Token | bottle |
Token ID | 9294 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ippers | +1.020 |
banks | +0.996 |
bles | +0.978 |
pouch | +0.939 |
ters | +0.908 |
Neg logit contributions | |
Acc | -1.575 |
Acc | -1.472 |
solved | -1.278 |
fixed | -1.277 |
collapse | -1.200 |
Token | of |
Token ID | 286 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
ippers | +1.983 |
banks | +1.970 |
bles | +1.899 |
pouch | +1.852 |
itive | +1.793 |
Neg logit contributions | |
Acc | -1.924 |
Acc | -1.752 |
fixed | -1.522 |
solved | -1.516 |
bell | -1.362 |
Token | lot |
Token ID | 1256 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
ippers | +4.399 |
bles | +4.380 |
itive | +4.253 |
banks | +4.244 |
sters | +4.121 |
Neg logit contributions | |
Acc | -4.384 |
Acc | -4.341 |
fixed | -3.815 |
solved | -3.492 |
cal | -3.194 |
Token | ion |
Token ID | 295 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ippers | +0.930 |
banks | +0.919 |
bles | +0.894 |
pouch | +0.885 |
itive | +0.835 |
Neg logit contributions | |
Acc | -1.041 |
Acc | -0.954 |
solved | -0.811 |
fixed | -0.804 |
collapse | -0.753 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ippers | +2.270 |
banks | +2.191 |
bles | +2.148 |
pouch | +2.122 |
ittens | +2.048 |
Neg logit contributions | |
Acc | -2.011 |
Acc | -1.811 |
fixed | -1.557 |
solved | -1.553 |
collapse | -1.388 |
Token | The |
Token ID | 383 |
Feature activation | +4.64e+00 |
Pos logit contributions | |
ippers | +1.688 |
banks | +1.645 |
bles | +1.636 |
pouch | +1.593 |
itive | +1.541 |
Neg logit contributions | |
Acc | -1.694 |
Acc | -1.557 |
fixed | -1.306 |
solved | -1.300 |
bell | -1.193 |
Token | lot |
Token ID | 1256 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ippers | +8.948 |
sters | +8.819 |
itive | +8.786 |
angers | +8.546 |
kins | +8.490 |
Neg logit contributions | |
Acc | -7.918 |
Acc | -7.317 |
fixed | -7.194 |
cal | -6.408 |
clean | -6.236 |
Token | ion |
Token ID | 295 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
ippers | +2.986 |
banks | +2.944 |
bles | +2.830 |
pouch | +2.791 |
itive | +2.757 |
Neg logit contributions | |
Acc | -2.552 |
Acc | -2.354 |
fixed | -2.022 |
solved | -1.975 |
collapse | -1.801 |
Token | was |
Token ID | 373 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ippers | +1.981 |
banks | +1.906 |
bles | +1.879 |
pouch | +1.843 |
ittens | +1.769 |
Neg logit contributions | |
Acc | -2.046 |
Acc | -1.867 |
solved | -1.615 |
fixed | -1.613 |
collapse | -1.439 |
Token | for |
Token ID | 329 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
ippers | +2.657 |
banks | +2.600 |
bles | +2.466 |
pouch | +2.424 |
itive | +2.416 |
Neg logit contributions | |
Acc | -2.501 |
Acc | -2.265 |
fixed | -2.041 |
solved | -1.969 |
gone | -1.789 |
Token | the |
Token ID | 262 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
ippers | +3.285 |
banks | +3.188 |
bles | +3.118 |
pouch | +3.113 |
itive | +3.002 |
Neg logit contributions | |
Acc | -3.165 |
Acc | -2.952 |
fixed | -2.566 |
solved | -2.506 |
collapse | -2.293 |
Token | lot |
Token ID | 1256 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ippers | +1.074 |
banks | +1.059 |
bles | +1.032 |
pouch | +1.018 |
itive | +0.963 |
Neg logit contributions | |
Acc | -1.154 |
Acc | -1.053 |
solved | -0.895 |
fixed | -0.889 |
collapse | -0.827 |
Token | ion |
Token ID | 295 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
ippers | +2.748 |
banks | +2.698 |
bles | +2.647 |
pouch | +2.567 |
itive | +2.564 |
Neg logit contributions | |
Acc | -2.309 |
Acc | -2.074 |
fixed | -1.795 |
solved | -1.757 |
cal | -1.597 |
Token | . |
Token ID | 13 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
ippers | +0.947 |
banks | +0.934 |
bles | +0.912 |
pouch | +0.902 |
ittens | +0.853 |
Neg logit contributions | |
Acc | -0.976 |
Acc | -0.889 |
solved | -0.749 |
fixed | -0.743 |
collapse | -0.695 |
Token | Anna |
Token ID | 11735 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
ippers | +1.280 |
banks | +1.264 |
bles | +1.242 |
pouch | +1.226 |
itive | +1.174 |
Neg logit contributions | |
Acc | -1.008 |
Acc | -0.908 |
solved | -0.736 |
fixed | -0.728 |
collapse | -0.668 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
ippers | +0.936 |
banks | +0.915 |
bles | +0.891 |
pouch | +0.878 |
ters | +0.824 |
Neg logit contributions | |
Acc | -1.298 |
Acc | -1.196 |
solved | -1.038 |
fixed | -1.029 |
collapse | -0.963 |
Token | the |
Token ID | 262 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
ippers | +3.102 |
bles | +2.974 |
banks | +2.951 |
kins | +2.816 |
itive | +2.813 |
Neg logit contributions | |
Acc | -2.552 |
Acc | -2.284 |
fixed | -1.927 |
solved | -1.880 |
cal | -1.727 |
Token | bottle |
Token ID | 9294 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
ippers | +8.552 |
sters | +8.405 |
itive | +8.378 |
bles | +8.169 |
kins | +7.718 |
Neg logit contributions | |
Acc | -7.105 |
Acc | -6.745 |
fixed | -6.589 |
cal | -6.020 |
solved | -5.808 |
Token | so |
Token ID | 523 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
ippers | +0.037 |
banks | +0.037 |
bles | +0.036 |
pouch | +0.036 |
ters | +0.034 |
Neg logit contributions | |
Acc | -0.050 |
Acc | -0.046 |
solved | -0.039 |
fixed | -0.038 |
collapse | -0.037 |
Token | her |
Token ID | 607 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
Acc | +0.015 |
Acc | +0.013 |
solved | +0.010 |
fixed | +0.010 |
collapse | +0.009 |
Neg logit contributions | |
ippers | -0.027 |
banks | -0.026 |
bles | -0.026 |
pouch | -0.026 |
ittens | -0.025 |
Token | mom |
Token ID | 1995 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
banks | +4.280 |
ippers | +4.240 |
bles | +4.236 |
itive | +4.194 |
ters | +3.998 |
Neg logit contributions | |
Acc | -3.610 |
Acc | -3.425 |
fixed | -2.890 |
bell | -2.702 |
solved | -2.641 |
Token | bought |
Token ID | 5839 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Acc | +0.104 |
Acc | +0.097 |
solved | +0.082 |
collapse | +0.081 |
fixed | +0.080 |
Neg logit contributions | |
ippers | -0.063 |
banks | -0.063 |
bles | -0.062 |
pouch | -0.061 |
ittens | -0.057 |
Token | mouth |
Token ID | 5422 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
ippers | +2.088 |
bles | +2.072 |
banks | +2.017 |
itive | +1.963 |
ters | +1.910 |
Neg logit contributions | |
Acc | -2.918 |
Acc | -2.772 |
fixed | -2.446 |
solved | -2.399 |
bell | -2.270 |
Token | and |
Token ID | 290 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
Acc | +0.032 |
Acc | +0.029 |
solved | +0.024 |
fixed | +0.024 |
collapse | +0.023 |
Neg logit contributions | |
ippers | -0.027 |
banks | -0.027 |
pouch | -0.026 |
bles | -0.026 |
ters | -0.025 |
Token | dropped |
Token ID | 5710 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
Acc | +0.564 |
Acc | +0.522 |
solved | +0.429 |
collapse | +0.424 |
fixed | +0.419 |
Neg logit contributions | |
banks | -0.404 |
ippers | -0.397 |
bles | -0.393 |
pouch | -0.391 |
kins | -0.373 |
Token | one |
Token ID | 530 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
ippers | +1.305 |
banks | +1.272 |
bles | +1.254 |
pouch | +1.224 |
ittens | +1.174 |
Neg logit contributions | |
Acc | -1.404 |
Acc | -1.285 |
fixed | -1.095 |
solved | -1.094 |
collapse | -1.019 |
Token | of |
Token ID | 286 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ippers | +2.150 |
banks | +2.108 |
bles | +2.055 |
itive | +1.969 |
pouch | +1.966 |
Neg logit contributions | |
Acc | -2.406 |
Acc | -2.216 |
fixed | -1.978 |
solved | -1.940 |
collapse | -1.822 |
Token | the |
Token ID | 262 |
Feature activation | +4.30e+00 |
Pos logit contributions | |
ippers | +1.878 |
banks | +1.840 |
bles | +1.808 |
pouch | +1.768 |
itive | +1.703 |
Neg logit contributions | |
Acc | -1.622 |
Acc | -1.464 |
fixed | -1.226 |
solved | -1.218 |
collapse | -1.121 |
Token | pills |
Token ID | 19521 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
ippers | +8.366 |
itive | +8.028 |
owers | +7.623 |
sters | +7.567 |
bles | +7.495 |
Neg logit contributions | |
fixed | -6.583 |
Acc | -6.408 |
Acc | -5.872 |
solved | -5.828 |
ruined | -5.439 |
Token | in |
Token ID | 287 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
ippers | +0.820 |
banks | +0.809 |
bles | +0.789 |
pouch | +0.780 |
ittens | +0.736 |
Neg logit contributions | |
Acc | -0.903 |
Acc | -0.825 |
solved | -0.699 |
fixed | -0.693 |
collapse | -0.651 |
Token | . |
Token ID | 13 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
ippers | +1.971 |
banks | +1.938 |
bles | +1.922 |
pouch | +1.840 |
sters | +1.802 |
Neg logit contributions | |
Acc | -1.627 |
Acc | -1.475 |
fixed | -1.255 |
solved | -1.252 |
collapse | -1.155 |
Token | |
Token ID | 220 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
ippers | +0.065 |
banks | +0.064 |
pouch | +0.062 |
bles | +0.061 |
ters | +0.058 |
Neg logit contributions | |
Acc | -0.077 |
Acc | -0.070 |
solved | -0.060 |
fixed | -0.059 |
collapse | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
ippers | +0.289 |
banks | +0.282 |
bles | +0.278 |
pouch | +0.272 |
ittens | +0.258 |
Neg logit contributions | |
Acc | -0.373 |
Acc | -0.343 |
solved | -0.294 |
fixed | -0.291 |
collapse | -0.277 |
Token | " |
Token ID | 1 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
ippers | +0.128 |
banks | +0.128 |
bles | +0.124 |
pouch | +0.122 |
ittens | +0.117 |
Neg logit contributions | |
Acc | -0.166 |
Acc | -0.153 |
solved | -0.130 |
fixed | -0.129 |
collapse | -0.122 |
Token | Help |
Token ID | 22087 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
Acc | +1.051 |
Acc | +1.016 |
solved | +0.848 |
fixed | +0.843 |
bell | +0.812 |
Neg logit contributions | |
banks | -0.767 |
ippers | -0.749 |
bles | -0.736 |
pouch | -0.734 |
ters | -0.727 |
Token | ! |
Token ID | 0 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
Acc | +0.566 |
Acc | +0.513 |
solved | +0.424 |
fixed | +0.417 |
bell | +0.397 |
Neg logit contributions | |
ippers | -0.498 |
banks | -0.493 |
bles | -0.483 |
pouch | -0.481 |
ters | -0.464 |
Token | Help |
Token ID | 10478 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
Acc | +1.082 |
Acc | +0.931 |
solved | +0.805 |
fixed | +0.779 |
aling | +0.724 |
Neg logit contributions | |
ippers | -1.138 |
banks | -1.125 |
pouch | -1.108 |
bles | -1.087 |
ters | -1.057 |
Token | !" |
Token ID | 2474 |
Feature activation | +9.10e-01 |
Pos logit contributions | |
Acc | +0.091 |
Acc | +0.083 |
solved | +0.068 |
fixed | +0.068 |
collapse | +0.063 |
Neg logit contributions | |
banks | -0.085 |
ippers | -0.085 |
bles | -0.083 |
pouch | -0.082 |
ters | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | +4.28e+00 |
Pos logit contributions | |
ippers | +1.992 |
banks | +1.951 |
bles | +1.945 |
pouch | +1.913 |
itive | +1.836 |
Neg logit contributions | |
Acc | -1.047 |
Acc | -0.928 |
solved | -0.714 |
fixed | -0.709 |
collapse | -0.636 |
Token | sheep |
Token ID | 15900 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ippers | +6.788 |
angers | +6.335 |
robe | +5.950 |
bles | +5.856 |
kins | +5.738 |
Neg logit contributions | |
Acc | -8.009 |
cal | -7.135 |
gone | -7.093 |
fixed | -7.039 |
Acc | -7.035 |
Token | ble |
Token ID | 7245 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
ippers | +2.372 |
banks | +2.197 |
bles | +2.194 |
pouch | +2.141 |
angers | +2.102 |
Neg logit contributions | |
Acc | -2.559 |
Acc | -2.322 |
fixed | -2.132 |
solved | -2.111 |
collapse | -1.978 |
Token | ated |
Token ID | 515 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
Acc | +0.243 |
Acc | +0.226 |
solved | +0.195 |
fixed | +0.194 |
collapse | +0.182 |
Neg logit contributions | |
ippers | -0.168 |
bles | -0.162 |
banks | -0.162 |
pouch | -0.156 |
kins | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
Acc | +1.058 |
Acc | +0.986 |
solved | +0.813 |
fixed | +0.788 |
bell | +0.749 |
Neg logit contributions | |
banks | -0.829 |
ippers | -0.798 |
pouch | -0.785 |
bles | -0.782 |
ters | -0.769 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
ippers | +0.771 |
banks | +0.761 |
bles | +0.743 |
pouch | +0.735 |
kins | +0.695 |
Neg logit contributions | |
Acc | -0.797 |
Acc | -0.726 |
solved | -0.611 |
fixed | -0.605 |
collapse | -0.567 |
Token | to |
Token ID | 284 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
Acc | +0.950 |
Acc | +0.868 |
solved | +0.686 |
fixed | +0.671 |
collapse | +0.654 |
Neg logit contributions | |
banks | -1.047 |
pouch | -1.010 |
ippers | -1.006 |
friends | -0.981 |
bles | -0.977 |
Token | sleep |
Token ID | 3993 |
Feature activation | +9.50e-01 |
Pos logit contributions | |
ippers | +0.400 |
banks | +0.399 |
pouch | +0.384 |
bles | +0.383 |
ittens | +0.362 |
Neg logit contributions | |
Acc | -0.425 |
Acc | -0.385 |
solved | -0.322 |
fixed | -0.317 |
collapse | -0.298 |
Token | on |
Token ID | 319 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
ippers | +1.580 |
banks | +1.555 |
bles | +1.516 |
pouch | +1.505 |
ittens | +1.426 |
Neg logit contributions | |
Acc | -1.596 |
Acc | -1.441 |
solved | -1.229 |
fixed | -1.228 |
collapse | -1.127 |
Token | it |
Token ID | 340 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
ippers | +0.609 |
banks | +0.604 |
bles | +0.587 |
pouch | +0.582 |
ittens | +0.550 |
Neg logit contributions | |
Acc | -0.651 |
Acc | -0.595 |
solved | -0.501 |
fixed | -0.495 |
collapse | -0.464 |
Token | . |
Token ID | 13 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
ippers | +1.165 |
banks | +1.141 |
bles | +1.119 |
pouch | +1.105 |
ittens | +1.050 |
Neg logit contributions | |
Acc | -1.149 |
Acc | -1.044 |
solved | -0.883 |
fixed | -0.877 |
collapse | -0.817 |
Token | The |
Token ID | 383 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
ippers | +1.638 |
banks | +1.592 |
bles | +1.586 |
pouch | +1.547 |
itive | +1.499 |
Neg logit contributions | |
Acc | -1.540 |
Acc | -1.415 |
solved | -1.180 |
fixed | -1.168 |
collapse | -1.079 |
Token | snail |
Token ID | 47374 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
ippers | +5.983 |
angers | +5.558 |
banks | +5.416 |
re | +5.211 |
itive | +5.113 |
Neg logit contributions | |
Acc | -8.579 |
cal | -7.624 |
Acc | -7.356 |
solved | -7.294 |
fixed | -7.147 |
Token | stretched |
Token ID | 19110 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
ippers | +2.163 |
banks | +2.015 |
bles | +2.008 |
pouch | +1.934 |
angers | +1.894 |
Neg logit contributions | |
Acc | -2.575 |
Acc | -2.346 |
solved | -2.134 |
fixed | -2.133 |
bell | -1.913 |
Token | its |
Token ID | 663 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
Acc | +0.286 |
Acc | +0.263 |
solved | +0.218 |
fixed | +0.211 |
collapse | +0.202 |
Neg logit contributions | |
ippers | -0.275 |
banks | -0.273 |
bles | -0.267 |
pouch | -0.264 |
ters | -0.250 |
Token | long |
Token ID | 890 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ippers | +3.837 |
bles | +3.667 |
angers | +3.656 |
itive | +3.549 |
ters | +3.534 |
Neg logit contributions | |
Acc | -4.897 |
Acc | -4.456 |
fixed | -4.170 |
solved | -3.990 |
gone | -3.911 |
Token | neck |
Token ID | 7393 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
bles | +3.082 |
ippers | +3.066 |
angers | +2.846 |
ters | +2.839 |
ancer | +2.834 |
Neg logit contributions | |
Acc | -5.434 |
Acc | -5.063 |
fixed | -4.792 |
solved | -4.668 |
bell | -4.481 |
Token | fox |
Token ID | 21831 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
ippers | +0.301 |
banks | +0.299 |
pouch | +0.289 |
bles | +0.288 |
ittens | +0.264 |
Neg logit contributions | |
Acc | -0.501 |
Acc | -0.462 |
solved | -0.402 |
fixed | -0.397 |
collapse | -0.380 |
Token | coming |
Token ID | 2406 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
ippers | +0.360 |
banks | +0.357 |
bles | +0.347 |
pouch | +0.344 |
ters | +0.319 |
Neg logit contributions | |
Acc | -0.521 |
Acc | -0.483 |
solved | -0.414 |
fixed | -0.411 |
collapse | -0.391 |
Token | towards |
Token ID | 3371 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
Acc | +1.431 |
Acc | +1.334 |
solved | +1.111 |
fixed | +1.090 |
collapse | +1.009 |
Neg logit contributions | |
banks | -1.277 |
bles | -1.237 |
ippers | -1.237 |
pouch | -1.236 |
ters | -1.177 |
Token | her |
Token ID | 607 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ippers | +1.349 |
banks | +1.310 |
bles | +1.308 |
pouch | +1.269 |
kins | +1.218 |
Neg logit contributions | |
Acc | -1.300 |
Acc | -1.180 |
solved | -0.992 |
fixed | -0.989 |
collapse | -0.912 |
Token | . |
Token ID | 13 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ippers | +1.915 |
bles | +1.840 |
banks | +1.828 |
pouch | +1.761 |
kins | +1.751 |
Neg logit contributions | |
Acc | -1.805 |
Acc | -1.639 |
solved | -1.397 |
fixed | -1.386 |
collapse | -1.303 |
Token | The |
Token ID | 383 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
ippers | +2.170 |
bles | +2.127 |
banks | +2.067 |
kins | +1.980 |
itive | +1.968 |
Neg logit contributions | |
Acc | -2.124 |
Acc | -1.949 |
fixed | -1.644 |
solved | -1.615 |
bell | -1.484 |
Token | fox |
Token ID | 21831 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
ippers | +7.686 |
angers | +7.152 |
kins | +6.912 |
bles | +6.667 |
banks | +6.594 |
Neg logit contributions | |
Acc | -7.666 |
cal | -6.895 |
solved | -6.511 |
gone | -6.446 |
fixed | -6.433 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
ippers | +0.897 |
banks | +0.880 |
bles | +0.856 |
pouch | +0.844 |
ittens | +0.795 |
Neg logit contributions | |
Acc | -1.099 |
Acc | -1.010 |
solved | -0.868 |
fixed | -0.860 |
collapse | -0.809 |
Token | to |
Token ID | 284 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
Acc | +0.409 |
Acc | +0.371 |
solved | +0.292 |
fixed | +0.286 |
collapse | +0.280 |
Neg logit contributions | |
banks | -0.481 |
pouch | -0.470 |
ippers | -0.469 |
bles | -0.455 |
friends | -0.449 |
Token | catch |
Token ID | 4929 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
banks | +0.065 |
ippers | +0.064 |
pouch | +0.062 |
bles | +0.061 |
ittens | +0.058 |
Neg logit contributions | |
Acc | -0.073 |
Acc | -0.066 |
solved | -0.055 |
fixed | -0.054 |
collapse | -0.051 |
Token | Betsy |
Token ID | 45775 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ippers | +0.987 |
banks | +0.968 |
bles | +0.950 |
pouch | +0.935 |
kins | +0.889 |
Neg logit contributions | |
Acc | -1.022 |
Acc | -0.930 |
solved | -0.785 |
fixed | -0.779 |
collapse | -0.727 |
Token | a |
Token ID | 257 |
Feature activation | +7.92e-01 |
Pos logit contributions | |
ippers | +1.775 |
banks | +1.739 |
bles | +1.684 |
pouch | +1.652 |
angers | +1.580 |
Neg logit contributions | |
Acc | -1.756 |
Acc | -1.598 |
fixed | -1.361 |
solved | -1.358 |
collapse | -1.218 |
Token | small |
Token ID | 1402 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
ippers | +0.864 |
banks | +0.842 |
bles | +0.809 |
pouch | +0.784 |
kins | +0.733 |
Neg logit contributions | |
Acc | -1.782 |
Acc | -1.667 |
solved | -1.470 |
fixed | -1.467 |
collapse | -1.387 |
Token | bear |
Token ID | 6842 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
ippers | +2.016 |
bles | +1.907 |
banks | +1.874 |
kins | +1.759 |
angers | +1.748 |
Neg logit contributions | |
Acc | -3.643 |
Acc | -3.510 |
fixed | -3.106 |
solved | -3.049 |
cal | -2.881 |
Token | . |
Token ID | 13 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
ippers | +1.523 |
banks | +1.493 |
bles | +1.464 |
pouch | +1.434 |
kins | +1.375 |
Neg logit contributions | |
Acc | -1.420 |
Acc | -1.295 |
solved | -1.091 |
fixed | -1.087 |
collapse | -0.987 |
Token | The |
Token ID | 383 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
ippers | +1.327 |
banks | +1.289 |
bles | +1.283 |
pouch | +1.252 |
kins | +1.198 |
Neg logit contributions | |
Acc | -1.477 |
Acc | -1.366 |
solved | -1.149 |
fixed | -1.143 |
collapse | -1.068 |
Token | small |
Token ID | 1402 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
ippers | +4.670 |
angers | +4.171 |
banks | +4.074 |
sters | +4.003 |
bles | +3.957 |
Neg logit contributions | |
Acc | -5.867 |
Acc | -5.552 |
fixed | -5.215 |
solved | -4.885 |
cal | -4.861 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
ippers | +7.612 |
angers | +7.088 |
ittens | +6.722 |
bles | +6.696 |
banks | +6.595 |
Neg logit contributions | |
fixed | -6.234 |
Acc | -6.227 |
Acc | -6.172 |
bug | -5.678 |
over | -5.673 |
Token | was |
Token ID | 373 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
banks | +0.044 |
ippers | +0.043 |
bles | +0.042 |
pouch | +0.042 |
kins | +0.039 |
Neg logit contributions | |
Acc | -0.057 |
Acc | -0.053 |
solved | -0.044 |
fixed | -0.044 |
collapse | -0.043 |
Token | always |
Token ID | 1464 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ippers | +1.701 |
banks | +1.660 |
bles | +1.610 |
pouch | +1.581 |
kins | +1.529 |
Neg logit contributions | |
Acc | -1.778 |
Acc | -1.622 |
fixed | -1.399 |
solved | -1.396 |
collapse | -1.265 |
Token | sad |
Token ID | 6507 |
Feature activation | +8.87e-01 |
Pos logit contributions | |
ippers | +1.949 |
banks | +1.913 |
bles | +1.828 |
pouch | +1.803 |
ittens | +1.758 |
Neg logit contributions | |
Acc | -2.203 |
Acc | -2.024 |
fixed | -1.766 |
solved | -1.753 |
collapse | -1.582 |
Token | , |
Token ID | 11 |
Feature activation | +6.79e-01 |
Pos logit contributions | |
ippers | +1.467 |
banks | +1.441 |
bles | +1.411 |
pouch | +1.400 |
kins | +1.333 |
Neg logit contributions | |
Acc | -1.500 |
Acc | -1.379 |
solved | -1.156 |
fixed | -1.152 |
collapse | -1.067 |
Token |
|
Token ID | 198 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
Acc | +1.528 |
Acc | +1.384 |
fixed | +1.221 |
solved | +1.200 |
collapse | +1.184 |
Neg logit contributions | |
ippers | -1.148 |
banks | -1.126 |
bles | -1.077 |
pouch | -1.074 |
angers | -1.033 |
Token |
|
Token ID | 198 |
Feature activation | -9.73e-01 |
Pos logit contributions | |
Acc | +1.497 |
Acc | +1.394 |
fixed | +1.213 |
solved | +1.206 |
collapse | +1.167 |
Neg logit contributions | |
ippers | -0.925 |
bles | -0.893 |
banks | -0.892 |
pouch | -0.837 |
kins | -0.815 |
Token | So |
Token ID | 2396 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Acc | +1.568 |
Acc | +1.489 |
fixed | +1.237 |
solved | +1.164 |
collapse | +1.151 |
Neg logit contributions | |
ippers | -1.595 |
bles | -1.592 |
banks | -1.584 |
kins | -1.507 |
pouch | -1.473 |
Token | Jill |
Token ID | 22772 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
Acc | +0.210 |
Acc | +0.190 |
fixed | +0.158 |
solved | +0.157 |
collapse | +0.149 |
Neg logit contributions | |
ippers | -0.176 |
banks | -0.173 |
bles | -0.170 |
pouch | -0.167 |
kins | -0.162 |
Token | fed |
Token ID | 11672 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
Acc | +0.552 |
Acc | +0.515 |
solved | +0.408 |
fixed | +0.405 |
collapse | +0.404 |
Neg logit contributions | |
banks | -0.459 |
ippers | -0.443 |
bles | -0.442 |
pouch | -0.441 |
kins | -0.417 |
Token | the |
Token ID | 262 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
ippers | +0.972 |
banks | +0.957 |
bles | +0.936 |
pouch | +0.926 |
ittens | +0.878 |
Neg logit contributions | |
Acc | -0.959 |
