Token | The |
Token ID | 383 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
confir | +0.301 |
contro | +0.288 |
discharge | +0.281 |
corpor | +0.277 |
Jude | +0.277 |
Neg logit contributions | |
acs | -0.302 |
Gravity | -0.294 |
assic | -0.293 |
ikhail | -0.292 |
Tokens | -0.286 |
Token | rising |
Token ID | 7396 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
confir | +0.190 |
contro | +0.180 |
discharge | +0.175 |
corpor | +0.172 |
Jude | +0.172 |
Neg logit contributions | |
acs | -0.236 |
Gravity | -0.231 |
assic | -0.229 |
ikhail | -0.229 |
Tokens | -0.225 |
Token | practice |
Token ID | 3357 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
acs | +0.026 |
Gravity | +0.025 |
assic | +0.025 |
ikhail | +0.025 |
Tokens | +0.024 |
Neg logit contributions | |
confir | -0.018 |
contro | -0.017 |
discharge | -0.016 |
Jude | -0.016 |
corpor | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
confir | +0.049 |
contro | +0.046 |
discharge | +0.045 |
corpor | +0.044 |
Jude | +0.044 |
Neg logit contributions | |
acs | -0.075 |
Gravity | -0.073 |
assic | -0.073 |
ikhail | -0.073 |
Tokens | -0.072 |
Token | qu |
Token ID | 627 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
confir | +0.032 |
contro | +0.030 |
discharge | +0.029 |
Jude | +0.029 |
corpor | +0.029 |
Neg logit contributions | |
acs | -0.041 |
Gravity | -0.041 |
assic | -0.040 |
ikhail | -0.040 |
Tokens | -0.039 |
Token | ack |
Token ID | 441 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
confir | +0.217 |
contro | +0.208 |
discharge | +0.203 |
corpor | +0.201 |
Jude | +0.201 |
Neg logit contributions | |
acs | -0.161 |
Gravity | -0.156 |
assic | -0.156 |
ikhail | -0.155 |
Tokens | -0.151 |
Token | ery |
Token ID | 1924 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
acs | +2.573 |
Gravity | +2.501 |
assic | +2.486 |
ikhail | +2.471 |
Tokens | +2.443 |
Neg logit contributions | |
confir | -2.190 |
contro | -2.064 |
discharge | -2.023 |
Jude | -1.986 |
cumulative | -1.982 |
Token | and |
Token ID | 290 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
acs | +0.394 |
Gravity | +0.385 |
assic | +0.383 |
ikhail | +0.382 |
Tokens | +0.374 |
Neg logit contributions | |
confir | -0.348 |
contro | -0.332 |
discharge | -0.323 |
corpor | -0.319 |
Jude | -0.318 |
Token | sectarian |
Token ID | 28005 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
confir | +0.054 |
contro | +0.051 |
discharge | +0.050 |
corpor | +0.049 |
Jude | +0.049 |
Neg logit contributions | |
acs | -0.064 |
Gravity | -0.063 |
assic | -0.062 |
ikhail | -0.062 |
Tokens | -0.061 |
Token | medicine |
Token ID | 9007 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
confir | +0.610 |
contro | +0.580 |
discharge | +0.564 |
corpor | +0.557 |
Jude | +0.556 |
Neg logit contributions | |
acs | -0.703 |
Gravity | -0.688 |
assic | -0.684 |
ikhail | -0.682 |
Tokens | -0.670 |
Token | fractured |
Token ID | 34051 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
acs | +0.322 |
Gravity | +0.315 |
assic | +0.313 |
ikhail | +0.312 |
Tokens | +0.306 |
Neg logit contributions | |
confir | -0.295 |
contro | -0.280 |
discharge | -0.273 |
corpor | -0.270 |
Jude | -0.269 |
Token | possession |
Token ID | 7797 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
acs | +0.329 |
Gravity | +0.321 |
assic | +0.320 |
ikhail | +0.319 |
Tokens | +0.313 |
Neg logit contributions | |
confir | -0.302 |
contro | -0.288 |
discharge | -0.280 |
corpor | -0.277 |
Jude | -0.276 |
Token | and |
Token ID | 290 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
acs | +1.905 |
Gravity | +1.855 |
assic | +1.842 |
ikhail | +1.830 |
Tokens | +1.789 |
Neg logit contributions | |
confir | -1.684 |
contro | -1.589 |
Jude | -1.527 |
discharge | -1.527 |
corpor | -1.521 |
Token | possession |
Token ID | 7797 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
acs | +0.350 |
Gravity | +0.345 |
assic | +0.340 |
ikhail | +0.339 |
Tokens | +0.331 |
Neg logit contributions | |
confir | -0.303 |
contro | -0.286 |
discharge | -0.275 |
Jude | -0.275 |
corpor | -0.272 |
Token | of |
Token ID | 286 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
acs | +1.833 |
Gravity | +1.785 |
assic | +1.770 |
ikhail | +1.761 |
Tokens | +1.733 |
Neg logit contributions | |
confir | -1.362 |
contro | -1.280 |
discharge | -1.233 |
Jude | -1.231 |
corpor | -1.222 |
Token | drug |
Token ID | 2563 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
acs | +0.464 |
Gravity | +0.455 |
assic | +0.448 |
ikhail | +0.446 |
Tokens | +0.441 |
Neg logit contributions | |
confir | -0.380 |
contro | -0.357 |
discharge | -0.347 |
Jude | -0.344 |
corpor | -0.340 |
Token | parap |
Token ID | 41406 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
acs | +0.402 |
Gravity | +0.393 |
assic | +0.391 |
ikhail | +0.390 |
Tokens | +0.383 |
Neg logit contributions | |
confir | -0.329 |
contro | -0.313 |
discharge | -0.304 |
corpor | -0.300 |
Jude | -0.299 |
Token | hern |
Token ID | 2881 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
acs | +1.952 |
Gravity | +1.849 |
ikhail | +1.831 |
assic | +1.818 |
ochem | +1.792 |
Neg logit contributions | |
confir | -2.938 |
contro | -2.748 |
corpor | -2.663 |
Jude | -2.662 |
discharge | -2.656 |
Token | alia |
Token ID | 9752 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
acs | +0.738 |
Gravity | +0.726 |
assic | +0.723 |
ikhail | +0.722 |
Tokens | +0.711 |
Neg logit contributions | |
confir | -0.385 |
contro | -0.360 |
discharge | -0.348 |
corpor | -0.341 |
Jude | -0.340 |
Token | . |
Token ID | 13 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
acs | +1.276 |
Gravity | +1.247 |
ikhail | +1.233 |
assic | +1.228 |
Tokens | +1.208 |
Neg logit contributions | |
confir | -1.047 |
contro | -0.979 |
discharge | -0.949 |
Jude | -0.948 |
corpor | -0.940 |
Token | It |
Token ID | 632 |
Feature activation | +6.91e-02 |
Pos logit contributions | |
acs | +0.428 |
Gravity | +0.423 |
assic | +0.414 |
ikhail | +0.414 |
Tokens | +0.408 |
Neg logit contributions | |
confir | -0.333 |
contro | -0.312 |
discharge | -0.302 |
corpor | -0.297 |
cumulative | -0.296 |
Token | � |
Token ID | 447 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
acs | +0.120 |
Gravity | +0.118 |
assic | +0.117 |
ikhail | +0.116 |
Tokens | +0.114 |
Neg logit contributions | |
confir | -0.124 |
contro | -0.119 |
discharge | -0.116 |
corpor | -0.115 |
Jude | -0.114 |
Token | had |
Token ID | 550 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
acs | +0.069 |
Gravity | +0.068 |
assic | +0.067 |
ikhail | +0.067 |
Tokens | +0.066 |
Neg logit contributions | |
confir | -0.062 |
contro | -0.059 |
discharge | -0.058 |
corpor | -0.057 |
Jude | -0.057 |
Token | said |
Token ID | 531 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
acs | +0.459 |
Gravity | +0.450 |
assic | +0.447 |
ikhail | +0.445 |
Tokens | +0.437 |
Neg logit contributions | |
confir | -0.391 |
contro | -0.372 |
discharge | -0.362 |
corpor | -0.357 |
Jude | -0.356 |
Token | , |
Token ID | 11 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
confir | +0.086 |
contro | +0.082 |
discharge | +0.079 |
corpor | +0.078 |
Jude | +0.078 |
Neg logit contributions | |
acs | -0.112 |
Gravity | -0.110 |
assic | -0.109 |
ikhail | -0.109 |
Tokens | -0.107 |
Token | " |
Token ID | 366 |
Feature activation | -8.17e-02 |
Pos logit contributions | |
confir | +0.040 |
contro | +0.038 |
discharge | +0.037 |
corpor | +0.036 |
Jude | +0.036 |
Neg logit contributions | |
acs | -0.060 |
Gravity | -0.059 |
assic | -0.059 |
ikhail | -0.059 |
Tokens | -0.058 |
Token | How |
Token ID | 2437 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
confir | +0.150 |
contro | +0.144 |
discharge | +0.141 |
corpor | +0.139 |
Jude | +0.139 |
Neg logit contributions | |
acs | -0.139 |
Gravity | -0.135 |
assic | -0.135 |
ikhail | -0.134 |
Tokens | -0.131 |
Token | cum |
Token ID | 36340 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
confir | +0.118 |
contro | +0.112 |
discharge | +0.109 |
corpor | +0.108 |
Jude | +0.107 |
Neg logit contributions | |
acs | -0.130 |
Gravity | -0.127 |
assic | -0.127 |
ikhail | -0.126 |
Tokens | -0.124 |
Token | nobody |
Token ID | 8168 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
acs | +2.467 |
Gravity | +2.406 |
assic | +2.379 |
ikhail | +2.373 |
ochem | +2.308 |
Neg logit contributions | |
confir | -2.014 |
contro | -1.904 |
discharge | -1.846 |
Jude | -1.819 |
corpor | -1.819 |
Token | just |
Token ID | 655 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
confir | +0.131 |
contro | +0.123 |
discharge | +0.121 |
corpor | +0.119 |
Jude | +0.119 |
Neg logit contributions | |
acs | -0.145 |
Gravity | -0.143 |
assic | -0.141 |
ikhail | -0.140 |
Tokens | -0.137 |
Token | does |
Token ID | 857 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
acs | +0.189 |
Gravity | +0.186 |
assic | +0.184 |
ikhail | +0.182 |
Tokens | +0.179 |
Neg logit contributions | |
confir | -0.166 |
contro | -0.157 |
discharge | -0.154 |
corpor | -0.152 |
Jude | -0.151 |
Token | simple |
Token ID | 2829 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
confir | +0.168 |
contro | +0.159 |
discharge | +0.155 |
corpor | +0.152 |
Jude | +0.152 |
Neg logit contributions | |
acs | -0.218 |
Gravity | -0.213 |
assic | -0.212 |
ikhail | -0.212 |
Tokens | -0.208 |
Token | stories |
Token ID | 3923 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
acs | +0.362 |
Gravity | +0.355 |
assic | +0.352 |
ikhail | +0.351 |
Tokens | +0.344 |
Neg logit contributions | |
confir | -0.358 |
contro | -0.341 |
discharge | -0.333 |
corpor | -0.328 |
Jude | -0.328 |
Token | a |
Token ID | 257 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
confir | +0.126 |
contro | +0.120 |
discharge | +0.117 |
corpor | +0.115 |
Jude | +0.115 |
Neg logit contributions | |
acs | -0.140 |
Gravity | -0.137 |
assic | -0.136 |
ikhail | -0.136 |
Tokens | -0.133 |
Token | little |
Token ID | 1310 |
Feature activation | +6.18e-02 |
Pos logit contributions | |
confir | +0.023 |
contro | +0.022 |
discharge | +0.021 |
Jude | +0.021 |
corpor | +0.021 |
Neg logit contributions | |
acs | -0.029 |
Gravity | -0.028 |
ikhail | -0.028 |
assic | -0.028 |
Tokens | -0.027 |
Token | - |
Token ID | 12 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
acs | +0.114 |
Gravity | +0.111 |
assic | +0.111 |
ikhail | +0.110 |
Tokens | +0.108 |
Neg logit contributions | |
confir | -0.105 |
contro | -0.100 |
discharge | -0.097 |
corpor | -0.096 |
Jude | -0.096 |
Token | known |
Token ID | 4002 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
confir | +0.099 |
contro | +0.093 |
discharge | +0.091 |
Jude | +0.089 |
corpor | +0.089 |
Neg logit contributions | |
acs | -0.135 |
Gravity | -0.130 |
assic | -0.130 |
ikhail | -0.130 |
Tokens | -0.128 |
Token | hedge |
Token ID | 19859 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
confir | +0.816 |
contro | +0.780 |
discharge | +0.759 |
corpor | +0.752 |
Jude | +0.747 |
Neg logit contributions | |
acs | -0.936 |
Gravity | -0.914 |
assic | -0.912 |
ikhail | -0.908 |
Tokens | -0.891 |
Token | fund |
Token ID | 1814 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
acs | +0.678 |
Gravity | +0.661 |
assic | +0.657 |
ikhail | +0.655 |
Tokens | +0.640 |
Neg logit contributions | |
confir | -0.799 |
contro | -0.765 |
discharge | -0.749 |
corpor | -0.741 |
Jude | -0.738 |
Token | manager |
Token ID | 4706 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
acs | +2.369 |
Gravity | +2.327 |
ikhail | +2.322 |
assic | +2.297 |
ochem | +2.253 |
Neg logit contributions | |
confir | -1.741 |
contro | -1.641 |
discharge | -1.577 |
Jude | -1.573 |
cumulative | -1.551 |
Token | who |
Token ID | 508 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
acs | +0.915 |
Gravity | +0.897 |
ikhail | +0.890 |
assic | +0.887 |
Tokens | +0.870 |
Neg logit contributions | |
confir | -0.701 |
contro | -0.660 |
discharge | -0.643 |
Jude | -0.632 |
corpor | -0.629 |
Token | crossed |
Token ID | 12606 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
confir | +0.379 |
contro | +0.364 |
discharge | +0.355 |
corpor | +0.352 |
Jude | +0.350 |
Neg logit contributions | |
acs | -0.370 |
Gravity | -0.361 |
assic | -0.360 |
ikhail | -0.360 |
Tokens | -0.352 |
Token | Goldman |
Token ID | 18740 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
acs | +1.062 |
Gravity | +1.039 |
assic | +1.034 |
ikhail | +1.031 |
Tokens | +1.012 |
Neg logit contributions | |
confir | -0.867 |
contro | -0.825 |
discharge | -0.803 |
corpor | -0.792 |
Jude | -0.789 |
Token | � |
Token ID | 447 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
acs | +0.555 |
Gravity | +0.543 |
assic | +0.540 |
ikhail | +0.538 |
Tokens | +0.528 |
Neg logit contributions | |
confir | -0.513 |
contro | -0.489 |
discharge | -0.476 |
corpor | -0.470 |
Jude | -0.469 |
Token | . |
Token ID | 13 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
confir | +0.028 |
contro | +0.026 |
discharge | +0.026 |
corpor | +0.025 |
Jude | +0.025 |
Neg logit contributions | |
acs | -0.035 |
Gravity | -0.034 |
assic | -0.034 |
ikhail | -0.034 |
Tokens | -0.033 |
Token | When |
Token ID | 1649 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
acs | +0.022 |
Gravity | +0.021 |
assic | +0.021 |
ikhail | +0.021 |
Tokens | +0.021 |
Neg logit contributions | |
confir | -0.022 |
contro | -0.021 |
discharge | -0.021 |
corpor | -0.020 |
cumulative | -0.020 |
Token | Trump |
Token ID | 1301 |
Feature activation | +4.67e-03 |
Pos logit contributions | |
acs | +0.301 |
Gravity | +0.295 |
assic | +0.292 |
ikhail | +0.292 |
Tokens | +0.287 |
Neg logit contributions | |
confir | -0.253 |
contro | -0.241 |
discharge | -0.234 |
corpor | -0.231 |
Jude | -0.230 |
Token | named |
Token ID | 3706 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
acs | +0.009 |
Gravity | +0.009 |
ikhail | +0.009 |
assic | +0.009 |
Tokens | +0.008 |
Neg logit contributions | |
confir | -0.008 |
contro | -0.007 |
discharge | -0.007 |
corpor | -0.007 |
Jude | -0.007 |
Token | his |
Token ID | 465 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
acs | +0.078 |
Gravity | +0.076 |
assic | +0.076 |
ikhail | +0.076 |
Tokens | +0.074 |
Neg logit contributions | |
confir | -0.082 |
contro | -0.079 |
discharge | -0.077 |
corpor | -0.076 |
Jude | -0.076 |
Token | vice |
Token ID | 7927 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
acs | +0.355 |
Gravity | +0.349 |
assic | +0.345 |
ikhail | +0.345 |
Tokens | +0.338 |
Neg logit contributions | |
confir | -0.287 |
contro | -0.272 |
discharge | -0.264 |
corpor | -0.260 |
Jude | -0.260 |
Token | presidential |
Token ID | 4787 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
acs | +1.795 |
Gravity | +1.739 |
assic | +1.727 |
ikhail | +1.723 |
Tokens | +1.682 |
Neg logit contributions | |
confir | -2.310 |
contro | -2.220 |
discharge | -2.166 |
Jude | -2.136 |
corpor | -2.135 |
Token | pick |
Token ID | 2298 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
acs | +0.586 |
Gravity | +0.575 |
ikhail | +0.572 |
assic | +0.572 |
Tokens | +0.560 |
Neg logit contributions | |
confir | -0.509 |
contro | -0.481 |
discharge | -0.465 |
Jude | -0.461 |
corpor | -0.461 |
Token | , |
Token ID | 11 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
acs | +0.037 |
Gravity | +0.037 |
assic | +0.036 |
ikhail | +0.036 |
Tokens | +0.036 |
Neg logit contributions | |
confir | -0.031 |
contro | -0.030 |
discharge | -0.029 |
corpor | -0.029 |
Jude | -0.028 |
Token | Gov |
Token ID | 10964 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
confir | +0.109 |
contro | +0.104 |
discharge | +0.102 |
corpor | +0.100 |
Jude | +0.100 |
Neg logit contributions | |
acs | -0.096 |
Gravity | -0.094 |
assic | -0.093 |
ikhail | -0.093 |
Tokens | -0.091 |
Token | . |
Token ID | 13 |
Feature activation | -5.20e-02 |
Pos logit contributions | |
confir | +0.145 |
contro | +0.137 |
discharge | +0.133 |
corpor | +0.131 |
Jude | +0.131 |
Neg logit contributions | |
acs | -0.211 |
Gravity | -0.207 |
assic | -0.206 |
ikhail | -0.205 |
Tokens | -0.202 |
Token | with |
Token ID | 351 |
Feature activation | -7.12e-03 |
Pos logit contributions | |
confir | +0.139 |
contro | +0.133 |
discharge | +0.129 |
corpor | +0.128 |
Jude | +0.127 |
Neg logit contributions | |
acs | -0.155 |
Gravity | -0.152 |
assic | -0.151 |
ikhail | -0.151 |
Tokens | -0.148 |
Token | sect |
Token ID | 15951 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
confir | +0.012 |
contro | +0.011 |
discharge | +0.011 |
Jude | +0.011 |
corpor | +0.011 |
Neg logit contributions | |
acs | -0.013 |
Gravity | -0.013 |
assic | -0.013 |
ikhail | -0.013 |
Tokens | -0.013 |
Token | arians |
Token ID | 13517 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
confir | +0.867 |
contro | +0.838 |
discharge | +0.817 |
corpor | +0.812 |
Jude | +0.808 |
Neg logit contributions | |
acs | -0.753 |
Gravity | -0.741 |
assic | -0.735 |
ikhail | -0.734 |
Tokens | -0.717 |
Token | and |
Token ID | 290 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
confir | +0.194 |
contro | +0.186 |
discharge | +0.181 |
corpor | +0.179 |
Jude | +0.177 |
Neg logit contributions | |
acs | -0.216 |
Gravity | -0.212 |
assic | -0.211 |
ikhail | -0.211 |
Tokens | -0.207 |
Token | qu |
Token ID | 627 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
confir | +0.064 |
contro | +0.061 |
discharge | +0.059 |
corpor | +0.059 |
Jude | +0.059 |
Neg logit contributions | |
acs | -0.072 |
Gravity | -0.071 |
assic | -0.070 |
ikhail | -0.070 |
Tokens | -0.069 |
Token | acks |
Token ID | 4595 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
confir | +0.247 |
contro | +0.238 |
discharge | +0.234 |
corpor | +0.231 |
Jude | +0.231 |
Neg logit contributions | |
acs | -0.169 |
Gravity | -0.164 |
assic | -0.163 |
ikhail | -0.163 |
Tokens | -0.158 |
Token | could |
Token ID | 714 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
acs | +2.086 |
Gravity | +2.026 |
assic | +2.018 |
ikhail | +2.009 |
Tokens | +1.964 |
Neg logit contributions | |
confir | -1.888 |
contro | -1.783 |
discharge | -1.745 |
Jude | -1.723 |
cumulative | -1.717 |
Token | not |
Token ID | 407 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
confir | +0.173 |
contro | +0.165 |
discharge | +0.161 |
corpor | +0.159 |
Jude | +0.159 |
Neg logit contributions | |
acs | -0.192 |
Gravity | -0.188 |
assic | -0.187 |
ikhail | -0.186 |
Tokens | -0.183 |
Token | support |
Token ID | 1104 |
Feature activation | +8.62e-03 |
Pos logit contributions | |
acs | +0.082 |
Gravity | +0.080 |
assic | +0.079 |
ikhail | +0.079 |
Tokens | +0.077 |
Neg logit contributions | |
confir | -0.080 |
contro | -0.076 |
discharge | -0.074 |
corpor | -0.073 |
Jude | -0.073 |
Token | an |
Token ID | 281 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
acs | +0.016 |
Gravity | +0.016 |
assic | +0.015 |
ikhail | +0.015 |
Tokens | +0.015 |
Neg logit contributions | |
confir | -0.015 |
contro | -0.014 |
discharge | -0.014 |
corpor | -0.013 |
Jude | -0.013 |
Token | influx |
Token ID | 25065 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
acs | +0.592 |
Gravity | +0.578 |
assic | +0.576 |
ikhail | +0.573 |
Tokens | +0.562 |
Neg logit contributions | |
confir | -0.527 |
contro | -0.499 |
discharge | -0.487 |
corpor | -0.480 |
Jude | -0.480 |
Token | the |
Token ID | 262 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
acs | +0.065 |
Gravity | +0.063 |
assic | +0.063 |
ikhail | +0.062 |
Tokens | +0.062 |
Neg logit contributions | |
confir | -0.050 |
contro | -0.048 |
discharge | -0.046 |
corpor | -0.046 |
Jude | -0.046 |
Token | office |
Token ID | 2607 |
Feature activation | +1.03e-02 |
Pos logit contributions | |
confir | +0.113 |
contro | +0.105 |
discharge | +0.100 |
Jude | +0.099 |
corpor | +0.099 |
Neg logit contributions | |
acs | -0.204 |
Gravity | -0.201 |
assic | -0.198 |
ikhail | -0.197 |
Tokens | -0.194 |
Token | said |
Token ID | 531 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
acs | +0.020 |
Gravity | +0.020 |
assic | +0.020 |
ikhail | +0.020 |
Tokens | +0.019 |
Neg logit contributions | |
confir | -0.016 |
contro | -0.015 |
discharge | -0.015 |
Jude | -0.015 |
corpor | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
confir | +0.051 |
contro | +0.048 |
discharge | +0.046 |
Jude | +0.046 |
corpor | +0.046 |
Neg logit contributions | |
acs | -0.074 |
Gravity | -0.072 |
assic | -0.071 |
ikhail | -0.071 |
Tokens | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
confir | +0.100 |
contro | +0.095 |
discharge | +0.092 |
corpor | +0.091 |
Jude | +0.091 |
Neg logit contributions | |
acs | -0.118 |
Gravity | -0.116 |
assic | -0.115 |
ikhail | -0.114 |
Tokens | -0.112 |
Token | draft |
Token ID | 4538 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
confir | +0.185 |
contro | +0.175 |
corpor | +0.167 |
discharge | +0.166 |
Jude | +0.166 |
Neg logit contributions | |
acs | -0.231 |
Gravity | -0.229 |
ikhail | -0.224 |
assic | -0.223 |
Tokens | -0.219 |
Token | Strategic |
Token ID | 24999 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
Gravity | +2.323 |
acs | +2.311 |
assic | +2.253 |
ikhail | +2.229 |
Tokens | +2.217 |
Neg logit contributions | |
confir | -1.573 |
contro | -1.491 |
Jude | -1.408 |
corpor | -1.406 |
discharge | -1.365 |
Token | Plan |
Token ID | 5224 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
acs | +0.362 |
Gravity | +0.360 |
assic | +0.349 |
ikhail | +0.347 |
Tokens | +0.345 |
Neg logit contributions | |
confir | -0.385 |
contro | -0.366 |
corpor | -0.350 |
discharge | -0.349 |
cumulative | -0.346 |
Token | released |
Token ID | 2716 |
Feature activation | +7.10e-02 |
Pos logit contributions | |
confir | +0.547 |
contro | +0.520 |
discharge | +0.501 |
corpor | +0.496 |
Jude | +0.496 |
Neg logit contributions | |
acs | -0.728 |
Gravity | -0.714 |
assic | -0.708 |
ikhail | -0.706 |
Tokens | -0.695 |
Token | last |
Token ID | 938 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
acs | +0.149 |
Gravity | +0.146 |
assic | +0.144 |
ikhail | +0.144 |
Tokens | +0.142 |
Neg logit contributions | |
confir | -0.103 |
contro | -0.097 |
discharge | -0.094 |
corpor | -0.092 |
Jude | -0.092 |
Token | week |
Token ID | 1285 |
Feature activation | -5.95e-02 |
Pos logit contributions | |
confir | +0.548 |
contro | +0.518 |
discharge | +0.503 |
corpor | +0.495 |
Jude | +0.493 |
Neg logit contributions | |
acs | -0.818 |
Gravity | -0.801 |
assic | -0.799 |
ikhail | -0.798 |
Tokens | -0.783 |
Token | been |
Token ID | 587 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
confir | +0.062 |
contro | +0.057 |
discharge | +0.056 |
Jude | +0.056 |
corpor | +0.055 |
Neg logit contributions | |
acs | -0.072 |
Gravity | -0.071 |
assic | -0.070 |
ikhail | -0.070 |
Tokens | -0.068 |
Token | trained |
Token ID | 8776 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
confir | +0.156 |
contro | +0.145 |
discharge | +0.141 |
Jude | +0.139 |
cumulative | +0.138 |
Neg logit contributions | |
acs | -0.246 |
Gravity | -0.242 |
assic | -0.239 |
ikhail | -0.238 |
Tokens | -0.234 |
Token | to |
Token ID | 284 |
Feature activation | -9.42e-02 |
Pos logit contributions | |
confir | +0.071 |
contro | +0.067 |
discharge | +0.066 |
Jude | +0.065 |
corpor | +0.064 |
Neg logit contributions | |
acs | -0.092 |
Gravity | -0.090 |
assic | -0.089 |
ikhail | -0.089 |
Tokens | -0.088 |
Token | navigate |
Token ID | 16500 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
confir | +0.144 |
contro | +0.134 |
discharge | +0.130 |
Jude | +0.130 |
corpor | +0.129 |
Neg logit contributions | |
acs | -0.190 |
Gravity | -0.187 |
assic | -0.184 |
ikhail | -0.183 |
Tokens | -0.180 |
Token | a |
Token ID | 257 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
acs | +0.335 |
Gravity | +0.329 |
ikhail | +0.322 |
assic | +0.322 |
Tokens | +0.318 |
Neg logit contributions | |
confir | -0.284 |
contro | -0.264 |
discharge | -0.258 |
Jude | -0.255 |
corpor | -0.253 |
Token | wheelchair |
Token ID | 27181 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
confir | +0.157 |
