Token | his |
Token ID | 465 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ratulations | +1.308 |
chart | +1.242 |
claw | +1.235 |
hess | +1.123 |
=~ | +1.114 |
Neg logit contributions | |
Noise | -1.309 |
Faculty | -1.213 |
grad | -1.201 |
Madness | -1.176 |
graduate | -1.109 |
Token | initial |
Token ID | 4238 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
ratulations | +1.288 |
chart | +1.217 |
claw | +1.210 |
hess | +1.095 |
=~ | +1.085 |
Neg logit contributions | |
Noise | -1.423 |
Faculty | -1.322 |
grad | -1.311 |
Madness | -1.284 |
graduate | -1.214 |
Token | pick |
Token ID | 2298 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Noise | +0.159 |
Faculty | +0.145 |
grad | +0.145 |
Madness | +0.141 |
graduate | +0.132 |
Neg logit contributions | |
ratulations | -0.192 |
chart | -0.181 |
claw | -0.181 |
hess | -0.166 |
=~ | -0.166 |
Token | from |
Token ID | 422 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ratulations | +1.144 |
chart | +1.084 |
claw | +1.080 |
hess | +0.993 |
=~ | +0.984 |
Neg logit contributions | |
Noise | -0.943 |
Faculty | -0.866 |
grad | -0.854 |
Madness | -0.837 |
graduate | -0.779 |
Token | Lone |
Token ID | 23405 |
Feature activation | +2.93e+00 |
Pos logit contributions | |
ratulations | +1.505 |
chart | +1.440 |
claw | +1.432 |
hess | +1.307 |
=~ | +1.297 |
Neg logit contributions | |
Noise | -1.335 |
Faculty | -1.233 |
grad | -1.216 |
Madness | -1.187 |
graduate | -1.115 |
Token | Druid |
Token ID | 32685 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
ratulations | +2.091 |
claw | +2.024 |
chart | +1.994 |
=~ | +1.788 |
hess | +1.784 |
Neg logit contributions | |
Noise | -2.355 |
Faculty | -2.207 |
grad | -2.194 |
Madness | -2.117 |
graduate | -2.022 |
Token | to |
Token ID | 284 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
claw | +7.659 |
ratulations | +7.616 |
chart | +7.450 |
hess | +6.804 |
=~ | +6.756 |
Neg logit contributions | |
Noise | -7.480 |
Faculty | -7.039 |
grad | -6.826 |
Madness | -6.524 |
Amend | -6.423 |
Token | Templar |
Token ID | 41741 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
ratulations | +2.031 |
chart | +1.945 |
claw | +1.936 |
hess | +1.769 |
=~ | +1.756 |
Neg logit contributions | |
Noise | -1.761 |
Faculty | -1.628 |
grad | -1.605 |
Madness | -1.561 |
graduate | -1.470 |
Token | Assassin |
Token ID | 21401 |
Feature activation | +5.18e+00 |
Pos logit contributions | |
ratulations | +2.554 |
chart | +2.411 |
claw | +2.400 |
hess | +2.090 |
=~ | +2.050 |
Neg logit contributions | |
Noise | -4.629 |
Faculty | -4.378 |
grad | -4.331 |
Madness | -4.210 |
graduate | -4.069 |
Token | . |
Token ID | 13 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ratulations | +4.078 |
chart | +3.944 |
claw | +3.941 |
hess | +3.560 |
=~ | +3.547 |
Neg logit contributions | |
Noise | -3.754 |
Faculty | -3.493 |
grad | -3.410 |
Madness | -3.303 |
graduate | -3.144 |
Token | He |
Token ID | 679 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ratulations | +0.933 |
chart | +0.887 |
claw | +0.882 |
hess | +0.807 |
=~ | +0.798 |
Neg logit contributions | |
Noise | -0.820 |
Faculty | -0.756 |
grad | -0.744 |
Madness | -0.731 |
graduate | -0.682 |
Token | _ |
Token ID | 62 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ratulations | +0.856 |
chart | +0.812 |
claw | +0.808 |
hess | +0.742 |
=~ | +0.738 |
Neg logit contributions | |
Noise | -0.689 |
Faculty | -0.632 |
grad | -0.628 |
Madness | -0.609 |
graduate | -0.572 |
Token | re |
Token ID | 260 |
Feature activation | -3.12e-03 |
Pos logit contributions | |
ratulations | +1.060 |
chart | +1.024 |
claw | +1.016 |
hess | +0.924 |
=~ | +0.913 |
Neg logit contributions | |
Noise | -0.949 |
Faculty | -0.878 |
grad | -0.855 |
Madness | -0.847 |
graduate | -0.788 |
Token | pos |
Token ID | 1930 |
Feature activation | +7.83e+00 |
Pos logit contributions | |
Noise | +0.002 |
grad | +0.002 |
Faculty | +0.002 |
Madness | +0.002 |
graduate | +0.002 |
Neg logit contributions | |
ratulations | -0.003 |
chart | -0.002 |
claw | -0.002 |
=~ | -0.002 |
hess | -0.002 |
Token | . |
Token ID | 764 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
chart | +5.463 |
ratulations | +5.434 |
claw | +5.311 |
=~ | +4.777 |
hess | +4.757 |
Neg logit contributions | |
Noise | -6.098 |
Faculty | -5.676 |
Madness | -5.561 |
grad | -5.455 |
Amend | -5.190 |
Token | contents |
Token ID | 10154 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
Noise | +1.488 |
grad | +1.378 |
Faculty | +1.354 |
Madness | +1.287 |
graduate | +1.213 |
Neg logit contributions | |
ratulations | -2.360 |
chart | -2.256 |
claw | -2.254 |
hess | -2.084 |
=~ | -2.054 |
Token | ] |
Token ID | 2361 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
ratulations | +2.895 |
chart | +2.793 |
claw | +2.766 |
hess | +2.492 |
=~ | +2.446 |
Neg logit contributions | |
Noise | -3.347 |
Faculty | -3.118 |
grad | -3.035 |
Madness | -3.015 |
graduate | -2.850 |
Token | WHERE |
Token ID | 33411 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
ratulations | +6.141 |
chart | +6.049 |
claw | +5.843 |
=~ | +5.293 |
hess | +5.236 |
Neg logit contributions | |
Noise | -8.938 |
Faculty | -8.415 |
grad | -8.212 |
Madness | -8.042 |
bris | -7.819 |
Token | id |
Token ID | 4686 |
Feature activation | +4.29e+00 |
Pos logit contributions | |
ratulations | +3.468 |
chart | +3.313 |
claw | +3.248 |
=~ | +2.967 |
hess | +2.942 |
Neg logit contributions | |
Noise | -4.046 |
Faculty | -3.778 |
grad | -3.709 |
Madness | -3.668 |
graduate | -3.445 |
Token | IN |
Token ID | 3268 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
ratulations | +3.101 |
chart | +2.972 |
claw | +2.943 |
hess | +2.657 |
=~ | +2.649 |
Neg logit contributions | |
Noise | -3.452 |
Faculty | -3.210 |
grad | -3.167 |
Madness | -3.111 |
graduate | -2.944 |
Token | ( |
Token ID | 357 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
ratulations | +2.209 |
chart | +2.125 |
claw | +2.115 |
hess | +1.957 |
=~ | +1.943 |
Neg logit contributions | |
Noise | -1.381 |
Faculty | -1.251 |
grad | -1.230 |
Madness | -1.195 |
graduate | -1.102 |
Token | SELECT |
Token ID | 33493 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
ratulations | +2.400 |
chart | +2.309 |
claw | +2.282 |
hess | +2.095 |
=~ | +2.082 |
Neg logit contributions | |
Noise | -2.061 |
Faculty | -1.898 |
grad | -1.856 |
Madness | -1.827 |
graduate | -1.705 |
Token | Nigel |
Token ID | 28772 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
Noise | +4.861 |
Faculty | +4.517 |
Madness | +4.293 |
grad | +4.176 |
Amend | +3.877 |
Neg logit contributions | |
ratulations | -5.924 |
chart | -5.548 |
claw | -5.524 |
hess | -5.016 |
=~ | -5.015 |
Token | Farage |
Token ID | 33904 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ratulations | +1.065 |
chart | +1.011 |
claw | +1.001 |
hess | +0.896 |
=~ | +0.891 |
Neg logit contributions | |
Noise | -1.298 |
Faculty | -1.210 |
grad | -1.196 |
Madness | -1.176 |
graduate | -1.113 |
Token | : |
Token ID | 25 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
ratulations | +1.544 |
chart | +1.479 |
claw | +1.472 |
hess | +1.349 |
=~ | +1.332 |
Neg logit contributions | |
Noise | -1.268 |
Faculty | -1.166 |
grad | -1.145 |
Madness | -1.121 |
graduate | -1.044 |
Token | " |
Token ID | 366 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
ratulations | +1.620 |
chart | +1.543 |
claw | +1.533 |
hess | +1.402 |
=~ | +1.387 |
Neg logit contributions | |
Noise | -1.474 |
Faculty | -1.368 |
grad | -1.361 |
Madness | -1.314 |
graduate | -1.243 |
Token | Lord |
Token ID | 22438 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
Noise | +0.101 |
Faculty | +0.093 |
grad | +0.092 |
Madness | +0.090 |
graduate | +0.085 |
Neg logit contributions | |
ratulations | -0.104 |
chart | -0.099 |
claw | -0.098 |
hess | -0.090 |
=~ | -0.089 |
Token | Hes |
Token ID | 44773 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
ratulations | +1.510 |
chart | +1.446 |
claw | +1.439 |
hess | +1.315 |
=~ | +1.304 |
Neg logit contributions | |
Noise | -1.277 |
Faculty | -1.175 |
grad | -1.159 |
Madness | -1.130 |
graduate | -1.060 |
Token | elt |
Token ID | 2120 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
chart | +6.592 |
ratulations | +6.584 |
claw | +6.498 |
hess | +5.855 |
=~ | +5.474 |
Neg logit contributions | |
Noise | -8.463 |
Faculty | -7.700 |
grad | -7.569 |
Madness | -7.446 |
Amend | -7.185 |
Token | ine |
Token ID | 500 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Noise | +1.461 |
Faculty | +1.344 |
grad | +1.341 |
Madness | +1.272 |
graduate | +1.218 |
Neg logit contributions | |
ratulations | -1.749 |
claw | -1.659 |
chart | -1.643 |
=~ | -1.502 |
hess | -1.479 |
Token | fears |
Token ID | 10251 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
ratulations | +1.247 |
chart | +1.191 |
claw | +1.185 |
hess | +1.078 |
=~ | +1.067 |
Neg logit contributions | |
Noise | -1.180 |
Faculty | -1.092 |
grad | -1.077 |
Madness | -1.053 |
graduate | -0.990 |
Token | what |
Token ID | 644 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ratulations | +2.691 |
claw | +2.581 |
chart | +2.578 |
hess | +2.352 |
=~ | +2.322 |
Neg logit contributions | |
Noise | -2.386 |
Faculty | -2.210 |
grad | -2.162 |
Madness | -2.115 |
graduate | -1.974 |
Token | the |
Token ID | 262 |
Feature activation | +9.70e-01 |
Pos logit contributions | |
ratulations | +1.043 |
chart | +0.995 |
claw | +0.990 |
hess | +0.899 |
=~ | +0.890 |
Neg logit contributions | |
Noise | -1.027 |
Faculty | -0.952 |
grad | -0.939 |
Madness | -0.919 |
graduate | -0.865 |
Token | really |
Token ID | 1107 |
Feature activation | +6.36e-01 |
Pos logit contributions | |
Noise | +0.469 |
Faculty | +0.432 |
grad | +0.429 |
Madness | +0.416 |
graduate | +0.395 |
Neg logit contributions | |
ratulations | -0.554 |
chart | -0.528 |
claw | -0.524 |
hess | -0.484 |
=~ | -0.479 |
Token | wanted |
Token ID | 2227 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
ratulations | +0.485 |
chart | +0.461 |
claw | +0.458 |
hess | +0.417 |
=~ | +0.413 |
Neg logit contributions | |
Noise | -0.495 |
Faculty | -0.459 |
grad | -0.454 |
Madness | -0.444 |
graduate | -0.420 |
Token | to |
Token ID | 284 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
Noise | +0.408 |
Faculty | +0.374 |
grad | +0.369 |
Madness | +0.360 |
graduate | +0.337 |
Neg logit contributions | |
ratulations | -0.510 |
chart | -0.488 |
claw | -0.485 |
hess | -0.445 |
=~ | -0.441 |
Token | include |
Token ID | 2291 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
Noise | +1.720 |
grad | +1.581 |
Faculty | +1.565 |
Madness | +1.523 |
graduate | +1.484 |
Neg logit contributions | |
ratulations | -2.038 |
chart | -1.912 |
claw | -1.902 |
hess | -1.763 |
=~ | -1.745 |
Token | from |
Token ID | 422 |
Feature activation | +8.71e-01 |
Pos logit contributions | |
ratulations | +0.346 |
chart | +0.330 |
claw | +0.328 |
hess | +0.301 |
=~ | +0.297 |
Neg logit contributions | |
Noise | -0.300 |
Faculty | -0.277 |
grad | -0.273 |
Madness | -0.267 |
graduate | -0.251 |
Token | release |
Token ID | 2650 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
ratulations | +0.707 |
chart | +0.675 |
claw | +0.671 |
hess | +0.612 |
=~ | +0.607 |
Neg logit contributions | |
Noise | -0.635 |
Faculty | -0.586 |
grad | -0.578 |
Madness | -0.565 |
graduate | -0.531 |
Token | - |
Token ID | 532 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
claw | +6.944 |
ratulations | +6.901 |
chart | +6.859 |
=~ | +6.093 |
hess | +6.078 |
Neg logit contributions | |
Noise | -7.349 |
Faculty | -7.014 |
grad | -6.621 |
Madness | -6.494 |
Amend | -6.383 |
Token | but |
Token ID | 475 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
Noise | +0.738 |
Faculty | +0.675 |
grad | +0.670 |
Madness | +0.652 |
graduate | +0.613 |
Neg logit contributions | |
ratulations | -0.876 |
chart | -0.832 |
claw | -0.828 |
hess | -0.763 |
=~ | -0.754 |
Token | because |
Token ID | 780 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
Noise | +1.135 |
Faculty | +1.036 |
grad | +1.032 |
Madness | +1.003 |
graduate | +0.945 |
Neg logit contributions | |
ratulations | -1.354 |
chart | -1.277 |
claw | -1.272 |
hess | -1.172 |
=~ | -1.157 |
Token | of |
Token ID | 286 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
Noise | +0.371 |
Faculty | +0.342 |
grad | +0.338 |
Madness | +0.330 |
graduate | +0.310 |
Neg logit contributions | |
ratulations | -0.417 |
chart | -0.398 |
claw | -0.396 |
hess | -0.362 |
=~ | -0.358 |
Token | scheduling |
Token ID | 26925 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
ratulations | +1.189 |
chart | +1.140 |
claw | +1.132 |
hess | +1.031 |
=~ | +1.024 |
Neg logit contributions | |
Noise | -1.105 |
Faculty | -1.025 |
grad | -1.009 |
Madness | -0.986 |
graduate | -0.928 |
Token | vim |
Token ID | 43907 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
ratulations | +1.127 |
chart | +1.082 |
claw | +1.071 |
hess | +0.977 |
=~ | +0.973 |
Neg logit contributions | |
Noise | -1.041 |
Faculty | -0.964 |
grad | -0.943 |
Madness | -0.928 |
graduate | -0.870 |
Token | 600 |
Token ID | 8054 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Noise | +0.390 |
Faculty | +0.361 |
grad | +0.354 |
Madness | +0.348 |
graduate | +0.326 |
Neg logit contributions | |
ratulations | -0.416 |
chart | -0.400 |
claw | -0.395 |
hess | -0.361 |
=~ | -0.359 |
Token | : |
Token ID | 25 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
Noise | +0.084 |
Faculty | +0.077 |
grad | +0.077 |
Madness | +0.075 |
graduate | +0.070 |
Neg logit contributions | |
ratulations | -0.093 |
claw | -0.088 |
chart | -0.088 |
hess | -0.081 |
=~ | -0.079 |
Token | set |
Token ID | 900 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
ratulations | +0.637 |
chart | +0.608 |
claw | +0.605 |
hess | +0.554 |
=~ | +0.549 |
Neg logit contributions | |
Noise | -0.541 |
Faculty | -0.498 |
grad | -0.492 |
Madness | -0.480 |
graduate | -0.450 |
Token | fold |
Token ID | 5591 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
ratulations | +0.847 |
chart | +0.812 |
claw | +0.806 |
hess | +0.736 |
=~ | +0.733 |
Neg logit contributions | |
Noise | -0.756 |
Faculty | -0.699 |
grad | -0.685 |
Madness | -0.673 |
graduate | -0.629 |
Token | method |
Token ID | 24396 |
Feature activation | +9.87e+00 |
Pos logit contributions | |
ratulations | +2.868 |
chart | +2.765 |
claw | +2.743 |
=~ | +2.498 |
hess | +2.488 |
Neg logit contributions | |
Noise | -2.916 |
Faculty | -2.733 |
grad | -2.665 |
Madness | -2.635 |
graduate | -2.476 |
Token | = |
Token ID | 28 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
ratulations | +7.072 |
chart | +6.969 |
claw | +6.899 |
=~ | +6.617 |
hess | +6.266 |
Neg logit contributions | |
Noise | -7.211 |
Faculty | -6.830 |
Madness | -6.584 |
grad | -6.422 |
Amend | -6.221 |
Token | mark |
Token ID | 4102 |
Feature activation | -9.97e-01 |
Pos logit contributions | |
ratulations | +1.699 |
chart | +1.631 |
claw | +1.621 |
hess | +1.470 |
=~ | +1.464 |
Neg logit contributions | |
Noise | -1.651 |
Faculty | -1.534 |
grad | -1.505 |
Madness | -1.479 |
graduate | -1.391 |
Token | er |
Token ID | 263 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
Noise | +0.725 |
Faculty | +0.665 |
grad | +0.665 |
Madness | +0.643 |
graduate | +0.606 |
Neg logit contributions | |
ratulations | -0.811 |
chart | -0.766 |
claw | -0.764 |
hess | -0.698 |
=~ | -0.688 |
Token | : |
Token ID | 25 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
ratulations | +2.109 |
chart | +2.029 |
claw | +2.013 |
hess | +1.829 |
=~ | +1.828 |
Neg logit contributions | |
Noise | -1.993 |
Faculty | -1.850 |
grad | -1.808 |
Madness | -1.780 |
graduate | -1.671 |
Token | */ |
Token ID | 9466 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
ratulations | +2.915 |
chart | +2.826 |
claw | +2.798 |
=~ | +2.545 |
hess | +2.539 |
Neg logit contributions | |
Noise | -2.665 |
Faculty | -2.483 |
grad | -2.402 |
Madness | -2.376 |
graduate | -2.222 |
Token | that |
Token ID | 326 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
ratulations | +1.055 |
chart | +1.009 |
claw | +1.003 |
hess | +0.917 |
=~ | +0.910 |
Neg logit contributions | |
Noise | -0.910 |
Faculty | -0.839 |
grad | -0.828 |
Madness | -0.808 |
graduate | -0.758 |
Token | to |
Token ID | 284 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
ratulations | +0.082 |
chart | +0.077 |
claw | +0.077 |
hess | +0.071 |
=~ | +0.070 |
Neg logit contributions | |
Noise | -0.072 |
Faculty | -0.067 |
grad | -0.066 |
Madness | -0.064 |
graduate | -0.061 |
Token | happen |
Token ID | 1645 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ratulations | +1.232 |
chart | +1.187 |
claw | +1.177 |
hess | +1.068 |
=~ | +1.063 |
Neg logit contributions | |
Noise | -1.073 |
Faculty | -0.984 |
grad | -0.961 |
Madness | -0.952 |
alum | -0.875 |
Token | ," |
Token ID | 553 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ratulations | +1.042 |
chart | +0.990 |
claw | +0.984 |
hess | +0.895 |
=~ | +0.889 |
Neg logit contributions | |
Noise | -1.024 |
Faculty | -0.950 |
grad | -0.939 |
Madness | -0.916 |
graduate | -0.867 |
Token | Donald |
Token ID | 3759 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
ratulations | +1.128 |
chart | +1.075 |
claw | +1.067 |
hess | +0.966 |
=~ | +0.956 |
Neg logit contributions | |
Noise | -1.192 |
Faculty | -1.107 |
grad | -1.094 |
Madness | -1.072 |
graduate | -1.011 |
Token | Ars |
Token ID | 24230 |
Feature activation | +9.42e+00 |
Pos logit contributions | |
ratulations | +1.814 |
chart | +1.727 |
claw | +1.708 |
hess | +1.545 |
=~ | +1.517 |
Neg logit contributions | |
Noise | -2.119 |
Faculty | -1.976 |
grad | -1.957 |
Madness | -1.924 |
graduate | -1.813 |
Token | ene |
Token ID | 1734 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ratulations | +5.033 |
chart | +4.895 |
claw | +4.825 |
hess | +4.323 |
enne | +4.145 |
Neg logit contributions | |
Noise | -9.066 |
Faculty | -8.565 |
grad | -8.481 |
Madness | -8.391 |
graduate | -7.921 |
Token | ault |
Token ID | 1721 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
Noise | +1.245 |
Faculty | +1.147 |
grad | +1.146 |
Madness | +1.130 |
graduate | +1.050 |
Neg logit contributions | |
ratulations | -1.807 |
claw | -1.698 |
chart | -1.688 |
=~ | -1.573 |
hess | -1.525 |
Token | said |
Token ID | 531 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
ratulations | +3.177 |
chart | +3.106 |
claw | +3.083 |
hess | +2.845 |
=~ | +2.798 |
Neg logit contributions | |
Noise | -2.149 |
Faculty | -1.946 |
grad | -1.909 |
Madness | -1.879 |
Amend | -1.709 |
Token | in |
Token ID | 287 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
ratulations | +0.340 |
chart | +0.323 |
claw | +0.322 |
hess | +0.295 |
=~ | +0.294 |
Neg logit contributions | |
Noise | -0.277 |
Faculty | -0.255 |
grad | -0.251 |
Madness | -0.244 |
graduate | -0.230 |
Token | response |
Token ID | 2882 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
ratulations | +1.166 |
chart | +1.117 |
claw | +1.109 |
hess | +1.003 |
=~ | +0.992 |
Neg logit contributions | |
Noise | -1.192 |
Faculty | -1.104 |
grad | -1.092 |
Madness | -1.071 |
graduate | -1.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
Noise | +1.749 |
grad | +1.611 |
Faculty | +1.604 |
Madness | +1.547 |
graduate | +1.465 |
Neg logit contributions | |
ratulations | -2.157 |
chart | -2.022 |
claw | -2.016 |
hess | -1.839 |
=~ | -1.829 |
Token |
|
Token ID | 198 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
ratulations | +1.071 |
chart | +1.035 |
claw | +1.028 |
hess | +0.940 |
=~ | +0.934 |
Neg logit contributions | |
Noise | -0.878 |
Faculty | -0.806 |
grad | -0.801 |
Madness | -0.778 |
graduate | -0.731 |
Token |
|
Token ID | 198 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
ratulations | +1.305 |
chart | +1.223 |
claw | +1.218 |
hess | +1.129 |
=~ | +1.125 |
Neg logit contributions | |
Noise | -0.866 |
Faculty | -0.796 |
grad | -0.782 |
Madness | -0.760 |
graduate | -0.706 |
Token | Article |
Token ID | 14906 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
ratulations | +0.313 |
chart | +0.296 |
claw | +0.295 |
hess | +0.270 |
=~ | +0.267 |
Neg logit contributions | |
Noise | -0.285 |
Faculty | -0.264 |
grad | -0.259 |
Madness | -0.253 |
graduate | -0.238 |
Token | Continued |
Token ID | 22168 |
Feature activation | +6.59e+00 |
Pos logit contributions | |
ratulations | +1.195 |
claw | +1.151 |
chart | +1.148 |
hess | +1.022 |
=~ | +1.007 |
Neg logit contributions | |
Noise | -1.429 |
Faculty | -1.328 |
grad | -1.315 |
Madness | -1.299 |
graduate | -1.224 |
Token | Below |
Token ID | 10383 |
Feature activation | +9.31e+00 |
Pos logit contributions | |
ratulations | +2.183 |
claw | +2.124 |
chart | +2.118 |
=~ | +1.625 |
hess | +1.616 |
Neg logit contributions | |
Noise | -7.586 |
grad | -7.209 |
Faculty | -7.197 |
Madness | -7.043 |
graduate | -6.866 |
Token |
|
Token ID | 198 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
claw | +6.756 |
chart | +6.726 |
ratulations | +6.509 |
hess | +6.027 |
=~ | +5.959 |
Neg logit contributions | |
Noise | -6.820 |
grad | -6.280 |
Faculty | -6.151 |
Madness | -6.006 |
charism | -5.927 |
Token |
|
Token ID | 198 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ratulations | +1.535 |
chart | +1.446 |
claw | +1.435 |
hess | +1.331 |
=~ | +1.322 |
Neg logit contributions | |
Noise | -1.042 |
Faculty | -0.955 |
grad | -0.936 |
Madness | -0.914 |
graduate | -0.846 |
Token | The |
Token ID | 464 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ratulations | +0.829 |
chart | +0.790 |
claw | +0.785 |
hess | +0.714 |
=~ | +0.708 |
Neg logit contributions | |
Noise | -0.793 |
Faculty | -0.734 |
grad | -0.724 |
Madness | -0.708 |
graduate | -0.666 |
Token | T |
Token ID | 309 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ratulations | +1.151 |
chart | +1.096 |
claw | +1.089 |
hess | +0.987 |
=~ | +0.977 |
Neg logit contributions | |
Noise | -1.195 |
Faculty | -1.109 |
grad | -1.096 |
Madness | -1.073 |
graduate | -1.012 |
Token | PA |
Token ID | 4537 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
ratulations | +0.143 |
chart | +0.136 |
claw | +0.136 |
hess | +0.126 |
=~ | +0.125 |
Neg logit contributions | |
Noise | -0.083 |
Faculty | -0.077 |
grad | -0.075 |
Madness | -0.072 |
graduate | -0.067 |
Token | large |
Token ID | 1588 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
ratulations | +1.328 |
chart | +1.269 |
claw | +1.264 |
hess | +1.159 |
=~ | +1.148 |
Neg logit contributions | |
Noise | -1.052 |
Faculty | -0.967 |
grad | -0.948 |
Madness | -0.927 |
graduate | -0.864 |
Token | horses |
Token ID | 14260 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
ratulations | +0.269 |
chart | +0.259 |
claw | +0.257 |
hess | +0.235 |
=~ | +0.234 |
Neg logit contributions | |
Noise | -0.216 |
Faculty | -0.198 |
grad | -0.197 |
Madness | -0.192 |
graduate | -0.180 |
Token | ," |
Token ID | 553 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
ratulations | +0.141 |
chart | +0.134 |
claw | +0.133 |
hess | +0.122 |
=~ | +0.121 |
Neg logit contributions | |
Noise | -0.128 |
Faculty | -0.118 |
grad | -0.117 |
Madness | -0.114 |
graduate | -0.107 |
Token | joked |
Token ID | 27582 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
ratulations | +2.100 |
chart | +2.012 |
claw | +2.008 |
hess | +1.854 |
=~ | +1.833 |
Neg logit contributions | |
Noise | -1.501 |
Faculty | -1.371 |
grad | -1.345 |
Madness | -1.318 |
graduate | -1.215 |
Token | Ginny |
Token ID | 37877 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
ratulations | +2.218 |
chart | +2.126 |
claw | +2.125 |
hess | +1.950 |
=~ | +1.928 |
Neg logit contributions | |
Noise | -1.753 |
Faculty | -1.609 |
grad | -1.577 |
Madness | -1.552 |
graduate | -1.431 |
Token | Schro |
Token ID | 47349 |
Feature activation | +9.25e+00 |
Pos logit contributions | |
ratulations | +2.317 |
