Token | Tim |
Token ID | 5045 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
riage | +0.361 |
ipples | +0.354 |
ry | +0.346 |
flower | +0.344 |
ac | +0.338 |
Neg logit contributions | |
hy | -0.358 |
Mic | -0.319 |
rides | -0.279 |
Red | -0.276 |
Roger | -0.270 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
hy | +1.502 |
Mic | +1.356 |
rides | +1.176 |
Red | +1.171 |
Roger | +1.169 |
Neg logit contributions | |
riage | -1.194 |
flower | -1.183 |
ipples | -1.183 |
ry | -1.168 |
andy | -1.141 |
Token | to |
Token ID | 284 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
hy | +0.337 |
Mic | +0.306 |
Red | +0.276 |
rides | +0.273 |
Roger | +0.270 |
Neg logit contributions | |
riage | -0.201 |
ipples | -0.196 |
flower | -0.190 |
ry | -0.189 |
department | -0.185 |
Token | play |
Token ID | 711 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
hy | +0.216 |
Mic | +0.197 |
Red | +0.174 |
rides | +0.173 |
Roger | +0.172 |
Neg logit contributions | |
riage | -0.162 |
ipples | -0.158 |
ry | -0.153 |
flower | -0.153 |
andy | -0.150 |
Token | all |
Token ID | 477 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
hy | +2.919 |
Mic | +2.472 |
Robert | +2.178 |
nam | +2.175 |
Roger | +2.136 |
Neg logit contributions | |
ipples | -2.339 |
ry | -2.327 |
flower | -2.267 |
riage | -2.244 |
andy | -2.240 |
Token | day |
Token ID | 1110 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
hy | +0.676 |
Mic | +0.622 |
Red | +0.554 |
rides | +0.553 |
Roger | +0.550 |
Neg logit contributions | |
riage | -0.397 |
ipples | -0.391 |
flower | -0.378 |
ry | -0.378 |
ac | -0.369 |
Token | . |
Token ID | 13 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
hy | +10.413 |
Mic | +9.533 |
Red | +9.028 |
Blue | +8.707 |
Bow | +8.472 |
Neg logit contributions | |
ry | -5.171 |
again | -4.872 |
andy | -4.644 |
ipples | -4.565 |
en | -4.441 |
Token | One |
Token ID | 1881 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
riage | +0.705 |
ipples | +0.691 |
ry | +0.676 |
flower | +0.671 |
ac | +0.662 |
Neg logit contributions | |
hy | -0.677 |
Mic | -0.601 |
rides | -0.526 |
Red | -0.521 |
Roger | -0.506 |
Token | day |
Token ID | 1110 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
riage | +1.331 |
ipples | +1.296 |
department | +1.253 |
ry | +1.250 |
flower | +1.225 |
Neg logit contributions | |
hy | -1.713 |
Mic | -1.517 |
rides | -1.374 |
Red | -1.346 |
Roger | -1.316 |
Token | , |
Token ID | 11 |
Feature activation | +7.90e-01 |
Pos logit contributions | |
hy | +2.547 |
Mic | +2.355 |
Red | +2.095 |
Robert | +2.043 |
rides | +2.041 |
Neg logit contributions | |
ry | -1.784 |
riage | -1.755 |
ipples | -1.685 |
flower | -1.676 |
andy | -1.658 |
Token | he |
Token ID | 339 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
hy | +1.456 |
Mic | +1.280 |
rides | +1.107 |
Red | +1.104 |
Robert | +1.083 |
Neg logit contributions | |
riage | -1.394 |
ipples | -1.361 |
flower | -1.358 |
ry | -1.339 |
department | -1.318 |
Token | I |
Token ID | 314 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
riage | +0.683 |
ipples | +0.680 |
ry | +0.656 |
flower | +0.649 |
andy | +0.641 |
Neg logit contributions | |
hy | -0.847 |
Mic | -0.764 |
rides | -0.678 |
Red | -0.672 |
Roger | -0.663 |
Token | can |
Token ID | 460 |
Feature activation | +3.45e-01 |
Pos logit contributions | |
riage | +1.848 |
ipples | +1.747 |
flower | +1.679 |
department | +1.639 |
andy | +1.636 |
Neg logit contributions | |
hy | -2.352 |
Mic | -2.088 |
rides | -1.911 |
Red | -1.840 |
Roger | -1.804 |
Token | 't |
Token ID | 470 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
hy | +0.543 |
Mic | +0.485 |
rides | +0.407 |
Red | +0.405 |
Roger | +0.398 |
Neg logit contributions | |
riage | -0.687 |
ipples | -0.681 |
ry | -0.669 |
flower | -0.668 |
andy | -0.652 |
Token | get |
Token ID | 651 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
hy | +0.634 |
Mic | +0.571 |
Red | +0.498 |
rides | +0.498 |
Roger | +0.491 |
Neg logit contributions | |
riage | -0.531 |
ipples | -0.525 |
ry | -0.512 |
flower | -0.510 |
ac | -0.492 |
Token | out |
Token ID | 503 |
Feature activation | -8.94e-01 |
Pos logit contributions | |
hy | +0.691 |
Mic | +0.624 |
Red | +0.543 |
rides | +0.541 |
Roger | +0.528 |
Neg logit contributions | |
riage | -0.603 |
ipples | -0.592 |
flower | -0.577 |
ry | -0.576 |
department | -0.563 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
riage | +1.572 |
ipples | +1.538 |
flower | +1.486 |
ry | +1.476 |
andy | +1.474 |
Neg logit contributions | |
hy | -1.645 |
Mic | -1.427 |
rides | -1.300 |
Red | -1.241 |
Roger | -1.233 |
Token | cried |
Token ID | 16896 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
hy | +8.623 |
Mic | +7.437 |
blonde | +6.820 |
hell | +6.803 |
nam | +6.613 |
Neg logit contributions | |
Duck | -6.177 |
again | -6.010 |
Mama | -5.617 |
antly | -5.519 |
department | -5.454 |
Token | Blue |
Token ID | 4518 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
hy | +3.411 |
Mic | +2.962 |
rides | +2.741 |
Robert | +2.666 |
nam | +2.608 |
Neg logit contributions | |
department | -2.786 |
again | -2.743 |
riage | -2.721 |
ipples | -2.618 |
flower | -2.614 |
Token | . |
Token ID | 13 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
riage | +0.247 |
ipples | +0.241 |
ry | +0.235 |
flower | +0.233 |
ac | +0.229 |
Neg logit contributions | |
hy | -0.310 |
Mic | -0.280 |
rides | -0.249 |
Red | -0.246 |
Roger | -0.242 |
Token | He |
Token ID | 679 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
hy | +5.169 |
Mic | +4.473 |
rides | +4.153 |
nam | +4.151 |
hell | +4.102 |
Neg logit contributions | |
department | -2.647 |
riage | -2.552 |
Duck | -2.505 |
again | -2.394 |
flower | -2.359 |
Token | tried |
Token ID | 3088 |
Feature activation | -9.76e-01 |
Pos logit contributions | |
riage | +0.573 |
ipples | +0.557 |
ry | +0.539 |
flower | +0.537 |
ac | +0.526 |
Neg logit contributions | |
hy | -0.688 |
Mic | -0.621 |
rides | -0.553 |
Red | -0.545 |
Roger | -0.535 |
Token | Max |
Token ID | 5436 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
hy | +2.409 |
Mic | +2.050 |
ve | +1.879 |
nam | +1.872 |
rides | +1.802 |
Neg logit contributions | |
riage | -2.686 |
department | -2.595 |
flower | -2.548 |
ipples | -2.543 |
andy | -2.516 |
Token | loved |
Token ID | 6151 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
hy | +4.372 |
Mic | +4.055 |
nam | +3.743 |
Robert | +3.642 |
Roger | +3.627 |
Neg logit contributions | |
flower | -1.802 |
ry | -1.760 |
riage | -1.749 |
ipples | -1.730 |
andy | -1.678 |
Token | to |
Token ID | 284 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
hy | +1.486 |
Mic | +1.347 |
Red | +1.246 |
Roger | +1.212 |
Robert | +1.200 |
Neg logit contributions | |
riage | -0.705 |
ipples | -0.705 |
flower | -0.682 |
department | -0.672 |
andy | -0.656 |
Token | play |
Token ID | 711 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
hy | +0.648 |
Mic | +0.589 |
Red | +0.522 |
Roger | +0.515 |
rides | +0.508 |
Neg logit contributions | |
riage | -0.533 |
ipples | -0.524 |
flower | -0.510 |
ry | -0.503 |
andy | -0.498 |
Token | all |
Token ID | 477 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
hy | +2.249 |
Mic | +1.900 |
nam | +1.720 |
Red | +1.681 |
Robert | +1.665 |
Neg logit contributions | |
ipples | -1.859 |
ry | -1.786 |
ac | -1.784 |
riage | -1.783 |
flower | -1.776 |
Token | day |
Token ID | 1110 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
hy | +0.599 |
Mic | +0.550 |
Red | +0.491 |
rides | +0.491 |
Roger | +0.488 |
Neg logit contributions | |
riage | -0.355 |
ipples | -0.352 |
flower | -0.339 |
ry | -0.338 |
ac | -0.332 |
Token | long |
Token ID | 890 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
hy | +10.220 |
Mic | +8.996 |
Red | +8.640 |
Robert | +8.003 |
Blue | +7.969 |
Neg logit contributions | |
ry | -4.974 |
ipples | -4.636 |
andy | -4.590 |
again | -4.582 |
boil | -4.400 |
Token | . |
Token ID | 13 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
hy | +6.748 |
Mic | +6.211 |
Red | +5.664 |
Robert | +5.601 |
Roger | +5.598 |
Neg logit contributions | |
again | -2.713 |
ry | -2.680 |
ipples | -2.642 |
en | -2.622 |
ac | -2.437 |
Token | He |
Token ID | 679 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
hy | +0.783 |
Mic | +0.697 |
rides | +0.613 |
Red | +0.607 |
Roger | +0.603 |
Neg logit contributions | |
riage | -0.678 |
ipples | -0.658 |
ry | -0.640 |
flower | -0.637 |
andy | -0.633 |
Token | had |
Token ID | 550 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
hy | +0.328 |
Mic | +0.300 |
rides | +0.270 |
Red | +0.270 |
Roger | +0.267 |
Neg logit contributions | |
riage | -0.172 |
ipples | -0.170 |
ry | -0.164 |
flower | -0.164 |
ac | -0.160 |
Token | many |
Token ID | 867 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
hy | +2.463 |
Mic | +2.217 |
Red | +1.983 |
Roger | +1.949 |
Robert | +1.919 |
Neg logit contributions | |
ipples | -1.668 |
ac | -1.604 |
riage | -1.556 |
flower | -1.532 |
department | -1.528 |
Token | uffy |
Token ID | 15352 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
riage | +0.513 |
ipples | +0.497 |
ry | +0.480 |
flower | +0.478 |
department | +0.469 |
Neg logit contributions | |
hy | -0.709 |
Mic | -0.644 |
rides | -0.575 |
Red | -0.571 |
Roger | -0.561 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
hy | +0.692 |
Mic | +0.625 |
Red | +0.543 |
rides | +0.541 |
Roger | +0.535 |
Neg logit contributions | |
riage | -0.580 |
ipples | -0.568 |
ry | -0.562 |
flower | -0.557 |
andy | -0.549 |
Token | to |
Token ID | 284 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
hy | +0.338 |
Mic | +0.309 |
Red | +0.279 |
rides | +0.277 |
Roger | +0.274 |
Neg logit contributions | |
riage | -0.186 |
ipples | -0.182 |
flower | -0.175 |
ry | -0.175 |
department | -0.171 |
Token | play |
Token ID | 711 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
riage | +0.162 |
ipples | +0.159 |
ry | +0.154 |
flower | +0.153 |
ac | +0.150 |
Neg logit contributions | |
hy | -0.205 |
Mic | -0.185 |
rides | -0.164 |
Red | -0.163 |
Roger | -0.160 |
Token | all |
Token ID | 477 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
hy | +2.014 |
Mic | +1.718 |
nam | +1.512 |
Robert | +1.484 |
Red | +1.480 |
Neg logit contributions | |
ipples | -1.690 |
ry | -1.667 |
riage | -1.657 |
flower | -1.633 |
ac | -1.630 |
Token | day |
Token ID | 1110 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
riage | +0.155 |
ipples | +0.151 |
ry | +0.146 |
flower | +0.145 |
ac | +0.142 |
Neg logit contributions | |
hy | -0.241 |
Mic | -0.220 |
rides | -0.197 |
Red | -0.196 |
Roger | -0.193 |
Token | and |
Token ID | 290 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
hy | +9.893 |
Mic | +9.066 |
Red | +8.445 |
Blue | +7.996 |
Bow | +7.914 |
Neg logit contributions | |
ry | -4.876 |
again | -4.724 |
andy | -4.491 |
ipples | -4.273 |
M | -4.220 |
Token | drink |
Token ID | 4144 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
hy | +0.364 |
Mic | +0.331 |
Red | +0.292 |
rides | +0.289 |
Roger | +0.286 |
Neg logit contributions | |
riage | -0.294 |
ipples | -0.286 |
flower | -0.278 |
ry | -0.278 |
andy | -0.271 |
Token | milk |
Token ID | 7545 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
hy | +1.154 |
Mic | +1.052 |
Red | +0.929 |
rides | +0.920 |
Roger | +0.917 |
Neg logit contributions | |
ipples | -0.778 |
riage | -0.768 |
ry | -0.753 |
flower | -0.741 |
andy | -0.735 |
Token | from |
Token ID | 422 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
hy | +1.614 |
Mic | +1.466 |
Red | +1.306 |
rides | +1.290 |
Roger | +1.289 |
Neg logit contributions | |
ipples | -1.056 |
flower | -1.037 |
riage | -1.035 |
ry | -1.032 |
department | -0.979 |
Token | his |
Token ID | 465 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
hy | +0.741 |
Mic | +0.673 |
Red | +0.597 |
rides | +0.594 |
Roger | +0.587 |
Neg logit contributions | |
riage | -0.508 |
ipples | -0.494 |
ry | -0.480 |
flower | -0.479 |
andy | -0.470 |
Token | the |
Token ID | 262 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
hy | +0.141 |
Mic | +0.129 |
rides | +0.117 |
Red | +0.116 |
Roger | +0.114 |
Neg logit contributions | |
riage | -0.074 |
ipples | -0.072 |
flower | -0.070 |
ry | -0.069 |
ac | -0.067 |
Token | middle |
Token ID | 3504 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
riage | +1.342 |
ipples | +1.291 |
ry | +1.277 |
ac | +1.247 |
flower | +1.244 |
Neg logit contributions | |
hy | -1.614 |
Mic | -1.434 |
rides | -1.281 |
Red | -1.266 |
Roger | -1.227 |
Token | of |
Token ID | 286 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
riage | +1.246 |
ipples | +1.187 |
department | +1.152 |
flower | +1.148 |
ry | +1.140 |
Neg logit contributions | |
hy | -1.563 |
Mic | -1.363 |
rides | -1.251 |
Red | -1.197 |
Robert | -1.180 |
Token | the |
Token ID | 262 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
riage | +0.187 |
ipples | +0.182 |
ry | +0.176 |
flower | +0.174 |
ac | +0.170 |
Neg logit contributions | |
hy | -0.344 |
Mic | -0.314 |
rides | -0.285 |
Red | -0.283 |
Roger | -0.278 |
Token | swamp |
Token ID | 26837 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
riage | +0.865 |
ipples | +0.806 |
ry | +0.793 |
ac | +0.777 |
flower | +0.774 |
Neg logit contributions | |
hy | -1.915 |
Mic | -1.763 |
rides | -1.626 |
Red | -1.602 |
Roger | -1.571 |
Token | ! |
Token ID | 0 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
hy | +0.313 |
Mic | +0.288 |
Red | +0.253 |
rides | +0.251 |
Roger | +0.249 |
Neg logit contributions | |
riage | -0.241 |
ipples | -0.237 |
ry | -0.233 |
flower | -0.229 |
department | -0.224 |
Token |
|
Token ID | 198 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
hy | +8.637 |
Mic | +7.283 |
hell | +7.127 |
nam | +7.105 |
thel | +7.045 |
Neg logit contributions | |
again | -4.938 |
department | -4.901 |
Duck | -4.887 |
riage | -4.555 |
Sometimes | -4.385 |
Token |
|
Token ID | 198 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
hy | +5.221 |
nam | +4.358 |
Mic | +4.347 |
brace | +4.140 |
blonde | +4.134 |
Neg logit contributions | |
ull | -2.425 |
ry | -2.413 |
ucks | -2.366 |
ac | -2.341 |
again | -2.327 |
Token | Blue |
Token ID | 14573 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
hy | +7.497 |
nam | +5.918 |
con | +5.867 |
blonde | +5.774 |
brace | +5.633 |
Neg logit contributions | |
ac | -4.278 |
department | -4.183 |
again | -4.145 |
riage | -4.142 |
end | -4.135 |
Token | w |
Token ID | 266 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
riage | +0.266 |
ipples | +0.259 |
ry | +0.251 |
flower | +0.249 |
ac | +0.244 |
Neg logit contributions | |
hy | -0.399 |
Mic | -0.363 |
rides | -0.327 |
Red | -0.323 |
Roger | -0.318 |
Token | aded |
Token ID | 5286 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
hy | +1.075 |
Mic | +0.977 |
rides | +0.884 |
Red | +0.880 |
Roger | +0.866 |
Neg logit contributions | |
ipples | -0.602 |
riage | -0.580 |
andy | -0.572 |
ry | -0.567 |
ac | -0.567 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
hy | +0.726 |
Mic | +0.660 |
Red | +0.568 |
Roger | +0.562 |
rides | +0.560 |
Neg logit contributions | |
riage | -0.763 |
ipples | -0.743 |
flower | -0.718 |
ry | -0.714 |
department | -0.709 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
hy | +2.803 |
Mic | +2.585 |
Roger | +2.274 |
nam | +2.271 |
Robert | +2.258 |
Neg logit contributions | |
flower | -1.834 |
riage | -1.821 |
ipples | -1.810 |
ry | -1.803 |
ac | -1.759 |
Token | to |
Token ID | 284 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
hy | +0.646 |
Mic | +0.587 |
Red | +0.533 |
rides | +0.525 |
Roger | +0.524 |
Neg logit contributions | |
riage | -0.346 |
ipples | -0.340 |
flower | -0.328 |
ry | -0.324 |
department | -0.322 |
Token | play |
Token ID | 711 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
hy | +0.137 |
Mic | +0.125 |
Red | +0.110 |
rides | +0.110 |
Roger | +0.109 |
Neg logit contributions | |
riage | -0.103 |
ipples | -0.101 |
flower | -0.097 |
ry | -0.097 |
andy | -0.095 |
Token | all |
Token ID | 477 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
hy | +0.856 |
Mic | +0.739 |
Red | +0.634 |
rides | +0.632 |
Roger | +0.626 |
Neg logit contributions | |
riage | -0.845 |
ipples | -0.843 |
ry | -0.827 |
flower | -0.815 |
ac | -0.808 |
Token | day |
Token ID | 1110 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
hy | +0.182 |
Mic | +0.167 |
rides | +0.149 |
Red | +0.149 |
Roger | +0.147 |
Neg logit contributions | |
riage | -0.112 |
ipples | -0.110 |
ry | -0.106 |
flower | -0.106 |
ac | -0.104 |
Token | long |
Token ID | 890 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
hy | +9.813 |
Mic | +8.839 |
Red | +8.302 |
Blue | +7.850 |
Robert | +7.767 |
Neg logit contributions | |
ry | -4.639 |
again | -4.297 |
ipples | -4.231 |
andy | -4.097 |
boil | -4.078 |
Token | . |
Token ID | 13 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
hy | +5.618 |
Mic | +5.263 |
Red | +4.675 |
Roger | +4.644 |
Robert | +4.627 |
Neg logit contributions | |
ry | -2.462 |
ipples | -2.397 |
again | -2.375 |
en | -2.330 |
ac | -2.263 |
Token | One |
Token ID | 1881 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
hy | +0.152 |
Mic | +0.137 |
rides | +0.118 |
Red | +0.118 |
Roger | +0.116 |
Neg logit contributions | |
riage | -0.151 |
ipples | -0.148 |
ry | -0.143 |
flower | -0.142 |
andy | -0.140 |
Token | day |
Token ID | 1110 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
riage | +1.206 |
ipples | +1.170 |
ry | +1.135 |
department | +1.120 |
flower | +1.115 |
Neg logit contributions | |
hy | -1.566 |
Mic | -1.399 |
rides | -1.255 |
Red | -1.237 |
Roger | -1.206 |
Token | , |
Token ID | 11 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
hy | +1.029 |
Mic | +0.956 |
Red | +0.835 |
rides | +0.818 |
Roger | +0.818 |
Neg logit contributions | |
riage | -0.872 |
ry | -0.847 |
ipples | -0.836 |
flower | -0.817 |
ac | -0.815 |
Token | pe |
Token ID | 613 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
riage | +1.553 |
ipples | +1.518 |
department | +1.471 |
ry | +1.453 |
ac | +1.437 |
Neg logit contributions | |
hy | -1.575 |
Mic | -1.355 |
rides | -1.244 |
Red | -1.185 |
brace | -1.157 |
Token | b |
Token ID | 65 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
hy | +0.041 |
Mic | +0.037 |
rides | +0.032 |
Red | +0.032 |
Roger | +0.032 |
Neg logit contributions | |
riage | -0.039 |
ipples | -0.038 |
ry | -0.037 |
flower | -0.037 |
ac | -0.036 |
Token | ble |
Token ID | 903 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
hy | +0.469 |
Mic | +0.439 |
rides | +0.404 |
Red | +0.404 |
Roger | +0.399 |
Neg logit contributions | |
riage | -0.126 |
ipples | -0.122 |
ry | -0.117 |
flower | -0.116 |
ac | -0.111 |
Token | I |
Token ID | 314 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
hy | +2.917 |
Mic | +2.708 |
Red | +2.391 |
Roger | +2.349 |
Robert | +2.306 |
Neg logit contributions | |
ry | -2.204 |
flower | -2.141 |
riage | -2.141 |
ipples | -2.111 |
ac | -2.043 |
Token | found |
Token ID | 1043 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
riage | +0.754 |
ipples | +0.732 |
ry | +0.704 |
flower | +0.703 |
department | +0.693 |
Neg logit contributions | |
hy | -1.098 |
Mic | -0.988 |
rides | -0.903 |
Red | -0.880 |
Roger | -0.862 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
hy | +2.882 |
Mic | +2.693 |
Red | +2.366 |
Roger | +2.312 |
blonde | +2.284 |
Neg logit contributions | |
riage | -2.021 |
ry | -1.970 |
ac | -1.942 |
flower | -1.916 |
ipples | -1.909 |
Token | said |
Token ID | 531 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
hy | +8.240 |
Mic | +7.060 |
hell | +6.799 |
blonde | +6.651 |
nam | +6.423 |
Neg logit contributions | |
Duck | -5.554 |
again | -5.266 |
riage | -4.963 |
Mama | -4.886 |
M | -4.849 |
Token | Blue |
Token ID | 4518 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
hy | +2.566 |
Mic | +2.168 |
rides | +1.845 |
hell | +1.810 |
blonde | +1.783 |
Neg logit contributions | |
riage | -3.943 |
flower | -3.749 |
department | -3.743 |
Duck | -3.703 |
again | -3.674 |
Token | . |
Token ID | 13 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
riage | +0.171 |
ipples | +0.167 |
ry | +0.163 |
flower | +0.162 |
ac | +0.158 |
Neg logit contributions | |
hy | -0.218 |
Mic | -0.197 |
rides | -0.176 |
Red | -0.174 |
Roger | -0.171 |
Token | " |
Token ID | 366 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
hy | +5.608 |
Mic | +4.885 |
hell | +4.610 |
nam | +4.603 |
blonde | +4.529 |
Neg logit contributions | |
department | -2.263 |
riage | -2.190 |
Duck | -2.096 |
flower | -2.076 |
again | -2.032 |
Token | Wow |
Token ID | 22017 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
hy | +2.732 |
Mic | +2.377 |
rides | +2.273 |
Robert | +2.158 |
nam | +2.148 |
Neg logit contributions | |
ipples | -1.504 |
riage | -1.503 |
flower | -1.492 |
ac | -1.418 |
andy | -1.361 |
Token | Max |
Token ID | 5436 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
hy | +2.433 |
Mic | +2.099 |
ve | +1.906 |
nam | +1.869 |
Roger | +1.830 |
Neg logit contributions | |
riage | -2.757 |
department | -2.617 |
ipples | -2.579 |
flower | -2.575 |
andy | -2.562 |
Token | loved |
Token ID | 6151 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
hy | +3.548 |
Mic | +3.257 |
nam | +2.946 |
Robert | +2.918 |
Roger | +2.891 |
Neg logit contributions | |
riage | -1.992 |
flower | -1.978 |
ry | -1.954 |
ipples | -1.917 |
andy | -1.897 |
Token | to |
Token ID | 284 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
hy | +1.275 |
Mic | +1.155 |
Red | +1.063 |
Roger | +1.037 |
Robert | +1.029 |
Neg logit contributions | |
riage | -0.646 |
ipples | -0.642 |
flower | -0.623 |
department | -0.611 |
andy | -0.602 |
Token | play |
Token ID | 711 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
hy | +1.139 |
Mic | +1.047 |
Red | +0.934 |
Roger | +0.926 |
Robert | +0.916 |
Neg logit contributions | |
riage | -0.857 |
ipples | -0.843 |
flower | -0.826 |
ry | -0.807 |
andy | -0.805 |
Token | all |
Token ID | 477 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
hy | +2.231 |
Mic | +1.881 |
nam | +1.689 |
Red | +1.662 |
Robert | +1.660 |
Neg logit contributions | |
ipples | -1.872 |
flower | -1.810 |
ry | -1.807 |
riage | -1.805 |
ac | -1.798 |
Token | day |
Token ID | 1110 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
hy | +0.183 |
Mic | +0.167 |
rides | +0.150 |
Red | +0.149 |
Roger | +0.148 |
Neg logit contributions | |
riage | -0.113 |
ipples | -0.111 |
