Token | � |
Token ID | 129 |
Feature activation | +5.76e+02 |
Pos logit contributions | |
Jones | +0.253 |
White | +0.233 |
faded | +0.230 |
Ancient | +0.227 |
blank | +0.222 |
Neg logit contributions | |
n | -0.366 |
ered | -0.328 |
pause | -0.327 |
hey | -0.318 |
spend | -0.313 |
Token | � |
Token ID | 241 |
Feature activation | +4.76e+02 |
Pos logit contributions | |
interests | +0.659 |
pieces | +0.652 |
victory | +0.601 |
occasions | +0.582 |
nels | +0.572 |
Neg logit contributions | |
whispered | -0.925 |
needed | -0.890 |
said | -0.883 |
meant | -0.881 |
Jones | -0.831 |
Token | I |
Token ID | 40 |
Feature activation | +3.93e+02 |
Pos logit contributions | |
Majesty | +0.849 |
disl | +0.848 |
pieces | +0.847 |
n | +0.838 |
arrest | +0.827 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.428 |
meant | -0.428 |
Jones | -0.380 |
said | -0.375 |
Token | â |
Token ID | 22940 |
Feature activation | +4.40e+02 |
Pos logit contributions | |
n | +0.467 |
disl | +0.459 |
pieces | +0.451 |
interests | +0.433 |
Majesty | +0.432 |
Neg logit contributions | |
needed | -0.614 |
whispered | -0.605 |
meant | -0.586 |
Jones | -0.562 |
said | -0.545 |
Token | € |
Token ID | 26391 |
Feature activation | +4.47e+02 |
Pos logit contributions | |
disl | +0.439 |
pieces | +0.434 |
n | +0.427 |
Majesty | +0.425 |
interests | +0.418 |
Neg logit contributions | |
needed | -0.760 |
whispered | -0.753 |
meant | -0.739 |
Jones | -0.701 |
said | -0.691 |
Token | ™ |
Token ID | 8151 |
Feature activation | +5.09e+02 |
Pos logit contributions | |
pieces | +0.628 |
disl | +0.613 |
victory | +0.606 |
occasions | +0.597 |
iu | +0.594 |
Neg logit contributions | |
needed | -0.606 |
whispered | -0.587 |
Jones | -0.565 |
said | -0.550 |
meant | -0.549 |
Token | m |
Token ID | 76 |
Feature activation | +3.93e+02 |
Pos logit contributions | |
n | +0.863 |
ric | +0.807 |
pieces | +0.777 |
disl | +0.777 |
ieved | +0.769 |
Neg logit contributions | |
whispered | -0.555 |
needed | -0.511 |
faded | -0.509 |
Jones | -0.505 |
White | -0.489 |
Token | sorry |
Token ID | 7926 |
Feature activation | +3.96e+02 |
Pos logit contributions | |
n | +0.585 |
disl | +0.570 |
pieces | +0.550 |
Majesty | +0.548 |
ric | +0.545 |
Neg logit contributions | |
needed | -0.499 |
whispered | -0.494 |
meant | -0.467 |
Jones | -0.452 |
faded | -0.442 |
Token | Jane |
Token ID | 12091 |
Feature activation | +3.86e+02 |
Pos logit contributions | |
n | +0.640 |
disl | +0.633 |
pieces | +0.626 |
interests | +0.606 |
Majesty | +0.605 |
Neg logit contributions | |
needed | -0.451 |
whispered | -0.441 |
meant | -0.423 |
Jones | -0.398 |
said | -0.382 |
Token | . |
Token ID | 13 |
Feature activation | +3.13e+02 |
Pos logit contributions | |
n | +0.607 |
disl | +0.539 |
pieces | +0.534 |
Majesty | +0.532 |
spend | +0.528 |
Neg logit contributions | |
Fuel | -0.422 |
Ancient | -0.414 |
whispered | -0.412 |
blank | -0.406 |
aisle | -0.404 |
Token | I |
Token ID | 314 |
Feature activation | +3.19e+02 |
Pos logit contributions | |
n | +0.438 |
disl | +0.430 |
pieces | +0.410 |
Majesty | +0.406 |
interests | +0.398 |
Neg logit contributions | |
needed | -0.431 |
whispered | -0.422 |
meant | -0.407 |
faded | -0.384 |
chy | -0.379 |
Token | � |
Token ID | 129 |
Feature activation | +5.76e+02 |
Pos logit contributions | |
needed | +0.236 |
whispered | +0.230 |
meant | +0.220 |
Jones | +0.206 |
said | +0.196 |
Neg logit contributions | |
n | -0.377 |
disl | -0.372 |
pieces | -0.368 |
interests | -0.358 |
Majesty | -0.357 |
Token | � |
Token ID | 241 |
Feature activation | +4.76e+02 |
Pos logit contributions | |
interests | +0.659 |
pieces | +0.652 |
victory | +0.601 |
occasions | +0.582 |
nels | +0.572 |
Neg logit contributions | |
whispered | -0.925 |
needed | -0.890 |
said | -0.883 |
meant | -0.881 |
Jones | -0.831 |
Token | That |
Token ID | 2504 |
Feature activation | +3.77e+02 |
Pos logit contributions | |
Majesty | +0.849 |
disl | +0.848 |
pieces | +0.847 |
n | +0.838 |
arrest | +0.827 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.428 |
meant | -0.428 |
Jones | -0.380 |
said | -0.375 |
Token | â |
Token ID | 22940 |
Feature activation | +4.36e+02 |
Pos logit contributions | |
n | +0.544 |
disl | +0.537 |
pieces | +0.530 |
Majesty | +0.512 |
interests | +0.512 |
Neg logit contributions | |
needed | -0.495 |
whispered | -0.485 |
meant | -0.469 |
Jones | -0.443 |
said | -0.429 |
Token | € |
Token ID | 26391 |
Feature activation | +4.43e+02 |
Pos logit contributions | |
n | +0.392 |
disl | +0.382 |
pieces | +0.373 |
interests | +0.353 |
Majesty | +0.351 |
Neg logit contributions | |
needed | -0.809 |
whispered | -0.799 |
meant | -0.778 |
Jones | -0.751 |
said | -0.732 |
Token | ™ |
Token ID | 8151 |
Feature activation | +5.06e+02 |
Pos logit contributions | |
pieces | +0.612 |
disl | +0.599 |
n | +0.588 |
victory | +0.583 |
interests | +0.583 |
Neg logit contributions | |
needed | -0.605 |
whispered | -0.579 |
meant | -0.558 |
Jones | -0.550 |
said | -0.534 |
Token | s |
Token ID | 82 |
Feature activation | +4.05e+02 |
Pos logit contributions | |
n | +0.838 |
ric | +0.776 |
disl | +0.761 |
pieces | +0.746 |
ieved | +0.727 |
Neg logit contributions | |
whispered | -0.602 |
Jones | -0.557 |
faded | -0.544 |
White | -0.535 |
needed | -0.530 |
Token | a |
Token ID | 257 |
Feature activation | +3.82e+02 |
Pos logit contributions | |
pieces | +0.532 |
n | +0.528 |
disl | +0.525 |
iu | +0.517 |
arrest | +0.516 |
Neg logit contributions | |
needed | -0.583 |
whispered | -0.560 |
meant | -0.557 |
said | -0.513 |
Jones | -0.512 |
Token | special |
Token ID | 2041 |
Feature activation | +3.86e+02 |
Pos logit contributions | |
n | +0.562 |
disl | +0.556 |
pieces | +0.547 |
Majesty | +0.532 |
interests | +0.530 |
Neg logit contributions | |
needed | -0.481 |
whispered | -0.476 |
meant | -0.458 |
Jones | -0.435 |
said | -0.414 |
Token | necklace |
Token ID | 38987 |
Feature activation | +3.75e+02 |
Pos logit contributions | |
n | +0.509 |
disl | +0.500 |
pieces | +0.490 |
interests | +0.477 |
Majesty | +0.474 |
Neg logit contributions | |
needed | -0.549 |
whispered | -0.541 |
meant | -0.522 |
Jones | -0.498 |
said | -0.480 |
Token | , |
Token ID | 11 |
Feature activation | +3.09e+02 |
Pos logit contributions | |
n | +0.524 |
pieces | +0.459 |
disl | +0.458 |
interests | +0.450 |
nels | +0.440 |
Neg logit contributions | |
needed | -0.484 |
whispered | -0.481 |
Jones | -0.461 |
bored | -0.458 |
White | -0.452 |
Token | � |
Token ID | 129 |
Feature activation | +5.76e+02 |
Pos logit contributions | |
needed | +0.194 |
whispered | +0.187 |
meant | +0.179 |
Jones | +0.165 |
chy | +0.156 |
Neg logit contributions | |
n | -0.411 |
pieces | -0.407 |
disl | -0.405 |
interests | -0.396 |
iu | -0.395 |
Token | � |
Token ID | 241 |
Feature activation | +4.76e+02 |
Pos logit contributions | |
interests | +0.659 |
pieces | +0.652 |
victory | +0.601 |
occasions | +0.582 |
nels | +0.572 |
Neg logit contributions | |
whispered | -0.925 |
needed | -0.890 |
said | -0.883 |
meant | -0.881 |
Jones | -0.831 |
Token | It |
Token ID | 1026 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
Majesty | +0.849 |
disl | +0.848 |
pieces | +0.847 |
n | +0.838 |
arrest | +0.827 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.428 |
meant | -0.428 |
Jones | -0.380 |
said | -0.375 |
Token | â |
Token ID | 22940 |
Feature activation | +4.36e+02 |
Pos logit contributions | |
n | +0.552 |
disl | +0.533 |
pieces | +0.520 |
interests | +0.506 |
Majesty | +0.502 |
Neg logit contributions | |
whispered | -0.516 |
needed | -0.508 |
Jones | -0.488 |
meant | -0.476 |
faded | -0.465 |
Token | € |
Token ID | 26391 |
Feature activation | +4.43e+02 |
Pos logit contributions | |
n | +0.439 |
disl | +0.435 |
pieces | +0.427 |
Majesty | +0.408 |
interests | +0.407 |
Neg logit contributions | |
needed | -0.758 |
whispered | -0.747 |
meant | -0.728 |
Jones | -0.699 |
said | -0.682 |
Token | ™ |
Token ID | 8151 |
Feature activation | +5.06e+02 |
Pos logit contributions | |
pieces | +0.596 |
disl | +0.585 |
n | +0.573 |
iu | +0.567 |
victory | +0.566 |
Neg logit contributions | |
needed | -0.619 |
whispered | -0.599 |
meant | -0.572 |
Jones | -0.566 |
said | -0.550 |
Token | s |
Token ID | 82 |
Feature activation | +4.04e+02 |
Pos logit contributions | |
n | +0.881 |
ric | +0.828 |
disl | +0.783 |
pieces | +0.782 |
ieved | +0.765 |
Neg logit contributions | |
whispered | -0.549 |
Jones | -0.514 |
faded | -0.496 |
needed | -0.490 |
blank | -0.480 |
Token | a |
Token ID | 257 |
Feature activation | +3.81e+02 |
Pos logit contributions | |
n | +0.531 |
disl | +0.519 |
pieces | +0.505 |
interests | +0.491 |
Majesty | +0.489 |
Neg logit contributions | |
needed | -0.580 |
whispered | -0.577 |
meant | -0.551 |
Jones | -0.536 |
faded | -0.511 |
Token | magic |
Token ID | 5536 |
Feature activation | +3.82e+02 |
Pos logit contributions | |
n | +0.557 |
disl | +0.553 |
pieces | +0.544 |
victory | +0.534 |
Majesty | +0.534 |
Neg logit contributions | |
needed | -0.471 |
whispered | -0.471 |
meant | -0.456 |
Jones | -0.430 |
faded | -0.405 |
Token | book |
Token ID | 1492 |
Feature activation | +3.67e+02 |
Pos logit contributions | |
n | +0.523 |
disl | +0.506 |
Majesty | +0.497 |
pieces | +0.494 |
interests | +0.491 |
Neg logit contributions | |
needed | -0.509 |
whispered | -0.508 |
meant | -0.487 |
Jones | -0.468 |
bored | -0.450 |
Token | ! |
Token ID | 0 |
Feature activation | +3.00e+02 |
Pos logit contributions | |
n | +0.469 |
pieces | +0.386 |
victory | +0.381 |
occasions | +0.378 |
disl | +0.371 |
Neg logit contributions | |
Jones | -0.513 |
whispered | -0.512 |
needed | -0.498 |
Fuel | -0.494 |
bored | -0.490 |
Token | � |
Token ID | 129 |
Feature activation | +5.76e+02 |
Pos logit contributions | |
needed | +0.370 |
whispered | +0.364 |
meant | +0.355 |
Jones | +0.342 |
faded | +0.329 |
Neg logit contributions | |
n | -0.291 |
disl | -0.276 |
pieces | -0.269 |
interests | -0.261 |
iu | -0.257 |
Token | � |
Token ID | 241 |
Feature activation | +4.76e+02 |
Pos logit contributions | |
interests | +0.659 |
pieces | +0.652 |
victory | +0.601 |
occasions | +0.582 |
nels | +0.572 |
Neg logit contributions | |
whispered | -0.925 |
needed | -0.890 |
said | -0.883 |
meant | -0.881 |
Jones | -0.831 |
Token | It |
Token ID | 1026 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
Majesty | +0.849 |
disl | +0.848 |
pieces | +0.847 |
n | +0.838 |
arrest | +0.827 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.428 |
meant | -0.428 |
Jones | -0.380 |
said | -0.375 |
Token | â |
Token ID | 22940 |
Feature activation | +4.36e+02 |
Pos logit contributions | |
n | +0.552 |
disl | +0.533 |
pieces | +0.520 |
interests | +0.506 |
Majesty | +0.502 |
Neg logit contributions | |
whispered | -0.516 |
needed | -0.508 |
Jones | -0.488 |
meant | -0.476 |
faded | -0.465 |
Token | € |
Token ID | 26391 |
Feature activation | +4.43e+02 |
Pos logit contributions | |
n | +0.439 |
disl | +0.435 |
pieces | +0.427 |
Majesty | +0.408 |
interests | +0.407 |
Neg logit contributions | |
needed | -0.758 |
whispered | -0.747 |
meant | -0.728 |
Jones | -0.699 |
said | -0.682 |
Token | ™ |
Token ID | 8151 |
Feature activation | +5.06e+02 |
Pos logit contributions | |
pieces | +0.596 |
disl | +0.585 |
n | +0.573 |
iu | +0.567 |
victory | +0.566 |
Neg logit contributions | |
needed | -0.619 |
whispered | -0.599 |
meant | -0.572 |
Jones | -0.566 |
said | -0.550 |
Token | s |
Token ID | 82 |
Feature activation | +4.04e+02 |
Pos logit contributions | |
n | +0.881 |
ric | +0.828 |
disl | +0.783 |
pieces | +0.782 |
ieved | +0.765 |
Neg logit contributions | |
whispered | -0.549 |
Jones | -0.514 |
faded | -0.496 |
needed | -0.490 |
blank | -0.480 |
Token | almost |
Token ID | 2048 |
Feature activation | +3.87e+02 |
Pos logit contributions | |
n | +0.531 |
disl | +0.519 |
pieces | +0.505 |
interests | +0.491 |
Majesty | +0.489 |
Neg logit contributions | |
needed | -0.580 |
whispered | -0.577 |
meant | -0.551 |
Jones | -0.536 |
faded | -0.511 |
Token | just |
Token ID | 655 |
Feature activation | +3.68e+02 |
Pos logit contributions | |
n | +0.490 |
disl | +0.478 |
pieces | +0.443 |
Majesty | +0.440 |
interests | +0.439 |
Neg logit contributions | |
needed | -0.558 |
whispered | -0.555 |
Jones | -0.544 |
meant | -0.535 |
White | -0.515 |
Token | as |
Token ID | 355 |
Feature activation | +3.33e+02 |
Pos logit contributions | |
n | +0.475 |
disl | +0.462 |
pieces | +0.446 |
interests | +0.437 |
Majesty | +0.433 |
Neg logit contributions | |
needed | -0.525 |
whispered | -0.521 |
meant | -0.503 |
Jones | -0.502 |
White | -0.473 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +3.48e+02 |
Pos logit contributions | |
n | +0.467 |
disl | +0.439 |
victory | +0.437 |
interests | +0.419 |
pieces | +0.418 |
Neg logit contributions | |
whispered | -0.430 |
needed | -0.415 |
Jones | -0.414 |
meant | -0.407 |
spoke | -0.383 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+02 |
Pos logit contributions | |
pieces | +0.954 |
disl | +0.915 |
nels | +0.911 |
latable | +0.901 |
interests | +0.896 |
Neg logit contributions | |
needed | -0.668 |
whispered | -0.652 |
said | -0.621 |
Jones | -0.595 |
meant | -0.584 |
Token | � |
Token ID | 241 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
pieces | +0.575 |
interests | +0.555 |
n | +0.549 |
disl | +0.542 |
victory | +0.525 |
Neg logit contributions | |
whispered | -0.880 |
needed | -0.878 |
meant | -0.849 |
Jones | -0.817 |
said | -0.811 |
Token | That |
Token ID | 2504 |
Feature activation | +3.82e+02 |
Pos logit contributions | |
n | +0.785 |
disl | +0.783 |
pieces | +0.780 |
Majesty | +0.764 |
iu | +0.756 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.466 |
meant | -0.453 |
Jones | -0.417 |
said | -0.404 |
Token | â |
Token ID | 22940 |
Feature activation | +4.36e+02 |
Pos logit contributions | |
n | +0.557 |
disl | +0.550 |
pieces | +0.544 |
iu | +0.526 |
Majesty | +0.525 |
Neg logit contributions | |
needed | -0.495 |
whispered | -0.484 |
meant | -0.469 |
Jones | -0.441 |
said | -0.428 |
Token | € |
Token ID | 26391 |
Feature activation | +4.43e+02 |
Pos logit contributions | |
n | +0.391 |
disl | +0.379 |
pieces | +0.368 |
interests | +0.347 |
iu | +0.346 |
Neg logit contributions | |
needed | -0.813 |
whispered | -0.804 |
meant | -0.780 |
Jones | -0.757 |
said | -0.735 |
Token | ™ |
Token ID | 8151 |
Feature activation | +5.05e+02 |
Pos logit contributions | |
pieces | +0.585 |
disl | +0.579 |
n | +0.577 |
interests | +0.562 |
iu | +0.560 |
Neg logit contributions | |
needed | -0.624 |
whispered | -0.603 |
meant | -0.582 |
Jones | -0.568 |
said | -0.550 |
Token | s |
Token ID | 82 |
Feature activation | +4.03e+02 |
Pos logit contributions | |
n | +0.826 |
ric | +0.768 |
disl | +0.761 |
pieces | +0.741 |
ieved | +0.711 |
Neg logit contributions | |
whispered | -0.614 |
Jones | -0.564 |
faded | -0.543 |
needed | -0.537 |
White | -0.532 |
Token | right |
Token ID | 826 |
Feature activation | +3.82e+02 |
Pos logit contributions | |
pieces | +0.536 |
disl | +0.527 |
n | +0.525 |
iu | +0.523 |
arrest | +0.519 |
Neg logit contributions | |
needed | -0.579 |
meant | -0.554 |
whispered | -0.550 |
said | -0.506 |
Jones | -0.501 |
Token | , |
Token ID | 11 |
Feature activation | +3.32e+02 |
Pos logit contributions | |
n | +0.558 |
disl | +0.546 |
Majesty | +0.519 |
interests | +0.516 |
pieces | +0.514 |
Neg logit contributions | |
needed | -0.490 |
whispered | -0.486 |
meant | -0.454 |
Jones | -0.438 |
faded | -0.426 |
Token | Emma |
Token ID | 18966 |
Feature activation | +3.64e+02 |
Pos logit contributions | |
n | +0.436 |
disl | +0.431 |
pieces | +0.426 |
iu | +0.411 |
interests | +0.411 |
Neg logit contributions | |
needed | -0.475 |
whispered | -0.466 |
meant | -0.452 |
Jones | -0.428 |
said | -0.419 |
Token | . |
Token ID | 13 |
Feature activation | +2.98e+02 |
Pos logit contributions | |
n | +0.505 |
disl | +0.472 |
Majesty | +0.464 |
spend | +0.436 |
interests | +0.434 |
Neg logit contributions | |
whispered | -0.448 |
needed | -0.441 |
faded | -0.440 |
dark | -0.436 |
Fuel | -0.429 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+02 |
Pos logit contributions | |
pieces | +0.954 |
disl | +0.915 |
nels | +0.911 |
latable | +0.901 |
interests | +0.896 |
Neg logit contributions | |
needed | -0.668 |
whispered | -0.652 |
said | -0.621 |
Jones | -0.595 |
meant | -0.584 |
Token | � |
Token ID | 241 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
pieces | +0.575 |
interests | +0.555 |
n | +0.549 |
disl | +0.542 |
victory | +0.525 |
Neg logit contributions | |
whispered | -0.880 |
needed | -0.878 |
meant | -0.849 |
Jones | -0.817 |
said | -0.811 |
Token | That |
Token ID | 2504 |
Feature activation | +3.82e+02 |
Pos logit contributions | |
n | +0.785 |
disl | +0.783 |
pieces | +0.780 |
Majesty | +0.764 |
iu | +0.756 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.466 |
meant | -0.453 |
Jones | -0.417 |
said | -0.404 |
Token | â |
Token ID | 22940 |
Feature activation | +4.36e+02 |
Pos logit contributions | |
n | +0.557 |
disl | +0.550 |
pieces | +0.544 |
iu | +0.526 |
Majesty | +0.525 |
Neg logit contributions | |
needed | -0.495 |
whispered | -0.484 |
meant | -0.469 |
Jones | -0.441 |
said | -0.428 |
Token | € |
Token ID | 26391 |
Feature activation | +4.43e+02 |
Pos logit contributions | |
n | +0.391 |
disl | +0.379 |
pieces | +0.368 |
interests | +0.347 |
iu | +0.346 |
Neg logit contributions | |
needed | -0.813 |
whispered | -0.804 |
meant | -0.780 |
Jones | -0.757 |
said | -0.735 |
Token | ™ |
Token ID | 8151 |
Feature activation | +5.05e+02 |
Pos logit contributions | |
pieces | +0.585 |
disl | +0.579 |
n | +0.577 |
interests | +0.562 |
iu | +0.560 |
Neg logit contributions | |
needed | -0.624 |
whispered | -0.603 |
meant | -0.582 |
Jones | -0.568 |
said | -0.550 |
Token | s |
Token ID | 82 |
Feature activation | +4.03e+02 |
Pos logit contributions | |
n | +0.826 |
ric | +0.768 |
disl | +0.761 |
pieces | +0.741 |
ieved | +0.711 |
Neg logit contributions | |
whispered | -0.614 |
Jones | -0.564 |
faded | -0.543 |
needed | -0.537 |
White | -0.532 |
Token | the |
Token ID | 262 |
Feature activation | +3.89e+02 |
Pos logit contributions | |
pieces | +0.536 |
disl | +0.527 |
n | +0.525 |
iu | +0.523 |
arrest | +0.519 |
Neg logit contributions | |
needed | -0.579 |
meant | -0.554 |
whispered | -0.550 |
said | -0.506 |
Jones | -0.501 |
Token | gas |
Token ID | 3623 |
Feature activation | +3.89e+02 |
Pos logit contributions | |
n | +0.569 |
disl | +0.559 |
pieces | +0.550 |
Majesty | +0.534 |
interests | +0.533 |
Neg logit contributions | |
needed | -0.499 |
whispered | -0.493 |
meant | -0.472 |
Jones | -0.451 |
faded | -0.430 |
Token | from |
Token ID | 422 |
Feature activation | +3.39e+02 |
Pos logit contributions | |
n | +0.565 |
disl | +0.536 |
ric | +0.530 |
interests | +0.517 |
nels | +0.516 |
Neg logit contributions | |
whispered | -0.499 |
needed | -0.478 |
meant | -0.463 |
Jones | -0.451 |
faded | -0.440 |
Token | the |
Token ID | 262 |
Feature activation | +3.40e+02 |
Pos logit contributions | |
n | +0.415 |
disl | +0.403 |
pieces | +0.392 |
arrest | +0.383 |
Majesty | +0.379 |
Neg logit contributions | |
needed | -0.517 |
whispered | -0.510 |
meant | -0.494 |
Jones | -0.478 |
faded | -0.457 |
Token | � |
Token ID | 129 |
Feature activation | +5.76e+02 |
Pos logit contributions | |
whispered | +0.431 |
Ancient | +0.400 |
meant | +0.390 |
faded | +0.387 |
Fuel | +0.386 |
Neg logit contributions | |
n | -0.238 |
victory | -0.218 |
job | -0.212 |
blast | -0.202 |
spend | -0.197 |
Token | � |
Token ID | 241 |
Feature activation | +4.76e+02 |
Pos logit contributions | |
interests | +0.659 |
pieces | +0.652 |
victory | +0.601 |
occasions | +0.582 |
nels | +0.572 |
Neg logit contributions | |
whispered | -0.925 |
needed | -0.890 |
said | -0.883 |
meant | -0.881 |
Jones | -0.831 |
Token | Let |
Token ID | 5756 |
Feature activation | +3.69e+02 |
Pos logit contributions | |
Majesty | +0.849 |
disl | +0.848 |
pieces | +0.847 |
n | +0.838 |
arrest | +0.827 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.428 |
meant | -0.428 |
Jones | -0.380 |
said | -0.375 |
Token | â |
Token ID | 22940 |
Feature activation | +4.17e+02 |
Pos logit contributions | |
ric | +0.670 |
iu | +0.662 |
Majesty | +0.659 |
disl | +0.650 |
latable | +0.648 |
Neg logit contributions | |
said | -0.389 |
meant | -0.388 |
needed | -0.386 |
whispered | -0.378 |
spoke | -0.314 |
Token | € |
Token ID | 26391 |
Feature activation | +4.28e+02 |
Pos logit contributions | |
Majesty | +0.403 |
disl | +0.390 |
pieces | +0.388 |
interests | +0.386 |
arrest | +0.376 |
Neg logit contributions | |
needed | -0.727 |
whispered | -0.721 |
meant | -0.715 |
said | -0.670 |
Jones | -0.665 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.93e+02 |
Pos logit contributions | |
pieces | +0.671 |
occasions | +0.666 |
disl | +0.660 |
victory | +0.657 |
iu | +0.654 |
Neg logit contributions | |
needed | -0.505 |
whispered | -0.479 |
said | -0.463 |
Jones | -0.461 |
meant | -0.446 |
Token | s |
Token ID | 82 |
Feature activation | +3.92e+02 |
Pos logit contributions | |
n | +0.904 |
pieces | +0.833 |
ric | +0.825 |
disl | +0.813 |
ieved | +0.794 |
Neg logit contributions | |
whispered | -0.483 |
needed | -0.455 |
faded | -0.435 |
Jones | -0.430 |
meant | -0.426 |
Token | go |
Token ID | 467 |
Feature activation | +3.58e+02 |
Pos logit contributions | |
iu | +0.539 |
latable | +0.513 |
pieces | +0.506 |
disl | +0.500 |
arrest | +0.494 |
Neg logit contributions | |
needed | -0.596 |
said | -0.577 |
whispered | -0.573 |
meant | -0.573 |
Jones | -0.495 |
Token | find |
Token ID | 1064 |
Feature activation | +3.41e+02 |
Pos logit contributions | |
disl | +0.587 |
iu | +0.573 |
Majesty | +0.559 |
interests | +0.558 |
pieces | +0.557 |
Neg logit contributions | |
whispered | -0.429 |
needed | -0.420 |
said | -0.400 |
meant | -0.393 |
Jones | -0.382 |
Token | something |
Token ID | 1223 |
Feature activation | +3.33e+02 |
Pos logit contributions | |
disl | +0.475 |
n | +0.468 |
pieces | +0.463 |
iu | +0.461 |
Majesty | +0.453 |
Neg logit contributions | |
needed | -0.466 |
whispered | -0.465 |
meant | -0.449 |
said | -0.428 |
Jones | -0.417 |
Token | else |
Token ID | 2073 |
Feature activation | +3.53e+02 |
Pos logit contributions | |
n | +0.439 |
disl | +0.421 |