chart | +2.219 |
claw | +2.206 |
hess | +1.987 |
=~ | +1.959 |
Neg logit contributions | |
Noise | -2.715 |
Faculty | -2.540 |
grad | -2.497 |
Madness | -2.470 |
graduate | -2.315 |
Token | eder |
Token ID | 5702 |
Feature activation | +6.30e+00 |
Pos logit contributions | |
ratulations | +5.171 |
chart | +5.033 |
claw | +5.011 |
hess | +4.429 |
=~ | +4.175 |
Neg logit contributions | |
Noise | -8.830 |
Faculty | -8.248 |
Madness | -8.029 |
grad | -8.008 |
Amend | -7.592 |
Token | . |
Token ID | 13 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
ratulations | +4.277 |
claw | +4.256 |
chart | +4.156 |
hess | +3.741 |
=~ | +3.645 |
Neg logit contributions | |
Noise | -5.003 |
Faculty | -4.711 |
grad | -4.519 |
Madness | -4.512 |
Amend | -4.277 |
Token | " |
Token ID | 366 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
ratulations | +1.396 |
chart | +1.345 |
claw | +1.342 |
hess | +1.234 |
=~ | +1.207 |
Neg logit contributions | |
Noise | -1.276 |
Faculty | -1.179 |
grad | -1.179 |
Madness | -1.137 |
graduate | -1.079 |
Token | And |
Token ID | 1870 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ratulations | +1.098 |
chart | +1.049 |
claw | +1.042 |
hess | +0.949 |
=~ | +0.941 |
Neg logit contributions | |
Noise | -1.027 |
Faculty | -0.949 |
grad | -0.937 |
Madness | -0.916 |
graduate | -0.861 |
Token | we |
Token ID | 356 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ratulations | +1.600 |
chart | +1.524 |
claw | +1.520 |
hess | +1.382 |
=~ | +1.368 |
Neg logit contributions | |
Noise | -1.478 |
Faculty | -1.372 |
grad | -1.345 |
Madness | -1.316 |
graduate | -1.236 |
Token | The |
Token ID | 383 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
ratulations | +3.513 |
claw | +3.453 |
chart | +3.448 |
hess | +3.100 |
=~ | +3.077 |
Neg logit contributions | |
Noise | -2.824 |
Faculty | -2.590 |
grad | -2.583 |
Madness | -2.508 |
graduate | -2.326 |
Token | Associated |
Token ID | 10575 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
ratulations | +1.737 |
chart | +1.675 |
claw | +1.668 |
hess | +1.543 |
=~ | +1.529 |
Neg logit contributions | |
Noise | -1.093 |
Faculty | -0.989 |
grad | -0.977 |
Madness | -0.949 |
graduate | -0.875 |
Token | Press |
Token ID | 4332 |
Feature activation | +6.86e+00 |
Pos logit contributions | |
ratulations | +5.194 |
claw | +5.118 |
chart | +5.108 |
hess | +4.695 |
=~ | +4.677 |
Neg logit contributions | |
Noise | -3.145 |
grad | -2.851 |
Faculty | -2.801 |
Madness | -2.707 |
graduate | -2.538 |
Token | . |
Token ID | 13 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
claw | +4.234 |
ratulations | +4.233 |
chart | +4.214 |
hess | +3.737 |
=~ | +3.604 |
Neg logit contributions | |
Noise | -5.832 |
Faculty | -5.329 |
grad | -5.323 |
Madness | -5.273 |
graduate | -4.943 |
Token | All |
Token ID | 1439 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
claw | +3.233 |
chart | +3.233 |
ratulations | +3.211 |
=~ | +2.864 |
hess | +2.827 |
Neg logit contributions | |
Noise | -3.658 |
grad | -3.477 |
Faculty | -3.363 |
Madness | -3.294 |
graduate | -3.195 |
Token | Rights |
Token ID | 6923 |
Feature activation | +9.20e+00 |
Pos logit contributions | |
ratulations | +6.365 |
claw | +6.154 |
chart | +6.122 |
=~ | +5.597 |
hess | +5.556 |
Neg logit contributions | |
Noise | -4.462 |
grad | -4.045 |
Faculty | -4.019 |
Madness | -3.969 |
alum | -3.662 |
Token | Reserved |
Token ID | 33876 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
ratulations | +4.768 |
chart | +4.725 |
claw | +4.650 |
=~ | +3.998 |
hess | +3.997 |
Neg logit contributions | |
Noise | -8.659 |
grad | -8.207 |
Faculty | -8.097 |
Madness | -8.011 |
graduate | -7.702 |
Token | . |
Token ID | 13 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
chart | +5.806 |
claw | +5.797 |
ratulations | +5.703 |
hess | +5.156 |
=~ | +5.048 |
Neg logit contributions | |
Noise | -7.325 |
grad | -6.932 |
Faculty | -6.800 |
Madness | -6.660 |
graduate | -6.399 |
Token | This |
Token ID | 770 |
Feature activation | +6.23e+00 |
Pos logit contributions | |
ratulations | +2.352 |
chart | +2.297 |
claw | +2.272 |
hess | +2.007 |
=~ | +1.984 |
Neg logit contributions | |
Noise | -2.939 |
grad | -2.773 |
Faculty | -2.739 |
Madness | -2.673 |
graduate | -2.569 |
Token | material |
Token ID | 2587 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
ratulations | +4.811 |
chart | +4.702 |
claw | +4.646 |
=~ | +4.180 |
hess | +4.144 |
Neg logit contributions | |
Noise | -4.568 |
Faculty | -4.191 |
grad | -4.188 |
Madness | -4.105 |
graduate | -3.830 |
Token | may |
Token ID | 743 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ratulations | +3.881 |
chart | +3.778 |
claw | +3.758 |
hess | +3.425 |
=~ | +3.390 |
Neg logit contributions | |
Noise | -3.290 |
Faculty | -3.029 |
grad | -2.979 |
Madness | -2.917 |
graduate | -2.711 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
ratulations | +2.311 |
chart | +2.217 |
claw | +2.200 |
hess | +1.986 |
=~ | +1.980 |
Neg logit contributions | |
Noise | -2.454 |
Faculty | -2.273 |
grad | -2.219 |
Madness | -2.190 |
alum | -2.059 |
Token |
|
Token ID | 198 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
Noise | +0.746 |
Faculty | +0.686 |
grad | +0.680 |
Madness | +0.660 |
graduate | +0.616 |
Neg logit contributions | |
ratulations | -0.916 |
chart | -0.880 |
claw | -0.874 |
hess | -0.803 |
=~ | -0.794 |
Token |
|
Token ID | 198 |
Feature activation | +3.91e-01 |
Pos logit contributions | |
ratulations | +1.116 |
chart | +1.046 |
claw | +1.038 |
hess | +0.963 |
=~ | +0.959 |
Neg logit contributions | |
Noise | -0.743 |
Faculty | -0.680 |
grad | -0.667 |
Madness | -0.652 |
graduate | -0.604 |
Token | Analysis |
Token ID | 32750 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
ratulations | +0.309 |
chart | +0.292 |
claw | +0.290 |
hess | +0.264 |
=~ | +0.262 |
Neg logit contributions | |
Noise | -0.293 |
Faculty | -0.272 |
grad | -0.267 |
Madness | -0.262 |
graduate | -0.246 |
Token | Lord |
Token ID | 4453 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
ratulations | +0.859 |
chart | +0.821 |
claw | +0.816 |
hess | +0.744 |
=~ | +0.738 |
Neg logit contributions | |
Noise | -0.782 |
Faculty | -0.723 |
grad | -0.713 |
Madness | -0.697 |
graduate | -0.655 |
Token | Hes |
Token ID | 44773 |
Feature activation | +9.18e+00 |
Pos logit contributions | |
ratulations | +1.341 |
chart | +1.272 |
claw | +1.262 |
hess | +1.132 |
=~ | +1.122 |
Neg logit contributions | |
Noise | -1.613 |
Faculty | -1.506 |
grad | -1.489 |
Madness | -1.461 |
graduate | -1.384 |
Token | elt |
Token ID | 2120 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
ratulations | +5.680 |
chart | +5.624 |
claw | +5.590 |
hess | +5.101 |
=~ | +4.759 |
Neg logit contributions | |
Noise | -8.123 |
Faculty | -7.438 |
Madness | -7.304 |
grad | -7.289 |
Amend | -6.933 |
Token | ine |
Token ID | 500 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ratulations | +1.509 |
claw | +1.408 |
chart | +1.407 |
=~ | +1.285 |
hess | +1.272 |
Neg logit contributions | |
Noise | -1.381 |
Faculty | -1.287 |
grad | -1.278 |
Madness | -1.241 |
graduate | -1.177 |
Token | 's |
Token ID | 338 |
Feature activation | +8.52e-01 |
Pos logit contributions | |
ratulations | +1.218 |
chart | +1.168 |
claw | +1.161 |
hess | +1.062 |
=~ | +1.051 |
Neg logit contributions | |
Noise | -1.048 |
Faculty | -0.965 |
grad | -0.950 |
Madness | -0.930 |
graduate | -0.869 |
Token | intervention |
Token ID | 9572 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ratulations | +0.752 |
chart | +0.720 |
claw | +0.717 |
hess | +0.660 |
=~ | +0.655 |
Neg logit contributions | |
Noise | -0.561 |
Faculty | -0.514 |
grad | -0.506 |
Madness | -0.493 |
graduate | -0.460 |
Token | reminds |
Token ID | 17603 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
ratulations | +1.731 |
chart | +1.665 |
claw | +1.651 |
hess | +1.502 |
=~ | +1.490 |
Neg logit contributions | |
Noise | -1.600 |
Faculty | -1.478 |
grad | -1.453 |
Madness | -1.423 |
graduate | -1.333 |
Token | receives |
Token ID | 11583 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
ratulations | +0.964 |
chart | +0.925 |
claw | +0.921 |
hess | +0.842 |
=~ | +0.836 |
Neg logit contributions | |
Noise | -0.797 |
Faculty | -0.735 |
grad | -0.723 |
Madness | -0.706 |
graduate | -0.660 |
Token | secret |
Token ID | 3200 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
ratulations | +0.685 |
chart | +0.655 |
claw | +0.653 |
hess | +0.594 |
=~ | +0.589 |
Neg logit contributions | |
Noise | -0.621 |
Faculty | -0.576 |
grad | -0.568 |
Madness | -0.554 |
graduate | -0.521 |
Token | voices |
Token ID | 10839 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
ratulations | +2.672 |
chart | +2.567 |
claw | +2.553 |
hess | +2.340 |
=~ | +2.326 |
Neg logit contributions | |
Noise | -2.034 |
Faculty | -1.877 |
grad | -1.843 |
Madness | -1.787 |
graduate | -1.676 |
Token | in |
Token ID | 287 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
Noise | +1.819 |
Faculty | +1.633 |
grad | +1.629 |
Madness | +1.584 |
graduate | +1.470 |
Neg logit contributions | |
ratulations | -2.532 |
chart | -2.391 |
claw | -2.384 |
hess | -2.212 |
=~ | -2.190 |
Token | his |
Token ID | 465 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
Noise | +1.220 |
Faculty | +1.115 |
grad | +1.102 |
Madness | +1.078 |
graduate | +1.003 |
Neg logit contributions | |
ratulations | -1.550 |
chart | -1.479 |
claw | -1.469 |
hess | -1.351 |
=~ | -1.340 |
Token | or |
Token ID | 393 |
Feature activation | +9.18e+00 |
Pos logit contributions | |
Noise | +0.352 |
Faculty | +0.325 |
grad | +0.321 |
Madness | +0.315 |
graduate | +0.295 |
Neg logit contributions | |
ratulations | -0.383 |
chart | -0.365 |
claw | -0.362 |
hess | -0.332 |
=~ | -0.328 |
Token | her |
Token ID | 607 |
Feature activation | -7.65e-01 |
Pos logit contributions | |
ratulations | +4.954 |
claw | +4.780 |
chart | +4.766 |
hess | +4.215 |
=~ | +4.122 |
Neg logit contributions | |
Noise | -8.700 |
Faculty | -8.271 |
grad | -8.026 |
Madness | -7.932 |
Amend | -7.605 |
Token | head |
Token ID | 1182 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Noise | +0.560 |
Faculty | +0.516 |
grad | +0.509 |
Madness | +0.502 |
graduate | +0.468 |
Neg logit contributions | |
ratulations | -0.623 |
chart | -0.593 |
claw | -0.587 |
hess | -0.539 |
=~ | -0.534 |
Token | , |
Token ID | 11 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
ratulations | +1.379 |
chart | +1.329 |
claw | +1.323 |
hess | +1.201 |
=~ | +1.191 |
Neg logit contributions | |
Noise | -1.222 |
Faculty | -1.132 |
grad | -1.114 |
Madness | -1.087 |
graduate | -1.022 |
Token | and |
Token ID | 290 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
Noise | +0.254 |
Faculty | +0.233 |
grad | +0.230 |
Madness | +0.225 |
graduate | +0.210 |
Neg logit contributions | |
ratulations | -0.315 |
chart | -0.301 |
claw | -0.299 |
hess | -0.275 |
=~ | -0.272 |
Token | must |
Token ID | 1276 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ratulations | +1.094 |
chart | +1.047 |
claw | +1.043 |
hess | +0.949 |
=~ | +0.941 |
Neg logit contributions | |
Noise | -0.987 |
Faculty | -0.914 |
grad | -0.900 |
Madness | -0.879 |
graduate | -0.826 |
Token | vote |
Token ID | 3015 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
ratulations | +1.730 |
chart | +1.654 |
claw | +1.638 |
hess | +1.453 |
=~ | +1.432 |
Neg logit contributions | |
Noise | -2.269 |
Faculty | -2.116 |
grad | -2.095 |
Madness | -2.050 |
graduate | -1.944 |
Token | ". |
Token ID | 1911 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
ratulations | +0.021 |
chart | +0.020 |
claw | +0.020 |
hess | +0.017 |
=~ | +0.017 |
Neg logit contributions | |
Noise | -0.034 |
Faculty | -0.032 |
grad | -0.032 |
Madness | -0.031 |
graduate | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
ratulations | +0.269 |
chart | +0.262 |
claw | +0.260 |
hess | +0.237 |
=~ | +0.234 |
Neg logit contributions | |
Noise | -0.234 |
grad | -0.219 |
Faculty | -0.216 |
Madness | -0.207 |
graduate | -0.199 |
Token |
|
Token ID | 198 |
Feature activation | +3.51e-03 |
Pos logit contributions | |
ratulations | +1.730 |
chart | +1.636 |
claw | +1.624 |
hess | +1.502 |
=~ | +1.495 |
Neg logit contributions | |
Noise | -1.206 |
Faculty | -1.104 |
grad | -1.084 |
Madness | -1.059 |
graduate | -0.983 |
Token | Lord |
Token ID | 22438 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
ratulations | +0.003 |
chart | +0.003 |
claw | +0.003 |
hess | +0.002 |
=~ | +0.002 |
Neg logit contributions | |
Noise | -0.003 |
Faculty | -0.002 |
grad | -0.002 |
Madness | -0.002 |
graduate | -0.002 |
Token | Hes |
Token ID | 44773 |
Feature activation | +9.15e+00 |
Pos logit contributions | |
ratulations | +1.258 |
chart | +1.195 |
claw | +1.189 |
hess | +1.064 |
=~ | +1.045 |
Neg logit contributions | |
Noise | -1.607 |
Faculty | -1.502 |
grad | -1.485 |
Madness | -1.454 |
graduate | -1.382 |
Token | elt |
Token ID | 2120 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
ratulations | +5.667 |
chart | +5.625 |
claw | +5.588 |
hess | +5.086 |
=~ | +4.741 |
Neg logit contributions | |
Noise | -8.090 |
Faculty | -7.416 |
Madness | -7.290 |
grad | -7.280 |
Amend | -6.910 |
Token | ine |
Token ID | 500 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
ratulations | +0.663 |
claw | +0.615 |
chart | +0.613 |
=~ | +0.565 |
hess | +0.556 |
Neg logit contributions | |
Noise | -0.580 |
Faculty | -0.542 |
grad | -0.539 |
Madness | -0.521 |
graduate | -0.496 |
Token | said |
Token ID | 531 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ratulations | +0.790 |
chart | +0.757 |
claw | +0.752 |
hess | +0.691 |
=~ | +0.686 |
Neg logit contributions | |
Noise | -0.615 |
Faculty | -0.564 |
grad | -0.556 |
Madness | -0.542 |
graduate | -0.506 |
Token | : |
Token ID | 25 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
Noise | +1.634 |
Faculty | +1.503 |
grad | +1.500 |
Madness | +1.447 |
graduate | +1.386 |
Neg logit contributions | |
ratulations | -1.861 |
chart | -1.777 |
claw | -1.771 |
=~ | -1.605 |
hess | -1.602 |
Token | " |
Token ID | 366 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
ratulations | +1.395 |
chart | +1.323 |
claw | +1.314 |
hess | +1.204 |
=~ | +1.199 |
Neg logit contributions | |
Noise | -1.138 |
Faculty | -1.048 |
grad | -1.031 |
Madness | -1.008 |
graduate | -0.944 |
Token |
|
Token ID | 198 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
Noise | +0.800 |
Faculty | +0.735 |
grad | +0.734 |
Madness | +0.710 |
graduate | +0.669 |
Neg logit contributions | |
ratulations | -0.987 |
claw | -0.941 |
chart | -0.939 |
hess | -0.860 |
=~ | -0.855 |
Token |
|
Token ID | 198 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ratulations | +0.538 |
chart | +0.503 |
claw | +0.500 |
hess | +0.465 |
=~ | +0.464 |
Neg logit contributions | |
Noise | -0.350 |
Faculty | -0.321 |
grad | -0.316 |
Madness | -0.308 |
graduate | -0.285 |
Token | EDIT |
Token ID | 24706 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
Noise | +0.989 |
Faculty | +0.918 |
grad | +0.904 |
Madness | +0.888 |
graduate | +0.834 |
Neg logit contributions | |
ratulations | -1.010 |
claw | -0.943 |
chart | -0.943 |
hess | -0.861 |
=~ | -0.851 |
Token | OR |
Token ID | 1581 |
Feature activation | +6.89e+00 |
Pos logit contributions | |
ratulations | +0.738 |
chart | +0.702 |
claw | +0.698 |
hess | +0.629 |
=~ | +0.624 |
Neg logit contributions | |
Noise | -0.811 |
Faculty | -0.754 |
grad | -0.745 |
Madness | -0.730 |
graduate | -0.690 |
Token | ' |
Token ID | 6 |
Feature activation | +5.11e+00 |
Pos logit contributions | |
ratulations | +4.747 |
claw | +4.668 |
chart | +4.632 |
=~ | +4.204 |
hess | +4.060 |
Neg logit contributions | |
Noise | -5.292 |
grad | -4.945 |
Faculty | -4.880 |
Madness | -4.832 |
bris | -4.589 |
Token | S |
Token ID | 50 |
Feature activation | +9.11e+00 |
Pos logit contributions | |
ratulations | +3.059 |
chart | +2.953 |
claw | +2.907 |
hess | +2.564 |
=~ | +2.555 |
Neg logit contributions | |
Noise | -4.630 |
Faculty | -4.364 |
grad | -4.339 |
Madness | -4.250 |
graduate | -4.056 |
Token | NOTE |
Token ID | 24550 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
ratulations | +7.099 |
chart | +6.937 |
claw | +6.700 |
=~ | +6.163 |
cuff | +6.035 |
Neg logit contributions | |
Noise | -6.244 |
Faculty | -5.964 |
Madness | -5.812 |
grad | -5.740 |
bris | -5.363 |
Token | : |
Token ID | 25 |
Feature activation | +8.91e-01 |
Pos logit contributions | |
ratulations | +3.087 |
chart | +3.015 |
claw | +2.973 |
=~ | +2.647 |
hess | +2.622 |
Neg logit contributions | |
Noise | -3.427 |
Faculty | -3.192 |
grad | -3.131 |
Madness | -3.062 |
bris | -2.919 |
Token | The |
Token ID | 383 |
Feature activation | +9.27e-01 |
Pos logit contributions | |
ratulations | +0.726 |
chart | +0.694 |
claw | +0.690 |
hess | +0.630 |
=~ | +0.624 |
Neg logit contributions | |
Noise | -0.645 |
Faculty | -0.596 |
grad | -0.587 |
Madness | -0.574 |
graduate | -0.539 |
Token | lineup |
Token ID | 12750 |
Feature activation | -9.41e-01 |
Pos logit contributions | |
ratulations | +0.775 |
chart | +0.740 |
claw | +0.737 |
hess | +0.675 |
=~ | +0.670 |
Neg logit contributions | |
Noise | -0.652 |
Faculty | -0.601 |
grad | -0.593 |
Madness | -0.578 |
graduate | -0.542 |
Token | below |
Token ID | 2174 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
Noise | +0.678 |
Faculty | +0.625 |
grad | +0.623 |
Madness | +0.605 |
graduate | +0.570 |
Neg logit contributions | |
ratulations | -0.777 |
claw | -0.735 |
chart | -0.730 |
hess | -0.673 |
=~ | -0.672 |
Token | the |
Token ID | 262 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ratulations | +1.002 |
chart | +0.964 |
claw | +0.960 |
hess | +0.878 |
=~ | +0.872 |
Neg logit contributions | |
Noise | -0.774 |
Faculty | -0.707 |
grad | -0.703 |
Madness | -0.681 |
graduate | -0.637 |
Token | Director |
Token ID | 5890 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ratulations | +1.027 |
chart | +0.984 |
claw | +0.977 |
hess | +0.890 |
=~ | +0.884 |
Neg logit contributions | |
Noise | -0.910 |
Faculty | -0.837 |
grad | -0.830 |
Madness | -0.809 |
graduate | -0.759 |
Token | , |
Token ID | 11 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ratulations | +0.842 |
chart | +0.811 |
claw | +0.805 |
hess | +0.730 |
=~ | +0.725 |
Neg logit contributions | |
Noise | -0.806 |
Faculty | -0.740 |
grad | -0.735 |
Madness | -0.719 |
graduate | -0.674 |
Token | CDC |
Token ID | 20434 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
ratulations | +0.995 |
chart | +0.956 |
claw | +0.950 |
hess | +0.868 |
=~ | +0.861 |
Neg logit contributions | |
Noise | -0.818 |
Faculty | -0.750 |
grad | -0.745 |
Madness | -0.725 |
graduate | -0.678 |
Token | . |
Token ID | 13 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
ratulations | +2.973 |
chart | +2.922 |
claw | +2.893 |
hess | +2.609 |
=~ | +2.582 |
Neg logit contributions | |
Noise | -2.956 |
grad | -2.702 |
Faculty | -2.695 |
Madness | -2.639 |
graduate | -2.473 |
Token | Correspond |
Token ID | 34428 |
Feature activation | +9.04e+00 |
Pos logit contributions | |
Noise | +0.477 |
Faculty | +0.439 |
grad | +0.431 |
Madness | +0.421 |
graduate | +0.393 |
Neg logit contributions | |
ratulations | -0.608 |
chart | -0.581 |
claw | -0.577 |
hess | -0.531 |
=~ | -0.526 |
Token | ing |
Token ID | 278 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
ratulations | +6.296 |
chart | +5.956 |
claw | +5.861 |
hess | +5.351 |
=~ | +5.221 |
Neg logit contributions | |
Noise | -7.211 |
grad | -6.508 |
Madness | -6.429 |
Faculty | -6.390 |
Amend | -6.015 |
Token | contributor |
Token ID | 18920 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
ratulations | +1.925 |
chart | +1.848 |
claw | +1.834 |
hess | +1.676 |
=~ | +1.658 |
Neg logit contributions | |
Noise | -1.639 |
Faculty | -1.492 |
grad | -1.480 |
Madness | -1.454 |
graduate | -1.347 |
Token | : |
Token ID | 25 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ratulations | +1.559 |
chart | +1.493 |
claw | +1.481 |
hess | +1.355 |
=~ | +1.342 |
Neg logit contributions | |
Noise | -1.433 |
Faculty | -1.316 |
grad | -1.304 |
Madness | -1.279 |
graduate | -1.196 |
Token | David |
Token ID | 3271 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
ratulations | +1.166 |
chart | +1.123 |
claw | +1.118 |
hess | +1.028 |
=~ | +1.019 |
Neg logit contributions | |
Noise | -0.835 |
Faculty | -0.758 |
grad | -0.751 |
Madness | -0.731 |
graduate | -0.679 |
Token | A |
Token ID | 317 |
Feature activation | +7.04e+00 |
Pos logit contributions | |
Noise | +0.043 |
Faculty | +0.040 |
grad | +0.039 |
Madness | +0.038 |
graduate | +0.036 |
Neg logit contributions | |
ratulations | -0.042 |
chart | -0.040 |
claw | -0.040 |
hess | -0.036 |
=~ | -0.036 |
Token | Associated |
Token ID | 10575 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
ratulations | +1.737 |
chart | +1.675 |
claw | +1.668 |
hess | +1.543 |
=~ | +1.529 |
Neg logit contributions | |
Noise | -1.093 |
Faculty | -0.989 |
grad | -0.977 |
Madness | -0.949 |
graduate | -0.875 |
Token | Press |
Token ID | 4332 |
Feature activation | +6.86e+00 |
Pos logit contributions | |
ratulations | +5.194 |
claw | +5.118 |
chart | +5.108 |
hess | +4.695 |
=~ | +4.677 |
Neg logit contributions | |
Noise | -3.145 |
grad | -2.851 |
Faculty | -2.801 |
Madness | -2.707 |
graduate | -2.538 |
Token | . |
Token ID | 13 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
claw | +4.234 |
ratulations | +4.233 |
chart | +4.214 |
hess | +3.737 |
=~ | +3.604 |
Neg logit contributions | |
Noise | -5.832 |
Faculty | -5.329 |
grad | -5.323 |
Madness | -5.273 |
graduate | -4.943 |
Token | All |
Token ID | 1439 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
claw | +3.233 |
chart | +3.233 |
ratulations | +3.211 |
=~ | +2.864 |
hess | +2.827 |
Neg logit contributions | |
Noise | -3.658 |
grad | -3.477 |
Faculty | -3.363 |
Madness | -3.294 |
graduate | -3.195 |
Token | Rights |
Token ID | 6923 |
Feature activation | +9.20e+00 |
Pos logit contributions | |
ratulations | +6.365 |
claw | +6.154 |
chart | +6.122 |
=~ | +5.597 |
hess | +5.556 |
Neg logit contributions | |
Noise | -4.462 |
grad | -4.045 |
Faculty | -4.019 |
Madness | -3.969 |
alum | -3.662 |
Token | Reserved |
Token ID | 33876 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
ratulations | +4.768 |
chart | +4.725 |
claw | +4.650 |
=~ | +3.998 |
hess | +3.997 |
Neg logit contributions | |
Noise | -8.659 |
grad | -8.207 |
Faculty | -8.097 |
Madness | -8.011 |
graduate | -7.702 |
Token | . |
Token ID | 13 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
chart | +5.806 |
claw | +5.797 |
ratulations | +5.703 |
hess | +5.156 |
=~ | +5.048 |
Neg logit contributions | |
Noise | -7.325 |
grad | -6.932 |
Faculty | -6.800 |
Madness | -6.660 |
graduate | -6.399 |
Token | This |
Token ID | 770 |
Feature activation | +6.23e+00 |
Pos logit contributions | |
ratulations | +2.352 |
chart | +2.297 |
claw | +2.272 |
hess | +2.007 |
=~ | +1.984 |
Neg logit contributions | |
Noise | -2.939 |
grad | -2.773 |
Faculty | -2.739 |
Madness | -2.673 |
graduate | -2.569 |
Token | material |
Token ID | 2587 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
ratulations | +4.811 |
chart | +4.702 |
claw | +4.646 |
=~ | +4.180 |
hess | +4.144 |
Neg logit contributions | |
Noise | -4.568 |
Faculty | -4.191 |
grad | -4.188 |
Madness | -4.105 |
graduate | -3.830 |
Token | may |
Token ID | 743 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ratulations | +3.881 |
chart | +3.778 |
claw | +3.758 |
hess | +3.425 |
=~ | +3.390 |
Neg logit contributions | |
Noise | -3.290 |
Faculty | -3.029 |
grad | -2.979 |
Madness | -2.917 |
graduate | -2.711 |
Token | not |
Token ID | 407 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
ratulations | +2.198 |
chart | +2.098 |
claw | +2.082 |
hess | +1.911 |
=~ | +1.890 |
Neg logit contributions | |