ry | -0.107 |
flower | -0.107 |
ac | -0.104 |
Token | long |
Token ID | 890 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
hy | +9.717 |
Mic | +8.394 |
Red | +8.165 |
Robert | +7.558 |
Blue | +7.544 |
Neg logit contributions | |
ry | -4.671 |
ipples | -4.422 |
andy | -4.414 |
again | -4.358 |
boil | -4.159 |
Token | with |
Token ID | 351 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
hy | +6.315 |
Mic | +5.817 |
Red | +5.273 |
Robert | +5.221 |
Roger | +5.219 |
Neg logit contributions | |
again | -2.517 |
ry | -2.512 |
ipples | -2.471 |
en | -2.453 |
ac | -2.315 |
Token | his |
Token ID | 465 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
hy | +2.827 |
Mic | +2.505 |
Red | +2.317 |
Roger | +2.277 |
Robert | +2.265 |
Neg logit contributions | |
ipples | -1.666 |
ac | -1.624 |
flower | -1.622 |
andy | -1.618 |
riage | -1.578 |
Token | friends |
Token ID | 2460 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
hy | +4.827 |
Mic | +4.405 |
Red | +4.224 |
Roger | +4.156 |
Robert | +4.145 |
Neg logit contributions | |
flower | -1.684 |
ipples | -1.630 |
andy | -1.624 |
ac | -1.540 |
department | -1.502 |
Token | in |
Token ID | 287 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
hy | +0.297 |
Mic | +0.268 |
rides | +0.237 |
Red | +0.237 |
Roger | +0.232 |
Neg logit contributions | |
riage | -0.231 |
ipples | -0.226 |
ry | -0.223 |
flower | -0.222 |
ac | -0.216 |
Token | He |
Token ID | 679 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
riage | +0.550 |
ipples | +0.540 |
ry | +0.529 |
flower | +0.525 |
ac | +0.519 |
Neg logit contributions | |
hy | -0.448 |
Mic | -0.392 |
rides | -0.338 |
Red | -0.333 |
Roger | -0.325 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
hy | +1.309 |
Mic | +1.193 |
rides | +1.028 |
Robert | +1.027 |
Red | +1.026 |
Neg logit contributions | |
riage | -1.054 |
ipples | -1.044 |
flower | -1.027 |
ry | -1.022 |
andy | -1.008 |
Token | to |
Token ID | 284 |
Feature activation | +8.02e-02 |
Pos logit contributions | |
hy | +0.489 |
Mic | +0.448 |
Red | +0.405 |
rides | +0.399 |
Roger | +0.396 |
Neg logit contributions | |
riage | -0.262 |
ipples | -0.257 |
flower | -0.248 |
ry | -0.246 |
department | -0.242 |
Token | play |
Token ID | 711 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
hy | +0.162 |
Mic | +0.147 |
Red | +0.130 |
rides | +0.130 |
Roger | +0.128 |
Neg logit contributions | |
riage | -0.125 |
ipples | -0.122 |
flower | -0.119 |
ry | -0.118 |
andy | -0.116 |
Token | all |
Token ID | 477 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
hy | +1.077 |
Mic | +0.935 |
Red | +0.805 |
Robert | +0.798 |
Roger | +0.794 |
Neg logit contributions | |
ipples | -0.995 |
riage | -0.990 |
ry | -0.975 |
flower | -0.961 |
ac | -0.947 |
Token | day |
Token ID | 1110 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
riage | +0.618 |
ipples | +0.598 |
ry | +0.580 |
flower | +0.575 |
department | +0.573 |
Neg logit contributions | |
hy | -0.901 |
Mic | -0.813 |
rides | -0.739 |
Red | -0.726 |
Roger | -0.711 |
Token | in |
Token ID | 287 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
hy | +9.961 |
Mic | +9.053 |
Red | +8.435 |
Blue | +8.042 |
Bow | +8.000 |
Neg logit contributions | |
ry | -4.291 |
again | -3.956 |
ipples | -3.855 |
boil | -3.697 |
en | -3.695 |
Token | the |
Token ID | 262 |
Feature activation | +7.44e-02 |
Pos logit contributions | |
riage | +0.586 |
ipples | +0.569 |
ry | +0.548 |
flower | +0.544 |
department | +0.535 |
Neg logit contributions | |
hy | -1.025 |
Mic | -0.935 |
rides | -0.858 |
Red | -0.838 |
Roger | -0.829 |
Token | sunshine |
Token ID | 34488 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
hy | +0.157 |
Mic | +0.142 |
rides | +0.126 |
Red | +0.126 |
Roger | +0.124 |
Neg logit contributions | |
riage | -0.111 |
ipples | -0.109 |
flower | -0.106 |
ry | -0.105 |
andy | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
hy | +0.479 |
Mic | +0.439 |
Red | +0.387 |
rides | +0.383 |
Roger | +0.381 |
Neg logit contributions | |
riage | -0.354 |
ipples | -0.351 |
flower | -0.343 |
ry | -0.342 |
ac | -0.328 |
Token | One |
Token ID | 1881 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
riage | +1.083 |
ipples | +1.063 |
ry | +1.040 |
flower | +1.038 |
ac | +1.024 |
Neg logit contributions | |
hy | -1.027 |
Mic | -0.908 |
rides | -0.803 |
Red | -0.791 |
Roger | -0.767 |
Token | through |
Token ID | 832 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
riage | +0.601 |
ipples | +0.582 |
ry | +0.565 |
flower | +0.560 |
ac | +0.550 |
Neg logit contributions | |
hy | -0.840 |
Mic | -0.754 |
rides | -0.684 |
Red | -0.674 |
Roger | -0.662 |
Token | the |
Token ID | 262 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
hy | +0.266 |
Mic | +0.244 |
Red | +0.215 |
rides | +0.215 |
Roger | +0.212 |
Neg logit contributions | |
riage | -0.178 |
ipples | -0.173 |
ry | -0.170 |
flower | -0.168 |
ac | -0.164 |
Token | pages |
Token ID | 5468 |
Feature activation | +7.07e-01 |
Pos logit contributions | |
hy | +4.172 |
Mic | +3.996 |
Roger | +3.521 |
Robert | +3.450 |
Bow | +3.439 |
Neg logit contributions | |
riage | -4.262 |
ac | -4.188 |
department | -4.187 |
flower | -4.102 |
liness | -4.046 |
Token | of |
Token ID | 286 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
hy | +1.499 |
Mic | +1.434 |
Red | +1.245 |
Roger | +1.237 |
rides | +1.226 |
Neg logit contributions | |
ipples | -1.010 |
ry | -0.999 |
riage | -0.995 |
flower | -0.965 |
ac | -0.941 |
Token | all |
Token ID | 477 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
hy | +2.546 |
Mic | +2.360 |
Red | +2.116 |
Roger | +2.068 |
rides | +2.060 |
Neg logit contributions | |
riage | -1.319 |
ipples | -1.235 |
flower | -1.226 |
andy | -1.182 |
department | -1.181 |
Token | the |
Token ID | 262 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
riage | +1.088 |
ipples | +1.065 |
ry | +1.021 |
department | +1.012 |
flower | +1.010 |
Neg logit contributions | |
hy | -1.402 |
Mic | -1.228 |
rides | -1.134 |
Red | -1.096 |
Roger | -1.074 |
Token | different |
Token ID | 1180 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
hy | +7.279 |
Mic | +6.718 |
Roger | +6.119 |
Blue | +6.012 |
Robert | +5.934 |
Neg logit contributions | |
ac | -6.146 |
en | -5.898 |
liness | -5.735 |
riage | -5.712 |
department | -5.581 |
Token | magazines |
Token ID | 16695 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
hy | +3.183 |
Mic | +3.093 |
Roger | +2.711 |
Red | +2.695 |
Blue | +2.642 |
Neg logit contributions | |
riage | -2.082 |
ry | -2.015 |
en | -2.006 |
ac | -1.993 |
ipples | -1.990 |
Token | . |
Token ID | 13 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
Mic | +4.139 |
hy | +4.072 |
Red | +3.505 |
Roger | +3.494 |
Robert | +3.414 |
Neg logit contributions | |
ry | -2.414 |
riage | -2.348 |
flower | -2.329 |
ipples | -2.295 |
again | -2.204 |
Token | She |
Token ID | 1375 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
hy | +1.879 |
Mic | +1.703 |
rides | +1.483 |
Roger | +1.424 |
Red | +1.422 |
Neg logit contributions | |
riage | -1.691 |
ry | -1.602 |
ipples | -1.576 |
department | -1.570 |
flower | -1.553 |
Token | looked |
Token ID | 3114 |
Feature activation | +8.21e-01 |
Pos logit contributions | |
hy | +1.308 |
Mic | +1.213 |
Red | +1.072 |
Roger | +1.063 |
rides | +1.047 |
Neg logit contributions | |
riage | -0.796 |
ipples | -0.785 |
ry | -0.784 |
flower | -0.762 |
andy | -0.752 |
Token | play |
Token ID | 711 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
hy | +0.615 |
Mic | +0.561 |
Red | +0.497 |
Roger | +0.491 |
rides | +0.491 |
Neg logit contributions | |
riage | -0.459 |
ipples | -0.448 |
flower | -0.438 |
ry | -0.435 |
andy | -0.428 |
Token | with |
Token ID | 351 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
hy | +2.755 |
Mic | +2.323 |
Robert | +2.052 |
Roger | +2.012 |
nam | +2.011 |
Neg logit contributions | |
flower | -2.235 |
ry | -2.233 |
ipples | -2.232 |
riage | -2.183 |
andy | -2.166 |
Token | her |
Token ID | 607 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
hy | +3.309 |
Mic | +2.854 |
Robert | +2.585 |
Roger | +2.576 |
Red | +2.556 |
Neg logit contributions | |
flower | -2.287 |
ry | -2.214 |
andy | -2.213 |
en | -2.165 |
ipples | -2.165 |
Token | toys |
Token ID | 14958 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
hy | +3.004 |
Mic | +2.619 |
Robert | +2.374 |
Red | +2.357 |
Roger | +2.341 |
Neg logit contributions | |
flower | -2.298 |
ry | -2.151 |
riage | -2.129 |
ipples | -2.114 |
andy | -2.111 |
Token | all |
Token ID | 477 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
hy | +3.103 |
Mic | +2.877 |
Roger | +2.543 |
Robert | +2.527 |
Red | +2.491 |
Neg logit contributions | |
flower | -1.907 |
ry | -1.903 |
ipples | -1.846 |
riage | -1.842 |
ac | -1.790 |
Token | day |
Token ID | 1110 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
hy | +0.326 |
Mic | +0.300 |
rides | +0.268 |
Red | +0.267 |
Roger | +0.264 |
Neg logit contributions | |
riage | -0.200 |
ipples | -0.195 |
flower | -0.189 |
ry | -0.189 |
ac | -0.184 |
Token | long |
Token ID | 890 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
hy | +8.870 |
Mic | +8.054 |
Red | +7.492 |
Bow | +7.185 |
Blue | +7.141 |
Neg logit contributions | |
ry | -5.211 |
andy | -4.907 |
again | -4.860 |
ipples | -4.680 |
ac | -4.459 |
Token | . |
Token ID | 13 |
Feature activation | +9.00e-02 |
Pos logit contributions | |
hy | +5.373 |
Mic | +5.174 |
Roger | +4.623 |
Red | +4.608 |
Robert | +4.563 |
Neg logit contributions | |
ry | -2.449 |
again | -2.366 |
en | -2.344 |
ac | -2.294 |
ipples | -2.285 |
Token | One |
Token ID | 1881 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
hy | +0.177 |
Mic | +0.158 |
rides | +0.140 |
Red | +0.138 |
Roger | +0.137 |
Neg logit contributions | |
riage | -0.147 |
ipples | -0.143 |
ry | -0.140 |
flower | -0.140 |
andy | -0.137 |
Token | day |
Token ID | 1110 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
riage | +1.123 |
ipples | +1.102 |
ry | +1.048 |
department | +1.047 |
flower | +1.034 |
Neg logit contributions | |
hy | -1.495 |
Mic | -1.325 |
rides | -1.200 |
Red | -1.179 |
Roger | -1.150 |
Token | , |
Token ID | 11 |
Feature activation | +9.92e-01 |
Pos logit contributions | |
hy | +3.082 |
Mic | +2.843 |
Red | +2.497 |
Robert | +2.472 |
rides | +2.465 |
Neg logit contributions | |
ry | -2.072 |
riage | -2.028 |
andy | -1.932 |
flower | -1.923 |
ac | -1.881 |
Token | ? |
Token ID | 30 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
riage | +1.130 |
ipples | +1.121 |
ry | +1.042 |
flower | +1.037 |
department | +1.035 |
Neg logit contributions | |
hy | -1.967 |
Mic | -1.762 |
rides | -1.627 |
Red | -1.584 |
Roger | -1.571 |
Token | They |
Token ID | 1119 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
riage | +0.874 |
ipples | +0.869 |
ry | +0.837 |
flower | +0.834 |
andy | +0.825 |
Neg logit contributions | |
hy | -0.958 |
Mic | -0.853 |
Red | -0.754 |
rides | -0.754 |
Roger | -0.734 |
Token | look |
Token ID | 804 |
Feature activation | +9.40e-01 |
Pos logit contributions | |
hy | +0.557 |
Mic | +0.513 |
rides | +0.450 |
Red | +0.449 |
Roger | +0.444 |
Neg logit contributions | |
riage | -0.394 |
ipples | -0.382 |
ry | -0.373 |
flower | -0.373 |
ac | -0.365 |
Token | so |
Token ID | 523 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
hy | +1.704 |
Mic | +1.573 |
rides | +1.317 |
Red | +1.307 |
Roger | +1.297 |
Neg logit contributions | |
riage | -1.673 |
ry | -1.604 |
ipples | -1.602 |
ac | -1.595 |
department | -1.554 |
Token | amazing |
Token ID | 4998 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
hy | +1.896 |
Mic | +1.793 |
Roger | +1.435 |
rides | +1.422 |
Robert | +1.416 |
Neg logit contributions | |
riage | -2.193 |
ipples | -2.145 |
ry | -2.106 |
ac | -2.077 |
andy | -2.047 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
hy | +0.887 |
Mic | +0.790 |
Red | +0.676 |
rides | +0.670 |
Roger | +0.654 |
Neg logit contributions | |
riage | -0.883 |
ry | -0.854 |
ipples | -0.852 |
flower | -0.833 |
ac | -0.820 |
Token |
|
Token ID | 198 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
hy | +8.246 |
Mic | +6.685 |
hell | +6.458 |
Red | +5.939 |
brace | +5.870 |
Neg logit contributions | |
Mama | -5.902 |
again | -5.700 |
Duck | -5.529 |
Mum | -5.496 |
M | -5.393 |
Token |
|
Token ID | 198 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
hy | +5.318 |
Mic | +4.261 |
brace | +4.114 |
rides | +4.095 |
nam | +4.062 |
Neg logit contributions | |
riage | -2.681 |
ry | -2.670 |
ucks | -2.635 |
ull | -2.626 |
again | -2.618 |
Token | " |
Token ID | 1 |
Feature activation | -3.61e-04 |
Pos logit contributions | |
hy | +3.713 |
Mic | +2.897 |
rides | +2.820 |
con | +2.721 |
brace | +2.718 |
Neg logit contributions | |
riage | -2.657 |
ac | -2.562 |
ry | -2.532 |
department | -2.531 |
ipples | -2.500 |
Token | Yes |
Token ID | 5297 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
riage | +0.001 |
ipples | +0.001 |
flower | +0.001 |
ry | +0.001 |
ac | +0.001 |
Neg logit contributions | |
hy | -0.001 |
Mic | -0.001 |
rides | -0.001 |
Red | -0.001 |
Roger | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
hy | +1.319 |
Mic | +1.191 |
rides | +1.051 |
Red | +1.024 |
nam | +0.985 |
Neg logit contributions | |
riage | -1.276 |
ry | -1.230 |
department | -1.205 |
ipples | -1.203 |
flower | -1.194 |
Token | we |
Token ID | 356 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
riage | +1.084 |
ipples | +1.054 |
ry | +1.018 |
flower | +1.014 |
andy | +1.008 |
Neg logit contributions | |
hy | -1.293 |
Mic | -1.133 |
rides | -1.033 |
Red | -1.004 |
Roger | -0.988 |
Token | breathe |
Token ID | 18044 |
Feature activation | -3.23e-01 |
Pos logit contributions | |
hy | +0.528 |
Mic | +0.482 |
rides | +0.425 |
Red | +0.425 |
Roger | +0.418 |
Neg logit contributions | |
riage | -0.388 |
ipples | -0.383 |
ry | -0.372 |
flower | -0.368 |
ac | -0.361 |
Token | . |
Token ID | 13 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
riage | +0.521 |
ipples | +0.507 |
flower | +0.493 |
ry | +0.492 |
andy | +0.482 |
Neg logit contributions | |
hy | -0.634 |
Mic | -0.571 |
rides | -0.511 |
Red | -0.504 |
Roger | -0.494 |
Token | |
Token ID | 220 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
hy | +5.659 |
Mic | +4.695 |
blonde | +4.400 |
nam | +4.383 |
hell | +4.335 |
Neg logit contributions | |
riage | -3.552 |
department | -3.402 |
flower | -3.378 |
again | -3.367 |
Duck | -3.365 |
Token |
|
Token ID | 198 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
hy | +0.776 |
Mic | +0.689 |
rides | +0.615 |
Red | +0.601 |
Roger | +0.591 |
Neg logit contributions | |
riage | -0.635 |
ipples | -0.630 |
ry | -0.619 |
flower | -0.601 |
ac | -0.601 |
Token |
|
Token ID | 198 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
hy | +3.557 |
Mic | +3.070 |
nam | +2.934 |
rides | +2.792 |
blonde | +2.783 |
Neg logit contributions | |
ry | -2.040 |
riage | -2.021 |
andy | -2.016 |
ipples | -2.015 |
ull | -1.953 |
Token | Blue |
Token ID | 14573 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
hy | +7.359 |
nam | +5.622 |
con | +5.490 |
dream | +5.379 |
blonde | +5.349 |
Neg logit contributions | |
flower | -5.784 |
department | -5.741 |
end | -5.719 |
ac | -5.631 |
Duck | -5.576 |
Token | sw |
Token ID | 1509 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
riage | +1.389 |
ipples | +1.221 |
department | +1.097 |
ours | +1.092 |
ucks | +1.086 |
Neg logit contributions | |
hy | -3.300 |
Mic | -3.098 |
rides | -2.777 |
Red | -2.769 |
Roger | -2.745 |
Token | am |
Token ID | 321 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
riage | +2.654 |
department | +2.307 |
flower | +2.291 |
ipples | +2.268 |
ry | +2.196 |
Neg logit contributions | |
hy | -4.564 |
Mic | -4.172 |
rides | -3.886 |
Red | -3.683 |
Bow | -3.642 |
Token | up |
Token ID | 510 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
riage | +0.779 |
ipples | +0.752 |
ry | +0.726 |
flower | +0.725 |
andy | +0.709 |
Neg logit contributions | |
hy | -1.048 |
Mic | -0.946 |
rides | -0.852 |
Red | -0.838 |
Roger | -0.830 |
Token | to |
Token ID | 284 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
hy | +0.069 |
Mic | +0.063 |
rides | +0.057 |
Red | +0.056 |
Roger | +0.055 |
Neg logit contributions | |
riage | -0.041 |
ipples | -0.040 |
ry | -0.039 |
flower | -0.039 |
department | -0.038 |
Token | play |
Token ID | 711 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
hy | +0.666 |
Mic | +0.608 |
Red | +0.540 |
Roger | +0.532 |
rides | +0.531 |
Neg logit contributions | |
riage | -0.486 |
ipples | -0.473 |
flower | -0.460 |
ry | -0.456 |
andy | -0.452 |
Token | with |
Token ID | 351 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
hy | +2.149 |
Mic | +1.800 |
nam | +1.571 |
Robert | +1.567 |
Red | +1.562 |
Neg logit contributions | |
ipples | -1.862 |
ry | -1.845 |
riage | -1.840 |
flower | -1.836 |
andy | -1.814 |
Token | her |
Token ID | 607 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
hy | +4.310 |
Mic | +3.702 |
Robert | +3.426 |
Roger | +3.411 |
Red | +3.368 |
Neg logit contributions | |
andy | -2.674 |
flower | -2.630 |
en | -2.559 |
ipples | -2.513 |
ucks | -2.475 |
Token | toys |
Token ID | 14958 |
Feature activation | +8.89e-01 |
Pos logit contributions | |
hy | +3.848 |
Mic | +3.341 |
Robert | +3.069 |
Red | +3.007 |
Roger | +2.998 |
Neg logit contributions | |
flower | -2.643 |
department | -2.406 |
ipples | -2.391 |
ry | -2.383 |
andy | -2.373 |
Token | all |
Token ID | 477 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
hy | +1.909 |
Mic | +1.736 |
Red | +1.530 |
Roger | +1.523 |
Robert | +1.511 |
Neg logit contributions | |
ry | -1.283 |
flower | -1.281 |
ipples | -1.277 |
riage | -1.275 |
ac | -1.233 |
Token | day |
Token ID | 1110 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
riage | +0.738 |
ipples | +0.719 |
ry | +0.696 |
department | +0.690 |
flower | +0.686 |
Neg logit contributions | |
hy | -1.053 |
Mic | -0.947 |
rides | -0.858 |
Red | -0.844 |
Roger | -0.828 |
Token | long |
Token ID | 890 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
hy | +8.006 |
Mic | +6.787 |
Red | +6.479 |
Bow | +6.127 |
Robert | +6.099 |
Neg logit contributions | |
ry | -5.918 |
andy | -5.667 |
again | -5.543 |
boil | -5.506 |
ipples | -5.469 |
Token | . |
Token ID | 13 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
hy | +4.726 |
Mic | +4.521 |
Roger | +4.027 |
Red | +4.020 |
Robert | +3.992 |
Neg logit contributions | |
ry | -2.245 |
en | -2.130 |
ipples | -2.125 |
ac | -2.110 |
again | -2.104 |
Token | One |
Token ID | 1881 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
riage | +0.443 |
ipples | +0.435 |
ry | +0.423 |
flower | +0.420 |
ac | +0.413 |
Neg logit contributions | |
hy | -0.479 |
Mic | -0.429 |
rides | -0.378 |
Red | -0.374 |
Roger | -0.366 |
Token | day |
Token ID | 1110 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
riage | +0.959 |
ipples | +0.935 |
ry | +0.902 |
department | +0.886 |
flower | +0.884 |
Neg logit contributions | |
hy | -1.298 |
Mic | -1.164 |
rides | -1.043 |
Red | -1.028 |
Roger | -1.008 |
Token | , |
Token ID | 11 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
hy | +2.317 |
Mic | +2.132 |
Red | +1.892 |
rides | +1.875 |
Robert | +1.857 |
Neg logit contributions | |
ry | -1.658 |
riage | -1.615 |
flower | -1.553 |
andy | -1.550 |
ac | -1.517 |
Token | She |
Token ID | 1375 |
Feature activation | +9.29e-01 |
Pos logit contributions | |
riage | +1.086 |
ipples | +1.076 |
ry | +1.049 |
flower | +1.041 |
ac | +1.028 |
Neg logit contributions | |
hy | -0.766 |
Mic | -0.661 |
rides | -0.564 |
Red | -0.561 |
Roger | -0.531 |
Token | would |
Token ID | 561 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
hy | +2.041 |
Mic | +1.823 |
Red | +1.632 |
Roger | +1.630 |
rides | +1.621 |
Neg logit contributions | |
riage | -1.268 |
ipples | -1.239 |
ry | -1.236 |
flower | -1.236 |
ac | -1.227 |
Token | eat |
Token ID | 4483 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
hy | +2.242 |
Mic | +2.022 |
Red | +1.820 |
Roger | +1.802 |
Robert | +1.753 |
Neg logit contributions | |
riage | -1.848 |
ipples | -1.818 |
flower | -1.811 |
ac | -1.769 |
andy | -1.757 |
Token | candy |
Token ID | 18550 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
hy | +4.154 |
Mic | +3.406 |
powers | +3.006 |
Robert | +3.003 |
Roger | +2.998 |
Neg logit contributions | |
again | -3.170 |
flower | -3.157 |
riage | -3.145 |
andy | -3.114 |
en | -3.107 |
Token | all |
Token ID | 477 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
hy | +5.253 |
Mic | +4.641 |
Roger | +4.214 |
blonde | +4.191 |
Red | +4.179 |
Neg logit contributions | |
ry | -2.968 |
flower | -2.904 |
again | -2.875 |
ipples | -2.821 |
ac | -2.777 |
Token | day |
Token ID | 1110 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
riage | +0.608 |
ipples | +0.591 |
ry | +0.571 |
flower | +0.565 |
department | +0.562 |
Neg logit contributions | |
hy | -0.894 |
Mic | -0.811 |
rides | -0.732 |
Red | -0.722 |
Roger | -0.710 |
Token | and |
Token ID | 290 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
hy | +8.784 |
Mic | +7.417 |
Red | +7.398 |
Bow | +6.996 |
Blue | +6.787 |
Neg logit contributions | |
ry | -5.375 |
andy | -5.021 |
boil | -4.946 |
again | -4.871 |
ipples | -4.852 |
Token | all |
Token ID | 477 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
hy | +2.293 |
Mic | +2.093 |
Red | +1.886 |
Roger | +1.851 |
Robert | +1.797 |
Neg logit contributions | |
riage | -1.668 |
ipples | -1.606 |
flower | -1.604 |
ry | -1.559 |
andy | -1.545 |
Token | night |
Token ID | 1755 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
riage | +1.031 |
ipples | +1.015 |
ry | +0.980 |
department | +0.974 |
flower | +0.969 |
Neg logit contributions | |
hy | -1.363 |
Mic | -1.203 |
rides | -1.099 |
Red | -1.066 |
nam | -1.045 |
Token | . |
Token ID | 13 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
hy | +7.603 |
Mic | +7.179 |
Robert | +6.345 |
Red | +6.280 |
Bow | +6.273 |
Neg logit contributions | |
ry | -4.098 |
ipples | -3.766 |
boil | -3.591 |
again | -3.516 |
en | -3.468 |
Token | One |
Token ID | 1881 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
hy | +0.383 |
Mic | +0.342 |
rides | +0.301 |
Red | +0.296 |
Roger | +0.295 |
Neg logit contributions | |
riage | -0.336 |
ipples | -0.326 |
ry | -0.320 |
flower | -0.319 |
andy | -0.315 |
Token | toy |
Token ID | 13373 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
hy | +2.560 |
