Token | All |
Token ID | 1439 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
Stub | +1.015 |
stra | +1.005 |
Amen | +0.989 |
stop | +0.973 |
Tel | +0.972 |
Neg logit contributions | |
ithing | -1.284 |
andom | -1.139 |
nesota | -1.127 |
hing | -1.119 |
arning | -1.116 |
Token | Rights |
Token ID | 6923 |
Feature activation | +6.00e+00 |
Pos logit contributions | |
Stub | +2.816 |
stra | +2.748 |
Amen | +2.693 |
Tel | +2.654 |
stop | +2.615 |
Neg logit contributions | |
ithing | -4.018 |
andom | -3.584 |
nesota | -3.584 |
hing | -3.530 |
arning | -3.526 |
Token | Reserved |
Token ID | 33876 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
Stub | +2.994 |
Amen | +2.706 |
reopen | +2.567 |
stra | +2.564 |
iest | +2.546 |
Neg logit contributions | |
ithing | -5.416 |
nesota | -4.835 |
andom | -4.768 |
evin | -4.714 |
ammy | -4.710 |
Token | . |
Token ID | 13 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
Stub | +1.383 |
stra | +1.366 |
Amen | +1.355 |
Tel | +1.324 |
stop | +1.316 |
Neg logit contributions | |
ithing | -1.995 |
andom | -1.778 |
nesota | -1.766 |
arning | -1.753 |
hing | -1.752 |
Token | Rep |
Token ID | 1432 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
Stub | +1.191 |
stra | +1.173 |
Amen | +1.159 |
Tel | +1.135 |
stop | +1.135 |
Neg logit contributions | |
ithing | -1.633 |
andom | -1.454 |
nesota | -1.439 |
hing | -1.431 |
arning | -1.427 |
Token | rint |
Token ID | 22272 |
Feature activation | +7.99e+00 |
Pos logit contributions | |
Stub | +3.541 |
stra | +3.508 |
Tel | +3.481 |
stop | +3.466 |
Amen | +3.448 |
Neg logit contributions | |
ithing | -2.069 |
nesota | -1.779 |
andom | -1.772 |
arning | -1.723 |
ammy | -1.696 |
Token | ed |
Token ID | 276 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
Stub | +4.474 |
Amen | +4.412 |
Tel | +4.354 |
pub | +4.330 |
stop | +4.312 |
Neg logit contributions | |
ithing | -5.933 |
ammy | -5.365 |
andom | -5.364 |
hing | -5.290 |
nesota | -5.289 |
Token | with |
Token ID | 351 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
Stub | +3.026 |
Amen | +2.960 |
stra | +2.932 |
Tel | +2.909 |
stop | +2.878 |
Neg logit contributions | |
ithing | -2.997 |
andom | -2.619 |
nesota | -2.594 |
hing | -2.589 |
arning | -2.571 |
Token | permission |
Token ID | 7170 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
Stub | +1.430 |
Amen | +1.364 |
stra | +1.356 |
stop | +1.335 |
reopen | +1.307 |
Neg logit contributions | |
ithing | -2.854 |
nesota | -2.607 |
andom | -2.595 |
arning | -2.557 |
hing | -2.542 |
Token | . |
Token ID | 13 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
Stub | +2.526 |
Amen | +2.494 |
stra | +2.457 |
Tel | +2.420 |
stop | +2.389 |
Neg logit contributions | |
ithing | -3.857 |
andom | -3.417 |
nesota | -3.393 |
hing | -3.393 |
arning | -3.388 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
Stub | +1.088 |
stra | +1.070 |
Amen | +1.058 |
Tel | +1.037 |
stop | +1.035 |
Neg logit contributions | |
ithing | -1.731 |
andom | -1.552 |
nesota | -1.538 |
hing | -1.531 |
arning | -1.526 |
Token | South |
Token ID | 2520 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
Stub | +1.044 |
stra | +1.042 |
Amen | +1.022 |
stop | +1.008 |
Tel | +1.006 |
Neg logit contributions | |
ithing | -1.102 |
andom | -0.970 |
nesota | -0.961 |
hing | -0.953 |
arning | -0.946 |
Token | Korea |
Token ID | 4969 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
Stub | +1.300 |
stra | +1.292 |
Amen | +1.263 |
stop | +1.246 |
Tel | +1.242 |
Neg logit contributions | |
ithing | -1.938 |
andom | -1.736 |
nesota | -1.716 |
hing | -1.713 |
arning | -1.700 |
Token | invested |
Token ID | 13356 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
stra | +0.274 |
Stub | +0.274 |
Amen | +0.264 |
reopen | +0.262 |
stop | +0.261 |
Neg logit contributions | |
ithing | -0.287 |
andom | -0.251 |
nesota | -0.250 |
hing | -0.245 |
arning | -0.245 |
Token | in |
Token ID | 287 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
Stub | +0.862 |
stra | +0.862 |
Amen | +0.841 |
stop | +0.830 |
reopen | +0.829 |
Neg logit contributions | |
ithing | -0.926 |
andom | -0.819 |
nesota | -0.812 |
hing | -0.795 |
arning | -0.792 |
Token | R |
Token ID | 371 |
Feature activation | -7.20e-01 |
Pos logit contributions | |
Stub | +1.294 |
stra | +1.277 |
Amen | +1.263 |
reopen | +1.246 |
stop | +1.242 |
Neg logit contributions | |
ithing | -1.376 |
andom | -1.215 |
nesota | -1.203 |
hing | -1.185 |
arning | -1.180 |
Token | & |
Token ID | 5 |
Feature activation | +7.84e+00 |
Pos logit contributions | |
ithing | +0.623 |
andom | +0.562 |
nesota | +0.558 |
hing | +0.554 |
arning | +0.552 |
Neg logit contributions | |
Stub | -0.345 |
stra | -0.345 |
Amen | -0.334 |
stop | -0.331 |
Tel | -0.330 |
Token | D |
Token ID | 35 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
Tel | +3.340 |
stop | +3.254 |
Num | +3.233 |
pub | +3.123 |
Rap | +3.076 |
Neg logit contributions | |
ithing | -7.358 |
nesota | -6.614 |
ammy | -6.514 |
arning | -6.493 |
andom | -6.463 |
Token | in |
Token ID | 287 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Stub | +1.435 |
stra | +1.431 |
Amen | +1.408 |
stop | +1.389 |
reopen | +1.388 |
Neg logit contributions | |
ithing | -1.546 |
andom | -1.388 |
nesota | -1.369 |
ammy | -1.337 |
arning | -1.336 |
Token | 2014 |
Token ID | 1946 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
Stub | +0.592 |
stra | +0.587 |
Amen | +0.578 |
stop | +0.567 |
Tel | +0.564 |
Neg logit contributions | |
ithing | -0.894 |
andom | -0.804 |
nesota | -0.798 |
hing | -0.792 |
arning | -0.786 |
Token | out |
Token ID | 503 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
stra | +1.516 |
Stub | +1.516 |
Amen | +1.488 |
stop | +1.471 |
reopen | +1.463 |
Neg logit contributions | |
ithing | -1.450 |
andom | -1.282 |
nesota | -1.269 |
hing | -1.229 |
arning | -1.226 |
Token | stri |
Token ID | 33565 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
ithing | +0.736 |
andom | +0.643 |
nesota | +0.635 |
hing | +0.631 |
arning | +0.628 |
Neg logit contributions | |
Stub | -0.745 |
stra | -0.742 |
Amen | -0.727 |
stop | -0.720 |
Tel | -0.719 |
Token | , |
Token ID | 11 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
stra | +1.725 |
pub | +1.682 |
Tel | +1.660 |
Stub | +1.659 |
Amen | +1.656 |
Neg logit contributions | |
ithing | -3.246 |
andom | -2.965 |
nesota | -2.945 |
arning | -2.911 |
ammy | -2.869 |
Token | check |
Token ID | 2198 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
Stub | +1.091 |
stra | +1.085 |
Amen | +1.056 |
stop | +1.040 |
Tel | +1.039 |
Neg logit contributions | |
ithing | -1.941 |
andom | -1.754 |
nesota | -1.735 |
hing | -1.724 |
arning | -1.721 |
Token | out |
Token ID | 503 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
Stub | +2.956 |
stra | +2.897 |
Amen | +2.808 |
stop | +2.759 |
Num | +2.759 |
Neg logit contributions | |
ithing | -3.617 |
andom | -3.297 |
arning | -3.263 |
nesota | -3.204 |
ammy | -3.093 |
Token | our |
Token ID | 674 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
Stub | +2.519 |
stra | +2.484 |
stop | +2.411 |
Amen | +2.400 |
reopen | +2.377 |
Neg logit contributions | |
ithing | -2.577 |
andom | -2.327 |
arning | -2.266 |
nesota | -2.265 |
hing | -2.225 |
Token | latest |
Token ID | 3452 |
Feature activation | +5.53e+00 |
Pos logit contributions | |
Stub | +1.318 |
stra | +1.297 |
stop | +1.256 |
Amen | +1.254 |
reopen | +1.235 |
Neg logit contributions | |
ithing | -2.358 |
andom | -2.165 |
nesota | -2.160 |
arning | -2.120 |
hing | -2.113 |
Token | issue |
Token ID | 2071 |
Feature activation | +7.65e+00 |
Pos logit contributions | |
stra | +2.805 |
Stub | +2.774 |
stop | +2.691 |
reopen | +2.632 |
Num | +2.581 |
Neg logit contributions | |
ithing | -4.554 |
nesota | -4.270 |
ammy | -4.251 |
andom | -4.216 |
arning | -4.121 |
Token |
|
Token ID | 198 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
Stub | +4.007 |
stra | +3.874 |
stop | +3.742 |
Tel | +3.706 |
Amen | +3.691 |
Neg logit contributions | |
ithing | -6.134 |
andom | -5.783 |
nesota | -5.775 |
arning | -5.480 |
ammy | -5.476 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
stra | +0.732 |
Stub | +0.722 |
stop | +0.714 |
Tel | +0.706 |
Amen | +0.697 |
Neg logit contributions | |
ithing | -0.984 |
andom | -0.881 |
nesota | -0.869 |
arning | -0.864 |
ammy | -0.851 |
Token | Subscribe |
Token ID | 27125 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
stra | +0.043 |
Stub | +0.043 |
Tel | +0.041 |
stop | +0.041 |
Amen | +0.041 |
Neg logit contributions | |
ithing | -0.109 |
andom | -0.100 |
nesota | -0.099 |
arning | -0.098 |
hing | -0.098 |
Token | now |
Token ID | 783 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
Stub | +1.155 |
stra | +1.125 |
Amen | +1.101 |
reopen | +1.087 |
stop | +1.084 |
Neg logit contributions | |
ithing | -1.346 |
andom | -1.215 |
nesota | -1.184 |
arning | -1.161 |
ammy | -1.146 |
Token | for |
Token ID | 329 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
ithing | +0.190 |
andom | +0.167 |
nesota | +0.165 |
ammy | +0.162 |
arning | +0.162 |
Neg logit contributions | |
Stub | -0.210 |
stra | -0.207 |
Amen | -0.204 |
reopen | -0.202 |
stop | -0.202 |
Token | . |
Token ID | 13 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
Stub | +0.391 |
stra | +0.389 |
Amen | +0.381 |
Tel | +0.377 |
stop | +0.377 |
Neg logit contributions | |
ithing | -0.492 |
andom | -0.434 |
nesota | -0.431 |
hing | -0.430 |
arning | -0.426 |
Token | � |
Token ID | 564 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ithing | +1.008 |
andom | +0.871 |
hing | +0.858 |
nesota | +0.854 |
arning | +0.845 |
Neg logit contributions | |
stra | -1.197 |
Stub | -1.193 |
reopen | -1.166 |
stop | -1.165 |
Amen | -1.165 |
Token | � |
Token ID | 250 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
ithing | +1.031 |
andom | +0.933 |
nesota | +0.924 |
hing | +0.919 |
arning | +0.916 |
Neg logit contributions | |
Stub | -0.489 |
stra | -0.482 |
Amen | -0.469 |
stop | -0.461 |
Tel | -0.460 |
Token | We |
Token ID | 1135 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
ithing | +0.275 |
andom | +0.243 |
nesota | +0.240 |
hing | +0.239 |
arning | +0.238 |
Neg logit contributions | |
Stub | -0.233 |
stra | -0.232 |
Amen | -0.226 |
stop | -0.224 |
Tel | -0.224 |
Token | would |
Token ID | 561 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
ithing | +1.082 |
andom | +0.948 |
nesota | +0.935 |
hing | +0.932 |
arning | +0.926 |
Neg logit contributions | |
Stub | -1.052 |
stra | -1.047 |
Amen | -1.028 |
Tel | -1.016 |
stop | -1.015 |
Token | rather |
Token ID | 2138 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
ithing | +1.743 |
andom | +1.516 |
hing | +1.498 |
nesota | +1.491 |
arning | +1.479 |
Neg logit contributions | |
Stub | -1.805 |
stra | -1.795 |
Amen | -1.766 |
Tel | -1.760 |
stop | -1.739 |
Token | answer |
Token ID | 3280 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
Stub | +4.658 |
reopen | +4.612 |
stra | +4.541 |
stop | +4.469 |
Amen | +4.467 |
Neg logit contributions | |
ithing | -5.162 |
nesota | -4.654 |
andom | -4.618 |
arning | -4.565 |
hing | -4.523 |
Token | to |
Token ID | 284 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Stub | +1.023 |
stra | +1.013 |
Amen | +0.991 |
stop | +0.979 |
reopen | +0.979 |
Neg logit contributions | |
ithing | -1.121 |
nesota | -0.985 |
andom | -0.983 |
hing | -0.971 |
arning | -0.969 |
Token | Ken |
Token ID | 7148 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Stub | +0.674 |
stra | +0.667 |
Amen | +0.655 |
stop | +0.644 |
reopen | +0.643 |
Neg logit contributions | |
ithing | -0.730 |
nesota | -0.644 |
andom | -0.639 |
arning | -0.635 |
hing | -0.633 |
Token | yon |
Token ID | 19181 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
ithing | +1.084 |
andom | +0.953 |
hing | +0.948 |
nesota | +0.941 |
arning | +0.936 |
Neg logit contributions | |
Stub | -0.878 |
stra | -0.876 |
Amen | -0.858 |
reopen | -0.852 |
stop | -0.850 |
Token | College |
Token ID | 5535 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
ithing | +1.292 |
andom | +1.143 |
hing | +1.127 |
nesota | +1.124 |
arning | +1.111 |
Neg logit contributions | |
Stub | -1.078 |
stra | -1.072 |
Amen | -1.053 |
stop | -1.049 |
reopen | -1.045 |
Token | , |
Token ID | 11 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
Stub | +1.627 |
stra | +1.626 |
stop | +1.591 |
Amen | +1.557 |
pub | +1.550 |
Neg logit contributions | |
ithing | -3.098 |
andom | -2.808 |
arning | -2.792 |
nesota | -2.772 |
ammy | -2.736 |
Token | check |
Token ID | 2198 |
Feature activation | +4.25e+00 |
Pos logit contributions | |
Stub | +0.882 |
stra | +0.872 |
stop | +0.856 |
Tel | +0.835 |
reopen | +0.832 |
Neg logit contributions | |
ithing | -2.000 |
andom | -1.839 |
nesota | -1.811 |
arning | -1.810 |
hing | -1.783 |
Token | out |
Token ID | 503 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
Stub | +2.405 |
stra | +2.366 |
Amen | +2.358 |
stop | +2.323 |
iest | +2.294 |
Neg logit contributions | |
ithing | -3.340 |
andom | -2.963 |
arning | -2.919 |
nesota | -2.885 |
ammy | -2.795 |
Token | our |
Token ID | 674 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
Stub | +2.450 |
stra | +2.411 |
stop | +2.361 |
Tel | +2.346 |
reopen | +2.309 |
Neg logit contributions | |
ithing | -2.706 |
andom | -2.387 |
nesota | -2.352 |
arning | -2.350 |
hing | -2.321 |
Token | latest |
Token ID | 3452 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
Stub | +1.215 |
stra | +1.196 |
stop | +1.174 |
Amen | +1.154 |
Tel | +1.153 |
Neg logit contributions | |
ithing | -2.126 |
andom | -1.930 |
nesota | -1.929 |
arning | -1.893 |
hing | -1.877 |
Token | issue |
Token ID | 2071 |
Feature activation | +7.42e+00 |
Pos logit contributions | |
Stub | +2.620 |
stop | +2.606 |
stra | +2.573 |
pub | +2.507 |
reopen | +2.462 |
Neg logit contributions | |
ithing | -4.490 |
nesota | -4.090 |
andom | -4.013 |
arning | -4.010 |
ammy | -3.985 |
Token |
|
Token ID | 198 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
stra | +3.656 |
Stub | +3.635 |
stop | +3.615 |
Tel | +3.540 |
Num | +3.466 |
Neg logit contributions | |
ithing | -6.225 |
nesota | -5.646 |
andom | -5.558 |
arning | -5.544 |
ammy | -5.525 |
Token |
|
Token ID | 198 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
stra | +1.600 |
Stub | +1.586 |
stop | +1.568 |
Tel | +1.550 |
pub | +1.521 |
Neg logit contributions | |
ithing | -2.374 |
andom | -2.123 |
arning | -2.099 |
nesota | -2.091 |
ammy | -2.064 |
Token | Travel |
Token ID | 33074 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
Stub | +0.754 |
stra | +0.749 |
Tel | +0.721 |
stop | +0.717 |
Amen | +0.701 |
Neg logit contributions | |
ithing | -2.115 |
andom | -1.947 |
nesota | -1.932 |
hing | -1.908 |
arning | -1.902 |
Token | With |
Token ID | 2080 |
Feature activation | +9.93e-01 |
Pos logit contributions | |
Stub | +0.395 |
stra | +0.387 |
Amen | +0.387 |
stop | +0.380 |
Tel | +0.380 |
Neg logit contributions | |
ithing | -0.271 |
andom | -0.231 |
nesota | -0.227 |
arning | -0.221 |
hing | -0.221 |
Token | The |
Token ID | 383 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
Stub | +0.669 |
stra | +0.665 |
Amen | +0.652 |
stop | +0.644 |
Tel | +0.642 |
Neg logit contributions | |
ithing | -0.671 |
andom | -0.587 |
nesota | -0.581 |
hing | -0.576 |
arning | -0.575 |
Token | maintained |
Token ID | 9456 |
Feature activation | +9.44e-01 |
Pos logit contributions | |
ithing | +0.034 |
andom | +0.030 |
nesota | +0.030 |
hing | +0.029 |
arning | +0.029 |
Neg logit contributions | |
Stub | -0.029 |
stra | -0.029 |
Amen | -0.029 |
stop | -0.028 |
Tel | -0.028 |
Token | by |
Token ID | 416 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
Stub | +0.624 |
stra | +0.620 |
Amen | +0.608 |
stop | +0.601 |
Tel | +0.601 |
Neg logit contributions | |
ithing | -0.650 |
andom | -0.569 |
nesota | -0.564 |
hing | -0.559 |
arning | -0.558 |
Token | Inter |
Token ID | 4225 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Stub | +0.854 |
stra | +0.848 |
Amen | +0.831 |
Tel | +0.821 |
stop | +0.819 |
Neg logit contributions | |
ithing | -1.035 |
andom | -0.918 |
nesota | -0.908 |
hing | -0.903 |
arning | -0.899 |
Token | pol |
Token ID | 16104 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
stra | +0.585 |
Stub | +0.569 |
stop | +0.561 |
Tel | +0.554 |
Amen | +0.547 |
Neg logit contributions | |
ithing | -1.379 |
andom | -1.250 |
nesota | -1.232 |
hing | -1.228 |
arning | -1.228 |
Token | itself |
Token ID | 2346 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Stub | +1.287 |
stra | +1.278 |
Amen | +1.253 |
stop | +1.237 |
Tel | +1.235 |
Neg logit contributions | |
ithing | -1.624 |
andom | -1.438 |
nesota | -1.425 |
hing | -1.412 |
arning | -1.409 |
Token | rather |
Token ID | 2138 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
Stub | +0.908 |
stra | +0.904 |
Amen | +0.883 |
stop | +0.872 |
Tel | +0.871 |
Neg logit contributions | |
ithing | -1.152 |
andom | -1.024 |
nesota | -1.013 |
hing | -1.004 |
arning | -1.002 |
Token | than |
Token ID | 621 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
Stub | +2.504 |
stra | +2.485 |
Amen | +2.357 |
reopen | +2.330 |
Tel | +2.273 |
Neg logit contributions | |
ithing | -7.122 |
andom | -6.516 |
arning | -6.495 |
ammy | -6.460 |
nesota | -6.458 |
Token | the |
Token ID | 262 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Stub | +1.842 |
stra | +1.806 |
Tel | +1.785 |
Amen | +1.770 |
reopen | +1.740 |
Neg logit contributions | |
ithing | -1.991 |
andom | -1.767 |
nesota | -1.756 |
hing | -1.742 |
arning | -1.731 |
Token | Australian |
Token ID | 6638 |
Feature activation | +4.15e-03 |
Pos logit contributions | |
Stub | +0.663 |
stra | +0.656 |
Amen | +0.641 |
Tel | +0.635 |
stop | +0.631 |
Neg logit contributions | |
ithing | -0.992 |
andom | -0.892 |
nesota | -0.883 |
hing | -0.877 |
arning | -0.876 |
Token | Federal |
Token ID | 5618 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Stub | +0.002 |
stra | +0.002 |
Amen | +0.002 |
stop | +0.002 |
reopen | +0.002 |
Neg logit contributions | |
ithing | -0.003 |
andom | -0.003 |
nesota | -0.003 |
hing | -0.003 |
arning | -0.003 |
Token | Police |
Token ID | 4287 |
Feature activation | +6.90e-01 |
Pos logit contributions | |
ithing | +0.841 |
andom | +0.732 |
nesota | +0.727 |
arning | +0.722 |
hing | +0.721 |
Neg logit contributions | |
Stub | -0.852 |
stra | -0.848 |
Amen | -0.832 |
Tel | -0.824 |
stop | -0.824 |
Token | to |
Token ID | 284 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
Stub | +1.135 |
stra | +1.132 |
Amen | +1.105 |
stop | +1.103 |
reopen | +1.093 |
Neg logit contributions | |
ithing | -1.251 |
andom | -1.105 |
nesota | -1.101 |
arning | -1.082 |
hing | -1.081 |
Token | play |
Token ID | 711 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ithing | +0.089 |
andom | +0.078 |
nesota | +0.078 |
hing | +0.077 |
arning | +0.077 |
Neg logit contributions | |
Stub | -0.080 |
stra | -0.079 |
Amen | -0.078 |
reopen | -0.077 |
stop | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
Stub | +0.745 |
stra | +0.740 |
Amen | +0.725 |
stop | +0.716 |
reopen | +0.714 |
Neg logit contributions | |
ithing | -0.765 |
andom | -0.670 |
nesota | -0.669 |
hing | -0.657 |
arning | -0.656 |
Token | I |
Token ID | 314 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
ithing | +0.544 |
andom | +0.477 |
nesota | +0.471 |
hing | +0.470 |
arning | +0.466 |
Neg logit contributions | |
Stub | -0.504 |
stra | -0.503 |
Amen | -0.492 |
Tel | -0.488 |
stop | -0.486 |
Token | would |
Token ID | 561 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
Stub | +0.462 |
stra | +0.461 |
Amen | +0.450 |
stop | +0.448 |
reopen | +0.447 |
Neg logit contributions | |
ithing | -0.497 |
andom | -0.440 |
nesota | -0.433 |
hing | -0.429 |
arning | -0.428 |
Token | rather |
Token ID | 2138 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
ithing | +0.910 |
andom | +0.799 |
nesota | +0.790 |
hing | +0.789 |
arning | +0.781 |
Neg logit contributions | |
Stub | -0.815 |
stra | -0.810 |
Amen | -0.793 |
Tel | -0.787 |
stop | -0.781 |
Token | go |
Token ID | 467 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
Stub | +4.182 |
reopen | +4.176 |
Amen | +4.082 |
stra | +4.048 |
stop | +4.047 |
Neg logit contributions | |
ithing | -5.447 |
andom | -4.900 |
arning | -4.877 |
nesota | -4.811 |
ammy | -4.801 |
Token | to |
Token ID | 284 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
Stub | +0.713 |
stra | +0.704 |
Amen | +0.694 |
reopen | +0.687 |
stop | +0.686 |
Neg logit contributions | |
ithing | -0.815 |
nesota | -0.721 |
andom | -0.721 |
hing | -0.710 |
arning | -0.709 |
Token | ruins |
Token ID | 20073 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
Stub | +0.349 |
stra | +0.347 |
Amen | +0.341 |
stop | +0.337 |
Tel | +0.336 |
Neg logit contributions | |
ithing | -0.327 |
andom | -0.285 |
nesota | -0.282 |
hing | -0.280 |
arning | -0.278 |
Token | to |
Token ID | 284 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
ithing | +0.517 |
andom | +0.455 |
hing | +0.449 |
nesota | +0.449 |
arning | +0.445 |
Neg logit contributions | |
Stub | -0.463 |
stra | -0.461 |
Amen | -0.450 |
Tel | -0.446 |
stop | -0.446 |
Token | play |
Token ID | 711 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
ithing | +0.255 |
andom | +0.225 |
nesota | +0.223 |
hing | +0.221 |
arning | +0.221 |
Neg logit contributions | |
Stub | -0.221 |
stra | -0.219 |
Amen | -0.215 |
reopen | -0.213 |
stop | -0.212 |
Token | Saturday |
Token ID | 3909 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
Stub | +0.690 |
stra | +0.677 |
Amen | +0.667 |
stop | +0.655 |
reopen | +0.654 |
Neg logit contributions | |
ithing | -0.912 |
andom | -0.815 |
nesota | -0.808 |
hing | -0.799 |
arning | -0.794 |
Token | ahead |
Token ID | 4058 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
Stub | +1.507 |
stra | +1.494 |
Amen | +1.475 |
stop | +1.446 |
reopen | +1.433 |
Neg logit contributions | |
ithing | -1.765 |
nesota | -1.577 |
andom | -1.567 |
hing | -1.535 |
arning | -1.508 |
Token | of |
Token ID | 286 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
stra | +0.584 |
Stub | +0.577 |
Amen | +0.568 |
Tel | +0.565 |
reopen | +0.559 |
Neg logit contributions | |
ithing | -0.607 |
andom | -0.535 |
nesota | -0.528 |
hing | -0.526 |
arning | -0.523 |
Token | his |
Token ID | 465 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
Stub | +1.686 |
stra | +1.668 |
Amen | +1.640 |
stop | +1.621 |
reopen | +1.616 |
Neg logit contributions | |
ithing | -1.748 |
andom | -1.535 |
nesota | -1.511 |
hing | -1.505 |
arning | -1.493 |
Token | final |
Token ID | 2457 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Stub | +0.027 |
stra | +0.027 |
Amen | +0.026 |
Tel | +0.026 |
stop | +0.026 |
Neg logit contributions | |
ithing | -0.032 |
andom | -0.028 |
nesota | -0.028 |
hing | -0.028 |
arning | -0.028 |
Token | match |
Token ID | 2872 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
Stub | +0.099 |
stra | +0.099 |
Amen | +0.097 |
Tel | +0.096 |
stop | +0.096 |
