Token | < |
Token ID | 1279 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
viol | +1.068 |
NPR | +1.010 |
containment | +0.986 |
IRO | +0.980 |
Carnegie | +0.940 |
Neg logit contributions | |
ommel | -0.978 |
reviewed | -0.911 |
atal | -0.800 |
eson | -0.790 |
ento | -0.788 |
Token | T |
Token ID | 309 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
viol | +0.279 |
NPR | +0.264 |
containment | +0.259 |
IRO | +0.258 |
Carnegie | +0.250 |
Neg logit contributions | |
ommel | -0.183 |
reviewed | -0.167 |
atal | -0.143 |
eson | -0.140 |
ento | -0.140 |
Token | > |
Token ID | 1875 |
Feature activation | +4.56e+00 |
Pos logit contributions | |
viol | +0.289 |
NPR | +0.272 |
containment | +0.266 |
IRO | +0.265 |
Carnegie | +0.256 |
Neg logit contributions | |
ommel | -0.245 |
reviewed | -0.227 |
atal | -0.199 |
eson | -0.197 |
ento | -0.196 |
Token | { |
Token ID | 1391 |
Feature activation | +7.40e+00 |
Pos logit contributions | |
ommel | +2.749 |
reviewed | +2.486 |
atal | +2.180 |
eson | +2.116 |
ento | +2.081 |
Neg logit contributions | |
viol | -4.216 |
NPR | -4.010 |
IRO | -3.902 |
containment | -3.871 |
Carnegie | -3.778 |
Token | ... |
Token ID | 2644 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
ommel | +5.130 |
reviewed | +4.738 |
atal | +4.226 |
eson | +4.164 |
MAC | +4.028 |
Neg logit contributions | |
viol | -5.918 |
NPR | -5.645 |
IRO | -5.527 |
containment | -5.306 |
pees | -5.278 |
Token | } |
Token ID | 1782 |
Feature activation | +1.16e+01 |
Pos logit contributions | |
ommel | +1.774 |
reviewed | +1.603 |
atal | +1.360 |
eson | +1.326 |
ento | +1.307 |
Neg logit contributions | |
viol | -3.111 |
NPR | -2.961 |
IRO | -2.903 |
containment | -2.887 |
Carnegie | -2.804 |
Token | impl |
Token ID | 4114 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
ommel | +7.323 |
reviewed | +6.704 |
atal | +6.106 |
eson | +5.968 |
alloc | +5.779 |
Neg logit contributions | |
viol | -9.170 |
NPR | -8.808 |
IRO | -8.643 |
pees | -8.396 |
Sov | -8.369 |
Token | < |
Token ID | 1279 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
ommel | +4.073 |
reviewed | +3.766 |
atal | +3.456 |
eson | +3.392 |
ento | +3.378 |
Neg logit contributions | |
viol | -3.890 |
NPR | -3.650 |
IRO | -3.573 |
containment | -3.501 |
pees | -3.336 |
Token | T |
Token ID | 309 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
ommel | +1.838 |
reviewed | +1.677 |
atal | +1.442 |
eson | +1.411 |
ento | +1.405 |
Neg logit contributions | |
viol | -2.894 |
NPR | -2.745 |
containment | -2.685 |
IRO | -2.685 |
Carnegie | -2.590 |
Token | > |
Token ID | 1875 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
ommel | +2.179 |
reviewed | +1.994 |
atal | +1.755 |
ento | +1.704 |
eson | +1.699 |
Neg logit contributions | |
viol | -2.864 |
NPR | -2.702 |
IRO | -2.647 |
containment | -2.630 |
Carnegie | -2.539 |
Token | Own |
Token ID | 11744 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
ommel | +2.266 |
reviewed | +2.116 |
atal | +1.922 |
eson | +1.895 |
ento | +1.886 |
Neg logit contributions | |
viol | -1.795 |
NPR | -1.669 |
IRO | -1.617 |
containment | -1.612 |
Carnegie | -1.526 |
Token | must |
Token ID | 1276 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
ommel | +3.617 |
reviewed | +3.360 |
eson | +2.952 |
atal | +2.940 |
ento | +2.859 |
Neg logit contributions | |
viol | -4.741 |
NPR | -4.601 |
IRO | -4.470 |
containment | -4.377 |
pees | -4.235 |
Token | roll |
Token ID | 4836 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
ommel | +5.430 |
reviewed | +5.080 |
atal | +4.612 |
eson | +4.591 |
ento | +4.477 |
Neg logit contributions | |
viol | -3.658 |
NPR | -3.560 |
IRO | -3.480 |
containment | -3.348 |
Carnegie | -3.176 |
Token | to |
Token ID | 284 |
Feature activation | +9.63e-01 |
Pos logit contributions | |
ommel | +2.032 |
reviewed | +1.897 |
atal | +1.663 |
eson | +1.651 |
ento | +1.631 |
Neg logit contributions | |
viol | -2.312 |
NPR | -2.185 |
IRO | -2.130 |
containment | -2.129 |
Carnegie | -2.027 |
Token | avoid |
Token ID | 3368 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
ommel | +0.769 |
reviewed | +0.720 |
atal | +0.646 |
eson | +0.639 |
ento | +0.636 |
Neg logit contributions | |
viol | -0.707 |
NPR | -0.668 |
IRO | -0.652 |
containment | -0.642 |
Carnegie | -0.618 |
Token | following |
Token ID | 1708 |
Feature activation | +6.96e+00 |
Pos logit contributions | |
ommel | +1.076 |
reviewed | +1.006 |
atal | +0.889 |
eson | +0.876 |
ento | +0.872 |
Neg logit contributions | |
viol | -1.150 |
NPR | -1.088 |
IRO | -1.066 |
containment | -1.056 |
Carnegie | -1.017 |
Token | the |
Token ID | 262 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
ommel | +4.745 |
reviewed | +4.460 |
eson | +3.899 |
atal | +3.849 |
Thu | +3.769 |
Neg logit contributions | |
viol | -5.674 |
NPR | -5.318 |
IRO | -5.224 |
containment | -5.207 |
pees | -5.070 |
Token | commands |
Token ID | 9729 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
ommel | +5.229 |
reviewed | +4.766 |
atal | +4.112 |
eson | +4.037 |
certs | +4.008 |
Neg logit contributions | |
viol | -10.250 |
IRO | -10.160 |
NPR | -10.003 |
containment | -9.588 |
Carnegie | -9.537 |
Token | in |
Token ID | 287 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
ommel | +3.207 |
reviewed | +3.032 |
eson | +2.568 |
atal | +2.561 |
ento | +2.547 |
Neg logit contributions | |
viol | -4.098 |
NPR | -3.850 |
containment | -3.809 |
IRO | -3.768 |
pees | -3.649 |
Token | his |
Token ID | 465 |
Feature activation | +5.26e+00 |
Pos logit contributions | |
ommel | +1.898 |
reviewed | +1.779 |
atal | +1.590 |
eson | +1.584 |
ento | +1.552 |
Neg logit contributions | |
viol | -1.755 |
NPR | -1.633 |
IRO | -1.605 |
containment | -1.584 |
pees | -1.528 |
Token | head |
Token ID | 1182 |
Feature activation | +6.48e+00 |
Pos logit contributions | |
ommel | +3.798 |
reviewed | +3.513 |
atal | +3.124 |
eson | +3.118 |
ento | +3.054 |
Neg logit contributions | |
viol | -4.140 |
NPR | -3.941 |
IRO | -3.880 |
containment | -3.735 |
pees | -3.720 |
Token | ). |
Token ID | 737 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
ommel | +3.765 |
reviewed | +3.535 |
atal | +2.899 |
eson | +2.870 |
ento | +2.838 |
Neg logit contributions | |
viol | -6.185 |
NPR | -5.776 |
containment | -5.716 |
IRO | -5.673 |
pees | -5.450 |
Token | etc |
Token ID | 3503 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
ommel | +3.149 |
reviewed | +2.905 |
eson | +2.631 |
atal | +2.594 |
ento | +2.570 |
Neg logit contributions | |
viol | -3.102 |
NPR | -2.955 |
IRO | -2.932 |
containment | -2.836 |
Carnegie | -2.726 |
Token | .) |
Token ID | 2014 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
ommel | +2.015 |
reviewed | +1.856 |
atal | +1.593 |
eson | +1.578 |
ento | +1.562 |
Neg logit contributions | |
viol | -2.858 |
NPR | -2.694 |
containment | -2.652 |
IRO | -2.650 |
Carnegie | -2.535 |
Token | and |
Token ID | 290 |
Feature activation | +4.98e+00 |
Pos logit contributions | |
ommel | +2.744 |
reviewed | +2.529 |
atal | +2.231 |
eson | +2.207 |
ento | +2.181 |
Neg logit contributions | |
viol | -3.264 |
NPR | -3.124 |
IRO | -3.049 |
containment | -2.991 |
Carnegie | -2.904 |
Token | what |
Token ID | 644 |
Feature activation | +6.13e+00 |
Pos logit contributions | |
ommel | +3.452 |
reviewed | +3.164 |
atal | +2.765 |
eson | +2.752 |
ento | +2.732 |
Neg logit contributions | |
viol | -4.130 |
NPR | -3.959 |
IRO | -3.833 |
containment | -3.776 |
Carnegie | -3.675 |
Token | scope |
Token ID | 8354 |
Feature activation | +5.94e+00 |
Pos logit contributions | |
ommel | +3.493 |
reviewed | +3.208 |
eson | +2.687 |
atal | +2.623 |
ento | +2.569 |
Neg logit contributions | |
viol | -5.750 |
NPR | -5.543 |
IRO | -5.440 |
pees | -5.341 |
Carnegie | -5.175 |
Token | do |
Token ID | 466 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
ommel | +4.001 |
reviewed | +3.704 |
atal | +3.206 |
eson | +3.202 |
ento | +3.196 |
Neg logit contributions | |
viol | -4.951 |
NPR | -4.826 |
IRO | -4.740 |
containment | -4.598 |
Carnegie | -4.479 |
Token | those |
Token ID | 883 |
Feature activation | +5.46e+00 |
Pos logit contributions | |
ommel | +6.430 |
reviewed | +6.023 |
atal | +5.420 |
certs | +5.158 |
ederal | +5.063 |
Neg logit contributions | |
viol | -8.705 |
NPR | -8.542 |
IRO | -8.507 |
containment | -8.105 |
pees | -7.951 |
Token | parameters |
Token ID | 10007 |
Feature activation | +5.98e+00 |
Pos logit contributions | |
ommel | +2.742 |
reviewed | +2.476 |
atal | +2.069 |
eson | +2.060 |
ento | +1.945 |
Neg logit contributions | |
viol | -5.523 |
NPR | -5.283 |
IRO | -5.195 |
containment | -5.096 |
Carnegie | -5.029 |
Token | have |
Token ID | 423 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
ommel | +3.877 |
reviewed | +3.582 |
atal | +3.085 |
eson | +3.079 |
ento | +3.017 |
Neg logit contributions | |
viol | -5.259 |
NPR | -4.998 |
IRO | -4.838 |
containment | -4.775 |
pees | -4.680 |
Token | ? |
Token ID | 30 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ommel | +1.305 |
reviewed | +1.195 |
atal | +1.020 |
eson | +1.004 |
ento | +1.000 |
Neg logit contributions | |
viol | -1.991 |
NPR | -1.880 |
containment | -1.848 |
IRO | -1.839 |
Carnegie | -1.777 |
Token | can |
Token ID | 460 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
ommel | +1.012 |
reviewed | +0.940 |
atal | +0.848 |
eson | +0.840 |
ento | +0.830 |
Neg logit contributions | |
viol | -0.941 |
NPR | -0.892 |
IRO | -0.865 |
containment | -0.853 |
Carnegie | -0.821 |
Token | Share |
Token ID | 8734 |
Feature activation | +9.75e+00 |
Pos logit contributions | |
ommel | +3.826 |
reviewed | +3.565 |
atal | +3.074 |
Thu | +2.940 |
ento | +2.901 |
Neg logit contributions | |
viol | -5.819 |
IRO | -5.443 |
NPR | -5.421 |
containment | -5.354 |
Sov | -5.207 |
Token | this |
Token ID | 428 |
Feature activation | +4.97e+00 |
Pos logit contributions | |
ommel | +5.099 |
reviewed | +4.489 |
atal | +3.827 |
Thu | +3.764 |
eson | +3.724 |
Neg logit contributions | |
viol | -9.289 |
NPR | -8.850 |
IRO | -8.685 |
containment | -8.565 |
pees | -8.481 |
Token | comment |
Token ID | 2912 |
Feature activation | +6.88e+00 |
Pos logit contributions | |
ommel | +3.039 |
reviewed | +2.789 |
atal | +2.352 |
eson | +2.315 |
Thu | +2.250 |
Neg logit contributions | |
viol | -4.504 |
NPR | -4.315 |
IRO | -4.251 |
containment | -4.141 |
pees | -4.103 |
Token | on |
Token ID | 319 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
ommel | +4.015 |
reviewed | +3.751 |
atal | +3.082 |
eson | +3.070 |
Thu | +3.011 |
Neg logit contributions | |
viol | -6.386 |
NPR | -5.975 |
containment | -5.973 |
IRO | -5.937 |
pees | -5.740 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +5.99e+00 |
Pos logit contributions | |
ommel | +1.057 |
reviewed | +0.982 |
atal | +0.873 |
eson | +0.861 |
ento | +0.856 |
Neg logit contributions | |
viol | -1.070 |
NPR | -1.007 |
IRO | -0.982 |
containment | -0.980 |
Carnegie | -0.933 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
ommel | +3.703 |
reviewed | +3.368 |
atal | +2.896 |
eson | +2.847 |
ento | +2.810 |
Neg logit contributions | |
viol | -5.438 |
NPR | -5.155 |
IRO | -5.107 |
containment | -4.973 |
pees | -4.863 |
Token | this |
Token ID | 428 |
Feature activation | +6.99e+00 |
Pos logit contributions | |
ommel | +5.384 |
reviewed | +4.951 |
eson | +3.978 |
Thu | +3.969 |
ento | +3.921 |
Neg logit contributions | |
viol | -10.175 |
NPR | -9.541 |
IRO | -9.479 |
containment | -9.467 |
�士 | -9.323 |
Token | comment |
Token ID | 2912 |
Feature activation | +8.66e+00 |
Pos logit contributions | |
ommel | +4.084 |
reviewed | +3.744 |
atal | +3.073 |
eson | +3.057 |
ento | +2.974 |
Neg logit contributions | |
viol | -6.402 |
NPR | -6.172 |
IRO | -6.103 |
containment | -5.912 |
pees | -5.898 |
Token | on |
Token ID | 319 |
Feature activation | +8.64e+00 |
Pos logit contributions | |
ommel | +4.736 |
reviewed | +4.468 |
eson | +3.616 |
atal | +3.515 |
Thu | +3.493 |
Neg logit contributions | |
viol | -8.164 |
containment | -7.668 |
NPR | -7.623 |
IRO | -7.583 |
�士 | -7.394 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
ommel | +5.155 |
reviewed | +4.924 |
Thu | +4.199 |
Thu | +4.196 |
Elm | +4.150 |
Neg logit contributions | |
viol | -7.440 |
containment | -7.019 |
NPR | -6.976 |
IRO | -6.892 |
pees | -6.875 |
Token |
|
Token ID | 198 |
Feature activation | +5.49e+00 |
Pos logit contributions | |
ommel | +0.964 |
reviewed | +0.890 |
atal | +0.788 |
eson | +0.776 |
ento | +0.772 |
Neg logit contributions | |
viol | -1.067 |
NPR | -1.005 |
containment | -0.986 |
IRO | -0.981 |
Carnegie | -0.941 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
ommel | +3.546 |
reviewed | +3.308 |
atal | +2.831 |
ento | +2.709 |
Thu | +2.707 |
Neg logit contributions | |
viol | -5.601 |
IRO | -5.269 |
NPR | -5.262 |
containment | -5.160 |
Sov | -5.102 |
Token | this |
Token ID | 428 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
ommel | +5.074 |
reviewed | +4.328 |
ech | +3.759 |
ederal | +3.721 |
Thu | +3.720 |
Neg logit contributions | |
viol | -9.890 |
NPR | -9.298 |
Sov | -9.203 |
containment | -9.102 |
IRO | -9.091 |
Token | comment |
Token ID | 2912 |
Feature activation | +7.13e+00 |
Pos logit contributions | |
ommel | +3.155 |
reviewed | +2.968 |
atal | +2.426 |
eson | +2.368 |
alloc | +2.302 |
Neg logit contributions | |
viol | -4.877 |
NPR | -4.560 |
IRO | -4.503 |
containment | -4.382 |
Sov | -4.320 |
Token | on |
Token ID | 319 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
ommel | +4.445 |
reviewed | +4.078 |
atal | +3.480 |
eson | +3.404 |
MAC | +3.307 |
Neg logit contributions | |
viol | -6.410 |
NPR | -6.075 |
containment | -5.982 |
IRO | -5.980 |
pees | -5.663 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
ommel | +3.227 |
reviewed | +2.976 |
atal | +2.610 |
Thu | +2.498 |
ederal | +2.497 |
Neg logit contributions | |
viol | -3.877 |
IRO | -3.638 |
NPR | -3.625 |
containment | -3.566 |
pees | -3.403 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
ommel | +1.740 |
reviewed | +1.567 |
atal | +1.357 |
eson | +1.331 |
ento | +1.326 |
Neg logit contributions | |
viol | -2.292 |
NPR | -2.142 |
IRO | -2.117 |
containment | -2.106 |
pees | -2.039 |
Token | this |
Token ID | 428 |
Feature activation | +5.29e+00 |
Pos logit contributions | |
ommel | +5.486 |
reviewed | +4.858 |
ourke | +4.059 |
Thu | +3.989 |
ederal | +3.964 |
Neg logit contributions | |
viol | -9.996 |
containment | -9.267 |
NPR | -9.225 |
Sov | -9.059 |
IRO | -8.986 |
Token | comment |
Token ID | 2912 |
Feature activation | +7.64e+00 |
Pos logit contributions | |
ommel | +3.298 |
reviewed | +3.064 |
atal | +2.546 |
eson | +2.517 |
Thu | +2.466 |
Neg logit contributions | |
viol | -4.801 |
NPR | -4.527 |
IRO | -4.468 |
containment | -4.392 |
pees | -4.292 |
Token | on |
Token ID | 319 |
Feature activation | +6.65e+00 |
Pos logit contributions | |
ommel | +4.358 |
reviewed | +4.030 |
eson | +3.278 |
atal | +3.256 |
Thu | +3.189 |
Neg logit contributions | |
viol | -7.090 |
NPR | -6.683 |
containment | -6.664 |
IRO | -6.611 |
pees | -6.380 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
ommel | +4.353 |
reviewed | +4.038 |
atal | +3.541 |
Thu | +3.528 |
eson | +3.488 |
Neg logit contributions | |
viol | -5.518 |
IRO | -5.261 |
NPR | -5.248 |
containment | -5.186 |
Sov | -5.034 |
Token |
|
Token ID | 198 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
ommel | +2.139 |
reviewed | +1.961 |
atal | +1.746 |
eson | +1.709 |
ento | +1.706 |
Neg logit contributions | |
viol | -2.378 |
NPR | -2.239 |
IRO | -2.193 |
containment | -2.190 |
Carnegie | -2.084 |
Token | l |
Token ID | 300 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
viol | +1.804 |
NPR | +1.688 |
containment | +1.656 |
IRO | +1.650 |
Carnegie | +1.589 |
Neg logit contributions | |
ommel | -1.398 |
reviewed | -1.281 |
atal | -1.111 |
eson | -1.099 |
ento | -1.087 |
Token | ' |
Token ID | 6 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
ommel | +0.373 |
reviewed | +0.345 |
atal | +0.305 |
eson | +0.300 |
ento | +0.299 |
Neg logit contributions | |
viol | -0.426 |
NPR | -0.401 |
containment | -0.392 |
IRO | -0.390 |
Carnegie | -0.376 |
Token | eng |
Token ID | 1516 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ommel | +3.507 |
reviewed | +3.283 |
atal | +2.930 |
ento | +2.854 |
eson | +2.803 |
Neg logit contributions | |
viol | -3.958 |
NPR | -3.698 |
containment | -3.682 |
IRO | -3.614 |
Carnegie | -3.538 |
Token | agement |
Token ID | 5082 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ommel | +0.642 |
reviewed | +0.553 |
atal | +0.424 |
eson | +0.407 |
ento | +0.404 |
Neg logit contributions | |
viol | -1.901 |
NPR | -1.820 |
containment | -1.793 |
IRO | -1.788 |
Carnegie | -1.743 |
Token | qu |
Token ID | 627 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
ommel | +1.335 |
reviewed | +1.241 |
atal | +1.108 |
ento | +1.091 |
eson | +1.086 |
Neg logit contributions | |
viol | -1.348 |
NPR | -1.285 |
IRO | -1.245 |
containment | -1.241 |
Carnegie | -1.196 |
Token | ' |
Token ID | 6 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
ommel | +1.511 |
reviewed | +1.384 |
atal | +1.223 |
eson | +1.196 |
ento | +1.196 |
Neg logit contributions | |
viol | -1.911 |
NPR | -1.810 |
containment | -1.779 |
IRO | -1.770 |
Carnegie | -1.709 |
Token | ils |
Token ID | 4487 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
ommel | +6.838 |
reviewed | +6.326 |
atal | +5.700 |
ento | +5.599 |
eson | +5.535 |
Neg logit contributions | |
viol | -8.601 |
containment | -8.235 |
NPR | -8.030 |
IRO | -7.863 |
Carnegie | -7.799 |
Token | av |
Token ID | 1196 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
ommel | +1.867 |
reviewed | +1.732 |
atal | +1.556 |
ento | +1.546 |
eson | +1.538 |
Neg logit contributions | |
viol | -1.745 |
NPR | -1.664 |
IRO | -1.618 |
containment | -1.606 |
Carnegie | -1.550 |
Token | a |
Token ID | 64 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
viol | +0.671 |
NPR | +0.632 |
IRO | +0.615 |
containment | +0.614 |
Carnegie | +0.592 |
Neg logit contributions | |
ommel | -0.539 |
reviewed | -0.497 |
atal | -0.433 |
eson | -0.428 |
ento | -0.423 |
Token | ient |
Token ID | 1153 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
ommel | +2.537 |
reviewed | +2.292 |
eson | +2.040 |
atal | +2.026 |
ento | +2.013 |
Neg logit contributions | |
viol | -3.418 |
NPR | -3.285 |
containment | -3.214 |
IRO | -3.211 |
Carnegie | -3.148 |
Token | pr |
Token ID | 778 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
ommel | +2.959 |
reviewed | +2.744 |
atal | +2.472 |
ento | +2.437 |
eson | +2.419 |
Neg logit contributions | |
viol | -2.751 |
NPR | -2.666 |
IRO | -2.590 |
containment | -2.536 |
Carnegie | -2.466 |
Token | Share |
Token ID | 8734 |
Feature activation | +9.62e+00 |
Pos logit contributions | |
ommel | +3.811 |
reviewed | +3.547 |
atal | +3.066 |
eson | +2.925 |
Thu | +2.920 |
Neg logit contributions | |
viol | -5.520 |
NPR | -5.167 |
IRO | -5.154 |
containment | -5.071 |
pees | -4.899 |
Token | this |
Token ID | 428 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
ommel | +4.994 |
reviewed | +4.498 |
atal | +3.727 |
Thu | +3.718 |
eson | +3.685 |
Neg logit contributions | |
viol | -9.197 |
NPR | -8.728 |
IRO | -8.606 |
containment | -8.520 |
pees | -8.455 |
Token | comment |
Token ID | 2912 |
Feature activation | +7.11e+00 |
Pos logit contributions | |
ommel | +2.960 |
reviewed | +2.721 |
atal | +2.292 |
eson | +2.267 |
ento | +2.210 |
Neg logit contributions | |
viol | -4.287 |
NPR | -4.108 |
IRO | -4.045 |
containment | -3.940 |
pees | -3.903 |
Token | on |
Token ID | 319 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ommel | +4.107 |
reviewed | +3.849 |
atal | +3.140 |
eson | +3.135 |
Thu | +3.086 |
Neg logit contributions | |
viol | -6.615 |
containment | -6.189 |
NPR | -6.183 |
IRO | -6.146 |
pees | -5.962 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +6.73e+00 |
Pos logit contributions | |
ommel | +1.318 |
reviewed | +1.224 |
atal | +1.085 |
eson | +1.072 |
ento | +1.063 |
Neg logit contributions | |
viol | -1.354 |
NPR | -1.274 |
IRO | -1.244 |
containment | -1.241 |
Carnegie | -1.178 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
ommel | +4.082 |
reviewed | +3.713 |
atal | +3.182 |
eson | +3.128 |
ento | +3.084 |
Neg logit contributions | |
viol | -6.150 |
NPR | -5.847 |
IRO | -5.759 |
containment | -5.643 |
pees | -5.503 |
Token | this |
Token ID | 428 |
Feature activation | +6.80e+00 |
Pos logit contributions | |
ommel | +5.299 |
reviewed | +4.886 |
eson | +3.931 |
Thu | +3.924 |
ento | +3.867 |
Neg logit contributions | |
viol | -9.929 |
NPR | -9.323 |
IRO | -9.263 |
containment | -9.239 |
pees | -9.115 |
Token | comment |
Token ID | 2912 |
Feature activation | +8.55e+00 |
Pos logit contributions | |
ommel | +4.042 |
reviewed | +3.698 |
atal | +3.053 |
eson | +3.043 |
ento | +2.964 |
Neg logit contributions | |
viol | -6.182 |
NPR | -5.962 |
IRO | -5.886 |
containment | -5.709 |
pees | -5.697 |
Token | on |
Token ID | 319 |
Feature activation | +7.80e+00 |
Pos logit contributions | |
ommel | +4.709 |
reviewed | +4.423 |
eson | +3.595 |
atal | +3.504 |
Thu | +3.476 |
Neg logit contributions | |
viol | -8.052 |
containment | -7.566 |
NPR | -7.526 |
IRO | -7.482 |
pees | -7.294 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
ommel | +4.876 |
reviewed | +4.644 |
Thu | +3.953 |
Thu | +3.910 |
atal | +3.882 |
Neg logit contributions | |
viol | -6.636 |
containment | -6.239 |
NPR | -6.205 |
IRO | -6.101 |
pees | -6.075 |
Token |
|
Token ID | 198 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
ommel | +0.877 |
reviewed | +0.810 |
atal | +0.717 |
eson | +0.706 |
ento | +0.703 |
Neg logit contributions | |
viol | -0.973 |
NPR | -0.916 |
containment | -0.899 |
IRO | -0.894 |
Carnegie | -0.859 |
Token | 14 |
Token ID | 1415 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
ommel | +3.104 |
reviewed | +2.926 |
atal | +2.424 |
Thu | +2.406 |
MAC | +2.369 |
Neg logit contributions | |
viol | -5.095 |
containment | -4.723 |
NPR | -4.666 |
IRO | -4.577 |
Carnegie | -4.556 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +6.10e+00 |
Pos logit contributions | |
ommel | +1.510 |
reviewed | +1.370 |
atal | +1.105 |
Thu | +1.047 |
ento | +1.043 |
Neg logit contributions | |
viol | -3.381 |
NPR | -3.237 |
containment | -3.178 |
IRO | -3.166 |
Carnegie | -3.086 |
Token |
|
Token ID | 198 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
ommel | +3.854 |
reviewed | +3.618 |
atal | +3.091 |
eson | +3.011 |
Thu | +2.984 |
Neg logit contributions | |
viol | -5.433 |
containment | -5.026 |
NPR | -5.018 |
IRO | -4.941 |
pees | -4.807 |
Token |
|
Token ID | 198 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
ommel | +2.924 |
reviewed | +2.756 |
atal | +2.361 |
eson | +2.355 |
ento | +2.313 |
Neg logit contributions | |
viol | -3.641 |
NPR | -3.370 |
containment | -3.357 |
IRO | -3.315 |
pees | -3.184 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
ommel | +2.683 |
reviewed | +2.500 |
atal | +2.085 |
eson | +2.028 |
ento | +2.017 |
Neg logit contributions | |
viol | -4.485 |
NPR | -4.173 |
containment | -4.168 |
IRO | -4.123 |
Carnegie | -4.036 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
ommel | +3.546 |
reviewed | +3.308 |
atal | +2.831 |
ento | +2.709 |
Thu | +2.707 |
Neg logit contributions | |
viol | -5.601 |
IRO | -5.269 |
NPR | -5.262 |
containment | -5.160 |
Sov | -5.102 |
Token | this |
Token ID | 428 |
Feature activation | +5.22e+00 |
Pos logit contributions | |