Noise | -1.858 |
Faculty | -1.712 |
grad | -1.687 |
Madness | -1.648 |
graduate | -1.547 |
Token | UK |
Token ID | 3482 |
Feature activation | +5.97e-01 |
Pos logit contributions | |
Noise | +0.635 |
Faculty | +0.592 |
grad | +0.582 |
Madness | +0.572 |
graduate | +0.540 |
Neg logit contributions | |
ratulations | -0.574 |
chart | -0.542 |
claw | -0.538 |
hess | -0.489 |
=~ | -0.482 |
Token | , |
Token ID | 11 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ratulations | +0.499 |
chart | +0.475 |
claw | +0.473 |
hess | +0.432 |
=~ | +0.429 |
Neg logit contributions | |
Noise | -0.421 |
Faculty | -0.388 |
grad | -0.382 |
Madness | -0.374 |
graduate | -0.350 |
Token | with |
Token ID | 351 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
ratulations | +1.271 |
chart | +1.217 |
claw | +1.210 |
hess | +1.110 |
=~ | +1.100 |
Neg logit contributions | |
Noise | -1.030 |
Faculty | -0.946 |
grad | -0.932 |
Madness | -0.910 |
graduate | -0.851 |
Token | the |
Token ID | 262 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
ratulations | +0.207 |
chart | +0.198 |
claw | +0.196 |
hess | +0.180 |
=~ | +0.177 |
Neg logit contributions | |
Noise | -0.187 |
Faculty | -0.172 |
grad | -0.170 |
Madness | -0.166 |
graduate | -0.156 |
Token | US |
Token ID | 1294 |
Feature activation | +4.92e-01 |
Pos logit contributions | |
Noise | +0.134 |
Faculty | +0.123 |
grad | +0.122 |
Madness | +0.118 |
graduate | +0.111 |
Neg logit contributions | |
ratulations | -0.164 |
claw | -0.156 |
chart | -0.156 |
hess | -0.144 |
=~ | -0.141 |
Token | auction |
Token ID | 14389 |
Feature activation | +9.02e+00 |
Pos logit contributions | |
ratulations | +0.401 |
chart | +0.380 |
claw | +0.380 |
hess | +0.347 |
=~ | +0.342 |
Neg logit contributions | |
Noise | -0.358 |
Faculty | -0.330 |
grad | -0.327 |
Madness | -0.318 |
graduate | -0.300 |
Token | said |
Token ID | 531 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
chart | +6.789 |
ratulations | +6.788 |
claw | +6.634 |
hess | +5.916 |
=~ | +5.894 |
Neg logit contributions | |
Noise | -6.506 |
Faculty | -5.966 |
Madness | -5.752 |
grad | -5.692 |
Lank | -5.330 |
Token | to |
Token ID | 284 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
ratulations | +1.248 |
chart | +1.199 |
claw | +1.187 |
hess | +1.094 |
=~ | +1.082 |
Neg logit contributions | |
Noise | -1.066 |
Faculty | -0.986 |
grad | -0.967 |
Madness | -0.943 |
graduate | -0.881 |
Token | be |
Token ID | 307 |
Feature activation | +3.56e+00 |
Pos logit contributions | |
ratulations | +1.526 |
chart | +1.452 |
claw | +1.440 |
hess | +1.292 |
=~ | +1.280 |
Neg logit contributions | |
Noise | -1.825 |
Faculty | -1.704 |
grad | -1.682 |
Madness | -1.651 |
graduate | -1.560 |
Token | topping |
Token ID | 34366 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ratulations | +3.001 |
chart | +2.892 |
claw | +2.871 |
hess | +2.621 |
=~ | +2.600 |
Neg logit contributions | |
Noise | -2.435 |
Faculty | -2.249 |
grad | -2.199 |
Madness | -2.150 |
graduate | -1.996 |
Token | $ |
Token ID | 720 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
ratulations | +1.074 |
chart | +1.023 |
claw | +1.020 |
hess | +0.933 |
=~ | +0.923 |
Neg logit contributions | |
Noise | -0.912 |
Faculty | -0.838 |
grad | -0.828 |
Madness | -0.809 |
graduate | -0.758 |
Token | communities |
Token ID | 5348 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
Noise | +1.244 |
grad | +1.180 |
Faculty | +1.167 |
Madness | +1.084 |
graduate | +1.063 |
Neg logit contributions | |
ratulations | -1.738 |
chart | -1.645 |
claw | -1.623 |
hess | -1.510 |
=~ | -1.507 |
Token | ," |
Token ID | 553 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
Noise | +0.471 |
Faculty | +0.436 |
grad | +0.432 |
Madness | +0.415 |
graduate | +0.395 |
Neg logit contributions | |
ratulations | -0.577 |
claw | -0.548 |
chart | -0.547 |
hess | -0.500 |
=~ | -0.500 |
Token | d |
Token ID | 288 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
ratulations | +1.403 |
chart | +1.330 |
claw | +1.318 |
hess | +1.180 |
=~ | +1.163 |
Neg logit contributions | |
Noise | -1.851 |
Faculty | -1.730 |
grad | -1.713 |
Madness | -1.682 |
graduate | -1.596 |
Token | ' |
Token ID | 6 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
ratulations | +2.376 |
chart | +2.265 |
claw | +2.246 |
=~ | +2.062 |
hess | +2.055 |
Neg logit contributions | |
Noise | -1.722 |
Faculty | -1.588 |
grad | -1.563 |
Madness | -1.519 |
graduate | -1.420 |
Token | Ent |
Token ID | 14539 |
Feature activation | +4.41e+00 |
Pos logit contributions | |
Noise | +0.530 |
Faculty | +0.492 |
grad | +0.478 |
Madness | +0.462 |
Amend | +0.428 |
Neg logit contributions | |
ratulations | -0.818 |
chart | -0.772 |
claw | -0.766 |
=~ | -0.705 |
hess | -0.700 |
Token | rem |
Token ID | 2787 |
Feature activation | +9.01e+00 |
Pos logit contributions | |
ratulations | +3.360 |
chart | +3.212 |
claw | +3.191 |
hess | +2.890 |
=~ | +2.860 |
Neg logit contributions | |
Noise | -3.418 |
Faculty | -3.170 |
grad | -3.129 |
Madness | -3.067 |
graduate | -2.886 |
Token | ont |
Token ID | 756 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
ratulations | +7.129 |
chart | +6.967 |
claw | +6.893 |
hess | +6.369 |
=~ | +6.197 |
Neg logit contributions | |
Noise | -6.522 |
Faculty | -5.951 |
grad | -5.820 |
Madness | -5.820 |
graduate | -5.355 |
Token | states |
Token ID | 2585 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ratulations | +2.540 |
chart | +2.469 |
claw | +2.447 |
hess | +2.257 |
=~ | +2.233 |
Neg logit contributions | |
Noise | -1.714 |
Faculty | -1.547 |
grad | -1.521 |
Madness | -1.490 |
graduate | -1.370 |
Token | in |
Token ID | 287 |
Feature activation | +4.61e+00 |
Pos logit contributions | |
ratulations | +1.331 |
chart | +1.284 |
claw | +1.264 |
hess | +1.147 |
=~ | +1.132 |
Neg logit contributions | |
Noise | -1.367 |
Faculty | -1.267 |
grad | -1.246 |
Madness | -1.225 |
graduate | -1.150 |
Token | her |
Token ID | 607 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
ratulations | +2.895 |
chart | +2.848 |
claw | +2.743 |
hess | +2.457 |
=~ | +2.390 |
Neg logit contributions | |
Noise | -4.000 |
Faculty | -3.729 |
grad | -3.697 |
Madness | -3.663 |
Amend | -3.450 |
Token | report |
Token ID | 989 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
ratulations | +0.882 |
chart | +0.856 |
claw | +0.851 |
hess | +0.797 |
=~ | +0.793 |
Neg logit contributions | |
Noise | -0.322 |
Faculty | -0.277 |
grad | -0.271 |
Madness | -0.260 |
graduate | -0.227 |
Token | just |
Token ID | 655 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
ratulations | +0.535 |
chart | +0.512 |
claw | +0.508 |
hess | +0.464 |
=~ | +0.461 |
Neg logit contributions | |
Noise | -0.482 |
Faculty | -0.445 |
grad | -0.442 |
Madness | -0.430 |
graduate | -0.405 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
ratulations | +2.412 |
chart | +2.301 |
claw | +2.293 |
hess | +2.082 |
=~ | +2.059 |
Neg logit contributions | |
Noise | -2.312 |
Faculty | -2.142 |
grad | -2.100 |
Madness | -2.062 |
graduate | -1.933 |
Token | animals |
Token ID | 4695 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
ratulations | +0.434 |
chart | +0.415 |
claw | +0.411 |
hess | +0.374 |
=~ | +0.372 |
Neg logit contributions | |
Noise | -0.407 |
Faculty | -0.376 |
grad | -0.374 |
Madness | -0.364 |
graduate | -0.343 |
Token | ," |
Token ID | 553 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
ratulations | +0.192 |
chart | +0.184 |
claw | +0.182 |
hess | +0.166 |
=~ | +0.165 |
Neg logit contributions | |
Noise | -0.178 |
Faculty | -0.164 |
grad | -0.163 |
Madness | -0.159 |
graduate | -0.150 |
Token | Brad |
Token ID | 8114 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
ratulations | +1.880 |
chart | +1.805 |
claw | +1.798 |
hess | +1.658 |
=~ | +1.639 |
Neg logit contributions | |
Noise | -1.384 |
Faculty | -1.264 |
grad | -1.239 |
Madness | -1.216 |
graduate | -1.124 |
Token | Schro |
Token ID | 47349 |
Feature activation | +8.99e+00 |
Pos logit contributions | |
ratulations | +3.164 |
chart | +3.062 |
claw | +3.034 |
hess | +2.798 |
=~ | +2.754 |
Neg logit contributions | |
Noise | -2.636 |
Faculty | -2.433 |
grad | -2.379 |
Madness | -2.351 |
graduate | -2.174 |
Token | eder |
Token ID | 5702 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
ratulations | +5.114 |
chart | +4.955 |
claw | +4.915 |
hess | +4.342 |
=~ | +4.125 |
Neg logit contributions | |
Noise | -8.556 |
Faculty | -7.982 |
Madness | -7.811 |
grad | -7.790 |
Amend | -7.371 |
Token | said |
Token ID | 531 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ratulations | +3.972 |
claw | +3.891 |
chart | +3.890 |
hess | +3.587 |
=~ | +3.508 |
Neg logit contributions | |
Noise | -2.818 |
Faculty | -2.595 |
Madness | -2.484 |
grad | -2.461 |
Amend | -2.289 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ratulations | +1.644 |
chart | +1.575 |
claw | +1.568 |
hess | +1.415 |
=~ | +1.396 |
Neg logit contributions | |
Noise | -1.675 |
Faculty | -1.557 |
grad | -1.523 |
Madness | -1.504 |
Amend | -1.402 |
Token |
|
Token ID | 198 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
ratulations | +0.832 |
chart | +0.798 |
claw | +0.794 |
hess | +0.727 |
=~ | +0.715 |
Neg logit contributions | |
Noise | -0.774 |
Faculty | -0.715 |
grad | -0.711 |
Madness | -0.691 |
graduate | -0.652 |
Token |
|
Token ID | 198 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
ratulations | +0.319 |
chart | +0.296 |
claw | +0.294 |
hess | +0.274 |
=~ | +0.273 |
Neg logit contributions | |
Noise | -0.198 |
Faculty | -0.181 |
grad | -0.178 |
Madness | -0.173 |
graduate | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
Noise | +0.416 |
Faculty | +0.384 |
grad | +0.383 |
Madness | +0.371 |
graduate | +0.353 |
Neg logit contributions | |
ratulations | -0.461 |
chart | -0.436 |
claw | -0.435 |
hess | -0.400 |
=~ | -0.395 |
Token |
|
Token ID | 198 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
Noise | +0.011 |
Faculty | +0.010 |
grad | +0.010 |
Madness | +0.010 |
graduate | +0.009 |
Neg logit contributions | |
ratulations | -0.014 |
chart | -0.013 |
claw | -0.013 |
hess | -0.012 |
=~ | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
ratulations | +0.615 |
chart | +0.570 |
claw | +0.565 |
hess | +0.528 |
=~ | +0.526 |
Neg logit contributions | |
Noise | -0.382 |
Faculty | -0.351 |
grad | -0.344 |
Madness | -0.335 |
graduate | -0.310 |
Token | Article |
Token ID | 14906 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
Noise | +0.257 |
Faculty | +0.240 |
grad | +0.236 |
Madness | +0.231 |
graduate | +0.218 |
Neg logit contributions | |
ratulations | -0.249 |
chart | -0.233 |
claw | -0.232 |
hess | -0.211 |
=~ | -0.209 |
Token | Continued |
Token ID | 22168 |
Feature activation | +6.22e+00 |
Pos logit contributions | |
Noise | +0.292 |
Faculty | +0.272 |
grad | +0.268 |
Madness | +0.261 |
graduate | +0.248 |
Neg logit contributions | |
ratulations | -0.293 |
chart | -0.280 |
claw | -0.277 |
hess | -0.251 |
=~ | -0.251 |
Token | Below |
Token ID | 10383 |
Feature activation | +8.98e+00 |
Pos logit contributions | |
ratulations | +2.171 |
chart | +2.074 |
claw | +2.062 |
hess | +1.599 |
=~ | +1.591 |
Neg logit contributions | |
Noise | -7.128 |
Faculty | -6.766 |
grad | -6.753 |
Madness | -6.616 |
graduate | -6.426 |
Token |
|
Token ID | 198 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
claw | +6.576 |
chart | +6.555 |
ratulations | +6.345 |
hess | +5.857 |
=~ | +5.783 |
Neg logit contributions | |
Noise | -6.581 |
grad | -6.051 |
Faculty | -5.945 |
Madness | -5.808 |
charism | -5.677 |
Token |
|
Token ID | 198 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ratulations | +0.587 |
chart | +0.545 |
claw | +0.540 |
hess | +0.504 |
=~ | +0.502 |
Neg logit contributions | |
Noise | -0.369 |
Faculty | -0.339 |
grad | -0.331 |
Madness | -0.324 |
graduate | -0.299 |
Token | T |
Token ID | 51 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
Noise | +0.944 |
Faculty | +0.884 |
grad | +0.862 |
Madness | +0.846 |
graduate | +0.793 |
Neg logit contributions | |
ratulations | -0.991 |
chart | -0.918 |
claw | -0.917 |
hess | -0.838 |
=~ | -0.825 |
Token | od |
Token ID | 375 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Noise | +2.273 |
Faculty | +2.075 |
grad | +2.051 |
Madness | +1.983 |
graduate | +1.847 |
Neg logit contributions | |
ratulations | -3.527 |
chart | -3.378 |
claw | -3.363 |
=~ | -3.127 |
hess | -3.107 |
Token | as |
Token ID | 292 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Noise | +0.520 |
Faculty | +0.475 |
grad | +0.472 |
Madness | +0.444 |
graduate | +0.417 |
Neg logit contributions | |
ratulations | -1.049 |
chart | -0.993 |
claw | -0.984 |
=~ | -0.937 |
hess | -0.910 |
Token | along |
Token ID | 1863 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
ratulations | +3.332 |
chart | +3.228 |
claw | +3.204 |
hess | +2.916 |
=~ | +2.852 |
Neg logit contributions | |
Noise | -3.102 |
Faculty | -2.868 |
grad | -2.765 |
Madness | -2.753 |
Amend | -2.561 |
Token | the |
Token ID | 262 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ratulations | +2.575 |
chart | +2.515 |
claw | +2.470 |
hess | +2.233 |
=~ | +2.205 |
Neg logit contributions | |
Noise | -2.683 |
Faculty | -2.478 |
grad | -2.431 |
Madness | -2.412 |
graduate | -2.257 |
Token | city |
Token ID | 1748 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ratulations | +0.730 |
chart | +0.697 |
claw | +0.685 |
hess | +0.609 |
=~ | +0.598 |
Neg logit contributions | |
Noise | -1.053 |
Faculty | -0.984 |
grad | -0.969 |
Madness | -0.963 |
graduate | -0.909 |
Token | 's |
Token ID | 338 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ratulations | +1.092 |
chart | +1.052 |
claw | +1.048 |
hess | +0.964 |
=~ | +0.943 |
Neg logit contributions | |
Noise | -0.942 |
Faculty | -0.862 |
grad | -0.842 |
Madness | -0.834 |
graduate | -0.772 |
Token | broad |
Token ID | 3154 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ratulations | +1.022 |
chart | +0.985 |
claw | +0.975 |
hess | +0.901 |
=~ | +0.888 |
Neg logit contributions | |
Noise | -0.782 |
Faculty | -0.712 |
grad | -0.694 |
Madness | -0.688 |
graduate | -0.633 |
Token | bou |
Token ID | 35833 |
Feature activation | +8.96e+00 |
Pos logit contributions | |
Noise | +0.639 |
grad | +0.613 |
Faculty | +0.595 |
Madness | +0.554 |
graduate | +0.546 |
Neg logit contributions | |
ratulations | -0.956 |
claw | -0.902 |
chart | -0.900 |
=~ | -0.845 |
hess | -0.824 |
Token | lev |
Token ID | 2768 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
chart | +4.971 |
ratulations | +4.961 |
claw | +4.835 |
hess | +4.216 |
=~ | +4.109 |
Neg logit contributions | |
Noise | -8.556 |
Faculty | -7.982 |
Madness | -7.753 |
grad | -7.703 |
Amend | -7.462 |
Token | ards |
Token ID | 1371 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
ratulations | +2.936 |
chart | +2.843 |
claw | +2.821 |
hess | +2.644 |
=~ | +2.620 |
Neg logit contributions | |
Noise | -1.370 |
Faculty | -1.206 |
grad | -1.173 |
Madness | -1.144 |
graduate | -1.024 |
Token | in |
Token ID | 287 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
ratulations | +3.003 |
chart | +2.889 |
claw | +2.869 |
hess | +2.606 |
=~ | +2.539 |
Neg logit contributions | |
Noise | -2.822 |
Faculty | -2.608 |
grad | -2.520 |
Madness | -2.501 |
Amend | -2.337 |
Token | support |
Token ID | 1104 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
ratulations | +1.444 |
chart | +1.375 |
claw | +1.360 |
hess | +1.228 |
=~ | +1.210 |
Neg logit contributions | |
Noise | -1.650 |
Faculty | -1.539 |
grad | -1.510 |
Madness | -1.489 |
graduate | -1.401 |
Token | , |
Token ID | 11 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
ratulations | +1.655 |
chart | +1.577 |
claw | +1.570 |
hess | +1.434 |
=~ | +1.396 |
Neg logit contributions | |
Noise | -1.575 |
Faculty | -1.460 |
grad | -1.432 |
Madness | -1.421 |
graduate | -1.314 |
Token | temperature |
Token ID | 5951 |
Feature activation | +8.91e-02 |
Pos logit contributions | |
Noise | +0.380 |
grad | +0.349 |
Faculty | +0.348 |
Madness | +0.333 |
graduate | +0.317 |
Neg logit contributions | |
ratulations | -0.476 |
chart | -0.454 |
claw | -0.452 |
hess | -0.414 |
=~ | -0.412 |
Token | feedback |
Token ID | 7538 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
ratulations | +0.076 |
chart | +0.072 |
claw | +0.072 |
hess | +0.066 |
=~ | +0.066 |
Neg logit contributions | |
Noise | -0.061 |
grad | -0.057 |
Faculty | -0.056 |
Madness | -0.054 |
graduate | -0.051 |
Token | s |
Token ID | 82 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
Noise | +2.490 |
grad | +2.334 |
Faculty | +2.303 |
Madness | +2.250 |
graduate | +2.154 |
Neg logit contributions | |
ratulations | -1.927 |
claw | -1.796 |
chart | -1.792 |
hess | -1.595 |
=~ | -1.573 |
Token | nor |
Token ID | 4249 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
Noise | +2.074 |
grad | +1.900 |
Faculty | +1.884 |
Madness | +1.806 |
graduate | +1.722 |
Neg logit contributions | |
ratulations | -2.984 |
chart | -2.826 |
claw | -2.824 |
hess | -2.607 |
=~ | -2.581 |
Token | their |
Token ID | 511 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
Noise | +2.354 |
grad | +2.198 |
Faculty | +2.141 |
Madness | +2.061 |
graduate | +1.982 |
Neg logit contributions | |
ratulations | -2.918 |
chart | -2.768 |
claw | -2.748 |
hess | -2.521 |
=~ | -2.508 |
Token | sum |
Token ID | 2160 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
Noise | +1.746 |
grad | +1.620 |
Faculty | +1.593 |
Madness | +1.536 |
graduate | +1.463 |
Neg logit contributions | |
ratulations | -2.085 |
chart | -1.989 |
claw | -1.985 |
hess | -1.802 |
=~ | -1.798 |
Token | can |
Token ID | 460 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
Noise | +6.923 |
grad | +6.721 |
Faculty | +6.378 |
Madness | +6.138 |
graduate | +6.013 |
Neg logit contributions | |
ratulations | -8.345 |
claw | -7.943 |
chart | -7.939 |
hess | -7.069 |
akeru | -6.988 |
Token | be |
Token ID | 307 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
Noise | +1.498 |
grad | +1.393 |
Faculty | +1.378 |
Madness | +1.340 |
graduate | +1.299 |
Neg logit contributions | |
ratulations | -1.534 |
claw | -1.451 |
chart | -1.443 |
hess | -1.320 |
=~ | -1.304 |
Token | directly |
Token ID | 3264 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
Noise | +2.326 |
grad | +2.129 |
Faculty | +2.099 |
Madness | +2.017 |
graduate | +1.919 |
Neg logit contributions | |
ratulations | -3.078 |
claw | -2.953 |
chart | -2.910 |
hess | -2.685 |
=~ | -2.674 |
Token | measured |
Token ID | 8630 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
Noise | +0.301 |
Faculty | +0.275 |
grad | +0.273 |
Madness | +0.265 |
graduate | +0.248 |
Neg logit contributions | |
ratulations | -0.390 |
claw | -0.373 |
chart | -0.372 |
hess | -0.341 |
=~ | -0.339 |
Token | ; |
Token ID | 26 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
Noise | +1.712 |
grad | +1.570 |
Faculty | +1.547 |
Madness | +1.501 |
graduate | +1.422 |
Neg logit contributions | |
ratulations | -2.267 |
claw | -2.137 |
chart | -2.132 |
=~ | -1.956 |
hess | -1.949 |
Token | , |
Token ID | 11 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ratulations | +0.955 |
chart | +0.916 |
claw | +0.910 |
hess | +0.832 |
=~ | +0.825 |
Neg logit contributions | |
Noise | -0.799 |
Faculty | -0.736 |
grad | -0.724 |
Madness | -0.707 |
graduate | -0.661 |
Token | a |
Token ID | 257 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
Noise | +0.800 |
Faculty | +0.738 |
grad | +0.733 |
Madness | +0.716 |
graduate | +0.677 |
Neg logit contributions | |
ratulations | -0.791 |
chart | -0.751 |
claw | -0.744 |
hess | -0.686 |
=~ | -0.673 |
Token | purge |
Token ID | 35714 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
Noise | +1.081 |
Faculty | +0.988 |
grad | +0.981 |
Madness | +0.951 |
graduate | +0.897 |
Neg logit contributions | |
ratulations | -1.370 |
chart | -1.305 |
claw | -1.295 |
hess | -1.194 |
=~ | -1.186 |
Token | — |
Token ID | 960 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
Noise | +2.247 |
grad | +2.066 |
Faculty | +2.060 |
Madness | +2.015 |
graduate | +1.898 |
Neg logit contributions | |
ratulations | -2.594 |
chart | -2.422 |
claw | -2.402 |
hess | -2.229 |
=~ | -2.199 |
Token | as |
Token ID | 292 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
Noise | +1.598 |
Faculty | +1.468 |
grad | +1.450 |
Madness | +1.425 |
graduate | +1.329 |
Neg logit contributions | |
ratulations | -1.815 |
chart | -1.695 |
claw | -1.682 |
hess | -1.566 |
=~ | -1.543 |
Token | defined |
Token ID | 5447 |
Feature activation | -9.89e+00 |
Pos logit contributions | |
Noise | +1.927 |
Faculty | +1.757 |
grad | +1.737 |
Madness | +1.708 |
graduate | +1.589 |
Neg logit contributions | |
ratulations | -2.311 |
chart | -2.211 |
claw | -2.197 |
hess | -2.036 |
=~ | -2.005 |
Token | by |
Token ID | 416 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
Noise | +5.325 |
grad | +4.889 |
Faculty | +4.685 |
Madness | +4.673 |
graduate | +4.354 |
Neg logit contributions | |
ratulations | -9.159 |
claw | -8.712 |
chart | -8.544 |
hess | -8.172 |
=~ | -8.014 |
Token | the |
Token ID | 262 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
Noise | +1.368 |
Faculty | +1.249 |
grad | +1.233 |
Madness | +1.208 |
graduate | +1.133 |
Neg logit contributions | |
ratulations | -1.637 |
chart | -1.557 |
claw | -1.541 |
hess | -1.437 |
=~ | -1.409 |
Token | movie |
Token ID | 3807 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
Noise | +0.232 |
Faculty | +0.214 |
grad | +0.210 |
Madness | +0.206 |
graduate | +0.193 |
Neg logit contributions | |
ratulations | -0.276 |
chart | -0.262 |
claw | -0.262 |
hess | -0.241 |
=~ | -0.238 |
Token | The |
Token ID | 383 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
ratulations | +1.476 |
chart | +1.417 |
claw | +1.409 |
hess | +1.287 |
=~ | +1.277 |
Neg logit contributions | |
Noise | -1.265 |
Faculty | -1.164 |
grad | -1.149 |
Madness | -1.123 |
graduate | -1.050 |
Token | Pur |
Token ID | 9330 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
Noise | +0.318 |
Faculty | +0.292 |
grad | +0.284 |
Madness | +0.282 |
graduate | +0.260 |
Neg logit contributions | |
ratulations | -0.394 |
chart | -0.376 |
claw | -0.372 |
hess | -0.343 |
=~ | -0.341 |
Token | adolescence |
Token ID | 37258 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
ratulations | +0.313 |
chart | +0.299 |
claw | +0.297 |
hess | +0.273 |
=~ | +0.270 |
Neg logit contributions | |
Noise | -0.259 |
Faculty | -0.239 |
grad | -0.237 |
Madness | -0.230 |
graduate | -0.217 |
Token | . |
Token ID | 764 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
Noise | +1.595 |
Faculty | +1.475 |
grad | +1.464 |
Madness | +1.411 |
graduate | +1.333 |
Neg logit contributions | |
ratulations | -2.038 |
chart | -1.924 |
claw | -1.916 |
hess | -1.761 |
=~ | -1.755 |
Token | Journal |
Token ID | 4913 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
Noise | +1.615 |
Faculty | +1.501 |
grad | +1.421 |
Madness | +1.408 |
graduate | +1.307 |
Neg logit contributions | |
ratulations | -2.241 |
chart | -2.096 |
claw | -2.075 |
hess | -1.941 |
=~ | -1.911 |
Token | of |
Token ID | 286 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
ratulations | +2.134 |
chart | +2.014 |
claw | +1.999 |
hess | +1.853 |
=~ | +1.840 |
Neg logit contributions | |
Noise | -1.542 |
Faculty | -1.421 |
grad | -1.389 |
Madness | -1.361 |
graduate | -1.268 |
Token | Youth |
Token ID | 17582 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
Noise | +3.221 |
Faculty | +3.014 |
grad | +2.884 |
Madness | +2.839 |
graduate | +2.678 |
Neg logit contributions | |
ratulations | -4.223 |
chart | -3.988 |
claw | -3.936 |
hess | -3.694 |
=~ | -3.614 |
Token | and |
Token ID | 290 |
Feature activation | -9.73e+00 |
Pos logit contributions | |
Noise | +2.272 |
Faculty | +2.085 |