Neg logit contributions | |
ithing | -0.107 |
andom | -0.094 |
nesota | -0.093 |
hing | -0.093 |
arning | -0.092 |
Token | in |
Token ID | 287 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
Stub | +4.318 |
stop | +4.199 |
stra | +4.197 |
Amen | +4.165 |
reopen | +4.086 |
Neg logit contributions | |
ithing | -5.218 |
nesota | -4.587 |
andom | -4.582 |
ammy | -4.487 |
hing | -4.484 |
Token | charge |
Token ID | 3877 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Stub | +0.416 |
stra | +0.412 |
Amen | +0.404 |
stop | +0.397 |
Tel | +0.396 |
Neg logit contributions | |
ithing | -0.571 |
andom | -0.510 |
nesota | -0.504 |
hing | -0.502 |
arning | -0.498 |
Token | at |
Token ID | 379 |
Feature activation | +5.94e-01 |
Pos logit contributions | |
Stub | +0.952 |
stra | +0.945 |
Amen | +0.927 |
stop | +0.918 |
Tel | +0.914 |
Neg logit contributions | |
ithing | -1.029 |
andom | -0.904 |
nesota | -0.897 |
hing | -0.890 |
arning | -0.883 |
Token | Old |
Token ID | 5706 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
stra | +0.362 |
Stub | +0.360 |
Amen | +0.353 |
Tel | +0.350 |
stop | +0.350 |
Neg logit contributions | |
ithing | -0.435 |
andom | -0.383 |
nesota | -0.380 |
hing | -0.379 |
arning | -0.377 |
Token | Trafford |
Token ID | 41107 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
Stub | +0.855 |
stra | +0.854 |
Amen | +0.829 |
reopen | +0.815 |
Tel | +0.814 |
Neg logit contributions | |
ithing | -1.903 |
andom | -1.729 |
nesota | -1.716 |
hing | -1.711 |
arning | -1.703 |
Token | money |
Token ID | 1637 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
Stub | +1.760 |
stra | +1.744 |
Amen | +1.707 |
stop | +1.688 |
reopen | +1.672 |
Neg logit contributions | |
ithing | -2.405 |
nesota | -2.162 |
andom | -2.149 |
hing | -2.113 |
arning | -2.112 |
Token | to |
Token ID | 284 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
Stub | +1.207 |
stra | +1.205 |
Amen | +1.174 |
reopen | +1.167 |
stop | +1.164 |
Neg logit contributions | |
ithing | -1.429 |
andom | -1.269 |
nesota | -1.268 |
hing | -1.239 |
arning | -1.231 |
Token | flow |
Token ID | 5202 |
Feature activation | +9.85e-01 |
Pos logit contributions | |
ithing | +0.913 |
andom | +0.805 |
hing | +0.793 |
nesota | +0.793 |
arning | +0.785 |
Neg logit contributions | |
Stub | -0.789 |
stra | -0.784 |
Amen | -0.768 |
Tel | -0.762 |
stop | -0.758 |
Token | to |
Token ID | 284 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
Stub | +0.617 |
stra | +0.612 |
Amen | +0.599 |
stop | +0.594 |
reopen | +0.591 |
Neg logit contributions | |
ithing | -0.714 |
andom | -0.634 |
nesota | -0.627 |
arning | -0.618 |
hing | -0.617 |
Token | par |
Token ID | 1582 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
Stub | +0.210 |
stra | +0.209 |
Amen | +0.205 |
stop | +0.202 |
Tel | +0.202 |
Neg logit contributions | |
ithing | -0.270 |
andom | -0.240 |
nesota | -0.237 |
hing | -0.236 |
arning | -0.235 |
Token | och |
Token ID | 5374 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
stra | +0.683 |
Stub | +0.673 |
Amen | +0.643 |
stop | +0.630 |
reopen | +0.615 |
Neg logit contributions | |
ithing | -2.998 |
andom | -2.764 |
nesota | -2.752 |
hing | -2.734 |
arning | -2.732 |
Token | ial |
Token ID | 498 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
stra | +4.278 |
stop | +4.130 |
Stub | +4.121 |
iest | +4.074 |
pub | +4.067 |
Neg logit contributions | |
ithing | -4.971 |
nesota | -4.462 |
andom | -4.398 |
hing | -4.347 |
ammy | -4.305 |
Token | schools |
Token ID | 4266 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
Stub | +1.126 |
stra | +1.123 |
Amen | +1.099 |
stop | +1.085 |
reopen | +1.083 |
Neg logit contributions | |
ithing | -1.191 |
andom | -1.048 |
nesota | -1.037 |
hing | -1.029 |
arning | -1.023 |
Token | . |
Token ID | 13 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
Stub | +1.233 |
stra | +1.229 |
Amen | +1.197 |
reopen | +1.186 |
stop | +1.185 |
Neg logit contributions | |
ithing | -1.581 |
andom | -1.415 |
nesota | -1.402 |
arning | -1.379 |
hing | -1.379 |
Token | It |
Token ID | 632 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
ithing | +0.643 |
andom | +0.562 |
nesota | +0.556 |
hing | +0.552 |
arning | +0.548 |
Neg logit contributions | |
Stub | -0.650 |
stra | -0.648 |
Amen | -0.634 |
stop | -0.629 |
Tel | -0.628 |
Token | 's |
Token ID | 338 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ithing | +1.386 |
andom | +1.208 |
hing | +1.202 |
nesota | +1.196 |
arning | +1.177 |
Neg logit contributions | |
Stub | -1.422 |
stra | -1.416 |
Amen | -1.395 |
Tel | -1.391 |
stop | -1.378 |
Token | know |
Token ID | 760 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
Stub | +0.055 |
stra | +0.055 |
Amen | +0.054 |
reopen | +0.054 |
stop | +0.054 |
Neg logit contributions | |
ithing | -0.040 |
nesota | -0.034 |
andom | -0.034 |
arning | -0.033 |
hing | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
ithing | +1.476 |
andom | +1.292 |
hing | +1.276 |
nesota | +1.265 |
arning | +1.256 |
Neg logit contributions | |
Stub | -1.481 |
stra | -1.474 |
Amen | -1.434 |
Tel | -1.427 |
stop | -1.424 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.19e-02 |
Pos logit contributions | |
ithing | +0.131 |
andom | +0.115 |
nesota | +0.114 |
arning | +0.112 |
hing | +0.112 |
Neg logit contributions | |
Stub | -0.128 |
stra | -0.127 |
Amen | -0.126 |
stop | -0.123 |
Tel | -0.123 |
Token | Why |
Token ID | 5195 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
Stub | +0.049 |
stra | +0.049 |
Amen | +0.048 |
stop | +0.047 |
reopen | +0.047 |
Neg logit contributions | |
ithing | -0.061 |
andom | -0.054 |
nesota | -0.054 |
hing | -0.053 |
arning | -0.053 |
Token | automation |
Token ID | 22771 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
ithing | +1.469 |
andom | +1.282 |
hing | +1.256 |
nesota | +1.256 |
arning | +1.242 |
Neg logit contributions | |
stra | -1.581 |
Stub | -1.579 |
Amen | -1.540 |
Tel | -1.536 |
stop | -1.528 |
Token | rather |
Token ID | 2138 |
Feature activation | +6.92e+00 |
Pos logit contributions | |
ithing | +1.857 |
andom | +1.620 |
hing | +1.596 |
nesota | +1.593 |
arning | +1.578 |
Neg logit contributions | |
Stub | -1.963 |
stra | -1.959 |
Amen | -1.922 |
Tel | -1.904 |
stop | -1.895 |
Token | than |
Token ID | 621 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
stra | +2.906 |
Stub | +2.897 |
stop | +2.798 |
Amen | +2.791 |
reopen | +2.750 |
Neg logit contributions | |
ithing | -6.280 |
andom | -5.714 |
nesota | -5.611 |
arning | -5.594 |
hing | -5.556 |
Token | Brexit |
Token ID | 11435 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
Stub | +1.390 |
stra | +1.375 |
Amen | +1.356 |
reopen | +1.338 |
stop | +1.336 |
Neg logit contributions | |
ithing | -1.537 |
andom | -1.354 |
nesota | -1.347 |
hing | -1.332 |
arning | -1.321 |
Token | is |
Token ID | 318 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
ithing | +0.170 |
andom | +0.148 |
nesota | +0.147 |
hing | +0.146 |
arning | +0.145 |
Neg logit contributions | |
Stub | -0.171 |
stra | -0.170 |
Amen | -0.167 |
stop | -0.165 |
reopen | -0.165 |
Token | the |
Token ID | 262 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
Stub | +0.444 |
stra | +0.440 |
Amen | +0.432 |
reopen | +0.425 |
stop | +0.425 |
Neg logit contributions | |
ithing | -0.468 |
andom | -0.409 |
nesota | -0.407 |
hing | -0.399 |
arning | -0.399 |
Token | prime |
Token ID | 6994 |
Feature activation | -6.96e-02 |
Pos logit contributions | |
Stub | +0.543 |
stra | +0.536 |
Amen | +0.525 |
stop | +0.519 |
reopen | +0.517 |
Neg logit contributions | |
ithing | -0.702 |
andom | -0.623 |
nesota | -0.621 |
hing | -0.611 |
arning | -0.609 |
Token | to |
Token ID | 284 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
Stub | +1.026 |
stra | +1.022 |
stop | +0.996 |
Amen | +0.989 |
reopen | +0.979 |
Neg logit contributions | |
ithing | -1.424 |
andom | -1.265 |
arning | -1.258 |
nesota | -1.250 |
ammy | -1.236 |
Token | utilize |
Token ID | 17624 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
ithing | +0.153 |
andom | +0.135 |
nesota | +0.133 |
hing | +0.132 |
arning | +0.131 |
Neg logit contributions | |
Stub | -0.144 |
stra | -0.144 |
Amen | -0.141 |
stop | -0.139 |
Tel | -0.139 |
Token | fat |
Token ID | 3735 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ithing | +0.206 |
andom | +0.182 |
nesota | +0.180 |
hing | +0.179 |
arning | +0.178 |
Neg logit contributions | |
Stub | -0.171 |
stra | -0.170 |
Amen | -0.166 |
stop | -0.164 |
Tel | -0.164 |
Token | for |
Token ID | 329 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
stra | +0.736 |
Stub | +0.735 |
Amen | +0.712 |
stop | +0.709 |
reopen | +0.709 |
Neg logit contributions | |
ithing | -0.769 |
andom | -0.675 |
nesota | -0.668 |
arning | -0.662 |
hing | -0.661 |
Token | energy |
Token ID | 2568 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
ithing | +0.326 |
andom | +0.288 |
nesota | +0.285 |
hing | +0.283 |
arning | +0.282 |
Neg logit contributions | |
Stub | -0.283 |
stra | -0.280 |
Amen | -0.275 |
stop | -0.271 |
reopen | -0.271 |
Token | rather |
Token ID | 2138 |
Feature activation | +6.85e+00 |
Pos logit contributions | |
Stub | +1.739 |
stra | +1.731 |
stop | +1.662 |
Amen | +1.662 |
reopen | +1.655 |
Neg logit contributions | |
ithing | -2.172 |
andom | -1.938 |
nesota | -1.906 |
hing | -1.896 |
arning | -1.883 |
Token | than |
Token ID | 621 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
Stub | +2.241 |
stra | +2.163 |
Amen | +2.074 |
stop | +2.063 |
reopen | +2.024 |
Neg logit contributions | |
ithing | -6.844 |
andom | -6.325 |
arning | -6.294 |
ammy | -6.244 |
nesota | -6.127 |
Token | glucose |
Token ID | 15701 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
Stub | +1.222 |
stra | +1.200 |
Amen | +1.179 |
reopen | +1.176 |
stop | +1.169 |
Neg logit contributions | |
ithing | -1.443 |
andom | -1.288 |
nesota | -1.277 |
hing | -1.259 |
arning | -1.253 |
Token | , |
Token ID | 11 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Stub | +0.829 |
stra | +0.826 |
Amen | +0.806 |
stop | +0.797 |
reopen | +0.795 |
Neg logit contributions | |
ithing | -0.971 |
andom | -0.861 |
nesota | -0.850 |
hing | -0.845 |
arning | -0.841 |
Token | as |
Token ID | 355 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Stub | +0.071 |
stra | +0.071 |
Amen | +0.070 |
stop | +0.069 |
reopen | +0.069 |
Neg logit contributions | |
ithing | -0.071 |
andom | -0.062 |
nesota | -0.061 |
hing | -0.061 |
arning | -0.061 |
Token | there |
Token ID | 612 |
Feature activation | -9.59e-01 |
Pos logit contributions | |
ithing | +1.257 |
andom | +1.087 |
hing | +1.076 |
nesota | +1.073 |
arning | +1.059 |
Neg logit contributions | |
stra | -1.509 |
Stub | -1.504 |
Amen | -1.479 |
Tel | -1.468 |
stop | -1.466 |
Token | message |
Token ID | 3275 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ithing | +0.668 |
andom | +0.579 |
nesota | +0.570 |
hing | +0.567 |
arning | +0.563 |
Neg logit contributions | |
Stub | -0.766 |
stra | -0.763 |
Amen | -0.749 |
stop | -0.741 |
Tel | -0.741 |
Token | to |
Token ID | 284 |
Feature activation | -8.96e-01 |
Pos logit contributions | |
ithing | +1.631 |
andom | +1.441 |
hing | +1.424 |
nesota | +1.412 |
arning | +1.406 |
Neg logit contributions | |
Stub | -1.495 |
stra | -1.488 |
Amen | -1.456 |
Tel | -1.443 |
reopen | -1.435 |
Token | those |
Token ID | 883 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
ithing | +0.625 |
andom | +0.550 |
nesota | +0.543 |
hing | +0.540 |
arning | +0.537 |
Neg logit contributions | |
Stub | -0.583 |
stra | -0.580 |
Amen | -0.568 |
stop | -0.562 |
Tel | -0.561 |
Token | in |
Token ID | 287 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ithing | +0.225 |
andom | +0.198 |
nesota | +0.195 |
hing | +0.194 |
arning | +0.193 |
Neg logit contributions | |
Stub | -0.220 |
stra | -0.219 |
Amen | -0.214 |
stop | -0.212 |
Tel | -0.212 |
Token | power |
Token ID | 1176 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
Stub | +1.348 |
stra | +1.343 |
Amen | +1.321 |
stop | +1.310 |
Tel | +1.310 |
Neg logit contributions | |
ithing | -0.761 |
andom | -0.627 |
nesota | -0.617 |
hing | -0.612 |
arning | -0.608 |
Token | rather |
Token ID | 2138 |
Feature activation | +6.81e+00 |
Pos logit contributions | |
Stub | +1.195 |
stra | +1.190 |
Amen | +1.168 |
stop | +1.157 |
reopen | +1.146 |
Neg logit contributions | |
ithing | -1.565 |
andom | -1.390 |
nesota | -1.382 |
arning | -1.361 |
hing | -1.349 |
Token | than |
Token ID | 621 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Stub | +2.280 |
stra | +2.220 |
Amen | +2.196 |
reopen | +2.164 |
stop | +2.134 |
Neg logit contributions | |
ithing | -6.721 |
andom | -6.181 |
arning | -6.099 |
nesota | -6.079 |
ammy | -6.079 |
Token | cont |
Token ID | 542 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
Stub | +1.053 |
stra | +1.046 |
Amen | +1.026 |
reopen | +1.018 |
stop | +1.017 |
Neg logit contributions | |
ithing | -1.124 |
andom | -0.987 |
nesota | -0.981 |
hing | -0.969 |
arning | -0.967 |
Token | ending |
Token ID | 1571 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ithing | +1.650 |
andom | +1.436 |
hing | +1.415 |
arning | +1.406 |
nesota | +1.402 |
Neg logit contributions | |
Stub | -1.669 |
stra | -1.647 |
Amen | -1.634 |
reopen | -1.610 |
Tel | -1.606 |
Token | for |
Token ID | 329 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
Stub | +0.923 |
stra | +0.914 |
Amen | +0.900 |
stop | +0.893 |
reopen | +0.891 |
Neg logit contributions | |
ithing | -0.830 |
andom | -0.717 |
nesota | -0.712 |
hing | -0.704 |
arning | -0.703 |
Token | power |
Token ID | 1176 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
ithing | +0.198 |
andom | +0.173 |
nesota | +0.172 |
hing | +0.171 |
arning | +0.170 |
Neg logit contributions | |
Stub | -0.192 |
stra | -0.190 |
Amen | -0.187 |
stop | -0.185 |
reopen | -0.185 |
Token | climate |
Token ID | 4258 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
Stub | +1.488 |
stra | +1.461 |
Amen | +1.453 |
reopen | +1.421 |
stop | +1.418 |
Neg logit contributions | |
ithing | -1.953 |
andom | -1.739 |
nesota | -1.726 |
arning | -1.712 |
hing | -1.687 |
Token | sensitivity |
Token ID | 14233 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
ithing | +1.899 |
andom | +1.735 |
hing | +1.722 |
nesota | +1.701 |
arning | +1.673 |
Neg logit contributions | |
stra | -1.185 |
Stub | -1.179 |
Amen | -1.143 |
Tel | -1.133 |
stop | -1.122 |
Token | by |
Token ID | 416 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
Stub | +0.352 |
stra | +0.351 |
Amen | +0.344 |
stop | +0.339 |
Tel | +0.339 |
Neg logit contributions | |
ithing | -0.396 |
andom | -0.349 |
nesota | -0.345 |
hing | -0.344 |
arning | -0.341 |
Token | empirical |
Token ID | 21594 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
Stub | +0.027 |
stra | +0.026 |
Amen | +0.026 |
stop | +0.025 |
Tel | +0.025 |
Neg logit contributions | |
ithing | -0.031 |
andom | -0.027 |
nesota | -0.027 |
hing | -0.027 |
arning | -0.027 |
Token | methods |
Token ID | 5050 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
Stub | +0.754 |
stra | +0.743 |
Amen | +0.720 |
reopen | +0.712 |
stop | +0.711 |
Neg logit contributions | |
ithing | -1.124 |
andom | -1.005 |
nesota | -1.001 |
arning | -0.983 |
ammy | -0.979 |
Token | rather |
Token ID | 2138 |
Feature activation | +6.76e+00 |
Pos logit contributions | |
Stub | +0.694 |
stra | +0.691 |
Amen | +0.673 |
stop | +0.665 |
reopen | +0.664 |
Neg logit contributions | |
ithing | -0.881 |
andom | -0.782 |
nesota | -0.775 |
hing | -0.767 |
arning | -0.766 |
Token | than |
Token ID | 621 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
Stub | +2.270 |
stra | +2.208 |
reopen | +2.131 |
Amen | +2.119 |
stop | +2.074 |
Neg logit contributions | |
ithing | -6.659 |
andom | -6.139 |
arning | -6.085 |
ammy | -6.069 |
nesota | -6.012 |
Token | by |
Token ID | 416 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Stub | +1.270 |
stra | +1.251 |
Amen | +1.233 |
reopen | +1.228 |
stop | +1.215 |
Neg logit contributions | |
ithing | -1.333 |
andom | -1.170 |
nesota | -1.166 |
arning | -1.146 |
hing | -1.144 |
Token | the |
Token ID | 262 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
ithing | +0.790 |
andom | +0.695 |
hing | +0.685 |
nesota | +0.684 |
arning | +0.679 |
Neg logit contributions | |
Stub | -0.734 |
stra | -0.734 |
Amen | -0.717 |
stop | -0.711 |
Tel | -0.710 |
Token | use |
Token ID | 779 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
Stub | +0.321 |
stra | +0.318 |
Amen | +0.310 |
stop | +0.306 |
reopen | +0.305 |
Neg logit contributions | |
ithing | -0.500 |
andom | -0.450 |
nesota | -0.446 |
hing | -0.442 |
arning | -0.441 |
Token | of |
Token ID | 286 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ithing | +1.423 |
andom | +1.258 |
nesota | +1.243 |
hing | +1.237 |
arning | +1.230 |
Neg logit contributions | |
Stub | -1.223 |
stra | -1.217 |
Amen | -1.191 |
stop | -1.177 |
Tel | -1.176 |
Token | talk |
Token ID | 1561 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
reopen | +1.907 |
Stub | +1.854 |
stra | +1.846 |
stop | +1.825 |
Amen | +1.818 |
Neg logit contributions | |
ithing | -2.327 |
andom | -2.067 |
arning | -2.054 |
nesota | -2.047 |
ammy | -1.995 |
Token | more |
Token ID | 517 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
Stub | +0.393 |
stra | +0.390 |
Amen | +0.383 |
stop | +0.379 |
reopen | +0.378 |
Neg logit contributions | |
ithing | -0.417 |
andom | -0.368 |
nesota | -0.363 |
arning | -0.359 |
hing | -0.358 |
Token | but |
Token ID | 475 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
ithing | +0.305 |
andom | +0.265 |
nesota | +0.260 |
arning | +0.258 |
hing | +0.258 |
Neg logit contributions | |
Stub | -0.345 |
stra | -0.343 |
Amen | -0.337 |
stop | -0.334 |
reopen | -0.333 |
Token | I |
Token ID | 314 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
ithing | +0.426 |
andom | +0.374 |
nesota | +0.370 |
hing | +0.368 |
arning | +0.366 |
Neg logit contributions | |
Stub | -0.399 |
stra | -0.397 |
Amen | -0.389 |
stop | -0.385 |
Tel | -0.384 |
Token | 'd |
Token ID | 1549 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
Stub | +0.279 |
stra | +0.278 |
reopen | +0.273 |
Amen | +0.272 |
stop | +0.271 |
Neg logit contributions | |
ithing | -0.283 |
andom | -0.250 |
nesota | -0.247 |
arning | -0.243 |
hing | -0.242 |
Token | rather |
Token ID | 2138 |
Feature activation | +6.74e+00 |
Pos logit contributions | |
ithing | +0.684 |
andom | +0.595 |
nesota | +0.587 |
hing | +0.583 |
arning | +0.580 |
Neg logit contributions | |
Stub | -0.745 |
stra | -0.742 |
Amen | -0.727 |
stop | -0.721 |
reopen | -0.719 |
Token | have |
Token ID | 423 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
reopen | +4.113 |
Stub | +4.017 |
stra | +3.962 |
stop | +3.961 |
Amen | +3.899 |
Neg logit contributions | |
ithing | -4.927 |
arning | -4.378 |
andom | -4.369 |
nesota | -4.342 |
ammy | -4.294 |
Token | someone |
Token ID | 2130 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
Stub | +1.684 |
stra | +1.662 |
Amen | +1.633 |
reopen | +1.630 |
stop | +1.618 |
Neg logit contributions | |
ithing | -1.792 |
andom | -1.566 |
arning | -1.559 |
nesota | -1.550 |
hing | -1.541 |
Token | here |
Token ID | 994 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
Stub | +0.778 |
stra | +0.776 |
reopen | +0.761 |
Amen | +0.760 |
stop | +0.758 |
Neg logit contributions | |
ithing | -0.836 |
andom | -0.736 |
nesota | -0.731 |
arning | -0.722 |
hing | -0.718 |
Token | with |
Token ID | 351 |
Feature activation | +5.92e-01 |
Pos logit contributions | |
Stub | +0.117 |
stra | +0.116 |
Amen | +0.114 |
stop | +0.113 |
reopen | +0.113 |
Neg logit contributions | |
ithing | -0.126 |
andom | -0.111 |
nesota | -0.110 |
hing | -0.109 |
arning | -0.109 |
Token | me |
Token ID | 502 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Stub | +0.394 |
stra | +0.391 |
Amen | +0.383 |
reopen | +0.379 |
stop | +0.379 |
Neg logit contributions | |
ithing | -0.405 |
andom | -0.356 |
nesota | -0.351 |
arning | -0.348 |
hing | -0.348 |
Token | be |
Token ID | 307 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
ithing | +0.245 |
andom | +0.210 |
nesota | +0.206 |
hing | +0.205 |
arning | +0.204 |
Neg logit contributions | |
Stub | -0.322 |
stra | -0.321 |
Amen | -0.315 |
stop | -0.312 |
Tel | -0.312 |
Token | so |
Token ID | 523 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
ithing | +0.179 |
andom | +0.157 |
nesota | +0.155 |
hing | +0.154 |
arning | +0.153 |
Neg logit contributions | |
Stub | -0.180 |
stra | -0.180 |
Amen | -0.176 |
stop | -0.174 |
Tel | -0.174 |
Token | I |
Token ID | 314 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
ithing | +0.265 |
andom | +0.232 |
nesota | +0.230 |
arning | +0.227 |
hing | +0.226 |
Neg logit contributions | |
Stub | -0.268 |
stra | -0.265 |
Amen | -0.261 |
stop | -0.259 |
reopen | -0.258 |
Token | actually |
Token ID | 1682 |
Feature activation | -9.24e-01 |
Pos logit contributions | |
ithing | +0.112 |
andom | +0.098 |
nesota | +0.097 |
arning | +0.095 |
hing | +0.094 |
Neg logit contributions | |
Stub | -0.121 |
stra | -0.120 |
Amen | -0.118 |
reopen | -0.117 |
stop | -0.117 |
Token | would |
Token ID | 561 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ithing | +0.640 |
andom | +0.563 |
nesota | +0.557 |
hing | +0.551 |
arning | +0.550 |
Neg logit contributions | |
Stub | -0.606 |
stra | -0.602 |
Amen | -0.590 |
stop | -0.584 |
reopen | -0.583 |
Token | rather |
Token ID | 2138 |
Feature activation | +6.74e+00 |
Pos logit contributions | |
ithing | +0.861 |
andom | +0.756 |
hing | +0.746 |
nesota | +0.744 |
arning | +0.738 |
Neg logit contributions | |
Stub | -0.814 |
stra | -0.810 |
Amen | -0.794 |
Tel | -0.786 |
stop | -0.782 |
Token | the |
Token ID | 262 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
Stub | +4.016 |
reopen | +3.967 |
stop | +3.857 |
Amen | +3.807 |
stra | +3.796 |
Neg logit contributions | |
ithing | -4.982 |
nesota | -4.499 |
andom | -4.445 |
arning | -4.409 |
ammy | -4.273 |
Token | screen |
Token ID | 3159 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ithing | +0.255 |
andom | +0.223 |
nesota | +0.221 |
hing | +0.220 |
arning | +0.218 |
Neg logit contributions | |
Stub | -0.248 |
stra | -0.247 |
Amen | -0.242 |
stop | -0.239 |
Tel | -0.239 |
Token | door |
Token ID | 3420 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
ithing | +0.091 |
andom | +0.080 |
hing | +0.079 |
nesota | +0.079 |
arning | +0.078 |
Neg logit contributions | |
Stub | -0.082 |
stra | -0.081 |
Amen | -0.080 |
Tel | -0.079 |
stop | -0.079 |
Token | effect |
Token ID | 1245 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ithing | +0.006 |
andom | +0.004 |
nesota | +0.004 |
hing | +0.004 |
arning | +0.004 |
Neg logit contributions | |
Stub | -0.018 |
stra | -0.018 |
Amen | -0.017 |
Tel | -0.017 |
stop | -0.017 |
Token | image |
Token ID | 2939 |
Feature activation | +8.39e-01 |
Pos logit contributions | |
ithing | +0.927 |
andom | +0.800 |
hing | +0.788 |
nesota | +0.786 |
arning | +0.778 |