ommel | +5.074 |
reviewed | +4.328 |
ech | +3.759 |
ederal | +3.721 |
Thu | +3.720 |
Neg logit contributions | |
viol | -9.890 |
NPR | -9.298 |
Sov | -9.203 |
containment | -9.102 |
IRO | -9.091 |
Token | comment |
Token ID | 2912 |
Feature activation | +7.13e+00 |
Pos logit contributions | |
ommel | +3.155 |
reviewed | +2.968 |
atal | +2.426 |
eson | +2.368 |
alloc | +2.302 |
Neg logit contributions | |
viol | -4.877 |
NPR | -4.560 |
IRO | -4.503 |
containment | -4.382 |
Sov | -4.320 |
Token | on |
Token ID | 319 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
ommel | +4.445 |
reviewed | +4.078 |
atal | +3.480 |
eson | +3.404 |
MAC | +3.307 |
Neg logit contributions | |
viol | -6.410 |
NPR | -6.075 |
containment | -5.982 |
IRO | -5.980 |
pees | -5.663 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
ommel | +3.227 |
reviewed | +2.976 |
atal | +2.610 |
Thu | +2.498 |
ederal | +2.497 |
Neg logit contributions | |
viol | -3.877 |
IRO | -3.638 |
NPR | -3.625 |
containment | -3.566 |
pees | -3.403 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
ommel | +1.740 |
reviewed | +1.567 |
atal | +1.357 |
eson | +1.331 |
ento | +1.326 |
Neg logit contributions | |
viol | -2.292 |
NPR | -2.142 |
IRO | -2.117 |
containment | -2.106 |
pees | -2.039 |
Token | ents |
Token ID | 658 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
viol | +1.060 |
NPR | +1.009 |
IRO | +0.988 |
containment | +0.972 |
Carnegie | +0.939 |
Neg logit contributions | |
ommel | -0.881 |
reviewed | -0.830 |
atal | -0.716 |
eson | -0.704 |
ento | -0.702 |
Token | ind |
Token ID | 773 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
ommel | +0.193 |
reviewed | +0.179 |
atal | +0.160 |
eson | +0.158 |
ento | +0.157 |
Neg logit contributions | |
viol | -0.191 |
NPR | -0.180 |
containment | -0.175 |
IRO | -0.175 |
Carnegie | -0.168 |
Token | iqu |
Token ID | 1557 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
viol | +0.057 |
NPR | +0.054 |
containment | +0.053 |
IRO | +0.053 |
Carnegie | +0.051 |
Neg logit contributions | |
ommel | -0.043 |
reviewed | -0.040 |
atal | -0.034 |
eson | -0.034 |
ento | -0.034 |
Token | ent |
Token ID | 298 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
ommel | +0.887 |
reviewed | +0.814 |
atal | +0.710 |
eson | +0.697 |
ento | +0.694 |
Neg logit contributions | |
viol | -1.224 |
NPR | -1.157 |
containment | -1.132 |
IRO | -1.130 |
Carnegie | -1.092 |
Token | qu |
Token ID | 627 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
ommel | +2.092 |
reviewed | +1.946 |
atal | +1.739 |
ento | +1.709 |
eson | +1.704 |
Neg logit contributions | |
viol | -2.180 |
NPR | -2.070 |
IRO | -2.012 |
containment | -2.004 |
Carnegie | -1.947 |
Token | ' |
Token ID | 6 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
ommel | +1.403 |
reviewed | +1.285 |
atal | +1.142 |
eson | +1.117 |
ento | +1.114 |
Neg logit contributions | |
viol | -1.733 |
NPR | -1.637 |
containment | -1.612 |
IRO | -1.598 |
Carnegie | -1.545 |
Token | ils |
Token ID | 4487 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
ommel | +6.347 |
reviewed | +5.808 |
atal | +5.224 |
ento | +5.116 |
eson | +5.060 |
Neg logit contributions | |
viol | -8.752 |
containment | -8.386 |
NPR | -8.172 |
IRO | -7.987 |
Carnegie | -7.960 |
Token | ont |
Token ID | 39585 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ommel | +4.436 |
reviewed | +4.105 |
ento | +3.724 |
atal | +3.716 |
eson | +3.667 |
Neg logit contributions | |
viol | -3.951 |
NPR | -3.856 |
IRO | -3.743 |
containment | -3.687 |
Carnegie | -3.578 |
Token | mont |
Token ID | 40689 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
viol | +1.417 |
NPR | +1.312 |
containment | +1.284 |
IRO | +1.276 |
Carnegie | +1.225 |
Neg logit contributions | |
ommel | -1.501 |
reviewed | -1.402 |
atal | -1.250 |
eson | -1.235 |
ento | -1.228 |
Token | é |
Token ID | 2634 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
viol | +0.570 |
NPR | +0.535 |
containment | +0.522 |
IRO | +0.521 |
Carnegie | +0.501 |
Neg logit contributions | |
ommel | -0.513 |
reviewed | -0.476 |
atal | -0.419 |
eson | -0.413 |
ento | -0.411 |
Token | une |
Token ID | 17809 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ommel | +1.071 |
reviewed | +0.996 |
atal | +0.890 |
ento | +0.873 |
eson | +0.870 |
Neg logit contributions | |
viol | -1.102 |
NPR | -1.043 |
containment | -1.013 |
IRO | -1.011 |
Carnegie | -0.975 |
Token | > |
Token ID | 1875 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
ommel | +2.132 |
reviewed | +1.953 |
atal | +1.714 |
eson | +1.668 |
ento | +1.668 |
Neg logit contributions | |
viol | -2.870 |
NPR | -2.714 |
IRO | -2.653 |
containment | -2.643 |
Carnegie | -2.549 |
Token | Atomic |
Token ID | 28976 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
ommel | +3.027 |
reviewed | +2.834 |
atal | +2.571 |
eson | +2.524 |
ento | +2.498 |
Neg logit contributions | |
viol | -2.690 |
NPR | -2.494 |
IRO | -2.435 |
containment | -2.415 |
pees | -2.289 |
Token | < |
Token ID | 1279 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
ommel | +2.086 |
reviewed | +1.952 |
atal | +1.766 |
eson | +1.743 |
ento | +1.738 |
Neg logit contributions | |
viol | -1.684 |
NPR | -1.559 |
containment | -1.516 |
IRO | -1.512 |
Carnegie | -1.443 |
Token | T |
Token ID | 309 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
ommel | +0.915 |
reviewed | +0.836 |
atal | +0.722 |
eson | +0.710 |
ento | +0.707 |
Neg logit contributions | |
viol | -1.300 |
NPR | -1.231 |
containment | -1.207 |
IRO | -1.202 |
Carnegie | -1.167 |
Token | > |
Token ID | 1875 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
ommel | +1.433 |
reviewed | +1.318 |
atal | +1.152 |
eson | +1.132 |
ento | +1.128 |
Neg logit contributions | |
viol | -1.846 |
NPR | -1.742 |
containment | -1.703 |
IRO | -1.701 |
Carnegie | -1.639 |
Token | { |
Token ID | 1391 |
Feature activation | +9.80e+00 |
Pos logit contributions | |
ommel | +3.097 |
reviewed | +2.808 |
atal | +2.426 |
eson | +2.376 |
ento | +2.374 |
Neg logit contributions | |
viol | -4.913 |
NPR | -4.664 |
IRO | -4.555 |
containment | -4.527 |
Carnegie | -4.394 |
Token | /// |
Token ID | 34013 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
ommel | +6.358 |
reviewed | +5.860 |
atal | +5.211 |
eson | +5.178 |
alloc | +4.986 |
Neg logit contributions | |
viol | -7.980 |
NPR | -7.626 |
IRO | -7.415 |
pees | -7.247 |
containment | -7.137 |
Token | Create |
Token ID | 13610 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ommel | +2.762 |
reviewed | +2.583 |
atal | +2.308 |
eson | +2.268 |
ento | +2.235 |
Neg logit contributions | |
viol | -2.807 |
NPR | -2.647 |
IRO | -2.587 |
containment | -2.545 |
pees | -2.451 |
Token | a |
Token ID | 257 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
ommel | +1.018 |
reviewed | +0.951 |
atal | +0.853 |
eson | +0.841 |
ento | +0.836 |
Neg logit contributions | |
viol | -0.931 |
NPR | -0.870 |
IRO | -0.845 |
containment | -0.843 |
Carnegie | -0.808 |
Token | new |
Token ID | 649 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
ommel | +1.012 |
reviewed | +0.948 |
atal | +0.855 |
eson | +0.844 |
ento | +0.838 |
Neg logit contributions | |
viol | -0.848 |
NPR | -0.791 |
IRO | -0.767 |
containment | -0.764 |
Carnegie | -0.731 |
Token | , |
Token ID | 11 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
ommel | +0.246 |
reviewed | +0.230 |
atal | +0.206 |
eson | +0.203 |
ento | +0.203 |
Neg logit contributions | |
viol | -0.218 |
NPR | -0.203 |
containment | -0.198 |
IRO | -0.197 |
Carnegie | -0.189 |
Token | 55 |
Token ID | 2816 |
Feature activation | +6.42e+00 |
Pos logit contributions | |
ommel | +2.456 |
reviewed | +2.341 |
atal | +1.883 |
Thu | +1.877 |
MAC | +1.851 |
Neg logit contributions | |
viol | -4.524 |
containment | -4.241 |
NPR | -4.181 |
IRO | -4.107 |
Carnegie | -4.058 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +5.91e+00 |
Pos logit contributions | |
ommel | +2.630 |
reviewed | +2.348 |
atal | +1.855 |
Thu | +1.682 |
eson | +1.613 |
Neg logit contributions | |
viol | -7.046 |
NPR | -6.762 |
IRO | -6.708 |
containment | -6.639 |
pees | -6.484 |
Token |
|
Token ID | 198 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
ommel | +3.718 |
reviewed | +3.511 |
atal | +2.974 |
eson | +2.931 |
Thu | +2.872 |
Neg logit contributions | |
viol | -5.254 |
containment | -4.861 |
NPR | -4.859 |
IRO | -4.822 |
pees | -4.677 |
Token |
|
Token ID | 198 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
ommel | +3.056 |
reviewed | +2.886 |
eson | +2.463 |
atal | +2.456 |
Thu | +2.408 |
Neg logit contributions | |
viol | -3.949 |
NPR | -3.650 |
containment | -3.646 |
IRO | -3.606 |
pees | -3.480 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +6.36e+00 |
Pos logit contributions | |
ommel | +2.927 |
reviewed | +2.761 |
atal | +2.266 |
eson | +2.203 |
ento | +2.175 |
Neg logit contributions | |
viol | -5.085 |
containment | -4.731 |
NPR | -4.701 |
IRO | -4.669 |
Carnegie | -4.571 |
Token | Share |
Token ID | 8734 |
Feature activation | +9.75e+00 |
Pos logit contributions | |
ommel | +3.826 |
reviewed | +3.565 |
atal | +3.074 |
Thu | +2.940 |
ento | +2.901 |
Neg logit contributions | |
viol | -5.819 |
IRO | -5.443 |
NPR | -5.421 |
containment | -5.354 |
Sov | -5.207 |
Token | this |
Token ID | 428 |
Feature activation | +4.97e+00 |
Pos logit contributions | |
ommel | +5.099 |
reviewed | +4.489 |
atal | +3.827 |
Thu | +3.764 |
eson | +3.724 |
Neg logit contributions | |
viol | -9.289 |
NPR | -8.850 |
IRO | -8.685 |
containment | -8.565 |
pees | -8.481 |
Token | comment |
Token ID | 2912 |
Feature activation | +6.88e+00 |
Pos logit contributions | |
ommel | +3.039 |
reviewed | +2.789 |
atal | +2.352 |
eson | +2.315 |
Thu | +2.250 |
Neg logit contributions | |
viol | -4.504 |
NPR | -4.315 |
IRO | -4.251 |
containment | -4.141 |
pees | -4.103 |
Token | on |
Token ID | 319 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
ommel | +4.015 |
reviewed | +3.751 |
atal | +3.082 |
eson | +3.070 |
Thu | +3.011 |
Neg logit contributions | |
viol | -6.386 |
NPR | -5.975 |
containment | -5.973 |
IRO | -5.937 |
pees | -5.740 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +5.99e+00 |
Pos logit contributions | |
ommel | +1.057 |
reviewed | +0.982 |
atal | +0.873 |
eson | +0.861 |
ento | +0.856 |
Neg logit contributions | |
viol | -1.070 |
NPR | -1.007 |
IRO | -0.982 |
containment | -0.980 |
Carnegie | -0.933 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
ommel | +3.703 |
reviewed | +3.368 |
atal | +2.896 |
eson | +2.847 |
ento | +2.810 |
Neg logit contributions | |
viol | -5.438 |
NPR | -5.155 |
IRO | -5.107 |
containment | -4.973 |
pees | -4.863 |
Token | among |
Token ID | 1871 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
ommel | +0.435 |
reviewed | +0.402 |
atal | +0.349 |
eson | +0.344 |
ento | +0.341 |
Neg logit contributions | |
viol | -0.569 |
NPR | -0.536 |
containment | -0.524 |
IRO | -0.521 |
Carnegie | -0.505 |
Token | all |
Token ID | 477 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
ommel | +2.012 |
reviewed | +1.866 |
atal | +1.664 |
eson | +1.641 |
ento | +1.631 |
Neg logit contributions | |
viol | -1.938 |
NPR | -1.846 |
IRO | -1.795 |
containment | -1.766 |
Carnegie | -1.701 |
Token | edge |
Token ID | 5743 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
ommel | +2.617 |
reviewed | +2.416 |
atal | +2.141 |
eson | +2.110 |
ento | +2.089 |
Neg logit contributions | |
viol | -2.896 |
NPR | -2.775 |
IRO | -2.703 |
containment | -2.652 |
Carnegie | -2.575 |
Token | r |
Token ID | 374 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ommel | +0.223 |
reviewed | +0.212 |
atal | +0.194 |
eson | +0.193 |
ento | +0.192 |
Neg logit contributions | |
viol | -0.114 |
NPR | -0.104 |
IRO | -0.100 |
containment | -0.100 |
Carnegie | -0.094 |
Token | us |
Token ID | 385 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
ommel | +1.875 |
reviewed | +1.770 |
atal | +1.622 |
eson | +1.609 |
ento | +1.596 |
Neg logit contributions | |
viol | -0.993 |
NPR | -0.899 |
containment | -0.867 |
IRO | -0.864 |
Carnegie | -0.808 |
Token | hers |
Token ID | 7084 |
Feature activation | +9.71e+00 |
Pos logit contributions | |
ommel | +1.594 |
reviewed | +1.428 |
eson | +1.135 |
atal | +1.132 |
ento | +1.086 |
Neg logit contributions | |
viol | -3.639 |
NPR | -3.473 |
containment | -3.431 |
IRO | -3.418 |
Carnegie | -3.294 |
Token | *) |
Token ID | 28104 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
ommel | +5.024 |
reviewed | +4.530 |
atal | +3.827 |
ento | +3.750 |
eson | +3.708 |
Neg logit contributions | |
viol | -9.077 |
NPR | -8.763 |
IRO | -8.715 |
containment | -8.627 |
Sov | -8.340 |
Token |
|
Token ID | 198 |
Feature activation | +4.41e+00 |
Pos logit contributions | |
ommel | +3.673 |
reviewed | +3.400 |
atal | +2.972 |
ento | +2.921 |
eson | +2.915 |
Neg logit contributions | |
viol | -4.497 |
NPR | -4.257 |
IRO | -4.143 |
containment | -4.131 |
Carnegie | -3.962 |
Token |
|
Token ID | 198 |
Feature activation | +5.34e+00 |
Pos logit contributions | |
ommel | +2.917 |
reviewed | +2.777 |
atal | +2.354 |
eson | +2.343 |
Thu | +2.296 |
Neg logit contributions | |
viol | -3.834 |
containment | -3.548 |
NPR | -3.535 |
IRO | -3.489 |
pees | -3.381 |
Token | Total |
Token ID | 14957 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
ommel | +3.828 |
reviewed | +3.604 |
atal | +3.153 |
MAC | +3.087 |
Thu | +3.068 |
Neg logit contributions | |
viol | -4.353 |
containment | -3.999 |
NPR | -3.979 |
IRO | -3.886 |
Carnegie | -3.843 |
Token | guaranteed |
Token ID | 11462 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ommel | +1.576 |
reviewed | +1.443 |
atal | +1.246 |
eson | +1.225 |
ento | +1.214 |
Neg logit contributions | |
viol | -2.249 |
NPR | -2.134 |
containment | -2.082 |
IRO | -2.079 |
Carnegie | -2.009 |
Token | , |
Token ID | 11 |
Feature activation | +8.61e-01 |
Pos logit contributions | |
viol | +1.141 |
NPR | +1.061 |
containment | +1.039 |
IRO | +1.033 |
Carnegie | +0.992 |
Neg logit contributions | |
ommel | -1.144 |
reviewed | -1.068 |
atal | -0.948 |
eson | -0.938 |
ento | -0.930 |
Token | le |
Token ID | 443 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
ommel | +0.699 |
reviewed | +0.653 |
atal | +0.587 |
eson | +0.579 |
ento | +0.578 |
Neg logit contributions | |
viol | -0.621 |
NPR | -0.586 |
IRO | -0.569 |
containment | -0.568 |
Carnegie | -0.544 |
Token | contr |
Token ID | 3445 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ommel | +2.284 |
reviewed | +2.095 |
atal | +1.901 |
ento | +1.871 |
eson | +1.871 |
Neg logit contributions | |
viol | -2.145 |
NPR | -2.082 |
IRO | -2.001 |
containment | -1.999 |
Carnegie | -1.915 |
Token | at |
Token ID | 265 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
viol | +1.022 |
NPR | +0.958 |
containment | +0.939 |
IRO | +0.934 |
Carnegie | +0.901 |
Neg logit contributions | |
ommel | -0.885 |
reviewed | -0.823 |
atal | -0.718 |
eson | -0.712 |
ento | -0.708 |
Token | qu |
Token ID | 627 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
ommel | +0.253 |
reviewed | +0.233 |
atal | +0.208 |
ento | +0.205 |
eson | +0.204 |
Neg logit contributions | |
viol | -0.266 |
NPR | -0.251 |
containment | -0.244 |
IRO | -0.244 |
Carnegie | -0.235 |
Token | ' |
Token ID | 6 |
Feature activation | +9.63e+00 |
Pos logit contributions | |
ommel | +1.396 |
reviewed | +1.278 |
atal | +1.132 |
eson | +1.109 |
ento | +1.107 |
Neg logit contributions | |
viol | -1.744 |
NPR | -1.652 |
containment | -1.621 |
IRO | -1.613 |
Carnegie | -1.557 |
Token | ils |
Token ID | 4487 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
ommel | +6.140 |
reviewed | +5.610 |
atal | +5.063 |
ento | +4.926 |
eson | +4.868 |
Neg logit contributions | |
viol | -8.282 |
containment | -7.942 |
NPR | -7.774 |
IRO | -7.613 |
Carnegie | -7.547 |
Token | ont |
Token ID | 39585 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
ommel | +3.735 |
reviewed | +3.457 |
ento | +3.127 |
atal | +3.122 |
eson | +3.088 |
Neg logit contributions | |
viol | -3.318 |
NPR | -3.232 |
IRO | -3.135 |
containment | -3.076 |
Carnegie | -3.000 |
Token | sign |
Token ID | 1051 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
viol | +1.769 |
NPR | +1.633 |
containment | +1.601 |
IRO | +1.586 |
Carnegie | +1.523 |
Neg logit contributions | |
ommel | -1.883 |
reviewed | -1.760 |
atal | -1.568 |
eson | -1.551 |
ento | -1.540 |
Token | é |
Token ID | 2634 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
ommel | +1.915 |
reviewed | +1.736 |
atal | +1.536 |
eson | +1.504 |
ento | +1.497 |
Neg logit contributions | |
viol | -2.585 |
NPR | -2.458 |
containment | -2.419 |
IRO | -2.409 |
Carnegie | -2.328 |
Token | , |
Token ID | 11 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ommel | +1.271 |
reviewed | +1.182 |
atal | +1.057 |
ento | +1.039 |
eson | +1.035 |
Neg logit contributions | |
viol | -1.289 |
NPR | -1.221 |
containment | -1.186 |
IRO | -1.184 |
Carnegie | -1.142 |
Token | 19 |
Token ID | 1129 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
ommel | +1.496 |
reviewed | +1.396 |
atal | +1.151 |
eson | +1.119 |
MAC | +1.112 |
Neg logit contributions | |
viol | -2.617 |
containment | -2.449 |
NPR | -2.447 |
IRO | -2.401 |
Carnegie | -2.354 |
Token | Recommend |
Token ID | 30361 |
Feature activation | +4.75e+00 |
Pos logit contributions | |
ommel | +2.262 |
reviewed | +2.070 |
atal | +1.636 |
Thu | +1.506 |
ento | +1.491 |
Neg logit contributions | |
viol | -5.323 |
NPR | -5.103 |
IRO | -5.039 |
containment | -5.035 |
pees | -4.856 |
Token |
|
Token ID | 198 |
Feature activation | +4.41e+00 |
Pos logit contributions | |
ommel | +3.110 |
reviewed | +2.928 |
atal | +2.504 |
eson | +2.469 |
ento | +2.420 |
Neg logit contributions | |
viol | -4.139 |
NPR | -3.843 |
containment | -3.824 |
IRO | -3.803 |
pees | -3.665 |
Token |
|
Token ID | 198 |
Feature activation | +5.25e+00 |
Pos logit contributions | |
ommel | +2.965 |
reviewed | +2.799 |
eson | +2.392 |
atal | +2.390 |
ento | +2.338 |
Neg logit contributions | |
viol | -3.789 |
NPR | -3.506 |
containment | -3.495 |
IRO | -3.461 |
pees | -3.331 |
Token | Recommend |
Token ID | 41248 |
Feature activation | +6.14e+00 |
Pos logit contributions | |
ommel | +2.912 |
reviewed | +2.769 |
atal | +2.256 |
eson | +2.179 |
Thu | +2.171 |
Neg logit contributions | |
viol | -5.173 |
containment | -4.820 |
NPR | -4.764 |
IRO | -4.732 |
Carnegie | -4.639 |
Token | Share |
Token ID | 8734 |
Feature activation | +9.62e+00 |
Pos logit contributions | |
ommel | +3.811 |
reviewed | +3.547 |
atal | +3.066 |
eson | +2.925 |
Thu | +2.920 |
Neg logit contributions | |
viol | -5.520 |
NPR | -5.167 |
IRO | -5.154 |
containment | -5.071 |
pees | -4.899 |
Token | this |
Token ID | 428 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
ommel | +4.994 |
reviewed | +4.498 |
atal | +3.727 |
Thu | +3.718 |
eson | +3.685 |
Neg logit contributions | |
viol | -9.197 |
NPR | -8.728 |
IRO | -8.606 |
containment | -8.520 |
pees | -8.455 |
Token | comment |
Token ID | 2912 |
Feature activation | +7.11e+00 |
Pos logit contributions | |
ommel | +2.960 |
reviewed | +2.721 |
atal | +2.292 |
eson | +2.267 |
ento | +2.210 |
Neg logit contributions | |
viol | -4.287 |
NPR | -4.108 |
IRO | -4.045 |
containment | -3.940 |
pees | -3.903 |
Token | on |
Token ID | 319 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ommel | +4.107 |
reviewed | +3.849 |
atal | +3.140 |
eson | +3.135 |
Thu | +3.086 |
Neg logit contributions | |
viol | -6.615 |
containment | -6.189 |
NPR | -6.183 |
IRO | -6.146 |
pees | -5.962 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +6.73e+00 |
Pos logit contributions | |
ommel | +1.318 |
reviewed | +1.224 |
atal | +1.085 |
eson | +1.072 |
ento | +1.063 |
Neg logit contributions | |
viol | -1.354 |
NPR | -1.274 |
IRO | -1.244 |
containment | -1.241 |
Carnegie | -1.178 |
Token | Share |
Token ID | 8734 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
ommel | +4.082 |
reviewed | +3.713 |
atal | +3.182 |
eson | +3.128 |
ento | +3.084 |
Neg logit contributions | |
viol | -6.150 |
NPR | -5.847 |
IRO | -5.759 |
containment | -5.643 |
pees | -5.503 |
Token | Share |
Token ID | 8734 |
Feature activation | +9.12e+00 |
Pos logit contributions | |
ommel | +3.950 |
reviewed | +3.664 |
atal | +3.166 |
Thu | +3.012 |
ento | +3.001 |
Neg logit contributions | |
viol | -5.789 |
IRO | -5.425 |
NPR | -5.421 |
containment | -5.326 |
pees | -5.142 |
Token | this |
Token ID | 428 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
ommel | +4.833 |
reviewed | +4.355 |
atal | +3.632 |
Thu | +3.614 |
eson | +3.597 |
Neg logit contributions | |
viol | -8.668 |
NPR | -8.213 |
IRO | -8.117 |
containment | -8.031 |
pees | -7.947 |
Token | comment |
Token ID | 2912 |
Feature activation | +7.03e+00 |
Pos logit contributions | |
ommel | +2.411 |
reviewed | +2.216 |
atal | +1.893 |
eson | +1.869 |
ento | +1.832 |
Neg logit contributions | |
viol | -3.292 |
NPR | -3.142 |
IRO | -3.091 |
containment | -3.027 |
pees | -2.964 |
Token | on |
Token ID | 319 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ommel | +4.085 |
reviewed | +3.822 |
atal | +3.129 |
eson | +3.124 |
Thu | +3.059 |
Neg logit contributions | |
viol | -6.528 |
NPR | -6.103 |
containment | -6.103 |
IRO | -6.066 |
pees | -5.873 |
Token | Facebook |
Token ID | 3203 |
Feature activation | +6.31e+00 |
Pos logit contributions | |
ommel | +1.335 |
reviewed | +1.240 |
atal | +1.100 |
eson | +1.086 |
ento | +1.076 |
Neg logit contributions | |
viol | -1.374 |
NPR | -1.294 |
IRO | -1.263 |
containment | -1.260 |
Carnegie | -1.196 |
Token | Share |
Token ID | 8734 |
Feature activation | +9.59e+00 |
Pos logit contributions | |
ommel | +3.880 |
reviewed | +3.530 |
atal | +3.036 |
eson | +2.986 |
ento | +2.937 |
Neg logit contributions | |
viol | -5.725 |
NPR | -5.441 |
IRO | -5.365 |
containment | -5.257 |
pees | -5.125 |
Token | this |
Token ID | 428 |
Feature activation | +5.96e+00 |
Pos logit contributions | |
ommel | +5.096 |
reviewed | +4.675 |
eson | +3.805 |
Thu | +3.769 |
ento | +3.729 |
Neg logit contributions | |
viol | -9.142 |
NPR | -8.606 |
IRO | -8.543 |
containment | -8.494 |
pees | -8.369 |
Token | comment |
Token ID | 2912 |
Feature activation | +8.39e+00 |
Pos logit contributions | |
ommel | +3.658 |
reviewed | +3.350 |
atal | +2.800 |
eson | +2.781 |
ento | +2.718 |
Neg logit contributions | |
viol | -5.349 |
NPR | -5.151 |
IRO | -5.079 |
containment | -4.942 |
pees | -4.900 |
Token | on |
Token ID | 319 |
Feature activation | +7.76e+00 |
Pos logit contributions | |
ommel | +4.663 |
reviewed | +4.372 |
eson | +3.560 |
atal | +3.484 |
Thu | +3.433 |
Neg logit contributions | |
viol | -7.867 |
containment | -7.389 |
NPR | -7.364 |
IRO | -7.317 |
pees | -7.115 |
Token | Twitter |
Token ID | 3009 |
Feature activation | +6.91e-01 |
Pos logit contributions | |
ommel | +4.871 |
reviewed | +4.627 |
Thu | +3.937 |
Thu | +3.897 |
atal | +3.885 |
Neg logit contributions | |
viol | -6.573 |
containment | -6.176 |
NPR | -6.174 |
IRO | -6.061 |
pees | -6.030 |
Token |
|
Token ID | 198 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
ommel | +0.507 |
reviewed | +0.469 |
atal | +0.415 |
eson | +0.409 |
ento | +0.407 |
Neg logit contributions | |
viol | -0.556 |
NPR | -0.523 |
containment | -0.512 |
IRO | -0.510 |
Carnegie | -0.490 |
Token |
|
Token ID | 198 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ommel | +0.882 |
reviewed | +0.822 |
atal | +0.723 |
eson | +0.714 |
ento | +0.709 |
Neg logit contributions | |
viol | -0.977 |
NPR | -0.914 |
