Token | Where |
Token ID | 6350 |
Feature activation | -8.80e-02 |
Pos logit contributions | |
unden | +1.029 |
rall | +1.004 |
arrang | +0.940 |
reluct | +0.933 |
fort | +0.915 |
Neg logit contributions | |
ocket | -0.895 |
adiator | -0.858 |
Explore | -0.854 |
replay | -0.844 |
rium | -0.838 |
Token | Primary |
Token ID | 21087 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
unden | +0.196 |
rall | +0.192 |
arrang | +0.182 |
reluct | +0.181 |
fort | +0.178 |
Neg logit contributions | |
ocket | -0.102 |
adiator | -0.096 |
Explore | -0.095 |
replay | -0.094 |
rium | -0.093 |
Token | | |
Token ID | 930 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
ocket | +0.503 |
adiator | +0.474 |
Explore | +0.469 |
rium | +0.459 |
replay | +0.458 |
Neg logit contributions | |
unden | -1.031 |
rall | -1.011 |
arrang | -0.959 |
reluct | -0.953 |
fort | -0.936 |
Token | Select |
Token ID | 9683 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
unden | +0.917 |
rall | +0.893 |
arrang | +0.837 |
reluct | +0.831 |
fort | +0.814 |
Neg logit contributions | |
ocket | -0.773 |
adiator | -0.739 |
Explore | -0.736 |
replay | -0.726 |
rium | -0.723 |
Token | - |
Token ID | 532 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
unden | +0.904 |
rall | +0.884 |
arrang | +0.828 |
reluct | +0.817 |
'. | +0.811 |
Neg logit contributions | |
ocket | -0.731 |
adiator | -0.704 |
Explore | -0.695 |
replay | -0.687 |
rium | -0.684 |
Token | Exp |
Token ID | 16870 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
unden | +0.575 |
rall | +0.564 |
arrang | +0.537 |
reluct | +0.532 |
fort | +0.526 |
Neg logit contributions | |
ocket | -0.254 |
adiator | -0.239 |
Explore | -0.236 |
replay | -0.231 |
rium | -0.230 |
Token | and |
Token ID | 392 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ocket | +2.190 |
Explore | +2.083 |
adiator | +2.009 |
rium | +1.994 |
replay | +1.963 |
Neg logit contributions | |
unden | -5.212 |
rall | -5.092 |
reluct | -4.908 |
arrang | -4.883 |
Ern | -4.695 |
Token | Property |
Token ID | 21746 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
unden | +0.304 |
rall | +0.300 |
arrang | +0.283 |
'. | +0.282 |
fort | +0.280 |
Neg logit contributions | |
ocket | -0.142 |
adiator | -0.136 |
Explore | -0.130 |
replay | -0.129 |
rium | -0.129 |
Token | B |
Token ID | 347 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
ocket | +2.207 |
Explore | +2.111 |
adiator | +2.090 |
paces | +2.087 |
replay | +2.081 |
Neg logit contributions | |
unden | -2.214 |
rall | -2.121 |
reluct | -2.003 |
arrang | -1.996 |
fort | -1.935 |
Token | ounds |
Token ID | 3733 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ocket | +0.201 |
adiator | +0.193 |
Explore | +0.192 |
rium | +0.191 |
paces | +0.190 |
Neg logit contributions | |
unden | -0.165 |
rall | -0.161 |
arrang | -0.148 |
reluct | -0.147 |
cir | -0.142 |
Token | | |
Token ID | 930 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
ocket | +1.553 |
Explore | +1.479 |
adiator | +1.466 |
paces | +1.460 |
replay | +1.443 |
Neg logit contributions | |
unden | -1.865 |
rall | -1.801 |
reluct | -1.702 |
arrang | -1.699 |
fort | -1.648 |
Token | get |
Token ID | 651 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
ocket | +0.116 |
replay | +0.111 |
adiator | +0.111 |
Explore | +0.110 |
rium | +0.109 |
Neg logit contributions | |
unden | -0.169 |
rall | -0.165 |
arrang | -0.156 |
reluct | -0.154 |
fort | -0.150 |
Token | worse |
Token ID | 4785 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
ocket | +0.327 |
adiator | +0.310 |
replay | +0.308 |
Explore | +0.307 |
rium | +0.305 |
Neg logit contributions | |
unden | -0.455 |
rall | -0.448 |
arrang | -0.420 |
reluct | -0.415 |
fort | -0.410 |
Token | ." |
Token ID | 526 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
ocket | +0.360 |
adiator | +0.337 |
Explore | +0.331 |
replay | +0.330 |
rium | +0.329 |
Neg logit contributions | |
unden | -0.629 |
rall | -0.619 |
arrang | -0.581 |
reluct | -0.578 |
'. | -0.565 |
Token |
|
Token ID | 198 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
ocket | +0.947 |
Explore | +0.901 |
adiator | +0.893 |
replay | +0.883 |
rium | +0.876 |
Neg logit contributions | |
unden | -1.769 |
rall | -1.714 |
reluct | -1.634 |
arrang | -1.632 |
'. | -1.532 |
Token |
|
Token ID | 198 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
ocket | +0.359 |
adiator | +0.341 |
Explore | +0.340 |
replay | +0.334 |
rium | +0.332 |
Neg logit contributions | |
unden | -0.608 |
rall | -0.593 |
arrang | -0.562 |
reluct | -0.559 |
'. | -0.538 |
Token | "… |
Token ID | 24426 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
ocket | +0.336 |
Explore | +0.321 |
adiator | +0.319 |
rium | +0.311 |
replay | +0.310 |
Neg logit contributions | |
unden | -0.562 |
rall | -0.549 |
arrang | -0.519 |
reluct | -0.516 |
'. | -0.504 |
Token | Alright |
Token ID | 31442 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
ocket | +3.056 |
Explore | +2.958 |
rium | +2.901 |
Weak | +2.883 |
adiator | +2.877 |
Neg logit contributions | |
unden | -4.207 |
rall | -4.128 |
arrang | -3.858 |
reluct | -3.855 |
'. | -3.664 |
Token | ." |
Token ID | 526 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
ocket | +1.060 |
adiator | +0.992 |
Explore | +0.980 |
rium | +0.972 |
replay | +0.972 |
Neg logit contributions | |
unden | -1.903 |
rall | -1.864 |
arrang | -1.749 |
reluct | -1.746 |
'. | -1.722 |
Token |
|
Token ID | 198 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
ocket | +0.726 |
Explore | +0.690 |
adiator | +0.685 |
replay | +0.676 |
rium | +0.672 |
Neg logit contributions | |
unden | -1.317 |
rall | -1.279 |
arrang | -1.216 |
reluct | -1.214 |
'. | -1.156 |
Token |
|
Token ID | 198 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ocket | +0.170 |
adiator | +0.162 |
Explore | +0.160 |
replay | +0.158 |
rium | +0.157 |
Neg logit contributions | |
unden | -0.277 |
rall | -0.271 |
arrang | -0.256 |
reluct | -0.254 |
'. | -0.248 |
Token | Ste |
Token ID | 7447 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
ocket | +0.230 |
Explore | +0.219 |
adiator | +0.219 |
rium | +0.213 |
replay | +0.213 |
Neg logit contributions | |
unden | -0.374 |
rall | -0.366 |
arrang | -0.345 |
reluct | -0.343 |
'. | -0.336 |
Token | Del |
Token ID | 4216 |
Feature activation | +6.62e-01 |
Pos logit contributions | |
unden | +0.246 |
rall | +0.240 |
arrang | +0.225 |
reluct | +0.222 |
fort | +0.221 |
Neg logit contributions | |
ocket | -0.204 |
adiator | -0.198 |
Explore | -0.196 |
rium | -0.192 |
replay | -0.192 |
Token | Sur |
Token ID | 4198 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
ocket | +1.366 |
adiator | +1.323 |
Explore | +1.318 |
replay | +1.308 |
rium | +1.302 |
Neg logit contributions | |
unden | -0.878 |
rall | -0.849 |
arrang | -0.782 |
reluct | -0.759 |
'. | -0.739 |
Token | featuring |
Token ID | 9593 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
ocket | +0.127 |
adiator | +0.122 |
Explore | +0.120 |
replay | +0.119 |
rium | +0.118 |
Neg logit contributions | |
unden | -0.195 |
rall | -0.191 |
arrang | -0.180 |
reluct | -0.178 |
fort | -0.176 |
Token | Coach |
Token ID | 16393 |
Feature activation | +7.34e-02 |
Pos logit contributions | |
ocket | +0.205 |
adiator | +0.197 |
replay | +0.195 |
Explore | +0.194 |
rium | +0.193 |
Neg logit contributions | |
unden | -0.243 |
rall | -0.236 |
arrang | -0.222 |
reluct | -0.219 |
fort | -0.215 |
Token | ( |
Token ID | 357 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
ocket | +0.120 |
adiator | +0.116 |
Explore | +0.115 |
replay | +0.114 |
rium | +0.113 |
Neg logit contributions | |
unden | -0.129 |
rall | -0.125 |
arrang | -0.117 |
reluct | -0.116 |
'. | -0.113 |
Token | Surv |
Token ID | 34652 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
unden | +0.226 |
rall | +0.222 |
arrang | +0.209 |
reluct | +0.207 |
fort | +0.204 |
Neg logit contributions | |
ocket | -0.149 |
adiator | -0.142 |
Explore | -0.141 |
replay | -0.139 |
rium | -0.138 |
Token | iv |
Token ID | 452 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
ocket | +2.529 |
adiator | +2.372 |
iversity | +2.335 |
replay | +2.322 |
Explore | +2.317 |
Neg logit contributions | |
unden | -4.226 |
rall | -4.131 |
arrang | -3.896 |
reluct | -3.885 |
'. | -3.753 |
Token | or |
Token ID | 273 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
unden | +0.747 |
rall | +0.726 |
arrang | +0.672 |
reluct | +0.661 |
'. | +0.650 |
Neg logit contributions | |
ocket | -0.881 |
adiator | -0.852 |
Explore | -0.845 |
replay | -0.837 |
rium | -0.834 |
Token | T |
Token ID | 309 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
unden | +0.656 |
rall | +0.643 |
arrang | +0.601 |
reluct | +0.594 |
fort | +0.591 |
Neg logit contributions | |
ocket | -0.503 |
adiator | -0.485 |
Explore | -0.479 |
rium | -0.472 |
paces | -0.471 |
Token | oc |
Token ID | 420 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
ocket | +0.440 |
adiator | +0.423 |
Explore | +0.418 |
replay | +0.412 |
rium | +0.410 |
Neg logit contributions | |
unden | -0.608 |
rall | -0.595 |
arrang | -0.559 |
reluct | -0.552 |
fort | -0.547 |
Token | ant |
Token ID | 415 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
unden | +0.360 |
rall | +0.352 |
arrang | +0.327 |
fort | +0.325 |
reluct | +0.324 |
Neg logit contributions | |
ocket | -0.324 |
adiator | -0.311 |
Explore | -0.310 |
replay | -0.304 |
paces | -0.303 |
Token | card |
Token ID | 2657 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
unden | +0.404 |
rall | +0.395 |
arrang | +0.371 |
reluct | +0.368 |
'. | +0.363 |
Neg logit contributions | |
ocket | -0.284 |
adiator | -0.272 |
Explore | -0.269 |
replay | -0.265 |
rium | -0.264 |
Token | ) |
Token ID | 8 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
ocket | +0.325 |
adiator | +0.310 |
Explore | +0.306 |
replay | +0.302 |
rium | +0.300 |
Neg logit contributions | |
unden | -0.530 |
rall | -0.519 |
arrang | -0.490 |
reluct | -0.485 |
fort | -0.480 |
Token |
|
Token ID | 198 |
Feature activation | +2.77e-01 |
Pos logit contributions | |
unden | +0.370 |
rall | +0.361 |
arrang | +0.341 |
reluct | +0.337 |
'. | +0.333 |
Neg logit contributions | |
ocket | -0.245 |
adiator | -0.235 |
Explore | -0.232 |
replay | -0.230 |
rium | -0.228 |
Token |
|
Token ID | 198 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
ocket | +0.354 |
adiator | +0.337 |
Explore | +0.335 |
replay | +0.330 |
rium | +0.328 |
Neg logit contributions | |
unden | -0.596 |
rall | -0.581 |
arrang | -0.551 |
reluct | -0.547 |
'. | -0.528 |
Token | R |
Token ID | 49 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
ocket | +0.299 |
adiator | +0.286 |
Explore | +0.285 |
replay | +0.278 |
rium | +0.277 |
Neg logit contributions | |
unden | -0.495 |
rall | -0.483 |
arrang | -0.458 |
reluct | -0.455 |
'. | -0.445 |
Token | attle |
Token ID | 1999 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
unden | +0.268 |
rall | +0.262 |
arrang | +0.248 |
reluct | +0.245 |
'. | +0.242 |
Neg logit contributions | |
ocket | -0.172 |
adiator | -0.165 |
Explore | -0.161 |
replay | -0.159 |
rium | -0.159 |
Token | Snake |
Token ID | 49795 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
ocket | +2.807 |
adiator | +2.717 |
Explore | +2.669 |
replay | +2.646 |
rium | +2.627 |
Neg logit contributions | |
unden | -3.656 |
rall | -3.457 |
arrang | -3.359 |
reluct | -3.337 |
'. | -3.158 |
Token | : |
Token ID | 25 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ocket | +0.312 |
adiator | +0.298 |
Explore | +0.295 |
replay | +0.292 |
rium | +0.290 |
Neg logit contributions | |
unden | -0.430 |
rall | -0.419 |
arrang | -0.396 |
reluct | -0.391 |
'. | -0.386 |
Token | La |
Token ID | 4689 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
ocket | +0.350 |
adiator | +0.337 |
Explore | +0.334 |
replay | +0.332 |
rium | +0.328 |
Neg logit contributions | |
unden | -0.415 |
rall | -0.401 |
arrang | -0.379 |
reluct | -0.375 |
fort | -0.366 |
Token | N |
Token ID | 45 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
unden | +0.205 |
rall | +0.199 |
arrang | +0.187 |
reluct | +0.186 |
'. | +0.182 |
Neg logit contributions | |
ocket | -0.162 |
adiator | -0.156 |
Explore | -0.155 |
replay | -0.153 |
rium | -0.152 |
Token | m |
Token ID | 76 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
unden | +0.136 |
rall | +0.132 |
arrang | +0.125 |
reluct | +0.124 |
'. | +0.121 |
Neg logit contributions | |
ocket | -0.099 |
adiator | -0.095 |
Explore | -0.094 |
rium | -0.092 |
replay | -0.092 |
Token | ) |
Token ID | 8 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
ocket | +0.667 |
adiator | +0.628 |
Explore | +0.627 |
replay | +0.625 |
rium | +0.612 |
Neg logit contributions | |
unden | -0.931 |
rall | -0.896 |
arrang | -0.850 |
reluct | -0.847 |
'. | -0.831 |
Token |
|
Token ID | 198 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
unden | +0.595 |
rall | +0.580 |
arrang | +0.550 |
reluct | +0.546 |
'. | +0.536 |
Neg logit contributions | |
ocket | -0.372 |
adiator | -0.353 |
Explore | -0.352 |
replay | -0.343 |
rium | -0.340 |
Token |
|
Token ID | 198 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
unden | +0.120 |
rall | +0.118 |
arrang | +0.111 |
reluct | +0.110 |
'. | +0.108 |
Neg logit contributions | |
ocket | -0.075 |
adiator | -0.072 |
Explore | -0.071 |
replay | -0.070 |
rium | -0.070 |
Token | ( |
Token ID | 7 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
unden | +0.189 |
rall | +0.185 |
arrang | +0.175 |
reluct | +0.173 |
'. | +0.169 |
Neg logit contributions | |
ocket | -0.128 |
Explore | -0.123 |
adiator | -0.122 |
rium | -0.118 |
paces | -0.118 |
Token | K |
Token ID | 42 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
unden | +0.421 |
rall | +0.411 |
arrang | +0.388 |
reluct | +0.384 |
'. | +0.378 |
Neg logit contributions | |
ocket | -0.290 |
adiator | -0.276 |
Explore | -0.275 |
replay | -0.270 |
rium | -0.269 |
Token | ath |
Token ID | 776 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
unden | +0.538 |
rall | +0.525 |
arrang | +0.491 |
reluct | +0.486 |
'. | +0.479 |
Neg logit contributions | |
ocket | -0.460 |
adiator | -0.442 |
Explore | -0.438 |
replay | -0.431 |
rium | -0.430 |
Token | leen |
Token ID | 20042 |
Feature activation | +9.79e-01 |
Pos logit contributions | |
ocket | +1.893 |
adiator | +1.744 |
Explore | +1.700 |
rium | +1.697 |
replay | +1.683 |
Neg logit contributions | |
unden | -4.615 |
rall | -4.475 |
arrang | -4.341 |
reluct | -4.269 |
'. | -4.179 |
Token | Flynn |
Token ID | 14654 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
ocket | +1.409 |
adiator | +1.334 |
replay | +1.328 |
Explore | +1.327 |
rium | +1.305 |
Neg logit contributions | |
unden | -1.920 |
rall | -1.847 |
arrang | -1.774 |
reluct | -1.751 |
'. | -1.696 |
Token | ) |
Token ID | 8 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
ocket | +0.442 |
adiator | +0.418 |
Explore | +0.414 |
replay | +0.408 |
rium | +0.403 |
Neg logit contributions | |
unden | -0.695 |
rall | -0.678 |
arrang | -0.640 |
reluct | -0.638 |
'. | -0.630 |
Token |
|
Token ID | 198 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
unden | +0.337 |
rall | +0.329 |
arrang | +0.311 |
reluct | +0.307 |
'. | +0.302 |
Neg logit contributions | |
ocket | -0.234 |
adiator | -0.221 |
Explore | -0.220 |
replay | -0.215 |
rium | -0.215 |
Token |
|
Token ID | 198 |
Feature activation | +9.23e-02 |
Pos logit contributions | |
ocket | +0.117 |
adiator | +0.111 |
Explore | +0.111 |
replay | +0.109 |
rium | +0.108 |
Neg logit contributions | |
unden | -0.191 |
rall | -0.187 |
arrang | -0.177 |
reluct | -0.175 |
'. | -0.171 |
Token | customers |
Token ID | 4297 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
unden | +0.270 |
rall | +0.262 |
arrang | +0.247 |
reluct | +0.244 |
fort | +0.242 |
Neg logit contributions | |
ocket | -0.210 |
adiator | -0.201 |
Explore | -0.200 |
replay | -0.199 |
rium | -0.196 |
Token | might |
Token ID | 1244 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
unden | +0.664 |
rall | +0.648 |
arrang | +0.610 |
reluct | +0.603 |
fort | +0.595 |
Neg logit contributions | |
ocket | -0.471 |
adiator | -0.451 |
Explore | -0.446 |
replay | -0.441 |
rium | -0.439 |
Token | need |
Token ID | 761 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
ocket | +0.074 |
replay | +0.072 |
adiator | +0.071 |
Explore | +0.071 |
paces | +0.070 |
Neg logit contributions | |
unden | -0.098 |
rall | -0.095 |
arrang | -0.089 |
reluct | -0.088 |
fort | -0.086 |
Token | to |
Token ID | 284 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ocket | +0.263 |
adiator | +0.253 |
replay | +0.251 |
Explore | +0.250 |
rium | +0.244 |
Neg logit contributions | |
unden | -0.411 |
rall | -0.402 |
arrang | -0.379 |
reluct | -0.375 |
'. | -0.370 |
Token | be |
Token ID | 307 |
Feature activation | +9.54e-02 |
Pos logit contributions | |
ocket | +0.187 |
replay | +0.182 |
adiator | +0.179 |
Explore | +0.177 |
paces | +0.176 |
Neg logit contributions | |
unden | -0.284 |
rall | -0.277 |
arrang | -0.260 |
reluct | -0.259 |
fort | -0.253 |
Token | dial |
Token ID | 5980 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
ocket | +0.146 |
adiator | +0.140 |
replay | +0.139 |
Explore | +0.138 |
iversity | +0.136 |
Neg logit contributions | |
unden | -0.177 |
rall | -0.174 |
arrang | -0.162 |
reluct | -0.161 |
fort | -0.158 |
Token | ed |
Token ID | 276 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
ocket | +2.187 |
replay | +2.070 |
adiator | +2.026 |
Explore | +2.019 |
iversity | +1.972 |
Neg logit contributions | |
unden | -4.206 |
rall | -4.113 |
arrang | -3.932 |
reluct | -3.905 |
'. | -3.774 |
Token | back |
Token ID | 736 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
ocket | +0.377 |
adiator | +0.363 |
replay | +0.361 |
Explore | +0.358 |
rium | +0.352 |
Neg logit contributions | |
unden | -0.491 |
rall | -0.481 |
arrang | -0.451 |
reluct | -0.447 |
fort | -0.436 |
Token | . |
Token ID | 13 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
ocket | +0.184 |
adiator | +0.176 |
replay | +0.173 |
Explore | +0.173 |
paces | +0.170 |
Neg logit contributions | |
unden | -0.284 |
rall | -0.278 |
arrang | -0.262 |
reluct | -0.259 |
fort | -0.253 |
Token | In |
Token ID | 554 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
unden | +0.079 |
rall | +0.077 |
arrang | +0.073 |
reluct | +0.072 |
fort | +0.070 |
Neg logit contributions | |
ocket | -0.051 |
Explore | -0.049 |
adiator | -0.049 |
replay | -0.048 |
rium | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | +8.73e-03 |
Pos logit contributions | |
unden | +0.258 |
rall | +0.250 |
arrang | +0.235 |
reluct | +0.233 |
'. | +0.228 |
Neg logit contributions | |
ocket | -0.223 |
adiator | -0.214 |
Explore | -0.211 |
replay | -0.211 |
rium | -0.208 |
Token | addition |
Token ID | 3090 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
ocket | +0.150 |
replay | +0.144 |
adiator | +0.144 |
Explore | +0.143 |
rium | +0.141 |
Neg logit contributions | |
unden | -0.187 |
rall | -0.182 |
arrang | -0.171 |
reluct | -0.170 |
fort | -0.167 |
Token | of |
Token ID | 286 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
unden | +0.335 |
rall | +0.328 |
arrang | +0.311 |
reluct | +0.308 |
fort | +0.304 |
Neg logit contributions | |
ocket | -0.180 |
adiator | -0.172 |
Explore | -0.169 |
replay | -0.167 |
rium | -0.166 |
Token | crisp |
Token ID | 23453 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
ocket | +0.131 |
replay | +0.128 |
adiator | +0.127 |
Explore | +0.126 |
rium | +0.124 |
Neg logit contributions | |
unden | -0.164 |
rall | -0.159 |
arrang | -0.150 |
reluct | -0.149 |
fort | -0.146 |
Token | high |
Token ID | 1029 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
ocket | +1.158 |
replay | +1.116 |
adiator | +1.111 |
Explore | +1.095 |
iversity | +1.072 |
Neg logit contributions | |
unden | -1.848 |
rall | -1.804 |
reluct | -1.706 |
arrang | -1.705 |
'. | -1.663 |
Token | - |
Token ID | 12 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
ocket | +0.189 |
adiator | +0.179 |
replay | +0.177 |
Explore | +0.176 |
rium | +0.172 |
Neg logit contributions | |
unden | -0.438 |
rall | -0.431 |
arrang | -0.409 |
reluct | -0.406 |
fort | -0.400 |
Token | definition |
Token ID | 46758 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ocket | +0.248 |
adiator | +0.235 |
Explore | +0.232 |
replay | +0.231 |
rium | +0.228 |
Neg logit contributions | |
unden | -0.564 |
rall | -0.556 |
arrang | -0.526 |
reluct | -0.524 |
'. | -0.511 |
Token | graphics |
Token ID | 9382 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
replay | +2.125 |
ocket | +2.075 |
adiator | +1.984 |
Explore | +1.956 |
rium | +1.898 |
Neg logit contributions | |
unden | -4.115 |
rall | -4.002 |
reluct | -3.856 |
arrang | -3.808 |
fort | -3.727 |
Token | and |
Token ID | 290 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
unden | +0.274 |
rall | +0.269 |
arrang | +0.252 |
reluct | +0.252 |
'. | +0.247 |
Neg logit contributions | |
ocket | -0.176 |
adiator | -0.169 |
Explore | -0.168 |
replay | -0.167 |
paces | -0.165 |
Token | Wii |
Token ID | 16591 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
ocket | +0.095 |
replay | +0.094 |
adiator | +0.092 |
Explore | +0.092 |
rium | +0.090 |
Neg logit contributions | |
unden | -0.114 |
rall | -0.111 |
arrang | -0.104 |
reluct | -0.104 |
fort | -0.101 |
Token | Motion |
Token ID | 20843 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
unden | +0.206 |
rall | +0.202 |
arrang | +0.189 |
reluct | +0.187 |
fort | +0.185 |
Neg logit contributions | |
ocket | -0.155 |
adiator | -0.148 |
Explore | -0.147 |
replay | -0.145 |
rium | -0.144 |
Token | Plus |
Token ID | 17860 |
Feature activation | +6.05e-01 |
Pos logit contributions | |
ocket | +0.763 |
adiator | +0.732 |
Explore | +0.732 |
replay | +0.725 |
rium | +0.715 |
Neg logit contributions | |
unden | -0.961 |
rall | -0.938 |
arrang | -0.882 |
reluct | -0.873 |
'. | -0.855 |
Token |
|
Token ID | 198 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
ocket | +1.049 |
adiator | +0.979 |
Explore | +0.975 |
rium | +0.968 |
replay | +0.963 |
Neg logit contributions | |
unden | -1.833 |
rall | -1.799 |
arrang | -1.696 |
reluct | -1.674 |
'. | -1.662 |
Token |
|
Token ID | 198 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
ocket | +0.421 |
adiator | +0.400 |
Explore | +0.399 |
replay | +0.392 |
rium | +0.390 |
Neg logit contributions | |
unden | -0.712 |
rall | -0.695 |
arrang | -0.658 |
reluct | -0.654 |
'. | -0.631 |
Token | 「 |
Token ID | 13697 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
ocket | +0.447 |
Explore | +0.429 |
adiator | +0.425 |
rium | +0.417 |
replay | +0.414 |
Neg logit contributions | |
unden | -0.677 |
rall | -0.662 |
arrang | -0.624 |
reluct | -0.620 |
'. | -0.604 |
Token | Oh |
Token ID | 5812 |
Feature activation | +5.68e-01 |
Pos logit contributions | |
ocket | +0.029 |
adiator | +0.028 |
Explore | +0.028 |
replay | +0.027 |
rium | +0.027 |
Neg logit contributions | |
unden | -0.047 |
rall | -0.046 |
arrang | -0.043 |
reluct | -0.043 |
fort | -0.042 |
Token | my |
Token ID | 616 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
ocket | +0.724 |
adiator | +0.688 |
Explore | +0.676 |
rium | +0.670 |
replay | +0.668 |
Neg logit contributions | |
unden | -1.201 |
rall | -1.180 |
arrang | -1.109 |
reluct | -1.103 |
'. | -1.084 |
Token | …. |
Token ID | 11580 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ocket | +0.174 |
adiator | +0.166 |
Explore | +0.164 |
replay | +0.162 |