contro | +0.146 |
discharge | +0.142 |
Jude | +0.141 |
corpor | +0.139 |
Neg logit contributions | |
acs | -0.212 |
Gravity | -0.209 |
ikhail | -0.206 |
assic | -0.205 |
Mov | -0.201 |
Token | through |
Token ID | 832 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
acs | +2.093 |
Gravity | +2.051 |
ikhail | +1.999 |
assic | +1.978 |
Tokens | +1.966 |
Neg logit contributions | |
confir | -1.853 |
contro | -1.689 |
Jude | -1.668 |
cumulative | -1.652 |
discharge | -1.648 |
Token | a |
Token ID | 257 |
Feature activation | -8.30e-02 |
Pos logit contributions | |
confir | +0.286 |
contro | +0.269 |
discharge | +0.262 |
corpor | +0.258 |
Jude | +0.258 |
Neg logit contributions | |
acs | -0.344 |
Gravity | -0.339 |
assic | -0.334 |
ikhail | -0.334 |
Tokens | -0.328 |
Token | maze |
Token ID | 31237 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
confir | +0.125 |
contro | +0.116 |
discharge | +0.113 |
Jude | +0.112 |
corpor | +0.112 |
Neg logit contributions | |
acs | -0.168 |
Gravity | -0.166 |
assic | -0.163 |
ikhail | -0.163 |
Mov | -0.160 |
Token | and |
Token ID | 290 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
acs | +0.330 |
Gravity | +0.324 |
ikhail | +0.319 |
assic | +0.318 |
Tokens | +0.312 |
Neg logit contributions | |
confir | -0.270 |
contro | -0.251 |
discharge | -0.245 |
Jude | -0.244 |
corpor | -0.241 |
Token | have |
Token ID | 423 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
confir | +0.055 |
contro | +0.051 |
discharge | +0.050 |
Jude | +0.050 |
corpor | +0.049 |
Neg logit contributions | |
acs | -0.070 |
Gravity | -0.069 |
assic | -0.068 |
ikhail | -0.067 |
Tokens | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
acs | +0.336 |
Gravity | +0.331 |
ikhail | +0.327 |
assic | +0.327 |
Tokens | +0.322 |
Neg logit contributions | |
confir | -0.221 |
contro | -0.207 |
discharge | -0.201 |
corpor | -0.197 |
cumulative | -0.195 |
Token |
|
Token ID | 198 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
acs | +0.327 |
Gravity | +0.321 |
ikhail | +0.316 |
assic | +0.315 |
Tokens | +0.315 |
Neg logit contributions | |
confir | -0.201 |
contro | -0.179 |
discharge | -0.173 |
corpor | -0.171 |
cumulative | -0.169 |
Token | Read |
Token ID | 5569 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
acs | +0.163 |
Gravity | +0.160 |
assic | +0.159 |
ikhail | +0.158 |
Tokens | +0.154 |
Neg logit contributions | |
confir | -0.171 |
contro | -0.164 |
discharge | -0.160 |
corpor | -0.158 |
Jude | -0.158 |
Token | more |
Token ID | 517 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
acs | +0.438 |
Gravity | +0.428 |
assic | +0.425 |
ikhail | +0.425 |
Tokens | +0.416 |
Neg logit contributions | |
confir | -0.420 |
contro | -0.400 |
discharge | -0.390 |
corpor | -0.385 |
Jude | -0.384 |
Token | about |
Token ID | 546 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
acs | +0.111 |
Gravity | +0.109 |
assic | +0.108 |
ikhail | +0.108 |
Tokens | +0.106 |
Neg logit contributions | |
confir | -0.079 |
contro | -0.075 |
discharge | -0.073 |
corpor | -0.072 |
Jude | -0.071 |
Token | Sud |
Token ID | 14818 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
acs | +0.194 |
Gravity | +0.191 |
assic | +0.189 |
ikhail | +0.189 |
Tokens | +0.184 |
Neg logit contributions | |
confir | -0.193 |
contro | -0.184 |
discharge | -0.180 |
corpor | -0.177 |
Jude | -0.177 |
Token | h |
Token ID | 71 |
Feature activation | +8.89e-02 |
Pos logit contributions | |
acs | +1.839 |
Gravity | +1.789 |
ikhail | +1.782 |
assic | +1.781 |
Tokens | +1.726 |
Neg logit contributions | |
confir | -2.085 |
contro | -1.996 |
discharge | -1.944 |
corpor | -1.925 |
Jude | -1.913 |
Token | ir |
Token ID | 343 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
acs | +0.161 |
Gravity | +0.157 |
assic | +0.156 |
ikhail | +0.156 |
Tokens | +0.153 |
Neg logit contributions | |
confir | -0.153 |
contro | -0.147 |
discharge | -0.143 |
corpor | -0.141 |
Jude | -0.141 |
Token | Ch |
Token ID | 609 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
confir | +0.571 |
contro | +0.544 |
discharge | +0.530 |
corpor | +0.524 |
Jude | +0.522 |
Neg logit contributions | |
acs | -0.643 |
Gravity | -0.629 |
assic | -0.627 |
ikhail | -0.625 |
Tokens | -0.614 |
Token | oud |
Token ID | 2778 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
acs | +0.514 |
Gravity | +0.499 |
assic | +0.499 |
ikhail | +0.498 |
Tokens | +0.486 |
Neg logit contributions | |
confir | -0.518 |
contro | -0.492 |
discharge | -0.480 |
corpor | -0.475 |
Jude | -0.473 |
Token | h |
Token ID | 71 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
confir | +0.091 |
contro | +0.086 |
discharge | +0.084 |
corpor | +0.084 |
Jude | +0.083 |
Neg logit contributions | |
acs | -0.082 |
Gravity | -0.079 |
assic | -0.079 |
ikhail | -0.079 |
Tokens | -0.077 |
Token | speeches |
Token ID | 18138 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
acs | +0.257 |
Gravity | +0.252 |
assic | +0.251 |
ikhail | +0.250 |
Tokens | +0.246 |
Neg logit contributions | |
confir | -0.196 |
contro | -0.186 |
discharge | -0.180 |
corpor | -0.178 |
Jude | -0.177 |
Token | . |
Token ID | 13 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
acs | +0.204 |
Gravity | +0.200 |
assic | +0.198 |
ikhail | +0.198 |
Tokens | +0.194 |
Neg logit contributions | |
confir | -0.153 |
contro | -0.145 |
discharge | -0.140 |
corpor | -0.138 |
Jude | -0.138 |
Token |
|
Token ID | 198 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
confir | +0.087 |
contro | +0.082 |
discharge | +0.079 |
corpor | +0.078 |
Jude | +0.077 |
Neg logit contributions | |
acs | -0.153 |
Gravity | -0.151 |
assic | -0.150 |
ikhail | -0.150 |
Tokens | -0.147 |
Token |
|
Token ID | 198 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
acs | +0.036 |
Gravity | +0.036 |
ikhail | +0.035 |
assic | +0.035 |
Tokens | +0.035 |
Neg logit contributions | |
confir | -0.022 |
contro | -0.019 |
discharge | -0.019 |
corpor | -0.019 |
cumulative | -0.018 |
Token | Newsletter |
Token ID | 33031 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
confir | +0.005 |
contro | +0.005 |
discharge | +0.005 |
corpor | +0.005 |
Jude | +0.005 |
Neg logit contributions | |
acs | -0.006 |
Gravity | -0.006 |
assic | -0.006 |
ikhail | -0.006 |
Tokens | -0.006 |
Token | Sign |
Token ID | 5865 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
acs | +0.411 |
Gravity | +0.404 |
ikhail | +0.398 |
assic | +0.395 |
Tokens | +0.389 |
Neg logit contributions | |
confir | -0.368 |
contro | -0.344 |
discharge | -0.337 |
corpor | -0.331 |
cumulative | -0.330 |
Token | Up |
Token ID | 3205 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
acs | +0.899 |
Gravity | +0.859 |
ikhail | +0.812 |
assic | +0.808 |
Tokens | +0.794 |
Neg logit contributions | |
confir | -3.063 |
contro | -2.927 |
discharge | -2.885 |
corpor | -2.859 |
cumulative | -2.840 |
Token | Continue |
Token ID | 10054 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
acs | +0.120 |
Gravity | +0.118 |
ikhail | +0.115 |
assic | +0.114 |
Tokens | +0.112 |
Neg logit contributions | |
confir | -0.236 |
contro | -0.226 |
discharge | -0.221 |
corpor | -0.221 |
cumulative | -0.219 |
Token | reading |
Token ID | 3555 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
acs | +0.685 |
Gravity | +0.675 |
ikhail | +0.653 |
assic | +0.653 |
Tokens | +0.635 |
Neg logit contributions | |
confir | -1.020 |
contro | -0.955 |
discharge | -0.937 |
corpor | -0.934 |
Jude | -0.926 |
Token | the |
Token ID | 262 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
acs | +0.079 |
Gravity | +0.078 |
ikhail | +0.077 |
assic | +0.077 |
Tokens | +0.076 |
Neg logit contributions | |
confir | -0.063 |
contro | -0.058 |
discharge | -0.056 |
corpor | -0.055 |
Jude | -0.055 |
Token | main |
Token ID | 1388 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
confir | +0.019 |
contro | +0.018 |
discharge | +0.017 |
corpor | +0.017 |
Jude | +0.017 |
Neg logit contributions | |
acs | -0.025 |
Gravity | -0.024 |
assic | -0.024 |
ikhail | -0.024 |
Tokens | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +9.24e-02 |
Pos logit contributions | |
acs | +0.271 |
Gravity | +0.265 |
assic | +0.263 |
ikhail | +0.263 |
Tokens | +0.258 |
Neg logit contributions | |
confir | -0.226 |
contro | -0.214 |
discharge | -0.209 |
corpor | -0.206 |
Jude | -0.205 |
Token | broadcast |
Token ID | 7025 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
acs | +0.176 |
Gravity | +0.173 |
ikhail | +0.169 |
assic | +0.167 |
Bro | +0.166 |
Neg logit contributions | |
confir | -0.153 |
contro | -0.137 |
Jude | -0.137 |
cumulative | -0.133 |
discharge | -0.132 |
Token | , |
Token ID | 11 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
acs | +0.403 |
Gravity | +0.394 |
assic | +0.392 |
ikhail | +0.392 |
Tokens | +0.384 |
Neg logit contributions | |
confir | -0.325 |
contro | -0.310 |
discharge | -0.302 |
corpor | -0.298 |
Jude | -0.297 |
Token | rewritten |
Token ID | 30101 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
acs | +0.235 |
Gravity | +0.233 |
ikhail | +0.230 |
assic | +0.229 |
Tokens | +0.225 |
Neg logit contributions | |
confir | -0.154 |
contro | -0.141 |
discharge | -0.136 |
Jude | -0.135 |
cumulative | -0.133 |
Token | or |
Token ID | 393 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
confir | +0.502 |
contro | +0.486 |
discharge | +0.474 |
corpor | +0.468 |
cumulative | +0.461 |
Neg logit contributions | |
acs | -0.675 |
Gravity | -0.664 |
ikhail | -0.662 |
assic | -0.662 |
Tokens | -0.650 |
Token | redistributed |
Token ID | 38913 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
acs | +0.131 |
Gravity | +0.128 |
assic | +0.126 |
ikhail | +0.125 |
Tokens | +0.123 |
Neg logit contributions | |
confir | -0.196 |
contro | -0.187 |
discharge | -0.183 |
Jude | -0.181 |
corpor | -0.180 |
Token | .) |
Token ID | 2014 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
acs | +1.929 |
Gravity | +1.876 |
ikhail | +1.858 |
assic | +1.849 |
ochem | +1.805 |
Neg logit contributions | |
confir | -1.934 |
contro | -1.825 |
discharge | -1.777 |
corpor | -1.756 |
Jude | -1.755 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.25e-04 |
Pos logit contributions | |
acs | +0.554 |
Gravity | +0.543 |
assic | +0.541 |
ikhail | +0.540 |
Tokens | +0.531 |
Neg logit contributions | |
confir | -0.348 |
contro | -0.328 |
discharge | -0.317 |
corpor | -0.312 |
Jude | -0.311 |
Token | New |
Token ID | 3791 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
acs | +0.000 |
Gravity | +0.000 |
assic | +0.000 |
ikhail | +0.000 |
Tokens | +0.000 |
Neg logit contributions | |
confir | -0.000 |
contro | -0.000 |
discharge | -0.000 |
corpor | -0.000 |
Jude | -0.000 |
Token | castle |
Token ID | 18676 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
acs | +0.402 |
Gravity | +0.391 |
assic | +0.389 |
ikhail | +0.388 |
Tokens | +0.379 |
Neg logit contributions | |
confir | -0.484 |
contro | -0.464 |
discharge | -0.454 |
corpor | -0.449 |
Jude | -0.448 |
Token | United |
Token ID | 1578 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
confir | +0.357 |
contro | +0.341 |
discharge | +0.333 |
corpor | +0.329 |
Jude | +0.328 |
Neg logit contributions | |
acs | -0.342 |
Gravity | -0.334 |
assic | -0.332 |
ikhail | -0.331 |
Tokens | -0.324 |
Token | be |
Token ID | 307 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
confir | +0.271 |
contro | +0.257 |
discharge | +0.249 |
corpor | +0.247 |
Jude | +0.247 |
Neg logit contributions | |
acs | -0.329 |
Gravity | -0.322 |
assic | -0.319 |
ikhail | -0.318 |
Tokens | -0.313 |
Token | chosen |
Token ID | 7147 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
acs | +0.187 |
Gravity | +0.183 |
assic | +0.182 |
ikhail | +0.181 |
Tokens | +0.178 |
Neg logit contributions | |
confir | -0.148 |
contro | -0.140 |
discharge | -0.137 |
corpor | -0.135 |
Jude | -0.134 |
Token | as |
Token ID | 355 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
acs | +0.045 |
Gravity | +0.044 |
assic | +0.044 |
ikhail | +0.044 |
Tokens | +0.043 |
Neg logit contributions | |
confir | -0.035 |
contro | -0.033 |
discharge | -0.032 |
corpor | -0.032 |
Jude | -0.031 |
Token | Romney |
Token ID | 8431 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
confir | +0.314 |
contro | +0.299 |
discharge | +0.291 |
corpor | +0.287 |
Jude | +0.286 |
Neg logit contributions | |
acs | -0.386 |
Gravity | -0.378 |
assic | -0.377 |
ikhail | -0.376 |
Tokens | -0.368 |
Token | 's |
Token ID | 338 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
confir | +0.032 |
contro | +0.031 |
discharge | +0.030 |
corpor | +0.029 |
Jude | +0.029 |
Neg logit contributions | |
acs | -0.034 |
Gravity | -0.033 |
assic | -0.033 |
ikhail | -0.033 |
Tokens | -0.032 |
Token | vice |
Token ID | 7927 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
confir | +0.304 |
contro | +0.290 |
discharge | +0.283 |
corpor | +0.279 |
Jude | +0.279 |
Neg logit contributions | |
acs | -0.302 |
Gravity | -0.296 |
assic | -0.293 |
ikhail | -0.293 |
Tokens | -0.287 |
Token | presidential |
Token ID | 4787 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
acs | +1.555 |
Gravity | +1.506 |
assic | +1.486 |
ikhail | +1.483 |
Tokens | +1.441 |
Neg logit contributions | |
confir | -2.252 |
contro | -2.159 |
discharge | -2.106 |
corpor | -2.081 |
Jude | -2.081 |
Token | candidate |
Token ID | 4540 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
acs | +0.611 |
Gravity | +0.601 |
assic | +0.595 |
ikhail | +0.593 |
Tokens | +0.582 |
Neg logit contributions | |
confir | -0.570 |
contro | -0.538 |
discharge | -0.519 |
Jude | -0.516 |
corpor | -0.516 |
Token | . |
Token ID | 13 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
acs | +0.350 |
Gravity | +0.344 |
ikhail | +0.340 |
assic | +0.339 |
Tokens | +0.334 |
Neg logit contributions | |
confir | -0.270 |
contro | -0.254 |
discharge | -0.247 |
corpor | -0.244 |
Jude | -0.242 |
Token |
|
Token ID | 198 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
confir | +0.044 |
contro | +0.041 |
discharge | +0.040 |
corpor | +0.039 |
Jude | +0.039 |
Neg logit contributions | |
acs | -0.067 |
Gravity | -0.066 |
assic | -0.065 |
ikhail | -0.065 |
Tokens | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
confir | +0.022 |
contro | +0.020 |
discharge | +0.019 |
corpor | +0.019 |
cumulative | +0.019 |
Neg logit contributions | |
acs | -0.037 |
Gravity | -0.037 |
ikhail | -0.036 |
assic | -0.036 |
Tokens | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
confir | +0.309 |
contro | +0.292 |
discharge | +0.285 |
corpor | +0.281 |
Jude | +0.280 |
Neg logit contributions | |
acs | -0.397 |
Gravity | -0.388 |
assic | -0.386 |
ikhail | -0.386 |
Tokens | -0.378 |
Token | Ma |
Token ID | 6669 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
confir | +0.102 |
contro | +0.096 |
discharge | +0.095 |
corpor | +0.093 |
Jude | +0.093 |
Neg logit contributions | |
acs | -0.106 |
Gravity | -0.104 |
ikhail | -0.103 |
assic | -0.103 |
Tokens | -0.100 |
Token | urer |
Token ID | 15051 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
acs | +0.604 |
Gravity | +0.591 |
assic | +0.588 |
ikhail | +0.586 |
Tokens | +0.575 |
Neg logit contributions | |
confir | -0.541 |
contro | -0.515 |
discharge | -0.502 |
corpor | -0.495 |
Jude | -0.494 |
Token | and |
Token ID | 290 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
acs | +0.435 |
Gravity | +0.426 |
assic | +0.423 |
ikhail | +0.422 |
Tokens | +0.414 |
Neg logit contributions | |
confir | -0.368 |
contro | -0.349 |
discharge | -0.340 |
corpor | -0.336 |
Jude | -0.335 |
Token | Qu |
Token ID | 2264 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
confir | +0.098 |
contro | +0.093 |
discharge | +0.090 |
corpor | +0.089 |
Jude | +0.089 |
Neg logit contributions | |
acs | -0.127 |
Gravity | -0.125 |
ikhail | -0.124 |
assic | -0.123 |
Tokens | -0.121 |
Token | ack |
Token ID | 441 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
confir | +0.452 |
contro | +0.438 |
discharge | +0.430 |
corpor | +0.424 |
Jude | +0.423 |
Neg logit contributions | |
acs | -0.348 |
Gravity | -0.342 |
assic | -0.337 |
ikhail | -0.337 |
Tokens | -0.331 |
Token | en |
Token ID | 268 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
acs | +1.530 |
assic | +1.462 |
Gravity | +1.455 |
ikhail | +1.439 |
ochem | +1.403 |
Neg logit contributions | |
confir | -2.345 |
contro | -2.209 |
discharge | -2.152 |
Jude | -2.140 |
corpor | -2.137 |
Token | bush |
Token ID | 50231 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
confir | +0.059 |
contro | +0.054 |
discharge | +0.052 |
corpor | +0.052 |
Jude | +0.051 |
Neg logit contributions | |
acs | -0.102 |
Gravity | -0.099 |
assic | -0.099 |
ikhail | -0.099 |
Tokens | -0.097 |
Token | , |
Token ID | 11 |
Feature activation | +6.91e-02 |
Pos logit contributions | |
confir | +0.621 |
contro | +0.590 |
discharge | +0.573 |
corpor | +0.566 |
Jude | +0.564 |
Neg logit contributions | |
acs | -0.780 |
Gravity | -0.764 |
assic | -0.760 |
ikhail | -0.759 |
Tokens | -0.745 |
Token | to |
Token ID | 284 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
acs | +0.135 |
Gravity | +0.133 |
ikhail | +0.131 |
assic | +0.131 |
Tokens | +0.128 |
Neg logit contributions | |
confir | -0.110 |
contro | -0.103 |
discharge | -0.101 |
corpor | -0.099 |
Jude | -0.099 |
Token | Triple |
Token ID | 19817 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
confir | +0.082 |
contro | +0.078 |
discharge | +0.076 |
Jude | +0.075 |
corpor | +0.075 |
Neg logit contributions | |
acs | -0.096 |
Gravity | -0.095 |
assic | -0.094 |
ikhail | -0.094 |
Tokens | -0.092 |
Token | the |
Token ID | 262 |
Feature activation | -1.65e-03 |
Pos logit contributions | |
confir | +0.023 |
contro | +0.022 |
discharge | +0.021 |
corpor | +0.021 |
Jude | +0.021 |
Neg logit contributions | |
acs | -0.028 |
assic | -0.027 |
Gravity | -0.027 |
ikhail | -0.027 |
Tokens | -0.027 |
Token | country |
Token ID | 1499 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
confir | +0.003 |
contro | +0.002 |
discharge | +0.002 |
corpor | +0.002 |
Jude | +0.002 |
Neg logit contributions | |
acs | -0.003 |
Gravity | -0.003 |
assic | -0.003 |
ikhail | -0.003 |
Tokens | -0.003 |
Token | knew |
Token ID | 2993 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
acs | +0.349 |
Gravity | +0.341 |
assic | +0.337 |
ikhail | +0.334 |
Tokens | +0.328 |
Neg logit contributions | |
confir | -0.370 |
contro | -0.350 |
discharge | -0.343 |
Jude | -0.338 |
corpor | -0.338 |
Token | who |
Token ID | 508 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
confir | +0.211 |
contro | +0.200 |
discharge | +0.194 |
corpor | +0.191 |
Jude | +0.191 |
Neg logit contributions | |
acs | -0.284 |
Gravity | -0.278 |
assic | -0.277 |
ikhail | -0.276 |
Tokens | -0.271 |
Token | the |
Token ID | 262 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
confir | +0.311 |
contro | +0.296 |
discharge | +0.287 |
corpor | +0.284 |
Jude | +0.283 |
Neg logit contributions | |
acs | -0.408 |
Gravity | -0.399 |
assic | -0.398 |
ikhail | -0.397 |
Tokens | -0.390 |
Token | Koch |
Token ID | 17009 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
confir | +0.029 |
contro | +0.027 |
discharge | +0.026 |
corpor | +0.026 |
Jude | +0.026 |
Neg logit contributions | |
acs | -0.046 |
Gravity | -0.045 |
assic | -0.045 |
ikhail | -0.045 |
Tokens | -0.044 |
Token | brothers |
Token ID | 9397 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
acs | +2.257 |
Gravity | +2.206 |
ikhail | +2.187 |
assic | +2.185 |
Tokens | +2.147 |
Neg logit contributions | |
confir | -1.537 |
contro | -1.440 |
discharge | -1.400 |
Jude | -1.373 |
corpor | -1.370 |
Token | are |
Token ID | 389 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
acs | +1.445 |
Gravity | +1.411 |
assic | +1.401 |
ikhail | +1.400 |
Tokens | +1.378 |
Neg logit contributions | |
confir | -0.957 |
contro | -0.889 |
discharge | -0.860 |
cumulative | -0.848 |
Jude | -0.847 |
Token | . |
Token ID | 13 |
Feature activation | +9.09e-03 |
Pos logit contributions | |
confir | +0.050 |
contro | +0.048 |
discharge | +0.046 |
corpor | +0.046 |
Jude | +0.046 |
Neg logit contributions | |
acs | -0.067 |
Gravity | -0.066 |
assic | -0.066 |
ikhail | -0.065 |
Tokens | -0.064 |
Token | With |
Token ID | 2080 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
acs | +0.016 |
Gravity | +0.016 |
assic | +0.016 |
ikhail | +0.016 |
Tokens | +0.016 |
Neg logit contributions | |
confir | -0.016 |
contro | -0.015 |
discharge | -0.015 |
corpor | -0.015 |
cumulative | -0.014 |
Token | 25 |
Token ID | 1679 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
confir | +0.309 |
contro | +0.296 |
discharge | +0.288 |
corpor | +0.285 |
Jude | +0.284 |
Neg logit contributions | |
acs | -0.346 |
Gravity | -0.338 |
assic | -0.338 |
ikhail | -0.337 |
Tokens | -0.330 |
Token | years |
Token ID | 812 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
confir | +0.149 |
contro | +0.144 |
discharge | +0.141 |
cumulative | +0.139 |
Jude | +0.139 |
Neg logit contributions | |
acs | -0.118 |
Gravity | -0.114 |
ikhail | -0.113 |
assic | -0.113 |
Tokens | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | +7.60e-02 |
Pos logit contributions | |
acs | +0.251 |
Gravity | +0.247 |
assic | +0.245 |
ikhail | +0.245 |
Tokens | +0.240 |
Neg logit contributions | |
confir | -0.186 |
contro | -0.176 |
discharge | -0.171 |
corpor | -0.169 |
Jude | -0.168 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.02e-02 |
Pos logit contributions | |
acs | +0.154 |
Gravity | +0.152 |
assic | +0.150 |
ikhail | +0.150 |
Tokens | +0.147 |
Neg logit contributions | |
confir | -0.116 |
contro | -0.109 |
discharge | -0.106 |
corpor | -0.104 |
cumulative | -0.103 |
Token | Image |
Token ID | 5159 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
acs | +0.136 |
Gravity | +0.133 |
assic | +0.133 |
ikhail | +0.132 |
Tokens | +0.129 |
Neg logit contributions | |
confir | -0.147 |
contro | -0.141 |
discharge | -0.137 |
corpor | -0.136 |
Jude | -0.136 |
Token | copyright |
Token ID | 6634 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
acs | +0.582 |
Gravity | +0.563 |
assic | +0.557 |
ikhail | +0.556 |
Tokens | +0.537 |
Neg logit contributions | |
confir | -1.160 |
contro | -1.121 |
discharge | -1.101 |
corpor | -1.090 |
Jude | -1.088 |
Token | AFP |
Token ID | 8916 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
acs | +0.406 |
Gravity | +0.400 |
assic | +0.396 |
ikhail | +0.394 |
Tokens | +0.386 |
Neg logit contributions | |
confir | -0.366 |
contro | -0.347 |
discharge | -0.339 |
corpor | -0.333 |
cumulative | -0.331 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
acs | +1.666 |
Gravity | +1.623 |
ikhail | +1.609 |
assic | +1.604 |
Tokens | +1.562 |
Neg logit contributions | |
confir | -2.102 |
contro | -2.017 |
discharge | -1.971 |
corpor | -1.942 |
cumulative | -1.940 |
Token |
|
Token ID | 198 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
acs | +0.396 |
Gravity | +0.391 |
ikhail | +0.385 |
assic | +0.384 |
Tokens | +0.382 |
Neg logit contributions | |
confir | -0.237 |
contro | -0.213 |
discharge | -0.206 |
corpor | -0.202 |
cumulative | -0.201 |
Token | The |
Token ID | 464 |
Feature activation | -9.02e-02 |
Pos logit contributions | |
acs | +0.184 |
Gravity | +0.179 |
assic | +0.178 |
ikhail | +0.178 |
Tokens | +0.173 |
Neg logit contributions | |
confir | -0.208 |
contro | -0.199 |
discharge | -0.195 |
corpor | -0.193 |
Jude | -0.193 |
Token | Bank |
Token ID | 5018 |
Feature activation | -5.10e-03 |
Pos logit contributions | |
confir | +0.157 |
contro | +0.150 |
discharge | +0.146 |
corpor | +0.145 |
Jude | +0.144 |
Neg logit contributions | |
acs | -0.162 |
Gravity | -0.158 |
assic | -0.158 |
ikhail | -0.157 |
Tokens | -0.154 |
Token | of |
Token ID | 286 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
confir | +0.007 |
contro | +0.007 |
discharge | +0.007 |
corpor | +0.006 |
Jude | +0.006 |
Neg logit contributions | |
acs | -0.011 |
Gravity | -0.011 |
assic | -0.011 |