containment | -0.897 |
IRO | -0.892 |
Carnegie | -0.857 |
Token | City |
Token ID | 14941 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ommel | +0.866 |
reviewed | +0.813 |
atal | +0.715 |
eson | +0.707 |
ento | +0.701 |
Neg logit contributions | |
viol | -0.901 |
NPR | -0.836 |
containment | -0.828 |
IRO | -0.817 |
Carnegie | -0.790 |
Token | Council |
Token ID | 4281 |
Feature activation | -5.14e+00 |
Pos logit contributions | |
viol | +0.913 |
NPR | +0.828 |
containment | +0.799 |
IRO | +0.799 |
Carnegie | +0.759 |
Neg logit contributions | |
ommel | -1.561 |
reviewed | -1.473 |
atal | -1.353 |
eson | -1.335 |
ento | -1.335 |
Token | — |
Token ID | 960 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
viol | +3.488 |
NPR | +3.201 |
containment | +3.104 |
IRO | +3.080 |
Carnegie | +2.989 |
Neg logit contributions | |
ommel | -4.385 |
reviewed | -4.089 |
atal | -3.720 |
ento | -3.684 |
eson | -3.684 |
Token | Point |
Token ID | 12727 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
viol | +0.490 |
NPR | +0.463 |
containment | +0.451 |
IRO | +0.451 |
Carnegie | +0.435 |
Neg logit contributions | |
ommel | -0.400 |
reviewed | -0.368 |
atal | -0.323 |
eson | -0.318 |
ento | -0.317 |
Token | of |
Token ID | 286 |
Feature activation | +9.58e+00 |
Pos logit contributions | |
ommel | +2.426 |
reviewed | +2.257 |
atal | +1.974 |
ento | +1.953 |
eson | +1.943 |
Neg logit contributions | |
viol | -2.838 |
NPR | -2.685 |
IRO | -2.607 |
containment | -2.590 |
Carnegie | -2.506 |
Token | Sale |
Token ID | 16467 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
ommel | +4.205 |
reviewed | +3.700 |
atal | +3.078 |
ento | +3.051 |
eson | +2.935 |
Neg logit contributions | |
viol | -9.676 |
NPR | -9.549 |
IRO | -9.409 |
containment | -8.963 |
Sov | -8.941 |
Token | Plan |
Token ID | 5224 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
ommel | +1.618 |
reviewed | +1.486 |
atal | +1.312 |
eson | +1.294 |
ento | +1.285 |
Neg logit contributions | |
viol | -1.936 |
NPR | -1.837 |
containment | -1.797 |
IRO | -1.792 |
Carnegie | -1.707 |
Token |
|
Token ID | 198 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
ommel | +1.567 |
reviewed | +1.455 |
atal | +1.286 |
eson | +1.261 |
ento | +1.259 |
Neg logit contributions | |
viol | -1.777 |
NPR | -1.679 |
IRO | -1.639 |
containment | -1.639 |
Carnegie | -1.566 |
Token |
|
Token ID | 198 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
ommel | +2.371 |
reviewed | +2.249 |
atal | +1.913 |
eson | +1.911 |
Thu | +1.870 |
Neg logit contributions | |
viol | -3.024 |
containment | -2.802 |
NPR | -2.796 |
IRO | -2.750 |
pees | -2.662 |
Token | Last |
Token ID | 5956 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
ommel | +2.271 |
reviewed | +2.169 |
atal | +1.876 |
eson | +1.857 |
Thu | +1.850 |
Neg logit contributions | |
viol | -2.453 |
containment | -2.279 |
NPR | -2.212 |
IRO | -2.185 |
Carnegie | -2.144 |
Token | - |
Token ID | 12 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
ommel | +2.733 |
reviewed | +2.561 |
atal | +2.258 |
eson | +2.224 |
ento | +2.169 |
Neg logit contributions | |
viol | -3.418 |
NPR | -3.213 |
IRO | -3.184 |
containment | -3.175 |
Carnegie | -3.042 |
Token | m |
Token ID | 76 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
ommel | +3.719 |
reviewed | +3.518 |
atal | +3.182 |
eson | +3.160 |
ento | +3.065 |
Neg logit contributions | |
viol | -3.983 |
containment | -3.703 |
NPR | -3.662 |
IRO | -3.652 |
Carnegie | -3.510 |
Token | eth |
Token ID | 2788 |
Feature activation | +6.06e+00 |
Pos logit contributions | |
ommel | +2.883 |
reviewed | +2.621 |
atal | +2.333 |
eson | +2.293 |
ento | +2.214 |
Neg logit contributions | |
viol | -4.748 |
containment | -4.486 |
NPR | -4.482 |
IRO | -4.383 |
Carnegie | -4.263 |
Token | ox |
Token ID | 1140 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
ommel | +3.833 |
reviewed | +3.538 |
atal | +3.175 |
eson | +3.089 |
ento | +3.078 |
Neg logit contributions | |
viol | -5.420 |
containment | -5.104 |
NPR | -5.038 |
IRO | -5.001 |
Carnegie | -4.827 |
Token | ym |
Token ID | 4948 |
Feature activation | +5.48e+00 |
Pos logit contributions | |
ommel | +3.373 |
reviewed | +3.174 |
atal | +2.852 |
eson | +2.809 |
ento | +2.767 |
Neg logit contributions | |
viol | -3.327 |
NPR | -3.089 |
containment | -3.058 |
IRO | -3.011 |
Carnegie | -2.875 |
Token | ethyl |
Token ID | 21610 |
Feature activation | +9.56e+00 |
Pos logit contributions | |
ommel | +3.184 |
reviewed | +2.934 |
atal | +2.601 |
eson | +2.520 |
ento | +2.499 |
Neg logit contributions | |
viol | -5.197 |
NPR | -4.958 |
containment | -4.915 |
IRO | -4.846 |
Carnegie | -4.665 |
Token | phen |
Token ID | 31024 |
Feature activation | +5.13e+00 |
Pos logit contributions | |
ommel | +5.679 |
reviewed | +5.334 |
atal | +4.795 |
eson | +4.635 |
enic | +4.494 |
Neg logit contributions | |
viol | -8.339 |
IRO | -7.907 |
containment | -7.892 |
NPR | -7.887 |
Carnegie | -7.523 |
Token | yl |
Token ID | 2645 |
Feature activation | +7.36e+00 |
Pos logit contributions | |
ommel | +4.267 |
reviewed | +3.956 |
atal | +3.665 |
eson | +3.613 |
ento | +3.528 |
Neg logit contributions | |
viol | -3.607 |
NPR | -3.349 |
containment | -3.320 |
IRO | -3.295 |
Carnegie | -3.095 |
Token | )- |
Token ID | 13219 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
ommel | +3.861 |
reviewed | +3.655 |
atal | +3.109 |
eson | +3.005 |
ento | +2.875 |
Neg logit contributions | |
viol | -7.138 |
NPR | -6.787 |
containment | -6.757 |
IRO | -6.746 |
Carnegie | -6.494 |
Token | 2 |
Token ID | 17 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
ommel | +1.674 |
reviewed | +1.551 |
atal | +1.341 |
eson | +1.316 |
ento | +1.285 |
Neg logit contributions | |
viol | -2.406 |
NPR | -2.255 |
containment | -2.230 |
IRO | -2.216 |
Carnegie | -2.149 |
Token | - |
Token ID | 12 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
ommel | +2.293 |
reviewed | +2.131 |
atal | +1.873 |
eson | +1.852 |
ento | +1.821 |
Neg logit contributions | |
viol | -2.775 |
NPR | -2.607 |
containment | -2.574 |
IRO | -2.566 |
Carnegie | -2.466 |
Token | of |
Token ID | 286 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ommel | +0.257 |
reviewed | +0.238 |
atal | +0.209 |
eson | +0.206 |
ento | +0.205 |
Neg logit contributions | |
viol | -0.297 |
NPR | -0.280 |
containment | -0.273 |
IRO | -0.272 |
Carnegie | -0.262 |
Token | such |
Token ID | 884 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
viol | +0.912 |
NPR | +0.844 |
containment | +0.824 |
IRO | +0.813 |
Carnegie | +0.785 |
Neg logit contributions | |
ommel | -1.005 |
reviewed | -0.932 |
atal | -0.837 |
eson | -0.828 |
ento | -0.822 |
Token | other |
Token ID | 584 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
viol | +1.767 |
NPR | +1.642 |
containment | +1.609 |
IRO | +1.595 |
Carnegie | +1.538 |
Neg logit contributions | |
ommel | -1.734 |
reviewed | -1.612 |
atal | -1.428 |
eson | -1.409 |
ento | -1.406 |
Token | State |
Token ID | 1812 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
ommel | +0.802 |
reviewed | +0.745 |
atal | +0.661 |
eson | +0.652 |
ento | +0.644 |
Neg logit contributions | |
viol | -0.903 |
NPR | -0.861 |
IRO | -0.841 |
containment | -0.832 |
Carnegie | -0.804 |
Token | , |
Token ID | 11 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
ommel | +1.571 |
reviewed | +1.451 |
atal | +1.270 |
eson | +1.250 |
ento | +1.240 |
Neg logit contributions | |
viol | -1.983 |
NPR | -1.885 |
IRO | -1.839 |
containment | -1.838 |
Carnegie | -1.776 |
Token | territory |
Token ID | 7674 |
Feature activation | +9.49e+00 |
Pos logit contributions | |
ommel | +2.562 |
reviewed | +2.407 |
atal | +2.213 |
eson | +2.177 |
ederal | +2.160 |
Neg logit contributions | |
viol | -1.877 |
NPR | -1.761 |
IRO | -1.714 |
containment | -1.670 |
Carnegie | -1.623 |
Token | , |
Token ID | 11 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
ommel | +5.106 |
reviewed | +4.623 |
eson | +3.940 |
atal | +3.913 |
ento | +3.825 |
Neg logit contributions | |
viol | -8.697 |
NPR | -8.472 |
IRO | -8.407 |
containment | -8.111 |
pees | -8.029 |
Token | possession |
Token ID | 7797 |
Feature activation | +6.46e+00 |
Pos logit contributions | |
ommel | +2.543 |
reviewed | +2.327 |
atal | +2.086 |
eson | +2.071 |
ento | +2.046 |
Neg logit contributions | |
viol | -2.703 |
NPR | -2.581 |
IRO | -2.518 |
containment | -2.448 |
Carnegie | -2.407 |
Token | , |
Token ID | 11 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ommel | +3.989 |
reviewed | +3.645 |
atal | +3.129 |
ento | +3.093 |
eson | +3.088 |
Neg logit contributions | |
viol | -5.635 |
NPR | -5.460 |
IRO | -5.367 |
containment | -5.292 |
Carnegie | -5.160 |
Token | or |
Token ID | 393 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
ommel | +0.855 |
reviewed | +0.790 |
atal | +0.699 |
eson | +0.688 |
ento | +0.686 |
Neg logit contributions | |
viol | -0.972 |
NPR | -0.916 |
IRO | -0.894 |
containment | -0.892 |
Carnegie | -0.860 |
Token | tribe |
Token ID | 14893 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ommel | +2.855 |
reviewed | +2.508 |
atal | +2.159 |
ento | +2.099 |
eson | +2.098 |
Neg logit contributions | |
viol | -4.901 |
NPR | -4.790 |
IRO | -4.727 |
pees | -4.574 |
Carnegie | -4.498 |
Token | farewell |
Token ID | 33458 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
viol | +1.526 |
NPR | +1.430 |
containment | +1.402 |
IRO | +1.396 |
Carnegie | +1.355 |
Neg logit contributions | |
ommel | -1.273 |
reviewed | -1.178 |
atal | -1.034 |
eson | -1.015 |
ento | -1.014 |
Token | ... |
Token ID | 986 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
ommel | +0.136 |
reviewed | +0.126 |
atal | +0.112 |
eson | +0.111 |
ento | +0.110 |
Neg logit contributions | |
viol | -0.140 |
NPR | -0.131 |
containment | -0.128 |
IRO | -0.127 |
Carnegie | -0.122 |
Token |
|
Token ID | 198 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
ommel | +1.759 |
reviewed | +1.643 |
atal | +1.456 |
eson | +1.447 |
ento | +1.428 |
Neg logit contributions | |
viol | -1.849 |
NPR | -1.739 |
containment | -1.698 |
IRO | -1.690 |
Carnegie | -1.615 |
Token |
|
Token ID | 198 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ommel | +0.557 |
reviewed | +0.517 |
atal | +0.458 |
eson | +0.451 |
ento | +0.449 |
Neg logit contributions | |
viol | -0.600 |
NPR | -0.563 |
containment | -0.549 |
IRO | -0.548 |
Carnegie | -0.527 |
Token | Follow |
Token ID | 7155 |
Feature activation | +6.48e+00 |
Pos logit contributions | |
ommel | +1.332 |
reviewed | +1.252 |
atal | +1.105 |
eson | +1.096 |
ento | +1.082 |
Neg logit contributions | |
viol | -1.344 |
NPR | -1.237 |
containment | -1.233 |
IRO | -1.205 |
Carnegie | -1.171 |
Token | Sir |
Token ID | 7361 |
Feature activation | +9.42e+00 |
Pos logit contributions | |
ommel | +4.836 |
reviewed | +4.391 |
atal | +3.980 |
eson | +3.936 |
ento | +3.816 |
Neg logit contributions | |
viol | -5.000 |
NPR | -4.736 |
containment | -4.594 |
IRO | -4.567 |
pees | -4.360 |
Token | Alex |
Token ID | 4422 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
ommel | +6.384 |
reviewed | +5.679 |
atal | +5.173 |
eson | +5.013 |
ento | +4.878 |
Neg logit contributions | |
viol | -7.653 |
NPR | -7.366 |
containment | -7.229 |
IRO | -7.032 |
pees | -6.776 |
Token | Ferguson |
Token ID | 15031 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
ommel | +1.332 |
reviewed | +1.233 |
atal | +1.095 |
eson | +1.085 |
ento | +1.076 |
Neg logit contributions | |
viol | -1.391 |
NPR | -1.307 |
containment | -1.275 |
IRO | -1.271 |
Carnegie | -1.214 |
Token | 's |
Token ID | 338 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ommel | +1.689 |
reviewed | +1.545 |
atal | +1.375 |
eson | +1.350 |
ento | +1.335 |
Neg logit contributions | |
viol | -1.999 |
NPR | -1.882 |
containment | -1.838 |
IRO | -1.832 |
Carnegie | -1.758 |
Token | final |
Token ID | 2457 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
ommel | +1.306 |
reviewed | +1.207 |
atal | +1.081 |
eson | +1.065 |
ento | +1.052 |
Neg logit contributions | |
viol | -1.345 |
NPR | -1.264 |
containment | -1.228 |
IRO | -1.227 |
Carnegie | -1.173 |
Token | game |
Token ID | 983 |
Feature activation | +5.15e-01 |
Pos logit contributions | |
viol | +1.469 |
NPR | +1.376 |
containment | +1.346 |
IRO | +1.335 |
Carnegie | +1.295 |
Neg logit contributions | |
ommel | -1.453 |
reviewed | -1.366 |
atal | -1.194 |
ento | -1.188 |
eson | -1.178 |
Token | based |
Token ID | 3106 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
ommel | +1.464 |
reviewed | +1.369 |
atal | +1.195 |
eson | +1.157 |
ento | +1.157 |
Neg logit contributions | |
viol | -1.780 |
NPR | -1.662 |
containment | -1.646 |
IRO | -1.630 |
Carnegie | -1.571 |
Token | Anat |
Token ID | 28880 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
ommel | +1.895 |
reviewed | +1.772 |
atal | +1.594 |
eson | +1.567 |
ento | +1.549 |
Neg logit contributions | |
viol | -1.790 |
NPR | -1.711 |
IRO | -1.662 |
containment | -1.657 |
pees | -1.582 |
Token | ol |
Token ID | 349 |
Feature activation | +7.03e-02 |
Pos logit contributions | |
viol | +2.822 |
NPR | +2.611 |
IRO | +2.575 |
containment | +2.534 |
Carnegie | +2.492 |
Neg logit contributions | |
ommel | -2.531 |
reviewed | -2.352 |
eson | -2.036 |
atal | -2.036 |
ento | -2.030 |
Token | iy |
Token ID | 7745 |
Feature activation | +8.46e-02 |
Pos logit contributions | |
ommel | +0.052 |
reviewed | +0.048 |
atal | +0.042 |
eson | +0.042 |
ento | +0.042 |
Neg logit contributions | |
viol | -0.057 |
NPR | -0.053 |
containment | -0.052 |
IRO | -0.052 |
Carnegie | -0.050 |
Token | D |
Token ID | 360 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
ommel | +0.063 |
reviewed | +0.058 |
atal | +0.051 |
eson | +0.051 |
ento | +0.051 |
Neg logit contributions | |
viol | -0.068 |
NPR | -0.064 |
containment | -0.062 |
IRO | -0.062 |
Carnegie | -0.060 |
Token | erk |
Token ID | 9587 |
Feature activation | +9.27e+00 |
Pos logit contributions | |
ommel | +0.257 |
reviewed | +0.237 |
atal | +0.211 |
eson | +0.207 |
ento | +0.206 |
Neg logit contributions | |
viol | -0.290 |
NPR | -0.271 |
containment | -0.267 |
IRO | -0.265 |
Carnegie | -0.255 |
Token | ach |
Token ID | 620 |
Feature activation | +4.22e-01 |
Pos logit contributions | |
ommel | +2.483 |
reviewed | +1.857 |
atal | +1.451 |
eson | +1.267 |
ento | +1.088 |
Neg logit contributions | |
viol | -11.767 |
NPR | -11.291 |
containment | -11.138 |
IRO | -11.110 |
Civ | -10.671 |
Token | , |
Token ID | 11 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
ommel | +0.328 |
reviewed | +0.306 |
atal | +0.273 |
eson | +0.269 |
ento | +0.268 |
Neg logit contributions | |
viol | -0.322 |
NPR | -0.302 |
containment | -0.294 |
IRO | -0.293 |
Carnegie | -0.281 |
Token | the |
Token ID | 262 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
ommel | +0.212 |
reviewed | +0.198 |
atal | +0.178 |
eson | +0.175 |
ento | +0.175 |
Neg logit contributions | |
viol | -0.188 |
NPR | -0.176 |
containment | -0.171 |
IRO | -0.171 |
Carnegie | -0.163 |
Token | owner |
Token ID | 4870 |
Feature activation | -9.10e-01 |
Pos logit contributions | |
viol | +0.255 |
NPR | +0.238 |
containment | +0.231 |
IRO | +0.231 |
Carnegie | +0.221 |
Neg logit contributions | |
ommel | -0.297 |
reviewed | -0.278 |
atal | -0.250 |
eson | -0.247 |
ento | -0.246 |
Token | of |
Token ID | 286 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
viol | +0.746 |
NPR | +0.699 |
containment | +0.684 |
IRO | +0.681 |
Carnegie | +0.658 |
Neg logit contributions | |
ommel | -0.658 |
reviewed | -0.610 |
atal | -0.539 |
eson | -0.531 |
ento | -0.528 |
Token | s |
Token ID | 82 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
ommel | +1.876 |
reviewed | +1.729 |
atal | +1.547 |
eson | +1.517 |
ento | +1.515 |
Neg logit contributions | |
viol | -2.163 |
NPR | -2.031 |
IRO | -1.986 |
containment | -1.984 |
Carnegie | -1.921 |
Token | assume |
Token ID | 7048 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
ommel | +1.316 |
reviewed | +1.222 |
atal | +1.107 |
eson | +1.094 |
ento | +1.079 |
Neg logit contributions | |
viol | -1.132 |
NPR | -1.062 |
containment | -1.030 |
IRO | -1.029 |
Carnegie | -0.982 |
Token | u |
Token ID | 334 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
viol | +2.157 |
NPR | +1.981 |
containment | +1.970 |
IRO | +1.940 |
Carnegie | +1.864 |
Neg logit contributions | |
ommel | -2.344 |
reviewed | -2.202 |
atal | -1.962 |
eson | -1.934 |
ento | -1.926 |
Token | = |
Token ID | 796 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
viol | +0.886 |
NPR | +0.829 |
containment | +0.808 |
IRO | +0.806 |
Carnegie | +0.773 |
Neg logit contributions | |
ommel | -0.911 |
reviewed | -0.849 |
atal | -0.757 |
eson | -0.747 |
ento | -0.744 |
Token | [ |
Token ID | 685 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ommel | +0.281 |
reviewed | +0.262 |
atal | +0.234 |
eson | +0.231 |
ento | +0.230 |
Neg logit contributions | |
viol | -0.271 |
NPR | -0.254 |
IRO | -0.246 |
containment | -0.246 |
Carnegie | -0.236 |
Token | CO |
Token ID | 8220 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
viol | +1.065 |
NPR | +1.008 |
IRO | +0.985 |
containment | +0.983 |
Carnegie | +0.948 |
Neg logit contributions | |
ommel | -0.828 |
reviewed | -0.761 |
atal | -0.664 |
eson | -0.655 |
ento | -0.651 |
Token | 2 |
Token ID | 362 |
Feature activation | -8.13e+00 |
Pos logit contributions | |
viol | +6.674 |
IRO | +6.202 |
NPR | +6.024 |
containment | +5.935 |
Civ | +5.754 |
Neg logit contributions | |
ommel | -9.751 |
reviewed | -9.205 |
eson | -8.476 |
ento | -8.308 |
atal | -8.078 |
Token | ] |
Token ID | 2361 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
viol | +6.083 |
containment | +5.712 |
NPR | +5.662 |
IRO | +5.650 |
Civ | +5.362 |
Neg logit contributions | |
ommel | -6.079 |
reviewed | -5.597 |
eson | -5.004 |
atal | -4.925 |
ento | -4.920 |
Token | / |
Token ID | 14 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
viol | +1.039 |
NPR | +0.973 |
containment | +0.951 |
IRO | +0.949 |
Carnegie | +0.910 |
Neg logit contributions | |
ommel | -1.005 |
reviewed | -0.933 |
atal | -0.827 |
eson | -0.818 |
ento | -0.814 |
Token | f |
Token ID | 277 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
viol | +1.560 |
NPR | +1.474 |
IRO | +1.437 |
containment | +1.433 |
Carnegie | +1.370 |
Neg logit contributions | |
ommel | -1.431 |
reviewed | -1.322 |
atal | -1.164 |
eson | -1.154 |
ento | -1.146 |
Token | absorbed |
Token ID | 19233 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
viol | +2.158 |
NPR | +2.020 |
containment | +1.992 |
IRO | +1.966 |
Carnegie | +1.886 |
Neg logit contributions | |
ommel | -2.130 |
reviewed | -1.986 |
atal | -1.745 |
eson | -1.735 |
ento | -1.718 |
Token | ) |
Token ID | 1267 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
ommel | +1.575 |
reviewed | +1.439 |
atal | +1.252 |
ento | +1.225 |
eson | +1.220 |
Neg logit contributions | |
viol | -2.204 |
NPR | -2.086 |
containment | -2.033 |
IRO | -2.029 |
Carnegie | -1.969 |
Token | / |
Token ID | 1220 |
Feature activation | +6.44e-01 |
Pos logit contributions | |
ommel | +0.250 |
reviewed | +0.233 |
atal | +0.206 |
eson | +0.203 |
ento | +0.203 |
Neg logit contributions | |
viol | -0.264 |
NPR | -0.247 |
containment | -0.241 |
IRO | -0.241 |
Carnegie | -0.232 |
Token | u |
Token ID | 84 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
ommel | +0.490 |
reviewed | +0.456 |
atal | +0.406 |
eson | +0.400 |
ento | +0.399 |
Neg logit contributions | |
viol | -0.503 |
NPR | -0.471 |
containment | -0.459 |
IRO | -0.458 |
Carnegie | -0.441 |
Token | ) |
Token ID | 8 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
viol | +3.186 |
NPR | +3.015 |
containment | +2.962 |
IRO | +2.957 |
Civ | +2.837 |
Neg logit contributions | |
ommel | -2.349 |
reviewed | -2.156 |
atal | -1.863 |
eson | -1.852 |
ento | -1.826 |
Token | [ |
Token ID | 685 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
viol | +1.215 |
NPR | +1.140 |
containment | +1.114 |
IRO | +1.111 |
Carnegie | +1.066 |
Neg logit contributions | |
ommel | -1.149 |
reviewed | -1.068 |
atal | -0.943 |
eson | -0.933 |
ento | -0.925 |
Token | CO |
Token ID | 8220 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
viol | +2.968 |
NPR | +2.833 |
IRO | +2.771 |
containment | +2.732 |
Carnegie | +2.632 |
Neg logit contributions | |
ommel | -2.717 |
reviewed | -2.470 |
atal | -2.202 |
eson | -2.190 |
ento | -2.161 |
Token | 2 |
Token ID | 362 |
Feature activation | -7.31e+00 |
Pos logit contributions | |
viol | +5.494 |
NPR | +5.066 |
IRO | +5.048 |
Civ | +4.816 |
containment | +4.764 |
Neg logit contributions | |
ommel | -10.608 |
reviewed | -9.910 |
ento | -9.250 |
eson | -9.191 |
enic | -9.041 |
Token | ] |
Token ID | 2361 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
viol | +4.842 |
NPR | +4.483 |
containment | +4.470 |
IRO | +4.437 |
Civ | +4.152 |
Neg logit contributions | |
ommel | -6.137 |
reviewed | -5.701 |
eson | -5.196 |
atal | -5.133 |
ento | -5.110 |
Token | ( |
Token ID | 357 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
viol | +1.851 |
NPR | +1.734 |
containment | +1.695 |
IRO | +1.693 |
Carnegie | +1.613 |
Neg logit contributions | |
ommel | -1.877 |
reviewed | -1.749 |
atal | -1.548 |
eson | -1.541 |
ento | -1.532 |
Token | t |
Token ID | 83 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
viol | +1.595 |
NPR | +1.505 |
IRO | +1.471 |
containment | +1.468 |
Carnegie | +1.407 |
Neg logit contributions | |
ommel | -1.366 |
reviewed | -1.258 |
atal | -1.107 |
eson | -1.096 |
ento | -1.080 |
Token | 0 |
Token ID | 15 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
viol | +1.172 |
NPR | +1.108 |
containment | +1.085 |
IRO | +1.084 |
Carnegie | +1.040 |
Neg logit contributions | |
ommel | -0.884 |
reviewed | -0.814 |
atal | -0.705 |
eson | -0.696 |
ento | -0.692 |
Token | official |
Token ID | 1743 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
viol | +0.927 |
NPR | +0.861 |
containment | +0.835 |
IRO | +0.833 |
Carnegie | +0.795 |
Neg logit contributions | |
ommel | -1.199 |
reviewed | -1.125 |
atal | -1.016 |
eson | -1.004 |
ento | -1.001 |
Token | statement |
Token ID | 2643 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
viol | +2.638 |
NPR | +2.383 |
containment | +2.291 |
IRO | +2.184 |
Carnegie | +2.134 |
Neg logit contributions | |
ommel | -5.800 |
reviewed | -5.530 |
atal | -5.131 |
ento | -5.087 |
eson | -5.012 |
Token | soon |
Token ID | 2582 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
viol | +1.102 |
NPR | +1.021 |
containment | +0.996 |
IRO | +0.989 |
Carnegie | +0.956 |
Neg logit contributions | |
ommel | -1.335 |
reviewed | -1.249 |
atal | -1.124 |
ento | -1.099 |
eson | -1.098 |
Token | . |
Token ID | 13 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
ommel | +0.057 |
reviewed | +0.053 |
atal | +0.047 |
eson | +0.046 |
ento | +0.046 |
Neg logit contributions | |
viol | -0.066 |
NPR | -0.062 |
containment | -0.060 |
IRO | -0.060 |
Carnegie | -0.058 |
Token | As |
Token ID | 1081 |
Feature activation | -6.86e+00 |
Pos logit contributions | |
viol | +2.001 |
NPR | +1.881 |
IRO | +1.821 |
containment | +1.809 |
Carnegie | +1.794 |
Neg logit contributions | |
ommel | -2.527 |
reviewed | -2.318 |
atal | -2.133 |
ento | -2.089 |
eson | -2.088 |
Token | well |
Token ID | 880 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
viol | +3.455 |
NPR | +2.997 |
containment | +2.978 |
IRO | +2.847 |
Carnegie | +2.846 |
Neg logit contributions | |
ommel | -7.015 |
reviewed | -6.590 |
atal | -6.124 |