Mic | +2.256 |
Red | +2.034 |
Robert | +2.017 |
Roger | +1.996 |
Neg logit contributions | |
flower | -1.834 |
ipples | -1.759 |
riage | -1.753 |
department | -1.735 |
ry | -1.728 |
Token | cars |
Token ID | 5006 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
hy | +0.981 |
Mic | +0.877 |
Red | +0.770 |
Robert | +0.769 |
Roger | +0.768 |
Neg logit contributions | |
riage | -0.769 |
ry | -0.755 |
flower | -0.750 |
ipples | -0.748 |
department | -0.733 |
Token | and |
Token ID | 290 |
Feature activation | +9.02e-01 |
Pos logit contributions | |
hy | +4.044 |
Mic | +3.856 |
Roger | +3.503 |
Robert | +3.441 |
blonde | +3.427 |
Neg logit contributions | |
flower | -2.456 |
ry | -2.399 |
ipples | -2.383 |
ac | -2.369 |
again | -2.280 |
Token | trucks |
Token ID | 13960 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
hy | +1.832 |
Mic | +1.623 |
Red | +1.452 |
Roger | +1.447 |
Robert | +1.442 |
Neg logit contributions | |
riage | -1.389 |
flower | -1.342 |
department | -1.328 |
ipples | -1.327 |
ric | -1.302 |
Token | all |
Token ID | 477 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
Mic | +4.850 |
hy | +4.777 |
Roger | +4.336 |
Robert | +4.206 |
blonde | +4.181 |
Neg logit contributions | |
ipples | -2.632 |
flower | -2.583 |
ry | -2.495 |
department | -2.483 |
again | -2.474 |
Token | day |
Token ID | 1110 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
riage | +0.669 |
ipples | +0.650 |
ry | +0.634 |
flower | +0.624 |
department | +0.621 |
Neg logit contributions | |
hy | -0.957 |
Mic | -0.857 |
rides | -0.779 |
Red | -0.766 |
Roger | -0.751 |
Token | long |
Token ID | 890 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
hy | +7.448 |
Mic | +6.922 |
Red | +6.197 |
Blue | +5.919 |
Bow | +5.837 |
Neg logit contributions | |
ry | -5.823 |
ipples | -5.762 |
again | -5.601 |
ull | -5.388 |
andy | -5.310 |
Token | . |
Token ID | 13 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
hy | +4.427 |
Mic | +4.280 |
Red | +3.792 |
Roger | +3.783 |
Robert | +3.768 |
Neg logit contributions | |
ry | -2.181 |
ipples | -2.157 |
en | -2.110 |
again | -2.074 |
riage | -2.072 |
Token | One |
Token ID | 1881 |
Feature activation | -9.17e-01 |
Pos logit contributions | |
riage | +0.338 |
ipples | +0.331 |
ry | +0.323 |
flower | +0.321 |
ac | +0.315 |
Neg logit contributions | |
hy | -0.374 |
Mic | -0.336 |
rides | -0.296 |
Red | -0.293 |
Roger | -0.287 |
Token | day |
Token ID | 1110 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
riage | +1.472 |
ipples | +1.442 |
ry | +1.374 |
department | +1.372 |
flower | +1.356 |
Neg logit contributions | |
hy | -1.868 |
Mic | -1.643 |
rides | -1.494 |
Red | -1.463 |
Roger | -1.426 |
Token | , |
Token ID | 11 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
hy | +2.702 |
Mic | +2.476 |
Red | +2.199 |
rides | +2.159 |
Robert | +2.147 |
Neg logit contributions | |
ry | -1.849 |
riage | -1.824 |
flower | -1.746 |
ipples | -1.731 |
department | -1.727 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
riage | +0.957 |
ipples | +0.943 |
ry | +0.923 |
flower | +0.914 |
ac | +0.905 |
Neg logit contributions | |
hy | -0.656 |
Mic | -0.568 |
Red | -0.475 |
rides | -0.473 |
Roger | -0.453 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
hy | +2.174 |
Mic | +1.958 |
Roger | +1.704 |
Robert | +1.703 |
rides | +1.693 |
Neg logit contributions | |
riage | -1.769 |
flower | -1.768 |
ry | -1.752 |
ipples | -1.734 |
ac | -1.696 |
Token | to |
Token ID | 284 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
hy | +0.413 |
Mic | +0.376 |
Red | +0.338 |
rides | +0.335 |
Roger | +0.333 |
Neg logit contributions | |
riage | -0.245 |
ipples | -0.237 |
flower | -0.232 |
ry | -0.230 |
department | -0.225 |
Token | play |
Token ID | 711 |
Feature activation | +9.60e-01 |
Pos logit contributions | |
hy | +0.212 |
Mic | +0.193 |
Red | +0.171 |
rides | +0.170 |
Roger | +0.168 |
Neg logit contributions | |
riage | -0.160 |
ipples | -0.156 |
ry | -0.152 |
flower | -0.151 |
andy | -0.148 |
Token | all |
Token ID | 477 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
hy | +1.814 |
Mic | +1.544 |
Robert | +1.342 |
Red | +1.330 |
Roger | +1.326 |
Neg logit contributions | |
ry | -1.623 |
ipples | -1.620 |
riage | -1.609 |
flower | -1.608 |
andy | -1.576 |
Token | day |
Token ID | 1110 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
riage | +0.222 |
ipples | +0.216 |
ry | +0.209 |
flower | +0.208 |
ac | +0.203 |
Neg logit contributions | |
hy | -0.341 |
Mic | -0.310 |
rides | -0.279 |
Red | -0.276 |
Roger | -0.272 |
Token | long |
Token ID | 890 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
hy | +9.069 |
Mic | +8.164 |
Red | +7.618 |
Bow | +7.352 |
Blue | +7.339 |
Neg logit contributions | |
ry | -4.903 |
andy | -4.415 |
again | -4.370 |
boil | -4.251 |
ipples | -4.244 |
Token | and |
Token ID | 290 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
hy | +6.286 |
Mic | +5.869 |
Roger | +5.270 |
Robert | +5.264 |
Red | +5.219 |
Neg logit contributions | |
ry | -2.524 |
again | -2.495 |
en | -2.462 |
ipples | -2.282 |
andy | -2.233 |
Token | never |
Token ID | 1239 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
hy | +0.408 |
Mic | +0.369 |
rides | +0.324 |
Red | +0.322 |
Roger | +0.318 |
Neg logit contributions | |
riage | -0.359 |
ipples | -0.347 |
ry | -0.340 |
flower | -0.340 |
ac | -0.333 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
riage | +0.964 |
ipples | +0.947 |
ry | +0.921 |
flower | +0.914 |
andy | +0.906 |
Neg logit contributions | |
hy | -1.093 |
Mic | -0.957 |
rides | -0.868 |
Red | -0.846 |
Roger | -0.818 |
Token | to |
Token ID | 284 |
Feature activation | -6.95e-01 |
Pos logit contributions | |
hy | +0.261 |
Mic | +0.241 |
Red | +0.215 |
rides | +0.215 |
Roger | +0.212 |
Neg logit contributions | |
riage | -0.145 |
ipples | -0.141 |
ry | -0.137 |
flower | -0.135 |
department | -0.134 |
Token | play |
Token ID | 711 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
riage | +0.306 |
ipples | +0.299 |
ry | +0.291 |
flower | +0.289 |
ac | +0.283 |
Neg logit contributions | |
hy | -0.406 |
Mic | -0.367 |
rides | -0.328 |
Red | -0.324 |
Roger | -0.319 |
Token | with |
Token ID | 351 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
hy | +0.858 |
Mic | +0.747 |
Red | +0.632 |
rides | +0.631 |
Robert | +0.628 |
Neg logit contributions | |
riage | -0.842 |
ipples | -0.840 |
ry | -0.824 |
flower | -0.817 |
ac | -0.804 |
Token | his |
Token ID | 465 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
hy | +0.326 |
Mic | +0.293 |
rides | +0.258 |
Red | +0.258 |
Roger | +0.254 |
Neg logit contributions | |
riage | -0.243 |
ipples | -0.239 |
ry | -0.234 |
flower | -0.232 |
andy | -0.227 |
Token | toys |
Token ID | 14958 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
hy | +1.528 |
Mic | +1.340 |
Red | +1.198 |
rides | +1.191 |
Robert | +1.188 |
Neg logit contributions | |
flower | -1.155 |
riage | -1.144 |
ipples | -1.140 |
ry | -1.119 |
department | -1.113 |
Token | all |
Token ID | 477 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
hy | +1.627 |
Mic | +1.484 |
Roger | +1.294 |
Red | +1.293 |
Robert | +1.288 |
Neg logit contributions | |
riage | -1.060 |
ry | -1.053 |
ipples | -1.052 |
flower | -1.034 |
ac | -1.009 |
Token | day |
Token ID | 1110 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
hy | +0.110 |
Mic | +0.100 |
rides | +0.090 |
Red | +0.089 |
Roger | +0.088 |
Neg logit contributions | |
riage | -0.069 |
ipples | -0.067 |
ry | -0.065 |
flower | -0.065 |
ac | -0.063 |
Token | long |
Token ID | 890 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
hy | +8.636 |
Mic | +7.798 |
Red | +7.134 |
Blue | +6.915 |
Bow | +6.893 |
Neg logit contributions | |
ry | -5.126 |
ipples | -4.708 |
again | -4.698 |
andy | -4.676 |
boil | -4.417 |
Token | . |
Token ID | 13 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
hy | +4.177 |
Mic | +3.948 |
Red | +3.491 |
Roger | +3.460 |
Robert | +3.447 |
Neg logit contributions | |
ry | -2.150 |
ipples | -2.093 |
riage | -2.068 |
ac | -2.048 |
en | -2.024 |
Token | One |
Token ID | 1881 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
riage | +0.618 |
ipples | +0.608 |
ry | +0.592 |
flower | +0.588 |
ac | +0.580 |
Neg logit contributions | |
hy | -0.638 |
Mic | -0.572 |
rides | -0.502 |
Red | -0.498 |
Roger | -0.487 |
Token | day |
Token ID | 1110 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
riage | +1.648 |
ipples | +1.617 |
department | +1.545 |
ry | +1.533 |
ric | +1.525 |
Neg logit contributions | |
hy | -2.014 |
Mic | -1.776 |
rides | -1.610 |
Red | -1.576 |
Roger | -1.534 |
Token | , |
Token ID | 11 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
hy | +3.001 |
Mic | +2.763 |
Red | +2.445 |
rides | +2.397 |
nam | +2.358 |
Neg logit contributions | |
ry | -2.040 |
riage | -1.987 |
department | -1.890 |
flower | -1.887 |
ipples | -1.885 |
Token | play |
Token ID | 711 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
riage | +0.306 |
ipples | +0.299 |
ry | +0.291 |
flower | +0.289 |
ac | +0.283 |
Neg logit contributions | |
hy | -0.406 |
Mic | -0.367 |
rides | -0.328 |
Red | -0.324 |
Roger | -0.319 |
Token | with |
Token ID | 351 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
hy | +0.858 |
Mic | +0.747 |
Red | +0.632 |
rides | +0.631 |
Robert | +0.628 |
Neg logit contributions | |
riage | -0.842 |
ipples | -0.840 |
ry | -0.824 |
flower | -0.817 |
ac | -0.804 |
Token | his |
Token ID | 465 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
hy | +0.326 |
Mic | +0.293 |
rides | +0.258 |
Red | +0.258 |
Roger | +0.254 |
Neg logit contributions | |
riage | -0.243 |
ipples | -0.239 |
ry | -0.234 |
flower | -0.232 |
andy | -0.227 |
Token | toys |
Token ID | 14958 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
hy | +1.528 |
Mic | +1.340 |
Red | +1.198 |
rides | +1.191 |
Robert | +1.188 |
Neg logit contributions | |
flower | -1.155 |
riage | -1.144 |
ipples | -1.140 |
ry | -1.119 |
department | -1.113 |
Token | all |
Token ID | 477 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
hy | +1.627 |
Mic | +1.484 |
Roger | +1.294 |
Red | +1.293 |
Robert | +1.288 |
Neg logit contributions | |
riage | -1.060 |
ry | -1.053 |
ipples | -1.052 |
flower | -1.034 |
ac | -1.009 |
Token | day |
Token ID | 1110 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
hy | +0.110 |
Mic | +0.100 |
rides | +0.090 |
Red | +0.089 |
Roger | +0.088 |
Neg logit contributions | |
riage | -0.069 |
ipples | -0.067 |
ry | -0.065 |
flower | -0.065 |
ac | -0.063 |
Token | long |
Token ID | 890 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
hy | +8.636 |
Mic | +7.798 |
Red | +7.134 |
Blue | +6.915 |
Bow | +6.893 |
Neg logit contributions | |
ry | -5.126 |
ipples | -4.708 |
again | -4.698 |
andy | -4.676 |
boil | -4.417 |
Token | . |
Token ID | 13 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
hy | +4.177 |
Mic | +3.948 |
Red | +3.491 |
Roger | +3.460 |
Robert | +3.447 |
Neg logit contributions | |
ry | -2.150 |
ipples | -2.093 |
riage | -2.068 |
ac | -2.048 |
en | -2.024 |
Token | One |
Token ID | 1881 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
riage | +0.618 |
ipples | +0.608 |
ry | +0.592 |
flower | +0.588 |
ac | +0.580 |
Neg logit contributions | |
hy | -0.638 |
Mic | -0.572 |
rides | -0.502 |
Red | -0.498 |
Roger | -0.487 |
Token | day |
Token ID | 1110 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
riage | +1.648 |
ipples | +1.617 |
department | +1.545 |
ry | +1.533 |
ric | +1.525 |
Neg logit contributions | |
hy | -2.014 |
Mic | -1.776 |
rides | -1.610 |
Red | -1.576 |
Roger | -1.534 |
Token | , |
Token ID | 11 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
hy | +3.001 |
Mic | +2.763 |
Red | +2.445 |
rides | +2.397 |
nam | +2.358 |
Neg logit contributions | |
ry | -2.040 |
riage | -1.987 |
department | -1.890 |
flower | -1.887 |
ipples | -1.885 |
Token | illy |
Token ID | 6548 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
riage | +0.506 |
ipples | +0.489 |
ry | +0.468 |
flower | +0.466 |
department | +0.457 |
Neg logit contributions | |
hy | -0.927 |
Mic | -0.852 |
rides | -0.772 |
Red | -0.764 |
Roger | -0.754 |
Token | liked |
Token ID | 8288 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
hy | +1.199 |
Mic | +1.101 |
Red | +0.947 |
rides | +0.946 |
Roger | +0.942 |
Neg logit contributions | |
riage | -1.016 |
ry | -0.995 |
ipples | -0.994 |
flower | -0.992 |
andy | -0.973 |
Token | to |
Token ID | 284 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
hy | +0.663 |
Mic | +0.608 |
Red | +0.550 |
rides | +0.541 |
Roger | +0.540 |
Neg logit contributions | |
riage | -0.350 |
ipples | -0.342 |
flower | -0.332 |
ry | -0.327 |
department | -0.323 |
Token | play |
Token ID | 711 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
riage | +0.224 |
ipples | +0.219 |
ry | +0.212 |
flower | +0.211 |
ac | +0.207 |
Neg logit contributions | |
hy | -0.300 |
Mic | -0.272 |
rides | -0.243 |
Red | -0.240 |
Roger | -0.236 |
Token | all |
Token ID | 477 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
hy | +1.096 |
Mic | +0.946 |
Red | +0.812 |
Robert | +0.801 |
Roger | +0.801 |
Neg logit contributions | |
ipples | -1.063 |
riage | -1.061 |
ry | -1.048 |
flower | -1.036 |
andy | -1.025 |
Token | day |
Token ID | 1110 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
riage | +0.300 |
ipples | +0.292 |
ry | +0.283 |
flower | +0.281 |
department | +0.275 |
Neg logit contributions | |
hy | -0.457 |
Mic | -0.416 |
rides | -0.375 |
Red | -0.371 |
Roger | -0.364 |
Token | long |
Token ID | 890 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
hy | +8.842 |
Mic | +8.141 |
Red | +7.570 |
Blue | +7.225 |
Roger | +7.041 |
Neg logit contributions | |
ry | -4.993 |
andy | -4.610 |
again | -4.579 |
en | -4.550 |
ipples | -4.502 |
Token | . |
Token ID | 13 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
hy | +5.838 |
Mic | +5.486 |
Red | +4.912 |
Roger | +4.874 |
Robert | +4.868 |
Neg logit contributions | |
ry | -2.590 |
en | -2.537 |
again | -2.469 |
ipples | -2.450 |
andy | -2.368 |
Token | She |
Token ID | 1375 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
riage | +1.354 |
ipples | +1.339 |
ry | +1.300 |
flower | +1.294 |
department | +1.288 |
Neg logit contributions | |
hy | -1.130 |
Mic | -0.988 |
rides | -0.866 |
Red | -0.846 |
Roger | -0.816 |
Token | was |
Token ID | 373 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
riage | +0.029 |
ipples | +0.029 |
ry | +0.028 |
flower | +0.028 |
ac | +0.027 |
Neg logit contributions | |
hy | -0.039 |
Mic | -0.035 |
rides | -0.031 |
Red | -0.031 |
Roger | -0.031 |
Token | very |
Token ID | 845 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
riage | +0.748 |
ipples | +0.728 |
ry | +0.718 |
flower | +0.706 |
andy | +0.696 |
Neg logit contributions | |
hy | -0.693 |
Mic | -0.610 |
rides | -0.537 |
Red | -0.526 |
Roger | -0.514 |
Token |
|
Token ID | 198 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
hy | +2.212 |
Mic | +1.963 |
nam | +1.719 |
rides | +1.712 |
Red | +1.695 |
Neg logit contributions | |
riage | -1.482 |
ry | -1.452 |
ipples | -1.437 |
ucks | -1.417 |
andy | -1.407 |
Token |
|
Token ID | 198 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
hy | +0.312 |
Mic | +0.278 |
rides | +0.245 |
Red | +0.242 |
Roger | +0.238 |
Neg logit contributions | |
riage | -0.271 |
ipples | -0.266 |
ry | -0.262 |
flower | -0.256 |
andy | -0.254 |
Token | S |
Token ID | 50 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
riage | +0.396 |
ipples | +0.386 |
ry | +0.373 |
flower | +0.372 |
ac | +0.364 |
Neg logit contributions | |
hy | -0.561 |
Mic | -0.511 |
rides | -0.458 |
Red | -0.455 |
Roger | -0.446 |
Token | ara |
Token ID | 3301 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
riage | +1.261 |
ipples | +1.247 |
flower | +1.168 |
ac | +1.128 |
ry | +1.123 |
Neg logit contributions | |
hy | -2.862 |
Mic | -2.580 |
rides | -2.444 |
Red | -2.392 |
Roger | -2.317 |
Token | offered |
Token ID | 4438 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
riage | +1.371 |
ipples | +1.311 |
ry | +1.257 |
flower | +1.252 |
ric | +1.245 |
Neg logit contributions | |
hy | -1.359 |
Mic | -1.213 |
rides | -1.084 |
Red | -1.057 |
Roger | -1.022 |
Token | her |
Token ID | 607 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
riage | +2.589 |
ipples | +2.562 |
ry | +2.434 |
flower | +2.423 |
ric | +2.360 |
Neg logit contributions | |
hy | -3.221 |
Mic | -2.896 |
rides | -2.719 |
Bow | -2.621 |
Red | -2.579 |
Token | purple |
Token ID | 14032 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
riage | +9.835 |
ipples | +8.619 |
ry | +8.541 |
department | +8.514 |
pour | +8.470 |
Neg logit contributions | |
long | -6.397 |
hy | -6.179 |
brace | -5.970 |
rides | -5.878 |
curly | -5.407 |
Token | bubble |
Token ID | 14310 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
riage | +4.445 |
ipples | +3.967 |
department | +3.686 |
iz | +3.659 |
ac | +3.641 |
Neg logit contributions | |
hy | -5.874 |
Mic | -5.317 |
rides | -5.145 |
brace | -5.137 |
long | -5.023 |
Token | wand |
Token ID | 11569 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
riage | +2.995 |
ipples | +2.688 |
ric | +2.605 |
ry | +2.602 |
flower | +2.590 |
Neg logit contributions | |
hy | -3.237 |
Mic | -2.760 |
rides | -2.588 |
brace | -2.521 |
hell | -2.464 |
Token | to |
Token ID | 284 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
riage | +1.867 |
ipples | +1.797 |
andy | +1.718 |
flower | +1.717 |
ry | +1.716 |
Neg logit contributions | |
hy | -1.870 |
Mic | -1.586 |
rides | -1.457 |
Red | -1.391 |
Bow | -1.373 |
Token | Max |
Token ID | 5436 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
riage | +1.297 |
ipples | +1.259 |
ry | +1.226 |
flower | +1.209 |
ac | +1.194 |
Neg logit contributions | |
hy | -1.996 |
Mic | -1.824 |
rides | -1.661 |
Red | -1.641 |
Roger | -1.600 |
Token | butterfly |
Token ID | 35113 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
hy | +1.319 |
Mic | +1.188 |
Red | +1.072 |
rides | +1.048 |
Roger | +1.045 |
Neg logit contributions | |
riage | -0.908 |
flower | -0.883 |
ipples | -0.877 |
andy | -0.844 |
ac | -0.836 |
Token | flew |
Token ID | 13112 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
hy | +1.329 |
Mic | +1.228 |
Red | +1.083 |
nam | +1.060 |
Roger | +1.058 |
Neg logit contributions | |
riage | -1.035 |
ry | -1.013 |
ipples | -1.012 |
flower | -1.004 |
andy | -0.983 |
Token | away |
Token ID | 1497 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
riage | +0.963 |
ipples | +0.929 |
ry | +0.900 |
flower | +0.883 |
ac | +0.869 |
Neg logit contributions | |
hy | -1.530 |
Mic | -1.374 |
rides | -1.253 |
Red | -1.238 |
Roger | -1.227 |
Token | , |
Token ID | 11 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
riage | +0.330 |
ipples | +0.323 |
ry | +0.315 |
flower | +0.313 |
ac | +0.308 |
Neg logit contributions | |
hy | -0.342 |
Mic | -0.305 |
rides | -0.268 |
Red | -0.265 |
Roger | -0.260 |
Token | Joe |
Token ID | 5689 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
riage | +6.133 |
ucks | +6.047 |
ac | +5.962 |
ric | +5.930 |
flower | +5.913 |
Neg logit contributions | |
hy | -3.637 |
long | -3.136 |
rides | -2.986 |
Bow | -2.975 |
Robert | -2.935 |
Token | and |
Token ID | 290 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
hy | +0.349 |
Mic | +0.315 |
Red | +0.276 |
rides | +0.275 |
Roger | +0.270 |
Neg logit contributions | |
riage | -0.292 |
ipples | -0.288 |
ry | -0.283 |
flower | -0.282 |
andy | -0.275 |
Token | Linda |
Token ID | 21776 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
riage | +5.600 |
ipples | +5.487 |
ric | +5.435 |
flower | +5.158 |
department | +5.146 |
Neg logit contributions | |
hy | -8.356 |
Mic | -7.379 |
long | -7.308 |
Ellie | -7.163 |
Bow | -7.006 |
Token | decided |
Token ID | 3066 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
hy | +1.725 |
Mic | +1.619 |
Red | +1.420 |
Roger | +1.395 |
Robert | +1.371 |
Neg logit contributions | |
riage | -1.259 |
ipples | -1.232 |
ry | -1.221 |
flower | -1.210 |
andy | -1.179 |
Token | to |
Token ID | 284 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
riage | +0.672 |
ipples | +0.664 |
flower | +0.640 |
ry | +0.638 |
ac | +0.626 |
Neg logit contributions | |
hy | -1.040 |
Mic | -0.935 |
rides | -0.862 |
Red | -0.839 |
Roger | -0.833 |
Token | share |
Token ID | 2648 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
riage | +0.347 |
ipples | +0.338 |
ry | +0.328 |
flower | +0.325 |
ac | +0.318 |
Neg logit contributions | |
hy | -0.535 |
Mic | -0.487 |
rides | -0.439 |
Red | -0.434 |
Roger | -0.427 |
Token | a |
Token ID | 257 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
riage | +0.358 |
ipples | +0.348 |
ry | +0.336 |
flower | +0.333 |
department | +0.325 |
Neg logit contributions | |
hy | -0.658 |
Mic | -0.603 |
rides | -0.550 |
Red | -0.543 |
Roger | -0.535 |
Token | share |
Token ID | 2648 |
Feature activation | -2.83e-01 |
Pos logit contributions | |
riage | +0.347 |
ipples | +0.338 |
ry | +0.328 |
flower | +0.325 |
ac | +0.318 |
Neg logit contributions | |
hy | -0.535 |
Mic | -0.487 |
rides | -0.439 |
Red | -0.434 |
Roger | -0.427 |
Token | a |
Token ID | 257 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
riage | +0.358 |
ipples | +0.348 |
ry | +0.336 |
flower | +0.333 |
department | +0.325 |
Neg logit contributions | |
hy | -0.658 |
Mic | -0.603 |
rides | -0.550 |
Red | -0.543 |
Roger | -0.535 |
Token | hug |
Token ID | 16225 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
hy | +1.270 |
Mic | +1.120 |
Red | +0.990 |
rides | +0.983 |
Roger | +0.965 |
Neg logit contributions | |
riage | -1.135 |
ipples | -1.119 |
ry | -1.113 |
flower | -1.110 |
ac | -1.100 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
riage | +1.999 |
ipples | +1.990 |
andy | +1.908 |
ry | +1.894 |
ric | +1.878 |
Neg logit contributions | |
hy | -2.170 |
Mic | -1.853 |
rides | -1.719 |
Red | -1.638 |
Bow | -1.617 |
Token | Joe |
Token ID | 5689 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
riage | +4.894 |
flower | +4.818 |
ac | +4.722 |
ipples | +4.637 |
ric | +4.496 |
Neg logit contributions | |
hy | -4.184 |
Mic | -3.518 |
rides | -3.317 |
Robert | -3.207 |
Bow | -3.140 |
Token | and |
Token ID | 290 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
hy | +0.584 |
Mic | +0.528 |
Red | +0.464 |
rides | +0.461 |
Roger | +0.456 |
Neg logit contributions | |
riage | -0.462 |
ipples | -0.455 |
ry | -0.450 |