Acc | -0.871 |
solved | -0.730 |
fixed | -0.724 |
collapse | -0.675 |
Token | bunny |
Token ID | 44915 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
ippers | +6.968 |
angers | +6.587 |
banks | +6.195 |
bles | +5.927 |
ittens | +5.856 |
Neg logit contributions | |
Acc | -7.961 |
fixed | -7.339 |
cal | -7.247 |
gone | -7.050 |
solved | -6.968 |
Token | some |
Token ID | 617 |
Feature activation | +9.48e-01 |
Pos logit contributions | |
ippers | +0.806 |
banks | +0.791 |
bles | +0.773 |
pouch | +0.761 |
ittens | +0.716 |
Neg logit contributions | |
Acc | -1.044 |
Acc | -0.960 |
solved | -0.825 |
fixed | -0.818 |
collapse | -0.772 |
Token | carrots |
Token ID | 34397 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
ippers | +0.812 |
banks | +0.770 |
bles | +0.751 |
pouch | +0.722 |
itive | +0.654 |
Neg logit contributions | |
Acc | -2.343 |
Acc | -2.202 |
solved | -1.991 |
fixed | -1.990 |
collapse | -1.886 |
Token | . |
Token ID | 13 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
ippers | +0.680 |
banks | +0.673 |
bles | +0.656 |
pouch | +0.649 |
kins | +0.614 |
Neg logit contributions | |
Acc | -0.739 |
Acc | -0.675 |
solved | -0.569 |
fixed | -0.563 |
collapse | -0.530 |
Token | The |
Token ID | 383 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
Acc | +1.764 |
Acc | +1.584 |
fixed | +1.408 |
solved | +1.402 |
collapse | +1.364 |
Neg logit contributions | |
ippers | -1.216 |
banks | -1.208 |
pouch | -1.148 |
bles | -1.129 |
angers | -1.097 |
Token | but |
Token ID | 475 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ippers | +0.484 |
banks | +0.478 |
bles | +0.468 |
pouch | +0.461 |
ittens | +0.437 |
Neg logit contributions | |
Acc | -0.600 |
Acc | -0.549 |
solved | -0.469 |
fixed | -0.462 |
collapse | -0.439 |
Token | only |
Token ID | 691 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ippers | +2.652 |
banks | +2.537 |
bles | +2.524 |
pouch | +2.460 |
itive | +2.415 |
Neg logit contributions | |
Acc | -1.986 |
Acc | -1.832 |
fixed | -1.493 |
solved | -1.485 |
collapse | -1.351 |
Token | a |
Token ID | 257 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
ippers | +1.885 |
banks | +1.837 |
bles | +1.793 |
pouch | +1.753 |
ittens | +1.684 |
Neg logit contributions | |
Acc | -2.062 |
Acc | -1.920 |
fixed | -1.658 |
solved | -1.646 |
collapse | -1.519 |
Token | little |
Token ID | 1310 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
ippers | +2.383 |
bles | +2.278 |
banks | +2.269 |
ters | +2.110 |
angers | +2.094 |
Neg logit contributions | |
Acc | -3.634 |
Acc | -3.374 |
fixed | -3.046 |
solved | -3.013 |
cal | -2.874 |
Token | . |
Token ID | 13 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
ippers | +3.681 |
bles | +3.594 |
banks | +3.451 |
angers | +3.336 |
kins | +3.325 |
Neg logit contributions | |
Acc | -2.866 |
Acc | -2.656 |
fixed | -2.304 |
solved | -2.229 |
cal | -2.177 |
Token | The |
Token ID | 383 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
ippers | +0.631 |
banks | +0.625 |
bles | +0.609 |
pouch | +0.606 |
ters | +0.574 |
Neg logit contributions | |
Acc | -0.582 |
Acc | -0.524 |
solved | -0.439 |
fixed | -0.433 |
collapse | -0.404 |
Token | oats |
Token ID | 47009 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ippers | +6.623 |
bles | +6.232 |
angers | +5.960 |
robe | +5.725 |
ters | +5.661 |
Neg logit contributions | |
Acc | -7.744 |
cal | -7.004 |
fixed | -6.861 |
Acc | -6.799 |
solved | -6.544 |
Token | are |
Token ID | 389 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
ippers | +2.628 |
banks | +2.537 |
bles | +2.493 |
pouch | +2.466 |
itive | +2.429 |
Neg logit contributions | |
Acc | -2.559 |
Acc | -2.378 |
fixed | -2.114 |
solved | -2.072 |
collapse | -1.928 |
Token | for |
Token ID | 329 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ippers | +1.514 |
banks | +1.474 |
bles | +1.440 |
pouch | +1.415 |
itive | +1.363 |
Neg logit contributions | |
Acc | -1.511 |
Acc | -1.374 |
solved | -1.180 |
fixed | -1.178 |
collapse | -1.088 |
Token | the |
Token ID | 262 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ippers | +1.860 |
banks | +1.815 |
bles | +1.784 |
pouch | +1.762 |
itive | +1.688 |
Neg logit contributions | |
Acc | -1.790 |
Acc | -1.651 |
solved | -1.394 |
fixed | -1.393 |
collapse | -1.289 |
Token | horses |
Token ID | 14260 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ippers | +3.234 |
bles | +3.029 |
kins | +2.738 |
angers | +2.722 |
banks | +2.722 |
Neg logit contributions | |
Acc | -5.268 |
Acc | -4.863 |
fixed | -4.567 |
solved | -4.492 |
cal | -4.340 |
Token |
|
Token ID | 198 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ippers | +1.868 |
banks | +1.837 |
bles | +1.797 |
pouch | +1.770 |
itive | +1.706 |
Neg logit contributions | |
Acc | -1.909 |
Acc | -1.719 |
solved | -1.459 |
fixed | -1.445 |
collapse | -1.332 |
Token | The |
Token ID | 464 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
ippers | +2.126 |
banks | +2.029 |
bles | +2.013 |
pouch | +1.983 |
kins | +1.901 |
Neg logit contributions | |
Acc | -2.105 |
Acc | -1.855 |
solved | -1.632 |
fixed | -1.569 |
collapse | -1.468 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
ippers | +4.996 |
banks | +4.908 |
pouch | +4.718 |
bles | +4.580 |
angers | +4.431 |
Neg logit contributions | |
Acc | -4.944 |
Acc | -4.443 |
fixed | -3.965 |
solved | -3.853 |
cal | -3.704 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
Acc | +0.463 |
Acc | +0.426 |
collapse | +0.351 |
solved | +0.351 |
fixed | +0.347 |
Neg logit contributions | |
banks | -0.338 |
ippers | -0.327 |
pouch | -0.323 |
bles | -0.318 |
friends | -0.302 |
Token | at |
Token ID | 379 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
ippers | +0.201 |
banks | +0.198 |
bles | +0.195 |
pouch | +0.192 |
itive | +0.182 |
Neg logit contributions | |
Acc | -0.233 |
Acc | -0.215 |
solved | -0.181 |
fixed | -0.178 |
collapse | -0.172 |
Token | the |
Token ID | 262 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
Acc | +0.033 |
Acc | +0.030 |
solved | +0.026 |
fixed | +0.025 |
collapse | +0.024 |
Neg logit contributions | |
ippers | -0.029 |
banks | -0.028 |
bles | -0.028 |
pouch | -0.027 |
friends | -0.026 |
Token | w |
Token ID | 266 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
ippers | +8.267 |
banks | +7.630 |
pouch | +7.578 |
kins | +7.560 |
bles | +7.528 |
Neg logit contributions | |
Acc | -6.243 |
Acc | -5.675 |
cal | -5.261 |
fixed | -5.232 |
over | -4.880 |
Token | affle |
Token ID | 30697 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
pouch | +1.961 |
banks | +1.941 |
ippers | +1.729 |
ters | +1.653 |
couple | +1.632 |
Neg logit contributions | |
Acc | -4.671 |
Acc | -4.407 |
fixed | -3.816 |
solved | -3.816 |
aling | -3.717 |
Token | maker |
Token ID | 16009 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
ippers | +1.491 |
banks | +1.464 |
bles | +1.429 |
pouch | +1.411 |
ittens | +1.339 |
Neg logit contributions | |
Acc | -1.524 |
Acc | -1.393 |
solved | -1.177 |
fixed | -1.167 |
collapse | -1.079 |
Token | , |
Token ID | 11 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
ippers | +0.416 |
banks | +0.413 |
bles | +0.401 |
pouch | +0.399 |
ters | +0.377 |
Neg logit contributions | |
Acc | -0.461 |
Acc | -0.420 |
solved | -0.353 |
fixed | -0.349 |
collapse | -0.332 |
Token | who |
Token ID | 508 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
Acc | +0.557 |
Acc | +0.517 |
solved | +0.444 |
fixed | +0.437 |
collapse | +0.431 |
Neg logit contributions | |
banks | -0.391 |
ippers | -0.387 |
pouch | -0.381 |
bles | -0.377 |
couple | -0.355 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ippers | +0.433 |
banks | +0.432 |
bles | +0.418 |
pouch | +0.414 |
kins | +0.385 |
Neg logit contributions | |
Acc | -0.630 |
Acc | -0.582 |
solved | -0.500 |
fixed | -0.496 |
collapse | -0.476 |
Token | her |
Token ID | 607 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
Acc | +0.376 |
Acc | +0.344 |
solved | +0.296 |
fixed | +0.289 |
collapse | +0.284 |
Neg logit contributions | |
ippers | -0.251 |
banks | -0.250 |
bles | -0.244 |
pouch | -0.238 |
ters | -0.230 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
ippers | +0.356 |
banks | +0.350 |
bles | +0.344 |
pouch | +0.339 |
kins | +0.317 |
Neg logit contributions | |
Acc | -0.523 |
Acc | -0.480 |
solved | -0.413 |
fixed | -0.407 |
collapse | -0.391 |
Token | my |
Token ID | 1820 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
ippers | +0.675 |
banks | +0.670 |
bles | +0.662 |
pouch | +0.653 |
ters | +0.630 |
Neg logit contributions | |
Acc | -0.316 |
Acc | -0.270 |
solved | -0.197 |
fixed | -0.190 |
collapse | -0.172 |
Token | what |
Token ID | 644 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
ippers | +0.956 |
banks | +0.942 |
bles | +0.918 |
pouch | +0.909 |
ittens | +0.860 |
Neg logit contributions | |
Acc | -1.040 |
Acc | -0.953 |
solved | -0.804 |
fixed | -0.798 |
collapse | -0.746 |
Token | the |
Token ID | 262 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
Acc | +0.298 |
Acc | +0.271 |
solved | +0.228 |
fixed | +0.226 |
collapse | +0.221 |
Neg logit contributions | |
ippers | -0.226 |
banks | -0.223 |
pouch | -0.219 |
bles | -0.216 |
kins | -0.209 |
Token | sign |
Token ID | 1051 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
bles | +7.562 |
ippers | +7.308 |
banks | +7.258 |
ters | +7.074 |
kins | +6.928 |
Neg logit contributions | |
Acc | -6.229 |
Acc | -6.016 |
cal | -6.010 |
fixed | -5.861 |
solved | -5.502 |
Token | meant |
Token ID | 4001 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
ippers | +0.365 |
banks | +0.362 |
bles | +0.350 |
pouch | +0.347 |
kins | +0.323 |
Neg logit contributions | |
Acc | -0.604 |
Acc | -0.556 |
solved | -0.481 |
fixed | -0.476 |
collapse | -0.457 |
Token | . |
Token ID | 13 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
ippers | +1.523 |
banks | +1.508 |
bles | +1.469 |
pouch | +1.460 |
ittens | +1.380 |
Neg logit contributions | |
Acc | -1.492 |
Acc | -1.373 |
solved | -1.164 |
fixed | -1.153 |
collapse | -1.055 |
Token | Her |
Token ID | 2332 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
Acc | +1.434 |
Acc | +1.262 |
solved | +1.140 |
collapse | +1.095 |
fixed | +1.095 |
Neg logit contributions | |
banks | -1.110 |
ippers | -1.095 |
pouch | -1.058 |
bles | -1.017 |
angers | -0.988 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
ippers | +0.596 |
banks | +0.586 |
bles | +0.571 |
pouch | +0.564 |
ittens | +0.528 |
Neg logit contributions | |
Acc | -0.835 |
Acc | -0.768 |
solved | -0.663 |
fixed | -0.656 |
collapse | -0.623 |
Token | of |
Token ID | 286 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
Acc | +0.627 |
Acc | +0.578 |
solved | +0.498 |
fixed | +0.493 |
collapse | +0.476 |
Neg logit contributions | |
banks | -0.339 |
ippers | -0.338 |
pouch | -0.326 |
bles | -0.323 |
friends | -0.305 |
Token | water |
Token ID | 1660 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
ippers | +2.025 |
bles | +1.906 |
banks | +1.861 |
pouch | +1.802 |
angers | +1.757 |
Neg logit contributions | |
Acc | -3.508 |
Acc | -3.292 |
solved | -2.906 |
fixed | -2.896 |
cal | -2.749 |
Token | l |
Token ID | 300 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
ippers | +0.734 |
banks | +0.725 |
bles | +0.707 |
pouch | +0.699 |
kins | +0.660 |
Neg logit contributions | |
Acc | -0.816 |
Acc | -0.746 |
solved | -0.631 |
fixed | -0.625 |
collapse | -0.588 |
Token | ilies |
Token ID | 3922 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
ippers | +0.703 |
banks | +0.701 |
bles | +0.646 |
pouch | +0.636 |
itive | +0.581 |
Neg logit contributions | |
Acc | -2.340 |
Acc | -2.192 |
solved | -1.984 |
fixed | -1.954 |
collapse | -1.885 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ippers | +0.865 |
banks | +0.853 |
bles | +0.832 |
pouch | +0.822 |
ittens | +0.777 |
Neg logit contributions | |
Acc | -0.962 |
Acc | -0.879 |
solved | -0.746 |
fixed | -0.739 |
collapse | -0.694 |
Token | The |
Token ID | 383 |
Feature activation | +4.06e+00 |
Pos logit contributions | |
ippers | +0.266 |
banks | +0.264 |
pouch | +0.256 |
bles | +0.255 |
ters | +0.242 |
Neg logit contributions | |
Acc | -0.271 |
Acc | -0.246 |
solved | -0.207 |
fixed | -0.206 |
collapse | -0.195 |
Token | red |
Token ID | 2266 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
ippers | +7.244 |
itive | +7.171 |
ittens | +6.654 |
angers | +6.643 |
banks | +6.604 |
Neg logit contributions | |
Acc | -6.754 |
cal | -6.188 |
Acc | -5.907 |
solved | -5.728 |
bug | -5.429 |
Token | l |
Token ID | 300 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
banks | +4.995 |
ippers | +4.942 |
itive | +4.911 |
bles | +4.768 |
kins | +4.691 |
Neg logit contributions | |
Acc | -6.071 |
Acc | -5.545 |
bug | -5.295 |
cal | -5.206 |
bell | -5.057 |
Token | ily |
Token ID | 813 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
banks | +2.721 |
ippers | +2.581 |
itive | +2.443 |
pouch | +2.401 |
bles | +2.342 |
Neg logit contributions | |
Acc | -4.195 |
Acc | -3.830 |
solved | -3.523 |
bug | -3.302 |
fixed | -3.299 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
ippers | +0.412 |
banks | +0.410 |
bles | +0.400 |
pouch | +0.396 |
kins | +0.372 |
Neg logit contributions | |
Acc | -0.498 |
Acc | -0.459 |
solved | -0.387 |
fixed | -0.383 |
collapse | -0.366 |
Token | all |
Token ID | 477 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
ippers | +0.532 |
banks | +0.527 |
bles | +0.514 |
pouch | +0.507 |
ters | +0.479 |
Neg logit contributions | |
Acc | -0.642 |
Acc | -0.589 |
solved | -0.501 |
fixed | -0.497 |
collapse | -0.471 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
ippers | +4.674 |
bles | +4.366 |
kins | +4.184 |
itive | +4.060 |
sters | +4.050 |
Neg logit contributions | |
Acc | -6.427 |
Acc | -6.042 |
fixed | -5.576 |
solved | -5.321 |
cal | -5.171 |
Token | 's |
Token ID | 338 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
ippers | +0.614 |
banks | +0.609 |
bles | +0.595 |
pouch | +0.591 |
kins | +0.561 |
Neg logit contributions | |
Acc | -0.512 |
Acc | -0.461 |
solved | -0.374 |
fixed | -0.369 |
collapse | -0.347 |
Token | big |
Token ID | 1263 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
bles | +4.258 |
ippers | +4.244 |
itive | +4.218 |
banks | +4.088 |
kins | +4.032 |
Neg logit contributions | |
Acc | -4.999 |
Acc | -4.881 |
fixed | -4.247 |
bell | -4.009 |
solved | -3.986 |
Token | paws |
Token ID | 49193 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ippers | +2.119 |
bles | +2.064 |
banks | +1.989 |
kins | +1.918 |
ittens | +1.877 |
Neg logit contributions | |
Acc | -2.856 |
Acc | -2.726 |
fixed | -2.412 |
solved | -2.373 |
cal | -2.184 |
Token | . |
Token ID | 13 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
ippers | +1.889 |
banks | +1.827 |
bles | +1.805 |
pouch | +1.783 |
ittens | +1.707 |
Neg logit contributions | |
Acc | -1.737 |
Acc | -1.578 |
fixed | -1.352 |
solved | -1.344 |
collapse | -1.227 |
Token | The |
Token ID | 383 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
ippers | +1.549 |
banks | +1.513 |
bles | +1.501 |
pouch | +1.466 |
itive | +1.419 |
Neg logit contributions | |
Acc | -1.493 |
Acc | -1.365 |
fixed | -1.131 |
solved | -1.131 |
collapse | -1.035 |
Token | bear |
Token ID | 6842 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
ippers | +7.007 |
angers | +6.602 |
sters | +6.597 |
bles | +6.560 |
itive | +6.456 |
Neg logit contributions | |
Acc | -7.531 |
Acc | -7.032 |
fixed | -6.748 |
cal | -6.148 |
repaired | -5.644 |
Token | liked |
Token ID | 8288 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
ippers | +0.365 |
banks | +0.365 |
bles | +0.354 |
pouch | +0.350 |
kins | +0.329 |
Neg logit contributions | |
Acc | -0.479 |
Acc | -0.441 |
solved | -0.374 |
fixed | -0.369 |
collapse | -0.358 |
Token | to |
Token ID | 284 |
Feature activation | +7.01e-01 |
Pos logit contributions | |
ippers | +0.628 |
banks | +0.624 |
bles | +0.610 |
pouch | +0.606 |
ittens | +0.579 |
Neg logit contributions | |
Acc | -0.473 |
Acc | -0.423 |
solved | -0.341 |
fixed | -0.336 |
collapse | -0.313 |
Token | roll |
Token ID | 4836 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ippers | +1.173 |
banks | +1.150 |
bles | +1.136 |
pouch | +1.113 |
ters | +1.062 |
Neg logit contributions | |
Acc | -1.160 |
Acc | -1.059 |
solved | -0.888 |
fixed | -0.885 |
collapse | -0.824 |
Token | around |
Token ID | 1088 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
ippers | +1.854 |
banks | +1.807 |
bles | +1.753 |
pouch | +1.742 |
itive | +1.692 |
Neg logit contributions | |
Acc | -1.954 |
Acc | -1.773 |
fixed | -1.546 |
solved | -1.526 |
collapse | -1.421 |
Token | unk |
Token ID | 2954 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
banks | +1.633 |
ippers | +1.530 |
pouch | +1.505 |
bles | +1.444 |
couple | +1.410 |
Neg logit contributions | |
Acc | -3.419 |
Acc | -3.145 |
solved | -2.813 |
fixed | -2.744 |
collapse | -2.631 |
Token | . |
Token ID | 13 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
Acc | +0.843 |
Acc | +0.768 |
solved | +0.631 |
fixed | +0.619 |
collapse | +0.601 |
Neg logit contributions | |
banks | -0.708 |
ippers | -0.706 |
pouch | -0.705 |
bles | -0.689 |
ters | -0.657 |
Token | He |
Token ID | 679 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
Acc | +0.547 |
Acc | +0.485 |
solved | +0.406 |
fixed | +0.402 |
collapse | +0.390 |
Neg logit contributions | |
banks | -0.539 |
ippers | -0.529 |
pouch | -0.513 |
bles | -0.510 |
ters | -0.491 |
Token | hopped |
Token ID | 45230 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
banks | +0.152 |
ippers | +0.150 |
bles | +0.148 |
pouch | +0.146 |
kins | +0.136 |
Neg logit contributions | |
Acc | -0.210 |
Acc | -0.195 |
solved | -0.163 |
fixed | -0.162 |
collapse | -0.157 |
Token | towards |
Token ID | 3371 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
Acc | +0.219 |
Acc | +0.201 |
solved | +0.154 |
fixed | +0.152 |
collapse | +0.145 |
Neg logit contributions | |
banks | -0.266 |
ippers | -0.264 |
bles | -0.260 |
pouch | -0.259 |
ters | -0.247 |
Token | the |
Token ID | 262 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
ippers | +0.472 |
banks | +0.470 |
bles | +0.455 |
pouch | +0.452 |
ters | +0.427 |
Neg logit contributions | |
Acc | -0.503 |
Acc | -0.459 |
solved | -0.384 |
fixed | -0.379 |
collapse | -0.358 |
Token | sk |
Token ID | 1341 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
ippers | +7.791 |
kins | +7.110 |
bles | +6.881 |
angers | +6.855 |
itive | +6.706 |
Neg logit contributions | |
Acc | -6.663 |
fixed | -5.907 |
solved | -5.904 |
cal | -5.864 |
Acc | -5.784 |
Token | unk |
Token ID | 2954 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
banks | +2.683 |
pouch | +2.483 |
couple | +2.424 |
ippers | +2.398 |
ranc | +2.360 |
Neg logit contributions | |
Acc | -4.872 |
Acc | -4.442 |
solved | -3.986 |
fixed | -3.809 |
bug | -3.736 |
Token | and |
Token ID | 290 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
Acc | +0.707 |
Acc | +0.646 |
solved | +0.537 |
fixed | +0.528 |
collapse | +0.508 |
Neg logit contributions | |
banks | -0.554 |
ippers | -0.539 |
pouch | -0.534 |
bles | -0.519 |
ters | -0.493 |
Token | he |
Token ID | 339 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
ippers | +0.703 |
banks | +0.694 |
bles | +0.675 |
pouch | +0.667 |
kins | +0.627 |
Neg logit contributions | |
Acc | -0.884 |
Acc | -0.813 |
solved | -0.695 |
fixed | -0.689 |
collapse | -0.651 |
Token | recognized |
Token ID | 8018 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
ippers | +0.858 |
banks | +0.844 |
bles | +0.823 |
pouch | +0.813 |
ittens | +0.768 |
Neg logit contributions | |
Acc | -0.987 |
Acc | -0.903 |
solved | -0.771 |
fixed | -0.764 |
collapse | -0.717 |
Token | stick |
Token ID | 4859 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
ippers | +0.703 |
banks | +0.692 |
bles | +0.672 |
pouch | +0.661 |
kins | +0.618 |
Neg logit contributions | |
Acc | -1.070 |
Acc | -0.991 |
solved | -0.861 |
fixed | -0.855 |
collapse | -0.811 |
Token | in |
Token ID | 287 |
Feature activation | +9.18e-01 |
Pos logit contributions | |
ippers | +0.561 |
banks | +0.555 |
bles | +0.542 |
pouch | +0.537 |
ters | +0.505 |
Neg logit contributions | |
Acc | -0.622 |
Acc | -0.569 |
solved | -0.478 |
fixed | -0.472 |
collapse | -0.446 |
Token | its |
Token ID | 663 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
ippers | +1.844 |
banks | +1.801 |
bles | +1.789 |
pouch | +1.744 |
ittens | +1.701 |
Neg logit contributions | |
Acc | -1.233 |
Acc | -1.103 |
solved | -0.895 |
fixed | -0.889 |
collapse | -0.826 |
Token | mouth |
Token ID | 5422 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
ippers | +0.956 |
banks | +0.925 |
bles | +0.912 |
pouch | +0.857 |
itive | +0.836 |
Neg logit contributions | |
Acc | -1.901 |
Acc | -1.784 |
solved | -1.584 |
fixed | -1.578 |
collapse | -1.496 |
Token | ! |
Token ID | 0 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
ippers | +0.747 |
banks | +0.738 |
bles | +0.721 |
pouch | +0.712 |
kins | +0.674 |
Neg logit contributions | |
Acc | -0.773 |
Acc | -0.705 |
solved | -0.591 |
fixed | -0.585 |
collapse | -0.549 |
Token | The |
Token ID | 383 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
ippers | +1.572 |
banks | +1.537 |
bles | +1.524 |
pouch | +1.483 |
itive | +1.452 |
Neg logit contributions | |
Acc | -1.517 |
Acc | -1.396 |
solved | -1.154 |
fixed | -1.148 |
collapse | -1.057 |
Token | two |
Token ID | 734 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
ippers | +6.261 |
angers | +5.821 |
banks | +5.539 |
sters | +5.529 |
ates | +5.364 |
Neg logit contributions | |
Acc | -7.911 |
fixed | -7.071 |
Acc | -7.008 |
cal | -6.905 |
thunder | -6.653 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
ippers | +3.508 |
banks | +3.318 |
robe | +3.229 |
angers | +3.133 |
pouch | +3.034 |
Neg logit contributions | |
Acc | -7.002 |
fixed | -6.445 |
Acc | -6.381 |
solved | -6.280 |
cal | -6.075 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
banks | +0.229 |
ippers | +0.229 |
bles | +0.222 |
pouch | +0.218 |
ters | +0.204 |
Neg logit contributions | |
Acc | -0.361 |
Acc | -0.333 |
solved | -0.286 |
fixed | -0.284 |
collapse | -0.274 |
Token | and |
Token ID | 290 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
Acc | +0.459 |
Acc | +0.424 |
solved | +0.357 |
fixed | +0.351 |
collapse | +0.348 |
Neg logit contributions | |
banks | -0.354 |
ippers | -0.352 |
bles | -0.348 |
pouch | -0.334 |
ters | -0.328 |
Token | laughed |
Token ID | 13818 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
ippers | +1.221 |
banks | +1.194 |
bles | +1.170 |
pouch | +1.156 |
ittens | +1.104 |
Neg logit contributions | |
Acc | -1.177 |
Acc | -1.070 |
solved | -0.904 |
fixed | -0.896 |
collapse | -0.825 |
Token | fox |
Token ID | 21831 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
ippers | +0.496 |
banks | +0.490 |
bles | +0.474 |
pouch | +0.471 |
ittens | +0.436 |
Neg logit contributions | |
Acc | -0.788 |
Acc | -0.728 |