rium | +0.162 |
Neg logit contributions | |
unden | -0.260 |
rall | -0.254 |
arrang | -0.239 |
reluct | -0.237 |
'. | -0.234 |
Token | How |
Token ID | 1374 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
ocket | +2.656 |
Explore | +2.551 |
adiator | +2.531 |
rium | +2.520 |
paces | +2.493 |
Neg logit contributions | |
unden | -3.491 |
rall | -3.416 |
arrang | -3.173 |
reluct | -3.169 |
'. | -3.084 |
Token | bad |
Token ID | 2089 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
unden | +0.308 |
rall | +0.302 |
arrang | +0.280 |
reluct | +0.278 |
'. | +0.273 |
Neg logit contributions | |
ocket | -0.234 |
adiator | -0.222 |
Explore | -0.221 |
replay | -0.220 |
rium | -0.218 |
Token | is |
Token ID | 318 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
unden | +1.128 |
rall | +1.104 |
arrang | +1.042 |
reluct | +1.030 |
fort | +1.019 |
Neg logit contributions | |
ocket | -0.743 |
adiator | -0.712 |
Explore | -0.703 |
replay | -0.693 |
rium | -0.690 |
Token | it |
Token ID | 340 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
ocket | +0.128 |
adiator | +0.120 |
Explore | +0.120 |
replay | +0.118 |
rium | +0.118 |
Neg logit contributions | |
unden | -0.212 |
rall | -0.208 |
arrang | -0.195 |
reluct | -0.194 |
fort | -0.191 |
Token | ?」 |
Token ID | 42943 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
ocket | +0.111 |
adiator | +0.104 |
Explore | +0.103 |
rium | +0.102 |
replay | +0.102 |
Neg logit contributions | |
unden | -0.191 |
rall | -0.187 |
arrang | -0.175 |
reluct | -0.175 |
'. | -0.172 |
Token | streaming |
Token ID | 11305 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
ocket | +0.142 |
replay | +0.137 |
adiator | +0.136 |
Explore | +0.136 |
rium | +0.134 |
Neg logit contributions | |
unden | -0.154 |
rall | -0.149 |
arrang | -0.138 |
reluct | -0.138 |
fort | -0.135 |
Token | five |
Token ID | 1936 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
ocket | +0.732 |
replay | +0.713 |
adiator | +0.707 |
Explore | +0.704 |
paces | +0.688 |
Neg logit contributions | |
unden | -0.937 |
rall | -0.911 |
reluct | -0.861 |
arrang | -0.854 |
fort | -0.832 |
Token | videos |
Token ID | 5861 |
Feature activation | +8.24e-01 |
Pos logit contributions | |
ocket | +0.035 |
adiator | +0.034 |
replay | +0.034 |
Explore | +0.034 |
paces | +0.033 |
Neg logit contributions | |
unden | -0.042 |
rall | -0.041 |
arrang | -0.039 |
reluct | -0.038 |
fort | -0.038 |
Token | of |
Token ID | 286 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
ocket | +1.150 |
replay | +1.095 |
adiator | +1.089 |
paces | +1.083 |
Explore | +1.074 |
Neg logit contributions | |
unden | -1.650 |
rall | -1.608 |
reluct | -1.515 |
arrang | -1.513 |
fort | -1.451 |
Token | high |
Token ID | 1029 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
unden | +0.068 |
rall | +0.066 |
reluct | +0.063 |
arrang | +0.063 |
fort | +0.061 |
Neg logit contributions | |
ocket | -0.053 |
replay | -0.052 |
adiator | -0.052 |
Explore | -0.051 |
rium | -0.051 |
Token | definition |
Token ID | 6770 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
ocket | +0.289 |
replay | +0.283 |
adiator | +0.275 |
Explore | +0.273 |
iversity | +0.268 |
Neg logit contributions | |
unden | -0.564 |
rall | -0.557 |
arrang | -0.529 |
reluct | -0.525 |
fort | -0.513 |
Token | video |
Token ID | 2008 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
replay | +2.488 |
ocket | +2.436 |
adiator | +2.362 |
Explore | +2.352 |
crawl | +2.307 |
Neg logit contributions | |
unden | -3.561 |
rall | -3.472 |
reluct | -3.338 |
arrang | -3.295 |
fort | -3.209 |
Token | a |
Token ID | 257 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
ocket | +0.899 |
replay | +0.875 |
adiator | +0.863 |
Explore | +0.855 |
paces | +0.845 |
Neg logit contributions | |
unden | -1.247 |
rall | -1.224 |
reluct | -1.145 |
arrang | -1.143 |
fort | -1.114 |
Token | day |
Token ID | 1110 |
Feature activation | -7.43e-02 |
Pos logit contributions | |
ocket | +0.210 |
replay | +0.200 |
adiator | +0.200 |
Explore | +0.198 |
rium | +0.196 |
Neg logit contributions | |
unden | -0.255 |
rall | -0.247 |
arrang | -0.234 |
reluct | -0.230 |
fort | -0.225 |
Token | . |
Token ID | 13 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
unden | +0.154 |
rall | +0.151 |
arrang | +0.142 |
reluct | +0.141 |
fort | +0.138 |
Neg logit contributions | |
ocket | -0.098 |
adiator | -0.094 |
replay | -0.092 |
Explore | -0.092 |
paces | -0.091 |
Token |
|
Token ID | 198 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
ocket | +0.265 |
Explore | +0.253 |
adiator | +0.252 |
replay | +0.250 |
rium | +0.246 |
Neg logit contributions | |
unden | -0.391 |
rall | -0.382 |
arrang | -0.359 |
reluct | -0.357 |
fort | -0.345 |
Token | mon |
Token ID | 937 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
ocket | +0.094 |
adiator | +0.090 |
replay | +0.090 |
Explore | +0.089 |
rium | +0.088 |
Neg logit contributions | |
unden | -0.114 |
rall | -0.111 |
arrang | -0.104 |
reluct | -0.103 |
fort | -0.101 |
Token | ologue |
Token ID | 39795 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
ocket | +0.033 |
adiator | +0.032 |
replay | +0.032 |
rium | +0.032 |
Explore | +0.031 |
Neg logit contributions | |
unden | -0.028 |
rall | -0.027 |
arrang | -0.025 |
reluct | -0.025 |
'. | -0.024 |
Token | given |
Token ID | 1813 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
unden | +0.602 |
rall | +0.589 |
arrang | +0.554 |
reluct | +0.547 |
fort | +0.541 |
Neg logit contributions | |
ocket | -0.433 |
adiator | -0.416 |
Explore | -0.411 |
replay | -0.405 |
rium | -0.404 |
Token | by |
Token ID | 416 |
Feature activation | -1.70e-03 |
Pos logit contributions | |
unden | +0.096 |
rall | +0.094 |
arrang | +0.088 |
reluct | +0.087 |
'. | +0.086 |
Neg logit contributions | |
ocket | -0.072 |
adiator | -0.069 |
Explore | -0.068 |
replay | -0.068 |
rium | -0.067 |
Token | Jimmy |
Token ID | 12963 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
unden | +0.003 |
rall | +0.003 |
arrang | +0.003 |
reluct | +0.003 |
fort | +0.003 |
Neg logit contributions | |
ocket | -0.003 |
adiator | -0.003 |
Explore | -0.003 |
replay | -0.003 |
rium | -0.003 |
Token | Kimmel |
Token ID | 46552 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
ocket | +0.457 |
adiator | +0.421 |
replay | +0.413 |
Explore | +0.407 |
iversity | +0.404 |
Neg logit contributions | |
unden | -1.181 |
rall | -1.159 |
arrang | -1.113 |
reluct | -1.100 |
'. | -1.073 |
Token | concerning |
Token ID | 9305 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ocket | +2.537 |
adiator | +2.398 |
replay | +2.386 |
Explore | +2.356 |
iversity | +2.321 |
Neg logit contributions | |
unden | -3.596 |
rall | -3.506 |
arrang | -3.291 |
reluct | -3.253 |
'. | -3.170 |
Token | his |
Token ID | 465 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
ocket | +0.230 |
adiator | +0.221 |
Explore | +0.218 |
replay | +0.218 |
rium | +0.216 |
Neg logit contributions | |
unden | -0.302 |
rall | -0.295 |
arrang | -0.277 |
reluct | -0.274 |
fort | -0.271 |
Token | newborn |
Token ID | 23856 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
ocket | +0.227 |
adiator | +0.217 |
replay | +0.217 |
Explore | +0.215 |
rium | +0.212 |
Neg logit contributions | |
unden | -0.296 |
rall | -0.289 |
arrang | -0.270 |
reluct | -0.267 |
fort | -0.265 |
Token | son |
Token ID | 3367 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
ocket | +1.015 |
adiator | +0.972 |
replay | +0.965 |
Explore | +0.952 |
rium | +0.948 |
Neg logit contributions | |
unden | -1.456 |
rall | -1.432 |
arrang | -1.338 |
reluct | -1.328 |
'. | -1.306 |
Token | . |
Token ID | 13 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
unden | +0.389 |
rall | +0.381 |
arrang | +0.358 |
reluct | +0.354 |
'. | +0.349 |
Neg logit contributions | |
ocket | -0.282 |
adiator | -0.269 |
Explore | -0.265 |
replay | -0.264 |
rium | -0.262 |
Token | and |
Token ID | 290 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
ocket | +0.150 |
adiator | +0.143 |
Explore | +0.141 |
replay | +0.141 |
rium | +0.139 |
Neg logit contributions | |
unden | -0.210 |
rall | -0.205 |
arrang | -0.193 |
reluct | -0.191 |
fort | -0.189 |
Token | spread |
Token ID | 4104 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
unden | +0.381 |
rall | +0.372 |
arrang | +0.348 |
reluct | +0.344 |
fort | +0.340 |
Neg logit contributions | |
ocket | -0.310 |
adiator | -0.298 |
Explore | -0.295 |
replay | -0.292 |
rium | -0.291 |
Token | them |
Token ID | 606 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
ocket | +0.770 |
adiator | +0.736 |
Explore | +0.729 |
replay | +0.726 |
rium | +0.722 |
Neg logit contributions | |
unden | -1.159 |
rall | -1.122 |
arrang | -1.060 |
reluct | -1.057 |
fort | -1.037 |
Token | near |
Token ID | 1474 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
unden | +0.228 |
rall | +0.223 |
arrang | +0.208 |
reluct | +0.207 |
fort | +0.203 |
Neg logit contributions | |
ocket | -0.182 |
adiator | -0.174 |
Explore | -0.172 |
replay | -0.172 |
rium | -0.169 |
Token | the |
Token ID | 262 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ocket | +0.483 |
adiator | +0.460 |
Explore | +0.458 |
rium | +0.456 |
replay | +0.455 |
Neg logit contributions | |
unden | -0.597 |
rall | -0.581 |
arrang | -0.547 |
reluct | -0.539 |
fort | -0.532 |
Token | sunk |
Token ID | 24790 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
ocket | +0.053 |
adiator | +0.051 |
Explore | +0.050 |
replay | +0.050 |
rium | +0.050 |
Neg logit contributions | |
unden | -0.071 |
rall | -0.070 |
arrang | -0.066 |
reluct | -0.065 |
fort | -0.064 |
Token | en |
Token ID | 268 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
ocket | +3.049 |
adiator | +2.919 |
rium | +2.881 |
Explore | +2.862 |
iversity | +2.856 |
Neg logit contributions | |
unden | -3.125 |
rall | -3.057 |
arrang | -2.857 |
reluct | -2.831 |
'. | -2.723 |
Token | fireplace |
Token ID | 44250 |
Feature activation | +6.70e-01 |
Pos logit contributions | |
ocket | +0.546 |
adiator | +0.517 |
Explore | +0.507 |
replay | +0.505 |
rium | +0.501 |
Neg logit contributions | |
unden | -0.964 |
rall | -0.946 |
arrang | -0.892 |
reluct | -0.885 |
'. | -0.874 |
Token | . |
Token ID | 13 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
ocket | +0.852 |
adiator | +0.797 |
Explore | +0.790 |
replay | +0.787 |
rium | +0.781 |
Neg logit contributions | |
unden | -1.425 |
rall | -1.398 |
reluct | -1.309 |
arrang | -1.309 |
'. | -1.285 |
Token | After |
Token ID | 2293 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
ocket | +0.212 |
Explore | +0.201 |
adiator | +0.199 |
replay | +0.198 |
rium | +0.197 |
Neg logit contributions | |
unden | -0.316 |
rall | -0.308 |
arrang | -0.290 |
reluct | -0.289 |
fort | -0.279 |
Token | that |
Token ID | 326 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
unden | +0.070 |
rall | +0.068 |
arrang | +0.064 |
reluct | +0.064 |
'. | +0.063 |
Neg logit contributions | |
ocket | -0.051 |
adiator | -0.049 |
replay | -0.049 |
Explore | -0.048 |
rium | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -3.24e-02 |
Pos logit contributions | |
unden | +0.320 |
rall | +0.314 |
arrang | +0.296 |
reluct | +0.292 |
fort | +0.291 |
Neg logit contributions | |
ocket | -0.207 |
adiator | -0.198 |
Explore | -0.195 |
replay | -0.192 |
rium | -0.192 |
Token | Chart |
Token ID | 45488 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
unden | +0.067 |
rall | +0.066 |
arrang | +0.062 |
reluct | +0.062 |
'. | +0.061 |
Neg logit contributions | |
ocket | -0.042 |
adiator | -0.040 |
Explore | -0.040 |
replay | -0.039 |
rium | -0.039 |
Token | ered |
Token ID | 1068 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
ocket | +0.178 |
adiator | +0.171 |
Explore | +0.171 |
replay | +0.169 |
rium | +0.169 |
Neg logit contributions | |
unden | -0.271 |
rall | -0.265 |
arrang | -0.249 |
reluct | -0.246 |
'. | -0.243 |
Token | Institute |
Token ID | 5136 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
unden | +0.220 |
rall | +0.216 |
arrang | +0.202 |
reluct | +0.200 |
fort | +0.197 |
Neg logit contributions | |
ocket | -0.159 |
Explore | -0.151 |
adiator | -0.151 |
replay | -0.150 |
rium | -0.148 |
Token | of |
Token ID | 286 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
unden | +1.280 |
rall | +1.253 |
arrang | +1.183 |
reluct | +1.170 |
'. | +1.156 |
Neg logit contributions | |
ocket | -0.786 |
adiator | -0.750 |
Explore | -0.742 |
replay | -0.732 |
rium | -0.727 |
Token | Personnel |
Token ID | 41623 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
ocket | +0.370 |
adiator | +0.352 |
Explore | +0.351 |
rium | +0.347 |
replay | +0.345 |
Neg logit contributions | |
unden | -0.576 |
rall | -0.564 |
arrang | -0.533 |
reluct | -0.529 |
fort | -0.518 |
Token | and |
Token ID | 290 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
ocket | +2.402 |
Explore | +2.299 |
replay | +2.260 |
rium | +2.231 |
iversity | +2.225 |
Neg logit contributions | |
unden | -3.644 |
rall | -3.587 |
arrang | -3.356 |
reluct | -3.333 |
fort | -3.245 |
Token | Development |
Token ID | 7712 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
ocket | +0.282 |
adiator | +0.270 |
Explore | +0.268 |
replay | +0.266 |
rium | +0.265 |
Neg logit contributions | |
unden | -0.351 |
rall | -0.344 |
arrang | -0.321 |
reluct | -0.319 |
fort | -0.313 |
Token | " |
Token ID | 366 |
Feature activation | -9.29e-02 |
Pos logit contributions | |
unden | +0.639 |
rall | +0.626 |
arrang | +0.587 |
reluct | +0.581 |
fort | +0.574 |
Neg logit contributions | |
ocket | -0.450 |
adiator | -0.430 |
Explore | -0.427 |
replay | -0.422 |
rium | -0.419 |
Token | Overall |
Token ID | 16350 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
unden | +0.202 |
rall | +0.198 |
arrang | +0.188 |
reluct | +0.186 |
fort | +0.183 |
Neg logit contributions | |
ocket | -0.112 |
adiator | -0.107 |
Explore | -0.106 |
replay | -0.104 |
rium | -0.103 |
Token | , |
Token ID | 11 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
ocket | +0.467 |
adiator | +0.449 |
Explore | +0.443 |
replay | +0.440 |
iversity | +0.436 |
Neg logit contributions | |
unden | -0.681 |
rall | -0.669 |
arrang | -0.627 |
reluct | -0.623 |
fort | -0.615 |
Token |
|
Token ID | 198 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
ocket | +0.226 |
adiator | +0.216 |
Explore | +0.214 |
replay | +0.211 |
rium | +0.210 |
Neg logit contributions | |
unden | -0.378 |
rall | -0.369 |
arrang | -0.349 |
reluct | -0.347 |
'. | -0.336 |
Token | 7 |
Token ID | 22 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
ocket | +0.263 |
adiator | +0.250 |
Explore | +0.249 |
replay | +0.242 |
rium | +0.241 |
Neg logit contributions | |
unden | -0.500 |
rall | -0.490 |
arrang | -0.465 |
reluct | -0.461 |
'. | -0.454 |
Token | . |
Token ID | 13 |
Feature activation | -5.02e-02 |
Pos logit contributions | |
ocket | +0.122 |
adiator | +0.116 |
Explore | +0.115 |
replay | +0.113 |
rium | +0.112 |
Neg logit contributions | |
unden | -0.219 |
rall | -0.214 |
arrang | -0.203 |
reluct | -0.203 |
fort | -0.194 |
Token | Dw |
Token ID | 19113 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
unden | +0.087 |
rall | +0.085 |
arrang | +0.079 |
reluct | +0.079 |
fort | +0.076 |
Neg logit contributions | |
ocket | -0.084 |
adiator | -0.081 |
Explore | -0.080 |
rium | -0.079 |
replay | -0.079 |
Token | ark |
Token ID | 668 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
ocket | +0.687 |
adiator | +0.660 |
Explore | +0.646 |
rium | +0.642 |
iversity | +0.639 |
Neg logit contributions | |
unden | -0.956 |
rall | -0.937 |
arrang | -0.881 |
reluct | -0.874 |
'. | -0.859 |
Token | adh |
Token ID | 24411 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ocket | +0.244 |
adiator | +0.231 |
Explore | +0.226 |
rium | +0.223 |
replay | +0.222 |
Neg logit contributions | |
unden | -0.478 |
rall | -0.468 |
arrang | -0.444 |
reluct | -0.440 |
'. | -0.435 |
Token | ish |
Token ID | 680 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
ocket | +2.544 |
adiator | +2.413 |
rium | +2.402 |
Explore | +2.348 |
iversity | +2.332 |
Neg logit contributions | |
unden | -3.514 |
rall | -3.452 |
arrang | -3.227 |
reluct | -3.209 |
'. | -3.121 |
Token | Temple |
Token ID | 10857 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
ocket | +0.779 |
adiator | +0.747 |
rium | +0.737 |
Explore | +0.735 |
replay | +0.729 |
Neg logit contributions | |
unden | -0.923 |
rall | -0.907 |
arrang | -0.843 |
reluct | -0.837 |
'. | -0.817 |
Token |
|
Token ID | 198 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
ocket | +0.367 |
adiator | +0.350 |
Explore | +0.345 |
rium | +0.341 |
replay | +0.339 |
Neg logit contributions | |
unden | -0.601 |
rall | -0.589 |
arrang | -0.556 |
reluct | -0.552 |
'. | -0.541 |
Token |
|
Token ID | 198 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
ocket | +0.504 |
adiator | +0.479 |
Explore | +0.478 |
replay | +0.470 |
rium | +0.467 |
Neg logit contributions | |
unden | -0.863 |
rall | -0.842 |
arrang | -0.798 |
reluct | -0.794 |
'. | -0.762 |
Token | Address |
Token ID | 20231 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
ocket | +0.608 |
Explore | +0.585 |
adiator | +0.583 |
rium | +0.571 |
replay | +0.569 |
Neg logit contributions | |
unden | -0.762 |
rall | -0.743 |
arrang | -0.697 |
reluct | -0.695 |
'. | -0.670 |
Token | May |
Token ID | 1737 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
ocket | +0.431 |
adiator | +0.416 |
Explore | +0.414 |
replay | +0.411 |
iversity | +0.405 |
Neg logit contributions | |
unden | -0.501 |
rall | -0.489 |
arrang | -0.459 |
reluct | -0.456 |
'. | -0.443 |
Token | 5 |
Token ID | 642 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
ocket | +0.308 |
adiator | +0.295 |
Explore | +0.292 |
rium | +0.288 |
replay | +0.288 |
Neg logit contributions | |
unden | -0.401 |
rall | -0.392 |
arrang | -0.369 |
reluct | -0.365 |
'. | -0.358 |
Token | ). |
Token ID | 737 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
ocket | +0.362 |
adiator | +0.343 |
Explore | +0.339 |
replay | +0.336 |
iversity | +0.332 |
Neg logit contributions | |
unden | -0.631 |
rall | -0.616 |
arrang | -0.585 |
reluct | -0.579 |
'. | -0.569 |
Token | The |
Token ID | 383 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
ocket | +0.308 |
Explore | +0.293 |
adiator | +0.293 |
replay | +0.289 |
rium | +0.287 |
Neg logit contributions | |
unden | -0.443 |
rall | -0.433 |
arrang | -0.408 |
reluct | -0.406 |
fort | -0.392 |
Token | most |
Token ID | 749 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
ocket | +0.295 |
adiator | +0.282 |
replay | +0.281 |
Explore | +0.280 |
rium | +0.277 |
Neg logit contributions | |
unden | -0.334 |
rall | -0.326 |
arrang | -0.304 |
reluct | -0.300 |
fort | -0.295 |
Token | recent |
Token ID | 2274 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ocket | +0.321 |
adiator | +0.306 |
replay | +0.305 |
Explore | +0.304 |
rium | +0.300 |
Neg logit contributions | |
unden | -0.387 |
rall | -0.380 |
arrang | -0.355 |
reluct | -0.352 |
fort | -0.347 |
Token | example |
Token ID | 1672 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
ocket | +2.638 |
replay | +2.553 |
Explore | +2.503 |
adiator | +2.480 |
rium | +2.444 |
Neg logit contributions | |
unden | -3.339 |
rall | -3.234 |
reluct | -3.014 |
arrang | -3.011 |
fort | -2.960 |
Token | of |
Token ID | 286 |
Feature activation | +5.49e-02 |
Pos logit contributions | |
ocket | +0.642 |
adiator | +0.611 |
Explore | +0.603 |
replay | +0.598 |
rium | +0.593 |
Neg logit contributions | |
unden | -1.039 |
rall | -1.018 |
arrang | -0.960 |
reluct | -0.950 |
'. | -0.938 |
Token | the |
Token ID | 262 |
Feature activation | +8.82e-02 |
Pos logit contributions | |
ocket | +0.082 |
adiator | +0.079 |
replay | +0.078 |
Explore | +0.078 |
rium | +0.077 |
Neg logit contributions | |
unden | -0.103 |
rall | -0.101 |
arrang | -0.095 |
reluct | -0.094 |
fort | -0.093 |
Token | wrong |
Token ID | 2642 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
ocket | +0.145 |
replay | +0.139 |
adiator | +0.139 |
Explore | +0.137 |
rium | +0.136 |
Neg logit contributions | |
unden | -0.153 |
rall | -0.150 |
arrang | -0.139 |
reluct | -0.137 |
fort | -0.136 |
Token | ness |
Token ID | 1108 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
ocket | +0.070 |
adiator | +0.066 |
replay | +0.066 |
Explore | +0.065 |
rium | +0.065 |
Neg logit contributions | |
unden | -0.102 |
rall | -0.099 |
arrang | -0.093 |
reluct | -0.093 |
'. | -0.091 |
Token | up |
Token ID | 510 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
unden | +0.037 |
rall | +0.036 |
arrang | +0.034 |
reluct | +0.033 |
fort | +0.033 |
Neg logit contributions | |
ocket | -0.027 |
adiator | -0.026 |
Explore | -0.026 |
replay | -0.026 |
rium | -0.025 |
Token | but |
Token ID | 475 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
ocket | +0.014 |
adiator | +0.013 |
replay | +0.013 |
Explore | +0.013 |
paces | +0.013 |
Neg logit contributions | |
unden | -0.021 |
rall | -0.020 |
arrang | -0.019 |
reluct | -0.019 |
fort | -0.019 |
Token | I |
Token ID | 314 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
unden | +0.086 |
rall | +0.084 |
arrang | +0.079 |
reluct | +0.078 |
fort | +0.078 |
Neg logit contributions | |
ocket | -0.065 |
adiator | -0.062 |
replay | -0.061 |
Explore | -0.061 |
rium | -0.060 |
Token | would |
Token ID | 561 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
ocket | +0.168 |
replay | +0.163 |
Explore | +0.159 |
adiator | +0.159 |
rium | +0.157 |
Neg logit contributions | |
unden | -0.226 |
rall | -0.220 |
arrang | -0.205 |
reluct | -0.202 |
fort | -0.199 |
Token | still |
Token ID | 991 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ocket | +0.243 |
replay | +0.237 |
adiator | +0.231 |
Explore | +0.231 |
paces | +0.228 |
Neg logit contributions | |
unden | -0.302 |
rall | -0.293 |
arrang | -0.274 |
reluct | -0.272 |
fort | -0.264 |
Token | recommend |
Token ID | 4313 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ocket | +0.181 |
replay | +0.174 |
adiator | +0.172 |
Explore | +0.172 |
paces | +0.169 |
Neg logit contributions | |
unden | -0.212 |
rall | -0.205 |
arrang | -0.192 |
reluct | -0.191 |
fort | -0.187 |
Token | we |
Token ID | 356 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
ocket | +2.419 |
replay | +2.318 |
adiator | +2.281 |
Explore | +2.258 |
iversity | +2.204 |
Neg logit contributions | |
unden | -3.505 |
rall | -3.397 |
arrang | -3.215 |
reluct | -3.192 |
fort | -3.168 |
Token | sign |
Token ID | 1051 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
ocket | +0.481 |
replay | +0.463 |
Explore | +0.455 |
adiator | +0.454 |
paces | +0.444 |
Neg logit contributions | |
unden | -0.716 |
rall | -0.695 |
arrang | -0.655 |
reluct | -0.654 |
fort | -0.644 |
Token | him |
Token ID | 683 |
Feature activation | +3.64e-03 |
Pos logit contributions | |
unden | +0.485 |
rall | +0.473 |
arrang | +0.445 |
reluct | +0.443 |
fort | +0.436 |
Neg logit contributions | |
ocket | -0.332 |
adiator | -0.318 |
Explore | -0.313 |
replay | -0.312 |
rium | -0.306 |
Token | for |
Token ID | 329 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
ocket | +0.005 |
adiator | +0.004 |
replay | +0.004 |
Explore | +0.004 |
paces | +0.004 |
Neg logit contributions | |
unden | -0.008 |
rall | -0.007 |
arrang | -0.007 |
reluct | -0.007 |
fort | -0.007 |
Token | up |
Token ID | 510 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
ocket | +0.076 |
adiator | +0.073 |
replay | +0.073 |
Explore | +0.073 |