ikhail | -0.011 |
Tokens | -0.010 |
Token | Chicago |
Token ID | 4842 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
confir | +0.281 |
contro | +0.268 |
discharge | +0.262 |
corpor | +0.259 |
Jude | +0.258 |
Neg logit contributions | |
acs | -0.304 |
Gravity | -0.297 |
assic | -0.296 |
ikhail | -0.295 |
Tokens | -0.289 |
Token | Bears |
Token ID | 15682 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
confir | +0.105 |
contro | +0.101 |
discharge | +0.098 |
corpor | +0.097 |
Jude | +0.097 |
Neg logit contributions | |
acs | -0.111 |
Gravity | -0.108 |
ikhail | -0.108 |
assic | -0.108 |
Tokens | -0.106 |
Token | have |
Token ID | 423 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
confir | +0.409 |
contro | +0.388 |
discharge | +0.377 |
corpor | +0.372 |
Jude | +0.372 |
Neg logit contributions | |
acs | -0.535 |
Gravity | -0.524 |
assic | -0.519 |
ikhail | -0.518 |
Tokens | -0.510 |
Token | blocked |
Token ID | 10226 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
confir | +0.051 |
contro | +0.048 |
discharge | +0.047 |
corpor | +0.046 |
Jude | +0.046 |
Neg logit contributions | |
acs | -0.066 |
Gravity | -0.065 |
assic | -0.064 |
ikhail | -0.064 |
Tokens | -0.063 |
Token | outside |
Token ID | 2354 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
acs | +0.314 |
Gravity | +0.309 |
assic | +0.305 |
ikhail | +0.305 |
Tokens | +0.299 |
Neg logit contributions | |
confir | -0.273 |
contro | -0.259 |
discharge | -0.251 |
corpor | -0.248 |
Jude | -0.248 |
Token | linebackers |
Token ID | 43081 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
acs | +0.164 |
Gravity | +0.161 |
assic | +0.160 |
ikhail | +0.160 |
Tokens | +0.157 |
Neg logit contributions | |
confir | -0.124 |
contro | -0.117 |
discharge | -0.114 |
corpor | -0.113 |
Jude | -0.112 |
Token | coach |
Token ID | 3985 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
acs | +1.952 |
Gravity | +1.921 |
ikhail | +1.889 |
assic | +1.869 |
Mov | +1.848 |
Neg logit contributions | |
confir | -1.795 |
contro | -1.710 |
discharge | -1.646 |
corpor | -1.621 |
cumulative | -1.616 |
Token | Clint |
Token ID | 21549 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
acs | +0.836 |
Gravity | +0.821 |
assic | +0.809 |
ikhail | +0.809 |
Tokens | +0.788 |
Neg logit contributions | |
confir | -0.881 |
contro | -0.837 |
discharge | -0.816 |
corpor | -0.805 |
cumulative | -0.802 |
Token | Hur |
Token ID | 10077 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
acs | +0.119 |
Gravity | +0.116 |
assic | +0.115 |
ikhail | +0.115 |
Tokens | +0.112 |
Neg logit contributions | |
confir | -0.122 |
contro | -0.116 |
discharge | -0.113 |
corpor | -0.111 |
cumulative | -0.111 |
Token | tt |
Token ID | 926 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
confir | +0.202 |
contro | +0.194 |
discharge | +0.190 |
corpor | +0.187 |
Jude | +0.187 |
Neg logit contributions | |
acs | -0.192 |
Gravity | -0.189 |
assic | -0.188 |
ikhail | -0.187 |
Tokens | -0.183 |
Token | from |
Token ID | 422 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
confir | +0.075 |
contro | +0.072 |
discharge | +0.070 |
corpor | +0.069 |
Jude | +0.068 |
Neg logit contributions | |
acs | -0.096 |
Gravity | -0.094 |
assic | -0.093 |
ikhail | -0.093 |
Tokens | -0.091 |
Token | Indiana |
Token ID | 10278 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
confir | +0.204 |
contro | +0.197 |
discharge | +0.192 |
Jude | +0.191 |
corpor | +0.190 |
Neg logit contributions | |
acs | -0.186 |
assic | -0.181 |
Gravity | -0.180 |
ikhail | -0.180 |
Tokens | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
acs | +0.288 |
Gravity | +0.282 |
assic | +0.281 |
ikhail | +0.280 |
Tokens | +0.275 |
Neg logit contributions | |
confir | -0.201 |
contro | -0.190 |
discharge | -0.184 |
corpor | -0.181 |
Jude | -0.181 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
acs | +0.229 |
Gravity | +0.225 |
assic | +0.223 |
ikhail | +0.223 |
Tokens | +0.219 |
Neg logit contributions | |
confir | -0.170 |
contro | -0.161 |
discharge | -0.157 |
corpor | -0.154 |
Jude | -0.154 |
Token | The |
Token ID | 464 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
confir | +0.269 |
contro | +0.258 |
discharge | +0.252 |
corpor | +0.249 |
Jude | +0.249 |
Neg logit contributions | |
acs | -0.251 |
Gravity | -0.246 |
assic | -0.244 |
ikhail | -0.244 |
Tokens | -0.237 |
Token | Bengals |
Token ID | 24037 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
confir | +0.254 |
contro | +0.243 |
discharge | +0.237 |
corpor | +0.234 |
Jude | +0.233 |
Neg logit contributions | |
acs | -0.276 |
Gravity | -0.270 |
assic | -0.268 |
ikhail | -0.268 |
Tokens | -0.263 |
Token | signed |
Token ID | 4488 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
confir | +0.094 |
contro | +0.089 |
discharge | +0.087 |
corpor | +0.086 |
Jude | +0.086 |
Neg logit contributions | |
acs | -0.122 |
Gravity | -0.119 |
assic | -0.118 |
ikhail | -0.118 |
Tokens | -0.116 |
Token | kicker |
Token ID | 38297 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
acs | +2.008 |
Gravity | +1.994 |
ikhail | +1.955 |
assic | +1.944 |
Mov | +1.917 |
Neg logit contributions | |
confir | -1.711 |
contro | -1.586 |
discharge | -1.559 |
corpor | -1.525 |
Jude | -1.523 |
Token | Jonathan |
Token ID | 11232 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
acs | +0.400 |
Gravity | +0.392 |
ikhail | +0.387 |
assic | +0.387 |
Tokens | +0.377 |
Neg logit contributions | |
confir | -0.403 |
contro | -0.382 |
discharge | -0.372 |
corpor | -0.367 |
cumulative | -0.366 |
Token | Brown |
Token ID | 4373 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
confir | +1.010 |
contro | +0.973 |
discharge | +0.951 |
Jude | +0.944 |
corpor | +0.944 |
Neg logit contributions | |
acs | -1.084 |
assic | -1.060 |
ikhail | -1.057 |
Gravity | -1.055 |
Tokens | -1.037 |
Token | , |
Token ID | 11 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
confir | +1.069 |
contro | +1.014 |
discharge | +0.985 |
corpor | +0.972 |
Jude | +0.969 |
Neg logit contributions | |
acs | -1.376 |
Gravity | -1.347 |
assic | -1.341 |
ikhail | -1.338 |
Tokens | -1.313 |
Token | far |
Token ID | 1290 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
confir | +0.065 |
contro | +0.061 |
discharge | +0.059 |
Jude | +0.058 |
corpor | +0.058 |
Neg logit contributions | |
acs | -0.083 |
Gravity | -0.081 |
ikhail | -0.080 |
assic | -0.080 |
Tokens | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
acs | +0.035 |
assic | +0.035 |
Gravity | +0.035 |
ikhail | +0.034 |
Tokens | +0.034 |
Neg logit contributions | |
confir | -0.026 |
contro | -0.025 |
discharge | -0.024 |
corpor | -0.024 |
Jude | -0.024 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
acs | +0.255 |
Gravity | +0.250 |
assic | +0.249 |
ikhail | +0.249 |
Tokens | +0.244 |
Neg logit contributions | |
confir | -0.162 |
contro | -0.153 |
discharge | -0.148 |
Jude | -0.146 |
corpor | -0.146 |
Token | Media |
Token ID | 13152 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
confir | +0.081 |
contro | +0.078 |
discharge | +0.076 |
corpor | +0.075 |
Jude | +0.075 |
Neg logit contributions | |
acs | -0.076 |
Gravity | -0.074 |
assic | -0.073 |
ikhail | -0.073 |
Tokens | -0.071 |
Token | playback |
Token ID | 16388 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
confir | +0.021 |
contro | +0.020 |
discharge | +0.019 |
corpor | +0.019 |
Jude | +0.019 |
Neg logit contributions | |
acs | -0.029 |
Gravity | -0.028 |
assic | -0.028 |
ikhail | -0.028 |
Tokens | -0.027 |
Token | is |
Token ID | 318 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
acs | +0.959 |
Gravity | +0.948 |
assic | +0.940 |
ikhail | +0.934 |
Tokens | +0.910 |
Neg logit contributions | |
confir | -0.964 |
contro | -0.896 |
discharge | -0.878 |
corpor | -0.865 |
cumulative | -0.860 |
Token | unsupported |
Token ID | 24222 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
acs | +0.182 |
ikhail | +0.175 |
assic | +0.175 |
Gravity | +0.174 |
Tokens | +0.170 |
Neg logit contributions | |
confir | -0.343 |
contro | -0.334 |
discharge | -0.326 |
corpor | -0.325 |
Jude | -0.324 |
Token | on |
Token ID | 319 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
acs | +1.754 |
Gravity | +1.753 |
ikhail | +1.692 |
assic | +1.685 |
Tokens | +1.665 |
Neg logit contributions | |
confir | -2.003 |
contro | -1.841 |
discharge | -1.809 |
corpor | -1.786 |
Jude | -1.782 |
Token | your |
Token ID | 534 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
acs | +0.275 |
Gravity | +0.273 |
assic | +0.268 |
ikhail | +0.267 |
Tokens | +0.261 |
Neg logit contributions | |
confir | -0.280 |
contro | -0.261 |
discharge | -0.255 |
Jude | -0.253 |
corpor | -0.252 |
Token | device |
Token ID | 3335 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
acs | +0.789 |
Gravity | +0.787 |
ikhail | +0.775 |
assic | +0.773 |
Mov | +0.760 |
Neg logit contributions | |
confir | -0.485 |
contro | -0.436 |
discharge | -0.425 |
Jude | -0.423 |
corpor | -0.415 |
Token | Media |
Token ID | 6343 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
acs | +1.064 |
Gravity | +1.063 |
ikhail | +1.029 |
assic | +1.025 |
Mov | +1.010 |
Neg logit contributions | |
confir | -1.024 |
contro | -0.949 |
discharge | -0.931 |
corpor | -0.920 |
cumulative | -0.914 |
Token | caption |
Token ID | 8305 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
acs | +0.031 |
Gravity | +0.030 |
assic | +0.030 |
ikhail | +0.030 |
Tokens | +0.028 |
Neg logit contributions | |
confir | -0.079 |
contro | -0.077 |
discharge | -0.075 |
corpor | -0.075 |
Jude | -0.075 |
Token | solve |
Token ID | 8494 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
confir | +0.458 |
contro | +0.435 |
discharge | +0.422 |
corpor | +0.417 |
Jude | +0.416 |
Neg logit contributions | |
acs | -0.560 |
Gravity | -0.549 |
assic | -0.544 |
ikhail | -0.542 |
Tokens | -0.533 |
Token | the |
Token ID | 262 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
acs | +0.506 |
Gravity | +0.498 |
assic | +0.491 |
ikhail | +0.491 |
Tokens | +0.480 |
Neg logit contributions | |
confir | -0.449 |
contro | -0.426 |
discharge | -0.415 |
corpor | -0.409 |
Jude | -0.408 |
Token | Pittsburgh |
Token ID | 12767 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
acs | +0.053 |
Gravity | +0.052 |
assic | +0.052 |
ikhail | +0.051 |
Tokens | +0.050 |
Neg logit contributions | |
confir | -0.051 |
contro | -0.048 |
discharge | -0.047 |
corpor | -0.047 |
Jude | -0.047 |
Token | Penguins |
Token ID | 29755 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
acs | +0.574 |
Gravity | +0.559 |
ikhail | +0.552 |
assic | +0.551 |
Tokens | +0.535 |
Neg logit contributions | |
confir | -0.813 |
contro | -0.778 |
discharge | -0.762 |
corpor | -0.757 |
Jude | -0.753 |
Token | expansion |
Token ID | 7118 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
confir | +0.022 |
contro | +0.021 |
discharge | +0.021 |
corpor | +0.020 |
Jude | +0.020 |
Neg logit contributions | |
acs | -0.026 |
Gravity | -0.025 |
ikhail | -0.025 |
assic | -0.025 |
Tokens | -0.024 |
Token | draft |
Token ID | 4538 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
acs | +0.116 |
Gravity | +0.115 |
ikhail | +0.113 |
assic | +0.113 |
Tokens | +0.111 |
Neg logit contributions | |
confir | -0.079 |
contro | -0.073 |
discharge | -0.070 |
corpor | -0.070 |
Jude | -0.070 |
Token | problem |
Token ID | 1917 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
acs | +2.066 |
Gravity | +2.063 |
assic | +2.027 |
ikhail | +2.019 |
Tokens | +1.977 |
Neg logit contributions | |
confir | -1.637 |
contro | -1.511 |
discharge | -1.462 |
Jude | -1.444 |
corpor | -1.442 |
Token | ? |
Token ID | 30 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
confir | +0.206 |
contro | +0.196 |
discharge | +0.191 |
corpor | +0.188 |
Jude | +0.188 |
Neg logit contributions | |
acs | -0.234 |
Gravity | -0.229 |
assic | -0.228 |
ikhail | -0.227 |
Tokens | -0.223 |
Token | If |
Token ID | 1002 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
confir | +0.022 |
contro | +0.020 |
discharge | +0.020 |
corpor | +0.020 |
cumulative | +0.019 |
Neg logit contributions | |
acs | -0.028 |
Gravity | -0.027 |
ikhail | -0.027 |
assic | -0.027 |
Tokens | -0.026 |
Token | you |
Token ID | 345 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
confir | +0.124 |
contro | +0.119 |
discharge | +0.116 |
corpor | +0.115 |
Jude | +0.114 |
Neg logit contributions | |
acs | -0.112 |
Gravity | -0.110 |
ikhail | -0.108 |
assic | -0.108 |
Tokens | -0.106 |
Token | believe |
Token ID | 1975 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
confir | +0.033 |
contro | +0.031 |
discharge | +0.030 |
Jude | +0.030 |
corpor | +0.030 |
Neg logit contributions | |
acs | -0.038 |
Gravity | -0.038 |
assic | -0.037 |
ikhail | -0.037 |
Tokens | -0.036 |
Token | is |
Token ID | 318 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
acs | +0.135 |
Gravity | +0.131 |
assic | +0.131 |
ikhail | +0.130 |
Tokens | +0.128 |
Neg logit contributions | |
confir | -0.121 |
contro | -0.115 |
discharge | -0.112 |
Jude | -0.111 |
corpor | -0.111 |
Token | able |
Token ID | 1498 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
acs | +0.136 |
Gravity | +0.133 |
assic | +0.132 |
ikhail | +0.131 |
Tokens | +0.129 |
Neg logit contributions | |
confir | -0.117 |
contro | -0.111 |
discharge | -0.108 |
Jude | -0.107 |
corpor | -0.106 |
Token | to |
Token ID | 284 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
acs | +0.456 |
Gravity | +0.447 |
assic | +0.445 |
ikhail | +0.444 |
Tokens | +0.436 |
Neg logit contributions | |
confir | -0.324 |
contro | -0.306 |
discharge | -0.297 |
corpor | -0.293 |
Jude | -0.292 |
Token | meet |
Token ID | 1826 |
Feature activation | -3.62e-03 |
Pos logit contributions | |
acs | +0.221 |
Gravity | +0.216 |
assic | +0.213 |
ikhail | +0.212 |
Tokens | +0.210 |
Neg logit contributions | |
confir | -0.190 |
contro | -0.179 |
Jude | -0.174 |
discharge | -0.173 |
corpor | -0.173 |
Token | its |
Token ID | 663 |
Feature activation | +8.53e-02 |
Pos logit contributions | |
confir | +0.006 |
contro | +0.005 |
discharge | +0.005 |
corpor | +0.005 |
Jude | +0.005 |
Neg logit contributions | |
acs | -0.007 |
Gravity | -0.007 |
assic | -0.007 |
ikhail | -0.007 |
Tokens | -0.007 |
Token | deficit |
Token ID | 11807 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
acs | +0.156 |
Gravity | +0.154 |
assic | +0.152 |
ikhail | +0.151 |
Tokens | +0.149 |
Neg logit contributions | |
confir | -0.146 |
contro | -0.138 |
discharge | -0.134 |
Jude | -0.133 |
corpor | -0.132 |
Token | targets |
Token ID | 6670 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
acs | +2.278 |
Gravity | +2.232 |
assic | +2.194 |
ikhail | +2.188 |
Tokens | +2.170 |
Neg logit contributions | |
confir | -1.409 |
contro | -1.302 |
Jude | -1.260 |
corpor | -1.250 |
discharge | -1.242 |
Token | , |
Token ID | 11 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
confir | +0.441 |
contro | +0.419 |
discharge | +0.407 |
corpor | +0.401 |
Jude | +0.400 |
Neg logit contributions | |
acs | -0.563 |
Gravity | -0.551 |
assic | -0.548 |
ikhail | -0.547 |
Tokens | -0.537 |
Token | given |
Token ID | 1813 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
acs | +0.023 |
Gravity | +0.022 |
assic | +0.022 |
ikhail | +0.022 |
Tokens | +0.022 |
Neg logit contributions | |
confir | -0.017 |
contro | -0.016 |
discharge | -0.016 |
corpor | -0.016 |
Jude | -0.016 |
Token | that |
Token ID | 326 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
acs | +0.003 |
Gravity | +0.003 |
ikhail | +0.003 |
assic | +0.003 |
Tokens | +0.003 |
Neg logit contributions | |
confir | -0.003 |
contro | -0.003 |
discharge | -0.003 |
corpor | -0.003 |
Jude | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
acs | +0.118 |
Gravity | +0.115 |
assic | +0.114 |
ikhail | +0.113 |
Tokens | +0.112 |
Neg logit contributions | |
confir | -0.097 |
contro | -0.091 |
discharge | -0.089 |
Jude | -0.088 |
corpor | -0.087 |
Token | d |
Token ID | 67 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
acs | +0.397 |
assic | +0.385 |
Gravity | +0.384 |
ikhail | +0.381 |
Tokens | +0.372 |
Neg logit contributions | |
confir | -0.401 |
contro | -0.382 |
discharge | -0.371 |
corpor | -0.366 |
cumulative | -0.366 |
Token | ong |
Token ID | 506 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
acs | +0.283 |
assic | +0.273 |
Gravity | +0.273 |
ikhail | +0.271 |
Tokens | +0.260 |
Neg logit contributions | |
confir | -0.387 |
contro | -0.370 |
discharge | -0.361 |
corpor | -0.358 |
Jude | -0.357 |
Token | province |
Token ID | 8473 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
acs | +0.410 |
Gravity | +0.403 |
assic | +0.401 |
ikhail | +0.397 |
Tokens | +0.390 |
Neg logit contributions | |
confir | -0.269 |
contro | -0.254 |
discharge | -0.245 |
corpor | -0.241 |
Jude | -0.240 |
Token | turns |
Token ID | 4962 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
confir | +0.341 |
contro | +0.325 |
discharge | +0.316 |
corpor | +0.312 |
Jude | +0.311 |
Neg logit contributions | |
acs | -0.389 |
Gravity | -0.380 |
assic | -0.379 |
ikhail | -0.378 |
Tokens | -0.370 |
Token | redd |
Token ID | 14688 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
acs | +0.225 |
Gravity | +0.222 |
assic | +0.219 |
ikhail | +0.218 |
Tokens | +0.214 |
Neg logit contributions | |
confir | -0.180 |
contro | -0.169 |
discharge | -0.164 |
corpor | -0.162 |
Jude | -0.161 |
Token | ish |
Token ID | 680 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
confir | +0.583 |
contro | +0.543 |
discharge | +0.527 |
Jude | +0.508 |
corpor | +0.507 |
Neg logit contributions | |
acs | -1.449 |
Gravity | -1.432 |
assic | -1.423 |
ikhail | -1.419 |
Tokens | -1.401 |
Token | brown |
Token ID | 7586 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
confir | +1.926 |
contro | +1.880 |
discharge | +1.855 |
corpor | +1.801 |
Jude | +1.797 |
Neg logit contributions | |
acs | -2.486 |
Gravity | -2.436 |
assic | -2.432 |
ikhail | -2.429 |
Tokens | -2.370 |
Token | after |
Token ID | 706 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
confir | +0.261 |
contro | +0.248 |
discharge | +0.241 |
corpor | +0.238 |
Jude | +0.237 |
Neg logit contributions | |
acs | -0.330 |
Gravity | -0.323 |
assic | -0.321 |
ikhail | -0.320 |
Tokens | -0.314 |
Token | the |
Token ID | 262 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
confir | +0.170 |
contro | +0.162 |
discharge | +0.157 |
corpor | +0.155 |
Jude | +0.155 |
Neg logit contributions | |
acs | -0.206 |
Gravity | -0.202 |
assic | -0.201 |
ikhail | -0.200 |
Tokens | -0.196 |
Token | water |
Token ID | 1660 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
confir | +0.259 |
contro | +0.246 |
discharge | +0.239 |
corpor | +0.236 |
Jude | +0.235 |
Neg logit contributions | |
acs | -0.313 |
Gravity | -0.306 |
assic | -0.305 |
ikhail | -0.304 |
Tokens | -0.298 |
Token | was |
Token ID | 373 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
confir | +0.326 |
contro | +0.312 |
discharge | +0.307 |
corpor | +0.301 |
cumulative | +0.299 |
Neg logit contributions | |
acs | -0.363 |
Gravity | -0.354 |
ikhail | -0.353 |
assic | -0.352 |
Tokens | -0.347 |
Token | atics |
Token ID | 23372 |
Feature activation | -5.86e-01 |
Pos logit contributions | |
acs | +0.081 |
Gravity | +0.079 |
assic | +0.079 |
ikhail | +0.078 |
Tokens | +0.077 |
Neg logit contributions | |
confir | -0.115 |
contro | -0.111 |
discharge | -0.108 |
corpor | -0.107 |
Jude | -0.107 |
Token | and |
Token ID | 290 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
confir | +0.883 |
contro | +0.837 |
discharge | +0.813 |
corpor | +0.801 |
Jude | +0.799 |
Neg logit contributions | |
acs | -1.192 |
Gravity | -1.167 |
assic | -1.162 |
ikhail | -1.159 |
Tokens | -1.138 |
Token | over |
Token ID | 625 |
Feature activation | +9.75e-02 |
Pos logit contributions | |
acs | +0.045 |
Gravity | +0.044 |
assic | +0.044 |
ikhail | +0.044 |
Tokens | +0.043 |
Neg logit contributions | |
confir | -0.038 |
contro | -0.036 |
discharge | -0.035 |
corpor | -0.035 |
Jude | -0.035 |
Token | 90 |
Token ID | 4101 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
acs | +0.190 |
Gravity | +0.187 |
assic | +0.185 |
ikhail | +0.185 |
Tokens | +0.182 |
Neg logit contributions | |
confir | -0.155 |
contro | -0.147 |
discharge | -0.143 |
corpor | -0.141 |
Jude | -0.140 |
Token | 000 |
Token ID | 12877 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
confir | +0.424 |
contro | +0.401 |
discharge | +0.391 |
Jude | +0.386 |
corpor | +0.386 |
Neg logit contributions | |
acs | -0.498 |
Gravity | -0.489 |
assic | -0.484 |
ikhail | -0.482 |
Tokens | -0.473 |
Token | rounds |
Token ID | 9196 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
confir | +0.901 |
contro | +0.853 |
discharge | +0.829 |
corpor | +0.817 |
Jude | +0.815 |
Neg logit contributions | |
acs | -1.197 |
Gravity | -1.173 |
assic | -1.167 |
ikhail | -1.164 |
Tokens | -1.143 |
Token | of |
Token ID | 286 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
confir | +1.678 |
contro | +1.603 |
discharge | +1.552 |
cumulative | +1.519 |
corpor | +1.510 |
Neg logit contributions | |
acs | -2.697 |
assic | -2.647 |
Gravity | -2.645 |
ikhail | -2.640 |
Tokens | -2.594 |
Token | blank |
Token ID | 9178 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
confir | +0.171 |
contro | +0.158 |
discharge | +0.154 |
Jude | +0.153 |
corpor | +0.152 |
Neg logit contributions | |
acs | -0.243 |
Gravity | -0.242 |
assic | -0.237 |
ikhail | -0.236 |
Tokens | -0.233 |
Token | ammunition |
Token ID | 14271 |
Feature activation | -7.70e-03 |
Pos logit contributions | |
acs | +0.573 |
Gravity | +0.566 |
assic | +0.561 |
ikhail | +0.559 |
Tokens | +0.551 |
Neg logit contributions | |
confir | -0.391 |
contro | -0.363 |
discharge | -0.350 |
corpor | -0.348 |
Jude | -0.348 |
Token | . |
Token ID | 13 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
confir | +0.012 |
contro | +0.011 |
discharge | +0.011 |
corpor | +0.011 |
Jude | +0.011 |
Neg logit contributions | |
acs | -0.015 |
Gravity | -0.015 |
assic | -0.015 |
ikhail | -0.015 |
Tokens | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
confir | +0.021 |
contro | +0.020 |
discharge | +0.020 |
corpor | +0.019 |
Jude | +0.019 |
Neg logit contributions | |
acs | -0.024 |
Gravity | -0.024 |
assic | -0.024 |
ikhail | -0.024 |
Tokens | -0.023 |
Token | ongoing |
Token ID | 7044 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
confir | +0.164 |
contro | +0.154 |
discharge | +0.150 |
corpor | +0.148 |
cumulative | +0.146 |
Neg logit contributions | |
acs | -0.199 |
Gravity | -0.196 |
assic | -0.194 |
ikhail | -0.194 |
Tokens | -0.190 |
Token | violations |
Token ID | 11734 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
confir | +0.236 |
contro | +0.221 |
discharge | +0.214 |
corpor | +0.210 |
Jude | +0.208 |
Neg logit contributions | |
acs | -0.341 |
Gravity | -0.335 |
assic | -0.332 |
ikhail | -0.332 |
Tokens | -0.326 |
Token | and |
Token ID | 290 |
Feature activation | -9.64e-02 |
Pos logit contributions | |
confir | +0.514 |
contro | +0.486 |
discharge | +0.472 |
corpor | +0.464 |
Jude | +0.463 |
Neg logit contributions | |
acs | -0.763 |
Gravity | -0.748 |
assic | -0.745 |
ikhail | -0.743 |
Tokens | -0.730 |
Token | increase |
Token ID | 2620 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
confir | +0.154 |
contro | +0.145 |
discharge | +0.141 |
corpor | +0.139 |
Jude | +0.139 |
Neg logit contributions | |
acs | -0.187 |
Gravity | -0.184 |
assic | -0.182 |
ikhail | -0.182 |
Tokens | -0.178 |
Token | in |
Token ID | 287 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
confir | +0.374 |
contro | +0.352 |
discharge | +0.343 |
corpor | +0.337 |
Jude | +0.334 |
Neg logit contributions | |
acs | -0.473 |
Gravity | -0.465 |
ikhail | -0.460 |
assic | -0.459 |
Tokens | -0.450 |
Token | settlement |
Token ID | 9443 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
confir | +0.095 |
contro | +0.089 |
discharge | +0.087 |