victory | +0.402 |
pieces | +0.397 |
interests | +0.394 |
Neg logit contributions | |
Jones | -0.472 |
whispered | -0.465 |
needed | -0.450 |
White | -0.443 |
faded | -0.425 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+02 |
Pos logit contributions | |
pieces | +0.954 |
disl | +0.915 |
nels | +0.911 |
latable | +0.901 |
interests | +0.896 |
Neg logit contributions | |
needed | -0.668 |
whispered | -0.652 |
said | -0.621 |
Jones | -0.595 |
meant | -0.584 |
Token | � |
Token ID | 241 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
pieces | +0.575 |
interests | +0.555 |
n | +0.549 |
disl | +0.542 |
victory | +0.525 |
Neg logit contributions | |
whispered | -0.880 |
needed | -0.878 |
meant | -0.849 |
Jones | -0.817 |
said | -0.811 |
Token | Let |
Token ID | 5756 |
Feature activation | +3.77e+02 |
Pos logit contributions | |
n | +0.785 |
disl | +0.783 |
pieces | +0.780 |
Majesty | +0.764 |
iu | +0.756 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.466 |
meant | -0.453 |
Jones | -0.417 |
said | -0.404 |
Token | â |
Token ID | 22940 |
Feature activation | +4.18e+02 |
Pos logit contributions | |
ric | +0.681 |
iu | +0.662 |
latable | +0.645 |
Majesty | +0.640 |
disl | +0.627 |
Neg logit contributions | |
meant | -0.427 |
said | -0.425 |
needed | -0.412 |
whispered | -0.410 |
spoke | -0.348 |
Token | € |
Token ID | 26391 |
Feature activation | +4.29e+02 |
Pos logit contributions | |
Majesty | +0.383 |
pieces | +0.381 |
disl | +0.377 |
interests | +0.376 |
arrest | +0.367 |
Neg logit contributions | |
needed | -0.742 |
whispered | -0.735 |
meant | -0.725 |
said | -0.682 |
Jones | -0.681 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.92e+02 |
Pos logit contributions | |
pieces | +0.609 |
disl | +0.599 |
iu | +0.596 |
occasions | +0.595 |
victory | +0.591 |
Neg logit contributions | |
needed | -0.565 |
whispered | -0.537 |
Jones | -0.518 |
said | -0.511 |
meant | -0.508 |
Token | s |
Token ID | 82 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
n | +0.892 |
pieces | +0.831 |
ric | +0.819 |
disl | +0.815 |
interests | +0.789 |
Neg logit contributions | |
whispered | -0.491 |
needed | -0.461 |
Jones | -0.434 |
meant | -0.432 |
faded | -0.432 |
Token | explore |
Token ID | 7301 |
Feature activation | +3.69e+02 |
Pos logit contributions | |
iu | +0.521 |
latable | +0.479 |
pieces | +0.473 |
disl | +0.465 |
earned | +0.457 |
Neg logit contributions | |
needed | -0.616 |
meant | -0.599 |
said | -0.594 |
whispered | -0.590 |
Jones | -0.520 |
Token | and |
Token ID | 290 |
Feature activation | +3.20e+02 |
Pos logit contributions | |
n | +0.528 |
disl | +0.525 |
pieces | +0.515 |
iu | +0.504 |
Majesty | +0.502 |
Neg logit contributions | |
needed | -0.486 |
whispered | -0.479 |
meant | -0.461 |
Jones | -0.436 |
said | -0.427 |
Token | have |
Token ID | 423 |
Feature activation | +3.15e+02 |
Pos logit contributions | |
iu | +0.476 |
disl | +0.476 |
Majesty | +0.456 |
latable | +0.455 |
ric | +0.453 |
Neg logit contributions | |
needed | -0.441 |
whispered | -0.436 |
said | -0.418 |
meant | -0.401 |
spoke | -0.371 |
Token | some |
Token ID | 617 |
Feature activation | +3.08e+02 |
Pos logit contributions | |
disl | +0.466 |
n | +0.461 |
pieces | +0.459 |
iu | +0.458 |
ric | +0.448 |
Neg logit contributions | |
needed | -0.404 |
whispered | -0.398 |
meant | -0.380 |
said | -0.359 |
dark | -0.346 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+02 |
Pos logit contributions | |
pieces | +0.954 |
disl | +0.915 |
nels | +0.911 |
latable | +0.901 |
interests | +0.896 |
Neg logit contributions | |
needed | -0.668 |
whispered | -0.652 |
said | -0.621 |
Jones | -0.595 |
meant | -0.584 |
Token | � |
Token ID | 241 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
pieces | +0.575 |
interests | +0.555 |
n | +0.549 |
disl | +0.542 |
victory | +0.525 |
Neg logit contributions | |
whispered | -0.880 |
needed | -0.878 |
meant | -0.849 |
Jones | -0.817 |
said | -0.811 |
Token | Let |
Token ID | 5756 |
Feature activation | +3.77e+02 |
Pos logit contributions | |
n | +0.785 |
disl | +0.783 |
pieces | +0.780 |
Majesty | +0.764 |
iu | +0.756 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.466 |
meant | -0.453 |
Jones | -0.417 |
said | -0.404 |
Token | â |
Token ID | 22940 |
Feature activation | +4.18e+02 |
Pos logit contributions | |
ric | +0.681 |
iu | +0.662 |
latable | +0.645 |
Majesty | +0.640 |
disl | +0.627 |
Neg logit contributions | |
meant | -0.427 |
said | -0.425 |
needed | -0.412 |
whispered | -0.410 |
spoke | -0.348 |
Token | € |
Token ID | 26391 |
Feature activation | +4.29e+02 |
Pos logit contributions | |
Majesty | +0.383 |
pieces | +0.381 |
disl | +0.377 |
interests | +0.376 |
arrest | +0.367 |
Neg logit contributions | |
needed | -0.742 |
whispered | -0.735 |
meant | -0.725 |
said | -0.682 |
Jones | -0.681 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.92e+02 |
Pos logit contributions | |
pieces | +0.609 |
disl | +0.599 |
iu | +0.596 |
occasions | +0.595 |
victory | +0.591 |
Neg logit contributions | |
needed | -0.565 |
whispered | -0.537 |
Jones | -0.518 |
said | -0.511 |
meant | -0.508 |
Token | s |
Token ID | 82 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
n | +0.892 |
pieces | +0.831 |
ric | +0.819 |
disl | +0.815 |
interests | +0.789 |
Neg logit contributions | |
whispered | -0.491 |
needed | -0.461 |
Jones | -0.434 |
meant | -0.432 |
faded | -0.432 |
Token | have |
Token ID | 423 |
Feature activation | +3.60e+02 |
Pos logit contributions | |
iu | +0.521 |
latable | +0.479 |
pieces | +0.473 |
disl | +0.465 |
earned | +0.457 |
Neg logit contributions | |
needed | -0.616 |
meant | -0.599 |
said | -0.594 |
whispered | -0.590 |
Jones | -0.520 |
Token | a |
Token ID | 257 |
Feature activation | +3.53e+02 |
Pos logit contributions | |
pieces | +0.539 |
iu | +0.538 |
disl | +0.535 |
ric | +0.532 |
n | +0.531 |
Neg logit contributions | |
needed | -0.452 |
whispered | -0.445 |
meant | -0.424 |
said | -0.409 |
dark | -0.391 |
Token | stretching |
Token ID | 20880 |
Feature activation | +3.54e+02 |
Pos logit contributions | |
n | +0.483 |
disl | +0.477 |
pieces | +0.468 |
iu | +0.457 |
arrest | +0.453 |
Neg logit contributions | |
needed | -0.491 |
whispered | -0.481 |
meant | -0.461 |
Jones | -0.439 |
said | -0.431 |
Token | race |
Token ID | 3234 |
Feature activation | +3.32e+02 |
Pos logit contributions | |
disl | +0.528 |
pieces | +0.519 |
n | +0.518 |
iu | +0.504 |
Majesty | +0.501 |
Neg logit contributions | |
needed | -0.456 |
whispered | -0.443 |
meant | -0.429 |
said | -0.400 |
Jones | -0.399 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+02 |
Pos logit contributions | |
pieces | +0.954 |
disl | +0.915 |
nels | +0.911 |
latable | +0.901 |
interests | +0.896 |
Neg logit contributions | |
needed | -0.668 |
whispered | -0.652 |
said | -0.621 |
Jones | -0.595 |
meant | -0.584 |
Token | � |
Token ID | 241 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
pieces | +0.575 |
interests | +0.555 |
n | +0.549 |
disl | +0.542 |
victory | +0.525 |
Neg logit contributions | |
whispered | -0.880 |
needed | -0.878 |
meant | -0.849 |
Jones | -0.817 |
said | -0.811 |
Token | Let |
Token ID | 5756 |
Feature activation | +3.77e+02 |
Pos logit contributions | |
n | +0.785 |
disl | +0.783 |
pieces | +0.780 |
Majesty | +0.764 |
iu | +0.756 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.466 |
meant | -0.453 |
Jones | -0.417 |
said | -0.404 |
Token | â |
Token ID | 22940 |
Feature activation | +4.18e+02 |
Pos logit contributions | |
ric | +0.681 |
iu | +0.662 |
latable | +0.645 |
Majesty | +0.640 |
disl | +0.627 |
Neg logit contributions | |
meant | -0.427 |
said | -0.425 |
needed | -0.412 |
whispered | -0.410 |
spoke | -0.348 |
Token | € |
Token ID | 26391 |
Feature activation | +4.29e+02 |
Pos logit contributions | |
Majesty | +0.383 |
pieces | +0.381 |
disl | +0.377 |
interests | +0.376 |
arrest | +0.367 |
Neg logit contributions | |
needed | -0.742 |
whispered | -0.735 |
meant | -0.725 |
said | -0.682 |
Jones | -0.681 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.92e+02 |
Pos logit contributions | |
pieces | +0.609 |
disl | +0.599 |
iu | +0.596 |
occasions | +0.595 |
victory | +0.591 |
Neg logit contributions | |
needed | -0.565 |
whispered | -0.537 |
Jones | -0.518 |
said | -0.511 |
meant | -0.508 |
Token | s |
Token ID | 82 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
n | +0.892 |
pieces | +0.831 |
ric | +0.819 |
disl | +0.815 |
interests | +0.789 |
Neg logit contributions | |
whispered | -0.491 |
needed | -0.461 |
Jones | -0.434 |
meant | -0.432 |
faded | -0.432 |
Token | play |
Token ID | 711 |
Feature activation | +3.71e+02 |
Pos logit contributions | |
iu | +0.521 |
latable | +0.479 |
pieces | +0.473 |
disl | +0.465 |
earned | +0.457 |
Neg logit contributions | |
needed | -0.616 |
meant | -0.599 |
said | -0.594 |
whispered | -0.590 |
Jones | -0.520 |
Token | a |
Token ID | 257 |
Feature activation | +3.52e+02 |
Pos logit contributions | |
disl | +0.558 |
iu | +0.545 |
n | +0.539 |
pieces | +0.538 |
Majesty | +0.531 |
Neg logit contributions | |
whispered | -0.457 |
needed | -0.454 |
meant | -0.432 |
said | -0.418 |
Jones | -0.403 |
Token | joke |
Token ID | 9707 |
Feature activation | +3.52e+02 |
Pos logit contributions | |
n | +0.493 |
disl | +0.493 |
pieces | +0.478 |
iu | +0.470 |
interests | +0.468 |
Neg logit contributions | |
needed | -0.472 |
whispered | -0.463 |
meant | -0.439 |
Jones | -0.420 |
said | -0.414 |
Token | . |
Token ID | 13 |
Feature activation | +2.88e+02 |
Pos logit contributions | |
n | +0.454 |
pieces | +0.397 |
disl | +0.388 |
nels | +0.379 |
victory | +0.377 |
Neg logit contributions | |
needed | -0.538 |
whispered | -0.510 |
faded | -0.496 |
Jones | -0.489 |
White | -0.479 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+02 |
Pos logit contributions | |
pieces | +0.954 |
disl | +0.915 |
nels | +0.911 |
latable | +0.901 |
interests | +0.896 |
Neg logit contributions | |
needed | -0.668 |
whispered | -0.652 |
said | -0.621 |
Jones | -0.595 |
meant | -0.584 |
Token | � |
Token ID | 241 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
pieces | +0.575 |
interests | +0.555 |
n | +0.549 |
disl | +0.542 |
victory | +0.525 |
Neg logit contributions | |
whispered | -0.880 |
needed | -0.878 |
meant | -0.849 |
Jones | -0.817 |
said | -0.811 |
Token | Let |
Token ID | 5756 |
Feature activation | +3.77e+02 |
Pos logit contributions | |
n | +0.785 |
disl | +0.783 |
pieces | +0.780 |
Majesty | +0.764 |
iu | +0.756 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.466 |
meant | -0.453 |
Jones | -0.417 |
said | -0.404 |
Token | â |
Token ID | 22940 |
Feature activation | +4.18e+02 |
Pos logit contributions | |
ric | +0.681 |
iu | +0.662 |
latable | +0.645 |
Majesty | +0.640 |
disl | +0.627 |
Neg logit contributions | |
meant | -0.427 |
said | -0.425 |
needed | -0.412 |
whispered | -0.410 |
spoke | -0.348 |
Token | € |
Token ID | 26391 |
Feature activation | +4.29e+02 |
Pos logit contributions | |
Majesty | +0.383 |
pieces | +0.381 |
disl | +0.377 |
interests | +0.376 |
arrest | +0.367 |
Neg logit contributions | |
needed | -0.742 |
whispered | -0.735 |
meant | -0.725 |
said | -0.682 |
Jones | -0.681 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.92e+02 |
Pos logit contributions | |
pieces | +0.609 |
disl | +0.599 |
iu | +0.596 |
occasions | +0.595 |
victory | +0.591 |
Neg logit contributions | |
needed | -0.565 |
whispered | -0.537 |
Jones | -0.518 |
said | -0.511 |
meant | -0.508 |
Token | s |
Token ID | 82 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
n | +0.892 |
pieces | +0.831 |
ric | +0.819 |
disl | +0.815 |
interests | +0.789 |
Neg logit contributions | |
whispered | -0.491 |
needed | -0.461 |
Jones | -0.434 |
meant | -0.432 |
faded | -0.432 |
Token | not |
Token ID | 407 |
Feature activation | +3.72e+02 |
Pos logit contributions | |
iu | +0.521 |
latable | +0.479 |
pieces | +0.473 |
disl | +0.465 |
earned | +0.457 |
Neg logit contributions | |
needed | -0.616 |
meant | -0.599 |
said | -0.594 |
whispered | -0.590 |
Jones | -0.520 |
Token | think |
Token ID | 892 |
Feature activation | +3.53e+02 |
Pos logit contributions | |
iu | +0.470 |
latable | +0.461 |
Majesty | +0.459 |
disl | +0.457 |
pieces | +0.455 |
Neg logit contributions | |
needed | -0.578 |
meant | -0.559 |
whispered | -0.551 |
said | -0.511 |
dark | -0.492 |
Token | about |
Token ID | 546 |
Feature activation | +3.23e+02 |
Pos logit contributions | |
n | +0.465 |
disl | +0.443 |
pieces | +0.442 |
interests | +0.424 |
victory | +0.416 |
Neg logit contributions | |
needed | -0.510 |
whispered | -0.504 |
meant | -0.481 |
Jones | -0.471 |
faded | -0.450 |
Token | the |
Token ID | 262 |
Feature activation | +3.24e+02 |
Pos logit contributions | |
pieces | +0.393 |
iu | +0.391 |
n | +0.390 |
disl | +0.388 |
Majesty | +0.375 |
Neg logit contributions | |
needed | -0.482 |
whispered | -0.477 |
meant | -0.466 |
Jones | -0.440 |
said | -0.438 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+02 |
Pos logit contributions | |
pieces | +0.954 |
disl | +0.915 |
nels | +0.911 |
latable | +0.901 |
interests | +0.896 |
Neg logit contributions | |
needed | -0.668 |
whispered | -0.652 |
said | -0.621 |
Jones | -0.595 |
meant | -0.584 |
Token | � |
Token ID | 241 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
pieces | +0.575 |
interests | +0.555 |
n | +0.549 |
disl | +0.542 |
victory | +0.525 |
Neg logit contributions | |
whispered | -0.880 |
needed | -0.878 |
meant | -0.849 |
Jones | -0.817 |
said | -0.811 |
Token | Who |
Token ID | 8241 |
Feature activation | +3.80e+02 |
Pos logit contributions | |
n | +0.785 |
disl | +0.783 |
pieces | +0.780 |
Majesty | +0.764 |
iu | +0.756 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.466 |
meant | -0.453 |
Jones | -0.417 |
said | -0.404 |
Token | â |
Token ID | 22940 |
Feature activation | +4.15e+02 |
Pos logit contributions | |
n | +0.566 |
disl | +0.565 |
pieces | +0.556 |
iu | +0.545 |
Majesty | +0.540 |
Neg logit contributions | |
needed | -0.480 |
whispered | -0.467 |
meant | -0.455 |
Jones | -0.416 |
said | -0.416 |
Token | € |
Token ID | 26391 |
Feature activation | +4.27e+02 |
Pos logit contributions | |
n | +0.388 |
disl | +0.385 |
pieces | +0.381 |
interests | +0.364 |
Majesty | +0.363 |
Neg logit contributions | |
needed | -0.749 |
whispered | -0.738 |
meant | -0.722 |
Jones | -0.692 |
said | -0.677 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.90e+02 |
Pos logit contributions | |
n | +0.603 |
disl | +0.602 |
pieces | +0.599 |
interests | +0.579 |
iu | +0.578 |
Neg logit contributions | |
needed | -0.565 |
whispered | -0.550 |
meant | -0.529 |
Jones | -0.510 |
said | -0.492 |
Token | s |
Token ID | 82 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
n | +0.870 |
ric | +0.821 |
pieces | +0.799 |
disl | +0.792 |
ieved | +0.758 |
Neg logit contributions | |
whispered | -0.502 |
faded | -0.455 |
needed | -0.452 |
Jones | -0.447 |
White | -0.429 |
Token | there |
Token ID | 612 |
Feature activation | +3.59e+02 |
Pos logit contributions | |
n | +0.466 |
disl | +0.464 |
pieces | +0.460 |
iu | +0.448 |
interests | +0.442 |
Neg logit contributions | |
needed | -0.611 |
whispered | -0.595 |
meant | -0.585 |
Jones | -0.553 |
said | -0.545 |
Token | ? |
Token ID | 30 |
Feature activation | +3.15e+02 |
Pos logit contributions | |
n | +0.378 |
disl | +0.327 |
ric | +0.326 |
pieces | +0.316 |
pause | +0.312 |
Neg logit contributions | |
whispered | -0.610 |
needed | -0.594 |
faded | -0.579 |
Jones | -0.560 |
meant | -0.556 |
Token | â |
Token ID | 22940 |
Feature activation | +3.54e+02 |
Pos logit contributions | |
n | +0.435 |
pieces | +0.404 |
disl | +0.401 |
Majesty | +0.379 |
ric | +0.378 |
Neg logit contributions | |
needed | -0.442 |
whispered | -0.441 |
meant | -0.421 |
faded | -0.415 |
chy | -0.408 |
Token | € |
Token ID | 26391 |
Feature activation | +3.64e+02 |
Pos logit contributions | |
n | +0.302 |
disl | +0.269 |
ric | +0.226 |
iu | +0.225 |
ieved | +0.223 |
Neg logit contributions | |
needed | -0.717 |
whispered | -0.697 |
meant | -0.675 |
faded | -0.674 |
Jones | -0.663 |
Token | � |
Token ID | 241 |
Feature activation | +4.76e+02 |
Pos logit contributions | |
interests | +0.659 |
pieces | +0.652 |
victory | +0.601 |
occasions | +0.582 |
nels | +0.572 |
Neg logit contributions | |
whispered | -0.925 |
needed | -0.890 |
said | -0.883 |
meant | -0.881 |
Jones | -0.831 |
Token | Please |
Token ID | 5492 |
Feature activation | +4.03e+02 |
Pos logit contributions | |
Majesty | +0.849 |
disl | +0.848 |
pieces | +0.847 |
n | +0.838 |
arrest | +0.827 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.428 |
meant | -0.428 |
Jones | -0.380 |
said | -0.375 |
Token | don |
Token ID | 836 |
Feature activation | +3.88e+02 |
Pos logit contributions | |
n | +0.510 |
disl | +0.509 |
pieces | +0.504 |
iu | +0.484 |
Majesty | +0.482 |
Neg logit contributions | |
needed | -0.597 |
whispered | -0.585 |
meant | -0.567 |
Jones | -0.541 |
said | -0.529 |
Token | â |
Token ID | 22940 |
Feature activation | +4.22e+02 |
Pos logit contributions | |
Majesty | +0.628 |
disl | +0.605 |
interests | +0.595 |
pieces | +0.578 |
iu | +0.564 |
Neg logit contributions | |
needed | -0.503 |
meant | -0.501 |
whispered | -0.470 |
said | -0.458 |
Jones | -0.427 |
Token | € |
Token ID | 26391 |
Feature activation | +4.28e+02 |
Pos logit contributions | |
disl | +0.420 |
Majesty | +0.418 |
pieces | +0.418 |
interests | +0.401 |
arrest | +0.397 |
Neg logit contributions | |
needed | -0.733 |
whispered | -0.721 |
meant | -0.719 |
said | -0.670 |
Jones | -0.665 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.89e+02 |
Pos logit contributions | |
pieces | +0.628 |
disl | +0.622 |
n | +0.608 |
iu | +0.603 |
interests | +0.600 |
Neg logit contributions | |
needed | -0.549 |
whispered | -0.529 |
meant | -0.508 |
Jones | -0.500 |
said | -0.488 |
Token | t |
Token ID | 83 |
Feature activation | +3.87e+02 |
Pos logit contributions | |
n | +0.812 |
pieces | +0.741 |
ric | +0.728 |
disl | +0.722 |
ieved | +0.709 |
Neg logit contributions | |
whispered | -0.568 |
needed | -0.525 |
faded | -0.502 |
Jones | -0.496 |
meant | -0.496 |
Token | eat |
Token ID | 4483 |
Feature activation | +3.70e+02 |
Pos logit contributions | |
n | +0.446 |
disl | +0.441 |
pieces | +0.437 |
Majesty | +0.421 |
iu | +0.419 |
Neg logit contributions | |
needed | -0.621 |
whispered | -0.608 |
meant | -0.595 |
Jones | -0.562 |
said | -0.553 |
Token | me |
Token ID | 502 |
Feature activation | +3.69e+02 |
Pos logit contributions | |
n | +0.412 |
disl | +0.406 |
pieces | +0.399 |
interests | +0.381 |
Majesty | +0.381 |
Neg logit contributions | |
needed | -0.604 |
whispered | -0.595 |
meant | -0.578 |
Jones | -0.554 |
said | -0.539 |
Token | , |
Token ID | 11 |
Feature activation | +3.14e+02 |
Pos logit contributions | |
n | +0.488 |
disl | +0.459 |
pieces | +0.451 |
interests | +0.448 |
Majesty | +0.438 |
Neg logit contributions | |
needed | -0.518 |
whispered | -0.515 |
meant | -0.498 |
Jones | -0.486 |
faded | -0.479 |
Token | â |
Token ID | 22940 |
Feature activation | +3.62e+02 |
Pos logit contributions | |
n | +0.436 |
disl | +0.432 |
pieces | +0.425 |
Majesty | +0.411 |
interests | +0.410 |
Neg logit contributions | |
needed | -0.425 |
whispered | -0.417 |
meant | -0.403 |
Jones | -0.382 |
said | -0.371 |
Token | � |
Token ID | 129 |
Feature activation | +4.59e+02 |
Pos logit contributions | |
pieces | +0.838 |
disl | +0.838 |
n | +0.822 |
interests | +0.811 |
iu | +0.806 |
Neg logit contributions | |
needed | -0.412 |
whispered | -0.404 |
meant | -0.367 |
Jones | -0.361 |
said | -0.353 |
Token | � |
Token ID | 241 |
Feature activation | +4.22e+02 |
Pos logit contributions | |
n | +0.495 |
disl | +0.481 |
pieces | +0.454 |
Majesty | +0.442 |
iu | +0.438 |
Neg logit contributions | |
needed | -0.790 |
whispered | -0.771 |
meant | -0.753 |
Jones | -0.723 |
faded | -0.704 |
Token | I |
Token ID | 40 |
Feature activation | +3.74e+02 |
Pos logit contributions | |
pieces | +0.731 |
disl | +0.728 |
n | +0.721 |
Majesty | +0.721 |
arrest | +0.710 |
Neg logit contributions | |
needed | -0.423 |
whispered | -0.400 |
meant | -0.394 |
Jones | -0.357 |
said | -0.350 |
Token | â |
Token ID | 22940 |
Feature activation | +4.17e+02 |
Pos logit contributions | |
disl | +0.442 |
n | +0.441 |
pieces | +0.440 |
iu | +0.424 |
Majesty | +0.420 |
Neg logit contributions | |
needed | -0.588 |
whispered | -0.578 |
meant | -0.562 |
Jones | -0.528 |
said | -0.525 |
Token | € |
Token ID | 26391 |
Feature activation | +4.24e+02 |
Pos logit contributions | |
n | +0.369 |
disl | +0.355 |
pieces | +0.345 |
iu | +0.325 |
interests | +0.324 |
Neg logit contributions | |
needed | -0.788 |
whispered | -0.776 |
meant | -0.756 |
Jones | -0.731 |
faded | -0.713 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.85e+02 |
Pos logit contributions | |
pieces | +0.585 |
disl | +0.580 |
victory | +0.561 |
interests | +0.560 |
n | +0.559 |
Neg logit contributions | |
needed | -0.578 |
whispered | -0.565 |
Jones | -0.541 |
meant | -0.529 |
said | -0.524 |
Token | m |
Token ID | 76 |
Feature activation | +3.77e+02 |
Pos logit contributions | |
n | +0.817 |
ric | +0.760 |
pieces | +0.742 |
disl | +0.740 |
ieved | +0.722 |
Neg logit contributions | |
whispered | -0.535 |
needed | -0.494 |
faded | -0.483 |
Jones | -0.481 |
White | -0.465 |
Token | here |
Token ID | 994 |
Feature activation | +3.73e+02 |
Pos logit contributions | |
n | +0.547 |
disl | +0.531 |
pieces | +0.515 |
Majesty | +0.511 |
ric | +0.505 |
Neg logit contributions | |
needed | -0.496 |
whispered | -0.491 |
meant | -0.466 |
Jones | -0.447 |
faded | -0.438 |
Token | to |
Token ID | 284 |
Feature activation | +3.38e+02 |
Pos logit contributions | |
n | +0.478 |
disl | +0.457 |
pieces | +0.442 |
interests | +0.435 |
victory | +0.435 |
Neg logit contributions | |
needed | -0.545 |
whispered | -0.541 |
meant | -0.514 |
Jones | -0.495 |
faded | -0.489 |
Token | give |
Token ID | 1577 |
Feature activation | +3.18e+02 |
Pos logit contributions | |
iu | +0.441 |
nels | +0.410 |
pieces | +0.409 |
disl | +0.401 |
interests | +0.397 |
Neg logit contributions | |
whispered | -0.517 |
said | -0.507 |
needed | -0.500 |
meant | -0.486 |
dark | -0.459 |
Token | you |
Token ID | 345 |
Feature activation | +3.20e+02 |
