions | |
tti | -4.374 |
etz | -3.809 |
neb | -3.680 |
earchers | -3.642 |
Probe | -3.615 |
Token | , |
Token ID | 11 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
BELOW | +4.011 |
idated | +3.862 |
productive | +3.825 |
arrass | +3.540 |
ridge | +3.451 |
Neg logit contributions | |
tti | -4.661 |
etz | -4.147 |
neb | -4.115 |
Probe | -4.092 |
earchers | -3.939 |
Token | this |
Token ID | 428 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
tti | +0.308 |
etz | +0.272 |
neb | +0.271 |
Probe | +0.268 |
mysteries | +0.260 |
Neg logit contributions | |
BELOW | -0.268 |
idated | -0.259 |
productive | -0.256 |
arrass | -0.236 |
ridge | -0.231 |
Token | will |
Token ID | 481 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
BELOW | +0.931 |
idated | +0.904 |
productive | +0.896 |
arrass | +0.832 |
ridge | +0.816 |
Neg logit contributions | |
tti | -0.882 |
etz | -0.775 |
neb | -0.763 |
Probe | -0.759 |
mysteries | -0.731 |
Token | give |
Token ID | 1577 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
BELOW | +3.277 |
idated | +3.072 |
productive | +3.042 |
arrass | +2.876 |
ridge | +2.818 |
Neg logit contributions | |
tti | -3.526 |
etz | -3.167 |
neb | -3.041 |
Probe | -3.026 |
earchers | -3.003 |
Token | anyone |
Token ID | 2687 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
tti | +3.688 |
neb | +3.268 |
etz | +3.214 |
Probe | +3.191 |
mysteries | +3.179 |
Neg logit contributions | |
BELOW | -4.086 |
idated | -4.047 |
productive | -3.977 |
arrass | -3.680 |
ridge | -3.620 |
Token | looking |
Token ID | 2045 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
tti | +1.721 |
neb | +1.491 |
etz | +1.490 |
Probe | +1.473 |
mysteries | +1.428 |
Neg logit contributions | |
BELOW | -1.944 |
idated | -1.903 |
productive | -1.884 |
arrass | -1.750 |
ridge | -1.713 |
Token | to |
Token ID | 284 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
tti | +0.908 |
etz | +0.781 |
neb | +0.780 |
Probe | +0.770 |
mysteries | +0.745 |
Neg logit contributions | |
BELOW | -1.123 |
idated | -1.101 |
productive | -1.091 |
arrass | -1.018 |
ridge | -0.999 |
Token | receive |
Token ID | 3328 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
tti | +1.964 |
neb | +1.716 |
etz | +1.715 |
Probe | +1.700 |
mysteries | +1.649 |
Neg logit contributions | |
BELOW | -1.975 |
idated | -1.928 |
productive | -1.911 |
arrass | -1.765 |
ridge | -1.729 |
Token | their |
Token ID | 511 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
tti | +2.032 |
neb | +1.794 |
etz | +1.792 |
Probe | +1.775 |
mysteries | +1.728 |
Neg logit contributions | |
BELOW | -1.801 |
idated | -1.754 |
productive | -1.736 |
arrass | -1.597 |
ridge | -1.559 |
Token | CPR |
Token ID | 42920 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
BELOW | +2.629 |
idated | +2.505 |
productive | +2.473 |
arrass | +2.239 |
ridge | +2.232 |
Neg logit contributions | |
tti | -3.433 |
etz | -3.088 |
Probe | -2.987 |
neb | -2.971 |
earchers | -2.948 |
Token | consent |
Token ID | 8281 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
BELOW | +2.256 |
idated | +2.179 |
productive | +2.162 |
arrass | +1.984 |
ridge | +1.961 |
Neg logit contributions | |
tti | -2.198 |
etz | -1.907 |
neb | -1.883 |
Probe | -1.872 |
earchers | -1.823 |
Token | or |
Token ID | 393 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
tti | +3.801 |
etz | +3.340 |
neb | +3.313 |
Probe | +3.250 |
recognition | +3.186 |
Neg logit contributions | |
BELOW | -4.407 |
idated | -4.349 |
productive | -4.277 |
arrass | -4.018 |
ridge | -3.932 |
Token | approval |
Token ID | 7546 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
tti | +0.488 |
etz | +0.425 |
neb | +0.422 |
Probe | +0.418 |
mysteries | +0.404 |
Neg logit contributions | |
BELOW | -0.507 |
idated | -0.493 |
productive | -0.488 |
arrass | -0.452 |
ridge | -0.444 |
Token | from |
Token ID | 422 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
tti | +1.847 |
etz | +1.599 |
neb | +1.596 |
Probe | +1.574 |
mysteries | +1.522 |
Neg logit contributions | |
BELOW | -2.205 |
idated | -2.159 |
productive | -2.138 |
arrass | -1.998 |
ridge | -1.958 |
Token | the |
Token ID | 262 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
BELOW | +1.813 |
idated | +1.735 |
productive | +1.715 |
arrass | +1.588 |
ridge | +1.567 |
Neg logit contributions | |
tti | -1.753 |
etz | -1.520 |
neb | -1.505 |
Probe | -1.479 |
mysteries | -1.438 |
Token | Korean |
Token ID | 6983 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
BELOW | +2.267 |
idated | +2.179 |
productive | +2.148 |
arrass | +1.979 |
ridge | +1.960 |
Neg logit contributions | |
tti | -2.105 |
etz | -1.825 |
neb | -1.788 |
Probe | -1.759 |
emis | -1.717 |
Token | Communication |
Token ID | 26117 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
tti | +0.814 |
etz | +0.692 |
neb | +0.687 |
Probe | +0.678 |
mysteries | +0.653 |
Neg logit contributions | |
BELOW | -1.141 |
idated | -1.115 |
productive | -1.104 |
arrass | -1.034 |
ridge | -1.016 |
Token | Commission |
Token ID | 4513 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
tti | +2.763 |
Probe | +2.324 |
etz | +2.309 |