solved | -0.632 |
fixed | -0.626 |
collapse | -0.597 |
Token | coming |
Token ID | 2406 |
Feature activation | -8.45e-01 |
Pos logit contributions | |
ippers | +0.385 |
banks | +0.381 |
bles | +0.370 |
pouch | +0.368 |
ters | +0.341 |
Neg logit contributions | |
Acc | -0.548 |
Acc | -0.507 |
solved | -0.435 |
fixed | -0.431 |
collapse | -0.410 |
Token | towards |
Token ID | 3371 |
Feature activation | +6.39e-01 |
Pos logit contributions | |
Acc | +1.507 |
Acc | +1.405 |
solved | +1.171 |
fixed | +1.147 |
collapse | +1.060 |
Neg logit contributions | |
banks | -1.329 |
pouch | -1.290 |
bles | -1.285 |
ippers | -1.284 |
ters | -1.220 |
Token | her |
Token ID | 607 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ippers | +1.068 |
banks | +1.046 |
bles | +1.032 |
pouch | +1.012 |
kins | +0.963 |
Neg logit contributions | |
Acc | -1.074 |
Acc | -0.977 |
solved | -0.822 |
fixed | -0.817 |
collapse | -0.760 |
Token | . |
Token ID | 13 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ippers | +1.668 |
bles | +1.594 |
banks | +1.593 |
pouch | +1.529 |
kins | +1.513 |
Neg logit contributions | |
Acc | -1.832 |
Acc | -1.674 |
solved | -1.444 |
fixed | -1.434 |
collapse | -1.353 |
Token | The |
Token ID | 383 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
ippers | +2.293 |
bles | +2.248 |
banks | +2.179 |
kins | +2.096 |
itive | +2.083 |
Neg logit contributions | |
Acc | -2.229 |
Acc | -2.048 |
fixed | -1.725 |
solved | -1.690 |
bell | -1.558 |
Token | fox |
Token ID | 21831 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
ippers | +7.026 |
angers | +6.463 |
kins | +6.304 |
banks | +6.126 |
bles | +6.014 |
Neg logit contributions | |
Acc | -7.671 |
cal | -6.824 |
solved | -6.556 |
fixed | -6.478 |
Acc | -6.456 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
ippers | +1.089 |
banks | +1.062 |
bles | +1.033 |
pouch | +1.017 |
ittens | +0.962 |
Neg logit contributions | |
Acc | -1.306 |
Acc | -1.200 |
solved | -1.037 |
fixed | -1.027 |
collapse | -0.960 |
Token | to |
Token ID | 284 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
Acc | +0.533 |
Acc | +0.484 |
solved | +0.380 |
fixed | +0.373 |
collapse | +0.366 |
Neg logit contributions | |
banks | -0.618 |
pouch | -0.603 |
ippers | -0.599 |
bles | -0.582 |
friends | -0.578 |
Token | catch |
Token ID | 4929 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
banks | +0.036 |
ippers | +0.036 |
pouch | +0.035 |
bles | +0.034 |
ittens | +0.032 |
Neg logit contributions | |
Acc | -0.041 |
Acc | -0.037 |
solved | -0.031 |
fixed | -0.030 |
collapse | -0.028 |
Token | Betsy |
Token ID | 45775 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ippers | +0.927 |
banks | +0.912 |
bles | +0.893 |
pouch | +0.880 |
kins | +0.835 |
Neg logit contributions | |
Acc | -0.966 |
Acc | -0.880 |
solved | -0.742 |
fixed | -0.736 |
collapse | -0.688 |
Token | wanted |
Token ID | 2227 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
ippers | +1.231 |
banks | +1.195 |
bles | +1.163 |
pouch | +1.148 |
ittens | +1.092 |
Neg logit contributions | |
Acc | -1.482 |
Acc | -1.364 |
solved | -1.180 |
fixed | -1.168 |
collapse | -1.092 |
Token | to |
Token ID | 284 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
Acc | +1.588 |
Acc | +1.451 |
solved | +1.121 |
fixed | +1.120 |
collapse | +1.096 |
Neg logit contributions | |
banks | -1.584 |
friends | -1.526 |
pouch | -1.525 |
ippers | -1.523 |
bles | -1.496 |
Token | catch |
Token ID | 4929 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
Acc | +0.344 |
Acc | +0.311 |
solved | +0.257 |
fixed | +0.254 |
collapse | +0.236 |
Neg logit contributions | |
banks | -0.294 |
ippers | -0.289 |
pouch | -0.283 |
bles | -0.278 |
friends | -0.265 |
Token | him |
Token ID | 683 |
Feature activation | +7.69e-01 |
Pos logit contributions | |
ippers | +0.870 |
banks | +0.858 |
bles | +0.838 |
pouch | +0.829 |
kins | +0.785 |
Neg logit contributions | |
Acc | -0.901 |
Acc | -0.821 |
solved | -0.691 |
fixed | -0.684 |
collapse | -0.641 |
Token | . |
Token ID | 13 |
Feature activation | +2.71e-03 |
Pos logit contributions | |
ippers | +1.324 |
banks | +1.297 |
bles | +1.266 |
pouch | +1.255 |
ittens | +1.201 |
Neg logit contributions | |
Acc | -1.238 |
Acc | -1.120 |
solved | -0.944 |
fixed | -0.935 |
collapse | -0.872 |
Token | The |
Token ID | 383 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
banks | +0.004 |
ippers | +0.004 |
pouch | +0.004 |
bles | +0.004 |
ters | +0.004 |
Neg logit contributions | |
Acc | -0.005 |
Acc | -0.004 |
solved | -0.004 |
fixed | -0.004 |
collapse | -0.003 |
Token | duck |
Token ID | 22045 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
ippers | +6.388 |
angers | +6.169 |
banks | +5.652 |
kins | +5.427 |
owers | +5.400 |
Neg logit contributions | |
Acc | -7.836 |
cal | -7.527 |
Acc | -6.930 |
fixed | -6.852 |
solved | -6.711 |
Token | was |
Token ID | 373 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
ippers | +1.361 |
banks | +1.304 |
bles | +1.276 |
pouch | +1.259 |
ittens | +1.205 |
Neg logit contributions | |
Acc | -1.610 |
Acc | -1.471 |
solved | -1.279 |
fixed | -1.272 |
collapse | -1.179 |
Token | scared |
Token ID | 12008 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
ippers | +1.007 |
banks | +0.986 |
bles | +0.958 |
pouch | +0.938 |
ittens | +0.893 |
Neg logit contributions | |
Acc | -1.336 |
Acc | -1.229 |
solved | -1.069 |
fixed | -1.061 |
collapse | -0.999 |
Token | and |
Token ID | 290 |
Feature activation | +5.63e-01 |
Pos logit contributions | |
Acc | +0.830 |
Acc | +0.770 |
solved | +0.661 |
fixed | +0.649 |
collapse | +0.625 |
Neg logit contributions | |
ippers | -0.517 |
banks | -0.510 |
bles | -0.503 |
pouch | -0.492 |
ters | -0.482 |
Token | did |
Token ID | 750 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
ippers | +0.847 |
banks | +0.831 |
bles | +0.811 |
pouch | +0.800 |
ittens | +0.754 |
Neg logit contributions | |
Acc | -1.037 |
Acc | -0.952 |
solved | -0.815 |
fixed | -0.808 |
collapse | -0.763 |
Token | actus |
Token ID | 34144 |
Feature activation | +8.28e-01 |
Pos logit contributions | |
Acc | +0.929 |
Acc | +0.878 |
solved | +0.785 |
fixed | +0.773 |
collapse | +0.759 |
Neg logit contributions | |
ippers | -0.323 |
banks | -0.297 |
bles | -0.293 |
pouch | -0.288 |
ittens | -0.270 |
Token | as |
Token ID | 355 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
ippers | +1.442 |
banks | +1.418 |
bles | +1.386 |
pouch | +1.368 |
kins | +1.306 |
Neg logit contributions | |
Acc | -1.314 |
Acc | -1.184 |
solved | -1.002 |
fixed | -0.995 |
collapse | -0.911 |
Token | a |
Token ID | 257 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
Acc | +0.166 |
Acc | +0.154 |
solved | +0.128 |
fixed | +0.128 |
collapse | +0.124 |
Neg logit contributions | |
ippers | -0.130 |
banks | -0.126 |
bles | -0.126 |
pouch | -0.122 |
friends | -0.117 |
Token | pet |
Token ID | 4273 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Acc | +0.159 |
Acc | +0.148 |
solved | +0.129 |
fixed | +0.127 |
collapse | +0.123 |
Neg logit contributions | |
ippers | -0.086 |
banks | -0.086 |
pouch | -0.084 |
bles | -0.082 |
couple | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
ippers | +2.244 |
banks | +2.189 |
bles | +2.176 |
pouch | +2.055 |
kins | +2.048 |
Neg logit contributions | |
Acc | -2.104 |
Acc | -1.916 |
solved | -1.641 |
fixed | -1.621 |
bell | -1.482 |
Token | The |
Token ID | 383 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
ippers | +0.685 |
banks | +0.676 |
bles | +0.657 |
pouch | +0.650 |
ittens | +0.610 |
Neg logit contributions | |
Acc | -0.866 |
Acc | -0.795 |
solved | -0.681 |
fixed | -0.675 |
collapse | -0.638 |
Token | fox |
Token ID | 21831 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
ippers | +8.437 |
angers | +7.679 |
bles | +7.578 |
kins | +7.544 |
ters | +7.254 |
Neg logit contributions | |
Acc | -6.313 |
Acc | -5.406 |
cal | -5.030 |
fixed | -4.715 |
solved | -4.697 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
ippers | +0.806 |
banks | +0.792 |
bles | +0.772 |
pouch | +0.762 |
ittens | +0.716 |
Neg logit contributions | |
Acc | -1.017 |
Acc | -0.934 |
solved | -0.803 |
fixed | -0.796 |
collapse | -0.749 |
Token | really |
Token ID | 1107 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
Acc | +0.203 |
Acc | +0.184 |
solved | +0.148 |
fixed | +0.145 |
collapse | +0.138 |
Neg logit contributions | |
banks | -0.245 |
ippers | -0.244 |
bles | -0.239 |
pouch | -0.237 |
friends | -0.227 |
Token | hard |
Token ID | 1327 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
ippers | +1.288 |
banks | +1.274 |
bles | +1.246 |
pouch | +1.238 |
kins | +1.184 |
Neg logit contributions | |
Acc | -0.834 |
Acc | -0.735 |
solved | -0.584 |
fixed | -0.575 |
collapse | -0.525 |
Token | to |
Token ID | 284 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
ippers | +3.203 |
banks | +3.170 |
pouch | +3.121 |
bles | +3.114 |
kins | +3.030 |
Neg logit contributions | |
Acc | -2.312 |
Acc | -2.065 |
fixed | -1.760 |
solved | -1.746 |
collapsing | -1.513 |
Token | to |
Token ID | 284 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
Acc | +1.547 |
Acc | +1.443 |
solved | +1.157 |
fixed | +1.118 |
bell | +1.102 |
Neg logit contributions | |
banks | -1.645 |
ippers | -1.642 |
bles | -1.621 |
pouch | -1.592 |
friends | -1.562 |
Token | grab |
Token ID | 5552 |
Feature activation | -4.32e-01 |
Pos logit contributions | |
Acc | +0.340 |
Acc | +0.310 |
solved | +0.252 |
fixed | +0.246 |
bell | +0.233 |
Neg logit contributions | |
banks | -0.319 |
ippers | -0.313 |
bles | -0.305 |
pouch | -0.304 |
ters | -0.287 |
Token | it |
Token ID | 340 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
Acc | +0.841 |
Acc | +0.778 |
solved | +0.636 |
fixed | +0.620 |
bell | +0.605 |
Neg logit contributions | |
ippers | -0.632 |
banks | -0.618 |
bles | -0.612 |
pouch | -0.599 |
ters | -0.574 |
Token | . |
Token ID | 13 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
Acc | +0.577 |
Acc | +0.531 |
solved | +0.436 |
fixed | +0.423 |
collapse | +0.403 |
Neg logit contributions | |
banks | -0.448 |
ippers | -0.443 |
bles | -0.434 |
pouch | -0.430 |
ters | -0.410 |
Token | But |
Token ID | 887 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
Acc | +1.508 |
Acc | +1.336 |
solved | +1.152 |
fixed | +1.131 |
collapse | +1.103 |
Neg logit contributions | |
ippers | -1.028 |
banks | -1.011 |
bles | -0.942 |
pouch | -0.930 |
angers | -0.908 |
Token | the |
Token ID | 262 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
Acc | +0.025 |
Acc | +0.022 |
solved | +0.018 |
fixed | +0.018 |
collapse | +0.017 |
Neg logit contributions | |
ippers | -0.022 |
banks | -0.021 |
bles | -0.021 |
pouch | -0.020 |
kins | -0.020 |
Token | sk |
Token ID | 1341 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ippers | +7.096 |
kins | +7.021 |
bles | +6.718 |
angers | +6.671 |
itive | +6.666 |
Neg logit contributions | |
Acc | -6.864 |
Acc | -6.429 |
cal | -6.166 |
collapse | -6.046 |
� | -5.989 |
Token | unk |
Token ID | 2954 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
banks | +1.909 |
ippers | +1.814 |
pouch | +1.796 |
couple | +1.758 |
bles | +1.733 |
Neg logit contributions | |
Acc | -3.951 |
Acc | -3.684 |
solved | -3.205 |
fixed | -3.116 |
collapse | -3.054 |
Token | was |
Token ID | 373 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
Acc | +0.285 |
Acc | +0.263 |
solved | +0.216 |
fixed | +0.213 |
collapse | +0.206 |
Neg logit contributions | |
banks | -0.218 |
ippers | -0.216 |
bles | -0.212 |
pouch | -0.211 |
ters | -0.197 |
Token | too |
Token ID | 1165 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
Acc | +0.262 |
Acc | +0.246 |
solved | +0.198 |
fixed | +0.196 |
bell | +0.190 |
Neg logit contributions | |
ippers | -0.208 |
banks | -0.207 |
bles | -0.206 |
pouch | -0.204 |
ters | -0.191 |
Token | fast |
Token ID | 3049 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
Acc | +0.574 |
Acc | +0.532 |
solved | +0.425 |
fixed | +0.413 |
collapse | +0.393 |
Neg logit contributions | |
ippers | -0.565 |
bles | -0.557 |
banks | -0.553 |
pouch | -0.548 |
friends | -0.514 |
Token | and |
Token ID | 290 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ippers | +0.040 |
banks | +0.040 |
bles | +0.039 |
pouch | +0.038 |
ters | +0.036 |
Neg logit contributions | |
Acc | -0.046 |
Acc | -0.042 |
solved | -0.035 |
fixed | -0.034 |
collapse | -0.033 |
Token | hands |
Token ID | 2832 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
ippers | +2.200 |
banks | +2.137 |
bles | +2.094 |
pouch | +2.044 |
ittens | +1.976 |
Neg logit contributions | |
Acc | -2.140 |
Acc | -1.992 |
solved | -1.703 |
fixed | -1.695 |
collapse | -1.531 |
Token | . |
Token ID | 13 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
ippers | +1.143 |
banks | +1.125 |
bles | +1.099 |
pouch | +1.089 |
ittens | +1.033 |
Neg logit contributions | |
Acc | -1.161 |
Acc | -1.056 |
solved | -0.894 |
fixed | -0.889 |
collapse | -0.827 |
Token | Rachel |
Token ID | 15984 |
Feature activation | -5.62e-02 |
Pos logit contributions | |
ippers | +1.348 |
banks | +1.327 |
bles | +1.297 |
pouch | +1.288 |
itive | +1.229 |
Neg logit contributions | |
Acc | -1.357 |
Acc | -1.242 |
solved | -1.045 |
fixed | -1.024 |
collapse | -0.955 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Acc | +0.102 |
Acc | +0.094 |
solved | +0.078 |
fixed | +0.077 |
collapse | +0.076 |
Neg logit contributions | |
banks | -0.084 |
ippers | -0.084 |
bles | -0.082 |
pouch | -0.082 |
kins | -0.077 |
Token | the |
Token ID | 262 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
ippers | +1.858 |
banks | +1.835 |
bles | +1.789 |
pouch | +1.754 |
kins | +1.687 |
Neg logit contributions | |
Acc | -1.800 |
Acc | -1.634 |
solved | -1.354 |
fixed | -1.353 |
collapse | -1.226 |
Token | feeling |
Token ID | 4203 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
sters | +7.496 |
ippers | +7.469 |
bles | +7.228 |
itive | +6.887 |
kins | +6.823 |
Neg logit contributions | |
Acc | -6.948 |
Acc | -6.365 |
fixed | -6.153 |
cal | -5.967 |
solved | -5.704 |
Token | of |
Token ID | 286 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
Acc | +0.024 |
Acc | +0.022 |
solved | +0.019 |
fixed | +0.019 |
collapse | +0.018 |
Neg logit contributions | |
ippers | -0.015 |
banks | -0.014 |
bles | -0.014 |
pouch | -0.014 |
kins | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
ippers | +1.502 |
banks | +1.477 |
bles | +1.452 |
pouch | +1.422 |
kins | +1.360 |
Neg logit contributions | |
Acc | -1.480 |
Acc | -1.351 |
solved | -1.125 |
fixed | -1.123 |
collapse | -1.050 |
Token | lot |
Token ID | 1256 |
Feature activation | -7.72e-02 |
Pos logit contributions | |
ippers | +5.286 |
bles | +5.195 |
sters | +5.187 |
angers | +4.877 |
itive | +4.857 |
Neg logit contributions | |
Acc | -6.629 |
Acc | -6.122 |
fixed | -5.704 |
cal | -5.602 |
solved | -5.302 |
Token | ion |
Token ID | 295 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
Acc | +0.140 |
Acc | +0.128 |
solved | +0.109 |
fixed | +0.106 |
collapse | +0.101 |
Neg logit contributions | |
ippers | -0.118 |
banks | -0.117 |
bles | -0.114 |
pouch | -0.113 |
ters | -0.108 |
Token | She |
Token ID | 1375 |
Feature activation | -6.33e-01 |
Pos logit contributions | |
ippers | +0.250 |
banks | +0.244 |
bles | +0.238 |
pouch | +0.237 |
ittens | +0.223 |
Neg logit contributions | |
Acc | -0.332 |
Acc | -0.303 |
solved | -0.263 |
fixed | -0.260 |
collapse | -0.250 |
Token | said |
Token ID | 531 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
Acc | +1.446 |
Acc | +1.355 |
collapse | +1.176 |
collapsing | +1.132 |
solved | +1.130 |
Neg logit contributions | |
bles | -0.640 |
pouch | -0.624 |
ippers | -0.622 |
friends | -0.611 |
banks | -0.609 |
Token | they |
Token ID | 484 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
ippers | +0.504 |
banks | +0.496 |
bles | +0.484 |
pouch | +0.475 |
ittens | +0.447 |
Neg logit contributions | |
Acc | -0.727 |
Acc | -0.670 |
solved | -0.581 |
fixed | -0.575 |
collapse | -0.548 |
Token | could |
Token ID | 714 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
ippers | +0.583 |
banks | +0.575 |
bles | +0.563 |
pouch | +0.554 |
kins | +0.523 |
Neg logit contributions | |
Acc | -0.706 |
Acc | -0.649 |
solved | -0.550 |
fixed | -0.544 |
collapse | -0.520 |
Token | have |
Token ID | 423 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
Acc | +1.544 |
Acc | +1.372 |
solved | +1.155 |
fixed | +1.100 |
anne | +1.084 |
Neg logit contributions | |
banks | -1.295 |
ippers | -1.294 |
bles | -1.245 |
pouch | -1.209 |
friends | -1.201 |
Token | a |
Token ID | 257 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
Acc | +1.858 |
Acc | +1.745 |
bell | +1.420 |
aling | +1.408 |
collapse | +1.403 |
Neg logit contributions | |
banks | -1.294 |
friends | -1.255 |
ippers | -1.230 |
bles | -1.223 |
pouch | -1.185 |
Token | snack |
Token ID | 26906 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
Acc | +7.159 |
Acc | +6.660 |
aling | +6.222 |
anne | +5.874 |
collapse | +5.765 |
Neg logit contributions | |
friends | -3.676 |
pair | -3.617 |
couple | -3.386 |
pouch | -3.220 |
ones | -3.191 |
Token | and |
Token ID | 290 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
Acc | +0.968 |
Acc | +0.891 |
solved | +0.723 |
fixed | +0.697 |
collapse | +0.695 |
Neg logit contributions | |
ippers | -0.778 |
banks | -0.768 |
bles | -0.757 |
pouch | -0.737 |
kins | -0.719 |
Token | read |
Token ID | 1100 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Acc | +0.992 |
Acc | +0.903 |
solved | +0.749 |
fixed | +0.731 |
collapse | +0.722 |
Neg logit contributions | |
banks | -0.709 |
ippers | -0.690 |
pouch | -0.670 |
bles | -0.667 |
friends | -0.656 |
Token | a |
Token ID | 257 |
Feature activation | -9.27e-01 |
Pos logit contributions | |
Acc | +2.773 |
Acc | +2.532 |
solved | +2.130 |
collapse | +2.094 |
aling | +2.078 |
Neg logit contributions | |
banks | -1.759 |
ippers | -1.735 |
bles | -1.677 |
ones | -1.670 |
friends | -1.670 |
Token | story |
Token ID | 1621 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
Acc | +2.266 |
Acc | +2.102 |
solved | +1.859 |
fixed | +1.817 |
aling | +1.800 |
Neg logit contributions | |
ippers | -0.869 |
pouch | -0.838 |
banks | -0.816 |
couple | -0.807 |
pair | -0.803 |
Token | my |
Token ID | 1820 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
Acc | +1.189 |
Acc | +1.067 |
collapse | +0.788 |
solved | +0.783 |
collapsing | +0.753 |
Neg logit contributions | |
ippers | -1.853 |
bles | -1.834 |
banks | -1.829 |
pouch | -1.824 |
friends | -1.796 |
Token | was |
Token ID | 373 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
Acc | +1.247 |
Acc | +1.164 |
collapse | +0.953 |
solved | +0.949 |
bell | +0.938 |
Neg logit contributions | |
banks | -0.882 |
pouch | -0.861 |
ippers | -0.858 |
bles | -0.844 |
friends | -0.824 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
Acc | +0.836 |
Acc | +0.779 |
solved | +0.654 |
fixed | +0.635 |
collapse | +0.633 |
Neg logit contributions | |
ippers | -0.582 |
pouch | -0.576 |
bles | -0.575 |
banks | -0.571 |
friends | -0.556 |
Token | . |
Token ID | 13 |
Feature activation | -7.89e-01 |
Pos logit contributions | |
Acc | +0.712 |
Acc | +0.648 |
solved | +0.529 |
fixed | +0.520 |
collapse | +0.501 |
Neg logit contributions | |
ippers | -0.743 |
banks | -0.727 |
bles | -0.721 |
pouch | -0.711 |
friends | -0.704 |
Token | The |
Token ID | 383 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
Acc | +1.554 |
Acc | +1.387 |
solved | +1.260 |
fixed | +1.220 |
collapse | +1.204 |
Neg logit contributions | |
ippers | -1.054 |
banks | -1.038 |
pouch | -1.021 |
bles | -0.995 |
ittens | -0.983 |
Token | end |
Token ID | 886 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
Acc | +0.454 |
Acc | +0.420 |
solved | +0.360 |
fixed | +0.354 |
collapse | +0.342 |
Neg logit contributions | |
ippers | -0.315 |
banks | -0.314 |
bles | -0.306 |
pouch | -0.303 |
friends | -0.292 |
Token | . |
Token ID | 13 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
Acc | +7.718 |
Acc | +7.014 |
aling | +6.256 |
solved | +5.993 |
collapse | +5.992 |
Neg logit contributions | |
friends | -2.758 |
things | -2.703 |
ippers | -2.497 |
pouch | -2.481 |
ittens | -2.478 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
Acc | +1.731 |
Acc | +1.585 |
solved | +1.405 |
fixed | +1.363 |
collapse | +1.354 |
Neg logit contributions | |
ippers | -0.998 |
pouch | -0.980 |
banks | -0.961 |
bles | -0.933 |
ittens | -0.926 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
ippers | +0.208 |
banks | +0.205 |
bles | +0.199 |
pouch | +0.198 |
ters | +0.185 |
Neg logit contributions | |
Acc | -0.281 |
Acc | -0.259 |
solved | -0.220 |
fixed | -0.217 |
collapse | -0.208 |
Token | upon |
Token ID | 2402 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
ippers | +0.023 |
bles | +0.021 |
banks | +0.021 |
pouch | +0.020 |
ittens | +0.019 |
Neg logit contributions | |
Acc | -0.071 |
Acc | -0.067 |
solved | -0.060 |
fixed | -0.059 |
collapse | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
Acc | +0.124 |
Acc | +0.116 |
solved | +0.097 |
fixed | +0.096 |
collapse | +0.093 |
Neg logit contributions | |
ippers | -0.095 |
banks | -0.094 |
bles | -0.092 |
pouch | -0.090 |
friends | -0.085 |
Token | eating |
Token ID | 6600 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
ippers | +0.181 |
banks | +0.181 |
bles | +0.175 |
pouch | +0.172 |
kins | +0.163 |
Neg logit contributions | |
Acc | -0.244 |
Acc | -0.223 |
solved | -0.191 |
fixed | -0.189 |
collapse | -0.181 |
Token | cher |
Token ID | 21382 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
Acc | +1.587 |
Acc | +1.465 |
solved | +1.255 |
fixed | +1.205 |
collapse | +1.195 |
Neg logit contributions | |
banks | -1.021 |
ippers | -1.001 |
bles | -0.972 |
pouch | -0.938 |
kins | -0.933 |
Token | ries |
Token ID | 1678 |
Feature activation | -9.17e-01 |
Pos logit contributions | |
ippers | +0.502 |
banks | +0.492 |
bles | +0.480 |
pouch | +0.469 |
kins | +0.441 |
Neg logit contributions | |
Acc | -0.799 |
Acc | -0.743 |
solved | -0.645 |
fixed | -0.638 |
collapse | -0.610 |
Token | . |
Token ID | 13 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
Acc | +1.816 |
Acc | +1.711 |
solved | +1.376 |
fixed | +1.320 |
collapse | +1.313 |
Neg logit contributions | |
banks | -1.316 |
ippers | -1.294 |
bles | -1.292 |
pouch | -1.242 |
ters | -1.221 |
Token | The |
Token ID | 383 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
Acc | +1.218 |
Acc | +1.103 |
solved | +0.996 |
fixed | +0.972 |
collapse | +0.957 |
Neg logit contributions | |
banks | -0.804 |
ippers | -0.797 |
pouch | -0.780 |
ittens | -0.752 |
bles | -0.748 |
Token | end |
Token ID | 886 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
banks | +0.049 |
ippers | +0.048 |
bles | +0.047 |
pouch | +0.046 |
kins | +0.044 |
Neg logit contributions | |
Acc | -0.065 |
Acc | -0.060 |
solved | -0.051 |
fixed | -0.050 |
collapse | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -9.81e-01 |
Pos logit contributions | |