grad | +2.007 |
Madness | +1.993 |
graduate | +1.807 |
Neg logit contributions | |
ratulations | -3.460 |
chart | -3.278 |
claw | -3.216 |
hess | -3.087 |
=~ | -2.998 |
Token | Ad |
Token ID | 1215 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
Noise | +7.092 |
Faculty | +6.729 |
Madness | +6.306 |
grad | +6.242 |
Substance | +6.152 |
Neg logit contributions | |
ratulations | -7.705 |
chart | -7.297 |
hess | -7.140 |
claw | -6.982 |
cuff | -6.807 |
Token | olesc |
Token ID | 16850 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
Noise | +4.077 |
Faculty | +3.832 |
grad | +3.722 |
Madness | +3.649 |
yg | +3.578 |
Neg logit contributions | |
ratulations | -4.388 |
chart | -4.226 |
claw | -4.199 |
hess | -3.924 |
cuff | -3.793 |
Token | ence |
Token ID | 594 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
Noise | +2.901 |
Faculty | +2.799 |
grad | +2.721 |
Madness | +2.623 |
graduate | +2.462 |
Neg logit contributions | |
ratulations | -3.788 |
chart | -3.667 |
claw | -3.637 |
hess | -3.380 |
=~ | -3.302 |
Token | , |
Token ID | 11 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
Noise | +2.443 |
Faculty | +2.246 |
grad | +2.216 |
Madness | +2.163 |
graduate | +2.024 |
Neg logit contributions | |
ratulations | -2.939 |
chart | -2.815 |
claw | -2.799 |
hess | -2.561 |
=~ | -2.540 |
Token | 39 |
Token ID | 5014 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Noise | +3.038 |
Faculty | +2.826 |
grad | +2.683 |
Madness | +2.658 |
graduate | +2.440 |
Neg logit contributions | |
ratulations | -4.940 |
chart | -4.540 |
claw | -4.499 |
hess | -4.237 |
=~ | -4.132 |
Token | ESPN |
Token ID | 10409 |
Feature activation | -5.12e+00 |
Pos logit contributions | |
Noise | +1.237 |
Faculty | +1.130 |
grad | +1.112 |
Madness | +1.095 |
graduate | +1.021 |
Neg logit contributions | |
ratulations | -1.531 |
chart | -1.450 |
claw | -1.446 |
hess | -1.327 |
=~ | -1.317 |
Token | analyst |
Token ID | 12499 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
Noise | +3.558 |
Faculty | +3.286 |
grad | +3.226 |
Madness | +3.207 |
graduate | +2.971 |
Neg logit contributions | |
ratulations | -4.386 |
claw | -4.041 |
chart | -4.013 |
hess | -3.781 |
=~ | -3.743 |
Token | Fran |
Token ID | 12323 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
Noise | +1.501 |
Faculty | +1.372 |
grad | +1.347 |
Madness | +1.329 |
graduate | +1.228 |
Neg logit contributions | |
ratulations | -2.033 |
chart | -1.923 |
claw | -1.912 |
hess | -1.774 |
=~ | -1.750 |
Token | Fr |
Token ID | 1305 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
Noise | +1.816 |
Faculty | +1.672 |
grad | +1.630 |
Madness | +1.613 |
graduate | +1.498 |
Neg logit contributions | |
ratulations | -2.163 |
chart | -2.057 |
claw | -2.052 |
=~ | -1.879 |
hess | -1.875 |
Token | as |
Token ID | 292 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ratulations | +0.118 |
chart | +0.113 |
claw | +0.112 |
hess | +0.103 |
=~ | +0.103 |
Neg logit contributions | |
Noise | -0.088 |
Faculty | -0.080 |
grad | -0.079 |
Madness | -0.077 |
graduate | -0.071 |
Token | ch |
Token ID | 354 |
Feature activation | -9.53e+00 |
Pos logit contributions | |
Noise | +1.325 |
Faculty | +1.219 |
grad | +1.186 |
Madness | +1.174 |
graduate | +1.113 |
Neg logit contributions | |
ratulations | -1.505 |
claw | -1.363 |
chart | -1.342 |
hess | -1.272 |
=~ | -1.262 |
Token | illa |
Token ID | 5049 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
Noise | +6.587 |
Faculty | +6.293 |
grad | +6.213 |
Madness | +5.892 |
graduate | +5.763 |
Neg logit contributions | |
ratulations | -7.676 |
chart | -7.032 |
claw | -6.998 |
hess | -6.794 |
=~ | -6.657 |
Token | .[ |
Token ID | 3693 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
Noise | +1.155 |
Faculty | +1.061 |
grad | +1.050 |
Madness | +1.032 |
graduate | +0.962 |
Neg logit contributions | |
ratulations | -1.473 |
claw | -1.378 |
chart | -1.377 |
hess | -1.279 |
=~ | -1.267 |
Token | 94 |
Token ID | 5824 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
ratulations | +0.866 |
chart | +0.843 |
claw | +0.831 |
hess | +0.722 |
=~ | +0.721 |
Neg logit contributions | |
Noise | -1.304 |
Faculty | -1.224 |
grad | -1.206 |
Madness | -1.191 |
graduate | -1.131 |
Token | ] |
Token ID | 60 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
Noise | +2.067 |
Faculty | +1.908 |
grad | +1.882 |
Madness | +1.839 |
graduate | +1.726 |
Neg logit contributions | |
ratulations | -2.318 |
chart | -2.215 |
claw | -2.201 |
hess | -2.009 |
=~ | -1.993 |
Token | O |
Token ID | 440 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
ratulations | +2.058 |
chart | +2.009 |
claw | +1.998 |
hess | +1.810 |
=~ | +1.799 |
Neg logit contributions | |
Noise | -1.967 |
grad | -1.813 |
Faculty | -1.812 |
Madness | -1.750 |
graduate | -1.655 |
Token | ota |
Token ID | 4265 |
Feature activation | -7.87e-01 |
Pos logit contributions | |
Noise | +1.803 |
Faculty | +1.669 |
grad | +1.635 |
Madness | +1.601 |
graduate | +1.496 |
Neg logit contributions | |
ratulations | -2.050 |
claw | -1.924 |
chart | -1.923 |
=~ | -1.772 |
hess | -1.754 |
Token | , |
Token ID | 11 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
Noise | +0.569 |
Faculty | +0.524 |
grad | +0.518 |
Madness | +0.506 |
graduate | +0.475 |
Neg logit contributions | |
ratulations | -0.644 |
chart | -0.612 |
claw | -0.609 |
hess | -0.557 |
=~ | -0.553 |
Token | a |
Token ID | 257 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ratulations | +1.377 |
chart | +1.317 |
claw | +1.309 |
hess | +1.194 |
=~ | +1.184 |
Neg logit contributions | |
Noise | -1.235 |
Faculty | -1.140 |
grad | -1.124 |
Madness | -1.099 |
graduate | -1.031 |
Token | P |
Token ID | 350 |
Feature activation | +6.97e-01 |
Pos logit contributions | |
Noise | +0.947 |
Faculty | +0.871 |
grad | +0.870 |
Madness | +0.832 |
graduate | +0.802 |
Neg logit contributions | |
ratulations | -1.180 |
chart | -1.119 |
claw | -1.113 |
hess | -1.031 |
=~ | -1.015 |
Token | izz |
Token ID | 6457 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ratulations | +0.561 |
chart | +0.537 |
claw | +0.533 |
hess | +0.486 |
=~ | +0.482 |
Neg logit contributions | |
Noise | -0.512 |
Faculty | -0.473 |
grad | -0.467 |
Madness | -0.455 |
graduate | -0.429 |
Token | ag |
Token ID | 363 |
Feature activation | -9.47e+00 |
Pos logit contributions | |
ratulations | +1.457 |
chart | +1.411 |
claw | +1.401 |
hess | +1.296 |
=~ | +1.278 |
Neg logit contributions | |
Noise | -0.860 |
Faculty | -0.775 |
grad | -0.758 |
Madness | -0.742 |
graduate | -0.675 |
Token | ater |
Token ID | 729 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
Noise | +7.758 |
grad | +7.415 |
Faculty | +7.322 |
Madness | +7.120 |
alum | +7.004 |
Neg logit contributions | |
ratulations | -6.130 |
chart | -5.659 |
claw | -5.538 |
=~ | -5.196 |
wcs | -4.991 |
Token | conspiracy |
Token ID | 10086 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Noise | +2.344 |
grad | +2.169 |
Faculty | +2.159 |
Madness | +2.077 |
alum | +2.001 |
Neg logit contributions | |
ratulations | -2.800 |
chart | -2.637 |
claw | -2.610 |
hess | -2.432 |
=~ | -2.406 |
Token | theorist |
Token ID | 46410 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
Noise | +0.149 |
Faculty | +0.135 |
grad | +0.134 |
Madness | +0.129 |
graduate | +0.122 |
Neg logit contributions | |
ratulations | -0.225 |
chart | -0.215 |
claw | -0.214 |
hess | -0.198 |
=~ | -0.197 |
Token | and |
Token ID | 290 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
Noise | +0.240 |
Faculty | +0.221 |
grad | +0.220 |
Madness | +0.213 |
graduate | +0.202 |
Neg logit contributions | |
ratulations | -0.284 |
chart | -0.269 |
claw | -0.268 |
hess | -0.246 |
=~ | -0.244 |
Token | affiliate |
Token ID | 17375 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
Noise | +0.449 |
Faculty | +0.412 |
grad | +0.408 |
Madness | +0.397 |
graduate | +0.374 |
Neg logit contributions | |
ratulations | -0.537 |
chart | -0.512 |
claw | -0.509 |
hess | -0.468 |
=~ | -0.463 |
Token | incorrect |
Token ID | 11491 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
Noise | +0.458 |
Faculty | +0.420 |
grad | +0.416 |
Madness | +0.406 |
graduate | +0.381 |
Neg logit contributions | |
ratulations | -0.538 |
chart | -0.514 |
claw | -0.510 |
hess | -0.468 |
=~ | -0.464 |
Token | handling |
Token ID | 9041 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ratulations | +0.233 |
chart | +0.223 |
claw | +0.221 |
hess | +0.203 |
=~ | +0.202 |
Neg logit contributions | |
Noise | -0.190 |
Faculty | -0.175 |
grad | -0.173 |
Madness | -0.169 |
graduate | -0.158 |
Token | of |
Token ID | 286 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
Noise | +0.822 |
Faculty | +0.753 |
grad | +0.743 |
Madness | +0.724 |
graduate | +0.676 |
Neg logit contributions | |
ratulations | -1.075 |
chart | -1.025 |
claw | -1.019 |
hess | -0.938 |
=~ | -0.932 |
Token | Mam |
Token ID | 29926 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
Noise | +0.236 |
Faculty | +0.217 |
grad | +0.215 |
Madness | +0.209 |
graduate | +0.197 |
Neg logit contributions | |
ratulations | -0.277 |
chart | -0.265 |
claw | -0.263 |
hess | -0.241 |
=~ | -0.239 |
Token | ad |
Token ID | 324 |
Feature activation | -6.27e+00 |
Pos logit contributions | |
Noise | +2.305 |
Faculty | +2.132 |
Madness | +2.111 |
grad | +2.087 |
graduate | +1.919 |
Neg logit contributions | |
ratulations | -3.540 |
claw | -3.343 |
chart | -3.330 |
=~ | -3.070 |
hess | -3.036 |
Token | ou |
Token ID | 280 |
Feature activation | -9.30e+00 |
Pos logit contributions | |
Noise | +2.056 |
Faculty | +1.754 |
grad | +1.694 |
Madness | +1.628 |
Amend | +1.356 |
Neg logit contributions | |
ratulations | -7.561 |
chart | -7.217 |
claw | -7.212 |
hess | -6.830 |
=~ | -6.719 |
Token | Sak |
Token ID | 13231 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
Noise | +3.331 |
grad | +2.795 |
Madness | +2.737 |
Faculty | +2.475 |
graduate | +2.428 |
Neg logit contributions | |
ratulations | -10.403 |
chart | -9.942 |
claw | -9.932 |
wcs | -9.579 |
hess | -9.555 |
Token | ho |
Token ID | 8873 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Noise | +2.512 |
grad | +2.283 |
Faculty | +2.185 |
Amend | +2.151 |
yg | +2.057 |
Neg logit contributions | |
ratulations | -4.954 |
claw | -4.678 |
chart | -4.641 |
=~ | -4.501 |
wcs | -4.491 |
Token | drugs |
Token ID | 5010 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
ratulations | +1.317 |
chart | +1.266 |
claw | +1.261 |
hess | +1.154 |
=~ | +1.141 |
Neg logit contributions | |
Noise | -1.066 |
Faculty | -0.980 |
grad | -0.963 |
Madness | -0.941 |
graduate | -0.876 |
Token | test |
Token ID | 1332 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ratulations | +1.512 |
chart | +1.454 |
claw | +1.447 |
hess | +1.328 |
=~ | +1.316 |
Neg logit contributions | |
Noise | -1.151 |
Faculty | -1.054 |
grad | -1.038 |
Madness | -1.011 |
graduate | -0.941 |
Token | Read |
Token ID | 4149 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
Noise | +1.366 |
Faculty | +1.254 |
grad | +1.234 |
Madness | +1.210 |
graduate | +1.128 |
Neg logit contributions | |
ratulations | -1.750 |
chart | -1.666 |
claw | -1.659 |
hess | -1.522 |
=~ | -1.518 |
Token | it |
Token ID | 340 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
ratulations | +0.583 |
claw | +0.545 |
chart | +0.543 |
hess | +0.500 |
=~ | +0.494 |
Neg logit contributions | |
Noise | -0.387 |
grad | -0.361 |
Faculty | -0.352 |
Madness | -0.348 |
graduate | -0.326 |
Token | be |
Token ID | 307 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
ratulations | +0.128 |
claw | +0.117 |
chart | +0.116 |
hess | +0.107 |
=~ | +0.104 |
Neg logit contributions | |
Noise | -0.130 |
grad | -0.124 |
Faculty | -0.122 |
Madness | -0.118 |
graduate | -0.117 |
Token | expanding |
Token ID | 11581 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
Noise | +0.055 |
Faculty | +0.049 |
grad | +0.049 |
Madness | +0.046 |
graduate | +0.042 |
Neg logit contributions | |
ratulations | -0.120 |
chart | -0.115 |
claw | -0.115 |
hess | -0.107 |
=~ | -0.106 |
Token | ? |
Token ID | 30 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
Noise | +3.504 |
grad | +3.247 |
Faculty | +3.224 |
Madness | +3.198 |
graduate | +2.944 |
Neg logit contributions | |
ratulations | -4.730 |
claw | -4.309 |
chart | -4.214 |
hess | -3.919 |
cuff | -3.851 |
Token | Like |
Token ID | 4525 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
Noise | +0.490 |
Faculty | +0.451 |
grad | +0.444 |
Madness | +0.437 |
graduate | +0.407 |
Neg logit contributions | |
ratulations | -0.562 |
chart | -0.534 |
claw | -0.532 |
hess | -0.486 |
=~ | -0.483 |
Token | … |
Token ID | 1399 |
Feature activation | -9.21e+00 |
Pos logit contributions | |
ratulations | +3.407 |
chart | +3.256 |
claw | +3.224 |
hess | +2.871 |
=~ | +2.869 |
Neg logit contributions | |
Noise | -3.829 |
Faculty | -3.589 |
grad | -3.504 |
Madness | -3.432 |
graduate | -3.265 |
Token | what |
Token ID | 644 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
Noise | +6.230 |
grad | +5.795 |
Faculty | +5.727 |
Madness | +5.481 |
graduate | +5.240 |
Neg logit contributions | |
ratulations | -7.589 |
claw | -7.155 |
chart | -6.999 |
hess | -6.699 |
=~ | -6.467 |
Token | � |
Token ID | 447 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
Noise | +2.437 |
Faculty | +2.270 |
grad | +2.253 |
Madness | +2.215 |
graduate | +2.075 |
Neg logit contributions | |
ratulations | -2.289 |
chart | -2.164 |
claw | -2.136 |
hess | -1.928 |
=~ | -1.922 |
Token | � |
Token ID | 247 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
Noise | +1.651 |
grad | +1.592 |
Faculty | +1.553 |
Madness | +1.513 |
graduate | +1.444 |
Neg logit contributions | |
ratulations | -2.397 |
chart | -2.265 |
claw | -2.237 |
hess | -2.064 |
=~ | -2.014 |
Token | s |
Token ID | 82 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
Noise | +3.505 |
grad | +3.311 |
Faculty | +3.272 |
Madness | +3.201 |
yg | +3.063 |
Neg logit contributions | |
ratulations | -3.261 |
chart | -2.954 |
claw | -2.910 |
=~ | -2.639 |
hess | -2.587 |
Token | past |
Token ID | 1613 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
Noise | +2.136 |
grad | +2.018 |
Faculty | +1.996 |
Madness | +1.871 |
graduate | +1.848 |
Neg logit contributions | |
ratulations | -2.499 |
claw | -2.328 |
chart | -2.304 |
=~ | -2.129 |
hess | -2.116 |
Token | will |
Token ID | 481 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
Noise | +0.824 |
grad | +0.764 |
Faculty | +0.756 |
Madness | +0.736 |
graduate | +0.699 |
Neg logit contributions | |
ratulations | -0.941 |
claw | -0.881 |
chart | -0.880 |
hess | -0.816 |
=~ | -0.790 |
Token | actually |
Token ID | 1682 |
Feature activation | +6.13e-01 |
Pos logit contributions | |
ratulations | +0.055 |
claw | +0.052 |
chart | +0.052 |
hess | +0.048 |
=~ | +0.047 |
Neg logit contributions | |
Noise | -0.048 |
grad | -0.044 |
Faculty | -0.044 |
Madness | -0.043 |
graduate | -0.041 |
Token | cost |
Token ID | 1575 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
ratulations | +0.502 |
claw | +0.474 |
chart | +0.473 |
hess | +0.435 |
=~ | +0.427 |
Neg logit contributions | |
Noise | -0.445 |
grad | -0.409 |
Faculty | -0.408 |
Madness | -0.395 |
graduate | -0.378 |
Token | more |
Token ID | 517 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Noise | +0.371 |
Faculty | +0.338 |
grad | +0.337 |
Madness | +0.325 |
graduate | +0.307 |
Neg logit contributions | |
ratulations | -0.489 |
claw | -0.465 |
chart | -0.465 |
hess | -0.427 |
=~ | -0.416 |
Token | than |
Token ID | 621 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
Noise | +1.048 |
Faculty | +0.960 |
grad | +0.959 |
Madness | +0.927 |
graduate | +0.874 |
Neg logit contributions | |
ratulations | -1.287 |
chart | -1.215 |
claw | -1.214 |
hess | -1.113 |
=~ | -1.097 |
Token | fuel |
Token ID | 5252 |
Feature activation | -9.21e+00 |
Pos logit contributions | |
Noise | +0.217 |
Faculty | +0.199 |
grad | +0.198 |
Madness | +0.191 |
graduate | +0.182 |
Neg logit contributions | |
ratulations | -0.267 |
chart | -0.254 |
claw | -0.253 |
hess | -0.233 |
=~ | -0.229 |
Token | oil |
Token ID | 3056 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
Noise | +5.081 |
grad | +4.838 |
Faculty | +4.494 |
Madness | +4.455 |
alum | +4.256 |
Neg logit contributions | |
ratulations | -8.577 |
claw | -8.147 |
chart | -7.958 |
hess | -7.638 |
=~ | -7.364 |
Token | . |
Token ID | 13 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
Noise | +2.285 |
grad | +2.118 |
Faculty | +2.068 |
Madness | +2.014 |
graduate | +1.904 |
Neg logit contributions | |
ratulations | -3.018 |
claw | -2.855 |
chart | -2.817 |
hess | -2.636 |
=~ | -2.558 |
Token |
|
Token ID | 198 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
Noise | +0.258 |
Faculty | +0.237 |
grad | +0.232 |
Madness | +0.228 |
graduate | +0.212 |
Neg logit contributions | |
ratulations | -0.313 |
chart | -0.295 |
claw | -0.294 |
hess | -0.271 |
=~ | -0.267 |
Token |
|
Token ID | 198 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
ratulations | +0.673 |
chart | +0.624 |
claw | +0.620 |
hess | +0.579 |
=~ | +0.577 |
Neg logit contributions | |
Noise | -0.421 |
Faculty | -0.386 |
grad | -0.378 |
Madness | -0.368 |
graduate | -0.341 |
Token | The |
Token ID | 464 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
Noise | +0.259 |
Faculty | +0.241 |
grad | +0.236 |
Madness | +0.232 |
graduate | +0.218 |
Neg logit contributions | |
ratulations | -0.268 |
claw | -0.249 |
chart | -0.249 |
hess | -0.228 |
=~ | -0.224 |
Token | 2011 |
Token ID | 2813 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
ratulations | +1.033 |
chart | +1.000 |
claw | +0.992 |
hess | +0.920 |
=~ | +0.917 |
Neg logit contributions | |
Noise | -0.549 |
Faculty | -0.492 |
grad | -0.483 |
Madness | -0.468 |
graduate | -0.426 |
Token | book |
Token ID | 1492 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
ratulations | +0.799 |
chart | +0.776 |
claw | +0.771 |
hess | +0.720 |
=~ | +0.717 |
Neg logit contributions | |
Noise | -0.322 |
Faculty | -0.281 |
grad | -0.274 |
Madness | -0.264 |
graduate | -0.233 |
Token | � |
Token ID | 564 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ratulations | +1.892 |
chart | +1.822 |
claw | +1.810 |
hess | +1.645 |
=~ | +1.639 |
Neg logit contributions | |
Noise | -1.637 |
Faculty | -1.511 |
grad | -1.496 |
Madness | -1.451 |
graduate | -1.366 |
Token | � |
Token ID | 250 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
ratulations | +1.181 |
chart | +1.117 |
claw | +1.107 |
hess | +0.994 |
=~ | +0.977 |
Neg logit contributions | |
Noise | -1.474 |
Faculty | -1.378 |
grad | -1.365 |
Madness | -1.338 |
graduate | -1.270 |
Token | Money |
Token ID | 26788 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
Noise | +0.228 |
Faculty | +0.211 |
grad | +0.208 |
Madness | +0.204 |
graduate | +0.192 |
Neg logit contributions | |
ratulations | -0.230 |
chart | -0.217 |
claw | -0.216 |
hess | -0.197 |
=~ | -0.195 |
Token | and |
Token ID | 290 |
Feature activation | -9.16e+00 |
Pos logit contributions | |
Noise | +2.285 |
Faculty | +2.066 |
Madness | +2.056 |
grad | +2.037 |
graduate | +1.848 |
Neg logit contributions | |
ratulations | -2.814 |
chart | -2.642 |
claw | -2.626 |
hess | -2.440 |
=~ | -2.425 |
Token | Power |
Token ID | 4333 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
Noise | +5.960 |
Madness | +5.633 |
Faculty | +5.341 |
grad | +5.168 |
graduate | +4.739 |
Neg logit contributions | |
ratulations | -8.159 |
claw | -7.481 |
chart | -7.463 |
hess | -7.226 |
=~ | -6.962 |
Token | : |
Token ID | 25 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
Noise | +4.156 |
Faculty | +3.771 |
grad | +3.746 |
Madness | +3.719 |
graduate | +3.376 |
Neg logit contributions | |
ratulations | -5.329 |
chart | -5.004 |
claw | -4.972 |
hess | -4.673 |
=~ | -4.616 |
Token | How |
Token ID | 1374 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
ratulations | +1.452 |
chart | +1.400 |
claw | +1.385 |
hess | +1.250 |
=~ | +1.246 |
Neg logit contributions | |
Noise | -1.402 |
Faculty | -1.296 |
grad | -1.289 |
Madness | -1.244 |
graduate | -1.183 |
Token | Goldman |
Token ID | 18740 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
Noise | +1.488 |
Faculty | +1.374 |
grad | +1.346 |
Madness | +1.338 |
graduate | +1.235 |
Neg logit contributions | |
ratulations | -1.714 |
claw | -1.619 |
chart | -1.619 |
hess | -1.487 |
=~ | -1.469 |
Token | Sachs |
Token ID | 25158 |
Feature activation | -8.13e-01 |
Pos logit contributions | |
Noise | +0.431 |
Faculty | +0.401 |
grad | +0.396 |
Madness | +0.389 |
graduate | +0.366 |
Neg logit contributions | |
ratulations | -0.406 |
chart | -0.383 |
claw | -0.381 |
hess | -0.345 |
=~ | -0.342 |
Token | with |
Token ID | 351 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
Noise | +0.764 |
grad | +0.692 |
Faculty | +0.677 |
Madness | +0.661 |
graduate | +0.614 |
Neg logit contributions | |
ratulations | -1.479 |
chart | -1.364 |
claw | -1.362 |
=~ | -1.283 |
hess | -1.273 |
Token | berries |
Token ID | 36322 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
Noise | +1.663 |
grad | +1.597 |
Faculty | +1.521 |
Madness | +1.459 |
graduate | +1.411 |
Neg logit contributions | |
ratulations | -2.759 |
chart | -2.673 |
claw | -2.670 |
hess | -2.472 |
=~ | -2.408 |
Token | , |
Token ID | 11 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
Noise | +0.178 |
Faculty | +0.163 |
grad | +0.160 |
Madness | +0.157 |
graduate | +0.145 |
Neg logit contributions | |
ratulations | -0.261 |
chart | -0.242 |
claw | -0.239 |
hess | -0.222 |
=~ | -0.220 |
Token | but |
Token ID | 475 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
Noise | +0.365 |
grad | +0.331 |
Faculty | +0.322 |
Madness | +0.312 |
graduate | +0.292 |
Neg logit contributions | |
ratulations | -0.710 |
chart | -0.667 |
claw | -0.665 |
=~ | -0.623 |
hess | -0.621 |
Token | not |
Token ID | 407 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
Noise | +0.100 |
grad | +0.092 |
Faculty | +0.090 |
Madness | +0.087 |
graduate | +0.083 |
Neg logit contributions | |
ratulations | -0.146 |
chart | -0.135 |
claw | -0.135 |
hess | -0.126 |
=~ | -0.124 |
Token | too |
Token ID | 1165 |
Feature activation | -8.96e+00 |
Pos logit contributions | |
Noise | +1.991 |
grad | +1.891 |
Faculty | +1.848 |
Madness | +1.809 |
graduate | +1.723 |
Neg logit contributions | |
ratulations | -2.333 |
claw | -2.161 |
chart | -2.146 |
hess | -1.953 |
=~ | -1.940 |
Token | many |
Token ID | 867 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
Noise | +3.942 |
grad | +3.885 |
Faculty | +3.532 |
graduate | +3.285 |
alum | +3.182 |
Neg logit contributions | |
ratulations | -9.613 |
chart | -9.079 |
claw | -9.046 |
hess | -8.383 |
=~ | -8.331 |
Token | ) |
Token ID | 8 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
Noise | +1.326 |
Faculty | +1.214 |
grad | +1.208 |
Madness | +1.168 |
graduate | +1.099 |
Neg logit contributions | |
ratulations | -1.897 |
chart | -1.731 |
claw | -1.710 |
hess | -1.606 |
=~ | -1.584 |
Token | 200 |
Token ID | 939 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
ratulations | +2.107 |
chart | +2.021 |
claw | +2.011 |
hess | +1.834 |
=~ | +1.822 |
Neg logit contributions | |
Noise | -1.803 |
Faculty | -1.662 |
grad | -1.640 |
Madness | -1.602 |
graduate | -1.501 |
Token | g |
Token ID | 70 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
ratulations | +3.241 |
chart | +3.136 |
claw | +3.111 |
hess | +2.843 |
=~ | +2.837 |
Neg logit contributions | |
Noise | -2.775 |
Faculty | -2.545 |
grad | -2.501 |
Madness | -2.471 |
Amend | -2.290 |
Token | of |
Token ID | 286 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
Noise | +2.057 |
grad | +1.899 |
Faculty | +1.897 |
Madness | +1.817 |
alum | +1.742 |
Neg logit contributions | |
ratulations | -2.496 |
chart | -2.343 |