neb | +2.299 |
Television | +2.224 |
Neg logit contributions | |
BELOW | -4.447 |
idated | -4.376 |
productive | -4.354 |
arrass | -4.105 |
ridge | -4.037 |
Token | ." |
Token ID | 526 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
tti | +1.720 |
etz | +1.493 |
neb | +1.488 |
Probe | +1.476 |
mysteries | +1.422 |
Neg logit contributions | |
BELOW | -1.958 |
idated | -1.915 |
productive | -1.898 |
arrass | -1.767 |
ridge | -1.727 |
Token |
|
Token ID | 198 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
tti | +3.199 |
Probe | +2.810 |
etz | +2.787 |
neb | +2.744 |
Television | +2.702 |
Neg logit contributions | |
BELOW | -3.047 |
idated | -3.006 |
productive | -2.981 |
arrass | -2.740 |
ridge | -2.662 |
Token |
|
Token ID | 198 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
tti | +3.659 |
etz | +3.178 |
Probe | +3.134 |
neb | +3.097 |
recognition | +3.034 |
Neg logit contributions | |
BELOW | -3.505 |
idated | -3.485 |
productive | -3.429 |
arrass | -3.244 |
覚醒 | -3.109 |
Token | emergency |
Token ID | 6334 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
BELOW | +0.006 |
idated | +0.006 |
productive | +0.006 |
arrass | +0.005 |
ridge | +0.005 |
Neg logit contributions | |
tti | -0.005 |
etz | -0.005 |
neb | -0.005 |
Probe | -0.005 |
mysteries | -0.005 |
Token | alert |
Token ID | 7995 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
BELOW | +2.384 |
productive | +2.278 |
idated | +2.263 |
arrass | +2.102 |
ridge | +2.061 |
Neg logit contributions | |
tti | -2.347 |
etz | -2.051 |
Probe | -2.006 |
neb | -1.995 |
earchers | -1.977 |
Token | via |
Token ID | 2884 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
BELOW | +1.357 |
idated | +1.299 |
productive | +1.292 |
arrass | +1.194 |
ridge | +1.181 |
Neg logit contributions | |
tti | -1.282 |
etz | -1.114 |
neb | -1.105 |
Probe | -1.103 |
mysteries | -1.068 |
Token | its |
Token ID | 663 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
BELOW | +2.065 |
idated | +1.972 |
productive | +1.958 |
arrass | +1.813 |
ridge | +1.793 |
Neg logit contributions | |
tti | -2.026 |
etz | -1.788 |
neb | -1.755 |
Probe | -1.747 |
mysteries | -1.696 |
Token | website |
Token ID | 3052 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
tti | +1.724 |
etz | +1.506 |
neb | +1.505 |
Probe | +1.490 |
mysteries | +1.444 |
Neg logit contributions | |
BELOW | -1.735 |
idated | -1.694 |
productive | -1.677 |
arrass | -1.552 |
ridge | -1.518 |
Token | early |
Token ID | 1903 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
BELOW | +6.923 |
productive | +6.332 |
idated | +6.236 |
ridge | +5.894 |
arrass | +5.786 |
Neg logit contributions | |
tti | -6.723 |
etz | -5.812 |
neb | -5.753 |
Probe | -5.749 |
earchers | -5.746 |
Token | on |
Token ID | 319 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
BELOW | +2.798 |
idated | +2.662 |
productive | +2.636 |
arrass | +2.405 |
ridge | +2.400 |
Neg logit contributions | |
tti | -3.177 |
etz | -2.801 |
neb | -2.784 |
Probe | -2.762 |
earchers | -2.671 |
Token | . |
Token ID | 13 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
BELOW | +2.067 |
idated | +1.976 |
productive | +1.970 |
arrass | +1.805 |
ridge | +1.800 |
Neg logit contributions | |
tti | -2.133 |
etz | -1.874 |
neb | -1.865 |
Probe | -1.843 |
earchers | -1.787 |
Token | It |
Token ID | 632 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
BELOW | +0.519 |
idated | +0.502 |
productive | +0.497 |
arrass | +0.462 |
ridge | +0.455 |
Neg logit contributions | |
tti | -0.442 |
etz | -0.384 |
neb | -0.384 |
Probe | -0.376 |
mysteries | -0.366 |
Token | read |
Token ID | 1100 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
BELOW | +4.271 |
idated | +4.120 |
productive | +4.033 |
arrass | +3.746 |
ridge | +3.719 |
Neg logit contributions | |
tti | -4.352 |
neb | -3.834 |
Probe | -3.811 |
etz | -3.745 |
earchers | -3.702 |
Token | , |
Token ID | 11 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
BELOW | +4.060 |
idated | +3.852 |
productive | +3.820 |
arrass | +3.547 |
ridge | +3.541 |
Neg logit contributions | |
tti | -3.997 |
etz | -3.519 |
Probe | -3.460 |
neb | -3.439 |
mysteries | -3.319 |
Token | ly |
Token ID | 306 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
BELOW | +0.421 |
idated | +0.411 |
productive | +0.405 |
arrass | +0.371 |
ridge | +0.362 |
Neg logit contributions | |
tti | -0.521 |
etz | -0.462 |
neb | -0.460 |
Probe | -0.457 |
mysteries | -0.443 |
Token | marked |
Token ID | 7498 |
Feature activation | +6.21e-02 |
Pos logit contributions | |
tti | +1.498 |
neb | +1.320 |
etz | +1.319 |
Probe | +1.300 |
mysteries | +1.267 |
Neg logit contributions | |
BELOW | -1.369 |
idated | -1.335 |
productive | -1.322 |
arrass | -1.220 |
ridge | -1.191 |
Token | by |
Token ID | 416 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
BELOW | +0.421 |
idated | +0.410 |
productive | +0.406 |
arrass | +0.376 |
ridge | +0.369 |
Neg logit contributions | |
tti | -0.391 |
etz | -0.340 |
neb | -0.339 |
Probe | -0.336 |
mysteries | -0.324 |
Token | the |
Token ID | 262 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
tti | +0.625 |