Acc | +7.297 |
Acc | +6.622 |
aling | +5.869 |
solved | +5.584 |
collapse | +5.572 |
Neg logit contributions | |
things | -3.082 |
friends | -3.035 |
ones | -2.766 |
ippers | -2.713 |
ittens | -2.697 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
Acc | +2.062 |
Acc | +1.902 |
solved | +1.689 |
collapse | +1.646 |
fixed | +1.640 |
Neg logit contributions | |
pouch | -1.128 |
ippers | -1.118 |
banks | -1.100 |
ittens | -1.077 |
friends | -1.031 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
Acc | +0.296 |
Acc | +0.275 |
solved | +0.232 |
fixed | +0.229 |
collapse | +0.222 |
Neg logit contributions | |
ippers | -0.202 |
banks | -0.199 |
pouch | -0.193 |
bles | -0.192 |
robe | -0.179 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.88e-04 |
Pos logit contributions | |
ippers | +0.153 |
banks | +0.146 |
bles | +0.145 |
pouch | +0.138 |
ittens | +0.125 |
Neg logit contributions | |
Acc | -0.471 |
Acc | -0.444 |
solved | -0.396 |
fixed | -0.392 |
collapse | -0.380 |
Token | a |
Token ID | 257 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
ippers | +0.000 |
banks | +0.000 |
bles | +0.000 |
pouch | +0.000 |
itive | +0.000 |
Neg logit contributions | |
Acc | -0.000 |
Acc | -0.000 |
solved | -0.000 |
fixed | -0.000 |
collapse | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
ippers | +0.672 |
banks | +0.663 |
bles | +0.647 |
pouch | +0.641 |
kins | +0.606 |
Neg logit contributions | |
Acc | -0.684 |
Acc | -0.622 |
solved | -0.522 |
fixed | -0.516 |
collapse | -0.485 |
Token | patience |
Token ID | 16336 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
ippers | +0.075 |
banks | +0.075 |
pouch | +0.072 |
bles | +0.072 |
friends | +0.068 |
Neg logit contributions | |
Acc | -0.080 |
Acc | -0.072 |
solved | -0.061 |
fixed | -0.060 |
collapse | -0.056 |
Token | ." |
Token ID | 526 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
ippers | +0.387 |
banks | +0.382 |
bles | +0.373 |
pouch | +0.368 |
ters | +0.349 |
Neg logit contributions | |
Acc | -0.428 |
Acc | -0.389 |
solved | -0.329 |
fixed | -0.323 |
collapse | -0.308 |
Token |
|
Token ID | 198 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
Acc | +0.279 |
Acc | +0.253 |
solved | +0.224 |
fixed | +0.222 |
bell | +0.206 |
Neg logit contributions | |
ippers | -0.221 |
banks | -0.221 |
pouch | -0.213 |
bles | -0.209 |
ittens | -0.203 |
Token |
|
Token ID | 198 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
Acc | +1.093 |
Acc | +1.001 |
solved | +0.887 |
fixed | +0.875 |
bell | +0.842 |
Neg logit contributions | |
banks | -0.697 |
ippers | -0.693 |
bles | -0.672 |
pouch | -0.667 |
ittens | -0.648 |
Token | S |
Token ID | 50 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
Acc | +0.465 |
Acc | +0.422 |
fixed | +0.349 |
solved | +0.348 |
bell | +0.326 |
Neg logit contributions | |
banks | -0.539 |
ippers | -0.531 |
bles | -0.522 |
pouch | -0.522 |
ittens | -0.504 |
Token | ue |
Token ID | 518 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Acc | +4.576 |
Acc | +4.160 |
aling | +3.464 |
fixed | +3.287 |
solved | +3.273 |
Neg logit contributions | |
bles | -5.661 |
ittens | -5.660 |
ippers | -5.623 |
banks | -5.242 |
sters | -5.191 |
Token | nods |
Token ID | 34833 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Acc | +2.280 |
Acc | +2.127 |
solved | +1.790 |
fixed | +1.786 |
collapse | +1.722 |
Neg logit contributions | |
banks | -1.748 |
bles | -1.689 |
ippers | -1.676 |
pouch | -1.655 |
friends | -1.642 |
Token | and |
Token ID | 290 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
Acc | +0.444 |
Acc | +0.412 |
solved | +0.357 |
fixed | +0.349 |
collapse | +0.332 |
Neg logit contributions | |
ippers | -0.364 |
banks | -0.352 |
bles | -0.351 |
pouch | -0.342 |
kins | -0.329 |
Token | says |
Token ID | 1139 |
Feature activation | -7.26e-01 |
Pos logit contributions | |
Acc | +2.211 |
Acc | +1.987 |
solved | +1.722 |
fixed | +1.676 |
bell | +1.662 |
Neg logit contributions | |
banks | -1.932 |
ippers | -1.866 |
pouch | -1.848 |
bles | -1.812 |
friends | -1.804 |
Token | , |
Token ID | 11 |
Feature activation | -5.77e-03 |
Pos logit contributions | |
Acc | +1.481 |
Acc | +1.340 |
solved | +1.207 |
fixed | +1.186 |
aling | +1.141 |
Neg logit contributions | |
ippers | -0.963 |
bles | -0.926 |
banks | -0.871 |
pouch | -0.864 |
friends | -0.862 |
Token | , |
Token ID | 11 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
Acc | +0.476 |
Acc | +0.429 |
solved | +0.362 |
fixed | +0.359 |
collapse | +0.348 |
Neg logit contributions | |
ippers | -0.403 |
bles | -0.396 |
banks | -0.394 |
pouch | -0.382 |
friends | -0.372 |
Token | " |
Token ID | 366 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
Acc | +0.384 |
Acc | +0.356 |
solved | +0.313 |
fixed | +0.309 |
collapse | +0.302 |
Neg logit contributions | |
ippers | -0.227 |
banks | -0.225 |
bles | -0.220 |
pouch | -0.218 |
couple | -0.201 |
Token | That |
Token ID | 2504 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
Acc | +1.043 |
Acc | +0.992 |
fixed | +0.812 |
solved | +0.809 |
collapse | +0.802 |
Neg logit contributions | |
ippers | -0.904 |
bles | -0.879 |
banks | -0.875 |
pouch | -0.863 |
ters | -0.860 |
Token | was |
Token ID | 373 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
Acc | +1.644 |
Acc | +1.486 |
solved | +1.240 |
fixed | +1.219 |
collapse | +1.218 |
Neg logit contributions | |
banks | -1.119 |
pouch | -1.062 |
ippers | -1.040 |
bles | -1.039 |
friends | -1.022 |
Token | such |
Token ID | 884 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Acc | +3.423 |
Acc | +3.124 |
bell | +2.614 |
anne | +2.526 |
collapse | +2.525 |
Neg logit contributions | |
friends | -1.865 |
bles | -1.848 |
banks | -1.823 |
pouch | -1.797 |
ters | -1.732 |
Token | a |
Token ID | 257 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
ippers | +0.343 |
banks | +0.339 |
bles | +0.331 |
pouch | +0.328 |
ters | +0.309 |
Neg logit contributions | |
Acc | -0.405 |
Acc | -0.372 |
solved | -0.311 |
fixed | -0.310 |
collapse | -0.295 |
Token | brave |
Token ID | 14802 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Acc | +6.337 |
Acc | +5.720 |
anne | +5.194 |
collapse | +5.166 |
bell | +5.113 |
Neg logit contributions | |
friends | -3.851 |
pouch | -3.674 |
pair | -3.589 |
couple | -3.585 |
things | -3.472 |
Token | brick |
Token ID | 17214 |
Feature activation | -8.60e-01 |
Pos logit contributions | |
Acc | +3.577 |
Acc | +3.226 |
solved | +2.806 |
aling | +2.786 |
fixed | +2.751 |
Neg logit contributions | |
banks | -1.422 |
ippers | -1.396 |
friends | -1.395 |
pouch | -1.373 |
pair | -1.354 |
Token | to |
Token ID | 284 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
Acc | +1.585 |
Acc | +1.416 |
solved | +1.146 |
fixed | +1.111 |
collapse | +1.102 |
Neg logit contributions | |
banks | -1.329 |
ippers | -1.310 |
pouch | -1.297 |
bles | -1.293 |
friends | -1.252 |
Token | get |
Token ID | 651 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
Acc | +0.761 |
Acc | +0.686 |
solved | +0.555 |
fixed | +0.547 |
bell | +0.529 |
Neg logit contributions | |
banks | -0.635 |
ippers | -0.630 |
bles | -0.614 |
pouch | -0.603 |
friends | -0.579 |
Token | so |
Token ID | 523 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
Acc | +0.859 |
Acc | +0.777 |
solved | +0.642 |
fixed | +0.621 |
collapse | +0.613 |
Neg logit contributions | |
ippers | -0.683 |
banks | -0.677 |
bles | -0.664 |
pouch | -0.655 |
friends | -0.634 |
Token | . |
Token ID | 13 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
Acc | +1.019 |
Acc | +0.938 |
solved | +0.808 |
fixed | +0.800 |
collapse | +0.771 |
Neg logit contributions | |
ippers | -0.714 |
banks | -0.701 |
bles | -0.692 |
friends | -0.683 |
pouch | -0.672 |
Token | They |
Token ID | 1119 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
Acc | +1.086 |
Acc | +0.987 |
solved | +0.883 |
fixed | +0.880 |
collapse | +0.858 |
Neg logit contributions | |
ippers | -0.642 |
banks | -0.639 |
pouch | -0.620 |
bles | -0.604 |
friends | -0.568 |
Token | thought |
Token ID | 1807 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
Acc | +0.268 |
Acc | +0.250 |
solved | +0.206 |
fixed | +0.204 |
collapse | +0.203 |
Neg logit contributions | |
banks | -0.180 |
ippers | -0.177 |
bles | -0.176 |
pouch | -0.174 |
kins | -0.162 |
Token | it |
Token ID | 340 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
Acc | +1.884 |
Acc | +1.713 |
solved | +1.459 |
fixed | +1.430 |
collapse | +1.385 |
Neg logit contributions | |
bles | -1.235 |
ippers | -1.234 |
friends | -1.198 |
banks | -1.185 |
kins | -1.160 |
Token | was |
Token ID | 373 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Acc | +1.440 |
Acc | +1.356 |
collapse | +1.118 |
solved | +1.108 |
aling | +1.101 |
Neg logit contributions | |
banks | -0.857 |
bles | -0.812 |
ippers | -0.800 |
pouch | -0.791 |
robe | -0.776 |
Token | a |
Token ID | 257 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
Acc | +2.297 |
Acc | +2.121 |
bell | +1.801 |
aling | +1.736 |
collapse | +1.735 |
Neg logit contributions | |
bles | -1.635 |
friends | -1.599 |
banks | -1.565 |
kins | -1.552 |
ippers | -1.538 |
Token | good |
Token ID | 922 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
Acc | +6.578 |
Acc | +6.046 |
aling | +5.756 |
anne | +5.682 |
collapse | +5.484 |
Neg logit contributions | |
friends | -3.735 |
pair | -3.446 |
ones | -3.335 |
things | -3.332 |
couple | -3.289 |
Token | day |
Token ID | 1110 |
Feature activation | -7.47e-01 |
Pos logit contributions | |
Acc | +1.089 |
Acc | +1.012 |
solved | +0.898 |
fixed | +0.889 |
collapse | +0.882 |
Neg logit contributions | |
ippers | -0.335 |
banks | -0.318 |
pouch | -0.301 |
bles | -0.298 |
friends | -0.282 |
Token | at |
Token ID | 379 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Acc | +1.490 |
Acc | +1.359 |
solved | +1.113 |
collapse | +1.106 |
fixed | +1.093 |
Neg logit contributions | |
ippers | -1.042 |
banks | -1.016 |
pouch | -1.004 |
friends | -0.978 |
couple | -0.973 |
Token | the |
Token ID | 262 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Acc | +2.506 |
Acc | +2.300 |
bell | +1.922 |
anne | +1.899 |
solved | +1.859 |
Neg logit contributions | |
banks | -1.696 |
bles | -1.686 |
ippers | -1.648 |
friends | -1.639 |
kins | -1.607 |
Token | park |
Token ID | 3952 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
Acc | +2.888 |
Acc | +2.675 |
anne | +2.276 |
solved | +2.251 |
fixed | +2.249 |
Neg logit contributions | |
banks | -1.577 |
pouch | -1.427 |
ones | -1.404 |
ippers | -1.380 |
bles | -1.359 |
Token | . |
Token ID | 13 |
Feature activation | -8.69e-01 |
Pos logit contributions | |
ippers | +0.173 |
banks | +0.170 |
bles | +0.168 |
pouch | +0.164 |
ters | +0.159 |
Neg logit contributions | |
Acc | -0.218 |
Acc | -0.199 |
solved | -0.169 |
fixed | -0.167 |
collapse | -0.158 |
Token | I |
Token ID | 314 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
Acc | +1.537 |
Acc | +1.303 |
solved | +1.239 |
fixed | +1.232 |
collapse | +1.188 |
Neg logit contributions | |
ippers | -1.340 |
pouch | -1.275 |
banks | -1.248 |
bles | -1.235 |
ters | -1.199 |
Token | wanted |
Token ID | 2227 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
Acc | +0.834 |
Acc | +0.767 |
solved | +0.649 |
fixed | +0.635 |
collapse | +0.630 |
Neg logit contributions | |
banks | -0.582 |
ippers | -0.576 |
bles | -0.558 |
pouch | -0.554 |
friends | -0.519 |
Token | to |
Token ID | 284 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Acc | +3.956 |
Acc | +3.531 |
solved | +2.955 |
aling | +2.895 |
collapse | +2.872 |
Neg logit contributions | |
friends | -3.297 |
things | -3.124 |
ones | -3.085 |
pills | -3.065 |
animals | -2.996 |
Token | have |
Token ID | 423 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
Acc | +2.416 |
Acc | +2.125 |
solved | +1.781 |
aling | +1.719 |
fixed | +1.707 |
Neg logit contributions | |
banks | -1.660 |
friends | -1.628 |
ippers | -1.607 |
pouch | -1.553 |
bles | -1.551 |
Token | a |
Token ID | 257 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
Acc | +1.628 |
Acc | +1.494 |
bell | +1.216 |
solved | +1.212 |
collapse | +1.194 |
Neg logit contributions | |
banks | -1.279 |
friends | -1.277 |
ippers | -1.246 |
pouch | -1.217 |
bles | -1.209 |
Token | nice |
Token ID | 3621 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
Acc | +7.081 |
Acc | +6.563 |
aling | +6.168 |
anne | +5.652 |
collapse | +5.648 |
Neg logit contributions | |
friends | -3.243 |
pair | -3.095 |
pouch | -2.944 |
ones | -2.922 |
couple | -2.747 |
Token | view |
Token ID | 1570 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Acc | +2.332 |
Acc | +2.103 |
solved | +1.845 |
aling | +1.816 |
fixed | +1.778 |
Neg logit contributions | |
ippers | -1.109 |
friends | -1.075 |
banks | -1.072 |
pouch | -1.071 |
pair | -1.045 |
Token | too |
Token ID | 1165 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
Acc | +2.597 |
Acc | +2.400 |
solved | +1.980 |
collapse | +1.906 |
fixed | +1.897 |
Neg logit contributions | |
ippers | -1.507 |
bles | -1.490 |
banks | -1.458 |
ters | -1.452 |
pouch | -1.427 |
Token | . |
Token ID | 13 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
Acc | +0.515 |
Acc | +0.473 |
solved | +0.401 |
fixed | +0.393 |
collapse | +0.379 |
Neg logit contributions | |
banks | -0.330 |
ippers | -0.329 |
bles | -0.320 |
pouch | -0.317 |
ters | -0.297 |
Token | But |
Token ID | 887 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
Acc | +1.414 |
Acc | +1.208 |
solved | +1.121 |
fixed | +1.111 |
collapse | +1.086 |
Neg logit contributions | |
ippers | -1.247 |
pouch | -1.207 |
banks | -1.182 |
bles | -1.149 |
ters | -1.121 |
Token | jumped |
Token ID | 11687 |
Feature activation | -9.43e-01 |
Pos logit contributions | |
Acc | +1.414 |
Acc | +1.329 |
solved | +1.048 |
fixed | +1.036 |
collapse | +1.036 |
Neg logit contributions | |
banks | -0.961 |
pouch | -0.929 |
bles | -0.908 |
ippers | -0.892 |
friends | -0.870 |
Token | on |
Token ID | 319 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Acc | +1.955 |
Acc | +1.862 |
solved | +1.510 |
fixed | +1.468 |
collapse | +1.426 |
Neg logit contributions | |
banks | -1.259 |
bles | -1.257 |
ippers | -1.241 |
pouch | -1.204 |
ters | -1.175 |
Token | Zip |
Token ID | 38636 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
Acc | +2.732 |
Acc | +2.549 |
solved | +2.095 |
aling | +2.042 |
fixed | +2.004 |
Neg logit contributions | |
banks | -1.843 |
pouch | -1.720 |
ippers | -1.693 |
friends | -1.669 |
ones | -1.666 |
Token | and |
Token ID | 290 |
Feature activation | -5.70e-04 |
Pos logit contributions | |
Acc | +1.442 |
Acc | +1.353 |
solved | +1.093 |
fixed | +1.061 |
collapse | +1.028 |
Neg logit contributions | |
banks | -1.227 |
pouch | -1.155 |
ippers | -1.153 |
bles | -1.132 |
ters | -1.073 |
Token | they |
Token ID | 484 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
Acc | +0.001 |
Acc | +0.001 |
solved | +0.001 |
fixed | +0.001 |
collapse | +0.001 |
Neg logit contributions | |
banks | -0.001 |
ippers | -0.001 |
pouch | -0.001 |
bles | -0.001 |
ters | -0.001 |
Token | fought |
Token ID | 8350 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
ippers | +0.469 |
banks | +0.469 |
bles | +0.454 |
pouch | +0.449 |
kins | +0.419 |
Neg logit contributions | |
Acc | -0.621 |
Acc | -0.573 |
solved | -0.486 |
fixed | -0.481 |
collapse | -0.457 |
Token | . |
Token ID | 13 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
Acc | +6.705 |
Acc | +6.420 |
anne | +5.045 |
bell | +5.024 |
aling | +4.772 |
Neg logit contributions | |
things | -3.764 |
friends | -3.758 |
banks | -3.676 |
ters | -3.641 |
bles | -3.618 |
Token | The |
Token ID | 383 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
Acc | +0.768 |
Acc | +0.670 |
solved | +0.562 |
fixed | +0.558 |
bell | +0.517 |
Neg logit contributions | |
banks | -0.813 |
ippers | -0.790 |
pouch | -0.776 |
bles | -0.768 |
ittens | -0.741 |
Token | cat |
Token ID | 3797 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
ippers | +0.291 |
banks | +0.289 |
bles | +0.279 |
pouch | +0.278 |
ittens | +0.256 |
Neg logit contributions | |
Acc | -0.527 |
Acc | -0.490 |
solved | -0.427 |
fixed | -0.422 |
collapse | -0.404 |
Token | was |
Token ID | 373 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
Acc | +0.724 |
Acc | +0.675 |
solved | +0.544 |
fixed | +0.537 |
collapse | +0.522 |
Neg logit contributions | |
banks | -0.513 |
bles | -0.494 |
pouch | -0.489 |
ippers | -0.486 |
friends | -0.468 |
Token | very |
Token ID | 845 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
Acc | +0.934 |
Acc | +0.879 |
solved | +0.703 |
bell | +0.695 |
fixed | +0.687 |
Neg logit contributions | |
bles | -0.648 |
banks | -0.641 |
pouch | -0.633 |
ippers | -0.631 |
friends | -0.619 |
Token | The |
Token ID | 464 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
Acc | +0.500 |
Acc | +0.463 |
fixed | +0.391 |
solved | +0.386 |
collapse | +0.370 |
Neg logit contributions | |
ippers | -0.443 |
banks | -0.437 |
bles | -0.426 |
pouch | -0.424 |
ittens | -0.404 |
Token | voice |
Token ID | 3809 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
ippers | +0.400 |
banks | +0.396 |
bles | +0.382 |
pouch | +0.378 |
kins | +0.349 |
Neg logit contributions | |
Acc | -0.675 |
Acc | -0.625 |
solved | -0.545 |
fixed | -0.541 |
collapse | -0.516 |
Token | said |
Token ID | 531 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
Acc | +0.881 |
Acc | +0.806 |
solved | +0.667 |
fixed | +0.660 |
collapse | +0.643 |
Neg logit contributions | |
banks | -0.564 |
ippers | -0.548 |
pouch | -0.541 |
bles | -0.538 |
ters | -0.512 |
Token | , |
Token ID | 11 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Acc | +0.876 |
Acc | +0.789 |
solved | +0.673 |
fixed | +0.667 |
collapse | +0.646 |
Neg logit contributions | |
ippers | -0.645 |
banks | -0.636 |
bles | -0.627 |
pouch | -0.621 |
ters | -0.593 |
Token | " |
Token ID | 366 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
Acc | +0.225 |
Acc | +0.205 |
solved | +0.176 |
fixed | +0.174 |
collapse | +0.167 |
Neg logit contributions | |
ippers | -0.176 |
banks | -0.176 |
bles | -0.172 |
pouch | -0.170 |
ters | -0.161 |
Token | cong |
Token ID | 36801 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
Acc | +0.815 |
Acc | +0.769 |
fixed | +0.646 |
solved | +0.639 |
collapse | +0.615 |
Neg logit contributions | |
ippers | -0.740 |
banks | -0.726 |
bles | -0.716 |
pouch | -0.714 |
ters | -0.702 |
Token | ratulations |
Token ID | 30167 |
Feature activation | +8.67e-01 |
Pos logit contributions | |
Acc | +5.199 |
Acc | +4.254 |
gone | +4.086 |
fixed | +3.979 |
anne | +3.975 |
Neg logit contributions | |
ippers | -5.109 |
banks | -4.968 |
pouch | -4.743 |
bles | -4.701 |
angers | -4.628 |
Token | , |
Token ID | 11 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
ippers | +1.640 |
banks | +1.637 |
bles | +1.585 |
pouch | +1.582 |
ittens | +1.504 |
Neg logit contributions | |
Acc | -1.224 |
Acc | -1.103 |
solved | -0.898 |
fixed | -0.888 |
collapse | -0.810 |
Token | you |
Token ID | 345 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
Acc | +0.088 |
Acc | +0.079 |
solved | +0.067 |
fixed | +0.066 |
collapse | +0.062 |
Neg logit contributions | |
ippers | -0.073 |
banks | -0.072 |
bles | -0.071 |
pouch | -0.070 |
ters | -0.067 |
Token | can |
Token ID | 460 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
ippers | +0.938 |
banks | +0.925 |
bles | +0.901 |
pouch | +0.891 |
ittens | +0.843 |
Neg logit contributions | |
Acc | -1.071 |
Acc | -0.985 |
solved | -0.843 |
fixed | -0.834 |
collapse | -0.783 |
Token | have |
Token ID | 423 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
Acc | +0.856 |
Acc | +0.775 |
solved | +0.619 |
fixed | +0.616 |
collapse | +0.584 |
Neg logit contributions | |
banks | -0.694 |
ippers | -0.682 |
bles | -0.667 |
pouch | -0.655 |
friends | -0.624 |
Token | have |
Token ID | 423 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Acc | +1.154 |
Acc | +1.026 |
solved | +0.858 |
fixed | +0.823 |
collapse | +0.815 |
Neg logit contributions | |
banks | -0.984 |
ippers | -0.978 |
pouch | -0.942 |
bles | -0.934 |
friends | -0.917 |
Token | some |
Token ID | 617 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
Acc | +2.726 |
Acc | +2.567 |
collapse | +2.165 |
solved | +2.162 |
bell | +2.162 |
Neg logit contributions | |
banks | -1.199 |
friends | -1.189 |
ippers | -1.102 |
ones | -1.099 |
pouch | -1.094 |
Token | ?" |
Token ID | 1701 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
ippers | +0.539 |
banks | +0.530 |
bles | +0.514 |
pouch | +0.509 |
kins | +0.477 |
Neg logit contributions | |
Acc | -0.758 |
Acc | -0.698 |
solved | -0.602 |
fixed | -0.595 |
collapse | -0.567 |
Token |
|
Token ID | 198 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
Acc | +1.230 |
Acc | +1.133 |
solved | +0.995 |
fixed | +0.976 |
aling | +0.930 |
Neg logit contributions | |
ippers | -0.771 |
banks | -0.754 |
pouch | -0.721 |
bles | -0.720 |
ittens | -0.700 |
Token |
|
Token ID | 198 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
Acc | +3.737 |
Acc | +3.502 |
solved | +3.083 |
fixed | +3.011 |
aling | +2.979 |
Neg logit contributions | |
friends | -1.645 |
ippers | -1.631 |
bles | -1.571 |
pouch | -1.571 |
ittens | -1.567 |
Token | S |
Token ID | 50 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
Acc | +4.518 |
Acc | +4.423 |
fixed | +3.650 |
bell | +3.555 |
aling | +3.527 |
Neg logit contributions | |
friends | -2.452 |
couple | -2.425 |
pouch | -2.414 |
bles | -2.354 |
banks | -2.348 |
Token | ue |
Token ID | 518 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Acc | +4.101 |
Acc | +3.766 |
aling | +2.760 |
cal | +2.507 |
fixed | +2.375 |
Neg logit contributions | |
bles | -5.975 |
ippers | -5.877 |
ittens | -5.852 |
owers | -5.812 |
kins | -5.699 |
Token | said |
Token ID | 531 |
Feature activation | -9.42e-01 |
Pos logit contributions | |
Acc | +2.886 |
Acc | +2.741 |
collapse | +2.332 |
solved | +2.280 |
aling | +2.262 |
Neg logit contributions | |
bles | -1.499 |
banks | -1.493 |
pouch | -1.484 |
friends | -1.456 |
ippers | -1.426 |
Token | , |
Token ID | 11 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
Acc | +2.123 |
Acc | +1.950 |
solved | +1.767 |
fixed | +1.712 |
aling | +1.698 |
Neg logit contributions | |
ippers | -1.057 |
bles | -1.023 |
friends | -0.963 |
pouch | -0.956 |
banks | -0.955 |
Token | " |
Token ID | 366 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
Acc | +0.573 |
Acc | +0.535 |
solved | +0.476 |
fixed | +0.468 |
collapse | +0.458 |
Neg logit contributions | |
ippers | -0.272 |
bles | -0.265 |
banks | -0.261 |
pouch | -0.257 |
ters | -0.241 |
Token | Yes |
Token ID | 5297 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
Acc | +0.307 |
Acc | +0.282 |
solved | +0.241 |
fixed | +0.238 |
collapse | +0.225 |
Neg logit contributions | |
ippers | -0.301 |
banks | -0.298 |
pouch | -0.294 |
bles | -0.290 |
ters | -0.279 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
Acc | +0.215 |
Acc | +0.199 |
solved | +0.165 |