claw | -2.338 |
hess | -2.156 |
=~ | -2.113 |
Token | HA |
Token ID | 7801 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
ratulations | +0.057 |
chart | +0.054 |
claw | +0.054 |
hess | +0.048 |
=~ | +0.048 |
Neg logit contributions | |
Noise | -0.072 |
Faculty | -0.067 |
grad | -0.066 |
Madness | -0.065 |
graduate | -0.062 |
Token | ( |
Token ID | 357 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
Noise | +0.327 |
Faculty | +0.302 |
grad | +0.298 |
Madness | +0.290 |
graduate | +0.273 |
Neg logit contributions | |
ratulations | -0.380 |
chart | -0.362 |
claw | -0.359 |
hess | -0.330 |
=~ | -0.327 |
Token | Occup |
Token ID | 47658 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
ratulations | +0.033 |
chart | +0.031 |
claw | +0.031 |
hess | +0.028 |
=~ | +0.028 |
Neg logit contributions | |
Noise | -0.029 |
Faculty | -0.027 |
grad | -0.027 |
Madness | -0.026 |
graduate | -0.025 |
Token | ational |
Token ID | 864 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
ratulations | +0.149 |
claw | +0.144 |
chart | +0.142 |
hess | +0.130 |
=~ | +0.128 |
Neg logit contributions | |
Noise | -0.166 |
Faculty | -0.154 |
grad | -0.153 |
Madness | -0.148 |
graduate | -0.140 |
Token | Health |
Token ID | 3893 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
Noise | +0.160 |
Faculty | +0.128 |
grad | +0.123 |
Madness | +0.115 |
graduate | +0.093 |
Neg logit contributions | |
ratulations | -0.689 |
chart | -0.670 |
claw | -0.667 |
hess | -0.630 |
=~ | -0.627 |
Token | and |
Token ID | 290 |
Feature activation | -8.87e+00 |
Pos logit contributions | |
ratulations | +0.985 |
chart | +0.946 |
claw | +0.938 |
hess | +0.859 |
=~ | +0.852 |
Neg logit contributions | |
Noise | -0.796 |
Faculty | -0.732 |
grad | -0.723 |
Madness | -0.706 |
graduate | -0.660 |
Token | Safety |
Token ID | 11233 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
Noise | +1.960 |
Faculty | +1.380 |
Madness | +1.075 |
grad | +1.045 |
Substance | +0.750 |
Neg logit contributions | |
ratulations | -11.985 |
chart | -11.588 |
claw | -11.456 |
wcs | -11.369 |
hess | -11.249 |
Token | Administration |
Token ID | 8694 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Noise | +2.746 |
Faculty | +2.485 |
grad | +2.446 |
Madness | +2.347 |
graduate | +2.199 |
Neg logit contributions | |
ratulations | -3.689 |
claw | -3.524 |
chart | -3.502 |
hess | -3.266 |
=~ | -3.235 |
Token | ) |
Token ID | 8 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
Noise | +0.792 |
Faculty | +0.732 |
grad | +0.724 |
Madness | +0.709 |
graduate | +0.666 |
Neg logit contributions | |
ratulations | -0.822 |
chart | -0.785 |
claw | -0.780 |
hess | -0.708 |
=~ | -0.702 |
Token | laws |
Token ID | 3657 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
Noise | +0.157 |
Faculty | +0.145 |
grad | +0.142 |
Madness | +0.139 |
graduate | +0.130 |
Neg logit contributions | |
ratulations | -0.183 |
chart | -0.174 |
claw | -0.173 |
hess | -0.159 |
=~ | -0.158 |
Token | are |
Token ID | 389 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
Noise | +0.141 |
Faculty | +0.131 |
grad | +0.129 |
Madness | +0.126 |
graduate | +0.119 |
Neg logit contributions | |
ratulations | -0.158 |
chart | -0.151 |
claw | -0.150 |
hess | -0.137 |
=~ | -0.136 |
Token | offer |
Token ID | 2897 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
ratulations | +2.270 |
chart | +2.188 |
claw | +2.163 |
hess | +1.961 |
=~ | +1.938 |
Neg logit contributions | |
Noise | -2.214 |
Faculty | -2.060 |
grad | -2.009 |
Madness | -1.983 |
Amend | -1.849 |
Token | the |
Token ID | 262 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
ratulations | +1.354 |
chart | +1.293 |
claw | +1.286 |
hess | +1.172 |
=~ | +1.163 |
Neg logit contributions | |
Noise | -1.222 |
Faculty | -1.128 |
grad | -1.109 |
Madness | -1.089 |
graduate | -1.016 |
Token | Coconut |
Token ID | 46943 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
Noise | +0.452 |
Faculty | +0.418 |
grad | +0.412 |
Madness | +0.402 |
graduate | +0.380 |
Neg logit contributions | |
ratulations | -0.491 |
chart | -0.469 |
claw | -0.466 |
hess | -0.426 |
=~ | -0.420 |
Token | Bay |
Token ID | 4696 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
ratulations | +2.259 |
chart | +2.177 |
claw | +2.163 |
hess | +1.971 |
=~ | +1.953 |
Neg logit contributions | |
Noise | -1.943 |
Faculty | -1.797 |
grad | -1.772 |
Madness | -1.725 |
graduate | -1.624 |
Token | Resort |
Token ID | 28345 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
ratulations | +1.851 |
chart | +1.774 |
claw | +1.766 |
hess | +1.605 |
=~ | +1.593 |
Neg logit contributions | |
Noise | -1.644 |
Faculty | -1.518 |
grad | -1.495 |
Madness | -1.462 |
graduate | -1.369 |
Token | & |
Token ID | 1222 |
Feature activation | -8.79e+00 |
Pos logit contributions | |
Noise | +0.540 |
Faculty | +0.497 |
grad | +0.490 |
Madness | +0.479 |
graduate | +0.447 |
Neg logit contributions | |
ratulations | -0.666 |
chart | -0.635 |
claw | -0.631 |
hess | -0.581 |
=~ | -0.575 |
Token | Spa |
Token ID | 40208 |
Feature activation | -8.17e-01 |
Pos logit contributions | |
Noise | +5.072 |
Faculty | +4.597 |
Madness | +4.496 |
grad | +4.372 |
graduate | +3.960 |
Neg logit contributions | |
ratulations | -8.323 |
chart | -7.860 |
claw | -7.732 |
wcs | -7.311 |
hess | -7.286 |
Token | in |
Token ID | 287 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
Noise | +0.566 |
Faculty | +0.521 |
grad | +0.514 |
Madness | +0.501 |
graduate | +0.469 |
Neg logit contributions | |
ratulations | -0.694 |
chart | -0.663 |
claw | -0.659 |
hess | -0.606 |
=~ | -0.600 |
Token | St |
Token ID | 520 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
Noise | +0.205 |
Faculty | +0.189 |
grad | +0.186 |
Madness | +0.182 |
graduate | +0.171 |
Neg logit contributions | |
ratulations | -0.235 |
chart | -0.224 |
claw | -0.223 |
hess | -0.204 |
=~ | -0.202 |
Token | Lucia |
Token ID | 48158 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ratulations | +1.657 |
chart | +1.600 |
claw | +1.581 |
hess | +1.453 |
=~ | +1.423 |
Neg logit contributions | |
Noise | -1.354 |
Faculty | -1.241 |
grad | -1.223 |
Madness | -1.194 |
graduate | -1.116 |
Token | from |
Token ID | 422 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
Noise | +0.761 |
Faculty | +0.698 |
grad | +0.688 |
Madness | +0.670 |
graduate | +0.626 |
Neg logit contributions | |
ratulations | -1.010 |
chart | -0.963 |
claw | -0.958 |
hess | -0.882 |
=~ | -0.875 |
Token | , |
Token ID | 11 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
ratulations | +0.229 |
chart | +0.220 |
claw | +0.219 |
hess | +0.200 |
=~ | +0.198 |
Neg logit contributions | |
Noise | -0.196 |
Faculty | -0.180 |
grad | -0.178 |
Madness | -0.174 |
graduate | -0.162 |
Token | Ministry |
Token ID | 9475 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
ratulations | +0.482 |
chart | +0.460 |
claw | +0.457 |
hess | +0.416 |
=~ | +0.413 |
Neg logit contributions | |
Noise | -0.439 |
Faculty | -0.406 |
grad | -0.401 |
Madness | -0.392 |
graduate | -0.368 |
Token | of |
Token ID | 286 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
Noise | +0.105 |
Faculty | +0.096 |
grad | +0.095 |
Madness | +0.093 |
graduate | +0.087 |
Neg logit contributions | |
ratulations | -0.123 |
chart | -0.119 |
claw | -0.117 |
hess | -0.107 |
=~ | -0.107 |
Token | Road |
Token ID | 5567 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
Noise | +3.556 |
Faculty | +3.159 |
Madness | +3.089 |
grad | +3.056 |
Amend | +2.775 |
Neg logit contributions | |
ratulations | -5.135 |
chart | -4.885 |
claw | -4.863 |
hess | -4.550 |
=~ | -4.441 |
Token | Transport |
Token ID | 19940 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
Noise | +2.177 |
Faculty | +1.973 |
grad | +1.939 |
Madness | +1.876 |
graduate | +1.732 |
Neg logit contributions | |
ratulations | -3.636 |
chart | -3.504 |
claw | -3.490 |
hess | -3.230 |
=~ | -3.211 |
Token | & |
Token ID | 1222 |
Feature activation | -8.77e+00 |
Pos logit contributions | |
Noise | +3.580 |
Faculty | +3.285 |
grad | +3.215 |
Madness | +3.109 |
graduate | +2.923 |
Neg logit contributions | |
ratulations | -4.997 |
claw | -4.708 |
chart | -4.702 |
=~ | -4.339 |
hess | -4.318 |
Token | High |
Token ID | 3334 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
Noise | +4.859 |
Faculty | +4.206 |
grad | +4.032 |
Madness | +4.000 |
Amend | +3.623 |
Neg logit contributions | |
ratulations | -8.806 |
chart | -8.400 |
claw | -8.290 |
hess | -7.801 |
=~ | -7.651 |
Token | ways |
Token ID | 1322 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
ratulations | +1.186 |
chart | +1.106 |
claw | +1.098 |
hess | +0.910 |
=~ | +0.889 |
Neg logit contributions | |
Noise | -2.740 |
Faculty | -2.609 |
grad | -2.585 |
Madness | -2.547 |
graduate | -2.447 |
Token | constituted |
Token ID | 27276 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
Noise | +3.327 |
Faculty | +3.054 |
grad | +2.979 |
Madness | +2.882 |
Amend | +2.709 |
Neg logit contributions | |
ratulations | -4.745 |
claw | -4.468 |
chart | -4.468 |
=~ | -4.127 |
hess | -4.127 |
Token | a |
Token ID | 257 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
ratulations | +0.296 |
chart | +0.283 |
claw | +0.281 |
hess | +0.256 |
=~ | +0.254 |
Neg logit contributions | |
Noise | -0.278 |
Faculty | -0.257 |
grad | -0.254 |
Madness | -0.248 |
graduate | -0.233 |
Token | Group |
Token ID | 4912 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
ratulations | +0.310 |
chart | +0.297 |
claw | +0.295 |
hess | +0.270 |
=~ | +0.268 |
Neg logit contributions | |
Noise | -0.253 |
Faculty | -0.232 |
grad | -0.229 |
Madness | -0.224 |
graduate | -0.209 |
Token | takeover |
Token ID | 27011 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
ratulations | +0.108 |
chart | +0.103 |
claw | +0.102 |
hess | +0.094 |
=~ | +0.093 |
Neg logit contributions | |
Noise | -0.093 |
Faculty | -0.086 |
grad | -0.085 |
Madness | -0.083 |
graduate | -0.078 |
Token | of |
Token ID | 286 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ratulations | +1.699 |
chart | +1.645 |
claw | +1.627 |
hess | +1.481 |
=~ | +1.472 |
Neg logit contributions | |
Noise | -1.469 |
Faculty | -1.346 |
grad | -1.316 |
Madness | -1.286 |
graduate | -1.197 |
Token | V |
Token ID | 569 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
Noise | +0.838 |
Faculty | +0.778 |
grad | +0.769 |
Madness | +0.751 |
graduate | +0.705 |
Neg logit contributions | |
ratulations | -1.008 |
claw | -0.955 |
chart | -0.955 |
hess | -0.878 |
=~ | -0.866 |
Token | enez |
Token ID | 11437 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
Noise | +1.261 |
Faculty | +1.177 |
grad | +1.174 |
Madness | +1.129 |
graduate | +1.077 |
Neg logit contributions | |
ratulations | -1.357 |
chart | -1.293 |
claw | -1.276 |
=~ | -1.182 |
hess | -1.167 |
Token | ue |
Token ID | 518 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
Noise | +0.519 |
Faculty | +0.487 |
grad | +0.483 |
Madness | +0.474 |
graduate | +0.448 |
Neg logit contributions | |
ratulations | -0.409 |
chart | -0.384 |
claw | -0.380 |
=~ | -0.339 |
hess | -0.337 |
Token | ... |
Token ID | 986 |
Feature activation | -8.76e+00 |
Pos logit contributions | |
Noise | +0.157 |
Faculty | +0.145 |
grad | +0.143 |
Madness | +0.140 |
graduate | +0.131 |
Neg logit contributions | |
ratulations | -0.176 |
chart | -0.167 |
claw | -0.167 |
hess | -0.152 |
=~ | -0.151 |
Token |
|
Token ID | 198 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
Noise | +5.164 |
Faculty | +4.765 |
grad | +4.673 |
Madness | +4.525 |
graduate | +4.216 |
Neg logit contributions | |
ratulations | -8.094 |
claw | -7.614 |
chart | -7.603 |
hess | -7.115 |
=~ | -7.014 |
Token |
|
Token ID | 198 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
Noise | +1.609 |
Faculty | +1.484 |
grad | +1.441 |
Madness | +1.415 |
graduate | +1.288 |
Neg logit contributions | |
ratulations | -3.235 |
chart | -2.898 |
claw | -2.869 |
=~ | -2.724 |
hess | -2.718 |
Token | From |
Token ID | 4863 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
Noise | +1.694 |
Faculty | +1.582 |
grad | +1.554 |
Madness | +1.522 |
graduate | +1.430 |
Neg logit contributions | |
ratulations | -1.894 |
claw | -1.764 |
chart | -1.760 |
hess | -1.627 |
=~ | -1.597 |
Token | Vox |
Token ID | 28035 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
Noise | +3.111 |
Faculty | +2.873 |
grad | +2.842 |
Madness | +2.761 |
graduate | +2.597 |
Neg logit contributions | |
ratulations | -3.808 |
chart | -3.602 |
claw | -3.582 |
hess | -3.293 |
=~ | -3.250 |
Token | : |
Token ID | 25 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
Noise | +0.168 |
Faculty | +0.155 |
grad | +0.153 |
Madness | +0.149 |
graduate | +0.140 |
Neg logit contributions | |
ratulations | -0.196 |
chart | -0.187 |
claw | -0.186 |
hess | -0.171 |
=~ | -0.169 |
Token | Facebook |
Token ID | 3203 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
ratulations | +0.668 |
chart | +0.644 |
claw | +0.639 |
hess | +0.584 |
=~ | +0.579 |
Neg logit contributions | |
Noise | -0.528 |
Faculty | -0.485 |
grad | -0.479 |
Madness | -0.469 |
graduate | -0.435 |
Token | ) |
Token ID | 8 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
Noise | +1.624 |
Faculty | +1.501 |
grad | +1.482 |
Madness | +1.448 |
graduate | +1.364 |
Neg logit contributions | |
ratulations | -1.788 |
chart | -1.697 |
claw | -1.692 |
hess | -1.546 |
=~ | -1.537 |
Token | I |
Token ID | 40 |
Feature activation | -6.81e-01 |
Pos logit contributions | |
Noise | +0.282 |
Faculty | +0.259 |
grad | +0.256 |
Madness | +0.250 |
graduate | +0.234 |
Neg logit contributions | |
ratulations | -0.338 |
chart | -0.324 |
claw | -0.323 |
hess | -0.295 |
=~ | -0.293 |
Token | hor |
Token ID | 17899 |
Feature activation | +7.87e-01 |
Pos logit contributions | |
Noise | +0.380 |
Faculty | +0.341 |
grad | +0.334 |
Madness | +0.326 |
graduate | +0.295 |
Neg logit contributions | |
ratulations | -0.671 |
chart | -0.646 |
claw | -0.643 |
hess | -0.596 |
=~ | -0.591 |
Token | Mal |
Token ID | 4434 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
ratulations | +0.594 |
chart | +0.567 |
claw | +0.567 |
hess | +0.510 |
=~ | +0.506 |
Neg logit contributions | |
Noise | -0.611 |
Faculty | -0.563 |
grad | -0.554 |
Madness | -0.543 |
graduate | -0.512 |
Token | anch |
Token ID | 3702 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
ratulations | +0.299 |
chart | +0.280 |
claw | +0.279 |
=~ | +0.259 |
hess | +0.258 |
Neg logit contributions | |
Noise | -0.225 |
Faculty | -0.208 |
grad | -0.205 |
Madness | -0.198 |
graduate | -0.186 |
Token | ak |
Token ID | 461 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
Noise | +1.325 |
Faculty | +1.048 |
grad | +1.017 |
Madness | +1.006 |
yg | +0.745 |
Neg logit contributions | |
ratulations | -11.458 |
chart | -10.547 |
claw | -10.529 |
=~ | -10.297 |
wcs | -10.135 |
Token | is |
Token ID | 318 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
ratulations | +0.666 |
chart | +0.636 |
claw | +0.633 |
hess | +0.571 |
=~ | +0.566 |
Neg logit contributions | |
Noise | -0.702 |
Faculty | -0.651 |
grad | -0.641 |
Madness | -0.630 |
graduate | -0.592 |
Token | the |
Token ID | 262 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
ratulations | +2.522 |
chart | +2.441 |
claw | +2.428 |
hess | +2.176 |
=~ | +2.161 |
Neg logit contributions | |
Noise | -2.518 |
Faculty | -2.326 |
grad | -2.273 |
Madness | -2.259 |
Amend | -2.106 |
Token | CEO |
Token ID | 6123 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ratulations | +1.363 |
chart | +1.307 |
claw | +1.299 |
hess | +1.180 |
=~ | +1.172 |
Neg logit contributions | |
Noise | -1.264 |
Faculty | -1.166 |
grad | -1.148 |
Madness | -1.127 |
graduate | -1.053 |
Token | of |
Token ID | 286 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
Noise | +0.798 |
Faculty | +0.729 |
grad | +0.718 |
Madness | +0.707 |
graduate | +0.653 |
Neg logit contributions | |
ratulations | -0.969 |
chart | -0.932 |
claw | -0.927 |
hess | -0.846 |
=~ | -0.839 |
Token |
|
Token ID | 198 |
Feature activation | -9.84e-02 |
Pos logit contributions | |
Noise | +0.162 |
Faculty | +0.148 |
grad | +0.146 |
Madness | +0.143 |
graduate | +0.133 |
Neg logit contributions | |
ratulations | -0.207 |
chart | -0.198 |
claw | -0.197 |
hess | -0.181 |
=~ | -0.179 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
Noise | +0.058 |
Faculty | +0.053 |
grad | +0.052 |
Madness | +0.051 |
graduate | +0.047 |
Neg logit contributions | |
ratulations | -0.096 |
chart | -0.089 |
claw | -0.088 |
=~ | -0.083 |
hess | -0.083 |
Token | Also |
Token ID | 7583 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
Noise | +1.030 |
Faculty | +0.958 |
grad | +0.944 |
Madness | +0.917 |
graduate | +0.867 |
Neg logit contributions | |
ratulations | -1.125 |
claw | -1.054 |
chart | -1.052 |
hess | -0.960 |
=~ | -0.954 |
Token | in |
Token ID | 287 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Noise | +0.426 |
Faculty | +0.391 |
grad | +0.387 |
Madness | +0.375 |
graduate | +0.353 |
Neg logit contributions | |
ratulations | -0.528 |
chart | -0.503 |
claw | -0.502 |
hess | -0.461 |
=~ | -0.459 |
Token | Energy |
Token ID | 6682 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
ratulations | +1.222 |
chart | +1.175 |
claw | +1.165 |
hess | +1.061 |
=~ | +1.045 |
Neg logit contributions | |
Noise | -1.148 |
Faculty | -1.065 |
grad | -1.048 |
Madness | -1.027 |
graduate | -0.963 |
Token | & |
Token ID | 1222 |
Feature activation | -8.68e+00 |
Pos logit contributions | |
Noise | +1.819 |
Faculty | +1.632 |
grad | +1.619 |
Madness | +1.585 |
graduate | +1.448 |
Neg logit contributions | |
ratulations | -2.750 |
chart | -2.620 |
claw | -2.597 |
=~ | -2.421 |
hess | -2.403 |
Token | Environment |
Token ID | 9344 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
Noise | +4.666 |
Faculty | +3.970 |
Madness | +3.950 |
grad | +3.819 |
graduate | +3.331 |
Neg logit contributions | |
ratulations | -8.913 |
claw | -8.334 |
chart | -8.272 |
hess | -7.970 |
=~ | -7.816 |
Token | : |
Token ID | 25 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
Noise | +2.300 |
Faculty | +2.121 |
grad | +2.090 |
Madness | +2.041 |
graduate | +1.907 |
Neg logit contributions | |
ratulations | -2.855 |
chart | -2.713 |
claw | -2.698 |
=~ | -2.508 |
hess | -2.484 |
Token |
|
Token ID | 198 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
ratulations | +1.572 |
chart | +1.529 |
claw | +1.501 |
hess | +1.373 |
=~ | +1.348 |
Neg logit contributions | |
Noise | -1.313 |
Faculty | -1.209 |
grad | -1.201 |
Madness | -1.162 |
graduate | -1.097 |
Token |
|
Token ID | 198 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
Noise | +0.556 |
Faculty | +0.511 |
grad | +0.501 |
Madness | +0.488 |
graduate | +0.451 |
Neg logit contributions | |
ratulations | -0.968 |
chart | -0.889 |
claw | -0.882 |
=~ | -0.827 |
hess | -0.825 |
Token | Nearly |
Token ID | 27927 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
Noise | +1.072 |
Faculty | +0.998 |
grad | +0.981 |
Madness | +0.960 |
graduate | +0.903 |
Neg logit contributions | |
ratulations | -1.101 |
claw | -1.028 |
chart | -1.026 |
hess | -0.935 |
=~ | -0.929 |
Token | Energy |
Token ID | 6682 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Noise | +1.755 |
Faculty | +1.559 |
Madness | +1.518 |
grad | +1.486 |
graduate | +1.317 |
Neg logit contributions | |
ratulations | -3.368 |
claw | -3.202 |
chart | -3.185 |
hess | -3.030 |
=~ | -2.950 |
Token | , |
Token ID | 11 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
Noise | +0.943 |
Faculty | +0.866 |
grad | +0.854 |
Madness | +0.832 |
graduate | +0.778 |
Neg logit contributions | |
ratulations | -1.185 |
chart | -1.135 |
claw | -1.128 |
hess | -1.036 |
=~ | -1.027 |
Token | his |
Token ID | 465 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
ratulations | +1.752 |
chart | +1.674 |
claw | +1.655 |
hess | +1.496 |
=~ | +1.491 |
Neg logit contributions | |
Noise | -1.819 |
Faculty | -1.694 |
grad | -1.671 |
Madness | -1.638 |
graduate | -1.535 |
Token | Mitchell |
Token ID | 14526 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
ratulations | +0.627 |
chart | +0.601 |
claw | +0.596 |
hess | +0.544 |
=~ | +0.541 |
Neg logit contributions | |
Noise | -0.534 |
Faculty | -0.491 |
grad | -0.484 |
Madness | -0.474 |
graduate | -0.442 |
Token | Energy |
Token ID | 6682 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
Noise | +0.517 |
Faculty | +0.477 |
grad | +0.471 |
Madness | +0.459 |
graduate | +0.431 |
Neg logit contributions | |
ratulations | -0.621 |
chart | -0.592 |
claw | -0.589 |
hess | -0.542 |
=~ | -0.535 |
Token | & |
Token ID | 1222 |
Feature activation | -8.64e+00 |
Pos logit contributions | |
Noise | +1.998 |
Faculty | +1.853 |
grad | +1.842 |
Madness | +1.785 |
graduate | +1.685 |
Neg logit contributions | |
ratulations | -2.530 |
chart | -2.393 |
claw | -2.393 |
hess | -2.231 |
=~ | -2.188 |
Token | Development |
Token ID | 7712 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
Noise | +5.330 |
Faculty | +4.804 |
Madness | +4.683 |
grad | +4.600 |
graduate | +4.226 |
Neg logit contributions | |
ratulations | -8.037 |
claw | -7.546 |
chart | -7.429 |
hess | -7.257 |
erker | -6.914 |
Token | was |
Token ID | 373 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Noise | +2.357 |
Faculty | +2.209 |
grad | +2.183 |
Madness | +2.108 |
graduate | +2.012 |
Neg logit contributions | |
ratulations | -2.836 |
claw | -2.680 |
chart | -2.666 |
hess | -2.486 |
=~ | -2.453 |
Token | experimenting |
Token ID | 27826 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
ratulations | +0.883 |
chart | +0.844 |
claw | +0.840 |
hess | +0.770 |
=~ | +0.763 |
Neg logit contributions | |
Noise | -0.727 |
Faculty | -0.668 |
grad | -0.659 |
Madness | -0.643 |
graduate | -0.602 |
Token | with |
Token ID | 351 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
ratulations | +1.109 |
chart | +1.068 |
claw | +1.061 |
hess | +0.975 |
=~ | +0.968 |
Neg logit contributions | |
Noise | -0.809 |
Faculty | -0.740 |
grad | -0.723 |
Madness | -0.707 |
graduate | -0.654 |
Token | drilling |
Token ID | 19747 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
ratulations | +2.062 |
chart | +1.979 |
claw | +1.964 |
hess | +1.795 |
=~ | +1.780 |
Neg logit contributions | |
Noise | -1.778 |
Faculty | -1.641 |
grad | -1.607 |
Madness | -1.576 |
graduate | -1.470 |
Token | M |
Token ID | 44 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
ratulations | +0.101 |
chart | +0.095 |
claw | +0.095 |
hess | +0.086 |
=~ | +0.085 |
Neg logit contributions | |
Noise | -0.098 |
Faculty | -0.090 |
grad | -0.089 |
Madness | -0.087 |
graduate | -0.082 |
Token | ISS |
Token ID | 16744 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
Noise | +1.375 |
Faculty | +1.273 |
grad | +1.259 |
Madness | +1.227 |
graduate | +1.160 |
Neg logit contributions | |
ratulations | -1.467 |
chart | -1.392 |
claw | -1.392 |
hess | -1.256 |
=~ | -1.253 |
Token | ING |
Token ID | 2751 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
ratulations | +1.464 |
chart | +1.396 |
claw | +1.385 |
hess | +1.235 |
=~ | +1.228 |
Neg logit contributions | |
Noise | -1.833 |
Faculty | -1.713 |
grad | -1.694 |
Madness | -1.662 |
graduate | -1.575 |
Token | 2 |
Token ID | 362 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ratulations | +0.909 |
chart | +0.867 |
claw | +0.861 |
hess | +0.784 |
=~ | +0.777 |
Neg logit contributions | |
Noise | -0.866 |
Faculty | -0.801 |
grad | -0.791 |
Madness | -0.773 |
graduate | -0.728 |
Token | - |
Token ID | 12 |
Feature activation | -9.32e-02 |
Pos logit contributions | |
Noise | +1.487 |
Faculty | +1.372 |
grad | +1.363 |
Madness | +1.322 |
graduate | +1.252 |
Neg logit contributions | |
ratulations | -1.620 |
chart | -1.541 |
claw | -1.529 |
hess | -1.401 |
=~ | -1.383 |
Token | Y |
Token ID | 56 |
Feature activation | -8.51e+00 |