MAC | -6.011 |
eson | -6.003 |
Token | as |
Token ID | 355 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
viol | +5.716 |
NPR | +5.288 |
containment | +5.149 |
Civ | +4.927 |
Carnegie | +4.892 |
Neg logit contributions | |
ommel | -9.069 |
reviewed | -8.345 |
atal | -7.995 |
ento | -7.808 |
ederal | -7.745 |
Token | updates |
Token ID | 5992 |
Feature activation | -5.81e+00 |
Pos logit contributions | |
viol | +2.854 |
NPR | +2.683 |
containment | +2.622 |
IRO | +2.599 |
Carnegie | +2.515 |
Neg logit contributions | |
ommel | -2.442 |
reviewed | -2.239 |
atal | -1.966 |
eson | -1.946 |
ento | -1.939 |
Token | regarding |
Token ID | 5115 |
Feature activation | -4.82e+00 |
Pos logit contributions | |
viol | +3.934 |
NPR | +3.667 |
containment | +3.577 |
IRO | +3.465 |
Carnegie | +3.380 |
Neg logit contributions | |
ommel | -4.956 |
reviewed | -4.512 |
atal | -4.245 |
eson | -4.134 |
ento | -4.086 |
Token | the |
Token ID | 262 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
viol | +3.024 |
NPR | +2.791 |
containment | +2.718 |
IRO | +2.692 |
Carnegie | +2.636 |
Neg logit contributions | |
ommel | -4.349 |
reviewed | -4.035 |
atal | -3.714 |
ento | -3.666 |
eson | -3.649 |
Token | future |
Token ID | 2003 |
Feature activation | -4.82e+00 |
Pos logit contributions | |
viol | +2.486 |
containment | +2.233 |
NPR | +2.201 |
IRO | +2.126 |
Carnegie | +2.076 |
Neg logit contributions | |
ommel | -3.717 |
reviewed | -3.484 |
atal | -3.177 |
eson | -3.141 |
ento | -3.132 |
Token | image |
Token ID | 2939 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
viol | +0.652 |
NPR | +0.585 |
containment | +0.577 |
IRO | +0.562 |
Carnegie | +0.536 |
Neg logit contributions | |
ommel | -1.182 |
reviewed | -1.121 |
atal | -1.022 |
eson | -1.015 |
ento | -1.012 |
Token | = |
Token ID | 796 |
Feature activation | -5.22e+00 |
Pos logit contributions | |
ommel | +0.402 |
reviewed | +0.374 |
atal | +0.333 |
eson | +0.329 |
ento | +0.327 |
Neg logit contributions | |
viol | -0.395 |
NPR | -0.370 |
containment | -0.361 |
IRO | -0.360 |
Carnegie | -0.346 |
Token | im |
Token ID | 545 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
viol | +2.753 |
containment | +2.395 |
NPR | +2.374 |
IRO | +2.274 |
Carnegie | +2.204 |
Neg logit contributions | |
ommel | -5.305 |
reviewed | -5.040 |
atal | -4.627 |
ento | -4.590 |
eson | -4.560 |
Token | utils |
Token ID | 26791 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
viol | +0.080 |
NPR | +0.075 |
containment | +0.073 |
IRO | +0.073 |
Carnegie | +0.070 |
Neg logit contributions | |
ommel | -0.076 |
reviewed | -0.071 |
atal | -0.063 |
eson | -0.062 |
ento | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
viol | +2.198 |
NPR | +2.095 |
IRO | +2.027 |
containment | +2.011 |
Carnegie | +1.972 |
Neg logit contributions | |
ommel | -2.306 |
reviewed | -2.109 |
atal | -1.939 |
ento | -1.902 |
eson | -1.886 |
Token | res |
Token ID | 411 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
viol | +2.713 |
NPR | +2.633 |
containment | +2.520 |
IRO | +2.500 |
pees | +2.434 |
Neg logit contributions | |
ommel | -3.117 |
reviewed | -2.871 |
atal | -2.637 |
ento | -2.619 |
eson | -2.544 |
Token | ize |
Token ID | 1096 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
NPR | +5.015 |
viol | +4.916 |
pees | +4.826 |
IRO | +4.692 |
containment | +4.592 |
Neg logit contributions | |
ommel | -8.755 |
reviewed | -8.372 |
atal | -7.694 |
Thu | -7.687 |
ento | -7.678 |
Token | ( |
Token ID | 7 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
viol | +1.786 |
NPR | +1.724 |
IRO | +1.665 |
containment | +1.620 |
Carnegie | +1.595 |
Neg logit contributions | |
ommel | -1.798 |
reviewed | -1.648 |
atal | -1.488 |
ento | -1.483 |
eson | -1.466 |
Token | image |
Token ID | 9060 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
viol | +1.949 |
NPR | +1.884 |
IRO | +1.804 |
containment | +1.772 |
Carnegie | +1.653 |
Neg logit contributions | |
ommel | -3.619 |
reviewed | -3.385 |
eson | -3.097 |
atal | -3.081 |
ento | -3.080 |
Token | , |
Token ID | 11 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
viol | +2.462 |
NPR | +2.396 |
IRO | +2.288 |
containment | +2.260 |
Carnegie | +2.223 |
Neg logit contributions | |
ommel | -2.606 |
reviewed | -2.358 |
atal | -2.198 |
ento | -2.143 |
ederal | -2.125 |
Token | height |
Token ID | 6001 |
Feature activation | -4.33e+00 |
Pos logit contributions | |
viol | +3.139 |
containment | +2.929 |
NPR | +2.902 |
Carnegie | +2.803 |
IRO | +2.793 |
Neg logit contributions | |
ommel | -2.863 |
reviewed | -2.713 |
atal | -2.379 |
ento | -2.360 |
eson | -2.339 |
Token | Lo |
Token ID | 6706 |
Feature activation | -5.67e+00 |
Pos logit contributions | |
viol | +0.214 |
NPR | +0.200 |
containment | +0.195 |
IRO | +0.194 |
Carnegie | +0.187 |
Neg logit contributions | |
ommel | -0.219 |
reviewed | -0.205 |
atal | -0.182 |
eson | -0.180 |
ento | -0.179 |
Token | a |
Token ID | 64 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
viol | +5.078 |
NPR | +4.774 |
IRO | +4.636 |
containment | +4.612 |
Civ | +4.570 |
Neg logit contributions | |
ommel | -3.617 |
reviewed | -3.331 |
atal | -2.839 |
ento | -2.816 |
Thu | -2.775 |
Token | series |
Token ID | 2168 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
viol | +1.005 |
NPR | +0.935 |
containment | +0.912 |
IRO | +0.908 |
Carnegie | +0.871 |
Neg logit contributions | |
ommel | -1.161 |
reviewed | -1.086 |
atal | -0.974 |
eson | -0.963 |
ento | -0.960 |
Token | of |
Token ID | 286 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
ommel | +1.118 |
reviewed | +1.038 |
atal | +0.928 |
eson | +0.913 |
ento | +0.906 |
Neg logit contributions | |
viol | -1.148 |
NPR | -1.084 |
IRO | -1.052 |
containment | -1.048 |
Carnegie | -1.009 |
Token | [ |
Token ID | 685 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ommel | +1.173 |
reviewed | +1.102 |
atal | +0.990 |
eson | +0.973 |
ento | +0.967 |
Neg logit contributions | |
viol | -1.046 |
NPR | -0.982 |
IRO | -0.951 |
containment | -0.945 |
pees | -0.901 |
Token | CO |
Token ID | 8220 |
Feature activation | -9.98e+00 |
Pos logit contributions | |
viol | +1.437 |
NPR | +1.365 |
IRO | +1.334 |
containment | +1.319 |
Carnegie | +1.270 |
Neg logit contributions | |
ommel | -1.414 |
reviewed | -1.292 |
atal | -1.159 |
eson | -1.147 |
ento | -1.141 |
Token | 2 |
Token ID | 362 |
Feature activation | -7.30e+00 |
Pos logit contributions | |
viol | +5.282 |
IRO | +4.913 |
NPR | +4.911 |
containment | +4.627 |
Civ | +4.541 |
Neg logit contributions | |
ommel | -9.472 |
reviewed | -8.697 |
eson | -8.138 |
ento | -8.083 |
enic | -7.931 |
Token | ] |
Token ID | 2361 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
viol | +4.920 |
containment | +4.551 |
NPR | +4.523 |
IRO | +4.504 |
Civ | +4.217 |
Neg logit contributions | |
ommel | -6.117 |
reviewed | -5.577 |
eson | -5.120 |
atal | -5.068 |
ento | -5.049 |
Token | are |
Token ID | 389 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
viol | +0.268 |
NPR | +0.252 |
containment | +0.245 |
IRO | +0.245 |
Carnegie | +0.235 |
Neg logit contributions | |
ommel | -0.264 |
reviewed | -0.245 |
atal | -0.218 |
eson | -0.215 |
ento | -0.214 |
Token | displayed |
Token ID | 9066 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ommel | +0.052 |
reviewed | +0.048 |
atal | +0.043 |
eson | +0.043 |
ento | +0.043 |
Neg logit contributions | |
viol | -0.047 |
NPR | -0.044 |
containment | -0.043 |
IRO | -0.043 |
Carnegie | -0.041 |
Token | on |
Token ID | 319 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
ommel | +0.840 |
reviewed | +0.780 |
atal | +0.694 |
eson | +0.683 |
ento | +0.680 |
Neg logit contributions | |
viol | -0.884 |
NPR | -0.833 |
IRO | -0.812 |
containment | -0.811 |
Carnegie | -0.779 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ommel | +0.662 |
reviewed | +0.616 |
atal | +0.548 |
eson | +0.540 |
ento | +0.537 |
Neg logit contributions | |
viol | -0.679 |
NPR | -0.637 |
containment | -0.621 |
IRO | -0.620 |
Carnegie | -0.595 |
Token |
|
Token ID | 198 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
ommel | +0.833 |
reviewed | +0.776 |
atal | +0.683 |
eson | +0.675 |
ento | +0.670 |
Neg logit contributions | |
viol | -0.917 |
NPR | -0.859 |
containment | -0.841 |
IRO | -0.837 |
Carnegie | -0.805 |
Token | EDIT |
Token ID | 24706 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
ommel | +0.476 |
reviewed | +0.442 |
atal | +0.385 |
eson | +0.379 |
ento | +0.377 |
Neg logit contributions | |
viol | -0.584 |
NPR | -0.546 |
containment | -0.538 |
IRO | -0.533 |
Carnegie | -0.517 |
Token | OR |
Token ID | 1581 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
viol | +1.846 |
NPR | +1.718 |
IRO | +1.674 |
containment | +1.644 |
Civ | +1.571 |
Neg logit contributions | |
ommel | -2.493 |
reviewed | -2.321 |
atal | -2.114 |
eson | -2.090 |
ento | -2.074 |
Token | ' |
Token ID | 6 |
Feature activation | -7.57e+00 |
Pos logit contributions | |
viol | +2.404 |
NPR | +2.290 |
IRO | +2.229 |
containment | +2.216 |
Carnegie | +2.126 |
Neg logit contributions | |
ommel | -2.353 |
reviewed | -2.140 |
atal | -1.923 |
ento | -1.897 |
eson | -1.894 |
Token | S |
Token ID | 50 |
Feature activation | -9.98e+00 |
Pos logit contributions | |
viol | +6.349 |
NPR | +6.123 |
IRO | +5.950 |
Carnegie | +5.766 |
containment | +5.722 |
Neg logit contributions | |
ommel | -5.221 |
reviewed | -4.541 |
atal | -4.210 |
ento | -4.208 |
ederal | -4.115 |
Token | NOTE |
Token ID | 24550 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
viol | +6.433 |
NPR | +5.749 |
containment | +5.724 |
IRO | +5.688 |
Civ | +5.420 |
Neg logit contributions | |
ommel | -8.633 |
reviewed | -7.875 |
atal | -7.401 |
ederal | -7.399 |
eson | -7.265 |
Token | : |
Token ID | 25 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
viol | +1.711 |
NPR | +1.576 |
IRO | +1.547 |
containment | +1.526 |
Carnegie | +1.469 |
Neg logit contributions | |
ommel | -2.217 |
reviewed | -2.079 |
atal | -1.887 |
eson | -1.872 |
ento | -1.868 |
Token | The |
Token ID | 383 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
viol | +1.432 |
NPR | +1.346 |
IRO | +1.311 |
containment | +1.303 |
Carnegie | +1.270 |
Neg logit contributions | |
ommel | -1.478 |
reviewed | -1.364 |
atal | -1.223 |
eson | -1.207 |
ento | -1.199 |
Token | lineup |
Token ID | 12750 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
ommel | +0.247 |
reviewed | +0.230 |
atal | +0.205 |
eson | +0.202 |
ento | +0.201 |
Neg logit contributions | |
viol | -0.241 |
NPR | -0.226 |
containment | -0.220 |
IRO | -0.219 |
Carnegie | -0.210 |
Token | below |
Token ID | 2174 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
viol | +0.769 |
NPR | +0.716 |
containment | +0.701 |
IRO | +0.696 |
Carnegie | +0.670 |
Neg logit contributions | |
ommel | -0.784 |
reviewed | -0.728 |
atal | -0.652 |
eson | -0.641 |
ento | -0.639 |
Token | until |
Token ID | 1566 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
viol | +0.508 |
NPR | +0.458 |
containment | +0.457 |
IRO | +0.434 |
Carnegie | +0.428 |
Neg logit contributions | |
ommel | -0.803 |
reviewed | -0.757 |
atal | -0.697 |
eson | -0.684 |
ento | -0.679 |
Token | golden |
Token ID | 10861 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
ommel | +1.789 |
reviewed | +1.640 |
atal | +1.430 |
eson | +1.407 |
ento | +1.397 |
Neg logit contributions | |
viol | -2.242 |
NPR | -2.134 |
IRO | -2.077 |
containment | -2.068 |
Carnegie | -1.993 |
Token | brown |
Token ID | 7586 |
Feature activation | -9.60e-01 |
Pos logit contributions | |
ommel | +1.723 |
reviewed | +1.567 |
atal | +1.363 |
eson | +1.336 |
ento | +1.328 |
Neg logit contributions | |
viol | -2.381 |
NPR | -2.274 |
IRO | -2.221 |
containment | -2.204 |
Carnegie | -2.132 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
viol | +0.752 |
NPR | +0.707 |
containment | +0.688 |
IRO | +0.688 |
Carnegie | +0.667 |
Neg logit contributions | |
ommel | -0.721 |
reviewed | -0.668 |
atal | -0.600 |
ento | -0.588 |
eson | -0.587 |
Token | These |
Token ID | 2312 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
ommel | +1.076 |
reviewed | +0.997 |
atal | +0.862 |
ento | +0.842 |
eson | +0.840 |
Neg logit contributions | |
viol | -1.508 |
NPR | -1.426 |
IRO | -1.404 |
containment | -1.400 |
Carnegie | -1.334 |
Token | are |
Token ID | 389 |
Feature activation | -9.87e+00 |
Pos logit contributions | |
viol | +1.058 |
NPR | +0.980 |
containment | +0.955 |
IRO | +0.939 |
Carnegie | +0.911 |
Neg logit contributions | |
ommel | -1.384 |
reviewed | -1.297 |
atal | -1.174 |
ento | -1.160 |
eson | -1.159 |
Token | lightly |
Token ID | 15376 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
viol | +6.797 |
NPR | +6.153 |
containment | +6.145 |
IRO | +6.138 |
Carnegie | +5.871 |
Neg logit contributions | |
ommel | -7.812 |
reviewed | -7.216 |
atal | -6.863 |
eson | -6.760 |
ederal | -6.632 |
Token | fried |
Token ID | 23018 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
viol | +1.323 |
NPR | +1.207 |
containment | +1.169 |
IRO | +1.162 |
Carnegie | +1.107 |
Neg logit contributions | |
ommel | -1.851 |
reviewed | -1.728 |
atal | -1.588 |
eson | -1.571 |
ento | -1.553 |
Token | and |
Token ID | 290 |
Feature activation | -3.90e+00 |
Pos logit contributions | |
viol | +1.134 |
NPR | +1.056 |
containment | +1.028 |
IRO | +1.018 |
Carnegie | +0.996 |
Neg logit contributions | |
ommel | -1.232 |
reviewed | -1.146 |
atal | -1.033 |
ento | -1.012 |
eson | -1.009 |
Token | not |
Token ID | 407 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
viol | +2.820 |
NPR | +2.631 |
containment | +2.556 |
IRO | +2.497 |
Carnegie | +2.435 |
Neg logit contributions | |
ommel | -3.175 |
reviewed | -2.938 |
atal | -2.697 |
eson | -2.619 |
ento | -2.619 |
Token | deep |
Token ID | 2769 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
viol | +2.378 |
containment | +2.123 |
NPR | +2.082 |
IRO | +2.022 |
Carnegie | +1.983 |
Neg logit contributions | |
ommel | -3.696 |
reviewed | -3.492 |
atal | -3.229 |
ento | -3.169 |
eson | -3.168 |
Token | , |
Token ID | 11 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
ommel | +0.965 |
reviewed | +0.892 |
atal | +0.793 |
eson | +0.779 |
ento | +0.777 |
Neg logit contributions | |
viol | -1.054 |
NPR | -0.994 |
IRO | -0.967 |
containment | -0.966 |
Carnegie | -0.932 |
Token | then |
Token ID | 788 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
viol | +0.009 |
NPR | +0.009 |
containment | +0.009 |
IRO | +0.009 |
Carnegie | +0.008 |
Neg logit contributions | |
ommel | -0.009 |
reviewed | -0.009 |
atal | -0.008 |
eson | -0.008 |
ento | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
ommel | +0.115 |
reviewed | +0.106 |
atal | +0.095 |
eson | +0.094 |
ento | +0.094 |
Neg logit contributions | |
viol | -0.112 |
NPR | -0.106 |
IRO | -0.103 |
containment | -0.102 |
Carnegie | -0.099 |
Token |
|
Token ID | 198 |
Feature activation | +8.73e-01 |
Pos logit contributions | |
ommel | +1.360 |
reviewed | +1.272 |
atal | +1.111 |
eson | +1.101 |
ento | +1.089 |
Neg logit contributions | |
viol | -1.557 |
NPR | -1.454 |
containment | -1.431 |
IRO | -1.422 |
Carnegie | -1.364 |
Token | [ |
Token ID | 58 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ommel | +0.611 |
reviewed | +0.566 |
atal | +0.498 |
eson | +0.489 |
ento | +0.487 |
Neg logit contributions | |
viol | -0.736 |
NPR | -0.690 |
containment | -0.677 |
IRO | -0.673 |
Carnegie | -0.653 |
Token | CO |
Token ID | 8220 |
Feature activation | -9.67e+00 |
Pos logit contributions | |
viol | +1.040 |
NPR | +0.981 |
IRO | +0.957 |
containment | +0.955 |
Carnegie | +0.918 |
Neg logit contributions | |
ommel | -0.917 |
reviewed | -0.845 |
atal | -0.746 |
eson | -0.737 |
ento | -0.732 |
Token | 2 |
Token ID | 362 |
Feature activation | -6.31e+00 |
Pos logit contributions | |
viol | +5.200 |
NPR | +4.790 |
IRO | +4.759 |
containment | +4.600 |
Civ | +4.515 |
Neg logit contributions | |
ommel | -9.162 |
reviewed | -8.537 |
eson | -7.890 |
ento | -7.877 |
atal | -7.685 |
Token | ] |
Token ID | 2361 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
viol | +4.375 |
containment | +4.051 |
NPR | +4.045 |
IRO | +3.999 |
Civ | +3.771 |
Neg logit contributions | |
ommel | -5.180 |
reviewed | -4.808 |
eson | -4.342 |
atal | -4.302 |
ento | -4.284 |
Token | ( |
Token ID | 7 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
ommel | +0.511 |
reviewed | +0.474 |
atal | +0.421 |
eson | +0.414 |
ento | +0.412 |
Neg logit contributions | |
viol | -0.551 |
NPR | -0.518 |
containment | -0.504 |
IRO | -0.503 |
Carnegie | -0.485 |
Token | t |
Token ID | 83 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
ommel | +0.503 |
reviewed | +0.465 |
atal | +0.407 |
eson | +0.400 |
ento | +0.399 |
Neg logit contributions | |
viol | -0.627 |
NPR | -0.591 |
containment | -0.578 |
IRO | -0.575 |
Carnegie | -0.557 |
Token | ) |
Token ID | 8 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ommel | +0.494 |
reviewed | +0.453 |
atal | +0.392 |
eson | +0.386 |
ento | +0.384 |
Neg logit contributions | |
viol | -0.695 |
NPR | -0.657 |
containment | -0.643 |
IRO | -0.642 |
Carnegie | -0.621 |
Token | aw |
Token ID | 707 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
ommel | +1.201 |
reviewed | +1.100 |
atal | +0.991 |
eson | +0.979 |
ento | +0.970 |
Neg logit contributions | |
viol | -1.232 |
NPR | -1.131 |
containment | -1.130 |
IRO | -1.111 |
Carnegie | -1.071 |
Token | , |
Token ID | 11 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
ommel | +1.831 |
reviewed | +1.691 |
atal | +1.498 |
eson | +1.482 |
ento | +1.474 |
Neg logit contributions | |
viol | -1.929 |
NPR | -1.830 |
IRO | -1.785 |
containment | -1.776 |
Carnegie | -1.704 |
Token | and |
Token ID | 290 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
viol | +0.217 |
NPR | +0.203 |
containment | +0.198 |
IRO | +0.198 |
Carnegie | +0.190 |
Neg logit contributions | |
ommel | -0.209 |
reviewed | -0.194 |
atal | -0.173 |
eson | -0.170 |
ento | -0.169 |
Token | R |
Token ID | 371 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
viol | +0.164 |
NPR | +0.153 |
containment | +0.149 |
IRO | +0.149 |
Carnegie | +0.142 |
Neg logit contributions | |
ommel | -0.177 |
reviewed | -0.165 |
atal | -0.147 |
eson | -0.145 |
ento | -0.145 |
Token | ors |
Token ID | 669 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
viol | +0.530 |
NPR | +0.502 |
IRO | +0.489 |
containment | +0.489 |
Carnegie | +0.471 |
Neg logit contributions | |
ommel | -0.418 |
reviewed | -0.388 |
atal | -0.338 |
eson | -0.333 |
ento | -0.332 |
Token | h |
Token ID | 71 |
Feature activation | -9.52e+00 |
Pos logit contributions | |
ommel | +0.220 |
reviewed | +0.199 |
atal | +0.170 |
eson | +0.166 |
ento | +0.166 |
Neg logit contributions | |
viol | -0.362 |
NPR | -0.344 |
containment | -0.337 |
IRO | -0.336 |
Carnegie | -0.326 |
Token | & |
Token ID | 1222 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
viol | +8.187 |
NPR | +7.897 |
IRO | +7.631 |
containment | +7.483 |
pees | +7.461 |
Neg logit contributions | |
ommel | -5.815 |
reviewed | -5.570 |
atal | -4.655 |
ederal | -4.620 |
ento | -4.599 |
Token | Br |
Token ID | 1709 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
viol | +1.182 |
NPR | +1.103 |
containment | +1.075 |
IRO | +1.071 |
Carnegie | +1.040 |
Neg logit contributions | |
ommel | -1.177 |
reviewed | -1.096 |
atal | -0.976 |
ento | -0.962 |
eson | -0.962 |
Token | ine |
Token ID | 500 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
viol | +1.324 |
NPR | +1.255 |
IRO | +1.223 |
containment | +1.213 |
Carnegie | +1.172 |
Neg logit contributions | |
ommel | -1.156 |
reviewed | -1.080 |
atal | -0.939 |
eson | -0.930 |
ento | -0.923 |
Token | . |
Token ID | 13 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
ommel | +1.181 |
reviewed | +1.094 |
atal | +0.969 |
eson | +0.957 |
ento | +0.950 |
Neg logit contributions | |
viol | -1.250 |
NPR | -1.182 |
containment | -1.148 |
IRO | -1.145 |
Carnegie | -1.101 |
Token | The |
Token ID | 383 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
ommel | +0.348 |
reviewed | +0.326 |
atal | +0.288 |
eson | +0.285 |
ento | +0.283 |
Neg logit contributions | |
viol | -0.353 |
NPR | -0.330 |
containment | -0.325 |
IRO | -0.322 |
Carnegie | -0.306 |
Token | # |
Token ID | 1303 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
viol | +1.508 |
NPR | +1.411 |
containment | +1.385 |
IRO | +1.374 |
Carnegie | +1.326 |
Neg logit contributions | |
ommel | -1.371 |
reviewed | -1.277 |
atal | -1.127 |
eson | -1.108 |
ento | -1.103 |
Token | import |
Token ID | 1330 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
viol | +2.351 |
NPR | +2.138 |
containment | +2.097 |
IRO | +2.053 |
Carnegie | +1.957 |
Neg logit contributions | |
ommel | -3.711 |
reviewed | -3.512 |
atal | -3.191 |
ento | -3.170 |
eson | -3.146 |
Token | the |
Token ID | 262 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
viol | +1.382 |
NPR | +1.269 |
containment | +1.254 |
IRO | +1.224 |
Carnegie | +1.192 |
Neg logit contributions | |
ommel | -1.725 |
reviewed | -1.621 |
atal | -1.455 |
ento | -1.441 |
eson | -1.432 |
Token | necessary |
Token ID | 3306 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
viol | +1.751 |
NPR | +1.641 |
containment | +1.624 |
IRO | +1.590 |
Carnegie | +1.557 |
Neg logit contributions | |
ommel | -1.420 |
reviewed | -1.301 |
atal | -1.144 |
ento | -1.120 |
eson | -1.118 |
Token | packages |
Token ID | 10392 |
Feature activation | -6.39e+00 |
Pos logit contributions | |
viol | +1.413 |
NPR | +1.347 |
containment | +1.332 |
IRO | +1.300 |
Carnegie | +1.279 |
Neg logit contributions | |
ommel | -0.804 |
reviewed | -0.723 |
atal | -0.610 |
ento | -0.600 |
eson | -0.590 |
Token | from |
Token ID | 422 |
Feature activation | -9.29e+00 |
Pos logit contributions | |
viol | +4.846 |
NPR | +4.533 |
containment | +4.526 |
Carnegie | +4.310 |
IRO | +4.309 |
Neg logit contributions | |
ommel | -4.850 |
reviewed | -4.514 |
atal | -3.981 |
eson | -3.942 |
ento | -3.920 |
Token | im |
Token ID | 545 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
viol | +4.717 |
containment | +4.122 |
NPR | +3.934 |
ig | +3.841 |
IRO | +3.810 |
Neg logit contributions | |
ommel | -9.279 |
reviewed | -8.947 |
Thu | -8.103 |
atal | -8.089 |
ento | -8.058 |
Token | utils |
Token ID | 26791 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
ommel | +1.023 |
reviewed | +0.945 |
atal | +0.838 |
eson | +0.828 |
ento | +0.820 |
Neg logit contributions | |
viol | -1.137 |
NPR | -1.064 |
containment | -1.040 |
IRO | -1.036 |
Carnegie | -0.996 |
Token | import |
Token ID | 1330 |
Feature activation | +7.35e-01 |
Pos logit contributions | |
viol | +1.208 |
NPR | +1.106 |
containment | +1.079 |
IRO | +1.066 |
Carnegie | +1.015 |
Neg logit contributions | |
ommel | -1.867 |
reviewed | -1.766 |
atal | -1.605 |
eson | -1.587 |
ento | -1.586 |
Token | paths |
Token ID | 13532 |
Feature activation | -4.25e+00 |
Pos logit contributions | |
ommel | +0.584 |
reviewed | +0.544 |
atal | +0.487 |
eson | +0.482 |
ento | +0.477 |
Neg logit contributions | |
viol | -0.547 |
NPR | -0.514 |
IRO | -0.502 |
containment | -0.498 |
Carnegie | -0.477 |
Token | import |
Token ID | 1330 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
viol | +2.412 |
NPR | +2.192 |
containment | +2.169 |
IRO | +2.090 |
Carnegie | +2.007 |
Neg logit contributions | |
ommel | -4.100 |
reviewed | -3.889 |
atal | -3.550 |
ento | -3.504 |
eson | -3.500 |
Token | , |
Token ID | 837 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
viol | +0.011 |
NPR | +0.011 |