fixed | +0.163 |
collapse | +0.160 |
Neg logit contributions | |
banks | -0.141 |
ippers | -0.139 |
bles | -0.137 |
pouch | -0.134 |
kins | -0.126 |
Token | lots |
Token ID | 6041 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
Acc | +0.298 |
Acc | +0.273 |
solved | +0.233 |
fixed | +0.231 |
collapse | +0.222 |
Neg logit contributions | |
banks | -0.195 |
ippers | -0.194 |
bles | -0.188 |
pouch | -0.185 |
ters | -0.176 |
Token | of |
Token ID | 286 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
Acc | +0.702 |
Acc | +0.648 |
solved | +0.553 |
fixed | +0.545 |
collapse | +0.530 |
Neg logit contributions | |
ippers | -0.377 |
banks | -0.377 |
bles | -0.365 |
pouch | -0.354 |
ters | -0.344 |
Token | sweet |
Token ID | 6029 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
ippers | +0.287 |
banks | +0.285 |
bles | +0.272 |
pouch | +0.268 |
ters | +0.246 |
Neg logit contributions | |
Acc | -0.596 |
Acc | -0.556 |
solved | -0.488 |
fixed | -0.484 |
collapse | -0.465 |
Token | - |
Token ID | 12 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ippers | +0.183 |
banks | +0.181 |
bles | +0.168 |
pouch | +0.164 |
ters | +0.146 |
Neg logit contributions | |
Acc | -0.629 |
Acc | -0.590 |
solved | -0.529 |
fixed | -0.525 |
collapse | -0.508 |
Token | sm |
Token ID | 5796 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
Acc | +1.696 |
Acc | +1.567 |
solved | +1.248 |
collapse | +1.218 |
fixed | +1.214 |
Neg logit contributions | |
banks | -1.760 |
ippers | -1.727 |
bles | -1.711 |
ters | -1.661 |
pouch | -1.641 |
Token | elling |
Token ID | 9417 |
Feature activation | +4.29e-01 |
Pos logit contributions | |
Acc | +5.841 |
Acc | +5.383 |
solved | +4.702 |
collapse | +4.622 |
fixed | +4.600 |
Neg logit contributions | |
ippers | -4.043 |
banks | -4.024 |
angers | -4.004 |
ters | -3.990 |
unn | -3.837 |
Token | flowers |
Token ID | 12734 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
ippers | +0.376 |
banks | +0.369 |
bles | +0.351 |
pouch | +0.344 |
ters | +0.307 |
Neg logit contributions | |
Acc | -1.056 |
Acc | -0.990 |
solved | -0.883 |
fixed | -0.877 |
collapse | -0.844 |
Token | ! |
Token ID | 0 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
banks | +0.103 |
ippers | +0.103 |
bles | +0.100 |
pouch | +0.099 |
ters | +0.094 |
Neg logit contributions | |
Acc | -0.119 |
Acc | -0.109 |
solved | -0.090 |
fixed | -0.089 |
collapse | -0.085 |
Token | |
Token ID | 220 |
Feature activation | -9.29e-03 |
Pos logit contributions | |
ippers | +0.497 |
banks | +0.491 |
bles | +0.475 |
pouch | +0.474 |
ittens | +0.443 |
Neg logit contributions | |
Acc | -0.647 |
Acc | -0.594 |
solved | -0.511 |
fixed | -0.507 |
collapse | -0.481 |
Token |
|
Token ID | 198 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
Acc | +0.018 |
Acc | +0.016 |
solved | +0.014 |
fixed | +0.014 |
collapse | +0.013 |
Neg logit contributions | |
ippers | -0.013 |
banks | -0.013 |
bles | -0.013 |
pouch | -0.012 |
ters | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
Acc | +0.023 |
Acc | +0.021 |
solved | +0.018 |
fixed | +0.018 |
collapse | +0.017 |
Neg logit contributions | |
banks | -0.020 |
ippers | -0.020 |
pouch | -0.019 |
bles | -0.019 |
ters | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
Acc | +0.676 |
Acc | +0.608 |
solved | +0.544 |
fixed | +0.536 |
collapse | +0.518 |
Neg logit contributions | |
ippers | -0.487 |
banks | -0.476 |
pouch | -0.469 |
bles | -0.458 |
ittens | -0.438 |
Token |
|
Token ID | 198 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
Acc | +0.734 |
Acc | +0.674 |
solved | +0.592 |
fixed | +0.585 |
collapse | +0.561 |
Neg logit contributions | |
ippers | -0.545 |
banks | -0.534 |
bles | -0.520 |
pouch | -0.517 |
ittens | -0.499 |
Token | M |
Token ID | 44 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
Acc | +0.732 |
Acc | +0.679 |
fixed | +0.586 |
solved | +0.583 |
collapse | +0.556 |
Neg logit contributions | |
ippers | -0.585 |
banks | -0.571 |
bles | -0.558 |
pouch | -0.556 |
ittens | -0.535 |
Token | addy |
Token ID | 13218 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
Acc | +1.578 |
Acc | +1.448 |
solved | +1.278 |
fixed | +1.269 |
collapse | +1.199 |
Neg logit contributions | |
ippers | -0.980 |
bles | -0.936 |
ittens | -0.926 |
banks | -0.911 |
pouch | -0.898 |
Token | never |
Token ID | 1239 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
Acc | +1.053 |
Acc | +0.968 |
solved | +0.793 |
collapse | +0.783 |
fixed | +0.777 |
Neg logit contributions | |
banks | -0.838 |
ippers | -0.818 |
pouch | -0.809 |
bles | -0.805 |
friends | -0.762 |
Token | worried |
Token ID | 7960 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
Acc | +6.752 |
Acc | +6.331 |
collapsing | +5.069 |
collapse | +4.892 |
Lisa | +4.748 |
Neg logit contributions | |
friends | -3.259 |
banks | -3.037 |
pair | -3.007 |
pouch | -2.950 |
legs | -2.875 |
Token | about |
Token ID | 546 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
Acc | +1.556 |
Acc | +1.401 |
solved | +1.150 |
fixed | +1.137 |
collapse | +1.047 |
Neg logit contributions | |
banks | -1.586 |
ippers | -1.586 |
bles | -1.544 |
itive | -1.514 |
friends | -1.513 |
Token | losing |
Token ID | 6078 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Acc | +0.707 |
Acc | +0.643 |
solved | +0.550 |
fixed | +0.539 |
bell | +0.511 |
Neg logit contributions | |
banks | -0.601 |
ippers | -0.597 |
bles | -0.580 |
pouch | -0.577 |
friends | -0.555 |
Token | her |
Token ID | 607 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
Acc | +0.187 |
Acc | +0.172 |
solved | +0.145 |
fixed | +0.142 |
collapse | +0.136 |
Neg logit contributions | |
ippers | -0.160 |
banks | -0.158 |
bles | -0.154 |
pouch | -0.153 |
ters | -0.145 |
Token | special |
Token ID | 2041 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
Acc | +1.002 |
Acc | +0.918 |
solved | +0.808 |
fixed | +0.781 |
collapse | +0.761 |
Neg logit contributions | |
ippers | -0.628 |
banks | -0.628 |
pouch | -0.599 |
bles | -0.592 |
friends | -0.571 |
Token | and |
Token ID | 290 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
Acc | +2.021 |
Acc | +1.849 |
solved | +1.493 |
fixed | +1.458 |
collapse | +1.458 |
Neg logit contributions | |
ippers | -1.380 |
banks | -1.362 |
pouch | -1.342 |
bles | -1.326 |
friends | -1.323 |
Token | was |
Token ID | 373 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
Acc | +0.745 |
Acc | +0.677 |
solved | +0.568 |
fixed | +0.561 |
collapse | +0.554 |
Neg logit contributions | |
banks | -0.545 |
ippers | -0.542 |
pouch | -0.531 |
bles | -0.530 |
friends | -0.517 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
Acc | +1.149 |
Acc | +1.074 |
collapse | +0.878 |
solved | +0.873 |
fixed | +0.862 |
Neg logit contributions | |
ippers | -0.941 |
bles | -0.928 |
pouch | -0.926 |
friends | -0.910 |
banks | -0.907 |
Token | to |
Token ID | 284 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
Acc | +0.818 |
Acc | +0.743 |
solved | +0.619 |
fixed | +0.618 |
collapse | +0.596 |
Neg logit contributions | |
ippers | -0.714 |
banks | -0.694 |
bles | -0.688 |
pouch | -0.681 |
friends | -0.679 |
Token | have |
Token ID | 423 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
Acc | +0.039 |
Acc | +0.036 |
solved | +0.030 |
fixed | +0.029 |
collapse | +0.028 |
Neg logit contributions | |
ippers | -0.036 |
banks | -0.036 |
bles | -0.035 |
pouch | -0.035 |
kins | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
Acc | +1.255 |
Acc | +1.177 |
collapse | +0.936 |
bell | +0.932 |
solved | +0.907 |
Neg logit contributions | |
ippers | -0.952 |
banks | -0.926 |
friends | -0.923 |
pouch | -0.916 |
bles | -0.909 |
Token | unique |
Token ID | 3748 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Acc | +6.828 |
Acc | +6.098 |
aling | +5.990 |
collapse | +5.660 |
collapsing | +5.323 |
Neg logit contributions | |
friends | -3.431 |
couple | -3.321 |
pair | -3.166 |
pouch | -3.158 |
bag | -2.876 |
Token | piece |
Token ID | 3704 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
Acc | +3.153 |
Acc | +2.824 |
aling | +2.576 |
collapse | +2.530 |
solved | +2.494 |
Neg logit contributions | |
ippers | -1.456 |
pouch | -1.450 |
friends | -1.440 |
banks | -1.412 |
pair | -1.401 |
Token | of |
Token ID | 286 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Acc | +1.801 |
Acc | +1.682 |
solved | +1.385 |
fixed | +1.358 |
collapse | +1.356 |
Neg logit contributions | |
ippers | -1.031 |
bles | -0.994 |
pouch | -0.991 |
banks | -0.982 |
friends | -0.957 |
Token | art |
Token ID | 1242 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Acc | +2.078 |
Acc | +1.826 |
solved | +1.576 |
collapse | +1.545 |
aling | +1.543 |
Neg logit contributions | |
banks | -1.480 |
ippers | -1.454 |
pouch | -1.440 |
friends | -1.431 |
ones | -1.371 |
Token | to |
Token ID | 284 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
Acc | +2.006 |
Acc | +1.849 |
collapse | +1.516 |
solved | +1.467 |
collapsing | +1.436 |
Neg logit contributions | |
ippers | -1.470 |
bles | -1.400 |
banks | -1.397 |
kins | -1.391 |
pouch | -1.356 |
Token | lighter |
Token ID | 14871 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
ippers | +0.039 |
banks | +0.038 |
bles | +0.036 |
pouch | +0.036 |
kins | +0.032 |
Neg logit contributions | |
Acc | -0.109 |
Acc | -0.102 |
solved | -0.092 |
fixed | -0.090 |
collapse | -0.088 |
Token | from |
Token ID | 422 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
ippers | +0.131 |
banks | +0.130 |
bles | +0.126 |
pouch | +0.125 |
ters | +0.119 |
Neg logit contributions | |
Acc | -0.141 |
Acc | -0.128 |
solved | -0.108 |
fixed | -0.106 |
collapse | -0.100 |
Token | Tom |
Token ID | 4186 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
Acc | +1.103 |
Acc | +0.996 |
solved | +0.840 |
fixed | +0.818 |
bell | +0.794 |
Neg logit contributions | |
banks | -0.924 |
ippers | -0.913 |
pouch | -0.888 |
bles | -0.881 |
friends | -0.855 |
Token | . |
Token ID | 13 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
Acc | +0.052 |
Acc | +0.047 |
solved | +0.040 |
fixed | +0.039 |
collapse | +0.037 |
Neg logit contributions | |
ippers | -0.046 |
banks | -0.046 |
pouch | -0.044 |
bles | -0.044 |
ters | -0.042 |
Token | They |
Token ID | 1119 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
Acc | +1.075 |
Acc | +0.957 |
solved | +0.871 |
fixed | +0.859 |
collapse | +0.817 |
Neg logit contributions | |
ippers | -0.705 |
banks | -0.685 |
pouch | -0.675 |
bles | -0.656 |
ittens | -0.625 |
Token | fought |
Token ID | 8350 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
Acc | +0.734 |
Acc | +0.671 |
solved | +0.548 |
fixed | +0.539 |
collapse | +0.524 |
Neg logit contributions | |
banks | -0.573 |
ippers | -0.552 |
pouch | -0.540 |
bles | -0.540 |
kins | -0.502 |
Token | and |
Token ID | 290 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
Acc | +6.177 |
Acc | +5.637 |
bug | +4.175 |
anne | +4.161 |
solved | +4.100 |
Neg logit contributions | |
bles | -4.220 |
things | -4.206 |
friends | -4.168 |
ters | -4.159 |
banks | -3.957 |
Token | pulled |
Token ID | 5954 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Acc | +1.082 |
Acc | +0.975 |
solved | +0.780 |
fixed | +0.761 |
bell | +0.746 |
Neg logit contributions | |
banks | -0.951 |
ippers | -0.924 |
bles | -0.914 |
pouch | -0.903 |
ters | -0.877 |
Token | and |
Token ID | 290 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Acc | +2.924 |
Acc | +2.654 |
solved | +2.232 |
fixed | +2.106 |
bell | +2.077 |
Neg logit contributions | |
banks | -2.147 |
ippers | -2.124 |
bles | -2.065 |
ters | -2.045 |
friends | -2.027 |
Token | pushed |
Token ID | 7121 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Acc | +2.573 |
Acc | +2.358 |
bell | +1.862 |
solved | +1.821 |
anne | +1.803 |
Neg logit contributions | |
banks | -1.931 |
bles | -1.853 |
ters | -1.823 |
ippers | -1.823 |
pouch | -1.806 |
Token | . |
Token ID | 13 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Acc | +2.286 |
Acc | +2.095 |
solved | +1.793 |
fixed | +1.723 |
bell | +1.665 |
Neg logit contributions | |
banks | -1.585 |
ippers | -1.583 |
bles | -1.538 |
ters | -1.518 |
friends | -1.465 |
Token | them |
Token ID | 606 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
Acc | +1.202 |
Acc | +1.090 |
solved | +0.956 |
fixed | +0.917 |
bell | +0.871 |
Neg logit contributions | |
ippers | -0.909 |
banks | -0.908 |
bles | -0.907 |
ters | -0.856 |
kins | -0.844 |
Token | back |
Token ID | 736 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
ippers | +0.279 |
banks | +0.279 |
bles | +0.273 |
pouch | +0.263 |
kins | +0.253 |
Neg logit contributions | |
Acc | -0.340 |
Acc | -0.312 |
solved | -0.264 |
fixed | -0.260 |
collapse | -0.245 |
Token | ." |
Token ID | 526 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
Acc | +0.977 |
Acc | +0.899 |
solved | +0.764 |
fixed | +0.741 |
collapse | +0.712 |
Neg logit contributions | |
banks | -0.696 |
bles | -0.687 |
ippers | -0.683 |
pouch | -0.643 |
ters | -0.637 |
Token |
|
Token ID | 198 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
ippers | +0.026 |
banks | +0.025 |
bles | +0.024 |
pouch | +0.024 |
ittens | +0.023 |
Neg logit contributions | |
Acc | -0.035 |
Acc | -0.032 |
solved | -0.028 |
fixed | -0.028 |
collapse | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Acc | +1.757 |
Acc | +1.603 |
solved | +1.447 |
fixed | +1.407 |
collapse | +1.340 |
Neg logit contributions | |
ippers | -1.016 |
bles | -0.965 |
banks | -0.956 |
pouch | -0.941 |
ittens | -0.917 |
Token | S |
Token ID | 50 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
Acc | +2.438 |
Acc | +2.268 |
fixed | +1.957 |
solved | +1.932 |
bell | +1.798 |
Neg logit contributions | |
ippers | -1.647 |
banks | -1.600 |
bles | -1.600 |
pouch | -1.580 |
friends | -1.514 |
Token | ue |
Token ID | 518 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Acc | +4.013 |
Acc | +3.698 |
aling | +2.731 |
cal | +2.468 |
fixed | +2.388 |
Neg logit contributions | |
bles | -5.931 |
ippers | -5.871 |
ittens | -5.719 |
owers | -5.700 |
kins | -5.596 |
Token | looked |
Token ID | 3114 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
Acc | +2.877 |
Acc | +2.750 |
collapse | +2.290 |
solved | +2.242 |
aling | +2.242 |
Neg logit contributions | |
banks | -1.698 |
bles | -1.655 |
pouch | -1.576 |
ippers | -1.574 |
friends | -1.556 |
Token | at |
Token ID | 379 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Acc | +0.050 |
Acc | +0.046 |
solved | +0.039 |
fixed | +0.038 |
collapse | +0.037 |
Neg logit contributions | |
ippers | -0.044 |
banks | -0.043 |
bles | -0.042 |
pouch | -0.042 |
friends | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Acc | +1.951 |
Acc | +1.792 |
solved | +1.491 |
fixed | +1.386 |
bell | +1.309 |
Neg logit contributions | |
banks | -2.437 |
ippers | -2.409 |
bles | -2.385 |
friends | -2.307 |
ones | -2.300 |
Token | rake |
Token ID | 47885 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
ippers | +1.505 |
banks | +1.438 |
bles | +1.394 |
pouch | +1.394 |
itive | +1.313 |
Neg logit contributions | |
Acc | -2.390 |
Acc | -2.213 |
solved | -1.953 |
fixed | -1.935 |
collapse | -1.848 |
Token | learned |
Token ID | 4499 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
ippers | +0.711 |
banks | +0.704 |
bles | +0.689 |
pouch | +0.682 |
kins | +0.646 |
Neg logit contributions | |
Acc | -0.681 |
Acc | -0.619 |
solved | -0.512 |
fixed | -0.506 |
collapse | -0.476 |
Token | their |
Token ID | 511 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
Acc | +1.726 |
Acc | +1.560 |
solved | +1.356 |
aling | +1.312 |
fixed | +1.274 |
Neg logit contributions | |
banks | -1.484 |
ippers | -1.475 |
friends | -1.445 |
bles | -1.422 |
pouch | -1.408 |
Token | lesson |
Token ID | 11483 |
Feature activation | -7.14e-01 |
Pos logit contributions | |
ippers | +0.036 |
banks | +0.036 |
bles | +0.035 |
pouch | +0.034 |
friends | +0.032 |
Neg logit contributions | |
Acc | -0.059 |
Acc | -0.055 |
solved | -0.048 |
fixed | -0.048 |
collapse | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
Acc | +1.381 |
Acc | +1.283 |
solved | +1.089 |
fixed | +1.072 |
collapse | +1.032 |
Neg logit contributions | |
banks | -0.994 |
ippers | -0.979 |
bles | -0.946 |
pouch | -0.925 |
ters | -0.913 |
Token | They |
Token ID | 1119 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Acc | +0.738 |
Acc | +0.668 |
solved | +0.602 |
fixed | +0.592 |
collapse | +0.569 |
Neg logit contributions | |
ippers | -0.456 |
banks | -0.440 |
pouch | -0.432 |
bles | -0.419 |
ittens | -0.396 |
Token | never |
Token ID | 1239 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
Acc | +0.232 |
Acc | +0.216 |
solved | +0.176 |
fixed | +0.173 |
collapse | +0.171 |
Neg logit contributions | |
banks | -0.188 |
ippers | -0.183 |
pouch | -0.181 |
bles | -0.180 |
friends | -0.169 |
Token | played |
Token ID | 2826 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
Acc | +6.451 |
Acc | +5.931 |
bell | +4.668 |
anne | +4.596 |
collapse | +4.592 |
Neg logit contributions | |
friends | -3.632 |
banks | -3.393 |
animals | -3.183 |
pair | -3.164 |
pouch | -3.123 |
Token | with |
Token ID | 351 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
Acc | +0.700 |
Acc | +0.648 |
solved | +0.542 |
fixed | +0.533 |
bell | +0.524 |
Neg logit contributions | |
ippers | -0.517 |
banks | -0.511 |
bles | -0.503 |
pouch | -0.477 |
ters | -0.476 |
Token | the |
Token ID | 262 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
Acc | +1.174 |
Acc | +1.082 |
solved | +0.952 |
fixed | +0.914 |
aling | +0.912 |
Neg logit contributions | |
banks | -0.712 |
ippers | -0.701 |
bles | -0.684 |
friends | -0.651 |
pouch | -0.649 |
Token | ladder |
Token ID | 18002 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ippers | +0.873 |
banks | +0.848 |
bles | +0.819 |
pouch | +0.811 |
ittens | +0.756 |
Neg logit contributions | |
Acc | -1.619 |
Acc | -1.506 |
solved | -1.327 |
fixed | -1.324 |
collapse | -1.262 |
Token | and |
Token ID | 290 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
Acc | +2.944 |
Acc | +2.759 |
solved | +1.877 |
anne | +1.774 |
OP | +1.718 |
Neg logit contributions | |
banks | -4.015 |
ters | -3.890 |
bles | -3.873 |
friends | -3.770 |
ones | -3.770 |
Token | said |
Token ID | 531 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
ippers | +1.127 |
banks | +1.097 |
bles | +1.062 |
pouch | +1.039 |
ittens | +0.974 |
Neg logit contributions | |
Acc | -1.746 |
Acc | -1.620 |
fixed | -1.431 |
solved | -1.428 |
collapse | -1.325 |
Token | sadly |
Token ID | 21098 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
Acc | +0.663 |
Acc | +0.602 |
solved | +0.515 |
fixed | +0.493 |
collapse | +0.485 |
Neg logit contributions | |
ippers | -0.520 |
banks | -0.500 |
bles | -0.493 |
pouch | -0.490 |
kins | -0.470 |
Token | . |
Token ID | 13 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
Acc | +0.006 |
Acc | +0.005 |
solved | +0.004 |
fixed | +0.004 |
collapse | +0.004 |
Neg logit contributions | |
ippers | -0.005 |
banks | -0.005 |
pouch | -0.005 |
bles | -0.005 |
ittens | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Acc | +1.742 |
Acc | +1.572 |
solved | +1.437 |
fixed | +1.401 |
collapse | +1.386 |
Neg logit contributions | |
ippers | -0.919 |
banks | -0.902 |
pouch | -0.884 |
bles | -0.859 |
ittens | -0.841 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Acc | +2.497 |
Acc | +2.301 |
solved | +2.057 |
collapse | +2.003 |
fixed | +1.979 |
Neg logit contributions | |
ippers | -1.498 |
banks | -1.453 |
ittens | -1.446 |
bles | -1.433 |
pouch | -1.412 |
Token | S |
Token ID | 50 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
Acc | +2.926 |
Acc | +2.790 |
solved | +2.421 |
fixed | +2.399 |
collapse | +2.391 |
Neg logit contributions | |
banks | -1.608 |
bles | -1.576 |
ippers | -1.576 |
ittens | -1.532 |
pouch | -1.499 |
Token | ally |
Token ID | 453 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
Acc | +4.972 |
Acc | +4.687 |
explode | +3.975 |
aling | +3.909 |
collapse | +3.871 |
Neg logit contributions | |
ittens | -4.585 |
ippers | -4.518 |
bles | -4.469 |
friends | -4.143 |
re | -4.076 |
Token | tried |
Token ID | 3088 |
Feature activation | -7.27e-01 |
Pos logit contributions | |
Acc | +1.363 |
Acc | +1.264 |
collapse | +1.024 |
solved | +0.995 |
collapsing | +0.982 |
Neg logit contributions | |
banks | -1.133 |
pouch | -1.100 |
bles | -1.095 |
ippers | -1.072 |
friends | -1.058 |
Token | to |
Token ID | 284 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Acc | +1.152 |
Acc | +1.057 |
solved | +0.859 |
fixed | +0.827 |
aling | +0.819 |
Neg logit contributions | |
banks | -1.235 |
ippers | -1.225 |
pouch | -1.195 |
bles | -1.185 |
friends | -1.183 |
Token | cheer |
Token ID | 14042 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
Acc | +0.411 |
Acc | +0.372 |
solved | +0.301 |
fixed | +0.291 |
collapse | +0.277 |
Neg logit contributions | |
banks | -0.423 |
ippers | -0.418 |
bles | -0.410 |
pouch | -0.403 |
friends | -0.389 |
Token | him |
Token ID | 683 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
Acc | +1.589 |
Acc | +1.433 |
solved | +1.266 |
fixed | +1.199 |
collapse | +1.145 |
Neg logit contributions | |
ippers | -1.503 |
bles | -1.470 |
banks | -1.445 |
ters | -1.382 |
friends | -1.367 |
Token | his |
Token ID | 465 |
Feature activation | +9.89e-01 |
Pos logit contributions | |
Acc | +1.053 |
Acc | +0.968 |
solved | +0.840 |
fixed | +0.815 |
bell | +0.792 |
Neg logit contributions | |
banks | -0.665 |
ippers | -0.664 |
bles | -0.640 |
pouch | -0.622 |
ters | -0.609 |
Token | juice |
Token ID | 13135 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
ippers | +1.498 |
banks | +1.473 |
bles | +1.448 |
pouch | +1.427 |
itive | +1.368 |
Neg logit contributions | |
Acc | -1.793 |
Acc | -1.659 |
fixed | -1.411 |
solved | -1.398 |
collapse | -1.300 |
Token | again |
Token ID | 757 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
Acc | +0.848 |
Acc | +0.785 |
solved | +0.644 |
fixed | +0.624 |
collapse | +0.605 |
Neg logit contributions | |
ippers | -0.709 |
banks | -0.706 |
bles | -0.695 |
pouch | -0.671 |
ters | -0.650 |
Token | . |
Token ID | 13 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Acc | +0.710 |
Acc | +0.660 |
solved | +0.544 |
fixed | +0.531 |
collapse | +0.523 |
Neg logit contributions | |
banks | -0.583 |
ippers | -0.580 |
bles | -0.559 |
pouch | -0.557 |
kins | -0.539 |
Token | The |
Token ID | 383 |
Feature activation | +9.23e-01 |
Pos logit contributions | |
Acc | +2.145 |
Acc | +1.931 |
solved | +1.753 |
collapse | +1.711 |
fixed | +1.698 |
Neg logit contributions | |
banks | -1.373 |
pouch | -1.368 |
ippers | -1.368 |
ittens | -1.305 |
friends | -1.299 |
Token | end |
Token ID | 886 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
ippers | +1.462 |
banks | +1.432 |
bles | +1.396 |
pouch | +1.377 |
ittens | +1.305 |
Neg logit contributions | |
Acc | -1.654 |
Acc | -1.511 |
solved | -1.286 |
fixed | -1.276 |
collapse | -1.195 |
Token | . |
Token ID | 13 |
Feature activation | -9.89e-01 |