Pos logit contributions | |
disl | +0.389 |
pieces | +0.381 |
n | +0.379 |
Majesty | +0.369 |
iu | +0.368 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.484 |
meant | -0.465 |
said | -0.447 |
Jones | -0.444 |
Token | � |
Token ID | 129 |
Feature activation | +4.59e+02 |
Pos logit contributions | |
pieces | +0.838 |
disl | +0.838 |
n | +0.822 |
interests | +0.811 |
iu | +0.806 |
Neg logit contributions | |
needed | -0.412 |
whispered | -0.404 |
meant | -0.367 |
Jones | -0.361 |
said | -0.353 |
Token | � |
Token ID | 241 |
Feature activation | +4.22e+02 |
Pos logit contributions | |
n | +0.495 |
disl | +0.481 |
pieces | +0.454 |
Majesty | +0.442 |
iu | +0.438 |
Neg logit contributions | |
needed | -0.790 |
whispered | -0.771 |
meant | -0.753 |
Jones | -0.723 |
faded | -0.704 |
Token | It |
Token ID | 1026 |
Feature activation | +3.71e+02 |
Pos logit contributions | |
pieces | +0.731 |
disl | +0.728 |
n | +0.721 |
Majesty | +0.721 |
arrest | +0.710 |
Neg logit contributions | |
needed | -0.423 |
whispered | -0.400 |
meant | -0.394 |
Jones | -0.357 |
said | -0.350 |
Token | â |
Token ID | 22940 |
Feature activation | +4.13e+02 |
Pos logit contributions | |
n | +0.539 |
disl | +0.531 |
pieces | +0.524 |
interests | +0.507 |
Majesty | +0.506 |
Neg logit contributions | |
needed | -0.483 |
whispered | -0.474 |
meant | -0.456 |
Jones | -0.434 |
said | -0.417 |
Token | € |
Token ID | 26391 |
Feature activation | +4.20e+02 |
Pos logit contributions | |
n | +0.389 |
disl | +0.373 |
pieces | +0.363 |
iu | +0.343 |
interests | +0.343 |
Neg logit contributions | |
needed | -0.758 |
whispered | -0.748 |
meant | -0.726 |
Jones | -0.702 |
faded | -0.683 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.81e+02 |
Pos logit contributions | |
pieces | +0.588 |
disl | +0.580 |
n | +0.565 |
interests | +0.562 |
iu | +0.562 |
Neg logit contributions | |
needed | -0.561 |
whispered | -0.546 |
Jones | -0.517 |
meant | -0.516 |
said | -0.500 |
Token | s |
Token ID | 82 |
Feature activation | +3.85e+02 |
Pos logit contributions | |
n | +0.822 |
ric | +0.769 |
disl | +0.741 |
pieces | +0.738 |
ieved | +0.710 |
Neg logit contributions | |
whispered | -0.538 |
Jones | -0.496 |
needed | -0.488 |
faded | -0.477 |
White | -0.463 |
Token | a |
Token ID | 257 |
Feature activation | +3.66e+02 |
Pos logit contributions | |
n | +0.508 |
disl | +0.504 |
pieces | +0.499 |
iu | +0.480 |
Majesty | +0.478 |
Neg logit contributions | |
needed | -0.551 |
whispered | -0.538 |
meant | -0.524 |
Jones | -0.494 |
said | -0.482 |
Token | surprise |
Token ID | 5975 |
Feature activation | +3.60e+02 |
Pos logit contributions | |
n | +0.524 |
disl | +0.519 |
pieces | +0.511 |
Majesty | +0.500 |
victory | +0.499 |
Neg logit contributions | |
needed | -0.467 |
whispered | -0.463 |
meant | -0.449 |
Jones | -0.426 |
faded | -0.401 |
Token | ! |
Token ID | 0 |
Feature activation | +2.99e+02 |
Pos logit contributions | |
n | +0.437 |
pieces | +0.394 |
disl | +0.384 |
occasions | +0.379 |
blast | +0.372 |
Neg logit contributions | |
needed | -0.552 |
whispered | -0.543 |
Jones | -0.515 |
meant | -0.511 |
bored | -0.504 |
Token | â |
Token ID | 22940 |
Feature activation | +3.46e+02 |
Pos logit contributions | |
n | +0.410 |
disl | +0.407 |
pieces | +0.400 |
interests | +0.386 |
Majesty | +0.386 |
Neg logit contributions | |
needed | -0.410 |
whispered | -0.403 |
meant | -0.389 |
Jones | -0.372 |
said | -0.359 |
Token | � |
Token ID | 129 |
Feature activation | +4.59e+02 |
Pos logit contributions | |
pieces | +0.838 |
disl | +0.838 |
n | +0.822 |
interests | +0.811 |
iu | +0.806 |
Neg logit contributions | |
needed | -0.412 |
whispered | -0.404 |
meant | -0.367 |
Jones | -0.361 |
said | -0.353 |
Token | � |
Token ID | 241 |
Feature activation | +4.22e+02 |
Pos logit contributions | |
n | +0.495 |
disl | +0.481 |
pieces | +0.454 |
Majesty | +0.442 |
iu | +0.438 |
Neg logit contributions | |
needed | -0.790 |
whispered | -0.771 |
meant | -0.753 |
Jones | -0.723 |
faded | -0.704 |
Token | It |
Token ID | 1026 |
Feature activation | +3.71e+02 |
Pos logit contributions | |
pieces | +0.731 |
disl | +0.728 |
n | +0.721 |
Majesty | +0.721 |
arrest | +0.710 |
Neg logit contributions | |
needed | -0.423 |
whispered | -0.400 |
meant | -0.394 |
Jones | -0.357 |
said | -0.350 |
Token | â |
Token ID | 22940 |
Feature activation | +4.13e+02 |
Pos logit contributions | |
n | +0.539 |
disl | +0.531 |
pieces | +0.524 |
interests | +0.507 |
Majesty | +0.506 |
Neg logit contributions | |
needed | -0.483 |
whispered | -0.474 |
meant | -0.456 |
Jones | -0.434 |
said | -0.417 |
Token | € |
Token ID | 26391 |
Feature activation | +4.20e+02 |
Pos logit contributions | |
n | +0.389 |
disl | +0.373 |
pieces | +0.363 |
iu | +0.343 |
interests | +0.343 |
Neg logit contributions | |
needed | -0.758 |
whispered | -0.748 |
meant | -0.726 |
Jones | -0.702 |
faded | -0.683 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.81e+02 |
Pos logit contributions | |
pieces | +0.588 |
disl | +0.580 |
n | +0.565 |
interests | +0.562 |
iu | +0.562 |
Neg logit contributions | |
needed | -0.561 |
whispered | -0.546 |
Jones | -0.517 |
meant | -0.516 |
said | -0.500 |
Token | s |
Token ID | 82 |
Feature activation | +3.85e+02 |
Pos logit contributions | |
n | +0.822 |
ric | +0.769 |
disl | +0.741 |
pieces | +0.738 |
ieved | +0.710 |
Neg logit contributions | |
whispered | -0.538 |
Jones | -0.496 |
needed | -0.488 |
faded | -0.477 |
White | -0.463 |
Token | a |
Token ID | 257 |
Feature activation | +3.66e+02 |
Pos logit contributions | |
n | +0.508 |
disl | +0.504 |
pieces | +0.499 |
iu | +0.480 |
Majesty | +0.478 |
Neg logit contributions | |
needed | -0.551 |
whispered | -0.538 |
meant | -0.524 |
Jones | -0.494 |
said | -0.482 |
Token | books |
Token ID | 3835 |
Feature activation | +3.70e+02 |
Pos logit contributions | |
n | +0.524 |
disl | +0.519 |
pieces | +0.511 |
Majesty | +0.500 |
victory | +0.499 |
Neg logit contributions | |
needed | -0.467 |
whispered | -0.463 |
meant | -0.449 |
Jones | -0.426 |
faded | -0.401 |
Token | he |
Token ID | 258 |
Feature activation | +3.85e+02 |
Pos logit contributions | |
n | +0.464 |
blast | +0.415 |
victory | +0.408 |
spend | +0.401 |
pieces | +0.391 |
Neg logit contributions | |
Jones | -0.533 |
whispered | -0.528 |
needed | -0.520 |
White | -0.508 |
spoke | -0.482 |
Token | lf |
Token ID | 1652 |
Feature activation | +3.98e+02 |
Pos logit contributions | |
n | +0.534 |
disl | +0.529 |
pieces | +0.522 |
interests | +0.504 |
Majesty | +0.502 |
Neg logit contributions | |
needed | -0.522 |
whispered | -0.512 |
meant | -0.494 |
Jones | -0.469 |
said | -0.456 |
Token | her |
Token ID | 607 |
Feature activation | +4.38e+02 |
Pos logit contributions | |
needed | +0.231 |
whispered | +0.229 |
Jones | +0.215 |
meant | +0.215 |
White | +0.194 |
Neg logit contributions | |
n | -0.388 |
disl | -0.385 |
pieces | -0.383 |
interests | -0.374 |
Majesty | -0.364 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.92e+02 |
Pos logit contributions | |
n | +0.607 |
disl | +0.598 |
pieces | +0.589 |
interests | +0.570 |
Majesty | +0.569 |
Neg logit contributions | |
needed | -0.595 |
whispered | -0.585 |
meant | -0.565 |
Jones | -0.537 |
said | -0.518 |
Token | my |
Token ID | 1820 |
Feature activation | +4.11e+02 |
Pos logit contributions | |
n | +0.640 |
pieces | +0.609 |
spend | +0.597 |
hey | +0.575 |
ric | +0.575 |
Neg logit contributions | |
whispered | -0.394 |
needed | -0.386 |
meant | -0.381 |
Jones | -0.381 |
faded | -0.373 |
Token | watched |
Token ID | 7342 |
Feature activation | +3.29e+02 |
Pos logit contributions | |
n | +0.643 |
spend | +0.615 |
pieces | +0.579 |
Majesty | +0.566 |
disl | +0.558 |
Neg logit contributions | |
Ancient | -0.430 |
dark | -0.426 |
whispered | -0.415 |
Fuel | -0.413 |
faded | -0.408 |
Token | in |
Token ID | 287 |
Feature activation | +3.14e+02 |
Pos logit contributions | |
n | +0.447 |
pieces | +0.434 |
disl | +0.429 |
iu | +0.410 |
interests | +0.406 |
Neg logit contributions | |
needed | -0.463 |
whispered | -0.455 |
meant | -0.439 |
Jones | -0.415 |
faded | -0.403 |
Token | amaz |
Token ID | 40642 |
Feature activation | +4.79e+02 |
Pos logit contributions | |
n | +0.441 |
disl | +0.436 |
pieces | +0.430 |
interests | +0.418 |
arrest | +0.415 |
Neg logit contributions | |
needed | -0.420 |
whispered | -0.412 |
meant | -0.397 |
Jones | -0.376 |
spoke | -0.361 |
Token | ement |
Token ID | 972 |
Feature activation | +3.53e+02 |
Pos logit contributions | |
blast | +0.979 |
arrest | +0.974 |
ric | +0.962 |
disl | +0.954 |
n | +0.941 |
Neg logit contributions | |
said | -0.362 |
meant | -0.345 |
whispered | -0.331 |
needed | -0.317 |
answered | -0.308 |
Token | as |
Token ID | 355 |
Feature activation | +3.05e+02 |
Pos logit contributions | |
n | +0.575 |
pieces | +0.563 |
disl | +0.562 |
interests | +0.541 |
Majesty | +0.538 |
Neg logit contributions | |
needed | -0.393 |
whispered | -0.384 |
meant | -0.366 |
Jones | -0.345 |
faded | -0.329 |
Token | the |
Token ID | 262 |
Feature activation | +3.10e+02 |
Pos logit contributions | |
n | +0.477 |
pieces | +0.470 |
disl | +0.468 |
victory | +0.455 |
interests | +0.447 |
Neg logit contributions | |
needed | -0.350 |
whispered | -0.346 |
meant | -0.327 |
Jones | -0.318 |
spoke | -0.294 |
Token | machine |
Token ID | 4572 |
Feature activation | +3.16e+02 |
Pos logit contributions | |
n | +0.433 |
disl | +0.425 |
pieces | +0.421 |
interests | +0.405 |
Majesty | +0.405 |
Neg logit contributions | |
needed | -0.420 |
whispered | -0.413 |
meant | -0.398 |
Jones | -0.380 |
said | -0.363 |
Token | flew |
Token ID | 13112 |
Feature activation | +2.88e+02 |
Pos logit contributions | |
n | +0.507 |
pieces | +0.440 |
spend | +0.421 |
join | +0.417 |
ered | +0.416 |
Neg logit contributions | |
Jones | -0.363 |
Fuel | -0.352 |
Ancient | -0.340 |
Pr | -0.336 |
mail | -0.336 |
Token | � |
Token ID | 129 |
Feature activation | +4.59e+02 |
Pos logit contributions | |
pieces | +0.838 |
disl | +0.838 |
n | +0.822 |
interests | +0.811 |
iu | +0.806 |
Neg logit contributions | |
needed | -0.412 |
whispered | -0.404 |
meant | -0.367 |
Jones | -0.361 |
said | -0.353 |
Token | � |
Token ID | 241 |
Feature activation | +4.22e+02 |
Pos logit contributions | |
n | +0.495 |
disl | +0.481 |
pieces | +0.454 |
Majesty | +0.442 |
iu | +0.438 |
Neg logit contributions | |
needed | -0.790 |
whispered | -0.771 |
meant | -0.753 |
Jones | -0.723 |
faded | -0.704 |
Token | You |
Token ID | 1639 |
Feature activation | +3.64e+02 |
Pos logit contributions | |
pieces | +0.731 |
disl | +0.728 |
n | +0.721 |
Majesty | +0.721 |
arrest | +0.710 |
Neg logit contributions | |
needed | -0.423 |
whispered | -0.400 |
meant | -0.394 |
Jones | -0.357 |
said | -0.350 |
Token | â |
Token ID | 22940 |
Feature activation | +4.07e+02 |
Pos logit contributions | |
disl | +0.470 |
pieces | +0.461 |
n | +0.458 |
iu | +0.454 |
Majesty | +0.443 |
Neg logit contributions | |
needed | -0.547 |
whispered | -0.532 |
meant | -0.523 |
said | -0.487 |
Jones | -0.486 |
Token | € |
Token ID | 26391 |
Feature activation | +4.16e+02 |
Pos logit contributions | |
n | +0.340 |
disl | +0.298 |
pieces | +0.265 |
ric | +0.254 |
iu | +0.252 |
Neg logit contributions | |
needed | -0.829 |
whispered | -0.816 |
meant | -0.777 |
faded | -0.774 |
Jones | -0.773 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.78e+02 |
Pos logit contributions | |
pieces | +0.548 |
disl | +0.547 |
n | +0.540 |
interests | +0.525 |
iu | +0.524 |
Neg logit contributions | |
needed | -0.586 |
whispered | -0.571 |
meant | -0.548 |
Jones | -0.540 |
said | -0.522 |
Token | ll |
Token ID | 297 |
Feature activation | +4.13e+02 |
Pos logit contributions | |
n | +0.804 |
ric | +0.763 |
disl | +0.733 |
pieces | +0.733 |
interests | +0.704 |
Neg logit contributions | |
whispered | -0.541 |
faded | -0.489 |
needed | -0.487 |
Jones | -0.476 |
dark | -0.461 |
Token | see |
Token ID | 766 |
Feature activation | +3.54e+02 |
Pos logit contributions | |
iu | +0.447 |
disl | +0.435 |
pieces | +0.434 |
arrest | +0.419 |
Majesty | +0.417 |
Neg logit contributions | |
needed | -0.698 |
meant | -0.683 |
whispered | -0.676 |
said | -0.645 |
chy | -0.611 |
Token | . |
Token ID | 13 |
Feature activation | +3.02e+02 |
Pos logit contributions | |
n | +0.393 |
disl | +0.387 |
pieces | +0.380 |
iu | +0.365 |
interests | +0.363 |
Neg logit contributions | |
needed | -0.579 |
whispered | -0.571 |
meant | -0.555 |
Jones | -0.531 |
said | -0.520 |
Token | â |
Token ID | 22940 |
Feature activation | +3.51e+02 |
Pos logit contributions | |
n | +0.411 |
disl | +0.405 |
pieces | +0.399 |
interests | +0.385 |
Majesty | +0.384 |
Neg logit contributions | |
needed | -0.418 |
whispered | -0.411 |
meant | -0.397 |
Jones | -0.377 |
said | -0.365 |
Token | € |
Token ID | 26391 |
Feature activation | +3.60e+02 |
Pos logit contributions | |
n | +0.294 |
disl | +0.271 |
pieces | +0.249 |
iu | +0.237 |
nels | +0.233 |
Neg logit contributions | |
needed | -0.699 |
whispered | -0.684 |
meant | -0.665 |
Jones | -0.652 |
faded | -0.643 |
Token | � |
Token ID | 129 |
Feature activation | +4.59e+02 |
Pos logit contributions | |
pieces | +0.838 |
disl | +0.838 |
n | +0.822 |
interests | +0.811 |
iu | +0.806 |
Neg logit contributions | |
needed | -0.412 |
whispered | -0.404 |
meant | -0.367 |
Jones | -0.361 |
said | -0.353 |
Token | � |
Token ID | 241 |
Feature activation | +4.22e+02 |
Pos logit contributions | |
n | +0.495 |
disl | +0.481 |
pieces | +0.454 |
Majesty | +0.442 |
iu | +0.438 |
Neg logit contributions | |
needed | -0.790 |
whispered | -0.771 |
meant | -0.753 |
Jones | -0.723 |
faded | -0.704 |
Token | Where |
Token ID | 8496 |
Feature activation | +3.72e+02 |
Pos logit contributions | |
pieces | +0.731 |
disl | +0.728 |
n | +0.721 |
Majesty | +0.721 |
arrest | +0.710 |
Neg logit contributions | |
needed | -0.423 |
whispered | -0.400 |
meant | -0.394 |
Jones | -0.357 |
said | -0.350 |
Token | â |
Token ID | 22940 |
Feature activation | +4.07e+02 |
Pos logit contributions | |
iu | +0.560 |
arrest | +0.552 |
disl | +0.546 |
pieces | +0.544 |
Majesty | +0.538 |
Neg logit contributions | |
needed | -0.503 |
meant | -0.493 |
whispered | -0.478 |
said | -0.457 |
smelled | -0.410 |
Token | € |
Token ID | 26391 |
Feature activation | +4.16e+02 |
Pos logit contributions | |
pieces | +0.363 |
disl | +0.361 |
n | +0.353 |
interests | +0.349 |
Majesty | +0.348 |
Neg logit contributions | |
needed | -0.745 |
whispered | -0.737 |
meant | -0.722 |
Jones | -0.688 |
said | -0.681 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.77e+02 |
Pos logit contributions | |
disl | +0.689 |
pieces | +0.686 |
n | +0.673 |
interests | +0.667 |
iu | +0.666 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.440 |
meant | -0.415 |
Jones | -0.407 |
said | -0.391 |
Token | s |
Token ID | 82 |
Feature activation | +3.81e+02 |
Pos logit contributions | |
n | +0.814 |
ric | +0.745 |
pieces | +0.734 |
disl | +0.723 |
ieved | +0.705 |
Neg logit contributions | |
whispered | -0.523 |
Jones | -0.482 |
needed | -0.479 |
faded | -0.477 |
White | -0.465 |
Token | my |
Token ID | 616 |
Feature activation | +3.78e+02 |
Pos logit contributions | |
pieces | +0.500 |
iu | +0.496 |
disl | +0.495 |
n | +0.489 |
interests | +0.482 |
Neg logit contributions | |
needed | -0.562 |
meant | -0.541 |
whispered | -0.540 |
said | -0.505 |
Jones | -0.482 |
Token | jelly |
Token ID | 27644 |
Feature activation | +3.83e+02 |
Pos logit contributions | |
n | +0.465 |
disl | +0.456 |
pieces | +0.448 |
Majesty | +0.435 |
interests | +0.433 |
Neg logit contributions | |
needed | -0.574 |
whispered | -0.565 |
meant | -0.548 |
Jones | -0.523 |
said | -0.506 |
Token | ? |
Token ID | 30 |
Feature activation | +3.08e+02 |
Pos logit contributions | |
n | +0.422 |
disl | +0.403 |
pieces | +0.388 |
Majesty | +0.380 |
interests | +0.376 |
Neg logit contributions | |
needed | -0.622 |
whispered | -0.620 |
meant | -0.593 |
Jones | -0.587 |
White | -0.565 |
Token | â |
Token ID | 22940 |
Feature activation | +3.45e+02 |
Pos logit contributions | |
n | +0.418 |
pieces | +0.380 |
disl | +0.380 |
ric | +0.356 |
Majesty | +0.355 |
Neg logit contributions | |
needed | -0.451 |
whispered | -0.444 |
meant | -0.421 |
faded | -0.418 |
chy | -0.417 |
Token | � |
Token ID | 241 |
Feature activation | +4.76e+02 |
Pos logit contributions | |
interests | +0.659 |
pieces | +0.652 |
victory | +0.601 |
occasions | +0.582 |
nels | +0.572 |
Neg logit contributions | |
whispered | -0.925 |
needed | -0.890 |
said | -0.883 |
meant | -0.881 |
Jones | -0.831 |
Token | Why |
Token ID | 5195 |
Feature activation | +3.81e+02 |
Pos logit contributions | |
Majesty | +0.849 |
disl | +0.848 |
pieces | +0.847 |
n | +0.838 |
arrest | +0.827 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.428 |
meant | -0.428 |
Jones | -0.380 |
said | -0.375 |
Token | don |
Token ID | 836 |
Feature activation | +3.73e+02 |
Pos logit contributions | |
pieces | +0.461 |
n | +0.459 |
disl | +0.453 |
iu | +0.449 |
arrest | +0.442 |
Neg logit contributions | |
needed | -0.586 |
whispered | -0.572 |
meant | -0.569 |
Jones | -0.530 |
said | -0.529 |
Token | â |
Token ID | 22940 |
Feature activation | +4.06e+02 |
Pos logit contributions | |
Majesty | +0.666 |
interests | +0.630 |
disl | +0.605 |
majesty | +0.593 |
iu | +0.592 |
Neg logit contributions | |
meant | -0.451 |
needed | -0.446 |
said | -0.403 |
whispered | -0.402 |
White | -0.371 |
Token | € |
Token ID | 26391 |
Feature activation | +4.14e+02 |
Pos logit contributions | |
disl | +0.389 |
pieces | +0.389 |
Majesty | +0.382 |
n | +0.378 |
interests | +0.375 |
Neg logit contributions | |
needed | -0.719 |
whispered | -0.709 |
meant | -0.701 |
Jones | -0.657 |
said | -0.656 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.75e+02 |
Pos logit contributions | |
disl | +0.576 |
pieces | +0.576 |
n | +0.574 |
iu | +0.555 |
interests | +0.555 |
Neg logit contributions | |
needed | -0.555 |
whispered | -0.540 |
meant | -0.522 |
Jones | -0.502 |
said | -0.489 |
Token | t |
Token ID | 83 |
Feature activation | +3.77e+02 |
Pos logit contributions | |
n | +0.767 |
pieces | +0.683 |
ric | +0.676 |
ieved | +0.662 |
disl | +0.661 |
Neg logit contributions | |
whispered | -0.579 |
Jones | -0.529 |
faded | -0.524 |
needed | -0.523 |
blank | -0.520 |
Token | you |
Token ID | 345 |
Feature activation | +3.68e+02 |
Pos logit contributions | |
n | +0.492 |
disl | +0.485 |
pieces | +0.480 |
Majesty | +0.463 |
iu | +0.462 |
Neg logit contributions | |
needed | -0.547 |
whispered | -0.536 |
meant | -0.521 |
Jones | -0.493 |
said | -0.481 |
Token | come |
Token ID | 1282 |
Feature activation | +3.37e+02 |
Pos logit contributions | |
n | +0.433 |
pause | +0.383 |
spend | +0.383 |
disl | +0.371 |
interests | +0.370 |
Neg logit contributions | |
whispered | -0.554 |
Jones | -0.553 |
needed | -0.545 |
aisle | -0.545 |
White | -0.529 |
Token | with |
Token ID | 351 |
Feature activation | +3.17e+02 |
Pos logit contributions | |
n | +0.477 |
disl | +0.475 |
pieces | +0.466 |
iu | +0.453 |
Majesty | +0.449 |
Neg logit contributions | |
needed | -0.450 |
whispered | -0.445 |
meant | -0.428 |
Jones | -0.410 |
said | -0.394 |
Token | me |
Token ID | 502 |
Feature activation | +3.27e+02 |
Pos logit contributions | |
n | +0.342 |
disl | +0.334 |
pieces | +0.325 |
interests | +0.314 |
Majesty | +0.310 |
Neg logit contributions | |
needed | -0.532 |
whispered | -0.523 |
meant | -0.508 |
Jones | -0.492 |
faded | -0.474 |
Token | ball |
Token ID | 2613 |
Feature activation | -2.32e+02 |
Pos logit contributions | |
needed | +0.280 |
whispered | +0.277 |
meant | +0.255 |
dark | +0.248 |
said | +0.248 |
Neg logit contributions | |
n | -0.381 |
disl | -0.381 |
iu | -0.365 |
Majesty | -0.364 |
pieces | -0.364 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e+02 |
Pos logit contributions | |
whispered | +0.286 |
needed | +0.281 |
Jones | +0.273 |
meant | +0.272 |
spoke | +0.256 |
Neg logit contributions | |
n | -0.352 |
victory | -0.344 |
pieces | -0.334 |
ap | -0.328 |
blast | -0.326 |
Token | They |
Token ID | 1119 |
Feature activation | -2.45e+02 |
Pos logit contributions | |
needed | +0.263 |
whispered | +0.259 |
meant | +0.247 |
Jones | +0.226 |
spoke | +0.220 |
Neg logit contributions | |
n | -0.420 |
disl | -0.410 |
pieces | -0.409 |
interests | -0.404 |
Majesty | -0.394 |
Token | are |
Token ID | 389 |
Feature activation | -2.47e+02 |
Pos logit contributions | |
needed | +0.444 |
whispered | +0.435 |
meant | +0.420 |
said | +0.403 |
Jones | +0.401 |
Neg logit contributions | |
disl | -0.234 |
n | -0.233 |
pieces | -0.228 |
iu | -0.221 |
Majesty | -0.220 |
Token | clever |
Token ID | 14169 |
Feature activation | -2.44e+02 |
Pos logit contributions | |
needed | +0.264 |
whispered | +0.259 |
meant | +0.249 |
Jones | +0.227 |
said | +0.223 |
Neg logit contributions | |
n | -0.417 |
disl | -0.415 |
pieces | -0.408 |
iu | -0.397 |
interests | -0.396 |
Token | . |
Token ID | 13 |
Feature activation | -2.55e+02 |
Pos logit contributions | |
needed | +0.250 |
whispered | +0.244 |
meant | +0.231 |
Jones | +0.219 |
faded | +0.206 |
Neg logit contributions | |
n | -0.418 |
disl | -0.412 |
pieces | -0.409 |
interests | -0.401 |
Majesty | -0.398 |
Token | They |
Token ID | 1119 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.257 |
whispered | +0.252 |
meant | +0.240 |
Jones | +0.219 |
faded | +0.213 |
Neg logit contributions | |
n | -0.447 |
disl | -0.437 |
pieces | -0.435 |
interests | -0.431 |
Majesty | -0.418 |
Token | know |
Token ID | 760 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
needed | +0.440 |
whispered | +0.431 |
meant | +0.416 |
said | +0.400 |
Jones | +0.390 |
Neg logit contributions | |
n | -0.258 |
disl | -0.256 |
pieces | -0.255 |
iu | -0.250 |
Majesty | -0.245 |
Token | how |
Token ID | 703 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
needed | +0.476 |
whispered | +0.470 |
meant | +0.459 |
said | +0.436 |
Jones | +0.435 |
Neg logit contributions | |
n | -0.234 |
disl | -0.228 |
pieces | -0.222 |
iu | -0.215 |
arrest | -0.212 |
Token | to |
Token ID | 284 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
needed | +0.253 |
dark | +0.204 |
sleepy | +0.201 |
bored | +0.200 |
whispered | +0.195 |
Neg logit contributions | |
iu | -0.514 |
Majesty | -0.499 |
han | -0.495 |