flower | -0.449 |
andy | -0.436 |
Token | Linda |
Token ID | 21776 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
riage | +5.590 |
ipples | +5.391 |
ric | +5.288 |
flower | +5.035 |
department | +4.981 |
Neg logit contributions | |
hy | -8.342 |
Mic | -7.548 |
long | -7.338 |
Ellie | -7.168 |
brace | -7.120 |
Token | closed |
Token ID | 4838 |
Feature activation | -8.20e-01 |
Pos logit contributions | |
hy | +1.253 |
Mic | +1.153 |
Red | +1.012 |
Roger | +1.002 |
rides | +0.985 |
Neg logit contributions | |
riage | -0.954 |
ipples | -0.930 |
ry | -0.920 |
flower | -0.915 |
andy | -0.896 |
Token | their |
Token ID | 511 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
riage | +1.029 |
ipples | +1.005 |
flower | +0.953 |
ry | +0.941 |
ac | +0.927 |
Neg logit contributions | |
hy | -1.937 |
Mic | -1.752 |
rides | -1.637 |
Roger | -1.588 |
Red | -1.584 |
Token | eyes |
Token ID | 2951 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
hy | +0.253 |
Mic | +0.234 |
Red | +0.213 |
rides | +0.212 |
Roger | +0.210 |
Neg logit contributions | |
riage | -0.100 |
ipples | -0.096 |
ry | -0.093 |
flower | -0.092 |
ac | -0.090 |
Token | and |
Token ID | 290 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
hy | +1.777 |
Mic | +1.680 |
Red | +1.463 |
Roger | +1.445 |
rides | +1.436 |
Neg logit contributions | |
ry | -1.256 |
ipples | -1.243 |
riage | -1.241 |
flower | -1.211 |
department | -1.199 |
Token | on |
Token ID | 319 |
Feature activation | -3.03e-01 |
Pos logit contributions | |
riage | +2.044 |
ipples | +2.036 |
ry | +1.932 |
andy | +1.919 |
ac | +1.911 |
Neg logit contributions | |
hy | -1.813 |
Mic | -1.599 |
rides | -1.451 |
Red | -1.388 |
Roger | -1.368 |
Token | , |
Token ID | 11 |
Feature activation | -6.14e-01 |
Pos logit contributions | |
riage | +0.547 |
ipples | +0.539 |
ry | +0.522 |
flower | +0.521 |
ac | +0.512 |
Neg logit contributions | |
hy | -0.545 |
Mic | -0.484 |
rides | -0.427 |
Red | -0.419 |
Roger | -0.412 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
riage | +1.051 |
ipples | +1.033 |
ry | +1.003 |
flower | +0.999 |
ac | +0.977 |
Neg logit contributions | |
hy | -1.158 |
Mic | -1.034 |
rides | -0.923 |
Red | -0.911 |
Roger | -0.888 |
Token | my |
Token ID | 1820 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
ipples | +1.411 |
riage | +1.393 |
ac | +1.295 |
ry | +1.288 |
andy | +1.264 |
Neg logit contributions | |
hy | -2.579 |
Mic | -2.334 |
rides | -2.149 |
Red | -2.106 |
Roger | -2.048 |
Token | ! |
Token ID | 0 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
hy | +0.664 |
Mic | +0.603 |
Red | +0.527 |
rides | +0.525 |
Roger | +0.517 |
Neg logit contributions | |
riage | -0.566 |
ry | -0.547 |
ipples | -0.545 |
flower | -0.535 |
department | -0.523 |
Token | Hur |
Token ID | 10077 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
ipples | +2.314 |
riage | +2.271 |
andy | +2.233 |
ry | +2.224 |
flower | +2.212 |
Neg logit contributions | |
hy | -2.223 |
Mic | -1.897 |
Red | -1.696 |
rides | -1.663 |
Bow | -1.618 |
Token | ry |
Token ID | 563 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
ipples | +4.948 |
ac | +4.155 |
ull | +3.992 |
andy | +3.750 |
ainer | +3.695 |
Neg logit contributions | |
hy | -9.593 |
rides | -7.860 |
ve | -7.591 |
Mic | -7.561 |
Red | -7.514 |
Token | up |
Token ID | 510 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
ipples | +3.622 |
riage | +3.606 |
flower | +3.474 |
ac | +3.409 |
ric | +3.324 |
Neg logit contributions | |
hy | -3.575 |
rides | -3.077 |
Mic | -2.982 |
Red | -2.729 |
Robert | -2.725 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
riage | +0.256 |
ipples | +0.251 |
ry | +0.244 |
flower | +0.243 |
ac | +0.238 |
Neg logit contributions | |
hy | -0.289 |
Mic | -0.260 |
rides | -0.229 |
Red | -0.227 |
Roger | -0.222 |
Token | he |
Token ID | 339 |
Feature activation | +3.67e-01 |
Pos logit contributions | |
hy | +4.090 |
Mic | +3.900 |
Red | +3.384 |
Robert | +3.380 |
Roger | +3.342 |
Neg logit contributions | |
ry | -2.731 |
ipples | -2.599 |
flower | -2.570 |
riage | -2.527 |
M | -2.515 |
Token | said |
Token ID | 531 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
hy | +0.603 |
Mic | +0.538 |
Red | +0.455 |
rides | +0.449 |
Roger | +0.443 |
Neg logit contributions | |
riage | -0.704 |
ipples | -0.697 |
ry | -0.689 |
flower | -0.677 |
department | -0.673 |
Token | tea |
Token ID | 8887 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
riage | +2.909 |
ipples | +2.866 |
ull | +2.715 |
ry | +2.680 |
flower | +2.679 |
Neg logit contributions | |
hy | -3.839 |
rides | -3.267 |
Mic | -3.257 |
nam | -3.066 |
brace | -3.025 |
Token | and |
Token ID | 290 |
Feature activation | -7.83e-01 |
Pos logit contributions | |
riage | +1.740 |
ipples | +1.652 |
ry | +1.586 |
ric | +1.584 |
flower | +1.572 |
Neg logit contributions | |
hy | -1.908 |
Mic | -1.691 |
rides | -1.547 |
Red | -1.467 |
brace | -1.430 |
Token | cookies |
Token ID | 14746 |
Feature activation | -4.97e-01 |
Pos logit contributions | |
riage | +1.170 |
ipples | +1.140 |
ry | +1.108 |
flower | +1.100 |
andy | +1.081 |
Neg logit contributions | |
hy | -1.638 |
Mic | -1.469 |
rides | -1.369 |
Red | -1.304 |
nam | -1.302 |
Token | to |
Token ID | 284 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
riage | +0.828 |
ipples | +0.803 |
ry | +0.776 |
flower | +0.773 |
andy | +0.767 |
Neg logit contributions | |
hy | -0.959 |
Mic | -0.853 |
rides | -0.768 |
Red | -0.750 |
Roger | -0.734 |
Token | the |
Token ID | 262 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
riage | +0.987 |
ipples | +0.964 |
flower | +0.936 |
ry | +0.935 |
department | +0.909 |
Neg logit contributions | |
hy | -1.658 |
Mic | -1.518 |
rides | -1.390 |
Red | -1.374 |
nam | -1.342 |
Token | living |
Token ID | 2877 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
riage | +1.823 |
ipples | +1.762 |
ry | +1.719 |
ac | +1.713 |
flower | +1.686 |
Neg logit contributions | |
hy | -2.793 |
Mic | -2.473 |
rides | -2.337 |
Red | -2.270 |
Roger | -2.182 |
Token | room |
Token ID | 2119 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
riage | +6.517 |
department | +6.303 |
flower | +6.083 |
ipples | +6.050 |
andy | +6.023 |
Neg logit contributions | |
hy | -6.267 |
rides | -5.861 |
nam | -5.274 |
wra | -5.232 |
long | -5.163 |
Token | . |
Token ID | 13 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
riage | +2.257 |
ipples | +2.186 |
flower | +2.138 |
andy | +2.137 |
ric | +2.095 |
Neg logit contributions | |
hy | -2.315 |
Mic | -1.881 |
rides | -1.844 |
Red | -1.686 |
nam | -1.672 |
Token | They |
Token ID | 1119 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
riage | +1.419 |
ipples | +1.393 |
flower | +1.374 |
ry | +1.359 |
ac | +1.343 |
Neg logit contributions | |
hy | -1.714 |
Mic | -1.526 |
rides | -1.387 |
Red | -1.381 |
Bow | -1.324 |
Token | sit |
Token ID | 1650 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
riage | +0.511 |
ipples | +0.498 |
ry | +0.484 |
flower | +0.481 |
ac | +0.471 |
Neg logit contributions | |
hy | -0.703 |
Mic | -0.632 |
rides | -0.573 |
Red | -0.562 |
Roger | -0.550 |
Token | on |
Token ID | 319 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
riage | +4.312 |
ipples | +4.185 |
andy | +4.108 |
department | +4.069 |
ric | +4.009 |
Neg logit contributions | |
hy | -2.953 |
Mic | -2.534 |
rides | -2.350 |
nam | -2.292 |
Red | -2.249 |
Token | They |
Token ID | 2990 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
hy | +0.312 |
Mic | +0.282 |
rides | +0.253 |
Red | +0.250 |
Roger | +0.246 |
Neg logit contributions | |
riage | -0.209 |
ipples | -0.206 |
ry | -0.200 |
flower | -0.198 |
andy | -0.195 |
Token | go |
Token ID | 467 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
riage | +0.293 |
ipples | +0.286 |
ry | +0.278 |
flower | +0.276 |
ac | +0.271 |
Neg logit contributions | |
hy | -0.398 |
Mic | -0.360 |
rides | -0.323 |
Red | -0.319 |
Roger | -0.313 |
Token | back |
Token ID | 736 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
riage | +2.160 |
ipples | +2.114 |
andy | +2.030 |
flower | +2.015 |
ac | +1.993 |
Neg logit contributions | |
hy | -2.890 |
Mic | -2.520 |
rides | -2.466 |
Red | -2.284 |
Robert | -2.256 |
Token | to |
Token ID | 284 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
riage | +1.435 |
ipples | +1.403 |
flower | +1.332 |
andy | +1.323 |
ac | +1.315 |
Neg logit contributions | |
hy | -1.922 |
Mic | -1.722 |
rides | -1.622 |
Red | -1.529 |
Roger | -1.502 |
Token | the |
Token ID | 262 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
ipples | +2.097 |
riage | +2.096 |
ac | +2.052 |
ry | +2.000 |
flower | +1.998 |
Neg logit contributions | |
hy | -3.084 |
Mic | -2.797 |
rides | -2.634 |
Robert | -2.546 |
nam | -2.516 |
Token | living |
Token ID | 2877 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
riage | +3.534 |
ipples | +3.340 |
ac | +3.327 |
flower | +3.220 |
ry | +3.158 |
Neg logit contributions | |
hy | -4.353 |
Mic | -3.992 |
rides | -3.840 |
Red | -3.581 |
long | -3.573 |
Token | room |
Token ID | 2119 |
Feature activation | -4.92e-01 |
Pos logit contributions | |
riage | +6.752 |
department | +6.429 |
andy | +6.176 |
flower | +6.114 |
ric | +6.080 |
Neg logit contributions | |
hy | -5.753 |
rides | -5.641 |
nam | -5.412 |
Mic | -5.095 |
long | -4.989 |
Token | and |
Token ID | 290 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
riage | +0.843 |
ipples | +0.822 |
flower | +0.798 |
ry | +0.790 |
andy | +0.785 |
Neg logit contributions | |
hy | -0.927 |
Mic | -0.811 |
rides | -0.738 |
Red | -0.713 |
Roger | -0.698 |
Token | pick |
Token ID | 2298 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
riage | +0.654 |
ipples | +0.644 |
flower | +0.621 |
ry | +0.619 |
ac | +0.610 |
Neg logit contributions | |
hy | -0.893 |
Mic | -0.798 |
rides | -0.731 |
Red | -0.710 |
Roger | -0.697 |
Token | up |
Token ID | 510 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
riage | +2.082 |
ipples | +1.976 |
flower | +1.913 |
department | +1.898 |
ric | +1.893 |
Neg logit contributions | |
hy | -2.699 |
Mic | -2.434 |
rides | -2.342 |
Red | -2.215 |
Robert | -2.173 |
Token | the |
Token ID | 262 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
riage | +1.375 |
ipples | +1.332 |
flower | +1.287 |
ry | +1.264 |
ac | +1.246 |
Neg logit contributions | |
hy | -2.394 |
Mic | -2.139 |
rides | -2.053 |
Red | -1.952 |
nam | -1.930 |
Token | to |
Token ID | 284 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
riage | +1.092 |
ipples | +1.071 |
flower | +1.037 |
ry | +1.036 |
andy | +1.013 |
Neg logit contributions | |
hy | -1.615 |
Mic | -1.453 |
rides | -1.345 |
Red | -1.298 |
Roger | -1.290 |
Token | bat |
Token ID | 7365 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
riage | +0.056 |
ipples | +0.054 |
ry | +0.052 |
flower | +0.052 |
ac | +0.051 |
Neg logit contributions | |
hy | -0.102 |
Mic | -0.093 |
rides | -0.084 |
Red | -0.084 |
Roger | -0.082 |
Token | he |
Token ID | 258 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
riage | +0.076 |
ipples | +0.075 |
ry | +0.074 |
flower | +0.074 |
ac | +0.073 |
Neg logit contributions | |
hy | -0.027 |
Mic | -0.022 |
rides | -0.016 |
Red | -0.015 |
Roger | -0.015 |
Token | first |
Token ID | 717 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
riage | +0.027 |
ipples | +0.027 |
ry | +0.026 |
flower | +0.026 |
ac | +0.025 |
Neg logit contributions | |
hy | -0.034 |
Mic | -0.031 |
rides | -0.027 |
Red | -0.027 |
Roger | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
riage | +0.944 |
ipples | +0.929 |
flower | +0.900 |
ry | +0.889 |
department | +0.882 |
Neg logit contributions | |
hy | -1.131 |
Mic | -1.008 |
rides | -0.913 |
Red | -0.894 |
Roger | -0.878 |
Token | Hur |
Token ID | 10077 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
riage | +1.151 |
ipples | +1.147 |
flower | +1.115 |
ry | +1.109 |
department | +1.091 |
Neg logit contributions | |
hy | -1.228 |
Mic | -1.089 |
rides | -0.960 |
Red | -0.958 |
Roger | -0.926 |
Token | ry |
Token ID | 563 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ipples | +3.454 |
ac | +2.942 |
ull | +2.689 |
flower | +2.637 |
andy | +2.544 |
Neg logit contributions | |
hy | -10.097 |
rides | -8.686 |
cases | -8.510 |
Mic | -8.404 |
ve | -8.284 |
Token | up |
Token ID | 510 |
Feature activation | -6.71e-01 |
Pos logit contributions | |
riage | +3.720 |
ipples | +3.693 |
flower | +3.564 |
ac | +3.446 |
ric | +3.412 |
Neg logit contributions | |
hy | -3.537 |
Mic | -2.999 |
rides | -2.973 |
Red | -2.708 |
Robert | -2.700 |
Token | , |
Token ID | 11 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
riage | +1.108 |
ipples | +1.091 |
flower | +1.068 |
ry | +1.039 |
department | +1.033 |
Neg logit contributions | |
hy | -1.282 |
Mic | -1.147 |
rides | -1.039 |
Red | -1.010 |
Roger | -1.000 |
Token | or |
Token ID | 393 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
riage | +0.497 |
ipples | +0.485 |
ry | +0.471 |
flower | +0.470 |
department | +0.459 |
Neg logit contributions | |
hy | -0.654 |
Mic | -0.587 |
rides | -0.529 |
Red | -0.521 |
Roger | -0.512 |
Token | the |
Token ID | 262 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
hy | +0.260 |
Mic | +0.236 |
rides | +0.209 |
Red | +0.208 |
Roger | +0.204 |
Neg logit contributions | |
riage | -0.196 |
ipples | -0.189 |
ry | -0.186 |
flower | -0.183 |
ac | -0.180 |
Token | Bob |
Token ID | 18861 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
hy | +2.581 |
Mic | +2.241 |
nam | +2.117 |
rides | +2.069 |
brace | +2.019 |
Neg logit contributions | |
riage | -1.752 |
ipples | -1.706 |
flower | -1.670 |
ry | -1.655 |
andy | -1.630 |
Token | said |
Token ID | 531 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
riage | +0.915 |
ipples | +0.892 |
ry | +0.861 |
flower | +0.857 |
ac | +0.849 |
Neg logit contributions | |
hy | -0.879 |
Mic | -0.784 |
rides | -0.688 |
Red | -0.674 |
Roger | -0.662 |
Token | , |
Token ID | 11 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
hy | +0.176 |
Mic | +0.159 |
rides | +0.140 |
Red | +0.139 |
Roger | +0.136 |
Neg logit contributions | |
riage | -0.157 |
ipples | -0.153 |
ry | -0.151 |
flower | -0.149 |
department | -0.147 |
Token | " |
Token ID | 366 |
Feature activation | +6.21e-02 |
Pos logit contributions | |
hy | +0.451 |
Mic | +0.412 |
rides | +0.377 |
Red | +0.372 |
Roger | +0.363 |
Neg logit contributions | |
riage | -0.239 |
ipples | -0.228 |
ry | -0.224 |
flower | -0.223 |
department | -0.219 |
Token | Mr |
Token ID | 5246 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
hy | +0.121 |
Mic | +0.108 |
rides | +0.097 |
Red | +0.094 |
Roger | +0.092 |
Neg logit contributions | |
riage | -0.103 |
ipples | -0.101 |
flower | -0.099 |
ry | -0.098 |
ac | -0.096 |
Token | . |
Token ID | 13 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
riage | +1.788 |
ipples | +1.732 |
ry | +1.678 |
ac | +1.669 |
flower | +1.650 |
Neg logit contributions | |
hy | -2.036 |
Mic | -1.793 |
rides | -1.581 |
Red | -1.565 |
Bow | -1.531 |
Token | Ant |
Token ID | 3738 |
Feature activation | -4.94e-01 |
Pos logit contributions | |
riage | +5.020 |
ipples | +4.745 |
ucks | +4.460 |
ac | +4.401 |
en | +4.345 |
Neg logit contributions | |
hy | -8.095 |
Mic | -7.638 |
Stretch | -6.722 |
Bow | -6.661 |
long | -6.532 |
Token | , |
Token ID | 11 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
riage | +0.851 |
ipples | +0.830 |
flower | +0.800 |
ry | +0.798 |
ac | +0.789 |
Neg logit contributions | |
hy | -0.925 |
Mic | -0.817 |
rides | -0.731 |
Red | -0.714 |
Roger | -0.700 |
Token | why |
Token ID | 1521 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
riage | +0.726 |
ipples | +0.715 |
ry | +0.689 |
flower | +0.686 |
ac | +0.673 |
Neg logit contributions | |
hy | -0.940 |
Mic | -0.842 |
rides | -0.757 |
Red | -0.743 |
Roger | -0.734 |
Token | is |
Token ID | 318 |
Feature activation | +2.72e-01 |
Pos logit contributions | |
riage | +0.368 |
ipples | +0.359 |
ry | +0.347 |
flower | +0.345 |
ac | +0.338 |
Neg logit contributions | |
hy | -0.550 |
Mic | -0.498 |
rides | -0.450 |
Red | -0.444 |
Roger | -0.437 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
hy | +0.568 |
Mic | +0.517 |
rides | +0.460 |
Red | +0.456 |
Roger | +0.445 |
Neg logit contributions | |
riage | -0.408 |
ipples | -0.394 |
ry | -0.382 |
flower | -0.381 |
department | -0.375 |
Token | surrender |
Token ID | 16908 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
riage | +0.187 |
ipples | +0.182 |
ry | +0.176 |
flower | +0.174 |
ac | +0.170 |
Neg logit contributions | |
hy | -0.330 |
Mic | -0.301 |
rides | -0.273 |
Red | -0.270 |
Roger | -0.266 |
Token | if |
Token ID | 611 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
riage | +0.557 |
ipples | +0.544 |
flower | +0.527 |
ry | +0.525 |
department | +0.515 |
Neg logit contributions | |
hy | -0.670 |
Mic | -0.602 |
rides | -0.537 |
Red | -0.529 |
Roger | -0.521 |
Token | it |
Token ID | 340 |
Feature activation | -5.30e-02 |
Pos logit contributions | |
riage | +0.655 |
ipples | +0.646 |
ry | +0.631 |
flower | +0.631 |
ac | +0.620 |
Neg logit contributions | |
hy | -0.442 |
Mic | -0.381 |
rides | -0.324 |
Red | -0.316 |
Roger | -0.309 |
Token | is |
Token ID | 318 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
riage | +0.067 |
ipples | +0.065 |
ry | +0.063 |
flower | +0.062 |
ac | +0.061 |
Neg logit contributions | |
hy | -0.123 |
Mic | -0.113 |
rides | -0.102 |
Red | -0.102 |
Roger | -0.100 |
Token | the |
Token ID | 262 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
riage | +0.383 |
ipples | +0.374 |
ry | +0.365 |
flower | +0.362 |
andy | +0.355 |
Neg logit contributions | |
hy | -0.451 |
Mic | -0.405 |
rides | -0.360 |
Red | -0.356 |
Roger | -0.350 |
Token | right |
Token ID | 826 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
ry | +2.502 |
riage | +2.477 |
ipples | +2.453 |
ucks | +2.375 |
flower | +2.352 |
Neg logit contributions | |
hy | -3.304 |
Mic | -2.892 |
rides | -2.696 |
Red | -2.579 |
nam | -2.529 |
Token | thing |
Token ID | 1517 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
ipples | +5.434 |
andy | +5.305 |
riage | +5.246 |
ric | +5.209 |
ucks | +5.095 |
Neg logit contributions | |
hy | -7.290 |
rides | -6.140 |
long | -5.990 |
Robert | -5.685 |
than | -5.685 |
Token | to |
Token ID | 284 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
hy | +0.901 |
Mic | +0.834 |
Red | +0.738 |
rides | +0.734 |
Roger | +0.723 |
Neg logit contributions | |
riage | -0.589 |
ry | -0.571 |
ipples | -0.568 |
flower | -0.556 |
department | -0.546 |
Token | do |
Token ID | 466 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
hy | +1.793 |
Mic | +1.647 |
Red | +1.464 |
rides | +1.442 |
nam | +1.435 |
Neg logit contributions | |
riage | -1.092 |
ipples | -1.048 |
ry | -1.000 |
flower | -0.999 |
andy | -0.983 |
Token | , |
Token ID | 11 |
Feature activation | -7.36e-01 |
Pos logit contributions | |
hy | +0.706 |
Mic | +0.647 |
Red | +0.564 |
rides | +0.559 |
Roger | +0.552 |
Neg logit contributions | |
riage | -0.515 |
ipples | -0.499 |
ry | -0.493 |
ac | -0.478 |
flower | -0.474 |
Token | without |
Token ID | 1231 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
riage | +1.281 |
ipples | +1.248 |
ry | +1.214 |
flower | +1.211 |
andy | +1.189 |
Neg logit contributions | |
hy | -1.359 |
Mic | -1.210 |
rides | -1.084 |
Red | -1.060 |
Roger | -1.044 |
Token | with |
Token ID | 351 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
riage | +2.335 |
ipples | +2.228 |
flower | +2.183 |
ry | +2.174 |
andy | +2.174 |
Neg logit contributions | |
hy | -1.682 |
Mic | -1.485 |
rides | -1.355 |
Red | -1.292 |
Roger | -1.230 |
Token | the |
Token ID | 262 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
riage | +1.520 |
ipples | +1.442 |
ry | +1.398 |
flower | +1.390 |
andy | +1.373 |
Neg logit contributions | |
hy | -2.334 |
Mic | -2.138 |
rides | -2.001 |
nam | -1.951 |
Red | -1.944 |
Token | ball |
Token ID | 2613 |
Feature activation | -9.03e-01 |
Pos logit contributions | |
riage | +0.667 |
ipples | +0.647 |
ry | +0.635 |
flower | +0.624 |
ac | +0.616 |
Neg logit contributions | |
hy | -0.777 |
Mic | -0.697 |
rides | -0.623 |
Red | -0.616 |
Roger | -0.601 |
Token | in |
Token ID | 287 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
riage | +1.630 |
ipples | +1.561 |
ry | +1.504 |
flower | +1.504 |
andy | +1.497 |
Neg logit contributions | |
hy | -1.627 |
Mic | -1.443 |
rides | -1.309 |
Red | -1.265 |
Roger | -1.213 |
Token | the |
Token ID | 262 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
riage | +2.451 |
ipples | +2.361 |
department | +2.322 |
flower | +2.297 |
ry | +2.294 |
Neg logit contributions | |
hy | -3.314 |
Mic | -3.046 |
rides | -2.816 |
Red | -2.768 |
nam | -2.712 |
Token | living |
Token ID | 2877 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
riage | +3.007 |
ipples | +2.890 |
ry | +2.874 |
ull | +2.742 |
ac | +2.736 |
Neg logit contributions | |
hy | -3.433 |
Mic | -3.124 |
rides | -2.863 |
Red | -2.818 |
Blue | -2.623 |
Token | room |
Token ID | 2119 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
riage | +6.597 |
department | +6.310 |
flower | +6.176 |
ipples | +6.107 |
andy | +5.908 |
Neg logit contributions | |
hy | -6.257 |
rides | -5.770 |
Mic | -5.228 |
nam | -5.096 |
long | -5.037 |
Token | . |
Token ID | 13 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
riage | +2.924 |
ipples | +2.851 |
flower | +2.808 |
ric | +2.804 |
andy | +2.779 |
Neg logit contributions | |
hy | -2.801 |
rides | -2.255 |
Mic | -2.247 |
Red | -2.034 |
nam | -2.002 |
Token | They |
Token ID | 1119 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
riage | +3.436 |
ac | +3.392 |
ry | +3.323 |
ipples | +3.301 |
flower | +3.277 |
Neg logit contributions | |
hy | -3.212 |
Mic | -2.705 |
rides | -2.507 |
Red | -2.452 |
Robert | -2.387 |
Token | forget |
Token ID | 6044 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
hy | +0.567 |
Mic | +0.516 |
Red | +0.452 |
rides | +0.450 |
Roger | +0.449 |
Neg logit contributions | |
riage | -0.434 |
ipples | -0.427 |
ry | -0.413 |
flower | -0.412 |
andy | -0.402 |
Token | what |
Token ID | 644 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ipples | +3.127 |
flower | +3.058 |
riage | +3.035 |
ac | +2.929 |
andy | +2.909 |
Neg logit contributions | |
hy | -2.956 |
Mic | -2.512 |
rides | -2.501 |
Bow | -2.349 |
Robert | -2.331 |
Token |
|
Token ID | 198 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
riage | +0.565 |