etz | +0.539 |
neb | +0.539 |
Probe | +0.532 |
mysteries | +0.515 |
Neg logit contributions | |
BELOW | -0.745 |
idated | -0.729 |
productive | -0.723 |
arrass | -0.673 |
ridge | -0.659 |
Token | over |
Token ID | 625 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
tti | +1.084 |
etz | +0.943 |
neb | +0.942 |
Probe | +0.931 |
mysteries | +0.902 |
Neg logit contributions | |
BELOW | -1.157 |
idated | -1.130 |
productive | -1.119 |
arrass | -1.039 |
ridge | -1.017 |
Token | hang |
Token ID | 33255 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
tti | +10.021 |
etz | +8.651 |
ft | +8.082 |
neb | +7.994 |
bes | +7.987 |
Neg logit contributions | |
BELOW | -13.735 |
idated | -13.607 |
productive | -12.887 |
arrass | -12.560 |
覚醒 | -12.253 |
Token | of |
Token ID | 286 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
tti | +5.261 |
neb | +4.750 |
etz | +4.583 |
recognition | +4.546 |
mysteries | +4.445 |
Neg logit contributions | |
idated | -6.067 |
BELOW | -6.054 |
productive | -5.966 |
arrass | -5.672 |
ridge | -5.387 |
Token | the |
Token ID | 262 |
Feature activation | -6.45e-02 |
Pos logit contributions | |
BELOW | +4.579 |
idated | +4.443 |
productive | +4.380 |
arrass | +4.047 |
ridge | +4.009 |
Neg logit contributions | |
tti | -4.178 |
etz | -3.586 |
neb | -3.555 |
Probe | -3.526 |
earchers | -3.471 |
Token | building |
Token ID | 2615 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
tti | +0.416 |
etz | +0.364 |
neb | +0.362 |
Probe | +0.359 |
mysteries | +0.348 |
Neg logit contributions | |
BELOW | -0.425 |
idated | -0.414 |
productive | -0.410 |
arrass | -0.380 |
ridge | -0.372 |
Token | , |
Token ID | 11 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
BELOW | +3.171 |
idated | +2.905 |
productive | +2.900 |
ridge | +2.577 |
arrass | +2.535 |
Neg logit contributions | |
tti | -4.763 |
etz | -4.252 |
Probe | -4.227 |
neb | -4.147 |
earchers | -4.123 |
Token | they |
Token ID | 484 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
tti | +1.146 |
etz | +0.984 |
neb | +0.983 |
Probe | +0.971 |
mysteries | +0.938 |
Neg logit contributions | |
BELOW | -1.429 |
idated | -1.397 |
productive | -1.385 |
arrass | -1.291 |
ridge | -1.267 |
Token | :// |
Token ID | 1378 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
BELOW | +1.163 |
idated | +1.107 |
productive | +1.097 |
ridge | +0.993 |
arrass | +0.988 |
Neg logit contributions | |
tti | -1.284 |
neb | -1.160 |
etz | -1.131 |
Probe | -1.129 |
earchers | -1.117 |
Token | www |
Token ID | 2503 |
Feature activation | +6.46e-02 |
Pos logit contributions | |
tti | +2.550 |
etz | +2.274 |
neb | +2.247 |
Probe | +2.225 |
mysteries | +2.169 |
Neg logit contributions | |
BELOW | -1.989 |
idated | -1.933 |
productive | -1.896 |
arrass | -1.740 |
ridge | -1.692 |
Token | . |
Token ID | 13 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
BELOW | +0.461 |
idated | +0.445 |
productive | +0.444 |
arrass | +0.408 |
ridge | +0.407 |
Neg logit contributions | |
tti | -0.383 |
neb | -0.339 |
etz | -0.332 |
Probe | -0.330 |
earchers | -0.322 |
Token | A |
Token ID | 32 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
tti | +0.938 |
etz | +0.817 |
neb | +0.816 |
Probe | +0.809 |
mysteries | +0.783 |
Neg logit contributions | |
BELOW | -0.987 |
idated | -0.962 |
productive | -0.954 |
arrass | -0.884 |
ridge | -0.866 |
Token | chieve |
Token ID | 24957 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
tti | +6.772 |
etz | +5.816 |
Probe | +5.762 |
neb | +5.644 |
recognition | +5.585 |
Neg logit contributions | |
BELOW | -7.737 |
idated | -7.498 |
productive | -7.435 |
arrass | -6.920 |
itably | -6.765 |
Token | ment |
Token ID | 434 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
tti | +4.235 |
etz | +3.453 |
Probe | +3.387 |
neb | +3.348 |
recognition | +3.240 |
Neg logit contributions | |
BELOW | -8.181 |
idated | -8.033 |
productive | -7.785 |
arrass | -7.356 |
ridge | -7.259 |
Token | . |
Token ID | 13 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
tti | +5.234 |
etz | +4.581 |
Probe | +4.536 |
neb | +4.472 |
recognition | +4.420 |
Neg logit contributions | |
BELOW | -4.931 |
idated | -4.755 |
productive | -4.649 |
arrass | -4.310 |
ridge | -4.219 |
Token | org |
Token ID | 2398 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
tti | +2.470 |
etz | +2.129 |
neb | +2.100 |
Probe | +2.091 |
mysteries | +2.017 |
Neg logit contributions | |
BELOW | -2.749 |
idated | -2.674 |
productive | -2.639 |
arrass | -2.455 |
ridge | -2.395 |
Token | / |
Token ID | 14 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
BELOW | +1.870 |
idated | +1.812 |
productive | +1.798 |
arrass | +1.653 |
ridge | +1.619 |
Neg logit contributions | |
tti | -2.137 |
etz | -1.883 |
neb | -1.883 |
Probe | -1.865 |
mysteries | -1.814 |
Token | aut |
Token ID | 2306 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
tti | +7.268 |
etz | +6.347 |
neb | +6.144 |
Probe | +6.138 |
recognition | +6.033 |
Neg logit contributions | |
BELOW | -6.745 |
idated | -6.551 |
productive | -6.332 |
arrass | -5.902 |
itably | -5.748 |
Token | od |
Token ID | 375 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