Neg logit contributions | |
Stub | -1.086 |
stra | -1.084 |
Amen | -1.063 |
Tel | -1.053 |
stop | -1.049 |
Token | SL |
Token ID | 12419 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
Stub | +0.767 |
stra | +0.764 |
Amen | +0.748 |
stop | +0.739 |
Tel | +0.738 |
Neg logit contributions | |
ithing | -0.881 |
andom | -0.779 |
nesota | -0.768 |
hing | -0.766 |
arning | -0.761 |
Token | AY |
Token ID | 4792 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Stub | +1.353 |
stop | +1.327 |
Tel | +1.322 |
Amen | +1.307 |
stra | +1.296 |
Neg logit contributions | |
ithing | -1.406 |
nesota | -1.235 |
andom | -1.207 |
arning | -1.193 |
ammy | -1.174 |
Token | ER |
Token ID | 1137 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
Stub | +0.907 |
stra | +0.887 |
Tel | +0.876 |
stop | +0.871 |
Amen | +0.859 |
Neg logit contributions | |
ithing | -1.340 |
andom | -1.191 |
arning | -1.175 |
nesota | -1.174 |
ammy | -1.161 |
Token | ( |
Token ID | 357 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
Stub | +1.369 |
stra | +1.360 |
Amen | +1.332 |
stop | +1.321 |
Tel | +1.309 |
Neg logit contributions | |
ithing | -1.620 |
nesota | -1.429 |
andom | -1.419 |
arning | -1.402 |
hing | -1.400 |
Token | Dark |
Token ID | 17367 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
stra | +0.871 |
Stub | +0.865 |
Tel | +0.846 |
stop | +0.842 |
Amen | +0.834 |
Neg logit contributions | |
ithing | -1.142 |
nesota | -1.006 |
andom | -1.005 |
arning | -0.989 |
hing | -0.986 |
Token | Horse |
Token ID | 18455 |
Feature activation | +6.70e+00 |
Pos logit contributions | |
Stub | +2.544 |
stra | +2.543 |
stop | +2.475 |
Tel | +2.455 |
Amen | +2.453 |
Neg logit contributions | |
ithing | -1.799 |
andom | -1.517 |
nesota | -1.516 |
hing | -1.470 |
arning | -1.454 |
Token | ) |
Token ID | 8 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Stub | +3.397 |
stra | +3.358 |
pub | +3.275 |
stop | +3.230 |
Amen | +3.227 |
Neg logit contributions | |
ithing | -5.599 |
nesota | -5.105 |
andom | -4.893 |
hing | -4.870 |
arning | -4.867 |
Token | – |
Token ID | 784 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
ithing | +0.084 |
andom | +0.075 |
nesota | +0.074 |
hing | +0.074 |
arning | +0.074 |
Neg logit contributions | |
Stub | -0.057 |
stra | -0.057 |
Amen | -0.055 |
Tel | -0.055 |
stop | -0.055 |
Token | J |
Token ID | 449 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Stub | +0.095 |
stra | +0.094 |
Amen | +0.092 |
stop | +0.091 |
Tel | +0.091 |
Neg logit contributions | |
ithing | -0.094 |
andom | -0.082 |
nesota | -0.081 |
arning | -0.081 |
hing | -0.080 |
Token | oss |
Token ID | 793 |
Feature activation | -2.73e-03 |
Pos logit contributions | |
Stub | +0.091 |
stra | +0.091 |
Amen | +0.089 |
stop | +0.088 |
Tel | +0.088 |
Neg logit contributions | |
ithing | -0.093 |
andom | -0.081 |
nesota | -0.081 |
hing | -0.080 |
arning | -0.079 |
Token | Whedon |
Token ID | 49891 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ithing | +0.002 |
andom | +0.002 |
nesota | +0.002 |
hing | +0.002 |
arning | +0.002 |
Neg logit contributions | |
Stub | -0.002 |
stra | -0.002 |
Amen | -0.002 |
stop | -0.002 |
Tel | -0.002 |
Token | research |
Token ID | 2267 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
Stub | +1.143 |
stra | +1.135 |
Amen | +1.117 |
Tel | +1.103 |
stop | +1.101 |
Neg logit contributions | |
ithing | -1.053 |
andom | -0.920 |
nesota | -0.909 |
hing | -0.898 |
arning | -0.894 |
Token | and |
Token ID | 290 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ithing | +0.396 |
andom | +0.346 |
nesota | +0.341 |
hing | +0.340 |
arning | +0.338 |
Neg logit contributions | |
Stub | -0.394 |
stra | -0.393 |
Amen | -0.385 |
stop | -0.381 |
Tel | -0.380 |
Token | development |
Token ID | 2478 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Stub | +0.681 |
stra | +0.674 |
Amen | +0.661 |
reopen | +0.655 |
Tel | +0.653 |
Neg logit contributions | |
ithing | -0.964 |
andom | -0.865 |
nesota | -0.855 |
arning | -0.846 |
hing | -0.845 |
Token | ( |
Token ID | 357 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Stub | +0.984 |
stra | +0.979 |
Amen | +0.959 |
reopen | +0.947 |
stop | +0.947 |
Neg logit contributions | |
ithing | -0.964 |
andom | -0.851 |
nesota | -0.836 |
arning | -0.828 |
hing | -0.828 |
Token | R |
Token ID | 49 |
Feature activation | -9.31e-01 |
Pos logit contributions | |
ithing | +1.136 |
andom | +0.991 |
hing | +0.984 |
arning | +0.974 |
nesota | +0.971 |
Neg logit contributions | |
Stub | -1.214 |
stra | -1.206 |
Amen | -1.198 |
reopen | -1.174 |
stop | -1.167 |
Token | & |
Token ID | 5 |
Feature activation | +6.69e+00 |
Pos logit contributions | |
ithing | +0.772 |
andom | +0.697 |
hing | +0.688 |
arning | +0.683 |
nesota | +0.681 |
Neg logit contributions | |
Stub | -0.481 |
stra | -0.477 |
Amen | -0.467 |
reopen | -0.461 |
stop | -0.457 |
Token | D |
Token ID | 35 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Tel | +3.060 |
stop | +3.032 |
Stub | +3.027 |
stra | +3.006 |
pub | +2.993 |
Neg logit contributions | |
ithing | -6.021 |
nesota | -5.416 |
andom | -5.298 |
ammy | -5.203 |
arning | -5.191 |
Token | ) |
Token ID | 8 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Stub | +0.778 |
stra | +0.773 |
Amen | +0.754 |
stop | +0.744 |
Tel | +0.743 |
Neg logit contributions | |
ithing | -1.177 |
andom | -1.055 |
nesota | -1.044 |
hing | -1.039 |
arning | -1.034 |
Token | tot |
Token ID | 2006 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
stra | +0.753 |
Stub | +0.752 |
Amen | +0.740 |
stop | +0.729 |
reopen | +0.727 |
Neg logit contributions | |
ithing | -0.727 |
andom | -0.644 |
nesota | -0.634 |
arning | -0.624 |
hing | -0.623 |
Token | alled |
Token ID | 4262 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ithing | +0.988 |
andom | +0.876 |
nesota | +0.876 |
arning | +0.855 |
ammy | +0.846 |
Neg logit contributions | |
stra | -0.843 |
Stub | -0.841 |
stop | -0.821 |
Amen | -0.819 |
Tel | -0.818 |
Token | 2 |
Token ID | 362 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ithing | +0.721 |
andom | +0.635 |
nesota | +0.627 |
hing | +0.623 |
arning | +0.620 |
Neg logit contributions | |
Stub | -0.671 |
stra | -0.668 |
Amen | -0.655 |
stop | -0.647 |
Tel | -0.646 |
Token | t |
Token ID | 83 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
ithing | +0.199 |
andom | +0.171 |
nesota | +0.170 |
hing | +0.167 |
arning | +0.167 |
Neg logit contributions | |
Stub | -0.243 |
stra | -0.243 |
Amen | -0.238 |
stop | -0.235 |
Tel | -0.235 |
Token | bad |
Token ID | 2089 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
ithing | +0.951 |
andom | +0.841 |
hing | +0.831 |
nesota | +0.824 |
arning | +0.819 |
Neg logit contributions | |
Stub | -0.895 |
stra | -0.893 |
Amen | -0.875 |
Tel | -0.868 |
stop | -0.864 |
Token | at |
Token ID | 379 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ithing | +0.136 |
andom | +0.118 |
nesota | +0.117 |
hing | +0.116 |
arning | +0.115 |
Neg logit contributions | |
Stub | -0.142 |
stra | -0.141 |
Amen | -0.139 |
Tel | -0.137 |
stop | -0.137 |
Token | all |
Token ID | 477 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
Stub | +1.295 |
stra | +1.288 |
Amen | +1.252 |
stop | +1.251 |
reopen | +1.241 |
Neg logit contributions | |
ithing | -1.035 |
nesota | -0.899 |
andom | -0.891 |
arning | -0.856 |
hing | -0.846 |
Token | . |
Token ID | 13 |
Feature activation | +7.27e-01 |
Pos logit contributions | |
Stub | +1.402 |
stra | +1.396 |
Amen | +1.363 |
stop | +1.346 |
Tel | +1.338 |
Neg logit contributions | |
ithing | -1.867 |
andom | -1.667 |
nesota | -1.651 |
hing | -1.628 |
arning | -1.627 |
Token | Rather |
Token ID | 11317 |
Feature activation | +6.69e+00 |
Pos logit contributions | |
Stub | +0.506 |
stra | +0.496 |
Amen | +0.495 |
stop | +0.483 |
Tel | +0.482 |
Neg logit contributions | |
ithing | -0.485 |
andom | -0.421 |
nesota | -0.418 |
arning | -0.415 |
hing | -0.410 |
Token | , |
Token ID | 11 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Stub | +3.856 |
stra | +3.801 |
Amen | +3.756 |
stop | +3.678 |
reopen | +3.660 |
Neg logit contributions | |
ithing | -5.033 |
andom | -4.495 |
nesota | -4.447 |
arning | -4.418 |
ammy | -4.373 |
Token | it |
Token ID | 340 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
Stub | +1.113 |
stra | +1.103 |
Amen | +1.089 |
stop | +1.072 |
reopen | +1.069 |
Neg logit contributions | |
ithing | -1.082 |
andom | -0.944 |
nesota | -0.938 |
arning | -0.928 |
hing | -0.924 |
Token | is |
Token ID | 318 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Stub | +0.559 |
stra | +0.557 |
Amen | +0.547 |
stop | +0.541 |
Tel | +0.540 |
Neg logit contributions | |
ithing | -0.510 |
andom | -0.443 |
nesota | -0.437 |
hing | -0.434 |
arning | -0.432 |
Token | … |
Token ID | 1399 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
Stub | +0.875 |
stra | +0.871 |
Amen | +0.854 |
stop | +0.842 |
Tel | +0.840 |
Neg logit contributions | |
ithing | -0.893 |
andom | -0.779 |
nesota | -0.772 |
hing | -0.764 |
arning | -0.763 |
Token | well |
Token ID | 4053 |
Feature activation | -8.38e-01 |
Pos logit contributions | |
Stub | +0.281 |
stra | +0.279 |
Amen | +0.274 |
stop | +0.271 |
Tel | +0.270 |
Neg logit contributions | |
ithing | -0.295 |
andom | -0.259 |
nesota | -0.256 |
hing | -0.255 |
arning | -0.253 |
Token | is |
Token ID | 318 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
Stub | +0.439 |
stra | +0.436 |
Amen | +0.429 |
stop | +0.423 |
Tel | +0.423 |
Neg logit contributions | |
ithing | -0.440 |
andom | -0.386 |
nesota | -0.381 |
hing | -0.378 |
arning | -0.376 |
Token | no |
Token ID | 645 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
Stub | +0.365 |
stra | +0.359 |
Amen | +0.355 |
reopen | +0.349 |
stop | +0.349 |
Neg logit contributions | |
ithing | -0.335 |
andom | -0.289 |
nesota | -0.286 |
arning | -0.284 |
hing | -0.283 |
Token | one |
Token ID | 530 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ithing | +1.063 |
andom | +0.934 |
nesota | +0.918 |
hing | +0.918 |
arning | +0.913 |
Neg logit contributions | |
stra | -1.059 |
Stub | -1.054 |
Amen | -1.033 |
Tel | -1.027 |
stop | -1.020 |
Token | I |
Token ID | 314 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
ithing | +0.907 |
andom | +0.806 |
hing | +0.797 |
nesota | +0.790 |
arning | +0.785 |
Neg logit contributions | |
stra | -0.843 |
Stub | -0.837 |
Tel | -0.820 |
Amen | -0.819 |
reopen | -0.809 |
Token | would |
Token ID | 561 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
ithing | +0.063 |
andom | +0.055 |
nesota | +0.054 |
hing | +0.054 |
arning | +0.053 |
Neg logit contributions | |
Stub | -0.072 |
stra | -0.072 |
Amen | -0.070 |
stop | -0.070 |
Tel | -0.070 |
Token | rather |
Token ID | 2138 |
Feature activation | +6.66e+00 |
Pos logit contributions | |
ithing | +0.723 |
andom | +0.637 |
hing | +0.627 |
nesota | +0.625 |
arning | +0.619 |
Neg logit contributions | |
Stub | -0.686 |
stra | -0.683 |
Amen | -0.668 |
Tel | -0.663 |
stop | -0.657 |
Token | be |
Token ID | 307 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
Stub | +4.150 |
reopen | +4.108 |
Amen | +4.073 |
stra | +4.044 |
stop | +4.040 |
Neg logit contributions | |
ithing | -4.731 |
andom | -4.118 |
nesota | -4.116 |
arning | -4.098 |
hing | -4.007 |
Token | working |
Token ID | 1762 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Stub | +1.835 |
stra | +1.820 |
Amen | +1.793 |
Tel | +1.765 |
reopen | +1.759 |
Neg logit contributions | |
ithing | -1.924 |
andom | -1.679 |
nesota | -1.661 |
arning | -1.646 |
hing | -1.632 |
Token | alongside |
Token ID | 7848 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ithing | +1.049 |
andom | +0.900 |
hing | +0.888 |
nesota | +0.884 |
arning | +0.866 |
Neg logit contributions | |
Stub | -1.457 |
stra | -1.449 |
Amen | -1.421 |
stop | -1.410 |
Tel | -1.410 |
Token | . |
Token ID | 13 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
ithing | +1.259 |
andom | +1.125 |
hing | +1.121 |
nesota | +1.114 |
arning | +1.093 |
Neg logit contributions | |
Stub | -1.018 |
stra | -1.013 |
Tel | -0.984 |
Amen | -0.984 |
stop | -0.976 |
Token |
|
Token ID | 198 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
Stub | +0.413 |
stra | +0.407 |
Amen | +0.404 |
stop | +0.396 |
Tel | +0.395 |
Neg logit contributions | |
ithing | -0.418 |
andom | -0.365 |
nesota | -0.359 |
arning | -0.359 |
hing | -0.357 |
Token | he |
Token ID | 339 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
Stub | +0.536 |
stra | +0.528 |
Amen | +0.521 |
stop | +0.512 |
reopen | +0.512 |
Neg logit contributions | |
ithing | -0.528 |
andom | -0.463 |
nesota | -0.457 |
hing | -0.452 |
arning | -0.451 |
Token | attends |
Token ID | 32743 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
Stub | +1.117 |
stra | +1.109 |
Amen | +1.081 |
reopen | +1.081 |
stop | +1.063 |
Neg logit contributions | |
ithing | -1.226 |
andom | -1.076 |
nesota | -1.059 |
arning | -1.058 |
hing | -1.045 |
Token | a |
Token ID | 257 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
Stub | +1.671 |
stra | +1.638 |
Amen | +1.616 |
stop | +1.593 |
reopen | +1.587 |
Neg logit contributions | |
ithing | -1.809 |
andom | -1.588 |
nesota | -1.566 |
hing | -1.559 |
arning | -1.555 |
Token | scheduled |
Token ID | 7530 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Stub | +0.315 |
stra | +0.314 |
Amen | +0.307 |
stop | +0.304 |
Tel | +0.304 |
Neg logit contributions | |
ithing | -0.346 |
andom | -0.305 |
nesota | -0.301 |
hing | -0.300 |
arning | -0.298 |
Token | formal |
Token ID | 8766 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
Stub | +0.835 |
stra | +0.825 |
Amen | +0.810 |
stop | +0.801 |
reopen | +0.797 |
Neg logit contributions | |
ithing | -1.007 |
andom | -0.891 |
nesota | -0.879 |
hing | -0.876 |
arning | -0.874 |
Token | meeting |
Token ID | 3249 |
Feature activation | +6.63e+00 |
Pos logit contributions | |
Stub | +2.432 |
stra | +2.372 |
Amen | +2.323 |
stop | +2.314 |
reopen | +2.295 |
Neg logit contributions | |
ithing | -2.888 |
nesota | -2.528 |
andom | -2.516 |
arning | -2.496 |
hing | -2.466 |
Token | with |
Token ID | 351 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
Stub | +4.017 |
Amen | +3.896 |
stra | +3.812 |
reopen | +3.781 |
stop | +3.763 |
Neg logit contributions | |
ithing | -4.851 |
andom | -4.372 |
arning | -4.273 |
nesota | -4.257 |
ammy | -4.190 |
Token | his |
Token ID | 465 |
Feature activation | -8.24e-03 |
Pos logit contributions | |
Stub | +1.070 |
stra | +1.037 |
Amen | +1.026 |
stop | +1.009 |
reopen | +1.006 |
Neg logit contributions | |
ithing | -1.195 |
andom | -1.054 |
nesota | -1.043 |
arning | -1.036 |
hing | -1.032 |
Token | boss |
Token ID | 6478 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
ithing | +0.006 |
andom | +0.005 |
nesota | +0.005 |
hing | +0.005 |
arning | +0.005 |
Neg logit contributions | |
Stub | -0.005 |
stra | -0.005 |
Amen | -0.005 |
stop | -0.005 |
reopen | -0.005 |
Token | this |
Token ID | 428 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Stub | +2.029 |
stra | +1.989 |
Amen | +1.956 |
stop | +1.913 |
reopen | +1.908 |
Neg logit contributions | |
ithing | -2.538 |
andom | -2.259 |
nesota | -2.218 |
arning | -2.211 |
hing | -2.183 |
Token | morning |
Token ID | 3329 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
Stub | +0.635 |
stra | +0.629 |
Amen | +0.617 |
stop | +0.610 |
reopen | +0.608 |
Neg logit contributions | |
ithing | -0.726 |
andom | -0.643 |
nesota | -0.634 |
hing | -0.629 |
arning | -0.628 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
ithing | +2.776 |
andom | +2.464 |
nesota | +2.436 |
hing | +2.433 |
arning | +2.410 |
Neg logit contributions | |
stra | -2.190 |
Stub | -2.179 |
Amen | -2.120 |
reopen | -2.116 |
ospace | -2.110 |
Token | United |
Token ID | 1578 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
ithing | +2.634 |
andom | +2.314 |
nesota | +2.292 |
hing | +2.283 |
arning | +2.257 |
Neg logit contributions | |
stra | -2.458 |
Stub | -2.418 |
ospace | -2.380 |
reopen | -2.379 |
stop | -2.369 |
Token | States |
Token ID | 1829 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
ithing | +1.052 |
andom | +0.887 |
nesota | +0.883 |
arning | +0.862 |
hing | +0.853 |
Neg logit contributions | |
stra | -1.601 |
Stub | -1.591 |
reopen | -1.580 |
ospace | -1.563 |
stop | -1.561 |
Token | 5 |
Token ID | 642 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
ithing | +2.928 |
andom | +2.571 |
nesota | +2.540 |
hing | +2.538 |
arning | +2.501 |
Neg logit contributions | |
stra | -2.819 |
Stub | -2.781 |
ospace | -2.739 |
reopen | -2.732 |
Amen | -2.723 |
Token | 667 |
Token ID | 28933 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
ithing | +2.067 |
andom | +1.831 |
nesota | +1.800 |
hing | +1.791 |
arning | +1.779 |
Neg logit contributions | |
stra | -1.985 |
Stub | -1.985 |
Amen | -1.942 |
reopen | -1.921 |
Tel | -1.916 |
Token | Posts |
Token ID | 12043 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
ithing | +1.000 |
andom | +0.751 |
nesota | +0.732 |
arning | +0.725 |
hing | +0.718 |
Neg logit contributions | |
Stub | -2.753 |
stra | -2.710 |
Amen | -2.701 |
Tel | -2.675 |
stop | -2.671 |
Token | # |
Token ID | 1303 |
Feature activation | -6.37e+00 |
Pos logit contributions | |
ithing | +4.682 |
andom | +3.795 |
hing | +3.490 |
arning | +3.426 |
ammy | +3.414 |
Neg logit contributions | |
stra | -8.884 |
Tel | -8.661 |
ospace | -8.620 |
stop | -8.589 |
Stub | -8.450 |
Token | 12 |
Token ID | 1065 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
ithing | +4.317 |
hing | +3.918 |
andom | +3.885 |
nesota | +3.735 |
arning | +3.722 |
Neg logit contributions | |
Stub | -4.085 |
stra | -4.010 |
ospace | -3.982 |
reopen | -3.960 |
Amen | -3.945 |
Token | I |
Token ID | 314 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
ithing | +1.622 |
andom | +1.411 |
hing | +1.398 |
nesota | +1.391 |
arning | +1.374 |
Neg logit contributions | |
stra | -1.728 |
Stub | -1.712 |
Tel | -1.672 |
Amen | -1.671 |
stop | -1.666 |
Token | 'm |
Token ID | 1101 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
ithing | +1.794 |
andom | +1.568 |
hing | +1.550 |
nesota | +1.543 |
arning | +1.533 |
Neg logit contributions | |
Stub | -1.776 |
stra | -1.771 |
Amen | -1.736 |
Tel | -1.719 |
stop | -1.716 |
Token | glad |
Token ID | 9675 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ithing | +0.444 |
andom | +0.390 |
nesota | +0.386 |
hing | +0.383 |
arning | +0.381 |
Neg logit contributions | |
Stub | -0.414 |
stra | -0.412 |
Amen | -0.404 |
stop | -0.399 |
Tel | -0.398 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
ithing | +2.423 |
andom | +2.153 |
nesota | +2.132 |
hing | +2.128 |
arning | +2.107 |
Neg logit contributions | |
stra | -1.916 |
Stub | -1.904 |
Amen | -1.853 |
reopen | -1.847 |
ospace | -1.842 |
Token | United |
Token ID | 1578 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ithing | +2.466 |
andom | +2.165 |
nesota | +2.148 |
hing | +2.136 |
arning | +2.115 |
Neg logit contributions | |
stra | -2.295 |
Stub | -2.269 |
Amen | -2.227 |
reopen | -2.225 |
ospace | -2.220 |
Token | States |
Token ID | 1829 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
ithing | +1.148 |
nesota | +0.970 |
andom | +0.968 |
arning | +0.939 |
hing | +0.928 |
Neg logit contributions | |
stra | -1.716 |
Stub | -1.705 |
reopen | -1.692 |
stop | -1.673 |
Amen | -1.673 |
Token | 21 |
Token ID | 2310 |
Feature activation | -9.83e-01 |
Pos logit contributions | |
ithing | +2.703 |
andom | +2.377 |
nesota | +2.348 |
hing | +2.345 |
arning | +2.313 |
Neg logit contributions | |
stra | -2.557 |
Stub | -2.534 |
Amen | -2.480 |
reopen | -2.473 |
stop | -2.472 |
Token | 50 |
Token ID | 1120 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
ithing | +0.618 |
andom | +0.536 |
nesota | +0.528 |
hing | +0.525 |
arning | +0.522 |
Neg logit contributions | |
Stub | -0.704 |
stra | -0.702 |
Amen | -0.689 |
stop | -0.682 |
Tel | -0.681 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.74e+00 |
Pos logit contributions | |
ithing | +0.365 |
andom | +0.281 |
arning | +0.274 |
nesota | +0.270 |
hing | +0.268 |
Neg logit contributions | |
Stub | -0.902 |
Amen | -0.881 |
stra | -0.876 |
Tel | -0.872 |
stop | -0.871 |
Token | # |
Token ID | 1303 |
Feature activation | -5.94e+00 |
Pos logit contributions | |
ithing | +4.479 |
andom | +3.645 |
hing | +3.409 |
arning | +3.373 |
ammy | +3.351 |
Neg logit contributions | |
stra | -8.202 |
Tel | -7.982 |
ospace | -7.981 |
stop | -7.936 |
Stub | -7.801 |
Token | 16 |
Token ID | 1433 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ithing | +4.033 |
andom | +3.565 |
hing | +3.564 |
arning | +3.492 |
nesota | +3.473 |
Neg logit contributions | |
Stub | -3.885 |
stra | -3.862 |
reopen | -3.769 |
Amen | -3.767 |
Tel | -3.708 |
Token | Glad |
Token ID | 28442 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ithing | +1.198 |
andom | +1.041 |
hing | +1.028 |
nesota | +1.027 |
arning | +1.014 |
Neg logit contributions | |
stra | -1.287 |
Stub | -1.283 |
Amen | -1.253 |
Tel | -1.248 |
stop | -1.245 |
Token | to |
Token ID | 284 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
ithing | +0.861 |
andom | +0.754 |
hing | +0.745 |
nesota | +0.741 |
arning | +0.735 |
Neg logit contributions | |
stra | -0.842 |
Stub | -0.842 |
Amen | -0.823 |
Tel | -0.816 |
stop | -0.813 |
Token | hear |
Token ID | 3285 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
Stub | +1.015 |
stra | +0.999 |
Amen | +0.984 |
reopen | +0.983 |
stop | +0.973 |
Neg logit contributions | |
ithing | -1.454 |
andom | -1.294 |
nesota | -1.290 |
arning | -1.280 |
hing | -1.263 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
ithing | +3.079 |
andom | +2.745 |
nesota | +2.716 |
hing | +2.711 |
arning | +2.672 |
Neg logit contributions | |
stra | -2.498 |
Stub | -2.472 |
reopen | -2.425 |
ospace | -2.418 |
Amen | -2.415 |
Token | United |
Token ID | 1578 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ithing | +2.340 |
andom | +2.068 |
nesota | +2.048 |
hing | +2.037 |
arning | +2.018 |
Neg logit contributions | |
stra | -2.020 |
Stub | -2.009 |
Amen | -1.965 |
stop | -1.952 |
reopen | -1.952 |
Token | States |
Token ID | 1829 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
ithing | +0.871 |
andom | +0.738 |
nesota | +0.733 |
arning | +0.717 |
hing | +0.710 |
Neg logit contributions | |
stra | -1.263 |
Stub | -1.261 |
reopen | -1.244 |
Amen | -1.235 |
stop | -1.231 |
Token | 10 |
Token ID | 838 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ithing | +2.358 |
andom | +2.076 |
nesota | +2.049 |
hing | +2.044 |
arning | +2.022 |
Neg logit contributions | |
stra | -2.192 |
Stub | -2.181 |
Amen | -2.130 |
reopen | -2.120 |
stop | -2.120 |
Token | 53 |
Token ID | 4310 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ithing | +1.088 |
andom | +0.946 |
nesota | +0.934 |
hing | +0.929 |
arning | +0.924 |
Neg logit contributions | |
Stub | -1.156 |
stra | -1.150 |
Amen | -1.128 |
stop | -1.116 |
Tel | -1.116 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.57e+00 |
Pos logit contributions | |