rium | +0.071 |
Neg logit contributions | |
unden | -0.098 |
rall | -0.096 |
arrang | -0.090 |
reluct | -0.089 |
fort | -0.088 |
Token | chain |
Token ID | 6333 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
ocket | +0.391 |
Explore | +0.371 |
adiator | +0.371 |
replay | +0.368 |
rium | +0.363 |
Neg logit contributions | |
unden | -0.511 |
rall | -0.500 |
reluct | -0.468 |
arrang | -0.467 |
'. | -0.455 |
Token | . |
Token ID | 13 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
ocket | +0.338 |
adiator | +0.322 |
Explore | +0.319 |
replay | +0.315 |
paces | +0.314 |
Neg logit contributions | |
unden | -0.556 |
rall | -0.548 |
arrang | -0.515 |
reluct | -0.510 |
fort | -0.498 |
Token |
|
Token ID | 198 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
ocket | +0.032 |
adiator | +0.031 |
Explore | +0.031 |
replay | +0.030 |
rium | +0.030 |
Neg logit contributions | |
unden | -0.045 |
rall | -0.044 |
arrang | -0.042 |
reluct | -0.041 |
fort | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
unden | +0.288 |
rall | +0.282 |
arrang | +0.266 |
reluct | +0.262 |
fort | +0.262 |
Neg logit contributions | |
ocket | -0.186 |
adiator | -0.179 |
Explore | -0.176 |
replay | -0.174 |
rium | -0.173 |
Token | M |
Token ID | 44 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
unden | +0.349 |
rall | +0.342 |
arrang | +0.324 |
reluct | +0.320 |
fort | +0.319 |
Neg logit contributions | |
ocket | -0.189 |
adiator | -0.180 |
Explore | -0.176 |
replay | -0.175 |
rium | -0.174 |
Token | ounds |
Token ID | 3733 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ocket | +0.239 |
adiator | +0.229 |
Explore | +0.225 |
rium | +0.221 |
replay | +0.221 |
Neg logit contributions | |
unden | -0.384 |
rall | -0.376 |
arrang | -0.356 |
reluct | -0.352 |
'. | -0.347 |
Token | of |
Token ID | 286 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
ocket | +2.093 |
Explore | +1.991 |
adiator | +1.972 |
paces | +1.950 |
replay | +1.943 |
Neg logit contributions | |
unden | -3.787 |
rall | -3.686 |
arrang | -3.514 |
reluct | -3.490 |
fort | -3.375 |
Token | tail |
Token ID | 7894 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
ocket | +0.050 |
adiator | +0.049 |
Explore | +0.048 |
replay | +0.048 |
rium | +0.047 |
Neg logit contributions | |
unden | -0.061 |
rall | -0.060 |
arrang | -0.056 |
reluct | -0.056 |
fort | -0.055 |
Token | ings |
Token ID | 654 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
unden | +0.557 |
rall | +0.544 |
arrang | +0.512 |
reluct | +0.506 |
fort | +0.499 |
Neg logit contributions | |
ocket | -0.409 |
adiator | -0.392 |
Explore | -0.388 |
replay | -0.383 |
rium | -0.381 |
Token | from |
Token ID | 422 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
ocket | +0.025 |
adiator | +0.024 |
Explore | +0.024 |
replay | +0.024 |
paces | +0.023 |
Neg logit contributions | |
unden | -0.033 |
rall | -0.032 |
arrang | -0.030 |
reluct | -0.030 |
'. | -0.029 |
Token | mineral |
Token ID | 18352 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
ocket | +0.087 |
adiator | +0.085 |
Explore | +0.083 |
replay | +0.082 |
rium | +0.082 |
Neg logit contributions | |
unden | -0.105 |
rall | -0.102 |
arrang | -0.097 |
reluct | -0.095 |
fort | -0.094 |
Token | secession |
Token ID | 43314 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
ocket | +0.265 |
replay | +0.255 |
adiator | +0.254 |
Explore | +0.253 |
rium | +0.250 |
Neg logit contributions | |
unden | -0.291 |
rall | -0.281 |
arrang | -0.262 |
reluct | -0.259 |
fort | -0.257 |
Token | ist |
Token ID | 396 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
ocket | +0.905 |
adiator | +0.873 |
replay | +0.863 |
Explore | +0.862 |
rium | +0.852 |
Neg logit contributions | |
unden | -1.103 |
rall | -1.065 |
arrang | -1.005 |
reluct | -0.990 |
fort | -0.979 |
Token | bloc |
Token ID | 24003 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
unden | +0.437 |
rall | +0.425 |
arrang | +0.400 |
reluct | +0.396 |
fort | +0.392 |
Neg logit contributions | |
ocket | -0.314 |
adiator | -0.301 |
Explore | -0.298 |
replay | -0.297 |
rium | -0.295 |
Token | after |
Token ID | 706 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
unden | +0.444 |
rall | +0.434 |
arrang | +0.408 |
reluct | +0.403 |
fort | +0.398 |
Neg logit contributions | |
ocket | -0.306 |
adiator | -0.293 |
Explore | -0.289 |
replay | -0.286 |
rium | -0.285 |
Token | the |
Token ID | 262 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
ocket | +0.040 |
adiator | +0.038 |
Explore | +0.038 |
replay | +0.038 |
rium | +0.037 |
Neg logit contributions | |
unden | -0.051 |
rall | -0.050 |
arrang | -0.047 |
reluct | -0.046 |
fort | -0.046 |
Token | recent |
Token ID | 2274 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ocket | +0.298 |
replay | +0.287 |
adiator | +0.285 |
Explore | +0.284 |
rium | +0.279 |
Neg logit contributions | |
unden | -0.350 |
rall | -0.340 |
arrang | -0.318 |
reluct | -0.316 |
fort | -0.313 |
Token | surge |
Token ID | 13853 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
ocket | +2.640 |
replay | +2.598 |
Explore | +2.507 |
adiator | +2.488 |
crawl | +2.453 |
Neg logit contributions | |
unden | -3.146 |
rall | -3.023 |
fort | -2.832 |
reluct | -2.828 |
arrang | -2.819 |
Token | of |
Token ID | 286 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
ocket | +0.321 |
adiator | +0.309 |
Explore | +0.305 |
replay | +0.302 |
rium | +0.300 |
Neg logit contributions | |
unden | -0.458 |
rall | -0.448 |
arrang | -0.421 |
reluct | -0.416 |
fort | -0.412 |
Token | popular |
Token ID | 2968 |
Feature activation | +3.81e-01 |
Pos logit contributions | |
ocket | +0.334 |
adiator | +0.321 |
replay | +0.320 |
Explore | +0.318 |
rium | +0.315 |
Neg logit contributions | |
unden | -0.380 |
rall | -0.370 |
arrang | -0.347 |
reluct | -0.343 |
fort | -0.341 |
Token | mobil |
Token ID | 17754 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ocket | +0.623 |
adiator | +0.598 |
replay | +0.598 |
Explore | +0.591 |
rium | +0.586 |
Neg logit contributions | |
unden | -0.658 |
rall | -0.643 |
arrang | -0.600 |
reluct | -0.590 |
fort | -0.586 |
Token | isation |
Token ID | 5612 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
unden | +0.244 |
rall | +0.237 |
arrang | +0.222 |
reluct | +0.221 |
'. | +0.217 |
Neg logit contributions | |
ocket | -0.188 |
adiator | -0.180 |
Explore | -0.178 |
replay | -0.176 |
rium | -0.175 |
Token | federal |
Token ID | 2717 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
ocket | +0.118 |
adiator | +0.114 |
Explore | +0.112 |
replay | +0.112 |
rium | +0.111 |
Neg logit contributions | |
unden | -0.142 |
rall | -0.137 |
arrang | -0.129 |
reluct | -0.127 |
fort | -0.126 |
Token | reviews |
Token ID | 8088 |
Feature activation | +8.80e-01 |
Pos logit contributions | |
unden | +0.516 |
rall | +0.503 |
arrang | +0.472 |
reluct | +0.466 |
fort | +0.460 |
Neg logit contributions | |
ocket | -0.430 |
adiator | -0.414 |
Explore | -0.409 |
replay | -0.406 |
rium | -0.403 |
Token | of |
Token ID | 286 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
ocket | +1.130 |
adiator | +1.068 |
Explore | +1.046 |
paces | +1.044 |
replay | +1.041 |
Neg logit contributions | |
unden | -1.878 |
rall | -1.837 |
arrang | -1.723 |
reluct | -1.702 |
fort | -1.680 |
Token | whether |
Token ID | 1771 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
ocket | +0.167 |
adiator | +0.160 |
Explore | +0.158 |
replay | +0.157 |
rium | +0.156 |
Neg logit contributions | |
unden | -0.214 |
rall | -0.209 |
arrang | -0.196 |
reluct | -0.194 |
fort | -0.192 |
Token | health |
Token ID | 1535 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
unden | +1.059 |
rall | +1.035 |
arrang | +0.972 |
reluct | +0.960 |
'. | +0.951 |
Neg logit contributions | |
ocket | -0.815 |
adiator | -0.785 |
Explore | -0.774 |
replay | -0.762 |
rium | -0.762 |
Token | plans |
Token ID | 3352 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ocket | +1.114 |
Explore | +1.058 |
adiator | +1.058 |
replay | +1.055 |
iversity | +1.042 |
Neg logit contributions | |
unden | -1.411 |
rall | -1.373 |
arrang | -1.283 |
reluct | -1.268 |
fort | -1.252 |
Token | in |
Token ID | 287 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
ocket | +2.399 |
replay | +2.281 |
Explore | +2.238 |
paces | +2.237 |
adiator | +2.209 |
Neg logit contributions | |
unden | -3.522 |
rall | -3.387 |
arrang | -3.195 |
reluct | -3.183 |
fort | -3.108 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ocket | +0.292 |
adiator | +0.279 |
Explore | +0.277 |
replay | +0.276 |
rium | +0.273 |
Neg logit contributions | |
unden | -0.352 |
rall | -0.341 |
arrang | -0.322 |
reluct | -0.317 |
fort | -0.312 |
Token | ACA |
Token ID | 27628 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
ocket | +0.182 |
adiator | +0.175 |
Explore | +0.173 |
replay | +0.172 |
rium | +0.171 |
Neg logit contributions | |
unden | -0.212 |
rall | -0.206 |
arrang | -0.193 |
reluct | -0.190 |
fort | -0.188 |
Token | market |
Token ID | 1910 |
Feature activation | +7.17e-01 |
Pos logit contributions | |
ocket | +0.448 |
adiator | +0.427 |
replay | +0.424 |
Explore | +0.424 |
rium | +0.418 |
Neg logit contributions | |
unden | -0.636 |
rall | -0.619 |
arrang | -0.584 |
reluct | -0.577 |
fort | -0.567 |
Token | places |
Token ID | 23625 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
ocket | +1.023 |
adiator | +0.968 |
Explore | +0.966 |
replay | +0.960 |
rium | +0.954 |
Neg logit contributions | |
unden | -1.419 |
rall | -1.382 |
arrang | -1.300 |
reluct | -1.290 |
'. | -1.256 |
Token | Temple |
Token ID | 10857 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
ocket | +0.367 |
adiator | +0.353 |
Explore | +0.348 |
rium | +0.347 |
replay | +0.344 |
Neg logit contributions | |
unden | -0.386 |
rall | -0.377 |
arrang | -0.351 |
reluct | -0.348 |
'. | -0.339 |
Token |
|
Token ID | 198 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
ocket | +0.337 |
adiator | +0.321 |
Explore | +0.316 |
rium | +0.312 |
replay | +0.311 |
Neg logit contributions | |
unden | -0.554 |
rall | -0.543 |
arrang | -0.512 |
reluct | -0.508 |
'. | -0.500 |
Token |
|
Token ID | 198 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
ocket | +0.583 |
adiator | +0.554 |
Explore | +0.554 |
replay | +0.544 |
rium | +0.540 |
Neg logit contributions | |
unden | -1.012 |
rall | -0.987 |
arrang | -0.936 |
reluct | -0.933 |
'. | -0.891 |
Token | Address |
Token ID | 20231 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
ocket | +0.692 |
Explore | +0.664 |
adiator | +0.664 |
rium | +0.649 |
replay | +0.648 |
Neg logit contributions | |
unden | -0.863 |
rall | -0.842 |
arrang | -0.790 |
reluct | -0.786 |
'. | -0.759 |
Token | : |
Token ID | 25 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
ocket | +0.289 |
adiator | +0.275 |
Explore | +0.273 |
replay | +0.268 |
rium | +0.266 |
Neg logit contributions | |
unden | -0.561 |
rall | -0.549 |
arrang | -0.521 |
reluct | -0.518 |
'. | -0.502 |
Token | Sector |
Token ID | 25892 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ocket | +0.605 |
adiator | +0.583 |
Explore | +0.579 |
rium | +0.574 |
replay | +0.573 |
Neg logit contributions | |
unden | -0.626 |
rall | -0.609 |
arrang | -0.570 |
reluct | -0.566 |
fort | -0.541 |
Token | 20 |
Token ID | 1160 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
ocket | +2.676 |
rium | +2.570 |
adiator | +2.563 |
Explore | +2.529 |
iversity | +2.519 |
Neg logit contributions | |
unden | -3.191 |
rall | -3.103 |
arrang | -2.927 |
reluct | -2.925 |
fort | -2.733 |
Token | , |
Token ID | 11 |
Feature activation | +4.81e-01 |
Pos logit contributions | |
unden | +0.525 |
rall | +0.514 |
arrang | +0.485 |
reluct | +0.481 |
'. | +0.474 |
Neg logit contributions | |
ocket | -0.324 |
adiator | -0.309 |
Explore | -0.305 |
replay | -0.300 |
rium | -0.299 |
Token | J |
Token ID | 41 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
ocket | +0.810 |
adiator | +0.781 |
Explore | +0.776 |
rium | +0.771 |
replay | +0.766 |
Neg logit contributions | |
unden | -0.820 |
rall | -0.800 |
arrang | -0.745 |
reluct | -0.738 |
fort | -0.712 |
Token | Road |
Token ID | 5567 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
unden | +0.059 |
rall | +0.058 |
arrang | +0.054 |
reluct | +0.054 |
'. | +0.053 |
Neg logit contributions | |
ocket | -0.044 |
adiator | -0.042 |
Explore | -0.042 |
rium | -0.041 |
replay | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
unden | +0.480 |
rall | +0.471 |
arrang | +0.444 |
reluct | +0.440 |
'. | +0.433 |
Neg logit contributions | |
ocket | -0.299 |
adiator | -0.285 |
Explore | -0.281 |
rium | -0.278 |
replay | -0.277 |
Token | priority |
Token ID | 8475 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
ocket | +0.122 |
adiator | +0.116 |
replay | +0.115 |
Explore | +0.114 |
rium | +0.114 |
Neg logit contributions | |
unden | -0.169 |
rall | -0.166 |
arrang | -0.155 |
reluct | -0.154 |
'. | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
unden | +0.226 |
rall | +0.223 |
arrang | +0.208 |
reluct | +0.208 |
fort | +0.204 |
Neg logit contributions | |
ocket | -0.145 |
adiator | -0.138 |
replay | -0.136 |
Explore | -0.136 |
rium | -0.135 |
Token |
|
Token ID | 198 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
ocket | +0.244 |
Explore | +0.233 |
adiator | +0.231 |
replay | +0.229 |
rium | +0.228 |
Neg logit contributions | |
unden | -0.331 |
rall | -0.322 |
arrang | -0.304 |
reluct | -0.303 |
fort | -0.289 |
Token |
|
Token ID | 198 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
ocket | +0.106 |
adiator | +0.101 |
Explore | +0.100 |
replay | +0.098 |
rium | +0.098 |
Neg logit contributions | |
unden | -0.169 |
rall | -0.165 |
arrang | -0.156 |
reluct | -0.154 |
'. | -0.152 |
Token | But |
Token ID | 1537 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
ocket | +0.068 |
adiator | +0.065 |
Explore | +0.065 |
replay | +0.063 |
rium | +0.063 |
Neg logit contributions | |
unden | -0.114 |
rall | -0.111 |
arrang | -0.105 |
reluct | -0.104 |
'. | -0.102 |
Token | …. |
Token ID | 11580 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ocket | +0.917 |
adiator | +0.870 |
replay | +0.867 |
Explore | +0.865 |
rium | +0.857 |
Neg logit contributions | |
unden | -1.121 |
rall | -1.088 |
arrang | -1.020 |
reluct | -1.009 |
'. | -0.997 |
Token |
|
Token ID | 198 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
ocket | +2.503 |
Explore | +2.373 |
adiator | +2.369 |
rium | +2.351 |
replay | +2.350 |
Neg logit contributions | |
unden | -3.308 |
rall | -3.211 |
arrang | -3.012 |
reluct | -2.988 |
'. | -2.938 |
Token |
|
Token ID | 198 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
unden | +0.041 |
rall | +0.040 |
arrang | +0.038 |
reluct | +0.038 |
fort | +0.037 |
Neg logit contributions | |
ocket | -0.026 |
adiator | -0.025 |
Explore | -0.025 |
replay | -0.025 |
rium | -0.024 |
Token | " |
Token ID | 1 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
ocket | +0.043 |
adiator | +0.040 |
Explore | +0.040 |
replay | +0.039 |
rium | +0.039 |
Neg logit contributions | |
unden | -0.071 |
rall | -0.070 |
arrang | -0.066 |
reluct | -0.065 |
'. | -0.064 |
Token | Ah |
Token ID | 10910 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
ocket | +0.653 |
Explore | +0.626 |
adiator | +0.620 |
rium | +0.608 |
replay | +0.605 |
Neg logit contributions | |
unden | -1.023 |
rall | -1.001 |
arrang | -0.944 |
reluct | -0.939 |
'. | -0.913 |
Token | ." |
Token ID | 526 |
Feature activation | +5.45e-01 |
Pos logit contributions | |
ocket | +0.510 |
adiator | +0.486 |
Explore | +0.479 |
rium | +0.474 |
replay | +0.474 |
Neg logit contributions | |
unden | -0.807 |
rall | -0.792 |
arrang | -0.744 |
reluct | -0.738 |
'. | -0.726 |
Token | // |
Token ID | 3373 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
ocket | +0.233 |
adiator | +0.224 |
Explore | +0.223 |
replay | +0.222 |
rium | +0.219 |
Neg logit contributions | |
unden | -0.239 |
rall | -0.232 |
arrang | -0.217 |
reluct | -0.214 |
fort | -0.211 |
Token | The |
Token ID | 383 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
unden | +0.273 |
rall | +0.266 |
arrang | +0.250 |
reluct | +0.247 |
fort | +0.244 |
Neg logit contributions | |
ocket | -0.222 |
adiator | -0.213 |
Explore | -0.212 |
replay | -0.210 |
rium | -0.208 |
Token | previous |
Token ID | 2180 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
ocket | +0.204 |
adiator | +0.195 |
replay | +0.195 |
Explore | +0.195 |
rium | +0.191 |
Neg logit contributions | |
unden | -0.249 |
rall | -0.242 |
arrang | -0.227 |
reluct | -0.225 |
fort | -0.223 |
Token | board |
Token ID | 3096 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
ocket | +0.435 |
adiator | +0.409 |
replay | +0.407 |
Explore | +0.405 |
rium | +0.395 |
Neg logit contributions | |
unden | -0.882 |
rall | -0.861 |
arrang | -0.817 |
reluct | -0.810 |
fort | -0.803 |
Token | | |
Token ID | 930 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
ocket | +0.556 |
adiator | +0.528 |
Explore | +0.521 |
replay | +0.521 |
rium | +0.511 |
Neg logit contributions | |
unden | -0.912 |
rall | -0.893 |
arrang | -0.843 |
reluct | -0.837 |
fort | -0.826 |
Token | ---------------- |
Token ID | 34400 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
unden | +0.450 |
rall | +0.439 |
arrang | +0.411 |
reluct | +0.406 |
fort | +0.401 |
Neg logit contributions | |
ocket | -0.379 |
adiator | -0.365 |
Explore | -0.361 |
replay | -0.357 |
rium | -0.355 |
Token | ------------- |
Token ID | 32501 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
unden | +3.876 |
rall | +3.827 |
arrang | +3.553 |
'. | +3.512 |
reluct | +3.493 |
Neg logit contributions | |
ocket | -3.194 |
adiator | -3.083 |
Explore | -3.020 |
rium | -3.000 |
paces | -2.975 |
Token | recursive |
Token ID | 45115 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
unden | +2.189 |
rall | +2.143 |
arrang | +1.997 |
reluct | +1.965 |
'. | +1.949 |
Neg logit contributions | |
ocket | -2.001 |
adiator | -1.934 |
Explore | -1.908 |
rium | -1.884 |
replay | -1.875 |
Token | without |
Token ID | 1231 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
ocket | +1.132 |
adiator | +1.079 |
replay | +1.079 |
Explore | +1.072 |
Weak | +1.046 |
Neg logit contributions | |
unden | -1.724 |
rall | -1.681 |
arrang | -1.582 |
reluct | -1.574 |
fort | -1.552 |
Token | ind |
Token ID | 773 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
ocket | +0.064 |
adiator | +0.061 |
Explore | +0.060 |
replay | +0.060 |
rium | +0.060 |
Neg logit contributions | |
unden | -0.081 |
rall | -0.079 |
arrang | -0.074 |
reluct | -0.073 |
fort | -0.072 |
Token | irection |
Token ID | 4154 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
ocket | +0.454 |
adiator | +0.419 |
replay | +0.413 |
Explore | +0.412 |
paces | +0.410 |
Neg logit contributions | |
unden | -1.149 |
rall | -1.127 |
arrang | -1.073 |
reluct | -1.069 |
'. | -1.045 |
Token | by |
Token ID | 416 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
unden | +0.178 |
rall | +0.174 |
arrang | +0.163 |
reluct | +0.161 |
'. | +0.160 |
Neg logit contributions | |
ocket | -0.150 |
adiator | -0.145 |
Explore | -0.143 |
replay | -0.141 |
rium | -0.141 |
Token | clicking |
Token ID | 12264 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
unden | +0.532 |
rall | +0.519 |
arrang | +0.480 |
reluct | +0.472 |
fort | +0.468 |
Neg logit contributions | |
ocket | -0.613 |
adiator | -0.595 |
Explore | -0.589 |
replay | -0.583 |
rium | -0.582 |
Token | the |
Token ID | 262 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
unden | +0.580 |
rall | +0.568 |
arrang | +0.528 |
'. | +0.521 |
fort | +0.520 |
Neg logit contributions | |
ocket | -0.545 |
adiator | -0.527 |
Explore | -0.520 |
replay | -0.514 |
rium | -0.513 |
Token | box |
Token ID | 3091 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
unden | +0.631 |
rall | +0.614 |
arrang | +0.572 |
reluct | +0.564 |
fort | +0.554 |
Neg logit contributions | |
ocket | -0.639 |
adiator | -0.617 |
Explore | -0.611 |
replay | -0.606 |
rium | -0.603 |
Token | . |
Token ID | 13 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
unden | +0.061 |
rall | +0.059 |
arrang | +0.056 |
reluct | +0.056 |
fort | +0.054 |
Neg logit contributions | |
ocket | -0.034 |
adiator | -0.032 |
Explore | -0.032 |
replay | -0.031 |
rium | -0.031 |
Token | Invalid |
Token ID | 17665 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
unden | +0.226 |
rall | +0.212 |
arrang | +0.180 |
reluct | +0.175 |
fort | +0.165 |
Neg logit contributions | |
ocket | -0.737 |
adiator | -0.720 |
Explore | -0.718 |
replay | -0.712 |
rium | -0.710 |
Token | email |
Token ID | 3053 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
unden | +2.745 |
rall | +2.731 |
'. | +2.559 |
cir | +2.530 |
fort | +2.515 |
Neg logit contributions | |
ocket | -3.938 |
adiator | -3.871 |
Explore | -3.840 |
paces | -3.793 |
Weak | -3.737 |
Token | address |
Token ID | 2209 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
unden | +0.213 |
rall | +0.206 |
arrang | +0.193 |
reluct | +0.192 |
fort | +0.185 |
Neg logit contributions | |
ocket | -0.177 |
adiator | -0.169 |
replay | -0.168 |
Explore | -0.167 |
rium | -0.167 |
Token | . |
Token ID | 13 |
Feature activation | -5.03e-01 |
Pos logit contributions | |
unden | +0.513 |
rall | +0.502 |
arrang | +0.474 |
reluct | +0.469 |
'. | +0.466 |
Neg logit contributions | |
ocket | -0.311 |
adiator | -0.298 |
Explore | -0.293 |
replay | -0.289 |
rium | -0.288 |
Token | Please |
Token ID | 4222 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
unden | +0.899 |
rall | +0.875 |
arrang | +0.818 |
reluct | +0.811 |
fort | +0.786 |
Neg logit contributions | |
ocket | -0.816 |
adiator | -0.783 |
Explore | -0.782 |
replay | -0.771 |
rium | -0.767 |
Token | re |
Token ID | 302 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
unden | +0.973 |
rall | +0.943 |
arrang | +0.868 |
reluct | +0.854 |
fort | +0.839 |
Neg logit contributions | |
ocket | -1.283 |
adiator | -1.245 |
Explore | -1.234 |
replay | -1.224 |
rium | -1.219 |
Token | by |
Token ID | 416 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
unden | +0.141 |
rall | +0.138 |
arrang | +0.129 |
'. | +0.127 |
reluct | +0.127 |
Neg logit contributions | |
ocket | -0.120 |
adiator | -0.116 |
Explore | -0.115 |
replay | -0.113 |
rium | -0.113 |
Token | clicking |
Token ID | 12264 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
unden | +0.468 |
rall | +0.457 |
arrang | +0.423 |
reluct | +0.415 |
fort | +0.413 |
Neg logit contributions | |
ocket | -0.547 |
adiator | -0.530 |
Explore | -0.526 |
replay | -0.520 |
rium | -0.519 |
Token | the |
Token ID | 262 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
unden | +0.792 |
rall | +0.775 |
arrang | +0.720 |
'. | +0.714 |
fort | +0.713 |
Neg logit contributions | |
ocket | -0.752 |
adiator | -0.728 |
Explore | -0.718 |
replay | -0.710 |
rium | -0.708 |
Token | box |
Token ID | 3091 |
Feature activation | -4.10e-03 |
Pos logit contributions | |
unden | +0.681 |
rall | +0.664 |
arrang | +0.616 |
reluct | +0.607 |
fort | +0.602 |
Neg logit contributions | |
ocket | -0.708 |
adiator | -0.685 |
Explore | -0.678 |
replay | -0.670 |
rium | -0.669 |
Token | . |
Token ID | 13 |
Feature activation | -3.87e-01 |
Pos logit contributions | |
unden | +0.009 |
rall | +0.009 |
arrang | +0.008 |
reluct | +0.008 |
fort | +0.008 |
Neg logit contributions | |
ocket | -0.005 |
adiator | -0.005 |
Explore | -0.005 |
replay | -0.005 |
rium | -0.005 |
Token | Invalid |