BELOW | +1.661 |
idated | +1.637 |
productive | +1.619 |
arrass | +1.517 |
ridge | +1.481 |
Neg logit contributions | |
tti | -1.854 |
neb | -1.650 |
Probe | -1.634 |
etz | -1.632 |
mysteries | -1.588 |
Token | op |
Token ID | 404 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
tti | +0.365 |
etz | +0.320 |
neb | +0.320 |
Probe | +0.317 |
mysteries | +0.307 |
Neg logit contributions | |
BELOW | -0.354 |
idated | -0.345 |
productive | -0.342 |
arrass | -0.316 |
ridge | -0.309 |
Token | ā |
Token ID | 10235 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
idated | +4.957 |
BELOW | +4.896 |
productive | +4.802 |
arrass | +4.548 |
ridge | +4.520 |
Neg logit contributions | |
tti | -5.307 |
Probe | -4.879 |
neb | -4.796 |
Television | -4.678 |
mysteries | -4.672 |
Token | p |
Token ID | 79 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
tti | +1.304 |
etz | +1.134 |
neb | +1.131 |
Probe | +1.119 |
mysteries | +1.084 |
Neg logit contributions | |
BELOW | -1.398 |
idated | -1.364 |
productive | -1.351 |
arrass | -1.252 |
ridge | -1.227 |
Token | ā |
Token ID | 10235 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
idated | +6.087 |
BELOW | +5.964 |
productive | +5.833 |
ridge | +5.487 |
arrass | +5.484 |
Neg logit contributions | |
tti | -7.249 |
Probe | -6.645 |
neb | -6.547 |
earchers | -6.423 |
mysteries | -6.374 |
Token | . |
Token ID | 13 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
tti | +1.595 |
etz | +1.382 |
neb | +1.380 |
Probe | +1.363 |
mysteries | +1.319 |
Neg logit contributions | |
BELOW | -1.750 |
idated | -1.706 |
productive | -1.689 |
arrass | -1.567 |
ridge | -1.535 |
Token | Pa |
Token ID | 11243 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
BELOW | +1.659 |
idated | +1.611 |
productive | +1.589 |
arrass | +1.468 |
ridge | +1.447 |
Neg logit contributions | |
tti | -1.644 |
etz | -1.441 |
neb | -1.439 |
Probe | -1.412 |
mysteries | -1.389 |
Token | � |
Token ID | 134 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
tti | +5.474 |
etz | +4.923 |
neb | +4.873 |
Probe | +4.711 |
mysteries | +4.650 |
Neg logit contributions | |
BELOW | -5.922 |
productive | -5.700 |
idated | -5.627 |
arrass | -5.256 |
ridge | -5.077 |
Token | � |
Token ID | 119 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
BELOW | +1.115 |
idated | +1.083 |
productive | +1.074 |
arrass | +1.008 |
ridge | +0.981 |
Neg logit contributions | |
tti | -0.912 |
neb | -0.777 |
Probe | -0.771 |
etz | -0.770 |
mysteries | -0.746 |
Token | ē |
Token ID | 27092 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
BELOW | +0.723 |
idated | +0.708 |
productive | +0.700 |
arrass | +0.652 |
ridge | +0.639 |
Neg logit contributions | |
tti | -0.715 |
neb | -0.630 |
Probe | -0.626 |
etz | -0.624 |
mysteries | -0.606 |
Token | ak |
Token ID | 47594 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
tti | +3.534 |
neb | +3.060 |
etz | +3.055 |
Probe | +2.998 |
mysteries | +2.915 |
Neg logit contributions | |
BELOW | -3.612 |
idated | -3.519 |
productive | -3.475 |
arrass | -3.202 |
ridge | -3.119 |
Token | ula |
Token ID | 4712 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
BELOW | +2.088 |
idated | +2.057 |
productive | +2.031 |
arrass | +1.894 |
ridge | +1.875 |
Neg logit contributions | |
tti | -2.294 |
Probe | -2.043 |
neb | -2.033 |
etz | -2.018 |
mysteries | -1.976 |
Token | earlier |
Token ID | 2961 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
tti | +0.724 |
etz | +0.629 |
neb | +0.627 |
Probe | +0.621 |
mysteries | +0.600 |
Neg logit contributions | |
BELOW | -0.796 |
idated | -0.778 |
productive | -0.770 |
arrass | -0.715 |
ridge | -0.700 |
Token | in |
Token ID | 287 |
Feature activation | -3.85e-01 |
Pos logit contributions | |
tti | +4.427 |
neb | +3.898 |
etz | +3.890 |
Probe | +3.865 |
mysteries | +3.766 |
Neg logit contributions | |
BELOW | -4.422 |
idated | -4.364 |
productive | -4.321 |
arrass | -3.994 |
ridge | -3.844 |
Token | the |
Token ID | 262 |
Feature activation | -4.90e-01 |
Pos logit contributions | |
tti | +2.076 |
neb | +1.788 |
etz | +1.779 |
Probe | +1.778 |
mysteries | +1.703 |
Neg logit contributions | |
BELOW | -2.907 |
idated | -2.838 |
productive | -2.817 |
arrass | -2.639 |
ridge | -2.567 |
Token | week |
Token ID | 1285 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
tti | +2.820 |
neb | +2.474 |
etz | +2.429 |
Probe | +2.427 |
mysteries | +2.360 |
Neg logit contributions | |
BELOW | -3.499 |
idated | -3.441 |
productive | -3.401 |
arrass | -3.176 |
ridge | -3.093 |
Token | to |
Token ID | 284 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
BELOW | +2.158 |
idated | +2.078 |
productive | +2.061 |
arrass | +1.905 |
ridge | +1.880 |
Neg logit contributions | |
tti | -2.310 |
neb | -2.032 |
etz | -2.026 |
Probe | -2.008 |
mysteries | -1.948 |
Token | a |
Token ID | 257 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
tti | +2.796 |
etz | +2.432 |
neb | +2.428 |
Probe | +2.416 |
recognition | +2.316 |
Neg logit contributions | |
BELOW | -3.263 |
idated | -3.196 |
productive | -3.186 |
arrass | -2.940 |
ridge | -2.875 |