corpor | +0.086 |
Jude | +0.085 |
Neg logit contributions | |
acs | -0.113 |
Gravity | -0.111 |
ikhail | -0.110 |
assic | -0.110 |
Tokens | -0.108 |
Token | expansion |
Token ID | 7118 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
confir | +1.944 |
contro | +1.859 |
discharge | +1.815 |
corpor | +1.790 |
Jude | +1.781 |
Neg logit contributions | |
acs | -2.418 |
Gravity | -2.364 |
assic | -2.362 |
ikhail | -2.362 |
Tokens | -2.313 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
confir | +0.399 |
contro | +0.377 |
discharge | +0.366 |
corpor | +0.362 |
Jude | +0.360 |
Neg logit contributions | |
acs | -0.527 |
Gravity | -0.516 |
ikhail | -0.512 |
assic | -0.512 |
Tokens | -0.502 |
Token | A |
Token ID | 317 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
confir | +0.025 |
contro | +0.024 |
discharge | +0.023 |
corpor | +0.023 |
Jude | +0.023 |
Neg logit contributions | |
acs | -0.024 |
Gravity | -0.023 |
assic | -0.023 |
ikhail | -0.023 |
Tokens | -0.022 |
Token | shift |
Token ID | 6482 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
confir | +0.223 |
contro | +0.210 |
discharge | +0.204 |
corpor | +0.201 |
Jude | +0.200 |
Neg logit contributions | |
acs | -0.302 |
Gravity | -0.297 |
ikhail | -0.295 |
assic | -0.294 |
Tokens | -0.288 |
Token | has |
Token ID | 468 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
confir | +0.679 |
contro | +0.642 |
discharge | +0.623 |
corpor | +0.614 |
Jude | +0.612 |
Neg logit contributions | |
acs | -0.950 |
Gravity | -0.931 |
assic | -0.927 |
ikhail | -0.925 |
Tokens | -0.908 |
Token | would |
Token ID | 561 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
confir | +0.261 |
contro | +0.247 |
discharge | +0.240 |
corpor | +0.237 |
Jude | +0.236 |
Neg logit contributions | |
acs | -0.325 |
Gravity | -0.319 |
assic | -0.316 |
ikhail | -0.315 |
Tokens | -0.310 |
Token | drive |
Token ID | 3708 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
confir | +0.275 |
contro | +0.261 |
discharge | +0.255 |
corpor | +0.251 |
Jude | +0.250 |
Neg logit contributions | |
acs | -0.334 |
Gravity | -0.327 |
assic | -0.326 |
ikhail | -0.325 |
Tokens | -0.319 |
Token | nearly |
Token ID | 3016 |
Feature activation | +7.80e-02 |
Pos logit contributions | |
acs | +0.127 |
Gravity | +0.125 |
assic | +0.124 |
ikhail | +0.123 |
Tokens | +0.121 |
Neg logit contributions | |
confir | -0.106 |
contro | -0.100 |
discharge | -0.097 |
corpor | -0.096 |
Jude | -0.096 |
Token | nine |
Token ID | 5193 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
acs | +0.148 |
Gravity | +0.146 |
assic | +0.144 |
ikhail | +0.143 |
Tokens | +0.141 |
Neg logit contributions | |
confir | -0.128 |
contro | -0.121 |
discharge | -0.118 |
corpor | -0.116 |
Jude | -0.116 |
Token | hours |
Token ID | 2250 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
confir | +0.432 |
contro | +0.416 |
discharge | +0.408 |
corpor | +0.403 |
Jude | +0.402 |
Neg logit contributions | |
acs | -0.339 |
Gravity | -0.330 |
assic | -0.328 |
ikhail | -0.328 |
Tokens | -0.319 |
Token | round |
Token ID | 2835 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
acs | +0.102 |
Gravity | +0.100 |
assic | +0.099 |
ikhail | +0.099 |
Tokens | +0.098 |
Neg logit contributions | |
confir | -0.080 |
contro | -0.076 |
discharge | -0.073 |
corpor | -0.073 |
Jude | -0.072 |
Token | trip |
Token ID | 5296 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
confir | +1.853 |
contro | +1.783 |
discharge | +1.761 |
Jude | +1.730 |
cumulative | +1.723 |
Neg logit contributions | |
acs | -2.384 |
ikhail | -2.330 |
assic | -2.327 |
Gravity | -2.306 |
Tokens | -2.273 |
Token | in |
Token ID | 287 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
confir | +0.193 |
contro | +0.183 |
discharge | +0.177 |
corpor | +0.175 |
Jude | +0.174 |
Neg logit contributions | |
acs | -0.255 |
Gravity | -0.250 |
assic | -0.249 |
ikhail | -0.248 |
Tokens | -0.244 |
Token | a |
Token ID | 257 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
confir | +0.010 |
contro | +0.010 |
discharge | +0.010 |
corpor | +0.009 |
Jude | +0.009 |
Neg logit contributions | |
acs | -0.011 |
Gravity | -0.011 |
assic | -0.011 |
ikhail | -0.011 |
Tokens | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
confir | +0.221 |
contro | +0.208 |
discharge | +0.203 |
corpor | +0.200 |
Jude | +0.199 |
Neg logit contributions | |
acs | -0.250 |
Gravity | -0.248 |
assic | -0.242 |
ikhail | -0.241 |
Mov | -0.237 |
Token | to |
Token ID | 284 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
acs | +0.082 |
Gravity | +0.080 |
ikhail | +0.079 |
assic | +0.079 |
Tokens | +0.078 |
Neg logit contributions | |
confir | -0.064 |
contro | -0.061 |
discharge | -0.059 |
Jude | -0.058 |
corpor | -0.058 |
Token | stated |
Token ID | 5081 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
acs | +0.538 |
Gravity | +0.528 |
ikhail | +0.523 |
assic | +0.523 |
Tokens | +0.512 |
Neg logit contributions | |
confir | -0.437 |
contro | -0.416 |
discharge | -0.406 |
corpor | -0.400 |
Jude | -0.398 |
Token | that |
Token ID | 326 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
acs | +0.256 |
Gravity | +0.251 |
assic | +0.249 |
ikhail | +0.249 |
Tokens | +0.244 |
Neg logit contributions | |
confir | -0.207 |
contro | -0.197 |
discharge | -0.191 |
corpor | -0.188 |
Jude | -0.188 |
Token | the |
Token ID | 262 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
confir | +0.206 |
contro | +0.195 |
discharge | +0.189 |
corpor | +0.186 |
Jude | +0.186 |
Neg logit contributions | |
acs | -0.258 |
Gravity | -0.254 |
ikhail | -0.251 |
assic | -0.251 |
Tokens | -0.247 |
Token | Ministry |
Token ID | 9475 |
Feature activation | +6.89e-02 |
Pos logit contributions | |
confir | +0.060 |
contro | +0.057 |
discharge | +0.055 |
corpor | +0.054 |
Jude | +0.054 |
Neg logit contributions | |
acs | -0.080 |
Gravity | -0.079 |
ikhail | -0.078 |
assic | -0.078 |
Tokens | -0.076 |
Token | of |
Token ID | 286 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
acs | +0.137 |
Gravity | +0.135 |
assic | +0.133 |
ikhail | +0.133 |
Tokens | +0.131 |
Neg logit contributions | |
confir | -0.106 |
contro | -0.100 |
discharge | -0.097 |
corpor | -0.096 |
Jude | -0.096 |
Token | Internal |
Token ID | 18628 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
confir | +0.177 |
contro | +0.167 |
discharge | +0.162 |
corpor | +0.159 |
cumulative | +0.158 |
Neg logit contributions | |
acs | -0.250 |
Gravity | -0.246 |
assic | -0.244 |
ikhail | -0.244 |
Tokens | -0.239 |
Token | Affairs |
Token ID | 10665 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
confir | +2.053 |
contro | +1.974 |
discharge | +1.936 |
Jude | +1.924 |
corpor | +1.906 |
Neg logit contributions | |
acs | -2.163 |
assic | -2.121 |
ikhail | -2.100 |
Gravity | -2.099 |
Tokens | -2.067 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
confir | +0.605 |
contro | +0.576 |
discharge | +0.560 |
corpor | +0.553 |
Jude | +0.551 |
Neg logit contributions | |
acs | -0.707 |
Gravity | -0.693 |
assic | -0.687 |
ikhail | -0.685 |
Tokens | -0.672 |
Token | expressed |
Token ID | 6241 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
acs | +0.189 |
Gravity | +0.185 |
assic | +0.182 |
ikhail | +0.181 |
Tokens | +0.178 |
Neg logit contributions | |
confir | -0.169 |
contro | -0.161 |
discharge | -0.157 |
Jude | -0.155 |
cumulative | -0.155 |
Token | its |
Token ID | 663 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
acs | +0.160 |
Gravity | +0.157 |
ikhail | +0.155 |
assic | +0.155 |
Tokens | +0.152 |
Neg logit contributions | |
confir | -0.135 |
contro | -0.128 |
discharge | -0.124 |
corpor | -0.122 |
Jude | -0.122 |
Token | willingness |
Token ID | 16826 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
confir | +0.236 |
contro | +0.224 |
discharge | +0.217 |
corpor | +0.214 |
Jude | +0.214 |
Neg logit contributions | |
acs | -0.282 |
Gravity | -0.277 |
assic | -0.275 |
ikhail | -0.274 |
Tokens | -0.269 |
Token | a |
Token ID | 257 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
confir | +0.205 |
contro | +0.195 |
discharge | +0.190 |
corpor | +0.188 |
Jude | +0.187 |
Neg logit contributions | |
acs | -0.239 |
Gravity | -0.233 |
assic | -0.232 |
ikhail | -0.232 |
Tokens | -0.227 |
Token | service |
Token ID | 2139 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
confir | +0.092 |
contro | +0.088 |
discharge | +0.085 |
corpor | +0.084 |
Jude | +0.084 |
Neg logit contributions | |
acs | -0.101 |
Gravity | -0.099 |
assic | -0.098 |
ikhail | -0.098 |
Tokens | -0.096 |
Token | pistol |
Token ID | 16790 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
acs | +1.106 |
Gravity | +1.080 |
assic | +1.074 |
ikhail | +1.067 |
Tokens | +1.046 |
Neg logit contributions | |
confir | -1.047 |
contro | -0.997 |
discharge | -0.966 |
Jude | -0.957 |
cumulative | -0.955 |
Token | and |
Token ID | 290 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
confir | +0.074 |
contro | +0.070 |
discharge | +0.068 |
corpor | +0.067 |
Jude | +0.067 |
Neg logit contributions | |
acs | -0.088 |
Gravity | -0.086 |
assic | -0.085 |
ikhail | -0.085 |
Tokens | -0.084 |
Token | 15 |
Token ID | 1315 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
acs | +0.102 |
Gravity | +0.101 |
assic | +0.099 |
ikhail | +0.099 |
Tokens | +0.097 |
Neg logit contributions | |
confir | -0.079 |
contro | -0.075 |
discharge | -0.072 |
Jude | -0.071 |
cumulative | -0.071 |
Token | rounds |
Token ID | 9196 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
confir | +0.413 |
contro | +0.394 |
discharge | +0.384 |
corpor | +0.379 |
Jude | +0.378 |
Neg logit contributions | |
acs | -0.445 |
Gravity | -0.434 |
assic | -0.432 |
ikhail | -0.431 |
Tokens | -0.423 |
Token | of |
Token ID | 286 |
Feature activation | -7.32e-03 |
Pos logit contributions | |
confir | +1.609 |
contro | +1.532 |
discharge | +1.492 |
cumulative | +1.456 |
corpor | +1.455 |
Neg logit contributions | |
acs | -2.596 |
Gravity | -2.547 |
assic | -2.540 |
ikhail | -2.527 |
Tokens | -2.491 |
Token | ammunition |
Token ID | 14271 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
confir | +0.011 |
contro | +0.010 |
discharge | +0.009 |
Jude | +0.009 |
corpor | +0.009 |
Neg logit contributions | |
acs | -0.015 |
Gravity | -0.015 |
assic | -0.015 |
ikhail | -0.015 |
Mov | -0.015 |
Token | that |
Token ID | 326 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
acs | +0.033 |
Gravity | +0.033 |
ikhail | +0.032 |
assic | +0.032 |
Tokens | +0.031 |
Neg logit contributions | |
confir | -0.029 |
contro | -0.027 |
discharge | -0.026 |
Jude | -0.026 |
corpor | -0.026 |
Token | disappeared |
Token ID | 12120 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
acs | +0.235 |
Gravity | +0.230 |
assic | +0.229 |
ikhail | +0.228 |
Tokens | +0.223 |
Neg logit contributions | |
confir | -0.229 |
contro | -0.218 |
discharge | -0.213 |
corpor | -0.210 |
Jude | -0.210 |
Token | last |
Token ID | 938 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
acs | +0.161 |
Gravity | +0.158 |
assic | +0.157 |
ikhail | +0.157 |
Tokens | +0.154 |
Neg logit contributions | |
confir | -0.122 |
contro | -0.115 |
discharge | -0.112 |
corpor | -0.110 |
Jude | -0.110 |
Token | found |
Token ID | 1043 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
confir | +0.094 |
contro | +0.089 |
discharge | +0.087 |
corpor | +0.085 |
Jude | +0.085 |
Neg logit contributions | |
acs | -0.132 |
Gravity | -0.129 |
assic | -0.129 |
ikhail | -0.129 |
Tokens | -0.126 |
Token | a |
Token ID | 257 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
confir | +0.598 |
contro | +0.572 |
discharge | +0.556 |
corpor | +0.550 |
Jude | +0.546 |
Neg logit contributions | |
acs | -0.739 |
assic | -0.719 |
Gravity | -0.719 |
ikhail | -0.718 |
Tokens | -0.705 |
Token | ' |
Token ID | 705 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
acs | +0.039 |
Gravity | +0.038 |
assic | +0.038 |
ikhail | +0.038 |
Tokens | +0.037 |
Neg logit contributions | |
confir | -0.029 |
contro | -0.027 |
discharge | -0.026 |
corpor | -0.026 |
Jude | -0.026 |
Token | p |
Token ID | 79 |
Feature activation | +5.40e-01 |
Pos logit contributions | |
confir | +0.982 |
contro | +0.935 |
discharge | +0.910 |
corpor | +0.899 |
Jude | +0.896 |
Neg logit contributions | |
acs | -1.140 |
Gravity | -1.115 |
assic | -1.109 |
ikhail | -1.107 |
Tokens | -1.085 |
Token | reg |
Token ID | 2301 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
acs | +1.000 |
Gravity | +0.974 |
assic | +0.973 |
ikhail | +0.969 |
Tokens | +0.947 |
Neg logit contributions | |
confir | -0.917 |
contro | -0.872 |
discharge | -0.847 |
corpor | -0.837 |
Jude | -0.836 |
Token | nant |
Token ID | 22057 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
confir | +0.165 |
contro | +0.158 |
discharge | +0.156 |
corpor | +0.155 |
Jude | +0.154 |
Neg logit contributions | |
acs | -0.180 |
Gravity | -0.178 |
ikhail | -0.176 |
Tokens | -0.174 |
assic | -0.173 |
Token | ' |
Token ID | 6 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
confir | +1.935 |
contro | +1.856 |
discharge | +1.825 |
corpor | +1.794 |
Jude | +1.787 |
Neg logit contributions | |
acs | -2.263 |
ikhail | -2.195 |
assic | -2.191 |
Gravity | -2.190 |
Tokens | -2.159 |
Token | T |
Token ID | 309 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
confir | +0.712 |
contro | +0.681 |
discharge | +0.665 |
corpor | +0.657 |
Jude | +0.655 |
Neg logit contributions | |
acs | -0.756 |
Gravity | -0.735 |
assic | -0.733 |
ikhail | -0.732 |
Tokens | -0.719 |
Token | . |
Token ID | 13 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
confir | +0.099 |
contro | +0.095 |
discharge | +0.092 |
corpor | +0.091 |
Jude | +0.090 |
Neg logit contributions | |
acs | -0.138 |
Gravity | -0.136 |
ikhail | -0.134 |
assic | -0.134 |
Tokens | -0.133 |
Token | re |
Token ID | 302 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
acs | +0.290 |
assic | +0.284 |
Gravity | +0.284 |
ikhail | +0.283 |
Tokens | +0.276 |
Neg logit contributions | |
confir | -0.195 |
contro | -0.184 |
discharge | -0.178 |
corpor | -0.175 |
Jude | -0.175 |
Token | x |
Token ID | 87 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
acs | +0.106 |
assic | +0.103 |
Gravity | +0.103 |
ikhail | +0.103 |
Tokens | +0.100 |
Neg logit contributions | |
confir | -0.103 |
contro | -0.097 |
discharge | -0.095 |
corpor | -0.094 |
Jude | -0.094 |
Token | 01 |
Token ID | 486 |
Feature activation | -9.33e-02 |
Pos logit contributions | |
confir | +0.281 |
contro | +0.271 |
discharge | +0.264 |
Jude | +0.262 |
corpor | +0.261 |
Neg logit contributions | |
acs | -0.320 |
assic | -0.314 |
Gravity | -0.313 |
ikhail | -0.312 |
Tokens | -0.304 |
Token | . |
Token ID | 13 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
confir | +0.133 |
contro | +0.127 |
discharge | +0.123 |
corpor | +0.121 |
Jude | +0.121 |
Neg logit contributions | |
acs | -0.196 |
Gravity | -0.192 |
assic | -0.192 |
ikhail | -0.191 |
Tokens | -0.187 |
Token | 29 |
Token ID | 1959 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
confir | +0.358 |
contro | +0.345 |
discharge | +0.336 |
Jude | +0.333 |
corpor | +0.332 |
Neg logit contributions | |
acs | -0.428 |
assic | -0.419 |
Gravity | -0.418 |
ikhail | -0.417 |
Tokens | -0.406 |
Token | "] |
Token ID | 8973 |
Feature activation | -6.04e-02 |
Pos logit contributions | |
confir | +0.063 |
contro | +0.062 |
discharge | +0.060 |
Jude | +0.060 |
corpor | +0.060 |
Neg logit contributions | |
acs | -0.030 |
assic | -0.029 |
Gravity | -0.029 |
ikhail | -0.029 |
Tokens | -0.028 |
Token | [ |
Token ID | 685 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
confir | +0.120 |
contro | +0.117 |
discharge | +0.114 |
Jude | +0.114 |
corpor | +0.113 |
Neg logit contributions | |
acs | -0.092 |
assic | -0.090 |
Gravity | -0.089 |
ikhail | -0.089 |
Tokens | -0.086 |
Token | Round |
Token ID | 22685 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Gravity | +1.508 |
acs | +1.505 |
Tokens | +1.496 |
ikhail | +1.482 |
Bro | +1.475 |
Neg logit contributions | |
confir | -0.829 |
contro | -0.750 |
discharge | -0.732 |
cumulative | -0.696 |
corpor | -0.685 |
Token | " |
Token ID | 366 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
confir | +1.593 |
contro | +1.524 |
discharge | +1.477 |
Jude | +1.471 |
corpor | +1.467 |
Neg logit contributions | |
acs | -2.596 |
Gravity | -2.510 |
assic | -2.505 |
ikhail | -2.502 |
Tokens | -2.444 |
Token | 11 |
Token ID | 1157 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
acs | +0.296 |
Gravity | +0.290 |
assic | +0.288 |
ikhail | +0.288 |
Tokens | +0.282 |
Neg logit contributions | |
confir | -0.257 |
contro | -0.244 |
discharge | -0.238 |
corpor | -0.235 |
Jude | -0.234 |
Token | "] |
Token ID | 8973 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
acs | +0.094 |
Gravity | +0.092 |
assic | +0.092 |
ikhail | +0.091 |
Tokens | +0.089 |
Neg logit contributions | |
confir | -0.101 |
contro | -0.097 |
discharge | -0.095 |
corpor | -0.094 |
Jude | -0.094 |
Token | [ |
Token ID | 685 |
Feature activation | +6.99e-01 |
Pos logit contributions | |
acs | +0.077 |
Gravity | +0.075 |
assic | +0.074 |
ikhail | +0.074 |
Tokens | +0.072 |
Neg logit contributions | |
confir | -0.111 |
contro | -0.106 |
discharge | -0.104 |
corpor | -0.103 |
Jude | -0.102 |
Token | White |
Token ID | 12256 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
acs | +1.546 |
Gravity | +1.526 |
ikhail | +1.512 |
assic | +1.506 |
Tokens | +1.501 |
Neg logit contributions | |
confir | -0.931 |
contro | -0.870 |
discharge | -0.843 |
corpor | -0.815 |
Jude | -0.813 |
Token | her |
Token ID | 607 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
confir | +0.155 |
contro | +0.146 |
discharge | +0.142 |
corpor | +0.140 |
Jude | +0.140 |
Neg logit contributions | |
acs | -0.201 |
Gravity | -0.197 |
assic | -0.195 |
ikhail | -0.195 |
Tokens | -0.191 |
Token | departure |
Token ID | 12928 |
Feature activation | -3.37e-01 |
Pos logit contributions | |
confir | +0.414 |
contro | +0.395 |
discharge | +0.385 |
corpor | +0.380 |
Jude | +0.378 |
Neg logit contributions | |
acs | -0.488 |
Gravity | -0.477 |
assic | -0.476 |
ikhail | -0.474 |
Tokens | -0.465 |
Token | time |
Token ID | 640 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
confir | +0.515 |
contro | +0.492 |
discharge | +0.478 |
corpor | +0.472 |
Jude | +0.469 |
Neg logit contributions | |
acs | -0.678 |
Gravity | -0.661 |
assic | -0.661 |
ikhail | -0.659 |
Tokens | -0.645 |
Token | carefully |
Token ID | 7773 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
acs | +0.044 |
Gravity | +0.043 |
assic | +0.043 |
ikhail | +0.043 |
Tokens | +0.042 |
Neg logit contributions | |
confir | -0.040 |
contro | -0.038 |
discharge | -0.037 |
Jude | -0.036 |
corpor | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
confir | +0.123 |
contro | +0.116 |
discharge | +0.113 |
corpor | +0.111 |
Jude | +0.111 |
Neg logit contributions | |
acs | -0.165 |
Gravity | -0.162 |
assic | -0.161 |
ikhail | -0.160 |
Tokens | -0.157 |
Token | Either |
Token ID | 15467 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
confir | +0.030 |
contro | +0.028 |
discharge | +0.027 |
corpor | +0.027 |
cumulative | +0.027 |
Neg logit contributions | |
acs | -0.036 |
Gravity | -0.035 |
ikhail | -0.035 |
assic | -0.035 |
Tokens | -0.034 |
Token | she |
Token ID | 673 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
confir | +1.638 |
contro | +1.551 |
discharge | +1.514 |
corpor | +1.486 |
Jude | +1.480 |
Neg logit contributions | |
acs | -2.511 |
assic | -2.452 |
Gravity | -2.452 |
ikhail | -2.448 |
Tokens | -2.404 |
Token | gets |
Token ID | 3011 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
confir | +0.405 |
contro | +0.384 |
discharge | +0.374 |
Jude | +0.370 |
corpor | +0.370 |
Neg logit contributions | |
acs | -0.439 |
Gravity | -0.428 |
assic | -0.425 |
ikhail | -0.423 |
Tokens | -0.415 |
Token | on |
Token ID | 319 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
confir | +0.357 |
contro | +0.339 |
discharge | +0.330 |
corpor | +0.326 |
Jude | +0.325 |
Neg logit contributions | |
acs | -0.410 |
Gravity | -0.401 |
assic | -0.399 |
ikhail | -0.398 |
Tokens | -0.391 |
Token | the |
Token ID | 262 |
Feature activation | -9.06e-02 |
Pos logit contributions | |
confir | +0.192 |
contro | +0.182 |
discharge | +0.177 |
corpor | +0.175 |
Jude | +0.174 |
Neg logit contributions | |
acs | -0.234 |
Gravity | -0.229 |
assic | -0.227 |
ikhail | -0.227 |
Tokens | -0.223 |
Token | London |
Token ID | 3576 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
confir | +0.141 |
contro | +0.134 |
discharge | +0.130 |
corpor | +0.129 |
Jude | +0.128 |
Neg logit contributions | |
acs | -0.179 |
Gravity | -0.176 |
assic | -0.175 |
ikhail | -0.174 |
Tokens | -0.171 |
Token | Ay |
Token ID | 13709 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
confir | +0.065 |
contro | +0.063 |
discharge | +0.061 |
corpor | +0.060 |
Jude | +0.060 |
Neg logit contributions | |
acs | -0.063 |
ikhail | -0.061 |
Gravity | -0.061 |
assic | -0.060 |
Tokens | -0.059 |
Token | at |
Token ID | 265 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
confir | +0.325 |
contro | +0.314 |
discharge | +0.306 |
corpor | +0.303 |
Jude | +0.303 |
Neg logit contributions | |
acs | -0.273 |
Gravity | -0.270 |
assic | -0.266 |
ikhail | -0.264 |
Tokens | -0.260 |
Token | ollah |
Token ID | 17999 |
Feature activation | -5.44e-01 |
Pos logit contributions | |
confir | +0.178 |
contro | +0.166 |
discharge | +0.161 |
corpor | +0.159 |
Jude | +0.158 |
Neg logit contributions | |
acs | -0.286 |
Gravity | -0.280 |
assic | -0.279 |
ikhail | -0.278 |
Tokens | -0.273 |
Token | Ali |
Token ID | 12104 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
confir | +0.522 |
contro | +0.479 |
discharge | +0.458 |
Jude | +0.454 |
corpor | +0.450 |
Neg logit contributions | |
acs | -1.402 |
Gravity | -1.383 |
assic | -1.382 |
ikhail | -1.371 |
Tokens | -1.358 |
Token | Kham |
Token ID | 45159 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
confir | +0.709 |
contro | +0.669 |
discharge | +0.652 |
corpor | +0.644 |
cumulative | +0.640 |
Neg logit contributions | |
acs | -0.839 |
Gravity | -0.819 |
assic | -0.814 |
ikhail | -0.813 |
Tokens | -0.794 |
Token | enei |
Token ID | 46009 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
confir | +0.883 |
contro | +0.851 |
discharge | +0.830 |
corpor | +0.822 |
Jude | +0.820 |
Neg logit contributions | |
acs | -0.786 |
Gravity | -0.769 |
assic | -0.763 |
ikhail | -0.760 |
Tokens | -0.745 |
Token | , |
Token ID | 11 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
confir | +1.655 |
contro | +1.561 |
discharge | +1.496 |
corpor | +1.490 |
Jude | +1.486 |
Neg logit contributions | |
acs | -2.441 |
Gravity | -2.414 |
assic | -2.402 |
ikhail | -2.394 |
Tokens | -2.349 |
Token | warned |
Token ID | 7728 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
acs | +0.047 |
Gravity | +0.046 |
assic | +0.046 |
ikhail | +0.046 |
Tokens | +0.045 |
Neg logit contributions | |
confir | -0.040 |
contro | -0.038 |
discharge | -0.037 |
corpor | -0.037 |
Jude | -0.037 |
Token | that |
Token ID | 326 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
confir | +0.584 |
contro | +0.555 |
discharge | +0.541 |
corpor | +0.533 |
Jude | +0.531 |
Neg logit contributions | |
acs | -0.697 |
Gravity | -0.682 |
assic | -0.679 |
ikhail | -0.677 |
Tokens | -0.664 |
Token | the |
Token ID | 262 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
confir | +0.228 |
contro | +0.217 |
discharge | +0.211 |
corpor | +0.208 |
Jude | +0.207 |
Neg logit contributions | |
acs | -0.263 |
Gravity | -0.258 |
assic | -0.255 |
ikhail | -0.255 |
Tokens | -0.250 |
Token | day |
Token ID | 1110 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
confir | +0.162 |
contro | +0.154 |
discharge | +0.150 |
corpor | +0.148 |
Jude | +0.147 |
Neg logit contributions | |
acs | -0.173 |
Gravity | -0.170 |
ikhail | -0.168 |
assic | -0.168 |
Tokens | -0.164 |
Token | Lew |