Token ID | 17665 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
unden | +0.305 |
rall | +0.287 |
arrang | +0.243 |
reluct | +0.236 |
fort | +0.224 |
Neg logit contributions | |
ocket | -1.009 |
adiator | -0.986 |
Explore | -0.982 |
replay | -0.974 |
rium | -0.972 |
Token | email |
Token ID | 3053 |
Feature activation | +9.40e-04 |
Pos logit contributions | |
unden | +2.708 |
rall | +2.690 |
'. | +2.504 |
cir | +2.487 |
fort | +2.466 |
Neg logit contributions | |
ocket | -3.868 |
adiator | -3.796 |
Explore | -3.760 |
paces | -3.725 |
Weak | -3.667 |
Token | address |
Token ID | 2209 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
ocket | +0.001 |
adiator | +0.001 |
replay | +0.001 |
rium | +0.001 |
Explore | +0.001 |
Neg logit contributions | |
unden | -0.002 |
rall | -0.002 |
arrang | -0.002 |
reluct | -0.002 |
fort | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
unden | +0.242 |
rall | +0.237 |
arrang | +0.224 |
reluct | +0.222 |
fort | +0.219 |
Neg logit contributions | |
ocket | -0.145 |
adiator | -0.138 |
Explore | -0.136 |
replay | -0.134 |
rium | -0.134 |
Token | Please |
Token ID | 4222 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
unden | +0.962 |
rall | +0.938 |
arrang | +0.877 |
reluct | +0.869 |
fort | +0.842 |
Neg logit contributions | |
ocket | -0.871 |
adiator | -0.836 |
Explore | -0.835 |
replay | -0.822 |
rium | -0.819 |
Token | re |
Token ID | 302 |
Feature activation | -7.29e-02 |
Pos logit contributions | |
unden | +0.970 |
rall | +0.941 |
arrang | +0.865 |
reluct | +0.851 |
fort | +0.837 |
Neg logit contributions | |
ocket | -1.284 |
adiator | -1.246 |
Explore | -1.235 |
replay | -1.224 |
rium | -1.220 |
Token | by |
Token ID | 416 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
unden | +0.177 |
rall | +0.173 |
arrang | +0.162 |
reluct | +0.160 |
fort | +0.159 |
Neg logit contributions | |
ocket | -0.151 |
adiator | -0.145 |
Explore | -0.144 |
replay | -0.141 |
rium | -0.141 |
Token | clicking |
Token ID | 12264 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
unden | +0.576 |
rall | +0.562 |
arrang | +0.519 |
reluct | +0.510 |
fort | +0.509 |
Neg logit contributions | |
ocket | -0.682 |
adiator | -0.662 |
Explore | -0.657 |
replay | -0.649 |
rium | -0.649 |
Token | the |
Token ID | 262 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
unden | +0.827 |
rall | +0.809 |
arrang | +0.752 |
'. | +0.744 |
fort | +0.743 |
Neg logit contributions | |
ocket | -0.783 |
adiator | -0.759 |
Explore | -0.749 |
replay | -0.740 |
rium | -0.738 |
Token | box |
Token ID | 3091 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
unden | +0.705 |
rall | +0.687 |
arrang | +0.638 |
reluct | +0.629 |
fort | +0.622 |
Neg logit contributions | |
ocket | -0.729 |
adiator | -0.705 |
Explore | -0.697 |
replay | -0.690 |
rium | -0.688 |
Token | . |
Token ID | 13 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
unden | +0.352 |
rall | +0.345 |
arrang | +0.326 |
reluct | +0.323 |
fort | +0.316 |
Neg logit contributions | |
ocket | -0.204 |
adiator | -0.194 |
Explore | -0.192 |
replay | -0.190 |
rium | -0.188 |
Token | Invalid |
Token ID | 17665 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
unden | +0.298 |
rall | +0.281 |
arrang | +0.237 |
reluct | +0.229 |
fort | +0.220 |
Neg logit contributions | |
ocket | -1.024 |
adiator | -1.002 |
Explore | -0.996 |
replay | -0.989 |
rium | -0.987 |
Token | email |
Token ID | 3053 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
unden | +2.669 |
rall | +2.647 |
'. | +2.460 |
cir | +2.446 |
fort | +2.426 |
Neg logit contributions | |
ocket | -3.777 |
adiator | -3.711 |
Explore | -3.676 |
paces | -3.640 |
Weak | -3.576 |
Token | address |
Token ID | 2209 |
Feature activation | -8.66e-02 |
Pos logit contributions | |
unden | +0.145 |
rall | +0.140 |
arrang | +0.131 |
reluct | +0.131 |
fort | +0.126 |
Neg logit contributions | |
ocket | -0.118 |
adiator | -0.113 |
replay | -0.113 |
rium | -0.112 |
Explore | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
unden | +0.183 |
rall | +0.180 |
arrang | +0.170 |
reluct | +0.168 |
fort | +0.166 |
Neg logit contributions | |
ocket | -0.110 |
adiator | -0.105 |
Explore | -0.103 |
replay | -0.102 |
rium | -0.102 |
Token | Please |
Token ID | 4222 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
unden | +0.926 |
rall | +0.902 |
arrang | +0.843 |
reluct | +0.833 |
fort | +0.814 |
Neg logit contributions | |
ocket | -0.862 |
adiator | -0.829 |
Explore | -0.825 |
replay | -0.814 |
rium | -0.811 |
Token | re |
Token ID | 302 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
unden | +0.861 |
rall | +0.835 |
arrang | +0.768 |
reluct | +0.756 |
fort | +0.742 |
Neg logit contributions | |
ocket | -1.129 |
adiator | -1.094 |
Explore | -1.085 |
replay | -1.076 |
rium | -1.072 |
Token | that |
Token ID | 326 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
unden | +1.023 |
rall | +1.001 |
arrang | +0.943 |
reluct | +0.932 |
'. | +0.918 |
Neg logit contributions | |
ocket | -0.694 |
adiator | -0.663 |
Explore | -0.655 |
replay | -0.655 |
rium | -0.644 |
Token | 's |
Token ID | 338 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
unden | +0.023 |
rall | +0.023 |
arrang | +0.021 |
reluct | +0.021 |
fort | +0.021 |
Neg logit contributions | |
ocket | -0.017 |
replay | -0.017 |
adiator | -0.016 |
Explore | -0.016 |
rium | -0.016 |
Token | had |
Token ID | 550 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
ocket | +0.048 |
replay | +0.047 |
adiator | +0.047 |
Explore | +0.046 |
rium | +0.045 |
Neg logit contributions | |
unden | -0.058 |
rall | -0.056 |
arrang | -0.053 |
reluct | -0.053 |
fort | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
ocket | +0.103 |
replay | +0.101 |
adiator | +0.099 |
Explore | +0.099 |
rium | +0.097 |
Neg logit contributions | |
unden | -0.137 |
rall | -0.133 |
arrang | -0.125 |
reluct | -0.124 |
fort | -0.122 |
Token | story |
Token ID | 1621 |
Feature activation | -5.84e-01 |
Pos logit contributions | |
ocket | +0.219 |
replay | +0.215 |
adiator | +0.210 |
Explore | +0.209 |
rium | +0.205 |
Neg logit contributions | |
unden | -0.255 |
rall | -0.248 |
arrang | -0.233 |
reluct | -0.231 |
fort | -0.227 |
Token | patch |
Token ID | 8529 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
unden | +1.123 |
rall | +1.097 |
arrang | +1.031 |
reluct | +1.018 |
'. | +1.007 |
Neg logit contributions | |
ocket | -0.854 |
adiator | -0.820 |
Explore | -0.810 |
replay | -0.799 |
rium | -0.797 |
Token | . |
Token ID | 13 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
unden | +3.793 |
rall | +3.703 |
arrang | +3.510 |
fort | +3.467 |
'. | +3.465 |
Neg logit contributions | |
ocket | -2.774 |
adiator | -2.687 |
Explore | -2.666 |
rium | -2.611 |
Weak | -2.606 |
Token | That |
Token ID | 1320 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
ocket | +0.049 |
Explore | +0.047 |
adiator | +0.047 |
replay | +0.047 |
rium | +0.046 |
Neg logit contributions | |
unden | -0.079 |
rall | -0.077 |
arrang | -0.073 |
reluct | -0.073 |
fort | -0.070 |
Token | 's |
Token ID | 338 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
ocket | +0.153 |
replay | +0.150 |
Explore | +0.146 |
adiator | +0.146 |
rium | +0.144 |
Neg logit contributions | |
unden | -0.204 |
rall | -0.199 |
arrang | -0.186 |
reluct | -0.184 |
'. | -0.181 |
Token | obviously |
Token ID | 6189 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
unden | +0.115 |
rall | +0.112 |
arrang | +0.105 |
reluct | +0.104 |
fort | +0.102 |
Neg logit contributions | |
ocket | -0.095 |
adiator | -0.092 |
replay | -0.091 |
Explore | -0.091 |
rium | -0.090 |
Token | not |
Token ID | 407 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
unden | +0.426 |
rall | +0.417 |
arrang | +0.390 |
reluct | +0.385 |
fort | +0.380 |
Neg logit contributions | |
ocket | -0.359 |
replay | -0.346 |
adiator | -0.345 |
Explore | -0.344 |
rium | -0.338 |
Token | SI |
Token ID | 11584 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
ocket | +0.097 |
adiator | +0.093 |
Explore | +0.092 |
replay | +0.091 |
rium | +0.090 |
Neg logit contributions | |
unden | -0.134 |
rall | -0.131 |
arrang | -0.124 |
reluct | -0.122 |
fort | -0.121 |
Token | military |
Token ID | 2422 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
ocket | +0.187 |
adiator | +0.182 |
replay | +0.179 |
Explore | +0.178 |
rium | +0.177 |
Neg logit contributions | |
unden | -0.224 |
rall | -0.217 |
arrang | -0.204 |
reluct | -0.201 |
fort | -0.199 |
Token | was |
Token ID | 373 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
ocket | +0.461 |
adiator | +0.443 |
replay | +0.443 |
Explore | +0.436 |
rium | +0.434 |
Neg logit contributions | |
unden | -0.538 |
rall | -0.520 |
arrang | -0.485 |
reluct | -0.480 |
'. | -0.475 |
Token | under |
Token ID | 739 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
unden | +0.535 |
rall | +0.524 |
arrang | +0.492 |
reluct | +0.485 |
'. | +0.479 |
Neg logit contributions | |
ocket | -0.433 |
adiator | -0.416 |
Explore | -0.410 |
replay | -0.409 |
rium | -0.404 |
Token | the |
Token ID | 262 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
unden | +0.267 |
rall | +0.260 |
arrang | +0.243 |
reluct | +0.240 |
'. | +0.236 |
Neg logit contributions | |
ocket | -0.244 |
adiator | -0.238 |
Explore | -0.233 |
replay | -0.232 |
rium | -0.231 |
Token | command |
Token ID | 3141 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
unden | +0.597 |
rall | +0.580 |
arrang | +0.546 |
reluct | +0.541 |
fort | +0.534 |
Neg logit contributions | |
ocket | -0.445 |
adiator | -0.430 |
Explore | -0.424 |
replay | -0.424 |
rium | -0.418 |
Token | of |
Token ID | 286 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
unden | +3.998 |
rall | +3.918 |
arrang | +3.733 |
reluct | +3.680 |
fort | +3.675 |
Neg logit contributions | |
ocket | -2.493 |
Explore | -2.393 |
adiator | -2.363 |
replay | -2.320 |
paces | -2.306 |
Token | General |
Token ID | 3611 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
unden | +0.181 |
rall | +0.176 |
arrang | +0.164 |
reluct | +0.162 |
'. | +0.159 |
Neg logit contributions | |
ocket | -0.179 |
adiator | -0.174 |
Explore | -0.170 |
replay | -0.170 |
rium | -0.169 |
Token | Alfred |
Token ID | 22044 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
unden | +0.378 |
rall | +0.367 |
arrang | +0.343 |
reluct | +0.339 |
'. | +0.333 |
Neg logit contributions | |
ocket | -0.362 |
adiator | -0.349 |
Explore | -0.344 |
replay | -0.342 |
rium | -0.340 |
Token | o |
Token ID | 78 |
Feature activation | -2.89e-03 |
Pos logit contributions | |
unden | +1.649 |
rall | +1.614 |
arrang | +1.518 |
reluct | +1.500 |
fort | +1.486 |
Neg logit contributions | |
ocket | -1.161 |
adiator | -1.113 |
Explore | -1.101 |
replay | -1.086 |
rium | -1.081 |
Token | Gu |
Token ID | 1962 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
unden | +0.005 |
rall | +0.005 |
arrang | +0.005 |
reluct | +0.004 |
'. | +0.004 |
Neg logit contributions | |
ocket | -0.005 |
adiator | -0.005 |
Explore | -0.005 |
replay | -0.005 |
rium | -0.005 |
Token | under |
Token ID | 739 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
unden | +2.019 |
rall | +1.975 |
arrang | +1.858 |
reluct | +1.835 |
fort | +1.817 |
Neg logit contributions | |
ocket | -1.510 |
adiator | -1.448 |
Explore | -1.432 |
replay | -1.412 |
rium | -1.404 |
Token | Zh |
Token ID | 10511 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
unden | +0.425 |
rall | +0.414 |
arrang | +0.387 |
reluct | +0.386 |
fort | +0.379 |
Neg logit contributions | |
ocket | -0.341 |
adiator | -0.330 |
Explore | -0.325 |
replay | -0.324 |
rium | -0.322 |
Token | uk |
Token ID | 2724 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ocket | +0.311 |
adiator | +0.298 |
Explore | +0.294 |
replay | +0.290 |
rium | +0.289 |
Neg logit contributions | |
unden | -0.465 |
rall | -0.454 |
arrang | -0.428 |
reluct | -0.424 |
'. | -0.417 |
Token | ov |
Token ID | 709 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
unden | +2.137 |
rall | +2.123 |
fort | +2.042 |
'. | +2.016 |
arrang | +1.991 |
Neg logit contributions | |
ocket | -1.244 |
adiator | -1.175 |
Explore | -1.174 |
replay | -1.149 |
paces | -1.137 |
Token | 's |
Token ID | 338 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
unden | +1.537 |
rall | +1.500 |
arrang | +1.408 |
reluct | +1.395 |
fort | +1.375 |
Neg logit contributions | |
ocket | -1.158 |
adiator | -1.115 |
Explore | -1.098 |
replay | -1.089 |
rium | -1.083 |
Token | command |
Token ID | 3141 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
unden | +0.504 |
rall | +0.491 |
arrang | +0.459 |
reluct | +0.458 |
fort | +0.452 |
Neg logit contributions | |
ocket | -0.375 |
adiator | -0.360 |
replay | -0.357 |
Explore | -0.356 |
rium | -0.351 |
Token | .[ |
Token ID | 3693 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
unden | +3.681 |
rall | +3.602 |
arrang | +3.404 |
reluct | +3.346 |
'. | +3.344 |
Neg logit contributions | |
ocket | -2.821 |
adiator | -2.699 |
Explore | -2.695 |
replay | -2.636 |
rium | -2.625 |
Token | 31 |
Token ID | 3132 |
Feature activation | +2.67e-02 |
Pos logit contributions | |
ocket | +0.258 |
Explore | +0.243 |
adiator | +0.243 |
replay | +0.236 |
paces | +0.233 |
Neg logit contributions | |
unden | -0.566 |
rall | -0.554 |
arrang | -0.524 |
reluct | -0.522 |
'. | -0.511 |
Token | ] |
Token ID | 60 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
ocket | +0.044 |
adiator | +0.042 |
Explore | +0.042 |
replay | +0.041 |
iversity | +0.041 |
Neg logit contributions | |
unden | -0.047 |
rall | -0.045 |
arrang | -0.043 |
reluct | -0.042 |
'. | -0.042 |
Token | This |
Token ID | 770 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
ocket | +0.153 |
Explore | +0.147 |
adiator | +0.146 |
replay | +0.145 |
rium | +0.143 |
Neg logit contributions | |
unden | -0.237 |
rall | -0.230 |
reluct | -0.219 |
arrang | -0.218 |
fort | -0.211 |
Token | front |
Token ID | 2166 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
unden | +0.128 |
rall | +0.124 |
reluct | +0.115 |
arrang | +0.115 |
fort | +0.114 |
Neg logit contributions | |
ocket | -0.102 |
replay | -0.099 |
adiator | -0.098 |
Explore | -0.097 |
rium | -0.096 |
Token | handed |
Token ID | 10158 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
unden | +0.027 |
rall | +0.026 |
arrang | +0.024 |
reluct | +0.024 |
fort | +0.024 |
Neg logit contributions | |
ocket | -0.021 |
adiator | -0.020 |
Explore | -0.020 |
replay | -0.020 |
rium | -0.020 |
Token | Libya |
Token ID | 13465 |
Feature activation | +8.38e-01 |
Pos logit contributions | |
unden | +0.674 |
rall | +0.656 |
arrang | +0.610 |
reluct | +0.601 |
fort | +0.592 |
Neg logit contributions | |
ocket | -0.707 |
adiator | -0.683 |
Explore | -0.676 |
replay | -0.670 |
rium | -0.668 |
Token | over |
Token ID | 625 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
ocket | +1.268 |
adiator | +1.223 |
replay | +1.216 |
Explore | +1.213 |
rium | +1.199 |
Neg logit contributions | |
unden | -1.562 |
rall | -1.523 |
arrang | -1.434 |
reluct | -1.414 |
fort | -1.384 |
Token | to |
Token ID | 284 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
unden | +0.432 |
rall | +0.424 |
arrang | +0.401 |
reluct | +0.396 |
'. | +0.392 |
Neg logit contributions | |
ocket | -0.248 |
adiator | -0.238 |
Explore | -0.234 |
replay | -0.232 |
rium | -0.229 |
Token | a |
Token ID | 257 |
Feature activation | +7.33e-02 |
Pos logit contributions | |
ocket | +0.012 |
adiator | +0.012 |
Explore | +0.012 |
replay | +0.012 |
rium | +0.011 |
Neg logit contributions | |
unden | -0.015 |
rall | -0.015 |
arrang | -0.014 |
reluct | -0.014 |
fort | -0.014 |
Token | patch |
Token ID | 8529 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ocket | +0.114 |
adiator | +0.109 |
Explore | +0.108 |
replay | +0.107 |
rium | +0.107 |
Neg logit contributions | |
unden | -0.134 |
rall | -0.131 |
arrang | -0.123 |
reluct | -0.121 |
fort | -0.120 |
Token | work |
Token ID | 1818 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
unden | +3.784 |
rall | +3.729 |
arrang | +3.561 |
fort | +3.528 |
'. | +3.498 |
Neg logit contributions | |
ocket | -2.489 |
adiator | -2.395 |
Explore | -2.366 |
rium | -2.362 |
paces | -2.356 |
Token | of |
Token ID | 286 |
Feature activation | -5.50e-03 |
Pos logit contributions | |
ocket | +0.184 |
adiator | +0.176 |
replay | +0.173 |
Explore | +0.172 |
rium | +0.170 |
Neg logit contributions | |
unden | -0.324 |
rall | -0.316 |
arrang | -0.299 |
reluct | -0.296 |
fort | -0.293 |
Token | tribal |
Token ID | 17054 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
unden | +0.010 |
rall | +0.010 |
arrang | +0.009 |
reluct | +0.009 |
fort | +0.009 |
Neg logit contributions | |
ocket | -0.008 |
adiator | -0.008 |
replay | -0.008 |
Explore | -0.008 |
rium | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
ocket | +0.402 |
adiator | +0.390 |
replay | +0.387 |
Explore | +0.381 |
iversity | +0.376 |
Neg logit contributions | |
unden | -0.569 |
rall | -0.549 |
arrang | -0.518 |
reluct | -0.515 |
'. | -0.505 |
Token | Al |
Token ID | 978 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
ocket | +0.020 |
adiator | +0.020 |
Explore | +0.019 |
replay | +0.019 |
rium | +0.019 |
Neg logit contributions | |
unden | -0.024 |
rall | -0.024 |
arrang | -0.022 |
reluct | -0.022 |
fort | -0.022 |
Token | by |
Token ID | 416 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
ocket | +0.074 |
adiator | +0.071 |
Explore | +0.070 |
replay | +0.069 |
rium | +0.069 |
Neg logit contributions | |
unden | -0.091 |
rall | -0.089 |
arrang | -0.083 |
reluct | -0.082 |
fort | -0.081 |
Token | clicking |
Token ID | 12264 |
Feature activation | -3.88e-01 |
Pos logit contributions | |
unden | +0.451 |
rall | +0.439 |
arrang | +0.407 |
reluct | +0.401 |
fort | +0.395 |
Neg logit contributions | |
ocket | -0.511 |
adiator | -0.495 |
Explore | -0.491 |
replay | -0.486 |
rium | -0.484 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
unden | +0.678 |
rall | +0.663 |
arrang | +0.617 |
reluct | +0.608 |
'. | +0.607 |
Neg logit contributions | |
ocket | -0.629 |
adiator | -0.608 |
Explore | -0.601 |
replay | -0.594 |
rium | -0.592 |
Token | box |
Token ID | 3091 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
unden | +0.597 |
rall | +0.579 |
arrang | +0.541 |
reluct | +0.535 |
fort | +0.520 |
Neg logit contributions | |
ocket | -0.589 |
adiator | -0.568 |
Explore | -0.564 |
replay | -0.559 |
rium | -0.556 |
Token | . |
Token ID | 13 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
ocket | +0.004 |
adiator | +0.003 |
Explore | +0.003 |
replay | +0.003 |
rium | +0.003 |
Neg logit contributions | |
unden | -0.007 |
rall | -0.007 |
arrang | -0.006 |
reluct | -0.006 |
fort | -0.006 |
Token | Invalid |
Token ID | 17665 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
unden | +0.339 |
rall | +0.316 |
arrang | +0.272 |
reluct | +0.266 |
fort | +0.245 |
Neg logit contributions | |
ocket | -1.035 |
adiator | -1.011 |
Explore | -1.010 |
replay | -1.000 |
rium | -0.998 |
Token | email |
Token ID | 3053 |
Feature activation | +7.02e-02 |
Pos logit contributions | |
unden | +2.601 |
rall | +2.571 |
'. | +2.382 |
cir | +2.366 |
fort | +2.358 |
Neg logit contributions | |
ocket | -3.582 |
adiator | -3.507 |
Explore | -3.483 |
paces | -3.443 |
rium | -3.385 |
Token | address |
Token ID | 2209 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
ocket | +0.106 |
replay | +0.101 |
rium | +0.101 |
adiator | +0.101 |
Explore | +0.100 |
Neg logit contributions | |
unden | -0.136 |
rall | -0.131 |
arrang | -0.124 |
reluct | -0.124 |
fort | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
unden | +0.048 |
rall | +0.047 |
arrang | +0.045 |
reluct | +0.044 |
fort | +0.043 |
Neg logit contributions | |
ocket | -0.028 |
adiator | -0.027 |
Explore | -0.026 |
replay | -0.026 |
rium | -0.026 |
Token | Please |
Token ID | 4222 |
Feature activation | -6.48e-01 |
Pos logit contributions | |
unden | +0.975 |
rall | +0.951 |
arrang | +0.889 |
reluct | +0.884 |
fort | +0.850 |
Neg logit contributions | |
ocket | -0.853 |
Explore | -0.819 |
adiator | -0.817 |
replay | -0.804 |
rium | -0.802 |
Token | re |
Token ID | 302 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
unden | +0.963 |
rall | +0.932 |
arrang | +0.858 |
reluct | +0.847 |
fort | +0.827 |
Neg logit contributions | |
ocket | -1.240 |
adiator | -1.201 |
Explore | -1.192 |
replay | -1.183 |
rium | -1.177 |
Token | to |
Token ID | 284 |
Feature activation | -5.93e-02 |
Pos logit contributions | |
unden | +0.538 |
rall | +0.528 |
arrang | +0.499 |
reluct | +0.495 |
fort | +0.487 |
Neg logit contributions | |
ocket | -0.329 |
adiator | -0.313 |
replay | -0.310 |
Explore | -0.309 |
rium | -0.303 |
Token | learn |
Token ID | 2193 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
unden | +0.117 |
rall | +0.114 |
reluct | +0.107 |
arrang | +0.107 |
fort | +0.104 |
Neg logit contributions | |
ocket | -0.084 |
replay | -0.082 |
Explore | -0.080 |
adiator | -0.079 |
paces | -0.078 |
Token | more |
Token ID | 517 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
unden | +1.317 |
rall | +1.290 |
arrang | +1.206 |
reluct | +1.191 |
fort | +1.175 |
Neg logit contributions | |
ocket | -1.076 |
adiator | -1.034 |
Explore | -1.024 |
replay | -1.019 |
rium | -1.009 |
Token | about |
Token ID | 546 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
unden | +0.705 |
rall | +0.691 |
arrang | +0.639 |
reluct | +0.632 |
fort | +0.622 |
Neg logit contributions | |
ocket | -0.688 |
adiator | -0.663 |
Explore | -0.656 |
replay | -0.653 |
rium | -0.648 |
Token | social |
Token ID | 1919 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
unden | +0.454 |
rall | +0.443 |
arrang | +0.414 |
reluct | +0.411 |
fort | +0.407 |
Neg logit contributions | |
ocket | -0.338 |
adiator | -0.322 |
replay | -0.321 |
Explore | -0.319 |
rium | -0.314 |
Token | justice |
Token ID | 5316 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
unden | +2.014 |
rall | +1.968 |
arrang | +1.841 |
reluct | +1.821 |
fort | +1.799 |
Neg logit contributions | |
ocket | -1.637 |
adiator | -1.572 |
Explore | -1.555 |
replay | -1.541 |
rium | -1.530 |
Token | from |
Token ID | 422 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
unden | +3.665 |
rall | +3.577 |
arrang | +3.373 |
reluct | +3.330 |
'. | +3.296 |
Neg logit contributions | |
ocket | -2.689 |
adiator | -2.584 |
Explore | -2.554 |
replay | -2.516 |
rium | -2.508 |
Token | atheists |
Token ID | 28473 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
unden | +0.433 |
rall | +0.422 |
arrang | +0.398 |
reluct | +0.394 |
fort | +0.389 |
Neg logit contributions | |
ocket | -0.323 |
adiator | -0.308 |
replay | -0.307 |
Explore | -0.307 |
rium | -0.301 |
Token | who |
Token ID | 508 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
ocket | +0.113 |
adiator | +0.107 |
Explore | +0.105 |
paces | +0.105 |
replay | +0.105 |
Neg logit contributions | |
unden | -0.183 |
rall | -0.179 |
arrang | -0.169 |
reluct | -0.167 |
fort | -0.164 |
Token | shared |
Token ID | 4888 |
Feature activation | -6.46e-01 |
Pos logit contributions | |
ocket | +0.062 |
replay | +0.060 |
Explore | +0.059 |
adiator | +0.059 |
paces | +0.058 |
Neg logit contributions | |
unden | -0.079 |
rall | -0.076 |
arrang | -0.072 |
reluct | -0.071 |
fort | -0.070 |
Token | their |
Token ID | 511 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
unden | +1.349 |
rall | +1.323 |
arrang | +1.247 |
reluct | +1.234 |
fort | +1.220 |
Neg logit contributions | |
ocket | -0.837 |
adiator | -0.796 |