Token | different |
Token ID | 1180 |
Feature activation | -4.59e-03 |
Pos logit contributions | |
tti | +3.110 |
neb | +2.750 |
Probe | +2.699 |
etz | +2.698 |
recognition | +2.620 |
Neg logit contributions | |
BELOW | -3.391 |
idated | -3.314 |
productive | -3.284 |
arrass | -3.049 |
ridge | -2.972 |
Token | flight |
Token ID | 5474 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
tti | +0.027 |
etz | +0.023 |
neb | +0.022 |
Probe | +0.022 |
mysteries | +0.022 |
Neg logit contributions | |
BELOW | -0.033 |
idated | -0.033 |
productive | -0.032 |
arrass | -0.030 |
ridge | -0.030 |
Token | ( |
Token ID | 357 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
tti | +0.562 |
etz | +0.489 |
neb | +0.488 |
Probe | +0.483 |
mysteries | +0.468 |
Neg logit contributions | |
BELOW | -0.599 |
idated | -0.584 |
productive | -0.578 |
arrass | -0.536 |
ridge | -0.525 |
Token | because |
Token ID | 13893 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
tti | +1.079 |
etz | +0.945 |
neb | +0.941 |
Probe | +0.933 |
mysteries | +0.904 |
Neg logit contributions | |
BELOW | -1.067 |
idated | -1.039 |
productive | -1.027 |
arrass | -0.951 |
ridge | -0.929 |
Token | Exp |
Token ID | 5518 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
tti | +1.489 |
etz | +1.274 |
neb | +1.274 |
Probe | +1.259 |
mysteries | +1.215 |
Neg logit contributions | |
BELOW | -1.898 |
idated | -1.860 |
productive | -1.844 |
arrass | -1.719 |
ridge | -1.686 |
Token | becoming |
Token ID | 5033 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
tti | +0.441 |
etz | +0.384 |
neb | +0.382 |
Probe | +0.379 |
mysteries | +0.366 |
Neg logit contributions | |
BELOW | -0.481 |
idated | -0.468 |
productive | -0.464 |
arrass | -0.431 |
ridge | -0.422 |
Token | one |
Token ID | 530 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
tti | +2.136 |
neb | +1.866 |
etz | +1.853 |
Probe | +1.841 |
recognition | +1.789 |
Neg logit contributions | |
BELOW | -2.491 |
idated | -2.434 |
productive | -2.409 |
arrass | -2.247 |
ridge | -2.186 |
Token | of |
Token ID | 286 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
tti | +3.536 |
neb | +3.132 |
Probe | +3.055 |
etz | +3.026 |
mysteries | +2.992 |
Neg logit contributions | |
BELOW | -4.252 |
idated | -4.083 |
productive | -4.077 |
arrass | -3.813 |
ridge | -3.655 |
Token | the |
Token ID | 262 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
BELOW | +3.745 |
idated | +3.636 |
productive | +3.567 |
arrass | +3.337 |
ridge | +3.329 |
Neg logit contributions | |
tti | -3.452 |
etz | -2.977 |
neb | -2.912 |
Probe | -2.858 |
earchers | -2.822 |
Token | world |
Token ID | 995 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
BELOW | +1.876 |
idated | +1.816 |
productive | +1.783 |
arrass | +1.653 |
ridge | +1.622 |
Neg logit contributions | |
tti | -1.943 |
etz | -1.717 |
Probe | -1.650 |
neb | -1.648 |
earchers | -1.598 |
Token | � |
Token ID | 447 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
BELOW | +3.684 |
idated | +3.564 |
productive | +3.516 |
ridge | +3.302 |
arrass | +3.289 |
Neg logit contributions | |
tti | -2.952 |
etz | -2.541 |
Probe | -2.506 |
earchers | -2.478 |
neb | -2.467 |
Token | � |
Token ID | 247 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
tti | +2.139 |
etz | +1.853 |
neb | +1.847 |
Probe | +1.821 |
mysteries | +1.774 |
Neg logit contributions | |
BELOW | -2.408 |
idated | -2.357 |
productive | -2.327 |
arrass | -2.179 |
ridge | -2.131 |
Token | s |
Token ID | 82 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
BELOW | +0.877 |
idated | +0.855 |
productive | +0.849 |
arrass | +0.777 |
ridge | +0.761 |
Neg logit contributions | |
tti | -1.014 |
etz | -0.897 |
neb | -0.896 |
Probe | -0.891 |
mysteries | -0.863 |
Token | most |
Token ID | 749 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
BELOW | +1.061 |
idated | +1.031 |
productive | +1.020 |
arrass | +0.944 |
ridge | +0.927 |
Neg logit contributions | |
tti | -1.036 |
etz | -0.903 |
Probe | -0.890 |
neb | -0.888 |
earchers | -0.844 |
Token | famous |
Token ID | 5863 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
tti | +5.126 |
neb | +4.652 |
recognition | +4.480 |
etz | +4.442 |
mysteries | +4.434 |
Neg logit contributions | |
BELOW | -5.879 |
idated | -5.818 |
productive | -5.607 |
arrass | -5.290 |
ridge | -5.205 |
Token | av |
Token ID | 1196 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
tti | +0.879 |
etz | +0.757 |
neb | +0.748 |
Probe | +0.743 |
mysteries | +0.710 |
Neg logit contributions | |
BELOW | -1.096 |
idated | -1.067 |
productive | -1.059 |
arrass | -0.988 |
ridge | -0.971 |
Token | separate |
Token ID | 4553 |
Feature activation | -8.07e-03 |
Pos logit contributions | |
tti | +2.868 |
neb | +2.509 |
etz | +2.469 |
Probe | +2.453 |
mysteries | +2.419 |
Neg logit contributions | |
BELOW | -3.119 |
idated | -3.083 |
productive | -3.040 |
arrass | -2.830 |
ridge | -2.772 |
Token | Russian |
Token ID | 3394 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
tti | +0.051 |
etz | +0.044 |
neb | +0.044 |
Probe | +0.043 |
mysteries | +0.042 |
Neg logit contributions | |
BELOW | -0.055 |