IRO | +0.011 |
containment | +0.010 |
Carnegie | +0.010 |
Neg logit contributions | |
ommel | -0.011 |
reviewed | -0.010 |
atal | -0.009 |
ento | -0.009 |
eson | -0.009 |
Token | c |
Token ID | 269 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
viol | +0.724 |
NPR | +0.679 |
containment | +0.667 |
IRO | +0.661 |
Carnegie | +0.642 |
Neg logit contributions | |
ommel | -0.594 |
reviewed | -0.549 |
atal | -0.482 |
eson | -0.473 |
ento | -0.473 |
Token | v |
Token ID | 85 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
viol | +1.484 |
NPR | +1.392 |
IRO | +1.340 |
containment | +1.338 |
Carnegie | +1.288 |
Neg logit contributions | |
ommel | -1.770 |
reviewed | -1.649 |
atal | -1.485 |
eson | -1.477 |
ento | -1.470 |
Token | 2 |
Token ID | 17 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
ommel | +1.506 |
reviewed | +1.384 |
atal | +1.165 |
eson | +1.140 |
ento | +1.128 |
Neg logit contributions | |
viol | -2.538 |
NPR | -2.389 |
containment | -2.365 |
IRO | -2.339 |
Carnegie | -2.268 |
Token | . |
Token ID | 764 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ommel | +0.672 |
reviewed | +0.629 |
atal | +0.564 |
eson | +0.557 |
ento | +0.555 |
Neg logit contributions | |
viol | -0.587 |
NPR | -0.547 |
containment | -0.532 |
IRO | -0.531 |
Carnegie | -0.508 |
Token | MOR |
Token ID | 35208 |
Feature activation | -9.26e+00 |
Pos logit contributions | |
viol | +1.258 |
NPR | +1.168 |
containment | +1.138 |
IRO | +1.137 |
Carnegie | +1.093 |
Neg logit contributions | |
ommel | -1.390 |
reviewed | -1.303 |
atal | -1.162 |
ento | -1.154 |
eson | -1.153 |
Token | PH |
Token ID | 11909 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
NPR | +4.814 |
IRO | +4.540 |
viol | +4.315 |
Carnegie | +3.870 |
Nev | +3.850 |
Neg logit contributions | |
ommel | -9.201 |
reviewed | -8.281 |
atal | -8.044 |
ederal | -7.939 |
ento | -7.839 |
Token | _ |
Token ID | 62 |
Feature activation | -4.90e+00 |
Pos logit contributions | |
viol | +1.387 |
NPR | +1.349 |
IRO | +1.322 |
containment | +1.287 |
Carnegie | +1.265 |
Neg logit contributions | |
ommel | -0.960 |
reviewed | -0.866 |
atal | -0.762 |
eson | -0.739 |
ento | -0.738 |
Token | BL |
Token ID | 9148 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
viol | +1.615 |
NPR | +1.579 |
IRO | +1.493 |
containment | +1.263 |
Carnegie | +1.177 |
Neg logit contributions | |
ommel | -5.860 |
reviewed | -5.472 |
ento | -5.222 |
atal | -5.207 |
eson | -5.202 |
Token | ACK |
Token ID | 8120 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
viol | +0.546 |
NPR | +0.515 |
IRO | +0.500 |
containment | +0.494 |
Carnegie | +0.473 |
Neg logit contributions | |
ommel | -0.648 |
reviewed | -0.606 |
atal | -0.545 |
eson | -0.541 |
ento | -0.539 |
Token | H |
Token ID | 39 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
viol | +2.206 |
NPR | +2.145 |
IRO | +2.102 |
containment | +2.023 |
Carnegie | +1.948 |
Neg logit contributions | |
ommel | -2.214 |
reviewed | -2.014 |
atal | -1.842 |
eson | -1.798 |
ento | -1.792 |
Token | he |
Token ID | 339 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
viol | +0.482 |
NPR | +0.449 |
containment | +0.440 |
IRO | +0.436 |
Carnegie | +0.421 |
Neg logit contributions | |
ommel | -0.495 |
reviewed | -0.461 |
atal | -0.411 |
eson | -0.406 |
ento | -0.404 |
Token | 'll |
Token ID | 1183 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
viol | +1.053 |
NPR | +0.967 |
containment | +0.950 |
IRO | +0.933 |
Carnegie | +0.901 |
Neg logit contributions | |
ommel | -1.281 |
reviewed | -1.198 |
atal | -1.077 |
eson | -1.062 |
ento | -1.062 |
Token | be |
Token ID | 307 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
viol | +1.316 |
NPR | +1.193 |
containment | +1.173 |
IRO | +1.145 |
Carnegie | +1.101 |
Neg logit contributions | |
ommel | -1.847 |
reviewed | -1.741 |
atal | -1.561 |
ento | -1.558 |
eson | -1.546 |
Token | in |
Token ID | 287 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
viol | +2.051 |
NPR | +1.886 |
containment | +1.854 |
IRO | +1.823 |
Carnegie | +1.754 |
Neg logit contributions | |
ommel | -2.195 |
reviewed | -2.040 |
atal | -1.827 |
eson | -1.790 |
ento | -1.781 |
Token | custody |
Token ID | 10804 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
ommel | +0.463 |
reviewed | +0.434 |
atal | +0.391 |
eson | +0.386 |
ento | +0.384 |
Neg logit contributions | |
viol | -0.386 |
NPR | -0.360 |
IRO | -0.349 |
containment | -0.347 |
Carnegie | -0.333 |
Token | pretty |
Token ID | 2495 |
Feature activation | -9.16e+00 |
Pos logit contributions | |
viol | +0.264 |
NPR | +0.246 |
containment | +0.240 |
IRO | +0.239 |
Carnegie | +0.230 |
Neg logit contributions | |
ommel | -0.282 |
reviewed | -0.263 |
atal | -0.235 |
eson | -0.232 |
ento | -0.231 |
Token | quickly |
Token ID | 2952 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
viol | +4.502 |
containment | +4.001 |
NPR | +3.908 |
IRO | +3.792 |
Carnegie | +3.571 |
Neg logit contributions | |
ommel | -9.485 |
reviewed | -8.884 |
ento | -8.231 |
MAC | -8.143 |
atal | -8.112 |
Token | ." |
Token ID | 526 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ommel | +1.064 |
reviewed | +0.992 |
atal | +0.886 |
eson | +0.874 |
ento | +0.865 |
Neg logit contributions | |
viol | -1.016 |
NPR | -0.966 |
IRO | -0.942 |
containment | -0.927 |
Carnegie | -0.896 |
Token |
|
Token ID | 198 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
viol | +0.884 |
NPR | +0.827 |
containment | +0.806 |
IRO | +0.804 |
Carnegie | +0.775 |
Neg logit contributions | |
ommel | -0.879 |
reviewed | -0.813 |
atal | -0.725 |
ento | -0.712 |
eson | -0.711 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
viol | +0.585 |
NPR | +0.557 |
IRO | +0.537 |
containment | +0.533 |
Carnegie | +0.519 |
Neg logit contributions | |
ommel | -0.608 |
reviewed | -0.557 |
atal | -0.504 |
ento | -0.496 |
eson | -0.494 |
Token | ' |
Token ID | 6 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
viol | +0.182 |
NPR | +0.174 |
IRO | +0.168 |
containment | +0.167 |
Carnegie | +0.161 |
Neg logit contributions | |
ommel | -0.174 |
reviewed | -0.160 |
atal | -0.143 |
eson | -0.141 |
ento | -0.141 |
Token | new |
Token ID | 649 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
ommel | +0.511 |
reviewed | +0.480 |
atal | +0.438 |
eson | +0.433 |
ento | +0.430 |
Neg logit contributions | |
viol | -0.352 |
NPR | -0.327 |
IRO | -0.314 |
containment | -0.313 |
Carnegie | -0.297 |
Token | - |
Token ID | 12 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
ommel | +2.862 |
reviewed | +2.648 |
atal | +2.305 |
eson | +2.273 |
Thu | +2.207 |
Neg logit contributions | |
viol | -4.024 |
NPR | -3.791 |
IRO | -3.683 |
containment | -3.674 |
pees | -3.539 |
Token | object |
Token ID | 15252 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
ommel | +0.214 |
reviewed | +0.187 |
atal | +0.148 |
eson | +0.143 |
ento | +0.142 |
Neg logit contributions | |
viol | -0.554 |
NPR | -0.531 |
containment | -0.521 |
IRO | -0.520 |
Carnegie | -0.506 |
Token | system |
Token ID | 1080 |
Feature activation | -8.58e-01 |
Pos logit contributions | |
viol | +2.712 |
containment | +2.417 |
NPR | +2.403 |
IRO | +2.328 |
Carnegie | +2.252 |
Neg logit contributions | |
ommel | -4.508 |
reviewed | -4.248 |
atal | -3.869 |
ento | -3.837 |
eson | -3.833 |
Token | . |
Token ID | 13 |
Feature activation | -7.98e+00 |
Pos logit contributions | |
viol | +0.672 |
NPR | +0.639 |
IRO | +0.621 |
containment | +0.617 |
Carnegie | +0.599 |
Neg logit contributions | |
ommel | -0.642 |
reviewed | -0.591 |
atal | -0.531 |
ento | -0.522 |
eson | -0.521 |
Token | draw |
Token ID | 19334 |
Feature activation | -9.15e+00 |
Pos logit contributions | |
viol | +4.648 |
NPR | +4.368 |
IRO | +4.262 |
containment | +4.122 |
pees | +4.016 |
Neg logit contributions | |
ommel | -7.105 |
reviewed | -6.588 |
eson | -6.066 |
ento | -6.059 |
atal | -6.038 |
Token | ing |
Token ID | 278 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
viol | +5.704 |
NPR | +5.606 |
IRO | +5.338 |
containment | +5.262 |
ig | +5.228 |
Neg logit contributions | |
ommel | -7.247 |
reviewed | -6.830 |
MAC | -6.021 |
ento | -6.018 |
Thu | -6.005 |
Token | . |
Token ID | 13 |
Feature activation | -7.84e+00 |
Pos logit contributions | |
viol | +3.717 |
NPR | +3.569 |
IRO | +3.486 |
containment | +3.419 |
Carnegie | +3.350 |
Neg logit contributions | |
ommel | -4.311 |
reviewed | -3.948 |
atal | -3.641 |
ento | -3.607 |
eson | -3.579 |
Token | bit |
Token ID | 2545 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
viol | +5.411 |
NPR | +5.209 |
IRO | +5.144 |
containment | +4.982 |
pees | +4.810 |
Neg logit contributions | |
ommel | -6.047 |
reviewed | -5.592 |
ento | -5.009 |
eson | -4.986 |
atal | -4.942 |
Token | map |
Token ID | 8899 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
viol | +3.435 |
NPR | +3.307 |
IRO | +3.224 |
containment | +3.207 |
Carnegie | +3.098 |
Neg logit contributions | |
ommel | -2.706 |
reviewed | -2.498 |
ento | -2.153 |
atal | -2.152 |
eson | -2.108 |
Token | ($ |
Token ID | 16763 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
viol | +1.896 |
NPR | +1.709 |
containment | +1.705 |
IRO | +1.680 |
Carnegie | +1.592 |
Neg logit contributions | |
ommel | -3.449 |
reviewed | -3.281 |
atal | -2.980 |
ento | -2.958 |
eson | -2.956 |
Token | New |
Token ID | 968 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
viol | +0.138 |
NPR | +0.128 |
containment | +0.124 |
IRO | +0.124 |
Carnegie | +0.119 |
Neg logit contributions | |
ommel | -0.160 |
reviewed | -0.150 |
atal | -0.135 |
eson | -0.133 |
ento | -0.133 |
Token | - |
Token ID | 12 |
Feature activation | +8.51e-01 |
Pos logit contributions | |
ommel | +0.589 |
reviewed | +0.546 |
atal | +0.479 |
eson | +0.473 |
ento | +0.469 |
Neg logit contributions | |
viol | -0.696 |
NPR | -0.652 |
containment | -0.639 |
IRO | -0.635 |
Carnegie | -0.611 |
Token | Object |
Token ID | 10267 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
ommel | +0.527 |
reviewed | +0.482 |
atal | +0.414 |
eson | +0.406 |
ento | +0.404 |
Neg logit contributions | |
viol | -0.788 |
NPR | -0.742 |
containment | -0.730 |
IRO | -0.726 |
Carnegie | -0.704 |
Token | System |
Token ID | 4482 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
viol | +2.563 |
NPR | +2.329 |
containment | +2.309 |
IRO | +2.273 |
Carnegie | +2.196 |
Neg logit contributions | |
ommel | -3.730 |
reviewed | -3.526 |
atal | -3.181 |
eson | -3.146 |
ento | -3.138 |
Token | . |
Token ID | 13 |
Feature activation | -5.98e+00 |
Pos logit contributions | |
ommel | +0.257 |
reviewed | +0.237 |
atal | +0.211 |
eson | +0.207 |
ento | +0.206 |
Neg logit contributions | |
viol | -0.286 |
NPR | -0.270 |
IRO | -0.263 |
containment | -0.263 |
Carnegie | -0.253 |
Token | Draw |
Token ID | 25302 |
Feature activation | -9.03e+00 |
Pos logit contributions | |
viol | +3.752 |
NPR | +3.579 |
IRO | +3.481 |
containment | +3.359 |
Civ | +3.255 |
Neg logit contributions | |
ommel | -5.244 |
reviewed | -4.894 |
atal | -4.474 |
eson | -4.460 |
ento | -4.423 |
Token | ing |
Token ID | 278 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
viol | +5.653 |
NPR | +5.578 |
IRO | +5.318 |
containment | +5.174 |
ig | +5.107 |
Neg logit contributions | |
ommel | -7.228 |
reviewed | -6.787 |
atal | -6.009 |
ederal | -5.998 |
ento | -5.982 |
Token | . |
Token ID | 13 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
viol | +3.309 |
NPR | +3.189 |
IRO | +3.130 |
containment | +3.030 |
Civ | +2.968 |
Neg logit contributions | |
ommel | -3.557 |
reviewed | -3.266 |
atal | -2.995 |
ento | -2.937 |
eson | -2.936 |
Token | Rect |
Token ID | 45474 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
viol | +4.030 |
NPR | +3.885 |
IRO | +3.815 |
containment | +3.682 |
Civ | +3.567 |
Neg logit contributions | |
ommel | -4.210 |
reviewed | -3.890 |
atal | -3.467 |
eson | -3.454 |
ento | -3.425 |
Token | angle |
Token ID | 9248 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
viol | +1.466 |
NPR | +1.352 |
IRO | +1.302 |
containment | +1.302 |
Civ | +1.206 |
Neg logit contributions | |
ommel | -2.637 |
reviewed | -2.512 |
eson | -2.273 |
atal | -2.273 |
ento | -2.241 |
Token | F |
Token ID | 37 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
viol | +3.525 |
NPR | +3.360 |
IRO | +3.311 |
containment | +3.278 |
Carnegie | +3.146 |
Neg logit contributions | |
ommel | -2.784 |
reviewed | -2.568 |
atal | -2.254 |
eson | -2.224 |
ento | -2.163 |
Token | . |
Token ID | 13 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
viol | +0.589 |
NPR | +0.557 |
IRO | +0.541 |
containment | +0.539 |
Carnegie | +0.523 |
Neg logit contributions | |
ommel | -0.546 |
reviewed | -0.501 |
atal | -0.449 |
ento | -0.442 |
eson | -0.441 |
Token | Weapon |
Token ID | 13072 |
Feature activation | -6.06e+00 |
Pos logit contributions | |
viol | +0.339 |
NPR | +0.317 |
containment | +0.309 |
IRO | +0.309 |
Carnegie | +0.298 |
Neg logit contributions | |
ommel | -0.339 |
reviewed | -0.315 |
atal | -0.280 |
eson | -0.277 |
ento | -0.276 |
Token | person |
Token ID | 1048 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
viol | +4.430 |
containment | +4.008 |
NPR | +3.939 |
IRO | +3.846 |
Carnegie | +3.713 |
Neg logit contributions | |
ommel | -4.841 |
reviewed | -4.599 |
atal | -4.094 |
ento | -4.049 |
eson | -4.016 |
Token | loaded |
Token ID | 9639 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
viol | +3.362 |
containment | +3.062 |
NPR | +3.048 |
IRO | +3.000 |
Carnegie | +2.906 |
Neg logit contributions | |
ommel | -3.413 |
reviewed | -3.225 |
atal | -2.854 |
ento | -2.822 |
eson | -2.812 |
Token | , |
Token ID | 11 |
Feature activation | -6.10e+00 |
Pos logit contributions | |
viol | +1.151 |
NPR | +1.074 |
containment | +1.056 |
IRO | +1.042 |
Carnegie | +1.009 |
Neg logit contributions | |
ommel | -1.037 |
reviewed | -0.966 |
atal | -0.848 |
eson | -0.837 |
ento | -0.836 |
Token | initially |
Token ID | 7317 |
Feature activation | -8.97e+00 |
Pos logit contributions | |
viol | +4.316 |
NPR | +3.980 |
containment | +3.898 |
IRO | +3.841 |
Carnegie | +3.643 |
Neg logit contributions | |
ommel | -4.993 |
reviewed | -4.614 |
ento | -4.127 |
atal | -4.087 |
MAC | -4.065 |
Token | failing |
Token ID | 9894 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
viol | +5.201 |
NPR | +4.471 |
containment | +4.457 |
IRO | +4.154 |
Carnegie | +4.044 |
Neg logit contributions | |
ommel | -8.347 |
reviewed | -7.930 |
atal | -7.169 |
ento | -7.135 |
ourke | -7.114 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
viol | +0.424 |
NPR | +0.396 |
containment | +0.388 |
IRO | +0.385 |
Carnegie | +0.372 |
Neg logit contributions | |
ommel | -0.373 |
reviewed | -0.347 |
atal | -0.305 |
eson | -0.302 |
ento | -0.301 |
Token | sec |
Token ID | 792 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
viol | +0.922 |
NPR | +0.853 |
containment | +0.836 |
IRO | +0.831 |
Carnegie | +0.795 |
Neg logit contributions | |
ommel | -0.983 |
reviewed | -0.916 |
atal | -0.819 |
ento | -0.808 |
eson | -0.805 |
Token | ures |
Token ID | 942 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
ommel | +0.118 |
reviewed | +0.110 |
atal | +0.097 |
ento | +0.096 |
eson | +0.096 |
Neg logit contributions | |
viol | -0.120 |
NPR | -0.112 |
containment | -0.110 |
IRO | -0.109 |
Carnegie | -0.106 |
Token | mark |
Token ID | 1317 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
viol | +0.945 |
NPR | +0.854 |
containment | +0.836 |
IRO | +0.822 |
Carnegie | +0.781 |
Neg logit contributions | |
ommel | -1.542 |
reviewed | -1.464 |
atal | -1.329 |
ento | -1.322 |
eson | -1.320 |
Token | New |
Token ID | 968 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
viol | +0.269 |
NPR | +0.248 |
containment | +0.241 |
IRO | +0.240 |
Carnegie | +0.228 |
Neg logit contributions | |
ommel | -0.384 |
reviewed | -0.362 |
atal | -0.328 |
eson | -0.325 |
ento | -0.323 |
Token | - |
Token ID | 12 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
ommel | +0.546 |
reviewed | +0.506 |
atal | +0.446 |
eson | +0.440 |
ento | +0.437 |
Neg logit contributions | |
viol | -0.625 |
NPR | -0.588 |
containment | -0.574 |
IRO | -0.572 |
Carnegie | -0.551 |
Token | Object |
Token ID | 10267 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
ommel | +0.851 |
reviewed | +0.779 |
atal | +0.667 |
eson | +0.656 |
ento | +0.651 |
Neg logit contributions | |
viol | -1.297 |
NPR | -1.219 |
containment | -1.204 |
IRO | -1.194 |
Carnegie | -1.158 |
Token | System |
Token ID | 4482 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
viol | +3.001 |
NPR | +2.705 |
containment | +2.678 |
IRO | +2.629 |
Carnegie | +2.549 |
Neg logit contributions | |
ommel | -4.770 |
reviewed | -4.505 |
atal | -4.090 |
eson | -4.037 |
ento | -4.031 |
Token | . |
Token ID | 13 |
Feature activation | -6.34e+00 |
Pos logit contributions | |
viol | +0.365 |
NPR | +0.347 |
IRO | +0.337 |
containment | +0.335 |
Carnegie | +0.325 |
Neg logit contributions | |
ommel | -0.343 |
reviewed | -0.315 |
atal | -0.282 |
ento | -0.277 |
eson | -0.277 |
Token | Draw |
Token ID | 25302 |
Feature activation | -8.96e+00 |
Pos logit contributions | |
viol | +3.837 |
NPR | +3.811 |
IRO | +3.652 |
Civ | +3.413 |
containment | +3.405 |
Neg logit contributions | |
ommel | -5.677 |
reviewed | -5.203 |
atal | -4.887 |
ento | -4.827 |
eson | -4.820 |
Token | ing |
Token ID | 278 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
viol | +5.575 |
NPR | +5.522 |
IRO | +5.242 |
containment | +5.089 |
ig | +5.045 |
Neg logit contributions | |
ommel | -7.188 |
reviewed | -6.735 |
atal | -5.966 |
ederal | -5.963 |
eson | -5.935 |
Token | . |
Token ID | 13 |
Feature activation | -4.89e+00 |
Pos logit contributions | |
viol | +2.994 |
NPR | +2.910 |
IRO | +2.836 |
containment | +2.736 |
Carnegie | +2.700 |
Neg logit contributions | |
ommel | -3.337 |
reviewed | -3.044 |
atal | -2.816 |
ento | -2.760 |
eson | -2.753 |
Token | Font |
Token ID | 23252 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
viol | +3.674 |
NPR | +3.557 |
IRO | +3.494 |
containment | +3.340 |
Carnegie | +3.257 |
Neg logit contributions | |
ommel | -3.718 |
reviewed | -3.435 |
atal | -3.060 |
eson | -3.034 |
ento | -3.009 |
Token | ($ |
Token ID | 16763 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
viol | +2.934 |
NPR | +2.812 |
IRO | +2.758 |
containment | +2.731 |
Carnegie | +2.623 |
Neg logit contributions | |
ommel | -2.124 |
reviewed | -1.960 |
atal | -1.690 |
eson | -1.647 |
ento | -1.628 |
Token | Font |
Token ID | 23252 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
ommel | +1.335 |
reviewed | +1.231 |
atal | +1.071 |
eson | +1.056 |
ento | +1.053 |
Neg logit contributions | |
viol | -1.693 |
NPR | -1.580 |
containment | -1.558 |
IRO | -1.542 |
Carnegie | -1.496 |
Token | expectation |
Token ID | 17507 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
ommel | +0.202 |
reviewed | +0.187 |
atal | +0.164 |
eson | +0.162 |
ento | +0.161 |
Neg logit contributions | |
viol | -0.237 |
NPR | -0.224 |
IRO | -0.218 |
containment | -0.218 |
Carnegie | -0.210 |
Token | of |
Token ID | 286 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
viol | +0.476 |
NPR | +0.447 |
containment | +0.437 |
IRO | +0.436 |
Carnegie | +0.420 |
Neg logit contributions | |
ommel | -0.415 |
reviewed | -0.385 |
atal | -0.339 |
eson | -0.334 |
ento | -0.333 |
Token | 6 |
Token ID | 718 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
viol | +0.845 |
NPR | +0.791 |
containment | +0.770 |
IRO | +0.768 |
Carnegie | +0.738 |
Neg logit contributions | |
ommel | -0.873 |
reviewed | -0.814 |
atal | -0.726 |
eson | -0.716 |
ento | -0.714 |
Token | , |
Token ID | 11 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
viol | +0.752 |
NPR | +0.707 |
containment | +0.689 |
IRO | +0.689 |
Carnegie | +0.664 |
Neg logit contributions | |
ommel | -0.656 |
reviewed | -0.608 |
atal | -0.535 |
eson | -0.528 |
ento | -0.527 |
Token | 000 |
Token ID | 830 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
viol | +0.448 |
NPR | +0.422 |
containment | +0.412 |
IRO | +0.412 |
Carnegie | +0.397 |
Neg logit contributions | |
ommel | -0.366 |
reviewed | -0.338 |
atal | -0.296 |
eson | -0.292 |
ento | -0.291 |
Token | h |
Token ID | 289 |
Feature activation | -8.86e+00 |
Pos logit contributions | |
viol | +0.444 |
NPR | +0.412 |
containment | +0.401 |
IRO | +0.399 |
Carnegie | +0.381 |
Neg logit contributions | |
ommel | -0.573 |
reviewed | -0.538 |
atal | -0.486 |
eson | -0.480 |
ento | -0.478 |
Token | ry |
Token ID | 563 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
viol | +5.988 |
NPR | +5.927 |
IRO | +5.687 |
pees | +5.428 |
Carnegie | +5.333 |
Neg logit contributions | |
ommel | -6.804 |
reviewed | -6.451 |
atal | -5.719 |
eson | -5.692 |
MAC | -5.682 |
Token | v |
Token ID | 85 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
viol | +1.741 |
NPR | +1.607 |
IRO | +1.570 |
containment | +1.560 |
Carnegie | +1.500 |
Neg logit contributions | |
ommel | -1.999 |
reviewed | -1.864 |
atal | -1.661 |
eson | -1.657 |
ento | -1.646 |
Token | nia |
Token ID | 18142 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
viol | +1.531 |
NPR | +1.424 |
IRO | +1.368 |
containment | +1.367 |
Carnegie | +1.302 |
Neg logit contributions | |
ommel | -2.148 |
reviewed | -2.022 |
atal | -1.828 |
eson | -1.815 |
ento | -1.807 |
Token | ($ |
Token ID | 7198 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
viol | +0.611 |
NPR | +0.573 |
containment | +0.559 |
IRO | +0.558 |
Carnegie | +0.537 |
Neg logit contributions | |
ommel | -0.563 |
reviewed | -0.523 |
atal | -0.463 |
eson | -0.456 |
ento | -0.454 |
Token | 226 |
Token ID | 24909 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
viol | +0.271 |
NPR | +0.258 |
IRO | +0.252 |
containment | +0.251 |
Carnegie | +0.243 |
Neg logit contributions | |
ommel | -0.188 |
reviewed | -0.171 |
atal | -0.148 |
eson | -0.146 |
ento | -0.145 |
Token | game |
Token ID | 983 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
viol | +1.323 |
NPR | +1.208 |
containment | +1.194 |
IRO | +1.165 |
Carnegie | +1.132 |
Neg logit contributions | |
ommel | -1.696 |
reviewed | -1.593 |
atal | -1.430 |
ento | -1.416 |
eson | -1.414 |
Token | . |
Token ID | 13 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
ommel | +0.086 |
reviewed | +0.080 |
atal | +0.071 |
eson | +0.070 |
ento | +0.069 |
Neg logit contributions | |
viol | -0.097 |
NPR | -0.091 |
containment | -0.089 |
IRO | -0.089 |
Carnegie | -0.086 |
Token | While |
Token ID | 2893 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
ommel | +1.992 |
reviewed | +1.879 |
atal | +1.666 |
eson | +1.648 |
ento | +1.621 |
Neg logit contributions | |
viol | -1.988 |
NPR | -1.865 |
containment | -1.821 |
IRO | -1.811 |
pees | -1.734 |
Token | Philip |
Token ID | 14576 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
ommel | +1.103 |
reviewed | +1.034 |
atal | +0.927 |
eson | +0.916 |
ento | +0.907 |
Neg logit contributions | |
viol | -0.982 |
NPR | -0.922 |
IRO | -0.893 |
containment | -0.890 |
Carnegie | -0.850 |
Token | Gr |
Token ID | 1902 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
viol | +2.115 |
NPR | +1.924 |
containment | +1.906 |
IRO | +1.882 |
Carnegie | +1.868 |
Neg logit contributions | |
ommel | -2.068 |
reviewed | -1.909 |
atal | -1.690 |
eson | -1.654 |
ento | -1.647 |
Token | ub |
Token ID | 549 |
Feature activation | -8.81e+00 |
Pos logit contributions | |
viol | +1.712 |
NPR | +1.644 |
IRO | +1.594 |
containment | +1.576 |
Carnegie | +1.526 |
Neg logit contributions | |
ommel | -1.460 |