Pos logit contributions | |
Acc | +6.866 |
Acc | +6.289 |
aling | +5.569 |
collapse | +5.459 |
solved | +5.365 |
Neg logit contributions | |
friends | -2.816 |
things | -2.769 |
ippers | -2.610 |
pouch | -2.514 |
robe | -2.484 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Acc | +2.070 |
Acc | +1.913 |
solved | +1.692 |
collapse | +1.645 |
fixed | +1.640 |
Neg logit contributions | |
pouch | -1.136 |
ippers | -1.131 |
banks | -1.101 |
ittens | -1.067 |
friends | -1.052 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
ippers | +0.147 |
banks | +0.145 |
bles | +0.140 |
pouch | +0.140 |
ters | +0.131 |
Neg logit contributions | |
Acc | -0.200 |
Acc | -0.185 |
solved | -0.157 |
fixed | -0.155 |
collapse | -0.149 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
ippers | +0.193 |
banks | +0.185 |
bles | +0.181 |
pouch | +0.174 |
ittens | +0.157 |
Neg logit contributions | |
Acc | -0.587 |
Acc | -0.552 |
solved | -0.494 |
fixed | -0.488 |
collapse | -0.473 |
Token | a |
Token ID | 257 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
ippers | +0.093 |
banks | +0.092 |
bles | +0.089 |
pouch | +0.088 |
ittens | +0.083 |
Neg logit contributions | |
Acc | -0.117 |
Acc | -0.109 |
solved | -0.091 |
fixed | -0.090 |
collapse | -0.087 |
Token | think |
Token ID | 892 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
ippers | +0.695 |
banks | +0.686 |
bles | +0.671 |
pouch | +0.662 |
kins | +0.628 |
Neg logit contributions | |
Acc | -0.718 |
Acc | -0.654 |
solved | -0.547 |
fixed | -0.541 |
collapse | -0.511 |
Token | now |
Token ID | 783 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
Acc | +1.183 |
Acc | +1.065 |
solved | +0.877 |
collapse | +0.848 |
fixed | +0.848 |
Neg logit contributions | |
ippers | -0.943 |
bles | -0.929 |
banks | -0.891 |
pouch | -0.865 |
friends | -0.862 |
Token | ?" |
Token ID | 1701 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
Acc | +0.310 |
Acc | +0.278 |
solved | +0.235 |
fixed | +0.228 |
collapse | +0.225 |
Neg logit contributions | |
ippers | -0.254 |
bles | -0.247 |
banks | -0.244 |
pouch | -0.235 |
kins | -0.228 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Acc | +1.403 |
Acc | +1.268 |
solved | +1.170 |
collapse | +1.127 |
fixed | +1.126 |
Neg logit contributions | |
ippers | -0.841 |
banks | -0.778 |
pouch | -0.755 |
bles | -0.751 |
ittens | -0.737 |
Token |
|
Token ID | 198 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
Acc | +2.154 |
Acc | +1.983 |
solved | +1.786 |
collapse | +1.734 |
fixed | +1.716 |
Neg logit contributions | |
ippers | -1.266 |
bles | -1.172 |
banks | -1.155 |
pouch | -1.132 |
friends | -1.124 |
Token | S |
Token ID | 50 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
Acc | +2.601 |
Acc | +2.438 |
solved | +2.166 |
fixed | +2.136 |
collapse | +2.092 |
Neg logit contributions | |
ippers | -1.506 |
bles | -1.415 |
banks | -1.395 |
pouch | -1.339 |
ittens | -1.329 |
Token | ister |
Token ID | 1694 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Acc | +4.816 |
Acc | +4.368 |
explode | +3.764 |
cal | +3.701 |
solved | +3.693 |
Neg logit contributions | |
ippers | -4.815 |
ittens | -4.723 |
bles | -4.701 |
owers | -4.254 |
kins | -4.241 |
Token | looked |
Token ID | 3114 |
Feature activation | -9.78e-01 |
Pos logit contributions | |
Acc | +3.903 |
Acc | +3.515 |
collapse | +3.018 |
solved | +2.898 |
aling | +2.896 |
Neg logit contributions | |
bles | -1.954 |
banks | -1.946 |
pouch | -1.891 |
couple | -1.834 |
ters | -1.832 |
Token | at |
Token ID | 379 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
Acc | +1.926 |
Acc | +1.791 |
collapse | +1.543 |
solved | +1.496 |
collapsing | +1.433 |
Neg logit contributions | |
ippers | -1.335 |
bles | -1.329 |
banks | -1.323 |
pouch | -1.298 |
friends | -1.268 |
Token | the |
Token ID | 262 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
Acc | +0.488 |
Acc | +0.443 |
solved | +0.370 |
fixed | +0.356 |
collapse | +0.338 |
Neg logit contributions | |
ippers | -0.499 |
banks | -0.499 |
bles | -0.489 |
pouch | -0.477 |
friends | -0.462 |
Token | drawing |
Token ID | 8263 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
Acc | +1.398 |
Acc | +1.282 |
solved | +1.092 |
fixed | +1.061 |
collapse | +1.049 |
Neg logit contributions | |
ippers | -0.935 |
banks | -0.928 |
bles | -0.918 |
pouch | -0.893 |
friends | -0.880 |
Token | back |
Token ID | 736 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
Acc | +0.854 |
Acc | +0.790 |
solved | +0.652 |
fixed | +0.623 |
bell | +0.608 |
Neg logit contributions | |
banks | -0.735 |
ippers | -0.734 |
bles | -0.727 |
pouch | -0.725 |
ters | -0.686 |
Token | . |
Token ID | 13 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
Acc | +1.263 |
Acc | +1.159 |
solved | +0.991 |
fixed | +0.961 |
collapse | +0.929 |
Neg logit contributions | |
ippers | -0.903 |
banks | -0.870 |
bles | -0.866 |
pouch | -0.854 |
friends | -0.830 |
Token | They |
Token ID | 1119 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
Acc | +1.099 |
Acc | +0.991 |
solved | +0.912 |
fixed | +0.893 |
collapse | +0.837 |
Neg logit contributions | |
ippers | -0.805 |
banks | -0.765 |
pouch | -0.744 |
bles | -0.741 |
ittens | -0.709 |
Token | started |
Token ID | 2067 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
banks | +0.297 |
ippers | +0.295 |
bles | +0.287 |
pouch | +0.285 |
kins | +0.266 |
Neg logit contributions | |
Acc | -0.369 |
Acc | -0.341 |
solved | -0.286 |
fixed | -0.282 |
collapse | -0.270 |
Token | to |
Token ID | 284 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
Acc | +1.403 |
Acc | +1.280 |
solved | +1.084 |
fixed | +1.024 |
aling | +1.015 |
Neg logit contributions | |
banks | -1.579 |
ippers | -1.543 |
pouch | -1.487 |
friends | -1.478 |
bles | -1.477 |
Token | fight |
Token ID | 1907 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
Acc | +0.650 |
Acc | +0.582 |
solved | +0.488 |
fixed | +0.463 |
bell | +0.446 |
Neg logit contributions | |
banks | -0.732 |
ippers | -0.700 |
bles | -0.685 |
pouch | -0.682 |
ters | -0.655 |
Token | . |
Token ID | 13 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Acc | +6.391 |
Acc | +5.927 |
aling | +4.591 |
anne | +4.591 |
solved | +4.548 |
Neg logit contributions | |
things | -3.776 |
friends | -3.679 |
ters | -3.663 |
animals | -3.490 |
people | -3.479 |
Token | They |
Token ID | 1119 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
Acc | +0.249 |
Acc | +0.227 |
solved | +0.202 |
fixed | +0.200 |
collapse | +0.186 |
Neg logit contributions | |
ippers | -0.191 |
banks | -0.187 |
pouch | -0.180 |
bles | -0.178 |
ittens | -0.169 |
Token | pushed |
Token ID | 7121 |
Feature activation | -9.82e-01 |
Pos logit contributions | |
banks | +0.040 |
ippers | +0.040 |
bles | +0.039 |
pouch | +0.039 |
ters | +0.036 |
Neg logit contributions | |
Acc | -0.049 |
Acc | -0.045 |
solved | -0.037 |
fixed | -0.037 |
collapse | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -9.64e-01 |
Pos logit contributions | |
Acc | +1.892 |
Acc | +1.761 |
solved | +1.514 |
fixed | +1.454 |
bell | +1.396 |
Neg logit contributions | |
ippers | -1.458 |
banks | -1.436 |
bles | -1.401 |
ters | -1.385 |
pouch | -1.359 |
Token | pulled |
Token ID | 5954 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Acc | +1.896 |
Acc | +1.793 |
solved | +1.427 |
bell | +1.412 |
collapse | +1.366 |
Neg logit contributions | |
banks | -1.348 |
bles | -1.287 |
ippers | -1.266 |
pouch | -1.252 |
ters | -1.248 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
ippers | +2.631 |
bles | +2.521 |
banks | +2.490 |
pouch | +2.441 |
ittens | +2.437 |
Neg logit contributions | |
Acc | -2.250 |
Acc | -2.042 |
solved | -1.775 |
fixed | -1.771 |
collapse | -1.617 |
Token | where |
Token ID | 810 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
Acc | +0.295 |
Acc | +0.274 |
solved | +0.233 |
fixed | +0.230 |
collapse | +0.225 |
Neg logit contributions | |
banks | -0.196 |
ippers | -0.191 |
pouch | -0.187 |
bles | -0.186 |
ters | -0.173 |
Token | he |
Token ID | 339 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
ippers | +0.155 |
banks | +0.155 |
bles | +0.150 |
pouch | +0.147 |
kins | +0.139 |
Neg logit contributions | |
Acc | -0.226 |
Acc | -0.209 |
solved | -0.177 |
fixed | -0.175 |
collapse | -0.167 |
Token | found |
Token ID | 1043 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
banks | +0.089 |
ippers | +0.088 |
bles | +0.086 |
pouch | +0.086 |
kins | +0.080 |
Neg logit contributions | |
Acc | -0.108 |
Acc | -0.099 |
solved | -0.081 |
fixed | -0.080 |
collapse | -0.078 |
Token | some |
Token ID | 617 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ippers | +1.273 |
banks | +1.245 |
bles | +1.223 |
pouch | +1.201 |
ittens | +1.150 |
Neg logit contributions | |
Acc | -1.328 |
Acc | -1.214 |
solved | -1.042 |
fixed | -1.034 |
collapse | -0.960 |
Token | other |
Token ID | 584 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
ippers | +2.406 |
bles | +2.320 |
banks | +2.243 |
pouch | +2.133 |
angers | +2.126 |
Neg logit contributions | |
Acc | -3.322 |
Acc | -3.110 |
fixed | -2.844 |
solved | -2.826 |
cal | -2.650 |
Token | horses |
Token ID | 14260 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ippers | +2.799 |
bles | +2.593 |
robe | +2.406 |
angers | +2.358 |
kins | +2.284 |
Neg logit contributions | |
Acc | -6.363 |
fixed | -5.934 |
Acc | -5.858 |
solved | -5.852 |
cal | -5.815 |
Token | living |
Token ID | 2877 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
ippers | +1.751 |
banks | +1.701 |
bles | +1.680 |
pouch | +1.646 |
ittens | +1.597 |
Neg logit contributions | |
Acc | -1.649 |
Acc | -1.503 |
solved | -1.288 |
fixed | -1.285 |
collapse | -1.172 |
Token | happily |
Token ID | 18177 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
banks | +3.023 |
ippers | +2.993 |
bles | +2.954 |
ittens | +2.896 |
pouch | +2.838 |
Neg logit contributions | |
Acc | -3.448 |
Acc | -3.239 |
fixed | -2.884 |
solved | -2.844 |
gone | -2.583 |
Token | . |
Token ID | 13 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
ippers | +0.934 |
banks | +0.922 |
bles | +0.900 |
pouch | +0.890 |
ittens | +0.844 |
Neg logit contributions | |
Acc | -0.945 |
Acc | -0.860 |
solved | -0.724 |
fixed | -0.717 |
collapse | -0.669 |
Token | The |
Token ID | 383 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
ippers | +1.756 |
bles | +1.703 |
banks | +1.694 |
pouch | +1.649 |
itive | +1.616 |
Neg logit contributions | |
Acc | -1.580 |
Acc | -1.445 |
solved | -1.204 |
fixed | -1.196 |
collapse | -1.092 |
Token | birds |
Token ID | 10087 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
banks | +0.022 |
ippers | +0.022 |
bles | +0.021 |
pouch | +0.021 |
ters | +0.020 |
Neg logit contributions | |
Acc | -0.028 |
Acc | -0.026 |
solved | -0.022 |
fixed | -0.022 |
collapse | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
ippers | +0.697 |
banks | +0.690 |
bles | +0.673 |
pouch | +0.666 |
ters | +0.628 |
Neg logit contributions | |
Acc | -0.740 |
Acc | -0.676 |
solved | -0.568 |
fixed | -0.562 |
collapse | -0.529 |
Token | One |
Token ID | 1881 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
ippers | +0.387 |
banks | +0.386 |
bles | +0.370 |
pouch | +0.369 |
ittens | +0.345 |
Neg logit contributions | |
Acc | -0.530 |
Acc | -0.488 |
solved | -0.422 |
fixed | -0.418 |
collapse | -0.398 |
Token | day |
Token ID | 1110 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
Acc | +0.476 |
Acc | +0.447 |
solved | +0.388 |
fixed | +0.386 |
collapse | +0.375 |
Neg logit contributions | |
banks | -0.213 |
ippers | -0.212 |
pouch | -0.208 |
bles | -0.200 |
friends | -0.192 |
Token | , |
Token ID | 11 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
ippers | +0.196 |
banks | +0.194 |
bles | +0.187 |
pouch | +0.185 |
ters | +0.174 |
Neg logit contributions | |
Acc | -0.301 |
Acc | -0.278 |
solved | -0.238 |
fixed | -0.236 |
collapse | -0.226 |
Token | the |
Token ID | 262 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
ippers | +1.498 |
banks | +1.465 |
bles | +1.442 |
pouch | +1.424 |
ittens | +1.349 |
Neg logit contributions | |
Acc | -1.382 |
Acc | -1.250 |
solved | -1.052 |
fixed | -1.049 |
collapse | -0.967 |
Token | c |
Token ID | 269 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ippers | +4.514 |
ittens | +4.148 |
bles | +4.075 |
angers | +3.945 |
banks | +3.935 |
Neg logit contributions | |
Acc | -5.287 |
Acc | -4.788 |
cal | -4.393 |
solved | -4.362 |
fixed | -4.322 |
Token | rows |
Token ID | 8516 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
banks | +1.280 |
ippers | +1.275 |
pouch | +1.215 |
bles | +1.198 |
kins | +1.128 |
Neg logit contributions | |
Acc | -2.509 |
Acc | -2.318 |
solved | -2.037 |
fixed | -2.015 |
collapse | -1.927 |
Token | were |
Token ID | 547 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ippers | +1.047 |
banks | +1.023 |
bles | +1.001 |
pouch | +0.989 |
ittens | +0.941 |
Neg logit contributions | |
Acc | -1.198 |
Acc | -1.095 |
solved | -0.942 |
fixed | -0.933 |
collapse | -0.877 |
Token | being |
Token ID | 852 |
Feature activation | +9.68e-01 |
Pos logit contributions | |
ippers | +2.180 |
banks | +2.093 |
bles | +2.013 |
pouch | +1.996 |
ittens | +1.954 |
Neg logit contributions | |
Acc | -2.314 |
Acc | -2.076 |
solved | -1.882 |
fixed | -1.859 |
collapse | -1.703 |
Token | extra |
Token ID | 3131 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ippers | +1.561 |
banks | +1.538 |
bles | +1.471 |
pouch | +1.469 |
ittens | +1.409 |
Neg logit contributions | |
Acc | -1.690 |
Acc | -1.524 |
solved | -1.338 |
fixed | -1.333 |
collapse | -1.206 |
Token | he |
Token ID | 339 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ippers | +1.070 |
banks | +1.049 |
bles | +1.019 |
pouch | +1.003 |
ittens | +0.941 |
Neg logit contributions | |
Acc | -1.538 |
Acc | -1.419 |
solved | -1.237 |
fixed | -1.228 |
collapse | -1.154 |
Token | knew |
Token ID | 2993 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
Acc | +2.288 |
Acc | +2.130 |
collapse | +1.771 |
collapsing | +1.704 |
solved | +1.680 |
Neg logit contributions | |
pouch | -1.258 |
bles | -1.247 |
ippers | -1.240 |
banks | -1.234 |
friends | -1.194 |
Token | Santa |
Token ID | 8909 |
Feature activation | +3.26e-01 |
Pos logit contributions | |
ippers | +0.452 |
banks | +0.446 |
bles | +0.437 |
pouch | +0.432 |
ters | +0.409 |
Neg logit contributions | |
Acc | -0.475 |
Acc | -0.433 |
solved | -0.363 |
fixed | -0.358 |
collapse | -0.338 |
Token | Claus |
Token ID | 33989 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
ippers | +0.520 |
banks | +0.514 |
bles | +0.503 |
pouch | +0.497 |
kins | +0.470 |
Neg logit contributions | |
Acc | -0.566 |
Acc | -0.515 |
solved | -0.433 |
fixed | -0.427 |
collapse | -0.406 |
Token | and |
Token ID | 290 |
Feature activation | +8.25e-01 |
Pos logit contributions | |
ippers | +0.717 |
banks | +0.708 |
bles | +0.691 |
pouch | +0.682 |
kins | +0.643 |
Neg logit contributions | |
Acc | -0.833 |
Acc | -0.763 |
solved | -0.647 |
fixed | -0.641 |
collapse | -0.605 |
Token | the |
Token ID | 262 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
ippers | +1.289 |
banks | +1.259 |
bles | +1.234 |
pouch | +1.207 |
ittens | +1.150 |
Neg logit contributions | |
Acc | -1.479 |
Acc | -1.365 |
solved | -1.169 |
fixed | -1.164 |
collapse | -1.078 |
Token | re |
Token ID | 302 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ippers | +4.196 |
angers | +4.057 |
bles | +4.049 |
ittens | +3.826 |
banks | +3.745 |
Neg logit contributions | |
Acc | -5.945 |
Acc | -5.638 |
fixed | -5.204 |
cal | -5.111 |
solved | -4.891 |
Token | ind |
Token ID | 521 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
banks | +2.472 |
ippers | +2.455 |
bles | +2.391 |
pouch | +2.362 |
ters | +2.301 |
Neg logit contributions | |
Acc | -1.608 |
Acc | -1.430 |
fixed | -1.087 |
solved | -1.072 |
bell | -0.992 |
Token | eer |
Token ID | 28153 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ippers | +1.468 |
banks | +1.454 |
bles | +1.393 |
pouch | +1.376 |
ittens | +1.284 |
Neg logit contributions | |
Acc | -2.614 |
Acc | -2.448 |
fixed | -2.124 |
solved | -2.110 |
bell | -2.017 |
Token | nodded |
Token ID | 14464 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
ippers | +1.495 |
banks | +1.456 |
bles | +1.416 |
pouch | +1.387 |
ittens | +1.327 |
Neg logit contributions | |
Acc | -1.969 |
Acc | -1.826 |
fixed | -1.607 |
solved | -1.604 |
collapse | -1.478 |
Token | his |
Token ID | 465 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Acc | +1.298 |
Acc | +1.213 |
solved | +1.037 |
fixed | +1.002 |
collapse | +1.001 |
Neg logit contributions | |
ippers | -1.001 |
bles | -0.968 |
banks | -0.954 |
pouch | -0.942 |
kins | -0.922 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
Acc | +1.736 |
Acc | +1.575 |
solved | +1.399 |
fixed | +1.335 |
collapse | +1.311 |
Neg logit contributions | |
ippers | -1.258 |
bles | -1.197 |
banks | -1.170 |
ters | -1.155 |
pouch | -1.152 |
Token |
|
Token ID | 198 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
ippers | +0.228 |
banks | +0.223 |
pouch | +0.218 |
bles | +0.217 |
ters | +0.203 |
Neg logit contributions | |
Acc | -0.310 |
Acc | -0.284 |
solved | -0.245 |
fixed | -0.242 |
collapse | -0.231 |
Token |
|
Token ID | 198 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
Acc | +0.427 |
Acc | +0.394 |
solved | +0.341 |
fixed | +0.337 |
collapse | +0.323 |
Neg logit contributions | |
ippers | -0.291 |
banks | -0.285 |
bles | -0.279 |
pouch | -0.278 |
ittens | -0.261 |
Token | Anna |
Token ID | 31160 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
Acc | +0.258 |
Acc | +0.235 |
fixed | +0.190 |
solved | +0.189 |
collapse | +0.177 |
Neg logit contributions | |
ippers | -0.306 |
banks | -0.300 |
bles | -0.297 |
pouch | -0.296 |
ters | -0.282 |
Token | takes |
Token ID | 2753 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
Acc | +0.959 |
Acc | +0.872 |
solved | +0.724 |
fixed | +0.712 |
collapse | +0.704 |
Neg logit contributions | |
banks | -0.807 |
ippers | -0.806 |
pouch | -0.782 |
bles | -0.782 |
friends | -0.726 |
Token | the |
Token ID | 262 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
Acc | +0.110 |
Acc | +0.101 |
solved | +0.083 |
fixed | +0.081 |
collapse | +0.078 |
Neg logit contributions | |
ippers | -0.093 |
banks | -0.091 |
pouch | -0.089 |
bles | -0.089 |
ters | -0.084 |
Token | plate |
Token ID | 7480 |
Feature activation | -8.38e-01 |
Pos logit contributions | |
itive | +4.104 |
ippers | +4.041 |
banks | +4.010 |
angers | +3.937 |
ittens | +3.802 |
Neg logit contributions | |
Acc | -4.901 |
Acc | -4.595 |
fixed | -4.484 |
solved | -4.099 |
gone | -3.890 |
Token | and |
Token ID | 290 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
Acc | +1.673 |
Acc | +1.559 |
solved | +1.260 |
fixed | +1.212 |
collapse | +1.176 |
Neg logit contributions | |
ippers | -1.183 |
banks | -1.179 |
pouch | -1.159 |
bles | -1.143 |
couple | -1.079 |
Token | goes |
Token ID | 2925 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
Acc | +1.329 |
Acc | +1.197 |
solved | +0.997 |
fixed | +0.959 |
collapse | +0.936 |
Neg logit contributions | |
banks | -1.157 |
ippers | -1.138 |
pouch | -1.116 |
bles | -1.092 |
socks | -1.054 |
Token | to |
Token ID | 284 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
ippers | +0.577 |
banks | +0.569 |
bles | +0.559 |
pouch | +0.555 |
ittens | +0.526 |
Neg logit contributions | |
Acc | -0.418 |
Acc | -0.375 |
solved | -0.298 |
fixed | -0.292 |
collapse | -0.272 |
Token | the |
Token ID | 262 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
ippers | +0.286 |
banks | +0.283 |
bles | +0.275 |
pouch | +0.275 |
ittens | +0.257 |
Neg logit contributions | |
Acc | -0.297 |
Acc | -0.271 |
solved | -0.227 |
fixed | -0.223 |
collapse | -0.211 |
Token | hill |
Token ID | 12788 |
Feature activation | +9.42e-02 |
Pos logit contributions | |
Acc | +0.696 |
Acc | +0.657 |
solved | +0.576 |
fixed | +0.570 |
collapse | +0.545 |
Neg logit contributions | |
banks | -0.258 |
ippers | -0.242 |
pouch | -0.234 |
bles | -0.232 |
pair | -0.206 |
Token | ." |
Token ID | 526 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
banks | +0.137 |
ippers | +0.136 |
bles | +0.132 |
pouch | +0.131 |
ters | +0.124 |
Neg logit contributions | |
Acc | -0.177 |
Acc | -0.164 |
solved | -0.138 |
fixed | -0.136 |
collapse | -0.131 |
Token |
|
Token ID | 198 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
bles | +5.226 |
kins | +5.175 |
itive | +5.147 |
ippers | +4.886 |
ters | +4.694 |
Neg logit contributions | |
Acc | -4.737 |
Acc | -4.730 |
fixed | -3.752 |
Bye | -3.723 |
� | -3.670 |
Token |
|
Token ID | 198 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
ippers | +3.636 |
banks | +3.619 |
bles | +3.551 |
itive | +3.542 |
sters | +3.480 |
Neg logit contributions | |
Acc | -2.954 |
Acc | -2.639 |
fixed | -2.231 |
solved | -2.194 |
� | -1.998 |
Token | So |
Token ID | 2396 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
ippers | +4.919 |
kins | +4.823 |
itive | +4.600 |
bles | +4.585 |
angers | +4.569 |
Neg logit contributions | |
Acc | -4.413 |
Acc | -3.799 |
solved | -3.496 |
fixed | -3.408 |
� | -3.380 |
Token | , |
Token ID | 11 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
ippers | +4.407 |
bles | +4.289 |
banks | +4.262 |
angers | +4.171 |
kins | +4.149 |
Neg logit contributions | |
Acc | -3.667 |
Acc | -3.511 |
fixed | -2.997 |
solved | -2.984 |
� | -2.905 |
Token | they |
Token ID | 484 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
ippers | +5.392 |
kins | +5.198 |
banks | +5.152 |
bles | +5.143 |
angers | +4.960 |
Neg logit contributions | |
Acc | -4.937 |
Acc | -4.485 |
fixed | -4.001 |
� | -3.993 |
solved | -3.920 |
Token | all |
Token ID | 477 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
ippers | +0.904 |
banks | +0.891 |
bles | +0.870 |
pouch | +0.861 |
kins | +0.815 |
Neg logit contributions | |
Acc | -0.914 |
Acc | -0.831 |
solved | -0.701 |
fixed | -0.695 |
collapse | -0.648 |
Token | started |
Token ID | 2067 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
Acc | +0.040 |
Acc | +0.037 |
solved | +0.030 |
fixed | +0.029 |
collapse | +0.028 |
Neg logit contributions | |
banks | -0.036 |
ippers | -0.036 |
bles | -0.035 |
pouch | -0.035 |
itive | -0.033 |
Token | climbing |
Token ID | 14281 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
ippers | +0.298 |
banks | +0.298 |
bles | +0.288 |
pouch | +0.287 |
ittens | +0.273 |
Neg logit contributions | |
Acc | -0.240 |
Acc | -0.217 |
solved | -0.175 |
fixed | -0.172 |
collapse | -0.159 |
Token | the |
Token ID | 262 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
banks | +0.110 |
ippers | +0.108 |
bles | +0.106 |
pouch | +0.104 |
ters | +0.099 |
Neg logit contributions | |
Acc | -0.127 |
Acc | -0.118 |
solved | -0.098 |
fixed | -0.097 |
collapse | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ippers | +2.214 |
banks | +2.136 |
bles | +2.122 |
pouch | +2.112 |
ittens | +2.029 |
Neg logit contributions | |
Acc | -2.005 |
Acc | -1.826 |