Explore | -0.788 |
replay | -0.783 |
rium | -0.773 |
Token | if |
Token ID | 611 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
ocket | +0.342 |
adiator | +0.326 |
replay | +0.323 |
Explore | +0.321 |
rium | +0.320 |
Neg logit contributions | |
unden | -0.437 |
rall | -0.427 |
arrang | -0.401 |
reluct | -0.398 |
fort | -0.388 |
Token | you |
Token ID | 345 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
unden | +0.285 |
rall | +0.278 |
arrang | +0.260 |
reluct | +0.258 |
fort | +0.254 |
Neg logit contributions | |
ocket | -0.239 |
adiator | -0.229 |
replay | -0.227 |
Explore | -0.227 |
rium | -0.225 |
Token | 're |
Token ID | 821 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
unden | +0.026 |
rall | +0.025 |
arrang | +0.024 |
reluct | +0.024 |
fort | +0.023 |
Neg logit contributions | |
ocket | -0.021 |
adiator | -0.020 |
replay | -0.020 |
Explore | -0.020 |
rium | -0.019 |
Token | really |
Token ID | 1107 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
unden | +0.355 |
rall | +0.346 |
arrang | +0.326 |
reluct | +0.324 |
fort | +0.318 |
Neg logit contributions | |
ocket | -0.243 |
adiator | -0.231 |
replay | -0.228 |
Explore | -0.228 |
paces | -0.226 |
Token | f |
Token ID | 277 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
ocket | +0.080 |
adiator | +0.076 |
replay | +0.076 |
Explore | +0.075 |
paces | +0.075 |
Neg logit contributions | |
unden | -0.104 |
rall | -0.101 |
arrang | -0.095 |
reluct | -0.094 |
fort | -0.092 |
Token | **** |
Token ID | 2466 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
unden | +0.575 |
rall | +0.559 |
arrang | +0.524 |
reluct | +0.517 |
'. | +0.509 |
Neg logit contributions | |
ocket | -0.490 |
adiator | -0.471 |
Explore | -0.466 |
replay | -0.460 |
rium | -0.459 |
Token | d |
Token ID | 67 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
unden | +3.598 |
rall | +3.563 |
arrang | +3.334 |
fort | +3.304 |
'. | +3.283 |
Neg logit contributions | |
ocket | -2.664 |
adiator | -2.552 |
Explore | -2.533 |
replay | -2.505 |
rium | -2.466 |
Token | . |
Token ID | 13 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
unden | +0.461 |
rall | +0.450 |
arrang | +0.421 |
reluct | +0.416 |
fort | +0.410 |
Neg logit contributions | |
ocket | -0.392 |
adiator | -0.378 |
Explore | -0.373 |
replay | -0.370 |
rium | -0.368 |
Token |
|
Token ID | 198 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
unden | +0.677 |
rall | +0.662 |
arrang | +0.623 |
reluct | +0.617 |
fort | +0.607 |
Neg logit contributions | |
ocket | -0.461 |
adiator | -0.440 |
Explore | -0.436 |
replay | -0.432 |
rium | -0.428 |
Token |
|
Token ID | 198 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
unden | +0.480 |
rall | +0.470 |
arrang | +0.443 |
reluct | +0.438 |
fort | +0.436 |
Neg logit contributions | |
ocket | -0.308 |
adiator | -0.295 |
Explore | -0.291 |
replay | -0.287 |
rium | -0.286 |
Token | Let |
Token ID | 5756 |
Feature activation | -3.96e-03 |
Pos logit contributions | |
unden | +0.403 |
rall | +0.395 |
arrang | +0.373 |
reluct | +0.370 |
'. | +0.365 |
Neg logit contributions | |
ocket | -0.224 |
adiator | -0.214 |
Explore | -0.211 |
replay | -0.207 |
rium | -0.206 |
Token | vm |
Token ID | 14761 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
ocket | +0.306 |
adiator | +0.291 |
Explore | +0.288 |
replay | +0.283 |
rium | +0.282 |
Neg logit contributions | |
unden | -0.534 |
rall | -0.523 |
arrang | -0.495 |
reluct | -0.490 |
fort | -0.484 |
Token | ash |
Token ID | 1077 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
unden | +0.616 |
rall | +0.602 |
arrang | +0.568 |
reluct | +0.563 |
fort | +0.555 |
Neg logit contributions | |
ocket | -0.399 |
adiator | -0.381 |
Explore | -0.376 |
rium | -0.371 |
replay | -0.370 |
Token | ship |
Token ID | 4074 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
unden | +0.558 |
rall | +0.545 |
arrang | +0.513 |
reluct | +0.508 |
fort | +0.500 |
Neg logit contributions | |
ocket | -0.392 |
adiator | -0.376 |
Explore | -0.372 |
replay | -0.367 |
rium | -0.367 |
Token | yard |
Token ID | 9413 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
ocket | +0.204 |
adiator | +0.194 |
Explore | +0.191 |
replay | +0.187 |
rium | +0.187 |
Neg logit contributions | |
unden | -0.428 |
rall | -0.420 |
arrang | -0.398 |
reluct | -0.395 |
'. | -0.390 |
Token | at |
Token ID | 379 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
unden | +0.622 |
rall | +0.607 |
arrang | +0.571 |
reluct | +0.565 |
fort | +0.557 |
Neg logit contributions | |
ocket | -0.456 |
adiator | -0.436 |
Explore | -0.432 |
replay | -0.427 |
rium | -0.425 |
Token | Sever |
Token ID | 26434 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
unden | +0.541 |
rall | +0.525 |
arrang | +0.492 |
reluct | +0.488 |
fort | +0.476 |
Neg logit contributions | |
ocket | -0.503 |
adiator | -0.485 |
Explore | -0.482 |
rium | -0.479 |
replay | -0.477 |
Token | od |
Token ID | 375 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
unden | +3.185 |
rall | +3.119 |
arrang | +2.889 |
fort | +2.840 |
reluct | +2.835 |
Neg logit contributions | |
ocket | -3.109 |
adiator | -3.019 |
Explore | -2.992 |
replay | -2.959 |
paces | -2.935 |
Token | v |
Token ID | 85 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
unden | +0.560 |
rall | +0.548 |
arrang | +0.515 |
reluct | +0.509 |
fort | +0.503 |
Neg logit contributions | |
ocket | -0.405 |
adiator | -0.389 |
Explore | -0.384 |
replay | -0.379 |
rium | -0.378 |
Token | insk |
Token ID | 35803 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
unden | +0.406 |
rall | +0.399 |
arrang | +0.380 |
reluct | +0.376 |
'. | +0.373 |
Neg logit contributions | |
ocket | -0.156 |
adiator | -0.146 |
Explore | -0.143 |
replay | -0.141 |
rium | -0.139 |
Token | , |
Token ID | 11 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
ocket | +0.898 |
adiator | +0.853 |
Explore | +0.847 |
rium | +0.832 |
replay | +0.829 |
Neg logit contributions | |
unden | -1.408 |
rall | -1.374 |
arrang | -1.295 |
reluct | -1.292 |
fort | -1.258 |
Token | which |
Token ID | 543 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
ocket | +0.124 |
Explore | +0.119 |
adiator | +0.118 |
replay | +0.117 |
rium | +0.116 |
Neg logit contributions | |
unden | -0.175 |
rall | -0.170 |
arrang | -0.160 |
reluct | -0.159 |
fort | -0.157 |
Token | series |
Token ID | 2168 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
ocket | +0.269 |
adiator | +0.259 |
Explore | +0.256 |
replay | +0.256 |
rium | +0.252 |
Neg logit contributions | |
unden | -0.320 |
rall | -0.311 |
arrang | -0.292 |
reluct | -0.289 |
fort | -0.284 |
Token | Natural |
Token ID | 12068 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
unden | +0.321 |
rall | +0.312 |
arrang | +0.293 |
reluct | +0.290 |
fort | +0.283 |
Neg logit contributions | |
ocket | -0.279 |
adiator | -0.268 |
Explore | -0.266 |
replay | -0.264 |
rium | -0.262 |
Token | World |
Token ID | 2159 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
ocket | +0.710 |
adiator | +0.679 |
Explore | +0.674 |
replay | +0.665 |
rium | +0.661 |
Neg logit contributions | |
unden | -1.050 |
rall | -1.026 |
arrang | -0.969 |
reluct | -0.959 |
fort | -0.944 |
Token | episode |
Token ID | 4471 |
Feature activation | +7.37e-01 |
Pos logit contributions | |
unden | +0.800 |
rall | +0.782 |
arrang | +0.737 |
reluct | +0.729 |
fort | +0.720 |
Neg logit contributions | |
ocket | -0.539 |
adiator | -0.515 |
Explore | -0.509 |
replay | -0.503 |
rium | -0.501 |
Token | " |
Token ID | 366 |
Feature activation | +3.75e-03 |
Pos logit contributions | |
ocket | +1.155 |
Explore | +1.105 |
adiator | +1.105 |
replay | +1.097 |
rium | +1.090 |
Neg logit contributions | |
unden | -1.350 |
rall | -1.306 |
arrang | -1.232 |
reluct | -1.222 |
fort | -1.178 |
Token | Snow |
Token ID | 28974 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
ocket | +0.005 |
Explore | +0.005 |
adiator | +0.005 |
replay | +0.005 |
rium | +0.005 |
Neg logit contributions | |
unden | -0.008 |
rall | -0.007 |
arrang | -0.007 |
reluct | -0.007 |
'. | -0.007 |
Token | Leopard |
Token ID | 49072 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
unden | +3.092 |
rall | +3.045 |
arrang | +2.809 |
fort | +2.788 |
'. | +2.774 |
Neg logit contributions | |
ocket | -3.168 |
adiator | -3.071 |
Explore | -3.021 |
replay | -2.996 |
rium | -2.992 |
Token | – |
Token ID | 784 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
unden | +0.716 |
rall | +0.700 |
arrang | +0.657 |
reluct | +0.650 |
fort | +0.641 |
Neg logit contributions | |
ocket | -0.547 |
adiator | -0.526 |
Explore | -0.520 |
replay | -0.514 |
rium | -0.511 |
Token | Beyond |
Token ID | 12197 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
unden | +0.218 |
rall | +0.212 |
arrang | +0.199 |
reluct | +0.197 |
fort | +0.193 |
Neg logit contributions | |
ocket | -0.186 |
adiator | -0.179 |
Explore | -0.178 |
replay | -0.176 |
rium | -0.175 |
Token | the |
Token ID | 262 |
Feature activation | -7.81e-02 |
Pos logit contributions | |
unden | +0.343 |
rall | +0.335 |
arrang | +0.313 |
reluct | +0.310 |
fort | +0.305 |
Neg logit contributions | |
ocket | -0.287 |
adiator | -0.276 |
Explore | -0.274 |
replay | -0.270 |
rium | -0.269 |
Token | Myth |
Token ID | 18900 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
unden | +0.134 |
rall | +0.131 |
arrang | +0.122 |
reluct | +0.120 |
fort | +0.118 |
Neg logit contributions | |
ocket | -0.130 |
adiator | -0.125 |
Explore | -0.125 |
replay | -0.123 |
rium | -0.123 |
Token | bronze |
Token ID | 22101 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
unden | +0.632 |
rall | +0.612 |
arrang | +0.577 |
reluct | +0.573 |
fort | +0.568 |
Neg logit contributions | |
ocket | -0.501 |
adiator | -0.482 |
Explore | -0.481 |
replay | -0.480 |
rium | -0.478 |
Token | and |
Token ID | 290 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
unden | +0.048 |
rall | +0.047 |
arrang | +0.044 |
reluct | +0.044 |
'. | +0.043 |
Neg logit contributions | |
ocket | -0.033 |
adiator | -0.032 |
Explore | -0.031 |
replay | -0.031 |
rium | -0.031 |
Token | steel |
Token ID | 7771 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
unden | +0.558 |
rall | +0.544 |
arrang | +0.513 |
reluct | +0.507 |
'. | +0.501 |
Neg logit contributions | |
ocket | -0.410 |
adiator | -0.394 |
Explore | -0.388 |
replay | -0.386 |
rium | -0.384 |
Token | have |
Token ID | 423 |
Feature activation | -4.53e-01 |
Pos logit contributions | |
ocket | +0.111 |
adiator | +0.106 |
Explore | +0.105 |
replay | +0.104 |
rium | +0.104 |
Neg logit contributions | |
unden | -0.149 |
rall | -0.146 |
arrang | -0.137 |
reluct | -0.136 |
'. | -0.134 |
Token | been |
Token ID | 587 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
unden | +0.825 |
rall | +0.805 |
arrang | +0.754 |
reluct | +0.745 |
fort | +0.735 |
Neg logit contributions | |
ocket | -0.707 |
adiator | -0.680 |
Explore | -0.674 |
replay | -0.667 |
rium | -0.664 |
Token | collected |
Token ID | 7723 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
unden | +0.235 |
rall | +0.228 |
arrang | +0.213 |
reluct | +0.212 |
fort | +0.209 |
Neg logit contributions | |
ocket | -0.182 |
adiator | -0.173 |
Explore | -0.172 |
replay | -0.171 |
rium | -0.170 |
Token | in |
Token ID | 287 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
unden | +3.362 |
rall | +3.282 |
arrang | +3.084 |
reluct | +3.039 |
'. | +3.018 |
Neg logit contributions | |
ocket | -2.822 |
adiator | -2.716 |
Explore | -2.684 |
replay | -2.648 |
rium | -2.641 |
Token | tribute |
Token ID | 17547 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
unden | +0.361 |
rall | +0.350 |
arrang | +0.329 |
reluct | +0.326 |
fort | +0.322 |
Neg logit contributions | |
ocket | -0.306 |
adiator | -0.294 |
Explore | -0.293 |
replay | -0.292 |
rium | -0.290 |
Token | to |
Token ID | 284 |
Feature activation | -9.46e-02 |
Pos logit contributions | |
ocket | +0.633 |
adiator | +0.604 |
Explore | +0.596 |
replay | +0.588 |
rium | +0.585 |
Neg logit contributions | |
unden | -1.084 |
rall | -1.062 |
arrang | -1.002 |
reluct | -0.992 |
fort | -0.981 |
Token | some |
Token ID | 617 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
unden | +0.170 |
rall | +0.164 |
arrang | +0.155 |
reluct | +0.153 |
fort | +0.151 |
Neg logit contributions | |
ocket | -0.149 |
adiator | -0.145 |
Explore | -0.143 |
replay | -0.143 |
rium | -0.142 |
Token | of |
Token ID | 286 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
unden | +0.232 |
rall | +0.226 |
arrang | +0.215 |
reluct | +0.212 |
fort | +0.210 |
Neg logit contributions | |
ocket | -0.139 |
adiator | -0.132 |
Explore | -0.130 |
replay | -0.129 |
rium | -0.128 |
Token | which |
Token ID | 543 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
unden | +0.137 |
rall | +0.134 |
arrang | +0.126 |
reluct | +0.125 |
fort | +0.123 |
Neg logit contributions | |
ocket | -0.093 |
adiator | -0.090 |
Explore | -0.088 |
replay | -0.087 |
rium | -0.087 |
Token | are |
Token ID | 389 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
ocket | +0.320 |
adiator | +0.306 |
Explore | +0.305 |
replay | +0.302 |
paces | +0.298 |
Neg logit contributions | |
unden | -0.426 |
rall | -0.415 |
arrang | -0.391 |
reluct | -0.387 |
fort | -0.381 |
Token | probably |
Token ID | 2192 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
unden | +0.093 |
rall | +0.091 |
arrang | +0.085 |
reluct | +0.084 |
fort | +0.083 |
Neg logit contributions | |
ocket | -0.073 |
adiator | -0.070 |
Explore | -0.070 |
replay | -0.069 |
rium | -0.068 |
Token | easy |
Token ID | 2562 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
unden | +0.239 |
rall | +0.234 |
arrang | +0.220 |
reluct | +0.218 |
fort | +0.214 |
Neg logit contributions | |
ocket | -0.178 |
adiator | -0.170 |
Explore | -0.169 |
replay | -0.168 |
rium | -0.166 |
Token | to |
Token ID | 284 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
ocket | +0.586 |
adiator | +0.563 |
Explore | +0.557 |
replay | +0.552 |
rium | +0.544 |
Neg logit contributions | |
unden | -0.970 |
rall | -0.954 |
arrang | -0.895 |
reluct | -0.894 |
fort | -0.876 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
unden | +0.121 |
rall | +0.119 |
reluct | +0.112 |
arrang | +0.111 |
fort | +0.108 |
Neg logit contributions | |
ocket | -0.092 |
replay | -0.089 |
adiator | -0.088 |
Explore | -0.088 |
paces | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
unden | +3.577 |
rall | +3.489 |
arrang | +3.309 |
'. | +3.275 |
fort | +3.260 |
Neg logit contributions | |
ocket | -2.525 |
adiator | -2.433 |
Explore | -2.406 |
rium | -2.365 |
replay | -2.350 |
Token | but |
Token ID | 475 |
Feature activation | -4.87e-03 |
Pos logit contributions | |
ocket | +0.016 |
adiator | +0.015 |
Explore | +0.015 |
replay | +0.015 |
rium | +0.015 |
Neg logit contributions | |
unden | -0.025 |
rall | -0.024 |
arrang | -0.023 |
reluct | -0.023 |
fort | -0.022 |
Token | it |
Token ID | 340 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
unden | +0.009 |
rall | +0.009 |
arrang | +0.009 |
reluct | +0.009 |
fort | +0.008 |
Neg logit contributions | |
ocket | -0.007 |
adiator | -0.007 |
Explore | -0.007 |
replay | -0.007 |
rium | -0.007 |
Token | just |
Token ID | 655 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
unden | +0.124 |
rall | +0.122 |
arrang | +0.115 |
reluct | +0.113 |
'. | +0.112 |
Neg logit contributions | |
ocket | -0.075 |
adiator | -0.071 |
Explore | -0.071 |
replay | -0.070 |
rium | -0.069 |
Token | takes |
Token ID | 2753 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
unden | +0.361 |
rall | +0.352 |
arrang | +0.332 |
reluct | +0.328 |
fort | +0.324 |
Neg logit contributions | |
ocket | -0.249 |
adiator | -0.239 |
Explore | -0.236 |
replay | -0.233 |
rium | -0.232 |
Token | bow |
Token ID | 9563 |
Feature activation | -9.20e-02 |
Pos logit contributions | |
unden | +0.111 |
rall | +0.107 |
reluct | +0.099 |
arrang | +0.099 |
fort | +0.098 |
Neg logit contributions | |
ocket | -0.100 |
replay | -0.096 |
adiator | -0.096 |
Explore | -0.096 |
rium | -0.094 |
Token | in |
Token ID | 287 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
unden | +0.182 |
rall | +0.178 |
arrang | +0.166 |
reluct | +0.166 |
fort | +0.161 |
Neg logit contributions | |
ocket | -0.131 |
adiator | -0.125 |
Explore | -0.123 |
replay | -0.123 |
rium | -0.122 |
Token | person |
Token ID | 1048 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
unden | +0.088 |
rall | +0.085 |
arrang | +0.080 |
reluct | +0.079 |
fort | +0.078 |
Neg logit contributions | |
ocket | -0.076 |
adiator | -0.073 |
replay | -0.072 |
Explore | -0.072 |
rium | -0.071 |
Token | but |
Token ID | 475 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
unden | +0.064 |
rall | +0.062 |
arrang | +0.058 |
reluct | +0.058 |
fort | +0.057 |
Neg logit contributions | |
ocket | -0.042 |
adiator | -0.040 |
Explore | -0.039 |
replay | -0.039 |
rium | -0.039 |
Token | their |
Token ID | 511 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
unden | +0.037 |
rall | +0.036 |
arrang | +0.034 |
reluct | +0.034 |
fort | +0.033 |
Neg logit contributions | |
ocket | -0.028 |
adiator | -0.027 |
Explore | -0.027 |
replay | -0.026 |
paces | -0.026 |
Token | collected |
Token ID | 7723 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
unden | +0.026 |
rall | +0.025 |
reluct | +0.023 |
arrang | +0.023 |
fort | +0.023 |
Neg logit contributions | |
ocket | -0.021 |
replay | -0.020 |
adiator | -0.020 |
Explore | -0.020 |
rium | -0.020 |
Token | correspondence |
Token ID | 22440 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
unden | +3.285 |
rall | +3.212 |
arrang | +3.009 |
reluct | +2.962 |
'. | +2.932 |
Neg logit contributions | |
ocket | -2.826 |
adiator | -2.725 |
Explore | -2.694 |
rium | -2.656 |
replay | -2.653 |
Token | contains |
Token ID | 4909 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
unden | +0.072 |
rall | +0.070 |
arrang | +0.066 |
reluct | +0.065 |
fort | +0.064 |
Neg logit contributions | |
ocket | -0.055 |
adiator | -0.053 |
Explore | -0.052 |
replay | -0.052 |
rium | -0.052 |
Token | no |
Token ID | 645 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
unden | +0.511 |
rall | +0.499 |
arrang | +0.468 |
reluct | +0.464 |
fort | +0.458 |
Neg logit contributions | |
ocket | -0.396 |
adiator | -0.382 |
replay | -0.378 |
Explore | -0.377 |
rium | -0.370 |
Token | letter |
Token ID | 3850 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
unden | +0.436 |
rall | +0.426 |
arrang | +0.398 |
reluct | +0.397 |
fort | +0.392 |
Neg logit contributions | |
ocket | -0.321 |
adiator | -0.307 |
replay | -0.305 |
Explore | -0.304 |
rium | -0.296 |
Token | saying |
Token ID | 2282 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
unden | +0.183 |
rall | +0.178 |
arrang | +0.167 |
reluct | +0.166 |
fort | +0.163 |
Neg logit contributions | |
ocket | -0.131 |
adiator | -0.125 |
Explore | -0.124 |
replay | -0.123 |
rium | -0.122 |
Token | This |
Token ID | 770 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
ocket | +0.092 |
Explore | +0.088 |
adiator | +0.087 |
replay | +0.087 |
paces | +0.086 |
Neg logit contributions | |
unden | -0.126 |
rall | -0.123 |
reluct | -0.115 |
arrang | -0.115 |
fort | -0.112 |
Token | is |
Token ID | 318 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
ocket | +0.637 |
replay | +0.613 |
adiator | +0.600 |
Explore | +0.597 |
rium | +0.590 |
Neg logit contributions | |
unden | -1.079 |
rall | -1.053 |
reluct | -0.986 |
arrang | -0.981 |
fort | -0.968 |
Token | explained |
Token ID | 4893 |
Feature activation | +6.66e-02 |
Pos logit contributions | |
unden | +0.438 |
rall | +0.428 |
arrang | +0.400 |
reluct | +0.397 |
fort | +0.389 |
Neg logit contributions | |
ocket | -0.356 |
adiator | -0.341 |
Explore | -0.338 |
replay | -0.337 |
rium | -0.334 |
Token | here |
Token ID | 994 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
ocket | +0.100 |
adiator | +0.096 |
replay | +0.095 |
Explore | +0.095 |
paces | +0.094 |
Neg logit contributions | |
unden | -0.126 |
rall | -0.123 |
arrang | -0.115 |
reluct | -0.114 |
fort | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
unden | +0.248 |
rall | +0.243 |
arrang | +0.228 |
reluct | +0.226 |
fort | +0.222 |
Neg logit contributions | |
ocket | -0.152 |
adiator | -0.145 |
Explore | -0.142 |
replay | -0.142 |
paces | -0.140 |
Token | * |
Token ID | 1635 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
unden | +0.491 |
rall | +0.480 |
arrang | +0.448 |
reluct | +0.446 |
fort | +0.438 |
Neg logit contributions | |
ocket | -0.362 |
adiator | -0.346 |
Explore | -0.345 |
replay | -0.343 |
paces | -0.340 |
Token | mod |
Token ID | 4666 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
unden | +3.430 |
rall | +3.355 |
arrang | +3.158 |
reluct | +3.107 |
'. | +3.095 |
Neg logit contributions | |
ocket | -2.587 |
adiator | -2.492 |
Explore | -2.447 |
replay | -2.423 |
rium | -2.417 |
Token | eline |
Token ID | 4470 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
ocket | +0.928 |
adiator | +0.889 |
Explore | +0.885 |
rium | +0.878 |
paces | +0.874 |
Neg logit contributions | |
unden | -0.872 |
rall | -0.840 |
reluct | -0.785 |
arrang | -0.783 |
cir | -0.750 |
Token | * |
Token ID | 9 |
Feature activation | -9.06e-01 |
Pos logit contributions | |
unden | +0.835 |
rall | +0.815 |
arrang | +0.766 |
reluct | +0.759 |
fort | +0.745 |
Neg logit contributions | |
ocket | -0.622 |
adiator | -0.596 |
Explore | -0.592 |
replay | -0.585 |
rium | -0.580 |
Token | * |
Token ID | 1635 |
Feature activation | -9.64e-01 |
Pos logit contributions | |
unden | +1.723 |
rall | +1.683 |
arrang | +1.580 |
reluct | +1.561 |
fort | +1.542 |
Neg logit contributions | |
ocket | -1.341 |
adiator | -1.289 |
Explore | -1.274 |
replay | -1.259 |
rium | -1.254 |
Token | vim |
Token ID | 31124 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
unden | +1.774 |
rall | +1.731 |
arrang | +1.621 |
reluct | +1.602 |
fort | +1.579 |
Neg logit contributions | |
ocket | -1.489 |
adiator | -1.433 |
Explore | -1.419 |
replay | -1.402 |
rium | -1.396 |
Token | month |
Token ID | 1227 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
ocket | +0.302 |
replay | +0.290 |
adiator | +0.287 |
Explore | +0.285 |
rium | +0.281 |
Neg logit contributions | |
unden | -0.388 |
rall | -0.373 |
arrang | -0.353 |
reluct | -0.350 |
fort | -0.343 |
Token | is |
Token ID | 318 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
unden | +0.917 |
rall | +0.893 |
arrang | +0.839 |
reluct | +0.830 |
'. | +0.818 |
Neg logit contributions | |
ocket | -0.685 |
adiator | -0.657 |
Explore | -0.648 |
replay | -0.642 |
rium | -0.637 |
Token | to |
Token ID | 284 |
Feature activation | +8.76e-03 |
Pos logit contributions | |
unden | +0.257 |
rall | +0.251 |
arrang | +0.236 |
reluct | +0.234 |
fort | +0.231 |
Neg logit contributions | |
ocket | -0.182 |
adiator | -0.174 |
Explore | -0.172 |
replay | -0.172 |
rium | -0.169 |
Token | try |
Token ID | 1949 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
ocket | +0.013 |
replay | +0.012 |
Explore | +0.012 |
adiator | +0.012 |
paces | +0.012 |
Neg logit contributions | |
unden | -0.017 |
rall | -0.017 |
reluct | -0.015 |
arrang | -0.015 |
fort | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
ocket | +0.448 |
adiator | +0.429 |
Explore | +0.423 |
replay | +0.423 |
iversity | +0.415 |
Neg logit contributions | |
unden | -0.670 |
rall | -0.655 |
arrang | -0.616 |
reluct | -0.612 |
fort | -0.602 |
Token | collect |
Token ID | 2824 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