ipples | +0.549 |
flower | +0.535 |
ry | +0.534 |
department | +0.523 |
Neg logit contributions | |
hy | -0.712 |
Mic | -0.643 |
rides | -0.575 |
Red | -0.568 |
Roger | -0.558 |
Token | Mom |
Token ID | 29252 |
Feature activation | -9.86e-01 |
Pos logit contributions | |
riage | +0.856 |
ipples | +0.825 |
flower | +0.805 |
ry | +0.803 |
department | +0.793 |
Neg logit contributions | |
hy | -1.172 |
Mic | -1.073 |
Red | -0.959 |
rides | -0.958 |
Roger | -0.939 |
Token | comes |
Token ID | 2058 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
riage | +1.665 |
ipples | +1.619 |
flower | +1.554 |
ac | +1.548 |
ry | +1.545 |
Neg logit contributions | |
hy | -1.893 |
Mic | -1.661 |
rides | -1.588 |
Red | -1.470 |
Robert | -1.442 |
Token | to |
Token ID | 284 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
riage | +2.508 |
ipples | +2.349 |
flower | +2.256 |
ric | +2.226 |
andy | +2.217 |
Neg logit contributions | |
hy | -3.234 |
Mic | -2.922 |
rides | -2.746 |
Roger | -2.651 |
Robert | -2.647 |
Token | the |
Token ID | 262 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
ipples | +1.075 |
riage | +1.068 |
ry | +1.034 |
flower | +1.027 |
ac | +1.011 |
Neg logit contributions | |
hy | -1.701 |
Mic | -1.553 |
rides | -1.424 |
Red | -1.394 |
Roger | -1.382 |
Token | living |
Token ID | 2877 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
riage | +4.426 |
ipples | +4.283 |
ac | +4.239 |
flower | +4.151 |
andy | +4.111 |
Neg logit contributions | |
hy | -4.623 |
rides | -4.066 |
Mic | -4.053 |
Red | -3.830 |
long | -3.601 |
Token | room |
Token ID | 2119 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
riage | +6.583 |
department | +6.188 |
andy | +6.042 |
flower | +6.028 |
ipples | +5.996 |
Neg logit contributions | |
hy | -5.801 |
rides | -5.640 |
nam | -5.209 |
Mic | -5.114 |
long | -4.886 |
Token | and |
Token ID | 290 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
riage | +0.890 |
ipples | +0.873 |
flower | +0.844 |
ry | +0.836 |
andy | +0.830 |
Neg logit contributions | |
hy | -0.972 |
Mic | -0.848 |
rides | -0.774 |
Red | -0.747 |
Roger | -0.731 |
Token | sees |
Token ID | 7224 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
riage | +0.486 |
ipples | +0.478 |
flower | +0.460 |
ry | +0.460 |
ac | +0.452 |
Neg logit contributions | |
hy | -0.649 |
Mic | -0.584 |
rides | -0.530 |
Red | -0.519 |
Roger | -0.509 |
Token | their |
Token ID | 511 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
riage | +1.128 |
ipples | +1.095 |
flower | +1.066 |
ry | +1.040 |
ac | +1.040 |
Neg logit contributions | |
hy | -1.906 |
Mic | -1.713 |
rides | -1.619 |
Red | -1.570 |
Roger | -1.535 |
Token | bridge |
Token ID | 7696 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
riage | +4.076 |
flower | +3.793 |
ipples | +3.775 |
ry | +3.758 |
ac | +3.702 |
Neg logit contributions | |
hy | -4.052 |
rides | -3.392 |
Mic | -3.252 |
long | -3.101 |
brace | -2.958 |
Token | and |
Token ID | 290 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
riage | +1.170 |
ipples | +1.133 |
flower | +1.102 |
ry | +1.086 |
ac | +1.079 |
Neg logit contributions | |
hy | -1.258 |
Mic | -1.102 |
rides | -1.008 |
Red | -0.971 |
Roger | -0.942 |
Token | sad |
Token ID | 6507 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
hy | +1.015 |
Mic | +0.948 |
Red | +0.807 |
Roger | +0.803 |
Robert | +0.797 |
Neg logit contributions | |
riage | -0.966 |
ipples | -0.953 |
ry | -0.931 |
flower | -0.920 |
department | -0.906 |
Token | . |
Token ID | 13 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
riage | +1.249 |
ipples | +1.217 |
flower | +1.188 |
ac | +1.167 |
ry | +1.160 |
Neg logit contributions | |
hy | -1.384 |
Mic | -1.212 |
rides | -1.109 |
Red | -1.077 |
Roger | -1.034 |
Token | She |
Token ID | 1375 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
riage | +0.136 |
ipples | +0.134 |
ry | +0.131 |
flower | +0.131 |
andy | +0.129 |
Neg logit contributions | |
hy | -0.095 |
Mic | -0.083 |
rides | -0.070 |
Red | -0.069 |
Roger | -0.067 |
Token | wants |
Token ID | 3382 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
hy | +1.674 |
Mic | +1.554 |
Roger | +1.376 |
Red | +1.369 |
Robert | +1.367 |
Neg logit contributions | |
riage | -0.993 |
ipples | -0.983 |
ry | -0.943 |
andy | -0.936 |
ric | -0.915 |
Token | her |
Token ID | 607 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
riage | +2.136 |
ipples | +2.073 |
ry | +2.025 |
flower | +2.018 |
ac | +1.963 |
Neg logit contributions | |
hy | -2.648 |
Mic | -2.364 |
rides | -2.204 |
Bow | -2.096 |
Red | -2.087 |
Token | tool |
Token ID | 2891 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
riage | +6.148 |
ric | +5.755 |
ipples | +5.397 |
ucks | +5.248 |
department | +5.104 |
Neg logit contributions | |
hy | -6.707 |
long | -6.651 |
Mic | -6.222 |
rides | -6.150 |
powers | -5.881 |
Token | back |
Token ID | 736 |
Feature activation | -8.09e-01 |
Pos logit contributions | |
riage | +2.016 |
ipples | +1.891 |
ry | +1.838 |
ric | +1.820 |
andy | +1.794 |
Neg logit contributions | |
hy | -2.675 |
Mic | -2.398 |
rides | -2.207 |
Red | -2.112 |
brace | -2.066 |
Token | . |
Token ID | 13 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
riage | +1.396 |
ipples | +1.369 |
flower | +1.306 |
ry | +1.291 |
andy | +1.289 |
Neg logit contributions | |
hy | -1.515 |
Mic | -1.338 |
rides | -1.212 |
Red | -1.171 |
Roger | -1.151 |
Token | She |
Token ID | 1375 |
Feature activation | +4.02e-01 |
Pos logit contributions | |
hy | +0.743 |
Mic | +0.653 |
rides | +0.559 |
Red | +0.545 |
Roger | +0.543 |
Neg logit contributions | |
riage | -0.931 |
ipples | -0.908 |
ry | -0.892 |
andy | -0.885 |
flower | -0.885 |
Token | dives |
Token ID | 46638 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
hy | +0.860 |
Mic | +0.786 |
Red | +0.693 |
Roger | +0.690 |
Robert | +0.683 |
Neg logit contributions | |
riage | -0.581 |
ipples | -0.566 |
ry | -0.554 |
andy | -0.544 |
flower | -0.538 |
Token | like |
Token ID | 588 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
hy | +0.450 |
Mic | +0.407 |
rides | +0.347 |
Red | +0.346 |
Roger | +0.338 |
Neg logit contributions | |
riage | -0.479 |
ipples | -0.471 |
ry | -0.459 |
flower | -0.456 |
ac | -0.452 |
Token | a |
Token ID | 257 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
riage | +0.478 |
ipples | +0.466 |
ry | +0.455 |
flower | +0.451 |
ac | +0.443 |
Neg logit contributions | |
hy | -0.618 |
Mic | -0.556 |
rides | -0.499 |
Red | -0.492 |
Roger | -0.484 |
Token | toy |
Token ID | 13373 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
riage | +0.681 |
ipples | +0.665 |
ry | +0.648 |
flower | +0.637 |
ac | +0.632 |
Neg logit contributions | |
hy | -0.795 |
Mic | -0.717 |
rides | -0.634 |
Red | -0.629 |
Roger | -0.617 |
Token | ! |
Token ID | 0 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
hy | +0.889 |
Mic | +0.823 |
Red | +0.723 |
rides | +0.719 |
Roger | +0.717 |
Neg logit contributions | |
riage | -0.652 |
ipples | -0.634 |
flower | -0.631 |
ry | -0.625 |
department | -0.606 |
Token | Joe |
Token ID | 5689 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
riage | +4.284 |
flower | +4.142 |
ac | +4.078 |
ipples | +4.035 |
ric | +3.944 |
Neg logit contributions | |
hy | -4.141 |
Mic | -3.626 |
rides | -3.386 |
Bow | -3.265 |
Robert | -3.246 |
Token | and |
Token ID | 290 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
hy | +0.968 |
Mic | +0.876 |
Red | +0.772 |
rides | +0.769 |
Roger | +0.757 |
Neg logit contributions | |
ipples | -0.733 |
riage | -0.731 |
ry | -0.729 |
flower | -0.726 |
andy | -0.710 |
Token | Linda |
Token ID | 21776 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
riage | +5.100 |
ipples | +5.026 |
ric | +4.732 |
department | +4.576 |
flower | +4.478 |
Neg logit contributions | |
hy | -8.005 |
Mic | -7.360 |
long | -6.924 |
Bow | -6.795 |
Ellie | -6.772 |
Token | were |
Token ID | 547 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
hy | +2.168 |
Mic | +2.050 |
Red | +1.793 |
Roger | +1.770 |
Robert | +1.736 |
Neg logit contributions | |
riage | -1.464 |
ipples | -1.456 |
ry | -1.445 |
flower | -1.424 |
andy | -1.389 |
Token | so |
Token ID | 523 |
Feature activation | +9.99e-01 |
Pos logit contributions | |
riage | +0.195 |
ipples | +0.192 |
ry | +0.187 |
flower | +0.186 |
ac | +0.183 |
Neg logit contributions | |
hy | -0.170 |
Mic | -0.150 |
rides | -0.130 |
Red | -0.128 |
Roger | -0.125 |
Token | amazed |
Token ID | 24562 |
Feature activation | -3.30e-01 |
Pos logit contributions | |
hy | +1.729 |
Mic | +1.620 |
rides | +1.351 |
Red | +1.350 |
Roger | +1.328 |
Neg logit contributions | |
ipples | -1.903 |
riage | -1.817 |
flower | -1.792 |
ac | -1.785 |
ry | -1.783 |
Token | and |
Token ID | 290 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
riage | +0.544 |
ipples | +0.530 |
ry | +0.513 |
flower | +0.512 |
ac | +0.505 |
Neg logit contributions | |
hy | -0.641 |
Mic | -0.575 |
rides | -0.513 |
Red | -0.505 |
Roger | -0.497 |
Token | the |
Token ID | 262 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
riage | +0.995 |
ipples | +0.986 |
ry | +0.939 |
flower | +0.938 |
ac | +0.921 |
Neg logit contributions | |
hy | -1.434 |
Mic | -1.293 |
rides | -1.187 |
Red | -1.153 |
Roger | -1.138 |
Token | f |
Token ID | 277 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
riage | +2.503 |
ac | +2.236 |
ipples | +2.221 |
flower | +2.197 |
ry | +2.184 |
Neg logit contributions | |
hy | -4.726 |
Mic | -4.075 |
rides | -4.032 |
long | -3.909 |
Red | -3.806 |
Token | auc |
Token ID | 14272 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
riage | +2.165 |
flower | +1.973 |
ac | +1.760 |
ipples | +1.752 |
department | +1.715 |
Neg logit contributions | |
hy | -6.405 |
Mic | -5.585 |
rides | -5.388 |
looking | -5.357 |
Red | -5.293 |
Token | et |
Token ID | 316 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
riage | +2.093 |
ipples | +1.996 |
ry | +1.955 |
flower | +1.948 |
andy | +1.891 |
Neg logit contributions | |
hy | -4.112 |
Mic | -3.703 |
rides | -3.495 |
Red | -3.359 |
Bow | -3.211 |
Token | . |
Token ID | 13 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
riage | +3.397 |
ipples | +3.311 |
andy | +3.160 |
ucks | +3.144 |
ac | +3.137 |
Neg logit contributions | |
hy | -3.443 |
Mic | -2.799 |
rides | -2.686 |
Red | -2.518 |
Bow | -2.419 |
Token | The |
Token ID | 383 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ipples | +2.207 |
riage | +2.182 |
flower | +2.122 |
ac | +2.087 |
ry | +2.079 |
Neg logit contributions | |
hy | -1.707 |
Mic | -1.457 |
rides | -1.291 |
Red | -1.272 |
Robert | -1.199 |
Token | water |
Token ID | 1660 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
riage | +6.544 |
ric | +6.016 |
ry | +5.955 |
ipples | +5.807 |
ill | +5.750 |
Neg logit contributions | |
hy | -6.952 |
long | -6.071 |
Mic | -5.637 |
rides | -5.632 |
curly | -5.537 |
Token | came |
Token ID | 1625 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
riage | +3.892 |
ipples | +3.449 |
ucks | +3.446 |
ric | +3.437 |
andy | +3.411 |
Neg logit contributions | |
hy | -4.664 |
Mic | -3.863 |
rides | -3.833 |
brace | -3.523 |
Red | -3.510 |
Token | out |
Token ID | 503 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
riage | +1.845 |
ipples | +1.726 |
ry | +1.656 |
ac | +1.620 |
andy | +1.618 |
Neg logit contributions | |
hy | -2.162 |
Mic | -1.948 |
Red | -1.720 |
rides | -1.714 |
Roger | -1.687 |
Token | and |
Token ID | 290 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
riage | +3.162 |
ipples | +2.977 |
ric | +2.924 |
flower | +2.867 |
ours | +2.857 |
Neg logit contributions | |
hy | -2.757 |
Mic | -2.339 |
rides | -2.057 |
Red | -2.017 |
long | -2.014 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
ipples | +2.769 |
riage | +2.741 |
ric | +2.629 |
andy | +2.627 |
ac | +2.625 |
Neg logit contributions | |
hy | -2.940 |
Mic | -2.627 |
rides | -2.408 |
Red | -2.374 |
nam | -2.362 |
Token | surrender |
Token ID | 16908 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
hy | +0.698 |
Mic | +0.643 |
rides | +0.582 |
Red | +0.576 |
Roger | +0.568 |
Neg logit contributions | |
riage | -0.388 |
ipples | -0.375 |
ry | -0.361 |
flower | -0.360 |
andy | -0.351 |
Token | if |
Token ID | 611 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
riage | +0.959 |
ipples | +0.939 |
flower | +0.906 |
ry | +0.899 |
department | +0.888 |
Neg logit contributions | |
hy | -1.147 |
Mic | -1.030 |
rides | -0.924 |
Red | -0.908 |
Roger | -0.895 |
Token | it |
Token ID | 340 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
riage | +0.420 |
ipples | +0.413 |
ry | +0.404 |
flower | +0.403 |
ac | +0.397 |
Neg logit contributions | |
hy | -0.278 |
Mic | -0.240 |
rides | -0.202 |
Red | -0.198 |
Roger | -0.193 |
Token | is |
Token ID | 318 |
Feature activation | -7.67e-02 |
Pos logit contributions | |
hy | +0.236 |
Mic | +0.218 |
Red | +0.196 |
rides | +0.196 |
Roger | +0.194 |
Neg logit contributions | |
riage | -0.115 |
ipples | -0.111 |
ry | -0.108 |
flower | -0.107 |
ac | -0.104 |
Token | the |
Token ID | 262 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
riage | +0.122 |
ipples | +0.119 |
ry | +0.116 |
flower | +0.115 |
ac | +0.113 |
Neg logit contributions | |
hy | -0.153 |
Mic | -0.138 |
rides | -0.123 |
Red | -0.122 |
Roger | -0.119 |
Token | right |
Token ID | 826 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
ry | +3.169 |
riage | +3.125 |
ipples | +3.073 |
ucks | +3.009 |
ull | +2.965 |
Neg logit contributions | |
hy | -4.052 |
Mic | -3.472 |
rides | -3.264 |
long | -3.146 |
Red | -3.122 |
Token | thing |
Token ID | 1517 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
ipples | +5.200 |
riage | +5.098 |
andy | +5.065 |
ric | +5.021 |
department | +4.852 |
Neg logit contributions | |
hy | -7.212 |
rides | -6.041 |
long | -5.942 |
blonde | -5.679 |
curly | -5.667 |
Token | to |
Token ID | 284 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
hy | +0.970 |
Mic | +0.897 |
Red | +0.795 |
rides | +0.791 |
Roger | +0.780 |
Neg logit contributions | |
riage | -0.623 |
ry | -0.599 |
ipples | -0.599 |
flower | -0.587 |
department | -0.578 |
Token | do |
Token ID | 466 |
Feature activation | +4.67e-01 |
Pos logit contributions | |
hy | +1.978 |
Mic | +1.829 |
Red | +1.615 |
rides | +1.594 |
nam | +1.590 |
Neg logit contributions | |
riage | -1.259 |
ipples | -1.199 |
ry | -1.163 |
flower | -1.156 |
andy | -1.133 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
hy | +0.987 |
Mic | +0.908 |
Red | +0.790 |
rides | +0.785 |
Roger | +0.775 |
Neg logit contributions | |
riage | -0.695 |
ipples | -0.666 |
ry | -0.659 |
ac | -0.641 |
flower | -0.634 |
Token |
|
Token ID | 198 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
hy | +0.734 |
Mic | +0.655 |
rides | +0.581 |
Red | +0.577 |
Roger | +0.557 |
Neg logit contributions | |
riage | -0.617 |
ipples | -0.599 |
ry | -0.594 |
andy | -0.584 |
flower | -0.582 |
Token | happily |
Token ID | 18177 |
Feature activation | +9.88e-01 |
Pos logit contributions | |
hy | +2.183 |
Mic | +2.025 |
Roger | +1.761 |
Robert | +1.759 |
Red | +1.758 |
Neg logit contributions | |
ipples | -1.347 |
riage | -1.336 |
ry | -1.335 |
flower | -1.296 |
ac | -1.270 |
Token | . |
Token ID | 13 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
hy | +2.173 |
Mic | +2.085 |
Red | +1.819 |
Roger | +1.802 |
Robert | +1.778 |
Neg logit contributions | |
ry | -1.377 |
ipples | -1.369 |
riage | -1.354 |
flower | -1.315 |
ac | -1.277 |
Token | Then |
Token ID | 3244 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
riage | +0.994 |
ipples | +0.988 |
flower | +0.958 |
ry | +0.955 |
ac | +0.942 |
Neg logit contributions | |
hy | -0.793 |
Mic | -0.692 |
rides | -0.603 |
Red | -0.593 |
Roger | -0.569 |
Token | Rose |
Token ID | 8049 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ipples | +1.403 |
riage | +1.389 |
flower | +1.333 |
ry | +1.323 |
ac | +1.322 |
Neg logit contributions | |
hy | -1.158 |
Mic | -1.010 |
rides | -0.897 |
Red | -0.863 |
Roger | -0.845 |
Token | showed |
Token ID | 3751 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
riage | +2.554 |
ipples | +2.446 |
ric | +2.372 |
flower | +2.326 |
ac | +2.319 |
Neg logit contributions | |
hy | -2.549 |
Mic | -2.305 |
rides | -2.163 |
Red | -2.009 |
Roger | -1.964 |
Token | her |
Token ID | 607 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
riage | +1.292 |
ipples | +1.281 |
flower | +1.237 |
ry | +1.220 |
department | +1.197 |
Neg logit contributions | |
hy | -1.203 |
Mic | -1.070 |
rides | -0.961 |
Red | -0.926 |
Roger | -0.902 |
Token | how |
Token ID | 703 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
riage | +8.095 |
ipples | +7.252 |
ucks | +7.048 |
flower | +7.030 |
department | +6.944 |
Neg logit contributions | |
hy | -5.492 |
rides | -4.752 |
long | -4.557 |
brace | -4.260 |
powers | -4.096 |
Token | to |
Token ID | 284 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
riage | +1.175 |
ipples | +1.125 |
flower | +1.105 |
ry | +1.087 |
department | +1.072 |
Neg logit contributions | |
hy | -1.669 |
Mic | -1.487 |
rides | -1.369 |
Red | -1.332 |
Roger | -1.314 |
Token | turn |
Token ID | 1210 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
riage | +2.701 |
department | +2.511 |
ac | +2.449 |
ipples | +2.449 |
ry | +2.408 |
Neg logit contributions | |
hy | -3.635 |
Mic | -3.183 |
rides | -3.021 |
Red | -2.924 |
Blue | -2.832 |
Token | it |
Token ID | 340 |
Feature activation | -1.87e-02 |
Pos logit contributions | |
riage | +2.375 |
ipples | +2.209 |
department | +2.147 |
flower | +2.137 |
ac | +2.103 |
Neg logit contributions | |
hy | -2.836 |
Mic | -2.481 |
rides | -2.397 |
Red | -2.268 |
brace | -2.236 |
Token | off |
Token ID | 572 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
riage | +0.029 |
ipples | +0.029 |
ry | +0.028 |
flower | +0.028 |
ac | +0.027 |
Neg logit contributions | |
hy | -0.038 |
Mic | -0.034 |
rides | -0.031 |
Red | -0.030 |
Roger | -0.030 |
Token | f |
Token ID | 277 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
riage | +2.929 |
ac | +2.748 |
ipples | +2.664 |
ry | +2.648 |
ucks | +2.632 |
Neg logit contributions | |
hy | -4.440 |
Mic | -3.980 |
rides | -3.788 |
Red | -3.709 |
Roger | -3.590 |
Token | auc |
Token ID | 14272 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
riage | +2.910 |
department | +2.409 |
flower | +2.370 |
mistake | +2.186 |
ac | +2.178 |
Neg logit contributions | |
hy | -7.474 |
Mic | -6.428 |
looking | -6.368 |
rides | -6.181 |
Red | -6.164 |
Token | et |
Token ID | 316 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
riage | +3.489 |
department | +3.279 |
ipples | +3.200 |
flower | +3.150 |
ric | +3.125 |
Neg logit contributions | |
hy | -5.817 |
Mic | -5.132 |
rides | -4.930 |
Red | -4.690 |
hell | -4.592 |
Token | to |
Token ID | 284 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
riage | +4.336 |
ipples | +4.095 |
department | +4.034 |
ucks | +3.938 |
ac | +3.884 |
Neg logit contributions | |
hy | -5.336 |
Mic | -4.367 |
rides | -4.348 |
Red | -3.952 |
dream | -3.907 |
Token | ' |
Token ID | 705 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
riage | +2.707 |
ipples | +2.636 |
flower | +2.612 |
department | +2.602 |
ac | +2.568 |
Neg logit contributions | |
hy | -3.282 |
Mic | -2.868 |
rides | -2.752 |
Red | -2.607 |
Roger | -2.546 |
Token | spl |
Token ID | 22018 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
riage | +0.780 |
ipples | +0.764 |
ry | +0.742 |
flower | +0.732 |
department | +0.730 |
Neg logit contributions | |
hy | -0.987 |
Mic | -0.892 |
rides | -0.796 |
Red | -0.791 |
Roger | -0.775 |
Token | ash |
Token ID | 1077 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
riage | +4.683 |
department | +4.666 |
ipples | +4.102 |
mistake | +4.076 |
flower | +4.052 |
Neg logit contributions | |
hy | -8.185 |
Mic | -7.264 |
hell | -6.911 |
looking | -6.763 |
Red | -6.727 |
Token | ' |
Token ID | 6 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
ipples | +3.833 |
riage | +3.755 |
department | +3.614 |
ucks | +3.586 |
andy | +3.506 |
Neg logit contributions | |
hy | -4.291 |
Mic | -3.741 |
Roger | -3.376 |
rides | -3.324 |
Red | -3.319 |
Token | around |
Token ID | 1088 |
Feature activation | -4.41e-01 |
Pos logit contributions | |
riage | +1.453 |
ipples | +1.411 |
ry | +1.358 |
flower | +1.354 |
andy | +1.345 |
Neg logit contributions | |
hy | -1.766 |
Mic | -1.554 |
rides | -1.420 |
Red | -1.382 |
Roger | -1.351 |
Token | and |
Token ID | 290 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
riage | +0.722 |
ipples | +0.708 |
ry | +0.682 |
flower | +0.680 |
department | +0.669 |
Neg logit contributions | |
hy | -0.857 |
Mic | -0.770 |
rides | -0.686 |
Red | -0.676 |
Roger | -0.667 |
Token | have |
Token ID | 423 |
Feature activation | -7.17e-01 |
Pos logit contributions | |
riage | +0.438 |
ipples | +0.428 |
ry | +0.414 |
flower | +0.412 |
ac | +0.404 |
Neg logit contributions | |
hy | -0.623 |
Mic | -0.563 |
rides | -0.508 |
Red | -0.500 |
Roger | -0.492 |
Token | together |
Token ID | 1978 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
hy | +1.932 |
Mic | +1.800 |
Red | +1.598 |
rides | +1.567 |
Roger | +1.552 |
Neg logit contributions | |
ipples | -1.241 |
riage | -1.241 |
ry | -1.224 |
flower | -1.200 |
ac | -1.183 |
Token | . |
Token ID | 13 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
riage | +0.165 |
ipples | +0.161 |
ry | +0.157 |
flower | +0.156 |
ac | +0.153 |
Neg logit contributions | |
hy | -0.213 |
Mic | -0.193 |
rides | -0.171 |
Red | -0.170 |
Roger | -0.167 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
riage | +4.780 |
flower | +4.624 |
ac | +4.621 |
ipples | +4.509 |
department | +4.497 |
Neg logit contributions | |
hy | -3.138 |
Mic | -2.782 |
rides | -2.515 |
Bow | -2.420 |
Red | -2.383 |
Token |
|
Token ID | 198 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
riage | +2.020 |
ipples | +1.938 |
flower | +1.929 |
department | +1.891 |
ry | +1.860 |
Neg logit contributions | |
hy | -1.652 |
Mic | -1.498 |
rides | -1.304 |
Red | -1.288 |
Roger | -1.272 |
Token | Joe |
Token ID | 19585 |
Feature activation | -2.46e-01 |
Pos logit contributions | |
riage | +4.199 |
flower | +3.911 |
department | +3.890 |
ipples | +3.868 |
ry | +3.812 |
Neg logit contributions | |
hy | -3.555 |
Mic | -3.314 |
Red | -2.923 |
rides | -2.846 |
Roger | -2.837 |
Token | and |
Token ID | 290 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
riage | +0.419 |
ipples | +0.409 |