ithing | +0.443 |
andom | +0.338 |
nesota | +0.328 |
arning | +0.324 |
hing | +0.323 |
Neg logit contributions | |
Stub | -1.117 |
Amen | -1.092 |
stra | -1.092 |
Tel | -1.090 |
stop | -1.083 |
Token | # |
Token ID | 1303 |
Feature activation | -5.90e+00 |
Pos logit contributions | |
ithing | +4.414 |
andom | +3.588 |
hing | +3.328 |
arning | +3.320 |
ammy | +3.266 |
Neg logit contributions | |
stra | -8.106 |
ospace | -7.876 |
Tel | -7.867 |
stop | -7.824 |
Stub | -7.710 |
Token | 15 |
Token ID | 1314 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
ithing | +3.831 |
hing | +3.349 |
andom | +3.345 |
arning | +3.289 |
nesota | +3.275 |
Neg logit contributions | |
Stub | -4.069 |
stra | -3.999 |
Amen | -3.956 |
reopen | -3.946 |
ospace | -3.866 |
Token | honestly |
Token ID | 12698 |
Feature activation | -9.78e-01 |
Pos logit contributions | |
ithing | +1.475 |
andom | +1.282 |
hing | +1.269 |
nesota | +1.264 |
arning | +1.247 |
Neg logit contributions | |
stra | -1.591 |
Stub | -1.580 |
Amen | -1.542 |
Tel | -1.542 |
stop | -1.536 |
Token | if |
Token ID | 611 |
Feature activation | -5.23e-01 |
Pos logit contributions | |
ithing | +0.678 |
andom | +0.595 |
nesota | +0.587 |
hing | +0.587 |
arning | +0.581 |
Neg logit contributions | |
Stub | -0.638 |
stra | -0.637 |
Amen | -0.623 |
Tel | -0.616 |
stop | -0.616 |
Token | we |
Token ID | 356 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
ithing | +0.353 |
andom | +0.309 |
nesota | +0.305 |
hing | +0.303 |
arning | +0.302 |
Neg logit contributions | |
Stub | -0.352 |
stra | -0.351 |
Amen | -0.344 |
stop | -0.340 |
Tel | -0.339 |
Token | June |
Token ID | 2795 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
ithing | +3.853 |
andom | +3.275 |
hing | +3.235 |
nesota | +3.167 |
arning | +3.152 |
Neg logit contributions | |
stra | -5.465 |
ospace | -5.456 |
Stub | -5.356 |
stop | -5.327 |
Tel | -5.312 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ithing | +3.727 |
andom | +3.292 |
hing | +3.271 |
nesota | +3.263 |
arning | +3.215 |
Neg logit contributions | |
stra | -3.148 |
ospace | -3.100 |
Stub | -3.097 |
reopen | -3.056 |
stop | -3.021 |
Token | Canada |
Token ID | 3340 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ithing | +2.464 |
andom | +2.168 |
nesota | +2.146 |
hing | +2.135 |
arning | +2.114 |
Neg logit contributions | |
stra | -2.247 |
Stub | -2.214 |
ospace | -2.178 |
reopen | -2.174 |
Amen | -2.169 |
Token | 16 |
Token ID | 1467 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ithing | +1.701 |
andom | +1.502 |
nesota | +1.483 |
hing | +1.476 |
arning | +1.467 |
Neg logit contributions | |
Stub | -1.510 |
stra | -1.505 |
Amen | -1.472 |
stop | -1.457 |
Tel | -1.454 |
Token | 10 |
Token ID | 940 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
ithing | +1.634 |
andom | +1.442 |
nesota | +1.421 |
hing | +1.416 |
arning | +1.407 |
Neg logit contributions | |
Stub | -1.497 |
stra | -1.496 |
Amen | -1.461 |
stop | -1.445 |
reopen | -1.443 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.51e+00 |
Pos logit contributions | |
ithing | +1.283 |
andom | +0.962 |
nesota | +0.933 |
hing | +0.921 |
arning | +0.909 |
Neg logit contributions | |
Stub | -3.847 |
stra | -3.832 |
Amen | -3.784 |
stop | -3.756 |
Tel | -3.754 |
Token | # |
Token ID | 1303 |
Feature activation | -7.30e+00 |
Pos logit contributions | |
ithing | +4.402 |
andom | +3.582 |
hing | +3.322 |
arning | +3.289 |
ammy | +3.250 |
Neg logit contributions | |
stra | -8.087 |
Tel | -7.844 |
ospace | -7.840 |
stop | -7.795 |
Stub | -7.676 |
Token | 14 |
Token ID | 1415 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
ithing | +4.918 |
hing | +4.559 |
andom | +4.360 |
arning | +4.255 |
nesota | +4.243 |
Neg logit contributions | |
Stub | -4.633 |
reopen | -4.537 |
Amen | -4.469 |
ospace | -4.441 |
stra | -4.432 |
Token | On |
Token ID | 1550 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
ithing | +1.761 |
andom | +1.526 |
hing | +1.515 |
nesota | +1.508 |
arning | +1.483 |
Neg logit contributions | |
stra | -1.910 |
Stub | -1.890 |
Tel | -1.849 |
Amen | -1.845 |
stop | -1.841 |
Token | September |
Token ID | 2693 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
ithing | +2.304 |
andom | +1.995 |
nesota | +1.965 |
hing | +1.962 |
arning | +1.943 |
Neg logit contributions | |
Stub | -2.581 |
stra | -2.574 |
Amen | -2.524 |
Tel | -2.498 |
stop | -2.498 |
Token | 20 |
Token ID | 1160 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ithing | +1.263 |
andom | +1.076 |
hing | +1.065 |
nesota | +1.064 |
arning | +1.034 |
Neg logit contributions | |
stra | -1.807 |
Stub | -1.785 |
stop | -1.753 |
Tel | -1.752 |
reopen | -1.740 |
Token | 2010 |
Token ID | 3050 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ithing | +2.279 |
andom | +2.026 |
nesota | +2.007 |
hing | +1.999 |
arning | +1.984 |
Neg logit contributions | |
stra | -1.818 |
Stub | -1.808 |
Amen | -1.759 |
reopen | -1.756 |
stop | -1.748 |
Token | United |
Token ID | 1578 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ithing | +1.713 |
andom | +1.515 |
nesota | +1.498 |
hing | +1.492 |
arning | +1.479 |
Neg logit contributions | |
stra | -1.469 |
Stub | -1.465 |
Amen | -1.431 |
stop | -1.419 |
reopen | -1.418 |
Token | States |
Token ID | 1829 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
ithing | +0.712 |
andom | +0.611 |
nesota | +0.602 |
hing | +0.597 |
arning | +0.592 |
Neg logit contributions | |
Stub | -0.927 |
stra | -0.925 |
Amen | -0.908 |
reopen | -0.900 |
stop | -0.900 |
Token | 6 |
Token ID | 718 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
ithing | +2.081 |
andom | +1.832 |
nesota | +1.810 |
hing | +1.804 |
arning | +1.785 |
Neg logit contributions | |
stra | -1.931 |
Stub | -1.925 |
Amen | -1.879 |
reopen | -1.867 |
stop | -1.867 |
Token | 83 |
Token ID | 5999 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
ithing | +1.848 |
andom | +1.631 |
nesota | +1.603 |
hing | +1.599 |
arning | +1.587 |
Neg logit contributions | |
Stub | -1.768 |
stra | -1.766 |
Amen | -1.729 |
reopen | -1.709 |
stop | -1.708 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.48e+00 |
Pos logit contributions | |
ithing | +0.243 |
andom | +0.184 |
nesota | +0.180 |
arning | +0.178 |
hing | +0.176 |
Neg logit contributions | |
Stub | -0.623 |
stra | -0.610 |
Amen | -0.609 |
Tel | -0.606 |
stop | -0.601 |
Token | Last |
Token ID | 4586 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
ithing | +4.401 |
andom | +3.609 |
hing | +3.324 |
arning | +3.311 |
ammy | +3.301 |
Neg logit contributions | |
stra | -7.973 |
ospace | -7.748 |
Tel | -7.745 |
stop | -7.688 |
Stub | -7.572 |
Token | Edited |
Token ID | 34212 |
Feature activation | -5.57e+00 |
Pos logit contributions | |
ithing | +0.999 |
andom | +0.871 |
nesota | +0.861 |
hing | +0.855 |
arning | +0.852 |
Neg logit contributions | |
Stub | -1.033 |
stra | -1.028 |
Amen | -1.009 |
stop | -0.998 |
Tel | -0.996 |
Token | : |
Token ID | 25 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
ithing | +4.121 |
andom | +3.731 |
hing | +3.694 |
nesota | +3.638 |
arning | +3.608 |
Neg logit contributions | |
stra | -3.210 |
Stub | -3.200 |
Amen | -3.110 |
ospace | -3.098 |
reopen | -3.096 |
Token | 2013 |
Token ID | 2211 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ithing | +2.241 |
andom | +2.022 |
hing | +2.002 |
nesota | +1.977 |
arning | +1.972 |
Neg logit contributions | |
stra | -1.497 |
Stub | -1.493 |
reopen | -1.451 |
Amen | -1.444 |
Tel | -1.436 |
Token | - |
Token ID | 12 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
Stub | +0.744 |
stra | +0.718 |
stop | +0.717 |
Amen | +0.710 |
Tel | +0.709 |
Neg logit contributions | |
ithing | -1.275 |
nesota | -1.141 |
andom | -1.137 |
arning | -1.136 |
ammy | -1.127 |
Token | 2011 |
Token ID | 2813 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
ithing | +2.253 |
andom | +2.007 |
nesota | +1.986 |
hing | +1.978 |
arning | +1.963 |
Neg logit contributions | |
stra | -1.733 |
Stub | -1.727 |
Amen | -1.680 |
reopen | -1.669 |
stop | -1.668 |
Token | United |
Token ID | 1578 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
ithing | +2.241 |
andom | +1.977 |
nesota | +1.956 |
hing | +1.952 |
arning | +1.930 |
Neg logit contributions | |
stra | -1.968 |
Stub | -1.955 |
Amen | -1.910 |
reopen | -1.903 |
stop | -1.901 |
Token | States |
Token ID | 1829 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
ithing | +0.858 |
andom | +0.727 |
nesota | +0.720 |
arning | +0.704 |
hing | +0.698 |
Neg logit contributions | |
stra | -1.232 |
Stub | -1.228 |
reopen | -1.212 |
stop | -1.201 |
Amen | -1.200 |
Token | 228 |
Token ID | 29041 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
ithing | +2.477 |
andom | +2.178 |
nesota | +2.152 |
hing | +2.147 |
arning | +2.121 |
Neg logit contributions | |
stra | -2.331 |
Stub | -2.313 |
Amen | -2.260 |
reopen | -2.254 |
stop | -2.252 |
Token | 65 |
Token ID | 2996 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
ithing | +0.260 |
andom | +0.224 |
nesota | +0.223 |
hing | +0.222 |
arning | +0.220 |
Neg logit contributions | |
Stub | -0.277 |
stra | -0.272 |
Amen | -0.269 |
Tel | -0.266 |
stop | -0.265 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.39e+00 |
Pos logit contributions | |
ithing | +0.178 |
andom | +0.137 |
nesota | +0.135 |
arning | +0.134 |
ammy | +0.132 |
Neg logit contributions | |
Stub | -0.423 |
Amen | -0.411 |
stra | -0.408 |
Tel | -0.408 |
stop | -0.406 |
Token | # |
Token ID | 1303 |
Feature activation | -6.51e+00 |
Pos logit contributions | |
ithing | +4.372 |
andom | +3.574 |
hing | +3.315 |
arning | +3.284 |
ammy | +3.250 |
Neg logit contributions | |
stra | -7.940 |
Tel | -7.754 |
stop | -7.710 |
ospace | -7.704 |
Stub | -7.572 |
Token | 13 |
Token ID | 1485 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ithing | +4.464 |
hing | +4.007 |
andom | +3.927 |
evin | +3.828 |
nesota | +3.817 |
Neg logit contributions | |
Stub | -4.119 |
Amen | -3.993 |
reopen | -3.981 |
stra | -3.978 |
ospace | -3.966 |
Token | On |
Token ID | 1550 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
ithing | +1.728 |
andom | +1.505 |
hing | +1.488 |
nesota | +1.482 |
arning | +1.460 |
Neg logit contributions | |
stra | -1.848 |
Stub | -1.832 |
Amen | -1.788 |
Tel | -1.787 |
stop | -1.781 |
Token | September |
Token ID | 2693 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ithing | +1.975 |
andom | +1.716 |
nesota | +1.692 |
hing | +1.684 |
arning | +1.672 |
Neg logit contributions | |
Stub | -2.162 |
stra | -2.154 |
Amen | -2.113 |
stop | -2.091 |
Tel | -2.089 |
Token | 20 |
Token ID | 1160 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
ithing | +1.071 |
andom | +0.916 |
hing | +0.904 |
nesota | +0.901 |
arning | +0.883 |
Neg logit contributions | |
stra | -1.423 |
Stub | -1.409 |
stop | -1.381 |
Tel | -1.379 |
Amen | -1.374 |
Token | June |
Token ID | 2795 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
ithing | +3.744 |
andom | +3.178 |
hing | +3.131 |
nesota | +3.085 |
arning | +3.063 |
Neg logit contributions | |
stra | -5.156 |
ospace | -5.131 |
Stub | -5.055 |
Tel | -5.017 |
stop | -5.014 |
Token | 2009 |
Token ID | 3717 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ithing | +3.338 |
andom | +2.953 |
hing | +2.929 |
nesota | +2.923 |
arning | +2.888 |
Neg logit contributions | |
stra | -2.755 |
Stub | -2.727 |
ospace | -2.679 |
reopen | -2.663 |
Amen | -2.650 |
Token | Germany |
Token ID | 4486 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ithing | +1.514 |
andom | +1.332 |
nesota | +1.315 |
hing | +1.309 |
arning | +1.299 |
Neg logit contributions | |
Stub | -1.417 |
stra | -1.416 |
Amen | -1.383 |
stop | -1.370 |
reopen | -1.368 |
Token | 247 |
Token ID | 30179 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
ithing | +1.279 |
andom | +1.129 |
nesota | +1.116 |
hing | +1.110 |
arning | +1.105 |
Neg logit contributions | |
Stub | -1.123 |
stra | -1.115 |
Amen | -1.094 |
stop | -1.079 |
Tel | -1.078 |
Token | 34 |
Token ID | 2682 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
Stub | +0.202 |
stra | +0.199 |
Amen | +0.197 |
Tel | +0.196 |
stop | +0.195 |
Neg logit contributions | |
ithing | -0.128 |
andom | -0.106 |
nesota | -0.105 |
hing | -0.104 |
arning | -0.103 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.29e+00 |
Pos logit contributions | |
ithing | +0.529 |
andom | +0.403 |
nesota | +0.395 |
arning | +0.393 |
hing | +0.387 |
Neg logit contributions | |
Stub | -1.351 |
Amen | -1.320 |
stra | -1.319 |
Tel | -1.313 |
stop | -1.310 |
Token | # |
Token ID | 1303 |
Feature activation | -5.80e+00 |
Pos logit contributions | |
ithing | +4.343 |
andom | +3.556 |
arning | +3.314 |
hing | +3.302 |
ammy | +3.249 |
Neg logit contributions | |
stra | -7.835 |
Tel | -7.599 |
ospace | -7.575 |
stop | -7.537 |
Stub | -7.465 |
Token | 20 |
Token ID | 1238 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ithing | +2.961 |
andom | +2.495 |
hing | +2.471 |
nesota | +2.428 |
arning | +2.413 |
Neg logit contributions | |
Stub | -4.794 |
stra | -4.789 |
Amen | -4.703 |
reopen | -4.676 |
stop | -4.643 |
Token | Nice |
Token ID | 18460 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ithing | +0.702 |
andom | +0.613 |
nesota | +0.605 |
hing | +0.603 |
arning | +0.597 |
Neg logit contributions | |
Stub | -0.721 |
stra | -0.720 |
Amen | -0.704 |
stop | -0.698 |
Tel | -0.697 |
Token | , |
Token ID | 11 |
Feature activation | +6.12e-01 |
Pos logit contributions | |
Stub | +0.746 |
stra | +0.741 |
Amen | +0.726 |
stop | +0.718 |
reopen | +0.716 |
Neg logit contributions | |
ithing | -0.852 |
andom | -0.752 |
nesota | -0.743 |
hing | -0.738 |
arning | -0.736 |
Token | happy |
Token ID | 3772 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
Stub | +0.401 |
stra | +0.397 |
Amen | +0.390 |
reopen | +0.386 |
stop | +0.386 |
Neg logit contributions | |
ithing | -0.425 |
andom | -0.375 |
nesota | -0.370 |
arning | -0.366 |
hing | -0.365 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
ithing | +3.285 |
andom | +2.922 |
hing | +2.885 |
nesota | +2.882 |
arning | +2.858 |
Neg logit contributions | |
stra | -2.602 |
Stub | -2.589 |
Amen | -2.518 |
reopen | -2.515 |
Tel | -2.502 |
Token | United |
Token ID | 1578 |
Feature activation | -8.45e-01 |
Pos logit contributions | |
ithing | +2.278 |
andom | +2.012 |
nesota | +1.989 |
hing | +1.982 |
arning | +1.962 |
Neg logit contributions | |
stra | -2.004 |
Stub | -1.984 |
Amen | -1.941 |
reopen | -1.934 |
stop | -1.933 |
Token | States |
Token ID | 1829 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
ithing | +0.477 |
andom | +0.406 |
nesota | +0.399 |
hing | +0.395 |
arning | +0.394 |
Neg logit contributions | |
stra | -0.658 |
Stub | -0.657 |
Amen | -0.645 |
reopen | -0.642 |
stop | -0.639 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
ithing | +2.485 |
andom | +2.190 |
nesota | +2.157 |
hing | +2.151 |
arning | +2.128 |
Neg logit contributions | |
stra | -2.308 |
Stub | -2.291 |
Amen | -2.239 |
reopen | -2.232 |
stop | -2.232 |
Token | 240 |
Token ID | 16102 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
ithing | +1.070 |
andom | +0.907 |
nesota | +0.890 |
hing | +0.885 |
arning | +0.880 |
Neg logit contributions | |
Stub | -1.563 |
stra | -1.561 |
Amen | -1.534 |
stop | -1.518 |
Tel | -1.517 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.24e+00 |
Pos logit contributions | |
ithing | +0.116 |
andom | +0.092 |
nesota | +0.090 |
hing | +0.089 |
arning | +0.087 |
Neg logit contributions | |
Stub | -0.285 |
stra | -0.276 |
Amen | -0.276 |
stop | -0.274 |
Tel | -0.274 |
Token | # |
Token ID | 1303 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
ithing | +4.403 |
andom | +3.584 |
hing | +3.479 |
arning | +3.328 |
ammy | +3.274 |
Neg logit contributions | |
stra | -7.768 |
Tel | -7.562 |
ospace | -7.535 |
stop | -7.499 |
Stub | -7.438 |
Token | 10 |
Token ID | 940 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
ithing | +3.723 |
andom | +3.294 |
hing | +3.284 |
arning | +3.209 |
nesota | +3.184 |
Neg logit contributions | |
stra | -3.976 |
Stub | -3.971 |
reopen | -3.899 |
Amen | -3.865 |
ospace | -3.845 |
Token | Happy |
Token ID | 14628 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
ithing | +1.796 |
andom | +1.565 |
hing | +1.550 |
nesota | +1.541 |
arning | +1.518 |
Neg logit contributions | |
stra | -1.918 |
Stub | -1.903 |
Amen | -1.861 |
Tel | -1.857 |
stop | -1.848 |
Token | it |
Token ID | 340 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
Stub | +0.218 |
stra | +0.217 |
Amen | +0.212 |
stop | +0.210 |
Tel | +0.210 |
Neg logit contributions | |
ithing | -0.232 |
andom | -0.204 |
nesota | -0.201 |
hing | -0.200 |
arning | -0.199 |
Token | worked |
Token ID | 3111 |
Feature activation | -9.68e-01 |
Pos logit contributions | |
ithing | +0.368 |
andom | +0.326 |
hing | +0.323 |
nesota | +0.321 |
arning | +0.319 |
Neg logit contributions | |
Stub | -0.300 |
stra | -0.298 |
Amen | -0.293 |
Tel | -0.289 |
stop | -0.288 |
Token | 2002 |
Token ID | 6244 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
ithing | +2.799 |
andom | +2.496 |
nesota | +2.469 |
hing | +2.462 |
arning | +2.441 |
Neg logit contributions | |
stra | -2.135 |
Stub | -2.128 |
Amen | -2.078 |
reopen | -2.052 |
Tel | -2.051 |
Token | United |
Token ID | 1578 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
ithing | +1.846 |
andom | +1.630 |
nesota | +1.609 |
hing | +1.602 |
arning | +1.591 |
Neg logit contributions | |
stra | -1.644 |
Stub | -1.636 |
Amen | -1.599 |
stop | -1.586 |
reopen | -1.584 |
Token | States |
Token ID | 1829 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
ithing | +1.115 |
andom | +0.942 |
nesota | +0.932 |
arning | +0.912 |
hing | +0.905 |
Neg logit contributions | |
stra | -1.680 |
Stub | -1.671 |
reopen | -1.654 |
Amen | -1.636 |
ospace | -1.634 |
Token | 3 |
Token ID | 513 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ithing | +2.885 |
andom | +2.537 |
nesota | +2.500 |
hing | +2.499 |
arning | +2.465 |
Neg logit contributions | |
stra | -2.747 |
Stub | -2.714 |
Amen | -2.658 |
ospace | -2.655 |
reopen | -2.653 |
Token | 472 |
Token ID | 37856 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
ithing | +1.373 |
andom | +1.220 |
nesota | +1.205 |
hing | +1.199 |
arning | +1.192 |
Neg logit contributions | |
Stub | -1.121 |
stra | -1.118 |
Amen | -1.092 |
stop | -1.079 |
Tel | -1.077 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.12e+00 |
Pos logit contributions | |
ithing | +0.141 |
andom | +0.110 |
arning | +0.107 |
nesota | +0.107 |
ammy | +0.106 |
Neg logit contributions | |
Stub | -0.328 |
Amen | -0.318 |
Tel | -0.316 |
stra | -0.314 |
stop | -0.314 |
Token | # |
Token ID | 1303 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
ithing | +4.303 |
andom | +3.524 |
hing | +3.342 |
arning | +3.253 |
ammy | +3.238 |
Neg logit contributions | |
stra | -7.703 |
Tel | -7.519 |
ospace | -7.473 |
stop | -7.435 |
erion | -7.311 |
Token | 11 |
Token ID | 1157 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
ithing | +3.662 |
andom | +3.202 |
hing | +3.198 |
arning | +3.117 |
nesota | +3.057 |
Neg logit contributions | |
stra | -3.679 |
Stub | -3.674 |
reopen | -3.566 |
Amen | -3.545 |
ospace | -3.511 |
Token | On |
Token ID | 1550 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
ithing | +1.878 |
andom | +1.637 |
hing | +1.621 |
nesota | +1.609 |
arning | +1.591 |
Neg logit contributions | |
stra | -1.965 |
Stub | -1.946 |
Tel | -1.903 |
Amen | -1.901 |
stop | -1.897 |
Token | September |
Token ID | 2693 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ithing | +1.232 |
andom | +1.067 |
nesota | +1.053 |
hing | +1.049 |
arning | +1.040 |
Neg logit contributions | |
Stub | -1.344 |
stra | -1.342 |
Amen | -1.317 |
stop | -1.301 |
Tel | -1.300 |
Token | 20 |
Token ID | 1160 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
ithing | +1.216 |
andom | +1.084 |
nesota | +1.076 |
hing | +1.067 |
arning | +1.066 |
Neg logit contributions | |
Stub | -0.877 |
stra | -0.869 |
Amen | -0.852 |
stop | -0.839 |
Tel | -0.837 |
Token | 2011 |
Token ID | 2813 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
ithing | +2.968 |
andom | +2.627 |
nesota | +2.604 |
hing | +2.601 |
arning | +2.575 |
Neg logit contributions | |
stra | -2.419 |
Stub | -2.406 |
ospace | -2.337 |
reopen | -2.335 |
Amen | -2.333 |
Token | United |
Token ID | 1578 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
ithing | +2.771 |
andom | +2.440 |
nesota | +2.418 |
hing | +2.415 |
arning | +2.383 |
Neg logit contributions | |
stra | -2.465 |
Stub | -2.436 |
ospace | -2.396 |
reopen | -2.391 |
Amen | -2.385 |
Token | States |
Token ID | 1829 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
ithing | +1.173 |
andom | +0.993 |
nesota | +0.991 |
arning | +0.965 |
hing | +0.951 |
Neg logit contributions | |
stra | -1.718 |
Stub | -1.717 |
reopen | -1.698 |
ospace | -1.683 |
stop | -1.679 |
Token | 6 |
Token ID | 718 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
ithing | +2.862 |
andom | +2.513 |
nesota | +2.485 |
hing | +2.481 |
arning | +2.448 |
Neg logit contributions | |
stra | -2.733 |
Stub | -2.707 |
ospace | -2.652 |
reopen | -2.649 |
Amen | -2.647 |
Token | 12 |
Token ID | 1065 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
ithing | +2.028 |
andom | +1.799 |
nesota | +1.770 |
hing | +1.764 |
arning | +1.751 |
Neg logit contributions | |
Stub | -1.848 |
stra | -1.846 |
Amen | -1.805 |
reopen | -1.785 |
stop | -1.784 |
Token | Posts |
Token ID | 12043 |
Feature activation | -8.90e+00 |
Pos logit contributions | |
ithing | +1.183 |
andom | +0.922 |
nesota | +0.898 |
hing | +0.889 |
arning | +0.878 |
Neg logit contributions | |
Stub | -3.003 |
stra | -2.992 |
Amen | -2.952 |
stop | -2.929 |
Tel | -2.928 |
Token | # |
Token ID | 1303 |
Feature activation | -5.83e+00 |
Pos logit contributions | |
ithing | +4.180 |
andom | +3.425 |
hing | +3.191 |
arning | +3.166 |
ammy | +3.127 |
Neg logit contributions | |
stra | -7.487 |
ospace | -7.314 |
Tel | -7.299 |
stop | -7.241 |
Stub | -7.146 |
Token | 17 |
Token ID | 1558 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ithing | +3.628 |
andom | +3.179 |
hing | +3.171 |
arning | +3.126 |
nesota | +3.088 |
Neg logit contributions | |
Stub | -4.144 |
stra | -4.058 |
Amen | -3.996 |
reopen | -3.971 |
ospace | -3.924 |
Token | Good |
Token ID | 4599 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
ithing | +1.169 |
andom | +1.018 |
hing | +1.005 |
nesota | +1.004 |
arning | +0.992 |
Neg logit contributions | |
stra | -1.224 |
Stub | -1.220 |
Amen | -1.190 |
Tel | -1.185 |
stop | -1.184 |
Token | to |
Token ID | 284 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
Stub | +0.342 |
stra | +0.340 |
Amen | +0.333 |
stop | +0.329 |
Tel | +0.328 |
Neg logit contributions | |
ithing | -0.436 |
andom | -0.388 |
nesota | -0.383 |
hing | -0.382 |
arning | -0.380 |
Token | hear |
Token ID | 3285 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
Stub | +0.941 |
stra | +0.924 |
Amen | +0.913 |
reopen | +0.911 |
stop | +0.900 |