n | -0.493 |
pieces | -0.488 |
Token | share |
Token ID | 2648 |
Feature activation | -2.62e+02 |
Pos logit contributions | |
needed | +0.352 |
whispered | +0.343 |
chy | +0.341 |
said | +0.328 |
meant | +0.328 |
Neg logit contributions | |
iu | -0.383 |
pieces | -0.369 |
disl | -0.363 |
Majesty | -0.363 |
interests | -0.338 |
Token | a |
Token ID | 257 |
Feature activation | -2.37e+02 |
Pos logit contributions | |
needed | +0.409 |
whispered | +0.404 |
meant | +0.392 |
Jones | +0.377 |
said | +0.370 |
Neg logit contributions | |
n | -0.236 |
disl | -0.236 |
pieces | -0.229 |
iu | -0.219 |
Majesty | -0.218 |
Token | spaceship |
Token ID | 40663 |
Feature activation | -2.29e+02 |
Pos logit contributions | |
needed | +0.349 |
whispered | +0.345 |
meant | +0.341 |
Jones | +0.317 |
spoke | +0.306 |
Neg logit contributions | |
n | -0.297 |
victory | -0.290 |
pieces | -0.290 |
Majesty | -0.290 |
blast | -0.289 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.47e+02 |
Pos logit contributions | |
needed | +0.331 |
whispered | +0.330 |
meant | +0.321 |
said | +0.307 |
Jones | +0.300 |
Neg logit contributions | |
disl | -0.304 |
n | -0.294 |
pieces | -0.291 |
iu | -0.289 |
interests | -0.281 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.36e+02 |
Pos logit contributions | |
needed | +0.233 |
meant | +0.224 |
whispered | +0.215 |
Jones | +0.202 |
chy | +0.201 |
Neg logit contributions | |
n | -0.435 |
pieces | -0.424 |
disl | -0.421 |
arrest | -0.416 |
blast | -0.412 |
Token | said |
Token ID | 531 |
Feature activation | -2.47e+02 |
Pos logit contributions | |
lessness | +0.227 |
chy | +0.226 |
oky | +0.224 |
faded | +0.216 |
dark | +0.214 |
Neg logit contributions | |
n | -0.431 |
ered | -0.357 |
interests | -0.344 |
iu | -0.343 |
how | -0.339 |
Token | . |
Token ID | 13 |
Feature activation | -2.54e+02 |
Pos logit contributions | |
needed | +0.261 |
whispered | +0.251 |
meant | +0.246 |
Jones | +0.227 |
faded | +0.221 |
Neg logit contributions | |
n | -0.412 |
pieces | -0.399 |
disl | -0.397 |
iu | -0.390 |
victory | -0.385 |
Token | " |
Token ID | 366 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
needed | +0.345 |
chy | +0.342 |
whispered | +0.334 |
meant | +0.334 |
faded | +0.310 |
Neg logit contributions | |
n | -0.327 |
join | -0.309 |
blast | -0.300 |
interests | -0.286 |
our | -0.283 |
Token | We |
Token ID | 1135 |
Feature activation | -2.54e+02 |
Pos logit contributions | |
needed | +0.346 |
whispered | +0.343 |
meant | +0.329 |
Jones | +0.312 |
faded | +0.307 |
Neg logit contributions | |
n | -0.358 |
pieces | -0.340 |
disl | -0.336 |
interests | -0.334 |
arrest | -0.323 |
Token | can |
Token ID | 460 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.360 |
whispered | +0.346 |
meant | +0.337 |
said | +0.312 |
Jones | +0.309 |
Neg logit contributions | |
disl | -0.347 |
pieces | -0.344 |
n | -0.344 |
iu | -0.337 |
Majesty | -0.330 |
Token | fly |
Token ID | 6129 |
Feature activation | -2.54e+02 |
Pos logit contributions | |
whispered | +0.426 |
needed | +0.419 |
said | +0.399 |
meant | +0.398 |
Jones | +0.380 |
Neg logit contributions | |
disl | -0.290 |
pieces | -0.274 |
iu | -0.274 |
earned | -0.264 |
interests | -0.262 |
Token | to |
Token ID | 284 |
Feature activation | -2.57e+02 |
Pos logit contributions | |
needed | +0.325 |
whispered | +0.321 |
meant | +0.307 |
Jones | +0.293 |
dark | +0.288 |
Neg logit contributions | |
disl | -0.375 |
n | -0.373 |
iu | -0.366 |
pieces | -0.363 |
Majesty | -0.356 |
Token | learned |
Token ID | 4499 |
Feature activation | -2.41e+02 |
Pos logit contributions | |
Jones | +0.302 |
aisle | +0.282 |
whispered | +0.277 |
White | +0.269 |
needed | +0.267 |
Neg logit contributions | |
n | -0.332 |
disl | -0.325 |
victory | -0.313 |
interests | -0.310 |
spend | -0.309 |
Token | that |
Token ID | 326 |
Feature activation | -2.43e+02 |
Pos logit contributions | |
needed | +0.334 |
whispered | +0.330 |
meant | +0.326 |
said | +0.309 |
Jones | +0.290 |
Neg logit contributions | |
pieces | -0.341 |
n | -0.340 |
Majesty | -0.333 |
iu | -0.331 |
disl | -0.320 |
Token | courage |
Token ID | 11917 |
Feature activation | -2.41e+02 |
Pos logit contributions | |
whispered | +0.407 |
needed | +0.399 |
Jones | +0.387 |
meant | +0.377 |
aisle | +0.367 |
Neg logit contributions | |
victory | -0.254 |
n | -0.252 |
interests | -0.252 |
disl | -0.250 |
pieces | -0.232 |
Token | is |
Token ID | 318 |
Feature activation | -2.44e+02 |
Pos logit contributions | |
whispered | +0.330 |
needed | +0.321 |
Jones | +0.315 |
meant | +0.301 |
White | +0.296 |
Neg logit contributions | |
n | -0.321 |
disl | -0.315 |
interests | -0.313 |
pieces | -0.300 |
victory | -0.298 |
Token | important |
Token ID | 1593 |
Feature activation | -2.43e+02 |
Pos logit contributions | |
needed | +0.370 |
whispered | +0.366 |
meant | +0.348 |
Jones | +0.347 |
faded | +0.327 |
Neg logit contributions | |
n | -0.300 |
disl | -0.294 |
pieces | -0.286 |
interests | -0.282 |
victory | -0.277 |
Token | . |
Token ID | 13 |
Feature activation | -2.54e+02 |
Pos logit contributions | |
needed | +0.230 |
whispered | +0.226 |
meant | +0.212 |
Jones | +0.199 |
said | +0.187 |
Neg logit contributions | |
n | -0.438 |
disl | -0.433 |
pieces | -0.426 |
interests | -0.419 |
iu | -0.414 |
Token | They |
Token ID | 1119 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
whispered | +0.272 |
needed | +0.272 |
meant | +0.250 |
spoke | +0.232 |
faded | +0.232 |
Neg logit contributions | |
disl | -0.430 |
n | -0.421 |
interests | -0.413 |
pieces | -0.401 |
ric | -0.396 |
Token | learned |
Token ID | 4499 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
Jones | +0.318 |
aisle | +0.300 |
whispered | +0.293 |
White | +0.286 |
needed | +0.285 |
Neg logit contributions | |
n | -0.355 |
disl | -0.348 |
victory | -0.335 |
spend | -0.333 |
interests | -0.332 |
Token | that |
Token ID | 326 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
needed | +0.364 |
whispered | +0.360 |
meant | +0.355 |
said | +0.336 |
Jones | +0.321 |
Neg logit contributions | |
n | -0.352 |
pieces | -0.351 |
Majesty | -0.342 |
iu | -0.340 |
disl | -0.333 |
Token | soccer |
Token ID | 11783 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
whispered | +0.433 |
needed | +0.422 |
Jones | +0.413 |
meant | +0.397 |
aisle | +0.396 |
Neg logit contributions | |
victory | -0.276 |
interests | -0.271 |
n | -0.267 |
disl | -0.264 |
pieces | -0.244 |
Token | is |
Token ID | 318 |
Feature activation | -2.60e+02 |
Pos logit contributions | |
whispered | +0.372 |
Jones | +0.363 |
needed | +0.352 |
White | +0.344 |
meant | +0.339 |
Neg logit contributions | |
interests | -0.301 |
n | -0.300 |
disl | -0.298 |
victory | -0.298 |
pieces | -0.282 |
Token | or |
Token ID | 393 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.281 |
whispered | +0.275 |
meant | +0.263 |
Jones | +0.249 |
said | +0.237 |
Neg logit contributions | |
n | -0.381 |
disl | -0.375 |
pieces | -0.371 |
interests | -0.360 |
Majesty | -0.359 |
Token | toys |
Token ID | 14958 |
Feature activation | -2.35e+02 |
Pos logit contributions | |
needed | +0.329 |
whispered | +0.317 |
meant | +0.308 |
said | +0.287 |
Jones | +0.281 |
Neg logit contributions | |
n | -0.335 |
disl | -0.330 |
pieces | -0.323 |
iu | -0.320 |
nels | -0.316 |
Token | , |
Token ID | 11 |
Feature activation | -2.45e+02 |
Pos logit contributions | |
needed | +0.360 |
whispered | +0.357 |
meant | +0.338 |
Jones | +0.334 |
faded | +0.327 |
Neg logit contributions | |
n | -0.289 |
disl | -0.269 |
pieces | -0.268 |
interests | -0.263 |
arrest | -0.261 |
Token | or |
Token ID | 393 |
Feature activation | -2.43e+02 |
Pos logit contributions | |
needed | +0.304 |
whispered | +0.298 |
meant | +0.287 |
Jones | +0.271 |
said | +0.262 |
Neg logit contributions | |
n | -0.368 |
disl | -0.364 |
pieces | -0.359 |
Majesty | -0.348 |
interests | -0.347 |
Token | books |
Token ID | 3835 |
Feature activation | -2.37e+02 |
Pos logit contributions | |
needed | +0.329 |
whispered | +0.311 |
meant | +0.304 |
said | +0.286 |
dark | +0.274 |
Neg logit contributions | |
n | -0.354 |
disl | -0.349 |
iu | -0.344 |
pieces | -0.342 |
nels | -0.341 |
Token | ," |
Token ID | 553 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.383 |
whispered | +0.377 |
Jones | +0.357 |
spoke | +0.353 |
White | +0.346 |
Neg logit contributions | |
n | -0.271 |
victory | -0.241 |
pieces | -0.234 |
our | -0.234 |
blast | -0.231 |
Token | Max |
Token ID | 5436 |
Feature activation | -2.41e+02 |
Pos logit contributions | |
needed | +0.385 |
meant | +0.372 |
faded | +0.359 |
whispered | +0.358 |
chy | +0.355 |
Neg logit contributions | |
n | -0.300 |
disl | -0.296 |
hey | -0.289 |
victory | -0.288 |
pieces | -0.285 |
Token | said |
Token ID | 531 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
lessness | +0.195 |
needed | +0.187 |
faded | +0.185 |
aisle | +0.180 |
bored | +0.177 |
Neg logit contributions | |
n | -0.481 |
disl | -0.437 |
interests | -0.419 |
ieved | -0.401 |
how | -0.396 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e+02 |
Pos logit contributions | |
needed | +0.305 |
meant | +0.284 |
whispered | +0.282 |
aisle | +0.275 |
Jones | +0.268 |
Neg logit contributions | |
n | -0.377 |
pieces | -0.357 |
iu | -0.356 |
victory | -0.354 |
disl | -0.352 |
Token |
|
Token ID | 198 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
needed | +0.386 |
whispered | +0.376 |
chy | +0.362 |
meant | +0.354 |
spoke | +0.351 |
Neg logit contributions | |
blast | -0.281 |
n | -0.277 |
interests | -0.277 |
victory | -0.276 |
join | -0.271 |
Token |
|
Token ID | 198 |
Feature activation | -2.57e+02 |
Pos logit contributions | |
needed | +0.238 |
whispered | +0.231 |
meant | +0.220 |
Jones | +0.203 |
faded | +0.191 |
Neg logit contributions | |
n | -0.468 |
disl | -0.461 |
pieces | -0.455 |
interests | -0.446 |
iu | -0.442 |
Token | but |
Token ID | 475 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
needed | +0.270 |
whispered | +0.264 |
meant | +0.255 |
Jones | +0.247 |
spoke | +0.235 |
Neg logit contributions | |
disl | -0.371 |
n | -0.366 |
interests | -0.365 |
arrest | -0.361 |
victory | -0.359 |
Token | they |
Token ID | 484 |
Feature activation | -2.37e+02 |
Pos logit contributions | |
needed | +0.298 |
whispered | +0.297 |
meant | +0.280 |
Jones | +0.277 |
chy | +0.257 |
Neg logit contributions | |
n | -0.354 |
disl | -0.349 |
pieces | -0.338 |
interests | -0.336 |
victory | -0.332 |
Token | were |
Token ID | 547 |
Feature activation | -2.41e+02 |
Pos logit contributions | |
Jones | +0.282 |
whispered | +0.263 |
White | +0.257 |
aisle | +0.253 |
meant | +0.251 |
Neg logit contributions | |
n | -0.363 |
disl | -0.340 |
interests | -0.327 |
spend | -0.326 |
hey | -0.325 |
Token | more |
Token ID | 517 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
needed | +0.271 |
whispered | +0.265 |
meant | +0.254 |
said | +0.230 |
Jones | +0.229 |
Neg logit contributions | |
n | -0.398 |
disl | -0.395 |
pieces | -0.391 |
iu | -0.379 |
Majesty | -0.378 |
Token | careful |
Token ID | 8161 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.361 |
Jones | +0.350 |
whispered | +0.347 |
meant | +0.339 |
White | +0.330 |
Neg logit contributions | |
disl | -0.298 |
interests | -0.285 |
n | -0.284 |
Majesty | -0.272 |
pieces | -0.271 |
Token | . |
Token ID | 13 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.188 |
meant | +0.175 |
whispered | +0.163 |
said | +0.147 |
Jones | +0.134 |
Neg logit contributions | |
Majesty | -0.500 |
disl | -0.498 |
pieces | -0.494 |
arrest | -0.484 |
latable | -0.481 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.291 |
whispered | +0.291 |
meant | +0.271 |
spoke | +0.259 |
chy | +0.257 |
Neg logit contributions | |
disl | -0.401 |
interests | -0.396 |
n | -0.395 |
blast | -0.375 |
pieces | -0.375 |
Token | Anna |
Token ID | 31160 |
Feature activation | -2.47e+02 |
Pos logit contributions | |
whispered | +0.280 |
needed | +0.263 |
meant | +0.244 |
chy | +0.242 |
faded | +0.239 |
Neg logit contributions | |
n | -0.427 |
blast | -0.390 |
victory | -0.380 |
ric | -0.376 |
interests | -0.373 |
Token | and |
Token ID | 290 |
Feature activation | -2.57e+02 |
Pos logit contributions | |
Pr | +0.309 |
lessness | +0.303 |
crooked | +0.297 |
darker | +0.297 |
Fuel | +0.296 |
Neg logit contributions | |
n | -0.337 |
arrest | -0.323 |
disl | -0.312 |
interests | -0.303 |
blast | -0.301 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.49e+02 |
Pos logit contributions | |
needed | +0.235 |
whispered | +0.227 |
said | +0.223 |
meant | +0.206 |
Jones | +0.187 |
Neg logit contributions | |
pieces | -0.519 |
ric | -0.511 |
n | -0.511 |
Majesty | -0.510 |
occasions | -0.508 |
Token | were |
Token ID | 547 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
Jones | +0.262 |
White | +0.261 |
whispered | +0.254 |
chy | +0.241 |
blank | +0.240 |
Neg logit contributions | |
spend | -0.380 |
n | -0.377 |
disl | -0.374 |
arrest | -0.372 |
blast | -0.363 |
Token | have |
Token ID | 423 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.438 |
whispered | +0.433 |
meant | +0.423 |
Jones | +0.412 |
faded | +0.396 |
Neg logit contributions | |
n | -0.210 |
disl | -0.205 |
pieces | -0.200 |
interests | -0.188 |
Majesty | -0.187 |
Token | fun |
Token ID | 1257 |
Feature activation | -2.36e+02 |
Pos logit contributions | |
needed | +0.371 |
whispered | +0.366 |
meant | +0.354 |
Jones | +0.347 |
faded | +0.327 |
Neg logit contributions | |
n | -0.284 |
disl | -0.279 |
pieces | -0.274 |
interests | -0.267 |
Majesty | -0.262 |
Token | . |
Token ID | 13 |
Feature activation | -2.47e+02 |
Pos logit contributions | |
needed | +0.148 |
whispered | +0.140 |
meant | +0.128 |
said | +0.122 |
dark | +0.104 |
Neg logit contributions | |
disl | -0.504 |
n | -0.496 |
Majesty | -0.495 |
earned | -0.494 |
pieces | -0.489 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.49e+02 |
Pos logit contributions | |
needed | +0.269 |
whispered | +0.269 |
meant | +0.248 |
Jones | +0.236 |
spoke | +0.230 |
Neg logit contributions | |
disl | -0.410 |
n | -0.409 |
interests | -0.404 |
blast | -0.390 |
pieces | -0.388 |
Token | Ben |
Token ID | 11696 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
whispered | +0.219 |
needed | +0.214 |
meant | +0.192 |
faded | +0.179 |
chy | +0.177 |
Neg logit contributions | |
n | -0.468 |
blast | -0.439 |
victory | -0.434 |
disl | -0.433 |
arrest | -0.431 |
Token | likes |
Token ID | 7832 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
whispered | +0.362 |
needed | +0.350 |
White | +0.346 |
Jones | +0.345 |
faded | +0.340 |
Neg logit contributions | |
disl | -0.286 |
n | -0.272 |
interests | -0.269 |
our | -0.269 |
arrest | -0.263 |
Token | to |
Token ID | 284 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.355 |
whispered | +0.348 |
meant | +0.337 |
Jones | +0.320 |
said | +0.311 |
Neg logit contributions | |
n | -0.341 |
disl | -0.336 |
pieces | -0.332 |
Majesty | -0.320 |
interests | -0.319 |
Token | play |
Token ID | 711 |
Feature activation | -2.55e+02 |
Pos logit contributions | |
needed | +0.339 |
whispered | +0.335 |
said | +0.328 |
meant | +0.304 |
spoke | +0.292 |
Neg logit contributions | |
ieved | -0.369 |
hey | -0.369 |
iu | -0.352 |
Majesty | -0.348 |
ric | -0.346 |
Token | with |
Token ID | 351 |
Feature activation | -2.57e+02 |
Pos logit contributions | |
needed | +0.245 |
whispered | +0.244 |
said | +0.222 |
meant | +0.220 |
dark | +0.211 |
Neg logit contributions | |
disl | -0.461 |
ric | -0.454 |
Majesty | -0.453 |
pieces | -0.447 |
iu | -0.447 |
Token | his |
Token ID | 465 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.312 |
whispered | +0.302 |
meant | +0.292 |
Jones | +0.272 |
said | +0.270 |
Neg logit contributions | |
n | -0.403 |
pieces | -0.397 |
disl | -0.390 |
Majesty | -0.387 |
iu | -0.380 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.350 |
whispered | +0.342 |
meant | +0.329 |
Jones | +0.316 |
said | +0.307 |
Neg logit contributions | |
n | -0.356 |
disl | -0.345 |
pieces | -0.343 |
Majesty | -0.331 |
iu | -0.330 |
Token | OK |
Token ID | 7477 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.321 |
whispered | +0.321 |
meant | +0.309 |
bored | +0.303 |
said | +0.299 |
Neg logit contributions | |
disl | -0.356 |
earned | -0.354 |
occasions | -0.353 |
pieces | -0.350 |
interests | -0.348 |
Token | ," |
Token ID | 553 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
needed | +0.205 |
Jones | +0.198 |
whispered | +0.195 |
meant | +0.191 |
said | +0.188 |
Neg logit contributions | |
disl | -0.472 |
pieces | -0.471 |
n | -0.453 |
earned | -0.453 |
latable | -0.449 |
Token | Tom |
Token ID | 4186 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
needed | +0.316 |
whispered | +0.300 |
meant | +0.296 |
Jones | +0.284 |
chy | +0.274 |
Neg logit contributions | |
n | -0.377 |
disl | -0.375 |
pieces | -0.367 |
interests | -0.355 |
arrest | -0.354 |
Token | and |
Token ID | 290 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
needed | +0.308 |
faded | +0.275 |
shone | +0.269 |
chy | +0.256 |
smelled | +0.255 |
Neg logit contributions | |
n | -0.360 |
interests | -0.342 |
disl | -0.334 |
our | -0.318 |
pause | -0.316 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.44e+02 |
Pos logit contributions | |
needed | +0.240 |
whispered | +0.235 |
meant | +0.217 |
said | +0.205 |
Jones | +0.196 |
Neg logit contributions | |
n | -0.460 |
pieces | -0.458 |
disl | -0.452 |
ric | -0.445 |
Majesty | -0.443 |
Token | say |
Token ID | 910 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.403 |
whispered | +0.374 |
meant | +0.374 |
faded | +0.364 |
Jones | +0.363 |
Neg logit contributions | |
n | -0.278 |
disl | -0.260 |
pieces | -0.253 |
interests | -0.247 |
hey | -0.243 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e+02 |
Pos logit contributions | |
needed | +0.219 |
whispered | +0.213 |
meant | +0.201 |
Jones | +0.185 |
said | +0.176 |
Neg logit contributions | |
n | -0.477 |
disl | -0.473 |
pieces | -0.468 |
Majesty | -0.456 |
interests | -0.455 |
Token | " |
Token ID | 366 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
needed | +0.341 |
chy | +0.331 |
whispered | +0.330 |
meant | +0.320 |
spoke | +0.300 |
Neg logit contributions | |
interests | -0.356 |
blast | -0.345 |
disl | -0.341 |
our | -0.340 |
n | -0.332 |
Token | But |
Token ID | 1537 |
Feature activation | -2.60e+02 |
Pos logit contributions | |
needed | +0.405 |
whispered | +0.400 |
meant | +0.387 |
Jones | +0.369 |
said | +0.354 |
Neg logit contributions | |
n | -0.316 |
disl | -0.307 |
interests | -0.293 |
pieces | -0.286 |
hey | -0.283 |
Token | we |
Token ID | 356 |
Feature activation | -2.59e+02 |
Pos logit contributions | |
needed | +0.314 |
said | +0.307 |
meant | +0.284 |
whispered | +0.270 |
smelled | +0.257 |
Neg logit contributions | |
pieces | -0.487 |
earned | -0.485 |
arrest | -0.476 |
latable | -0.462 |
impression | -0.455 |
Token | were |
Token ID | 547 |
Feature activation | -2.63e+02 |
Pos logit contributions | |
needed | +0.322 |
whispered | +0.301 |
meant | +0.297 |
said | +0.276 |
Jones | +0.252 |
Neg logit contributions | |
pieces | -0.432 |
Majesty | -0.419 |
iu | -0.414 |
disl | -0.414 |
n | -0.412 |
Token | go |
Token ID | 467 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
needed | +0.326 |
said | +0.322 |
whispered | +0.303 |
smelled | +0.283 |
meant | +0.281 |
Neg logit contributions | |
Majesty | -0.399 |
heit | -0.387 |
iu | -0.384 |
disl | -0.380 |
pieces | -0.369 |
Token | outside |
Token ID | 2354 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
whispered | +0.252 |
needed | +0.244 |
said | +0.243 |
dark | +0.225 |
bored | +0.209 |
Neg logit contributions | |
disl | -0.475 |
iu | -0.448 |
Majesty | -0.438 |
accuracy | -0.425 |
heit | -0.424 |
Token | now |
Token ID | 783 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.293 |
whispered | +0.288 |
meant | +0.276 |
Jones | +0.263 |
said | +0.255 |
Neg logit contributions | |
n | -0.364 |
disl | -0.362 |
pieces | -0.355 |
iu | -0.345 |
interests | -0.344 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
whispered | +0.317 |
needed | +0.311 |
meant | +0.302 |
said | +0.299 |
Jones | +0.298 |
Neg logit contributions | |
pieces | -0.367 |
disl | -0.363 |
n | -0.349 |
ric | -0.349 |
interests | -0.348 |
Token | Anna |
Token ID | 11735 |
Feature activation | -2.43e+02 |
Pos logit contributions | |
needed | +0.328 |
meant | +0.312 |
whispered | +0.311 |
chy | +0.304 |
Jones | +0.291 |
Neg logit contributions | |
n | -0.342 |
blast | -0.337 |
disl | -0.336 |
pieces | -0.333 |
our | -0.331 |
Token | and |
Token ID | 290 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
chy | +0.330 |
flowed | +0.310 |
needed | +0.308 |
shone | +0.303 |
faded | +0.301 |
Neg logit contributions | |
n | -0.329 |
interests | -0.288 |
iu | -0.276 |
han | -0.264 |
Amen | -0.253 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.45e+02 |
Pos logit contributions | |
needed | +0.233 |
meant | +0.217 |
whispered | +0.216 |
Jones | +0.188 |
faded | +0.184 |
Neg logit contributions | |
n | -0.450 |
pieces | -0.445 |
disl | -0.443 |
arrest | -0.440 |
iu | -0.436 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
chy | +0.313 |
oky | +0.288 |
faded | +0.286 |
Lover | +0.286 |
needed | +0.282 |
Neg logit contributions | |
n | -0.369 |
pause | -0.328 |
spend | -0.323 |
interests | -0.292 |
iu | -0.290 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e+02 |
Pos logit contributions | |
needed | +0.322 |
whispered | +0.307 |
meant | +0.307 |
faded | +0.288 |
spoke | +0.279 |
Neg logit contributions | |
n | -0.375 |
interests | -0.368 |
pieces | -0.363 |
ap | -0.362 |
victory | -0.360 |
Token |
|
Token ID | 198 |
Feature activation | -2.60e+02 |
Pos logit contributions | |
needed | +0.402 |
chy | +0.399 |
whispered | +0.380 |
meant | +0.376 |
faded | +0.355 |
Neg logit contributions | |
n | -0.301 |
interests | -0.298 |
blast | -0.294 |
join | -0.281 |
our | -0.280 |
Token |
|
Token ID | 198 |
Feature activation | -2.61e+02 |
Pos logit contributions | |
needed | +0.245 |
whispered | +0.236 |
meant | +0.228 |
Jones | +0.208 |
faded | +0.197 |
Neg logit contributions | |
n | -0.475 |
disl | -0.467 |
pieces | -0.460 |
interests | -0.451 |
Majesty | -0.448 |
Token | Can |
Token ID | 6090 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