reviewed | -1.363 |
eson | -1.163 |
atal | -1.162 |
MAC | -1.158 |
Token | auer |
Token ID | 16261 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
viol | +7.978 |
IRO | +7.396 |
NPR | +7.289 |
containment | +7.214 |
pees | +7.166 |
Neg logit contributions | |
ommel | -5.097 |
reviewed | -4.630 |
ento | -4.008 |
MAC | -3.848 |
ensual | -3.769 |
Token | was |
Token ID | 373 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
viol | +0.189 |
NPR | +0.177 |
containment | +0.173 |
IRO | +0.172 |
Carnegie | +0.165 |
Neg logit contributions | |
ommel | -0.194 |
reviewed | -0.181 |
atal | -0.161 |
eson | -0.159 |
ento | -0.158 |
Token | being |
Token ID | 852 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
viol | +0.670 |
NPR | +0.631 |
containment | +0.612 |
IRO | +0.612 |
Carnegie | +0.586 |
Neg logit contributions | |
ommel | -0.676 |
reviewed | -0.630 |
atal | -0.562 |
eson | -0.554 |
ento | -0.550 |
Token | pepp |
Token ID | 49038 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
ommel | +1.066 |
reviewed | +1.000 |
atal | +0.895 |
eson | +0.882 |
ento | +0.871 |
Neg logit contributions | |
viol | -0.989 |
NPR | -0.939 |
IRO | -0.907 |
containment | -0.902 |
Carnegie | -0.861 |
Token | ered |
Token ID | 1068 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
ommel | +0.200 |
reviewed | +0.184 |
atal | +0.164 |
ento | +0.161 |
eson | +0.161 |
Neg logit contributions | |
viol | -0.215 |
NPR | -0.201 |
IRO | -0.195 |
containment | -0.195 |
Carnegie | -0.190 |
Token | ABS |
Token ID | 29950 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
ommel | +1.573 |
reviewed | +1.471 |
atal | +1.328 |
eson | +1.302 |
ento | +1.294 |
Neg logit contributions | |
viol | -1.443 |
NPR | -1.360 |
IRO | -1.322 |
containment | -1.313 |
Carnegie | -1.249 |
Token | OL |
Token ID | 3535 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
viol | +2.744 |
NPR | +2.571 |
IRO | +2.516 |
containment | +2.471 |
Carnegie | +2.385 |
Neg logit contributions | |
ommel | -3.364 |
reviewed | -3.143 |
eson | -2.817 |
atal | -2.816 |
ento | -2.764 |
Token | UT |
Token ID | 3843 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
viol | +3.030 |
NPR | +2.935 |
IRO | +2.907 |
containment | +2.693 |
Civ | +2.663 |
Neg logit contributions | |
ommel | -3.622 |
reviewed | -3.315 |
eson | -2.998 |
atal | -2.955 |
ento | -2.904 |
Token | ELY |
Token ID | 30943 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
viol | +2.322 |
NPR | +2.141 |
IRO | +2.077 |
containment | +1.959 |
Carnegie | +1.863 |
Neg logit contributions | |
ommel | -4.891 |
reviewed | -4.533 |
eson | -4.182 |
ento | -4.174 |
atal | -4.165 |
Token | NOT |
Token ID | 5626 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
ommel | +0.173 |
reviewed | +0.161 |
atal | +0.144 |
eson | +0.142 |
ento | +0.141 |
Neg logit contributions | |
viol | -0.164 |
NPR | -0.153 |
containment | -0.149 |
IRO | -0.149 |
Carnegie | -0.143 |
Token | H |
Token ID | 39 |
Feature activation | -8.79e+00 |
Pos logit contributions | |
viol | +0.335 |
NPR | +0.314 |
containment | +0.306 |
IRO | +0.306 |
Carnegie | +0.294 |
Neg logit contributions | |
ommel | -0.314 |
reviewed | -0.292 |
atal | -0.259 |
eson | -0.255 |
ento | -0.254 |
Token | ING |
Token ID | 2751 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
viol | +6.064 |
IRO | +5.750 |
NPR | +5.576 |
containment | +5.305 |
Carnegie | +5.265 |
Neg logit contributions | |
ommel | -7.344 |
reviewed | -6.772 |
atal | -6.242 |
ento | -6.090 |
ederal | -6.062 |
Token | ? |
Token ID | 30 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
viol | +2.281 |
NPR | +2.136 |
containment | +2.090 |
IRO | +2.078 |
Carnegie | +2.016 |
Neg logit contributions | |
ommel | -1.928 |
reviewed | -1.766 |
atal | -1.564 |
ento | -1.545 |
eson | -1.542 |
Token | The |
Token ID | 383 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
viol | +1.553 |
NPR | +1.457 |
IRO | +1.414 |
containment | +1.413 |
Carnegie | +1.364 |
Neg logit contributions | |
ommel | -1.573 |
reviewed | -1.458 |
atal | -1.301 |
eson | -1.284 |
ento | -1.282 |
Token | secret |
Token ID | 3200 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
viol | +0.089 |
NPR | +0.083 |
containment | +0.081 |
IRO | +0.080 |
Carnegie | +0.077 |
Neg logit contributions | |
ommel | -0.093 |
reviewed | -0.087 |
atal | -0.078 |
eson | -0.077 |
ento | -0.076 |
Token | is |
Token ID | 318 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
viol | +0.370 |
NPR | +0.346 |
containment | +0.338 |
IRO | +0.336 |
Carnegie | +0.324 |
Neg logit contributions | |
ommel | -0.366 |
reviewed | -0.340 |
atal | -0.302 |
eson | -0.298 |
ento | -0.298 |
Token | 12 |
Token ID | 1105 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
viol | +3.771 |
containment | +3.429 |
NPR | +3.306 |
IRO | +3.217 |
Carnegie | +3.189 |
Neg logit contributions | |
ommel | -6.406 |
reviewed | -5.841 |
atal | -5.430 |
apult | -5.419 |
MAC | -5.404 |
Token | to |
Token ID | 284 |
Feature activation | -6.80e+00 |
Pos logit contributions | |
viol | +2.244 |
NPR | +2.075 |
containment | +2.030 |
IRO | +2.015 |
Carnegie | +1.987 |
Neg logit contributions | |
ommel | -2.627 |
reviewed | -2.401 |
atal | -2.187 |
ento | -2.162 |
eson | -2.153 |
Token | 50 |
Token ID | 2026 |
Feature activation | +3.00e+00 |
Pos logit contributions | |
viol | +4.222 |
containment | +3.778 |
Carnegie | +3.750 |
NPR | +3.743 |
Civ | +3.548 |
Neg logit contributions | |
ommel | -6.118 |
reviewed | -5.657 |
atal | -5.241 |
eson | -5.205 |
ento | -5.172 |
Token | , |
Token ID | 11 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ommel | +2.028 |
reviewed | +1.887 |
atal | +1.633 |
eson | +1.610 |
ento | +1.597 |
Neg logit contributions | |
viol | -2.594 |
NPR | -2.437 |
containment | -2.401 |
IRO | -2.386 |
Carnegie | -2.288 |
Token | characterized |
Token ID | 16264 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
ommel | +1.319 |
reviewed | +1.240 |
atal | +1.118 |
eson | +1.105 |
ento | +1.100 |
Neg logit contributions | |
viol | -1.030 |
NPR | -0.956 |
containment | -0.929 |
IRO | -0.926 |
Carnegie | -0.883 |
Token | by |
Token ID | 416 |
Feature activation | -8.76e+00 |
Pos logit contributions | |
viol | +2.916 |
NPR | +2.714 |
containment | +2.626 |
Carnegie | +2.608 |
IRO | +2.601 |
Neg logit contributions | |
ommel | -3.514 |
reviewed | -3.184 |
atal | -2.958 |
ento | -2.947 |
eson | -2.882 |
Token | multiple |
Token ID | 3294 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
viol | +3.609 |
containment | +3.110 |
NPR | +2.923 |
Carnegie | +2.874 |
IRO | +2.862 |
Neg logit contributions | |
ommel | -9.606 |
reviewed | -8.992 |
atal | -8.511 |
ento | -8.416 |
eson | -8.380 |
Token | tiny |
Token ID | 7009 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
viol | +2.142 |
NPR | +1.947 |
containment | +1.925 |
IRO | +1.907 |
Carnegie | +1.839 |
Neg logit contributions | |
ommel | -2.638 |
reviewed | -2.463 |
atal | -2.212 |
eson | -2.201 |
ento | -2.185 |
Token | ker |
Token ID | 41927 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ommel | +0.646 |
reviewed | +0.590 |
atal | +0.508 |
eson | +0.499 |
ento | +0.496 |
Neg logit contributions | |
viol | -0.961 |
NPR | -0.911 |
containment | -0.891 |
IRO | -0.890 |
Carnegie | -0.860 |
Token | at |
Token ID | 265 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ommel | +1.339 |
reviewed | +1.223 |
atal | +1.053 |
eson | +1.033 |
ento | +1.027 |
Neg logit contributions | |
viol | -2.004 |
NPR | -1.899 |
containment | -1.859 |
IRO | -1.855 |
Carnegie | -1.795 |
Token | otic |
Token ID | 6210 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ommel | +0.812 |
reviewed | +0.731 |
atal | +0.614 |
eson | +0.599 |
ento | +0.593 |
Neg logit contributions | |
viol | -1.540 |
NPR | -1.466 |
containment | -1.437 |
IRO | -1.436 |
Carnegie | -1.391 |
Token | intuitive |
Token ID | 19933 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
ommel | +0.524 |
reviewed | +0.491 |
atal | +0.438 |
eson | +0.433 |
ento | +0.430 |
Neg logit contributions | |
viol | -0.488 |
NPR | -0.459 |
IRO | -0.446 |
containment | -0.446 |
Carnegie | -0.426 |
Token | software |
Token ID | 3788 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
ommel | +0.042 |
reviewed | +0.039 |
atal | +0.035 |
eson | +0.035 |
ento | +0.034 |
Neg logit contributions | |
viol | -0.043 |
NPR | -0.040 |
containment | -0.039 |
IRO | -0.039 |
Carnegie | -0.038 |
Token | integration |
Token ID | 11812 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ommel | +2.283 |
reviewed | +2.126 |
atal | +1.891 |
eson | +1.870 |
ento | +1.865 |
Neg logit contributions | |
viol | -2.246 |
NPR | -2.121 |
IRO | -2.056 |
containment | -2.045 |
Carnegie | -1.976 |
Token | make |
Token ID | 787 |
Feature activation | +5.68e+00 |
Pos logit contributions | |
ommel | +0.917 |
reviewed | +0.853 |
atal | +0.762 |
eson | +0.754 |
ento | +0.750 |
Neg logit contributions | |
viol | -0.867 |
NPR | -0.813 |
IRO | -0.791 |
containment | -0.790 |
Carnegie | -0.757 |
Token | it |
Token ID | 340 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
ommel | +3.751 |
reviewed | +3.480 |
atal | +3.009 |
eson | +2.957 |
ento | +2.941 |
Neg logit contributions | |
viol | -4.801 |
NPR | -4.642 |
IRO | -4.568 |
containment | -4.476 |
pees | -4.332 |
Token | easy |
Token ID | 2562 |
Feature activation | +6.89e+00 |
Pos logit contributions | |
ommel | +1.946 |
reviewed | +1.814 |
atal | +1.620 |
eson | +1.599 |
ento | +1.592 |
Neg logit contributions | |
viol | -1.874 |
NPR | -1.765 |
containment | -1.721 |
IRO | -1.718 |
Carnegie | -1.639 |
Token | to |
Token ID | 284 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
ommel | +4.685 |
reviewed | +4.292 |
atal | +3.682 |
ento | +3.612 |
eson | +3.611 |
Neg logit contributions | |
viol | -5.631 |
NPR | -5.477 |
IRO | -5.347 |
containment | -5.291 |
pees | -5.102 |
Token | love |
Token ID | 1842 |
Feature activation | +4.66e+00 |
Pos logit contributions | |
ommel | +2.247 |
reviewed | +2.084 |
atal | +1.887 |
eson | +1.852 |
ento | +1.840 |
Neg logit contributions | |
viol | -1.886 |
NPR | -1.782 |
IRO | -1.730 |
containment | -1.725 |
Carnegie | -1.631 |
Token | . |
Token ID | 13 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
ommel | +3.403 |
reviewed | +3.151 |
atal | +2.799 |
eson | +2.723 |
ento | +2.721 |
Neg logit contributions | |
viol | -3.705 |
NPR | -3.534 |
IRO | -3.456 |
containment | -3.439 |
pees | -3.285 |
Token | More |
Token ID | 3125 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
ommel | +0.379 |
reviewed | +0.353 |
atal | +0.315 |
eson | +0.310 |
ento | +0.309 |
Neg logit contributions | |
viol | -0.371 |
NPR | -0.348 |
containment | -0.338 |
IRO | -0.338 |
Carnegie | -0.325 |
Token | important |
Token ID | 1593 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
ommel | +0.154 |
reviewed | +0.143 |
atal | +0.126 |
eson | +0.124 |
ento | +0.124 |
Neg logit contributions | |
viol | -0.173 |
NPR | -0.163 |
containment | -0.159 |
IRO | -0.159 |
Carnegie | -0.153 |
Token | powerful |
Token ID | 3665 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
viol | +0.072 |
NPR | +0.068 |
IRO | +0.066 |
containment | +0.066 |
Carnegie | +0.063 |
Neg logit contributions | |
ommel | -0.079 |
reviewed | -0.073 |
atal | -0.066 |
eson | -0.065 |
ento | -0.065 |
Token | Yam |
Token ID | 14063 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
viol | +3.078 |
NPR | +2.800 |
containment | +2.791 |
IRO | +2.731 |
Carnegie | +2.709 |
Neg logit contributions | |
ommel | -3.190 |
reviewed | -2.996 |
atal | -2.647 |
eson | -2.615 |
ento | -2.588 |
Token | ag |
Token ID | 363 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
viol | +0.975 |
NPR | +0.920 |
IRO | +0.902 |
containment | +0.900 |
Carnegie | +0.868 |
Neg logit contributions | |
ommel | -0.724 |
reviewed | -0.665 |
atal | -0.574 |
eson | -0.567 |
ento | -0.559 |
Token | uchi |
Token ID | 22200 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ommel | +0.505 |
reviewed | +0.421 |
atal | +0.319 |
eson | +0.299 |
ento | +0.298 |
Neg logit contributions | |
viol | -1.780 |
NPR | -1.710 |
containment | -1.684 |
IRO | -1.676 |
Carnegie | -1.639 |
Token | - |
Token ID | 12 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
viol | +1.285 |
NPR | +1.192 |
containment | +1.167 |
IRO | +1.162 |
Carnegie | +1.120 |
Neg logit contributions | |
ommel | -1.386 |
reviewed | -1.298 |
atal | -1.154 |
eson | -1.141 |
ento | -1.136 |
Token | g |
Token ID | 70 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
ommel | +0.816 |
reviewed | +0.754 |
atal | +0.661 |
eson | +0.648 |
ento | +0.648 |
Neg logit contributions | |
viol | -1.005 |
NPR | -0.945 |
containment | -0.927 |
IRO | -0.921 |
Carnegie | -0.891 |
Token | umi |
Token ID | 12994 |
Feature activation | -8.20e-01 |
Pos logit contributions | |
ommel | +3.956 |
reviewed | +3.492 |
atal | +3.164 |
ento | +3.110 |
ederal | +2.970 |
Neg logit contributions | |
viol | -6.601 |
containment | -6.198 |
NPR | -6.162 |
IRO | -6.027 |
Carnegie | -5.847 |
Token | crime |
Token ID | 4065 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
viol | +0.614 |
NPR | +0.574 |
containment | +0.559 |
IRO | +0.557 |
Carnegie | +0.536 |
Neg logit contributions | |
ommel | -0.650 |
reviewed | -0.607 |
atal | -0.542 |
eson | -0.535 |
ento | -0.533 |
Token | synd |
Token ID | 11150 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
viol | +1.027 |
NPR | +0.941 |
containment | +0.919 |
IRO | +0.911 |
Carnegie | +0.871 |
Neg logit contributions | |
ommel | -1.420 |
reviewed | -1.334 |
atal | -1.208 |
eson | -1.195 |
ento | -1.191 |
Token | icate |
Token ID | 5344 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
ommel | +0.123 |
reviewed | +0.112 |
atal | +0.095 |
eson | +0.095 |
ento | +0.094 |
Neg logit contributions | |
viol | -0.194 |
NPR | -0.184 |
containment | -0.181 |
IRO | -0.180 |
Carnegie | -0.173 |
Token | , |
Token ID | 11 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
viol | +0.021 |
NPR | +0.020 |
containment | +0.019 |
IRO | +0.019 |
Carnegie | +0.018 |
Neg logit contributions | |
ommel | -0.021 |
reviewed | -0.020 |
atal | -0.018 |
eson | -0.017 |
ento | -0.017 |
Token | state |
Token ID | 1181 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
viol | +0.070 |
NPR | +0.066 |
containment | +0.065 |
IRO | +0.064 |
Carnegie | +0.062 |
Neg logit contributions | |
ommel | -0.053 |
reviewed | -0.049 |
atal | -0.042 |
ento | -0.042 |
eson | -0.042 |
Token | ( |
Token ID | 357 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
ommel | +0.372 |
reviewed | +0.344 |
atal | +0.305 |
eson | +0.300 |
ento | +0.299 |
Neg logit contributions | |
viol | -0.409 |
NPR | -0.385 |
containment | -0.375 |
IRO | -0.374 |
Carnegie | -0.361 |
Token | such |
Token ID | 10508 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
ommel | +0.176 |
reviewed | +0.162 |
atal | +0.142 |
eson | +0.140 |
ento | +0.140 |
Neg logit contributions | |
viol | -0.209 |
NPR | -0.197 |
containment | -0.192 |
IRO | -0.191 |
Carnegie | -0.185 |
Token | as |
Token ID | 355 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
viol | +0.133 |
NPR | +0.125 |
containment | +0.121 |
IRO | +0.121 |
Carnegie | +0.117 |
Neg logit contributions | |
ommel | -0.140 |
reviewed | -0.130 |
atal | -0.118 |
eson | -0.115 |
ento | -0.115 |
Token | through |
Token ID | 832 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
viol | +3.392 |
NPR | +3.123 |
containment | +3.117 |
IRO | +3.021 |
Carnegie | +2.961 |
Neg logit contributions | |
ommel | -3.197 |
reviewed | -2.955 |
atal | -2.684 |
ento | -2.598 |
eson | -2.572 |
Token | a |
Token ID | 257 |
Feature activation | +7.53e+00 |
Pos logit contributions | |
ommel | +1.687 |
reviewed | +1.574 |
atal | +1.409 |
eson | +1.391 |
ento | +1.384 |
Neg logit contributions | |
viol | -1.551 |
NPR | -1.455 |
IRO | -1.410 |
containment | -1.407 |
Carnegie | -1.342 |
Token | or |
Token ID | 393 |
Feature activation | +8.74e+00 |
Pos logit contributions | |
ommel | +4.905 |
reviewed | +4.500 |
atal | +3.960 |
eson | +3.933 |
ento | +3.903 |
Neg logit contributions | |
viol | -6.428 |
NPR | -6.108 |
IRO | -5.954 |
containment | -5.844 |
pees | -5.684 |
Token | ) |
Token ID | 1267 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ommel | +5.808 |
reviewed | +5.273 |
atal | +4.721 |
eson | +4.644 |
MAC | +4.568 |
Neg logit contributions | |
viol | -7.222 |
NPR | -6.965 |
IRO | -6.814 |
containment | -6.599 |
Civ | -6.387 |
Token | must |
Token ID | 1276 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
ommel | +1.205 |
reviewed | +1.120 |
atal | +0.997 |
eson | +0.978 |
ento | +0.976 |
Neg logit contributions | |
viol | -1.240 |
NPR | -1.157 |
containment | -1.138 |
IRO | -1.126 |
Carnegie | -1.088 |
Token | not |
Token ID | 407 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
viol | +2.167 |
NPR | +1.978 |
containment | +1.947 |
IRO | +1.889 |
Carnegie | +1.845 |
Neg logit contributions | |
ommel | -2.451 |
reviewed | -2.272 |
atal | -2.043 |
MAC | -2.006 |
ederal | -2.004 |
Token | modify |
Token ID | 13096 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
viol | +1.482 |
NPR | +1.367 |
containment | +1.337 |
IRO | +1.315 |
Carnegie | +1.272 |
Neg logit contributions | |
ommel | -1.771 |
reviewed | -1.646 |
atal | -1.487 |
eson | -1.456 |
ento | -1.455 |
Token | Amin |
Token ID | 39869 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
viol | +2.095 |
NPR | +1.978 |
containment | +1.917 |
IRO | +1.905 |
Carnegie | +1.865 |
Neg logit contributions | |
ommel | -1.890 |
reviewed | -1.733 |
eson | -1.520 |
atal | -1.513 |
ento | -1.498 |
Token | u |
Token ID | 84 |
Feature activation | +8.27e-02 |
Pos logit contributions | |
ommel | +2.260 |
reviewed | +2.100 |
atal | +1.853 |
ento | +1.819 |
eson | +1.810 |
Neg logit contributions | |
viol | -2.914 |
NPR | -2.752 |
containment | -2.707 |
IRO | -2.702 |
pees | -2.572 |
Token | , |
Token ID | 11 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
ommel | +0.059 |
reviewed | +0.055 |
atal | +0.048 |
eson | +0.047 |
ento | +0.047 |
Neg logit contributions | |
viol | -0.069 |
NPR | -0.064 |
containment | -0.063 |
IRO | -0.063 |
Carnegie | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
viol | +0.048 |
NPR | +0.045 |
containment | +0.044 |
IRO | +0.044 |
Carnegie | +0.042 |
Neg logit contributions | |
ommel | -0.047 |
reviewed | -0.044 |
atal | -0.039 |
eson | -0.038 |
ento | -0.038 |
Token | J |
Token ID | 449 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
ommel | +1.986 |
reviewed | +1.851 |
atal | +1.645 |
eson | +1.622 |
ento | +1.587 |
Neg logit contributions | |
viol | -1.924 |
NPR | -1.835 |
IRO | -1.783 |
containment | -1.756 |
pees | -1.679 |
Token | . |
Token ID | 13 |
Feature activation | +7.82e+00 |
Pos logit contributions | |
ommel | +1.457 |
reviewed | +1.328 |
atal | +1.199 |
eson | +1.173 |
ento | +1.155 |
Neg logit contributions | |
viol | -1.719 |
NPR | -1.595 |
containment | -1.595 |
IRO | -1.570 |
Carnegie | -1.522 |
Token | J |
Token ID | 41 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
ommel | +3.567 |
reviewed | +3.549 |
Thu | +2.842 |
Adams | +2.782 |
MAC | +2.754 |
Neg logit contributions | |
viol | -7.774 |
containment | -7.602 |
IRO | -7.112 |
pees | -7.104 |
NPR | -7.096 |
Token | . |
Token ID | 13 |
Feature activation | +6.64e+00 |
Pos logit contributions | |
ommel | +1.582 |
reviewed | +1.476 |
atal | +1.286 |
eson | +1.271 |
ento | +1.239 |
Neg logit contributions | |
viol | -2.050 |
NPR | -1.930 |
containment | -1.901 |
IRO | -1.889 |
Carnegie | -1.799 |
Token | B |
Token ID | 347 |
Feature activation | +4.58e+00 |
Pos logit contributions | |
reviewed | +3.999 |
ommel | +3.993 |
eson | +3.438 |
atal | +3.336 |
ür | +3.304 |
Neg logit contributions | |
viol | -5.765 |
containment | -5.532 |
IRO | -5.522 |
NPR | -5.472 |
pees | -5.129 |
Token | area |
Token ID | 20337 |
Feature activation | +4.99e+00 |
Pos logit contributions | |
ommel | +3.038 |
reviewed | +2.672 |
atal | +2.445 |
eson | +2.288 |
ourke | +2.271 |
Neg logit contributions | |
viol | -4.258 |
containment | -3.952 |
IRO | -3.898 |
NPR | -3.880 |
Sov | -3.720 |
Token | . |
Token ID | 13 |
Feature activation | +9.78e-01 |
Pos logit contributions | |
ommel | +3.324 |
reviewed | +3.039 |
atal | +2.693 |
eson | +2.634 |
ento | +2.586 |
Neg logit contributions | |
viol | -4.222 |
NPR | -4.100 |
IRO | -3.971 |
containment | -3.889 |
Carnegie | -3.796 |
Token | , |
Token ID | 11 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
ommel | +1.417 |
reviewed | +1.305 |
atal | +1.170 |
eson | +1.165 |
ento | +1.150 |
Neg logit contributions | |
viol | -1.425 |
NPR | -1.343 |
IRO | -1.311 |
containment | -1.297 |
Carnegie | -1.252 |
Token | K |
Token ID | 509 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
ommel | +0.211 |
reviewed | +0.196 |
atal | +0.175 |
eson | +0.173 |
ento | +0.172 |
Neg logit contributions | |
viol | -0.204 |
NPR | -0.191 |
containment | -0.186 |
IRO | -0.185 |
Carnegie | -0.178 |
Token | ISS |
Token ID | 16744 |
Feature activation | +4.75e+00 |
Pos logit contributions | |
ommel | +2.476 |
reviewed | +2.240 |
atal | +2.056 |
eson | +2.036 |
ento | +1.981 |
Neg logit contributions | |
viol | -2.595 |
NPR | -2.400 |
containment | -2.382 |
IRO | -2.305 |
Carnegie | -2.280 |
Token | Air |
Token ID | 3701 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
ommel | +3.714 |
reviewed | +3.414 |
atal | +3.097 |
eson | +3.083 |
ento | +3.031 |
Neg logit contributions | |
viol | -3.482 |
NPR | -3.302 |
IRO | -3.158 |
containment | -3.144 |
Carnegie | -3.068 |
Token | ch |
Token ID | 442 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
ommel | +3.015 |
reviewed | +2.777 |
atal | +2.501 |
eson | +2.488 |
ento | +2.462 |
Neg logit contributions | |
viol | -3.029 |
NPR | -2.856 |
IRO | -2.771 |
containment | -2.763 |
Carnegie | -2.687 |
Token | am |
Token ID | 321 |
Feature activation | +6.23e+00 |
Pos logit contributions | |
ommel | +0.329 |
reviewed | +0.305 |
atal | +0.273 |
eson | +0.269 |
ento | +0.267 |
Neg logit contributions | |
viol | -0.334 |
NPR | -0.311 |
containment | -0.305 |
IRO | -0.303 |
Carnegie | -0.291 |
Token | ois |
Token ID | 10924 |
Feature activation | +4.50e+00 |
Pos logit contributions | |
ommel | +4.199 |
reviewed | +3.739 |
atal | +3.440 |
eson | +3.349 |
ento | +3.290 |
Neg logit contributions | |
viol | -5.294 |
NPR | -5.103 |
containment | -4.968 |
IRO | -4.868 |
Carnegie | -4.788 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
ommel | +3.422 |
reviewed | +3.126 |
atal | +2.810 |
eson | +2.810 |
ento | +2.783 |
Neg logit contributions | |
viol | -3.426 |
NPR | -3.261 |
IRO | -3.134 |
containment | -3.110 |
Carnegie | -3.053 |
Token | a |
Token ID | 257 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ommel | +0.884 |
reviewed | +0.824 |
atal | +0.736 |
eson | +0.728 |
ento | +0.722 |
Neg logit contributions | |
viol | -0.831 |
NPR | -0.784 |
IRO | -0.758 |
containment | -0.757 |
Carnegie | -0.725 |
Token | large |
Token ID | 1588 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
ommel | +1.311 |
reviewed | +1.218 |
atal | +1.086 |
eson | +1.078 |
ento | +1.066 |
Neg logit contributions | |
viol | -1.280 |
NPR | -1.214 |
IRO | -1.171 |
containment | -1.166 |
Carnegie | -1.119 |
Token | pull |
Token ID | 2834 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ommel | +2.234 |
reviewed | +2.058 |
atal | +1.857 |
eson | +1.840 |
ento | +1.814 |
Neg logit contributions | |
viol | -2.108 |
NPR | -2.023 |
IRO | -1.939 |
containment | -1.910 |
pees | -1.845 |