unden | +0.367 |
rall | +0.358 |
reluct | +0.336 |
arrang | +0.334 |
fort | +0.327 |
Neg logit contributions | |
ocket | -0.267 |
replay | -0.260 |
Explore | -0.255 |
adiator | -0.254 |
paces | -0.251 |
Token | the |
Token ID | 262 |
Feature activation | -7.73e-02 |
Pos logit contributions | |
unden | +3.226 |
rall | +3.151 |
arrang | +2.948 |
reluct | +2.907 |
'. | +2.876 |
Neg logit contributions | |
ocket | -2.802 |
adiator | -2.700 |
Explore | -2.671 |
replay | -2.632 |
rium | -2.631 |
Token | Cur |
Token ID | 4424 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
unden | +0.143 |
rall | +0.139 |
arrang | +0.130 |
reluct | +0.129 |
fort | +0.127 |
Neg logit contributions | |
ocket | -0.119 |
replay | -0.114 |
adiator | -0.114 |
Explore | -0.113 |
rium | -0.111 |
Token | ric |
Token ID | 1173 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
unden | +0.410 |
rall | +0.401 |
arrang | +0.379 |
reluct | +0.375 |
'. | +0.370 |
Neg logit contributions | |
ocket | -0.259 |
adiator | -0.247 |
Explore | -0.244 |
rium | -0.241 |
replay | -0.241 |
Token | ulum |
Token ID | 14452 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
unden | +1.018 |
rall | +0.999 |
arrang | +0.950 |
reluct | +0.942 |
'. | +0.931 |
Neg logit contributions | |
ocket | -0.435 |
adiator | -0.410 |
Explore | -0.403 |
replay | -0.396 |
rium | -0.393 |
Token | Vit |
Token ID | 18271 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ocket | +0.229 |
adiator | +0.220 |
Explore | +0.219 |
replay | +0.219 |
rium | +0.215 |
Neg logit contributions | |
unden | -0.282 |
rall | -0.276 |
arrang | -0.258 |
reluct | -0.256 |
fort | -0.251 |
Token | show |
Token ID | 905 |
Feature activation | -5.17e-01 |
Pos logit contributions | |
unden | +0.075 |
rall | +0.072 |
reluct | +0.068 |
arrang | +0.068 |
fort | +0.066 |
Neg logit contributions | |
ocket | -0.053 |
replay | -0.052 |
Explore | -0.051 |
adiator | -0.051 |
paces | -0.050 |
Token | you |
Token ID | 345 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
unden | +1.003 |
rall | +0.977 |
arrang | +0.916 |
reluct | +0.907 |
fort | +0.897 |
Neg logit contributions | |
ocket | -0.745 |
adiator | -0.717 |
replay | -0.714 |
Explore | -0.708 |
rium | -0.700 |
Token | a |
Token ID | 257 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
ocket | +0.291 |
adiator | +0.279 |
replay | +0.279 |
Explore | +0.276 |
rium | +0.274 |
Neg logit contributions | |
unden | -0.332 |
rall | -0.324 |
arrang | -0.302 |
reluct | -0.299 |
fort | -0.294 |
Token | way |
Token ID | 835 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
unden | +0.200 |
rall | +0.195 |
arrang | +0.181 |
reluct | +0.179 |
fort | +0.178 |
Neg logit contributions | |
ocket | -0.172 |
replay | -0.166 |
adiator | -0.166 |
Explore | -0.163 |
rium | -0.162 |
Token | to |
Token ID | 284 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
ocket | +0.093 |
adiator | +0.089 |
Explore | +0.088 |
replay | +0.087 |
rium | +0.086 |
Neg logit contributions | |
unden | -0.156 |
rall | -0.153 |
arrang | -0.144 |
reluct | -0.143 |
'. | -0.141 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
unden | +0.272 |
rall | +0.266 |
reluct | +0.249 |
arrang | +0.248 |
fort | +0.244 |
Neg logit contributions | |
ocket | -0.190 |
replay | -0.184 |
adiator | -0.181 |
Explore | -0.180 |
paces | -0.178 |
Token | it |
Token ID | 340 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
unden | +3.350 |
rall | +3.280 |
arrang | +3.079 |
reluct | +3.031 |
'. | +3.008 |
Neg logit contributions | |
ocket | -2.652 |
adiator | -2.557 |
Explore | -2.539 |
rium | -2.493 |
iversity | -2.485 |
Token | by |
Token ID | 416 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
ocket | +0.023 |
adiator | +0.022 |
replay | +0.021 |
Explore | +0.021 |
paces | +0.021 |
Neg logit contributions | |
unden | -0.035 |
rall | -0.035 |
arrang | -0.032 |
reluct | -0.032 |
fort | -0.031 |
Token | spending |
Token ID | 4581 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
unden | +0.461 |
rall | +0.450 |
arrang | +0.423 |
reluct | +0.420 |
fort | +0.416 |
Neg logit contributions | |
ocket | -0.323 |
replay | -0.306 |
adiator | -0.306 |
Explore | -0.301 |
rium | -0.298 |
Token | less |
Token ID | 1342 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
unden | +0.342 |
rall | +0.335 |
arrang | +0.311 |
reluct | +0.309 |
'. | +0.302 |
Neg logit contributions | |
ocket | -0.317 |
adiator | -0.304 |
replay | -0.301 |
Explore | -0.301 |
rium | -0.299 |
Token | than |
Token ID | 621 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
ocket | +0.164 |
adiator | +0.156 |
Explore | +0.154 |
replay | +0.152 |
rium | +0.151 |
Neg logit contributions | |
unden | -0.288 |
rall | -0.282 |
arrang | -0.266 |
reluct | -0.264 |
fort | -0.261 |
Token | 't |
Token ID | 470 |
Feature activation | -2.83e-03 |
Pos logit contributions | |
unden | +0.237 |
rall | +0.231 |
arrang | +0.218 |
reluct | +0.216 |
'. | +0.209 |
Neg logit contributions | |
ocket | -0.169 |
adiator | -0.162 |
Explore | -0.161 |
replay | -0.158 |
rium | -0.158 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
unden | +0.006 |
rall | +0.005 |
reluct | +0.005 |
arrang | +0.005 |
fort | +0.005 |
Neg logit contributions | |
ocket | -0.004 |
replay | -0.004 |
adiator | -0.004 |
Explore | -0.004 |
paces | -0.004 |
Token | They |
Token ID | 1119 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ocket | +0.271 |
Explore | +0.259 |
adiator | +0.258 |
replay | +0.255 |
paces | +0.252 |
Neg logit contributions | |
unden | -0.374 |
rall | -0.365 |
arrang | -0.345 |
reluct | -0.342 |
fort | -0.332 |
Token | need |
Token ID | 761 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
ocket | +0.179 |
replay | +0.172 |
adiator | +0.171 |
Explore | +0.170 |
rium | +0.169 |
Neg logit contributions | |
unden | -0.216 |
rall | -0.210 |
arrang | -0.198 |
reluct | -0.195 |
fort | -0.191 |
Token | a |
Token ID | 257 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
ocket | +0.358 |
adiator | +0.343 |
replay | +0.340 |
Explore | +0.337 |
iversity | +0.334 |
Neg logit contributions | |
unden | -0.547 |
rall | -0.533 |
arrang | -0.506 |
reluct | -0.501 |
fort | -0.493 |
Token | * |
Token ID | 1635 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
ocket | +0.379 |
replay | +0.365 |
adiator | +0.363 |
Explore | +0.356 |
paces | +0.354 |
Neg logit contributions | |
unden | -0.462 |
rall | -0.448 |
arrang | -0.423 |
reluct | -0.418 |
fort | -0.411 |
Token | lot |
Token ID | 26487 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
unden | +3.397 |
rall | +3.323 |
arrang | +3.118 |
fort | +3.117 |
reluct | +3.074 |
Neg logit contributions | |
ocket | -2.536 |
adiator | -2.439 |
Explore | -2.384 |
rium | -2.360 |
replay | -2.357 |
Token | * |
Token ID | 9 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
ocket | +0.839 |
adiator | +0.806 |
Explore | +0.797 |
replay | +0.793 |
rium | +0.785 |
Neg logit contributions | |
unden | -1.167 |
rall | -1.141 |
arrang | -1.073 |
reluct | -1.057 |
'. | -1.040 |
Token | more |
Token ID | 517 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
unden | +0.598 |
rall | +0.585 |
arrang | +0.551 |
reluct | +0.543 |
fort | +0.537 |
Neg logit contributions | |
ocket | -0.426 |
adiator | -0.409 |
Explore | -0.404 |
replay | -0.400 |
rium | -0.397 |
Token | populism |
Token ID | 47584 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
ocket | +0.154 |
adiator | +0.147 |
replay | +0.146 |
Explore | +0.145 |
iversity | +0.143 |
Neg logit contributions | |
unden | -0.225 |
rall | -0.220 |
arrang | -0.208 |
reluct | -0.206 |
fort | -0.204 |
Token | - |
Token ID | 532 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
ocket | +0.070 |
adiator | +0.067 |
Explore | +0.066 |
replay | +0.065 |
rium | +0.065 |
Neg logit contributions | |
unden | -0.117 |
rall | -0.115 |
arrang | -0.108 |
reluct | -0.107 |
fort | -0.105 |
Token | memo |
Token ID | 16155 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
unden | +0.107 |
rall | +0.104 |
arrang | +0.097 |
reluct | +0.096 |
fort | +0.095 |
Neg logit contributions | |
ocket | -0.099 |
replay | -0.096 |
adiator | -0.095 |
Explore | -0.095 |
rium | -0.093 |
Token | either |
Token ID | 2035 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
ocket | +0.568 |
adiator | +0.538 |
replay | +0.528 |
Explore | +0.527 |
rium | +0.523 |
Neg logit contributions | |
unden | -0.854 |
rall | -0.838 |
arrang | -0.783 |
reluct | -0.778 |
'. | -0.763 |
Token | . |
Token ID | 13 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
unden | +1.238 |
rall | +1.214 |
arrang | +1.146 |
reluct | +1.134 |
'. | +1.121 |
Neg logit contributions | |
ocket | -0.737 |
adiator | -0.702 |
Explore | -0.691 |
replay | -0.684 |
rium | -0.680 |
Token | In |
Token ID | 554 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
unden | +0.107 |
rall | +0.104 |
arrang | +0.098 |
reluct | +0.098 |
fort | +0.094 |
Neg logit contributions | |
ocket | -0.079 |
Explore | -0.076 |
adiator | -0.075 |
replay | -0.075 |
rium | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
unden | +0.201 |
rall | +0.195 |
arrang | +0.183 |
reluct | +0.181 |
fort | +0.178 |
Neg logit contributions | |
ocket | -0.179 |
adiator | -0.173 |
Explore | -0.171 |
replay | -0.169 |
rium | -0.168 |
Token | recent |
Token ID | 2274 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
unden | +0.292 |
rall | +0.285 |
arrang | +0.267 |
reluct | +0.264 |
fort | +0.261 |
Neg logit contributions | |
ocket | -0.236 |
adiator | -0.227 |
Explore | -0.225 |
replay | -0.225 |
rium | -0.222 |
Token | UN |
Token ID | 4725 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
ocket | +2.066 |
replay | +2.026 |
adiator | +1.989 |
Explore | +1.966 |
rium | +1.951 |
Neg logit contributions | |
unden | -2.416 |
rall | -2.323 |
arrang | -2.170 |
reluct | -2.160 |
fort | -2.138 |
Token | report |
Token ID | 989 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
ocket | +0.691 |
adiator | +0.660 |
replay | +0.653 |
Explore | +0.652 |
rium | +0.642 |
Neg logit contributions | |
unden | -1.029 |
rall | -1.002 |
arrang | -0.942 |
reluct | -0.937 |
'. | -0.923 |
Token | investigating |
Token ID | 10240 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
ocket | +0.273 |
adiator | +0.261 |
Explore | +0.258 |
replay | +0.255 |
rium | +0.253 |
Neg logit contributions | |
unden | -0.426 |
rall | -0.417 |
arrang | -0.392 |
reluct | -0.389 |
fort | -0.384 |
Token | Protective |
Token ID | 45004 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
unden | +0.651 |
rall | +0.635 |
arrang | +0.593 |
reluct | +0.587 |
'. | +0.579 |
Neg logit contributions | |
ocket | -0.563 |
adiator | -0.543 |
Explore | -0.536 |
replay | -0.534 |
rium | -0.529 |
Token | Edge |
Token ID | 13113 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
unden | +0.771 |
rall | +0.752 |
arrang | +0.701 |
reluct | +0.696 |
'. | +0.679 |
Neg logit contributions | |
ocket | -0.690 |
adiator | -0.664 |
Explore | -0.657 |
replay | -0.652 |
rium | -0.648 |
Token | 43 |
Token ID | 3559 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
ocket | +0.302 |
adiator | +0.289 |
Explore | +0.288 |
replay | +0.284 |
rium | +0.284 |
Neg logit contributions | |
unden | -0.401 |
rall | -0.389 |
arrang | -0.367 |
reluct | -0.365 |
fort | -0.352 |
Token | in |
Token ID | 287 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
unden | +0.092 |
rall | +0.090 |
arrang | +0.085 |
reluct | +0.084 |
fort | +0.083 |
Neg logit contributions | |
ocket | -0.069 |
adiator | -0.067 |
Explore | -0.066 |
replay | -0.065 |
rium | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | -5.58e-02 |
Pos logit contributions | |
ocket | +0.076 |
adiator | +0.073 |
replay | +0.072 |
Explore | +0.072 |
rium | +0.072 |
Neg logit contributions | |
unden | -0.097 |
rall | -0.094 |
arrang | -0.088 |
reluct | -0.088 |
fort | -0.086 |
Token | 2014 |
Token ID | 1946 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
unden | +0.121 |
rall | +0.118 |
arrang | +0.112 |
reluct | +0.111 |
fort | +0.110 |
Neg logit contributions | |
ocket | -0.068 |
adiator | -0.065 |
Explore | -0.064 |
replay | -0.063 |
rium | -0.063 |
Token | National |
Token ID | 2351 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
ocket | +0.882 |
adiator | +0.855 |
replay | +0.848 |
Explore | +0.848 |
rium | +0.842 |
Neg logit contributions | |
unden | -0.873 |
rall | -0.851 |
arrang | -0.790 |
reluct | -0.784 |
fort | -0.769 |
Token | Draft |
Token ID | 13650 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
unden | +0.677 |
rall | +0.661 |
arrang | +0.618 |
reluct | +0.611 |
fort | +0.601 |
Neg logit contributions | |
ocket | -0.590 |
adiator | -0.569 |
Explore | -0.563 |
replay | -0.558 |
rium | -0.556 |
Token | , |
Token ID | 11 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
ocket | +1.881 |
adiator | +1.791 |
replay | +1.760 |
Explore | +1.750 |
iversity | +1.735 |
Neg logit contributions | |
unden | -3.125 |
rall | -3.064 |
arrang | -2.905 |
reluct | -2.861 |
fort | -2.805 |
Token | Mc |
Token ID | 1982 |
Feature activation | +6.78e-01 |
Pos logit contributions | |
ocket | +0.090 |
adiator | +0.086 |
Explore | +0.086 |
replay | +0.085 |
rium | +0.084 |
Neg logit contributions | |
unden | -0.087 |
rall | -0.084 |
arrang | -0.079 |
reluct | -0.077 |
fort | -0.076 |
Token | Govern |
Token ID | 29168 |
Feature activation | -6.31e-01 |
Pos logit contributions | |
ocket | +1.216 |
adiator | +1.174 |
Explore | +1.168 |
replay | +1.153 |
rium | +1.149 |
Neg logit contributions | |
unden | -1.086 |
rall | -1.058 |
arrang | -0.981 |
reluct | -0.969 |
'. | -0.943 |
Token | made |
Token ID | 925 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
unden | +1.195 |
rall | +1.167 |
arrang | +1.094 |
reluct | +1.081 |
fort | +1.067 |
Neg logit contributions | |
ocket | -0.941 |
adiator | -0.905 |
Explore | -0.896 |
replay | -0.885 |
rium | -0.880 |
Token | his |
Token ID | 465 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
unden | +0.246 |
rall | +0.240 |
arrang | +0.224 |
reluct | +0.224 |
fort | +0.220 |
Neg logit contributions | |
ocket | -0.170 |
adiator | -0.162 |
replay | -0.161 |
Explore | -0.161 |
rium | -0.159 |
Token | Knowledge |
Token ID | 20414 |
Feature activation | +7.79e-02 |
Pos logit contributions | |
ocket | +0.170 |
adiator | +0.163 |
Explore | +0.161 |
replay | +0.159 |
rium | +0.159 |
Neg logit contributions | |
unden | -0.235 |
rall | -0.230 |
arrang | -0.216 |
reluct | -0.214 |
'. | -0.210 |
Token | Fund |
Token ID | 7557 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
ocket | +0.117 |
Explore | +0.112 |
adiator | +0.112 |
rium | +0.110 |
replay | +0.110 |
Neg logit contributions | |
unden | -0.147 |
rall | -0.143 |
arrang | -0.135 |
reluct | -0.133 |
'. | -0.131 |
Token | , |
Token ID | 11 |
Feature activation | -7.92e-02 |
Pos logit contributions | |
ocket | +0.134 |
adiator | +0.127 |
Explore | +0.126 |
replay | +0.124 |
rium | +0.123 |
Neg logit contributions | |
unden | -0.222 |
rall | -0.217 |
arrang | -0.205 |
reluct | -0.203 |
'. | -0.199 |
Token | a |
Token ID | 257 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
unden | +0.153 |
rall | +0.149 |
arrang | +0.140 |
reluct | +0.139 |
fort | +0.137 |
Neg logit contributions | |
ocket | -0.115 |
Explore | -0.110 |
adiator | -0.110 |
replay | -0.108 |
rium | -0.108 |
Token | venture |
Token ID | 13189 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
unden | +0.112 |
rall | +0.109 |
arrang | +0.103 |
reluct | +0.102 |
fort | +0.100 |
Neg logit contributions | |
ocket | -0.086 |
adiator | -0.083 |
Explore | -0.082 |
replay | -0.081 |
rium | -0.081 |
Token | capital |
Token ID | 3139 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ocket | +0.414 |
adiator | +0.396 |
Explore | +0.394 |
replay | +0.388 |
rium | +0.388 |
Neg logit contributions | |
unden | -0.595 |
rall | -0.581 |
arrang | -0.546 |
reluct | -0.541 |
'. | -0.532 |
Token | fund |
Token ID | 1814 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
ocket | +1.725 |
adiator | +1.641 |
Explore | +1.632 |
replay | +1.628 |
rium | +1.613 |
Neg logit contributions | |
unden | -2.193 |
rall | -2.126 |
reluct | -1.983 |
arrang | -1.980 |
fort | -1.941 |
Token | of |
Token ID | 286 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
ocket | +0.629 |
Explore | +0.596 |
adiator | +0.596 |
replay | +0.584 |
rium | +0.584 |
Neg logit contributions | |
unden | -0.923 |
rall | -0.905 |
arrang | -0.848 |
reluct | -0.841 |
fort | -0.825 |
Token | $ |
Token ID | 720 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
ocket | +0.038 |
Explore | +0.037 |
adiator | +0.037 |
rium | +0.037 |
replay | +0.037 |
Neg logit contributions | |
unden | -0.043 |
rall | -0.042 |
arrang | -0.039 |
reluct | -0.038 |
fort | -0.038 |
Token | 50 |
Token ID | 1120 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
ocket | +0.268 |
Explore | +0.254 |
adiator | +0.253 |
rium | +0.246 |
replay | +0.246 |
Neg logit contributions | |
unden | -0.578 |
rall | -0.563 |
arrang | -0.536 |
reluct | -0.535 |
'. | -0.515 |
Token | million |
Token ID | 1510 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
unden | +1.397 |
rall | +1.364 |
arrang | +1.283 |
reluct | +1.271 |
'. | +1.252 |
Neg logit contributions | |
ocket | -1.031 |
adiator | -0.989 |
Explore | -0.978 |
replay | -0.966 |
rium | -0.962 |
Token | it |
Token ID | 340 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
unden | +1.394 |
rall | +1.364 |
arrang | +1.284 |
reluct | +1.271 |
fort | +1.255 |
Neg logit contributions | |
ocket | -0.970 |
adiator | -0.933 |
Explore | -0.923 |
rium | -0.905 |
replay | -0.904 |
Token | was |
Token ID | 373 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
unden | +0.089 |
rall | +0.087 |
arrang | +0.081 |
reluct | +0.080 |
'. | +0.079 |
Neg logit contributions | |
ocket | -0.072 |
adiator | -0.069 |
replay | -0.069 |
Explore | -0.068 |
rium | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
ocket | +0.215 |
replay | +0.204 |
adiator | +0.203 |
Explore | +0.201 |
rium | +0.199 |
Neg logit contributions | |
unden | -0.273 |
rall | -0.267 |
arrang | -0.251 |
reluct | -0.248 |
fort | -0.245 |
Token | Caps |
Token ID | 23534 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
unden | +0.090 |
rall | +0.088 |
arrang | +0.082 |
reluct | +0.081 |
fort | +0.080 |
Neg logit contributions | |
ocket | -0.070 |
adiator | -0.067 |
replay | -0.066 |
Explore | -0.066 |
rium | -0.065 |
Token | who |
Token ID | 508 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
ocket | +0.689 |
adiator | +0.652 |
replay | +0.647 |
Explore | +0.645 |
rium | +0.637 |
Neg logit contributions | |
unden | -0.987 |
rall | -0.962 |
arrang | -0.909 |
reluct | -0.896 |
'. | -0.884 |
Token | struck |
Token ID | 7425 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ocket | +0.418 |
replay | +0.399 |
adiator | +0.394 |
Explore | +0.391 |
paces | +0.391 |
Neg logit contributions | |
unden | -0.527 |
rall | -0.509 |
arrang | -0.479 |
reluct | -0.474 |
'. | -0.464 |
Token | first |
Token ID | 717 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
ocket | +1.494 |
replay | +1.422 |
adiator | +1.411 |
rium | +1.397 |
Explore | +1.390 |
Neg logit contributions | |
unden | -2.385 |
rall | -2.337 |
arrang | -2.193 |
reluct | -2.177 |
'. | -2.139 |
Token | when |
Token ID | 618 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
unden | +0.597 |
rall | +0.585 |
arrang | +0.552 |
reluct | +0.546 |
fort | +0.540 |
Neg logit contributions | |
ocket | -0.381 |
adiator | -0.365 |
Explore | -0.361 |
replay | -0.355 |
rium | -0.354 |
Token | Alex |
Token ID | 4422 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
ocket | +0.798 |
replay | +0.764 |
adiator | +0.762 |
Explore | +0.753 |
rium | +0.747 |
Neg logit contributions | |
unden | -0.838 |
rall | -0.812 |
arrang | -0.759 |
reluct | -0.749 |
fort | -0.740 |
Token | O |
Token ID | 440 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
unden | +0.280 |
rall | +0.272 |
arrang | +0.251 |
reluct | +0.247 |
fort | +0.243 |
Neg logit contributions | |
ocket | -0.342 |
adiator | -0.331 |
Explore | -0.328 |
replay | -0.326 |
rium | -0.324 |
Token | ve |
Token ID | 303 |
Feature activation | -2.38e-01 |
Pos logit contributions | |
ocket | +0.064 |
adiator | +0.060 |
paces | +0.059 |
Explore | +0.059 |
replay | +0.059 |
Neg logit contributions | |
unden | -0.115 |
rall | -0.113 |
arrang | -0.106 |
reluct | -0.106 |
fort | -0.103 |
Token | projected |
Token ID | 13301 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
unden | +0.271 |
rall | +0.266 |
arrang | +0.248 |
reluct | +0.245 |
fort | +0.243 |
Neg logit contributions | |
ocket | -0.230 |
adiator | -0.220 |
Explore | -0.219 |
replay | -0.218 |
rium | -0.216 |
Token | costs |
Token ID | 3484 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
unden | +0.286 |
rall | +0.280 |
arrang | +0.262 |
reluct | +0.259 |
fort | +0.256 |
Neg logit contributions | |
ocket | -0.233 |
adiator | -0.224 |
Explore | -0.222 |
replay | -0.220 |
rium | -0.218 |
Token | for |
Token ID | 329 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
ocket | +0.191 |
adiator | +0.183 |
Explore | +0.181 |
replay | +0.179 |
rium | +0.178 |
Neg logit contributions | |
unden | -0.268 |
rall | -0.263 |
arrang | -0.247 |
reluct | -0.244 |
'. | -0.241 |
Token | power |
Token ID | 1176 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
ocket | +0.394 |
adiator | +0.377 |
Explore | +0.376 |
replay | +0.374 |
rium | +0.369 |
Neg logit contributions | |
unden | -0.511 |
rall | -0.502 |
arrang | -0.469 |
reluct | -0.465 |
fort | -0.460 |
Token | from |
Token ID | 422 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
ocket | +0.592 |
adiator | +0.568 |
Explore | +0.565 |
replay | +0.562 |
rium | +0.554 |
Neg logit contributions | |
unden | -0.764 |
rall | -0.748 |
arrang | -0.698 |
reluct | -0.693 |
fort | -0.682 |
Token | natural |
Token ID | 3288 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ocket | +0.152 |
adiator | +0.146 |
Explore | +0.145 |
replay | +0.144 |
rium | +0.143 |
Neg logit contributions | |
unden | -0.206 |
rall | -0.202 |
arrang | -0.190 |
reluct | -0.188 |
fort | -0.186 |
Token | gas |
Token ID | 3623 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
ocket | +1.515 |
adiator | +1.450 |
replay | +1.447 |
Explore | +1.447 |
rium | +1.411 |
Neg logit contributions | |
unden | -2.327 |
rall | -2.281 |
arrang | -2.160 |
reluct | -2.133 |
'. | -2.097 |
Token | supplied |
Token ID | 14275 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
unden | +0.431 |
rall | +0.421 |
arrang | +0.396 |
reluct | +0.392 |
fort | +0.387 |
Neg logit contributions | |
ocket | -0.311 |
adiator | -0.298 |
Explore | -0.295 |
replay | -0.292 |
rium | -0.290 |
Token | power |
Token ID | 1176 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
unden | +0.584 |
rall | +0.573 |
arrang | +0.540 |
reluct | +0.533 |
fort | +0.525 |
Neg logit contributions | |
ocket | -0.396 |
adiator | -0.380 |
Explore | -0.378 |
replay | -0.372 |
paces | -0.369 |
Token | plants |
Token ID | 6134 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ocket | +0.297 |
adiator | +0.285 |
Explore | +0.284 |
replay | +0.281 |
rium | +0.278 |
Neg logit contributions | |
unden | -0.372 |
rall | -0.363 |
arrang | -0.339 |
reluct | -0.337 |
'. | -0.330 |
Token | , |
Token ID | 11 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
ocket | +0.219 |
adiator | +0.210 |
Explore | +0.208 |
replay | +0.205 |
rium | +0.204 |
Neg logit contributions | |