fixed | -1.550 |
solved | -1.542 |
collapse | -1.401 |
Token | One |
Token ID | 1881 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
ippers | +2.462 |
banks | +2.441 |
itive | +2.394 |
bles | +2.375 |
pouch | +2.371 |
Neg logit contributions | |
Acc | -2.575 |
Acc | -2.380 |
solved | -1.982 |
fixed | -1.980 |
bell | -1.794 |
Token | day |
Token ID | 1110 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Acc | +1.198 |
Acc | +1.129 |
solved | +0.979 |
fixed | +0.965 |
collapse | +0.959 |
Neg logit contributions | |
banks | -0.489 |
ippers | -0.482 |
pouch | -0.473 |
bles | -0.449 |
friends | -0.447 |
Token | , |
Token ID | 11 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
banks | +2.327 |
ippers | +2.256 |
pouch | +2.216 |
bles | +2.174 |
itive | +2.098 |
Neg logit contributions | |
Acc | -2.560 |
Acc | -2.367 |
solved | -2.049 |
fixed | -2.041 |
collapse | -1.833 |
Token | P |
Token ID | 350 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
banks | +2.798 |
ippers | +2.770 |
pouch | +2.727 |
itive | +2.589 |
bles | +2.581 |
Neg logit contributions | |
Acc | -2.977 |
Acc | -2.689 |
fixed | -2.367 |
solved | -2.355 |
collapse | -2.085 |
Token | uff |
Token ID | 1648 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
ippers | +0.353 |
banks | +0.342 |
bles | +0.330 |
pouch | +0.318 |
ittens | +0.286 |
Neg logit contributions | |
Acc | -1.007 |
Acc | -0.946 |
solved | -0.845 |
fixed | -0.840 |
collapse | -0.807 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
ippers | +2.183 |
banks | +2.182 |
pouch | +2.111 |
bles | +2.033 |
ittens | +1.961 |
Neg logit contributions | |
Acc | -2.061 |
Acc | -1.844 |
solved | -1.597 |
fixed | -1.597 |
bell | -1.403 |
Token | his |
Token ID | 465 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
ippers | +0.169 |
banks | +0.168 |
bles | +0.165 |
pouch | +0.161 |
ters | +0.153 |
Neg logit contributions | |
Acc | -0.214 |
Acc | -0.196 |
solved | -0.167 |
fixed | -0.164 |
collapse | -0.159 |
Token | friend |
Token ID | 1545 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
ippers | +0.988 |
banks | +0.981 |
bles | +0.936 |
pouch | +0.909 |
kins | +0.855 |
Neg logit contributions | |
Acc | -2.153 |
Acc | -2.028 |
solved | -1.794 |
fixed | -1.793 |
collapse | -1.698 |
Token | , |
Token ID | 11 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
Acc | +0.717 |
Acc | +0.645 |
solved | +0.544 |
fixed | +0.532 |
collapse | +0.526 |
Neg logit contributions | |
ippers | -0.520 |
banks | -0.518 |
bles | -0.506 |
pouch | -0.504 |
friends | -0.488 |
Token | a |
Token ID | 257 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
Acc | +0.861 |
Acc | +0.783 |
solved | +0.635 |
fixed | +0.620 |
collapse | +0.614 |
Neg logit contributions | |
ippers | -0.806 |
bles | -0.800 |
banks | -0.796 |
pouch | -0.781 |
friends | -0.764 |
Token | They |
Token ID | 1119 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
bles | +3.184 |
ippers | +3.171 |
banks | +3.094 |
itive | +3.079 |
kins | +2.982 |
Neg logit contributions | |
Acc | -3.129 |
Acc | -2.941 |
fixed | -2.349 |
solved | -2.346 |
collapse | -2.249 |
Token | see |
Token ID | 766 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
Acc | +0.029 |
Acc | +0.026 |
solved | +0.022 |
fixed | +0.021 |
collapse | +0.021 |
Neg logit contributions | |
ippers | -0.029 |
banks | -0.029 |
bles | -0.028 |
pouch | -0.027 |
kins | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
ippers | +0.681 |
banks | +0.671 |
bles | +0.654 |
pouch | +0.646 |
kins | +0.607 |
Neg logit contributions | |
Acc | -0.848 |
Acc | -0.779 |
solved | -0.666 |
fixed | -0.659 |
collapse | -0.624 |
Token | big |
Token ID | 1263 |
Feature activation | -7.56e-01 |
Pos logit contributions | |
Acc | +1.205 |
Acc | +1.104 |
solved | +0.932 |
fixed | +0.915 |
aling | +0.897 |
Neg logit contributions | |
pouch | -0.738 |
ippers | -0.737 |
bles | -0.729 |
banks | -0.725 |
pair | -0.704 |
Token | box |
Token ID | 3091 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
Acc | +1.707 |
Acc | +1.534 |
solved | +1.328 |
fixed | +1.288 |
collapse | +1.261 |
Neg logit contributions | |
pouch | -0.888 |
banks | -0.876 |
ippers | -0.868 |
pair | -0.832 |
bles | -0.824 |
Token | with |
Token ID | 351 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ippers | +0.679 |
banks | +0.669 |
bles | +0.652 |
pouch | +0.644 |
kins | +0.605 |
Neg logit contributions | |
Acc | -0.875 |
Acc | -0.805 |
solved | -0.689 |
fixed | -0.683 |
collapse | -0.646 |
Token | a |
Token ID | 257 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
ippers | +1.768 |
banks | +1.695 |
bles | +1.654 |
pouch | +1.640 |
ittens | +1.579 |
Neg logit contributions | |
Acc | -2.318 |
Acc | -2.142 |
fixed | -1.938 |
solved | -1.935 |
collapse | -1.764 |
Token | picture |
Token ID | 4286 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
ippers | +1.298 |
banks | +1.279 |
bles | +1.236 |
pouch | +1.201 |
ters | +1.164 |
Neg logit contributions | |
Acc | -1.632 |
Acc | -1.510 |
solved | -1.317 |
fixed | -1.312 |
collapse | -1.226 |
Token | of |
Token ID | 286 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Acc | +0.324 |
Acc | +0.298 |
solved | +0.251 |
fixed | +0.248 |
collapse | +0.241 |
Neg logit contributions | |
ippers | -0.190 |
banks | -0.183 |
bles | -0.182 |
pouch | -0.179 |
ters | -0.169 |
Token | a |
Token ID | 257 |
Feature activation | +4.79e-02 |
Pos logit contributions | |
ippers | +1.840 |
banks | +1.788 |
bles | +1.759 |
pouch | +1.722 |
itive | +1.667 |
Neg logit contributions | |
Acc | -1.835 |
Acc | -1.698 |
fixed | -1.463 |
solved | -1.460 |
collapse | -1.333 |
Token | lamp |
Token ID | 20450 |
Feature activation | +4.78e-01 |
Pos logit contributions | |
ippers | +0.066 |
banks | +0.065 |
pouch | +0.064 |
bles | +0.064 |
kins | +0.059 |
Neg logit contributions | |
Acc | -0.094 |
Acc | -0.086 |
solved | -0.073 |
fixed | -0.072 |
collapse | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | +1.29e-02 |
Pos logit contributions | |
Acc | +0.465 |
Acc | +0.424 |
solved | +0.339 |
fixed | +0.333 |
collapse | +0.328 |
Neg logit contributions | |
ippers | -0.433 |
banks | -0.428 |
bles | -0.417 |
pouch | -0.410 |
ters | -0.392 |
Token | become |
Token ID | 1716 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
ippers | +0.020 |
banks | +0.020 |
bles | +0.019 |
pouch | +0.019 |
kins | +0.018 |
Neg logit contributions | |
Acc | -0.024 |
Acc | -0.021 |
solved | -0.018 |
fixed | -0.018 |
collapse | -0.017 |
Token | independent |
Token ID | 4795 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
ippers | +0.275 |
banks | +0.272 |
bles | +0.266 |
pouch | +0.260 |
kins | +0.249 |
Neg logit contributions | |
Acc | -0.329 |
Acc | -0.303 |
solved | -0.253 |
fixed | -0.251 |
collapse | -0.241 |
Token | . |
Token ID | 13 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
ippers | +0.877 |
banks | +0.865 |
bles | +0.845 |
pouch | +0.836 |
ittens | +0.790 |
Neg logit contributions | |
Acc | -0.925 |
Acc | -0.844 |
solved | -0.712 |
fixed | -0.705 |
collapse | -0.660 |
Token | |
Token ID | 220 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
ippers | +1.434 |
banks | +1.397 |
bles | +1.394 |
pouch | +1.349 |
itive | +1.322 |
Neg logit contributions | |
Acc | -1.658 |
Acc | -1.538 |
solved | -1.301 |
fixed | -1.294 |
collapse | -1.197 |
Token |
|
Token ID | 198 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ippers | +1.039 |
banks | +1.029 |
bles | +0.998 |
pouch | +0.992 |
itive | +0.939 |
Neg logit contributions | |
Acc | -1.207 |
Acc | -1.107 |
solved | -0.941 |
fixed | -0.935 |
collapse | -0.877 |
Token |
|
Token ID | 198 |
Feature activation | +8.82e-01 |
Pos logit contributions | |
ippers | +2.336 |
banks | +2.301 |
bles | +2.249 |
pouch | +2.225 |
itive | +2.183 |
Neg logit contributions | |
Acc | -2.238 |
Acc | -2.005 |
solved | -1.715 |
fixed | -1.712 |
bell | -1.544 |
Token | First |
Token ID | 5962 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
ippers | +1.481 |
banks | +1.442 |
bles | +1.417 |
pouch | +1.408 |
kins | +1.336 |
Neg logit contributions | |
Acc | -1.493 |
Acc | -1.340 |
solved | -1.157 |
fixed | -1.134 |
collapse | -1.053 |
Token | , |
Token ID | 11 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
ippers | +1.148 |
banks | +1.125 |
bles | +1.102 |
pouch | +1.081 |
itive | +1.018 |
Neg logit contributions | |
Acc | -1.483 |
Acc | -1.374 |
solved | -1.188 |
fixed | -1.182 |
collapse | -1.106 |
Token | they |
Token ID | 484 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
ippers | +2.851 |
banks | +2.715 |
bles | +2.709 |
pouch | +2.665 |
itive | +2.555 |
Neg logit contributions | |
Acc | -2.588 |
Acc | -2.374 |
solved | -2.081 |
fixed | -2.081 |
collapse | -1.881 |
Token | studied |
Token ID | 9713 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
banks | +0.020 |
ippers | +0.020 |
bles | +0.019 |
pouch | +0.019 |
kins | +0.018 |
Neg logit contributions | |
Acc | -0.024 |
Acc | -0.022 |
solved | -0.018 |
fixed | -0.018 |
collapse | -0.017 |
Token | ! |
Token ID | 0 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ippers | +0.700 |
banks | +0.693 |
bles | +0.676 |
pouch | +0.669 |
kins | +0.633 |
Neg logit contributions | |
Acc | -0.722 |
Acc | -0.658 |
solved | -0.552 |
fixed | -0.547 |
collapse | -0.512 |
Token | She |
Token ID | 1375 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
ippers | +2.946 |
bles | +2.857 |
kins | +2.812 |
banks | +2.780 |
itive | +2.744 |
Neg logit contributions | |
Acc | -2.783 |
Acc | -2.551 |
solved | -2.187 |
fixed | -2.137 |
bell | -2.102 |
Token | was |
Token ID | 373 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ippers | +0.555 |
banks | +0.551 |
bles | +0.536 |
pouch | +0.529 |
kins | +0.497 |
Neg logit contributions | |
Acc | -0.712 |
Acc | -0.656 |
solved | -0.559 |
fixed | -0.554 |
collapse | -0.527 |
Token | so |
Token ID | 523 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
ippers | +1.527 |
banks | +1.471 |
bles | +1.430 |
pouch | +1.409 |
kins | +1.356 |
Neg logit contributions | |
Acc | -2.012 |
Acc | -1.847 |
solved | -1.621 |
fixed | -1.608 |
collapse | -1.506 |
Token | excited |
Token ID | 6568 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
Acc | +0.098 |
Acc | +0.091 |
solved | +0.080 |
fixed | +0.079 |
collapse | +0.075 |
Neg logit contributions | |
ippers | -0.055 |
banks | -0.054 |
bles | -0.053 |
pouch | -0.051 |
friends | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
ippers | +0.546 |
banks | +0.542 |
bles | +0.529 |
pouch | +0.523 |
ittens | +0.498 |
Neg logit contributions | |
Acc | -0.502 |
Acc | -0.456 |
solved | -0.375 |
fixed | -0.371 |
collapse | -0.348 |
Token | She |
Token ID | 1375 |
Feature activation | -9.19e-02 |
Pos logit contributions | |
ippers | +2.290 |
bles | +2.232 |
banks | +2.189 |
kins | +2.158 |
itive | +2.128 |
Neg logit contributions | |
Acc | -2.376 |
Acc | -2.200 |
solved | -1.846 |
fixed | -1.809 |
bell | -1.739 |
Token | scoop |
Token ID | 23529 |
Feature activation | +9.43e-01 |
Pos logit contributions | |
Acc | +0.175 |
Acc | +0.163 |
solved | +0.135 |
fixed | +0.134 |
collapse | +0.132 |
Neg logit contributions | |
banks | -0.131 |
ippers | -0.129 |
bles | -0.126 |
pouch | -0.125 |
ittens | -0.117 |
Token | ed |
Token ID | 276 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
ippers | +0.924 |
banks | +0.898 |
pouch | +0.844 |
bles | +0.840 |
ittens | +0.773 |
Neg logit contributions | |
Acc | -2.219 |
Acc | -2.060 |
solved | -1.865 |
fixed | -1.855 |
collapse | -1.762 |
Token | up |
Token ID | 510 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
Acc | +1.493 |
Acc | +1.395 |
solved | +1.164 |
fixed | +1.138 |
collapse | +1.118 |
Neg logit contributions | |
banks | -0.988 |
ippers | -0.958 |
bles | -0.943 |
pouch | -0.930 |
ters | -0.902 |
Token | the |
Token ID | 262 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
Acc | +1.257 |
Acc | +1.192 |
solved | +0.989 |
fixed | +0.966 |
collapse | +0.955 |
Neg logit contributions | |
banks | -0.840 |
ippers | -0.828 |
bles | -0.819 |
pouch | -0.797 |
friends | -0.778 |
Token | see |
Token ID | 766 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
ippers | +0.562 |
banks | +0.556 |
bles | +0.543 |
pouch | +0.537 |
kins | +0.510 |
Neg logit contributions | |
Acc | -0.502 |
Acc | -0.452 |
solved | -0.370 |
fixed | -0.366 |
collapse | -0.342 |
Token | the |
Token ID | 262 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
Acc | +0.849 |
Acc | +0.768 |
solved | +0.639 |
fixed | +0.623 |
bell | +0.615 |
Neg logit contributions | |
banks | -0.776 |
ippers | -0.771 |
bles | -0.758 |
pouch | -0.751 |
friends | -0.734 |
Token | colors |
Token ID | 7577 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
ippers | +1.190 |
banks | +1.164 |
bles | +1.131 |
pouch | +1.127 |
ittens | +1.071 |
Neg logit contributions | |
Acc | -1.530 |
Acc | -1.406 |
solved | -1.224 |
fixed | -1.217 |
collapse | -1.137 |
Token | change |
Token ID | 1487 |
Feature activation | -6.85e-01 |
Pos logit contributions | |
Acc | +0.830 |
Acc | +0.767 |
solved | +0.637 |
fixed | +0.625 |
collapse | +0.589 |
Neg logit contributions | |
ippers | -0.575 |
bles | -0.574 |
banks | -0.567 |
pouch | -0.539 |
ters | -0.526 |
Token | . |
Token ID | 13 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Acc | +1.342 |
Acc | +1.213 |
solved | +1.009 |
fixed | +0.974 |
aling | +0.942 |
Neg logit contributions | |
ippers | -0.997 |
banks | -0.989 |
bles | -0.980 |
pouch | -0.931 |
ters | -0.926 |
Token |
|
Token ID | 198 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ippers | +2.712 |
bles | +2.614 |
itive | +2.596 |
banks | +2.578 |
kins | +2.538 |
Neg logit contributions | |
Acc | -2.659 |
Acc | -2.490 |
solved | -2.091 |
fixed | -2.048 |
collapse | -1.854 |
Token |
|
Token ID | 198 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
ippers | +1.730 |
banks | +1.691 |
bles | +1.654 |
pouch | +1.636 |
itive | +1.582 |
Neg logit contributions | |
Acc | -1.792 |
Acc | -1.623 |
solved | -1.391 |
fixed | -1.375 |
collapse | -1.273 |
Token | Fin |
Token ID | 18467 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
ippers | +3.383 |
banks | +3.084 |
bles | +3.081 |
pouch | +3.079 |
angers | +3.077 |
Neg logit contributions | |
Acc | -3.310 |
Acc | -2.822 |
solved | -2.699 |
fixed | -2.582 |
aling | -2.410 |
Token | gave |
Token ID | 2921 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
banks | +0.324 |
ippers | +0.322 |
bles | +0.315 |
pouch | +0.309 |
kins | +0.292 |
Neg logit contributions | |
Acc | -0.380 |
Acc | -0.350 |
solved | -0.291 |
fixed | -0.288 |
collapse | -0.279 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
ippers | +1.359 |
banks | +1.338 |
bles | +1.301 |
pouch | +1.290 |
itive | +1.239 |
Neg logit contributions | |
Acc | -1.323 |
Acc | -1.203 |
solved | -1.020 |
fixed | -1.020 |
collapse | -0.933 |
Token | shiny |
Token ID | 22441 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ippers | +1.829 |
banks | +1.784 |
bles | +1.733 |
pouch | +1.711 |
ittens | +1.664 |
Neg logit contributions | |
Acc | -2.541 |
Acc | -2.387 |
fixed | -2.110 |
solved | -2.099 |
collapse | -1.927 |
Token | . |
Token ID | 13 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
ippers | +0.588 |
banks | +0.583 |
bles | +0.568 |
pouch | +0.562 |
ters | +0.531 |
Neg logit contributions | |
Acc | -0.629 |
Acc | -0.574 |
solved | -0.482 |
fixed | -0.476 |
collapse | -0.448 |
Token | He |
Token ID | 679 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
ippers | +0.543 |
banks | +0.532 |
bles | +0.519 |
pouch | +0.511 |
kins | +0.479 |
Neg logit contributions | |
Acc | -0.763 |
Acc | -0.702 |
solved | -0.607 |
fixed | -0.602 |
collapse | -0.571 |
Token | says |
Token ID | 1139 |
Feature activation | -6.83e-01 |
Pos logit contributions | |
Acc | +0.528 |
Acc | +0.480 |
solved | +0.391 |
fixed | +0.388 |
collapse | +0.373 |
Neg logit contributions | |
banks | -0.486 |
ippers | -0.476 |
bles | -0.471 |
pouch | -0.464 |
friends | -0.441 |
Token | they |
Token ID | 484 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
Acc | +1.437 |
Acc | +1.302 |
solved | +1.154 |
fixed | +1.137 |
aling | +1.116 |
Neg logit contributions | |
ippers | -0.831 |
bles | -0.829 |
friends | -0.783 |
banks | -0.776 |
pouch | -0.761 |
Token | are |
Token ID | 389 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
ippers | +1.032 |
banks | +1.016 |
bles | +0.991 |
pouch | +0.981 |
kins | +0.930 |
Neg logit contributions | |
Acc | -1.094 |
Acc | -0.997 |
solved | -0.849 |
fixed | -0.841 |
collapse | -0.783 |
Token | bad |
Token ID | 2089 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
Acc | +1.273 |
Acc | +1.171 |
bell | +0.957 |
solved | +0.929 |
fixed | +0.920 |
Neg logit contributions | |
bles | -1.104 |
ippers | -1.080 |
friends | -1.059 |
banks | -1.039 |
ters | -1.016 |
Token | kids |
Token ID | 3988 |
Feature activation | -9.81e-01 |
Pos logit contributions | |
ippers | +0.650 |
banks | +0.640 |
bles | +0.616 |
pouch | +0.610 |
kins | +0.560 |
Neg logit contributions | |
Acc | -1.238 |
Acc | -1.153 |
solved | -1.014 |
fixed | -1.007 |
collapse | -0.960 |
Token | . |
Token ID | 13 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
Acc | +1.948 |
Acc | +1.808 |
solved | +1.491 |
bell | +1.460 |
collapse | +1.451 |
Neg logit contributions | |
ippers | -1.330 |
bles | -1.316 |
banks | -1.302 |
friends | -1.290 |
pouch | -1.279 |
Token | He |
Token ID | 679 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
ippers | +0.235 |
banks | +0.228 |
bles | +0.223 |
pouch | +0.219 |
ters | +0.206 |
Neg logit contributions | |
Acc | -0.346 |
Acc | -0.317 |
solved | -0.277 |
fixed | -0.274 |
collapse | -0.262 |
Token | says |
Token ID | 1139 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
Acc | +1.299 |
Acc | +1.189 |
solved | +0.966 |
collapse | +0.953 |
fixed | +0.949 |
Neg logit contributions | |
banks | -0.999 |
bles | -0.968 |
ippers | -0.967 |
pouch | -0.952 |
friends | -0.933 |
Token | they |
Token ID | 484 |
Feature activation | +7.25e-01 |
Pos logit contributions | |
Acc | +0.697 |
Acc | +0.641 |
solved | +0.565 |
fixed | +0.555 |
aling | +0.535 |
Neg logit contributions | |
ippers | -0.396 |
bles | -0.393 |
banks | -0.384 |
pouch | -0.366 |
friends | -0.364 |
Token | pirate |
Token ID | 25868 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
Acc | +0.254 |
Acc | +0.233 |
solved | +0.203 |
fixed | +0.200 |
collapse | +0.191 |
Neg logit contributions | |
banks | -0.140 |
ippers | -0.139 |
pouch | -0.134 |
bles | -0.132 |
itive | -0.123 |
Token | ship |
Token ID | 4074 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
ippers | +1.375 |
banks | +1.347 |
bles | +1.315 |
pouch | +1.299 |
ittens | +1.241 |
Neg logit contributions | |
Acc | -1.475 |
Acc | -1.353 |
solved | -1.167 |
fixed | -1.161 |
collapse | -1.081 |
Token | ! |
Token ID | 0 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
banks | +0.195 |
ippers | +0.195 |
bles | +0.191 |
pouch | +0.188 |
ters | +0.179 |
Neg logit contributions | |
Acc | -0.198 |
Acc | -0.180 |
solved | -0.147 |
fixed | -0.144 |
collapse | -0.136 |
Token | The |
Token ID | 383 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
Acc | +1.703 |
Acc | +1.507 |
solved | +1.302 |
fixed | +1.295 |
collapse | +1.266 |
Neg logit contributions | |
banks | -1.421 |
ippers | -1.397 |
pouch | -1.336 |
bles | -1.299 |
ters | -1.266 |
Token | pirate |
Token ID | 25868 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
ippers | +0.898 |
banks | +0.880 |
bles | +0.856 |
pouch | +0.843 |
ittens | +0.796 |
Neg logit contributions | |
Acc | -1.304 |
Acc | -1.209 |
solved | -1.051 |
fixed | -1.045 |
collapse | -0.987 |
Token | ship |
Token ID | 4074 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
ippers | +1.627 |
banks | +1.571 |
bles | +1.533 |
pouch | +1.525 |
ittens | +1.463 |
Neg logit contributions | |
Acc | -1.683 |
Acc | -1.540 |
solved | -1.349 |
fixed | -1.346 |
collapse | -1.229 |
Token | was |
Token ID | 373 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
ippers | +0.904 |
banks | +0.889 |
bles | +0.868 |
pouch | +0.859 |
ittens | +0.813 |
Neg logit contributions | |
Acc | -0.973 |
Acc | -0.888 |
solved | -0.755 |
fixed | -0.748 |
collapse | -0.701 |
Token | trying |
Token ID | 2111 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
banks | +0.186 |
ippers | +0.186 |
bles | +0.181 |
pouch | +0.180 |
ters | +0.168 |
Neg logit contributions | |
Acc | -0.239 |
Acc | -0.219 |
solved | -0.182 |
fixed | -0.180 |
collapse | -0.172 |
Token | to |
Token ID | 284 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
Acc | +0.918 |
Acc | +0.846 |
solved | +0.649 |
fixed | +0.640 |
bell | +0.620 |
Neg logit contributions | |
banks | -1.035 |
ippers | -1.004 |
bles | -0.995 |
pouch | -0.991 |
friends | -0.957 |
Token | get |
Token ID | 651 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
banks | +0.124 |
ippers | +0.122 |
bles | +0.119 |
pouch | +0.118 |
ters | +0.112 |
Neg logit contributions | |
Acc | -0.130 |
Acc | -0.118 |
solved | -0.096 |
fixed | -0.094 |
bell | -0.088 |
Token | away |
Token ID | 1497 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
Acc | +1.056 |
Acc | +0.964 |
solved | +0.766 |
fixed | +0.729 |
bell | +0.729 |
Neg logit contributions | |
banks | -0.906 |
ippers | -0.883 |
bles | -0.870 |
pouch | -0.870 |
friends | -0.839 |
Token | started |
Token ID | 2067 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
Acc | +0.873 |
Acc | +0.799 |
solved | +0.641 |
collapse | +0.633 |
fixed | +0.632 |
Neg logit contributions | |
banks | -0.674 |
ippers | -0.673 |
bles | -0.657 |
pouch | -0.651 |
pair | -0.615 |
Token | the |
Token ID | 262 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
ippers | +0.222 |
banks | +0.221 |
bles | +0.215 |
pouch | +0.212 |
ters | +0.202 |
Neg logit contributions | |
Acc | -0.206 |
Acc | -0.186 |
solved | -0.152 |
fixed | -0.150 |
collapse | -0.141 |
Token | race |
Token ID | 3234 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
ippers | +1.123 |
banks | +1.103 |
bles | +1.081 |
pouch | +1.073 |
kins | +1.020 |
Neg logit contributions | |
Acc | -1.042 |
Acc | -0.943 |
solved | -0.792 |
fixed | -0.785 |
collapse | -0.731 |
Token | . |
Token ID | 13 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
Acc | +0.028 |
Acc | +0.026 |
solved | +0.021 |
fixed | +0.021 |
collapse | +0.020 |
Neg logit contributions | |
ippers | -0.025 |
banks | -0.025 |
bles | -0.025 |
pouch | -0.024 |
ters | -0.023 |
Token | Sarah |
Token ID | 10490 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
Acc | +0.802 |
Acc | +0.704 |
solved | +0.586 |
fixed | +0.581 |
collapse | +0.557 |
Neg logit contributions | |
banks | -0.773 |
ippers | -0.765 |
pouch | -0.729 |
bles | -0.728 |
ters | -0.705 |
Token | was |
Token ID | 373 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
ippers | +0.361 |
banks | +0.359 |
bles | +0.349 |
pouch | +0.345 |
ters | +0.323 |
Neg logit contributions | |
Acc | -0.461 |
Acc | -0.423 |
solved | -0.357 |
fixed | -0.352 |