Token | off |
Token ID | 572 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ommel | +1.632 |
reviewed | +1.540 |
atal | +1.404 |
eson | +1.389 |
ento | +1.380 |
Neg logit contributions | |
viol | -1.040 |
NPR | -0.978 |
IRO | -0.944 |
containment | -0.936 |
Carnegie | -0.884 |
Token | - |
Token ID | 12 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
viol | +1.370 |
NPR | +1.276 |
containment | +1.254 |
IRO | +1.247 |
Carnegie | +1.198 |
Neg logit contributions | |
ommel | -1.303 |
reviewed | -1.212 |
atal | -1.071 |
eson | -1.055 |
ento | -1.054 |
Token | the |
Token ID | 1169 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
ommel | +0.147 |
reviewed | +0.135 |
atal | +0.118 |
eson | +0.116 |
ento | +0.115 |
Neg logit contributions | |
viol | -0.197 |
NPR | -0.186 |
containment | -0.182 |
IRO | -0.182 |
Carnegie | -0.175 |
Token | - |
Token ID | 12 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
viol | +0.093 |
NPR | +0.087 |
containment | +0.085 |
IRO | +0.085 |
Carnegie | +0.082 |
Neg logit contributions | |
ommel | -0.082 |
reviewed | -0.076 |
atal | -0.067 |
eson | -0.066 |
ento | -0.066 |
Token | field |
Token ID | 3245 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
viol | +0.062 |
NPR | +0.059 |
containment | +0.058 |
IRO | +0.058 |
Carnegie | +0.056 |
Neg logit contributions | |
ommel | -0.040 |
reviewed | -0.037 |
atal | -0.031 |
eson | -0.031 |
ento | -0.031 |
Token | punishments |
Token ID | 35518 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
ommel | +1.757 |
reviewed | +1.633 |
atal | +1.455 |
ento | +1.429 |
eson | +1.426 |
Neg logit contributions | |
viol | -1.811 |
NPR | -1.729 |
IRO | -1.690 |
containment | -1.664 |
Carnegie | -1.612 |
Token | , |
Token ID | 11 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
ommel | +0.254 |
reviewed | +0.237 |
atal | +0.211 |
eson | +0.208 |
ento | +0.207 |
Neg logit contributions | |
viol | -0.246 |
NPR | -0.231 |
IRO | -0.225 |
containment | -0.225 |
Carnegie | -0.215 |
Token | Goodell |
Token ID | 41648 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
viol | +0.038 |
NPR | +0.035 |
containment | +0.034 |
IRO | +0.034 |
Carnegie | +0.033 |
Neg logit contributions | |
ommel | -0.041 |
reviewed | -0.038 |
atal | -0.034 |
eson | -0.034 |
ento | -0.033 |
Token | has |
Token ID | 468 |
Feature activation | -9.89e-01 |
Pos logit contributions | |
viol | +2.258 |
NPR | +2.061 |
containment | +2.033 |
IRO | +1.994 |
Carnegie | +1.943 |
Neg logit contributions | |
ommel | -2.486 |
reviewed | -2.316 |
atal | -2.073 |
eson | -2.047 |
ento | -2.043 |
Token | access |
Token ID | 1895 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
viol | +0.710 |
NPR | +0.660 |
containment | +0.643 |
IRO | +0.640 |
Carnegie | +0.614 |
Neg logit contributions | |
ommel | -0.817 |
reviewed | -0.764 |
atal | -0.686 |
eson | -0.677 |
ento | -0.675 |
Token | to |
Token ID | 284 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
viol | +0.278 |
NPR | +0.260 |
containment | +0.254 |
IRO | +0.252 |
Carnegie | +0.243 |
Neg logit contributions | |
ommel | -0.274 |
reviewed | -0.255 |
atal | -0.227 |
eson | -0.224 |
ento | -0.223 |
Token | cannot |
Token ID | 2314 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
viol | +0.844 |
NPR | +0.787 |
containment | +0.768 |
IRO | +0.763 |
Carnegie | +0.732 |
Neg logit contributions | |
ommel | -0.938 |
reviewed | -0.874 |
atal | -0.784 |
eson | -0.774 |
ento | -0.772 |
Token | f |
Token ID | 277 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
viol | +0.767 |
NPR | +0.706 |
containment | +0.693 |
IRO | +0.683 |
Carnegie | +0.660 |
Neg logit contributions | |
ommel | -0.869 |
reviewed | -0.812 |
atal | -0.725 |
ento | -0.720 |
eson | -0.718 |
Token | athom |
Token ID | 32910 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
viol | +1.631 |
NPR | +1.489 |
containment | +1.429 |
IRO | +1.421 |
Carnegie | +1.352 |
Neg logit contributions | |
ommel | -2.824 |
reviewed | -2.689 |
eson | -2.432 |
atal | -2.431 |
ento | -2.417 |
Token | why |
Token ID | 1521 |
Feature activation | -6.99e-01 |
Pos logit contributions | |
viol | +3.226 |
containment | +2.910 |
NPR | +2.883 |
IRO | +2.819 |
Carnegie | +2.770 |
Neg logit contributions | |
ommel | -3.786 |
reviewed | -3.551 |
atal | -3.190 |
eson | -3.157 |
ento | -3.145 |
Token | you |
Token ID | 345 |
Feature activation | -9.78e-01 |
Pos logit contributions | |
viol | +0.511 |
NPR | +0.476 |
containment | +0.464 |
IRO | +0.462 |
Carnegie | +0.443 |
Neg logit contributions | |
ommel | -0.569 |
reviewed | -0.531 |
atal | -0.476 |
eson | -0.470 |
ento | -0.468 |
Token | , |
Token ID | 11 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
viol | +0.737 |
NPR | +0.690 |
containment | +0.671 |
IRO | +0.669 |
Carnegie | +0.643 |
Neg logit contributions | |
ommel | -0.773 |
reviewed | -0.719 |
atal | -0.642 |
eson | -0.633 |
ento | -0.631 |
Token | someone |
Token ID | 2130 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ommel | +0.123 |
reviewed | +0.115 |
atal | +0.103 |
eson | +0.101 |
ento | +0.101 |
Neg logit contributions | |
viol | -0.113 |
NPR | -0.106 |
containment | -0.103 |
IRO | -0.102 |
Carnegie | -0.098 |
Token | who |
Token ID | 508 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
viol | +1.069 |
NPR | +0.998 |
containment | +0.977 |
IRO | +0.971 |
Carnegie | +0.940 |
Neg logit contributions | |
ommel | -1.003 |
reviewed | -0.928 |
atal | -0.824 |
eson | -0.814 |
ento | -0.810 |
Token | blocks |
Token ID | 7021 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
viol | +0.191 |
NPR | +0.178 |
containment | +0.173 |
IRO | +0.172 |
Carnegie | +0.166 |
Neg logit contributions | |
ommel | -0.203 |
reviewed | -0.189 |
atal | -0.169 |
eson | -0.167 |
ento | -0.166 |
Token | people |
Token ID | 661 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
ommel | +1.134 |
reviewed | +1.063 |
atal | +0.952 |
ento | +0.937 |
eson | +0.937 |
Neg logit contributions | |
viol | -1.035 |
NPR | -0.976 |
IRO | -0.944 |
containment | -0.944 |
Carnegie | -0.901 |
Token | for |
Token ID | 329 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
viol | +0.439 |
NPR | +0.409 |
containment | +0.402 |
IRO | +0.399 |
Carnegie | +0.387 |
Neg logit contributions | |
ommel | -0.396 |
reviewed | -0.366 |
atal | -0.322 |
eson | -0.318 |
ento | -0.317 |
Token | a |
Token ID | 257 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
viol | +1.465 |
NPR | +1.357 |
containment | +1.329 |
IRO | +1.316 |
Carnegie | +1.269 |
Neg logit contributions | |
ommel | -1.609 |
reviewed | -1.500 |
atal | -1.342 |
eson | -1.327 |
ento | -1.325 |
Token | send |
Token ID | 3758 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
ommel | +0.295 |
reviewed | +0.275 |
atal | +0.246 |
eson | +0.244 |
ento | +0.241 |
Neg logit contributions | |
viol | -0.277 |
NPR | -0.261 |
IRO | -0.253 |
containment | -0.252 |
Carnegie | -0.242 |
Token | and |
Token ID | 290 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
ommel | +1.051 |
reviewed | +0.972 |
atal | +0.870 |
eson | +0.854 |
ento | +0.847 |
Neg logit contributions | |
viol | -1.158 |
NPR | -1.099 |
containment | -1.067 |
IRO | -1.066 |
Carnegie | -1.027 |
Token | an |
Token ID | 281 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
viol | +2.625 |
NPR | +2.423 |
containment | +2.363 |
IRO | +2.319 |
Carnegie | +2.240 |
Neg logit contributions | |
ommel | -3.301 |
reviewed | -3.076 |
atal | -2.767 |
eson | -2.752 |
ento | -2.728 |
Token | a |
Token ID | 257 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
viol | +2.373 |
NPR | +2.199 |
containment | +2.127 |
IRO | +2.121 |
Carnegie | +2.016 |
Neg logit contributions | |
ommel | -3.031 |
reviewed | -2.827 |
atal | -2.548 |
eson | -2.527 |
ento | -2.509 |
Token | ck |
Token ID | 694 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
viol | +1.569 |
NPR | +1.460 |
containment | +1.426 |
IRO | +1.416 |
Carnegie | +1.361 |
Neg logit contributions | |
ommel | -1.613 |
reviewed | -1.509 |
atal | -1.332 |
ento | -1.314 |
eson | -1.313 |
Token | is |
Token ID | 318 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
ommel | +1.018 |
reviewed | +0.946 |
atal | +0.852 |
eson | +0.844 |
ento | +0.834 |
Neg logit contributions | |
viol | -0.994 |
NPR | -0.943 |
IRO | -0.914 |
containment | -0.911 |
Carnegie | -0.873 |
Token | simply |
Token ID | 2391 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
ommel | +2.473 |
reviewed | +2.311 |
atal | +2.065 |
eson | +2.053 |
ento | +2.000 |
Neg logit contributions | |
viol | -2.251 |
NPR | -2.136 |
IRO | -2.058 |
containment | -2.055 |
pees | -1.964 |
Token | the |
Token ID | 262 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ommel | +2.302 |
reviewed | +2.133 |
atal | +1.916 |
eson | +1.883 |
ento | +1.853 |
Neg logit contributions | |
viol | -2.158 |
NPR | -2.041 |
IRO | -1.972 |
containment | -1.956 |
pees | -1.873 |
Token | amount |
Token ID | 2033 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
ommel | +1.711 |
reviewed | +1.600 |
atal | +1.449 |
eson | +1.431 |
ento | +1.417 |
Neg logit contributions | |
viol | -1.376 |
NPR | -1.292 |
IRO | -1.249 |
containment | -1.237 |
Carnegie | -1.183 |
Token | of |
Token ID | 286 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
ommel | +1.648 |
reviewed | +1.524 |
atal | +1.339 |
eson | +1.313 |
ento | +1.303 |
Neg logit contributions | |
viol | -1.995 |
NPR | -1.904 |
IRO | -1.855 |
containment | -1.837 |
Carnegie | -1.780 |
Token | . |
Token ID | 13 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
viol | +1.853 |
NPR | +1.735 |
containment | +1.693 |
IRO | +1.683 |
Carnegie | +1.626 |
Neg logit contributions | |
ommel | -1.760 |
reviewed | -1.636 |
atal | -1.457 |
eson | -1.432 |
ento | -1.428 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
viol | +2.024 |
NPR | +1.902 |
IRO | +1.838 |
containment | +1.837 |
Carnegie | +1.778 |
Neg logit contributions | |
ommel | -2.080 |
reviewed | -1.935 |
atal | -1.730 |
eson | -1.706 |
ento | -1.695 |
Token | L |
Token ID | 43 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
ommel | +1.261 |
reviewed | +1.187 |
atal | +1.047 |
eson | +1.033 |
ento | +1.027 |
Neg logit contributions | |
viol | -1.275 |
NPR | -1.169 |
containment | -1.166 |
IRO | -1.142 |
Carnegie | -1.114 |
Token | ords |
Token ID | 3669 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
viol | +0.380 |
NPR | +0.356 |
containment | +0.347 |
IRO | +0.346 |
Carnegie | +0.333 |
Neg logit contributions | |
ommel | -0.362 |
reviewed | -0.337 |
atal | -0.299 |
eson | -0.294 |
ento | -0.293 |
Token | Brexit |
Token ID | 11435 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
viol | +1.120 |
NPR | +1.051 |
containment | +1.026 |
IRO | +1.023 |
Carnegie | +0.986 |
Neg logit contributions | |
ommel | -1.014 |
reviewed | -0.940 |
atal | -0.832 |
eson | -0.819 |
ento | -0.815 |
Token | reports |
Token ID | 3136 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
viol | +1.414 |
NPR | +1.314 |
containment | +1.288 |
IRO | +1.283 |
Carnegie | +1.234 |
Neg logit contributions | |
ommel | -1.401 |
reviewed | -1.311 |
atal | -1.168 |
ento | -1.146 |
eson | -1.146 |
Token | : |
Token ID | 25 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
ommel | +1.052 |
reviewed | +0.976 |
atal | +0.867 |
ento | +0.855 |
eson | +0.855 |
Neg logit contributions | |
viol | -1.039 |
NPR | -0.978 |
containment | -0.950 |
IRO | -0.950 |
Carnegie | -0.906 |
Token | QC |
Token ID | 36070 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
ommel | +1.457 |
reviewed | +1.353 |
atal | +1.200 |
eson | +1.186 |
ento | +1.179 |
Neg logit contributions | |
viol | -1.522 |
NPR | -1.428 |
containment | -1.395 |
IRO | -1.389 |
Carnegie | -1.328 |
Token | advises |
Token ID | 34138 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
viol | +0.029 |
NPR | +0.027 |
containment | +0.026 |
IRO | +0.026 |
Carnegie | +0.025 |
Neg logit contributions | |
ommel | -0.030 |
reviewed | -0.028 |
atal | -0.025 |
eson | -0.024 |
ento | -0.024 |
Token | making |
Token ID | 1642 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ommel | +2.268 |
reviewed | +2.082 |
atal | +1.845 |
eson | +1.842 |
ento | +1.820 |
Neg logit contributions | |
viol | -2.372 |
NPR | -2.261 |
IRO | -2.190 |
containment | -2.187 |
pees | -2.104 |
Token | file |
Token ID | 2393 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
ommel | +1.194 |
reviewed | +1.109 |
atal | +0.985 |
eson | +0.972 |
ento | +0.968 |
Neg logit contributions | |
viol | -1.228 |
NPR | -1.163 |
IRO | -1.128 |
containment | -1.125 |
Carnegie | -1.076 |
Token | away |
Token ID | 1497 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
ommel | +0.269 |
reviewed | +0.251 |
atal | +0.224 |
eson | +0.221 |
ento | +0.220 |
Neg logit contributions | |
viol | -0.253 |
NPR | -0.236 |
containment | -0.230 |
IRO | -0.230 |
Carnegie | -0.220 |
Token | prospective |
Token ID | 17530 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
viol | +0.358 |
NPR | +0.333 |
containment | +0.325 |
IRO | +0.323 |
Carnegie | +0.311 |
Neg logit contributions | |
ommel | -0.380 |
reviewed | -0.354 |
atal | -0.316 |
eson | -0.312 |
ento | -0.311 |
Token | parents |
Token ID | 3397 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
viol | +0.213 |
NPR | +0.197 |
containment | +0.191 |
IRO | +0.190 |
Carnegie | +0.183 |
Neg logit contributions | |
ommel | -0.260 |
reviewed | -0.243 |
atal | -0.219 |
eson | -0.216 |
ento | -0.215 |
Token | if |
Token ID | 611 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
ommel | +1.541 |
reviewed | +1.436 |
atal | +1.273 |
eson | +1.250 |
ento | +1.247 |
Neg logit contributions | |
viol | -1.699 |
NPR | -1.612 |
IRO | -1.582 |
containment | -1.573 |
Carnegie | -1.509 |
Token | they |
Token ID | 484 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ommel | +0.249 |
reviewed | +0.233 |
atal | +0.209 |
eson | +0.206 |
ento | +0.205 |
Neg logit contributions | |
viol | -0.226 |
NPR | -0.211 |
containment | -0.205 |
IRO | -0.205 |
Carnegie | -0.196 |
Token | objected |
Token ID | 29811 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
ommel | +0.994 |
reviewed | +0.932 |
atal | +0.837 |
ento | +0.822 |
eson | +0.822 |
Neg logit contributions | |
viol | -0.906 |
NPR | -0.857 |
IRO | -0.837 |
containment | -0.830 |
Carnegie | -0.792 |
Token | to |
Token ID | 284 |
Feature activation | -7.34e-01 |
Pos logit contributions | |
ommel | +0.185 |
reviewed | +0.172 |
atal | +0.153 |
eson | +0.151 |
ento | +0.150 |
Neg logit contributions | |
viol | -0.189 |
NPR | -0.176 |
containment | -0.172 |
IRO | -0.171 |
Carnegie | -0.165 |
Token | their |
Token ID | 511 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
viol | +0.514 |
NPR | +0.472 |
containment | +0.463 |
IRO | +0.457 |
Carnegie | +0.442 |
Neg logit contributions | |
ommel | -0.625 |
reviewed | -0.584 |
atal | -0.525 |
ento | -0.518 |
eson | -0.518 |
Token | sexual |
Token ID | 3206 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
viol | +1.243 |
NPR | +1.137 |
containment | +1.122 |
IRO | +1.098 |
Carnegie | +1.065 |
Neg logit contributions | |
ommel | -1.497 |
reviewed | -1.396 |
atal | -1.255 |
ento | -1.241 |
eson | -1.237 |
Token | orientation |
Token ID | 12852 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
ommel | +1.774 |
reviewed | +1.633 |
atal | +1.462 |
eson | +1.438 |
ento | +1.419 |
Neg logit contributions | |
viol | -2.010 |
NPR | -1.937 |
IRO | -1.880 |
containment | -1.842 |
Carnegie | -1.804 |
Token | or |
Token ID | 393 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
ommel | +0.633 |
reviewed | +0.586 |
atal | +0.518 |
eson | +0.510 |
ento | +0.508 |
Neg logit contributions | |
viol | -0.711 |
NPR | -0.669 |
containment | -0.652 |
IRO | -0.651 |
Carnegie | -0.627 |
Token | I |
Token ID | 314 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ommel | +0.783 |
reviewed | +0.731 |
atal | +0.654 |
eson | +0.645 |
ento | +0.642 |
Neg logit contributions | |
viol | -0.729 |
NPR | -0.682 |
containment | -0.664 |
IRO | -0.662 |
Carnegie | -0.635 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
ommel | +1.242 |
reviewed | +1.160 |
atal | +1.032 |
eson | +1.018 |
ento | +1.013 |
Neg logit contributions | |
viol | -1.198 |
NPR | -1.124 |
containment | -1.096 |
IRO | -1.093 |
Carnegie | -1.041 |
Token | 't |
Token ID | 470 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
ommel | +0.985 |
reviewed | +0.913 |
atal | +0.802 |
eson | +0.790 |
ento | +0.785 |
Neg logit contributions | |
viol | -1.153 |
NPR | -1.083 |
containment | -1.061 |
IRO | -1.057 |
Carnegie | -1.017 |
Token | have |
Token ID | 423 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
viol | +0.448 |
NPR | +0.415 |
containment | +0.407 |
IRO | +0.404 |
Carnegie | +0.392 |
Neg logit contributions | |
ommel | -0.458 |
reviewed | -0.427 |
atal | -0.380 |
ento | -0.376 |
eson | -0.374 |
Token | sworn |
Token ID | 20990 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
viol | +1.728 |
NPR | +1.573 |
containment | +1.543 |
IRO | +1.526 |
Carnegie | +1.488 |
Neg logit contributions | |
ommel | -2.175 |
reviewed | -2.025 |
atal | -1.849 |
eson | -1.823 |
ento | -1.820 |
Token | this |
Token ID | 428 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
viol | +0.092 |
NPR | +0.085 |
containment | +0.083 |
IRO | +0.083 |
Carnegie | +0.080 |
Neg logit contributions | |
ommel | -0.107 |
reviewed | -0.100 |
atal | -0.090 |
eson | -0.089 |
ento | -0.088 |
Token | was |
Token ID | 373 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
viol | +0.023 |
NPR | +0.021 |
containment | +0.021 |
IRO | +0.020 |
Carnegie | +0.020 |
Neg logit contributions | |
ommel | -0.024 |
reviewed | -0.022 |
atal | -0.020 |
ento | -0.020 |
eson | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
viol | +1.690 |
NPR | +1.540 |
containment | +1.526 |
IRO | +1.480 |
Carnegie | +1.469 |
Neg logit contributions | |
ommel | -2.041 |
reviewed | -1.895 |
atal | -1.706 |
ento | -1.688 |
eson | -1.686 |
Token | conversation |
Token ID | 5273 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
ommel | +0.820 |
reviewed | +0.765 |
atal | +0.682 |
eson | +0.675 |
ento | +0.667 |
Neg logit contributions | |
viol | -0.829 |
NPR | -0.785 |
IRO | -0.761 |
containment | -0.758 |
Carnegie | -0.725 |
Token | I |
Token ID | 314 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
ommel | +0.151 |
reviewed | +0.140 |
atal | +0.125 |
eson | +0.124 |
ento | +0.123 |
Neg logit contributions | |
viol | -0.147 |
NPR | -0.138 |
containment | -0.134 |
IRO | -0.134 |
Carnegie | -0.128 |
Token | would |
Token ID | 561 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
ommel | +1.407 |
reviewed | +1.301 |
atal | +1.164 |
eson | +1.140 |
ento | +1.137 |
Neg logit contributions | |
viol | -1.527 |
NPR | -1.441 |
IRO | -1.404 |
containment | -1.398 |
Civ | -1.330 |
Token | 's |
Token ID | 338 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
viol | +0.737 |
NPR | +0.688 |
containment | +0.670 |
IRO | +0.667 |
Carnegie | +0.642 |
Neg logit contributions | |
ommel | -0.789 |
reviewed | -0.737 |
atal | -0.659 |
eson | -0.650 |
ento | -0.647 |
Token | handed |
Token ID | 10158 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
viol | +0.938 |
NPR | +0.876 |
containment | +0.851 |
IRO | +0.849 |
Carnegie | +0.812 |
Neg logit contributions | |
ommel | -1.076 |
reviewed | -1.006 |
atal | -0.904 |
eson | -0.892 |
ento | -0.889 |
Token | from |
Token ID | 422 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
viol | +0.418 |
NPR | +0.392 |
containment | +0.386 |
IRO | +0.383 |
Carnegie | +0.369 |
Neg logit contributions | |
ommel | -0.327 |
reviewed | -0.301 |
atal | -0.263 |
ento | -0.260 |
eson | -0.259 |
Token | generation |
Token ID | 5270 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ommel | +2.013 |
reviewed | +1.900 |
atal | +1.740 |
eson | +1.719 |
ento | +1.712 |
Neg logit contributions | |
viol | -1.150 |
NPR | -1.063 |
IRO | -1.017 |
containment | -1.010 |
Carnegie | -0.944 |
Token | to |
Token ID | 284 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
ommel | +1.295 |
reviewed | +1.202 |
atal | +1.064 |
eson | +1.048 |
ento | +1.040 |
Neg logit contributions | |
viol | -1.374 |
NPR | -1.294 |
containment | -1.260 |
IRO | -1.260 |
Carnegie | -1.208 |
Token | generation |
Token ID | 5270 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
viol | +1.444 |
NPR | +1.290 |
containment | +1.247 |
IRO | +1.233 |
Carnegie | +1.157 |
Neg logit contributions | |
ommel | -3.086 |
reviewed | -2.927 |
atal | -2.696 |
eson | -2.667 |
ento | -2.662 |
Token | . |
Token ID | 13 |
Feature activation | +7.45e-01 |
Pos logit contributions | |
viol | +0.036 |
NPR | +0.034 |
IRO | +0.033 |
containment | +0.033 |
Carnegie | +0.032 |
Neg logit contributions | |
ommel | -0.031 |
reviewed | -0.029 |
atal | -0.025 |
eson | -0.025 |
ento | -0.025 |
Token | Hunting |
Token ID | 22059 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
ommel | +0.595 |
reviewed | +0.559 |
atal | +0.498 |
eson | +0.492 |
ento | +0.489 |
Neg logit contributions | |
viol | -0.551 |
NPR | -0.513 |
containment | -0.504 |
IRO | -0.499 |
Carnegie | -0.474 |
Token | and |
Token ID | 290 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
viol | +2.237 |
NPR | +2.053 |
containment | +2.027 |
IRO | +2.008 |
Carnegie | +1.942 |
Neg logit contributions | |
ommel | -2.191 |
reviewed | -2.048 |
atal | -1.829 |
ento | -1.796 |
eson | -1.796 |
Token | shooting |
Token ID | 4395 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
viol | +1.827 |
NPR | +1.683 |
containment | +1.665 |
IRO | +1.639 |
Carnegie | +1.597 |
Neg logit contributions | |
ommel | -1.824 |
reviewed | -1.698 |
atal | -1.515 |
eson | -1.493 |
ento | -1.486 |
Token | are |
Token ID | 389 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
viol | +0.077 |
NPR | +0.073 |
containment | +0.071 |
IRO | +0.071 |
Carnegie | +0.068 |
Neg logit contributions | |
ommel | -0.078 |
reviewed | -0.073 |
atal | -0.065 |
eson | -0.064 |
ento | -0.063 |
Token | Democratic |
Token ID | 4390 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
ommel | +0.546 |
reviewed | +0.509 |
atal | +0.456 |
eson | +0.449 |
ento | +0.447 |
Neg logit contributions | |
viol | -0.519 |
NPR | -0.487 |
IRO | -0.473 |
containment | -0.472 |
Carnegie | -0.451 |
Token | Front |
Token ID | 8880 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
ommel | +1.993 |
reviewed | +1.877 |
atal | +1.719 |
eson | +1.685 |
ento | +1.684 |
Neg logit contributions | |
viol | -1.418 |
NPR | -1.322 |
IRO | -1.273 |
containment | -1.269 |
Carnegie | -1.190 |
Token | for |
Token ID | 329 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
ommel | +2.436 |
reviewed | +2.255 |
atal | +1.995 |
ento | +1.936 |
eson | +1.934 |
Neg logit contributions | |
viol | -2.990 |
NPR | -2.837 |
containment | -2.770 |
IRO | -2.768 |
Carnegie | -2.635 |
Token | the |
Token ID | 262 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
ommel | +2.597 |
reviewed | +2.432 |
atal | +2.189 |
eson | +2.143 |
ento | +2.131 |
Neg logit contributions | |
viol | -2.672 |
NPR | -2.522 |
IRO | -2.460 |
containment | -2.420 |
Carnegie | -2.287 |
Token | Re |
Token ID | 797 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
viol | +1.529 |
NPR | +1.409 |
containment | +1.388 |
IRO | +1.364 |
Carnegie | +1.335 |
Neg logit contributions | |
ommel | -1.890 |
reviewed | -1.757 |
atal | -1.574 |
eson | -1.568 |
ento | -1.561 |
Token | un |
Token ID | 403 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
ommel | +1.868 |
reviewed | +1.744 |