unden | -0.290 |
rall | -0.283 |
arrang | -0.266 |
reluct | -0.263 |
fort | -0.259 |
Token | and |
Token ID | 290 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
ocket | +0.433 |
adiator | +0.412 |
Explore | +0.408 |
replay | +0.407 |
rium | +0.402 |
Neg logit contributions | |
unden | -0.664 |
rall | -0.652 |
reluct | -0.606 |
arrang | -0.605 |
fort | -0.597 |
Token | found |
Token ID | 1043 |
Feature activation | +9.35e-01 |
Pos logit contributions | |
unden | +0.054 |
rall | +0.053 |
arrang | +0.049 |
reluct | +0.049 |
fort | +0.049 |
Neg logit contributions | |
ocket | -0.044 |
adiator | -0.042 |
Explore | -0.042 |
replay | -0.042 |
rium | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
ocket | +1.285 |
Explore | +1.213 |
replay | +1.211 |
adiator | +1.210 |
rium | +1.192 |
Neg logit contributions | |
unden | -1.862 |
rall | -1.821 |
arrang | -1.709 |
reluct | -1.691 |
fort | -1.672 |
Token | private |
Token ID | 2839 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
unden | +0.059 |
rall | +0.057 |
arrang | +0.054 |
reluct | +0.053 |
fort | +0.052 |
Neg logit contributions | |
ocket | -0.049 |
adiator | -0.047 |
replay | -0.046 |
Explore | -0.046 |
rium | -0.046 |
Token | route |
Token ID | 6339 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
ocket | +0.595 |
adiator | +0.567 |
Explore | +0.560 |
rium | +0.559 |
replay | +0.559 |
Neg logit contributions | |
unden | -0.746 |
rall | -0.728 |
arrang | -0.676 |
reluct | -0.673 |
fort | -0.664 |
Token | was |
Token ID | 373 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
unden | +0.437 |
rall | +0.429 |
arrang | +0.402 |
reluct | +0.397 |
fort | +0.392 |
Neg logit contributions | |
ocket | -0.317 |
adiator | -0.303 |
Explore | -0.301 |
replay | -0.297 |
rium | -0.294 |
Token | faster |
Token ID | 5443 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
ocket | +0.144 |
adiator | +0.137 |
Explore | +0.137 |
replay | +0.135 |
rium | +0.135 |
Neg logit contributions | |
unden | -0.171 |
rall | -0.168 |
arrang | -0.156 |
reluct | -0.155 |
fort | -0.152 |
Token | . |
Token ID | 13 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
ocket | +0.053 |
adiator | +0.051 |
Explore | +0.050 |
replay | +0.050 |
rium | +0.049 |
Neg logit contributions | |
unden | -0.083 |
rall | -0.081 |
arrang | -0.076 |
reluct | -0.076 |
fort | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
unden | +0.043 |
rall | +0.042 |
arrang | +0.040 |
reluct | +0.040 |
fort | +0.039 |
Neg logit contributions | |
ocket | -0.026 |
adiator | -0.024 |
Explore | -0.024 |
replay | -0.024 |
rium | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
unden | +0.301 |
rall | +0.295 |
arrang | +0.278 |
reluct | +0.275 |
fort | +0.272 |
Neg logit contributions | |
ocket | -0.191 |
adiator | -0.183 |
Explore | -0.181 |
replay | -0.178 |
rium | -0.177 |
Token | Their |
Token ID | 14574 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
unden | +0.384 |
rall | +0.377 |
arrang | +0.357 |
reluct | +0.354 |
fort | +0.350 |
Neg logit contributions | |
ocket | -0.183 |
adiator | -0.173 |
Explore | -0.170 |
replay | -0.167 |
rium | -0.166 |
Token | fit |
Token ID | 4197 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
unden | +1.274 |
rall | +1.244 |
arrang | +1.170 |
reluct | +1.158 |
'. | +1.142 |
Neg logit contributions | |
ocket | -0.952 |
adiator | -0.914 |
Explore | -0.903 |
replay | -0.895 |
rium | -0.889 |
Token | their |
Token ID | 511 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
unden | +0.783 |
rall | +0.766 |
arrang | +0.721 |
reluct | +0.715 |
fort | +0.707 |
Neg logit contributions | |
ocket | -0.505 |
adiator | -0.483 |
Explore | -0.477 |
replay | -0.475 |
rium | -0.469 |
Token | locker |
Token ID | 17355 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ocket | +0.121 |
replay | +0.117 |
adiator | +0.116 |
Explore | +0.115 |
rium | +0.112 |
Neg logit contributions | |
unden | -0.157 |
rall | -0.153 |
arrang | -0.143 |
reluct | -0.142 |
fort | -0.141 |
Token | room |
Token ID | 2119 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
ocket | +1.359 |
adiator | +1.281 |
replay | +1.265 |
Explore | +1.252 |
rium | +1.230 |
Neg logit contributions | |
unden | -3.148 |
rall | -3.083 |
reluct | -2.927 |
arrang | -2.895 |
'. | -2.842 |
Token | and |
Token ID | 290 |
Feature activation | +3.26e-02 |
Pos logit contributions | |
ocket | +1.314 |
replay | +1.254 |
adiator | +1.253 |
Explore | +1.240 |
rium | +1.216 |
Neg logit contributions | |
unden | -2.063 |
rall | -2.038 |
reluct | -1.896 |
arrang | -1.874 |
'. | -1.851 |
Token | culture |
Token ID | 3968 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
ocket | +0.046 |
adiator | +0.044 |
Explore | +0.044 |
replay | +0.044 |
paces | +0.043 |
Neg logit contributions | |
unden | -0.064 |
rall | -0.063 |
arrang | -0.059 |
reluct | -0.059 |
fort | -0.058 |
Token | , |
Token ID | 11 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ocket | +0.866 |
adiator | +0.824 |
replay | +0.818 |
Explore | +0.818 |
paces | +0.811 |
Neg logit contributions | |
unden | -1.534 |
rall | -1.510 |
reluct | -1.409 |
arrang | -1.408 |
fort | -1.378 |
Token | and |
Token ID | 290 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
ocket | +0.159 |
adiator | +0.153 |
Explore | +0.152 |
replay | +0.151 |
rium | +0.148 |
Neg logit contributions | |
unden | -0.215 |
rall | -0.210 |
arrang | -0.198 |
reluct | -0.195 |
fort | -0.194 |
Token | they |
Token ID | 484 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
unden | +0.026 |
rall | +0.026 |
arrang | +0.024 |
reluct | +0.024 |
fort | +0.024 |
Neg logit contributions | |
ocket | -0.021 |
adiator | -0.020 |
Explore | -0.020 |
replay | -0.019 |
rium | -0.019 |
Token | 're |
Token ID | 821 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
ocket | +0.394 |
replay | +0.379 |
Explore | +0.374 |
adiator | +0.373 |
rium | +0.368 |
Neg logit contributions | |
unden | -0.567 |
rall | -0.550 |
arrang | -0.519 |
reluct | -0.513 |
fort | -0.504 |
Token | desperately |
Token ID | 16459 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
ocket | +0.106 |
adiator | +0.101 |
replay | +0.100 |
Explore | +0.100 |
paces | +0.098 |
Neg logit contributions | |
unden | -0.156 |
rall | -0.152 |
arrang | -0.144 |
reluct | -0.141 |
fort | -0.140 |
Token | border |
Token ID | 4865 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
unden | +0.303 |
rall | +0.294 |
arrang | +0.275 |
reluct | +0.273 |
fort | +0.270 |
Neg logit contributions | |
ocket | -0.275 |
adiator | -0.265 |
replay | -0.264 |
Explore | -0.262 |
rium | -0.261 |
Token | of |
Token ID | 286 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
unden | +0.500 |
rall | +0.490 |
arrang | +0.463 |
reluct | +0.458 |
fort | +0.453 |
Neg logit contributions | |
ocket | -0.287 |
adiator | -0.274 |
Explore | -0.270 |
replay | -0.267 |
rium | -0.265 |
Token | the |
Token ID | 262 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
unden | +0.101 |
rall | +0.098 |
arrang | +0.092 |
reluct | +0.091 |
fort | +0.090 |
Neg logit contributions | |
ocket | -0.082 |
adiator | -0.079 |
replay | -0.079 |
rium | -0.078 |
Explore | -0.078 |
Token | hands |
Token ID | 2832 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
unden | +0.022 |
rall | +0.021 |
arrang | +0.020 |
reluct | +0.019 |
fort | +0.019 |
Neg logit contributions | |
ocket | -0.018 |
replay | -0.017 |
adiator | -0.017 |
Explore | -0.017 |
rium | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -9.56e-03 |
Pos logit contributions | |
ocket | +0.371 |
adiator | +0.349 |
paces | +0.344 |
Explore | +0.343 |
replay | +0.342 |
Neg logit contributions | |
unden | -0.595 |
rall | -0.583 |
arrang | -0.548 |
reluct | -0.544 |
fort | -0.531 |
Token | feet |
Token ID | 3625 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
unden | +0.018 |
rall | +0.018 |
arrang | +0.017 |
reluct | +0.017 |
fort | +0.016 |
Neg logit contributions | |
ocket | -0.014 |
adiator | -0.013 |
Explore | -0.013 |
replay | -0.013 |
rium | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
ocket | +0.550 |
adiator | +0.515 |
Explore | +0.511 |
paces | +0.505 |
replay | +0.505 |
Neg logit contributions | |
unden | -0.980 |
rall | -0.964 |
arrang | -0.907 |
reluct | -0.902 |
'. | -0.887 |
Token | and |
Token ID | 290 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
unden | +0.179 |
rall | +0.175 |
arrang | +0.165 |
reluct | +0.162 |
fort | +0.161 |
Neg logit contributions | |
ocket | -0.127 |
adiator | -0.121 |
Explore | -0.120 |
replay | -0.120 |
rium | -0.120 |
Token | wrists |
Token ID | 38163 |
Feature activation | +5.13e-01 |
Pos logit contributions | |
unden | +0.296 |
rall | +0.288 |
arrang | +0.271 |
reluct | +0.267 |
fort | +0.265 |
Neg logit contributions | |
ocket | -0.235 |
adiator | -0.225 |
Explore | -0.223 |
replay | -0.223 |
rium | -0.222 |
Token | .[ |
Token ID | 3693 |
Feature activation | +6.19e-01 |
Pos logit contributions | |
ocket | +0.653 |
adiator | +0.614 |
paces | +0.606 |
Explore | +0.605 |
replay | +0.603 |
Neg logit contributions | |
unden | -1.094 |
rall | -1.075 |
arrang | -1.009 |
reluct | -1.002 |
fort | -0.981 |
Token | 1 |
Token ID | 16 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
ocket | +0.662 |
Explore | +0.624 |
adiator | +0.623 |
replay | +0.611 |
rium | +0.604 |
Neg logit contributions | |
unden | -1.443 |
rall | -1.411 |
arrang | -1.344 |
reluct | -1.334 |
'. | -1.305 |
Token |
|
Token ID | 198 |
Feature activation | -8.15e-02 |
Pos logit contributions | |
unden | +0.099 |
rall | +0.097 |
arrang | +0.091 |
reluct | +0.090 |
fort | +0.089 |
Neg logit contributions | |
ocket | -0.062 |
adiator | -0.060 |
Explore | -0.059 |
replay | -0.058 |
rium | -0.058 |
Token | � |
Token ID | 447 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
unden | +0.174 |
rall | +0.171 |
arrang | +0.162 |
reluct | +0.160 |
'. | +0.158 |
Neg logit contributions | |
ocket | -0.102 |
adiator | -0.097 |
Explore | -0.096 |
replay | -0.094 |
rium | -0.094 |
Token | � |
Token ID | 250 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
unden | +0.271 |
rall | +0.265 |
arrang | +0.251 |
reluct | +0.248 |
fort | +0.245 |
Neg logit contributions | |
ocket | -0.166 |
adiator | -0.159 |
Explore | -0.157 |
replay | -0.154 |
rium | -0.154 |
Token | A |
Token ID | 32 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
unden | +0.283 |
rall | +0.278 |
arrang | +0.262 |
reluct | +0.260 |
fort | +0.257 |
Neg logit contributions | |
ocket | -0.167 |
adiator | -0.160 |
Explore | -0.158 |
replay | -0.155 |
rium | -0.154 |
Token | lot |
Token ID | 1256 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
unden | +0.052 |
rall | +0.050 |
arrang | +0.047 |
reluct | +0.047 |
fort | +0.046 |
Neg logit contributions | |
ocket | -0.043 |
adiator | -0.041 |
Explore | -0.040 |
replay | -0.040 |
rium | -0.040 |
Token | of |
Token ID | 286 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
ocket | +0.119 |
adiator | +0.113 |
Explore | +0.112 |
replay | +0.110 |
rium | +0.109 |
Neg logit contributions | |
unden | -0.213 |
rall | -0.209 |
arrang | -0.198 |
reluct | -0.196 |
fort | -0.194 |
Token | what |
Token ID | 644 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ocket | +0.395 |
adiator | +0.378 |
Explore | +0.377 |
replay | +0.374 |
rium | +0.372 |
Neg logit contributions | |
unden | -0.505 |
rall | -0.494 |
arrang | -0.463 |
reluct | -0.457 |
fort | -0.452 |
Token | we |
Token ID | 356 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
unden | +0.246 |
rall | +0.240 |
arrang | +0.225 |
reluct | +0.222 |
fort | +0.220 |
Neg logit contributions | |
ocket | -0.187 |
adiator | -0.180 |
Explore | -0.179 |
replay | -0.177 |
rium | -0.176 |
Token | know |
Token ID | 760 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
ocket | +0.077 |
Explore | +0.074 |
adiator | +0.074 |
replay | +0.073 |
rium | +0.072 |
Neg logit contributions | |
unden | -0.105 |
rall | -0.102 |
arrang | -0.097 |
reluct | -0.095 |
fort | -0.094 |
Token | about |
Token ID | 546 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
unden | +0.283 |
rall | +0.276 |
arrang | +0.255 |
reluct | +0.252 |
'. | +0.246 |
Neg logit contributions | |
ocket | -0.321 |
adiator | -0.311 |
Explore | -0.309 |
replay | -0.305 |
rium | -0.305 |
Token | Philadelphia |
Token ID | 8857 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
unden | +0.318 |
rall | +0.309 |
arrang | +0.289 |
reluct | +0.286 |
fort | +0.282 |
Neg logit contributions | |
ocket | -0.289 |
adiator | -0.278 |
Explore | -0.276 |
replay | -0.274 |
rium | -0.272 |
Token | is |
Token ID | 318 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
ocket | +0.620 |
adiator | +0.593 |
Explore | +0.589 |
replay | +0.588 |
iversity | +0.578 |
Neg logit contributions | |
unden | -0.973 |
rall | -0.955 |
arrang | -0.895 |
reluct | -0.890 |
'. | -0.879 |
Token | a |
Token ID | 257 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
unden | +0.362 |
rall | +0.354 |
arrang | +0.331 |
reluct | +0.328 |
'. | +0.323 |
Neg logit contributions | |
ocket | -0.302 |
adiator | -0.290 |
Explore | -0.288 |
replay | -0.285 |
rium | -0.283 |
Token | breath |
Token ID | 8033 |
Feature activation | +4.88e-01 |
Pos logit contributions | |
ocket | +0.255 |
adiator | +0.245 |
Explore | +0.242 |
replay | +0.241 |
rium | +0.238 |
Neg logit contributions | |
unden | -0.302 |
rall | -0.295 |
arrang | -0.276 |
reluct | -0.273 |
fort | -0.269 |
Token | of |
Token ID | 286 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
ocket | +0.570 |
adiator | +0.541 |
Explore | +0.533 |
replay | +0.527 |
rium | +0.519 |
Neg logit contributions | |
unden | -1.082 |
rall | -1.061 |
arrang | -1.006 |
reluct | -0.996 |
'. | -0.985 |
Token | fresh |
Token ID | 4713 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
ocket | +0.089 |
adiator | +0.086 |
Explore | +0.085 |
replay | +0.085 |
rium | +0.084 |
Neg logit contributions | |
unden | -0.077 |
rall | -0.075 |
arrang | -0.069 |
reluct | -0.068 |
fort | -0.067 |
Token | air |
Token ID | 1633 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
unden | +0.619 |
rall | +0.605 |
arrang | +0.565 |
reluct | +0.557 |
fort | +0.549 |
Neg logit contributions | |
ocket | -0.576 |
adiator | -0.556 |
Explore | -0.551 |
replay | -0.546 |
rium | -0.544 |
Token | in |
Token ID | 287 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
ocket | +0.147 |
adiator | +0.141 |
Explore | +0.139 |
replay | +0.137 |
rium | +0.137 |
Neg logit contributions | |
unden | -0.200 |
rall | -0.196 |
arrang | -0.184 |
reluct | -0.182 |
fort | -0.180 |
Token | a |
Token ID | 257 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
ocket | +0.047 |
adiator | +0.045 |
Explore | +0.045 |
replay | +0.045 |
rium | +0.044 |
Neg logit contributions | |
unden | -0.056 |
rall | -0.055 |
arrang | -0.052 |
reluct | -0.051 |
fort | -0.050 |
Token | TV |
Token ID | 3195 |
Feature activation | +7.74e-01 |
Pos logit contributions | |
ocket | +0.184 |
adiator | +0.177 |
Explore | +0.175 |
replay | +0.173 |
rium | +0.172 |
Neg logit contributions | |
unden | -0.239 |
rall | -0.233 |
arrang | -0.219 |
reluct | -0.217 |
fort | -0.214 |
Token | landscape |
Token ID | 10747 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
ocket | +1.176 |
adiator | +1.131 |
Explore | +1.120 |
replay | +1.110 |
rium | +1.101 |
Neg logit contributions | |
unden | -1.437 |
rall | -1.408 |
arrang | -1.317 |
reluct | -1.303 |
'. | -1.288 |
Token | that |
Token ID | 326 |
Feature activation | -9.94e-02 |
Pos logit contributions | |
ocket | +0.024 |
adiator | +0.023 |
Explore | +0.022 |
replay | +0.022 |
rium | +0.022 |
Neg logit contributions | |
unden | -0.037 |
rall | -0.036 |
arrang | -0.034 |
reluct | -0.034 |
fort | -0.033 |
Token | --+ |
Token ID | 44785 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
unden | +0.522 |
rall | +0.513 |
arrang | +0.487 |
'. | +0.481 |
fort | +0.480 |
Neg logit contributions | |
ocket | -0.263 |
adiator | -0.249 |
Explore | -0.245 |
replay | -0.242 |
rium | -0.241 |
Token | | |
Token ID | 930 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
unden | +0.752 |
rall | +0.728 |
arrang | +0.678 |
reluct | +0.670 |
fort | +0.650 |
Neg logit contributions | |
ocket | -0.808 |
adiator | -0.781 |
Explore | -0.776 |
replay | -0.770 |
rium | -0.764 |
Token | ... |
Token ID | 2644 |
Feature activation | +7.92e-01 |
Pos logit contributions | |
unden | +0.027 |
rall | +0.026 |
arrang | +0.025 |
reluct | +0.024 |
fort | +0.024 |
Neg logit contributions | |
ocket | -0.015 |
adiator | -0.014 |
Explore | -0.014 |
replay | -0.014 |
rium | -0.013 |
Token | | |
Token ID | 930 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
ocket | +1.315 |
Explore | +1.270 |
replay | +1.265 |
adiator | +1.265 |
rium | +1.249 |
Neg logit contributions | |
unden | -1.437 |
rall | -1.374 |
reluct | -1.296 |
arrang | -1.295 |
fort | -1.185 |
Token | + |
Token ID | 1343 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
unden | +0.054 |
rall | +0.052 |
arrang | +0.048 |
reluct | +0.048 |
fort | +0.047 |
Neg logit contributions | |
ocket | -0.062 |
adiator | -0.060 |
replay | -0.060 |
Explore | -0.060 |
rium | -0.059 |
Token | -------------------- |
Token ID | 19351 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
unden | +0.691 |
rall | +0.675 |
arrang | +0.634 |
reluct | +0.626 |
fort | +0.618 |
Neg logit contributions | |
ocket | -0.538 |
adiator | -0.517 |
Explore | -0.511 |
replay | -0.505 |
rium | -0.503 |
Token | --+ |
Token ID | 44785 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
unden | +0.465 |
rall | +0.456 |
arrang | +0.433 |
'. | +0.428 |
fort | +0.428 |
Neg logit contributions | |
ocket | -0.235 |
adiator | -0.222 |
Explore | -0.219 |
replay | -0.216 |
rium | -0.215 |
Token | + |
Token ID | 1343 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
unden | +0.643 |
rall | +0.621 |
arrang | +0.576 |
reluct | +0.571 |
fort | +0.549 |
Neg logit contributions | |
ocket | -0.745 |
adiator | -0.720 |
Explore | -0.717 |
replay | -0.715 |
rium | -0.707 |
Token | -------------------- |
Token ID | 19351 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
unden | +0.486 |
rall | +0.475 |
arrang | +0.446 |
reluct | +0.441 |
'. | +0.433 |
Neg logit contributions | |
ocket | -0.371 |
adiator | -0.356 |
Explore | -0.354 |
replay | -0.349 |
rium | -0.347 |
Token | --+ |
Token ID | 44785 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
unden | +0.329 |
rall | +0.322 |
arrang | +0.307 |
'. | +0.303 |
fort | +0.303 |
Neg logit contributions | |
ocket | -0.166 |
adiator | -0.158 |
Explore | -0.155 |
replay | -0.153 |
rium | -0.152 |
Token | < |
Token ID | 1279 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
unden | +0.685 |
rall | +0.667 |
arrang | +0.619 |
reluct | +0.610 |
fort | +0.600 |
Neg logit contributions | |
ocket | -0.711 |
adiator | -0.685 |
Explore | -0.681 |
replay | -0.677 |
rium | -0.670 |
Token | International |
Token ID | 24274 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
ocket | +0.806 |
Explore | +0.775 |
adiator | +0.771 |
Weak | +0.750 |
replay | +0.747 |
Neg logit contributions | |
unden | -1.358 |
rall | -1.331 |
arrang | -1.258 |
reluct | -1.252 |
'. | -1.220 |
Token | goals |
Token ID | 4661 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
unden | +0.944 |
rall | +0.922 |
arrang | +0.867 |
reluct | +0.856 |
'. | +0.846 |
Neg logit contributions | |
ocket | -0.710 |
adiator | -0.683 |
Explore | -0.675 |
replay | -0.665 |
rium | -0.663 |
Token |
|
Token ID | 198 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
ocket | +0.125 |
adiator | +0.120 |
Explore | +0.118 |
replay | +0.118 |
paces | +0.117 |
Neg logit contributions | |
unden | -0.164 |
rall | -0.160 |
arrang | -0.151 |
reluct | -0.149 |
fort | -0.147 |
Token |
|
Token ID | 198 |
Feature activation | +2.71e-01 |
Pos logit contributions | |
ocket | +0.439 |
adiator | +0.418 |
Explore | +0.417 |
replay | +0.410 |
rium | +0.407 |
Neg logit contributions | |
unden | -0.753 |
rall | -0.734 |
arrang | -0.697 |
reluct | -0.693 |
'. | -0.665 |
Token | As |
Token ID | 1722 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ocket | +0.332 |
Explore | +0.317 |
adiator | +0.317 |
replay | +0.307 |
rium | +0.306 |
Neg logit contributions | |
unden | -0.587 |
rall | -0.575 |
arrang | -0.544 |
reluct | -0.540 |
'. | -0.529 |
Token | of |
Token ID | 286 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ocket | +0.164 |
adiator | +0.158 |
Explore | +0.156 |
replay | +0.154 |
rium | +0.154 |
Neg logit contributions | |
unden | -0.212 |
rall | -0.207 |
arrang | -0.194 |
reluct | -0.192 |
fort | -0.189 |
Token | 12 |
Token ID | 1105 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
ocket | +0.024 |
adiator | +0.023 |
replay | +0.023 |
Explore | +0.023 |
rium | +0.022 |
Neg logit contributions | |
unden | -0.029 |
rall | -0.028 |
arrang | -0.027 |
reluct | -0.027 |
fort | -0.026 |
Token | June |
Token ID | 2795 |
Feature activation | -2.66e-01 |
Pos logit contributions | |
unden | +0.069 |
rall | +0.067 |
arrang | +0.063 |
reluct | +0.063 |
fort | +0.061 |
Neg logit contributions | |
ocket | -0.063 |
adiator | -0.061 |
Explore | -0.060 |
replay | -0.060 |
iversity | -0.059 |
Token | 2016 |
Token ID | 1584 |
Feature activation | +2.77e-02 |
Pos logit contributions | |
unden | +0.640 |
rall | +0.628 |
arrang | +0.598 |
reluct | +0.592 |
fort | +0.586 |
Neg logit contributions | |
ocket | -0.260 |
adiator | -0.244 |
Explore | -0.241 |
replay | -0.237 |
rium | -0.234 |
Token | [ |
Token ID | 58 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
ocket | +0.038 |
adiator | +0.036 |
Explore | +0.036 |
replay | +0.035 |
rium | +0.035 |
Neg logit contributions | |
unden | -0.056 |
rall | -0.055 |
arrang | -0.052 |
reluct | -0.051 |
fort | -0.050 |
Token | 91 |
Token ID | 6420 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
unden | +0.519 |
rall | +0.506 |
arrang | +0.475 |
reluct | +0.470 |
'. | +0.460 |
Neg logit contributions | |
ocket | -0.408 |
adiator | -0.391 |
Explore | -0.389 |
replay | -0.383 |
rium | -0.381 |
Token | U |
Token ID | 471 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
unden | +0.681 |
rall | +0.665 |
arrang | +0.624 |
reluct | +0.618 |
fort | +0.611 |
Neg logit contributions | |
ocket | -0.498 |
adiator | -0.472 |
replay | -0.472 |
Explore | -0.469 |
rium | -0.462 |
Token | . |
Token ID | 13 |
Feature activation | +9.23e-04 |
Pos logit contributions | |
ocket | +0.265 |
adiator | +0.252 |
Explore | +0.251 |
replay | +0.246 |
rium | +0.245 |
Neg logit contributions | |
unden | -0.462 |
rall | -0.452 |
arrang | -0.428 |
reluct | -0.425 |
'. | -0.414 |
Token | S |
Token ID | 50 |
Feature activation | -6.53e-02 |
Pos logit contributions | |
ocket | +0.001 |
adiator | +0.001 |
Explore | +0.001 |
replay | +0.001 |
rium | +0.001 |
Neg logit contributions | |
unden | -0.002 |
rall | -0.002 |
arrang | -0.002 |
reluct | -0.002 |
fort | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
unden | +0.144 |
rall | +0.141 |
arrang | +0.134 |
reluct | +0.133 |
fort | +0.130 |
Neg logit contributions | |
ocket | -0.078 |
adiator | -0.074 |
Explore | -0.073 |
rium | -0.072 |
replay | -0.072 |
Token | energy |
Token ID | 2568 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
unden | +0.275 |
rall | +0.269 |
arrang | +0.254 |
reluct | +0.251 |
fort | +0.249 |
Neg logit contributions | |
ocket | -0.176 |
adiator | -0.167 |
Explore | -0.165 |
replay | -0.165 |
rium | -0.163 |
Token | policy |
Token ID | 2450 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