needed | +0.345 |
whispered | +0.313 |
Jones | +0.310 |
bored | +0.310 |
chy | +0.310 |
Neg logit contributions | |
n | -0.319 |
blast | -0.307 |
interests | -0.290 |
hey | -0.288 |
occasions | -0.282 |
Token | we |
Token ID | 356 |
Feature activation | -2.37e+02 |
Pos logit contributions | |
needed | +0.292 |
said | +0.266 |
whispered | +0.242 |
she | +0.239 |
could | +0.237 |
Neg logit contributions | |
Majesty | -0.442 |
impression | -0.429 |
pieces | -0.425 |
disl | -0.422 |
latable | -0.420 |
Token | go |
Token ID | 467 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
needed | +0.326 |
said | +0.322 |
whispered | +0.303 |
smelled | +0.283 |
meant | +0.281 |
Neg logit contributions | |
Majesty | -0.399 |
heit | -0.387 |
iu | -0.384 |
disl | -0.380 |
pieces | -0.369 |
Token | outside |
Token ID | 2354 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
whispered | +0.252 |
needed | +0.244 |
said | +0.243 |
dark | +0.225 |
bored | +0.209 |
Neg logit contributions | |
disl | -0.475 |
iu | -0.448 |
Majesty | -0.438 |
accuracy | -0.425 |
heit | -0.424 |
Token | now |
Token ID | 783 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.293 |
whispered | +0.288 |
meant | +0.276 |
Jones | +0.263 |
said | +0.255 |
Neg logit contributions | |
n | -0.364 |
disl | -0.362 |
pieces | -0.355 |
iu | -0.345 |
interests | -0.344 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
whispered | +0.317 |
needed | +0.311 |
meant | +0.302 |
said | +0.299 |
Jones | +0.298 |
Neg logit contributions | |
pieces | -0.367 |
disl | -0.363 |
n | -0.349 |
ric | -0.349 |
interests | -0.348 |
Token | Anna |
Token ID | 11735 |
Feature activation | -2.43e+02 |
Pos logit contributions | |
needed | +0.328 |
meant | +0.312 |
whispered | +0.311 |
chy | +0.304 |
Jones | +0.291 |
Neg logit contributions | |
n | -0.342 |
blast | -0.337 |
disl | -0.336 |
pieces | -0.333 |
our | -0.331 |
Token | and |
Token ID | 290 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
chy | +0.330 |
flowed | +0.310 |
needed | +0.308 |
shone | +0.303 |
faded | +0.301 |
Neg logit contributions | |
n | -0.329 |
interests | -0.288 |
iu | -0.276 |
han | -0.264 |
Amen | -0.253 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.45e+02 |
Pos logit contributions | |
needed | +0.233 |
meant | +0.217 |
whispered | +0.216 |
Jones | +0.188 |
faded | +0.184 |
Neg logit contributions | |
n | -0.450 |
pieces | -0.445 |
disl | -0.443 |
arrest | -0.440 |
iu | -0.436 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
chy | +0.313 |
oky | +0.288 |
faded | +0.286 |
Lover | +0.286 |
needed | +0.282 |
Neg logit contributions | |
n | -0.369 |
pause | -0.328 |
spend | -0.323 |
interests | -0.292 |
iu | -0.290 |
Token | . |
Token ID | 13 |
Feature activation | -2.63e+02 |
Pos logit contributions | |
needed | +0.322 |
whispered | +0.307 |
meant | +0.307 |
faded | +0.288 |
spoke | +0.279 |
Neg logit contributions | |
n | -0.375 |
interests | -0.368 |
pieces | -0.363 |
ap | -0.362 |
victory | -0.360 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e+02 |
Pos logit contributions | |
whispered | +0.286 |
needed | +0.281 |
Jones | +0.273 |
meant | +0.272 |
spoke | +0.256 |
Neg logit contributions | |
n | -0.352 |
victory | -0.344 |
pieces | -0.334 |
ap | -0.328 |
blast | -0.326 |
Token | They |
Token ID | 1119 |
Feature activation | -2.45e+02 |
Pos logit contributions | |
needed | +0.263 |
whispered | +0.259 |
meant | +0.247 |
Jones | +0.226 |
spoke | +0.220 |
Neg logit contributions | |
n | -0.420 |
disl | -0.410 |
pieces | -0.409 |
interests | -0.404 |
Majesty | -0.394 |
Token | are |
Token ID | 389 |
Feature activation | -2.47e+02 |
Pos logit contributions | |
needed | +0.444 |
whispered | +0.435 |
meant | +0.420 |
said | +0.403 |
Jones | +0.401 |
Neg logit contributions | |
disl | -0.234 |
n | -0.233 |
pieces | -0.228 |
iu | -0.221 |
Majesty | -0.220 |
Token | clever |
Token ID | 14169 |
Feature activation | -2.44e+02 |
Pos logit contributions | |
needed | +0.264 |
whispered | +0.259 |
meant | +0.249 |
Jones | +0.227 |
said | +0.223 |
Neg logit contributions | |
n | -0.417 |
disl | -0.415 |
pieces | -0.408 |
iu | -0.397 |
interests | -0.396 |
Token | . |
Token ID | 13 |
Feature activation | -2.55e+02 |
Pos logit contributions | |
needed | +0.250 |
whispered | +0.244 |
meant | +0.231 |
Jones | +0.219 |
faded | +0.206 |
Neg logit contributions | |
n | -0.418 |
disl | -0.412 |
pieces | -0.409 |
interests | -0.401 |
Majesty | -0.398 |
Token | They |
Token ID | 1119 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.257 |
whispered | +0.252 |
meant | +0.240 |
Jones | +0.219 |
faded | +0.213 |
Neg logit contributions | |
n | -0.447 |
disl | -0.437 |
pieces | -0.435 |
interests | -0.431 |
Majesty | -0.418 |
Token | know |
Token ID | 760 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
needed | +0.440 |
whispered | +0.431 |
meant | +0.416 |
said | +0.400 |
Jones | +0.390 |
Neg logit contributions | |
n | -0.258 |
disl | -0.256 |
pieces | -0.255 |
iu | -0.250 |
Majesty | -0.245 |
Token | how |
Token ID | 703 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
needed | +0.476 |
whispered | +0.470 |
meant | +0.459 |
said | +0.436 |
Jones | +0.435 |
Neg logit contributions | |
n | -0.234 |
disl | -0.228 |
pieces | -0.222 |
iu | -0.215 |
arrest | -0.212 |
Token | to |
Token ID | 284 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
needed | +0.253 |
dark | +0.204 |
sleepy | +0.201 |
bored | +0.200 |
whispered | +0.195 |
Neg logit contributions | |
iu | -0.514 |
Majesty | -0.499 |
han | -0.495 |
n | -0.493 |
pieces | -0.488 |
Token | share |
Token ID | 2648 |
Feature activation | -2.62e+02 |
Pos logit contributions | |
needed | +0.352 |
whispered | +0.343 |
chy | +0.341 |
said | +0.328 |
meant | +0.328 |
Neg logit contributions | |
iu | -0.383 |
pieces | -0.369 |
disl | -0.363 |
Majesty | -0.363 |
interests | -0.338 |
Token | and |
Token ID | 290 |
Feature activation | -2.64e+02 |
Pos logit contributions | |
whispered | +0.256 |
needed | +0.249 |
meant | +0.244 |
said | +0.227 |
Jones | +0.218 |
Neg logit contributions | |
n | -0.464 |
Majesty | -0.456 |
disl | -0.454 |
iu | -0.454 |
pieces | -0.454 |
Token | and |
Token ID | 290 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
needed | +0.210 |
whispered | +0.201 |
meant | +0.196 |
Jones | +0.180 |
said | +0.178 |
Neg logit contributions | |
n | -0.438 |
disl | -0.434 |
pieces | -0.434 |
iu | -0.426 |
ric | -0.424 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.260 |
whispered | +0.252 |
meant | +0.241 |
said | +0.231 |
dark | +0.206 |
Neg logit contributions | |
n | -0.429 |
ric | -0.413 |
iu | -0.410 |
pieces | -0.408 |
arrest | -0.397 |
Token | and |
Token ID | 290 |
Feature activation | -2.41e+02 |
Pos logit contributions | |
needed | +0.260 |
whispered | +0.252 |
meant | +0.243 |
said | +0.231 |
Jones | +0.227 |
Neg logit contributions | |
n | -0.405 |
pieces | -0.405 |
disl | -0.395 |
iu | -0.385 |
ric | -0.384 |
Token | good |
Token ID | 922 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.299 |
whispered | +0.293 |
meant | +0.281 |
said | +0.263 |
Jones | +0.256 |
Neg logit contributions | |
n | -0.377 |
pieces | -0.367 |
iu | -0.359 |
disl | -0.359 |
ric | -0.357 |
Token | things |
Token ID | 1243 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
needed | +0.278 |
whispered | +0.273 |
meant | +0.260 |
Jones | +0.246 |
faded | +0.235 |
Neg logit contributions | |
n | -0.378 |
disl | -0.374 |
pieces | -0.367 |
interests | -0.359 |
Majesty | -0.358 |
Token | ," |
Token ID | 553 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.325 |
whispered | +0.320 |
Jones | +0.308 |
White | +0.297 |
meant | +0.292 |
Neg logit contributions | |
n | -0.320 |
disl | -0.314 |
interests | -0.299 |
victory | -0.298 |
our | -0.292 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
needed | +0.368 |
whispered | +0.353 |
meant | +0.352 |
chy | +0.341 |
faded | +0.331 |
Neg logit contributions | |
disl | -0.328 |
pieces | -0.321 |
n | -0.320 |
arrest | -0.311 |
victory | -0.310 |
Token | says |
Token ID | 1139 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
needed | +0.278 |
faded | +0.249 |
shone | +0.234 |
chy | +0.232 |
smelled | +0.222 |
Neg logit contributions | |
n | -0.383 |
disl | -0.373 |
interests | -0.368 |
pieces | -0.354 |
Majesty | -0.350 |
Token | . |
Token ID | 13 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
needed | +0.240 |
whispered | +0.232 |
meant | +0.225 |
Jones | +0.205 |
faded | +0.194 |
Neg logit contributions | |
n | -0.448 |
pieces | -0.438 |
disl | -0.438 |
iu | -0.428 |
interests | -0.426 |
Token |
|
Token ID | 198 |
Feature activation | -2.55e+02 |
Pos logit contributions | |
chy | +0.428 |
needed | +0.420 |
meant | +0.401 |
whispered | +0.396 |
dusty | +0.376 |
Neg logit contributions | |
n | -0.249 |
interests | -0.249 |
blast | -0.248 |
disl | -0.244 |
join | -0.244 |
Token |
|
Token ID | 198 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
needed | +0.231 |
whispered | +0.224 |
meant | +0.212 |
Jones | +0.195 |
said | +0.183 |
Neg logit contributions | |
n | -0.472 |
disl | -0.467 |
pieces | -0.461 |
interests | -0.450 |
Majesty | -0.448 |
Token | . |
Token ID | 13 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.247 |
whispered | +0.240 |
meant | +0.229 |
Jones | +0.226 |
White | +0.208 |
Neg logit contributions | |
n | -0.388 |
pieces | -0.387 |
disl | -0.378 |
interests | -0.376 |
victory | -0.374 |
Token | They |
Token ID | 1119 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
needed | +0.309 |
whispered | +0.302 |
meant | +0.283 |
chy | +0.269 |
spoke | +0.268 |
Neg logit contributions | |
disl | -0.370 |
n | -0.369 |
interests | -0.366 |
blast | -0.350 |
victory | -0.340 |
Token | are |
Token ID | 389 |
Feature activation | -2.41e+02 |
Pos logit contributions | |
needed | +0.423 |
whispered | +0.417 |
meant | +0.404 |
said | +0.385 |
Jones | +0.382 |
Neg logit contributions | |
n | -0.234 |
disl | -0.232 |
pieces | -0.226 |
iu | -0.221 |
Majesty | -0.217 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.265 |
whispered | +0.259 |
meant | +0.251 |
said | +0.231 |
bored | +0.226 |
Neg logit contributions | |
n | -0.408 |
disl | -0.402 |
pieces | -0.390 |
iu | -0.389 |
arrest | -0.388 |
Token | friends |
Token ID | 2460 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.233 |
whispered | +0.228 |
meant | +0.217 |
Jones | +0.201 |
said | +0.188 |
Neg logit contributions | |
n | -0.433 |
disl | -0.430 |
pieces | -0.424 |
interests | -0.415 |
Majesty | -0.413 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.248 |
whispered | +0.245 |
meant | +0.231 |
Jones | +0.214 |
spoke | +0.206 |
Neg logit contributions | |
n | -0.400 |
pieces | -0.391 |
disl | -0.388 |
victory | -0.384 |
interests | -0.381 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.274 |
whispered | +0.272 |
meant | +0.253 |
Jones | +0.235 |
faded | +0.232 |
Neg logit contributions | |
disl | -0.419 |
n | -0.418 |
interests | -0.412 |
pieces | -0.395 |
blast | -0.395 |
Token | L |
Token ID | 43 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
whispered | +0.235 |
needed | +0.229 |
meant | +0.207 |
faded | +0.195 |
chy | +0.193 |
Neg logit contributions | |
n | -0.463 |
blast | -0.435 |
victory | -0.432 |
interests | -0.428 |
disl | -0.427 |
Token | ily |
Token ID | 813 |
Feature activation | -2.45e+02 |
Pos logit contributions | |
whispered | +0.256 |
needed | +0.255 |
meant | +0.241 |
said | +0.228 |
Jones | +0.223 |
Neg logit contributions | |
disl | -0.415 |
n | -0.394 |
Majesty | -0.391 |
pieces | -0.391 |
interests | -0.385 |
Token | and |
Token ID | 290 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
whispered | +0.354 |
needed | +0.347 |
meant | +0.339 |
Jones | +0.335 |
White | +0.329 |
Neg logit contributions | |
disl | -0.291 |
n | -0.290 |
arrest | -0.287 |
interests | -0.278 |
Majesty | -0.277 |
Token | Tom |
Token ID | 4186 |
Feature activation | -2.48e+02 |
Pos logit contributions | |
needed | +0.259 |
said | +0.257 |
whispered | +0.237 |
meant | +0.233 |
She | +0.210 |
Neg logit contributions | |
Majesty | -0.535 |
immers | -0.530 |
pieces | -0.522 |
ric | -0.520 |
occasions | -0.519 |
Token | swim |
Token ID | 9422 |
Feature activation | -2.33e+02 |
Pos logit contributions | |
needed | +0.277 |
whispered | +0.264 |
dark | +0.261 |
said | +0.260 |
sleepy | +0.255 |
Neg logit contributions | |
pieces | -0.386 |
iu | -0.377 |
interests | -0.377 |
han | -0.371 |
nels | -0.361 |
Token | with |
Token ID | 351 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.338 |
whispered | +0.330 |
meant | +0.313 |
Jones | +0.302 |
said | +0.301 |
Neg logit contributions | |
disl | -0.310 |
n | -0.304 |
pieces | -0.303 |
iu | -0.295 |
Majesty | -0.290 |
Token | you |
Token ID | 345 |
Feature activation | -2.35e+02 |
Pos logit contributions | |
needed | +0.375 |
whispered | +0.366 |
said | +0.354 |
dark | +0.342 |
meant | +0.338 |
Neg logit contributions | |
disl | -0.281 |
iu | -0.280 |
pieces | -0.278 |
ric | -0.277 |
interests | -0.272 |
Token | , |
Token ID | 11 |
Feature activation | -2.44e+02 |
Pos logit contributions | |
needed | +0.408 |
meant | +0.406 |
whispered | +0.404 |
faded | +0.381 |
spoke | +0.378 |
Neg logit contributions | |
n | -0.234 |
Majesty | -0.222 |
pieces | -0.216 |
victory | -0.216 |
interests | -0.207 |
Token | duck |
Token ID | 22045 |
Feature activation | -2.34e+02 |
Pos logit contributions | |
needed | +0.302 |
whispered | +0.296 |
meant | +0.284 |
Jones | +0.269 |
faded | +0.257 |
Neg logit contributions | |
n | -0.372 |
disl | -0.366 |
pieces | -0.361 |
Majesty | -0.350 |
interests | -0.349 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.380 |
whispered | +0.367 |
meant | +0.355 |
Jones | +0.342 |
said | +0.340 |
Neg logit contributions | |
n | -0.268 |
disl | -0.268 |
pieces | -0.265 |
iu | -0.260 |
interests | -0.255 |
Token | And |
Token ID | 843 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
meant | +0.330 |
chy | +0.328 |
needed | +0.323 |
faded | +0.312 |
dusty | +0.301 |
Neg logit contributions | |
n | -0.385 |
disl | -0.331 |
iu | -0.316 |
pieces | -0.308 |
arrest | -0.307 |
Token | they |
Token ID | 484 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
needed | +0.336 |
meant | +0.320 |
whispered | +0.317 |
Jones | +0.311 |
chy | +0.303 |
Neg logit contributions | |
n | -0.362 |
pieces | -0.347 |
victory | -0.329 |
disl | -0.326 |
interests | -0.325 |
Token | sw |
Token ID | 1509 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.342 |
meant | +0.342 |
Jones | +0.340 |
whispered | +0.337 |
aisle | +0.316 |
Neg logit contributions | |
n | -0.334 |
blast | -0.304 |
spend | -0.301 |
victory | -0.298 |
pieces | -0.295 |
Token | am |
Token ID | 321 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
needed | +0.291 |
whispered | +0.289 |
said | +0.272 |
meant | +0.254 |
ed | +0.253 |
Neg logit contributions | |
disl | -0.357 |
pieces | -0.350 |
hey | -0.337 |
blast | -0.333 |
risk | -0.325 |
Token | together |
Token ID | 1978 |
Feature activation | -2.54e+02 |
Pos logit contributions | |
needed | +0.272 |
whispered | +0.271 |
said | +0.262 |
dark | +0.240 |
meant | +0.240 |
Neg logit contributions | |
disl | -0.465 |
ric | -0.439 |
Majesty | -0.431 |
pieces | -0.425 |
hey | -0.415 |
Token | bad |
Token ID | 2089 |
Feature activation | -2.30e+02 |
Pos logit contributions | |
needed | +0.332 |
whispered | +0.326 |
meant | +0.317 |
Jones | +0.299 |
said | +0.292 |
Neg logit contributions | |
n | -0.308 |
disl | -0.304 |
pieces | -0.301 |
Majesty | -0.289 |
iu | -0.288 |
Token | . |
Token ID | 13 |
Feature activation | -2.41e+02 |
Pos logit contributions | |
needed | +0.269 |
whispered | +0.269 |
meant | +0.252 |
Jones | +0.250 |
White | +0.236 |
Neg logit contributions | |
n | -0.354 |
disl | -0.345 |
interests | -0.339 |
pieces | -0.333 |
arrest | -0.330 |
Token | It |
Token ID | 632 |
Feature activation | -2.35e+02 |
Pos logit contributions | |
needed | +0.281 |
meant | +0.271 |
whispered | +0.269 |
Jones | +0.260 |
White | +0.253 |
Neg logit contributions | |
pieces | -0.395 |
n | -0.375 |
victory | -0.367 |
Majesty | -0.366 |
nels | -0.365 |
Token | hurt |
Token ID | 5938 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
needed | +0.224 |
whispered | +0.222 |
meant | +0.208 |
Jones | +0.204 |
faded | +0.185 |
Neg logit contributions | |
n | -0.419 |
disl | -0.411 |
pieces | -0.401 |
arrest | -0.393 |
interests | -0.392 |
Token | me |
Token ID | 502 |
Feature activation | -2.35e+02 |
Pos logit contributions | |
needed | +0.457 |
whispered | +0.454 |
meant | +0.439 |
Jones | +0.430 |
spoke | +0.421 |
Neg logit contributions | |
n | -0.195 |
disl | -0.181 |
interests | -0.178 |
arrest | -0.172 |
victory | -0.170 |
Token | ." |
Token ID | 526 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.333 |
whispered | +0.323 |
spoke | +0.319 |
Jones | +0.315 |
faded | +0.314 |
Neg logit contributions | |
n | -0.318 |
interests | -0.274 |
our | -0.274 |
victory | -0.272 |
pieces | -0.270 |
Token | Anna |
Token ID | 11735 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
meant | +0.345 |
needed | +0.344 |
whispered | +0.343 |
chy | +0.320 |
Jones | +0.312 |
Neg logit contributions | |
blast | -0.316 |
victory | -0.312 |
disl | -0.306 |
n | -0.302 |
our | -0.300 |
Token | said |
Token ID | 531 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
chy | +0.248 |
aisle | +0.233 |
Today | +0.232 |
blank | +0.230 |
books | +0.229 |
Neg logit contributions | |
n | -0.401 |
iu | -0.354 |
hey | -0.347 |
arrest | -0.341 |
punish | -0.337 |
Token | . |
Token ID | 13 |
Feature activation | -2.57e+02 |
Pos logit contributions | |
needed | +0.229 |
whispered | +0.222 |
meant | +0.211 |
Jones | +0.194 |
said | +0.182 |
Neg logit contributions | |
n | -0.462 |
disl | -0.455 |
pieces | -0.451 |
interests | -0.440 |
iu | -0.440 |
Token |
|
Token ID | 198 |
Feature activation | -2.54e+02 |
Pos logit contributions | |
needed | +0.390 |
chy | +0.373 |
meant | +0.364 |
whispered | +0.355 |
shone | +0.348 |
Neg logit contributions | |
blast | -0.274 |
hey | -0.270 |
join | -0.267 |
n | -0.264 |
interests | -0.259 |
Token |
|
Token ID | 198 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
needed | +0.285 |
meant | +0.268 |
whispered | +0.268 |
Jones | +0.241 |
faded | +0.239 |
Neg logit contributions | |
n | -0.421 |
disl | -0.416 |
arrest | -0.398 |
interests | -0.397 |
pieces | -0.391 |
Token | . |
Token ID | 13 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
needed | +0.264 |
whispered | +0.254 |
meant | +0.252 |
spoke | +0.229 |
Jones | +0.227 |
Neg logit contributions | |
n | -0.362 |
pieces | -0.346 |
disl | -0.344 |
victory | -0.343 |
interests | -0.341 |
Token | They |
Token ID | 1119 |
Feature activation | -2.35e+02 |
Pos logit contributions | |
needed | +0.334 |
whispered | +0.318 |
meant | +0.314 |
chy | +0.309 |
spoke | +0.294 |
Neg logit contributions | |
disl | -0.321 |
blast | -0.316 |
our | -0.314 |
interests | -0.313 |
n | -0.306 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
Jones | +0.295 |
elly | +0.279 |
oky | +0.278 |
aisle | +0.274 |
Fuel | +0.271 |
Neg logit contributions | |
blast | -0.331 |
spend | -0.327 |
n | -0.325 |
stay | -0.310 |
pause | -0.309 |
Token | and |
Token ID | 290 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.310 |
aisle | +0.301 |
whispered | +0.297 |
meant | +0.283 |
spoke | +0.281 |
Neg logit contributions | |
victory | -0.326 |
interests | -0.321 |
our | -0.311 |
ap | -0.305 |
Majesty | -0.303 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.44e+02 |
Pos logit contributions | |
Jones | +0.237 |
aisle | +0.237 |
needed | +0.237 |
chy | +0.229 |
meant | +0.223 |
Neg logit contributions | |
blast | -0.382 |
n | -0.375 |
disl | -0.368 |
victory | -0.364 |
interests | -0.362 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.344 |
whispered | +0.327 |
aisle | +0.324 |
meant | +0.315 |
Jones | +0.312 |
Neg logit contributions | |
victory | -0.329 |
ap | -0.302 |
interests | -0.301 |
pieces | -0.300 |
our | -0.299 |
Token | They |
Token ID | 1119 |
Feature activation | -2.48e+02 |
Pos logit contributions | |
needed | +0.331 |
whispered | +0.323 |
meant | +0.309 |
chy | +0.301 |
spoke | +0.291 |
Neg logit contributions | |
disl | -0.356 |
interests | -0.355 |
blast | -0.345 |
n | -0.343 |
our | -0.337 |
Token | had |
Token ID | 550 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
Jones | +0.309 |
aisle | +0.297 |
Fuel | +0.285 |
oky | +0.283 |
elly | +0.273 |
Neg logit contributions | |
n | -0.349 |
blast | -0.346 |
spend | -0.345 |
pause | -0.324 |
stay | -0.323 |
Token | a |
Token ID | 257 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
Jones | +0.405 |
aisle | +0.403 |
needed | +0.391 |
White | +0.376 |
meant | +0.371 |
Neg logit contributions | |
n | -0.280 |
disl | -0.274 |
interests | -0.270 |
victory | -0.268 |
blast | -0.266 |
Token | great |
Token ID | 1049 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
needed | +0.384 |
whispered | +0.379 |
meant | +0.372 |
Jones | +0.353 |
said | +0.336 |
Neg logit contributions | |
n | -0.302 |
disl | -0.302 |
victory | -0.293 |
pieces | -0.293 |
blast | -0.291 |
Token | adventure |
Token ID | 8855 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.274 |
whispered | +0.252 |
meant | +0.241 |
dark | +0.236 |
Jones | +0.229 |
Neg logit contributions | |
pieces | -0.434 |
disl | -0.421 |
n | -0.421 |
ric | -0.411 |
ieved | -0.405 |
Token | He |
Token ID | 679 |
Feature activation | -2.33e+02 |
Pos logit contributions | |
needed | +0.279 |
whispered | +0.267 |
meant | +0.253 |
spoke | +0.247 |
chy | +0.242 |
Neg logit contributions | |
disl | -0.389 |
Majesty | -0.375 |
yeah | -0.353 |
n | -0.349 |
interests | -0.346 |
Token | gave |
Token ID | 2921 |
Feature activation | -2.41e+02 |
Pos logit contributions | |
Jones | +0.261 |
chy | +0.225 |
White | +0.225 |
meant | +0.223 |
aisle | +0.221 |
Neg logit contributions | |
n | -0.365 |
disl | -0.342 |
arrest | -0.341 |
interests | -0.337 |
blast | -0.328 |
Token | her |
Token ID | 607 |
Feature activation | -2.37e+02 |
Pos logit contributions | |
whispered | +0.240 |
needed | +0.232 |
said | +0.222 |
meant | +0.222 |
Jones | +0.204 |
Neg logit contributions | |
disl | -0.472 |
pieces | -0.444 |