idated | -0.053 |
productive | -0.053 |
arrass | -0.049 |
ridge | -0.048 |
Token | intelligence |
Token ID | 4430 |
Feature activation | -4.83e-01 |
Pos logit contributions | |
tti | +4.392 |
neb | +3.789 |
Probe | +3.758 |
etz | +3.712 |
mysteries | +3.632 |
Neg logit contributions | |
BELOW | -6.176 |
idated | -6.105 |
productive | -6.058 |
arrass | -5.678 |
ridge | -5.623 |
Token | groups |
Token ID | 2628 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
tti | +2.285 |
neb | +1.900 |
etz | +1.879 |
Probe | +1.859 |
mysteries | +1.792 |
Neg logit contributions | |
BELOW | -4.005 |
idated | -3.943 |
productive | -3.915 |
arrass | -3.684 |
ridge | -3.627 |
Token | had |
Token ID | 550 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
tti | +4.699 |
etz | +4.118 |
neb | +4.054 |
Probe | +3.999 |
mysteries | +3.869 |
Neg logit contributions | |
BELOW | -5.387 |
idated | -5.331 |
productive | -5.305 |
arrass | -4.924 |
ridge | -4.834 |
Token | gained |
Token ID | 8618 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
tti | +2.551 |
neb | +2.232 |
etz | +2.221 |
Probe | +2.193 |
mysteries | +2.143 |
Neg logit contributions | |
BELOW | -2.713 |
idated | -2.644 |
productive | -2.627 |
arrass | -2.444 |
ridge | -2.414 |
Token | access |
Token ID | 1895 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
tti | +8.441 |
recognition | +7.831 |
neb | +7.654 |
etz | +7.444 |
mysteries | +7.140 |
Neg logit contributions | |
idated | -13.828 |
BELOW | -13.758 |
productive | -13.664 |
覚醒 | -13.347 |
arrass | -12.844 |
Token | to |
Token ID | 284 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
tti | +3.013 |
etz | +2.605 |
neb | +2.565 |
Probe | +2.517 |
mysteries | +2.464 |
Neg logit contributions | |
BELOW | -3.901 |
idated | -3.851 |
productive | -3.775 |
arrass | -3.593 |
ridge | -3.490 |
Token | the |
Token ID | 262 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
tti | +1.398 |
etz | +1.209 |
neb | +1.205 |
Probe | +1.193 |
mysteries | +1.152 |
Neg logit contributions | |
BELOW | -1.614 |
idated | -1.574 |
productive | -1.560 |
arrass | -1.450 |
ridge | -1.422 |
Token | DNC |
Token ID | 17824 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
tti | +2.519 |
neb | +2.212 |
etz | +2.198 |
Probe | +2.186 |
mysteries | +2.123 |
Neg logit contributions | |
BELOW | -2.641 |
idated | -2.584 |
productive | -2.560 |
arrass | -2.380 |
ridge | -2.321 |
Token | 's |
Token ID | 338 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
tti | +2.077 |
neb | +1.810 |
etz | +1.809 |
Probe | +1.796 |
mysteries | +1.736 |
Neg logit contributions | |
BELOW | -2.260 |
idated | -2.213 |
productive | -2.190 |
arrass | -2.038 |
ridge | -1.989 |
Token | central |
Token ID | 4318 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
tti | +2.423 |
etz | +2.079 |
neb | +2.077 |
Probe | +2.064 |
mysteries | +1.987 |
Neg logit contributions | |
BELOW | -2.942 |
idated | -2.881 |
productive | -2.856 |
arrass | -2.656 |
ridge | -2.601 |
Token | government |
Token ID | 1230 |
Feature activation | -9.86e-01 |
Pos logit contributions | |
tti | +0.824 |
etz | +0.697 |
neb | +0.694 |
Probe | +0.686 |
mysteries | +0.657 |
Neg logit contributions | |
BELOW | -1.225 |
idated | -1.199 |
productive | -1.189 |
arrass | -1.114 |
ridge | -1.097 |
Token | . |
Token ID | 13 |
Feature activation | -9.49e-02 |
Pos logit contributions | |
tti | +5.577 |
etz | +4.694 |
neb | +4.687 |
Probe | +4.680 |
recognition | +4.594 |
Neg logit contributions | |
BELOW | -7.090 |
idated | -6.972 |
productive | -6.898 |
arrass | -6.454 |
ridge | -6.293 |
Token |
|
Token ID | 198 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
tti | +0.653 |
etz | +0.575 |
neb | +0.574 |
Probe | +0.569 |
mysteries | +0.552 |
Neg logit contributions | |
BELOW | -0.586 |
idated | -0.570 |
productive | -0.564 |
arrass | -0.520 |
ridge | -0.508 |
Token |
|
Token ID | 198 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
BELOW | +1.391 |
idated | +1.318 |
productive | +1.310 |
ridge | +1.177 |
arrass | +1.153 |
Neg logit contributions | |
tti | -1.992 |
neb | -1.828 |
etz | -1.798 |
earchers | -1.797 |
Probe | -1.783 |
Token | For |
Token ID | 1890 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
tti | +0.582 |
etz | +0.509 |
neb | +0.508 |
Probe | +0.503 |
mysteries | +0.488 |
Neg logit contributions | |
BELOW | -0.589 |
idated | -0.574 |
productive | -0.569 |
arrass | -0.526 |
ridge | -0.515 |
Token | cade |
Token ID | 46395 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
tti | +13.291 |
etz | +11.697 |
bes | +11.397 |
recognition | +11.300 |
bourg | +11.278 |
Neg logit contributions | |
BELOW | -11.840 |
idated | -11.547 |
productive | -11.385 |
覚醒 | -10.481 |
arrass | -10.095 |
Token | ll |
Token ID | 297 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
tti | +2.230 |
etz | +1.902 |
neb | +1.872 |
Probe | +1.839 |
mysteries | +1.784 |
Neg logit contributions | |
BELOW | -2.855 |
idated | -2.800 |
productive | -2.768 |
arrass | -2.579 |
ridge | -2.519 |
Token | was |
Token ID | 373 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
BELOW | +2.059 |
idated | +2.007 |