Neg logit contributions | |
ithing | -1.299 |
andom | -1.158 |
nesota | -1.151 |
arning | -1.144 |
hing | -1.125 |
Token |
|
Token ID | 198 |
Feature activation | -9.68e-01 |
Pos logit contributions | |
ithing | +2.312 |
andom | +2.024 |
nesota | +1.998 |
hing | +1.989 |
arning | +1.975 |
Neg logit contributions | |
Stub | -2.308 |
stra | -2.298 |
Amen | -2.252 |
stop | -2.227 |
Tel | -2.225 |
Token |
|
Token ID | 198 |
Feature activation | -7.53e-01 |
Pos logit contributions | |
ithing | +0.709 |
andom | +0.625 |
nesota | +0.618 |
arning | +0.612 |
ammy | +0.604 |
Neg logit contributions | |
Stub | -0.611 |
stra | -0.610 |
stop | -0.594 |
Amen | -0.594 |
Tel | -0.590 |
Token | It |
Token ID | 1026 |
Feature activation | -7.60e-01 |
Pos logit contributions | |
ithing | +0.551 |
andom | +0.487 |
hing | +0.481 |
nesota | +0.481 |
arning | +0.477 |
Neg logit contributions | |
Stub | -0.463 |
stra | -0.459 |
Amen | -0.451 |
reopen | -0.445 |
stop | -0.442 |
Token | 's |
Token ID | 338 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
ithing | +0.495 |
andom | +0.432 |
nesota | +0.425 |
hing | +0.423 |
arning | +0.420 |
Neg logit contributions | |
Stub | -0.530 |
stra | -0.528 |
Amen | -0.517 |
stop | -0.511 |
Tel | -0.511 |
Token | not |
Token ID | 407 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
Stub | +0.022 |
stra | +0.021 |
Amen | +0.021 |
stop | +0.021 |
reopen | +0.021 |
Neg logit contributions | |
ithing | -0.022 |
andom | -0.019 |
nesota | -0.019 |
hing | -0.019 |
arning | -0.019 |
Token | clear |
Token ID | 1598 |
Feature activation | -8.16e+00 |
Pos logit contributions | |
ithing | +0.067 |
andom | +0.058 |
nesota | +0.057 |
hing | +0.057 |
arning | +0.056 |
Neg logit contributions | |
Stub | -0.076 |
stra | -0.076 |
Amen | -0.074 |
stop | -0.073 |
Tel | -0.073 |
Token | why |
Token ID | 1521 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ithing | +4.509 |
hing | +4.099 |
andom | +3.760 |
arning | +3.755 |
nesota | +3.721 |
Neg logit contributions | |
stra | -5.933 |
ospace | -5.800 |
erion | -5.772 |
Tel | -5.772 |
Stub | -5.759 |
Token | Mr |
Token ID | 1770 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ithing | +2.365 |
andom | +2.044 |
hing | +2.043 |
nesota | +2.018 |
arning | +2.003 |
Neg logit contributions | |
stra | -2.343 |
Stub | -2.328 |
Amen | -2.295 |
stop | -2.266 |
Tel | -2.265 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
Stub | +0.700 |
stra | +0.684 |
Amen | +0.671 |
Tel | +0.669 |
stop | +0.667 |
Neg logit contributions | |
ithing | -1.093 |
andom | -0.972 |
nesota | -0.963 |
arning | -0.959 |
ammy | -0.952 |
Token | Mor |
Token ID | 3461 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
ithing | +0.097 |
andom | +0.086 |
hing | +0.085 |
nesota | +0.084 |
arning | +0.084 |
Neg logit contributions | |
stra | -0.078 |
Stub | -0.077 |
Amen | -0.075 |
reopen | -0.075 |
Tel | -0.075 |
Token | neau |
Token ID | 46533 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ithing | +0.436 |
andom | +0.375 |
nesota | +0.372 |
hing | +0.367 |
arning | +0.365 |
Neg logit contributions | |
stra | -0.524 |
Stub | -0.518 |
stop | -0.512 |
Amen | -0.510 |
Tel | -0.506 |
Token | TL |
Token ID | 24811 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
ithing | +1.881 |
andom | +1.655 |
nesota | +1.640 |
hing | +1.631 |
arning | +1.617 |
Neg logit contributions | |
stra | -1.719 |
Stub | -1.709 |
Amen | -1.671 |
reopen | -1.665 |
stop | -1.662 |
Token | AD |
Token ID | 2885 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ithing | +2.075 |
andom | +1.832 |
nesota | +1.802 |
hing | +1.801 |
arning | +1.795 |
Neg logit contributions | |
stra | -1.801 |
Stub | -1.772 |
Amen | -1.764 |
reopen | -1.738 |
ospace | -1.723 |
Token | T |
Token ID | 51 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
ithing | +0.953 |
andom | +0.824 |
nesota | +0.823 |
hing | +0.807 |
arning | +0.806 |
Neg logit contributions | |
Stub | -1.025 |
stra | -1.017 |
Amen | -0.997 |
stop | -0.996 |
Tel | -0.991 |
Token | 235 |
Token ID | 28878 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
ithing | +1.135 |
andom | +1.003 |
nesota | +0.991 |
hing | +0.986 |
arning | +0.980 |
Neg logit contributions | |
Stub | -0.988 |
stra | -0.983 |
Amen | -0.962 |
stop | -0.951 |
Tel | -0.950 |
Token | 21 |
Token ID | 2481 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
ithing | +0.597 |
andom | +0.517 |
nesota | +0.510 |
hing | +0.507 |
arning | +0.504 |
Neg logit contributions | |
Stub | -0.670 |
stra | -0.666 |
Amen | -0.654 |
stop | -0.647 |
Tel | -0.647 |
Token | Posts |
Token ID | 12043 |
Feature activation | -8.13e+00 |
Pos logit contributions | |
ithing | +0.054 |
andom | +0.042 |
nesota | +0.041 |
arning | +0.040 |
hing | +0.040 |
Neg logit contributions | |
Stub | -0.135 |
Amen | -0.132 |
stra | -0.131 |
Tel | -0.131 |
stop | -0.130 |
Token | # |
Token ID | 1303 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
ithing | +3.897 |
andom | +3.219 |
arning | +3.012 |
hing | +3.005 |
ammy | +2.941 |
Neg logit contributions | |
stra | -6.796 |
Tel | -6.644 |
ospace | -6.596 |
stop | -6.589 |
Stub | -6.515 |
Token | 18 |
Token ID | 1507 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
ithing | +2.636 |
andom | +2.187 |
hing | +2.151 |
nesota | +2.130 |
arning | +2.109 |
Neg logit contributions | |
Stub | -4.902 |
stra | -4.873 |
Amen | -4.801 |
reopen | -4.768 |
Tel | -4.749 |
Token | Good |
Token ID | 4599 |
Feature activation | +8.00e-02 |
Pos logit contributions | |
ithing | +0.914 |
andom | +0.797 |
nesota | +0.787 |
hing | +0.785 |
arning | +0.777 |
Neg logit contributions | |
stra | -0.949 |
Stub | -0.948 |
Amen | -0.925 |
stop | -0.919 |
Tel | -0.919 |
Token | stuff |
Token ID | 3404 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Stub | +0.047 |
stra | +0.047 |
Amen | +0.045 |
Tel | +0.045 |
stop | +0.045 |
Neg logit contributions | |
ithing | -0.061 |
andom | -0.054 |
nesota | -0.054 |
hing | -0.053 |
arning | -0.053 |
Token | ! |
Token ID | 0 |
Feature activation | -5.73e-01 |
Pos logit contributions | |
Stub | +0.706 |
stra | +0.701 |
Amen | +0.687 |
stop | +0.678 |
Tel | +0.676 |
Neg logit contributions | |
ithing | -0.835 |
andom | -0.739 |
nesota | -0.730 |
hing | -0.725 |
arning | -0.723 |
Token |
|
Token ID | 198 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
ithing | +0.077 |
andom | +0.067 |
hing | +0.066 |
nesota | +0.066 |
arning | +0.065 |
Neg logit contributions | |
Stub | -0.076 |
stra | -0.076 |
stop | -0.074 |
Tel | -0.074 |
reopen | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
ithing | +0.167 |
andom | +0.148 |
nesota | +0.146 |
arning | +0.145 |
ammy | +0.144 |
Neg logit contributions | |
stra | -0.133 |
Stub | -0.133 |
stop | -0.130 |
Amen | -0.129 |
Tel | -0.129 |
Token | It |
Token ID | 1026 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ithing | +0.026 |
andom | +0.023 |
nesota | +0.023 |
hing | +0.023 |
arning | +0.023 |
Neg logit contributions | |
Stub | -0.021 |
stra | -0.021 |
Amen | -0.021 |
stop | -0.020 |
reopen | -0.020 |
Token | 's |
Token ID | 338 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
ithing | +0.916 |
andom | +0.808 |
nesota | +0.800 |
hing | +0.796 |
arning | +0.791 |
Neg logit contributions | |
Stub | -0.810 |
stra | -0.807 |
Amen | -0.790 |
stop | -0.781 |
Tel | -0.781 |
Token | not |
Token ID | 407 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
ithing | +0.129 |
andom | +0.112 |
nesota | +0.111 |
hing | +0.110 |
arning | +0.110 |
Neg logit contributions | |
Stub | -0.131 |
stra | -0.129 |
Amen | -0.127 |
reopen | -0.126 |
stop | -0.126 |
Token | clear |
Token ID | 1598 |
Feature activation | -8.10e+00 |
Pos logit contributions | |
ithing | +0.184 |
andom | +0.160 |
nesota | +0.159 |
hing | +0.157 |
arning | +0.157 |
Neg logit contributions | |
Stub | -0.194 |
stra | -0.193 |
Amen | -0.190 |
stop | -0.188 |
Tel | -0.188 |
Token | what |
Token ID | 644 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
ithing | +4.510 |
hing | +4.157 |
andom | +3.921 |
ammy | +3.830 |
arning | +3.824 |
Neg logit contributions | |
stra | -5.769 |
Tel | -5.671 |
erion | -5.650 |
Stub | -5.645 |
ospace | -5.576 |
Token | long |
Token ID | 890 |
Feature activation | -7.63e-01 |
Pos logit contributions | |
ithing | +1.956 |
andom | +1.735 |
nesota | +1.714 |
hing | +1.711 |
arning | +1.677 |
Neg logit contributions | |
stra | -1.934 |
Stub | -1.923 |
Amen | -1.896 |
Tel | -1.885 |
stop | -1.875 |
Token | - |
Token ID | 12 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ithing | +0.566 |
andom | +0.502 |
nesota | +0.499 |
hing | +0.498 |
arning | +0.491 |
Neg logit contributions | |
Stub | -0.459 |
stra | -0.455 |
Amen | -0.447 |
Tel | -0.441 |
reopen | -0.439 |
Token | term |
Token ID | 4354 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
Stub | +0.934 |
stra | +0.933 |
Amen | +0.913 |
stop | +0.907 |
Tel | +0.905 |
Neg logit contributions | |
ithing | -0.803 |
andom | -0.694 |
nesota | -0.685 |
hing | -0.678 |
arning | -0.676 |
Token | good |
Token ID | 922 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
Stub | +0.252 |
stra | +0.250 |
Amen | +0.247 |
reopen | +0.243 |
stop | +0.243 |
Neg logit contributions | |
ithing | -0.244 |
andom | -0.212 |
nesota | -0.209 |
arning | -0.208 |
hing | -0.207 |
Token |
|
Token ID | 198 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Stub | +0.404 |
stra | +0.403 |
Amen | +0.394 |
stop | +0.391 |
reopen | +0.391 |
Neg logit contributions | |
ithing | -0.428 |
andom | -0.377 |
nesota | -0.371 |
hing | -0.371 |
arning | -0.366 |
Token |
|
Token ID | 198 |
Feature activation | +8.70e-01 |
Pos logit contributions | |
Stub | +0.084 |
stra | +0.084 |
stop | +0.082 |
Amen | +0.082 |
Tel | +0.082 |
Neg logit contributions | |
ithing | -0.106 |
andom | -0.094 |
nesota | -0.093 |
arning | -0.092 |
ammy | -0.091 |
Token | It |
Token ID | 1026 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Stub | +0.509 |
stra | +0.508 |
Amen | +0.495 |
Tel | +0.493 |
stop | +0.491 |
Neg logit contributions | |
ithing | -0.663 |
andom | -0.589 |
nesota | -0.584 |
hing | -0.579 |
arning | -0.577 |
Token | is |
Token ID | 318 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
ithing | +0.851 |
andom | +0.735 |
nesota | +0.725 |
hing | +0.723 |
arning | +0.716 |
Neg logit contributions | |
Stub | -1.002 |
stra | -0.996 |
Amen | -0.980 |
Tel | -0.970 |
stop | -0.969 |
Token | not |
Token ID | 407 |
Feature activation | +9.57e-02 |
Pos logit contributions | |
Stub | +0.090 |
stra | +0.089 |
Amen | +0.087 |
Tel | +0.086 |
stop | +0.086 |
Neg logit contributions | |
ithing | -0.085 |
andom | -0.074 |
nesota | -0.073 |
hing | -0.072 |
arning | -0.072 |
Token | clear |
Token ID | 1598 |
Feature activation | -7.76e+00 |
Pos logit contributions | |
Stub | +0.061 |
stra | +0.061 |
Amen | +0.059 |
Tel | +0.059 |
stop | +0.059 |
Neg logit contributions | |
ithing | -0.068 |
andom | -0.060 |
nesota | -0.059 |
hing | -0.059 |
arning | -0.059 |
Token | yet |
Token ID | 1865 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
ithing | +4.596 |
hing | +4.130 |
andom | +3.919 |
ammy | +3.824 |
arning | +3.821 |
Neg logit contributions | |
stra | -5.288 |
Stub | -5.282 |
erion | -5.201 |
ospace | -5.172 |
Tel | -5.151 |
Token | how |
Token ID | 703 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
ithing | +1.716 |
hing | +1.487 |
andom | +1.483 |
nesota | +1.454 |
arning | +1.440 |
Neg logit contributions | |
Stub | -1.879 |
stra | -1.871 |
Amen | -1.829 |
Tel | -1.826 |
stop | -1.822 |
Token | Tel |
Token ID | 12088 |
Feature activation | +9.63e-01 |
Pos logit contributions | |
ithing | +3.427 |
andom | +2.967 |
nesota | +2.936 |
hing | +2.923 |
ammy | +2.847 |
Neg logit contributions | |
Stub | -3.462 |
ospace | -3.461 |
stra | -3.436 |
Amen | -3.408 |
adium | -3.374 |
Token | stra |
Token ID | 12044 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
stra | +0.000 |
Stub | -0.019 |
stop | -0.031 |
Tel | -0.033 |
Amen | -0.034 |
Neg logit contributions | |
ithing | -1.316 |
andom | -1.235 |
nesota | -1.225 |
arning | -1.224 |
hing | -1.220 |
Token | � |
Token ID | 447 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
Stub | +0.535 |
stra | +0.533 |
Amen | +0.522 |
Tel | +0.518 |
stop | +0.517 |
Neg logit contributions | |
ithing | -0.482 |
andom | -0.418 |
nesota | -0.414 |
hing | -0.409 |
arning | -0.409 |
Token | so |
Token ID | 523 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
Stub | +0.362 |
stra | +0.359 |
Amen | +0.353 |
reopen | +0.349 |
stop | +0.348 |
Neg logit contributions | |
ithing | -0.397 |
andom | -0.352 |
nesota | -0.347 |
hing | -0.343 |
arning | -0.343 |
Token | it |
Token ID | 340 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
Stub | +0.463 |
stra | +0.454 |
Amen | +0.447 |
stop | +0.443 |
reopen | +0.442 |
Neg logit contributions | |
ithing | -0.464 |
andom | -0.409 |
nesota | -0.403 |
hing | -0.399 |
arning | -0.399 |
Token | 's |
Token ID | 338 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
Stub | +0.296 |
stra | +0.293 |
Amen | +0.287 |
reopen | +0.286 |
stop | +0.284 |
Neg logit contributions | |
ithing | -0.326 |
andom | -0.288 |
nesota | -0.283 |
hing | -0.281 |
arning | -0.281 |
Token | not |
Token ID | 407 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
Stub | +0.069 |
stra | +0.068 |
Amen | +0.067 |
reopen | +0.066 |
stop | +0.066 |
Neg logit contributions | |
ithing | -0.067 |
andom | -0.058 |
nesota | -0.057 |
arning | -0.057 |
hing | -0.057 |
Token | entirely |
Token ID | 5000 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ithing | +0.440 |
andom | +0.384 |
nesota | +0.381 |
hing | +0.376 |
arning | +0.375 |
Neg logit contributions | |
Stub | -0.432 |
stra | -0.428 |
Amen | -0.420 |
reopen | -0.415 |
stop | -0.415 |
Token | clear |
Token ID | 1598 |
Feature activation | -7.35e+00 |
Pos logit contributions | |
ithing | +0.821 |
andom | +0.708 |
nesota | +0.699 |
hing | +0.692 |
arning | +0.689 |
Neg logit contributions | |
Stub | -0.984 |
stra | -0.978 |
Amen | -0.962 |
stop | -0.952 |
Tel | -0.950 |
Token | what |
Token ID | 644 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
ithing | +4.123 |
hing | +3.721 |
andom | +3.518 |
nesota | +3.452 |
arning | +3.430 |
Neg logit contributions | |
stra | -5.391 |
Tel | -5.273 |
Stub | -5.260 |
ospace | -5.227 |
erion | -5.130 |
Token | he |
Token ID | 339 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ithing | +1.379 |
andom | +1.205 |
nesota | +1.198 |
hing | +1.194 |
arning | +1.173 |
Neg logit contributions | |
stra | -1.437 |
Stub | -1.430 |
Amen | -1.409 |
Tel | -1.396 |
stop | -1.388 |
Token | 's |
Token ID | 338 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
Stub | +0.755 |
stra | +0.752 |
Amen | +0.737 |
stop | +0.729 |
reopen | +0.729 |
Neg logit contributions | |
ithing | -0.712 |
andom | -0.620 |
nesota | -0.612 |
hing | -0.608 |
arning | -0.605 |
Token | going |
Token ID | 1016 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
ithing | +0.072 |
andom | +0.063 |
nesota | +0.062 |
hing | +0.062 |
arning | +0.062 |
Neg logit contributions | |
Stub | -0.072 |
stra | -0.072 |
Amen | -0.071 |
stop | -0.070 |
reopen | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
Stub | +0.586 |
stra | +0.584 |
Amen | +0.571 |
stop | +0.564 |
Tel | +0.564 |
Neg logit contributions | |
ithing | -0.656 |
andom | -0.578 |
nesota | -0.571 |
hing | -0.568 |
arning | -0.565 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ithing | +3.727 |
andom | +3.292 |
hing | +3.271 |
nesota | +3.263 |
arning | +3.215 |
Neg logit contributions | |
stra | -3.148 |
ospace | -3.100 |
Stub | -3.097 |
reopen | -3.056 |
stop | -3.021 |
Token | Canada |
Token ID | 3340 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ithing | +2.464 |
andom | +2.168 |
nesota | +2.146 |
hing | +2.135 |
arning | +2.114 |
Neg logit contributions | |
stra | -2.247 |
Stub | -2.214 |
ospace | -2.178 |
reopen | -2.174 |
Amen | -2.169 |
Token | 16 |
Token ID | 1467 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ithing | +1.701 |
andom | +1.502 |
nesota | +1.483 |
hing | +1.476 |
arning | +1.467 |
Neg logit contributions | |
Stub | -1.510 |
stra | -1.505 |
Amen | -1.472 |
stop | -1.457 |
Tel | -1.454 |
Token | 10 |
Token ID | 940 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
ithing | +1.634 |
andom | +1.442 |
nesota | +1.421 |
hing | +1.416 |
arning | +1.407 |
Neg logit contributions | |
Stub | -1.497 |
stra | -1.496 |
Amen | -1.461 |
stop | -1.445 |
reopen | -1.443 |
Token | Posts |
Token ID | 12043 |
Feature activation | -9.51e+00 |
Pos logit contributions | |
ithing | +1.283 |
andom | +0.962 |
nesota | +0.933 |
hing | +0.921 |
arning | +0.909 |
Neg logit contributions | |
Stub | -3.847 |
stra | -3.832 |
Amen | -3.784 |
stop | -3.756 |
Tel | -3.754 |
Token | # |
Token ID | 1303 |
Feature activation | -7.30e+00 |
Pos logit contributions | |
ithing | +4.402 |
andom | +3.582 |
hing | +3.322 |
arning | +3.289 |
ammy | +3.250 |
Neg logit contributions | |
stra | -8.087 |
Tel | -7.844 |
ospace | -7.840 |
stop | -7.795 |
Stub | -7.676 |
Token | 14 |
Token ID | 1415 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
ithing | +4.918 |
hing | +4.559 |
andom | +4.360 |
arning | +4.255 |
nesota | +4.243 |
Neg logit contributions | |
Stub | -4.633 |
reopen | -4.537 |
Amen | -4.469 |
ospace | -4.441 |
stra | -4.432 |
Token | On |
Token ID | 1550 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
ithing | +1.761 |
andom | +1.526 |
hing | +1.515 |
nesota | +1.508 |
arning | +1.483 |
Neg logit contributions | |
stra | -1.910 |
Stub | -1.890 |
Tel | -1.849 |
Amen | -1.845 |
stop | -1.841 |
Token | September |
Token ID | 2693 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
ithing | +2.304 |
andom | +1.995 |
nesota | +1.965 |
hing | +1.962 |
arning | +1.943 |
Neg logit contributions | |
Stub | -2.581 |
stra | -2.574 |
Amen | -2.524 |
Tel | -2.498 |
stop | -2.498 |
Token | 20 |
Token ID | 1160 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
ithing | +1.263 |
andom | +1.076 |
hing | +1.065 |
nesota | +1.064 |
arning | +1.034 |
Neg logit contributions | |
stra | -1.807 |
Stub | -1.785 |
stop | -1.753 |
Tel | -1.752 |
reopen | -1.740 |
Token | 2013 |
Token ID | 2211 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ithing | +0.988 |
andom | +0.892 |
nesota | +0.884 |
hing | +0.880 |
arning | +0.877 |
Neg logit contributions | |
Stub | -0.548 |
stra | -0.545 |
Amen | -0.529 |
stop | -0.521 |
Tel | -0.521 |
Token | of |
Token ID | 286 |
Feature activation | -6.33e-01 |
Pos logit contributions | |
Stub | +0.001 |
stra | +0.001 |
Amen | +0.001 |
reopen | +0.001 |
stop | +0.001 |
Neg logit contributions | |
ithing | -0.001 |
andom | -0.001 |
nesota | -0.001 |
hing | -0.001 |
arning | -0.001 |
Token | Dal |
Token ID | 12348 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
ithing | +0.425 |
andom | +0.372 |
nesota | +0.367 |
hing | +0.365 |
arning | +0.363 |
Neg logit contributions | |
Stub | -0.427 |
stra | -0.425 |
Amen | -0.416 |
stop | -0.412 |
Tel | -0.412 |
Token | K |
Token ID | 42 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ithing | +2.038 |
andom | +1.782 |
hing | +1.733 |
nesota | +1.721 |
arning | +1.707 |
Neg logit contributions | |
Stub | -2.246 |
stra | -2.234 |
Amen | -2.194 |
reopen | -2.175 |
Tel | -2.151 |
Token | Profile |
Token ID | 13118 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
ithing | +1.365 |
andom | +1.200 |
hing | +1.196 |
arning | +1.180 |
nesota | +1.179 |
Neg logit contributions | |
Stub | -1.200 |
stra | -1.186 |
Amen | -1.176 |
reopen | -1.166 |
stop | -1.155 |
Token | Blog |
Token ID | 14001 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
ithing | +1.402 |
andom | +1.154 |
hing | +1.129 |
arning | +1.117 |
nesota | +1.109 |
Neg logit contributions | |
stra | -2.525 |
Stub | -2.508 |
Amen | -2.478 |
reopen | -2.470 |
stop | -2.469 |
Token | Joined |
Token ID | 16709 |
Feature activation | -7.13e+00 |
Pos logit contributions | |
ithing | +1.172 |
andom | +0.942 |
nesota | +0.939 |
hing | +0.913 |
arning | +0.911 |
Neg logit contributions | |
Stub | -2.370 |
stra | -2.358 |
Amen | -2.319 |
stop | -2.297 |
Tel | -2.290 |
Token | June |
Token ID | 2795 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
ithing | +3.853 |
andom | +3.275 |
hing | +3.235 |
nesota | +3.167 |
arning | +3.152 |
Neg logit contributions | |
stra | -5.465 |
ospace | -5.456 |
Stub | -5.356 |
stop | -5.327 |
Tel | -5.312 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ithing | +3.727 |
andom | +3.292 |
hing | +3.271 |
nesota | +3.263 |
arning | +3.215 |
Neg logit contributions | |
stra | -3.148 |
ospace | -3.100 |
Stub | -3.097 |
reopen | -3.056 |
stop | -3.021 |
Token | Canada |
Token ID | 3340 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ithing | +2.464 |
andom | +2.168 |
nesota | +2.146 |
hing | +2.135 |
arning | +2.114 |
Neg logit contributions | |
stra | -2.247 |
Stub | -2.214 |
ospace | -2.178 |
reopen | -2.174 |
Amen | -2.169 |
Token | 16 |
Token ID | 1467 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ithing | +1.701 |
andom | +1.502 |
nesota | +1.483 |
hing | +1.476 |
arning | +1.467 |
Neg logit contributions | |
Stub | -1.510 |
stra | -1.505 |
Amen | -1.472 |
stop | -1.457 |
Tel | -1.454 |
Token | 10 |
Token ID | 940 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
ithing | +1.634 |
andom | +1.442 |
nesota | +1.421 |
hing | +1.416 |
arning | +1.407 |
Neg logit contributions | |
Stub | -1.497 |
stra | -1.496 |
Amen | -1.461 |
stop | -1.445 |
reopen | -1.443 |
Token | J |
Token ID | 41 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ithing | +1.384 |
andom | +1.235 |
hing | +1.223 |
nesota | +1.208 |
arning | +1.204 |
Neg logit contributions | |
Stub | -1.165 |
stra | -1.133 |
Amen | -1.131 |
reopen | -1.121 |
stop | -1.092 |
Token | aaaa |
Token ID | 24794 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
ithing | +0.711 |
andom | +0.626 |
nesota | +0.615 |
hing | +0.615 |
arning | +0.610 |
Neg logit contributions | |
Stub | -0.681 |
stra | -0.676 |
Amen | -0.664 |
reopen | -0.656 |
stop | -0.654 |
Token | asper |
Token ID | 32981 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ithing | +1.573 |
andom | +1.388 |
hing | +1.361 |
arning | +1.350 |
nesota | +1.347 |
Neg logit contributions | |
Stub | -1.571 |
stra | -1.555 |
Amen | -1.530 |
reopen | -1.515 |
stop | -1.506 |
Token | Profile |
Token ID | 13118 |
Feature activation | -3.44e+00 |
Pos logit contributions | |
ithing | +0.876 |
andom | +0.780 |
hing | +0.767 |
nesota | +0.764 |
arning | +0.759 |
Neg logit contributions | |
Stub | -0.754 |
stra | -0.746 |
Amen | -0.735 |
reopen | -0.729 |
Tel | -0.723 |
Token | Blog |
Token ID | 14001 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
ithing | +1.503 |
andom | +1.220 |
hing | +1.198 |
arning | +1.176 |
nesota | +1.167 |
Neg logit contributions | |
stra | -3.074 |
Stub | -3.052 |
reopen | -3.018 |
Amen | -3.015 |
Tel | -3.013 |
Token | Joined |
Token ID | 16709 |
Feature activation | -7.13e+00 |
Pos logit contributions | |