Token ID | 8260 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
confir | +0.012 |
contro | +0.012 |
discharge | +0.011 |
corpor | +0.011 |
Jude | +0.011 |
Neg logit contributions | |
acs | -0.012 |
Gravity | -0.011 |
assic | -0.011 |
ikhail | -0.011 |
Tokens | -0.011 |
Token | said |
Token ID | 531 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
acs | +1.161 |
Gravity | +1.132 |
assic | +1.122 |
ikhail | +1.118 |
Tokens | +1.095 |
Neg logit contributions | |
confir | -1.225 |
contro | -1.172 |
discharge | -1.145 |
corpor | -1.129 |
Jude | -1.127 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
acs | +0.270 |
Gravity | +0.265 |
assic | +0.261 |
ikhail | +0.261 |
Tokens | +0.257 |
Neg logit contributions | |
confir | -0.198 |
contro | -0.186 |
discharge | -0.181 |
corpor | -0.178 |
Jude | -0.177 |
Token |
|
Token ID | 198 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
confir | +0.026 |
contro | +0.024 |
discharge | +0.024 |
Jude | +0.023 |
corpor | +0.023 |
Neg logit contributions | |
acs | -0.042 |
Gravity | -0.041 |
assic | -0.041 |
ikhail | -0.041 |
Tokens | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
confir | +0.086 |
contro | +0.077 |
discharge | +0.075 |
corpor | +0.073 |
cumulative | +0.073 |
Neg logit contributions | |
acs | -0.145 |
Gravity | -0.143 |
ikhail | -0.141 |
assic | -0.141 |
Tokens | -0.140 |
Token | Neither |
Token ID | 27270 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
confir | +0.173 |
contro | +0.168 |
discharge | +0.164 |
Jude | +0.162 |
corpor | +0.162 |
Neg logit contributions | |
acs | -0.163 |
Gravity | -0.160 |
assic | -0.159 |
ikhail | -0.159 |
Tokens | -0.153 |
Token | action |
Token ID | 2223 |
Feature activation | +9.50e-02 |
Pos logit contributions | |
confir | +1.884 |
contro | +1.805 |
discharge | +1.766 |
corpor | +1.738 |
cumulative | +1.730 |
Neg logit contributions | |
acs | -2.179 |
assic | -2.139 |
ikhail | -2.129 |
Gravity | -2.115 |
Tokens | -2.094 |
Token | will |
Token ID | 481 |
Feature activation | +4.94e-03 |
Pos logit contributions | |
acs | +0.180 |
Gravity | +0.177 |
assic | +0.175 |
ikhail | +0.175 |
Tokens | +0.171 |
Neg logit contributions | |
confir | -0.156 |
contro | -0.148 |
discharge | -0.144 |
corpor | -0.143 |
Jude | -0.142 |
Token | extend |
Token ID | 9117 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
acs | +0.010 |
Gravity | +0.009 |
assic | +0.009 |
ikhail | +0.009 |
Tokens | +0.009 |
Neg logit contributions | |
confir | -0.008 |
contro | -0.007 |
discharge | -0.007 |
Jude | -0.007 |
corpor | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -6.32e-02 |
Pos logit contributions | |
confir | +0.193 |
contro | +0.184 |
discharge | +0.179 |
corpor | +0.176 |
Jude | +0.176 |
Neg logit contributions | |
acs | -0.244 |
Gravity | -0.238 |
assic | -0.238 |
ikhail | -0.237 |
Tokens | -0.232 |
Token | Oct |
Token ID | 2556 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
confir | +0.106 |
contro | +0.100 |
discharge | +0.097 |
corpor | +0.096 |
Jude | +0.096 |
Neg logit contributions | |
acs | -0.118 |
Gravity | -0.116 |
assic | -0.114 |
ikhail | -0.114 |
Tokens | -0.112 |
Token | in |
Token ID | 287 |
Feature activation | -2.83e-03 |
Pos logit contributions | |
acs | +0.349 |
Gravity | +0.342 |
assic | +0.339 |
ikhail | +0.338 |
Tokens | +0.334 |
Neg logit contributions | |
confir | -0.281 |
contro | -0.266 |
discharge | -0.258 |
Jude | -0.254 |
corpor | -0.254 |
Token | both |
Token ID | 1111 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
confir | +0.004 |
contro | +0.004 |
discharge | +0.004 |
corpor | +0.004 |
Jude | +0.004 |
Neg logit contributions | |
acs | -0.006 |
assic | -0.005 |
Gravity | -0.005 |
ikhail | -0.005 |
Tokens | -0.005 |
Token | official |
Token ID | 1743 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
confir | +1.257 |
contro | +1.211 |
discharge | +1.184 |
corpor | +1.179 |
Jude | +1.172 |
Neg logit contributions | |
acs | -1.018 |
assic | -0.993 |
Gravity | -0.989 |
ikhail | -0.987 |
Tokens | -0.961 |
Token | languages |
Token ID | 8950 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
confir | +0.086 |
contro | +0.084 |
discharge | +0.082 |
corpor | +0.082 |
Jude | +0.081 |
Neg logit contributions | |
acs | -0.046 |
assic | -0.045 |
Gravity | -0.045 |
ikhail | -0.044 |
Tokens | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -8.17e-02 |
Pos logit contributions | |
acs | +0.085 |
Gravity | +0.083 |
assic | +0.082 |
ikhail | +0.082 |
Tokens | +0.081 |
Neg logit contributions | |
confir | -0.068 |
contro | -0.064 |
discharge | -0.062 |
corpor | -0.062 |
Jude | -0.061 |
Token | Neither |
Token ID | 16126 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
confir | +0.136 |
contro | +0.129 |
discharge | +0.126 |
corpor | +0.124 |
Jude | +0.124 |
Neg logit contributions | |
acs | -0.154 |
Gravity | -0.151 |
assic | -0.150 |
ikhail | -0.149 |
Tokens | -0.146 |
Token | language |
Token ID | 3303 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
confir | +1.904 |
contro | +1.825 |
discharge | +1.771 |
corpor | +1.767 |
Jude | +1.741 |
Neg logit contributions | |
acs | -2.114 |
assic | -2.072 |
ikhail | -2.070 |
Gravity | -2.061 |
Tokens | -2.022 |
Token | takes |
Token ID | 2753 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
confir | +0.245 |
contro | +0.234 |
discharge | +0.228 |
corpor | +0.225 |
Jude | +0.225 |
Neg logit contributions | |
acs | -0.270 |
Gravity | -0.264 |
assic | -0.263 |
ikhail | -0.262 |
Tokens | -0.257 |
Token | precedence |
Token ID | 38177 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
confir | +0.526 |
contro | +0.499 |
discharge | +0.486 |
Jude | +0.479 |
corpor | +0.479 |
Neg logit contributions | |
acs | -0.590 |
Gravity | -0.577 |
assic | -0.573 |
ikhail | -0.570 |
Tokens | -0.560 |
Token | over |
Token ID | 625 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
acs | +0.126 |
Gravity | +0.123 |
assic | +0.122 |
ikhail | +0.122 |
Tokens | +0.120 |
Neg logit contributions | |
confir | -0.095 |
contro | -0.090 |
discharge | -0.088 |
corpor | -0.086 |
Jude | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
acs | +0.288 |
Gravity | +0.282 |
assic | +0.280 |
ikhail | +0.280 |
Tokens | +0.274 |
Neg logit contributions | |
confir | -0.256 |
contro | -0.243 |
discharge | -0.237 |
corpor | -0.234 |
Jude | -0.233 |
Token | Pupp |
Token ID | 20926 |
Feature activation | -8.53e-01 |
Pos logit contributions | |
acs | +0.097 |
Gravity | +0.095 |
assic | +0.094 |
ikhail | +0.094 |
Tokens | +0.092 |
Neg logit contributions | |
confir | -0.093 |
contro | -0.089 |
discharge | -0.086 |
corpor | -0.085 |
Jude | -0.085 |
Token | ies |
Token ID | 444 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
confir | +1.548 |
contro | +1.516 |
discharge | +1.478 |
corpor | +1.464 |
Jude | +1.460 |
Neg logit contributions | |
acs | -1.406 |
Gravity | -1.397 |
assic | -1.382 |
ikhail | -1.378 |
Tokens | -1.347 |
Token | stole |
Token ID | 18240 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
acs | +0.124 |
Gravity | +0.121 |
assic | +0.120 |
ikhail | +0.120 |
Tokens | +0.118 |
Neg logit contributions | |
confir | -0.100 |
contro | -0.095 |
discharge | -0.093 |
corpor | -0.091 |
Jude | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
confir | +0.321 |
contro | +0.306 |
discharge | +0.297 |
corpor | +0.293 |
Jude | +0.293 |
Neg logit contributions | |
acs | -0.383 |
Gravity | -0.374 |
assic | -0.373 |
ikhail | -0.372 |
Tokens | -0.365 |
Token | Sad |
Token ID | 14668 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
acs | +0.331 |
Gravity | +0.325 |
assic | +0.322 |
ikhail | +0.321 |
Tokens | +0.315 |
Neg logit contributions | |
confir | -0.269 |
contro | -0.254 |
discharge | -0.247 |
corpor | -0.244 |
Jude | -0.243 |
Token | Pupp |
Token ID | 20926 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
acs | +0.666 |
Gravity | +0.651 |
assic | +0.647 |
ikhail | +0.646 |
Tokens | +0.632 |
Neg logit contributions | |
confir | -0.664 |
contro | -0.635 |
discharge | -0.620 |
corpor | -0.613 |
Jude | -0.612 |
Token | ies |
Token ID | 444 |
Feature activation | +8.44e-02 |
Pos logit contributions | |
confir | +2.025 |
contro | +1.987 |
discharge | +1.939 |
Jude | +1.917 |
corpor | +1.916 |
Neg logit contributions | |
acs | -1.893 |
Gravity | -1.886 |
ikhail | -1.869 |
assic | -1.866 |
Tokens | -1.821 |
Token | brand |
Token ID | 4508 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
acs | +0.166 |
Gravity | +0.162 |
assic | +0.161 |
ikhail | +0.161 |
Tokens | +0.158 |
Neg logit contributions | |
confir | -0.133 |
contro | -0.126 |
discharge | -0.123 |
corpor | -0.121 |
Jude | -0.121 |
Token | — |
Token ID | 960 |
Feature activation | -9.09e-02 |
Pos logit contributions | |
acs | +0.820 |
Gravity | +0.804 |
assic | +0.794 |
ikhail | +0.792 |
Tokens | +0.783 |
Neg logit contributions | |
confir | -0.667 |
contro | -0.628 |
discharge | -0.613 |
Jude | -0.603 |
corpor | -0.602 |
Token | as |
Token ID | 292 |
Feature activation | +7.33e-02 |
Pos logit contributions | |
confir | +0.148 |
contro | +0.141 |
discharge | +0.137 |
corpor | +0.135 |
Jude | +0.135 |
Neg logit contributions | |
acs | -0.174 |
Gravity | -0.170 |
assic | -0.169 |
ikhail | -0.169 |
Tokens | -0.166 |
Token | one |
Token ID | 530 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
acs | +0.140 |
Gravity | +0.137 |
assic | +0.136 |
ikhail | +0.135 |
Tokens | +0.133 |
Neg logit contributions | |
confir | -0.120 |
contro | -0.114 |
discharge | -0.111 |
corpor | -0.109 |
Jude | -0.109 |
Token | between |
Token ID | 1022 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
acs | +0.373 |
Gravity | +0.365 |
assic | +0.363 |
ikhail | +0.362 |
Tokens | +0.355 |
Neg logit contributions | |
confir | -0.318 |
contro | -0.303 |
discharge | -0.295 |
corpor | -0.291 |
Jude | -0.290 |
Token | sovereign |
Token ID | 18901 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
acs | +0.253 |
Gravity | +0.249 |
ikhail | +0.246 |
assic | +0.246 |
Tokens | +0.241 |
Neg logit contributions | |
confir | -0.208 |
contro | -0.195 |
discharge | -0.190 |
corpor | -0.187 |
cumulative | -0.186 |
Token | Israel |
Token ID | 2692 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
acs | +0.271 |
Gravity | +0.265 |
assic | +0.263 |
ikhail | +0.262 |
Tokens | +0.258 |
Neg logit contributions | |
confir | -0.236 |
contro | -0.224 |
discharge | -0.217 |
corpor | -0.214 |
cumulative | -0.213 |
Token | and |
Token ID | 290 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
confir | +0.548 |
contro | +0.522 |
discharge | +0.507 |
corpor | +0.501 |
Jude | +0.499 |
Neg logit contributions | |
acs | -0.705 |
Gravity | -0.692 |
assic | -0.690 |
ikhail | -0.689 |
Tokens | -0.675 |
Token | the |
Token ID | 262 |
Feature activation | +2.77e-03 |
Pos logit contributions | |
acs | +0.026 |
Gravity | +0.026 |
assic | +0.026 |
ikhail | +0.026 |
Tokens | +0.025 |
Neg logit contributions | |
confir | -0.023 |
contro | -0.022 |
discharge | -0.021 |
corpor | -0.021 |
cumulative | -0.021 |
Token | settlements |
Token ID | 18573 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
acs | +0.005 |
Gravity | +0.005 |
assic | +0.004 |
ikhail | +0.004 |
Tokens | +0.004 |
Neg logit contributions | |
confir | -0.005 |
contro | -0.005 |
discharge | -0.005 |
corpor | -0.005 |
cumulative | -0.005 |
Token | in |
Token ID | 287 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
confir | +1.727 |
contro | +1.642 |
discharge | +1.600 |
corpor | +1.575 |
Jude | +1.569 |
Neg logit contributions | |
acs | -2.259 |
Gravity | -2.223 |
ikhail | -2.220 |
assic | -2.214 |
Tokens | -2.176 |
Token | the |
Token ID | 262 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
confir | +0.022 |
contro | +0.021 |
discharge | +0.021 |
corpor | +0.020 |
cumulative | +0.020 |
Neg logit contributions | |
acs | -0.027 |
Gravity | -0.026 |
ikhail | -0.026 |
assic | -0.026 |
Tokens | -0.025 |
Token | territories |
Token ID | 16771 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
confir | +0.127 |
contro | +0.122 |
discharge | +0.118 |
corpor | +0.117 |
cumulative | +0.117 |
Neg logit contributions | |
acs | -0.104 |
Gravity | -0.102 |
assic | -0.100 |
ikhail | -0.100 |
Tokens | -0.098 |
Token | ended |
Token ID | 4444 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
acs | +0.113 |
Gravity | +0.111 |
assic | +0.110 |
ikhail | +0.110 |
Tokens | +0.108 |
Neg logit contributions | |
confir | -0.084 |
contro | -0.079 |
discharge | -0.077 |
corpor | -0.076 |
Jude | -0.075 |
Token | up |
Token ID | 510 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
acs | +0.958 |
Gravity | +0.935 |
assic | +0.926 |
ikhail | +0.926 |
Tokens | +0.906 |
Neg logit contributions | |
confir | -1.000 |
contro | -0.950 |
discharge | -0.930 |
corpor | -0.917 |
Jude | -0.915 |
Token | other |
Token ID | 584 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
acs | +0.003 |
Gravity | +0.003 |
assic | +0.003 |
ikhail | +0.003 |
Tokens | +0.002 |
Neg logit contributions | |
confir | -0.002 |
contro | -0.002 |
discharge | -0.002 |
corpor | -0.002 |
Jude | -0.002 |
Token | team |
Token ID | 1074 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
acs | +0.056 |
Gravity | +0.055 |
assic | +0.054 |
ikhail | +0.054 |
Tokens | +0.053 |
Neg logit contributions | |
confir | -0.031 |
contro | -0.029 |
discharge | -0.028 |
corpor | -0.028 |
Jude | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -9.55e-04 |
Pos logit contributions | |
acs | +0.639 |
Gravity | +0.627 |
assic | +0.620 |
ikhail | +0.618 |
Tokens | +0.607 |
Neg logit contributions | |
confir | -0.539 |
contro | -0.510 |
discharge | -0.497 |
corpor | -0.491 |
Jude | -0.490 |
Token | his |
Token ID | 465 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
confir | +0.002 |
contro | +0.001 |
discharge | +0.001 |
corpor | +0.001 |
Jude | +0.001 |
Neg logit contributions | |
acs | -0.002 |
Gravity | -0.002 |
assic | -0.002 |
ikhail | -0.002 |
Tokens | -0.002 |
Token | team |
Token ID | 1074 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
confir | +0.052 |
contro | +0.049 |
discharge | +0.048 |
corpor | +0.047 |
Jude | +0.047 |
Neg logit contributions | |
acs | -0.065 |
Gravity | -0.063 |
assic | -0.063 |
ikhail | -0.063 |
Tokens | -0.062 |
Token | fight |
Token ID | 15481 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
acs | +0.406 |
Gravity | +0.397 |
assic | +0.395 |
ikhail | +0.394 |
Tokens | +0.386 |
Neg logit contributions | |
confir | -0.356 |
contro | -0.339 |
discharge | -0.330 |
corpor | -0.326 |
Jude | -0.325 |
Token | tim |
Token ID | 4628 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
confir | +1.740 |
contro | +1.678 |
discharge | +1.641 |
corpor | +1.598 |
cumulative | +1.598 |
Neg logit contributions | |
acs | -2.236 |
ikhail | -2.189 |
assic | -2.176 |
Gravity | -2.171 |
Tokens | -2.133 |
Token | ings |
Token ID | 654 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
confir | +0.268 |
contro | +0.253 |
discharge | +0.245 |
corpor | +0.240 |
Jude | +0.239 |
Neg logit contributions | |
acs | -0.423 |
Gravity | -0.415 |
assic | -0.414 |
ikhail | -0.413 |
Tokens | -0.407 |
Token | are |
Token ID | 389 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
confir | +0.371 |
contro | +0.352 |
discharge | +0.342 |
corpor | +0.336 |
Jude | +0.335 |
Neg logit contributions | |
acs | -0.514 |
Gravity | -0.504 |
assic | -0.502 |
ikhail | -0.501 |
Tokens | -0.492 |
Token | going |
Token ID | 1016 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
confir | +0.183 |
contro | +0.175 |
discharge | +0.170 |
corpor | +0.168 |
Jude | +0.168 |
Neg logit contributions | |
acs | -0.191 |
Gravity | -0.186 |
assic | -0.185 |
ikhail | -0.185 |
Tokens | -0.181 |
Token | to |
Token ID | 284 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
confir | +0.338 |
contro | +0.322 |
discharge | +0.311 |
corpor | +0.306 |
Jude | +0.306 |
Neg logit contributions | |
acs | -0.464 |
Gravity | -0.454 |
assic | -0.453 |
ikhail | -0.452 |
Tokens | -0.445 |
Token | success |
Token ID | 1943 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
acs | +0.200 |
Gravity | +0.196 |
assic | +0.195 |
ikhail | +0.194 |
Tokens | +0.191 |
Neg logit contributions | |
confir | -0.154 |
contro | -0.146 |
discharge | -0.142 |
corpor | -0.140 |
Jude | -0.140 |
Token | ". |
Token ID | 1911 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
confir | +0.432 |
contro | +0.408 |
discharge | +0.394 |
corpor | +0.388 |
Jude | +0.386 |
Neg logit contributions | |
acs | -0.717 |
Gravity | -0.703 |
assic | -0.701 |
ikhail | -0.700 |
Tokens | -0.687 |
Token |
|
Token ID | 198 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
confir | +0.309 |
contro | +0.289 |
discharge | +0.278 |
corpor | +0.273 |
Jude | +0.272 |
Neg logit contributions | |
acs | -0.578 |
Gravity | -0.568 |
assic | -0.564 |
ikhail | -0.563 |
Tokens | -0.555 |
Token |
|
Token ID | 198 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
confir | +0.212 |
contro | +0.192 |
discharge | +0.185 |
corpor | +0.182 |
cumulative | +0.181 |
Neg logit contributions | |
acs | -0.364 |
Gravity | -0.357 |
assic | -0.353 |
ikhail | -0.353 |
Tokens | -0.350 |
Token | L |
Token ID | 43 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
confir | +0.282 |
contro | +0.267 |
discharge | +0.260 |
corpor | +0.256 |
Jude | +0.256 |
Neg logit contributions | |
acs | -0.320 |
Gravity | -0.313 |
assic | -0.311 |
ikhail | -0.310 |
Tokens | -0.305 |
Token | ever |
Token ID | 964 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
confir | +0.037 |
contro | +0.034 |
discharge | +0.032 |
Jude | +0.032 |
corpor | +0.032 |
Neg logit contributions | |
acs | -0.083 |
assic | -0.081 |
Gravity | -0.081 |
ikhail | -0.081 |
Tokens | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
confir | +1.858 |
contro | +1.774 |
discharge | +1.723 |
corpor | +1.702 |
cumulative | +1.689 |
Neg logit contributions | |
acs | -2.128 |
ikhail | -2.080 |
assic | -2.079 |
Gravity | -2.077 |
Tokens | -2.038 |
Token | like |
Token ID | 588 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
confir | +0.067 |
contro | +0.063 |
discharge | +0.062 |
Jude | +0.061 |
corpor | +0.061 |
Neg logit contributions | |
acs | -0.088 |
Gravity | -0.088 |
ikhail | -0.085 |
assic | -0.085 |
Mov | -0.084 |
Token | Gibbs |
Token ID | 41071 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
acs | +0.313 |
Gravity | +0.309 |
assic | +0.305 |
ikhail | +0.305 |
Tokens | +0.300 |
Neg logit contributions | |
confir | -0.171 |
contro | -0.159 |
discharge | -0.153 |
corpor | -0.151 |
Jude | -0.150 |
Token | , |
Token ID | 11 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
acs | +0.443 |
Gravity | +0.433 |
assic | +0.429 |
ikhail | +0.429 |
Tokens | +0.420 |
Neg logit contributions | |
confir | -0.375 |
contro | -0.355 |
discharge | -0.346 |
corpor | -0.342 |
Jude | -0.341 |
Token | was |
Token ID | 373 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
confir | +0.079 |
contro | +0.075 |
discharge | +0.074 |
Jude | +0.073 |
corpor | +0.073 |
Neg logit contributions | |
acs | -0.083 |
Gravity | -0.082 |
assic | -0.081 |
ikhail | -0.080 |
Tokens | -0.079 |
Token | " |
Token ID | 366 |
Feature activation | -2.21e-03 |
Pos logit contributions | |
acs | +0.072 |
Gravity | +0.070 |
assic | +0.070 |
ikhail | +0.070 |
Tokens | +0.069 |
Neg logit contributions | |
confir | -0.059 |
contro | -0.056 |
discharge | -0.055 |
corpor | -0.054 |
Jude | -0.054 |
Token | j |
Token ID | 73 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
confir | +0.004 |
contro | +0.003 |
discharge | +0.003 |
corpor | +0.003 |
Jude | +0.003 |
Neg logit contributions | |
acs | -0.004 |
Gravity | -0.004 |
assic | -0.004 |
ikhail | -0.004 |
Tokens | -0.004 |
Token | okes |
Token ID | 3369 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
acs | +0.134 |
Gravity | +0.131 |
assic | +0.130 |
ikhail | +0.130 |
Tokens | +0.127 |
Neg logit contributions | |
confir | -0.153 |
contro | -0.147 |
discharge | -0.144 |
corpor | -0.142 |
Jude | -0.142 |
Token | " |
Token ID | 1 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
acs | +0.524 |
Gravity | +0.513 |
assic | +0.511 |
ikhail | +0.509 |
Tokens | +0.500 |
Neg logit contributions | |
confir | -0.379 |
contro | -0.358 |
discharge | -0.348 |
corpor | -0.342 |
Jude | -0.341 |
Token | about |
Token ID | 546 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
confir | +0.237 |
contro | +0.225 |
discharge | +0.218 |
corpor | +0.215 |
Jude | +0.215 |
Neg logit contributions | |
acs | -0.316 |
Gravity | -0.310 |
assic | -0.308 |
ikhail | -0.307 |
Tokens | -0.302 |
Token | Internal |
Token ID | 18628 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
acs | +0.351 |
Gravity | +0.344 |
assic | +0.341 |
ikhail | +0.340 |
Tokens | +0.333 |
Neg logit contributions | |
confir | -0.331 |
contro | -0.314 |
discharge | -0.306 |
corpor | -0.302 |
Jude | -0.302 |
Token | Revenue |
Token ID | 20197 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
confir | +1.281 |
contro | +1.256 |
discharge | +1.255 |
Jude | +1.253 |
corpor | +1.198 |
Neg logit contributions | |
acs | -2.601 |
assic | -2.583 |
Tokens | -2.537 |
ikhail | -2.520 |
Gravity | -2.503 |
Token | Service |
Token ID | 4809 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
confir | +0.343 |
contro | +0.318 |
discharge | +0.308 |
corpor | +0.300 |
Jude | +0.298 |
Neg logit contributions | |
acs | -0.777 |
Gravity | -0.764 |
assic | -0.762 |
ikhail | -0.761 |
Tokens | -0.750 |
Token | audits |
Token ID | 41973 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
acs | +0.051 |
Gravity | +0.050 |
assic | +0.050 |
ikhail | +0.050 |
Tokens | +0.049 |
Neg logit contributions | |
confir | -0.040 |
contro | -0.038 |
discharge | -0.036 |
Jude | -0.036 |
corpor | -0.036 |
Token | to |
Token ID | 284 |
Feature activation | -4.78e-02 |
Pos logit contributions | |
confir | +0.574 |
contro | +0.544 |
discharge | +0.529 |
corpor | +0.521 |
Jude | +0.519 |
Neg logit contributions | |
acs | -0.790 |
Gravity | -0.775 |
assic | -0.771 |
ikhail | -0.770 |
Tokens | -0.756 |
Token | the |
Token ID | 262 |
Feature activation | -4.52e-03 |
Pos logit contributions | |
confir | +0.078 |
contro | +0.075 |
discharge | +0.073 |
corpor | +0.072 |
Jude | +0.072 |
Neg logit contributions | |
acs | -0.091 |
Gravity | -0.089 |
assic | -0.088 |
ikhail | -0.088 |
Tokens | -0.087 |
Token | is |
Token ID | 318 |
Feature activation | +4.61e-03 |
Pos logit contributions | |
confir | +0.438 |
contro | +0.418 |
discharge | +0.407 |
corpor | +0.402 |
Jude | +0.401 |
Neg logit contributions | |
acs | -0.465 |
Gravity | -0.453 |
assic | -0.452 |
ikhail | -0.451 |
Tokens | -0.442 |
Token | part |
Token ID | 636 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
acs | +0.009 |
Gravity | +0.009 |
assic | +0.009 |
ikhail | +0.009 |
Tokens | +0.009 |
Neg logit contributions | |
confir | -0.007 |
contro | -0.007 |
discharge | -0.007 |
corpor | -0.006 |
Jude | -0.006 |
Token | of |
Token ID | 286 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
acs | +0.217 |
Gravity | +0.212 |
assic | +0.211 |
ikhail | +0.211 |
Tokens | +0.207 |
Neg logit contributions | |
confir | -0.154 |
contro | -0.145 |
discharge | -0.141 |
corpor | -0.139 |
Jude | -0.138 |
Token | the |
Token ID | 262 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
confir | +0.138 |
contro | +0.131 |
discharge | +0.128 |
corpor | +0.126 |
Jude | +0.126 |
Neg logit contributions | |
acs | -0.154 |
Gravity | -0.150 |
assic | -0.150 |
ikhail | -0.150 |
Tokens | -0.147 |
Token | Virtual |
Token ID | 15595 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
confir | +0.095 |
contro | +0.091 |
discharge | +0.089 |
corpor | +0.088 |
Jude | +0.088 |
Neg logit contributions | |
acs | -0.088 |
Gravity | -0.086 |
assic | -0.085 |
ikhail | -0.085 |
Tokens | -0.083 |
Token | Console |
Token ID | 24371 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
confir | +1.023 |
contro | +0.986 |
Jude | +0.959 |
discharge | +0.954 |
cumulative | +0.951 |