blast | -0.434 |
arrest | -0.432 |
interests | -0.429 |
Token | the |
Token ID | 262 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
whispered | +0.228 |
needed | +0.227 |
meant | +0.214 |
said | +0.205 |
dark | +0.197 |
Neg logit contributions | |
n | -0.430 |
iu | -0.426 |
Majesty | -0.424 |
hey | -0.411 |
occasions | -0.410 |
Token | ball |
Token ID | 2613 |
Feature activation | -2.34e+02 |
Pos logit contributions | |
needed | +0.324 |
whispered | +0.317 |
meant | +0.307 |
Jones | +0.291 |
said | +0.282 |
Neg logit contributions | |
n | -0.337 |
disl | -0.332 |
pieces | -0.327 |
Majesty | -0.316 |
iu | -0.316 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
Jones | +0.395 |
whispered | +0.390 |
needed | +0.373 |
meant | +0.373 |
White | +0.370 |
Neg logit contributions | |
victory | -0.238 |
n | -0.220 |
respect | -0.217 |
our | -0.217 |
interests | -0.212 |
Token | " |
Token ID | 366 |
Feature activation | -2.48e+02 |
Pos logit contributions | |
needed | +0.311 |
whispered | +0.294 |
meant | +0.279 |
spoke | +0.271 |
chy | +0.269 |
Neg logit contributions | |
disl | -0.400 |
Majesty | -0.383 |
n | -0.372 |
yeah | -0.370 |
interests | -0.356 |
Token | I |
Token ID | 40 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
needed | +0.333 |
Jones | +0.316 |
whispered | +0.315 |
meant | +0.308 |
chy | +0.293 |
Neg logit contributions | |
n | -0.367 |
disl | -0.361 |
interests | -0.332 |
arrest | -0.330 |
hey | -0.328 |
Token | 'm |
Token ID | 1101 |
Feature activation | -2.49e+02 |
Pos logit contributions | |
needed | +0.417 |
whispered | +0.416 |
meant | +0.402 |
Jones | +0.401 |
White | +0.381 |
Neg logit contributions | |
n | -0.270 |
disl | -0.266 |
interests | -0.245 |
pieces | -0.240 |
victory | -0.236 |
Token | sorry |
Token ID | 7926 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
whispered | +0.331 |
needed | +0.327 |
meant | +0.316 |
said | +0.307 |
bored | +0.298 |
Neg logit contributions | |
interests | -0.355 |
pieces | -0.354 |
disl | -0.354 |
occasions | -0.352 |
n | -0.351 |
Token | , |
Token ID | 11 |
Feature activation | -2.58e+02 |
Pos logit contributions | |
needed | +0.199 |
whispered | +0.184 |
meant | +0.181 |
said | +0.179 |
Jones | +0.169 |
Neg logit contributions | |
pieces | -0.516 |
n | -0.505 |
disl | -0.505 |
arrest | -0.499 |
iu | -0.490 |
Token | it |
Token ID | 340 |
Feature activation | -2.29e+02 |
Pos logit contributions | |
needed | +0.387 |
whispered | +0.375 |
meant | +0.366 |
Jones | +0.348 |
said | +0.345 |
Neg logit contributions | |
n | -0.259 |
disl | -0.254 |
pieces | -0.252 |
iu | -0.245 |
Majesty | -0.242 |
Token | is |
Token ID | 318 |
Feature activation | -2.35e+02 |
Pos logit contributions | |
Jones | +0.291 |
whispered | +0.276 |
White | +0.264 |
aisle | +0.263 |
Fuel | +0.259 |
Neg logit contributions | |
n | -0.344 |
disl | -0.316 |
arrest | -0.305 |
spend | -0.300 |
Majesty | -0.297 |
Token | ! |
Token ID | 0 |
Feature activation | -2.43e+02 |
Pos logit contributions | |
whispered | +0.334 |
Jones | +0.322 |
aisle | +0.320 |
White | +0.316 |
Fuel | +0.314 |
Neg logit contributions | |
n | -0.302 |
victory | -0.280 |
Majesty | -0.277 |
disl | -0.275 |
our | -0.274 |
Token | Follow |
Token ID | 7281 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.333 |
whispered | +0.321 |
meant | +0.318 |
Jones | +0.306 |
White | +0.296 |
Neg logit contributions | |
pieces | -0.333 |
n | -0.332 |
disl | -0.328 |
iu | -0.324 |
arrest | -0.322 |
Token | me |
Token ID | 502 |
Feature activation | -2.37e+02 |
Pos logit contributions | |
whispered | +0.271 |
Jones | +0.253 |
said | +0.253 |
T | +0.243 |
needed | +0.238 |
Neg logit contributions | |
pieces | -0.404 |
iu | -0.396 |
disl | -0.396 |
ric | -0.387 |
latable | -0.380 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
needed | +0.308 |
whispered | +0.307 |
meant | +0.289 |
Jones | +0.280 |
said | +0.271 |
Neg logit contributions | |
disl | -0.353 |
pieces | -0.340 |
iu | -0.339 |
n | -0.333 |
interests | -0.322 |
Token |
|
Token ID | 198 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
needed | +0.376 |
whispered | +0.371 |
meant | +0.367 |
faded | +0.362 |
chy | +0.360 |
Neg logit contributions | |
n | -0.319 |
disl | -0.300 |
pieces | -0.299 |
arrest | -0.281 |
blast | -0.271 |
Token |
|
Token ID | 198 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
needed | +0.219 |
whispered | +0.213 |
meant | +0.200 |
Jones | +0.183 |
said | +0.178 |
Neg logit contributions | |
n | -0.469 |
disl | -0.467 |
pieces | -0.462 |
Majesty | -0.454 |
iu | -0.451 |
Token | The |
Token ID | 464 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
needed | +0.354 |
whispered | +0.348 |
meant | +0.335 |
Jones | +0.316 |
said | +0.312 |
Neg logit contributions | |
n | -0.339 |
disl | -0.335 |
pieces | -0.330 |
Majesty | -0.323 |
iu | -0.320 |
Token | squirrel |
Token ID | 33039 |
Feature activation | -2.46e+02 |
Pos logit contributions | |
needed | +0.276 |
whispered | +0.271 |
meant | +0.258 |
Jones | +0.244 |
said | +0.230 |
Neg logit contributions | |
n | -0.420 |
disl | -0.414 |
pieces | -0.408 |
interests | -0.398 |
Majesty | -0.397 |
Token | led |
Token ID | 2957 |
Feature activation | -2.59e+02 |
Pos logit contributions | |
Jones | +0.304 |
bored | +0.284 |
meant | +0.269 |
needed | +0.269 |
dark | +0.266 |
Neg logit contributions | |
n | -0.406 |
arrest | -0.336 |
victory | -0.332 |
pause | -0.330 |
ered | -0.329 |
Token | forgot |
Token ID | 16453 |
Feature activation | -2.38e+02 |
Pos logit contributions | |
Jones | +0.327 |
aisle | +0.315 |
oky | +0.303 |
Fuel | +0.297 |
elly | +0.294 |
Neg logit contributions | |
n | -0.291 |
spend | -0.285 |
enjoy | -0.284 |
share | -0.276 |
blast | -0.272 |
Token | about |
Token ID | 546 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
needed | +0.409 |
whispered | +0.404 |
meant | +0.391 |
Jones | +0.381 |
faded | +0.367 |
Neg logit contributions | |
n | -0.246 |
disl | -0.242 |
pieces | -0.233 |
interests | -0.225 |
Majesty | -0.221 |
Token | lunch |
Token ID | 9965 |
Feature activation | -2.35e+02 |
Pos logit contributions | |
needed | +0.317 |
whispered | +0.311 |
meant | +0.300 |
Jones | +0.285 |
said | +0.275 |
Neg logit contributions | |
n | -0.343 |
disl | -0.338 |
pieces | -0.334 |
interests | -0.323 |
Majesty | -0.322 |
Token | . |
Token ID | 13 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
White | +0.372 |
Dead | +0.362 |
whispered | +0.362 |
Ancient | +0.358 |
Jones | +0.355 |
Neg logit contributions | |
victory | -0.245 |
n | -0.245 |
occasions | -0.228 |
ap | -0.226 |
interests | -0.225 |
Token | They |
Token ID | 1119 |
Feature activation | -2.46e+02 |
Pos logit contributions | |
whispered | +0.298 |
needed | +0.295 |
meant | +0.277 |
spoke | +0.265 |
chy | +0.258 |
Neg logit contributions | |
disl | -0.395 |
n | -0.376 |
interests | -0.374 |
blast | -0.361 |
nels | -0.351 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
Jones | +0.353 |
oky | +0.327 |
elly | +0.323 |
Fuel | +0.323 |
Ancient | +0.323 |
Neg logit contributions | |
spend | -0.310 |
n | -0.306 |
enjoy | -0.304 |
share | -0.291 |
wait | -0.282 |
Token | grapes |
Token ID | 36656 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.367 |
whispered | +0.360 |
meant | +0.349 |
Jones | +0.333 |
said | +0.323 |
Neg logit contributions | |
n | -0.327 |
disl | -0.322 |
pieces | -0.318 |
Majesty | -0.305 |
interests | -0.305 |
Token | . |
Token ID | 13 |
Feature activation | -2.60e+02 |
Pos logit contributions | |
needed | +0.317 |
whispered | +0.304 |
Jones | +0.300 |
meant | +0.293 |
White | +0.290 |
Neg logit contributions | |
n | -0.334 |
victory | -0.321 |
pieces | -0.308 |
interests | -0.305 |
disl | -0.302 |
Token | And |
Token ID | 843 |
Feature activation | -2.60e+02 |
Pos logit contributions | |
needed | +0.276 |
whispered | +0.274 |
meant | +0.255 |
Jones | +0.237 |
spoke | +0.235 |
Neg logit contributions | |
disl | -0.440 |
n | -0.432 |
interests | -0.423 |
pieces | -0.413 |
blast | -0.408 |
Token | they |
Token ID | 484 |
Feature activation | -2.57e+02 |
Pos logit contributions | |
needed | +0.266 |
whispered | +0.256 |
meant | +0.247 |
said | +0.228 |
Jones | +0.223 |
Neg logit contributions | |
n | -0.458 |
disl | -0.457 |
pieces | -0.454 |
arrest | -0.444 |
Majesty | -0.442 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.63e+02 |
Pos logit contributions | |
Jones | +0.348 |
aisle | +0.334 |
Fuel | +0.321 |
Ancient | +0.311 |
oky | +0.310 |
Neg logit contributions | |
n | -0.347 |
disl | -0.326 |
spend | -0.323 |
enjoy | -0.311 |
stay | -0.308 |
Token | . |
Token ID | 13 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
needed | +0.211 |
whispered | +0.207 |
meant | +0.194 |
Jones | +0.181 |
said | +0.178 |
Neg logit contributions | |
n | -0.421 |
disl | -0.420 |
pieces | -0.418 |
iu | -0.408 |
Majesty | -0.406 |
Token | Let |
Token ID | 3914 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
needed | +0.236 |
White | +0.233 |
whispered | +0.227 |
meant | +0.222 |
Jones | +0.221 |
Neg logit contributions | |
aches | -0.420 |
pieces | -0.407 |
ered | -0.392 |
n | -0.390 |
arrest | -0.388 |
Token | 's |
Token ID | 338 |
Feature activation | -2.35e+02 |
Pos logit contributions | |
said | +0.182 |
meant | +0.172 |
whispered | +0.170 |
needed | +0.170 |
dark | +0.131 |
Neg logit contributions | |
ric | -0.528 |
disl | -0.517 |
pieces | -0.502 |
iu | -0.502 |
Majesty | -0.498 |
Token | play |
Token ID | 711 |
Feature activation | -2.39e+02 |
Pos logit contributions | |
said | +0.174 |
went | +0.141 |
had | +0.141 |
made | +0.136 |
wide | +0.135 |
Neg logit contributions | |
earned | -0.560 |
ocol | -0.553 |
pieces | -0.503 |
beh | -0.503 |
iu | -0.498 |
Token | again |
Token ID | 757 |
Feature activation | -2.40e+02 |
Pos logit contributions | |
whispered | +0.262 |
needed | +0.253 |
said | +0.249 |
meant | +0.236 |
dark | +0.232 |
Neg logit contributions | |
ric | -0.401 |
iu | -0.393 |
disl | -0.391 |
earned | -0.386 |
n | -0.379 |
Token | ." |
Token ID | 526 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
needed | +0.269 |
whispered | +0.265 |
meant | +0.251 |
Jones | +0.237 |
said | +0.234 |
Neg logit contributions | |
n | -0.392 |
disl | -0.391 |
pieces | -0.388 |
iu | -0.377 |
Majesty | -0.375 |
Token | They |
Token ID | 1119 |
Feature activation | -2.48e+02 |
Pos logit contributions | |
needed | +0.278 |
meant | +0.270 |
whispered | +0.269 |
Jones | +0.239 |
chy | +0.237 |
Neg logit contributions | |
disl | -0.408 |
n | -0.408 |
interests | -0.400 |
pieces | -0.393 |
iu | -0.386 |
Token | take |
Token ID | 1011 |
Feature activation | -2.52e+02 |
Pos logit contributions | |
whispered | +0.368 |
said | +0.361 |
needed | +0.356 |
chy | +0.329 |
smelled | +0.310 |
Neg logit contributions | |
disl | -0.371 |
occasions | -0.354 |
earned | -0.349 |
Majesty | -0.344 |
iu | -0.341 |
Token | their |
Token ID | 511 |
Feature activation | -2.48e+02 |
Pos logit contributions | |
said | +0.307 |
needed | +0.305 |
whispered | +0.300 |
dark | +0.298 |
meant | +0.291 |
Neg logit contributions | |
disl | -0.436 |
ric | -0.418 |
Majesty | -0.407 |
n | -0.372 |
witness | -0.368 |
Token | boats |
Token ID | 16659 |
Feature activation | -2.47e+02 |
Pos logit contributions | |
needed | +0.301 |
whispered | +0.286 |
said | +0.268 |
meant | +0.268 |
dark | +0.262 |
Neg logit contributions | |
Majesty | -0.403 |
disl | -0.390 |
n | -0.383 |
ric | -0.377 |
pieces | -0.375 |
Token | and |
Token ID | 290 |
Feature activation | -2.56e+02 |
Pos logit contributions | |
whispered | +0.161 |
needed | +0.161 |
said | +0.140 |
meant | +0.138 |
Jones | +0.123 |
Neg logit contributions | |
Majesty | -0.526 |
disl | -0.524 |
n | -0.518 |
pieces | -0.511 |
iu | -0.502 |
Token | , |
Token ID | 11 |
Feature activation | -2.36e+02 |
Pos logit contributions | |
whispered | +0.217 |
needed | +0.210 |
Jones | +0.210 |
said | +0.207 |
meant | +0.199 |
Neg logit contributions | |
disl | -0.429 |
pieces | -0.419 |
ric | -0.408 |
n | -0.404 |
interests | -0.401 |
Token | duck |
Token ID | 22045 |
Feature activation | -2.26e+02 |
Pos logit contributions | |
whispered | +0.277 |
said | +0.270 |
meant | +0.269 |
needed | +0.265 |
dark | +0.235 |
Neg logit contributions | |
disl | -0.386 |
pieces | -0.385 |
ric | -0.370 |
occasions | -0.369 |
earned | -0.369 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.45e+02 |
Pos logit contributions | |
needed | +0.356 |
whispered | +0.351 |
meant | +0.341 |
Jones | +0.327 |
faded | +0.317 |
Neg logit contributions | |
n | -0.267 |
disl | -0.261 |
pieces | -0.255 |
Majesty | -0.246 |
interests | -0.245 |
Token | Mia |
Token ID | 32189 |
Feature activation | -2.34e+02 |
Pos logit contributions | |
needed | +0.298 |
whispered | +0.294 |
meant | +0.280 |
said | +0.265 |
Jones | +0.260 |
Neg logit contributions | |
n | -0.383 |
disl | -0.378 |
pieces | -0.373 |
Majesty | -0.365 |
interests | -0.363 |
Token | says |
Token ID | 1139 |
Feature activation | -2.43e+02 |
Pos logit contributions | |
needed | +0.215 |
whispered | +0.193 |
meant | +0.190 |
Jones | +0.177 |
faded | +0.170 |
Neg logit contributions | |
n | -0.433 |
disl | -0.425 |
pieces | -0.422 |
interests | -0.412 |
Majesty | -0.409 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
needed | +0.208 |
whispered | +0.204 |
meant | +0.190 |
Jones | +0.177 |
said | +0.167 |
Neg logit contributions | |
n | -0.461 |
disl | -0.459 |
pieces | -0.452 |
Majesty | -0.442 |
interests | -0.441 |
Token |
|
Token ID | 198 |
Feature activation | -2.49e+02 |
Pos logit contributions | |
needed | +0.380 |
meant | +0.369 |
chy | +0.360 |
whispered | +0.351 |
faded | +0.330 |
Neg logit contributions | |
n | -0.298 |
arrest | -0.293 |
ext | -0.283 |
blast | -0.282 |
join | -0.280 |
Token |
|
Token ID | 198 |
Feature activation | -2.50e+02 |
Pos logit contributions | |
needed | +0.208 |
whispered | +0.204 |
meant | +0.190 |
Jones | +0.175 |
said | +0.169 |
Neg logit contributions | |
n | -0.475 |
disl | -0.472 |
pieces | -0.470 |
Majesty | -0.457 |
iu | -0.455 |
Token | The |
Token ID | 464 |
Feature activation | -2.51e+02 |
Pos logit contributions | |
whispered | +0.302 |
needed | +0.290 |
said | +0.276 |
meant | +0.272 |
Jones | +0.260 |
Neg logit contributions | |
pieces | -0.389 |
n | -0.387 |
disl | -0.382 |
Majesty | -0.376 |
iu | -0.370 |
Token | duck |
Token ID | 22045 |
Feature activation | -2.42e+02 |
Pos logit contributions | |
needed | +0.366 |
whispered | +0.360 |
meant | +0.348 |
Jones | +0.334 |
said | +0.321 |
Neg logit contributions | |
n | -0.328 |
disl | -0.322 |
pieces | -0.317 |
interests | -0.306 |
Majesty | -0.306 |
Token | smiles |
Token ID | 21845 |
Feature activation | -2.53e+02 |
Pos logit contributions | |
needed | +0.323 |
meant | +0.312 |
whispered | +0.309 |
Jones | +0.309 |
White | +0.281 |
Neg logit contributions | |
n | -0.339 |
disl | -0.331 |
Majesty | -0.323 |
pieces | -0.318 |
interests | -0.317 |
Token | � |
Token ID | 129 |
Feature activation | +5.76e+02 |
Pos logit contributions | |
needed | +0.194 |
whispered | +0.187 |
meant | +0.179 |
Jones | +0.165 |
chy | +0.156 |
Neg logit contributions | |
n | -0.411 |
pieces | -0.407 |
disl | -0.405 |
interests | -0.396 |
iu | -0.395 |
Token | � |
Token ID | 241 |
Feature activation | +4.76e+02 |
Pos logit contributions | |
interests | +0.659 |
pieces | +0.652 |
victory | +0.601 |
occasions | +0.582 |
nels | +0.572 |
Neg logit contributions | |
whispered | -0.925 |
needed | -0.890 |
said | -0.883 |
meant | -0.881 |
Jones | -0.831 |
Token | It |
Token ID | 1026 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
Majesty | +0.849 |
disl | +0.848 |
pieces | +0.847 |
n | +0.838 |
arrest | +0.827 |
Neg logit contributions | |
needed | -0.454 |
whispered | -0.428 |
meant | -0.428 |
Jones | -0.380 |
said | -0.375 |
Token | â |
Token ID | 22940 |
Feature activation | +4.36e+02 |
Pos logit contributions | |
n | +0.552 |
disl | +0.533 |
pieces | +0.520 |
interests | +0.506 |
Majesty | +0.502 |
Neg logit contributions | |
whispered | -0.516 |
needed | -0.508 |
Jones | -0.488 |
meant | -0.476 |
faded | -0.465 |
Token | € |
Token ID | 26391 |
Feature activation | +4.43e+02 |
Pos logit contributions | |
n | +0.439 |
disl | +0.435 |
pieces | +0.427 |
Majesty | +0.408 |
interests | +0.407 |
Neg logit contributions | |
needed | -0.758 |
whispered | -0.747 |
meant | -0.728 |
Jones | -0.699 |
said | -0.682 |
Token | ™ |
Token ID | 8151 |
Feature activation | +5.06e+02 |
Pos logit contributions | |
pieces | +0.596 |
disl | +0.585 |
n | +0.573 |
iu | +0.567 |
victory | +0.566 |
Neg logit contributions | |
needed | -0.619 |
whispered | -0.599 |
meant | -0.572 |
Jones | -0.566 |
said | -0.550 |
Token | s |
Token ID | 82 |
Feature activation | +4.04e+02 |
Pos logit contributions | |
n | +0.881 |
ric | +0.828 |
disl | +0.783 |
pieces | +0.782 |
ieved | +0.765 |
Neg logit contributions | |
whispered | -0.549 |
Jones | -0.514 |
faded | -0.496 |
needed | -0.490 |
blank | -0.480 |
Token | a |
Token ID | 257 |
Feature activation | +3.81e+02 |
Pos logit contributions | |
n | +0.531 |
disl | +0.519 |
pieces | +0.505 |
interests | +0.491 |
Majesty | +0.489 |
Neg logit contributions | |
needed | -0.580 |
whispered | -0.577 |
meant | -0.551 |
Jones | -0.536 |
faded | -0.511 |
Token | magic |
Token ID | 5536 |
Feature activation | +3.82e+02 |
Pos logit contributions | |
n | +0.557 |
disl | +0.553 |
pieces | +0.544 |
victory | +0.534 |
Majesty | +0.534 |
Neg logit contributions | |
needed | -0.471 |
whispered | -0.471 |
meant | -0.456 |
Jones | -0.430 |
faded | -0.405 |
Token | book |
Token ID | 1492 |
Feature activation | +3.67e+02 |
Pos logit contributions | |
n | +0.523 |
disl | +0.506 |
Majesty | +0.497 |
pieces | +0.494 |
interests | +0.491 |
Neg logit contributions | |
needed | -0.509 |
whispered | -0.508 |
meant | -0.487 |
Jones | -0.468 |
bored | -0.450 |
Token | ! |
Token ID | 0 |
Feature activation | +3.00e+02 |
Pos logit contributions | |
n | +0.469 |
pieces | +0.386 |
victory | +0.381 |
occasions | +0.378 |
disl | +0.371 |
Neg logit contributions | |
Jones | -0.513 |
whispered | -0.512 |
needed | -0.498 |
Fuel | -0.494 |
bored | -0.490 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+02 |
Pos logit contributions | |
pieces | +0.954 |
disl | +0.915 |
nels | +0.911 |
latable | +0.901 |
interests | +0.896 |
Neg logit contributions | |
needed | -0.668 |
whispered | -0.652 |
said | -0.621 |
Jones | -0.595 |
meant | -0.584 |
Token | � |
Token ID | 241 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
pieces | +0.575 |
interests | +0.555 |
n | +0.549 |
disl | +0.542 |
victory | +0.525 |
Neg logit contributions | |
whispered | -0.880 |
needed | -0.878 |
meant | -0.849 |
Jones | -0.817 |
said | -0.811 |
Token | That |
Token ID | 2504 |
Feature activation | +3.82e+02 |
Pos logit contributions | |
n | +0.785 |
disl | +0.783 |
pieces | +0.780 |
Majesty | +0.764 |
iu | +0.756 |
Neg logit contributions | |
needed | -0.485 |
whispered | -0.466 |
meant | -0.453 |
Jones | -0.417 |
said | -0.404 |
Token | â |
Token ID | 22940 |
Feature activation | +4.36e+02 |
Pos logit contributions | |
n | +0.557 |
disl | +0.550 |
pieces | +0.544 |
iu | +0.526 |
Majesty | +0.525 |
Neg logit contributions | |
needed | -0.495 |
whispered | -0.484 |
meant | -0.469 |
Jones | -0.441 |
said | -0.428 |
Token | € |
Token ID | 26391 |
Feature activation | +4.43e+02 |
Pos logit contributions | |
n | +0.391 |
disl | +0.379 |
pieces | +0.368 |
interests | +0.347 |
iu | +0.346 |
Neg logit contributions | |
needed | -0.813 |
whispered | -0.804 |
meant | -0.780 |
Jones | -0.757 |
said | -0.735 |
Token | ™ |
Token ID | 8151 |
Feature activation | +5.05e+02 |
Pos logit contributions | |
pieces | +0.585 |
disl | +0.579 |
n | +0.577 |
interests | +0.562 |
iu | +0.560 |
Neg logit contributions | |
needed | -0.624 |
whispered | -0.603 |
meant | -0.582 |
Jones | -0.568 |
said | -0.550 |
Token | s |
Token ID | 82 |
Feature activation | +4.03e+02 |
Pos logit contributions | |
n | +0.826 |
ric | +0.768 |
disl | +0.761 |
pieces | +0.741 |
ieved | +0.711 |
Neg logit contributions | |
whispered | -0.614 |
Jones | -0.564 |
faded | -0.543 |
needed | -0.537 |
White | -0.532 |
Token | right |
Token ID | 826 |
Feature activation | +3.82e+02 |
Pos logit contributions | |
pieces | +0.536 |
disl | +0.527 |
n | +0.525 |
iu | +0.523 |
arrest | +0.519 |
Neg logit contributions | |
needed | -0.579 |
meant | -0.554 |
whispered | -0.550 |
said | -0.506 |
Jones | -0.501 |
Token | , |
Token ID | 11 |
Feature activation | +3.32e+02 |
Pos logit contributions | |
n | +0.558 |
disl | +0.546 |
Majesty | +0.519 |
interests | +0.516 |
pieces | +0.514 |
Neg logit contributions | |
needed | -0.490 |
whispered | -0.486 |
meant | -0.454 |
Jones | -0.438 |
faded | -0.426 |
Token | Emma |
Token ID | 18966 |
Feature activation | +3.64e+02 |
Pos logit contributions | |
n | +0.436 |
disl | +0.431 |
pieces | +0.426 |
iu | +0.411 |
interests | +0.411 |
Neg logit contributions | |
needed | -0.475 |
whispered | -0.466 |
meant | -0.452 |
Jones | -0.428 |
said | -0.419 |
Token | . |
Token ID | 13 |
Feature activation | +2.98e+02 |
Pos logit contributions | |
n | +0.505 |
disl | +0.472 |
Majesty | +0.464 |
spend | +0.436 |
interests | +0.434 |
Neg logit contributions | |
whispered | -0.448 |
needed | -0.441 |
faded | -0.440 |
dark | -0.436 |
Fuel | -0.429 |
Token | his |
Token ID | 465 |
Feature activation | +4.53e+02 |
Pos logit contributions | |
needed | +0.337 |
whispered | +0.337 |
meant | +0.324 |
Jones | +0.310 |
said | +0.303 |
Neg logit contributions | |
disl | -0.266 |
n | -0.261 |
pieces | -0.259 |
iu | -0.253 |
latable | -0.248 |
Token | head |
Token ID | 1182 |
Feature activation | +3.98e+02 |
Pos logit contributions | |
n | +0.618 |
disl | +0.610 |
pieces | +0.601 |
iu | +0.577 |
interests | +0.576 |
Neg logit contributions | |
needed | -0.629 |
whispered | -0.617 |
meant | -0.594 |
Jones | -0.567 |
said | -0.550 |
Token | out |
Token ID | 503 |
Feature activation | +3.62e+02 |
Pos logit contributions | |
n | +0.552 |
spend | +0.491 |
interests | +0.491 |
pause | +0.484 |
pieces | +0.483 |
Neg logit contributions | |
Ancient | -0.477 |
Jones | -0.472 |
whispered | -0.472 |
Fuel | -0.468 |
needed | -0.467 |
Token | of |
Token ID | 286 |
Feature activation | +3.48e+02 |
Pos logit contributions | |
n | +0.456 |