ithing | +1.360 |
andom | +1.090 |
nesota | +1.073 |
hing | +1.056 |
arning | +1.048 |
Neg logit contributions | |
Stub | -2.894 |
stra | -2.883 |
Amen | -2.837 |
stop | -2.815 |
Tel | -2.809 |
Token | April |
Token ID | 3035 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
ithing | +3.811 |
andom | +3.248 |
hing | +3.210 |
nesota | +3.130 |
arning | +3.108 |
Neg logit contributions | |
stra | -5.479 |
ospace | -5.477 |
Stub | -5.362 |
Tel | -5.333 |
stop | -5.329 |
Token | 2012 |
Token ID | 2321 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
ithing | +2.776 |
andom | +2.464 |
nesota | +2.436 |
hing | +2.433 |
arning | +2.410 |
Neg logit contributions | |
stra | -2.190 |
Stub | -2.179 |
Amen | -2.120 |
reopen | -2.116 |
ospace | -2.110 |
Token | United |
Token ID | 1578 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
ithing | +2.634 |
andom | +2.314 |
nesota | +2.292 |
hing | +2.283 |
arning | +2.257 |
Neg logit contributions | |
stra | -2.458 |
Stub | -2.418 |
ospace | -2.380 |
reopen | -2.379 |
stop | -2.369 |
Token | States |
Token ID | 1829 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
ithing | +1.052 |
andom | +0.887 |
nesota | +0.883 |
arning | +0.862 |
hing | +0.853 |
Neg logit contributions | |
stra | -1.601 |
Stub | -1.591 |
reopen | -1.580 |
ospace | -1.563 |
stop | -1.561 |
Token | 5 |
Token ID | 642 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
ithing | +2.928 |
andom | +2.571 |
nesota | +2.540 |
hing | +2.538 |
arning | +2.501 |
Neg logit contributions | |
stra | -2.819 |
Stub | -2.781 |
ospace | -2.739 |
reopen | -2.732 |
Amen | -2.723 |
Token |
|
Token ID | 198 |
Feature activation | -9.82e-02 |
Pos logit contributions | |
ithing | +0.464 |
andom | +0.407 |
hing | +0.404 |
nesota | +0.399 |
arning | +0.398 |
Neg logit contributions | |
Stub | -0.469 |
stra | -0.466 |
stop | -0.455 |
reopen | -0.453 |
Amen | -0.453 |
Token |
|
Token ID | 198 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
ithing | +0.075 |
andom | +0.066 |
nesota | +0.065 |
arning | +0.065 |
ammy | +0.064 |
Neg logit contributions | |
stra | -0.060 |
Stub | -0.060 |
stop | -0.058 |
Amen | -0.058 |
Tel | -0.058 |
Token | It |
Token ID | 1026 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Stub | +0.313 |
stra | +0.311 |
Amen | +0.304 |
stop | +0.300 |
Tel | +0.300 |
Neg logit contributions | |
ithing | -0.401 |
andom | -0.357 |
nesota | -0.353 |
hing | -0.352 |
arning | -0.350 |
Token | was |
Token ID | 373 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
ithing | +0.672 |
andom | +0.577 |
nesota | +0.568 |
hing | +0.565 |
arning | +0.561 |
Neg logit contributions | |
Stub | -0.865 |
stra | -0.861 |
Amen | -0.846 |
stop | -0.838 |
Tel | -0.838 |
Token | not |
Token ID | 407 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
ithing | +0.115 |
andom | +0.100 |
nesota | +0.099 |
hing | +0.098 |
arning | +0.098 |
Neg logit contributions | |
Stub | -0.118 |
stra | -0.117 |
Amen | -0.115 |
Tel | -0.114 |
stop | -0.114 |
Token | clear |
Token ID | 1598 |
Feature activation | -7.09e+00 |
Pos logit contributions | |
Stub | +0.243 |
stra | +0.241 |
Amen | +0.236 |
Tel | +0.234 |
stop | +0.233 |
Neg logit contributions | |
ithing | -0.274 |
andom | -0.241 |
nesota | -0.240 |
arning | -0.236 |
hing | -0.236 |
Token | what |
Token ID | 644 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
ithing | +4.084 |
hing | +3.653 |
andom | +3.449 |
arning | +3.439 |
ammy | +3.415 |
Neg logit contributions | |
stra | -5.066 |
Stub | -5.056 |
Tel | -4.961 |
ospace | -4.925 |
erion | -4.925 |
Token | percentage |
Token ID | 5873 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ithing | +2.204 |
andom | +1.928 |
nesota | +1.918 |
hing | +1.912 |
arning | +1.872 |
Neg logit contributions | |
stra | -2.273 |
Stub | -2.260 |
Amen | -2.213 |
Tel | -2.212 |
stop | -2.210 |
Token | of |
Token ID | 286 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
ithing | +0.921 |
andom | +0.802 |
hing | +0.798 |
nesota | +0.792 |
arning | +0.785 |
Neg logit contributions | |
Stub | -0.911 |
stra | -0.900 |
Amen | -0.877 |
stop | -0.877 |
Tel | -0.874 |
Token | students |
Token ID | 2444 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Stub | +0.508 |
stra | +0.504 |
Amen | +0.495 |
Tel | +0.487 |
stop | +0.487 |
Neg logit contributions | |
ithing | -0.568 |
andom | -0.501 |
nesota | -0.495 |
hing | -0.491 |
arning | -0.489 |
Token | were |
Token ID | 547 |
Feature activation | -5.65e-01 |
Pos logit contributions | |
Stub | +0.691 |
stra | +0.686 |
Amen | +0.674 |
reopen | +0.664 |
Tel | +0.662 |
Neg logit contributions | |
ithing | -0.710 |
andom | -0.626 |
nesota | -0.620 |
hing | -0.610 |
arning | -0.608 |
Token | K |
Token ID | 42 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
ithing | +1.364 |
andom | +1.215 |
hing | +1.196 |
nesota | +1.187 |
arning | +1.183 |
Neg logit contributions | |
Stub | -1.172 |
stra | -1.158 |
Amen | -1.144 |
reopen | -1.135 |
stop | -1.124 |
Token | ier |
Token ID | 959 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
ithing | +2.032 |
andom | +1.805 |
hing | +1.793 |
arning | +1.775 |
nesota | +1.756 |
Neg logit contributions | |
Stub | -1.793 |
Amen | -1.753 |
stra | -1.745 |
reopen | -1.725 |
stop | -1.699 |
Token | an |
Token ID | 272 |
Feature activation | +8.94e-01 |
Pos logit contributions | |
Stub | +0.252 |
stra | +0.250 |
Amen | +0.245 |
stop | +0.242 |
Tel | +0.242 |
Neg logit contributions | |
ithing | -0.296 |
andom | -0.262 |
nesota | -0.259 |
hing | -0.257 |
arning | -0.256 |
Token | Profile |
Token ID | 13118 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
Stub | +0.541 |
stra | +0.539 |
Amen | +0.526 |
stop | +0.520 |
Tel | +0.520 |
Neg logit contributions | |
ithing | -0.663 |
andom | -0.588 |
nesota | -0.581 |
hing | -0.578 |
arning | -0.575 |
Token | Blog |
Token ID | 14001 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
ithing | +1.317 |
andom | +1.091 |
hing | +1.074 |
arning | +1.054 |
nesota | +1.051 |
Neg logit contributions | |
stra | -2.334 |
Stub | -2.317 |
reopen | -2.288 |
Amen | -2.287 |
Tel | -2.281 |
Token | Joined |
Token ID | 16709 |
Feature activation | -7.04e+00 |
Pos logit contributions | |
ithing | +1.311 |
andom | +1.054 |
nesota | +1.044 |
hing | +1.020 |
arning | +1.018 |
Neg logit contributions | |
Stub | -2.669 |
stra | -2.654 |
Amen | -2.614 |
stop | -2.591 |
Tel | -2.580 |
Token | August |
Token ID | 2932 |
Feature activation | -4.04e+00 |
Pos logit contributions | |
ithing | +3.834 |
andom | +3.238 |
hing | +3.214 |
nesota | +3.158 |
arning | +3.131 |
Neg logit contributions | |
stra | -5.381 |
ospace | -5.367 |
Stub | -5.277 |
stop | -5.240 |
Tel | -5.239 |
Token | 2011 |
Token ID | 2813 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
ithing | +2.968 |
andom | +2.627 |
nesota | +2.604 |
hing | +2.601 |
arning | +2.575 |
Neg logit contributions | |
stra | -2.419 |
Stub | -2.406 |
ospace | -2.337 |
reopen | -2.335 |
Amen | -2.333 |
Token | United |
Token ID | 1578 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
ithing | +2.771 |
andom | +2.440 |
nesota | +2.418 |
hing | +2.415 |
arning | +2.383 |
Neg logit contributions | |
stra | -2.465 |
Stub | -2.436 |
ospace | -2.396 |
reopen | -2.391 |
Amen | -2.385 |
Token | States |
Token ID | 1829 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
ithing | +1.173 |
andom | +0.993 |
nesota | +0.991 |
arning | +0.965 |
hing | +0.951 |
Neg logit contributions | |
stra | -1.718 |
Stub | -1.717 |
reopen | -1.698 |
ospace | -1.683 |
stop | -1.679 |
Token | 6 |
Token ID | 718 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
ithing | +2.862 |
andom | +2.513 |
nesota | +2.485 |
hing | +2.481 |
arning | +2.448 |
Neg logit contributions | |
stra | -2.733 |
Stub | -2.707 |
ospace | -2.652 |
reopen | -2.649 |
Amen | -2.647 |
Token | le |
Token ID | 293 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
Stub | +0.052 |
stra | +0.051 |
Amen | +0.050 |
Tel | +0.049 |
reopen | +0.049 |
Neg logit contributions | |
ithing | -0.112 |
andom | -0.101 |
nesota | -0.101 |
hing | -0.101 |
arning | -0.100 |
Token |
|
Token ID | 198 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
Stub | +1.799 |
stra | +1.791 |
Amen | +1.755 |
stop | +1.734 |
Tel | +1.731 |
Neg logit contributions | |
ithing | -1.988 |
andom | -1.752 |
nesota | -1.732 |
hing | -1.722 |
arning | -1.712 |
Token |
|
Token ID | 198 |
Feature activation | +7.12e-03 |
Pos logit contributions | |
Stub | +0.016 |
stra | +0.016 |
stop | +0.015 |
Amen | +0.015 |
Tel | +0.015 |
Neg logit contributions | |
ithing | -0.020 |
andom | -0.018 |
nesota | -0.017 |
arning | -0.017 |
ammy | -0.017 |
Token | 2012 |
Token ID | 6999 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
stra | +0.005 |
Stub | +0.005 |
stop | +0.005 |
Tel | +0.005 |
Amen | +0.005 |
Neg logit contributions | |
ithing | -0.004 |
andom | -0.004 |
nesota | -0.004 |
arning | -0.004 |
hing | -0.004 |
Token | Larry |
Token ID | 13633 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ithing | +0.914 |
andom | +0.801 |
nesota | +0.792 |
hing | +0.789 |
arning | +0.782 |
Neg logit contributions | |
stra | -0.886 |
Stub | -0.882 |
Amen | -0.863 |
reopen | -0.857 |
stop | -0.856 |
Token | David |
Token ID | 3271 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
Stub | +0.677 |
stra | +0.651 |
Amen | +0.649 |
Tel | +0.630 |
stop | +0.626 |
Neg logit contributions | |
ithing | -1.030 |
andom | -0.914 |
nesota | -0.911 |
arning | -0.909 |
hing | -0.901 |
Token |
|
Token ID | 198 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
Stub | +2.479 |
stra | +2.467 |
Amen | +2.420 |
stop | +2.387 |
Tel | +2.384 |
Neg logit contributions | |
ithing | -2.671 |
andom | -2.349 |
nesota | -2.327 |
hing | -2.311 |
arning | -2.297 |
Token |
|
Token ID | 198 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
Stub | +1.156 |
stra | +1.152 |
stop | +1.132 |
Amen | +1.127 |
Tel | +1.119 |
Neg logit contributions | |
ithing | -1.674 |
andom | -1.485 |
arning | -1.471 |
nesota | -1.463 |
ammy | -1.453 |
Token | 2012 |
Token ID | 6999 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
stra | +0.780 |
Stub | +0.773 |
stop | +0.765 |
Tel | +0.764 |
Amen | +0.753 |
Neg logit contributions | |
ithing | -0.689 |
andom | -0.593 |
nesota | -0.589 |
arning | -0.579 |
hing | -0.571 |
Token | Jim |
Token ID | 5395 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
ithing | +0.503 |
andom | +0.441 |
nesota | +0.435 |
hing | +0.433 |
arning | +0.431 |
Neg logit contributions | |
Stub | -0.484 |
stra | -0.482 |
Amen | -0.472 |
stop | -0.467 |
Tel | -0.466 |
Token | Parsons |
Token ID | 35925 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
Stub | +0.973 |
Amen | +0.939 |
stra | +0.912 |
Tel | +0.907 |
stop | +0.889 |
Neg logit contributions | |
ithing | -1.596 |
nesota | -1.430 |
andom | -1.421 |
arning | -1.410 |
ammy | -1.389 |
Token | during |
Token ID | 1141 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
ithing | +1.009 |
andom | +0.885 |
nesota | +0.871 |
hing | +0.868 |
arning | +0.862 |
Neg logit contributions | |
Stub | -1.017 |
stra | -1.014 |
Amen | -0.997 |
Tel | -0.988 |
stop | -0.985 |
Token | Washington |
Token ID | 2669 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Stub | +0.572 |
stra | +0.571 |
Amen | +0.558 |
Tel | +0.554 |
stop | +0.552 |
Neg logit contributions | |
ithing | -0.603 |
andom | -0.529 |
nesota | -0.525 |
hing | -0.521 |
arning | -0.519 |
Token | � |
Token ID | 447 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
ithing | +0.200 |
andom | +0.181 |
nesota | +0.180 |
hing | +0.178 |
arning | +0.177 |
Neg logit contributions | |
Stub | -0.133 |
stra | -0.132 |
Amen | -0.130 |
stop | -0.127 |
reopen | -0.127 |
Token | � |
Token ID | 247 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
Stub | +0.034 |
stra | +0.034 |
Amen | +0.033 |
stop | +0.032 |
Tel | +0.032 |
Neg logit contributions | |
ithing | -0.055 |
andom | -0.049 |
nesota | -0.049 |
hing | -0.049 |
arning | -0.048 |
Token | s |
Token ID | 82 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
Stub | +0.312 |
stra | +0.311 |
Amen | +0.304 |
stop | +0.300 |
Tel | +0.300 |
Neg logit contributions | |
ithing | -0.388 |
andom | -0.344 |
nesota | -0.340 |
hing | -0.339 |
arning | -0.337 |
Token | practice |
Token ID | 3357 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
ithing | +0.029 |
andom | +0.025 |
nesota | +0.025 |
hing | +0.025 |
arning | +0.025 |
Neg logit contributions | |
Stub | -0.027 |
stra | -0.027 |
Amen | -0.026 |
Tel | -0.026 |
stop | -0.026 |
Token | May |
Token ID | 1737 |
Feature activation | -8.76e-01 |
Pos logit contributions | |
Stub | +2.311 |
stra | +2.283 |
Amen | +2.245 |
stop | +2.222 |
reopen | +2.209 |
Neg logit contributions | |
ithing | -3.031 |
andom | -2.709 |
arning | -2.649 |
nesota | -2.645 |
hing | -2.639 |
Token | 25 |
Token ID | 1679 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
ithing | +0.642 |
andom | +0.568 |
nesota | +0.560 |
arning | +0.556 |
hing | +0.556 |
Neg logit contributions | |
Stub | -0.542 |
stra | -0.538 |
Amen | -0.528 |
stop | -0.521 |
Tel | -0.519 |
Token | . |
Token ID | 13 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
Stub | +0.021 |
stra | +0.020 |
Amen | +0.020 |
Tel | +0.020 |
stop | +0.020 |
Neg logit contributions | |
ithing | -0.025 |
andom | -0.022 |
nesota | -0.022 |
hing | -0.022 |
arning | -0.022 |
Token | ( |
Token ID | 357 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
ithing | +0.170 |
andom | +0.150 |
nesota | +0.148 |
hing | +0.147 |
arning | +0.146 |
Neg logit contributions | |
Stub | -0.161 |
stra | -0.161 |
Amen | -0.157 |
stop | -0.155 |
Tel | -0.155 |
Token | John |
Token ID | 7554 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
ithing | +0.593 |
andom | +0.525 |
hing | +0.518 |
nesota | +0.518 |
arning | +0.514 |
Neg logit contributions | |
Stub | -0.502 |
stra | -0.502 |
Amen | -0.491 |
reopen | -0.486 |
stop | -0.484 |
Token | come |
Token ID | 1282 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
Stub | +1.035 |
stra | +1.032 |
Amen | +1.009 |
reopen | +1.001 |
stop | +0.997 |
Neg logit contributions | |
ithing | -1.163 |
andom | -1.028 |
nesota | -1.017 |
arning | -1.011 |
hing | -1.003 |
Token | . |
Token ID | 13 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
Stub | +0.862 |
stra | +0.862 |
Amen | +0.844 |
stop | +0.829 |
reopen | +0.828 |
Neg logit contributions | |
ithing | -1.063 |
andom | -0.945 |
nesota | -0.934 |
arning | -0.926 |
hing | -0.924 |
Token | For |
Token ID | 1114 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
Stub | +1.197 |
stra | +1.180 |
Amen | +1.176 |
Tel | +1.144 |
stop | +1.144 |
Neg logit contributions | |
ithing | -1.543 |
andom | -1.371 |
arning | -1.350 |
nesota | -1.345 |
hing | -1.345 |
Token | more |
Token ID | 517 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
Stub | +1.168 |
stra | +1.155 |
Amen | +1.136 |
stop | +1.126 |
reopen | +1.121 |
Neg logit contributions | |
ithing | -1.218 |
andom | -1.072 |
nesota | -1.058 |
arning | -1.052 |
hing | -1.046 |
Token | information |
Token ID | 1321 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
Stub | +1.071 |
stra | +1.062 |
Amen | +1.038 |
stop | +1.032 |
reopen | +1.025 |
Neg logit contributions | |
ithing | -1.287 |
andom | -1.144 |
arning | -1.124 |
nesota | -1.121 |
hing | -1.118 |
Token | go |
Token ID | 467 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
Stub | +1.305 |
stra | +1.300 |
Amen | +1.271 |
stop | +1.256 |
reopen | +1.255 |
Neg logit contributions | |
ithing | -1.663 |
andom | -1.479 |
nesota | -1.459 |
hing | -1.452 |
arning | -1.448 |
Token | to |
Token ID | 284 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
Stub | +2.503 |
stra | +2.472 |
stop | +2.439 |
Amen | +2.425 |
Tel | +2.414 |
Neg logit contributions | |
ithing | -3.124 |
andom | -2.773 |
nesota | -2.744 |
arning | -2.731 |
hing | -2.723 |
Token | emergency |
Token ID | 6334 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Stub | +1.108 |
stra | +1.101 |
Amen | +1.077 |
stop | +1.066 |
Tel | +1.061 |
Neg logit contributions | |
ithing | -1.394 |
andom | -1.238 |
nesota | -1.223 |
hing | -1.217 |
arning | -1.216 |
Token | . |
Token ID | 13 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
Stub | +0.714 |
stra | +0.712 |
Amen | +0.694 |
stop | +0.688 |
Tel | +0.685 |
Neg logit contributions | |
ithing | -0.946 |
andom | -0.842 |
nesota | -0.833 |
hing | -0.828 |
arning | -0.825 |
Token | tt |
Token ID | 926 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
stra | +1.344 |
Stub | +1.332 |
stop | +1.301 |
Amen | +1.296 |
Tel | +1.284 |
Neg logit contributions | |
ithing | -1.851 |
andom | -1.649 |
nesota | -1.627 |
arning | -1.621 |
hing | -1.619 |
Token | u |
Token ID | 84 |
Feature activation | -9.86e-01 |
Pos logit contributions | |
ithing | +1.001 |
andom | +0.887 |
nesota | +0.874 |
hing | +0.871 |
arning | +0.867 |
Neg logit contributions | |
Stub | -0.856 |
stra | -0.846 |
Amen | -0.834 |
Tel | -0.826 |
reopen | -0.826 |
Token | suffered |
Token ID | 6989 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
Stub | +0.280 |
stra | +0.279 |
Amen | +0.273 |
stop | +0.271 |
Tel | +0.270 |
Neg logit contributions | |
ithing | -0.270 |
andom | -0.236 |
nesota | -0.232 |
hing | -0.231 |
arning | -0.230 |
Token | a |
Token ID | 257 |
Feature activation | +9.34e-02 |
Pos logit contributions | |
Stub | +0.030 |
stra | +0.030 |
Amen | +0.029 |
stop | +0.029 |
Tel | +0.029 |
Neg logit contributions | |
ithing | -0.026 |
andom | -0.022 |
nesota | -0.022 |
hing | -0.022 |
arning | -0.022 |
Token | rupt |
Token ID | 31580 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
Stub | +0.061 |
stra | +0.060 |
Amen | +0.059 |
stop | +0.058 |
Tel | +0.058 |
Neg logit contributions | |
ithing | -0.066 |
andom | -0.058 |
nesota | -0.057 |
hing | -0.057 |
arning | -0.056 |
Token | ured |
Token ID | 1522 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
stra | +0.681 |
Stub | +0.673 |
stop | +0.669 |
Tel | +0.655 |
ospace | +0.650 |
Neg logit contributions | |
ithing | -0.929 |
andom | -0.835 |
nesota | -0.832 |
arning | -0.811 |
hing | -0.803 |
Token | Achilles |
Token ID | 39561 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
Stub | +0.848 |
stra | +0.825 |
Amen | +0.820 |
Tel | +0.808 |
stop | +0.800 |
Neg logit contributions | |
ithing | -1.038 |
nesota | -0.920 |
andom | -0.920 |
hing | -0.904 |
arning | -0.903 |
Token | tendon |
Token ID | 42205 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
Stub | +1.971 |
stra | +1.933 |
Tel | +1.881 |
reopen | +1.861 |
Amen | +1.860 |
Neg logit contributions | |
ithing | -2.684 |
andom | -2.428 |
arning | -2.356 |
nesota | -2.352 |
hing | -2.342 |
Token | in |
Token ID | 287 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
Stub | +2.164 |
stra | +2.127 |
reopen | +2.072 |
stop | +2.067 |
Amen | +2.062 |
Neg logit contributions | |
ithing | -2.595 |
andom | -2.342 |
hing | -2.269 |
nesota | -2.268 |
arning | -2.250 |
Token | Thursday |
Token ID | 3635 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
Stub | +0.183 |
stra | +0.182 |
Amen | +0.179 |
stop | +0.177 |
Tel | +0.176 |
Neg logit contributions | |
ithing | -0.198 |
andom | -0.174 |
nesota | -0.172 |
hing | -0.171 |
arning | -0.170 |
Token | 's |
Token ID | 338 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
Stub | +1.411 |
stra | +1.403 |
Amen | +1.368 |
stop | +1.349 |
Tel | +1.347 |
Neg logit contributions | |
ithing | -2.165 |
andom | -1.942 |
nesota | -1.921 |
hing | -1.913 |
arning | -1.904 |
Token | game |
Token ID | 983 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
Stub | +1.154 |
stra | +1.132 |
Amen | +1.117 |
stop | +1.105 |
reopen | +1.102 |
Neg logit contributions | |
ithing | -1.184 |
andom | -1.037 |
nesota | -1.015 |
hing | -1.012 |
arning | -1.007 |
Token | against |
Token ID | 1028 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
Stub | +0.990 |
stra | +0.980 |
Amen | +0.958 |
stop | +0.952 |
Tel | +0.947 |
Neg logit contributions | |
ithing | -1.252 |
andom | -1.114 |
nesota | -1.097 |
hing | -1.091 |
arning | -1.089 |
Token | . |
Token ID | 13 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ithing | +0.462 |
andom | +0.411 |
nesota | +0.407 |
hing | +0.405 |
arning | +0.403 |
Neg logit contributions | |
Stub | -0.350 |
stra | -0.348 |
Amen | -0.340 |
stop | -0.336 |
Tel | -0.336 |
Token | S |
Token ID | 50 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
Stub | +1.237 |
stra | +1.231 |
Amen | +1.208 |
stop | +1.194 |
Tel | +1.193 |
Neg logit contributions | |
ithing | -1.227 |
andom | -1.073 |
nesota | -1.059 |
hing | -1.054 |
arning | -1.047 |
Token | . |
Token ID | 13 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Stub | +1.340 |
stra | +1.329 |
Amen | +1.302 |
stop | +1.282 |
Tel | +1.278 |
Neg logit contributions | |
ithing | -1.841 |
andom | -1.642 |
nesota | -1.623 |
hing | -1.620 |
arning | -1.609 |
Token | nuclear |
Token ID | 4523 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
stra | +1.186 |
Stub | +1.185 |
Amen | +1.155 |
stop | +1.143 |
reopen | +1.141 |
Neg logit contributions | |
ithing | -1.500 |
andom | -1.331 |
nesota | -1.323 |
arning | -1.307 |
hing | -1.306 |
Token | regulatory |
Token ID | 11344 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
Stub | +0.668 |
stra | +0.664 |
Amen | +0.645 |
stop | +0.638 |
reopen | +0.638 |
Neg logit contributions | |
ithing | -1.106 |
andom | -0.994 |
nesota | -0.987 |
arning | -0.979 |
hing | -0.979 |
Token | body |
Token ID | 1767 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
Stub | +1.100 |
stra | +1.083 |
Amen | +1.057 |
reopen | +1.048 |
stop | +1.037 |
Neg logit contributions | |
ithing | -2.768 |
andom | -2.503 |
nesota | -2.488 |
arning | -2.487 |
hing | -2.473 |
Token | said |
Token ID | 531 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Stub | +2.640 |
stra | +2.592 |
Amen | +2.554 |
reopen | +2.543 |
stop | +2.522 |
Neg logit contributions | |
ithing | -2.841 |
nesota | -2.498 |
andom | -2.494 |
arning | -2.451 |
hing | -2.432 |
Token | one |
Token ID | 530 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
Stub | +0.826 |
stra | +0.820 |
Amen | +0.803 |
reopen | +0.798 |
stop | +0.795 |
Neg logit contributions | |
ithing | -0.966 |
andom | -0.853 |
nesota | -0.845 |
arning | -0.835 |
hing | -0.833 |
Token | of |
Token ID | 286 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
ithing | +1.195 |
andom | +1.056 |
hing | +1.049 |
nesota | +1.042 |
arning | +1.032 |
Neg logit contributions | |
Stub | -1.098 |
stra | -1.093 |
Amen | -1.072 |
Tel | -1.060 |
stop | -1.052 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Stub | +0.207 |
stra | +0.206 |
Amen | +0.202 |
stop | +0.200 |
Tel | +0.200 |
Neg logit contributions | |
ithing | -0.206 |
andom | -0.180 |
nesota | -0.178 |
hing | -0.177 |
arning | -0.176 |
Token | pools |
Token ID | 20354 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ithing | +0.932 |
andom | +0.828 |
nesota | +0.817 |
hing | +0.816 |
arning | +0.810 |
Neg logit contributions | |