Pos logit contributions | |
Noise | +0.073 |
Faculty | +0.068 |
grad | +0.067 |
Madness | +0.066 |
graduate | +0.062 |
Neg logit contributions | |
ratulations | -0.070 |
chart | -0.067 |
claw | -0.067 |
hess | -0.060 |
=~ | -0.060 |
Token | EAR |
Token ID | 17133 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
Noise | +4.938 |
Faculty | +4.545 |
grad | +4.483 |
Madness | +4.287 |
alum | +4.087 |
Neg logit contributions | |
ratulations | -7.881 |
chart | -7.602 |
claw | -7.303 |
hess | -6.745 |
=~ | -6.575 |
Token | - |
Token ID | 12 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
ratulations | +0.060 |
chart | +0.056 |
claw | +0.056 |
hess | +0.050 |
=~ | +0.050 |
Neg logit contributions | |
Noise | -0.066 |
Faculty | -0.062 |
grad | -0.061 |
Madness | -0.060 |
graduate | -0.056 |
Token | OLD |
Token ID | 15173 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
ratulations | +1.306 |
chart | +1.248 |
claw | +1.238 |
hess | +1.100 |
=~ | +1.094 |
Neg logit contributions | |
Noise | -1.774 |
Faculty | -1.660 |
grad | -1.644 |
Madness | -1.610 |
graduate | -1.535 |
Token | W |
Token ID | 370 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
Noise | +0.023 |
Faculty | +0.022 |
grad | +0.021 |
Madness | +0.021 |
graduate | +0.020 |
Neg logit contributions | |
ratulations | -0.025 |
chart | -0.024 |
claw | -0.023 |
hess | -0.021 |
=~ | -0.021 |
Token | ASHINGTON |
Token ID | 19436 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
ratulations | +0.406 |
chart | +0.390 |
claw | +0.387 |
hess | +0.356 |
=~ | +0.353 |
Neg logit contributions | |
Noise | -0.308 |
Faculty | -0.282 |
grad | -0.279 |
Madness | -0.271 |
graduate | -0.253 |
Token | One |
Token ID | 1881 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ratulations | +0.125 |
chart | +0.119 |
claw | +0.118 |
hess | +0.108 |
=~ | +0.107 |
Neg logit contributions | |
Noise | -0.100 |
Faculty | -0.092 |
grad | -0.089 |
Madness | -0.088 |
graduate | -0.082 |
Token | # |
Token ID | 1303 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
ratulations | +1.752 |
chart | +1.684 |
claw | +1.676 |
hess | +1.531 |
=~ | +1.512 |
Neg logit contributions | |
Noise | -1.438 |
Faculty | -1.312 |
grad | -1.287 |
Madness | -1.273 |
Amend | -1.178 |
Token | We |
Token ID | 1135 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
Noise | +2.775 |
Faculty | +2.580 |
grad | +2.539 |
Madness | +2.463 |
graduate | +2.327 |
Neg logit contributions | |
ratulations | -3.039 |
chart | -2.863 |
claw | -2.842 |
=~ | -2.613 |
hess | -2.602 |
Token | � |
Token ID | 447 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
Noise | +3.420 |
Faculty | +3.206 |
grad | +3.191 |
Madness | +3.080 |
graduate | +2.954 |
Neg logit contributions | |
ratulations | -3.427 |
chart | -3.269 |
claw | -3.217 |
=~ | -2.930 |
hess | -2.925 |
Token | � |
Token ID | 247 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
Noise | +1.198 |
grad | +1.143 |
Faculty | +1.130 |
Madness | +1.088 |
graduate | +1.043 |
Neg logit contributions | |
ratulations | -1.646 |
chart | -1.561 |
claw | -1.542 |
hess | -1.425 |
=~ | -1.401 |
Token | ve |
Token ID | 303 |
Feature activation | -8.46e+00 |
Pos logit contributions | |
Noise | +1.648 |
Faculty | +1.538 |
grad | +1.520 |
Madness | +1.486 |
graduate | +1.417 |
Neg logit contributions | |
ratulations | -1.505 |
chart | -1.417 |
claw | -1.393 |
=~ | -1.269 |
hess | -1.267 |
Token | Had |
Token ID | 25383 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
Noise | +4.025 |
Faculty | +3.758 |
grad | +3.726 |
Madness | +3.479 |
graduate | +3.277 |
Neg logit contributions | |
ratulations | -8.422 |
chart | -8.072 |
claw | -7.930 |
=~ | -7.465 |
hess | -7.355 |
Token | Enough |
Token ID | 47323 |
Feature activation | -5.04e+00 |
Pos logit contributions | |
Noise | +0.719 |
Faculty | +0.665 |
grad | +0.659 |
Madness | +0.638 |
graduate | +0.590 |
Neg logit contributions | |
ratulations | -1.089 |
chart | -1.043 |
claw | -1.036 |
hess | -0.947 |
=~ | -0.937 |
Token | protester |
Token ID | 34678 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Noise | +3.582 |
grad | +3.374 |
Faculty | +3.342 |
Madness | +3.162 |
graduate | +3.130 |
Neg logit contributions | |
ratulations | -4.060 |
chart | -3.870 |
claw | -3.853 |
=~ | -3.545 |
hess | -3.545 |
Token | in |
Token ID | 287 |
Feature activation | +9.33e-02 |
Pos logit contributions | |
ratulations | +1.056 |
chart | +1.008 |
claw | +1.001 |
hess | +0.914 |
=~ | +0.908 |
Neg logit contributions | |
Noise | -0.961 |
Faculty | -0.887 |
grad | -0.877 |
Madness | -0.856 |
graduate | -0.806 |
Token | Petersburg |
Token ID | 27051 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
ratulations | +0.084 |
chart | +0.080 |
claw | +0.080 |
hess | +0.074 |
=~ | +0.073 |
Neg logit contributions | |
Noise | -0.060 |
Faculty | -0.055 |
grad | -0.054 |
Madness | -0.053 |
graduate | -0.049 |
Token | between |
Token ID | 1022 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
ratulations | +0.841 |
chart | +0.807 |
claw | +0.803 |
hess | +0.736 |
=~ | +0.730 |
Neg logit contributions | |
Noise | -0.662 |
Faculty | -0.607 |
grad | -0.598 |
Madness | -0.584 |
graduate | -0.545 |
Token | these |
Token ID | 777 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
Noise | +1.434 |
Faculty | +1.327 |
grad | +1.320 |
Madness | +1.291 |
graduate | +1.217 |
Neg logit contributions | |
ratulations | -1.542 |
chart | -1.456 |
claw | -1.442 |
hess | -1.331 |
=~ | -1.308 |
Token | two |
Token ID | 734 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
Noise | +0.217 |
Faculty | +0.203 |
grad | +0.202 |
Madness | +0.197 |
graduate | +0.188 |
Neg logit contributions | |
ratulations | -0.183 |
chart | -0.174 |
claw | -0.172 |
hess | -0.155 |
=~ | -0.154 |
Token | ? |
Token ID | 30 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
Noise | +0.132 |
Faculty | +0.123 |
grad | +0.122 |
Madness | +0.119 |
graduate | +0.112 |
Neg logit contributions | |
ratulations | -0.140 |
chart | -0.133 |
claw | -0.132 |
hess | -0.121 |
=~ | -0.120 |
Token |
|
Token ID | 198 |
Feature activation | -8.31e+00 |
Pos logit contributions | |
Noise | +2.149 |
Faculty | +1.965 |
grad | +1.900 |
Madness | +1.891 |
Amend | +1.729 |
Neg logit contributions | |
ratulations | -2.939 |
chart | -2.771 |
claw | -2.768 |
hess | -2.545 |
=~ | -2.542 |
Token |
|
Token ID | 198 |
Feature activation | -8.43e+00 |
Pos logit contributions | |
Noise | +3.201 |
Faculty | +2.998 |
grad | +2.902 |
Madness | +2.771 |
graduate | +2.511 |
Neg logit contributions | |
ratulations | -9.700 |
chart | -8.286 |
claw | -8.154 |
wcs | -8.032 |
akeru | -7.993 |
Token | T |
Token ID | 51 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
Noise | +5.256 |
Faculty | +4.890 |
grad | +4.760 |
Madness | +4.656 |
Loading | +4.653 |
Neg logit contributions | |
ratulations | -7.299 |
claw | -6.447 |
chart | -6.436 |
hess | -6.043 |
=~ | -5.992 |
Token | OM |
Token ID | 2662 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
Noise | +3.231 |
Faculty | +2.997 |
grad | +2.924 |
Madness | +2.879 |
alum | +2.732 |
Neg logit contributions | |
ratulations | -3.420 |
claw | -3.180 |
chart | -3.179 |
hess | -2.881 |
=~ | -2.862 |
Token | KING |
Token ID | 32957 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
ratulations | +2.594 |
chart | +2.553 |
claw | +2.538 |
hess | +2.285 |
=~ | +2.276 |
Neg logit contributions | |
Noise | -2.915 |
Faculty | -2.709 |
grad | -2.663 |
Madness | -2.588 |
graduate | -2.454 |
Token | : |
Token ID | 25 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
ratulations | +2.251 |
chart | +2.147 |
claw | +2.141 |
=~ | +1.911 |
hess | +1.908 |
Neg logit contributions | |
Noise | -2.354 |
Faculty | -2.186 |
grad | -2.157 |
Madness | -2.117 |
graduate | -1.987 |
Token | Well |
Token ID | 3894 |
Feature activation | -3.76e+00 |
Pos logit contributions | |
Noise | +3.053 |
Faculty | +2.796 |
Madness | +2.732 |
grad | +2.634 |
Amend | +2.458 |
Neg logit contributions | |
ratulations | -3.828 |
chart | -3.533 |
claw | -3.498 |
hess | -3.261 |
wcs | -3.221 |
Token | Hare |
Token ID | 42285 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
ratulations | +1.108 |
chart | +1.053 |
claw | +1.049 |
hess | +0.943 |
=~ | +0.936 |
Neg logit contributions | |
Noise | -1.235 |
Faculty | -1.151 |
grad | -1.136 |
Madness | -1.114 |
graduate | -1.052 |
Token | wood |
Token ID | 3822 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
ratulations | +2.418 |
chart | +2.296 |
claw | +2.274 |
hess | +2.089 |
=~ | +2.075 |
Neg logit contributions | |
Noise | -2.067 |
Faculty | -1.908 |
grad | -1.885 |
Madness | -1.840 |
graduate | -1.727 |
Token | to |
Token ID | 284 |
Feature activation | +3.73e+00 |
Pos logit contributions | |
ratulations | +1.020 |
chart | +0.979 |
claw | +0.977 |
hess | +0.890 |
=~ | +0.884 |
Neg logit contributions | |
Noise | -0.842 |
Faculty | -0.776 |
grad | -0.760 |
Madness | -0.743 |
graduate | -0.693 |
Token | tire |
Token ID | 15867 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
ratulations | +2.843 |
claw | +2.740 |
chart | +2.728 |
=~ | +2.464 |
hess | +2.452 |
Neg logit contributions | |
Noise | -2.804 |
Faculty | -2.609 |
grad | -2.544 |
Madness | -2.506 |
graduate | -2.335 |
Token | kicking |
Token ID | 17997 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
Noise | +0.936 |
Faculty | +0.844 |
grad | +0.840 |
Madness | +0.809 |
graduate | +0.752 |
Neg logit contributions | |
ratulations | -1.478 |
chart | -1.408 |
claw | -1.399 |
hess | -1.312 |
=~ | -1.296 |
Token | status |
Token ID | 3722 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
Noise | +1.246 |
grad | +1.150 |
Faculty | +1.146 |
Madness | +1.109 |
graduate | +1.049 |
Neg logit contributions | |
ratulations | -1.526 |
chart | -1.449 |
claw | -1.432 |
hess | -1.333 |
=~ | -1.313 |
Token | . |
Token ID | 13 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
ratulations | +4.035 |
chart | +3.943 |
claw | +3.914 |
hess | +3.487 |
=~ | +3.468 |
Neg logit contributions | |
Noise | -4.103 |
Faculty | -3.818 |
grad | -3.701 |
Madness | -3.653 |
Amend | -3.429 |
Token | We |
Token ID | 775 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
Noise | +0.241 |
Faculty | +0.221 |
grad | +0.217 |
Madness | +0.213 |
graduate | +0.199 |
Neg logit contributions | |
ratulations | -0.295 |
chart | -0.281 |
claw | -0.279 |
hess | -0.257 |
=~ | -0.254 |
Token | 'll |
Token ID | 1183 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
ratulations | +1.102 |
chart | +1.051 |
claw | +1.044 |
hess | +0.947 |
=~ | +0.938 |
Neg logit contributions | |
Noise | -1.093 |
Faculty | -1.013 |
grad | -0.999 |
Madness | -0.979 |
graduate | -0.921 |
Token | update |
Token ID | 4296 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
ratulations | +0.687 |
chart | +0.656 |
claw | +0.652 |
hess | +0.600 |
=~ | +0.594 |
Neg logit contributions | |
Noise | -0.568 |
Faculty | -0.522 |
grad | -0.516 |
Madness | -0.502 |
graduate | -0.473 |
Token | you |
Token ID | 345 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ratulations | +1.052 |
chart | +1.004 |
claw | +0.998 |
hess | +0.908 |
=~ | +0.899 |
Neg logit contributions | |
Noise | -1.011 |
Faculty | -0.935 |
grad | -0.923 |
Madness | -0.903 |
graduate | -0.850 |
Token | � |
Token ID | 447 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
ratulations | +1.478 |
chart | +1.409 |
claw | +1.400 |
hess | +1.266 |
=~ | +1.254 |
Neg logit contributions | |
Noise | -1.563 |
Faculty | -1.453 |
grad | -1.434 |
Madness | -1.405 |
graduate | -1.326 |
Token | � |
Token ID | 247 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Noise | +0.477 |
grad | +0.451 |
Faculty | +0.445 |
Madness | +0.431 |
graduate | +0.409 |
Neg logit contributions | |
ratulations | -0.631 |
chart | -0.599 |
claw | -0.589 |
hess | -0.543 |
=~ | -0.533 |
Token | s |
Token ID | 82 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
Noise | +0.224 |
Faculty | +0.210 |
grad | +0.209 |
Madness | +0.205 |
graduate | +0.195 |
Neg logit contributions | |
ratulations | -0.159 |
chart | -0.150 |
claw | -0.149 |
hess | -0.131 |
=~ | -0.130 |
Token | a |
Token ID | 257 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
ratulations | +0.870 |
chart | +0.832 |
claw | +0.826 |
hess | +0.756 |
=~ | +0.750 |
Neg logit contributions | |
Noise | -0.753 |
Faculty | -0.694 |
grad | -0.686 |
Madness | -0.669 |
graduate | -0.627 |
Token | hint |
Token ID | 9254 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
ratulations | +1.161 |
chart | +1.105 |
claw | +1.096 |
hess | +1.003 |
=~ | +0.997 |
Neg logit contributions | |
Noise | -1.005 |
Faculty | -0.928 |
grad | -0.920 |
Madness | -0.894 |
graduate | -0.842 |
Token | of |
Token ID | 286 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
ratulations | +0.601 |
chart | +0.576 |
claw | +0.572 |
hess | +0.525 |
=~ | +0.521 |
Neg logit contributions | |
Noise | -0.485 |
Faculty | -0.445 |
grad | -0.439 |
Madness | -0.429 |
graduate | -0.400 |
Token | that |
Token ID | 326 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
ratulations | +4.323 |
claw | +4.177 |
chart | +4.166 |
hess | +3.770 |
=~ | +3.730 |
Neg logit contributions | |
Noise | -4.015 |
Faculty | -3.726 |
grad | -3.625 |
Madness | -3.541 |
graduate | -3.353 |
Token | decon |
Token ID | 37431 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
ratulations | +2.387 |
chart | +2.290 |
claw | +2.275 |
hess | +2.074 |
=~ | +2.055 |
Neg logit contributions | |
Noise | -2.165 |
Faculty | -1.999 |
grad | -1.969 |
Madness | -1.928 |
graduate | -1.809 |
Token | struction |
Token ID | 15019 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
Noise | +0.255 |
Faculty | +0.238 |
grad | +0.235 |
Madness | +0.230 |
graduate | +0.217 |
Neg logit contributions | |
ratulations | -0.240 |
chart | -0.225 |
claw | -0.225 |
hess | -0.203 |
=~ | -0.202 |
Token | in |
Token ID | 287 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
ratulations | +1.741 |
chart | +1.665 |
claw | +1.655 |
hess | +1.514 |
=~ | +1.501 |
Neg logit contributions | |
Noise | -1.487 |
Faculty | -1.369 |
grad | -1.351 |
Madness | -1.319 |
graduate | -1.236 |
Token | Pre |
Token ID | 3771 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
ratulations | +2.880 |
chart | +2.782 |
claw | +2.780 |
hess | +2.525 |
=~ | +2.514 |
Neg logit contributions | |
Noise | -2.321 |
Faculty | -2.147 |
grad | -2.098 |
Madness | -2.048 |
graduate | -1.920 |
Token | department |
Token ID | 5011 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
Noise | +0.462 |
Faculty | +0.421 |
grad | +0.414 |
Madness | +0.407 |
graduate | +0.371 |
Neg logit contributions | |
ratulations | -0.665 |
chart | -0.627 |
claw | -0.626 |
hess | -0.580 |
=~ | -0.577 |
Token | " |
Token ID | 366 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
Noise | +0.419 |
Faculty | +0.387 |
grad | +0.382 |
Madness | +0.373 |
graduate | +0.350 |
Neg logit contributions | |
ratulations | -0.484 |
chart | -0.462 |
claw | -0.460 |
hess | -0.420 |
=~ | -0.417 |
Token | has |
Token ID | 10134 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
ratulations | +0.352 |
chart | +0.335 |
claw | +0.333 |
hess | +0.303 |
=~ | +0.300 |
Neg logit contributions | |
Noise | -0.340 |
Faculty | -0.315 |
grad | -0.311 |
Madness | -0.304 |
graduate | -0.286 |
Token | known |
Token ID | 1900 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
Noise | +0.287 |
Faculty | +0.265 |
grad | +0.261 |
Madness | +0.255 |
graduate | +0.239 |
Neg logit contributions | |
ratulations | -0.337 |
chart | -0.322 |
claw | -0.320 |
hess | -0.293 |
=~ | -0.291 |
Token | all |
Token ID | 477 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
ratulations | +0.505 |
chart | +0.482 |
claw | +0.479 |
hess | +0.437 |
=~ | +0.434 |
Neg logit contributions | |
Noise | -0.456 |
Faculty | -0.422 |
grad | -0.416 |
Madness | -0.406 |
graduate | -0.382 |
Token | along |
Token ID | 1863 |
Feature activation | +5.29e+00 |
Pos logit contributions | |
ratulations | +1.590 |
chart | +1.520 |
claw | +1.512 |
hess | +1.381 |
=~ | +1.370 |
Neg logit contributions | |
Noise | -1.374 |
Faculty | -1.265 |
grad | -1.247 |
Madness | -1.220 |
graduate | -1.141 |
Token | " |
Token ID | 1 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
ratulations | +3.888 |
chart | +3.738 |
claw | +3.672 |
hess | +3.317 |
=~ | +3.286 |
Neg logit contributions | |
Noise | -4.102 |
Faculty | -3.809 |
grad | -3.729 |
Madness | -3.695 |
Amend | -3.442 |
Token | that |
Token ID | 326 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
Noise | +0.139 |
Faculty | +0.129 |
grad | +0.127 |
Madness | +0.124 |
graduate | +0.117 |
Neg logit contributions | |
ratulations | -0.153 |
chart | -0.146 |
claw | -0.145 |
hess | -0.132 |
=~ | -0.131 |
Token | WikiLeaks |
Token ID | 19766 |
Feature activation | +5.81e-01 |
Pos logit contributions | |
Noise | +0.620 |
Faculty | +0.572 |
grad | +0.566 |
Madness | +0.551 |
graduate | +0.521 |
Neg logit contributions | |
ratulations | -0.681 |
chart | -0.649 |
claw | -0.645 |
hess | -0.589 |
=~ | -0.584 |
Token | possesses |
Token ID | 22194 |
Feature activation | +5.09e-01 |
Pos logit contributions | |
ratulations | +0.478 |
chart | +0.457 |
claw | +0.454 |
hess | +0.415 |
=~ | +0.411 |
Neg logit contributions | |
Noise | -0.417 |
Faculty | -0.385 |
grad | -0.380 |
Madness | -0.371 |
graduate | -0.348 |
Token | classified |
Token ID | 10090 |
Feature activation | +4.34e+00 |
Pos logit contributions | |
ratulations | +0.422 |
chart | +0.402 |
claw | +0.400 |
hess | +0.366 |
=~ | +0.364 |
Neg logit contributions | |
Noise | -0.364 |
Faculty | -0.337 |
grad | -0.333 |
Madness | -0.323 |
graduate | -0.307 |
Token | Mem |
Token ID | 4942 |
Feature activation | +5.26e+00 |
Pos logit contributions | |
chart | +5.199 |
ratulations | +5.181 |
claw | +5.159 |
=~ | +4.633 |
hess | +4.614 |
Neg logit contributions | |
Noise | -5.241 |
Faculty | -4.932 |
grad | -4.913 |
Madness | -4.748 |
graduate | -4.574 |
Token | c |
Token ID | 66 |
Feature activation | +5.30e+00 |
Pos logit contributions | |
ratulations | +3.720 |
chart | +3.584 |
claw | +3.567 |
hess | +3.186 |
=~ | +3.149 |
Neg logit contributions | |
Noise | -4.326 |
Faculty | -4.029 |
grad | -3.975 |
Madness | -3.910 |
graduate | -3.689 |
Token | py |
Token ID | 9078 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
ratulations | +2.490 |
chart | +2.347 |
claw | +2.321 |
hess | +1.935 |
=~ | +1.900 |
Neg logit contributions | |
Noise | -5.624 |
Faculty | -5.327 |
grad | -5.257 |
Madness | -5.217 |
Amend | -4.981 |
Token | Reading |
Token ID | 36120 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
ratulations | +1.857 |
chart | +1.764 |
claw | +1.743 |
=~ | +1.528 |
hess | +1.523 |
Neg logit contributions | |
Noise | -2.886 |
Faculty | -2.726 |
grad | -2.695 |
Madness | -2.653 |
graduate | -2.531 |
Token | 55 |
Token ID | 5996 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
ratulations | +0.568 |
chart | +0.549 |
claw | +0.546 |
hess | +0.502 |
=~ | +0.501 |
Neg logit contributions | |
Noise | -0.366 |
Faculty | -0.330 |
grad | -0.324 |
Madness | -0.318 |
graduate | -0.290 |
Token | 80 |
Token ID | 1795 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
ratulations | +3.309 |
chart | +3.150 |
claw | +3.128 |
hess | +2.763 |
=~ | +2.736 |
Neg logit contributions | |
Noise | -4.825 |
Faculty | -4.522 |
grad | -4.479 |
Madness | -4.392 |
graduate | -4.190 |
Token | 953 |
Token ID | 49649 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
ratulations | +2.619 |
chart | +2.410 |
claw | +2.397 |
hess | +1.892 |
=~ | +1.878 |
Neg logit contributions | |
Noise | -7.848 |
Faculty | -7.442 |
grad | -7.387 |
Madness | -7.287 |
graduate | -7.018 |
Token | 16 |
Token ID | 1467 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
Noise | +2.092 |
Faculty | +1.905 |
grad | +1.892 |
Madness | +1.839 |
graduate | +1.713 |
Neg logit contributions | |
ratulations | -2.884 |
chart | -2.758 |
claw | -2.748 |
hess | -2.525 |
=~ | -2.498 |
Token | 27 |
Token ID | 1983 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
ratulations | +0.028 |
chart | +0.026 |
claw | +0.025 |
hess | +0.024 |
=~ | +0.023 |
Neg logit contributions | |
Noise | -0.025 |
Faculty | -0.023 |
grad | -0.023 |
Madness | -0.022 |
graduate | -0.021 |
Token | 145 |
Token ID | 18781 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ratulations | +3.010 |
chart | +2.881 |
claw | +2.865 |
hess | +2.633 |
=~ | +2.618 |
Neg logit contributions | |
Noise | -2.221 |
Faculty | -2.035 |
grad | -2.003 |
Madness | -1.953 |
graduate | -1.819 |
Token | Flat |
Token ID | 21939 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
ratulations | +0.803 |
chart | +0.769 |
claw | +0.764 |
hess | +0.696 |
=~ | +0.690 |
Neg logit contributions | |
Noise | -0.744 |
Faculty | -0.693 |
grad | -0.681 |
Madness | -0.668 |
graduate | -0.628 |
Token | high |
Token ID | 1029 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Noise | +0.785 |
grad | +0.736 |
Faculty | +0.733 |
Madness | +0.691 |
graduate | +0.675 |
Neg logit contributions | |
ratulations | -0.964 |
chart | -0.912 |
claw | -0.911 |
hess | -0.838 |
=~ | -0.824 |
Token | schools |
Token ID | 4266 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
ratulations | +0.814 |
chart | +0.776 |
claw | +0.771 |
hess | +0.699 |
=~ | +0.693 |
Neg logit contributions | |
Noise | -0.824 |
Faculty | -0.765 |
grad | -0.756 |
Madness | -0.739 |
graduate | -0.697 |
Token | start |
Token ID | 923 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
Noise | +2.148 |
Faculty | +2.001 |
grad | +1.982 |
Madness | +1.893 |
graduate | +1.843 |
Neg logit contributions | |
ratulations | -2.937 |
claw | -2.757 |
chart | -2.756 |
hess | -2.558 |
=~ | -2.554 |
Token | at |
Token ID | 379 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
Noise | +2.988 |
grad | +2.883 |
Faculty | +2.815 |
Madness | +2.672 |
graduate | +2.607 |
Neg logit contributions | |
ratulations | -4.122 |
chart | -3.880 |
claw | -3.875 |
hess | -3.569 |
=~ | -3.533 |
Token | 8 |
Token ID | 807 |
Feature activation | +7.81e-01 |
Pos logit contributions | |
Noise | +0.519 |
Faculty | +0.481 |
grad | +0.480 |
Madness | +0.463 |
graduate | +0.439 |
Neg logit contributions | |
ratulations | -0.588 |
chart | -0.556 |
claw | -0.555 |
hess | -0.507 |
=~ | -0.501 |
Token | : |
Token ID | 25 |
Feature activation | +7.71e+00 |
Pos logit contributions | |
ratulations | +0.595 |
chart | +0.567 |
claw | +0.561 |
=~ | +0.508 |
hess | +0.506 |
Neg logit contributions | |
Noise | -0.604 |
Faculty | -0.560 |
grad | -0.548 |
Madness | -0.538 |
graduate | -0.507 |
Token | 30 |
Token ID | 1270 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
ratulations | +4.304 |
chart | +4.301 |
claw | +4.260 |
=~ | +3.501 |
hess | +3.369 |
Neg logit contributions | |
Noise | -7.051 |
Faculty | -6.400 |
grad | -6.350 |
Madness | -6.345 |
Amend | -6.044 |
Token | or |
Token ID | 393 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
ratulations | +1.238 |
chart | +1.180 |
claw | +1.167 |
hess | +1.049 |
=~ | +1.048 |
Neg logit contributions | |
Noise | -1.349 |
Faculty | -1.248 |
grad | -1.216 |
Madness | -1.200 |
graduate | -1.129 |
Token | later |
Token ID | 1568 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Noise | +0.241 |
Faculty | +0.222 |
grad | +0.220 |
Madness | +0.215 |
graduate | +0.202 |
Neg logit contributions | |
ratulations | -0.280 |
chart | -0.266 |
claw | -0.265 |
hess | -0.242 |
=~ | -0.239 |
Token | , |
Token ID | 11 |
Feature activation | -9.49e-02 |
Pos logit contributions | |
ratulations | +1.264 |
chart | +1.217 |
claw | +1.207 |
hess | +1.096 |
=~ | +1.087 |
Neg logit contributions | |
Noise | -1.156 |
Faculty | -1.062 |
grad | -1.040 |
Madness | -1.023 |
graduate | -0.953 |
Token | and |
Token ID | 290 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
Noise | +0.055 |