productive | +1.967 |
arrass | +1.853 |
ridge | +1.829 |
Neg logit contributions | |
tti | -1.910 |
neb | -1.674 |
etz | -1.652 |
Probe | -1.647 |
earchers | -1.618 |
Token | called |
Token ID | 1444 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
BELOW | +2.981 |
idated | +2.890 |
productive | +2.832 |
arrass | +2.661 |
ridge | +2.612 |
Neg logit contributions | |
tti | -3.032 |
etz | -2.632 |
earchers | -2.614 |
neb | -2.602 |
Probe | -2.600 |
Token | to |
Token ID | 284 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
BELOW | +4.930 |
idated | +4.756 |
productive | +4.650 |
arrass | +4.434 |
ridge | +4.408 |
Neg logit contributions | |
tti | -5.032 |
neb | -4.317 |
etz | -4.262 |
Probe | -4.229 |
earchers | -4.226 |
Token | being |
Token ID | 852 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
tti | +2.275 |
etz | +1.968 |
neb | +1.965 |
Probe | +1.948 |
mysteries | +1.882 |
Neg logit contributions | |
BELOW | -2.584 |
idated | -2.526 |
productive | -2.504 |
arrass | -2.325 |
ridge | -2.276 |
Token | around |
Token ID | 1088 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
tti | +4.234 |
neb | +3.727 |
etz | +3.692 |
Probe | +3.676 |
mysteries | +3.570 |
Neg logit contributions | |
BELOW | -4.256 |
idated | -4.179 |
productive | -4.122 |
arrass | -3.805 |
ridge | -3.717 |
Token | Bill |
Token ID | 3941 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
BELOW | +1.989 |
idated | +1.919 |
productive | +1.899 |
arrass | +1.763 |
ridge | +1.742 |
Neg logit contributions | |
tti | -1.913 |
etz | -1.690 |
neb | -1.653 |
Probe | -1.624 |
earchers | -1.594 |
Token | now |
Token ID | 783 |
Feature activation | -4.27e-01 |
Pos logit contributions | |
BELOW | +0.426 |
idated | +0.417 |
productive | +0.412 |
arrass | +0.384 |
ridge | +0.378 |
Neg logit contributions | |
tti | -0.380 |
etz | -0.330 |
neb | -0.330 |
Probe | -0.325 |
mysteries | -0.315 |
Token | , |
Token ID | 11 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
tti | +2.617 |
neb | +2.270 |
etz | +2.255 |
Probe | +2.226 |
mysteries | +2.179 |
Neg logit contributions | |
BELOW | -2.943 |
idated | -2.880 |
productive | -2.852 |
arrass | -2.640 |
ridge | -2.573 |
Token | for |
Token ID | 329 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
tti | +1.763 |
etz | +1.510 |
neb | +1.506 |
Probe | +1.489 |
mysteries | +1.435 |
Neg logit contributions | |
BELOW | -2.271 |
idated | -2.219 |
productive | -2.200 |
arrass | -2.054 |
ridge | -2.017 |
Token | weeks |
Token ID | 2745 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
tti | +13.277 |
recognition | +11.961 |
neb | +11.698 |
Probe | +11.607 |
mysteries | +11.411 |
Neg logit contributions | |
idated | -15.049 |
覚醒 | -14.827 |
BELOW | -14.771 |
productive | -14.498 |
arrass | -13.615 |
Token | and |
Token ID | 290 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
BELOW | +2.364 |
idated | +2.275 |
productive | +2.260 |
arrass | +2.090 |
ridge | +2.065 |
Neg logit contributions | |
tti | -2.355 |
neb | -2.040 |
etz | -2.038 |
Probe | -2.024 |
earchers | -1.966 |
Token | weeks |
Token ID | 2745 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
tti | +0.629 |
etz | +0.499 |
neb | +0.497 |
Probe | +0.489 |
mysteries | +0.460 |
Neg logit contributions | |
BELOW | -1.441 |
idated | -1.415 |
productive | -1.406 |
arrass | -1.330 |
ridge | -1.311 |
Token | , |
Token ID | 11 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
BELOW | +4.292 |
productive | +4.081 |
idated | +4.057 |
arrass | +3.757 |
ridge | +3.729 |
Neg logit contributions | |
tti | -4.451 |
Probe | -3.865 |
etz | -3.861 |
neb | -3.815 |
earchers | -3.782 |
Token | [ |
Token ID | 685 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
tti | +1.378 |
etz | +1.181 |
neb | +1.176 |
Probe | +1.164 |
mysteries | +1.120 |
Neg logit contributions | |
BELOW | -1.773 |
idated | -1.733 |
productive | -1.718 |
arrass | -1.604 |
ridge | -1.575 |
Token | True |
Token ID | 6407 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
tti | +0.909 |
etz | +0.787 |
neb | +0.784 |
Probe | +0.774 |
mysteries | +0.751 |
Neg logit contributions | |
BELOW | -1.053 |
idated | -1.026 |
productive | -1.016 |
arrass | -0.946 |
ridge | -0.927 |
Token | Finn |
Token ID | 15368 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
BELOW | +2.109 |
idated | +2.075 |
productive | +2.063 |
arrass | +1.910 |
ridge | +1.890 |
Neg logit contributions | |
tti | -1.770 |
neb | -1.545 |
etz | -1.523 |
Probe | -1.506 |
mysteries | -1.466 |
Token | s |
Token ID | 82 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
tti | +1.722 |
etz | +1.497 |
neb | +1.483 |
Probe | +1.479 |
Television | +1.443 |
Neg logit contributions | |
BELOW | -1.745 |
idated | -1.688 |
productive | -1.679 |
arrass | -1.547 |
ridge | -1.508 |
Token | , |
Token ID | 11 |
Feature activation | +4.55e-03 |
Pos logit contributions | |
tti | +0.761 |
etz | +0.665 |
neb | +0.658 |
Probe | +0.655 |
Television | +0.634 |
Neg logit contributions | |
BELOW | -0.784 |
idated | -0.762 |
productive | -0.758 |
arrass | -0.698 |
ridge | -0.684 |
Token | has |
Token ID | 468 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