pieces | +0.450 |
victory | +0.450 |
interests | +0.444 |
arrest | +0.441 |
Neg logit contributions | |
needed | -0.481 |
whispered | -0.480 |
meant | -0.468 |
Jones | -0.461 |
White | -0.448 |
Token | his |
Token ID | 465 |
Feature activation | +3.63e+02 |
Pos logit contributions | |
n | +0.513 |
disl | +0.507 |
pieces | +0.503 |
Majesty | +0.484 |
iu | +0.483 |
Neg logit contributions | |
needed | -0.448 |
whispered | -0.441 |
meant | -0.423 |
Jones | -0.402 |
said | -0.392 |
Token | bur |
Token ID | 4356 |
Feature activation | +4.65e+02 |
Pos logit contributions | |
n | +0.492 |
disl | +0.485 |
pieces | +0.480 |
interests | +0.465 |
Majesty | +0.464 |
Neg logit contributions | |
needed | -0.501 |
whispered | -0.493 |
meant | -0.478 |
Jones | -0.454 |
said | -0.436 |
Token | row |
Token ID | 808 |
Feature activation | +3.72e+02 |
Pos logit contributions | |
disl | +0.958 |
blast | +0.920 |
pieces | +0.917 |
victory | +0.897 |
hey | +0.888 |
Neg logit contributions | |
needed | -0.421 |
whispered | -0.408 |
meant | -0.367 |
Jones | -0.335 |
chy | -0.328 |
Token | and |
Token ID | 290 |
Feature activation | +3.09e+02 |
Pos logit contributions | |
n | +0.582 |
interests | +0.526 |
pieces | +0.524 |
Majesty | +0.523 |
disl | +0.517 |
Neg logit contributions | |
whispered | -0.421 |
needed | -0.409 |
meant | -0.400 |
Jones | -0.374 |
White | -0.370 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.06e+02 |
Pos logit contributions | |
n | +0.520 |
disl | +0.511 |
pieces | +0.499 |
Majesty | +0.495 |
interests | +0.492 |
Neg logit contributions | |
needed | -0.311 |
whispered | -0.302 |
meant | -0.298 |
Jones | -0.283 |
White | -0.263 |
Token | something |
Token ID | 1223 |
Feature activation | +3.03e+02 |
Pos logit contributions | |
n | +0.373 |
disl | +0.363 |
pieces | +0.353 |
Majesty | +0.342 |
victory | +0.341 |
Neg logit contributions | |
needed | -0.460 |
whispered | -0.454 |
meant | -0.435 |
Jones | -0.430 |
faded | -0.412 |
Token | new |
Token ID | 649 |
Feature activation | +3.06e+02 |
Pos logit contributions | |
n | +0.378 |
victory | +0.366 |
disl | +0.338 |
pieces | +0.336 |
exercise | +0.331 |
Neg logit contributions | |
Jones | -0.453 |
whispered | -0.416 |
White | -0.415 |
needed | -0.394 |
Factory | -0.390 |
Token | of |
Token ID | 286 |
Feature activation | +4.07e+02 |
Pos logit contributions | |
needed | +0.356 |
whispered | +0.354 |
meant | +0.338 |
chy | +0.334 |
faded | +0.326 |
Neg logit contributions | |
n | -0.273 |
victory | -0.270 |
arrest | -0.269 |
disl | -0.261 |
pause | -0.260 |
Token | other |
Token ID | 584 |
Feature activation | +3.99e+02 |
Pos logit contributions | |
n | +0.582 |
disl | +0.575 |
pieces | +0.570 |
iu | +0.557 |
nels | +0.548 |
Neg logit contributions | |
needed | -0.546 |
whispered | -0.525 |
meant | -0.510 |
said | -0.474 |
Jones | -0.474 |
Token | people |
Token ID | 661 |
Feature activation | +3.48e+02 |
Pos logit contributions | |
pieces | +0.467 |
n | +0.466 |
disl | +0.456 |
interests | +0.452 |
arrest | +0.452 |
Neg logit contributions | |
whispered | -0.606 |
needed | -0.591 |
meant | -0.578 |
Jones | -0.563 |
faded | -0.543 |
Token | â |
Token ID | 22940 |
Feature activation | +3.98e+02 |
Pos logit contributions | |
n | +0.539 |
pieces | +0.521 |
disl | +0.520 |
interests | +0.507 |
arrest | +0.497 |
Neg logit contributions | |
whispered | -0.391 |
needed | -0.381 |
meant | -0.377 |
Jones | -0.376 |
White | -0.360 |
Token | € |
Token ID | 26391 |
Feature activation | +4.08e+02 |
Pos logit contributions | |
n | +0.465 |
disl | +0.438 |
pieces | +0.424 |
iu | +0.409 |
interests | +0.404 |
Neg logit contributions | |
needed | -0.649 |
whispered | -0.642 |
meant | -0.615 |
Jones | -0.597 |
faded | -0.587 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.70e+02 |
Pos logit contributions | |
disl | +0.652 |
n | +0.650 |
pieces | +0.649 |
interests | +0.627 |
iu | +0.626 |
Neg logit contributions | |
needed | -0.463 |
whispered | -0.450 |
meant | -0.425 |
Jones | -0.414 |
said | -0.397 |
Token | s |
Token ID | 82 |
Feature activation | +3.83e+02 |
Pos logit contributions | |
n | +0.802 |
ric | +0.735 |
pieces | +0.722 |
disl | +0.722 |
ieved | +0.697 |
Neg logit contributions | |
whispered | -0.511 |
needed | -0.476 |
Jones | -0.472 |
meant | -0.465 |
faded | -0.463 |
Token | property |
Token ID | 3119 |
Feature activation | +4.12e+02 |
Pos logit contributions | |
n | +0.544 |
disl | +0.533 |
pieces | +0.516 |
Majesty | +0.515 |
interests | +0.509 |
Neg logit contributions | |
whispered | -0.501 |
needed | -0.493 |
meant | -0.474 |
Jones | -0.457 |
faded | -0.439 |
Token | . |
Token ID | 13 |
Feature activation | +3.09e+02 |
Pos logit contributions | |
n | +0.632 |
disl | +0.599 |
pieces | +0.581 |
interests | +0.574 |
victory | +0.565 |
Neg logit contributions | |
whispered | -0.490 |
needed | -0.474 |
Jones | -0.462 |
faded | -0.432 |
White | -0.428 |
Token | We |
Token ID | 775 |
Feature activation | +3.09e+02 |
Pos logit contributions | |
n | +0.485 |
disl | +0.466 |
pieces | +0.449 |
Majesty | +0.439 |
ric | +0.437 |
Neg logit contributions | |
needed | -0.380 |
whispered | -0.375 |
meant | -0.358 |
faded | -0.339 |
chy | -0.337 |
Token | should |
Token ID | 815 |
Feature activation | +2.92e+02 |
Pos logit contributions | |
n | +0.385 |
disl | +0.382 |
pieces | +0.375 |
iu | +0.362 |
Majesty | +0.362 |
Neg logit contributions | |
needed | -0.467 |
whispered | -0.457 |
meant | -0.444 |
Jones | -0.421 |
said | -0.412 |
Token | � |
Token ID | 241 |
Feature activation | +4.22e+02 |
Pos logit contributions | |
n | +0.495 |
disl | +0.481 |
pieces | +0.454 |
Majesty | +0.442 |
iu | +0.438 |
Neg logit contributions | |
needed | -0.790 |
whispered | -0.771 |
meant | -0.753 |
Jones | -0.723 |
faded | -0.704 |
Token | I |
Token ID | 40 |
Feature activation | +3.74e+02 |
Pos logit contributions | |
pieces | +0.731 |
disl | +0.728 |
n | +0.721 |
Majesty | +0.721 |
arrest | +0.710 |
Neg logit contributions | |
needed | -0.423 |
whispered | -0.400 |
meant | -0.394 |
Jones | -0.357 |
said | -0.350 |
Token | don |
Token ID | 836 |
Feature activation | +3.72e+02 |
Pos logit contributions | |
disl | +0.442 |
n | +0.441 |
pieces | +0.440 |
iu | +0.424 |
Majesty | +0.420 |
Neg logit contributions | |
needed | -0.588 |
whispered | -0.578 |
meant | -0.562 |
Jones | -0.528 |
said | -0.525 |
Token | â |
Token ID | 22940 |
Feature activation | +4.05e+02 |
Pos logit contributions | |
Majesty | +0.484 |
disl | +0.479 |
interests | +0.465 |
pieces | +0.464 |
iu | +0.454 |
Neg logit contributions | |
needed | -0.554 |
meant | -0.541 |
whispered | -0.537 |
said | -0.499 |
Jones | -0.492 |
Token | € |
Token ID | 26391 |
Feature activation | +4.10e+02 |
Pos logit contributions | |
n | +0.336 |
disl | +0.320 |
pieces | +0.310 |
iu | +0.290 |
interests | +0.289 |
Neg logit contributions | |
needed | -0.790 |
whispered | -0.778 |
meant | -0.757 |
Jones | -0.736 |
faded | -0.717 |
Token | ™ |
Token ID | 8151 |
Feature activation | +4.71e+02 |
Pos logit contributions | |
pieces | +0.507 |
disl | +0.505 |
n | +0.487 |
iu | +0.485 |
interests | +0.484 |
Neg logit contributions | |
needed | -0.616 |
whispered | -0.601 |
Jones | -0.576 |
meant | -0.574 |
said | -0.559 |
Token | t |
Token ID | 83 |
Feature activation | +3.73e+02 |
Pos logit contributions | |
n | +0.749 |
ric | +0.676 |
pieces | +0.674 |
disl | +0.657 |
ieved | +0.654 |
Neg logit contributions | |
whispered | -0.578 |
needed | -0.529 |
faded | -0.517 |
Jones | -0.513 |
White | -0.497 |
Token | think |
Token ID | 892 |
Feature activation | +3.55e+02 |
Pos logit contributions | |
disl | +0.416 |
pieces | +0.414 |
n | +0.414 |
Majesty | +0.402 |
iu | +0.400 |
Neg logit contributions | |
needed | -0.607 |
whispered | -0.593 |
meant | -0.585 |
Jones | -0.546 |
said | -0.543 |
Token | so |
Token ID | 523 |
Feature activation | +3.32e+02 |
Pos logit contributions | |
n | +0.445 |
disl | +0.428 |
pieces | +0.424 |
interests | +0.406 |
Majesty | +0.401 |
Neg logit contributions | |
needed | -0.536 |
whispered | -0.528 |
meant | -0.507 |
Jones | -0.493 |
faded | -0.473 |
Token | . |
Token ID | 13 |
Feature activation | +2.94e+02 |
Pos logit contributions | |
pieces | +0.429 |
disl | +0.428 |
n | +0.426 |
iu | +0.421 |
interests | +0.410 |
Neg logit contributions | |
needed | -0.487 |
whispered | -0.468 |
meant | -0.460 |
said | -0.432 |
Jones | -0.427 |
Token | It |
Token ID | 632 |
Feature activation | +3.08e+02 |
Pos logit contributions | |
n | +0.428 |
disl | +0.418 |
pieces | +0.414 |
Majesty | +0.400 |
interests | +0.399 |
Neg logit contributions | |
needed | -0.382 |
whispered | -0.375 |
meant | -0.360 |
Jones | -0.337 |
faded | -0.331 |
Token | put |
Token ID | 1234 |
Feature activation | +2.11e+02 |
Pos logit contributions | |
latable | +0.331 |
iu | +0.327 |
ric | +0.323 |
ext | +0.319 |
pieces | +0.309 |
Neg logit contributions | |
said | -0.326 |
needed | -0.311 |
whispered | -0.305 |
meant | -0.272 |
sleepy | -0.260 |
Token | some |
Token ID | 617 |
Feature activation | +2.14e+02 |
Pos logit contributions | |
disl | +0.235 |
n | +0.225 |
pieces | +0.222 |
Majesty | +0.218 |
interests | +0.217 |
Neg logit contributions | |
needed | -0.349 |
whispered | -0.345 |
meant | -0.335 |
Jones | -0.322 |
said | -0.321 |
Token | band |
Token ID | 4097 |
Feature activation | +2.27e+02 |
Pos logit contributions | |
disl | +0.294 |
n | +0.293 |
pieces | +0.286 |
Majesty | +0.282 |
iu | +0.279 |
Neg logit contributions | |
needed | -0.298 |
whispered | -0.291 |
meant | -0.281 |
said | -0.264 |
Jones | -0.258 |
Token | - |
Token ID | 12 |
Feature activation | +2.68e+02 |
Pos logit contributions | |
latable | +0.380 |
hey | +0.370 |
Majesty | +0.369 |
pieces | +0.364 |
victory | +0.363 |
Neg logit contributions | |
needed | -0.251 |
whispered | -0.244 |
meant | -0.227 |
said | -0.223 |
chy | -0.197 |
Token | aid |
Token ID | 1698 |
Feature activation | +2.39e+02 |
Pos logit contributions | |
latable | +0.392 |
Majesty | +0.392 |
blast | +0.374 |
disl | +0.372 |
nels | +0.364 |
Neg logit contributions | |
needed | -0.354 |
whispered | -0.340 |
meant | -0.327 |
said | -0.317 |
spoke | -0.305 |
Token | s |
Token ID | 82 |
Feature activation | +2.19e+02 |
Pos logit contributions | |
n | +0.443 |
pieces | +0.422 |
disl | +0.417 |
interests | +0.414 |
iu | +0.406 |
Neg logit contributions | |
needed | -0.214 |
whispered | -0.210 |
meant | -0.194 |
Jones | -0.191 |
faded | -0.181 |
Token | on |
Token ID | 319 |
Feature activation | +2.09e+02 |
Pos logit contributions | |
n | +0.324 |
interests | +0.299 |
disl | +0.293 |
arrest | +0.286 |
victory | +0.285 |
Neg logit contributions | |
whispered | -0.275 |
needed | -0.272 |
Jones | -0.267 |
meant | -0.266 |
spoke | -0.253 |
Token | them |
Token ID | 606 |
Feature activation | +2.23e+02 |
Pos logit contributions | |
n | +0.188 |
arrest | +0.172 |
pieces | +0.169 |
disl | +0.169 |
our | +0.163 |
Neg logit contributions | |
needed | -0.400 |
whispered | -0.392 |
meant | -0.380 |
spoke | -0.367 |
faded | -0.362 |
Token | . |
Token ID | 13 |
Feature activation | +1.82e+02 |
Pos logit contributions | |
n | +0.278 |
our | +0.234 |
interests | +0.234 |
pieces | +0.224 |
stay | +0.216 |
Neg logit contributions | |
needed | -0.344 |
spoke | -0.334 |
Jones | -0.333 |
faded | -0.333 |
whispered | -0.328 |
Token | Then |
Token ID | 3244 |
Feature activation | +1.94e+02 |
Pos logit contributions | |
n | +0.298 |
pieces | +0.296 |
disl | +0.293 |
iu | +0.288 |
arrest | +0.287 |
Neg logit contributions | |
whispered | -0.195 |
needed | -0.193 |
Jones | -0.188 |
meant | -0.186 |
White | -0.179 |
Token | you |
Token ID | 345 |
Feature activation | +1.99e+02 |
Pos logit contributions | |
disl | +0.267 |
pieces | +0.264 |
latable | +0.259 |
ric | +0.256 |
iu | +0.253 |
Neg logit contributions | |
needed | -0.275 |
whispered | -0.271 |
meant | -0.258 |
said | -0.257 |
Jones | -0.250 |
Token | ang |
Token ID | 3550 |
Feature activation | +3.31e+02 |
Pos logit contributions | |
n | +0.502 |
disl | +0.501 |
pieces | +0.496 |
interests | +0.477 |
Majesty | +0.474 |
Neg logit contributions | |
needed | -0.422 |
whispered | -0.412 |
meant | -0.399 |
Jones | -0.397 |
White | -0.367 |
Token | rier |
Token ID | 5277 |
Feature activation | +4.02e+02 |
Pos logit contributions | |
disl | +0.442 |
n | +0.434 |
pieces | +0.428 |
interests | +0.422 |
Majesty | +0.410 |
Neg logit contributions | |
needed | -0.480 |
whispered | -0.467 |
meant | -0.460 |
Jones | -0.440 |
said | -0.426 |
Token | and |
Token ID | 290 |
Feature activation | +2.83e+02 |
Pos logit contributions | |
n | +0.632 |
disl | +0.608 |
interests | +0.580 |
iu | +0.576 |
hey | +0.571 |
Neg logit contributions | |
whispered | -0.444 |
needed | -0.441 |
White | -0.409 |
Jones | -0.399 |
dark | -0.398 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.07e+02 |
Pos logit contributions | |
n | +0.531 |
disl | +0.525 |
pieces | +0.511 |
hey | +0.501 |
iu | +0.497 |
Neg logit contributions | |
needed | -0.235 |
whispered | -0.228 |
Jones | -0.215 |
meant | -0.215 |
White | -0.189 |
Token | felt |
Token ID | 2936 |
Feature activation | +2.82e+02 |
Pos logit contributions | |
n | +0.551 |
hey | +0.472 |
pieces | +0.471 |
spend | +0.468 |
ieved | +0.467 |
Neg logit contributions | |
Jones | -0.289 |
dark | -0.285 |
Ancient | -0.282 |
blank | -0.278 |
aisle | -0.277 |
Token | even |
Token ID | 772 |
Feature activation | +2.92e+02 |
Pos logit contributions | |
n | +0.521 |
disl | +0.516 |
pieces | +0.509 |
interests | +0.495 |
Majesty | +0.495 |
Neg logit contributions | |
needed | -0.253 |
whispered | -0.247 |
meant | -0.233 |
Jones | -0.219 |
said | -0.203 |
Token | sad |
Token ID | 6507 |
Feature activation | +2.84e+02 |
Pos logit contributions | |
n | +0.547 |
disl | +0.538 |
arrest | +0.535 |
pieces | +0.532 |
Majesty | +0.529 |
Neg logit contributions | |
needed | -0.278 |
whispered | -0.272 |
meant | -0.256 |
said | -0.238 |
spoke | -0.218 |
Token | der |
Token ID | 1082 |
Feature activation | +3.91e+02 |
Pos logit contributions | |
disl | +0.439 |
n | +0.438 |
interests | +0.418 |
pieces | +0.416 |
victory | +0.411 |
Neg logit contributions | |
whispered | -0.328 |
needed | -0.323 |
Jones | -0.301 |
meant | -0.298 |
spoke | -0.287 |
Token | . |
Token ID | 13 |
Feature activation | +2.48e+02 |
Pos logit contributions | |
disl | +0.606 |
n | +0.590 |
interests | +0.550 |
pieces | +0.547 |
arrest | +0.525 |
Neg logit contributions | |
Jones | -0.451 |
whispered | -0.441 |
White | -0.428 |
needed | -0.422 |
aisle | -0.398 |
Token | From |
Token ID | 3574 |
Feature activation | +2.51e+02 |
Pos logit contributions | |
disl | +0.370 |
n | +0.369 |
Majesty | +0.325 |
ric | +0.323 |
interests | +0.317 |
Neg logit contributions | |
faded | -0.317 |
whispered | -0.315 |
meant | -0.307 |
needed | -0.305 |
shone | -0.297 |
Token | that |
Token ID | 326 |
Feature activation | +2.51e+02 |
Pos logit contributions | |
n | +0.251 |
disl | +0.246 |
pieces | +0.228 |
nels | +0.222 |
iu | +0.221 |
Neg logit contributions | |
needed | -0.439 |
whispered | -0.427 |
meant | -0.420 |
Jones | -0.413 |
faded | -0.397 |
Token | . |
Token ID | 13 |
Feature activation | +3.01e+02 |
Pos logit contributions | |
n | +0.587 |
disl | +0.586 |
pieces | +0.571 |
iu | +0.558 |
interests | +0.553 |
Neg logit contributions | |
needed | -0.426 |
whispered | -0.420 |
meant | -0.392 |
Jones | -0.378 |
said | -0.369 |
Token | It |
Token ID | 632 |
Feature activation | +3.14e+02 |
Pos logit contributions | |
n | +0.484 |
disl | +0.466 |
pieces | +0.446 |
Majesty | +0.443 |
interests | +0.438 |
Neg logit contributions | |
needed | -0.344 |
meant | -0.333 |
whispered | -0.328 |
chy | -0.326 |
faded | -0.314 |
Token | was |
Token ID | 373 |
Feature activation | +2.91e+02 |
Pos logit contributions | |
n | +0.519 |
disl | +0.495 |
blast | +0.471 |
pieces | +0.470 |
interests | +0.465 |
Neg logit contributions | |
Jones | -0.337 |
needed | -0.315 |
whispered | -0.311 |
meant | -0.290 |
White | -0.287 |
Token | her |
Token ID | 607 |
Feature activation | +3.00e+02 |
Pos logit contributions | |
n | +0.357 |
victory | +0.350 |
blast | +0.336 |
disl | +0.332 |
our | +0.330 |
Neg logit contributions | |
needed | -0.415 |
whispered | -0.407 |
Jones | -0.405 |
Fuel | -0.378 |
aisle | -0.377 |
Token | friend |
Token ID | 1545 |
Feature activation | +2.96e+02 |
Pos logit contributions | |
n | +0.353 |
pieces | +0.343 |
Majesty | +0.340 |
disl | +0.339 |
victory | +0.338 |
Neg logit contributions | |
needed | -0.462 |
whispered | -0.457 |
meant | -0.447 |
Jones | -0.424 |
spoke | -0.416 |
Token | , |
Token ID | 11 |
Feature activation | +2.64e+02 |
Pos logit contributions | |
n | +0.348 |
pieces | +0.329 |
spend | +0.306 |
our | +0.301 |
join | +0.301 |
Neg logit contributions | |
Fuel | -0.431 |
needed | -0.421 |
Gas | -0.415 |
Ancient | -0.415 |
Dead | -0.413 |
Token | Tommy |
Token ID | 19919 |
Feature activation | +2.93e+02 |
Pos logit contributions | |
n | +0.328 |
pieces | +0.309 |
victory | +0.302 |
interests | +0.301 |
our | +0.299 |
Neg logit contributions | |
needed | -0.375 |
whispered | -0.353 |
meant | -0.343 |
faded | -0.336 |
chy | -0.330 |
Token | . |
Token ID | 13 |
Feature activation | +2.43e+02 |
Pos logit contributions | |
n | +0.387 |
disl | +0.335 |
our | +0.330 |
spend | +0.321 |
arrest | +0.316 |
Neg logit contributions | |
faded | -0.402 |
Ancient | -0.390 |
Dead | -0.384 |
dark | -0.382 |
needed | -0.382 |
Token | " |
Token ID | 366 |
Feature activation | +2.52e+02 |
Pos logit contributions | |
n | +0.390 |
disl | +0.389 |
interests | +0.366 |
pieces | +0.366 |
Majesty | +0.362 |
Neg logit contributions | |
needed | -0.279 |
whispered | -0.267 |
meant | -0.254 |
chy | -0.245 |
spoke | -0.234 |
Token | Hi |
Token ID | 17250 |
Feature activation | +2.58e+02 |
Pos logit contributions | |
n | +0.443 |
disl | +0.440 |
pieces | +0.438 |
Majesty | +0.429 |
iu | +0.424 |
Neg logit contributions | |
needed | -0.246 |
whispered | -0.243 |
meant | -0.231 |
Jones | -0.215 |
said | -0.204 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.63e+02 |
Pos logit contributions | |
pieces | +0.448 |
disl | +0.447 |
ric | +0.437 |
n | +0.436 |
iu | +0.432 |
Neg logit contributions | |
needed | -0.268 |
whispered | -0.266 |
meant | -0.259 |
said | -0.248 |
Jones | -0.246 |
Token | in |
Token ID | 287 |
Feature activation | +3.53e+02 |
Pos logit contributions | |
n | +0.675 |
arrest | +0.642 |
disl | +0.638 |
pieces | +0.638 |
interests | +0.629 |
Neg logit contributions | |
whispered | -0.404 |
needed | -0.398 |
Jones | -0.395 |
meant | -0.388 |
aisle | -0.378 |
Token | the |
Token ID | 262 |
Feature activation | +3.70e+02 |
Pos logit contributions | |
n | +0.522 |
disl | +0.515 |
pieces | +0.509 |
Majesty | +0.491 |
interests | +0.491 |
Neg logit contributions | |
needed | -0.451 |
whispered | -0.442 |
meant | -0.426 |
Jones | -0.404 |
said | -0.389 |
Token | circle |
Token ID | 9197 |
Feature activation | +3.54e+02 |
Pos logit contributions | |
n | +0.559 |
disl | +0.554 |
pieces | +0.550 |
Majesty | +0.538 |
interests | +0.531 |
Neg logit contributions | |
needed | -0.449 |
whispered | -0.446 |
meant | -0.431 |
Jones | -0.402 |
said | -0.386 |
Token | . |
Token ID | 13 |
Feature activation | +2.70e+02 |
Pos logit contributions | |
n | +0.543 |
pieces | +0.525 |
interests | +0.518 |
disl | +0.501 |
Majesty | +0.495 |
Neg logit contributions | |
needed | -0.404 |
meant | -0.397 |
whispered | -0.396 |
Jones | -0.394 |
White | -0.380 |
Token |
|
Token ID | 198 |
Feature activation | +2.74e+02 |
Pos logit contributions | |
n | +0.423 |
disl | +0.417 |
pieces | +0.410 |
Majesty | +0.392 |
interests | +0.381 |
Neg logit contributions | |
needed | -0.312 |
meant | -0.302 |
whispered | -0.301 |
chy | -0.291 |
faded | -0.291 |
Token |
|
Token ID | 198 |
Feature activation | +2.71e+02 |
Pos logit contributions | |
n | +0.479 |
disl | +0.463 |
pieces | +0.458 |
interests | +0.449 |
Majesty | +0.439 |
Neg logit contributions | |
needed | -0.283 |
whispered | -0.273 |
meant | -0.266 |
Jones | -0.246 |
faded | -0.239 |
Token | The |
Token ID | 464 |
Feature activation | +2.87e+02 |
Pos logit contributions | |
n | +0.393 |
disl | +0.386 |
pieces | +0.380 |
interests | +0.367 |
Majesty | +0.365 |
Neg logit contributions | |
needed | -0.359 |
whispered | -0.350 |
meant | -0.341 |
Jones | -0.323 |
faded | -0.309 |
Token | mouse |
Token ID | 10211 |
Feature activation | +3.02e+02 |
Pos logit contributions | |
Majesty | +0.443 |
victory | +0.443 |
pieces | +0.436 |
n | +0.426 |
disl | +0.423 |
Neg logit contributions | |
whispered | -0.334 |
meant | -0.329 |
needed | -0.321 |
spoke | -0.295 |
Jones | -0.290 |
Token | stopped |
Token ID | 5025 |
Feature activation | +2.69e+02 |
Pos logit contributions | |
n | +0.482 |
Majesty | +0.441 |
disl | +0.436 |
blast | +0.434 |
ered | +0.433 |
Neg logit contributions | |
Jones | -0.341 |
Fuel | -0.310 |
meant | -0.301 |
aisle | -0.301 |
White | -0.301 |
Token | and |
Token ID | 290 |
Feature activation | +2.49e+02 |
Pos logit contributions | |
disl | +0.372 |
ric | +0.362 |
n | +0.360 |
victory | +0.349 |
arrest | +0.347 |
Neg logit contributions | |
needed | -0.364 |
whispered | -0.344 |
meant | -0.338 |
aisle | -0.336 |
faded | -0.313 |
Token | the |
Token ID | 262 |
Feature activation | +2.64e+02 |
Pos logit contributions | |
n | +0.400 |
disl | +0.393 |
hey | +0.382 |
arrest | +0.377 |
spend | +0.374 |
Neg logit contributions | |
Jones | -0.241 |
meant | -0.236 |
aisle | -0.229 |
needed | -0.226 |
elly | -0.225 |
Token | ummy |
Token ID | 13513 |
Feature activation | +2.85e+02 |
Pos logit contributions | |
n | +0.475 |
pieces | +0.471 |
disl | +0.471 |
Majesty | +0.456 |
interests | +0.451 |
Neg logit contributions | |
needed | -0.319 |
whispered | -0.303 |
meant | -0.298 |
Jones | -0.285 |
said | -0.274 |
Token | food |
Token ID | 2057 |
Feature activation | +2.75e+02 |
Pos logit contributions | |
n | +0.356 |
disl | +0.352 |
pieces | +0.346 |
Majesty | +0.335 |
interests | +0.332 |
Neg logit contributions | |
needed | -0.429 |
whispered | -0.421 |
meant | -0.409 |
Jones | -0.389 |
said | -0.381 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e+02 |
Pos logit contributions | |
n | +0.347 |
occasions | +0.307 |
exercise | +0.307 |
victory | +0.306 |
our | +0.302 |