Stub | -0.742 |
stra | -0.739 |
Amen | -0.722 |
Tel | -0.714 |
stop | -0.713 |
Token | same |
Token ID | 976 |
Feature activation | -5.61e+00 |
Pos logit contributions | |
ithing | +0.007 |
andom | +0.006 |
nesota | +0.006 |
hing | +0.006 |
arning | +0.006 |
Neg logit contributions | |
Stub | -0.006 |
stra | -0.006 |
Amen | -0.006 |
stop | -0.006 |
Tel | -0.006 |
Token | problem |
Token ID | 1917 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
ithing | +3.689 |
nesota | +3.220 |
hing | +3.212 |
andom | +3.200 |
arning | +3.138 |
Neg logit contributions | |
stra | -3.699 |
Stub | -3.666 |
Amen | -3.596 |
Tel | -3.581 |
pub | -3.570 |
Token | hits |
Token ID | 7127 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
ithing | +0.424 |
andom | +0.366 |
nesota | +0.361 |
hing | +0.359 |
arning | +0.357 |
Neg logit contributions | |
Stub | -0.496 |
stra | -0.493 |
Amen | -0.485 |
stop | -0.480 |
Tel | -0.479 |
Token | black |
Token ID | 2042 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ithing | +2.512 |
andom | +2.201 |
nesota | +2.172 |
hing | +2.158 |
arning | +2.136 |
Neg logit contributions | |
stra | -2.651 |
Stub | -2.623 |
Amen | -2.574 |
stop | -2.562 |
Tel | -2.555 |
Token | people |
Token ID | 661 |
Feature activation | -6.45e-01 |
Pos logit contributions | |
Stub | +0.858 |
stra | +0.856 |
Amen | +0.837 |
stop | +0.828 |
reopen | +0.825 |
Neg logit contributions | |
ithing | -0.838 |
andom | -0.732 |
nesota | -0.724 |
hing | -0.719 |
arning | -0.717 |
Token | all |
Token ID | 477 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ithing | +0.434 |
andom | +0.380 |
nesota | +0.376 |
hing | +0.374 |
arning | +0.371 |
Neg logit contributions | |
Stub | -0.433 |
stra | -0.432 |
Amen | -0.423 |
stop | -0.419 |
Tel | -0.418 |
Token | over |
Token ID | 625 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Stub | +0.596 |
stra | +0.593 |
Amen | +0.579 |
stop | +0.571 |
Tel | +0.571 |
Neg logit contributions | |
ithing | -0.813 |
andom | -0.725 |
nesota | -0.717 |
hing | -0.714 |
arning | -0.710 |
Token | the |
Token ID | 262 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
ithing | +0.788 |
andom | +0.690 |
nesota | +0.681 |
hing | +0.674 |
arning | +0.673 |
Neg logit contributions | |
Stub | -0.783 |
stra | -0.781 |
Amen | -0.766 |
Tel | -0.761 |
stop | -0.757 |
Token | country |
Token ID | 1499 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
ithing | +1.724 |
andom | +1.496 |
nesota | +1.489 |
arning | +1.463 |
hing | +1.463 |
Neg logit contributions | |
Stub | -1.863 |
stra | -1.852 |
Amen | -1.823 |
Tel | -1.821 |
reopen | -1.808 |
Token | , |
Token ID | 11 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
ithing | +1.072 |
andom | +0.940 |
hing | +0.934 |
nesota | +0.930 |
arning | +0.925 |
Neg logit contributions | |
Stub | -0.958 |
stra | -0.953 |
Amen | -0.931 |
Tel | -0.926 |
stop | -0.922 |
Token | we |
Token ID | 356 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
Stub | +0.135 |
stra | +0.134 |
Amen | +0.131 |
stop | +0.130 |
Tel | +0.129 |
Neg logit contributions | |
ithing | -0.151 |
andom | -0.134 |
nesota | -0.132 |
hing | -0.131 |
arning | -0.130 |
Token | that |
Token ID | 326 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
Stub | +0.506 |
stra | +0.506 |
Amen | +0.492 |
stop | +0.490 |
Tel | +0.488 |
Neg logit contributions | |
ithing | -0.562 |
andom | -0.494 |
nesota | -0.491 |
hing | -0.490 |
arning | -0.485 |
Token | modern |
Token ID | 3660 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ithing | +0.305 |
andom | +0.267 |
nesota | +0.263 |
hing | +0.263 |
arning | +0.261 |
Neg logit contributions | |
Stub | -0.297 |
stra | -0.296 |
Amen | -0.289 |
stop | -0.288 |
Tel | -0.287 |
Token | reptiles |
Token ID | 46617 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
Stub | +0.639 |
stra | +0.634 |
Amen | +0.622 |
stop | +0.612 |
Tel | +0.611 |
Neg logit contributions | |
ithing | -0.840 |
andom | -0.750 |
nesota | -0.741 |
hing | -0.735 |
arning | -0.732 |
Token | also |
Token ID | 635 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
Stub | +0.637 |
stra | +0.634 |
Amen | +0.622 |
reopen | +0.618 |
stop | +0.616 |
Neg logit contributions | |
ithing | -0.667 |
andom | -0.588 |
nesota | -0.584 |
arning | -0.573 |
hing | -0.572 |
Token | take |
Token ID | 1011 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
ithing | +0.383 |
andom | +0.337 |
nesota | +0.333 |
hing | +0.330 |
arning | +0.329 |
Neg logit contributions | |
Stub | -0.369 |
stra | -0.367 |
Amen | -0.359 |
stop | -0.356 |
reopen | -0.355 |
Token | a |
Token ID | 257 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
Stub | +0.040 |
stra | +0.040 |
Amen | +0.040 |
stop | +0.039 |
Tel | +0.039 |
Neg logit contributions | |
ithing | -0.040 |
andom | -0.035 |
nesota | -0.034 |
hing | -0.034 |
arning | -0.034 |
Token | long |
Token ID | 890 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
Stub | +0.943 |
stra | +0.936 |
Amen | +0.921 |
stop | +0.909 |
Tel | +0.905 |
Neg logit contributions | |
ithing | -1.052 |
andom | -0.933 |
nesota | -0.923 |
arning | -0.913 |
hing | -0.911 |
Token | time |
Token ID | 640 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
Stub | +0.355 |
stra | +0.353 |
Amen | +0.341 |
Tel | +0.339 |
stop | +0.337 |
Neg logit contributions | |
ithing | -0.587 |
andom | -0.525 |
hing | -0.522 |
nesota | -0.519 |
arning | -0.516 |
Token | to |
Token ID | 284 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
Stub | +0.267 |
stra | +0.265 |
Amen | +0.260 |
stop | +0.257 |
Tel | +0.257 |
Neg logit contributions | |
ithing | -0.294 |
andom | -0.259 |
nesota | -0.255 |
hing | -0.255 |
arning | -0.253 |
Token | hatch |
Token ID | 25834 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
Stub | +0.438 |
stra | +0.434 |
Amen | +0.427 |
reopen | +0.425 |
stop | +0.423 |
Neg logit contributions | |
ithing | -0.453 |
andom | -0.399 |
nesota | -0.396 |
arning | -0.389 |
hing | -0.389 |
Token | . |
Token ID | 13 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
Stub | +0.195 |
stra | +0.193 |
Amen | +0.189 |
Tel | +0.187 |
stop | +0.187 |
Neg logit contributions | |
ithing | -0.237 |
andom | -0.209 |
hing | -0.207 |
nesota | -0.206 |
arning | -0.205 |
Token | filled |
Token ID | 5901 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
ithing | +0.712 |
andom | +0.613 |
nesota | +0.604 |
hing | +0.601 |
arning | +0.596 |
Neg logit contributions | |
Stub | -0.877 |
stra | -0.874 |
Amen | -0.858 |
stop | -0.850 |
Tel | -0.849 |
Token | with |
Token ID | 351 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ithing | +0.449 |
andom | +0.385 |
hing | +0.380 |
nesota | +0.380 |
arning | +0.374 |
Neg logit contributions | |
Stub | -0.547 |
stra | -0.547 |
Amen | -0.533 |
Tel | -0.531 |
stop | -0.530 |
Token | water |
Token ID | 1660 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
ithing | +0.924 |
andom | +0.808 |
nesota | +0.799 |
hing | +0.798 |
arning | +0.791 |
Neg logit contributions | |
Stub | -0.880 |
stra | -0.878 |
Amen | -0.857 |
stop | -0.851 |
Tel | -0.850 |
Token | , |
Token ID | 11 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
stra | +1.059 |
Stub | +1.058 |
Amen | +1.032 |
stop | +1.026 |
reopen | +1.017 |
Neg logit contributions | |
ithing | -1.347 |
andom | -1.209 |
nesota | -1.198 |
arning | -1.181 |
hing | -1.174 |
Token | her |
Token ID | 607 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
Stub | +0.657 |
stra | +0.650 |
Amen | +0.643 |
reopen | +0.631 |
stop | +0.630 |
Neg logit contributions | |
ithing | -0.685 |
andom | -0.603 |
nesota | -0.594 |
hing | -0.588 |
arning | -0.587 |
Token | skull |
Token ID | 14511 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Stub | +0.459 |
stra | +0.456 |
Amen | +0.448 |
stop | +0.442 |
reopen | +0.442 |
Neg logit contributions | |
ithing | -0.466 |
andom | -0.410 |
nesota | -0.403 |
hing | -0.399 |
arning | -0.398 |
Token | — |
Token ID | 851 |
Feature activation | +8.74e-02 |
Pos logit contributions | |
ithing | +0.698 |
andom | +0.599 |
nesota | +0.592 |
hing | +0.591 |
arning | +0.584 |
Neg logit contributions | |
Stub | -0.838 |
stra | -0.836 |
Amen | -0.821 |
Tel | -0.812 |
stop | -0.811 |
Token | upside |
Token ID | 17196 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
Stub | +0.061 |
stra | +0.060 |
Amen | +0.059 |
Tel | +0.058 |
stop | +0.058 |
Neg logit contributions | |
ithing | -0.057 |
andom | -0.050 |
nesota | -0.049 |
hing | -0.049 |
arning | -0.049 |
Token | down |
Token ID | 866 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ithing | +1.148 |
hing | +0.993 |
andom | +0.983 |
nesota | +0.974 |
arning | +0.968 |
Neg logit contributions | |
stra | -1.156 |
Stub | -1.153 |
Amen | -1.141 |
Tel | -1.128 |
reopen | -1.113 |
Token | , |
Token ID | 11 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
ithing | +1.749 |
hing | +1.529 |
andom | +1.523 |
nesota | +1.514 |
arning | +1.505 |
Neg logit contributions | |
Stub | -1.458 |
stra | -1.450 |
Amen | -1.424 |
stop | -1.408 |
Tel | -1.408 |
Token | teeth |
Token ID | 9941 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
Stub | +0.178 |
stra | +0.175 |
Amen | +0.172 |
reopen | +0.171 |
Tel | +0.170 |
Neg logit contributions | |
ithing | -0.168 |
andom | -0.149 |
nesota | -0.146 |
arning | -0.144 |
hing | -0.144 |
Token | Plus |
Token ID | 17860 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
Stub | +0.864 |
stra | +0.864 |
Tel | +0.841 |
stop | +0.839 |
Amen | +0.838 |
Neg logit contributions | |
ithing | -1.091 |
andom | -0.966 |
nesota | -0.955 |
arning | -0.946 |
hing | -0.946 |
Token | , |
Token ID | 11 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
ithing | +0.553 |
andom | +0.488 |
nesota | +0.486 |
hing | +0.483 |
arning | +0.479 |
Neg logit contributions | |
stra | -0.464 |
Stub | -0.461 |
Amen | -0.451 |
Tel | -0.449 |
stop | -0.446 |
Token | they |
Token ID | 484 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Stub | +0.348 |
stra | +0.343 |
Amen | +0.338 |
stop | +0.333 |
reopen | +0.332 |
Neg logit contributions | |
ithing | -0.376 |
andom | -0.332 |
nesota | -0.327 |
hing | -0.325 |
arning | -0.324 |
Token | had |
Token ID | 550 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Stub | +1.066 |
stra | +1.060 |
reopen | +1.039 |
Amen | +1.034 |
stop | +1.032 |
Neg logit contributions | |
ithing | -0.997 |
andom | -0.870 |
nesota | -0.861 |
hing | -0.846 |
arning | -0.842 |
Token | added |
Token ID | 2087 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
Stub | +0.906 |
stra | +0.895 |
Amen | +0.880 |
reopen | +0.868 |
stop | +0.867 |
Neg logit contributions | |
ithing | -0.938 |
andom | -0.823 |
nesota | -0.806 |
hing | -0.804 |
arning | -0.801 |
Token | two |
Token ID | 734 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
ithing | +0.126 |
andom | +0.111 |
nesota | +0.109 |
hing | +0.109 |
arning | +0.108 |
Neg logit contributions | |
Stub | -0.118 |
stra | -0.118 |
Amen | -0.115 |
stop | -0.114 |
Tel | -0.114 |
Token | fine |
Token ID | 3734 |
Feature activation | -3.85e-01 |
Pos logit contributions | |
Stub | +0.109 |
stra | +0.109 |
Amen | +0.107 |
stop | +0.105 |
Tel | +0.105 |
Neg logit contributions | |
ithing | -0.127 |
andom | -0.112 |
nesota | -0.111 |
hing | -0.110 |
arning | -0.110 |
Token | arms |
Token ID | 5101 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
ithing | +0.276 |
andom | +0.244 |
nesota | +0.241 |
hing | +0.240 |
arning | +0.238 |
Neg logit contributions | |
Stub | -0.243 |
stra | -0.242 |
Amen | -0.237 |
stop | -0.234 |
Tel | -0.233 |
Token | in |
Token ID | 287 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
ithing | +0.349 |
andom | +0.308 |
nesota | +0.305 |
hing | +0.304 |
arning | +0.302 |
Neg logit contributions | |
Stub | -0.310 |
stra | -0.308 |
Amen | -0.303 |
Tel | -0.300 |
stop | -0.299 |
Token | Brandon |
Token ID | 14328 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
Stub | +0.249 |
stra | +0.247 |
Amen | +0.242 |
stop | +0.239 |
Tel | +0.238 |
Neg logit contributions | |
ithing | -0.286 |
andom | -0.253 |
nesota | -0.249 |
hing | -0.248 |
arning | -0.247 |
Token | Ma |
Token ID | 6669 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Stub | +0.026 |
stra | +0.026 |
Amen | +0.025 |
stop | +0.025 |
Tel | +0.025 |
Neg logit contributions | |
ithing | -0.029 |
andom | -0.026 |
nesota | -0.025 |
hing | -0.025 |
arning | -0.025 |
Token | as |
Token ID | 355 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ithing | +0.185 |
andom | +0.163 |
nesota | +0.161 |
hing | +0.159 |
arning | +0.159 |
Neg logit contributions | |
Stub | -0.183 |
stra | -0.183 |
Amen | -0.179 |
stop | -0.177 |
reopen | -0.177 |
Token | inevitable |
Token ID | 13203 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
ithing | +1.554 |
andom | +1.361 |
hing | +1.345 |
nesota | +1.338 |
arning | +1.319 |
Neg logit contributions | |
stra | -1.609 |
Stub | -1.602 |
Amen | -1.571 |
Tel | -1.559 |
stop | -1.558 |
Token | design |
Token ID | 1486 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ithing | +1.261 |
andom | +1.114 |
nesota | +1.104 |
hing | +1.102 |
arning | +1.085 |
Neg logit contributions | |
stra | -1.123 |
Stub | -1.121 |
Amen | -1.091 |
stop | -1.087 |
Tel | -1.085 |
Token | changes |
Token ID | 2458 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
Stub | +0.665 |
stra | +0.662 |
Amen | +0.647 |
reopen | +0.638 |
stop | +0.637 |
Neg logit contributions | |
ithing | -0.803 |
andom | -0.710 |
nesota | -0.701 |
arning | -0.697 |
hing | -0.695 |
Token | are |
Token ID | 389 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
ithing | +0.736 |
andom | +0.640 |
nesota | +0.633 |
arning | +0.625 |
hing | +0.623 |
Neg logit contributions | |
Stub | -0.831 |
stra | -0.827 |
Amen | -0.811 |
reopen | -0.808 |
stop | -0.805 |
Token | forthcoming |
Token ID | 21220 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
ithing | +0.240 |
andom | +0.207 |
nesota | +0.204 |
arning | +0.200 |
hing | +0.197 |
Neg logit contributions | |
Stub | -0.297 |
stra | -0.295 |
Amen | -0.290 |
reopen | -0.287 |
Tel | -0.286 |
Token | before |
Token ID | 878 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ithing | +0.272 |
andom | +0.241 |
nesota | +0.238 |
hing | +0.237 |
arning | +0.236 |
Neg logit contributions | |
Stub | -0.222 |
stra | -0.221 |
Amen | -0.216 |
stop | -0.214 |
Tel | -0.213 |
Token | the |
Token ID | 262 |
Feature activation | -7.40e-02 |
Pos logit contributions | |
ithing | +0.740 |
andom | +0.649 |
hing | +0.639 |
nesota | +0.635 |
arning | +0.627 |
Neg logit contributions | |
Stub | -0.733 |
stra | -0.732 |
Amen | -0.713 |
stop | -0.713 |
Tel | -0.709 |
Token | development |
Token ID | 2478 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
ithing | +0.058 |
andom | +0.052 |
nesota | +0.051 |
hing | +0.051 |
arning | +0.050 |
Neg logit contributions | |
Stub | -0.042 |
stra | -0.041 |
Amen | -0.040 |
Tel | -0.040 |
stop | -0.040 |
Token | gets |
Token ID | 3011 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
Stub | +1.027 |
stra | +1.021 |
Amen | +1.004 |
reopen | +0.989 |
stop | +0.988 |
Neg logit contributions | |
ithing | -1.117 |
andom | -0.982 |
nesota | -0.970 |
arning | -0.964 |
hing | -0.963 |
Token | full |
Token ID | 1336 |
Feature activation | -5.18e+00 |
Pos logit contributions | |
ithing | +2.144 |
andom | +1.865 |
hing | +1.859 |
nesota | +1.821 |
arning | +1.792 |
Neg logit contributions | |
Stub | -2.670 |
stra | -2.666 |
Amen | -2.618 |
stop | -2.611 |
Tel | -2.607 |
Token | origins |
Token ID | 15587 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Stub | +0.195 |
stra | +0.194 |
Amen | +0.190 |
reopen | +0.188 |
stop | +0.188 |
Neg logit contributions | |
ithing | -0.207 |
andom | -0.182 |
nesota | -0.179 |
arning | -0.178 |
hing | -0.178 |
Token | of |
Token ID | 286 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
ithing | +0.687 |
andom | +0.606 |
hing | +0.601 |
nesota | +0.598 |
arning | +0.589 |
Neg logit contributions | |
Stub | -0.675 |
stra | -0.674 |
Amen | -0.660 |
Tel | -0.654 |
stop | -0.652 |
Token | K |
Token ID | 509 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
ithing | +0.018 |
andom | +0.016 |
nesota | +0.015 |
arning | +0.015 |
hing | +0.015 |
Neg logit contributions | |
Stub | -0.016 |
stra | -0.016 |
Amen | -0.016 |
stop | -0.016 |
reopen | -0.016 |
Token | ratom |
Token ID | 44616 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ithing | +0.144 |
andom | +0.117 |
nesota | +0.116 |
arning | +0.113 |
hing | +0.111 |
Neg logit contributions | |
stra | -0.271 |
Stub | -0.269 |
stop | -0.266 |
Amen | -0.265 |
Tel | -0.264 |
Token | and |
Token ID | 290 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
ithing | +1.354 |
andom | +1.185 |
hing | +1.168 |
nesota | +1.160 |
arning | +1.144 |
Neg logit contributions | |
Stub | -1.508 |
stra | -1.504 |
Amen | -1.477 |
Tel | -1.461 |
stop | -1.453 |
Token | its |
Token ID | 663 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
ithing | +0.438 |
andom | +0.380 |
nesota | +0.375 |
hing | +0.371 |
arning | +0.371 |
Neg logit contributions | |
Stub | -0.483 |
stra | -0.478 |
Amen | -0.472 |
stop | -0.467 |
reopen | -0.466 |
Token | many |
Token ID | 867 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ithing | +1.945 |
andom | +1.714 |
hing | +1.689 |
nesota | +1.686 |
arning | +1.664 |
Neg logit contributions | |
Stub | -1.932 |
stra | -1.929 |
Amen | -1.878 |
Tel | -1.876 |
stop | -1.862 |
Token | benefits |
Token ID | 4034 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
ithing | +1.058 |
andom | +0.929 |
nesota | +0.915 |
hing | +0.913 |
arning | +0.904 |
Neg logit contributions | |
Stub | -1.055 |
stra | -1.049 |
Amen | -1.028 |
Tel | -1.018 |
stop | -1.017 |
Token | with |
Token ID | 351 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
ithing | +0.177 |
andom | +0.155 |
nesota | +0.153 |
hing | +0.152 |
arning | +0.151 |
Neg logit contributions | |
Stub | -0.190 |
stra | -0.189 |
Amen | -0.185 |
Tel | -0.183 |
stop | -0.183 |
Token | respect |
Token ID | 2461 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
ithing | +0.315 |
andom | +0.277 |
nesota | +0.273 |
hing | +0.271 |
arning | +0.271 |
Neg logit contributions | |
Stub | -0.300 |
stra | -0.297 |
Amen | -0.292 |
stop | -0.288 |
reopen | -0.288 |
Token | , |
Token ID | 11 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Stub | +0.287 |
stra | +0.286 |
Amen | +0.280 |
Tel | +0.276 |
stop | +0.276 |
Neg logit contributions | |
ithing | -0.337 |
andom | -0.299 |
nesota | -0.294 |
hing | -0.293 |
arning | -0.291 |
Token | ming |
Token ID | 2229 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
ithing | +0.989 |
andom | +0.866 |
hing | +0.856 |
nesota | +0.854 |
arning | +0.848 |
Neg logit contributions | |
Stub | -0.956 |
stra | -0.945 |
Amen | -0.934 |
reopen | -0.921 |
Tel | -0.920 |
Token | led |
Token ID | 992 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
stra | +0.171 |
Stub | +0.170 |
stop | +0.166 |
Amen | +0.166 |
Tel | +0.165 |
Neg logit contributions | |
ithing | -0.159 |
andom | -0.139 |
nesota | -0.139 |
arning | -0.136 |
hing | -0.134 |
Token | as |
Token ID | 355 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
ithing | +0.297 |
andom | +0.262 |
nesota | +0.259 |
hing | +0.258 |
arning | +0.256 |
Neg logit contributions | |
Stub | -0.265 |
stra | -0.264 |
Amen | -0.258 |
stop | -0.255 |
Tel | -0.255 |
Token | new |
Token ID | 649 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
ithing | +0.038 |
andom | +0.033 |
nesota | +0.033 |
hing | +0.033 |
arning | +0.033 |
Neg logit contributions | |
Stub | -0.036 |
stra | -0.036 |
Amen | -0.035 |
stop | -0.035 |
Tel | -0.035 |
Token | values |
Token ID | 3815 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ithing | +1.706 |
andom | +1.512 |
hing | +1.491 |
nesota | +1.484 |
arning | +1.460 |
Neg logit contributions | |
Stub | -1.582 |
stra | -1.564 |
Amen | -1.535 |
Tel | -1.523 |
stop | -1.517 |
Token | are |
Token ID | 389 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
ithing | +0.814 |
andom | +0.716 |
nesota | +0.705 |
hing | +0.704 |
arning | +0.699 |
Neg logit contributions | |
Stub | -0.769 |
stra | -0.765 |
Amen | -0.749 |
Tel | -0.742 |
stop | -0.741 |
Token | created |
Token ID | 2727 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
ithing | +1.411 |
andom | +1.217 |
hing | +1.196 |
nesota | +1.195 |
arning | +1.183 |
Neg logit contributions | |
Stub | -1.734 |
stra | -1.730 |
Amen | -1.698 |
stop | -1.684 |
Tel | -1.682 |
Token | and |
Token ID | 290 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ithing | +1.572 |
andom | +1.387 |
hing | +1.368 |
nesota | +1.364 |
arning | +1.348 |
Neg logit contributions | |
Stub | -1.468 |
stra | -1.464 |
Amen | -1.426 |
Tel | -1.419 |
stop | -1.418 |
Token | old |
Token ID | 1468 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
ithing | +0.991 |
andom | +0.868 |
nesota | +0.853 |
hing | +0.851 |
arning | +0.844 |
Neg logit contributions | |
Stub | -1.011 |
stra | -1.008 |
Amen | -0.987 |
stop | -0.978 |
Tel | -0.978 |
Token | ones |
Token ID | 3392 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
Stub | +0.084 |
stra | +0.084 |
Amen | +0.082 |
stop | +0.081 |
Tel | +0.081 |
Neg logit contributions | |
ithing | -0.127 |
andom | -0.114 |
nesota | -0.112 |
hing | -0.112 |
arning | -0.112 |
Token | are |
Token ID | 389 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
ithing | +0.323 |
andom | +0.283 |
nesota | +0.281 |
hing | +0.278 |
arning | +0.277 |
Neg logit contributions | |
Stub | -0.310 |
stra | -0.307 |
Amen | -0.301 |
stop | -0.298 |
reopen | -0.298 |
Token | and |
Token ID | 290 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
stra | +0.853 |
Stub | +0.851 |
Amen | +0.830 |
stop | +0.823 |
Tel | +0.822 |
Neg logit contributions | |
ithing | -0.902 |
andom | -0.791 |
nesota | -0.783 |
hing | -0.775 |
arning | -0.774 |
Token | art |
Token ID | 1242 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
Stub | +1.193 |
stra | +1.187 |
Amen | +1.168 |
stop | +1.151 |
reopen | +1.150 |
Neg logit contributions | |
ithing | -1.305 |
andom | -1.139 |
nesota | -1.136 |
arning | -1.131 |
hing | -1.125 |
Token | are |
Token ID | 389 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
Stub | +0.209 |
stra | +0.209 |
Amen | +0.204 |
stop | +0.203 |
Tel | +0.202 |
Neg logit contributions | |
ithing | -0.208 |
andom | -0.182 |
nesota | -0.180 |
hing | -0.179 |
arning | -0.178 |
Token | often |
Token ID | 1690 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
Stub | +0.399 |
stra | +0.397 |
Amen | +0.390 |
reopen | +0.385 |
Tel | +0.385 |
Neg logit contributions | |
ithing | -0.412 |
andom | -0.361 |
nesota | -0.356 |
hing | -0.354 |
arning | -0.353 |
Token | closely |
Token ID | 7173 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
Stub | +1.488 |
stra | +1.477 |
Amen | +1.449 |
reopen | +1.436 |
Tel | +1.429 |
Neg logit contributions | |
ithing | -1.508 |
andom | -1.317 |
nesota | -1.300 |
arning | -1.291 |
hing | -1.285 |
Token | inter |