Faculty | +0.050 |
grad | +0.049 |
Madness | +0.047 |
graduate | +0.044 |
Neg logit contributions | |
ratulations | -0.091 |
chart | -0.088 |
claw | -0.087 |
hess | -0.081 |
=~ | -0.080 |
Token | you |
Token ID | 345 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
ratulations | +1.431 |
chart | +1.368 |
claw | +1.360 |
hess | +1.242 |
=~ | +1.231 |
Neg logit contributions | |
Noise | -1.273 |
Faculty | -1.174 |
grad | -1.159 |
Madness | -1.132 |
graduate | -1.063 |
Token | 're |
Token ID | 821 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ratulations | +1.403 |
chart | +1.332 |
claw | +1.324 |
hess | +1.171 |
=~ | +1.156 |
Neg logit contributions | |
Noise | -1.872 |
Faculty | -1.751 |
grad | -1.721 |
Madness | -1.700 |
Amend | -1.597 |
Token | not |
Token ID | 407 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
ratulations | +0.822 |
chart | +0.783 |
claw | +0.779 |
hess | +0.703 |
=~ | +0.697 |
Neg logit contributions | |
Noise | -0.874 |
Faculty | -0.811 |
grad | -0.800 |
Madness | -0.785 |
graduate | -0.739 |
Token | supposed |
Token ID | 4385 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
ratulations | +0.158 |
chart | +0.150 |
claw | +0.149 |
hess | +0.134 |
=~ | +0.133 |
Neg logit contributions | |
Noise | -0.173 |
Faculty | -0.161 |
grad | -0.160 |
Madness | -0.156 |
graduate | -0.148 |
Token | to |
Token ID | 284 |
Feature activation | +3.84e-01 |
Pos logit contributions | |
ratulations | +1.283 |
chart | +1.201 |
claw | +1.197 |
hess | +1.104 |
=~ | +1.101 |
Neg logit contributions | |
Noise | -0.869 |
Faculty | -0.795 |
grad | -0.794 |
Madness | -0.765 |
graduate | -0.723 |
Token | show |
Token ID | 905 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ratulations | +0.313 |
chart | +0.298 |
claw | +0.296 |
hess | +0.272 |
=~ | +0.269 |
Neg logit contributions | |
Noise | -0.279 |
Faculty | -0.257 |
grad | -0.256 |
Madness | -0.248 |
graduate | -0.237 |
Token | you |
Token ID | 345 |
Feature activation | +8.20e-01 |
Pos logit contributions | |
ratulations | +0.792 |
chart | +0.756 |
claw | +0.751 |
hess | +0.682 |
=~ | +0.676 |
Neg logit contributions | |
Noise | -0.793 |
Faculty | -0.736 |
grad | -0.728 |
Madness | -0.711 |
graduate | -0.672 |
Token | don |
Token ID | 836 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
ratulations | +0.586 |
chart | +0.557 |
claw | +0.552 |
hess | +0.499 |
=~ | +0.496 |
Neg logit contributions | |
Noise | -0.675 |
Faculty | -0.630 |
grad | -0.626 |
Madness | -0.613 |
graduate | -0.582 |
Token | 't |
Token ID | 470 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ratulations | +2.385 |
chart | +2.226 |
claw | +2.204 |
hess | +1.969 |
=~ | +1.966 |
Neg logit contributions | |
Noise | -3.112 |
Faculty | -2.915 |
grad | -2.884 |
Madness | -2.831 |
graduate | -2.694 |
Token | know |
Token ID | 760 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
ratulations | +0.889 |
chart | +0.846 |
claw | +0.840 |
hess | +0.763 |
=~ | +0.756 |
Neg logit contributions | |
Noise | -0.917 |
Faculty | -0.850 |
grad | -0.843 |
Madness | -0.823 |
graduate | -0.781 |
Token | what |
Token ID | 644 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
ratulations | +2.232 |
chart | +2.123 |
claw | +2.109 |
hess | +1.892 |
=~ | +1.866 |
Neg logit contributions | |
Noise | -2.498 |
Faculty | -2.324 |
grad | -2.278 |
Madness | -2.245 |
graduate | -2.109 |
Token | , |
Token ID | 11 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ratulations | +0.080 |
chart | +0.076 |
claw | +0.076 |
hess | +0.069 |
=~ | +0.069 |
Neg logit contributions | |
Noise | -0.070 |
Faculty | -0.065 |
grad | -0.064 |
Madness | -0.063 |
graduate | -0.059 |
Token | as |
Token ID | 355 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
Noise | +0.098 |
Faculty | +0.090 |
grad | +0.090 |
Madness | +0.087 |
graduate | +0.082 |
Neg logit contributions | |
ratulations | -0.114 |
chart | -0.109 |
claw | -0.108 |
hess | -0.100 |
=~ | -0.099 |
Token | opposed |
Token ID | 6886 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Noise | +0.375 |
Faculty | +0.343 |
grad | +0.340 |
Madness | +0.331 |
graduate | +0.310 |
Neg logit contributions | |
ratulations | -0.461 |
chart | -0.440 |
claw | -0.438 |
hess | -0.404 |
=~ | -0.398 |
Token | to |
Token ID | 284 |
Feature activation | +7.74e-01 |
Pos logit contributions | |
Noise | +0.912 |
grad | +0.824 |
Faculty | +0.819 |
Madness | +0.791 |
graduate | +0.745 |
Neg logit contributions | |
ratulations | -1.447 |
chart | -1.347 |
claw | -1.341 |
hess | -1.256 |
=~ | -1.236 |
Token | being |
Token ID | 852 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
ratulations | +0.634 |
chart | +0.605 |
claw | +0.601 |
hess | +0.551 |
=~ | +0.545 |
Neg logit contributions | |
Noise | -0.559 |
Faculty | -0.515 |
grad | -0.509 |
Madness | -0.496 |
graduate | -0.466 |
Token | , |
Token ID | 11 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
ratulations | +2.007 |
chart | +1.930 |
claw | +1.917 |
hess | +1.737 |
=~ | +1.732 |
Neg logit contributions | |
Noise | -1.770 |
Faculty | -1.635 |
grad | -1.605 |
Madness | -1.576 |
graduate | -1.473 |
Token | � |
Token ID | 564 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
ratulations | +1.273 |
chart | +1.215 |
claw | +1.207 |
hess | +1.098 |
=~ | +1.089 |
Neg logit contributions | |
Noise | -1.201 |
Faculty | -1.111 |
grad | -1.096 |
Madness | -1.073 |
graduate | -1.009 |
Token | � |
Token ID | 250 |
Feature activation | -9.12e-01 |
Pos logit contributions | |
Noise | +0.714 |
grad | +0.666 |
Faculty | +0.664 |
Madness | +0.647 |
graduate | +0.615 |
Neg logit contributions | |
ratulations | -0.636 |
chart | -0.606 |
claw | -0.598 |
hess | -0.545 |
=~ | -0.532 |
Token | Ah |
Token ID | 10910 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
Noise | +0.691 |
Faculty | +0.638 |
grad | +0.630 |
Madness | +0.617 |
graduate | +0.580 |
Neg logit contributions | |
ratulations | -0.713 |
chart | -0.674 |
claw | -0.671 |
hess | -0.615 |
=~ | -0.606 |
Token | , |
Token ID | 11 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
Noise | +0.612 |
Faculty | +0.560 |
grad | +0.558 |
Madness | +0.542 |
graduate | +0.508 |
Neg logit contributions | |
ratulations | -0.746 |
chart | -0.714 |
claw | -0.711 |
hess | -0.658 |
=~ | -0.649 |
Token | racism |
Token ID | 10713 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
Noise | +0.277 |
Faculty | +0.255 |
grad | +0.254 |
Madness | +0.248 |
graduate | +0.234 |
Neg logit contributions | |
ratulations | -0.284 |
chart | -0.271 |
claw | -0.269 |
hess | -0.248 |
=~ | -0.243 |
Token | ; |
Token ID | 26 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ratulations | +0.308 |
chart | +0.294 |
claw | +0.292 |
hess | +0.266 |
=~ | +0.264 |
Neg logit contributions | |
Noise | -0.286 |
Faculty | -0.265 |
grad | -0.261 |
Madness | -0.255 |
graduate | -0.240 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
Noise | +0.726 |
Faculty | +0.667 |
grad | +0.661 |
Madness | +0.641 |
graduate | +0.604 |
Neg logit contributions | |
ratulations | -0.885 |
chart | -0.844 |
claw | -0.840 |
hess | -0.772 |
=~ | -0.763 |
Token | , |
Token ID | 11 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
ratulations | +0.181 |
chart | +0.173 |
claw | +0.172 |
hess | +0.157 |
=~ | +0.155 |
Neg logit contributions | |
Noise | -0.163 |
Faculty | -0.151 |
grad | -0.149 |
Madness | -0.145 |
graduate | -0.137 |
Token | finally |
Token ID | 3443 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
Noise | +0.172 |
Faculty | +0.158 |
grad | +0.156 |
Madness | +0.151 |
graduate | +0.143 |
Neg logit contributions | |
ratulations | -0.218 |
claw | -0.207 |
chart | -0.207 |
hess | -0.191 |
=~ | -0.189 |
Token | , |
Token ID | 11 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ratulations | +1.184 |
chart | +1.122 |
claw | +1.116 |
hess | +1.029 |
=~ | +1.022 |
Neg logit contributions | |
Noise | -0.928 |
Faculty | -0.854 |
grad | -0.849 |
Madness | -0.822 |
graduate | -0.774 |
Token | rising |
Token ID | 7396 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
Noise | +1.035 |
Faculty | +0.955 |
grad | +0.951 |
Madness | +0.918 |
graduate | +0.873 |
Neg logit contributions | |
ratulations | -1.192 |
chart | -1.129 |
claw | -1.124 |
hess | -1.033 |
=~ | -1.020 |
Token | so |
Token ID | 523 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
ratulations | +0.544 |
chart | +0.518 |
claw | +0.516 |
hess | +0.471 |
=~ | +0.466 |
Neg logit contributions | |
Noise | -0.494 |
Faculty | -0.456 |
grad | -0.451 |
Madness | -0.439 |
graduate | -0.414 |
Token | ya |
Token ID | 3972 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
Noise | +0.850 |
grad | +0.789 |
Faculty | +0.779 |
Madness | +0.754 |
graduate | +0.718 |
Neg logit contributions | |
ratulations | -0.969 |
claw | -0.908 |
chart | -0.903 |
hess | -0.835 |
=~ | -0.825 |
Token | bean |
Token ID | 26394 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
ratulations | +3.536 |
chart | +3.372 |
claw | +3.369 |
hess | +3.042 |
=~ | +3.030 |
Neg logit contributions | |
Noise | -3.607 |
Faculty | -3.378 |
Madness | -3.249 |
grad | -3.229 |
Amend | -3.038 |
Token | and |
Token ID | 290 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
ratulations | +0.182 |
chart | +0.172 |
claw | +0.172 |
hess | +0.155 |
=~ | +0.153 |
Neg logit contributions | |
Noise | -0.205 |
Faculty | -0.190 |
grad | -0.189 |
Madness | -0.184 |
graduate | -0.175 |
Token | so |
Token ID | 523 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
ratulations | +1.780 |
chart | +1.697 |
claw | +1.686 |
hess | +1.525 |
=~ | +1.512 |
Neg logit contributions | |
Noise | -1.847 |
Faculty | -1.716 |
grad | -1.691 |
Madness | -1.660 |
graduate | -1.561 |
Token | { |
Token ID | 1391 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
ratulations | +2.789 |
chart | +2.685 |
claw | +2.647 |
hess | +2.436 |
=~ | +2.424 |
Neg logit contributions | |
Noise | -2.546 |
Faculty | -2.357 |
grad | -2.314 |
Madness | -2.262 |
graduate | -2.138 |
Token | 1 |
Token ID | 16 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
Noise | +0.589 |
Faculty | +0.545 |
grad | +0.538 |
Madness | +0.526 |
graduate | +0.494 |
Neg logit contributions | |
ratulations | -0.602 |
chart | -0.568 |
claw | -0.568 |
hess | -0.515 |
=~ | -0.507 |
Token | , |
Token ID | 11 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
Noise | +1.897 |
Faculty | +1.742 |
grad | +1.719 |
Madness | +1.671 |
graduate | +1.569 |
Neg logit contributions | |
ratulations | -2.316 |
chart | -2.214 |
claw | -2.201 |
hess | -2.015 |
=~ | -1.996 |
Token | 6 |
Token ID | 21 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
Noise | +2.322 |
Faculty | +2.105 |
grad | +2.060 |
Madness | +2.011 |
Amend | +1.859 |
Neg logit contributions | |
ratulations | -3.219 |
claw | -3.049 |
chart | -3.041 |
hess | -2.791 |
=~ | -2.751 |
Token | } |
Token ID | 92 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
Noise | +1.347 |
Faculty | +1.234 |
grad | +1.221 |
Madness | +1.183 |
graduate | +1.111 |
Neg logit contributions | |
ratulations | -1.657 |
chart | -1.579 |
claw | -1.574 |
hess | -1.441 |
=~ | -1.422 |
Token | has |
Token ID | 468 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
Noise | +0.386 |
Faculty | +0.356 |
grad | +0.353 |
Madness | +0.344 |
graduate | +0.323 |
Neg logit contributions | |
ratulations | -0.401 |
chart | -0.379 |
claw | -0.379 |
hess | -0.344 |
=~ | -0.335 |
Token | valid |
Token ID | 4938 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
ratulations | +1.471 |
chart | +1.406 |
claw | +1.395 |
hess | +1.276 |
=~ | +1.268 |
Neg logit contributions | |
Noise | -1.374 |
Faculty | -1.272 |
grad | -1.253 |
Madness | -1.229 |
graduate | -1.155 |
Token | N |
Token ID | 399 |
Feature activation | -4.74e-01 |
Pos logit contributions | |
Noise | +0.110 |
Faculty | +0.102 |
grad | +0.101 |
Madness | +0.098 |
graduate | +0.092 |
Neg logit contributions | |
ratulations | -0.122 |
chart | -0.117 |
claw | -0.116 |
hess | -0.105 |
=~ | -0.105 |
Token | Ns |
Token ID | 47503 |
Feature activation | +5.46e-01 |
Pos logit contributions | |
Noise | +0.397 |
Faculty | +0.371 |
grad | +0.364 |
Madness | +0.364 |
graduate | +0.342 |
Neg logit contributions | |
ratulations | -0.331 |
chart | -0.311 |
claw | -0.310 |
hess | -0.279 |
=~ | -0.278 |
Token | at |
Token ID | 379 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
ratulations | +0.465 |
chart | +0.444 |
claw | +0.443 |
hess | +0.404 |
=~ | +0.401 |
Neg logit contributions | |
Noise | -0.376 |
Faculty | -0.345 |
grad | -0.341 |
Madness | -0.332 |
graduate | -0.310 |
Token | { |
Token ID | 1391 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ratulations | +0.035 |
chart | +0.033 |
claw | +0.033 |
hess | +0.030 |
=~ | +0.030 |
Neg logit contributions | |
Noise | -0.031 |
Faculty | -0.029 |
grad | -0.029 |
Madness | -0.028 |
graduate | -0.026 |
Token | ek |
Token ID | 988 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
ratulations | +3.887 |
claw | +3.815 |
chart | +3.784 |
hess | +3.511 |
=~ | +3.410 |
Neg logit contributions | |
Noise | -2.685 |
Faculty | -2.452 |
grad | -2.413 |
Madness | -2.349 |
graduate | -2.171 |
Token | record |
Token ID | 1700 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
Noise | +2.392 |
Faculty | +2.153 |
grad | +2.135 |
Madness | +2.113 |
graduate | +1.983 |
Neg logit contributions | |
ratulations | -2.677 |
chart | -2.497 |
claw | -2.496 |
=~ | -2.279 |
hess | -2.269 |
Token | is |
Token ID | 318 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
ratulations | +3.272 |
chart | +3.161 |
claw | +3.135 |
hess | +2.851 |
=~ | +2.830 |
Neg logit contributions | |
Noise | -3.053 |
Faculty | -2.844 |
grad | -2.782 |
Madness | -2.717 |
graduate | -2.553 |
Token | next |
Token ID | 1306 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
ratulations | +0.352 |
chart | +0.336 |
claw | +0.334 |
hess | +0.305 |
=~ | +0.303 |
Neg logit contributions | |
Noise | -0.312 |
Faculty | -0.286 |
grad | -0.282 |
Madness | -0.277 |
graduate | -0.259 |
Token | . |
Token ID | 13 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
ratulations | +2.579 |
chart | +2.479 |
claw | +2.449 |
hess | +2.212 |
=~ | +2.194 |
Neg logit contributions | |
Noise | -2.499 |
Faculty | -2.326 |
grad | -2.273 |
Madness | -2.243 |
graduate | -2.094 |
Token | On |
Token ID | 1550 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
ratulations | +0.710 |
chart | +0.681 |
claw | +0.677 |
hess | +0.619 |
=~ | +0.614 |
Neg logit contributions | |
Noise | -0.600 |
Faculty | -0.553 |
grad | -0.546 |
Madness | -0.532 |
graduate | -0.499 |
Token | the |
Token ID | 262 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
ratulations | +0.983 |
chart | +0.939 |
claw | +0.933 |
hess | +0.852 |
=~ | +0.844 |
Neg logit contributions | |
Noise | -0.886 |
Faculty | -0.816 |
grad | -0.806 |
Madness | -0.789 |
graduate | -0.740 |
Token | cover |
Token ID | 3002 |
Feature activation | +6.76e-03 |
Pos logit contributions | |
Noise | +1.443 |
Faculty | +1.313 |
grad | +1.301 |
Madness | +1.279 |
graduate | +1.200 |
Neg logit contributions | |
ratulations | -1.630 |
claw | -1.535 |
chart | -1.532 |
hess | -1.417 |
=~ | -1.384 |
Token | there |
Token ID | 612 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
ratulations | +0.006 |
chart | +0.005 |
claw | +0.005 |
hess | +0.005 |
=~ | +0.005 |
Neg logit contributions | |
Noise | -0.005 |
Faculty | -0.005 |
grad | -0.004 |
Madness | -0.004 |
graduate | -0.004 |
Token | � |
Token ID | 447 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ratulations | +0.616 |
chart | +0.585 |
claw | +0.581 |
hess | +0.528 |
=~ | +0.523 |
Neg logit contributions | |
Noise | -0.610 |
Faculty | -0.563 |
grad | -0.558 |
Madness | -0.547 |
graduate | -0.515 |
Token | � |
Token ID | 247 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Noise | +1.502 |
grad | +1.451 |
Faculty | +1.419 |
Madness | +1.381 |
graduate | +1.312 |
Neg logit contributions | |
ratulations | -2.147 |
chart | -2.015 |
claw | -1.965 |
hess | -1.822 |
=~ | -1.790 |
Token | is |
Token ID | 318 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
ratulations | +0.863 |
chart | +0.825 |
claw | +0.820 |
hess | +0.747 |
=~ | +0.741 |
Neg logit contributions | |
Noise | -0.794 |
Faculty | -0.733 |
grad | -0.724 |
Madness | -0.708 |
graduate | -0.665 |
Token | programmed |
Token ID | 27402 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Noise | +1.225 |
Faculty | +1.119 |
grad | +1.112 |
Madness | +1.080 |
graduate | +1.018 |
Neg logit contributions | |
ratulations | -1.487 |
chart | -1.412 |
claw | -1.404 |
hess | -1.295 |
=~ | -1.288 |
Token | to |
Token ID | 284 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Noise | +0.772 |
Faculty | +0.707 |
grad | +0.698 |
Madness | +0.681 |
graduate | +0.636 |
Neg logit contributions | |
ratulations | -0.987 |
chart | -0.944 |
claw | -0.939 |
hess | -0.863 |
=~ | -0.856 |
Token | loudly |
Token ID | 23112 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ratulations | +0.917 |
chart | +0.879 |
claw | +0.873 |
hess | +0.799 |
=~ | +0.792 |
Neg logit contributions | |
Noise | -0.774 |
Faculty | -0.713 |
grad | -0.702 |
Madness | -0.687 |
graduate | -0.642 |
Token | fear |
Token ID | 3252 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
Noise | +1.083 |
grad | +0.980 |
Faculty | +0.974 |
Madness | +0.944 |
graduate | +0.893 |
Neg logit contributions | |
ratulations | -1.538 |
chart | -1.475 |
claw | -1.468 |
hess | -1.357 |
=~ | -1.352 |
Token | for |
Token ID | 329 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
ratulations | +1.747 |
claw | +1.678 |
chart | +1.674 |
hess | +1.519 |
=~ | +1.501 |
Neg logit contributions | |
Noise | -1.499 |
Faculty | -1.391 |
grad | -1.351 |
Madness | -1.332 |
Amend | -1.239 |
Token | her |
Token ID | 607 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
Noise | +1.727 |
grad | +1.577 |
Faculty | +1.574 |
Madness | +1.517 |
graduate | +1.432 |
Neg logit contributions | |
ratulations | -2.233 |
chart | -2.147 |
claw | -2.123 |
hess | -1.963 |
=~ | -1.944 |
Token | son |
Token ID | 3367 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
Noise | +1.746 |
grad | +1.614 |
Faculty | +1.590 |
Madness | +1.536 |
graduate | +1.468 |
Neg logit contributions | |
ratulations | -2.295 |
chart | -2.202 |
claw | -2.174 |
hess | -2.016 |
=~ | -1.999 |
Token | � |
Token ID | 447 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
Noise | +1.232 |
Faculty | +1.126 |
grad | +1.117 |
Madness | +1.088 |
graduate | +1.020 |
Neg logit contributions | |
ratulations | -1.518 |
chart | -1.450 |
claw | -1.436 |
hess | -1.318 |
=~ | -1.312 |
Token | � |
Token ID | 247 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
ratulations | +0.589 |
chart | +0.561 |
claw | +0.551 |
hess | +0.508 |
=~ | +0.495 |
Neg logit contributions | |
Noise | -0.572 |
Faculty | -0.531 |
grad | -0.530 |
Madness | -0.515 |
graduate | -0.485 |
Token | s |
Token ID | 82 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
Noise | +0.377 |
Faculty | +0.347 |
grad | +0.343 |
Madness | +0.335 |
graduate | +0.314 |
Neg logit contributions | |
ratulations | -0.438 |
chart | -0.419 |
claw | -0.415 |
hess | -0.381 |
=~ | -0.377 |
Token | of |
Token ID | 286 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ratulations | +1.231 |
chart | +1.196 |
claw | +1.173 |
hess | +1.074 |
=~ | +1.063 |
Neg logit contributions | |
Noise | -1.048 |
Faculty | -0.950 |
grad | -0.945 |
Madness | -0.925 |
graduate | -0.859 |
Token | the |
Token ID | 262 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
Noise | +0.802 |
Faculty | +0.741 |
grad | +0.727 |
Madness | +0.710 |
graduate | +0.668 |
Neg logit contributions | |
ratulations | -0.937 |
chart | -0.886 |
claw | -0.883 |
hess | -0.808 |
=~ | -0.798 |
Token | Rockefeller |
Token ID | 32764 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
Noise | +1.156 |
Faculty | +1.071 |
grad | +1.046 |
Madness | +1.022 |
graduate | +0.963 |
Neg logit contributions | |
ratulations | -1.358 |
claw | -1.280 |
chart | -1.280 |
hess | -1.174 |
=~ | -1.159 |
Token | Family |
Token ID | 7884 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
ratulations | +1.576 |
chart | +1.516 |
claw | +1.506 |
hess | +1.364 |
=~ | +1.353 |
Neg logit contributions | |
Noise | -1.501 |
Faculty | -1.375 |
grad | -1.364 |
Madness | -1.335 |
graduate | -1.250 |
Token | Fund |
Token ID | 7557 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
ratulations | +1.570 |
chart | +1.528 |
claw | +1.508 |
hess | +1.380 |
=~ | +1.366 |
Neg logit contributions | |
Noise | -1.371 |
Faculty | -1.251 |
grad | -1.240 |
Madness | -1.212 |
graduate | -1.131 |
Token | 's |
Token ID | 338 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
Noise | +0.167 |
Faculty | +0.154 |
grad | +0.152 |
Madness | +0.148 |
graduate | +0.139 |
Neg logit contributions | |
ratulations | -0.207 |
chart | -0.198 |
claw | -0.197 |
hess | -0.181 |
=~ | -0.179 |
Token | Environmental |
Token ID | 13272 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
Noise | +1.603 |
Faculty | +1.489 |
grad | +1.454 |
Madness | +1.422 |
graduate | +1.344 |
Neg logit contributions | |
ratulations | -1.958 |
claw | -1.839 |
chart | -1.835 |
hess | -1.696 |
=~ | -1.671 |
Token | Grant |
Token ID | 12181 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
Noise | +1.453 |
Faculty | +1.332 |
grad | +1.287 |
Madness | +1.264 |
graduate | +1.161 |
Neg logit contributions | |
ratulations | -2.217 |
chart | -2.092 |
claw | -2.087 |
hess | -1.957 |
=~ | -1.943 |
Token | makers |
Token ID | 6620 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
Noise | +0.446 |
Faculty | +0.411 |
grad | +0.402 |
Madness | +0.395 |
graduate | +0.366 |
Neg logit contributions | |
ratulations | -0.582 |
chart | -0.553 |
claw | -0.551 |
hess | -0.508 |
=~ | -0.504 |
Token | Association |
Token ID | 5396 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
ratulations | +1.185 |
chart | +1.139 |
claw | +1.131 |
hess | +1.036 |
=~ | +1.029 |
Neg logit contributions | |
Noise | -0.938 |
Faculty | -0.860 |
grad | -0.849 |
Madness | -0.827 |
graduate | -0.772 |
Token | and |
Token ID | 290 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
ratulations | +0.996 |
chart | +0.958 |
claw | +0.950 |
hess | +0.870 |
=~ | +0.863 |
Neg logit contributions | |
Noise | -0.823 |
Faculty | -0.755 |
grad | -0.745 |
Madness | -0.728 |
graduate | -0.680 |
Token | most |
Token ID | 749 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
ratulations | +0.874 |
chart | +0.835 |
claw | +0.830 |
hess | +0.758 |
=~ | +0.751 |
Neg logit contributions | |
Noise | -0.784 |
Faculty | -0.724 |
grad | -0.714 |
Madness | -0.698 |
graduate | -0.655 |
Token | current |
Token ID | 1459 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ratulations | +2.379 |
chart | +2.287 |
claw | +2.272 |
hess | +2.090 |
=~ | +2.071 |
Neg logit contributions | |
Noise | -1.757 |
Faculty | -1.608 |
grad | -1.580 |
Madness | -1.541 |
graduate | -1.432 |
Token | estimates |
Token ID | 7746 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
ratulations | +2.242 |
chart | +2.164 |
claw | +2.144 |
hess | +1.961 |
=~ | +1.944 |
Neg logit contributions | |
Noise | -1.787 |
Faculty | -1.642 |
grad | -1.613 |
Madness | -1.577 |
graduate | -1.468 |
Token | place |
Token ID | 1295 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
Noise | +0.682 |
Faculty | +0.627 |
grad | +0.623 |
Madness | +0.606 |
graduate | +0.571 |
Neg logit contributions | |
ratulations | -0.865 |
claw | -0.813 |
chart | -0.808 |
hess | -0.752 |
=~ | -0.741 |
Token | 34 |
Token ID | 4974 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
ratulations | +0.377 |
chart | +0.359 |
claw | +0.356 |
hess | +0.328 |
=~ | +0.323 |
Neg logit contributions | |
Noise | -0.328 |
Faculty | -0.301 |
grad | -0.299 |
Madness | -0.291 |
graduate | -0.274 |