ocket | +0.151 |
adiator | +0.144 |
replay | +0.143 |
Explore | +0.143 |
rium | +0.140 |
Neg logit contributions | |
unden | -0.200 |
rall | -0.195 |
arrang | -0.183 |
reluct | -0.183 |
'. | -0.180 |
Token | , |
Token ID | 11 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
unden | +0.827 |
rall | +0.809 |
arrang | +0.763 |
reluct | +0.756 |
fort | +0.746 |
Neg logit contributions | |
ocket | -0.519 |
adiator | -0.495 |
Explore | -0.489 |
replay | -0.485 |
rium | -0.480 |
Token | something |
Token ID | 1223 |
Feature activation | -7.45e-01 |
Pos logit contributions | |
ocket | +0.082 |
Explore | +0.079 |
replay | +0.079 |
adiator | +0.078 |
paces | +0.077 |
Neg logit contributions | |
unden | -0.102 |
rall | -0.099 |
arrang | -0.093 |
reluct | -0.093 |
fort | -0.091 |
Token | that |
Token ID | 326 |
Feature activation | -6.40e-02 |
Pos logit contributions | |
unden | +1.402 |
rall | +1.369 |
arrang | +1.283 |
reluct | +1.268 |
fort | +1.253 |
Neg logit contributions | |
ocket | -1.119 |
adiator | -1.078 |
Explore | -1.064 |
replay | -1.051 |
rium | -1.048 |
Token | the |
Token ID | 262 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
unden | +0.121 |
rall | +0.118 |
arrang | +0.111 |
reluct | +0.109 |
fort | +0.108 |
Neg logit contributions | |
ocket | -0.095 |
adiator | -0.091 |
Explore | -0.091 |
replay | -0.090 |
rium | -0.089 |
Token | country |
Token ID | 1499 |
Feature activation | -6.54e-03 |
Pos logit contributions | |
ocket | +0.047 |
adiator | +0.045 |
replay | +0.045 |
Explore | +0.045 |
rium | +0.045 |
Neg logit contributions | |
unden | -0.056 |
rall | -0.054 |
arrang | -0.051 |
reluct | -0.050 |
fort | -0.050 |
Token | period |
Token ID | 2278 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
ocket | +0.110 |
adiator | +0.105 |
replay | +0.104 |
Explore | +0.103 |
paces | +0.102 |
Neg logit contributions | |
unden | -0.143 |
rall | -0.140 |
arrang | -0.131 |
reluct | -0.130 |
fort | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
ocket | +0.292 |
adiator | +0.279 |
Explore | +0.272 |
replay | +0.271 |
iversity | +0.270 |
Neg logit contributions | |
unden | -0.522 |
rall | -0.513 |
arrang | -0.484 |
reluct | -0.480 |
'. | -0.473 |
Token | But |
Token ID | 887 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
unden | +0.116 |
rall | +0.112 |
arrang | +0.107 |
reluct | +0.107 |
fort | +0.102 |
Neg logit contributions | |
ocket | -0.082 |
Explore | -0.078 |
adiator | -0.077 |
replay | -0.077 |
paces | -0.076 |
Token | Section |
Token ID | 7275 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
ocket | +0.497 |
replay | +0.472 |
adiator | +0.466 |
Explore | +0.462 |
paces | +0.459 |
Neg logit contributions | |
unden | -0.680 |
rall | -0.659 |
arrang | -0.626 |
reluct | -0.621 |
fort | -0.611 |
Token | 2 |
Token ID | 362 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
ocket | +1.602 |
adiator | +1.535 |
replay | +1.528 |
Explore | +1.513 |
rium | +1.492 |
Neg logit contributions | |
unden | -2.009 |
rall | -1.938 |
arrang | -1.830 |
reluct | -1.815 |
fort | -1.783 |
Token | of |
Token ID | 286 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
ocket | +0.293 |
adiator | +0.277 |
Explore | +0.272 |
replay | +0.270 |
paces | +0.267 |
Neg logit contributions | |
unden | -0.496 |
rall | -0.486 |
arrang | -0.460 |
reluct | -0.455 |
'. | -0.450 |
Token | the |
Token ID | 262 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
ocket | +0.036 |
adiator | +0.034 |
replay | +0.034 |
Explore | +0.034 |
rium | +0.034 |
Neg logit contributions | |
unden | -0.040 |
rall | -0.038 |
arrang | -0.036 |
reluct | -0.036 |
fort | -0.035 |
Token | law |
Token ID | 1099 |
Feature activation | -6.43e-01 |
Pos logit contributions | |
unden | +0.080 |
rall | +0.077 |
arrang | +0.074 |
reluct | +0.073 |
fort | +0.072 |
Neg logit contributions | |
ocket | -0.057 |
adiator | -0.054 |
replay | -0.054 |
Explore | -0.053 |
rium | -0.053 |
Token | wasn |
Token ID | 2492 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
unden | +1.254 |
rall | +1.225 |
arrang | +1.152 |
reluct | +1.139 |
fort | +1.125 |
Neg logit contributions | |
ocket | -0.923 |
adiator | -0.885 |
Explore | -0.875 |
replay | -0.866 |
rium | -0.860 |
Token | 't |
Token ID | 470 |
Feature activation | -3.86e-01 |
Pos logit contributions | |
unden | +0.848 |
rall | +0.829 |
arrang | +0.779 |
reluct | +0.769 |
fort | +0.762 |
Neg logit contributions | |
ocket | -0.638 |
adiator | -0.613 |
Explore | -0.606 |
replay | -0.599 |
rium | -0.596 |
Token | affected |
Token ID | 5676 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
unden | +0.740 |
rall | +0.721 |
arrang | +0.680 |
reluct | +0.674 |
fort | +0.664 |
Neg logit contributions | |
ocket | -0.574 |
adiator | -0.542 |
replay | -0.539 |
Explore | -0.537 |
rium | -0.528 |
Token | the |
Token ID | 262 |
Feature activation | -5.92e-02 |
Pos logit contributions | |
ocket | +1.116 |
adiator | +1.077 |
replay | +1.073 |
Explore | +1.072 |
rium | +1.063 |
Neg logit contributions | |
unden | -1.315 |
rall | -1.282 |
arrang | -1.203 |
reluct | -1.196 |
'. | -1.164 |
Token | isolated |
Token ID | 11557 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
unden | +0.104 |
rall | +0.101 |
arrang | +0.095 |
reluct | +0.094 |
'. | +0.092 |
Neg logit contributions | |
ocket | -0.095 |
adiator | -0.092 |
Explore | -0.091 |
replay | -0.091 |
rium | -0.090 |
Token | settlement |
Token ID | 9443 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
unden | +0.747 |
rall | +0.726 |
arrang | +0.679 |
reluct | +0.675 |
'. | +0.664 |
Neg logit contributions | |
ocket | -0.609 |
adiator | -0.587 |
Explore | -0.582 |
replay | -0.580 |
rium | -0.572 |
Token | - |
Token ID | 12 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
ocket | +0.390 |
adiator | +0.374 |
Explore | +0.372 |
replay | +0.367 |
rium | +0.366 |
Neg logit contributions | |
unden | -0.526 |
rall | -0.514 |
reluct | -0.479 |
arrang | -0.479 |
'. | -0.472 |
Token | city |
Token ID | 19205 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
ocket | +0.041 |
Explore | +0.040 |
adiator | +0.039 |
rium | +0.039 |
paces | +0.039 |
Neg logit contributions | |
unden | -0.055 |
rall | -0.053 |
reluct | -0.050 |
arrang | -0.049 |
'. | -0.048 |
Token | of |
Token ID | 286 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
unden | +0.497 |
rall | +0.485 |
arrang | +0.455 |
reluct | +0.452 |
'. | +0.446 |
Neg logit contributions | |
ocket | -0.351 |
adiator | -0.337 |
Explore | -0.333 |
replay | -0.329 |
rium | -0.328 |
Token | Ariel |
Token ID | 33364 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
ocket | +0.022 |
adiator | +0.021 |
Explore | +0.021 |
rium | +0.021 |
replay | +0.021 |
Neg logit contributions | |
unden | -0.023 |
rall | -0.022 |
reluct | -0.021 |
arrang | -0.021 |
fort | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
ocket | +0.245 |
adiator | +0.235 |
Explore | +0.232 |
rium | +0.228 |
replay | +0.228 |
Neg logit contributions | |
unden | -0.354 |
rall | -0.347 |
arrang | -0.326 |
reluct | -0.323 |
'. | -0.319 |
Token | the |
Token ID | 262 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
unden | +0.079 |
rall | +0.078 |
arrang | +0.072 |
reluct | +0.072 |
fort | +0.070 |
Neg logit contributions | |
ocket | -0.069 |
adiator | -0.067 |
Explore | -0.066 |
replay | -0.065 |
rium | -0.065 |
Token | Green |
Token ID | 3469 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
ocket | +0.077 |
adiator | +0.075 |
Explore | +0.074 |
replay | +0.074 |
rium | +0.074 |
Neg logit contributions | |
unden | -0.085 |
rall | -0.083 |
arrang | -0.077 |
reluct | -0.077 |
fort | -0.075 |
Token | line |
Token ID | 1627 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
ocket | +0.164 |
adiator | +0.159 |
Explore | +0.157 |
replay | +0.156 |
rium | +0.155 |
Neg logit contributions | |
unden | -0.140 |
rall | -0.135 |
arrang | -0.125 |
reluct | -0.124 |
'. | -0.121 |
Token | do |
Token ID | 466 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
ocket | +0.047 |
replay | +0.046 |
Explore | +0.045 |
adiator | +0.045 |
paces | +0.044 |
Neg logit contributions | |
unden | -0.059 |
rall | -0.057 |
reluct | -0.054 |
arrang | -0.053 |
fort | -0.052 |
Token | nothing |
Token ID | 2147 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
unden | +0.059 |
rall | +0.058 |
arrang | +0.055 |
reluct | +0.054 |
fort | +0.053 |
Neg logit contributions | |
ocket | -0.037 |
adiator | -0.035 |
Explore | -0.035 |
replay | -0.034 |
rium | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
unden | +0.911 |
rall | +0.890 |
arrang | +0.838 |
reluct | +0.826 |
fort | +0.819 |
Neg logit contributions | |
ocket | -0.698 |
adiator | -0.672 |
Explore | -0.664 |
replay | -0.656 |
rium | -0.654 |
Token | continue |
Token ID | 2555 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
unden | +0.350 |
rall | +0.341 |
arrang | +0.320 |
reluct | +0.317 |
fort | +0.313 |
Neg logit contributions | |
ocket | -0.272 |
adiator | -0.259 |
replay | -0.258 |
Explore | -0.258 |
rium | -0.253 |
Token | making |
Token ID | 1642 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
unden | +1.471 |
rall | +1.437 |
arrang | +1.354 |
reluct | +1.340 |
fort | +1.325 |
Neg logit contributions | |
ocket | -0.996 |
adiator | -0.953 |
Explore | -0.942 |
replay | -0.930 |
rium | -0.924 |
Token | temporary |
Token ID | 8584 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
unden | +0.355 |
rall | +0.344 |
arrang | +0.319 |
reluct | +0.318 |
fort | +0.314 |
Neg logit contributions | |
ocket | -0.310 |
adiator | -0.297 |
Explore | -0.294 |
replay | -0.294 |
rium | -0.289 |
Token | repairs |
Token ID | 20097 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
unden | +1.716 |
rall | +1.664 |
arrang | +1.536 |
reluct | +1.522 |
fort | +1.499 |
Neg logit contributions | |
ocket | -1.930 |
adiator | -1.864 |
Explore | -1.847 |
replay | -1.833 |
rium | -1.818 |
Token | . |
Token ID | 13 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
unden | +1.826 |
rall | +1.785 |
arrang | +1.677 |
reluct | +1.657 |
fort | +1.638 |
Neg logit contributions | |
ocket | -1.384 |
adiator | -1.331 |
Explore | -1.316 |
replay | -1.299 |
rium | -1.294 |
Token |
|
Token ID | 198 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
unden | +0.137 |
rall | +0.134 |
arrang | +0.126 |
reluct | +0.125 |
fort | +0.122 |
Neg logit contributions | |
ocket | -0.090 |
Explore | -0.085 |
adiator | -0.085 |
replay | -0.084 |
rium | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
unden | +0.300 |
rall | +0.294 |
arrang | +0.277 |
reluct | +0.274 |
fort | +0.273 |
Neg logit contributions | |
ocket | -0.195 |
adiator | -0.187 |
Explore | -0.184 |
replay | -0.182 |
rium | -0.181 |
Token | City |
Token ID | 14941 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
unden | +0.388 |
rall | +0.380 |
arrang | +0.359 |
reluct | +0.356 |
'. | +0.351 |
Neg logit contributions | |
ocket | -0.220 |
adiator | -0.210 |
Explore | -0.208 |
replay | -0.204 |
rium | -0.203 |
Token | the |
Token ID | 262 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
unden | +0.339 |
rall | +0.331 |
arrang | +0.309 |
reluct | +0.307 |
fort | +0.303 |
Neg logit contributions | |
ocket | -0.281 |
adiator | -0.269 |
replay | -0.267 |
Explore | -0.267 |
rium | -0.264 |
Token | city |
Token ID | 1748 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
unden | +0.335 |
rall | +0.326 |
arrang | +0.305 |
reluct | +0.303 |
fort | +0.300 |
Neg logit contributions | |
ocket | -0.277 |
adiator | -0.266 |
replay | -0.263 |
Explore | -0.262 |
rium | -0.260 |
Token | is |
Token ID | 318 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
unden | +0.195 |
rall | +0.190 |
arrang | +0.179 |
reluct | +0.178 |
fort | +0.175 |
Neg logit contributions | |
ocket | -0.135 |
adiator | -0.129 |
Explore | -0.127 |
replay | -0.127 |
rium | -0.126 |
Token | supposed |
Token ID | 4385 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
unden | +0.252 |
rall | +0.245 |
arrang | +0.229 |
reluct | +0.226 |
fort | +0.222 |
Neg logit contributions | |
ocket | -0.218 |
adiator | -0.209 |
replay | -0.206 |
Explore | -0.206 |
rium | -0.204 |
Token | to |
Token ID | 284 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
unden | +1.179 |
rall | +1.160 |
fort | +1.100 |
'. | +1.097 |
arrang | +1.093 |
Neg logit contributions | |
ocket | -0.782 |
adiator | -0.749 |
Explore | -0.735 |
rium | -0.725 |
replay | -0.721 |
Token | fix |
Token ID | 4259 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
unden | +0.165 |
rall | +0.160 |
reluct | +0.151 |
arrang | +0.149 |
fort | +0.146 |
Neg logit contributions | |
ocket | -0.125 |
replay | -0.122 |
Explore | -0.118 |
adiator | -0.118 |
crawl | -0.117 |
Token | the |
Token ID | 262 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
unden | +3.344 |
rall | +3.278 |
arrang | +3.089 |
reluct | +3.033 |
'. | +3.020 |
Neg logit contributions | |
ocket | -2.529 |
adiator | -2.444 |
Explore | -2.415 |
replay | -2.371 |
rium | -2.366 |
Token | sidewalks |
Token ID | 38282 |
Feature activation | -6.45e-03 |
Pos logit contributions | |
unden | +0.282 |
rall | +0.273 |
arrang | +0.255 |
reluct | +0.254 |
fort | +0.252 |
Neg logit contributions | |
ocket | -0.233 |
replay | -0.221 |
adiator | -0.221 |
Explore | -0.219 |
rium | -0.218 |
Token | , |
Token ID | 11 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
unden | +0.013 |
rall | +0.013 |
arrang | +0.012 |
reluct | +0.012 |
fort | +0.012 |
Neg logit contributions | |
ocket | -0.009 |
adiator | -0.008 |
Explore | -0.008 |
replay | -0.008 |
paces | -0.008 |
Token | at |
Token ID | 379 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
unden | +0.199 |
rall | +0.193 |
arrang | +0.182 |
reluct | +0.182 |
fort | +0.179 |
Neg logit contributions | |
ocket | -0.146 |
adiator | -0.139 |
replay | -0.138 |
Explore | -0.138 |
paces | -0.136 |
Token | their |
Token ID | 511 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
unden | +0.303 |
rall | +0.295 |
arrang | +0.277 |
reluct | +0.274 |
fort | +0.270 |
Neg logit contributions | |
ocket | -0.258 |
adiator | -0.248 |
rium | -0.246 |
Explore | -0.245 |
replay | -0.245 |
Token | and |
Token ID | 290 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
ocket | +0.208 |
adiator | +0.199 |
Explore | +0.197 |
replay | +0.196 |
rium | +0.194 |
Neg logit contributions | |
unden | -0.278 |
rall | -0.272 |
arrang | -0.256 |
reluct | -0.253 |
'. | -0.250 |
Token | whiff |
Token ID | 48112 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
unden | +0.089 |
rall | +0.087 |
arrang | +0.082 |
reluct | +0.081 |
fort | +0.080 |
Neg logit contributions | |
ocket | -0.078 |
adiator | -0.075 |
Explore | -0.074 |
replay | -0.074 |
rium | -0.073 |
Token | ling |
Token ID | 1359 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
ocket | +0.096 |
adiator | +0.092 |
Explore | +0.090 |
replay | +0.089 |
rium | +0.089 |
Neg logit contributions | |
unden | -0.150 |
rall | -0.147 |
arrang | -0.138 |
reluct | -0.137 |
'. | -0.135 |
Token | and |
Token ID | 290 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ocket | +0.221 |
adiator | +0.211 |
Explore | +0.209 |
replay | +0.207 |
rium | +0.206 |
Neg logit contributions | |
unden | -0.357 |
rall | -0.349 |
arrang | -0.330 |
reluct | -0.326 |
'. | -0.323 |
Token | w |
Token ID | 266 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
ocket | +0.025 |
adiator | +0.024 |
Explore | +0.024 |
replay | +0.023 |
rium | +0.023 |
Neg logit contributions | |
unden | -0.028 |
rall | -0.028 |
arrang | -0.026 |
reluct | -0.025 |
fort | -0.025 |
Token | aff |
Token ID | 2001 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ocket | +0.029 |
adiator | +0.028 |
Explore | +0.027 |
replay | +0.027 |
rium | +0.027 |
Neg logit contributions | |
unden | -0.075 |
rall | -0.073 |
arrang | -0.070 |
reluct | -0.069 |
fort | -0.068 |
Token | ling |
Token ID | 1359 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
unden | +2.904 |
rall | +2.844 |
arrang | +2.671 |
reluct | +2.641 |
fort | +2.623 |
Neg logit contributions | |
ocket | -2.124 |
adiator | -2.039 |
Explore | -2.016 |
replay | -1.989 |
rium | -1.975 |
Token | of |
Token ID | 286 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
unden | +0.250 |
rall | +0.244 |
arrang | +0.230 |
reluct | +0.228 |
'. | +0.225 |
Neg logit contributions | |
ocket | -0.167 |
adiator | -0.160 |
Explore | -0.158 |
replay | -0.156 |
rium | -0.155 |
Token | the |
Token ID | 262 |
Feature activation | -6.35e-02 |
Pos logit contributions | |
unden | +0.086 |
rall | +0.084 |
arrang | +0.078 |
reluct | +0.077 |
fort | +0.077 |
Neg logit contributions | |
ocket | -0.073 |
adiator | -0.070 |
Explore | -0.069 |
replay | -0.069 |
rium | -0.069 |
Token | trolls |
Token ID | 24727 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
unden | +0.116 |
rall | +0.114 |
arrang | +0.107 |
reluct | +0.105 |
fort | +0.104 |
Neg logit contributions | |
ocket | -0.098 |
adiator | -0.094 |
Explore | -0.094 |
replay | -0.093 |
rium | -0.092 |
Token | ) |
Token ID | 8 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
ocket | +0.013 |
adiator | +0.013 |
Explore | +0.012 |
replay | +0.012 |
rium | +0.012 |
Neg logit contributions | |
unden | -0.021 |
rall | -0.021 |
arrang | -0.020 |
reluct | -0.019 |
'. | -0.019 |
Token | Ferrari |
Token ID | 23502 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
unden | +0.363 |
rall | +0.353 |
arrang | +0.331 |
reluct | +0.327 |
fort | +0.322 |
Neg logit contributions | |
ocket | -0.327 |
adiator | -0.315 |
Explore | -0.313 |
replay | -0.311 |
rium | -0.308 |
Token | , |
Token ID | 11 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
unden | +0.141 |
rall | +0.137 |
arrang | +0.130 |
reluct | +0.128 |
fort | +0.126 |
Neg logit contributions | |
ocket | -0.092 |
adiator | -0.088 |
Explore | -0.087 |
replay | -0.086 |
rium | -0.085 |
Token | believes |
Token ID | 5804 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
unden | +0.032 |
rall | +0.031 |
arrang | +0.030 |
reluct | +0.029 |
fort | +0.029 |
Neg logit contributions | |
ocket | -0.025 |
adiator | -0.024 |
Explore | -0.024 |
replay | -0.024 |
paces | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ocket | +0.434 |
replay | +0.417 |
Explore | +0.415 |
adiator | +0.415 |
rium | +0.404 |
Neg logit contributions | |
unden | -0.582 |
rall | -0.564 |
arrang | -0.532 |
reluct | -0.527 |
fort | -0.521 |
Token | true |
Token ID | 2081 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
ocket | +0.169 |
replay | +0.162 |
adiator | +0.162 |
Explore | +0.160 |
rium | +0.158 |
Neg logit contributions | |
unden | -0.222 |
rall | -0.215 |
arrang | -0.202 |
reluct | -0.201 |
fort | -0.198 |
Token | pe |
Token ID | 613 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
ocket | +0.420 |
adiator | +0.402 |
replay | +0.402 |
Explore | +0.400 |
rium | +0.392 |
Neg logit contributions | |
unden | -0.527 |
rall | -0.514 |
arrang | -0.484 |
reluct | -0.479 |
fort | -0.474 |
Token | cking |
Token ID | 44377 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
unden | +2.297 |
rall | +2.257 |
arrang | +2.141 |
fort | +2.118 |
'. | +2.111 |
Neg logit contributions | |
ocket | -1.289 |
Explore | -1.224 |
adiator | -1.215 |
replay | -1.198 |
rium | -1.192 |
Token | order |
Token ID | 1502 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
unden | +0.336 |
rall | +0.326 |
arrang | +0.301 |
reluct | +0.298 |
fort | +0.293 |
Neg logit contributions | |
ocket | -0.369 |
adiator | -0.356 |
Explore | -0.353 |
replay | -0.351 |
rium | -0.348 |
Token | is |
Token ID | 318 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
unden | +0.113 |
rall | +0.111 |
arrang | +0.104 |
reluct | +0.103 |
fort | +0.101 |
Neg logit contributions | |
ocket | -0.083 |
adiator | -0.080 |
Explore | -0.079 |
replay | -0.078 |
rium | -0.078 |
Token | yet |
Token ID | 1865 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
ocket | +0.274 |
adiator | +0.263 |
Explore | +0.262 |
replay | +0.260 |
rium | +0.256 |
Neg logit contributions | |
unden | -0.372 |
rall | -0.366 |
arrang | -0.344 |
reluct | -0.342 |
fort | -0.336 |
Token | to |
Token ID | 284 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
unden | +0.684 |
rall | +0.671 |
arrang | +0.634 |
reluct | +0.627 |
fort | +0.620 |
Neg logit contributions | |
ocket | -0.404 |
adiator | -0.385 |
Explore | -0.380 |
replay | -0.375 |
rium | -0.373 |
Token | |
Token ID | 220 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
unden | +0.135 |
rall | +0.132 |
arrang | +0.124 |
reluct | +0.123 |
fort | +0.121 |
Neg logit contributions | |
ocket | -0.110 |
adiator | -0.106 |
Explore | -0.105 |
replay | -0.104 |
rium | -0.103 |
Token | ers |
Token ID | 364 |
Feature activation | -9.46e-02 |
Pos logit contributions | |
unden | +0.703 |
rall | +0.687 |
arrang | +0.647 |
reluct | +0.641 |
'. | +0.632 |
Neg logit contributions | |
ocket | -0.492 |
adiator | -0.470 |
Explore | -0.465 |
replay | -0.457 |
rium | -0.456 |
Token | atz |
Token ID | 27906 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
unden | +0.161 |
rall | +0.157 |
arrang | +0.146 |
reluct | +0.144 |
fort | +0.142 |
Neg logit contributions | |
ocket | -0.159 |
adiator | -0.154 |
Explore | -0.152 |
replay | -0.151 |
rium | -0.150 |
Token | " |
Token ID | 366 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
ocket | +0.255 |
adiator | +0.247 |
replay | +0.245 |
Explore | +0.244 |
rium | +0.240 |
Neg logit contributions | |
unden | -0.324 |
rall | -0.315 |
arrang | -0.296 |
reluct | -0.295 |
fort | -0.289 |
Token | social |
Token ID | 14557 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ocket | +0.108 |
adiator | +0.104 |
Explore | +0.104 |
rium | +0.101 |
replay | +0.101 |
Neg logit contributions | |
unden | -0.145 |
rall | -0.142 |
arrang | -0.133 |
reluct | -0.133 |
'. | -0.129 |
Token | ism |
Token ID | 1042 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
unden | +2.604 |
rall | +2.547 |
arrang | +2.400 |
reluct | +2.368 |
'. | +2.344 |
Neg logit contributions | |
ocket | -1.806 |
adiator | -1.732 |
Explore | -1.710 |
replay | -1.685 |
rium | -1.682 |
Token | " |
Token ID | 1 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
unden | +2.846 |
rall | +2.786 |
arrang | +2.628 |
reluct | +2.588 |
'. | +2.581 |
Neg logit contributions | |
ocket | -1.796 |
adiator | -1.707 |
Explore | -1.682 |
replay | -1.660 |
rium | -1.656 |
Token | devoid |
Token ID | 34065 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
ocket | +0.200 |
adiator | +0.192 |
Explore | +0.191 |
replay | +0.189 |
rium | +0.187 |
Neg logit contributions | |
unden | -0.291 |
rall | -0.285 |
arrang | -0.268 |
reluct | -0.265 |
fort | -0.262 |
Token | of |
Token ID | 286 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ocket | +0.074 |
adiator | +0.070 |
Explore | +0.069 |
replay | +0.068 |
rium | +0.068 |
Neg logit contributions | |
unden | -0.140 |
rall | -0.137 |
arrang | -0.130 |
reluct | -0.129 |
fort | -0.128 |
Token | social |
Token ID | 1919 |
Feature activation | -7.39e-01 |
Pos logit contributions | |
ocket | +0.173 |
replay | +0.167 |
adiator | +0.167 |
Explore | +0.166 |
rium | +0.163 |
Neg logit contributions | |
unden | -0.243 |
rall | -0.238 |
arrang | -0.223 |
reluct | -0.223 |
fort | -0.220 |
Token | content |
Token ID | 2695 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
unden | +1.453 |
rall | +1.421 |
arrang | +1.336 |
reluct | +1.322 |
fort | +1.306 |
Neg logit contributions | |
ocket | -1.046 |
adiator | -1.003 |
Explore | -0.991 |
replay | -0.980 |
rium | -0.975 |