Neg logit contributions | |
acs | -0.969 |
ikhail | -0.939 |
assic | -0.930 |
Gravity | -0.923 |
ochem | -0.910 |
Token | service |
Token ID | 2139 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
confir | +1.030 |
contro | +0.953 |
discharge | +0.902 |
corpor | +0.881 |
Jude | +0.878 |
Neg logit contributions | |
acs | -2.902 |
Gravity | -2.852 |
assic | -2.846 |
ikhail | -2.841 |
Tokens | -2.798 |
Token | , |
Token ID | 11 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
acs | +0.195 |
Gravity | +0.193 |
ikhail | +0.189 |
assic | +0.189 |
Tokens | +0.186 |
Neg logit contributions | |
confir | -0.168 |
contro | -0.158 |
discharge | -0.154 |
corpor | -0.152 |
Jude | -0.151 |
Token | which |
Token ID | 543 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
acs | +0.051 |
Gravity | +0.050 |
ikhail | +0.050 |
assic | +0.049 |
Tokens | +0.049 |
Neg logit contributions | |
confir | -0.023 |
contro | -0.020 |
discharge | -0.020 |
corpor | -0.020 |
Jude | -0.019 |
Token | brings |
Token ID | 6774 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
confir | +0.088 |
contro | +0.083 |
discharge | +0.081 |
corpor | +0.080 |
Jude | +0.080 |
Neg logit contributions | |
acs | -0.122 |
Gravity | -0.119 |
assic | -0.119 |
ikhail | -0.118 |
Tokens | -0.116 |
Token | you |
Token ID | 345 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
confir | +0.629 |
contro | +0.593 |
discharge | +0.581 |
corpor | +0.572 |
Jude | +0.571 |
Neg logit contributions | |
acs | -0.707 |
Gravity | -0.694 |
assic | -0.691 |
ikhail | -0.687 |
Tokens | -0.672 |
Token | a |
Token ID | 257 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
confir | +0.272 |
contro | +0.259 |
discharge | +0.252 |
corpor | +0.248 |
Jude | +0.248 |
Neg logit contributions | |
acs | -0.314 |
Gravity | -0.307 |
assic | -0.305 |
ikhail | -0.304 |
Tokens | -0.298 |
Token | super |
Token ID | 2208 |
Feature activation | +7.14e-01 |
Pos logit contributions | |
confir | +0.205 |
contro | +0.194 |
discharge | +0.189 |
corpor | +0.186 |
Jude | +0.186 |
Neg logit contributions | |
acs | -0.252 |
Gravity | -0.248 |
assic | -0.246 |
ikhail | -0.246 |
Tokens | -0.240 |
Token | - |
Token ID | 12 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
acs | +1.300 |
Gravity | +1.270 |
assic | +1.260 |
ikhail | +1.259 |
Tokens | +1.224 |
Neg logit contributions | |
confir | -1.228 |
contro | -1.165 |
discharge | -1.139 |
Jude | -1.123 |
corpor | -1.120 |
Token | p |
Token ID | 79 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
acs | +0.206 |
Gravity | +0.201 |
assic | +0.200 |
ikhail | +0.199 |
Tokens | +0.195 |
Neg logit contributions | |
confir | -0.190 |
contro | -0.181 |
discharge | -0.177 |
corpor | -0.174 |
Jude | -0.174 |
Token | reg |
Token ID | 2301 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
acs | +0.681 |
Gravity | +0.660 |
assic | +0.660 |
ikhail | +0.657 |
Tokens | +0.641 |
Neg logit contributions | |
confir | -0.680 |
contro | -0.648 |
discharge | -0.630 |
corpor | -0.623 |
Jude | -0.623 |
Token | nant |
Token ID | 22057 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
confir | +0.265 |
contro | +0.252 |
discharge | +0.247 |
corpor | +0.245 |
Jude | +0.244 |
Neg logit contributions | |
acs | -0.283 |
Gravity | -0.277 |
ikhail | -0.274 |
assic | -0.272 |
Tokens | -0.272 |
Token | Ann |
Token ID | 5506 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
confir | +1.933 |
contro | +1.866 |
discharge | +1.820 |
Jude | +1.801 |
corpor | +1.801 |
Neg logit contributions | |
acs | -1.984 |
assic | -1.932 |
ikhail | -1.928 |
Gravity | -1.915 |
Tokens | -1.891 |
Token | al |
Token ID | 282 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
acs | +0.383 |
Gravity | +0.377 |
assic | +0.376 |
ikhail | +0.376 |
Tokens | +0.370 |
Neg logit contributions | |
confir | -0.101 |
contro | -0.090 |
discharge | -0.084 |
corpor | -0.081 |
cumulative | -0.080 |
Token | ise |
Token ID | 786 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
confir | +0.572 |
contro | +0.539 |
discharge | +0.522 |
corpor | +0.517 |
cumulative | +0.514 |
Neg logit contributions | |
acs | -0.724 |
Gravity | -0.708 |
assic | -0.706 |
ikhail | -0.705 |
Tokens | -0.686 |
Token | — |
Token ID | 851 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
confir | +0.488 |
contro | +0.464 |
discharge | +0.451 |
corpor | +0.445 |
Jude | +0.443 |
Neg logit contributions | |
acs | -0.624 |
Gravity | -0.611 |
assic | -0.609 |
ikhail | -0.607 |
Tokens | -0.596 |
Token | knowing |
Token ID | 6970 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
acs | +0.317 |
Gravity | +0.311 |
assic | +0.308 |
ikhail | +0.307 |
Tokens | +0.302 |
Neg logit contributions | |
confir | -0.268 |
contro | -0.254 |
discharge | -0.247 |
corpor | -0.244 |
Jude | -0.243 |
Token | choking |
Token ID | 39302 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
confir | +0.814 |
contro | +0.774 |
discharge | +0.757 |
corpor | +0.746 |
Jude | +0.744 |
Neg logit contributions | |
acs | -1.019 |
assic | -1.001 |
ikhail | -0.999 |
Gravity | -0.995 |
Tokens | -0.975 |
Token | Jennifer |
Token ID | 16348 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
confir | +0.761 |
contro | +0.731 |
discharge | +0.714 |
Jude | +0.709 |
corpor | +0.706 |
Neg logit contributions | |
acs | -0.770 |
assic | -0.747 |
ikhail | -0.747 |
Gravity | -0.746 |
Tokens | -0.732 |
Token | Lawrence |
Token ID | 13914 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
confir | +0.500 |
contro | +0.481 |
discharge | +0.471 |
corpor | +0.466 |
Jude | +0.465 |
Neg logit contributions | |
acs | -0.387 |
Gravity | -0.376 |
assic | -0.375 |
ikhail | -0.374 |
Tokens | -0.364 |
Token | , |
Token ID | 837 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
confir | +0.692 |
contro | +0.655 |
discharge | +0.634 |
corpor | +0.625 |
Jude | +0.622 |
Neg logit contributions | |
acs | -1.074 |
Gravity | -1.054 |
ikhail | -1.053 |
assic | -1.051 |
Tokens | -1.034 |
Token | was |
Token ID | 373 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
confir | +0.468 |
contro | +0.448 |
discharge | +0.436 |
corpor | +0.432 |
Jude | +0.430 |
Neg logit contributions | |
acs | -0.482 |
Gravity | -0.469 |
assic | -0.469 |
ikhail | -0.468 |
Tokens | -0.459 |
Token | round |
Token ID | 2835 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
confir | +0.368 |
contro | +0.352 |
discharge | +0.343 |
corpor | +0.339 |
Jude | +0.338 |
Neg logit contributions | |
acs | -0.383 |
Gravity | -0.375 |
assic | -0.373 |
ikhail | -0.373 |
Tokens | -0.365 |
Token | ly |
Token ID | 306 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
confir | +2.157 |
contro | +2.111 |
discharge | +2.079 |
corpor | +2.071 |
cumulative | +2.050 |
Neg logit contributions | |
acs | -1.684 |
ikhail | -1.640 |
assic | -1.623 |
Tokens | -1.611 |
Gravity | -1.610 |
Token | criticized |
Token ID | 12318 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
confir | +0.108 |
contro | +0.103 |
discharge | +0.100 |
corpor | +0.099 |
Jude | +0.099 |
Neg logit contributions | |
acs | -0.120 |
Gravity | -0.117 |
assic | -0.116 |
ikhail | -0.116 |
Tokens | -0.114 |
Token | for |
Token ID | 329 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
confir | +0.166 |
contro | +0.157 |
discharge | +0.153 |
corpor | +0.150 |
Jude | +0.150 |
Neg logit contributions | |
acs | -0.217 |
Gravity | -0.213 |
assic | -0.212 |
ikhail | -0.211 |
Tokens | -0.207 |
Token | showing |
Token ID | 4478 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
confir | +0.254 |
contro | +0.241 |
discharge | +0.234 |
corpor | +0.231 |
Jude | +0.230 |
Neg logit contributions | |
acs | -0.323 |
Gravity | -0.316 |
assic | -0.314 |
ikhail | -0.314 |
Tokens | -0.308 |
Token | excessive |
Token ID | 13181 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
acs | +0.185 |
assic | +0.180 |
Gravity | +0.180 |
ikhail | +0.180 |
Tokens | +0.176 |
Neg logit contributions | |
confir | -0.171 |
contro | -0.163 |
discharge | -0.159 |
Jude | -0.158 |
corpor | -0.157 |
Token | and |
Token ID | 290 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
confir | +0.131 |
contro | +0.124 |
discharge | +0.121 |
corpor | +0.119 |
Jude | +0.119 |
Neg logit contributions | |
acs | -0.159 |
Gravity | -0.156 |
assic | -0.155 |
ikhail | -0.155 |
Tokens | -0.152 |
Token | all |
Token ID | 477 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
confir | +0.088 |
contro | +0.084 |
discharge | +0.081 |
corpor | +0.080 |
Jude | +0.080 |
Neg logit contributions | |
acs | -0.115 |
Gravity | -0.112 |
assic | -0.112 |
ikhail | -0.112 |
Tokens | -0.110 |
Token | the |
Token ID | 262 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
confir | +0.087 |
contro | +0.083 |
discharge | +0.080 |
corpor | +0.079 |
Jude | +0.079 |
Neg logit contributions | |
acs | -0.117 |
Gravity | -0.115 |
assic | -0.114 |
ikhail | -0.114 |
Tokens | -0.112 |
Token | til |
Token ID | 21502 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
confir | +0.024 |
contro | +0.023 |
discharge | +0.022 |
corpor | +0.022 |
Jude | +0.022 |
Neg logit contributions | |
acs | -0.024 |
Gravity | -0.023 |
assic | -0.023 |
ikhail | -0.023 |
Tokens | -0.023 |
Token | ap |
Token ID | 499 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
confir | +0.862 |
contro | +0.831 |
discharge | +0.816 |
corpor | +0.797 |
cumulative | +0.793 |
Neg logit contributions | |
Gravity | -1.106 |
acs | -1.094 |
ikhail | -1.093 |
assic | -1.082 |
Tokens | -1.075 |
Token | ia |
Token ID | 544 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
acs | +0.593 |
Gravity | +0.584 |
assic | +0.579 |
ikhail | +0.578 |
Tokens | +0.569 |
Neg logit contributions | |
confir | -0.577 |
contro | -0.553 |
discharge | -0.540 |
corpor | -0.533 |
Jude | -0.531 |
Token | fil |
Token ID | 1226 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
acs | +1.554 |
Gravity | +1.516 |
assic | +1.504 |
ikhail | +1.500 |
Tokens | +1.471 |
Neg logit contributions | |
confir | -1.328 |
contro | -1.257 |
discharge | -1.219 |
Jude | -1.211 |
corpor | -1.205 |
Token | lets |
Token ID | 5289 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
acs | +0.055 |
Gravity | +0.054 |
assic | +0.054 |
ikhail | +0.054 |
Tokens | +0.053 |
Neg logit contributions | |
confir | -0.023 |
contro | -0.021 |
discharge | -0.021 |
corpor | -0.020 |
Jude | -0.020 |
Token | are |
Token ID | 389 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
acs | +0.835 |
Gravity | +0.819 |
assic | +0.808 |
ikhail | +0.806 |
Tokens | +0.790 |
Neg logit contributions | |
confir | -0.620 |
contro | -0.582 |
Jude | -0.560 |
discharge | -0.560 |
corpor | -0.556 |
Token | the |
Token ID | 262 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
confir | +0.017 |
contro | +0.016 |
discharge | +0.015 |
corpor | +0.015 |
Jude | +0.015 |
Neg logit contributions | |
acs | -0.020 |
Gravity | -0.019 |
assic | -0.019 |
ikhail | -0.019 |
Tokens | -0.019 |
Token | same |
Token ID | 976 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
acs | +0.114 |
Gravity | +0.111 |
assic | +0.111 |
ikhail | +0.110 |
Tokens | +0.108 |
Neg logit contributions | |
confir | -0.091 |
contro | -0.087 |
discharge | -0.084 |
corpor | -0.083 |
Jude | -0.083 |
Token | ings |
Token ID | 654 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
acs | +0.615 |
assic | +0.599 |
Gravity | +0.599 |
ikhail | +0.593 |
Tokens | +0.581 |
Neg logit contributions | |
confir | -0.500 |
contro | -0.474 |
discharge | -0.459 |
corpor | -0.453 |
cumulative | -0.451 |
Token | head |
Token ID | 2256 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
confir | +0.037 |
contro | +0.035 |
discharge | +0.034 |
corpor | +0.034 |
Jude | +0.034 |
Neg logit contributions | |
acs | -0.032 |
assic | -0.031 |
ikhail | -0.031 |
Gravity | -0.031 |
Tokens | -0.030 |
Token | [ |
Token ID | 685 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
confir | +0.432 |
contro | +0.413 |
discharge | +0.401 |
corpor | +0.397 |
Jude | +0.396 |
Neg logit contributions | |
acs | -0.488 |
assic | -0.477 |
ikhail | -0.476 |
Gravity | -0.476 |
Tokens | -0.467 |
Token | who |
Token ID | 8727 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
acs | +1.094 |
Gravity | +1.060 |
assic | +1.056 |
ikhail | +1.056 |
Tokens | +1.039 |
Neg logit contributions | |
confir | -0.963 |
contro | -0.910 |
discharge | -0.889 |
cumulative | -0.871 |
Jude | -0.871 |
Token | gave |
Token ID | 2921 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
acs | +0.125 |
Gravity | +0.121 |
assic | +0.121 |
ikhail | +0.121 |
Tokens | +0.118 |
Neg logit contributions | |
confir | -0.162 |
contro | -0.156 |
discharge | -0.153 |
corpor | -0.152 |
Jude | -0.151 |
Token | Lear |
Token ID | 8010 |
Feature activation | +7.18e-01 |
Pos logit contributions | |
acs | +0.162 |
Gravity | +0.158 |
assic | +0.158 |
ikhail | +0.157 |
Tokens | +0.154 |
Neg logit contributions | |
confir | -0.140 |
contro | -0.134 |
discharge | -0.130 |
corpor | -0.128 |
Jude | -0.128 |
Token | y |
Token ID | 88 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
acs | +1.312 |
Gravity | +1.283 |
assic | +1.273 |
ikhail | +1.272 |
Tokens | +1.242 |
Neg logit contributions | |
confir | -1.233 |
contro | -1.172 |
discharge | -1.145 |
corpor | -1.127 |
Jude | -1.125 |
Token | his |
Token ID | 465 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
confir | +0.313 |
contro | +0.300 |
discharge | +0.293 |
corpor | +0.289 |
Jude | +0.289 |
Neg logit contributions | |
acs | -0.347 |
assic | -0.338 |
Gravity | -0.338 |
ikhail | -0.337 |
Tokens | -0.332 |
Token | first |
Token ID | 717 |
Feature activation | -9.30e-03 |
Pos logit contributions | |
acs | +0.343 |
Gravity | +0.335 |
assic | +0.335 |
ikhail | +0.334 |
Tokens | +0.328 |
Neg logit contributions | |
confir | -0.278 |
contro | -0.265 |
discharge | -0.259 |
corpor | -0.255 |
Jude | -0.255 |
Token | LSD |
Token ID | 27483 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
confir | +0.016 |
contro | +0.015 |
discharge | +0.015 |
corpor | +0.015 |
Jude | +0.015 |
Neg logit contributions | |
acs | -0.017 |
Gravity | -0.016 |
assic | -0.016 |
ikhail | -0.016 |
Tokens | -0.016 |
Token | trip |
Token ID | 5296 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
acs | +0.278 |
Gravity | +0.271 |
assic | +0.271 |
ikhail | +0.269 |
Tokens | +0.264 |
Neg logit contributions | |
confir | -0.293 |
contro | -0.281 |
discharge | -0.275 |
corpor | -0.271 |
Jude | -0.270 |
Token | from |
Token ID | 422 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
confir | +0.782 |
contro | +0.743 |
discharge | +0.723 |
corpor | +0.710 |
Jude | +0.708 |
Neg logit contributions | |
acs | -1.061 |
Gravity | -1.039 |
assic | -1.035 |
ikhail | -1.032 |
Tokens | -1.014 |
Token | his |
Token ID | 465 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
confir | +0.032 |
contro | +0.030 |
discharge | +0.029 |
Jude | +0.029 |
cumulative | +0.029 |
Neg logit contributions | |
acs | -0.038 |
Gravity | -0.037 |
assic | -0.036 |
ikhail | -0.036 |
Tokens | -0.036 |
Token | boss |
Token ID | 6478 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
acs | +0.245 |
Gravity | +0.240 |
assic | +0.237 |
ikhail | +0.236 |
Tokens | +0.232 |
Neg logit contributions | |
confir | -0.209 |
contro | -0.197 |
discharge | -0.191 |
Jude | -0.190 |
corpor | -0.189 |
Token | Hood |
Token ID | 17233 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
acs | +0.242 |
Gravity | +0.237 |
assic | +0.235 |
ikhail | +0.234 |
Tokens | +0.229 |
Neg logit contributions | |
confir | -0.207 |
contro | -0.197 |
discharge | -0.191 |
Jude | -0.188 |
corpor | -0.188 |
Token | a |
Token ID | 64 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
confir | +0.563 |
contro | +0.536 |
discharge | +0.522 |
corpor | +0.515 |
Jude | +0.514 |
Neg logit contributions | |
acs | -0.677 |
Gravity | -0.664 |
assic | -0.661 |
ikhail | -0.659 |
Tokens | -0.647 |
Token | despite |
Token ID | 3805 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
acs | +0.342 |
Gravity | +0.334 |
assic | +0.333 |
ikhail | +0.332 |
Tokens | +0.325 |
Neg logit contributions | |
confir | -0.271 |
contro | -0.257 |
discharge | -0.250 |
Jude | -0.246 |
corpor | -0.246 |
Token | the |
Token ID | 262 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
acs | +1.386 |
Gravity | +1.367 |
assic | +1.341 |
ikhail | +1.334 |
Tokens | +1.313 |
Neg logit contributions | |
confir | -1.249 |
contro | -1.181 |
discharge | -1.145 |
Jude | -1.137 |
corpor | -1.129 |
Token | fact |
Token ID | 1109 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
acs | +0.111 |
Gravity | +0.109 |
assic | +0.107 |
ikhail | +0.107 |
Tokens | +0.105 |
Neg logit contributions | |
confir | -0.101 |
contro | -0.096 |
discharge | -0.093 |
Jude | -0.092 |
corpor | -0.092 |
Token | that |
Token ID | 326 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
confir | +0.305 |
contro | +0.290 |
discharge | +0.282 |
corpor | +0.278 |
Jude | +0.277 |
Neg logit contributions | |
acs | -0.371 |
Gravity | -0.363 |
assic | -0.361 |
ikhail | -0.360 |
Tokens | -0.353 |
Token | it |
Token ID | 340 |
Feature activation | +2.13e-02 |
Pos logit contributions | |
acs | +0.122 |
Gravity | +0.119 |
assic | +0.118 |
ikhail | +0.117 |
Tokens | +0.115 |
Neg logit contributions | |
confir | -0.101 |
contro | -0.096 |
discharge | -0.093 |
Jude | -0.092 |
corpor | -0.091 |
Token | is |
Token ID | 318 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
acs | +0.037 |
Gravity | +0.036 |
assic | +0.036 |
ikhail | +0.036 |
Tokens | +0.035 |
Neg logit contributions | |
confir | -0.038 |
contro | -0.036 |
discharge | -0.036 |
Jude | -0.035 |
corpor | -0.035 |
Token | Boston |
Token ID | 6182 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
acs | +0.043 |
assic | +0.042 |
Gravity | +0.042 |
ikhail | +0.042 |
Tokens | +0.041 |
Neg logit contributions | |
confir | -0.052 |
contro | -0.050 |
discharge | -0.049 |
Jude | -0.049 |
corpor | -0.049 |
Token | ( |
Token ID | 357 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
acs | +0.125 |
Gravity | +0.122 |
assic | +0.122 |
ikhail | +0.122 |
Tokens | +0.119 |
Neg logit contributions | |
confir | -0.103 |
contro | -0.098 |
discharge | -0.095 |
corpor | -0.094 |
Jude | -0.094 |
Token | pictured |
Token ID | 28852 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
confir | +0.307 |
contro | +0.292 |
discharge | +0.285 |
corpor | +0.281 |
Jude | +0.280 |
Neg logit contributions | |
acs | -0.353 |
Gravity | -0.345 |
assic | -0.343 |
ikhail | -0.342 |
Tokens | -0.336 |
Token | ), |
Token ID | 828 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
confir | +0.605 |
contro | +0.579 |
discharge | +0.563 |
corpor | +0.556 |
Jude | +0.554 |
Neg logit contributions | |
acs | -0.712 |
Gravity | -0.695 |
assic | -0.692 |
ikhail | -0.691 |
Tokens | -0.677 |
Token | the |
Token ID | 262 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
acs | +0.243 |
Gravity | +0.237 |
assic | +0.236 |
ikhail | +0.236 |
Tokens | +0.231 |
Neg logit contributions | |
confir | -0.212 |
contro | -0.202 |
discharge | -0.197 |
corpor | -0.195 |
Jude | -0.194 |
Token | Latin |
Token ID | 9133 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
confir | +0.061 |
contro | +0.059 |
discharge | +0.057 |
corpor | +0.057 |
Jude | +0.057 |
Neg logit contributions | |
acs | -0.056 |
Gravity | -0.054 |
assic | -0.054 |
ikhail | -0.054 |
Tokens | -0.053 |
Token | Academy |
Token ID | 8581 |
Feature activation | +7.51e-02 |
Pos logit contributions | |
acs | +1.558 |
Gravity | +1.528 |
assic | +1.510 |
ikhail | +1.498 |
Tokens | +1.464 |
Neg logit contributions | |
confir | -1.648 |
contro | -1.570 |
discharge | -1.533 |
corpor | -1.512 |
cumulative | -1.503 |
Token | , |
Token ID | 11 |
Feature activation | +9.85e-02 |
Pos logit contributions | |
acs | +0.147 |
Gravity | +0.144 |
assic | +0.143 |
ikhail | +0.143 |
Tokens | +0.140 |
Neg logit contributions | |
confir | -0.119 |
contro | -0.113 |
discharge | -0.110 |
corpor | -0.108 |
Jude | -0.108 |
Token | some |
Token ID | 617 |
Feature activation | -5.62e-02 |
Pos logit contributions | |
acs | +0.184 |
Gravity | +0.180 |
assic | +0.179 |
ikhail | +0.179 |
Tokens | +0.175 |
Neg logit contributions | |
confir | -0.164 |
contro | -0.157 |
discharge | -0.153 |
corpor | -0.151 |
Jude | -0.151 |
Token | areas |
Token ID | 3006 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
confir | +0.097 |
contro | +0.093 |
discharge | +0.091 |
corpor | +0.090 |
Jude | +0.089 |
Neg logit contributions | |
acs | -0.102 |
Gravity | -0.099 |
assic | -0.099 |
ikhail | -0.099 |
Tokens | -0.097 |
Token | in |
Token ID | 287 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
acs | +0.022 |
Gravity | +0.022 |
assic | +0.022 |
ikhail | +0.022 |
Tokens | +0.021 |
Neg logit contributions | |
confir | -0.016 |
contro | -0.015 |
discharge | -0.015 |
corpor | -0.015 |
Jude | -0.015 |
Token | pick |
Token ID | 2298 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
confir | +0.144 |
contro | +0.137 |
discharge | +0.133 |
corpor | +0.132 |
Jude | +0.131 |
Neg logit contributions | |
acs | -0.176 |
Gravity | -0.172 |
assic | -0.171 |
ikhail | -0.171 |
Tokens | -0.167 |
Token | in |
Token ID | 287 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
acs | +0.854 |
Gravity | +0.839 |
assic | +0.832 |
ikhail | +0.832 |
Tokens | +0.811 |
Neg logit contributions | |
confir | -0.636 |
contro | -0.596 |
discharge | -0.578 |
corpor | -0.573 |
Jude | -0.569 |
Token | the |
Token ID | 262 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
acs | +0.098 |
Gravity | +0.096 |
assic | +0.096 |
ikhail | +0.095 |
Tokens | +0.094 |
Neg logit contributions | |
confir | -0.073 |
contro | -0.069 |
discharge | -0.067 |
corpor | -0.066 |
Jude | -0.066 |
Token | 2015 |
Token ID | 1853 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
acs | +0.037 |
Gravity | +0.036 |
assic | +0.036 |
ikhail | +0.036 |
Tokens | +0.035 |
Neg logit contributions | |
confir | -0.037 |
contro | -0.035 |
discharge | -0.035 |
corpor | -0.034 |
Jude | -0.034 |
Token | NBA |
Token ID | 7403 |
Feature activation | +6.69e-02 |
Pos logit contributions | |
confir | +0.386 |
contro | +0.369 |
discharge | +0.359 |
corpor | +0.355 |
Jude | +0.354 |
Neg logit contributions | |
acs | -0.440 |
Gravity | -0.430 |
assic | -0.429 |
ikhail | -0.428 |
Tokens | -0.420 |
Token | draft |
Token ID | 4538 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
acs | +0.142 |
Gravity | +0.140 |
assic | +0.139 |
ikhail | +0.139 |
Tokens | +0.136 |
Neg logit contributions | |
confir | -0.095 |
contro | -0.089 |
discharge | -0.086 |
corpor | -0.085 |
Jude | -0.085 |
Token | by |
Token ID | 416 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
acs | +1.427 |
Gravity | +1.409 |
assic | +1.397 |
ikhail | +1.396 |
Tokens | +1.363 |
Neg logit contributions | |
confir | -1.272 |
contro | -1.198 |
discharge | -1.164 |
corpor | -1.157 |
Jude | -1.150 |
Token | the |
Token ID | 262 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
confir | +0.070 |
contro | +0.066 |
discharge | +0.064 |
corpor | +0.064 |
Jude | +0.063 |
Neg logit contributions | |
acs | -0.086 |
Gravity | -0.085 |
ikhail | -0.084 |
assic | -0.084 |
Tokens | -0.082 |
Token | Philadelphia |
Token ID | 8857 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
confir | +0.027 |
contro | +0.026 |
discharge | +0.025 |
corpor | +0.025 |
Jude | +0.025 |
Neg logit contributions | |
acs | -0.024 |
Gravity | -0.023 |
assic | -0.023 |
ikhail | -0.023 |
Tokens | -0.022 |
Token | 76 |
Token ID | 8684 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
acs | +0.046 |
Gravity | +0.045 |
ikhail | +0.044 |
assic | +0.044 |
Tokens | +0.044 |
Neg logit contributions | |
confir | -0.040 |
contro | -0.038 |
discharge | -0.037 |
corpor | -0.036 |
Jude | -0.036 |
Token | ers |
Token ID | 364 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
confir | +0.277 |
contro | +0.264 |
discharge | +0.258 |
corpor | +0.256 |
Jude | +0.255 |
Neg logit contributions | |
acs | -0.227 |