ry | +0.397 |
flower | +0.395 |
ac | +0.388 |
Neg logit contributions | |
hy | -0.465 |
Mic | -0.417 |
rides | -0.369 |
Red | -0.364 |
Roger | -0.357 |
Token | Linda |
Token ID | 21776 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
riage | +4.806 |
ipples | +4.583 |
department | +4.350 |
ric | +4.283 |
ucks | +4.148 |
Neg logit contributions | |
hy | -8.153 |
Mic | -7.573 |
long | -7.007 |
Ellie | -6.918 |
nam | -6.865 |
Token | then |
Token ID | 788 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
hy | +2.784 |
Mic | +2.637 |
Red | +2.303 |
Roger | +2.303 |
Robert | +2.247 |
Neg logit contributions | |
riage | -1.886 |
ry | -1.833 |
flower | -1.810 |
ipples | -1.808 |
andy | -1.761 |
Token | kept |
Token ID | 4030 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
riage | +0.002 |
ipples | +0.002 |
ry | +0.002 |
flower | +0.002 |
ac | +0.002 |
Neg logit contributions | |
hy | -0.003 |
Mic | -0.003 |
rides | -0.002 |
Red | -0.002 |
Roger | -0.002 |
Token | walking |
Token ID | 6155 |
Feature activation | -8.37e-01 |
Pos logit contributions | |
riage | +0.261 |
ipples | +0.255 |
ry | +0.248 |
flower | +0.247 |
ac | +0.242 |
Neg logit contributions | |
hy | -0.322 |
Mic | -0.290 |
rides | -0.258 |
Red | -0.255 |
Roger | -0.251 |
Token | in |
Token ID | 287 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
riage | +1.471 |
ipples | +1.431 |
flower | +1.386 |
ry | +1.385 |
andy | +1.382 |
Neg logit contributions | |
hy | -1.563 |
Mic | -1.351 |
rides | -1.245 |
Red | -1.189 |
Roger | -1.173 |
Token | were |
Token ID | 547 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
hy | +0.270 |
Mic | +0.245 |
rides | +0.215 |
Red | +0.214 |
Roger | +0.211 |
Neg logit contributions | |
riage | -0.220 |
ipples | -0.214 |
ry | -0.210 |
flower | -0.210 |
andy | -0.204 |
Token | walking |
Token ID | 6155 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
hy | +0.432 |
Mic | +0.387 |
Red | +0.333 |
rides | +0.330 |
Roger | +0.321 |
Neg logit contributions | |
riage | -0.457 |
ipples | -0.450 |
ry | -0.441 |
flower | -0.439 |
ac | -0.436 |
Token | down |
Token ID | 866 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
riage | +0.442 |
ipples | +0.432 |
ry | +0.419 |
flower | +0.416 |
andy | +0.409 |
Neg logit contributions | |
hy | -0.573 |
Mic | -0.516 |
rides | -0.463 |
Red | -0.456 |
Roger | -0.449 |
Token | the |
Token ID | 262 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
riage | +0.519 |
ipples | +0.505 |
ry | +0.487 |
flower | +0.484 |
ac | +0.475 |
Neg logit contributions | |
hy | -0.930 |
Mic | -0.844 |
rides | -0.776 |
Red | -0.761 |
Roger | -0.753 |
Token | street |
Token ID | 4675 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
hy | +3.012 |
Mic | +2.773 |
blonde | +2.280 |
hell | +2.269 |
Roger | +2.260 |
Neg logit contributions | |
riage | -3.070 |
flower | -3.062 |
ipples | -3.038 |
ry | -3.014 |
andy | -2.984 |
Token | . |
Token ID | 13 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
hy | +0.889 |
Mic | +0.840 |
Red | +0.730 |
Roger | +0.719 |
rides | +0.715 |
Neg logit contributions | |
riage | -0.648 |
ipples | -0.636 |
ry | -0.626 |
flower | -0.617 |
department | -0.600 |
Token | Sandy |
Token ID | 17581 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
riage | +7.472 |
flower | +7.293 |
department | +7.235 |
ac | +7.199 |
mistake | +7.045 |
Neg logit contributions | |
hy | -5.242 |
long | -4.158 |
blonde | -4.049 |
Mic | -3.984 |
curly | -3.983 |
Token | was |
Token ID | 373 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
hy | +3.826 |
Mic | +3.606 |
nam | +3.201 |
Red | +3.180 |
Robert | +3.141 |
Neg logit contributions | |
ry | -1.610 |
flower | -1.502 |
andy | -1.499 |
riage | -1.492 |
ipples | -1.478 |
Token | feeling |
Token ID | 4203 |
Feature activation | +6.31e-01 |
Pos logit contributions | |
hy | +0.275 |
Mic | +0.249 |
rides | +0.214 |
Red | +0.213 |
Roger | +0.208 |
Neg logit contributions | |
riage | -0.280 |
ipples | -0.278 |
flower | -0.271 |
ry | -0.268 |
ac | -0.264 |
Token | cold |
Token ID | 4692 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
hy | +1.133 |
Mic | +1.050 |
rides | +0.885 |
Red | +0.880 |
Roger | +0.857 |
Neg logit contributions | |
ipples | -1.110 |
riage | -1.099 |
ry | -1.058 |
flower | -1.054 |
ac | -1.047 |
Token | , |
Token ID | 11 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
hy | +2.572 |
Mic | +2.445 |
Red | +2.117 |
Robert | +2.082 |
Roger | +2.065 |
Neg logit contributions | |
ipples | -1.745 |
flower | -1.737 |
riage | -1.717 |
ry | -1.693 |
department | -1.656 |
Token | " |
Token ID | 366 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
riage | +0.086 |
ipples | +0.084 |
ry | +0.081 |
flower | +0.080 |
ac | +0.078 |
Neg logit contributions | |
hy | -0.151 |
Mic | -0.138 |
rides | -0.125 |
Red | -0.123 |
Roger | -0.121 |
Token | It |
Token ID | 1026 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
riage | +0.370 |
ipples | +0.361 |
ry | +0.353 |
flower | +0.349 |
department | +0.343 |
Neg logit contributions | |
hy | -0.445 |
Mic | -0.401 |
rides | -0.355 |
Red | -0.352 |
Roger | -0.346 |
Token | 's |
Token ID | 338 |
Feature activation | -9.53e-01 |
Pos logit contributions | |
riage | +1.711 |
ipples | +1.662 |
flower | +1.572 |
ry | +1.566 |
ucks | +1.559 |
Neg logit contributions | |
hy | -2.263 |
Mic | -2.024 |
rides | -1.867 |
Red | -1.812 |
Roger | -1.776 |
Token | alive |
Token ID | 6776 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
riage | +1.785 |
ipples | +1.756 |
ry | +1.744 |
flower | +1.713 |
department | +1.700 |
Neg logit contributions | |
hy | -1.623 |
Mic | -1.433 |
rides | -1.258 |
Red | -1.240 |
Roger | -1.224 |
Token | ! |
Token ID | 0 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
riage | +1.379 |
ipples | +1.350 |
ry | +1.305 |
flower | +1.288 |
andy | +1.277 |
Neg logit contributions | |
hy | -1.477 |
Mic | -1.306 |
rides | -1.167 |
Red | -1.146 |
Roger | -1.132 |
Token | Hur |
Token ID | 10077 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
ipples | +2.444 |
riage | +2.381 |
flower | +2.340 |
ry | +2.333 |
andy | +2.308 |
Neg logit contributions | |
hy | -2.264 |
Mic | -1.972 |
Red | -1.739 |
rides | -1.695 |
Roger | -1.677 |
Token | ray |
Token ID | 2433 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
ipples | +4.527 |
ac | +3.954 |
ull | +3.869 |
ucks | +3.603 |
riage | +3.596 |
Neg logit contributions | |
hy | -8.723 |
ve | -7.053 |
Mic | -7.005 |
rides | -6.680 |
Red | -6.664 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
riage | +1.558 |
ipples | +1.528 |
flower | +1.470 |
andy | +1.443 |
ry | +1.437 |
Neg logit contributions | |
hy | -1.706 |
Mic | -1.504 |
rides | -1.377 |
Red | -1.320 |
Roger | -1.303 |
Token | The |
Token ID | 383 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
hy | +1.335 |
Mic | +1.196 |
rides | +1.053 |
Red | +1.049 |
nam | +1.014 |
Neg logit contributions | |
riage | -1.092 |
ipples | -1.087 |
ry | -1.071 |
flower | -1.045 |
andy | -1.038 |
Token | grape |
Token ID | 30777 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
riage | +2.227 |
ry | +2.132 |
ipples | +2.049 |
ac | +2.003 |
ucks | +1.994 |
Neg logit contributions | |
hy | -3.236 |
Mic | -3.074 |
rides | -2.740 |
Red | -2.655 |
Roger | -2.645 |
Token | j |
Token ID | 474 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
riage | +0.609 |
ipples | +0.585 |
ry | +0.568 |
flower | +0.562 |
ac | +0.554 |
Neg logit contributions | |
hy | -0.869 |
Mic | -0.787 |
rides | -0.711 |
Red | -0.698 |
Roger | -0.688 |
Token | He |
Token ID | 679 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
hy | +0.071 |
Mic | +0.063 |
rides | +0.055 |
Red | +0.054 |
Roger | +0.053 |
Neg logit contributions | |
riage | -0.077 |
ipples | -0.075 |
ry | -0.073 |
flower | -0.073 |
andy | -0.072 |
Token | liked |
Token ID | 8288 |
Feature activation | +8.39e-03 |
Pos logit contributions | |
hy | +1.005 |
Mic | +0.906 |
rides | +0.783 |
Red | +0.773 |
Robert | +0.769 |
Neg logit contributions | |
riage | -0.934 |
ipples | -0.921 |
flower | -0.906 |
ry | -0.900 |
andy | -0.892 |
Token | to |
Token ID | 284 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
hy | +0.019 |
Mic | +0.018 |
Red | +0.016 |
rides | +0.016 |
Roger | +0.015 |
Neg logit contributions | |
riage | -0.011 |
ipples | -0.011 |
ry | -0.010 |
flower | -0.010 |
department | -0.010 |
Token | play |
Token ID | 711 |
Feature activation | +3.49e-01 |
Pos logit contributions | |
riage | +0.325 |
ipples | +0.317 |
ry | +0.308 |
flower | +0.306 |
ac | +0.300 |
Neg logit contributions | |
hy | -0.432 |
Mic | -0.390 |
rides | -0.349 |
Red | -0.345 |
Roger | -0.339 |
Token | all |
Token ID | 477 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
hy | +0.618 |
Mic | +0.535 |
Red | +0.456 |
rides | +0.453 |
Roger | +0.449 |
Neg logit contributions | |
riage | -0.639 |
ipples | -0.636 |
ry | -0.625 |
flower | -0.616 |
ac | -0.607 |
Token | day |
Token ID | 1110 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
riage | +0.414 |
ipples | +0.402 |
ry | +0.390 |
flower | +0.387 |
department | +0.381 |
Neg logit contributions | |
hy | -0.618 |
Mic | -0.560 |
rides | -0.507 |
Red | -0.500 |
Roger | -0.491 |
Token | in |
Token ID | 287 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
hy | +9.471 |
Mic | +8.632 |
Red | +8.000 |
Blue | +7.758 |
Bow | +7.597 |
Neg logit contributions | |
ry | -4.158 |
again | -3.829 |
ipples | -3.703 |
en | -3.668 |
andy | -3.566 |
Token | the |
Token ID | 262 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
hy | +0.166 |
Mic | +0.152 |
Red | +0.137 |
rides | +0.136 |
Roger | +0.134 |
Neg logit contributions | |
riage | -0.089 |
ipples | -0.086 |
ry | -0.083 |
flower | -0.083 |
andy | -0.081 |
Token | sun |
Token ID | 4252 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
riage | +0.796 |
ipples | +0.770 |
ry | +0.754 |
flower | +0.739 |
ac | +0.734 |
Neg logit contributions | |
hy | -1.054 |
Mic | -0.958 |
rides | -0.866 |
Red | -0.850 |
Roger | -0.833 |
Token | . |
Token ID | 13 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
riage | +0.452 |
ipples | +0.440 |
ry | +0.428 |
flower | +0.423 |
ac | +0.419 |
Neg logit contributions | |
hy | -0.554 |
Mic | -0.498 |
rides | -0.445 |
Red | -0.438 |
Roger | -0.430 |
Token | One |
Token ID | 1881 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
riage | +0.922 |
ipples | +0.904 |
ry | +0.887 |
flower | +0.886 |
ac | +0.869 |
Neg logit contributions | |
hy | -0.936 |
Mic | -0.840 |
rides | -0.745 |
Red | -0.732 |
Roger | -0.715 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.11e-01 |
Pos logit contributions | |
hy | +0.605 |
Mic | +0.554 |
rides | +0.501 |
Red | +0.496 |
Roger | +0.490 |
Neg logit contributions | |
riage | -0.359 |
ipples | -0.349 |
flower | -0.348 |
ry | -0.344 |
department | -0.333 |
Token | crying |
Token ID | 13774 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
hy | +0.401 |
Mic | +0.365 |
Red | +0.319 |
rides | +0.319 |
Roger | +0.314 |
Neg logit contributions | |
riage | -0.349 |
ipples | -0.341 |
ry | -0.340 |
flower | -0.339 |
andy | -0.327 |
Token | . |
Token ID | 13 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
riage | +0.418 |
ipples | +0.409 |
ry | +0.396 |
flower | +0.393 |
ac | +0.387 |
Neg logit contributions | |
hy | -0.515 |
Mic | -0.463 |
rides | -0.412 |
Red | -0.407 |
Roger | -0.400 |
Token | |
Token ID | 220 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
hy | +3.799 |
Mic | +3.388 |
ve | +3.098 |
hell | +3.072 |
nam | +3.035 |
Neg logit contributions | |
riage | -2.807 |
andy | -2.700 |
department | -2.688 |
ry | -2.685 |
flower | -2.674 |
Token |
|
Token ID | 198 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
hy | +1.060 |
Mic | +0.946 |
rides | +0.847 |
Red | +0.826 |
Roger | +0.813 |
Neg logit contributions | |
riage | -0.839 |
ipples | -0.830 |
ry | -0.822 |
flower | -0.797 |
ac | -0.797 |
Token |
|
Token ID | 198 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
hy | +2.424 |
Mic | +2.121 |
nam | +1.928 |
rides | +1.903 |
Red | +1.873 |
Neg logit contributions | |
ry | -1.607 |
riage | -1.602 |
andy | -1.560 |
ipples | -1.525 |
ac | -1.513 |
Token | Max |
Token ID | 11518 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
hy | +4.411 |
Mic | +3.682 |
nam | +3.662 |
rides | +3.497 |
con | +3.459 |
Neg logit contributions | |
riage | -2.735 |
ry | -2.693 |
andy | -2.640 |
flower | -2.588 |
ac | -2.575 |
Token | went |
Token ID | 1816 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
hy | +2.247 |
Mic | +2.008 |
rides | +1.804 |
Red | +1.797 |
nam | +1.791 |
Neg logit contributions | |
ry | -1.539 |
flower | -1.539 |
andy | -1.521 |
riage | -1.518 |
department | -1.469 |
Token | up |
Token ID | 510 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
hy | +0.657 |
Mic | +0.604 |
Red | +0.548 |
rides | +0.545 |
Roger | +0.537 |
Neg logit contributions | |
riage | -0.309 |
ipples | -0.298 |
flower | -0.291 |
ry | -0.291 |
department | -0.287 |
Token | to |
Token ID | 284 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
hy | +3.547 |
Mic | +3.266 |
Red | +3.072 |
blonde | +2.937 |
nam | +2.926 |
Neg logit contributions | |
flower | -1.326 |
ry | -1.314 |
department | -1.311 |
riage | -1.300 |
ipples | -1.291 |
Token | her |
Token ID | 607 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
hy | +1.008 |
Mic | +0.918 |
Red | +0.821 |
rides | +0.812 |
Roger | +0.799 |
Neg logit contributions | |
riage | -0.637 |
ipples | -0.596 |
ry | -0.596 |
flower | -0.593 |
andy | -0.588 |
Token | at |
Token ID | 379 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
riage | +0.391 |
ipples | +0.383 |
ry | +0.372 |
flower | +0.371 |
ac | +0.364 |
Neg logit contributions | |
hy | -0.399 |
Mic | -0.356 |
rides | -0.313 |
Red | -0.308 |
Roger | -0.303 |
Token | him |
Token ID | 683 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
riage | +0.464 |
ipples | +0.455 |
ry | +0.440 |
flower | +0.437 |
ac | +0.428 |
Neg logit contributions | |
hy | -0.649 |
Mic | -0.590 |
rides | -0.530 |
Red | -0.522 |
Roger | -0.515 |
Token | , |
Token ID | 11 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
hy | +0.308 |
Mic | +0.278 |
Red | +0.243 |
rides | +0.242 |
Roger | +0.238 |
Neg logit contributions | |
riage | -0.282 |
ipples | -0.276 |
ry | -0.270 |
flower | -0.270 |
ac | -0.264 |
Token | he |
Token ID | 339 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
riage | +1.096 |
ipples | +1.077 |
flower | +1.050 |
ry | +1.045 |
ac | +1.031 |
Neg logit contributions | |
hy | -1.188 |
Mic | -1.045 |
rides | -0.945 |
Red | -0.913 |
Roger | -0.901 |
Token | is |
Token ID | 318 |
Feature activation | +7.71e-01 |
Pos logit contributions | |
hy | +2.008 |
Mic | +1.909 |
Red | +1.707 |
Roger | +1.696 |
nam | +1.696 |
Neg logit contributions | |
riage | -0.905 |
ipples | -0.875 |
flower | -0.871 |
ry | -0.871 |
andy | -0.827 |
Token | so |
Token ID | 523 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
hy | +1.483 |
Mic | +1.353 |
rides | +1.177 |
Red | +1.176 |
Roger | +1.141 |
Neg logit contributions | |
ipples | -1.265 |
riage | -1.259 |
flower | -1.219 |
ac | -1.198 |
andy | -1.174 |
Token | fat |
Token ID | 3735 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
hy | +3.700 |
Mic | +3.633 |
rides | +3.091 |
Roger | +3.040 |
Robert | +2.981 |
Neg logit contributions | |
ipples | -3.694 |
andy | -3.567 |
ry | -3.510 |
ac | -3.507 |
riage | -3.505 |
Token | and |
Token ID | 290 |
Feature activation | +9.51e-01 |
Pos logit contributions | |
hy | +4.106 |
Mic | +3.925 |
nam | +3.448 |
Red | +3.437 |
Roger | +3.375 |
Neg logit contributions | |
ipples | -2.750 |
ry | -2.678 |
flower | -2.662 |
riage | -2.605 |
department | -2.575 |
Token | silly |
Token ID | 14397 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
hy | +1.580 |
Mic | +1.467 |
rides | +1.223 |
Red | +1.222 |
Robert | +1.219 |
Neg logit contributions | |
riage | -1.816 |
ipples | -1.761 |
ry | -1.704 |
flower | -1.697 |
ac | -1.685 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
hy | +2.987 |
Mic | +2.811 |
Red | +2.424 |
Roger | +2.364 |
nam | +2.321 |
Neg logit contributions | |
ipples | -2.470 |
riage | -2.427 |
ry | -2.396 |
flower | -2.388 |
department | -2.338 |
Token | Mia |
Token ID | 32189 |
Feature activation | +7.16e-01 |
Pos logit contributions | |
hy | +0.188 |
Mic | +0.174 |
rides | +0.159 |
Red | +0.159 |
Roger | +0.156 |
Neg logit contributions | |
riage | -0.073 |
ipples | -0.071 |
ry | -0.068 |
flower | -0.066 |
andy | -0.065 |
Token | all |
Token ID | 477 |
Feature activation | -3.62e-01 |
Pos logit contributions | |
riage | +0.851 |
ipples | +0.845 |
flower | +0.818 |
ry | +0.808 |
ac | +0.793 |
Neg logit contributions | |
hy | -1.008 |
Mic | -0.891 |
rides | -0.810 |
Red | -0.787 |
Roger | -0.775 |
Token | the |
Token ID | 262 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
riage | +0.503 |
ipples | +0.493 |
ry | +0.474 |
flower | +0.473 |
department | +0.461 |
Neg logit contributions | |
hy | -0.797 |
Mic | -0.721 |
rides | -0.658 |
Red | -0.646 |
Roger | -0.638 |
Token | interesting |
Token ID | 3499 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
hy | +3.750 |
Mic | +3.579 |
Red | +3.171 |
Roger | +3.088 |
Blue | +3.066 |
Neg logit contributions | |
riage | -2.505 |
ac | -2.385 |
ipples | -2.286 |
department | -2.239 |
ry | -2.209 |
Token | things |
Token ID | 1243 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
hy | +2.591 |
Mic | +2.504 |
Red | +2.211 |
Roger | +2.151 |
Blue | +2.131 |
Neg logit contributions | |
riage | -1.623 |
ry | -1.545 |
ipples | -1.508 |
ac | -1.503 |
andy | -1.454 |
Token | about |
Token ID | 546 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
hy | +2.084 |
Mic | +2.005 |
Red | +1.704 |
Roger | +1.654 |
rides | +1.630 |
Neg logit contributions | |
riage | -1.629 |
ry | -1.574 |
ipples | -1.548 |
flower | -1.514 |
department | -1.501 |
Token | the |
Token ID | 262 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
hy | +0.704 |
Mic | +0.650 |
Red | +0.583 |
rides | +0.578 |
Roger | +0.572 |
Neg logit contributions | |
riage | -0.371 |
ipples | -0.355 |
ry | -0.346 |
flower | -0.343 |
andy | -0.333 |
Token | past |
Token ID | 1613 |
Feature activation | +4.80e-01 |
Pos logit contributions | |
hy | +3.881 |
Mic | +3.751 |
Roger | +3.258 |
Red | +3.239 |
Blue | +3.136 |
Neg logit contributions | |
riage | -3.019 |
ac | -2.735 |
ipples | -2.708 |
department | -2.699 |
flower | -2.671 |
Token | . |
Token ID | 13 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
hy | +1.106 |
Mic | +1.039 |
Red | +0.925 |
Roger | +0.909 |
rides | +0.901 |
Neg logit contributions | |
riage | -0.628 |
ry | -0.599 |
ipples | -0.590 |
flower | -0.582 |
ac | -0.570 |
Token | Em |
Token ID | 2295 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
hy | +0.960 |
Mic | +0.887 |
rides | +0.766 |
Roger | +0.752 |
Red | +0.743 |
Neg logit contributions | |
riage | -0.836 |
ipples | -0.790 |
ry | -0.788 |
department | -0.770 |
flower | -0.769 |
Token | mi |
Token ID | 11632 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
hy | +1.316 |
Mic | +1.164 |
Red | +0.984 |
rides | +0.981 |
Roger | +0.976 |
Neg logit contributions | |
riage | -1.555 |
ry | -1.525 |
flower | -1.504 |
ac | -1.489 |
andy | -1.473 |
Token | was |
Token ID | 373 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
riage | +1.226 |
ipples | +1.190 |
flower | +1.125 |
ry | +1.115 |
department | +1.100 |
Neg logit contributions | |
hy | -1.713 |
Mic | -1.554 |
rides | -1.416 |
Red | -1.370 |
Robert | -1.353 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
hy | +1.459 |
Mic | +1.335 |
rides | +1.231 |
Red | +1.208 |
Robert | +1.204 |
Neg logit contributions | |
riage | -0.663 |
flower | -0.660 |
ipples | -0.646 |
ry | -0.643 |
andy | -0.616 |
Token | who |
Token ID | 508 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
riage | +0.044 |
ipples | +0.043 |
ry | +0.042 |
flower | +0.042 |
ac | +0.041 |
Neg logit contributions | |
hy | -0.051 |
Mic | -0.046 |
rides | -0.040 |
Red | -0.040 |
Roger | -0.039 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
hy | +1.580 |
Mic | +1.462 |
rides | +1.255 |
Robert | +1.253 |
Roger | +1.245 |
Neg logit contributions | |
riage | -1.545 |
ipples | -1.516 |
flower | -1.484 |
ry | -1.480 |
ac | -1.475 |
Token | to |
Token ID | 284 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
riage | +0.233 |
ipples | +0.227 |
ry | +0.219 |
flower | +0.217 |
ac | +0.212 |
Neg logit contributions | |
hy | -0.393 |
Mic | -0.359 |
rides | -0.325 |
Red | -0.321 |
Roger | -0.317 |
Token | take |
Token ID | 1011 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
riage | +0.052 |
ipples | +0.051 |
ry | +0.049 |
flower | +0.049 |
ac | +0.048 |
Neg logit contributions | |
hy | -0.072 |
Mic | -0.066 |
rides | -0.058 |
Red | -0.058 |
Roger | -0.057 |
Token | things |
Token ID | 1243 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
hy | +3.275 |
Mic | +2.980 |
Red | +2.619 |
Roger | +2.586 |
Robert | +2.577 |
Neg logit contributions | |
ipples | -2.341 |
riage | -2.298 |
en | -2.253 |
ac | -2.234 |
ry | -2.223 |
Token | that |
Token ID | 326 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
hy | +4.674 |
Mic | +4.460 |
blonde | +3.942 |
Roger | +3.866 |
Robert | +3.839 |
Neg logit contributions | |
ry | -3.059 |
ipples | -2.812 |
riage | -2.764 |
flower | -2.764 |
ac | -2.747 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
hy | +2.276 |
Mic | +2.156 |
Red | +1.883 |
Robert | +1.876 |
Roger | +1.874 |
Neg logit contributions | |
riage | -1.692 |
ipples | -1.610 |
ry | -1.604 |
flower | -1.601 |
ac | -1.528 |
Token | 't |
Token ID | 470 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
hy | +0.210 |
Mic | +0.183 |
rides | +0.153 |
Red | +0.151 |
Roger | +0.148 |
Neg logit contributions | |
riage | -0.291 |
ipples | -0.287 |
ry | -0.281 |
flower | -0.279 |
andy | -0.276 |
Token | belong |
Token ID | 5594 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
hy | +2.094 |
Mic | +1.971 |
Robert | +1.679 |
Roger | +1.676 |
rides | +1.672 |
Neg logit contributions | |
riage | -1.809 |
ipples | -1.800 |
ac | -1.715 |
flower | -1.679 |
ry | -1.670 |
Token | to |
Token ID | 284 |
Feature activation | +7.90e-01 |
Pos logit contributions | |
riage | +0.267 |
ipples | +0.260 |
ry | +0.251 |
flower | +0.250 |
ac | +0.244 |
Neg logit contributions | |
hy | -0.440 |
Mic | -0.401 |
rides | -0.363 |
Red | -0.359 |
Roger | -0.353 |