collapse | -0.340 |
Token | very |
Token ID | 845 |
Feature activation | +5.73e-01 |
Pos logit contributions | |
ippers | +0.939 |
banks | +0.926 |
bles | +0.898 |
pouch | +0.888 |
kins | +0.837 |
Neg logit contributions | |
Acc | -1.219 |
Acc | -1.120 |
solved | -0.970 |
fixed | -0.963 |
collapse | -0.905 |
Token | quick |
Token ID | 2068 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ippers | +0.875 |
banks | +0.866 |
bles | +0.841 |
pouch | +0.832 |
ittens | +0.783 |
Neg logit contributions | |
Acc | -1.043 |
Acc | -0.958 |
solved | -0.818 |
fixed | -0.811 |
collapse | -0.763 |
Token | , |
Token ID | 11 |
Feature activation | -2.02e-01 |
Pos logit contributions | |
ippers | +1.788 |
banks | +1.761 |
bles | +1.724 |
pouch | +1.706 |
kins | +1.631 |
Neg logit contributions | |
Acc | -1.660 |
Acc | -1.511 |
solved | -1.282 |
fixed | -1.276 |
collapse | -1.169 |
Token | but |
Token ID | 475 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
Acc | +0.444 |
Acc | +0.417 |
solved | +0.361 |
fixed | +0.355 |
collapse | +0.348 |
Neg logit contributions | |
banks | -0.224 |
ippers | -0.224 |
bles | -0.214 |
pouch | -0.214 |
ters | -0.199 |
Token | Sam |
Token ID | 3409 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
Acc | +0.496 |
Acc | +0.448 |
solved | +0.353 |
fixed | +0.352 |
collapse | +0.335 |
Neg logit contributions | |
ippers | -0.551 |
banks | -0.544 |
bles | -0.543 |
pouch | -0.530 |
friends | -0.518 |
Token | a |
Token ID | 257 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ippers | +0.458 |
banks | +0.453 |
bles | +0.442 |
pouch | +0.435 |
itive | +0.411 |
Neg logit contributions | |
Acc | -0.543 |
Acc | -0.502 |
solved | -0.423 |
fixed | -0.419 |
collapse | -0.398 |
Token | time |
Token ID | 640 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
Acc | +4.402 |
Acc | +4.357 |
collapse | +3.651 |
fixed | +3.590 |
solved | +3.578 |
Neg logit contributions | |
ittens | -2.229 |
ippers | -2.228 |
banks | -2.182 |
friends | -2.121 |
pouch | -2.074 |
Token | , |
Token ID | 11 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Acc | +0.979 |
Acc | +0.891 |
solved | +0.787 |
fixed | +0.782 |
aling | +0.746 |
Neg logit contributions | |
ippers | -0.540 |
bles | -0.518 |
banks | -0.513 |
pouch | -0.494 |
ittens | -0.476 |
Token | there |
Token ID | 612 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
Acc | +0.324 |
Acc | +0.306 |
solved | +0.269 |
fixed | +0.266 |
collapse | +0.259 |
Neg logit contributions | |
ippers | -0.125 |
bles | -0.122 |
banks | -0.122 |
pouch | -0.117 |
kins | -0.108 |
Token | was |
Token ID | 373 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
ippers | +0.023 |
banks | +0.023 |
bles | +0.022 |
pouch | +0.022 |
kins | +0.020 |
Neg logit contributions | |
Acc | -0.033 |
Acc | -0.030 |
solved | -0.026 |
fixed | -0.026 |
collapse | -0.025 |
Token | an |
Token ID | 281 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
Acc | +1.225 |
Acc | +1.172 |
solved | +0.980 |
collapse | +0.969 |
fixed | +0.966 |
Neg logit contributions | |
ippers | -0.820 |
banks | -0.818 |
bles | -0.789 |
pouch | -0.779 |
friends | -0.772 |
Token | ordinary |
Token ID | 8850 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Acc | +2.012 |
Acc | +1.857 |
solved | +1.669 |
collapse | +1.625 |
fixed | +1.614 |
Neg logit contributions | |
banks | -1.046 |
ippers | -1.014 |
bles | -1.001 |
pouch | -0.976 |
ranc | -0.965 |
Token | boy |
Token ID | 2933 |
Feature activation | -9.27e-01 |
Pos logit contributions | |
Acc | +0.601 |
Acc | +0.556 |
solved | +0.500 |
fixed | +0.492 |
collapse | +0.485 |
Neg logit contributions | |
ippers | -0.231 |
banks | -0.225 |
pouch | -0.221 |
bles | -0.215 |
couple | -0.201 |
Token | who |
Token ID | 508 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
Acc | +1.848 |
Acc | +1.695 |
collapse | +1.464 |
solved | +1.452 |
fixed | +1.413 |
Neg logit contributions | |
ippers | -1.245 |
banks | -1.199 |
pouch | -1.195 |
friends | -1.166 |
bles | -1.150 |
Token | liked |
Token ID | 8288 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
Acc | +0.447 |
Acc | +0.409 |
solved | +0.340 |
fixed | +0.335 |
collapse | +0.335 |
Neg logit contributions | |
ippers | -0.362 |
banks | -0.362 |
pouch | -0.350 |
bles | -0.344 |
robe | -0.328 |
Token | to |
Token ID | 284 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
Acc | +0.755 |
Acc | +0.680 |
solved | +0.566 |
fixed | +0.546 |
collapse | +0.536 |
Neg logit contributions | |
ippers | -0.879 |
banks | -0.878 |
pouch | -0.842 |
bles | -0.841 |
friends | -0.834 |
Token | . |
Token ID | 13 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
ippers | +0.103 |
banks | +0.101 |
bles | +0.099 |
pouch | +0.097 |
kins | +0.093 |
Neg logit contributions | |
Acc | -0.117 |
Acc | -0.107 |
solved | -0.089 |
fixed | -0.087 |
collapse | -0.084 |
Token |
|
Token ID | 198 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
ippers | +0.051 |
banks | +0.050 |
pouch | +0.049 |
bles | +0.048 |
ittens | +0.045 |
Neg logit contributions | |
Acc | -0.066 |
Acc | -0.060 |
solved | -0.052 |
fixed | -0.052 |
collapse | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
ippers | +0.435 |
banks | +0.428 |
bles | +0.417 |
pouch | +0.412 |
ittens | +0.389 |
Neg logit contributions | |
Acc | -0.553 |
Acc | -0.509 |
solved | -0.436 |
fixed | -0.432 |
collapse | -0.410 |
Token | When |
Token ID | 2215 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
ippers | +0.521 |
banks | +0.516 |
bles | +0.503 |
pouch | +0.497 |
ittens | +0.470 |
Neg logit contributions | |
Acc | -0.584 |
Acc | -0.538 |
solved | -0.452 |
fixed | -0.451 |
collapse | -0.424 |
Token | she |
Token ID | 673 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
ippers | +0.319 |
banks | +0.315 |
pouch | +0.304 |
bles | +0.304 |
kins | +0.284 |
Neg logit contributions | |
Acc | -0.455 |
Acc | -0.421 |
solved | -0.359 |
fixed | -0.357 |
collapse | -0.338 |
Token | got |
Token ID | 1392 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
Acc | +0.968 |
Acc | +0.900 |
solved | +0.714 |
collapse | +0.707 |
fixed | +0.696 |
Neg logit contributions | |
banks | -0.721 |
ippers | -0.721 |
pouch | -0.714 |
bles | -0.696 |
pair | -0.667 |
Token | home |
Token ID | 1363 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
Acc | +0.907 |
Acc | +0.851 |
solved | +0.724 |
collapse | +0.704 |
fixed | +0.702 |
Neg logit contributions | |
ippers | -0.484 |
pouch | -0.464 |
banks | -0.461 |
bles | -0.453 |
friends | -0.438 |
Token | she |
Token ID | 673 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
ippers | +0.437 |
banks | +0.429 |
bles | +0.418 |
pouch | +0.412 |
kins | +0.387 |
Neg logit contributions | |
Acc | -0.638 |
Acc | -0.587 |
solved | -0.508 |
fixed | -0.501 |
collapse | -0.477 |
Token | showed |
Token ID | 3751 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
Acc | +1.425 |
Acc | +1.318 |
collapse | +1.092 |
solved | +1.072 |
collapsing | +1.060 |
Neg logit contributions | |
banks | -1.001 |
ippers | -0.992 |
pouch | -0.980 |
bles | -0.952 |
pair | -0.920 |
Token | her |
Token ID | 607 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
ippers | +0.395 |
banks | +0.389 |
bles | +0.381 |
pouch | +0.377 |
ters | +0.355 |
Neg logit contributions | |
Acc | -0.445 |
Acc | -0.406 |
solved | -0.341 |
fixed | -0.336 |
collapse | -0.321 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
ippers | +1.258 |
banks | +1.243 |
bles | +1.198 |
pouch | +1.157 |
ters | +1.123 |
Neg logit contributions | |
Acc | -2.028 |
Acc | -1.894 |
fixed | -1.659 |
solved | -1.645 |
collapse | -1.538 |
Token | making |
Token ID | 1642 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
ippers | +1.088 |
banks | +1.077 |
bles | +1.048 |
pouch | +1.045 |
kins | +0.991 |
Neg logit contributions | |
Acc | -0.927 |
Acc | -0.837 |
solved | -0.687 |
fixed | -0.683 |
collapse | -0.630 |
Token | a |
Token ID | 257 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Acc | +1.056 |
Acc | +0.963 |
solved | +0.825 |
fixed | +0.790 |
collapse | +0.789 |
Neg logit contributions | |
banks | -0.721 |
ippers | -0.718 |
bles | -0.714 |
friends | -0.685 |
ters | -0.668 |
Token | mess |
Token ID | 2085 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
Acc | +2.080 |
Acc | +1.876 |
solved | +1.538 |
aling | +1.474 |
fixed | +1.463 |
Neg logit contributions | |
ippers | -1.906 |
banks | -1.885 |
bles | -1.866 |
friends | -1.845 |
pouch | -1.818 |
Token | ?" |
Token ID | 1701 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
Acc | +0.087 |
Acc | +0.079 |
solved | +0.067 |
fixed | +0.065 |
collapse | +0.063 |
Neg logit contributions | |
ippers | -0.081 |
bles | -0.079 |
banks | -0.078 |
pouch | -0.074 |
ters | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
Acc | +1.115 |
Acc | +1.013 |
solved | +0.903 |
fixed | +0.894 |
collapse | +0.840 |
Neg logit contributions | |
ippers | -0.743 |
banks | -0.707 |
bles | -0.695 |
pouch | -0.689 |
friends | -0.677 |
Token |
|
Token ID | 198 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Acc | +1.903 |
Acc | +1.735 |
solved | +1.541 |
fixed | +1.520 |
aling | +1.434 |
Neg logit contributions | |
ippers | -1.102 |
bles | -1.045 |
banks | -1.026 |
pouch | -1.015 |
friends | -0.995 |
Token | Tom |
Token ID | 13787 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
Acc | +2.231 |
Acc | +2.034 |
fixed | +1.766 |
solved | +1.736 |
bell | +1.609 |
Neg logit contributions | |
ippers | -1.601 |
bles | -1.537 |
banks | -1.503 |
pouch | -1.494 |
ittens | -1.473 |
Token | and |
Token ID | 290 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
Acc | +0.663 |
Acc | +0.613 |
solved | +0.521 |
fixed | +0.513 |
collapse | +0.499 |
Neg logit contributions | |
ippers | -0.523 |
bles | -0.517 |
banks | -0.517 |
pouch | -0.500 |
ters | -0.471 |
Token | Mia |
Token ID | 32189 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
Acc | +1.230 |
Acc | +1.084 |
solved | +0.896 |
fixed | +0.892 |
aling | +0.815 |
Neg logit contributions | |
ippers | -1.463 |
banks | -1.453 |
bles | -1.452 |
friends | -1.439 |
pair | -1.423 |
Token | stop |
Token ID | 2245 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
ippers | +0.503 |
banks | +0.499 |
bles | +0.485 |
pouch | +0.478 |
ters | +0.450 |
Neg logit contributions | |
Acc | -0.607 |
Acc | -0.556 |
solved | -0.475 |
fixed | -0.470 |
collapse | -0.443 |
Token | fighting |
Token ID | 4330 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
Acc | +0.653 |
Acc | +0.587 |
solved | +0.516 |
fixed | +0.495 |
bell | +0.462 |
Neg logit contributions | |
ippers | -0.636 |
banks | -0.624 |
bles | -0.611 |
pouch | -0.589 |
friends | -0.579 |
Token | was |
Token ID | 373 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
Acc | +0.347 |
Acc | +0.316 |
solved | +0.272 |
fixed | +0.268 |
aling | +0.258 |
Neg logit contributions | |
banks | -0.221 |
ippers | -0.221 |
bles | -0.213 |
pouch | -0.213 |
kins | -0.200 |
Token | a |
Token ID | 257 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
Acc | +1.034 |
Acc | +0.984 |
solved | +0.826 |
fixed | +0.814 |
collapse | +0.812 |
Neg logit contributions | |
ippers | -0.716 |
banks | -0.711 |
bles | -0.684 |
pouch | -0.678 |
friends | -0.666 |
Token | little |
Token ID | 1310 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
Acc | +2.282 |
Acc | +2.155 |
solved | +1.927 |
collapse | +1.914 |
fixed | +1.887 |
Neg logit contributions | |
ippers | -0.681 |
pouch | -0.676 |
bles | -0.671 |
couple | -0.655 |
pair | -0.647 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Acc | +0.457 |
Acc | +0.427 |
solved | +0.388 |
fixed | +0.380 |
collapse | +0.375 |
Neg logit contributions | |
ippers | -0.117 |
pouch | -0.112 |
banks | -0.112 |
bles | -0.108 |
couple | -0.094 |
Token | named |
Token ID | 3706 |
Feature activation | -9.87e-01 |
Pos logit contributions | |
Acc | +2.384 |
Acc | +2.169 |
collapse | +1.946 |
solved | +1.899 |
fixed | +1.845 |
Neg logit contributions | |
pouch | -1.484 |
ippers | -1.468 |
banks | -1.432 |
friends | -1.421 |
ittens | -1.361 |
Token | Anna |
Token ID | 11735 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
Acc | +1.604 |
Acc | +1.457 |
solved | +1.184 |
fixed | +1.183 |
aling | +1.171 |
Neg logit contributions | |
banks | -1.626 |
pouch | -1.589 |
ippers | -1.586 |
bles | -1.582 |
friends | -1.545 |
Token | . |
Token ID | 13 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
Acc | +3.191 |
Acc | +2.974 |
collapse | +2.668 |
aling | +2.562 |
solved | +2.552 |
Neg logit contributions | |
ippers | -1.783 |
pouch | -1.775 |
banks | -1.757 |
friends | -1.724 |
itive | -1.724 |
Token | She |
Token ID | 1375 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
ippers | +0.369 |
banks | +0.365 |
bles | +0.354 |
pouch | +0.353 |
ittens | +0.332 |
Neg logit contributions | |
Acc | -0.417 |
Acc | -0.381 |
solved | -0.322 |
fixed | -0.321 |
collapse | -0.302 |
Token | was |
Token ID | 373 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Acc | +0.327 |
Acc | +0.303 |
solved | +0.252 |
collapse | +0.250 |
fixed | +0.249 |
Neg logit contributions | |
ippers | -0.269 |
banks | -0.269 |
bles | -0.262 |
pouch | -0.259 |
kins | -0.246 |
Token | so |
Token ID | 523 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
ippers | +1.833 |
banks | +1.822 |
bles | +1.759 |
pouch | +1.725 |
kins | +1.675 |
Neg logit contributions | |
Acc | -1.937 |
Acc | -1.766 |
fixed | -1.505 |
solved | -1.501 |
collapse | -1.357 |
Token | shy |
Token ID | 15800 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
ippers | +0.556 |
banks | +0.544 |
bles | +0.532 |
pouch | +0.524 |
ittens | +0.492 |
Neg logit contributions | |
Acc | -0.758 |
Acc | -0.697 |
solved | -0.603 |
fixed | -0.597 |
collapse | -0.567 |
Token | roared |
Token ID | 45615 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
Acc | +0.394 |
Acc | +0.370 |
solved | +0.299 |
fixed | +0.296 |
collapse | +0.293 |
Neg logit contributions | |
banks | -0.293 |
ippers | -0.282 |
bles | -0.280 |
pouch | -0.279 |
friends | -0.260 |
Token | again |
Token ID | 757 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
Acc | +1.669 |
Acc | +1.565 |
solved | +1.235 |
fixed | +1.225 |
collapse | +1.198 |
Neg logit contributions | |
banks | -1.268 |
bles | -1.257 |
ippers | -1.240 |
ters | -1.178 |
pouch | -1.174 |
Token | , |
Token ID | 11 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
ippers | +0.310 |
banks | +0.309 |
bles | +0.300 |
pouch | +0.297 |
itive | +0.279 |
Neg logit contributions | |
Acc | -0.356 |
Acc | -0.325 |
solved | -0.272 |
fixed | -0.269 |
collapse | -0.255 |
Token | hoping |
Token ID | 7725 |
Feature activation | -5.10e-01 |
Pos logit contributions | |
Acc | +0.287 |
Acc | +0.269 |
solved | +0.233 |
fixed | +0.230 |
collapse | +0.221 |
Neg logit contributions | |
banks | -0.161 |
ippers | -0.156 |
bles | -0.153 |
pouch | -0.151 |
ters | -0.140 |
Token | that |
Token ID | 326 |
Feature activation | -4.88e-01 |
Pos logit contributions | |
Acc | +1.008 |
Acc | +0.928 |
solved | +0.768 |
fixed | +0.757 |
collapse | +0.723 |
Neg logit contributions | |
banks | -0.713 |
ippers | -0.702 |
bles | -0.684 |
pouch | -0.675 |
friends | -0.670 |
Token | someone |
Token ID | 2130 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Acc | +1.026 |
Acc | +0.947 |
solved | +0.800 |
fixed | +0.796 |
bell | +0.769 |
Neg logit contributions | |
banks | -0.609 |
ippers | -0.576 |
bles | -0.565 |
pouch | -0.564 |
friends | -0.560 |
Token | would |
Token ID | 561 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
Acc | +2.179 |
Acc | +2.047 |
collapse | +1.585 |
solved | +1.540 |
bell | +1.538 |
Neg logit contributions | |
banks | -1.573 |
pouch | -1.512 |
bles | -1.497 |
friends | -1.485 |
ippers | -1.469 |
Token | come |
Token ID | 1282 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
banks | +0.209 |
ippers | +0.208 |
bles | +0.202 |
pouch | +0.201 |
ters | +0.190 |
Neg logit contributions | |
Acc | -0.183 |
Acc | -0.165 |
solved | -0.132 |
fixed | -0.130 |
collapse | -0.120 |
Token | out |
Token ID | 503 |
Feature activation | -8.56e-01 |
Pos logit contributions | |
Acc | +1.241 |
Acc | +1.134 |
solved | +0.893 |
fixed | +0.851 |
collapse | +0.827 |
Neg logit contributions | |
ippers | -1.342 |
banks | -1.332 |
bles | -1.321 |
pouch | -1.293 |
friends | -1.243 |
Token | and |
Token ID | 290 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
Acc | +1.748 |
Acc | +1.613 |
solved | +1.344 |
fixed | +1.305 |
collapse | +1.276 |
Neg logit contributions | |
ippers | -1.124 |
banks | -1.117 |
bles | -1.084 |
pouch | -1.059 |
friends | -1.052 |
Token | play |
Token ID | 711 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
Acc | +3.186 |
Acc | +2.856 |
solved | +2.288 |
anne | +2.237 |
fixed | +2.197 |
Neg logit contributions | |
banks | -2.374 |
pouch | -2.192 |
friends | -2.167 |
bank | -2.105 |
ippers | -2.052 |
Token | for |
Token ID | 329 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
ippers | +1.124 |
banks | +1.109 |
bles | +1.082 |
pouch | +1.071 |
itive | +1.008 |
Neg logit contributions | |
Acc | -1.313 |
Acc | -1.212 |
solved | -1.035 |
fixed | -1.029 |
collapse | -0.964 |
Token | today |
Token ID | 1909 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
Acc | +0.297 |
Acc | +0.267 |
solved | +0.221 |
fixed | +0.219 |
collapse | +0.207 |
Neg logit contributions | |
ippers | -0.272 |
banks | -0.269 |
bles | -0.265 |
pouch | -0.258 |
friends | -0.249 |
Token | ." |
Token ID | 526 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
ippers | +0.315 |
banks | +0.311 |
bles | +0.302 |
pouch | +0.299 |
kins | +0.283 |
Neg logit contributions | |
Acc | -0.400 |
Acc | -0.366 |
solved | -0.311 |
fixed | -0.308 |
collapse | -0.292 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
ippers | +0.264 |
banks | +0.259 |
pouch | +0.252 |
bles | +0.251 |
ittens | +0.235 |
Neg logit contributions | |
Acc | -0.338 |
Acc | -0.310 |
solved | -0.268 |
fixed | -0.265 |
collapse | -0.253 |
Token | L |
Token ID | 43 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ippers | +0.491 |
banks | +0.480 |
bles | +0.462 |
pouch | +0.454 |
kins | +0.414 |
Neg logit contributions | |
Acc | -1.106 |
Acc | -1.033 |
solved | -0.915 |
fixed | -0.909 |
collapse | -0.871 |
Token | ily |
Token ID | 813 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
banks | +2.262 |
ippers | +2.225 |
pouch | +2.161 |
bles | +2.131 |
sters | +2.024 |
Neg logit contributions | |
Acc | -2.089 |
Acc | -1.858 |
solved | -1.606 |
fixed | -1.587 |
bell | -1.451 |
Token | and |
Token ID | 290 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
Acc | +2.558 |
Acc | +2.311 |
collapse | +2.005 |
collapsing | +1.908 |
explode | +1.879 |
Neg logit contributions | |
banks | -1.707 |
pouch | -1.629 |
ippers | -1.627 |
bles | -1.552 |
friends | -1.533 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Acc | +1.359 |
Acc | +1.172 |
solved | +0.942 |
collapse | +0.914 |
fixed | +0.898 |
Neg logit contributions | |
ippers | -1.683 |
banks | -1.662 |
pouch | -1.662 |
friends | -1.650 |
bles | -1.608 |
Token | are |
Token ID | 389 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
Acc | +3.399 |
Acc | +3.117 |
collapse | +2.638 |
collapsing | +2.606 |
explode | +2.508 |
Neg logit contributions | |
banks | -1.856 |
ippers | -1.688 |
pouch | -1.671 |
bles | -1.634 |
friends | -1.560 |
Token | twins |
Token ID | 20345 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
ippers | +0.872 |
banks | +0.865 |
bles | +0.836 |
pouch | +0.828 |
kins | +0.776 |
Neg logit contributions | |
Acc | -1.147 |
Acc | -1.055 |
solved | -0.906 |
fixed | -0.900 |
collapse | -0.844 |
Token | . |
Token ID | 13 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
Acc | +0.552 |
Acc | +0.503 |
solved | +0.428 |
collapse | +0.421 |
fixed | +0.418 |
Neg logit contributions | |
ippers | -0.445 |
banks | -0.443 |
pouch | -0.429 |
bles | -0.426 |
friends | -0.407 |
Token | They |
Token ID | 1119 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
Acc | +1.209 |
Acc | +1.107 |
solved | +0.983 |
fixed | +0.976 |
collapse | +0.952 |
Neg logit contributions | |
ippers | -0.677 |
banks | -0.672 |
pouch | -0.639 |
bles | -0.625 |
friends | -0.585 |
Token | are |
Token ID | 389 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Acc | +1.210 |
Acc | +1.130 |
solved | +0.916 |
fixed | +0.907 |
bell | +0.886 |
Neg logit contributions | |
banks | -0.894 |
ippers | -0.843 |
pouch | -0.839 |
bles | -0.829 |
friends | -0.777 |
Token | happy |
Token ID | 3772 |
Feature activation | -8.34e-01 |
Pos logit contributions | |
ippers | +0.232 |
banks | +0.227 |
bles | +0.225 |
pouch | +0.219 |
friends | +0.206 |
Neg logit contributions | |
Acc | -0.348 |
Acc | -0.324 |
solved | -0.277 |
fixed | -0.273 |
collapse | -0.264 |
Token | and |
Token ID | 290 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
Acc | +1.568 |
Acc | +1.445 |
solved | +1.251 |
fixed | +1.229 |
aling | +1.210 |
Neg logit contributions | |
friends | -1.196 |
ippers | -1.196 |
banks | -1.171 |
bles | -1.164 |
pouch | -1.115 |
Token | forget |
Token ID | 6044 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ippers | +0.932 |
banks | +0.920 |
bles | +0.899 |
pouch | +0.888 |
ittens | +0.843 |
Neg logit contributions | |
Acc | -0.932 |
Acc | -0.849 |
solved | -0.712 |
fixed | -0.705 |
collapse | -0.659 |
Token | about |
Token ID | 546 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
Acc | +2.132 |
Acc | +1.941 |
aling | +1.642 |
solved | +1.635 |
collapse | +1.619 |
Neg logit contributions | |
friends | -2.208 |
banks | -2.162 |
bles | -2.101 |
ippers | -2.068 |
things | -2.035 |
Token | the |
Token ID | 262 |
Feature activation | -6.95e-03 |
Pos logit contributions | |
Acc | +2.452 |
Acc | +2.221 |
bell | +1.986 |
solved | +1.970 |
fixed | +1.926 |
Neg logit contributions | |
banks | -2.037 |
friends | -2.031 |
ippers | -1.961 |
bles | -1.929 |
pouch | -1.905 |
Token | hedge |
Token ID | 19859 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Acc | +0.014 |
Acc | +0.013 |
solved | +0.012 |
fixed | +0.011 |
collapse | +0.011 |
Neg logit contributions | |
banks | -0.009 |
ippers | -0.009 |
pouch | -0.009 |
bles | -0.008 |
friends | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -5.52e-01 |
Pos logit contributions | |
Acc | +2.627 |
Acc | +2.436 |
solved | +1.969 |
fixed | +1.928 |
collapse | +1.906 |
Neg logit contributions | |
banks | -1.893 |
pouch | -1.805 |
ters | -1.788 |
bles | -1.784 |
friends | -1.758 |
Token | the |
Token ID | 262 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
Acc | +1.009 |
Acc | +0.922 |
solved | +0.774 |
bell | +0.764 |
fixed | +0.763 |
Neg logit contributions | |
banks | -0.838 |
ippers | -0.812 |
pouch | -0.809 |
bles | -0.796 |
friends | -0.786 |
Token | dog |
Token ID | 3290 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
Acc | +0.317 |
Acc | +0.295 |
solved | +0.258 |
fixed | +0.255 |
bell | +0.245 |
Neg logit contributions | |
banks | -0.171 |
ippers | -0.169 |
pouch | -0.164 |
bles | -0.161 |
friends | -0.151 |