Gravity | -0.220 |
assic | -0.219 |
ikhail | -0.218 |
Tokens | -0.213 |
Token | video |
Token ID | 2008 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
acs | +0.802 |
Gravity | +0.793 |
assic | +0.781 |
ikhail | +0.777 |
Tokens | +0.766 |
Neg logit contributions | |
confir | -0.709 |
contro | -0.666 |
discharge | -0.651 |
Jude | -0.642 |
corpor | -0.640 |
Token | of |
Token ID | 286 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
confir | +0.546 |
contro | +0.520 |
discharge | +0.506 |
corpor | +0.499 |
Jude | +0.498 |
Neg logit contributions | |
acs | -0.636 |
Gravity | -0.622 |
assic | -0.619 |
ikhail | -0.617 |
Tokens | -0.605 |
Token | your |
Token ID | 534 |
Feature activation | +1.35e-02 |
Pos logit contributions | |
acs | +0.145 |
Gravity | +0.143 |
assic | +0.141 |
ikhail | +0.141 |
Tokens | +0.138 |
Neg logit contributions | |
confir | -0.139 |
contro | -0.132 |
discharge | -0.128 |
corpor | -0.127 |
Jude | -0.127 |
Token | owner |
Token ID | 4870 |
Feature activation | +9.31e-02 |
Pos logit contributions | |
acs | +0.026 |
Gravity | +0.025 |
assic | +0.025 |
ikhail | +0.025 |
Tokens | +0.025 |
Neg logit contributions | |
confir | -0.022 |
contro | -0.021 |
discharge | -0.020 |
Jude | -0.020 |
corpor | -0.020 |
Token | dancing |
Token ID | 15360 |
Feature activation | -7.23e-02 |
Pos logit contributions | |
acs | +0.172 |
Gravity | +0.168 |
assic | +0.167 |
ikhail | +0.166 |
Tokens | +0.164 |
Neg logit contributions | |
confir | -0.157 |
contro | -0.150 |
discharge | -0.146 |
corpor | -0.144 |
Jude | -0.144 |
Token | in |
Token ID | 287 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
confir | +0.118 |
contro | +0.112 |
discharge | +0.109 |
corpor | +0.108 |
Jude | +0.107 |
Neg logit contributions | |
acs | -0.138 |
Gravity | -0.135 |
assic | -0.134 |
ikhail | -0.134 |
Tokens | -0.131 |
Token | the |
Token ID | 262 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
acs | +0.108 |
Gravity | +0.106 |
assic | +0.105 |
ikhail | +0.105 |
Tokens | +0.103 |
Neg logit contributions | |
confir | -0.092 |
contro | -0.088 |
discharge | -0.085 |
corpor | -0.084 |
Jude | -0.084 |
Token | locker |
Token ID | 17355 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
confir | +0.067 |
contro | +0.064 |
discharge | +0.062 |
corpor | +0.061 |
Jude | +0.061 |
Neg logit contributions | |
acs | -0.079 |
Gravity | -0.077 |
assic | -0.077 |
ikhail | -0.076 |
Tokens | -0.075 |
Token | room |
Token ID | 2119 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
acs | +0.295 |
Gravity | +0.287 |
assic | +0.285 |
ikhail | +0.284 |
Tokens | +0.278 |
Neg logit contributions | |
confir | -0.270 |
contro | -0.257 |
discharge | -0.249 |
Jude | -0.247 |
corpor | -0.247 |
Token | , |
Token ID | 11 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
acs | +0.272 |
Gravity | +0.267 |
assic | +0.264 |
ikhail | +0.264 |
Tokens | +0.259 |
Neg logit contributions | |
confir | -0.222 |
contro | -0.210 |
discharge | -0.204 |
corpor | -0.201 |
Jude | -0.201 |
Token | because |
Token ID | 780 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
confir | +0.092 |
contro | +0.088 |
discharge | +0.085 |
corpor | +0.084 |
Jude | +0.084 |
Neg logit contributions | |
acs | -0.105 |
Gravity | -0.103 |
assic | -0.102 |
ikhail | -0.102 |
Tokens | -0.100 |
Token | � |
Token ID | 250 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
confir | +0.161 |
contro | +0.154 |
discharge | +0.151 |
corpor | +0.150 |
Jude | +0.149 |
Neg logit contributions | |
acs | -0.125 |
Gravity | -0.121 |
assic | -0.121 |
ikhail | -0.120 |
Tokens | -0.117 |
Token | What |
Token ID | 2061 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
acs | +0.165 |
Gravity | +0.160 |
assic | +0.160 |
ikhail | +0.160 |
Tokens | +0.156 |
Neg logit contributions | |
confir | -0.169 |
contro | -0.162 |
discharge | -0.158 |
Jude | -0.157 |
corpor | -0.157 |
Token | has |
Token ID | 468 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
acs | +0.484 |
Gravity | +0.474 |
assic | +0.471 |
ikhail | +0.470 |
Tokens | +0.460 |
Neg logit contributions | |
confir | -0.433 |
contro | -0.413 |
discharge | -0.402 |
corpor | -0.397 |
Jude | -0.395 |
Token | the |
Token ID | 262 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
acs | +0.230 |
Gravity | +0.226 |
assic | +0.224 |
ikhail | +0.223 |
Tokens | +0.219 |
Neg logit contributions | |
confir | -0.195 |
contro | -0.185 |
discharge | -0.181 |
corpor | -0.178 |
Jude | -0.177 |
Token | universe |
Token ID | 6881 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
acs | +0.287 |
Gravity | +0.281 |
assic | +0.279 |
ikhail | +0.278 |
Tokens | +0.272 |
Neg logit contributions | |
confir | -0.233 |
contro | -0.220 |
discharge | -0.215 |
corpor | -0.211 |
Jude | -0.210 |
Token | got |
Token ID | 1392 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
acs | +0.215 |
Gravity | +0.212 |
assic | +0.208 |
ikhail | +0.208 |
Tokens | +0.204 |
Neg logit contributions | |
confir | -0.181 |
contro | -0.171 |
discharge | -0.167 |
corpor | -0.164 |
Jude | -0.164 |
Token | to |
Token ID | 284 |
Feature activation | +6.44e-02 |
Pos logit contributions | |
acs | +0.440 |
Gravity | +0.432 |
assic | +0.429 |
ikhail | +0.428 |
Tokens | +0.421 |
Neg logit contributions | |
confir | -0.332 |
contro | -0.313 |
discharge | -0.305 |
corpor | -0.300 |
Jude | -0.300 |
Token | do |
Token ID | 466 |
Feature activation | +6.56e-02 |
Pos logit contributions | |
acs | +0.117 |
Gravity | +0.114 |
assic | +0.114 |
ikhail | +0.114 |
Tokens | +0.111 |
Neg logit contributions | |
confir | -0.111 |
contro | -0.106 |
discharge | -0.103 |
corpor | -0.102 |
Jude | -0.102 |
Token | with |
Token ID | 351 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
acs | +0.119 |
Gravity | +0.116 |
assic | +0.116 |
ikhail | +0.116 |
Tokens | +0.113 |
Neg logit contributions | |
confir | -0.113 |
contro | -0.108 |
discharge | -0.105 |
corpor | -0.105 |
Jude | -0.104 |
Token | it |
Token ID | 340 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
acs | +0.167 |
Gravity | +0.162 |
assic | +0.162 |
ikhail | +0.162 |
Tokens | +0.158 |
Neg logit contributions | |
confir | -0.161 |
contro | -0.154 |
discharge | -0.150 |
corpor | -0.148 |
Jude | -0.148 |
Token | ? |
Token ID | 30 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
acs | +0.209 |
Gravity | +0.203 |
assic | +0.203 |
ikhail | +0.203 |
Tokens | +0.199 |
Neg logit contributions | |
confir | -0.182 |
contro | -0.174 |
discharge | -0.169 |
corpor | -0.167 |
Jude | -0.166 |
Token | bullying |
Token ID | 20714 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
acs | +0.198 |
Gravity | +0.194 |
assic | +0.192 |
ikhail | +0.192 |
Tokens | +0.188 |
Neg logit contributions | |
confir | -0.149 |
contro | -0.139 |
discharge | -0.135 |
Jude | -0.133 |
corpor | -0.132 |
Token | , |
Token ID | 11 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
confir | +0.258 |
contro | +0.240 |
discharge | +0.232 |
Jude | +0.231 |
corpor | +0.229 |
Neg logit contributions | |
acs | -0.365 |
Gravity | -0.357 |
assic | -0.354 |
ikhail | -0.353 |
Tokens | -0.347 |
Token | dominance |
Token ID | 18648 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
acs | +0.372 |
Gravity | +0.365 |
assic | +0.360 |
ikhail | +0.358 |
Tokens | +0.353 |
Neg logit contributions | |
confir | -0.290 |
contro | -0.269 |
discharge | -0.262 |
Jude | -0.260 |
cumulative | -0.255 |
Token | , |
Token ID | 11 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
confir | +0.458 |
contro | +0.437 |
discharge | +0.425 |
corpor | +0.420 |
Jude | +0.417 |
Neg logit contributions | |
acs | -0.561 |
Gravity | -0.550 |
assic | -0.547 |
ikhail | -0.546 |
Tokens | -0.536 |
Token | and |
Token ID | 290 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
acs | +0.346 |
Gravity | +0.339 |
assic | +0.336 |
ikhail | +0.335 |
Tokens | +0.328 |
Neg logit contributions | |
confir | -0.307 |
contro | -0.288 |
discharge | -0.281 |
Jude | -0.278 |
corpor | -0.275 |
Token | victim |
Token ID | 3117 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
acs | +0.249 |
Gravity | +0.245 |
assic | +0.241 |
ikhail | +0.241 |
Tokens | +0.237 |
Neg logit contributions | |
confir | -0.188 |
contro | -0.174 |
discharge | -0.170 |
Jude | -0.169 |
corpor | -0.166 |
Token | ization |
Token ID | 1634 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
acs | +0.598 |
Gravity | +0.585 |
assic | +0.581 |
ikhail | +0.578 |
Tokens | +0.569 |
Neg logit contributions | |
confir | -0.327 |
contro | -0.303 |
discharge | -0.290 |
Jude | -0.289 |
corpor | -0.286 |
Token | during |
Token ID | 1141 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
confir | +0.486 |
contro | +0.455 |
discharge | +0.440 |
Jude | +0.434 |
corpor | +0.434 |
Neg logit contributions | |
acs | -0.738 |
Gravity | -0.722 |
ikhail | -0.716 |
assic | -0.716 |
Tokens | -0.703 |
Token | the |
Token ID | 262 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
acs | +0.216 |
Gravity | +0.213 |
ikhail | +0.210 |
assic | +0.210 |
Tokens | +0.205 |
Neg logit contributions | |
confir | -0.193 |
contro | -0.181 |
discharge | -0.175 |
Jude | -0.174 |
corpor | -0.173 |
Token | transition |
Token ID | 6801 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
acs | +0.051 |
Gravity | +0.050 |
assic | +0.049 |
ikhail | +0.049 |
Tokens | +0.048 |
Neg logit contributions | |
confir | -0.045 |
contro | -0.042 |
discharge | -0.040 |
Jude | -0.040 |
corpor | -0.040 |
Token | from |
Token ID | 422 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
acs | +0.361 |
Gravity | +0.353 |
assic | +0.351 |
ikhail | +0.350 |
Tokens | +0.343 |
Neg logit contributions | |
confir | -0.271 |
contro | -0.254 |
discharge | -0.245 |
Jude | -0.243 |
corpor | -0.242 |
Token | Un |
Token ID | 791 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
acs | +0.495 |
Gravity | +0.485 |
assic | +0.480 |
ikhail | +0.479 |
Tokens | +0.469 |
Neg logit contributions | |
confir | -0.470 |
contro | -0.447 |
discharge | -0.437 |
corpor | -0.432 |
Jude | -0.432 |
Token | locked |
Token ID | 24162 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
confir | +0.444 |
contro | +0.423 |
discharge | +0.413 |
corpor | +0.409 |
Jude | +0.408 |
Neg logit contributions | |
acs | -0.422 |
Gravity | -0.411 |
assic | -0.408 |
ikhail | -0.407 |
Tokens | -0.399 |
Token | BIOS |
Token ID | 27553 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
acs | +0.059 |
Gravity | +0.058 |
assic | +0.058 |
ikhail | +0.057 |
Tokens | +0.056 |
Neg logit contributions | |
confir | -0.049 |
contro | -0.046 |
discharge | -0.045 |
corpor | -0.044 |
Jude | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -2.39e-03 |
Pos logit contributions | |
acs | +0.449 |
Gravity | +0.442 |
assic | +0.435 |
ikhail | +0.434 |
Tokens | +0.429 |
Neg logit contributions | |
confir | -0.369 |
contro | -0.349 |
discharge | -0.340 |
Jude | -0.338 |
corpor | -0.338 |
Token | Digital |
Token ID | 10231 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
confir | +0.004 |
contro | +0.004 |
discharge | +0.004 |
corpor | +0.004 |
Jude | +0.004 |
Neg logit contributions | |
acs | -0.005 |
Gravity | -0.004 |
assic | -0.004 |
ikhail | -0.004 |
Tokens | -0.004 |
Token | P |
Token ID | 47 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
confir | +1.182 |
contro | +1.133 |
discharge | +1.104 |
corpor | +1.089 |
Jude | +1.088 |
Neg logit contributions | |
acs | -1.119 |
assic | -1.087 |
ikhail | -1.086 |
Gravity | -1.085 |
Tokens | -1.052 |
Token | WM |
Token ID | 22117 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
acs | +0.614 |
Gravity | +0.600 |
assic | +0.597 |
ikhail | +0.595 |
Tokens | +0.584 |
Neg logit contributions | |
confir | -0.546 |
contro | -0.519 |
discharge | -0.505 |
corpor | -0.500 |
Jude | -0.498 |
Token | Controller |
Token ID | 22741 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
acs | +0.250 |
Gravity | +0.245 |
assic | +0.242 |
ikhail | +0.241 |
Tokens | +0.238 |
Neg logit contributions | |
confir | -0.216 |
contro | -0.204 |
discharge | -0.199 |
corpor | -0.198 |
Jude | -0.198 |
Token | , |
Token ID | 11 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
acs | +0.533 |
Gravity | +0.523 |
assic | +0.516 |
ikhail | +0.516 |
Tokens | +0.508 |
Neg logit contributions | |
confir | -0.453 |
contro | -0.428 |
discharge | -0.418 |
Jude | -0.414 |
corpor | -0.414 |
Token | and |
Token ID | 290 |
Feature activation | -8.66e-02 |
Pos logit contributions | |
confir | +0.047 |
contro | +0.044 |
discharge | +0.043 |
corpor | +0.043 |
Jude | +0.043 |
Neg logit contributions | |
acs | -0.050 |
Gravity | -0.049 |
assic | -0.048 |
ikhail | -0.048 |
Tokens | -0.047 |
Token | Enhanced |
Token ID | 22104 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
confir | +0.148 |
contro | +0.141 |
discharge | +0.137 |
corpor | +0.136 |
Jude | +0.136 |
Neg logit contributions | |
acs | -0.158 |
Gravity | -0.155 |
assic | -0.154 |
ikhail | -0.153 |
Tokens | -0.150 |
Token | of |
Token ID | 286 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
acs | +0.896 |
Gravity | +0.880 |
assic | +0.874 |
ikhail | +0.870 |
Tokens | +0.856 |
Neg logit contributions | |
confir | -0.624 |
contro | -0.586 |
discharge | -0.569 |
corpor | -0.561 |
Jude | -0.560 |
Token | factors |
Token ID | 5087 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
acs | +0.181 |
Gravity | +0.177 |
assic | +0.175 |
ikhail | +0.175 |
Tokens | +0.172 |
Neg logit contributions | |
confir | -0.184 |
contro | -0.175 |
discharge | -0.171 |
corpor | -0.169 |
Jude | -0.168 |
Token | means |
Token ID | 1724 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
confir | +0.416 |
contro | +0.389 |
discharge | +0.374 |
corpor | +0.367 |
Jude | +0.366 |
Neg logit contributions | |
acs | -0.811 |
Gravity | -0.797 |
assic | -0.794 |
ikhail | -0.792 |
Tokens | -0.780 |
Token | it |
Token ID | 340 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
acs | +0.521 |
Gravity | +0.513 |
assic | +0.506 |
ikhail | +0.506 |
Tokens | +0.498 |
Neg logit contributions | |
confir | -0.375 |
contro | -0.351 |
discharge | -0.340 |
corpor | -0.337 |
Jude | -0.336 |
Token | 's |
Token ID | 338 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
acs | +0.187 |
Gravity | +0.183 |
assic | +0.182 |
ikhail | +0.182 |
Tokens | +0.178 |
Neg logit contributions | |
confir | -0.180 |
contro | -0.172 |
discharge | -0.167 |
corpor | -0.165 |
Jude | -0.165 |
Token | closed |
Token ID | 4838 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
acs | +0.348 |
Gravity | +0.341 |
assic | +0.337 |
ikhail | +0.336 |
Tokens | +0.331 |
Neg logit contributions | |
confir | -0.301 |
contro | -0.283 |
discharge | -0.276 |
corpor | -0.273 |
Jude | -0.272 |
Token | . |
Token ID | 13 |
Feature activation | +6.69e-02 |
Pos logit contributions | |
acs | +0.253 |
Gravity | +0.249 |
assic | +0.247 |
ikhail | +0.247 |
Tokens | +0.242 |
Neg logit contributions | |
confir | -0.186 |
contro | -0.175 |
discharge | -0.170 |
corpor | -0.168 |
Jude | -0.167 |
Token | And |
Token ID | 843 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
acs | +0.129 |
Gravity | +0.128 |
ikhail | +0.125 |
assic | +0.125 |
Tokens | +0.124 |
Neg logit contributions | |
confir | -0.109 |
contro | -0.103 |
discharge | -0.100 |
corpor | -0.099 |
cumulative | -0.098 |
Token | the |
Token ID | 262 |
Feature activation | +8.69e-02 |
Pos logit contributions | |
acs | +0.086 |
Gravity | +0.085 |
assic | +0.083 |
ikhail | +0.083 |
Tokens | +0.082 |
Neg logit contributions | |
confir | -0.087 |
contro | -0.082 |
discharge | -0.081 |
corpor | -0.079 |
Jude | -0.079 |
Token | only |
Token ID | 691 |
Feature activation | -6.06e-02 |
Pos logit contributions | |
acs | +0.168 |
Gravity | +0.165 |
assic | +0.163 |
ikhail | +0.163 |
Tokens | +0.160 |
Neg logit contributions | |
confir | -0.140 |
contro | -0.132 |
discharge | -0.129 |
corpor | -0.127 |
Jude | -0.127 |
Token | numbers |
Token ID | 3146 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
confir | +0.093 |
contro | +0.088 |
discharge | +0.085 |
corpor | +0.084 |
Jude | +0.084 |
Neg logit contributions | |
acs | -0.121 |
Gravity | -0.119 |
assic | -0.118 |
ikhail | -0.118 |
Tokens | -0.116 |
Token | Metroid |
Token ID | 45430 |
Feature activation | +6.52e-01 |
Pos logit contributions | |
acs | +0.552 |
Gravity | +0.537 |
assic | +0.535 |
ikhail | +0.534 |
Tokens | +0.522 |
Neg logit contributions | |
confir | -0.630 |
contro | -0.605 |
discharge | -0.592 |
corpor | -0.585 |
Jude | -0.585 |
Token | speed |
Token ID | 2866 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
acs | +1.295 |
Gravity | +1.282 |
assic | +1.261 |
ikhail | +1.251 |
Tokens | +1.228 |
Neg logit contributions | |
confir | -1.010 |
contro | -0.949 |
discharge | -0.924 |
Jude | -0.918 |
corpor | -0.916 |
Token | run |
Token ID | 5143 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
confir | +0.025 |
contro | +0.024 |
discharge | +0.024 |
corpor | +0.024 |
Jude | +0.023 |
Neg logit contributions | |
acs | -0.023 |
Gravity | -0.023 |
assic | -0.023 |
ikhail | -0.023 |
Tokens | -0.022 |
Token | race |
Token ID | 3234 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
confir | +0.597 |
contro | +0.567 |
discharge | +0.552 |
corpor | +0.544 |
Jude | +0.542 |
Neg logit contributions | |
acs | -0.751 |
Gravity | -0.734 |
assic | -0.732 |
ikhail | -0.730 |
Tokens | -0.716 |
Token | , |
Token ID | 11 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
confir | +0.403 |
contro | +0.382 |
discharge | +0.371 |
corpor | +0.366 |
Jude | +0.365 |
Neg logit contributions | |
acs | -0.523 |
Gravity | -0.513 |
assic | -0.510 |
ikhail | -0.509 |
Tokens | -0.499 |
Token | from |
Token ID | 422 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
acs | +0.056 |
Gravity | +0.055 |
assic | +0.055 |
ikhail | +0.055 |
Tokens | +0.054 |
Neg logit contributions | |
confir | -0.047 |
contro | -0.044 |
discharge | -0.043 |
corpor | -0.042 |
Jude | -0.042 |
Token | Awesome |
Token ID | 25020 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
confir | +0.256 |
contro | +0.243 |
discharge | +0.237 |
corpor | +0.234 |
Jude | +0.233 |
Neg logit contributions | |
acs | -0.281 |
Gravity | -0.276 |
assic | -0.273 |
ikhail | -0.273 |
Tokens | -0.267 |
Token | Games |
Token ID | 5776 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
acs | +0.191 |
Gravity | +0.186 |
assic | +0.185 |
ikhail | +0.185 |
Tokens | +0.180 |
Neg logit contributions | |
confir | -0.203 |
contro | -0.194 |
discharge | -0.190 |
corpor | -0.187 |
Jude | -0.187 |
Token | Done |
Token ID | 24429 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
confir | +0.149 |
contro | +0.141 |
discharge | +0.137 |
corpor | +0.135 |
Jude | +0.134 |
Neg logit contributions | |
acs | -0.201 |
Gravity | -0.197 |
assic | -0.196 |
ikhail | -0.195 |
Tokens | -0.192 |
Token | Quick |
Token ID | 12029 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
acs | +0.184 |
Gravity | +0.180 |
assic | +0.178 |
ikhail | +0.178 |
Tokens | +0.174 |
Neg logit contributions | |
confir | -0.190 |
contro | -0.181 |
discharge | -0.177 |
corpor | -0.175 |
Jude | -0.174 |
Token | 2014 |
Token ID | 1946 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
acs | +0.413 |
Gravity | +0.407 |
assic | +0.401 |
ikhail | +0.399 |
Tokens | +0.393 |
Neg logit contributions | |
confir | -0.324 |
contro | -0.306 |
discharge | -0.298 |
corpor | -0.293 |
Jude | -0.292 |
Token | to |
Token ID | 284 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
confir | +0.878 |
contro | +0.821 |
discharge | +0.798 |
corpor | +0.788 |
Jude | +0.782 |
Neg logit contributions | |
acs | -1.305 |
Gravity | -1.273 |
assic | -1.272 |
ikhail | -1.265 |
Tokens | -1.247 |
Token | give |
Token ID | 1577 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
confir | +0.161 |
contro | +0.152 |
discharge | +0.148 |
Jude | +0.147 |
corpor | +0.146 |
Neg logit contributions | |
acs | -0.205 |
Gravity | -0.202 |
assic | -0.200 |
ikhail | -0.199 |
Tokens | -0.196 |
Token | third |
Token ID | 2368 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
confir | +0.090 |
contro | +0.086 |
discharge | +0.084 |
corpor | +0.082 |
Jude | +0.082 |
Neg logit contributions | |
acs | -0.092 |
Gravity | -0.090 |
assic | -0.089 |
ikhail | -0.089 |
Tokens | -0.087 |
Token | driver |
Token ID | 4639 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
confir | +0.128 |
contro | +0.121 |
discharge | +0.118 |
corpor | +0.117 |
Jude | +0.116 |
Neg logit contributions | |
acs | -0.151 |
Gravity | -0.148 |
assic | -0.147 |
ikhail | -0.147 |
Tokens | -0.144 |
Token | Bruno |
Token ID | 31045 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
acs | +0.777 |
Gravity | +0.764 |
ikhail | +0.749 |
assic | +0.748 |
Mov | +0.728 |
Neg logit contributions | |
confir | -0.917 |
contro | -0.875 |
discharge | -0.860 |
corpor | -0.845 |
cumulative | -0.843 |
Token | S |
Token ID | 311 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
acs | +0.676 |
Gravity | +0.659 |
assic | +0.657 |
ikhail | +0.655 |
Tokens | +0.640 |
Neg logit contributions | |
confir | -0.725 |
contro | -0.694 |
discharge | -0.678 |
corpor | -0.670 |
Jude | -0.669 |
Token | enna |
Token ID | 13713 |
Feature activation | -7.28e-02 |
Pos logit contributions | |
acs | +0.037 |
Gravity | +0.036 |
assic | +0.036 |
ikhail | +0.036 |
Tokens | +0.035 |
Neg logit contributions | |
confir | -0.027 |
contro | -0.026 |
discharge | -0.025 |
corpor | -0.025 |
Jude | -0.025 |
Token | some |
Token ID | 617 |
Feature activation | -8.58e-03 |
Pos logit contributions | |
confir | +0.118 |
contro | +0.112 |
discharge | +0.109 |
corpor | +0.107 |
Jude | +0.107 |
Neg logit contributions | |
acs | -0.140 |
Gravity | -0.137 |
assic | -0.136 |
ikhail | -0.136 |
Tokens | -0.133 |
Token | time |
Token ID | 640 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
confir | +0.013 |
contro | +0.013 |
discharge | +0.012 |
corpor | +0.012 |
Jude | +0.012 |
Neg logit contributions | |
acs | -0.017 |
Gravity | -0.017 |
assic | -0.017 |
ikhail | -0.016 |
Tokens | -0.016 |
Token | in |
Token ID | 287 |
Feature activation | -1.80e-02 |
Pos logit contributions | |
acs | +0.187 |
Gravity | +0.183 |
assic | +0.182 |
ikhail | +0.181 |
Tokens | +0.177 |
Neg logit contributions | |
confir | -0.170 |
contro | -0.161 |
discharge | -0.157 |
corpor | -0.156 |
Jude | -0.155 |
Token | the |
Token ID | 262 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
confir | +0.028 |
contro | +0.027 |
discharge | +0.026 |
corpor | +0.026 |
Jude | +0.026 |
Neg logit contributions | |
acs | -0.035 |
Gravity | -0.035 |
assic | -0.034 |
ikhail | -0.034 |
Tokens | -0.034 |
Token | AAA |
Token ID | 25734 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
acs | +0.098 |
Gravity | +0.096 |
assic | +0.096 |
ikhail | +0.095 |
Tokens | +0.093 |
Neg logit contributions | |
confir | -0.094 |
contro | -0.089 |
discharge | -0.087 |
corpor | -0.086 |
Jude | -0.086 |
Token | games |
Token ID | 1830 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
acs | +0.483 |
Gravity | +0.473 |
assic | +0.469 |
ikhail | +0.466 |
Tokens | +0.459 |
Neg logit contributions | |
confir | -0.472 |
contro | -0.449 |
discharge | -0.438 |
Jude | -0.432 |
corpor | -0.432 |
Token | , |
Token ID | 11 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
confir | +0.220 |
contro | +0.210 |
discharge | +0.205 |
corpor | +0.202 |
Jude | +0.201 |
Neg logit contributions | |
acs | -0.277 |
ikhail | -0.271 |
assic | -0.271 |
Gravity | -0.270 |
Tokens | -0.265 |
Token | and |
Token ID | 290 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
acs | +0.144 |
Gravity | +0.141 |
assic | +0.141 |
ikhail | +0.140 |
Tokens | +0.137 |
Neg logit contributions | |
confir | -0.128 |
contro | -0.122 |
discharge | -0.119 |
corpor | -0.117 |
Jude | -0.117 |
Token | you |
Token ID | 345 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
acs | +0.124 |
Gravity | +0.123 |
assic | +0.121 |
ikhail | +0.120 |
Tokens | +0.119 |
Neg logit contributions | |
confir | -0.112 |
contro | -0.106 |