Token ID | 987 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
stra | +1.314 |
Stub | +1.314 |
Amen | +1.283 |
stop | +1.264 |
reopen | +1.264 |
Neg logit contributions | |
ithing | -1.457 |
andom | -1.286 |
nesota | -1.273 |
arning | -1.255 |
hing | -1.255 |
Token | ming |
Token ID | 2229 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
ithing | +0.989 |
andom | +0.866 |
hing | +0.856 |
nesota | +0.854 |
arning | +0.848 |
Neg logit contributions | |
Stub | -0.956 |
stra | -0.945 |
Amen | -0.934 |
reopen | -0.921 |
Tel | -0.920 |
Token | led |
Token ID | 992 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
stra | +0.171 |
Stub | +0.170 |
stop | +0.166 |
Amen | +0.166 |
Tel | +0.165 |
Neg logit contributions | |
ithing | -0.159 |
andom | -0.139 |
nesota | -0.139 |
arning | -0.136 |
hing | -0.134 |
Token | as |
Token ID | 355 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
ithing | +0.297 |
andom | +0.262 |
nesota | +0.259 |
hing | +0.258 |
arning | +0.256 |
Neg logit contributions | |
Stub | -0.265 |
stra | -0.264 |
Amen | -0.258 |
stop | -0.255 |
Tel | -0.255 |
Token | new |
Token ID | 649 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
ithing | +0.038 |
andom | +0.033 |
nesota | +0.033 |
hing | +0.033 |
arning | +0.033 |
Neg logit contributions | |
Stub | -0.036 |
stra | -0.036 |
Amen | -0.035 |
stop | -0.035 |
Tel | -0.035 |
Token | values |
Token ID | 3815 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ithing | +1.706 |
andom | +1.512 |
hing | +1.491 |
nesota | +1.484 |
arning | +1.460 |
Neg logit contributions | |
Stub | -1.582 |
stra | -1.564 |
Amen | -1.535 |
Tel | -1.523 |
stop | -1.517 |
Token | lid |
Token ID | 19789 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
ithing | +0.203 |
andom | +0.178 |
nesota | +0.176 |
hing | +0.174 |
arning | +0.174 |
Neg logit contributions | |
Stub | -0.190 |
stra | -0.189 |
Amen | -0.185 |
stop | -0.183 |
reopen | -0.183 |
Token | you |
Token ID | 345 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
Stub | +0.462 |
stra | +0.460 |
Amen | +0.450 |
stop | +0.444 |
Tel | +0.444 |
Neg logit contributions | |
ithing | -0.577 |
andom | -0.512 |
nesota | -0.506 |
hing | -0.504 |
arning | -0.501 |
Token | removed |
Token ID | 4615 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
ithing | +0.164 |
andom | +0.141 |
nesota | +0.139 |
arning | +0.136 |
hing | +0.136 |
Neg logit contributions | |
Stub | -0.209 |
stra | -0.208 |
reopen | -0.204 |
Amen | -0.203 |
stop | -0.203 |
Token | ). |
Token ID | 737 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
ithing | +0.225 |
andom | +0.198 |
hing | +0.196 |
nesota | +0.195 |
arning | +0.193 |
Neg logit contributions | |
Stub | -0.204 |
stra | -0.203 |
Amen | -0.200 |
Tel | -0.197 |
stop | -0.197 |
Token | This |
Token ID | 770 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
Stub | +0.346 |
stra | +0.343 |
Amen | +0.337 |
Tel | +0.334 |
stop | +0.334 |
Neg logit contributions | |
ithing | -0.354 |
andom | -0.311 |
nesota | -0.306 |
hing | -0.303 |
arning | -0.303 |
Token | can |
Token ID | 460 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
ithing | +0.657 |
andom | +0.564 |
nesota | +0.555 |
hing | +0.552 |
arning | +0.548 |
Neg logit contributions | |
Stub | -0.832 |
stra | -0.829 |
Amen | -0.814 |
stop | -0.806 |
Tel | -0.805 |
Token | be |
Token ID | 307 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
ithing | +1.059 |
hing | +0.899 |
andom | +0.898 |
nesota | +0.894 |
arning | +0.875 |
Neg logit contributions | |
Stub | -1.351 |
stra | -1.349 |
Amen | -1.327 |
Tel | -1.313 |
stop | -1.304 |
Token | a |
Token ID | 257 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
ithing | +0.496 |
andom | +0.420 |
nesota | +0.414 |
hing | +0.407 |
arning | +0.406 |
Neg logit contributions | |
Stub | -0.718 |
stra | -0.715 |
Amen | -0.703 |
reopen | -0.696 |
stop | -0.695 |
Token | pain |
Token ID | 2356 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
ithing | +0.438 |
andom | +0.386 |
nesota | +0.381 |
hing | +0.378 |
arning | +0.376 |
Neg logit contributions | |
Stub | -0.405 |
stra | -0.403 |
Amen | -0.394 |
stop | -0.390 |
Tel | -0.389 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ithing | +0.190 |
andom | +0.166 |
hing | +0.164 |
nesota | +0.164 |
arning | +0.163 |
Neg logit contributions | |
Stub | -0.181 |
stra | -0.180 |
Amen | -0.177 |
Tel | -0.175 |
stop | -0.174 |
Token | just |
Token ID | 655 |
Feature activation | -9.82e-01 |
Pos logit contributions | |
ithing | +0.759 |
andom | +0.665 |
nesota | +0.655 |
hing | +0.649 |
arning | +0.647 |
Neg logit contributions | |
Stub | -0.796 |
stra | -0.792 |
Amen | -0.777 |
stop | -0.770 |
reopen | -0.768 |
Token | at |
Token ID | 379 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
Stub | +1.249 |
stra | +1.245 |
Amen | +1.212 |
Tel | +1.200 |
stop | +1.199 |
Neg logit contributions | |
ithing | -1.363 |
andom | -1.202 |
nesota | -1.200 |
arning | -1.177 |
hing | -1.175 |
Token | the |
Token ID | 262 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
Stub | +1.612 |
stra | +1.599 |
Amen | +1.561 |
stop | +1.547 |
Tel | +1.545 |
Neg logit contributions | |
ithing | -1.274 |
nesota | -1.098 |
andom | -1.095 |
hing | -1.065 |
arning | -1.063 |
Token | heart |
Token ID | 2612 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
Stub | +1.237 |
stra | +1.227 |
Amen | +1.197 |
stop | +1.181 |
Tel | +1.177 |
Neg logit contributions | |
ithing | -1.799 |
andom | -1.610 |
nesota | -1.597 |
hing | -1.582 |
arning | -1.575 |
Token | of |
Token ID | 286 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
Stub | +2.247 |
stra | +2.247 |
Amen | +2.187 |
stop | +2.171 |
Tel | +2.157 |
Neg logit contributions | |
ithing | -2.725 |
andom | -2.419 |
nesota | -2.389 |
arning | -2.373 |
hing | -2.371 |
Token | this |
Token ID | 428 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
Stub | +2.101 |
stra | +2.090 |
Amen | +2.061 |
Tel | +2.004 |
stop | +2.002 |
Neg logit contributions | |
ithing | -2.417 |
nesota | -2.147 |
andom | -2.114 |
arning | -2.098 |
hing | -2.083 |
Token | trilogy |
Token ID | 26298 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
ithing | +0.076 |
andom | +0.067 |
nesota | +0.066 |
hing | +0.066 |
arning | +0.065 |
Neg logit contributions | |
Stub | -0.076 |
stra | -0.076 |
Amen | -0.074 |
stop | -0.073 |
Tel | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
ithing | +1.012 |
andom | +0.892 |
hing | +0.887 |
nesota | +0.875 |
arning | +0.872 |
Neg logit contributions | |
Stub | -0.853 |
stra | -0.845 |
Amen | -0.830 |
Tel | -0.821 |
reopen | -0.817 |
Token | you |
Token ID | 345 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Stub | +0.317 |
stra | +0.315 |
Amen | +0.309 |
stop | +0.306 |
Tel | +0.304 |
Neg logit contributions | |
ithing | -0.333 |
andom | -0.293 |
nesota | -0.291 |
hing | -0.287 |
arning | -0.287 |
Token | should |
Token ID | 815 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
stra | +0.980 |
Stub | +0.979 |
Amen | +0.957 |
stop | +0.955 |
reopen | +0.949 |
Neg logit contributions | |
ithing | -0.830 |
andom | -0.720 |
nesota | -0.710 |
hing | -0.699 |
arning | -0.696 |
Token | pl |
Token ID | 458 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
Stub | +1.041 |
stra | +1.037 |
Amen | +1.016 |
stop | +1.008 |
reopen | +1.005 |
Neg logit contributions | |
ithing | -1.033 |
andom | -0.905 |
nesota | -0.895 |
hing | -0.887 |
arning | -0.883 |
Token | ump |
Token ID | 931 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
ithing | +0.926 |
andom | +0.814 |
hing | +0.799 |
arning | +0.791 |
nesota | +0.788 |
Neg logit contributions | |
Stub | -0.976 |
Amen | -0.960 |
stra | -0.959 |
reopen | -0.943 |
Tel | -0.943 |
Token | in |
Token ID | 287 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Stub | +0.216 |
stra | +0.214 |
Amen | +0.210 |
stop | +0.207 |
Tel | +0.207 |
Neg logit contributions | |
ithing | -0.246 |
andom | -0.217 |
nesota | -0.214 |
hing | -0.213 |
arning | -0.212 |
Token | conflict |
Token ID | 5358 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
Stub | +0.612 |
stra | +0.608 |
Amen | +0.593 |
stop | +0.586 |
Tel | +0.585 |
Neg logit contributions | |
ithing | -0.892 |
andom | -0.799 |
nesota | -0.790 |
hing | -0.787 |
arning | -0.782 |
Token | , |
Token ID | 11 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
Stub | +1.774 |
stra | +1.751 |
Amen | +1.732 |
Tel | +1.701 |
stop | +1.693 |
Neg logit contributions | |
ithing | -2.064 |
andom | -1.828 |
arning | -1.795 |
nesota | -1.793 |
hing | -1.774 |
Token | Z |
Token ID | 1168 |
Feature activation | -3.96e-01 |
Pos logit contributions | |
Stub | +0.524 |
stra | +0.518 |
Amen | +0.512 |
stop | +0.502 |
Tel | +0.502 |
Neg logit contributions | |
ithing | -0.581 |
andom | -0.514 |
nesota | -0.506 |
hing | -0.502 |
arning | -0.501 |
Token | ain |
Token ID | 391 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
ithing | +0.270 |
andom | +0.236 |
nesota | +0.233 |
hing | +0.232 |
arning | +0.231 |
Neg logit contributions | |
Stub | -0.264 |
stra | -0.263 |
Amen | -0.257 |
stop | -0.255 |
Tel | -0.254 |
Token | ab |
Token ID | 397 |
Feature activation | -5.81e+00 |
Pos logit contributions | |
Stub | +0.375 |
stra | +0.373 |
Amen | +0.366 |
Tel | +0.362 |
stop | +0.361 |
Neg logit contributions | |
ithing | -0.391 |
andom | -0.344 |
nesota | -0.339 |
hing | -0.337 |
arning | -0.336 |
Token | Haw |
Token ID | 6091 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
ithing | +4.205 |
andom | +3.758 |
hing | +3.707 |
arning | +3.647 |
hai | +3.636 |
Neg logit contributions | |
stra | -3.407 |
Stub | -3.375 |
reopen | -3.374 |
stop | -3.300 |
Tel | -3.289 |
Token | a |
Token ID | 64 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
stra | +0.776 |
Stub | +0.775 |
Amen | +0.759 |
stop | +0.751 |
Tel | +0.748 |
Neg logit contributions | |
ithing | -0.840 |
andom | -0.741 |
nesota | -0.734 |
hing | -0.726 |
arning | -0.722 |
Token | Bang |
Token ID | 9801 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
Stub | +0.394 |
stra | +0.392 |
Amen | +0.384 |
stop | +0.378 |
Tel | +0.378 |
Neg logit contributions | |
ithing | -0.471 |
andom | -0.418 |
nesota | -0.413 |
arning | -0.408 |
hing | -0.408 |
Token | ura |
Token ID | 5330 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
ithing | +0.504 |
andom | +0.443 |
hing | +0.439 |
nesota | +0.437 |
arning | +0.434 |
Neg logit contributions | |
Stub | -0.459 |
stra | -0.453 |
Amen | -0.445 |
reopen | -0.443 |
stop | -0.441 |
Token | , |
Token ID | 11 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
Stub | +0.225 |
stra | +0.224 |
Amen | +0.219 |
stop | +0.216 |
Tel | +0.216 |
Neg logit contributions | |
ithing | -0.290 |
andom | -0.258 |
nesota | -0.255 |
hing | -0.255 |
arning | -0.253 |
Token | away |
Token ID | 1497 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
ithing | +1.847 |
nesota | +1.600 |
hing | +1.590 |
andom | +1.573 |
arning | +1.559 |
Neg logit contributions | |
Stub | -1.961 |
stra | -1.945 |
Amen | -1.932 |
Tel | -1.898 |
reopen | -1.892 |
Token | . |
Token ID | 13 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
Stub | +0.447 |
stra | +0.445 |
Amen | +0.435 |
stop | +0.430 |
reopen | +0.429 |
Neg logit contributions | |
ithing | -0.576 |
andom | -0.515 |
nesota | -0.508 |
hing | -0.503 |
arning | -0.501 |
Token |
|
Token ID | 198 |
Feature activation | -9.31e-01 |
Pos logit contributions | |
Stub | +0.373 |
stra | +0.370 |
Amen | +0.365 |
stop | +0.358 |
Tel | +0.358 |
Neg logit contributions | |
ithing | -0.379 |
andom | -0.333 |
nesota | -0.327 |
arning | -0.325 |
hing | -0.325 |
Token |
|
Token ID | 198 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
ithing | +0.680 |
andom | +0.600 |
nesota | +0.591 |
arning | +0.588 |
ammy | +0.580 |
Neg logit contributions | |
Stub | -0.588 |
stra | -0.586 |
Amen | -0.573 |
stop | -0.572 |
Tel | -0.571 |
Token | The |
Token ID | 464 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
ithing | +0.594 |
andom | +0.526 |
nesota | +0.520 |
hing | +0.519 |
arning | +0.515 |
Neg logit contributions | |
Stub | -0.503 |
stra | -0.499 |
Amen | -0.490 |
reopen | -0.484 |
stop | -0.482 |
Token | same |
Token ID | 976 |
Feature activation | -5.95e+00 |
Pos logit contributions | |
ithing | +0.056 |
andom | +0.050 |
nesota | +0.049 |
hing | +0.049 |
arning | +0.049 |
Neg logit contributions | |
Stub | -0.049 |
stra | -0.049 |
Amen | -0.048 |
stop | -0.047 |
Tel | -0.047 |
Token | cannot |
Token ID | 2314 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
ithing | +3.418 |
nesota | +2.993 |
andom | +2.929 |
hing | +2.927 |
arning | +2.862 |
Neg logit contributions | |
stra | -4.367 |
Stub | -4.305 |
Amen | -4.278 |
Tel | -4.272 |
ospace | -4.270 |
Token | be |
Token ID | 307 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
ithing | +1.280 |
andom | +1.068 |
nesota | +1.058 |
hing | +1.056 |
arning | +1.039 |
Neg logit contributions | |
Stub | -1.963 |
stra | -1.954 |
Amen | -1.921 |
Tel | -1.893 |
stop | -1.879 |
Token | said |
Token ID | 531 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
ithing | +1.785 |
andom | +1.538 |
hing | +1.521 |
nesota | +1.506 |
arning | +1.494 |
Neg logit contributions | |
Stub | -2.426 |
stra | -2.399 |
Amen | -2.366 |
reopen | -2.342 |
Tel | -2.338 |
Token | for |
Token ID | 329 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
ithing | +2.122 |
nesota | +1.842 |
andom | +1.836 |
hing | +1.834 |
arning | +1.788 |
Neg logit contributions | |
Stub | -2.811 |
stra | -2.799 |
reopen | -2.735 |
Amen | -2.733 |
Tel | -2.705 |
Token | B |
Token ID | 347 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
ithing | +1.755 |
andom | +1.541 |
nesota | +1.536 |
hing | +1.528 |
arning | +1.512 |
Neg logit contributions | |
Stub | -1.526 |
stra | -1.525 |
Tel | -1.487 |
Amen | -1.487 |
stop | -1.483 |
Token | searched |
Token ID | 16499 |
Feature activation | -3.19e+00 |
Pos logit contributions | |
ithing | +1.872 |
andom | +1.668 |
hing | +1.636 |
nesota | +1.634 |
arning | +1.619 |
Neg logit contributions | |
Stub | -1.609 |
stra | -1.605 |
Amen | -1.562 |
Tel | -1.557 |
stop | -1.551 |
Token | tiles |
Token ID | 19867 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ithing | +2.482 |
andom | +2.231 |
hing | +2.202 |
nesota | +2.181 |
arning | +2.167 |
Neg logit contributions | |
Stub | -1.777 |
stra | -1.770 |
Amen | -1.708 |
stop | -1.704 |
Tel | -1.703 |
Token | is |
Token ID | 318 |
Feature activation | +4.06e-01 |
Pos logit contributions | |
ithing | +0.845 |
andom | +0.736 |
nesota | +0.728 |
hing | +0.722 |
arning | +0.718 |
Neg logit contributions | |
Stub | -0.896 |
stra | -0.893 |
Amen | -0.875 |
stop | -0.866 |
Tel | -0.864 |
Token | the |
Token ID | 262 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
Stub | +0.267 |
stra | +0.265 |
Amen | +0.259 |
reopen | +0.256 |
stop | +0.256 |
Neg logit contributions | |
ithing | -0.282 |
nesota | -0.247 |
andom | -0.245 |
hing | -0.242 |
arning | -0.242 |
Token | list |
Token ID | 1351 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Stub | +0.381 |
stra | +0.378 |
Amen | +0.371 |
stop | +0.368 |
reopen | +0.366 |
Neg logit contributions | |
ithing | -0.408 |
andom | -0.357 |
nesota | -0.356 |
hing | -0.352 |
arning | -0.350 |
Token | search |
Token ID | 2989 |
Feature activation | -6.40e+00 |
Pos logit contributions | |
ithing | +1.063 |
andom | +0.941 |
hing | +0.923 |
nesota | +0.920 |
arning | +0.913 |
Neg logit contributions | |
Stub | -0.988 |
stra | -0.981 |
Amen | -0.964 |
Tel | -0.952 |
reopen | -0.949 |
Token | Space |
Token ID | 14106 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ithing | +4.267 |
andom | +3.843 |
hing | +3.776 |
nesota | +3.696 |
arning | +3.649 |
Neg logit contributions | |
Stub | -4.056 |
Amen | -4.042 |
stra | -3.981 |
Tel | -3.926 |
reopen | -3.891 |
Token | ; |
Token ID | 2162 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
ithing | +2.368 |
andom | +2.117 |
hing | +2.079 |
nesota | +2.060 |
arning | +2.046 |
Neg logit contributions | |
Stub | -2.063 |
stra | -2.045 |
Amen | -2.018 |
reopen | -1.990 |
stop | -1.975 |
Token | an |
Token ID | 281 |
Feature activation | -7.13e-01 |
Pos logit contributions | |
ithing | +2.520 |
andom | +2.229 |
hing | +2.184 |
nesota | +2.149 |
arning | +2.147 |
Neg logit contributions | |
stra | -2.532 |
Stub | -2.520 |
Amen | -2.466 |
Tel | -2.443 |
stop | -2.430 |
Token | association |
Token ID | 8112 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
ithing | +0.491 |
andom | +0.431 |
nesota | +0.426 |
hing | +0.423 |
arning | +0.421 |
Neg logit contributions | |
Stub | -0.469 |
stra | -0.467 |
Amen | -0.458 |
stop | -0.453 |
Tel | -0.452 |
Token | between |
Token ID | 1022 |
Feature activation | -3.87e+00 |
Pos logit contributions | |
Stub | +0.152 |
stra | +0.151 |
Amen | +0.148 |
stop | +0.147 |
Tel | +0.147 |
Neg logit contributions | |
ithing | -0.157 |
andom | -0.137 |
nesota | -0.136 |
hing | -0.135 |
arning | -0.134 |
Token | TL |
Token ID | 24811 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
ithing | +1.881 |
andom | +1.655 |
nesota | +1.640 |
hing | +1.631 |
arning | +1.617 |
Neg logit contributions | |
stra | -1.719 |
Stub | -1.709 |
Amen | -1.671 |
reopen | -1.665 |
stop | -1.662 |
Token | AD |
Token ID | 2885 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ithing | +2.075 |
andom | +1.832 |
nesota | +1.802 |
hing | +1.801 |
arning | +1.795 |
Neg logit contributions | |
stra | -1.801 |
Stub | -1.772 |
Amen | -1.764 |
reopen | -1.738 |
ospace | -1.723 |
Token | T |
Token ID | 51 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
ithing | +0.953 |
andom | +0.824 |
nesota | +0.823 |
hing | +0.807 |
arning | +0.806 |
Neg logit contributions | |
Stub | -1.025 |
stra | -1.017 |
Amen | -0.997 |
stop | -0.996 |
Tel | -0.991 |
Token | 235 |
Token ID | 28878 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
ithing | +1.135 |
andom | +1.003 |
nesota | +0.991 |
hing | +0.986 |
arning | +0.980 |
Neg logit contributions | |
Stub | -0.988 |
stra | -0.983 |
Amen | -0.962 |
stop | -0.951 |
Tel | -0.950 |
Token | 21 |
Token ID | 2481 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
ithing | +0.597 |
andom | +0.517 |
nesota | +0.510 |
hing | +0.507 |
arning | +0.504 |
Neg logit contributions | |
Stub | -0.670 |
stra | -0.666 |
Amen | -0.654 |
stop | -0.647 |
Tel | -0.647 |
Token | Posts |
Token ID | 12043 |
Feature activation | -8.13e+00 |
Pos logit contributions | |
ithing | +0.054 |
andom | +0.042 |
nesota | +0.041 |
arning | +0.040 |
hing | +0.040 |
Neg logit contributions | |
Stub | -0.135 |
Amen | -0.132 |
stra | -0.131 |
Tel | -0.131 |
stop | -0.130 |
Token | # |
Token ID | 1303 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
ithing | +3.897 |
andom | +3.219 |
arning | +3.012 |
hing | +3.005 |
ammy | +2.941 |
Neg logit contributions | |
stra | -6.796 |
Tel | -6.644 |
ospace | -6.596 |
stop | -6.589 |
Stub | -6.515 |
Token | 18 |
Token ID | 1507 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
ithing | +2.636 |
andom | +2.187 |
hing | +2.151 |
nesota | +2.130 |
arning | +2.109 |
Neg logit contributions | |
Stub | -4.902 |
stra | -4.873 |
Amen | -4.801 |
reopen | -4.768 |
Tel | -4.749 |
Token | Good |
Token ID | 4599 |
Feature activation | +8.00e-02 |
Pos logit contributions | |
ithing | +0.914 |
andom | +0.797 |
nesota | +0.787 |
hing | +0.785 |
arning | +0.777 |
Neg logit contributions | |
stra | -0.949 |
Stub | -0.948 |
Amen | -0.925 |
stop | -0.919 |
Tel | -0.919 |
Token | stuff |
Token ID | 3404 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Stub | +0.047 |
stra | +0.047 |
Amen | +0.045 |
Tel | +0.045 |
stop | +0.045 |
Neg logit contributions | |
ithing | -0.061 |
andom | -0.054 |
nesota | -0.054 |
hing | -0.053 |
arning | -0.053 |
Token | ! |
Token ID | 0 |
Feature activation | -5.73e-01 |
Pos logit contributions | |
Stub | +0.706 |
stra | +0.701 |
Amen | +0.687 |
stop | +0.678 |
Tel | +0.676 |
Neg logit contributions | |
ithing | -0.835 |
andom | -0.739 |
nesota | -0.730 |
hing | -0.725 |
arning | -0.723 |
Token | U |
Token ID | 52 |
Feature activation | +6.09e-01 |
Pos logit contributions | |
ithing | +0.623 |
andom | +0.550 |
nesota | +0.544 |
hing | +0.540 |
arning | +0.539 |
Neg logit contributions | |
Stub | -0.537 |
stra | -0.535 |
Amen | -0.523 |
Tel | -0.518 |
stop | -0.518 |
Token | students |
Token ID | 2444 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Stub | +0.351 |
stra | +0.350 |
Amen | +0.342 |
stop | +0.337 |
Tel | +0.337 |
Neg logit contributions | |
ithing | -0.470 |
andom | -0.418 |
nesota | -0.414 |
hing | -0.410 |
arning | -0.409 |
Token | both |
Token ID | 1111 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
ithing | +0.837 |
andom | +0.732 |
hing | +0.726 |
nesota | +0.726 |
arning | +0.719 |
Neg logit contributions | |
Stub | -0.809 |
stra | -0.804 |
Amen | -0.788 |
Tel | -0.782 |
stop | -0.778 |
Token | inside |
Token ID | 2641 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
Stub | +0.049 |
stra | +0.049 |
Amen | +0.048 |
reopen | +0.047 |
stop | +0.047 |
Neg logit contributions | |
ithing | -0.053 |
andom | -0.047 |
nesota | -0.046 |
hing | -0.046 |
arning | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
ithing | +1.645 |
andom | +1.447 |
hing | +1.422 |
nesota | +1.410 |
arning | +1.394 |
Neg logit contributions | |
stra | -1.663 |
Stub | -1.661 |
Amen | -1.635 |
Tel | -1.621 |
stop | -1.620 |
Token | outside |
Token ID | 2354 |
Feature activation | -6.17e+00 |
Pos logit contributions | |
ithing | +0.493 |
andom | +0.436 |
nesota | +0.431 |
hing | +0.429 |
arning | +0.426 |
Neg logit contributions | |
Stub | -0.434 |
stra | -0.431 |
Amen | -0.423 |
stop | -0.417 |
Tel | -0.417 |
Token | the |
Token ID | 262 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ithing | +3.713 |
andom | +3.301 |
hing | +3.230 |
nesota | +3.151 |
arning | +3.128 |
Neg logit contributions | |
stra | -4.205 |
Stub | -4.148 |
Amen | -4.146 |
Tel | -4.145 |
erion | -4.134 |
Token | classroom |
Token ID | 15806 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
ithing | +1.146 |
andom | +1.007 |
nesota | +0.992 |
hing | +0.989 |
arning | +0.982 |
Neg logit contributions | |
Stub | -1.091 |
stra | -1.087 |
Amen | -1.065 |
stop | -1.054 |
Tel | -1.053 |
Token | . |
Token ID | 13 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
ithing | +0.406 |
andom | +0.357 |
hing | +0.353 |
nesota | +0.353 |
arning | +0.350 |
Neg logit contributions | |
Stub | -0.368 |
stra | -0.366 |
Amen | -0.358 |
Tel | -0.355 |
stop | -0.354 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ithing | +1.323 |
andom | +1.150 |
nesota | +1.136 |
hing | +1.131 |
arning | +1.119 |
Neg logit contributions | |
stra | -1.458 |
Stub | -1.454 |
Amen | -1.420 |
stop | -1.414 |
reopen | -1.414 |
Token | Five |
Token ID | 20029 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
Stub | +0.062 |
stra | +0.062 |
Amen | +0.060 |
stop | +0.059 |
Tel | +0.059 |
Neg logit contributions | |
ithing | -0.077 |
andom | -0.068 |
nesota | -0.067 |
hing | -0.067 |
arning | -0.067 |