Token | . |
Token ID | 13 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
hy | +0.018 |
Mic | +0.017 |
rides | +0.015 |
Red | +0.015 |
Roger | +0.014 |
Neg logit contributions | |
riage | -0.013 |
ipples | -0.013 |
ry | -0.013 |
flower | -0.012 |
ac | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
riage | +1.097 |
ipples | +1.082 |
ry | +1.049 |
flower | +1.049 |
ac | +1.044 |
Neg logit contributions | |
hy | -1.061 |
Mic | -0.954 |
rides | -0.836 |
Red | -0.829 |
Roger | -0.808 |
Token |
|
Token ID | 198 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
riage | +0.081 |
ipples | +0.079 |
ry | +0.077 |
flower | +0.076 |
andy | +0.075 |
Neg logit contributions | |
hy | -0.088 |
Mic | -0.078 |
rides | -0.069 |
Red | -0.068 |
Roger | -0.067 |
Token | Tim |
Token ID | 14967 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
riage | +1.029 |
ipples | +1.005 |
ry | +0.980 |
flower | +0.979 |
department | +0.957 |
Neg logit contributions | |
hy | -0.893 |
Mic | -0.806 |
Red | -0.696 |
rides | -0.688 |
Roger | -0.672 |
Token | my |
Token ID | 1820 |
Feature activation | -9.14e-01 |
Pos logit contributions | |
riage | +2.450 |
ipples | +2.384 |
ry | +2.199 |
ours | +2.156 |
ucks | +2.135 |
Neg logit contributions | |
hy | -3.723 |
Mic | -3.523 |
rides | -3.119 |
Red | -3.062 |
Roger | -2.993 |
Token | didn |
Token ID | 1422 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
riage | +1.555 |
ipples | +1.515 |
ry | +1.440 |
flower | +1.425 |
ric | +1.411 |
Neg logit contributions | |
hy | -1.749 |
Mic | -1.610 |
rides | -1.435 |
Red | -1.400 |
Roger | -1.375 |
Token | 't |
Token ID | 470 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
hy | +0.354 |
Mic | +0.307 |
rides | +0.258 |
Red | +0.255 |
Roger | +0.250 |
Neg logit contributions | |
riage | -0.478 |
ipples | -0.470 |
ry | -0.461 |
flower | -0.456 |
andy | -0.455 |
Token | understand |
Token ID | 1833 |
Feature activation | -9.09e-01 |
Pos logit contributions | |
hy | +0.336 |
Mic | +0.304 |
rides | +0.268 |
Red | +0.265 |
Roger | +0.262 |
Neg logit contributions | |
riage | -0.294 |
ipples | -0.289 |
ry | -0.279 |
flower | -0.279 |
ac | -0.276 |
Token | why |
Token ID | 1521 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
ipples | +1.016 |
riage | +1.009 |
flower | +0.948 |
ry | +0.922 |
department | +0.912 |
Neg logit contributions | |
hy | -2.278 |
Mic | -2.105 |
rides | -1.945 |
Red | -1.889 |
Roger | -1.876 |
Token | he |
Token ID | 339 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
riage | +0.241 |
ipples | +0.236 |
ry | +0.230 |
flower | +0.229 |
ac | +0.225 |
Neg logit contributions | |
hy | -0.231 |
Mic | -0.206 |
rides | -0.180 |
Red | -0.177 |
Roger | -0.174 |
Token | needed |
Token ID | 2622 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
hy | +0.974 |
Mic | +0.897 |
rides | +0.797 |
Red | +0.795 |
Roger | +0.792 |
Neg logit contributions | |
riage | -0.587 |
ipples | -0.571 |
ry | -0.560 |
flower | -0.560 |
andy | -0.548 |
Token | more |
Token ID | 517 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
hy | +1.816 |
Mic | +1.699 |
Roger | +1.498 |
Red | +1.475 |
Robert | +1.475 |
Neg logit contributions | |
riage | -1.289 |
ipples | -1.279 |
flower | -1.256 |
ac | -1.200 |
ry | -1.199 |
Token | . |
Token ID | 13 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
hy | +3.111 |
Mic | +3.010 |
Roger | +2.661 |
Red | +2.610 |
Robert | +2.599 |
Neg logit contributions | |
ipples | -1.959 |
riage | -1.891 |
flower | -1.856 |
ry | -1.856 |
ac | -1.803 |
Token | They |
Token ID | 1119 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
riage | +0.773 |
ipples | +0.755 |
ry | +0.743 |
flower | +0.737 |
ac | +0.724 |
Neg logit contributions | |
hy | -0.988 |
Mic | -0.885 |
rides | -0.801 |
Red | -0.792 |
Roger | -0.770 |
Token | make |
Token ID | 787 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
hy | +0.802 |
Mic | +0.744 |
Red | +0.655 |
Roger | +0.652 |
rides | +0.647 |
Neg logit contributions | |
riage | -0.549 |
ipples | -0.543 |
flower | -0.521 |
ry | -0.516 |
ac | -0.507 |
Token | the |
Token ID | 262 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
riage | +0.701 |
ipples | +0.677 |
ry | +0.664 |
flower | +0.660 |
department | +0.646 |
Neg logit contributions | |
hy | -0.878 |
Mic | -0.784 |
rides | -0.714 |
Red | -0.696 |
Roger | -0.679 |
Token | map |
Token ID | 3975 |
Feature activation | +3.33e-01 |
Pos logit contributions | |
riage | +0.162 |
ipples | +0.157 |
ry | +0.152 |
flower | +0.152 |
ac | +0.148 |
Neg logit contributions | |
hy | -0.252 |
Mic | -0.230 |
rides | -0.206 |
Red | -0.205 |
Roger | -0.202 |
Token | colorful |
Token ID | 20239 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
hy | +0.688 |
Mic | +0.644 |
Red | +0.562 |
Roger | +0.559 |
rides | +0.555 |
Neg logit contributions | |
riage | -0.499 |
ipples | -0.493 |
ry | -0.481 |
flower | -0.477 |
ac | -0.463 |
Token | and |
Token ID | 290 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
riage | +0.028 |
ipples | +0.027 |
ry | +0.027 |
flower | +0.026 |
ac | +0.026 |
Neg logit contributions | |
hy | -0.034 |
Mic | -0.031 |
rides | -0.027 |
Red | -0.027 |
Roger | -0.027 |
Token | happy |
Token ID | 3772 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
hy | +0.026 |
Mic | +0.023 |
rides | +0.020 |
Red | +0.020 |
Roger | +0.020 |
Neg logit contributions | |
riage | -0.023 |
ipples | -0.022 |
ry | -0.021 |
flower | -0.021 |
ac | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
riage | +1.503 |
ipples | +1.467 |
ry | +1.430 |
ac | +1.425 |
flower | +1.420 |
Neg logit contributions | |
hy | -1.662 |
Mic | -1.435 |
rides | -1.357 |
Red | -1.282 |
Bow | -1.255 |
Token |
|
Token ID | 198 |
Feature activation | -4.16e-01 |
Pos logit contributions | |
riage | +1.070 |
ipples | +1.050 |
ry | +1.038 |
flower | +1.029 |
ac | +1.010 |
Neg logit contributions | |
hy | -1.202 |
Mic | -1.069 |
rides | -0.960 |
Red | -0.951 |
Roger | -0.924 |
Token | had |
Token ID | 550 |
Feature activation | -4.28e-03 |
Pos logit contributions | |
hy | +0.123 |
Mic | +0.112 |
rides | +0.099 |
Red | +0.098 |
Roger | +0.096 |
Neg logit contributions | |
riage | -0.100 |
ipples | -0.098 |
ry | -0.096 |
flower | -0.095 |
ac | -0.093 |
Token | helped |
Token ID | 4193 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
riage | +0.007 |
ipples | +0.007 |
ry | +0.006 |
flower | +0.006 |
ac | +0.006 |
Neg logit contributions | |
hy | -0.009 |
Mic | -0.008 |
rides | -0.007 |
Red | -0.007 |
Roger | -0.007 |
Token | it |
Token ID | 340 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
hy | +0.632 |
Mic | +0.574 |
rides | +0.516 |
Red | +0.515 |
Roger | +0.504 |
Neg logit contributions | |
riage | -0.388 |
ipples | -0.378 |
ry | -0.370 |
flower | -0.365 |
ac | -0.358 |
Token | feel |
Token ID | 1254 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
riage | +0.717 |
ipples | +0.692 |
ry | +0.672 |
flower | +0.671 |
department | +0.660 |
Neg logit contributions | |
hy | -0.951 |
Mic | -0.852 |
rides | -0.774 |
Red | -0.757 |
Roger | -0.744 |
Token | brave |
Token ID | 14802 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
riage | +0.727 |
ipples | +0.702 |
flower | +0.687 |
ry | +0.686 |
department | +0.671 |
Neg logit contributions | |
hy | -0.941 |
Mic | -0.846 |
rides | -0.763 |
Red | -0.752 |
Roger | -0.743 |
Token | enough |
Token ID | 1576 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
riage | +0.112 |
ipples | +0.109 |
ry | +0.106 |
flower | +0.106 |
ac | +0.103 |
Neg logit contributions | |
hy | -0.151 |
Mic | -0.137 |
rides | -0.122 |
Red | -0.121 |
Roger | -0.119 |
Token | to |
Token ID | 284 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
hy | +0.537 |
Mic | +0.494 |
Red | +0.443 |
rides | +0.440 |
Roger | +0.431 |
Neg logit contributions | |
riage | -0.328 |
ipples | -0.320 |
ry | -0.311 |
flower | -0.304 |
ac | -0.301 |
Token | bloom |
Token ID | 29955 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
riage | +0.859 |
ipples | +0.837 |
ry | +0.817 |
flower | +0.812 |
department | +0.803 |
Neg logit contributions | |
hy | -1.045 |
Mic | -0.933 |
rides | -0.844 |
Red | -0.819 |
Roger | -0.810 |
Token | . |
Token ID | 13 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
riage | +2.042 |
ipples | +1.942 |
department | +1.890 |
flower | +1.889 |
andy | +1.885 |
Neg logit contributions | |
hy | -2.319 |
Mic | -1.997 |
rides | -1.863 |
Red | -1.777 |
Roger | -1.770 |
Token |
|
Token ID | 198 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
riage | +0.286 |
ipples | +0.280 |
ry | +0.273 |
flower | +0.272 |
ac | +0.267 |
Neg logit contributions | |
hy | -0.300 |
Mic | -0.268 |
rides | -0.235 |
Red | -0.233 |
Roger | -0.228 |
Token |
|
Token ID | 198 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
riage | +0.149 |
ipples | +0.146 |
ry | +0.142 |
flower | +0.141 |
ac | +0.139 |
Neg logit contributions | |
hy | -0.169 |
Mic | -0.152 |
rides | -0.134 |
Red | -0.133 |
Roger | -0.130 |
Token | , |
Token ID | 11 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
hy | +2.188 |
Mic | +2.026 |
Red | +1.780 |
rides | +1.764 |
Robert | +1.760 |
Neg logit contributions | |
ry | -1.589 |
riage | -1.555 |
flower | -1.488 |
andy | -1.478 |
ac | -1.461 |
Token | she |
Token ID | 673 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
hy | +1.132 |
Mic | +1.011 |
rides | +0.880 |
Red | +0.859 |
Robert | +0.856 |
Neg logit contributions | |
riage | -1.101 |
flower | -1.067 |
ry | -1.064 |
ipples | -1.052 |
andy | -1.036 |
Token | found |
Token ID | 1043 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
riage | +0.666 |
ipples | +0.647 |
ry | +0.624 |
flower | +0.622 |
department | +0.611 |
Neg logit contributions | |
hy | -0.797 |
Mic | -0.721 |
rides | -0.646 |
Red | -0.634 |
Roger | -0.622 |
Token | a |
Token ID | 257 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
riage | +0.694 |
ipples | +0.677 |
ry | +0.662 |
flower | +0.653 |
ac | +0.643 |
Neg logit contributions | |
hy | -0.752 |
Mic | -0.676 |
rides | -0.596 |
Red | -0.590 |
Roger | -0.578 |
Token | pretty |
Token ID | 2495 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
hy | +0.807 |
Mic | +0.714 |
rides | +0.633 |
Red | +0.617 |
Robert | +0.603 |
Neg logit contributions | |
riage | -0.824 |
flower | -0.810 |
ipples | -0.803 |
ry | -0.798 |
department | -0.791 |
Token | crystal |
Token ID | 15121 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
hy | +0.799 |
Mic | +0.730 |
rides | +0.648 |
Red | +0.643 |
Roger | +0.634 |
Neg logit contributions | |
riage | -0.577 |
flower | -0.567 |
ipples | -0.557 |
ry | -0.556 |
andy | -0.544 |
Token | on |
Token ID | 319 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
hy | +1.648 |
Mic | +1.520 |
Red | +1.343 |
Robert | +1.330 |
Roger | +1.329 |
Neg logit contributions | |
flower | -1.376 |
ry | -1.352 |
riage | -1.346 |
ipples | -1.312 |
andy | -1.272 |
Token | the |
Token ID | 262 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
hy | +0.071 |
Mic | +0.065 |
rides | +0.059 |
Red | +0.059 |
Roger | +0.058 |
Neg logit contributions | |
riage | -0.037 |
ipples | -0.036 |
flower | -0.034 |
ry | -0.034 |
andy | -0.033 |
Token | ground |
Token ID | 2323 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
hy | +2.365 |
Mic | +2.143 |
rides | +1.917 |
Roger | +1.903 |
Red | +1.891 |
Neg logit contributions | |
flower | -1.619 |
riage | -1.607 |
ipples | -1.595 |
andy | -1.512 |
department | -1.502 |
Token | . |
Token ID | 13 |
Feature activation | +4.14e-01 |
Pos logit contributions | |
riage | +0.349 |
ipples | +0.340 |
ry | +0.331 |
flower | +0.329 |
ac | +0.323 |
Neg logit contributions | |
hy | -0.430 |
Mic | -0.386 |
rides | -0.344 |
Red | -0.340 |
Roger | -0.334 |
Token | It |
Token ID | 632 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
hy | +0.750 |
Mic | +0.666 |
rides | +0.583 |
Red | +0.571 |
Roger | +0.569 |
Neg logit contributions | |
riage | -0.733 |
ry | -0.708 |
ipples | -0.708 |
flower | -0.700 |
andy | -0.693 |
Token | . |
Token ID | 13 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
riage | +0.447 |
ipples | +0.436 |
ry | +0.425 |
flower | +0.423 |
ac | +0.415 |
Neg logit contributions | |
hy | -0.499 |
Mic | -0.445 |
rides | -0.396 |
Red | -0.389 |
Roger | -0.383 |
Token | He |
Token ID | 679 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
hy | +0.624 |
Mic | +0.564 |
rides | +0.490 |
Red | +0.487 |
Roger | +0.476 |
Neg logit contributions | |
riage | -0.574 |
ipples | -0.559 |
ry | -0.546 |
flower | -0.535 |
andy | -0.533 |
Token | took |
Token ID | 1718 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
riage | +0.205 |
ipples | +0.200 |
ry | +0.194 |
flower | +0.193 |
ac | +0.188 |
Neg logit contributions | |
hy | -0.295 |
Mic | -0.267 |
rides | -0.240 |
Red | -0.237 |
Roger | -0.234 |
Token | the |
Token ID | 262 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
hy | +2.056 |
Mic | +1.907 |
Red | +1.697 |
nam | +1.659 |
Roger | +1.648 |
Neg logit contributions | |
riage | -1.209 |
ipples | -1.199 |
ac | -1.173 |
ry | -1.148 |
andy | -1.115 |
Token | blue |
Token ID | 4171 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
hy | +0.374 |
Mic | +0.340 |
Red | +0.299 |
rides | +0.299 |
Roger | +0.293 |
Neg logit contributions | |
riage | -0.317 |
ipples | -0.310 |
flower | -0.304 |
ry | -0.300 |
ac | -0.296 |
Token | berry |
Token ID | 8396 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
riage | +0.496 |
ipples | +0.463 |
ry | +0.429 |
ac | +0.411 |
department | +0.410 |
Neg logit contributions | |
hy | -1.671 |
Mic | -1.536 |
rides | -1.432 |
Red | -1.411 |
Roger | -1.390 |
Token | â |
Token ID | 22940 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
hy | +1.726 |
Mic | +1.559 |
Red | +1.421 |
rides | +1.386 |
Roger | +1.382 |
Neg logit contributions | |
ipples | -1.153 |
ry | -1.152 |
flower | -1.151 |
riage | -1.140 |
andy | -1.095 |
Token | € |
Token ID | 26391 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
hy | +0.448 |
Mic | +0.411 |
rides | +0.360 |
Red | +0.355 |
Roger | +0.351 |
Neg logit contributions | |
riage | -0.381 |
ipples | -0.368 |
ry | -0.357 |
flower | -0.356 |
ac | -0.350 |
Token | â„¢ |
Token ID | 8151 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
hy | +3.112 |
Mic | +2.974 |
nam | +2.573 |
Red | +2.546 |
Roger | +2.461 |
Neg logit contributions | |
ry | -2.770 |
riage | -2.749 |
ipples | -2.734 |
flower | -2.661 |
ric | -2.625 |
Token | s |
Token ID | 82 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
hy | +2.351 |
Mic | +2.228 |
rides | +1.994 |
Red | +1.971 |
brace | +1.916 |
Neg logit contributions | |
ry | -1.735 |
riage | -1.723 |
flower | -1.681 |
ipples | -1.677 |
ac | -1.649 |
Token | advice |
Token ID | 5608 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
hy | +0.376 |
Mic | +0.346 |
Red | +0.310 |
rides | +0.309 |
Roger | +0.304 |
Neg logit contributions | |
riage | -0.234 |
ipples | -0.229 |
ry | -0.224 |
flower | -0.222 |
ac | -0.218 |
Token | happily |
Token ID | 18177 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
riage | +0.875 |
ipples | +0.852 |
ry | +0.834 |
flower | +0.829 |
andy | +0.818 |
Neg logit contributions | |
hy | -0.841 |
Mic | -0.744 |
rides | -0.658 |
Red | -0.646 |
Roger | -0.635 |
Token | ever |
Token ID | 1683 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
hy | +1.542 |
Mic | +1.485 |
Red | +1.292 |
Roger | +1.264 |
rides | +1.246 |
Neg logit contributions | |
riage | -1.278 |
ipples | -1.257 |
ry | -1.242 |
flower | -1.221 |
ac | -1.205 |
Token | after |
Token ID | 706 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
hy | +1.742 |
Mic | +1.613 |
Red | +1.444 |
rides | +1.414 |
Roger | +1.410 |
Neg logit contributions | |
riage | -1.281 |
ipples | -1.234 |
ry | -1.203 |
flower | -1.192 |
ac | -1.161 |
Token | . |
Token ID | 13 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
hy | +0.983 |
Mic | +0.910 |
Red | +0.800 |
rides | +0.784 |
Roger | +0.778 |
Neg logit contributions | |
riage | -0.679 |
ipples | -0.661 |
ry | -0.647 |
flower | -0.633 |
andy | -0.626 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
riage | +0.650 |
ipples | +0.640 |
ry | +0.623 |
flower | +0.620 |
ac | +0.610 |
Neg logit contributions | |
hy | -0.673 |
Mic | -0.601 |
rides | -0.527 |
Red | -0.523 |
Roger | -0.512 |
Token | One |
Token ID | 3198 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
riage | +0.403 |
ipples | +0.394 |
ry | +0.384 |
flower | +0.382 |
ac | +0.376 |
Neg logit contributions | |
hy | -0.436 |
Mic | -0.391 |
rides | -0.345 |
Red | -0.341 |
Roger | -0.335 |
Token | day |
Token ID | 1110 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
riage | +1.792 |
department | +1.733 |
ipples | +1.732 |
ry | +1.712 |
flower | +1.690 |
Neg logit contributions | |
hy | -2.227 |
Mic | -1.884 |
rides | -1.786 |
Red | -1.682 |
Roger | -1.664 |
Token | , |
Token ID | 11 |
Feature activation | +4.53e-01 |
Pos logit contributions | |
hy | +2.492 |
Mic | +2.281 |
Red | +2.045 |
rides | +2.018 |
nam | +1.939 |
Neg logit contributions | |
ry | -1.818 |
riage | -1.766 |
ac | -1.747 |
andy | -1.738 |
ipples | -1.731 |
Token | Stanley |
Token ID | 14167 |
Feature activation | +3.38e-01 |
Pos logit contributions | |
hy | +0.925 |
Mic | +0.841 |
rides | +0.755 |
Red | +0.731 |
Roger | +0.716 |
Neg logit contributions | |
riage | -0.700 |
ipples | -0.682 |
flower | -0.672 |
ry | -0.671 |
ac | -0.663 |
Token | and |
Token ID | 290 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
hy | +0.781 |
Mic | +0.716 |
Red | +0.645 |
rides | +0.639 |
nam | +0.631 |
Neg logit contributions | |
riage | -0.429 |
ipples | -0.421 |
ry | -0.418 |
flower | -0.409 |
andy | -0.400 |
Token | his |
Token ID | 465 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
hy | +2.847 |
Mic | +2.503 |
Red | +2.331 |
rides | +2.321 |
con | +2.222 |
Neg logit contributions | |
riage | -1.742 |
flower | -1.661 |
ull | -1.651 |
ac | -1.640 |
ipples | -1.624 |
Token | decided |
Token ID | 3066 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
riage | +0.659 |
ipples | +0.646 |
ry | +0.626 |
flower | +0.624 |
andy | +0.616 |
Neg logit contributions | |
hy | -0.627 |
Mic | -0.553 |
rides | -0.488 |
Red | -0.478 |
Roger | -0.469 |
Token | to |
Token ID | 284 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
riage | +0.405 |
ipples | +0.396 |
ry | +0.382 |
flower | +0.380 |
ac | +0.372 |
Neg logit contributions | |
hy | -0.648 |
Mic | -0.588 |
rides | -0.535 |
Red | -0.526 |
Roger | -0.520 |
Token | lock |
Token ID | 5793 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
riage | +0.850 |
ipples | +0.826 |
ry | +0.806 |
flower | +0.796 |
ac | +0.781 |
Neg logit contributions | |
hy | -1.242 |
Mic | -1.123 |
rides | -1.022 |
Red | -1.000 |
Roger | -0.985 |
Token | it |
Token ID | 340 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
ipples | +1.831 |
riage | +1.771 |
flower | +1.719 |
andy | +1.711 |
ry | +1.640 |
Neg logit contributions | |
hy | -2.746 |
Mic | -2.420 |
rides | -2.319 |
Red | -2.214 |
Bow | -2.210 |
Token | away |
Token ID | 1497 |
Feature activation | -7.28e-01 |
Pos logit contributions | |
hy | +0.460 |
Mic | +0.421 |
Red | +0.374 |
rides | +0.372 |
Roger | +0.368 |
Neg logit contributions | |
riage | -0.306 |
ipples | -0.295 |
ry | -0.294 |
flower | -0.285 |
ac | -0.283 |
Token | so |
Token ID | 523 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
riage | +1.167 |
ipples | +1.149 |
flower | +1.116 |
andy | +1.102 |
ry | +1.093 |
Neg logit contributions | |
hy | -1.437 |
Mic | -1.269 |
rides | -1.156 |
Red | -1.124 |
Roger | -1.108 |
Token | that |
Token ID | 326 |
Feature activation | -9.13e-01 |
Pos logit contributions | |
riage | +0.974 |
ipples | +0.966 |
flower | +0.939 |
ry | +0.930 |
ac | +0.917 |
Neg logit contributions | |
hy | -1.042 |
Mic | -0.917 |
rides | -0.821 |
Red | -0.809 |
Roger | -0.793 |
Token | no |
Token ID | 645 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
ipples | +1.494 |
riage | +1.469 |
flower | +1.436 |
ry | +1.434 |
andy | +1.432 |
Neg logit contributions | |
hy | -1.772 |
Mic | -1.568 |
rides | -1.451 |
Red | -1.399 |
nam | -1.394 |
Token | one |
Token ID | 530 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
riage | +0.744 |
ipples | +0.733 |
ry | +0.698 |
flower | +0.687 |
andy | +0.680 |
Neg logit contributions | |
hy | -1.416 |
Mic | -1.290 |
rides | -1.188 |
Red | -1.167 |
Roger | -1.148 |
Token | could |
Token ID | 714 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
riage | +0.734 |
ipples | +0.716 |
ry | +0.689 |
flower | +0.688 |
andy | +0.678 |
Neg logit contributions | |
hy | -0.941 |
Mic | -0.843 |
rides | -0.760 |
Red | -0.742 |
Roger | -0.731 |
Token | get |
Token ID | 651 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
hy | +2.480 |
Mic | +2.368 |
Red | +2.089 |
nam | +2.077 |
Roger | +2.072 |
Neg logit contributions | |
riage | -1.469 |
ipples | -1.459 |
ry | -1.437 |
flower | -1.424 |
department | -1.419 |
Token | He |
Token ID | 679 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
hy | +0.354 |
Mic | +0.315 |
rides | +0.276 |
Red | +0.270 |
Roger | +0.266 |
Neg logit contributions | |
riage | -0.349 |
ipples | -0.343 |
ry | -0.338 |
flower | -0.329 |
department | -0.328 |
Token | did |
Token ID | 750 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
hy | +0.880 |
Mic | +0.806 |
Red | +0.703 |
rides | +0.700 |
Roger | +0.691 |
Neg logit contributions | |
riage | -0.710 |
ipples | -0.692 |
ry | -0.688 |
flower | -0.677 |
department | -0.677 |
Token | not |
Token ID | 407 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
riage | +0.501 |
ipples | +0.491 |
flower | +0.477 |
ry | +0.477 |
ac | +0.466 |
Neg logit contributions | |
hy | -0.575 |
Mic | -0.513 |
rides | -0.459 |
Red | -0.451 |
Roger | -0.443 |
Token | know |
Token ID | 760 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
riage | +0.052 |
ipples | +0.051 |
ry | +0.049 |
flower | +0.049 |
ac | +0.048 |
Neg logit contributions | |
hy | -0.060 |
Mic | -0.054 |
rides | -0.047 |
Red | -0.047 |
Roger | -0.046 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
riage | +0.469 |
ipples | +0.458 |
flower | +0.438 |
ry | +0.436 |
ac | +0.427 |
Neg logit contributions | |
hy | -0.879 |
Mic | -0.801 |
rides | -0.735 |
Red | -0.724 |
Roger | -0.712 |
Token | and |
Token ID | 290 |
Feature activation | -6.11e-01 |
Pos logit contributions | |
riage | +0.063 |
ipples | +0.062 |
ry | +0.060 |
flower | +0.060 |
ac | +0.059 |