Token | , |
Token ID | 11 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
ratulations | +2.006 |
chart | +1.936 |
claw | +1.926 |
hess | +1.750 |
=~ | +1.741 |
Neg logit contributions | |
Noise | -1.724 |
Faculty | -1.591 |
grad | -1.558 |
Madness | -1.532 |
graduate | -1.423 |
Token | 776 |
Token ID | 39509 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
Noise | +0.464 |
Faculty | +0.425 |
grad | +0.418 |
Madness | +0.408 |
graduate | +0.380 |
Neg logit contributions | |
ratulations | -0.605 |
chart | -0.581 |
claw | -0.578 |
hess | -0.530 |
=~ | -0.526 |
Token | people |
Token ID | 661 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
ratulations | +1.091 |
chart | +1.044 |
claw | +1.038 |
hess | +0.945 |
=~ | +0.937 |
Neg logit contributions | |
Noise | -1.011 |
Faculty | -0.936 |
grad | -0.922 |
Madness | -0.903 |
graduate | -0.846 |
Token | in |
Token ID | 287 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ratulations | +0.305 |
chart | +0.291 |
claw | +0.288 |
hess | +0.266 |
=~ | +0.264 |
Neg logit contributions | |
Noise | -0.236 |
Faculty | -0.216 |
grad | -0.214 |
Madness | -0.207 |
graduate | -0.195 |
Token | the |
Token ID | 262 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
ratulations | +1.015 |
chart | +0.965 |
claw | +0.956 |
hess | +0.872 |
=~ | +0.863 |
Neg logit contributions | |
Noise | -1.002 |
Faculty | -0.928 |
grad | -0.918 |
Madness | -0.897 |
graduate | -0.847 |
Token | shelter |
Token ID | 11772 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
ratulations | +0.614 |
chart | +0.580 |
claw | +0.573 |
hess | +0.524 |
=~ | +0.518 |
Neg logit contributions | |
Noise | -0.638 |
Faculty | -0.592 |
grad | -0.587 |
Madness | -0.572 |
graduate | -0.543 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
ratulations | +1.314 |
claw | +1.253 |
chart | +1.252 |
hess | +1.129 |
=~ | +1.117 |
Neg logit contributions | |
Noise | -1.344 |
Faculty | -1.252 |
grad | -1.226 |
Madness | -1.209 |
graduate | -1.131 |
Token | , |
Token ID | 11 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ratulations | +1.056 |
chart | +1.010 |
claw | +1.008 |
hess | +0.914 |
=~ | +0.904 |
Neg logit contributions | |
Noise | -0.974 |
Faculty | -0.904 |
grad | -0.884 |
Madness | -0.875 |
graduate | -0.811 |
Token | he |
Token ID | 339 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
Noise | +1.158 |
Faculty | +1.072 |
grad | +1.065 |
Madness | +1.035 |
graduate | +0.993 |
Neg logit contributions | |
ratulations | -0.959 |
chart | -0.901 |
claw | -0.895 |
hess | -0.811 |
=~ | -0.799 |
Token | has |
Token ID | 468 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Noise | +0.258 |
Faculty | +0.237 |
grad | +0.237 |
Madness | +0.228 |
graduate | +0.216 |
Neg logit contributions | |
ratulations | -0.315 |
chart | -0.299 |
claw | -0.295 |
hess | -0.272 |
=~ | -0.271 |
Token | a |
Token ID | 257 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
Noise | +0.825 |
Faculty | +0.748 |
grad | +0.746 |
Madness | +0.717 |
graduate | +0.683 |
Neg logit contributions | |
ratulations | -1.059 |
chart | -1.008 |
claw | -0.998 |
hess | -0.919 |
=~ | -0.913 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
Noise | +1.063 |
Faculty | +0.981 |
grad | +0.981 |
Madness | +0.936 |
graduate | +0.902 |
Neg logit contributions | |
ratulations | -1.256 |
chart | -1.178 |
claw | -1.161 |
hess | -1.076 |
=~ | -1.072 |
Token | , |
Token ID | 11 |
Feature activation | +7.60e-01 |
Pos logit contributions | |
ratulations | +0.015 |
chart | +0.015 |
claw | +0.014 |
hess | +0.013 |
=~ | +0.013 |
Neg logit contributions | |
Noise | -0.013 |
Faculty | -0.011 |
grad | -0.011 |
Madness | -0.011 |
graduate | -0.010 |
Token | Angel |
Token ID | 3905 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
ratulations | +0.632 |
chart | +0.605 |
claw | +0.601 |
hess | +0.550 |
=~ | +0.545 |
Neg logit contributions | |
Noise | -0.539 |
Faculty | -0.496 |
grad | -0.489 |
Madness | -0.477 |
graduate | -0.447 |
Token | , |
Token ID | 11 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
ratulations | +0.783 |
chart | +0.748 |
claw | +0.744 |
hess | +0.680 |
=~ | +0.675 |
Neg logit contributions | |
Noise | -0.671 |
Faculty | -0.618 |
grad | -0.609 |
Madness | -0.595 |
graduate | -0.558 |
Token | that |
Token ID | 326 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
ratulations | +2.253 |
chart | +2.159 |
claw | +2.157 |
hess | +1.960 |
=~ | +1.945 |
Neg logit contributions | |
Noise | -2.024 |
Faculty | -1.874 |
grad | -1.843 |
Madness | -1.805 |
graduate | -1.689 |
Token | might |
Token ID | 1244 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ratulations | +1.707 |
chart | +1.623 |
claw | +1.608 |
hess | +1.459 |
=~ | +1.443 |
Neg logit contributions | |
Noise | -1.775 |
Faculty | -1.643 |
grad | -1.627 |
Madness | -1.591 |
graduate | -1.504 |
Token | ation |
Token ID | 341 |
Feature activation | -8.07e-01 |
Pos logit contributions | |
Noise | +0.845 |
Faculty | +0.794 |
grad | +0.784 |
Madness | +0.766 |
graduate | +0.735 |
Neg logit contributions | |
ratulations | -0.592 |
claw | -0.557 |
chart | -0.557 |
=~ | -0.493 |
hess | -0.485 |
Token | Facility |
Token ID | 29118 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
Noise | +0.550 |
Faculty | +0.510 |
grad | +0.501 |
Madness | +0.488 |
graduate | +0.459 |
Neg logit contributions | |
ratulations | -0.698 |
chart | -0.666 |
claw | -0.661 |
hess | -0.609 |
=~ | -0.606 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Noise | +1.157 |
Faculty | +1.076 |
grad | +1.057 |
Madness | +1.029 |
graduate | +0.973 |
Neg logit contributions | |
ratulations | -1.368 |
chart | -1.300 |
claw | -1.296 |
hess | -1.185 |
=~ | -1.178 |
Token | Find |
Token ID | 9938 |
Feature activation | -4.93e+00 |
Pos logit contributions | |
Noise | +0.097 |
Faculty | +0.089 |
grad | +0.088 |
Madness | +0.086 |
graduate | +0.080 |
Neg logit contributions | |
ratulations | -0.124 |
chart | -0.119 |
claw | -0.119 |
hess | -0.109 |
=~ | -0.108 |
Token | out |
Token ID | 503 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
Noise | +3.097 |
Faculty | +2.902 |
grad | +2.848 |
Madness | +2.693 |
graduate | +2.602 |
Neg logit contributions | |
ratulations | -4.471 |
chart | -4.260 |
claw | -4.238 |
=~ | -3.955 |
hess | -3.922 |
Token | more |
Token ID | 517 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
Noise | +0.529 |
Faculty | +0.491 |
grad | +0.484 |
Madness | +0.470 |
graduate | +0.445 |
Neg logit contributions | |
ratulations | -0.599 |
chart | -0.571 |
claw | -0.568 |
hess | -0.520 |
=~ | -0.516 |
Token | at |
Token ID | 379 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
Noise | +1.400 |
Faculty | +1.308 |
grad | +1.287 |
Madness | +1.247 |
graduate | +1.188 |
Neg logit contributions | |
ratulations | -1.564 |
chart | -1.485 |
claw | -1.480 |
hess | -1.361 |
=~ | -1.349 |
Token | : |
Token ID | 25 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
Noise | +2.220 |
Faculty | +2.071 |
grad | +2.054 |
Madness | +1.942 |
graduate | +1.863 |
Neg logit contributions | |
ratulations | -3.205 |
claw | -3.059 |
chart | -3.053 |
hess | -2.806 |
=~ | -2.802 |
Token | http |
Token ID | 2638 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ratulations | +0.939 |
chart | +0.905 |
claw | +0.899 |
hess | +0.833 |
=~ | +0.825 |
Neg logit contributions | |
Noise | -0.597 |
Faculty | -0.538 |
grad | -0.527 |
Madness | -0.517 |
graduate | -0.473 |
Token | :// |
Token ID | 1378 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
ratulations | +1.443 |
chart | +1.348 |
claw | +1.339 |
hess | +1.238 |
=~ | +1.235 |
Neg logit contributions | |
Noise | -0.929 |
Faculty | -0.857 |
grad | -0.844 |
Madness | -0.814 |
graduate | -0.759 |
Token | www |
Token ID | 2503 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
Noise | +2.399 |
Faculty | +2.265 |
grad | +2.243 |
Madness | +2.138 |
graduate | +2.066 |
Neg logit contributions | |
ratulations | -2.729 |
claw | -2.571 |
chart | -2.557 |
=~ | -2.358 |
hess | -2.320 |
Token | x |
Token ID | 2124 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Noise | +0.093 |
Faculty | +0.085 |
grad | +0.085 |
Madness | +0.082 |
graduate | +0.077 |
Neg logit contributions | |
ratulations | -0.122 |
chart | -0.113 |
claw | -0.113 |
hess | -0.104 |
=~ | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
Noise | +0.758 |
Faculty | +0.698 |
grad | +0.692 |
Madness | +0.674 |
graduate | +0.631 |
Neg logit contributions | |
ratulations | -1.015 |
chart | -0.903 |
claw | -0.898 |
hess | -0.834 |
=~ | -0.826 |
Token | x |
Token ID | 87 |
Feature activation | -3.96e+00 |
Pos logit contributions | |
Noise | +1.239 |
Faculty | +1.077 |
grad | +1.066 |
Madness | +0.929 |
yg | +0.926 |
Neg logit contributions | |
ratulations | -5.049 |
chart | -4.533 |
claw | -4.533 |
hess | -4.338 |
cuff | -4.324 |
Token | . |
Token ID | 13 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
Noise | +2.452 |
Faculty | +2.254 |
grad | +2.225 |
Madness | +2.173 |
yg | +2.024 |
Neg logit contributions | |
ratulations | -3.802 |
chart | -3.294 |
claw | -3.269 |
hess | -3.068 |
=~ | -3.064 |
Token | x |
Token ID | 87 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
Noise | +1.411 |
Faculty | +1.210 |
grad | +1.203 |
yg | +1.094 |
Madness | +1.046 |
Neg logit contributions | |
ratulations | -5.869 |
chart | -5.299 |
claw | -5.293 |
cuff | -5.081 |
hess | -5.058 |
Token | . |
Token ID | 13 |
Feature activation | -6.40e+00 |
Pos logit contributions | |
Noise | +3.249 |
Faculty | +3.059 |
grad | +3.036 |
Madness | +2.945 |
graduate | +2.773 |
Neg logit contributions | |
ratulations | -5.022 |
chart | -4.411 |
claw | -4.368 |
hess | -4.142 |
akeru | -4.088 |
Token | 175 |
Token ID | 17430 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
Noise | +2.569 |
Faculty | +2.476 |
grad | +2.463 |
Madness | +2.231 |
graduate | +2.132 |
Neg logit contributions | |
ratulations | -7.018 |
chart | -6.539 |
claw | -6.464 |
hess | -6.179 |
enne | -6.108 |
Token | 99 |
Token ID | 7388 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
Noise | +2.554 |
Faculty | +2.309 |
grad | +2.231 |
Madness | +2.200 |
Amend | +1.985 |
Neg logit contributions | |
ratulations | -4.242 |
chart | -3.939 |
claw | -3.899 |
hess | -3.620 |
=~ | -3.583 |
Token | 33 |
Token ID | 2091 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
Noise | +0.621 |
Faculty | +0.575 |
grad | +0.561 |
Madness | +0.551 |
graduate | +0.513 |
Neg logit contributions | |
ratulations | -0.777 |
chart | -0.729 |
claw | -0.723 |
hess | -0.667 |
=~ | -0.658 |
Token | 87 |
Token ID | 5774 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
ratulations | +1.738 |
chart | +1.683 |
claw | +1.673 |
hess | +1.520 |
=~ | +1.509 |
Neg logit contributions | |
Noise | -1.603 |
Faculty | -1.476 |
grad | -1.462 |
Madness | -1.425 |
graduate | -1.340 |
Token | x |
Token ID | 2124 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
ratulations | +2.024 |
chart | +1.931 |
claw | +1.919 |
hess | +1.753 |
=~ | +1.737 |
Neg logit contributions | |
Noise | -1.796 |
Faculty | -1.656 |
grad | -1.635 |
Madness | -1.597 |
graduate | -1.499 |
Token | lasted |
Token ID | 15436 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
Noise | +0.681 |
Faculty | +0.632 |
grad | +0.626 |
Madness | +0.607 |
graduate | +0.577 |
Neg logit contributions | |
ratulations | -0.790 |
chart | -0.745 |
claw | -0.742 |
hess | -0.680 |
=~ | -0.677 |
Token | for |
Token ID | 329 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
ratulations | +0.541 |
chart | +0.514 |
claw | +0.512 |
hess | +0.467 |
=~ | +0.463 |
Neg logit contributions | |
Noise | -0.490 |
Faculty | -0.453 |
grad | -0.449 |
Madness | -0.439 |
graduate | -0.412 |
Token | 25 |
Token ID | 1679 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
Noise | +0.554 |
Faculty | +0.512 |
grad | +0.510 |
Madness | +0.498 |
graduate | +0.470 |
Neg logit contributions | |
ratulations | -0.606 |
chart | -0.576 |
claw | -0.572 |
hess | -0.523 |
=~ | -0.514 |
Token | days |
Token ID | 1528 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
ratulations | +0.752 |
chart | +0.716 |
claw | +0.711 |
hess | +0.645 |
=~ | +0.638 |
Neg logit contributions | |
Noise | -0.759 |
Faculty | -0.704 |
grad | -0.696 |
Madness | -0.681 |
graduate | -0.642 |
Token | .[ |
Token ID | 3693 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
Noise | +0.556 |
Faculty | +0.512 |
grad | +0.508 |
Madness | +0.495 |
graduate | +0.464 |
Neg logit contributions | |
ratulations | -0.660 |
chart | -0.625 |
claw | -0.622 |
hess | -0.571 |
=~ | -0.565 |
Token | 1 |
Token ID | 16 |
Feature activation | -6.55e+00 |
Pos logit contributions | |
Noise | +2.240 |
Faculty | +2.078 |
grad | +2.043 |
Madness | +2.007 |
graduate | +1.884 |
Neg logit contributions | |
ratulations | -2.442 |
claw | -2.274 |
chart | -2.271 |
hess | -2.082 |
=~ | -2.059 |
Token | ] |
Token ID | 60 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
Noise | +4.836 |
Faculty | +4.471 |
grad | +4.409 |
Madness | +4.302 |
graduate | +4.069 |
Neg logit contributions | |
ratulations | -5.176 |
chart | -4.872 |
claw | -4.834 |
hess | -4.398 |
=~ | -4.366 |
Token |
|
Token ID | 198 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ratulations | +0.521 |
chart | +0.504 |
claw | +0.500 |
hess | +0.457 |
=~ | +0.455 |
Neg logit contributions | |
Noise | -0.413 |
Faculty | -0.378 |
grad | -0.375 |
Madness | -0.362 |
graduate | -0.341 |
Token |
|
Token ID | 198 |
Feature activation | +7.21e-01 |
Pos logit contributions | |
ratulations | +1.266 |
chart | +1.188 |
claw | +1.179 |
hess | +1.095 |
=~ | +1.090 |
Neg logit contributions | |
Noise | -0.839 |
Faculty | -0.769 |
grad | -0.755 |
Madness | -0.737 |
graduate | -0.683 |
Token | Exec |
Token ID | 23002 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
ratulations | +0.583 |
chart | +0.556 |
claw | +0.553 |
hess | +0.505 |
=~ | +0.500 |
Neg logit contributions | |
Noise | -0.526 |
Faculty | -0.486 |
grad | -0.480 |
Madness | -0.469 |
graduate | -0.440 |
Token | ut |
Token ID | 315 |
Feature activation | +8.78e-01 |
Pos logit contributions | |
Noise | +0.391 |
Faculty | +0.353 |
grad | +0.347 |
Madness | +0.337 |
graduate | +0.310 |
Neg logit contributions | |
ratulations | -0.651 |
chart | -0.627 |
claw | -0.623 |
hess | -0.578 |
=~ | -0.574 |
Token | en |
Token ID | 268 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ratulations | +0.435 |
chart | +0.406 |
claw | +0.403 |
hess | +0.361 |
=~ | +0.360 |
Neg logit contributions | |
Noise | -0.514 |
Faculty | -0.482 |
grad | -0.478 |
Madness | -0.467 |
graduate | -0.446 |
Token | was |
Token ID | 373 |
Feature activation | +9.47e-01 |
Pos logit contributions | |
Noise | +1.041 |
Faculty | +0.962 |
grad | +0.956 |
Madness | +0.928 |
graduate | +0.877 |
Neg logit contributions | |
ratulations | -1.265 |
chart | -1.205 |
claw | -1.188 |
hess | -1.099 |
=~ | -1.088 |
Token | doing |
Token ID | 1804 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
ratulations | +0.765 |
chart | +0.731 |
claw | +0.726 |
hess | +0.663 |
=~ | +0.657 |
Neg logit contributions | |
Noise | -0.691 |
Faculty | -0.639 |
grad | -0.630 |
Madness | -0.616 |
graduate | -0.579 |
Token | what |
Token ID | 644 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
ratulations | +0.281 |
chart | +0.268 |
claw | +0.266 |
hess | +0.243 |
=~ | +0.241 |
Neg logit contributions | |
Noise | -0.261 |
Faculty | -0.241 |
grad | -0.239 |
Madness | -0.233 |
graduate | -0.220 |
Token | Coach |
Token ID | 16393 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
Noise | +0.284 |
Faculty | +0.265 |
grad | +0.263 |
Madness | +0.257 |
graduate | +0.245 |
Neg logit contributions | |
ratulations | -0.229 |
chart | -0.217 |
claw | -0.215 |
hess | -0.194 |
=~ | -0.191 |
Token | [ |
Token ID | 685 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
Noise | +1.431 |
Faculty | +1.324 |
grad | +1.309 |
Madness | +1.271 |
graduate | +1.203 |
Neg logit contributions | |
ratulations | -1.628 |
chart | -1.549 |
claw | -1.541 |
hess | -1.419 |
=~ | -1.400 |
Token | Doug |
Token ID | 42297 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
Noise | +3.550 |
Faculty | +3.283 |
grad | +3.246 |
Madness | +3.152 |
graduate | +2.952 |
Neg logit contributions | |
ratulations | -4.598 |
chart | -4.255 |
claw | -4.229 |
hess | -3.977 |
=~ | -3.930 |
Token | ] |
Token ID | 60 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
Noise | +2.355 |
Faculty | +2.162 |
grad | +2.116 |
Madness | +2.065 |
graduate | +1.934 |
Neg logit contributions | |
ratulations | -3.228 |
chart | -3.068 |
claw | -3.054 |
=~ | -2.842 |
hess | -2.828 |
Token | Mar |
Token ID | 1526 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
ratulations | +1.086 |
chart | +1.052 |
claw | +1.035 |
hess | +0.946 |
=~ | +0.928 |
Neg logit contributions | |
Noise | -0.883 |
Faculty | -0.805 |
grad | -0.802 |
Madness | -0.777 |
alum | -0.724 |
Token | rone |
Token ID | 33171 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
Noise | +1.640 |
Faculty | +1.549 |
grad | +1.523 |
Madness | +1.465 |
graduate | +1.394 |
Neg logit contributions | |
ratulations | -2.280 |
claw | -2.181 |
chart | -2.179 |
=~ | -2.056 |
hess | -2.009 |
Token | has |
Token ID | 468 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
Noise | +1.061 |
Faculty | +0.980 |
grad | +0.977 |
Madness | +0.943 |
graduate | +0.894 |
Neg logit contributions | |
ratulations | -1.324 |
chart | -1.262 |
claw | -1.253 |
hess | -1.162 |
=~ | -1.154 |
Token | : |
Token ID | 25 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
Noise | +0.251 |
Faculty | +0.230 |
grad | +0.227 |
Madness | +0.222 |
graduate | +0.207 |
Neg logit contributions | |
ratulations | -0.320 |
chart | -0.305 |
claw | -0.304 |
hess | -0.279 |
=~ | -0.277 |
Token |
|
Token ID | 198 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
ratulations | +1.676 |
chart | +1.621 |
claw | +1.592 |
hess | +1.449 |
=~ | +1.434 |
Neg logit contributions | |
Noise | -1.473 |
Faculty | -1.362 |
grad | -1.345 |
Madness | -1.320 |
graduate | -1.233 |
Token |
|
Token ID | 198 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
Noise | +0.359 |
Faculty | +0.329 |
grad | +0.323 |
Madness | +0.314 |
graduate | +0.291 |
Neg logit contributions | |
ratulations | -0.619 |
chart | -0.567 |
claw | -0.563 |
hess | -0.529 |
=~ | -0.528 |
Token | During |
Token ID | 7191 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
Noise | +1.685 |
Faculty | +1.565 |
grad | +1.536 |
Madness | +1.496 |
graduate | +1.405 |
Neg logit contributions | |
ratulations | -1.926 |
claw | -1.790 |
chart | -1.785 |
hess | -1.635 |
=~ | -1.621 |
Token | this |
Token ID | 428 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
Noise | +0.773 |
Faculty | +0.713 |
grad | +0.713 |
Madness | +0.693 |
graduate | +0.656 |
Neg logit contributions | |
ratulations | -0.861 |
chart | -0.818 |
claw | -0.814 |
hess | -0.745 |
=~ | -0.738 |
Token | caps |
Token ID | 11022 |
Feature activation | -5.42e+00 |
Pos logit contributions | |
Noise | +1.134 |
grad | +1.044 |
Faculty | +1.042 |
Madness | +1.012 |
graduate | +0.953 |
Neg logit contributions | |
ratulations | -1.425 |
chart | -1.362 |
claw | -1.348 |
hess | -1.237 |
=~ | -1.231 |
Token | izing |
Token ID | 2890 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
Noise | +3.540 |
grad | +3.241 |
Faculty | +3.222 |
Madness | +3.159 |
graduate | +2.963 |
Neg logit contributions | |
ratulations | -4.680 |
chart | -4.389 |
claw | -4.364 |
=~ | -4.047 |
hess | -3.979 |
Token | , |
Token ID | 11 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
Noise | +2.102 |
grad | +1.936 |
Faculty | +1.925 |
Madness | +1.869 |
graduate | +1.755 |
Neg logit contributions | |
ratulations | -2.785 |
chart | -2.652 |
claw | -2.627 |
hess | -2.413 |
=~ | -2.413 |
Token | two |
Token ID | 734 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ratulations | +0.590 |
chart | +0.565 |
claw | +0.561 |
hess | +0.512 |
=~ | +0.507 |
Neg logit contributions | |
Noise | -0.546 |
Faculty | -0.506 |
grad | -0.495 |
Madness | -0.488 |
graduate | -0.456 |
Token | massive |
Token ID | 4858 |
Feature activation | -4.70e+00 |
Pos logit contributions | |
Noise | +1.523 |
grad | +1.409 |
Faculty | +1.406 |
Madness | +1.354 |
graduate | +1.286 |
Neg logit contributions | |
ratulations | -1.823 |
chart | -1.744 |
claw | -1.732 |
=~ | -1.581 |
hess | -1.581 |
Token | forces |
Token ID | 3386 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
Noise | +3.233 |
grad | +3.068 |
Faculty | +2.987 |
Madness | +2.895 |
graduate | +2.781 |
Neg logit contributions | |
ratulations | -3.952 |
chart | -3.779 |
claw | -3.720 |
=~ | -3.446 |
hess | -3.394 |
Token | le |
Token ID | 293 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
Noise | +0.360 |
Faculty | +0.334 |
grad | +0.328 |
Madness | +0.320 |
graduate | +0.303 |
Neg logit contributions | |
ratulations | -0.408 |
chart | -0.383 |
claw | -0.381 |
=~ | -0.352 |
hess | -0.348 |
Token | k |
Token ID | 74 |
Feature activation | -7.83e-01 |
Pos logit contributions | |
Noise | +1.242 |
Faculty | +1.148 |
grad | +1.120 |
Madness | +1.074 |
Amend | +1.036 |
Neg logit contributions | |
ratulations | -1.521 |
claw | -1.375 |
chart | -1.374 |
hess | -1.271 |
=~ | -1.271 |
Token | ( |
Token ID | 357 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
Noise | +0.428 |
Faculty | +0.384 |
grad | +0.377 |
Madness | +0.366 |
graduate | +0.335 |
Neg logit contributions | |
ratulations | -0.775 |
chart | -0.750 |
claw | -0.745 |
hess | -0.692 |
=~ | -0.686 |
Token | O |
Token ID | 46 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
ratulations | +1.306 |
chart | +1.263 |
claw | +1.252 |
hess | +1.137 |
=~ | +1.130 |
Neg logit contributions | |
Noise | -1.243 |
Faculty | -1.145 |
grad | -1.135 |
Madness | -1.110 |
graduate | -1.041 |
Token | mega |
Token ID | 13731 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
Noise | +1.837 |
Faculty | +1.622 |
grad | +1.621 |
Madness | +1.535 |
graduate | +1.440 |
Neg logit contributions | |
ratulations | -3.876 |
chart | -3.669 |
claw | -3.654 |
=~ | -3.427 |
hess | -3.407 |
Token | Pharma |
Token ID | 41845 |
Feature activation | -6.05e+00 |
Pos logit contributions | |
ratulations | +1.406 |
chart | +1.353 |
claw | +1.343 |
hess | +1.270 |
=~ | +1.259 |
Neg logit contributions | |
Noise | -0.459 |
Faculty | -0.395 |
grad | -0.385 |
Madness | -0.362 |
graduate | -0.320 |
Token | - |
Token ID | 532 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
Noise | +4.111 |
grad | +3.788 |
Faculty | +3.776 |
Madness | +3.693 |
graduate | +3.490 |
Neg logit contributions | |
ratulations | -5.180 |
chart | -4.870 |
claw | -4.836 |
hess | -4.495 |
=~ | -4.475 |
Token | Quick |
Token ID | 12029 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
Noise | +2.395 |
Faculty | +2.181 |
grad | +2.114 |
Madness | +2.104 |
graduate | +1.927 |
Neg logit contributions | |
ratulations | -3.602 |
claw | -3.406 |
chart | -3.384 |
hess | -3.155 |
=~ | -3.116 |
Token | step |
Token ID | 9662 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
Noise | +0.295 |
grad | +0.271 |
Faculty | +0.269 |
Madness | +0.262 |
graduate | +0.245 |
Neg logit contributions | |
ratulations | -0.454 |
chart | -0.424 |
claw | -0.421 |
=~ | -0.397 |
hess | -0.393 |
Token | ) |
Token ID | 8 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
Noise | +0.530 |
Faculty | +0.493 |
grad | +0.483 |
Madness | +0.473 |
graduate | +0.443 |
Neg logit contributions | |
ratulations | -0.690 |
chart | -0.635 |
claw | -0.627 |
hess | -0.583 |
=~ | -0.579 |
Token | s |
Token ID | 264 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ratulations | +0.472 |
chart | +0.446 |
claw | +0.441 |
hess | +0.397 |
=~ | +0.390 |
Neg logit contributions | |
Noise | -0.609 |
Faculty | -0.569 |
grad | -0.559 |
Madness | -0.553 |
graduate | -0.521 |