discharge | -0.103 |
corpor | -0.101 |
Jude | -0.101 |
Token | all |
Token ID | 477 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
acs | +0.041 |
Gravity | +0.040 |
assic | +0.040 |
ikhail | +0.040 |
Tokens | +0.039 |
Neg logit contributions | |
confir | -0.031 |
contro | -0.030 |
discharge | -0.029 |
corpor | -0.029 |
Jude | -0.028 |
Token | know |
Token ID | 760 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
acs | +0.046 |
Gravity | +0.046 |
assic | +0.045 |
ikhail | +0.045 |
Tokens | +0.044 |
Neg logit contributions | |
confir | -0.035 |
contro | -0.033 |
discharge | -0.032 |
corpor | -0.031 |
Jude | -0.031 |
Token | that |
Token ID | 326 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
confir | +0.317 |
contro | +0.300 |
discharge | +0.292 |
corpor | +0.288 |
Jude | +0.287 |
Neg logit contributions | |
acs | -0.349 |
Gravity | -0.344 |
assic | -0.340 |
ikhail | -0.338 |
Tokens | -0.334 |
Token | the |
Token ID | 262 |
Feature activation | +4.86e-03 |
Pos logit contributions | |
acs | +0.050 |
Gravity | +0.049 |
assic | +0.049 |
ikhail | +0.049 |
Tokens | +0.048 |
Neg logit contributions | |
confir | -0.044 |
contro | -0.041 |
discharge | -0.040 |
corpor | -0.040 |
Jude | -0.040 |
Token | Nintendo |
Token ID | 9714 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
acs | +0.009 |
Gravity | +0.009 |
assic | +0.009 |
ikhail | +0.009 |
Tokens | +0.009 |
Neg logit contributions | |
confir | -0.008 |
contro | -0.008 |
discharge | -0.008 |
corpor | -0.007 |
Jude | -0.007 |
Token | booth |
Token ID | 18600 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
confir | +0.911 |
contro | +0.872 |
discharge | +0.848 |
corpor | +0.838 |
Jude | +0.834 |
Neg logit contributions | |
acs | -1.027 |
ikhail | -1.004 |
assic | -1.002 |
Gravity | -1.001 |
Tokens | -0.977 |
Token | had |
Token ID | 550 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
confir | +0.822 |
contro | +0.777 |
discharge | +0.754 |
corpor | +0.748 |
Jude | +0.746 |
Neg logit contributions | |
acs | -1.024 |
Gravity | -1.002 |
assic | -0.993 |
ikhail | -0.991 |
Tokens | -0.971 |
Token | wrought |
Token ID | 38733 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
acs | +0.415 |
Gravity | +0.408 |
ikhail | +0.394 |
assic | +0.394 |
Mov | +0.390 |
Neg logit contributions | |
confir | -0.332 |
contro | -0.308 |
discharge | -0.302 |
corpor | -0.301 |
Jude | -0.300 |
Token | , |
Token ID | 11 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
confir | +0.513 |
contro | +0.480 |
discharge | +0.469 |
corpor | +0.462 |
Jude | +0.462 |
Neg logit contributions | |
acs | -0.644 |
Gravity | -0.632 |
ikhail | -0.621 |
assic | -0.621 |
Tokens | -0.612 |
Token |
|
Token ID | 198 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
confir | +0.042 |
contro | +0.040 |
discharge | +0.038 |
corpor | +0.038 |
Jude | +0.038 |
Neg logit contributions | |
acs | -0.073 |
Gravity | -0.072 |
assic | -0.072 |
ikhail | -0.071 |
Tokens | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | +8.28e-02 |
Pos logit contributions | |
acs | +0.508 |
Gravity | +0.500 |
ikhail | +0.489 |
Tokens | +0.489 |
assic | +0.488 |
Neg logit contributions | |
confir | -0.322 |
contro | -0.284 |
discharge | -0.275 |
corpor | -0.271 |
cumulative | -0.268 |
Token | and |
Token ID | 392 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
acs | +0.154 |
Gravity | +0.150 |
assic | +0.149 |
ikhail | +0.149 |
Tokens | +0.146 |
Neg logit contributions | |
confir | -0.140 |
contro | -0.133 |
discharge | -0.130 |
corpor | -0.128 |
Jude | -0.128 |
Token | on |
Token ID | 319 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
confir | +0.032 |
contro | +0.030 |
discharge | +0.029 |
corpor | +0.029 |
Jude | +0.029 |
Neg logit contributions | |
acs | -0.037 |
Gravity | -0.037 |
ikhail | -0.036 |
assic | -0.036 |
Tokens | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
confir | +0.215 |
contro | +0.205 |
discharge | +0.200 |
corpor | +0.197 |
Jude | +0.197 |
Neg logit contributions | |
acs | -0.221 |
Gravity | -0.216 |
assic | -0.214 |
ikhail | -0.214 |
Tokens | -0.210 |
Token | labor |
Token ID | 4827 |
Feature activation | -5.74e-01 |
Pos logit contributions | |
confir | +0.194 |
contro | +0.185 |
discharge | +0.180 |
corpor | +0.178 |
Jude | +0.178 |
Neg logit contributions | |
acs | -0.211 |
Gravity | -0.206 |
assic | -0.204 |
ikhail | -0.204 |
Tokens | -0.200 |
Token | that |
Token ID | 326 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
confir | +0.825 |
contro | +0.787 |
discharge | +0.766 |
corpor | +0.752 |
Jude | +0.748 |
Neg logit contributions | |
acs | -1.195 |
Gravity | -1.175 |
assic | -1.173 |
ikhail | -1.172 |
Tokens | -1.151 |
Token | I |
Token ID | 314 |
Feature activation | -2.94e-02 |
Pos logit contributions | |
acs | +0.092 |
Gravity | +0.090 |
assic | +0.089 |
ikhail | +0.089 |
Tokens | +0.087 |
Neg logit contributions | |
confir | -0.084 |
contro | -0.079 |
discharge | -0.077 |
corpor | -0.076 |
Jude | -0.076 |
Token | placing |
Token ID | 12560 |
Feature activation | +7.19e-02 |
Pos logit contributions | |
confir | +0.036 |
contro | +0.034 |
discharge | +0.033 |
corpor | +0.033 |
Jude | +0.033 |
Neg logit contributions | |
acs | -0.038 |
Gravity | -0.038 |
assic | -0.037 |
ikhail | -0.037 |
Tokens | -0.037 |
Token | figures |
Token ID | 5538 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
acs | +0.118 |
Gravity | +0.115 |
assic | +0.114 |
ikhail | +0.113 |
Tokens | +0.111 |
Neg logit contributions | |
confir | -0.137 |
contro | -0.130 |
discharge | -0.128 |
corpor | -0.126 |
Jude | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
confir | +0.238 |
contro | +0.227 |
discharge | +0.220 |
corpor | +0.217 |
Jude | +0.216 |
Neg logit contributions | |
acs | -0.308 |
Gravity | -0.301 |
assic | -0.300 |
ikhail | -0.300 |
Tokens | -0.294 |
Token |
|
Token ID | 198 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
confir | +0.055 |
contro | +0.052 |
discharge | +0.050 |
corpor | +0.049 |
Jude | +0.049 |
Neg logit contributions | |
acs | -0.076 |
Gravity | -0.075 |
assic | -0.074 |
ikhail | -0.074 |
Tokens | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
confir | +0.025 |
contro | +0.023 |
discharge | +0.022 |
corpor | +0.021 |
cumulative | +0.021 |
Neg logit contributions | |
acs | -0.042 |
Gravity | -0.042 |
assic | -0.041 |
ikhail | -0.041 |
Tokens | -0.041 |
Token | Lastly |
Token ID | 37511 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
confir | +0.066 |
contro | +0.064 |
discharge | +0.062 |
corpor | +0.061 |
Jude | +0.061 |
Neg logit contributions | |
acs | -0.060 |
Gravity | -0.058 |
assic | -0.058 |
ikhail | -0.058 |
Tokens | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
confir | +0.390 |
contro | +0.371 |
discharge | +0.361 |
corpor | +0.356 |
Jude | +0.355 |
Neg logit contributions | |
acs | -0.476 |
Gravity | -0.466 |
assic | -0.464 |
ikhail | -0.463 |
Tokens | -0.454 |
Token | I |
Token ID | 314 |
Feature activation | -3.98e-02 |
Pos logit contributions | |
confir | +0.020 |
contro | +0.019 |
discharge | +0.019 |
corpor | +0.018 |
Jude | +0.018 |
Neg logit contributions | |
acs | -0.024 |
Gravity | -0.023 |
assic | -0.023 |
ikhail | -0.023 |
Tokens | -0.023 |
Token | tint |
Token ID | 34791 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
confir | +0.060 |
contro | +0.057 |
discharge | +0.056 |
corpor | +0.055 |
Jude | +0.055 |
Neg logit contributions | |
acs | -0.080 |
Gravity | -0.079 |
assic | -0.078 |
ikhail | -0.078 |
Tokens | -0.077 |
Token | ed |
Token ID | 276 |
Feature activation | +6.11e-02 |
Pos logit contributions | |
confir | +0.721 |
contro | +0.690 |
discharge | +0.673 |
corpor | +0.663 |
Jude | +0.663 |
Neg logit contributions | |
acs | -0.803 |
assic | -0.783 |
Gravity | -0.782 |
ikhail | -0.780 |
Tokens | -0.765 |
Token | around |
Token ID | 1088 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
acs | +0.119 |
Gravity | +0.117 |
assic | +0.116 |
ikhail | +0.116 |
Tokens | +0.113 |
Neg logit contributions | |
confir | -0.097 |
contro | -0.092 |
discharge | -0.090 |
corpor | -0.089 |
Jude | -0.088 |
Token | sorts |
Token ID | 10524 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
confir | +0.158 |
contro | +0.150 |
discharge | +0.146 |
corpor | +0.144 |
Jude | +0.143 |
Neg logit contributions | |
acs | -0.189 |
Gravity | -0.185 |
assic | -0.184 |
ikhail | -0.183 |
Tokens | -0.180 |
Token | in |
Token ID | 287 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
confir | +0.175 |
contro | +0.164 |
discharge | +0.160 |
corpor | +0.157 |
Jude | +0.157 |
Neg logit contributions | |
acs | -0.220 |
Gravity | -0.216 |
assic | -0.214 |
ikhail | -0.213 |
Tokens | -0.210 |
Token | a |
Token ID | 257 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
confir | +0.077 |
contro | +0.073 |
discharge | +0.071 |
corpor | +0.070 |
Jude | +0.070 |
Neg logit contributions | |
acs | -0.091 |
Gravity | -0.089 |
assic | -0.089 |
ikhail | -0.089 |
Tokens | -0.087 |
Token | blow |
Token ID | 6611 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
confir | +0.250 |
contro | +0.233 |
discharge | +0.226 |
corpor | +0.224 |
Jude | +0.223 |
Neg logit contributions | |
acs | -0.340 |
Gravity | -0.336 |
assic | -0.331 |
ikhail | -0.330 |
Tokens | -0.324 |
Token | out |
Token ID | 448 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
confir | +0.232 |
contro | +0.217 |
discharge | +0.209 |
corpor | +0.206 |
Jude | +0.205 |
Neg logit contributions | |
acs | -0.418 |
Gravity | -0.410 |
assic | -0.408 |
ikhail | -0.408 |
Tokens | -0.400 |
Token | victory |
Token ID | 5373 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
acs | +0.061 |
Gravity | +0.060 |
assic | +0.059 |
ikhail | +0.059 |
Tokens | +0.058 |
Neg logit contributions | |
confir | -0.032 |
contro | -0.030 |
Jude | -0.028 |
discharge | -0.028 |
corpor | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
confir | +1.015 |
contro | +0.960 |
discharge | +0.932 |
corpor | +0.921 |
Jude | +0.918 |
Neg logit contributions | |
acs | -1.259 |
Gravity | -1.232 |
assic | -1.221 |
ikhail | -1.220 |
Tokens | -1.198 |
Token | In |
Token ID | 554 |
Feature activation | +6.30e-02 |
Pos logit contributions | |
acs | +0.048 |
Gravity | +0.048 |
assic | +0.047 |
ikhail | +0.047 |
Tokens | +0.046 |
Neg logit contributions | |
confir | -0.042 |
contro | -0.039 |
discharge | -0.038 |
corpor | -0.038 |
cumulative | -0.038 |
Token | 2014 |
Token ID | 1946 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
acs | +0.118 |
Gravity | +0.115 |
assic | +0.115 |
ikhail | +0.115 |
Tokens | +0.112 |
Neg logit contributions | |
confir | -0.105 |
contro | -0.100 |
discharge | -0.097 |
corpor | -0.096 |
Jude | -0.096 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
acs | +0.396 |
Gravity | +0.387 |
assic | +0.386 |
ikhail | +0.385 |
Tokens | +0.377 |
Neg logit contributions | |
confir | -0.324 |
contro | -0.308 |
discharge | -0.300 |
corpor | -0.296 |
Jude | -0.295 |
Token | Texas |
Token ID | 3936 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
acs | +0.047 |
Gravity | +0.046 |
assic | +0.046 |
ikhail | +0.046 |
Tokens | +0.045 |
Neg logit contributions | |
confir | -0.037 |
contro | -0.035 |
discharge | -0.034 |
corpor | -0.033 |
Jude | -0.033 |
Token | path |
Token ID | 3108 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
acs | +0.839 |
Gravity | +0.825 |
assic | +0.817 |
ikhail | +0.812 |
Tokens | +0.802 |
Neg logit contributions | |
confir | -0.636 |
contro | -0.605 |
discharge | -0.586 |
Jude | -0.578 |
corpor | -0.574 |
Token | , |
Token ID | 837 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
acs | +0.452 |
Gravity | +0.444 |
ikhail | +0.440 |
assic | +0.439 |
Tokens | +0.435 |
Neg logit contributions | |
confir | -0.316 |
contro | -0.298 |
discharge | -0.290 |
Jude | -0.284 |
corpor | -0.284 |
Token | 4 |
Token ID | 604 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
acs | +0.218 |
Gravity | +0.214 |
assic | +0.212 |
ikhail | +0.211 |
Tokens | +0.207 |
Neg logit contributions | |
confir | -0.174 |
contro | -0.164 |
discharge | -0.159 |
Jude | -0.157 |
corpor | -0.156 |
Token | ) |
Token ID | 1267 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
confir | +0.230 |
contro | +0.214 |
discharge | +0.206 |
Jude | +0.202 |
corpor | +0.202 |
Neg logit contributions | |
acs | -0.458 |
Gravity | -0.449 |
assic | -0.447 |
ikhail | -0.446 |
Tokens | -0.440 |
Token | = |
Token ID | 796 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
confir | +0.068 |
contro | +0.065 |
discharge | +0.063 |
corpor | +0.062 |
Jude | +0.062 |
Neg logit contributions | |
acs | -0.075 |
Gravity | -0.074 |
assic | -0.073 |
ikhail | -0.073 |
Tokens | -0.072 |
Token | '. |
Token ID | 45302 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
acs | +1.007 |
Gravity | +0.991 |
assic | +0.977 |
ikhail | +0.972 |
Tokens | +0.962 |
Neg logit contributions | |
confir | -0.753 |
contro | -0.713 |
discharge | -0.688 |
Jude | -0.679 |
corpor | -0.675 |
Token | cl |
Token ID | 565 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
confir | +1.037 |
contro | +0.988 |
discharge | +0.961 |
corpor | +0.948 |
Jude | +0.947 |
Neg logit contributions | |
acs | -1.194 |
Gravity | -1.166 |
assic | -1.161 |
ikhail | -1.158 |
Tokens | -1.135 |
Token | j |
Token ID | 73 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
acs | +0.726 |
assic | +0.704 |
ikhail | +0.698 |
Gravity | +0.697 |
Tokens | +0.683 |
Neg logit contributions | |
confir | -0.669 |
contro | -0.635 |
discharge | -0.616 |
corpor | -0.610 |
Jude | -0.607 |
Token | ' |
Token ID | 6 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
confir | +0.220 |
contro | +0.209 |
discharge | +0.203 |
corpor | +0.200 |
Jude | +0.200 |
Neg logit contributions | |
acs | -0.273 |
Gravity | -0.267 |
assic | -0.265 |
ikhail | -0.264 |
Tokens | -0.260 |
Token | OR |
Token ID | 6375 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
confir | +0.178 |
contro | +0.168 |
discharge | +0.163 |
corpor | +0.160 |
Jude | +0.160 |
Neg logit contributions | |
acs | -0.231 |
Gravity | -0.227 |
assic | -0.224 |
ikhail | -0.223 |
Tokens | -0.220 |
Token | RIGHT |
Token ID | 33621 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
acs | +0.215 |
Gravity | +0.211 |
assic | +0.208 |
ikhail | +0.208 |
Tokens | +0.205 |
Neg logit contributions | |
confir | -0.171 |
contro | -0.162 |
discharge | -0.157 |
corpor | -0.155 |
Jude | -0.154 |
Token | however |
Token ID | 2158 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
confir | +0.091 |
contro | +0.087 |
discharge | +0.084 |
corpor | +0.083 |
Jude | +0.083 |
Neg logit contributions | |
acs | -0.107 |
Gravity | -0.105 |
assic | -0.104 |
ikhail | -0.104 |
Tokens | -0.102 |
Token | , |
Token ID | 11 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
acs | +1.089 |
Gravity | +1.073 |
ikhail | +1.059 |
assic | +1.054 |
Tokens | +1.042 |
Neg logit contributions | |
confir | -0.879 |
contro | -0.829 |
discharge | -0.811 |
Jude | -0.798 |
corpor | -0.796 |
Token | and |
Token ID | 290 |
Feature activation | -9.66e-02 |
Pos logit contributions | |
confir | +0.082 |
contro | +0.078 |
discharge | +0.076 |
corpor | +0.075 |
Jude | +0.075 |
Neg logit contributions | |
acs | -0.100 |
Gravity | -0.098 |
assic | -0.098 |
ikhail | -0.098 |
Tokens | -0.096 |
Token | his |
Token ID | 465 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
confir | +0.154 |
contro | +0.146 |
discharge | +0.142 |
corpor | +0.140 |
Jude | +0.140 |
Neg logit contributions | |
acs | -0.188 |
Gravity | -0.185 |
ikhail | -0.183 |
assic | -0.183 |
Tokens | -0.179 |
Token | preseason |
Token ID | 18913 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
acs | +0.336 |
Gravity | +0.330 |
assic | +0.328 |
ikhail | +0.327 |
Tokens | +0.321 |
Neg logit contributions | |
confir | -0.255 |
contro | -0.240 |
discharge | -0.233 |
corpor | -0.231 |
Jude | -0.230 |
Token | teased |
Token ID | 35380 |
Feature activation | -6.63e-01 |
Pos logit contributions | |
acs | +0.923 |
Gravity | +0.904 |
ikhail | +0.892 |
assic | +0.890 |
Tokens | +0.875 |
Neg logit contributions | |
confir | -0.748 |
contro | -0.708 |
discharge | -0.684 |
corpor | -0.683 |
Jude | -0.682 |
Token | us |
Token ID | 514 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
confir | +1.094 |
contro | +1.045 |
discharge | +1.017 |
corpor | +1.003 |
cumulative | +1.002 |
Neg logit contributions | |
acs | -1.258 |
assic | -1.228 |
Gravity | -1.221 |
ikhail | -1.217 |
Tokens | -1.193 |
Token | with |
Token ID | 351 |
Feature activation | -8.38e-02 |
Pos logit contributions | |
confir | +0.220 |
contro | +0.209 |
discharge | +0.204 |
corpor | +0.201 |
Jude | +0.201 |
Neg logit contributions | |
acs | -0.244 |
Gravity | -0.238 |
assic | -0.237 |
ikhail | -0.236 |
Tokens | -0.232 |
Token | previews |
Token ID | 41418 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
confir | +0.139 |
contro | +0.133 |
discharge | +0.129 |
corpor | +0.128 |
Jude | +0.127 |
Neg logit contributions | |
acs | -0.157 |
Gravity | -0.154 |
assic | -0.153 |
ikhail | -0.153 |
Tokens | -0.150 |
Token | of |
Token ID | 286 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
confir | +0.456 |
contro | +0.431 |
discharge | +0.418 |
corpor | +0.411 |
Jude | +0.410 |
Neg logit contributions | |
acs | -0.672 |
Gravity | -0.659 |
assic | -0.656 |
ikhail | -0.654 |
Tokens | -0.643 |
Token | greatness |
Token ID | 27951 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
confir | +0.094 |
contro | +0.090 |
discharge | +0.088 |
corpor | +0.087 |
Jude | +0.087 |
Neg logit contributions | |
acs | -0.097 |
Gravity | -0.095 |
assic | -0.094 |
ikhail | -0.094 |
Tokens | -0.092 |
Token | that |
Token ID | 326 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
confir | +0.218 |
contro | +0.206 |
discharge | +0.201 |
corpor | +0.198 |
Jude | +0.198 |
Neg logit contributions | |
acs | -0.239 |
Gravity | -0.235 |
assic | -0.232 |
ikhail | -0.231 |
Tokens | -0.227 |
Token | life |
Token ID | 1204 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
confir | +0.167 |
contro | +0.158 |
discharge | +0.153 |
corpor | +0.151 |
Jude | +0.150 |
Neg logit contributions | |
acs | -0.194 |
Gravity | -0.191 |
assic | -0.188 |
ikhail | -0.188 |
Tokens | -0.185 |
Token | is |
Token ID | 318 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
confir | +0.217 |
contro | +0.207 |
discharge | +0.202 |
corpor | +0.200 |
Jude | +0.200 |
Neg logit contributions | |
acs | -0.204 |
Gravity | -0.199 |
assic | -0.197 |
ikhail | -0.197 |
Tokens | -0.193 |
Token | meaningless |
Token ID | 24513 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
confir | +0.167 |
contro | +0.156 |
discharge | +0.153 |
Jude | +0.151 |
corpor | +0.151 |
Neg logit contributions | |
acs | -0.178 |
Gravity | -0.175 |
assic | -0.171 |
ikhail | -0.171 |
Tokens | -0.169 |
Token | . |
Token ID | 764 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
acs | +0.272 |
Gravity | +0.268 |
ikhail | +0.263 |
assic | +0.263 |
Tokens | +0.260 |
Neg logit contributions | |
confir | -0.228 |
contro | -0.214 |
discharge | -0.208 |
Jude | -0.206 |
corpor | -0.205 |
Token | . |
Token ID | 764 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
confir | +1.221 |
contro | +1.177 |
discharge | +1.142 |
Jude | +1.136 |
corpor | +1.128 |
Neg logit contributions | |
acs | -1.520 |
assic | -1.494 |
ikhail | -1.481 |
Gravity | -1.470 |
Tokens | -1.448 |
Token | . |
Token ID | 764 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
confir | +0.598 |
contro | +0.553 |
discharge | +0.523 |
corpor | +0.512 |
Jude | +0.509 |
Neg logit contributions | |
acs | -1.734 |
assic | -1.705 |
Gravity | -1.704 |
ikhail | -1.699 |
Tokens | -1.675 |
Token | The |
Token ID | 383 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
confir | +1.024 |
contro | +0.982 |
discharge | +0.955 |
corpor | +0.942 |
Jude | +0.941 |
Neg logit contributions | |
acs | -1.166 |
assic | -1.140 |
ikhail | -1.133 |
Gravity | -1.132 |
Tokens | -1.110 |
Token | only |
Token ID | 691 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
confir | +0.037 |
contro | +0.035 |
discharge | +0.034 |
corpor | +0.034 |
Jude | +0.034 |
Neg logit contributions | |
acs | -0.046 |
Gravity | -0.045 |
assic | -0.045 |
ikhail | -0.045 |
Tokens | -0.044 |
Token | thing |
Token ID | 1517 |
Feature activation | +6.26e-02 |
Pos logit contributions | |
confir | +0.162 |
contro | +0.151 |
discharge | +0.147 |
corpor | +0.145 |
Jude | +0.145 |
Neg logit contributions | |
acs | -0.223 |
Gravity | -0.219 |
assic | -0.216 |
ikhail | -0.216 |
Tokens | -0.213 |
Token | that |
Token ID | 326 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
acs | +0.121 |
Gravity | +0.119 |
ikhail | +0.116 |
assic | +0.116 |
Tokens | +0.114 |
Neg logit contributions | |
confir | -0.101 |
contro | -0.094 |
discharge | -0.092 |
Jude | -0.091 |
corpor | -0.091 |
Token | used |
Token ID | 1484 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
confir | +0.101 |
contro | +0.097 |
discharge | +0.094 |
Jude | +0.093 |
corpor | +0.093 |
Neg logit contributions | |
acs | -0.181 |
Gravity | -0.180 |
assic | -0.179 |
ikhail | -0.177 |
Tokens | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
confir | +0.091 |
contro | +0.087 |
discharge | +0.084 |
corpor | +0.083 |
Jude | +0.083 |
Neg logit contributions | |
acs | -0.121 |
Gravity | -0.118 |
assic | -0.118 |
ikhail | -0.118 |
Tokens | -0.115 |
Token | � |
Token ID | 447 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
acs | +0.070 |
Gravity | +0.069 |
assic | +0.068 |
ikhail | +0.068 |
Tokens | +0.067 |
Neg logit contributions | |
confir | -0.068 |
contro | -0.065 |
discharge | -0.063 |
corpor | -0.062 |
Jude | -0.062 |
Token | � |
Token ID | 251 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
acs | +0.151 |
Gravity | +0.148 |
assic | +0.146 |
ikhail | +0.146 |
Tokens | +0.143 |
Neg logit contributions | |
confir | -0.167 |
contro | -0.160 |
discharge | -0.156 |
corpor | -0.155 |
Jude | -0.154 |
Token | He |
Token ID | 679 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
acs | +0.090 |
Gravity | +0.088 |
assic | +0.088 |
ikhail | +0.088 |
Tokens | +0.086 |
Neg logit contributions | |
confir | -0.086 |
contro | -0.082 |
discharge | -0.080 |
corpor | -0.079 |
Jude | -0.079 |
Token | sometimes |
Token ID | 3360 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
acs | +0.280 |
Gravity | +0.274 |
assic | +0.273 |
ikhail | +0.272 |
Tokens | +0.267 |
Neg logit contributions | |
confir | -0.245 |
contro | -0.233 |
discharge | -0.227 |
corpor | -0.224 |
Jude | -0.223 |
Token | shot |
Token ID | 2823 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
confir | +1.007 |
contro | +0.960 |
discharge | +0.933 |
corpor | +0.921 |
Jude | +0.916 |
Neg logit contributions | |
acs | -1.238 |
Gravity | -1.211 |
assic | -1.207 |
ikhail | -1.207 |
Tokens | -1.183 |
Token | in |
Token ID | 287 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
acs | +0.375 |
Gravity | +0.368 |
assic | +0.364 |
ikhail | +0.364 |
Tokens | +0.357 |
Neg logit contributions | |
confir | -0.342 |
contro | -0.325 |
discharge | -0.316 |
Jude | -0.312 |
corpor | -0.312 |
Token | the |
Token ID | 262 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
acs | +0.255 |
Gravity | +0.250 |
assic | +0.249 |
ikhail | +0.248 |
Tokens | +0.243 |
Neg logit contributions | |
confir | -0.213 |
contro | -0.203 |
discharge | -0.197 |
corpor | -0.195 |
Jude | -0.194 |
Token | afternoon |
Token ID | 6672 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
acs | +0.076 |
Gravity | +0.074 |
assic | +0.074 |
ikhail | +0.073 |
Tokens | +0.072 |
Neg logit contributions | |
confir | -0.064 |
contro | -0.061 |
discharge | -0.060 |
corpor | -0.059 |
Jude | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | -7.75e-03 |
Pos logit contributions | |
acs | +0.071 |
Gravity | +0.070 |
assic | +0.069 |
ikhail | +0.069 |
Tokens | +0.068 |
Neg logit contributions | |
confir | -0.060 |
contro | -0.057 |
discharge | -0.055 |
corpor | -0.055 |
Jude | -0.054 |