Neg logit contributions | |
hy | -0.082 |
Mic | -0.074 |
rides | -0.066 |
Red | -0.066 |
Roger | -0.064 |
Token | Ben |
Token ID | 3932 |
Feature activation | -8.22e-02 |
Pos logit contributions | |
riage | +0.919 |
ipples | +0.901 |
ry | +0.873 |
flower | +0.869 |
ac | +0.854 |
Neg logit contributions | |
hy | -1.280 |
Mic | -1.146 |
rides | -1.045 |
Red | -1.015 |
Roger | -1.007 |
Token | were |
Token ID | 547 |
Feature activation | +8.00e-01 |
Pos logit contributions | |
riage | +0.127 |
ipples | +0.124 |
ry | +0.121 |
flower | +0.120 |
ac | +0.118 |
Neg logit contributions | |
hy | -0.167 |
Mic | -0.152 |
rides | -0.135 |
Red | -0.134 |
Roger | -0.131 |
Token | in |
Token ID | 287 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
hy | +1.519 |
Mic | +1.390 |
Red | +1.201 |
rides | +1.183 |
Roger | +1.149 |
Neg logit contributions | |
ipples | -1.355 |
riage | -1.337 |
ac | -1.286 |
flower | -1.285 |
department | -1.278 |
Token | the |
Token ID | 262 |
Feature activation | -7.86e-01 |
Pos logit contributions | |
hy | +0.260 |
Mic | +0.236 |
Red | +0.206 |
rides | +0.205 |
Roger | +0.201 |
Neg logit contributions | |
riage | -0.232 |
ipples | -0.226 |
ry | -0.219 |
flower | -0.218 |
ac | -0.217 |
Token | pile |
Token ID | 14540 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
riage | +1.229 |
ipples | +1.204 |
ry | +1.181 |
flower | +1.143 |
ac | +1.142 |
Neg logit contributions | |
hy | -1.644 |
Mic | -1.477 |
rides | -1.336 |
Red | -1.302 |
Roger | -1.274 |
Token | and |
Token ID | 290 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
hy | +2.892 |
Mic | +2.746 |
nam | +2.395 |
Red | +2.387 |
Robert | +2.334 |
Neg logit contributions | |
ipples | -1.987 |
ry | -1.899 |
riage | -1.861 |
andy | -1.836 |
en | -1.821 |
Token | ran |
Token ID | 4966 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
hy | +0.167 |
Mic | +0.150 |
Red | +0.131 |
rides | +0.130 |
Roger | +0.128 |
Neg logit contributions | |
riage | -0.155 |
ipples | -0.152 |
ry | -0.149 |
flower | -0.147 |
ac | -0.144 |
Token | . |
Token ID | 13 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
hy | +2.895 |
Mic | +2.765 |
nam | +2.400 |
Red | +2.394 |
Robert | +2.356 |
Neg logit contributions | |
ipples | -1.982 |
ry | -1.887 |
riage | -1.856 |
andy | -1.811 |
en | -1.802 |
Token | Soon |
Token ID | 15894 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
hy | +1.227 |
Mic | +1.132 |
Red | +0.965 |
Roger | +0.965 |
rides | +0.953 |
Neg logit contributions | |
riage | -1.132 |
ipples | -1.124 |
ry | -1.088 |
department | -1.086 |
flower | -1.049 |
Token | enough |
Token ID | 1576 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
hy | +1.139 |
Mic | +1.042 |
Red | +0.914 |
nam | +0.885 |
Roger | +0.883 |
Neg logit contributions | |
riage | -0.957 |
ipples | -0.947 |
ry | -0.932 |
department | -0.900 |
andy | -0.900 |
Token | he |
Token ID | 339 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
hy | +2.227 |
Mic | +2.063 |
Red | +1.861 |
Roger | +1.744 |
nam | +1.737 |
Neg logit contributions | |
riage | -1.661 |
ipples | -1.645 |
ry | -1.636 |
department | -1.597 |
flower | -1.556 |
Token | found |
Token ID | 1043 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
riage | +0.522 |
ipples | +0.506 |
ry | +0.491 |
flower | +0.490 |
ac | +0.479 |
Neg logit contributions | |
hy | -0.698 |
Mic | -0.629 |
rides | -0.569 |
Red | -0.558 |
Roger | -0.548 |
Token | himself |
Token ID | 2241 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
riage | +0.785 |
ipples | +0.757 |
ry | +0.744 |
flower | +0.741 |
andy | +0.724 |
Neg logit contributions | |
hy | -0.916 |
Mic | -0.818 |
rides | -0.741 |
Red | -0.718 |
Roger | -0.710 |
Token | lost |
Token ID | 2626 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
hy | +0.469 |
Mic | +0.430 |
Red | +0.376 |
rides | +0.369 |
Roger | +0.366 |
Neg logit contributions | |
riage | -0.397 |
ipples | -0.394 |
ry | -0.386 |
flower | -0.381 |
ac | -0.372 |
Token | in |
Token ID | 287 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
riage | +0.312 |
ipples | +0.304 |
ry | +0.296 |
flower | +0.294 |
ac | +0.288 |
Neg logit contributions | |
hy | -0.389 |
Mic | -0.351 |
rides | -0.313 |
Red | -0.309 |
Roger | -0.304 |
Token | a |
Token ID | 257 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
riage | +0.526 |
ipples | +0.511 |
ry | +0.497 |
flower | +0.492 |
ac | +0.481 |
Neg logit contributions | |
hy | -0.833 |
Mic | -0.757 |
rides | -0.690 |
Red | -0.675 |
Roger | -0.669 |
Token | me |
Token ID | 502 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
riage | +0.969 |
ipples | +0.965 |
flower | +0.928 |
ry | +0.913 |
department | +0.892 |
Neg logit contributions | |
hy | -1.310 |
Mic | -1.200 |
rides | -1.080 |
Red | -1.059 |
Roger | -1.037 |
Token | before |
Token ID | 878 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
ipples | +2.536 |
riage | +2.511 |
flower | +2.415 |
ull | +2.329 |
ric | +2.278 |
Neg logit contributions | |
hy | -3.690 |
Mic | -3.330 |
rides | -3.152 |
Red | -2.976 |
nam | -2.941 |
Token | you |
Token ID | 345 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
hy | +0.434 |
Mic | +0.375 |
Red | +0.313 |
rides | +0.309 |
Roger | +0.301 |
Neg logit contributions | |
riage | -0.616 |
ipples | -0.600 |
ry | -0.590 |
flower | -0.585 |
andy | -0.581 |
Token | use |
Token ID | 779 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
hy | +0.396 |
Mic | +0.362 |
rides | +0.322 |
Red | +0.321 |
Roger | +0.319 |
Neg logit contributions | |
riage | -0.260 |
ipples | -0.254 |
ry | -0.248 |
flower | -0.245 |
ac | -0.239 |
Token | something |
Token ID | 1223 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
hy | +0.358 |
Mic | +0.326 |
Red | +0.289 |
rides | +0.289 |
Roger | +0.285 |
Neg logit contributions | |
riage | -0.230 |
ipples | -0.223 |
ry | -0.219 |
flower | -0.215 |
ac | -0.213 |
Token | sharp |
Token ID | 7786 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
hy | +2.304 |
Mic | +2.141 |
Red | +1.854 |
Roger | +1.851 |
Robert | +1.845 |
Neg logit contributions | |
ry | -1.519 |
riage | -1.509 |
ipples | -1.476 |
ac | -1.419 |
department | -1.417 |
Token | . |
Token ID | 13 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
riage | +0.790 |
ipples | +0.770 |
flower | +0.746 |
ry | +0.741 |
andy | +0.734 |
Neg logit contributions | |
hy | -0.963 |
Mic | -0.861 |
rides | -0.778 |
Red | -0.760 |
Roger | -0.743 |
Token | I |
Token ID | 314 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
riage | +1.358 |
ipples | +1.357 |
ry | +1.318 |
flower | +1.318 |
andy | +1.293 |
Neg logit contributions | |
hy | -1.358 |
Mic | -1.213 |
Red | -1.059 |
rides | -1.048 |
Roger | -1.018 |
Token | will |
Token ID | 481 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
riage | +0.795 |
ipples | +0.769 |
flower | +0.734 |
ry | +0.730 |
andy | +0.721 |
Neg logit contributions | |
hy | -1.435 |
Mic | -1.302 |
rides | -1.195 |
Red | -1.177 |
Roger | -1.151 |
Token | help |
Token ID | 1037 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
hy | +0.578 |
Mic | +0.528 |
rides | +0.475 |
Red | +0.472 |
Roger | +0.467 |
Neg logit contributions | |
riage | -0.335 |
ipples | -0.328 |
ry | -0.317 |
flower | -0.314 |
andy | -0.307 |
Token | you |
Token ID | 345 |
Feature activation | -4.25e-01 |
Pos logit contributions | |
hy | +1.368 |
Mic | +1.159 |
rides | +0.966 |
Red | +0.960 |
Roger | +0.957 |
Neg logit contributions | |
riage | -1.748 |
ry | -1.714 |
ipples | -1.702 |
andy | -1.687 |
ac | -1.677 |
Token | Ben |
Token ID | 3932 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
riage | +1.109 |
ipples | +1.095 |
ry | +1.056 |
flower | +1.039 |
ac | +1.036 |
Neg logit contributions | |
hy | -1.467 |
Mic | -1.328 |
rides | -1.201 |
Red | -1.158 |
Roger | -1.145 |
Token | come |
Token ID | 1282 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
hy | +0.987 |
Mic | +0.910 |
Red | +0.802 |
Roger | +0.791 |
rides | +0.790 |
Neg logit contributions | |
riage | -0.699 |
ipples | -0.682 |
ry | -0.658 |
flower | -0.655 |
department | -0.639 |
Token | to |
Token ID | 284 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
riage | +3.180 |
ipples | +3.093 |
andy | +2.897 |
ric | +2.845 |
flower | +2.789 |
Neg logit contributions | |
hy | -3.494 |
Mic | -3.090 |
rides | -2.825 |
Red | -2.665 |
Robert | -2.642 |
Token | the |
Token ID | 262 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
riage | +0.639 |
ipples | +0.635 |
ry | +0.607 |
flower | +0.605 |
ac | +0.595 |
Neg logit contributions | |
hy | -1.182 |
Mic | -1.086 |
rides | -0.992 |
Red | -0.977 |
Roger | -0.968 |
Token | table |
Token ID | 3084 |
Feature activation | -9.28e-01 |
Pos logit contributions | |
riage | +1.035 |
ipples | +1.020 |
ry | +0.982 |
ac | +0.974 |
flower | +0.969 |
Neg logit contributions | |
hy | -1.423 |
Mic | -1.289 |
rides | -1.174 |
Red | -1.149 |
Roger | -1.128 |
Token | . |
Token ID | 13 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
riage | +1.659 |
ipples | +1.598 |
flower | +1.552 |
andy | +1.539 |
ric | +1.528 |
Neg logit contributions | |
hy | -1.687 |
Mic | -1.461 |
rides | -1.353 |
Red | -1.273 |
Robert | -1.262 |
Token | They |
Token ID | 1119 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
flower | +4.358 |
riage | +4.355 |
ipples | +4.323 |
ac | +4.288 |
andy | +4.218 |
Neg logit contributions | |
hy | -3.786 |
Mic | -3.118 |
Red | -2.934 |
rides | -2.918 |
curly | -2.848 |
Token | look |
Token ID | 804 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
hy | +1.293 |
Mic | +1.192 |
Red | +1.057 |
Roger | +1.047 |
rides | +1.036 |
Neg logit contributions | |
riage | -0.850 |
ipples | -0.837 |
ry | -0.810 |
flower | -0.798 |
department | -0.788 |
Token | at |
Token ID | 379 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
hy | +1.494 |
Mic | +1.369 |
Red | +1.189 |
Roger | +1.171 |
rides | +1.160 |
Neg logit contributions | |
riage | -1.133 |
ry | -1.108 |
ipples | -1.086 |
ac | -1.067 |
department | -1.045 |
Token | their |
Token ID | 511 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
riage | +2.039 |
ipples | +2.037 |
ry | +1.938 |
department | +1.926 |
flower | +1.897 |
Neg logit contributions | |
hy | -3.683 |
Mic | -3.391 |
rides | -3.193 |
nam | -3.173 |
Robert | -3.101 |
Token | plates |
Token ID | 13375 |
Feature activation | -3.48e-01 |
Pos logit contributions | |
riage | +0.550 |
ipples | +0.537 |
ry | +0.522 |
flower | +0.519 |
ac | +0.511 |
Neg logit contributions | |
hy | -0.628 |
Mic | -0.563 |
rides | -0.503 |
Red | -0.493 |
Roger | -0.483 |
Token | show |
Token ID | 905 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
riage | +3.064 |
flower | +2.915 |
ipples | +2.893 |
ry | +2.877 |
department | +2.874 |
Neg logit contributions | |
hy | -3.234 |
Mic | -2.899 |
rides | -2.768 |
Roger | -2.563 |
Red | -2.524 |
Token | his |
Token ID | 465 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
riage | +0.452 |
ipples | +0.442 |
flower | +0.428 |
ry | +0.428 |
ac | +0.416 |
Neg logit contributions | |
hy | -0.625 |
Mic | -0.564 |
rides | -0.510 |
Red | -0.500 |
Roger | -0.494 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
riage | +4.273 |
flower | +3.928 |
ipples | +3.923 |
ry | +3.850 |
ac | +3.843 |
Neg logit contributions | |
hy | -4.601 |
rides | -3.993 |
Mic | -3.848 |
long | -3.794 |
nam | -3.610 |
Token | how |
Token ID | 703 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
riage | +1.749 |
ipples | +1.745 |
ry | +1.639 |
flower | +1.629 |
department | +1.627 |
Neg logit contributions | |
hy | -2.264 |
Mic | -1.967 |
rides | -1.842 |
Robert | -1.742 |
Robert | -1.741 |
Token | purple |
Token ID | 14032 |
Feature activation | -9.57e-01 |
Pos logit contributions | |
riage | +0.683 |
ipples | +0.669 |
flower | +0.656 |
ry | +0.655 |
ac | +0.643 |
Neg logit contributions | |
hy | -0.513 |
Mic | -0.443 |
rides | -0.383 |
Red | -0.374 |
Roger | -0.366 |
Token | the |
Token ID | 262 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
riage | +2.232 |
ipples | +2.158 |
ac | +2.105 |
department | +2.104 |
ry | +2.099 |
Neg logit contributions | |
hy | -1.220 |
Mic | -1.002 |
rides | -0.891 |
Red | -0.823 |
Roger | -0.803 |
Token | bath |
Token ID | 7837 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
riage | +3.653 |
ry | +3.403 |
ac | +3.283 |
ric | +3.282 |
flower | +3.236 |
Neg logit contributions | |
hy | -4.510 |
rides | -3.901 |
Mic | -3.881 |
Red | -3.739 |
long | -3.729 |
Token | tub |
Token ID | 37995 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
riage | +1.765 |
ipples | +1.625 |
ac | +1.575 |
ry | +1.575 |
ric | +1.568 |
Neg logit contributions | |
hy | -2.733 |
Mic | -2.411 |
rides | -2.274 |
Red | -2.217 |
Roger | -2.174 |
Token | was |
Token ID | 373 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
hy | +0.555 |
Mic | +0.518 |
Red | +0.464 |
rides | +0.462 |
Roger | +0.457 |
Neg logit contributions | |
riage | -0.259 |
ipples | -0.253 |
ry | -0.244 |
flower | -0.242 |
department | -0.232 |
Token | when |
Token ID | 618 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
riage | +1.819 |
ipples | +1.754 |
ry | +1.706 |
ac | +1.691 |
andy | +1.690 |
Neg logit contributions | |
hy | -1.998 |
Mic | -1.726 |
rides | -1.603 |
Red | -1.543 |
Roger | -1.539 |
Token | it |
Token ID | 340 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
riage | +1.325 |
ipples | +1.309 |
flower | +1.283 |
ry | +1.283 |
ac | +1.271 |
Neg logit contributions | |
hy | -0.708 |
Mic | -0.586 |
rides | -0.500 |
Red | -0.470 |
Roger | -0.467 |
Token | the |
Token ID | 262 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
riage | +1.037 |
ipples | +1.021 |
flower | +0.987 |
ry | +0.987 |
andy | +0.975 |
Neg logit contributions | |
hy | -1.210 |
Mic | -1.082 |
rides | -0.983 |
Red | -0.961 |
Roger | -0.944 |
Token | big |
Token ID | 1263 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
riage | +3.565 |
ry | +3.400 |
flower | +3.340 |
ric | +3.319 |
ipples | +3.317 |
Neg logit contributions | |
hy | -2.426 |
Mic | -2.138 |
rides | -1.974 |
long | -1.852 |
Red | -1.806 |
Token | bubble |
Token ID | 14310 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
riage | +2.287 |
department | +2.106 |
ipples | +2.073 |
ric | +2.055 |
ry | +2.043 |
Neg logit contributions | |
hy | -4.476 |
Mic | -4.085 |
rides | -3.991 |
long | -3.804 |
Red | -3.716 |
Token | and |
Token ID | 290 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
riage | +1.325 |
ipples | +1.254 |
ry | +1.220 |
flower | +1.207 |
ac | +1.202 |
Neg logit contributions | |
hy | -1.677 |
Mic | -1.495 |
rides | -1.376 |
Red | -1.326 |
Roger | -1.311 |
Token | the |
Token ID | 262 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
riage | +2.269 |
ry | +2.233 |
ipples | +2.229 |
andy | +2.173 |
department | +2.097 |
Neg logit contributions | |
hy | -3.286 |
Mic | -2.950 |
rides | -2.751 |
nam | -2.692 |
Bow | -2.690 |
Token | small |
Token ID | 1402 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
riage | +4.034 |
ry | +3.871 |
flower | +3.743 |
department | +3.719 |
ucks | +3.714 |
Neg logit contributions | |
hy | -2.626 |
Mic | -2.301 |
rides | -2.126 |
long | -2.059 |
Red | -1.953 |
Token | bubble |
Token ID | 14310 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
riage | +3.722 |
ric | +3.226 |
department | +3.208 |
ipples | +2.992 |
ac | +2.964 |
Neg logit contributions | |
hy | -6.482 |
long | -6.019 |
rides | -5.862 |
Mic | -5.842 |
curly | -5.518 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
riage | +1.036 |
ipples | +0.984 |
ry | +0.962 |
flower | +0.955 |
ac | +0.944 |
Neg logit contributions | |
hy | -1.254 |
Mic | -1.115 |
rides | -1.020 |
Red | -0.987 |
Roger | -0.970 |
Token | that |
Token ID | 326 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
riage | +2.734 |
flower | +2.733 |
ipples | +2.703 |
andy | +2.632 |
ry | +2.623 |
Neg logit contributions | |
hy | -2.622 |
rides | -2.170 |
Mic | -2.162 |
Bow | -2.026 |
Red | -1.987 |
Token | it |
Token ID | 340 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
riage | +1.713 |
ipples | +1.690 |
ry | +1.638 |
flower | +1.630 |
andy | +1.630 |
Neg logit contributions | |
hy | -2.083 |
Mic | -1.866 |
rides | -1.744 |
Red | -1.660 |
Bow | -1.653 |
Token | was |
Token ID | 373 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
riage | +2.033 |
ipples | +1.919 |
flower | +1.835 |
ry | +1.820 |
ucks | +1.815 |
Neg logit contributions | |
hy | -2.823 |
Mic | -2.581 |
rides | -2.443 |
Red | -2.346 |
Roger | -2.281 |
Token | to |
Token ID | 284 |
Feature activation | -1.00e-02 |
Pos logit contributions | |
riage | +0.061 |
ipples | +0.060 |
ry | +0.058 |
flower | +0.058 |
department | +0.056 |
Neg logit contributions | |
hy | -0.107 |
Mic | -0.098 |
rides | -0.088 |
Red | -0.088 |
Roger | -0.087 |
Token | explore |
Token ID | 7301 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
riage | +0.016 |
ipples | +0.016 |
ry | +0.015 |
flower | +0.015 |
andy | +0.015 |
Neg logit contributions | |
hy | -0.020 |
Mic | -0.018 |
rides | -0.016 |
Red | -0.016 |
Roger | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
hy | +1.090 |
Mic | +0.980 |
Red | +0.877 |
rides | +0.864 |
Robert | +0.860 |
Neg logit contributions | |
riage | -0.634 |
ry | -0.618 |
ipples | -0.617 |
flower | -0.593 |
andy | -0.587 |
Token | He |
Token ID | 679 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
hy | +1.113 |
Mic | +1.015 |
rides | +0.888 |
Roger | +0.882 |
Red | +0.882 |
Neg logit contributions | |
riage | -0.886 |
ipples | -0.860 |
flower | -0.834 |
ry | -0.827 |
andy | -0.816 |
Token | sc |
Token ID | 629 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
riage | +0.418 |
ipples | +0.408 |
ry | +0.397 |
flower | +0.395 |
ac | +0.387 |
Neg logit contributions | |
hy | -0.490 |
Mic | -0.440 |
rides | -0.392 |
Red | -0.386 |
Roger | -0.379 |
Token | ur |
Token ID | 333 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
hy | +1.347 |
Mic | +1.279 |
Red | +1.114 |
Roger | +1.108 |
rides | +1.083 |
Neg logit contributions | |
ipples | -1.317 |
riage | -1.314 |
ry | -1.306 |
ac | -1.266 |
andy | -1.266 |
Token | ried |
Token ID | 2228 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
riage | +3.111 |
ipples | +2.664 |
andy | +2.627 |
ry | +2.591 |
ours | +2.590 |
Neg logit contributions | |
hy | -5.557 |
Mic | -5.143 |
rides | -4.814 |
nam | -4.745 |
ve | -4.652 |
Token | off |
Token ID | 572 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
hy | +1.303 |
Mic | +1.183 |
Red | +1.051 |
Roger | +1.016 |
rides | +1.011 |
Neg logit contributions | |
ipples | -0.953 |
riage | -0.951 |
ry | -0.937 |
flower | -0.925 |
ac | -0.911 |
Token | one |
Token ID | 530 |
Feature activation | -5.54e-01 |
Pos logit contributions | |
hy | +2.449 |
Mic | +2.268 |
Red | +2.075 |
Roger | +2.001 |
rides | +1.980 |
Neg logit contributions | |
ry | -1.223 |
ipples | -1.196 |
riage | -1.179 |
flower | -1.168 |
department | -1.167 |
Token | day |
Token ID | 1110 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
riage | +0.734 |
ipples | +0.708 |
ry | +0.677 |
flower | +0.672 |
department | +0.669 |
Neg logit contributions | |
hy | -1.258 |
Mic | -1.152 |
rides | -1.044 |
Red | -1.024 |
Roger | -1.008 |
Token | and |
Token ID | 290 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
hy | +4.107 |
Mic | +3.825 |
Red | +3.522 |
Roger | +3.392 |
Robert | +3.331 |
Neg logit contributions | |
ry | -2.430 |
ipples | -2.317 |
andy | -2.227 |
ac | -2.160 |
flower | -2.142 |
Token | pl |
Token ID | 458 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
riage | +2.278 |
ipples | +2.236 |
flower | +2.167 |
ry | +2.161 |
ac | +2.144 |
Neg logit contributions | |
hy | -2.788 |
Mic | -2.486 |
rides | -2.263 |
Red | -2.215 |
Roger | -2.192 |
Token | umber |
Token ID | 4494 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
riage | +2.818 |
department | +2.639 |
flower | +2.539 |
andy | +2.460 |
ipples | +2.433 |
Neg logit contributions | |
hy | -4.762 |
Mic | -4.122 |
rides | -3.810 |
Red | -3.694 |
looking | -3.664 |
Token | to |
Token ID | 284 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
riage | +0.420 |
ipples | +0.412 |
ry | +0.399 |
flower | +0.397 |
andy | +0.390 |
Neg logit contributions | |
hy | -0.509 |
Mic | -0.456 |
rides | -0.407 |
Red | -0.401 |
Roger | -0.394 |
Token | get |
Token ID | 651 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
hy | +0.115 |
Mic | +0.105 |
rides | +0.094 |
Red | +0.094 |
Roger | +0.092 |
Neg logit contributions | |
riage | -0.071 |
ipples | -0.068 |
ry | -0.066 |
flower | -0.065 |
ac | -0.064 |
Token | a |
Token ID | 257 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
riage | +1.404 |
ipples | +1.392 |
flower | +1.356 |
andy | +1.336 |
ry | +1.331 |
Neg logit contributions | |
hy | -1.779 |
Mic | -1.586 |
rides | -1.467 |
Red | -1.423 |
Roger | -1.403 |
Token | new |
Token ID | 649 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
riage | +1.663 |
ipples | +1.621 |
flower | +1.553 |
ac | +1.546 |
ry | +1.539 |
Neg logit contributions | |
hy | -2.208 |
Mic | -1.986 |
rides | -1.796 |
Red | -1.771 |
Roger | -1.749 |
Token | tap |
Token ID | 9814 |
Feature activation | +9.27e-02 |
Pos logit contributions | |
riage | +3.522 |
ipples | +3.239 |
ac | +3.166 |
andy | +3.102 |
ric | +3.077 |
Neg logit contributions | |
hy | -4.584 |
rides | -3.967 |
Mic | -3.933 |
long | -3.746 |
con | -3.597 |
Token | ," |
Token ID | 553 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
hy | +0.178 |
Mic | +0.162 |
rides | +0.142 |
Red | +0.141 |
Roger | +0.139 |
Neg logit contributions | |
riage | -0.153 |
ipples | -0.149 |
ry | -0.146 |
flower | -0.145 |
ac | -0.142 |
Token | Mom |
Token ID | 11254 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
hy | +0.450 |
Mic | +0.416 |
rides | +0.381 |
Red | +0.379 |
Roger | +0.374 |
Neg logit contributions | |
riage | -0.171 |
ipples | -0.162 |
ry | -0.156 |
flower | -0.155 |
andy | -0.152 |
Token | says |
Token ID | 1139 |
Feature activation | -1.35e-05 |
Pos logit contributions | |
hy | +0.362 |
Mic | +0.327 |
Red | +0.286 |
rides | +0.284 |
Roger | +0.282 |
Neg logit contributions | |
riage | -0.313 |
ipples | -0.306 |
ry | -0.301 |
flower | -0.298 |
andy | -0.294 |
Token | . |
Token ID | 13 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
riage | +0.000 |
ipples | +0.000 |
ry | +0.000 |
flower | +0.000 |
ac | +0.000 |
Neg logit contributions | |
hy | -0.000 |
Mic | -0.000 |
rides | -0.000 |
Red | -0.000 |
Roger | -0.000 |