atal | +1.595 |
eson | +1.556 |
ento | +1.543 |
Neg logit contributions | |
viol | -1.695 |
NPR | -1.580 |
IRO | -1.529 |
containment | -1.516 |
Carnegie | -1.446 |
Token | ification |
Token ID | 2649 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
viol | +2.196 |
NPR | +2.048 |
containment | +2.016 |
IRO | +1.986 |
Carnegie | +1.930 |
Neg logit contributions | |
ommel | -2.425 |
reviewed | -2.262 |
atal | -1.999 |
ento | -1.990 |
eson | -1.986 |
Token | of |
Token ID | 286 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
viol | +1.012 |
NPR | +0.946 |
containment | +0.927 |
IRO | +0.926 |
Carnegie | +0.897 |
Neg logit contributions | |
ommel | -0.929 |
reviewed | -0.859 |
atal | -0.758 |
eson | -0.755 |
ento | -0.748 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
viol | +0.127 |
NPR | +0.117 |
containment | +0.114 |
IRO | +0.114 |
Carnegie | +0.110 |
Neg logit contributions | |
ommel | -0.157 |
reviewed | -0.146 |
atal | -0.132 |
eson | -0.131 |
ento | -0.130 |
Token | Father |
Token ID | 9190 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
viol | +0.898 |
NPR | +0.826 |
containment | +0.807 |
IRO | +0.803 |
Carnegie | +0.779 |
Neg logit contributions | |
ommel | -1.108 |
reviewed | -1.035 |
atal | -0.929 |
eson | -0.923 |
ento | -0.917 |
Token | land |
Token ID | 1044 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
ommel | +0.256 |
reviewed | +0.221 |
atal | +0.169 |
eson | +0.163 |
ento | +0.162 |
Neg logit contributions | |
viol | -0.749 |
NPR | -0.717 |
containment | -0.705 |
IRO | -0.704 |
Carnegie | -0.687 |
Token | Catalonia |
Token ID | 33859 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ommel | +1.007 |
reviewed | +0.934 |
atal | +0.833 |
eson | +0.822 |
ento | +0.818 |
Neg logit contributions | |
viol | -1.002 |
NPR | -0.949 |
IRO | -0.922 |
containment | -0.917 |
Carnegie | -0.878 |
Token | . |
Token ID | 13 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
viol | +1.349 |
NPR | +1.263 |
containment | +1.236 |
IRO | +1.230 |
Carnegie | +1.187 |
Neg logit contributions | |
ommel | -1.203 |
reviewed | -1.120 |
atal | -0.984 |
eson | -0.975 |
ento | -0.968 |
Token | After |
Token ID | 2293 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
viol | +0.033 |
NPR | +0.031 |
containment | +0.030 |
IRO | +0.030 |
Carnegie | +0.029 |
Neg logit contributions | |
ommel | -0.034 |
reviewed | -0.031 |
atal | -0.028 |
eson | -0.027 |
ento | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
viol | +0.734 |
NPR | +0.681 |
containment | +0.669 |
IRO | +0.661 |
Carnegie | +0.637 |
Neg logit contributions | |
ommel | -0.815 |
reviewed | -0.762 |
atal | -0.684 |
eson | -0.674 |
ento | -0.672 |
Token | Ten |
Token ID | 9368 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
viol | +0.237 |
NPR | +0.221 |
containment | +0.216 |
IRO | +0.215 |
Carnegie | +0.206 |
Neg logit contributions | |
ommel | -0.254 |
reviewed | -0.237 |
atal | -0.212 |
eson | -0.209 |
ento | -0.208 |
Token | - |
Token ID | 12 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
viol | +0.398 |
NPR | +0.374 |
containment | +0.365 |
IRO | +0.364 |
Carnegie | +0.353 |
Neg logit contributions | |
ommel | -0.318 |
reviewed | -0.292 |
atal | -0.252 |
eson | -0.251 |
ento | -0.248 |
Token | Day |
Token ID | 12393 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
viol | +0.360 |
NPR | +0.340 |
containment | +0.331 |
IRO | +0.331 |
Carnegie | +0.319 |
Neg logit contributions | |
ommel | -0.283 |
reviewed | -0.261 |
atal | -0.227 |
eson | -0.225 |
ento | -0.223 |
Token | War |
Token ID | 1810 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
viol | +0.753 |
NPR | +0.694 |
containment | +0.675 |
IRO | +0.667 |
Carnegie | +0.640 |
Neg logit contributions | |
ommel | -1.107 |
reviewed | -1.046 |
atal | -0.948 |
eson | -0.939 |
ento | -0.934 |
Token | , |
Token ID | 11 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
ommel | +0.152 |
reviewed | +0.141 |
atal | +0.125 |
eson | +0.123 |
ento | +0.122 |
Neg logit contributions | |
viol | -0.166 |
NPR | -0.155 |
containment | -0.152 |
IRO | -0.151 |
Carnegie | -0.146 |
Token | a |
Token ID | 257 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
viol | +0.394 |
NPR | +0.365 |
containment | +0.359 |
IRO | +0.354 |
Carnegie | +0.343 |
Neg logit contributions | |
ommel | -0.416 |
reviewed | -0.390 |
atal | -0.348 |
eson | -0.343 |
ento | -0.342 |
Token | referendum |
Token ID | 11467 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
viol | +0.200 |
NPR | +0.187 |
containment | +0.182 |
IRO | +0.182 |
Carnegie | +0.174 |
Neg logit contributions | |
ommel | -0.223 |
reviewed | -0.208 |
atal | -0.187 |
eson | -0.184 |
ento | -0.184 |
Token | law |
Token ID | 1099 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ommel | +2.346 |
reviewed | +2.181 |
atal | +1.939 |
eson | +1.926 |
ento | +1.908 |
Neg logit contributions | |
viol | -2.388 |
NPR | -2.283 |
IRO | -2.203 |
containment | -2.189 |
Carnegie | -2.091 |
Token | and |
Token ID | 290 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
viol | +0.880 |
NPR | +0.826 |
containment | +0.805 |
IRO | +0.804 |
Carnegie | +0.773 |
Neg logit contributions | |
ommel | -0.843 |
reviewed | -0.784 |
atal | -0.695 |
eson | -0.685 |
ento | -0.682 |
Token | has |
Token ID | 468 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ommel | +0.359 |
reviewed | +0.336 |
atal | +0.299 |
eson | +0.296 |
ento | +0.295 |
Neg logit contributions | |
viol | -0.339 |
NPR | -0.324 |
IRO | -0.313 |
containment | -0.312 |
Carnegie | -0.298 |
Token | the |
Token ID | 262 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
viol | +1.060 |
NPR | +0.980 |
containment | +0.957 |
IRO | +0.953 |
Carnegie | +0.915 |
Neg logit contributions | |
ommel | -1.209 |
reviewed | -1.126 |
atal | -1.010 |
eson | -0.997 |
ento | -0.994 |
Token | right |
Token ID | 826 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
ommel | +0.391 |
reviewed | +0.368 |
atal | +0.332 |
eson | +0.329 |
ento | +0.327 |
Neg logit contributions | |
viol | -0.305 |
NPR | -0.285 |
containment | -0.275 |
IRO | -0.275 |
Carnegie | -0.262 |
Token | to |
Token ID | 284 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
viol | +0.043 |
NPR | +0.040 |
containment | +0.039 |
IRO | +0.039 |
Carnegie | +0.038 |
Neg logit contributions | |
ommel | -0.041 |
reviewed | -0.038 |
atal | -0.034 |
eson | -0.033 |
ento | -0.033 |
Token | the |
Token ID | 262 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
viol | +1.157 |
NPR | +1.073 |
containment | +1.047 |
IRO | +1.040 |
Carnegie | +0.999 |
Neg logit contributions | |
ommel | -1.341 |
reviewed | -1.254 |
atal | -1.125 |
eson | -1.110 |
ento | -1.106 |
Token | equal |
Token ID | 4961 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
viol | +1.199 |
NPR | +1.112 |
containment | +1.094 |
IRO | +1.080 |
Carnegie | +1.048 |
Neg logit contributions | |
ommel | -1.135 |
reviewed | -1.053 |
atal | -0.929 |
eson | -0.918 |
ento | -0.917 |
Token | protection |
Token ID | 4800 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
viol | +1.845 |
NPR | +1.693 |
containment | +1.668 |
IRO | +1.640 |
Carnegie | +1.579 |
Neg logit contributions | |
ommel | -2.192 |
reviewed | -2.053 |
atal | -1.839 |
eson | -1.815 |
ento | -1.811 |
Token | and |
Token ID | 290 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
viol | +0.423 |
NPR | +0.396 |
containment | +0.388 |
IRO | +0.386 |
Carnegie | +0.372 |
Neg logit contributions | |
ommel | -0.375 |
reviewed | -0.346 |
atal | -0.306 |
eson | -0.302 |
ento | -0.301 |
Token | equal |
Token ID | 4961 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
ommel | +1.707 |
reviewed | +1.582 |
atal | +1.415 |
eson | +1.402 |
ento | +1.392 |
Neg logit contributions | |
viol | -1.677 |
NPR | -1.595 |
IRO | -1.549 |
containment | -1.524 |
pees | -1.488 |
Token | are |
Token ID | 389 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
ommel | +0.460 |
reviewed | +0.430 |
atal | +0.387 |
eson | +0.382 |
ento | +0.381 |
Neg logit contributions | |
viol | -0.385 |
NPR | -0.358 |
containment | -0.348 |
IRO | -0.347 |
Carnegie | -0.332 |
Token | not |
Token ID | 407 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
viol | +2.292 |
NPR | +2.099 |
containment | +2.071 |
IRO | +2.017 |
Carnegie | +1.980 |
Neg logit contributions | |
ommel | -2.564 |
reviewed | -2.386 |
atal | -2.137 |
eson | -2.110 |
ento | -2.104 |
Token | too |
Token ID | 1165 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
viol | +2.739 |
containment | +2.456 |
NPR | +2.431 |
IRO | +2.378 |
Carnegie | +2.348 |
Neg logit contributions | |
ommel | -3.423 |
reviewed | -3.281 |
atal | -2.882 |
MAC | -2.876 |
ento | -2.875 |
Token | obvious |
Token ID | 3489 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
viol | +0.697 |
NPR | +0.651 |
containment | +0.638 |
IRO | +0.634 |
Carnegie | +0.613 |
Neg logit contributions | |
ommel | -0.643 |
reviewed | -0.599 |
atal | -0.522 |
ento | -0.518 |
eson | -0.516 |
Token | ( |
Token ID | 357 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
viol | +1.351 |
NPR | +1.268 |
IRO | +1.226 |
containment | +1.224 |
Carnegie | +1.193 |
Neg logit contributions | |
ommel | -1.326 |
reviewed | -1.241 |
atal | -1.097 |
eson | -1.093 |
ento | -1.089 |
Token | as |
Token ID | 292 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
viol | +1.824 |
NPR | +1.754 |
containment | +1.659 |
IRO | +1.656 |
Carnegie | +1.584 |
Neg logit contributions | |
ommel | -2.207 |
reviewed | -2.020 |
atal | -1.850 |
eson | -1.837 |
ento | -1.832 |
Token | opposed |
Token ID | 6886 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
viol | +3.688 |
NPR | +3.333 |
IRO | +3.165 |
containment | +3.113 |
Carnegie | +2.962 |
Neg logit contributions | |
ommel | -5.475 |
reviewed | -5.171 |
atal | -4.685 |
eson | -4.681 |
MAC | -4.617 |
Token | to |
Token ID | 284 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
viol | +1.967 |
NPR | +1.831 |
containment | +1.778 |
IRO | +1.778 |
Carnegie | +1.727 |
Neg logit contributions | |
ommel | -2.218 |
reviewed | -2.048 |
atal | -1.857 |
ento | -1.834 |
eson | -1.815 |
Token | back |
Token ID | 736 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
viol | +0.964 |
NPR | +0.899 |
containment | +0.878 |
IRO | +0.872 |
Carnegie | +0.838 |
Neg logit contributions | |
ommel | -1.027 |
reviewed | -0.958 |
atal | -0.857 |
eson | -0.846 |
ento | -0.844 |
Token | in |
Token ID | 287 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
viol | +2.457 |
NPR | +2.289 |
containment | +2.224 |
IRO | +2.221 |
Carnegie | +2.132 |
Neg logit contributions | |
ommel | -2.791 |
reviewed | -2.584 |
atal | -2.343 |
eson | -2.310 |
ento | -2.302 |
Token | the |
Token ID | 262 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
viol | +1.456 |
NPR | +1.348 |
containment | +1.313 |
IRO | +1.305 |
Carnegie | +1.266 |
Neg logit contributions | |
ommel | -1.895 |
reviewed | -1.768 |
atal | -1.603 |
ento | -1.590 |
eson | -1.579 |
Token | occasion |
Token ID | 6695 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
viol | +1.042 |
NPR | +0.970 |
containment | +0.944 |
IRO | +0.941 |
Carnegie | +0.901 |
Neg logit contributions | |
ommel | -1.224 |
reviewed | -1.146 |
atal | -1.030 |
eson | -1.017 |
ento | -1.013 |
Token | again |
Token ID | 757 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ommel | +1.557 |
reviewed | +1.451 |
atal | +1.278 |
eson | +1.262 |
ento | +1.253 |
Neg logit contributions | |
viol | -1.709 |
NPR | -1.614 |
IRO | -1.573 |
containment | -1.571 |
Carnegie | -1.506 |
Token | to |
Token ID | 284 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ommel | +1.119 |
reviewed | +1.039 |
atal | +0.918 |
eson | +0.906 |
ento | +0.899 |
Neg logit contributions | |
viol | -1.230 |
NPR | -1.162 |
IRO | -1.136 |
containment | -1.130 |
Carnegie | -1.083 |
Token | advocate |
Token ID | 12811 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ommel | +0.980 |
reviewed | +0.916 |
atal | +0.826 |
eson | +0.814 |
ento | +0.810 |
Neg logit contributions | |
viol | -0.824 |
NPR | -0.771 |
IRO | -0.750 |
containment | -0.746 |
Carnegie | -0.710 |
Token | closer |
Token ID | 5699 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
viol | +1.276 |
NPR | +1.183 |
containment | +1.171 |
IRO | +1.152 |
Carnegie | +1.117 |
Neg logit contributions | |
ommel | -1.270 |
reviewed | -1.184 |
atal | -1.056 |
ento | -1.037 |
eson | -1.037 |
Token | Turkish |
Token ID | 9663 |
Feature activation | -6.44e+00 |
Pos logit contributions | |
viol | +1.147 |
containment | +1.035 |
NPR | +1.021 |
IRO | +0.980 |
Carnegie | +0.962 |
Neg logit contributions | |
ommel | -1.874 |
reviewed | -1.769 |
atal | -1.612 |
ento | -1.600 |
eson | -1.597 |
Token | ties |
Token ID | 8470 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
viol | +4.101 |
containment | +3.863 |
NPR | +3.627 |
Carnegie | +3.504 |
IRO | +3.494 |
Neg logit contributions | |
ommel | -5.729 |
reviewed | -5.453 |
atal | -4.850 |
ento | -4.820 |
eson | -4.764 |
Token | to |
Token ID | 284 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
viol | +2.145 |
NPR | +1.993 |
containment | +1.963 |
IRO | +1.926 |
Carnegie | +1.879 |
Neg logit contributions | |
ommel | -1.962 |
reviewed | -1.808 |
atal | -1.607 |
ento | -1.579 |
eson | -1.577 |
Token | Israel |
Token ID | 2692 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
viol | +1.856 |
NPR | +1.725 |
containment | +1.710 |
IRO | +1.678 |
Carnegie | +1.644 |
Neg logit contributions | |
ommel | -1.576 |
reviewed | -1.455 |
atal | -1.283 |
eson | -1.261 |
ento | -1.253 |
Token | , |
Token ID | 11 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
viol | +0.977 |
NPR | +0.915 |
containment | +0.895 |
IRO | +0.890 |
Carnegie | +0.858 |
Neg logit contributions | |
ommel | -0.912 |
reviewed | -0.847 |
atal | -0.750 |
eson | -0.740 |
ento | -0.737 |
Token | arguing |
Token ID | 11810 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ommel | +1.447 |
reviewed | +1.353 |
atal | +1.210 |
eson | +1.194 |
ento | +1.185 |
Neg logit contributions | |
viol | -1.315 |
NPR | -1.242 |
IRO | -1.215 |
containment | -1.196 |
pees | -1.143 |
Token | standard |
Token ID | 3210 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
viol | +1.697 |
NPR | +1.573 |
containment | +1.553 |
IRO | +1.527 |
Carnegie | +1.479 |
Neg logit contributions | |
ommel | -1.907 |
reviewed | -1.783 |
atal | -1.596 |
eson | -1.582 |
ento | -1.580 |
Token | conditions |
Token ID | 3403 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
viol | +1.791 |
NPR | +1.664 |
containment | +1.644 |
IRO | +1.610 |
Carnegie | +1.562 |
Neg logit contributions | |
ommel | -1.942 |
reviewed | -1.813 |
atal | -1.620 |
ento | -1.605 |
eson | -1.603 |
Token | ( |
Token ID | 357 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
viol | +1.114 |
NPR | +1.039 |
containment | +1.021 |
IRO | +1.007 |
Carnegie | +0.974 |
Neg logit contributions | |
ommel | -1.061 |
reviewed | -0.987 |
atal | -0.875 |
eson | -0.868 |
ento | -0.861 |
Token | a |
Token ID | 64 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
viol | +1.664 |
NPR | +1.586 |
IRO | +1.543 |
containment | +1.538 |
Carnegie | +1.483 |
Neg logit contributions | |
ommel | -1.279 |
reviewed | -1.158 |
atal | -1.017 |
eson | -1.007 |
ento | -1.005 |
Token | 12 |
Token ID | 1105 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
viol | +1.178 |
NPR | +1.099 |
containment | +1.073 |
IRO | +1.068 |
Carnegie | +1.026 |
Neg logit contributions | |
ommel | -1.257 |
reviewed | -1.172 |
atal | -1.047 |
eson | -1.034 |
ento | -1.031 |
Token | h |
Token ID | 289 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
viol | +1.606 |
NPR | +1.504 |
containment | +1.472 |
IRO | +1.459 |
Carnegie | +1.412 |
Neg logit contributions | |
ommel | -1.481 |
reviewed | -1.375 |
atal | -1.212 |
ento | -1.198 |
eson | -1.197 |
Token | light |
Token ID | 1657 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
viol | +4.140 |
NPR | +3.856 |
containment | +3.842 |
IRO | +3.702 |
Carnegie | +3.605 |
Neg logit contributions | |
ommel | -4.997 |
reviewed | -4.745 |
ento | -4.223 |
eson | -4.210 |
atal | -4.192 |
Token | / |
Token ID | 14 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
viol | +0.425 |
NPR | +0.398 |
containment | +0.389 |
IRO | +0.387 |
Carnegie | +0.372 |
Neg logit contributions | |
ommel | -0.420 |
reviewed | -0.391 |
atal | -0.346 |
eson | -0.342 |
ento | -0.341 |
Token | dark |
Token ID | 21953 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
viol | +1.844 |
NPR | +1.761 |
containment | +1.705 |
IRO | +1.703 |
Carnegie | +1.641 |
Neg logit contributions | |
ommel | -1.414 |
reviewed | -1.286 |
eson | -1.111 |
atal | -1.109 |
ento | -1.100 |
Token | cycle |
Token ID | 6772 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
viol | +0.313 |
NPR | +0.293 |
IRO | +0.285 |
containment | +0.285 |
Carnegie | +0.274 |
Neg logit contributions | |
ommel | -0.314 |
reviewed | -0.292 |
atal | -0.261 |
eson | -0.258 |
ento | -0.256 |
Token | with |
Token ID | 351 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
viol | +2.854 |
NPR | +2.681 |
containment | +2.647 |
IRO | +2.602 |
Carnegie | +2.515 |
Neg logit contributions | |
ommel | -2.401 |
reviewed | -2.220 |
atal | -1.931 |
eson | -1.925 |
ento | -1.910 |
Token | a |
Token ID | 257 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
viol | +0.791 |
NPR | +0.735 |
containment | +0.718 |
IRO | +0.713 |
Carnegie | +0.689 |
Neg logit contributions | |
ommel | -0.879 |
reviewed | -0.822 |
atal | -0.737 |
eson | -0.727 |
ento | -0.724 |
Token | Brazilian |
Token ID | 17036 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
ommel | +1.973 |
reviewed | +1.845 |
atal | +1.646 |
eson | +1.636 |
ento | +1.631 |
Neg logit contributions | |
viol | -1.804 |
NPR | -1.700 |
IRO | -1.652 |
containment | -1.639 |
Carnegie | -1.562 |
Token | prosecutor |
Token ID | 13683 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ommel | +1.791 |
reviewed | +1.665 |
atal | +1.502 |
eson | +1.483 |
ento | +1.479 |
Neg logit contributions | |
viol | -1.535 |
NPR | -1.446 |
IRO | -1.398 |
containment | -1.397 |
pees | -1.340 |
Token | that |
Token ID | 326 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
ommel | +0.105 |
reviewed | +0.098 |
atal | +0.087 |
eson | +0.086 |
ento | +0.086 |
Neg logit contributions | |
viol | -0.102 |
NPR | -0.096 |
containment | -0.093 |
IRO | -0.093 |
Carnegie | -0.089 |
Token | Rolls |
Token ID | 40074 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
viol | +0.296 |
NPR | +0.274 |
containment | +0.268 |
IRO | +0.267 |
Carnegie | +0.257 |
Neg logit contributions | |
ommel | -0.317 |
reviewed | -0.297 |
atal | -0.265 |
eson | -0.261 |
ento | -0.261 |
Token | - |
Token ID | 12 |
Feature activation | -7.16e+00 |
Pos logit contributions | |
viol | +1.377 |
NPR | +1.287 |
containment | +1.258 |
IRO | +1.253 |
Carnegie | +1.208 |
Neg logit contributions | |
ommel | -1.331 |
reviewed | -1.238 |
atal | -1.097 |
eson | -1.081 |
ento | -1.078 |
Token | Roy |
Token ID | 32027 |
Feature activation | -5.88e-01 |
Pos logit contributions | |
viol | +5.435 |
NPR | +5.254 |
IRO | +5.052 |
containment | +4.882 |
Nev | +4.797 |
Neg logit contributions | |
ommel | -5.404 |
reviewed | -4.879 |
eson | -4.440 |
atal | -4.415 |
ento | -4.311 |
Token | ce |
Token ID | 344 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
viol | +0.546 |
NPR | +0.517 |
containment | +0.507 |
IRO | +0.506 |
Carnegie | +0.490 |
Neg logit contributions | |
ommel | -0.362 |
reviewed | -0.330 |
atal | -0.283 |
eson | -0.278 |
ento | -0.277 |
Token | is |
Token ID | 318 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
ommel | +1.573 |
reviewed | +1.459 |
atal | +1.309 |
eson | +1.286 |
ento | +1.278 |
Neg logit contributions | |
viol | -1.521 |
NPR | -1.453 |
containment | -1.394 |
IRO | -1.389 |
Carnegie | -1.337 |
Token | co |
Token ID | 763 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
viol | +0.162 |
NPR | +0.152 |
containment | +0.147 |
IRO | +0.147 |
Carnegie | +0.141 |
Neg logit contributions | |
ommel | -0.174 |
reviewed | -0.163 |
atal | -0.146 |
eson | -0.144 |
ento | -0.143 |
Token | - |
Token ID | 12 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
viol | +0.724 |
NPR | +0.673 |
containment | +0.659 |
IRO | +0.658 |
Carnegie | +0.631 |
Neg logit contributions | |
ommel | -0.678 |
reviewed | -0.628 |
atal | -0.556 |
eson | -0.552 |
ento | -0.549 |
Token | n |
Token ID | 77 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
viol | +1.668 |
NPR | +1.594 |
IRO | +1.546 |
containment | +1.529 |
Carnegie | +1.476 |
Neg logit contributions | |
ommel | -1.594 |
reviewed | -1.469 |
atal | -1.299 |
eson | -1.288 |
ento | -1.280 |
Token | uf |
Token ID | 3046 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
viol | +3.199 |
NPR | +3.094 |
IRO | +3.056 |
containment | +2.967 |
Carnegie | +2.910 |
Neg logit contributions | |
ommel | -2.540 |
reviewed | -2.368 |
eson | -2.046 |
ento | -2.015 |
atal | -1.999 |
Token | c |
Token ID | 66 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
ommel | +1.464 |
reviewed | +1.342 |
atal | +1.195 |
eson | +1.166 |
ento | +1.160 |
Neg logit contributions | |
viol | -1.736 |
NPR | -1.630 |
containment | -1.596 |
IRO | -1.587 |
Carnegie | -1.526 |
Token | . |
Token ID | 13 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
ommel | +2.193 |
reviewed | +2.020 |
atal | +1.796 |
eson | +1.739 |
ento | +1.738 |
Neg logit contributions | |
viol | -2.611 |
NPR | -2.453 |
containment | -2.406 |
IRO | -2.388 |
Carnegie | -2.304 |
Token | co |
Token ID | 1073 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
viol | +0.294 |
NPR | +0.278 |
containment | +0.271 |
IRO | +0.270 |
Carnegie | +0.262 |
Neg logit contributions | |
ommel | -0.240 |
reviewed | -0.220 |
atal | -0.193 |
eson | -0.190 |
ento | -0.190 |
Token | . |
Token ID | 13 |
Feature activation | -5.79e+00 |
Pos logit contributions | |
viol | +1.866 |
NPR | +1.727 |
containment | +1.702 |
IRO | +1.681 |
Carnegie | +1.630 |
Neg logit contributions | |
ommel | -1.836 |
reviewed | -1.694 |
ento | -1.487 |
eson | -1.486 |
atal | -1.469 |
Token | uk |
Token ID | 2724 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
viol | +4.524 |
NPR | +4.337 |
IRO | +4.234 |
containment | +4.203 |
Carnegie | +4.131 |
Neg logit contributions | |
ommel | -4.177 |
reviewed | -3.719 |
atal | -3.270 |
eson | -3.269 |
ento | -3.193 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
viol | +3.628 |
NPR | +3.373 |
containment | +3.341 |
IRO | +3.300 |
Carnegie | +3.203 |
Neg logit contributions | |
ommel | -3.590 |
reviewed | -3.370 |
atal | -2.962 |
eson | -2.951 |
ento | -2.906 |
Token | which |
Token ID | 543 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
viol | +0.910 |
NPR | +0.853 |
containment | +0.833 |
IRO | +0.829 |
Carnegie | +0.800 |
Neg logit contributions | |
ommel | -0.869 |
reviewed | -0.808 |
atal | -0.716 |
eson | -0.706 |
ento | -0.704 |
Token | will |
Token ID | 481 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
ommel | +0.909 |
reviewed | +0.853 |
atal | +0.770 |
eson | +0.762 |
ento | +0.758 |
Neg logit contributions | |
viol | -0.717 |
NPR | -0.669 |
containment | -0.647 |
IRO | -0.645 |
Carnegie | -0.614 |
Token | be |
Token ID | 307 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
ommel | +1.352 |
reviewed | +1.262 |
atal | +1.137 |
eson | +1.121 |
ento | +1.108 |
Neg logit contributions | |
viol | -1.223 |
NPR | -1.149 |
IRO | -1.113 |
containment | -1.111 |
Carnegie | -1.060 |