BELOW | +0.031 |
idated | +0.030 |
productive | +0.030 |
arrass | +0.028 |
ridge | +0.027 |
Neg logit contributions | |
tti | -0.028 |
neb | -0.025 |
etz | -0.025 |
Probe | -0.024 |
mysteries | -0.024 |
Token | demanded |
Token ID | 12284 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
tti | +3.454 |
neb | +2.999 |
etz | +2.973 |
Probe | +2.930 |
mysteries | +2.865 |
Neg logit contributions | |
BELOW | -3.781 |
productive | -3.679 |
idated | -3.677 |
arrass | -3.390 |
ridge | -3.343 |
Token | a |
Token ID | 257 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
tti | +10.809 |
recognition | +10.018 |
neb | +9.469 |
Probe | +9.252 |
Television | +9.125 |
Neg logit contributions | |
BELOW | -12.895 |
productive | -12.827 |
idated | -12.696 |
覚醒 | -11.662 |
arrass | -11.625 |
Token | ban |
Token ID | 3958 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
tti | +4.777 |
neb | +4.257 |
Probe | +4.176 |
recognition | +4.123 |
etz | +4.118 |
Neg logit contributions | |
BELOW | -5.218 |
idated | -5.173 |
productive | -5.105 |
arrass | -4.693 |
ridge | -4.591 |
Token | on |
Token ID | 319 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
tti | +1.628 |
etz | +1.387 |
neb | +1.365 |
Probe | +1.334 |
recognition | +1.287 |
Neg logit contributions | |
BELOW | -2.694 |
idated | -2.616 |
productive | -2.611 |
arrass | -2.457 |
ridge | -2.402 |
Token | the |
Token ID | 262 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
BELOW | +0.367 |
idated | +0.357 |
productive | +0.352 |
arrass | +0.328 |
ridge | +0.320 |
Neg logit contributions | |
tti | -0.348 |
etz | -0.302 |
neb | -0.299 |
Probe | -0.299 |
mysteries | -0.287 |
Token | ritual |
Token ID | 12146 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
tti | +1.828 |
neb | +1.588 |
etz | +1.586 |
Probe | +1.565 |
mysteries | +1.518 |
Neg logit contributions | |
BELOW | -2.066 |
idated | -2.016 |
productive | -2.001 |
arrass | -1.859 |
ridge | -1.822 |
Token | that |
Token ID | 326 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
tti | +3.491 |
neb | +3.053 |
etz | +2.977 |
Probe | +2.933 |
mysteries | +2.869 |
Neg logit contributions | |
BELOW | -4.682 |
idated | -4.618 |
productive | -4.577 |
arrass | -4.267 |
ridge | -4.197 |
Token | keeping |
Token ID | 5291 |
Feature activation | -6.59e-01 |
Pos logit contributions | |
tti | +1.576 |
neb | +1.362 |
etz | +1.355 |
Probe | +1.336 |
mysteries | +1.296 |
Neg logit contributions | |
BELOW | -1.956 |
idated | -1.915 |
productive | -1.898 |
arrass | -1.768 |
ridge | -1.738 |
Token | the |
Token ID | 262 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
tti | +3.894 |
neb | +3.447 |
etz | +3.347 |
Probe | +3.337 |
mysteries | +3.325 |
Neg logit contributions | |
BELOW | -4.475 |
idated | -4.425 |
productive | -4.366 |
arrass | -4.065 |
ridge | -4.007 |
Token | Hair |
Token ID | 23102 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
tti | +2.803 |
neb | +2.503 |
etz | +2.451 |
Probe | +2.439 |
mysteries | +2.395 |
Neg logit contributions | |
BELOW | -2.690 |
idated | -2.637 |
productive | -2.593 |
arrass | -2.417 |
ridge | -2.364 |
Token | Club |
Token ID | 6289 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
tti | +5.661 |
etz | +5.092 |
Probe | +5.077 |
neb | +4.953 |
recognition | +4.879 |
Neg logit contributions | |
BELOW | -5.702 |
idated | -5.620 |
productive | -5.473 |
arrass | -5.097 |
itably | -4.993 |
Token | for |
Token ID | 329 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
tti | +5.089 |
neb | +4.492 |
Probe | +4.420 |
etz | +4.406 |
mysteries | +4.281 |
Neg logit contributions | |
BELOW | -5.649 |
idated | -5.550 |
productive | -5.444 |
arrass | -5.096 |
itably | -4.922 |
Token | Growth |
Token ID | 19045 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
tti | +10.712 |
neb | +10.080 |
Television | +9.965 |
Probe | +9.905 |
ked | +9.781 |
Neg logit contributions | |
idated | -10.989 |
BELOW | -10.897 |
productive | -10.261 |
arrass | -9.897 |
覚醒 | -9.655 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
tti | +6.949 |
neb | +6.334 |
Probe | +6.168 |
recognition | +6.061 |
mysteries | +5.986 |
Neg logit contributions | |
idated | -7.874 |
BELOW | -7.850 |
productive | -7.678 |
arrass | -7.251 |
ridge | -7.158 |
Token | is |
Token ID | 318 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
tti | +3.071 |
neb | +2.661 |
etz | +2.642 |
Probe | +2.605 |
mysteries | +2.552 |
Neg logit contributions | |
BELOW | -3.411 |
idated | -3.319 |
productive | -3.273 |
arrass | -3.073 |
ridge | -2.999 |
Token | more |
Token ID | 517 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
tti | +1.627 |
neb | +1.412 |
etz | +1.400 |
Probe | +1.377 |
mysteries | +1.339 |
Neg logit contributions | |
BELOW | -1.908 |
idated | -1.867 |
productive | -1.841 |
arrass | -1.723 |
ridge | -1.689 |
Token | important |
Token ID | 1593 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
tti | +8.850 |
neb | +8.295 |
recognition | +7.588 |
mysteries | +7.456 |
etz | +7.212 |
Neg logit contributions | |
idated | -10.554 |
BELOW | -10.451 |
覚醒 | -9.617 |
productive | -9.595 |
arrass | -9.376 |