Neg logit contributions | |
needed | -0.393 |
whispered | -0.389 |
faded | -0.386 |
spoke | -0.370 |
White | -0.364 |
Token | One |
Token ID | 1881 |
Feature activation | +2.33e+02 |
Pos logit contributions | |
n | +0.347 |
disl | +0.337 |
pieces | +0.335 |
interests | +0.325 |
Majesty | +0.323 |
Neg logit contributions | |
needed | -0.286 |
whispered | -0.283 |
meant | -0.273 |
faded | -0.253 |
Jones | -0.251 |
Token | day |
Token ID | 1110 |
Feature activation | +2.41e+02 |
Pos logit contributions | |
n | +0.284 |
exercise | +0.259 |
earned | +0.258 |
pause | +0.247 |
spend | +0.244 |
Neg logit contributions | |
whispered | -0.360 |
Jones | -0.349 |
faded | -0.326 |
spoke | -0.322 |
White | -0.321 |
Token | , |
Token ID | 11 |
Feature activation | +2.23e+02 |
Pos logit contributions | |
n | +0.359 |
disl | +0.342 |
pieces | +0.337 |
arrest | +0.330 |
victory | +0.328 |
Neg logit contributions | |
needed | -0.305 |
whispered | -0.302 |
meant | -0.294 |
Jones | -0.275 |
faded | -0.265 |
Token | her |
Token ID | 607 |
Feature activation | +2.40e+02 |
Pos logit contributions | |
disl | +0.460 |
n | +0.457 |
pieces | +0.453 |
Majesty | +0.449 |
interests | +0.446 |
Neg logit contributions | |
needed | -0.161 |
whispered | -0.155 |
meant | -0.141 |
said | -0.128 |
Jones | -0.128 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.39e+02 |
Pos logit contributions | |
n | +0.230 |
Majesty | +0.229 |
interests | +0.227 |
victory | +0.221 |
arrest | +0.217 |
Neg logit contributions | |
whispered | -0.400 |
needed | -0.385 |
meant | -0.383 |
spoke | -0.373 |
faded | -0.368 |
Token | made |
Token ID | 925 |
Feature activation | +2.18e+02 |
Pos logit contributions | |
n | +0.381 |
spend | +0.346 |
pause | +0.317 |
interests | +0.314 |
hey | +0.307 |
Neg logit contributions | |
faded | -0.246 |
Ancient | -0.245 |
blank | -0.244 |
Jones | -0.240 |
crooked | -0.240 |
Token | some |
Token ID | 617 |
Feature activation | +2.22e+02 |
Pos logit contributions | |
n | +0.290 |
pieces | +0.264 |
disl | +0.261 |
iu | +0.248 |
victory | +0.248 |
Neg logit contributions | |
needed | -0.326 |
whispered | -0.318 |
faded | -0.300 |
Jones | -0.298 |
meant | -0.298 |
Token | toast |
Token ID | 27805 |
Feature activation | +2.43e+02 |
Pos logit contributions | |
victory | +0.238 |
exercise | +0.236 |
pieces | +0.236 |
n | +0.227 |
disl | +0.221 |
Neg logit contributions | |
whispered | -0.349 |
Jones | -0.346 |
needed | -0.343 |
faded | -0.338 |
spoke | -0.327 |
Token | end |
Token ID | 886 |
Feature activation | +1.55e+02 |
Pos logit contributions | |
n | +0.176 |
disl | +0.170 |
pieces | +0.168 |
iu | +0.163 |
Majesty | +0.163 |
Neg logit contributions | |
whispered | -0.243 |
needed | -0.241 |
meant | -0.237 |
Jones | -0.229 |
faded | -0.215 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e+02 |
Pos logit contributions | |
n | +0.306 |
disl | +0.304 |
pieces | +0.303 |
Majesty | +0.300 |
arrest | +0.296 |
Neg logit contributions | |
needed | -0.124 |
whispered | -0.119 |
meant | -0.115 |
Jones | -0.102 |
said | -0.099 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.29e+02 |
Pos logit contributions | |
n | +0.200 |
disl | +0.191 |
pieces | +0.186 |
blast | +0.184 |
interests | +0.183 |
Neg logit contributions | |
whispered | -0.153 |
needed | -0.151 |
meant | -0.143 |
chy | -0.135 |
spoke | -0.134 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.27e+02 |
Pos logit contributions | |
n | +0.185 |
blast | +0.165 |
arrest | +0.153 |
victory | +0.152 |
ric | +0.149 |
Neg logit contributions | |
whispered | -0.179 |
needed | -0.174 |
faded | -0.164 |
meant | -0.163 |
chy | -0.163 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.56e+02 |
Pos logit contributions | |
n | +0.193 |
disl | +0.184 |
pieces | +0.184 |
interests | +0.183 |
victory | +0.178 |
Neg logit contributions | |
whispered | -0.151 |
needed | -0.147 |
meant | -0.144 |
Jones | -0.137 |
faded | -0.131 |
Token | a |
Token ID | 257 |
Feature activation | +1.36e+02 |
Pos logit contributions | |
Majesty | +0.338 |
pieces | +0.335 |
nels | +0.328 |
disl | +0.328 |
han | +0.325 |
Neg logit contributions | |
needed | -0.120 |
dark | -0.115 |
whispered | -0.113 |
meant | -0.108 |
said | -0.105 |
Token | time |
Token ID | 640 |
Feature activation | +1.35e+02 |
Pos logit contributions | |
pieces | +0.237 |
disl | +0.232 |
Majesty | +0.222 |
ric | +0.219 |
nels | +0.215 |
Neg logit contributions | |
needed | -0.158 |
whispered | -0.148 |
dark | -0.144 |
said | -0.142 |
chy | -0.139 |
Token | , |
Token ID | 11 |
Feature activation | +1.22e+02 |
Pos logit contributions | |
pieces | +0.271 |
disl | +0.262 |
Majesty | +0.257 |
n | +0.254 |
iu | +0.252 |
Neg logit contributions | |
needed | -0.113 |
whispered | -0.104 |
meant | -0.100 |
said | -0.096 |
Jones | -0.095 |
Token | there |
Token ID | 612 |
Feature activation | +1.27e+02 |
Pos logit contributions | |
n | +0.197 |
disl | +0.195 |
pieces | +0.193 |
Majesty | +0.187 |
interests | +0.186 |
Neg logit contributions | |
needed | -0.140 |
whispered | -0.136 |
meant | -0.131 |
Jones | -0.123 |
said | -0.118 |
Token | was |
Token ID | 373 |
Feature activation | +1.23e+02 |
Pos logit contributions | |
n | +0.200 |
disl | +0.180 |
blast | +0.177 |
victory | +0.177 |
interests | +0.176 |
Neg logit contributions | |
whispered | -0.154 |
needed | -0.142 |
Jones | -0.140 |
meant | -0.132 |
faded | -0.130 |
Token | a |
Token ID | 257 |
Feature activation | +1.23e+02 |
Pos logit contributions | |
disl | +0.210 |
n | +0.206 |
Majesty | +0.205 |
pieces | +0.205 |
nels | +0.200 |
Neg logit contributions | |
needed | -0.135 |
whispered | -0.129 |
meant | -0.127 |
dark | -0.120 |
said | -0.117 |
Token | brought |
Token ID | 3181 |
Feature activation | +1.73e+02 |
Pos logit contributions | |
n | +0.252 |
spend | +0.217 |
interests | +0.201 |
our | +0.193 |
pause | +0.191 |
Neg logit contributions | |
faded | -0.253 |
Jones | -0.251 |
wicked | -0.249 |
Ancient | -0.248 |
crooked | -0.247 |
Token | home |
Token ID | 1363 |
Feature activation | +1.79e+02 |
Pos logit contributions | |
n | +0.178 |
our | +0.171 |
victory | +0.147 |
your | +0.145 |
occasions | +0.143 |
Neg logit contributions | |
faded | -0.295 |
needed | -0.292 |
whispered | -0.289 |
chy | -0.286 |
spoke | -0.281 |
Token | a |
Token ID | 257 |
Feature activation | +1.67e+02 |
Pos logit contributions | |
n | +0.244 |
disl | +0.239 |
pieces | +0.236 |
interests | +0.231 |
iu | +0.226 |
Neg logit contributions | |
needed | -0.245 |
whispered | -0.241 |
meant | -0.233 |
Jones | -0.223 |
faded | -0.215 |
Token | newspaper |
Token ID | 7533 |
Feature activation | +1.84e+02 |
Pos logit contributions | |
victory | +0.211 |
n | +0.204 |
blast | +0.203 |
pause | +0.201 |
pieces | +0.201 |
Neg logit contributions | |
needed | -0.235 |
meant | -0.235 |
whispered | -0.233 |
Jones | -0.216 |
faded | -0.209 |
Token | . |
Token ID | 13 |
Feature activation | +1.38e+02 |
Pos logit contributions | |
n | +0.196 |
victory | +0.176 |
spend | +0.175 |
han | +0.174 |
pieces | +0.171 |
Neg logit contributions | |
meant | -0.275 |
whispered | -0.275 |
needed | -0.273 |
White | -0.263 |
faded | -0.260 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.64e+02 |
Pos logit contributions | |
n | +0.194 |
disl | +0.185 |
blast | +0.178 |
pieces | +0.178 |
spend | +0.175 |
Neg logit contributions | |
whispered | -0.182 |
needed | -0.181 |
meant | -0.175 |
Jones | -0.167 |
faded | -0.165 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.46e+02 |
Pos logit contributions | |
n | +0.261 |
disl | +0.233 |
interests | +0.227 |
arrest | +0.225 |
spend | +0.222 |
Neg logit contributions | |
Jones | -0.197 |
faded | -0.179 |
lessness | -0.177 |
White | -0.173 |
dark | -0.172 |
Token | it |
Token ID | 340 |
Feature activation | +1.50e+02 |
Pos logit contributions | |
n | +0.137 |
disl | +0.137 |
how | +0.132 |
arrest | +0.131 |
our | +0.128 |
Neg logit contributions | |
needed | -0.249 |
Jones | -0.244 |
whispered | -0.243 |
faded | -0.243 |
chy | -0.243 |
Token | and |
Token ID | 290 |
Feature activation | +1.32e+02 |
Pos logit contributions | |
n | +0.135 |
pause | +0.121 |
how | +0.120 |
go | +0.114 |
exercise | +0.111 |
Neg logit contributions | |
Jones | -0.310 |
lessness | -0.308 |
irk | -0.303 |
Did | -0.274 |
Gas | -0.270 |
Token | said |
Token ID | 531 |
Feature activation | +1.33e+02 |
Pos logit contributions | |
n | +0.240 |
disl | +0.235 |
spend | +0.219 |
Majesty | +0.218 |
how | +0.216 |
Neg logit contributions | |
Jones | -0.118 |
chy | -0.113 |
oky | -0.101 |
meant | -0.100 |
faded | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +1.23e+02 |
Pos logit contributions | |
n | +0.209 |
disl | +0.207 |
pieces | +0.202 |
Majesty | +0.197 |
interests | +0.196 |
Neg logit contributions | |
needed | -0.156 |
whispered | -0.154 |
meant | -0.148 |
Jones | -0.139 |
faded | -0.132 |
Token | Tom |
Token ID | 4186 |
Feature activation | +2.62e+01 |
Pos logit contributions | |
n | +0.004 |
disl | +0.004 |
interests | +0.004 |
our | +0.004 |
blast | +0.004 |
Neg logit contributions | |
needed | -0.004 |
whispered | -0.004 |
chy | -0.004 |
shone | -0.004 |
meant | -0.004 |
Token | dreamed |
Token ID | 27947 |
Feature activation | +8.22e+00 |
Pos logit contributions | |
n | +0.031 |
arrest | +0.028 |
blast | +0.025 |
interests | +0.023 |
pause | +0.023 |
Neg logit contributions | |
Pr | -0.039 |
Ancient | -0.038 |
Gas | -0.038 |
pods | -0.037 |
Jones | -0.037 |
Token | that |
Token ID | 326 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
n | +0.005 |
victory | +0.005 |
hey | +0.004 |
pieces | +0.004 |
how | +0.004 |
Neg logit contributions | |
Jones | -0.017 |
whispered | -0.017 |
Fuel | -0.016 |
White | -0.016 |
needed | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +5.99e+00 |
Pos logit contributions | |
our | +0.004 |
exercise | +0.004 |
victory | +0.004 |
how | +0.003 |
n | +0.003 |
Neg logit contributions | |
opy | -0.006 |
oky | -0.005 |
Ancient | -0.005 |
Factory | -0.005 |
whispered | -0.005 |
Token | big |
Token ID | 1263 |
Feature activation | +1.34e+01 |
Pos logit contributions | |
victory | +0.008 |
blast | +0.008 |
Majesty | +0.007 |
disl | +0.007 |
n | +0.007 |
Neg logit contributions | |
whispered | -0.009 |
meant | -0.009 |
needed | -0.008 |
Jones | -0.008 |
Fuel | -0.008 |
Token | monster |
Token ID | 9234 |
Feature activation | +2.03e+01 |
Pos logit contributions | |
victory | +0.018 |
blast | +0.018 |
disl | +0.017 |
interests | +0.017 |
n | +0.017 |
Neg logit contributions | |
whispered | -0.018 |
needed | -0.018 |
meant | -0.018 |
Jones | -0.017 |
spoke | -0.016 |
Token | took |
Token ID | 1718 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
n | +0.031 |
disl | +0.029 |
arrest | +0.028 |
blast | +0.028 |
victory | +0.027 |
Neg logit contributions | |
Jones | -0.024 |
whispered | -0.023 |
Fuel | -0.022 |
Ancient | -0.022 |
White | -0.022 |
Token | all |
Token ID | 477 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
needed | +0.002 |
whispered | +0.001 |
meant | +0.001 |
Jones | +0.001 |
spoke | +0.001 |
Neg logit contributions | |
n | -0.001 |
victory | -0.001 |
our | -0.001 |
pieces | -0.001 |
blast | -0.001 |
Token | his |
Token ID | 465 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
n | +0.002 |
disl | +0.002 |
pieces | +0.002 |
interests | +0.002 |
arrest | +0.002 |
Neg logit contributions | |
needed | -0.002 |
whispered | -0.002 |
meant | -0.002 |
Jones | -0.002 |
faded | -0.002 |
Token | toys |
Token ID | 14958 |
Feature activation | +4.95e+00 |
Pos logit contributions | |
interests | +0.003 |
n | +0.003 |
pieces | +0.003 |
victory | +0.003 |
disl | +0.003 |
Neg logit contributions | |
whispered | -0.004 |
Jones | -0.004 |
needed | -0.004 |
meant | -0.004 |
faded | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
victory | +0.006 |
n | +0.006 |
our | +0.005 |
arrest | +0.005 |
stay | +0.005 |
Neg logit contributions | |
White | -0.007 |
Ancient | -0.007 |
Pr | -0.007 |
whispered | -0.007 |
Dead | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +1.58e+02 |
Pos logit contributions | |
n | +0.377 |
disl | +0.372 |
pieces | +0.366 |
interests | +0.360 |
iu | +0.358 |
Neg logit contributions | |
needed | -0.187 |
whispered | -0.183 |
meant | -0.173 |
Jones | -0.161 |
faded | -0.152 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.83e+02 |
Pos logit contributions | |
n | +0.132 |
disl | +0.131 |
latable | +0.128 |
ric | +0.117 |
spend | +0.115 |
Neg logit contributions | |
whispered | -0.313 |
needed | -0.295 |
faded | -0.289 |
spoke | -0.284 |
meant | -0.280 |
Token | my |
Token ID | 1820 |
Feature activation | +1.94e+02 |
Pos logit contributions | |
n | +0.226 |
interests | +0.211 |
disl | +0.207 |
ric | +0.207 |
latable | +0.187 |
Neg logit contributions | |
whispered | -0.266 |
faded | -0.263 |
Jones | -0.257 |
needed | -0.252 |
spoke | -0.250 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.63e+02 |
Pos logit contributions | |
n | +0.289 |
disl | +0.269 |
interests | +0.263 |
blast | +0.248 |
ric | +0.248 |
Neg logit contributions | |
whispered | -0.236 |
Jones | -0.227 |
faded | -0.225 |
meant | -0.218 |
needed | -0.212 |
Token | planes |
Token ID | 13016 |
Feature activation | +1.79e+02 |
Pos logit contributions | |
n | +0.236 |
pieces | +0.236 |
disl | +0.230 |
Majesty | +0.226 |
iu | +0.223 |
Neg logit contributions | |
needed | -0.215 |
whispered | -0.208 |
meant | -0.201 |
Jones | -0.190 |
said | -0.188 |
Token | and |
Token ID | 290 |
Feature activation | +1.47e+02 |
Pos logit contributions | |
disl | +0.330 |
n | +0.329 |
pieces | +0.324 |
Majesty | +0.320 |
iu | +0.317 |
Neg logit contributions | |
needed | -0.167 |
whispered | -0.161 |
meant | -0.155 |
Jones | -0.140 |
said | -0.137 |
Token | always |
Token ID | 1464 |
Feature activation | +1.51e+02 |
Pos logit contributions | |
hey | +0.313 |
ieved | +0.312 |
disl | +0.309 |
pieces | +0.306 |
iu | +0.303 |
Neg logit contributions | |
needed | -0.129 |
whispered | -0.108 |
meant | -0.104 |
said | -0.101 |
spoke | -0.085 |
Token | dreamed |
Token ID | 27947 |
Feature activation | +1.43e+02 |
Pos logit contributions | |
n | +0.271 |
disl | +0.264 |
spend | +0.260 |
interests | +0.260 |
blast | +0.250 |
Neg logit contributions | |
Jones | -0.122 |
whispered | -0.121 |
aisle | -0.119 |
needed | -0.118 |
meant | -0.114 |
Token | of |
Token ID | 286 |
Feature activation | +1.48e+02 |
Pos logit contributions | |
n | +0.144 |
victory | +0.140 |
pieces | +0.135 |
disl | +0.133 |
interests | +0.132 |
Neg logit contributions | |
whispered | -0.249 |
Jones | -0.243 |
needed | -0.243 |
meant | -0.238 |
White | -0.223 |
Token | flying |
Token ID | 7348 |
Feature activation | +1.47e+02 |
Pos logit contributions | |
n | +0.199 |
disl | +0.197 |
pieces | +0.192 |
interests | +0.187 |
victory | +0.187 |
Neg logit contributions | |
needed | -0.208 |
whispered | -0.204 |
meant | -0.196 |
Jones | -0.189 |
faded | -0.181 |
Token | one |
Token ID | 530 |
Feature activation | +1.38e+02 |
Pos logit contributions | |
n | +0.208 |
disl | +0.203 |
pieces | +0.200 |
interests | +0.198 |
victory | +0.198 |
Neg logit contributions | |
needed | -0.191 |
whispered | -0.191 |
meant | -0.181 |
Jones | -0.178 |
faded | -0.166 |
Token | slowly |
Token ID | 6364 |
Feature activation | +9.50e+00 |
Pos logit contributions | |
n | +0.024 |
pieces | +0.022 |
victory | +0.022 |
iu | +0.021 |
ap | +0.021 |
Neg logit contributions | |
whispered | -0.028 |
needed | -0.027 |
meant | -0.027 |
Jones | -0.027 |
spoke | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
n | +0.013 |
victory | +0.012 |
pieces | +0.011 |
join | +0.011 |
blast | +0.011 |
Neg logit contributions | |
Jones | -0.013 |
whispered | -0.013 |
needed | -0.013 |
meant | -0.012 |
spoke | -0.012 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
whispered | +0.018 |
faded | +0.018 |
needed | +0.018 |
chy | +0.017 |
meant | +0.017 |
Neg logit contributions | |
disl | -0.011 |
n | -0.010 |
arrest | -0.009 |
hey | -0.008 |
yeah | -0.008 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
n | +0.017 |
interests | +0.014 |
blast | +0.014 |
arrest | +0.014 |
disl | +0.014 |
Neg logit contributions | |
Jones | -0.014 |
dark | -0.013 |
meant | -0.012 |
White | -0.012 |
bored | -0.012 |
Token | that |
Token ID | 326 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
whispered | +0.010 |
meant | +0.010 |
said | +0.010 |
needed | +0.010 |
dark | +0.009 |
Neg logit contributions | |
pieces | -0.019 |
Majesty | -0.018 |
blast | -0.018 |
iu | -0.018 |
ieved | -0.017 |
Token | asking |
Token ID | 4737 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
whispered | +0.017 |
needed | +0.016 |
Jones | +0.015 |
meant | +0.015 |
faded | +0.015 |
Neg logit contributions | |
n | -0.013 |
disl | -0.012 |
victory | -0.012 |
interests | -0.012 |
iu | -0.011 |
Token | for |
Token ID | 329 |
Feature activation | -1.52e+01 |
Pos logit contributions | |
whispered | +0.022 |
faded | +0.022 |
needed | +0.021 |
spoke | +0.021 |
aisle | +0.021 |
Neg logit contributions | |
interests | -0.014 |
n | -0.014 |
our | -0.014 |
respect | -0.014 |
victory | -0.014 |
Token | help |
Token ID | 1037 |
Feature activation | -1.82e+01 |
Pos logit contributions | |
needed | +0.024 |
whispered | +0.023 |
meant | +0.022 |
said | +0.021 |
Jones | +0.021 |
Neg logit contributions | |
pieces | -0.019 |
n | -0.018 |
disl | -0.018 |
iu | -0.018 |
Majesty | -0.017 |
Token | is |
Token ID | 318 |
Feature activation | -2.12e+01 |
Pos logit contributions | |
needed | +0.021 |
whispered | +0.020 |
meant | +0.019 |
Jones | +0.018 |
faded | +0.017 |
Neg logit contributions | |
n | -0.030 |
disl | -0.029 |
pieces | -0.029 |
interests | -0.028 |
Majesty | -0.028 |
Token | good |
Token ID | 922 |
Feature activation | -1.85e+01 |
Pos logit contributions | |
whispered | +0.033 |
needed | +0.033 |
Jones | +0.032 |
meant | +0.032 |
faded | +0.030 |
Neg logit contributions | |
n | -0.025 |
disl | -0.024 |
interests | -0.023 |
victory | -0.023 |
pieces | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
needed | +0.024 |
meant | +0.024 |
whispered | +0.023 |
said | +0.021 |
Jones | +0.021 |
Neg logit contributions | |
pieces | -0.029 |
n | -0.028 |
disl | -0.028 |
Majesty | -0.027 |
ric | -0.027 |
Token | even |
Token ID | 772 |
Feature activation | -1.02e+02 |
Pos logit contributions | |
needed | +0.090 |
whispered | +0.086 |
said | +0.084 |
sleepy | +0.079 |
sleeping | +0.079 |
Neg logit contributions | |
earned | -0.201 |
pieces | -0.193 |
disl | -0.190 |
han | -0.179 |
ieved | -0.177 |
Token | though |
Token ID | 996 |
Feature activation | -1.06e+02 |
Pos logit contributions | |
needed | +0.108 |
whispered | +0.106 |
meant | +0.103 |
bored | +0.098 |
dark | +0.097 |
Neg logit contributions | |
ric | -0.189 |
n | -0.187 |
Majesty | -0.186 |
pieces | -0.185 |
arrest | -0.185 |
Token | it |
Token ID | 340 |
Feature activation | -1.06e+02 |
Pos logit contributions | |
needed | +0.163 |
meant | +0.150 |
whispered | +0.147 |
dark | +0.141 |
said | +0.140 |
Neg logit contributions | |
disl | -0.140 |
pieces | -0.138 |
iu | -0.136 |
n | -0.135 |
Majesty | -0.134 |
Token | r |
Token ID | 374 |
Feature activation | -1.04e+02 |
Pos logit contributions | |
Jones | +0.122 |
whispered | +0.120 |
needed | +0.112 |
meant | +0.107 |
Fuel | +0.106 |
Neg logit contributions | |
n | -0.170 |
disl | -0.155 |
interests | -0.150 |
arrest | -0.148 |
blast | -0.147 |
Token | ained |
Token ID | 1328 |
Feature activation | -1.08e+02 |
Pos logit contributions | |
needed | +0.092 |
meant | +0.087 |
whispered | +0.085 |
Jones | +0.079 |
bored | +0.073 |
Neg logit contributions | |
pieces | -0.196 |
disl | -0.194 |
n | -0.191 |
arrest | -0.187 |
Majesty | -0.186 |
Token | . |
Token ID | 13 |
Feature activation | -1.26e+02 |
Pos logit contributions | |
whispered | +0.157 |
needed | +0.156 |
meant | +0.149 |
Jones | +0.143 |
spoke | +0.138 |
Neg logit contributions | |
n | -0.142 |
disl | -0.136 |
pieces | -0.134 |
victory | -0.130 |
interests | -0.130 |
Token | I |
Token ID | 314 |
Feature activation | -1.20e+02 |
Pos logit contributions | |
White | +0.124 |
needed | +0.121 |
meant | +0.116 |
Jones | +0.113 |
T | +0.107 |
Neg logit contributions | |
aches | -0.242 |
pieces | -0.237 |
disl | -0.233 |
ered | -0.229 |
immers | -0.228 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.30e+02 |
Pos logit contributions | |
needed | +0.182 |
whispered | +0.177 |
meant | +0.172 |
said | +0.161 |
Jones | +0.161 |
Neg logit contributions | |
pieces | -0.151 |
n | -0.150 |
disl | -0.149 |
iu | -0.144 |
Majesty | -0.144 |
Token | that |
Token ID | 326 |
Feature activation | -1.30e+02 |
Pos logit contributions | |
meant | +0.150 |
said | +0.143 |
whispered | +0.142 |
needed | +0.140 |
dark | +0.133 |
Neg logit contributions | |
pieces | -0.235 |
iu | -0.229 |
Majesty | -0.225 |
arrest | -0.218 |
ocol | -0.217 |
Token | sometimes |
Token ID | 3360 |
Feature activation | -1.30e+02 |
Pos logit contributions | |
needed | +0.178 |
whispered | +0.176 |
meant | +0.168 |
Jones | +0.162 |
said | +0.155 |
Neg logit contributions | |
n | -0.181 |
disl | -0.177 |
pieces | -0.174 |
interests | -0.169 |
Majesty | -0.168 |
Token | we |
Token ID | 356 |
Feature activation | -1.29e+02 |
Pos logit contributions | |
needed | +0.184 |
meant | +0.174 |
whispered | +0.172 |
dark | +0.159 |
said | +0.158 |
Neg logit contributions | |
pieces | -0.184 |
disl | -0.183 |
n | -0.180 |
iu | -0.179 |
ric | -0.175 |
Token | surprised |
Token ID | 6655 |
Feature activation | -2.67e+01 |
Pos logit contributions | |
Jones | +0.044 |
aisle | +0.043 |
needed | +0.041 |
pods | +0.040 |
whispered | +0.039 |
Neg logit contributions | |
disl | -0.037 |
n | -0.035 |
victory | -0.034 |
ieved | -0.032 |
our | -0.032 |
Token | that |
Token ID | 326 |
Feature activation | -3.42e+01 |
Pos logit contributions | |
needed | +0.029 |
whispered | +0.028 |
meant | +0.028 |
said | +0.028 |
Jones | +0.025 |
Neg logit contributions | |
pieces | -0.047 |
Majesty | -0.045 |
disl | -0.045 |
ric | -0.044 |
blast | -0.044 |
Token | he |
Token ID | 339 |
Feature activation | -2.94e+01 |
Pos logit contributions | |
whispered | +0.034 |
needed | +0.034 |
meant | +0.032 |
Jones | +0.030 |
chy | +0.029 |
Neg logit contributions | |
n | -0.059 |
disl | -0.057 |
victory | -0.057 |
pieces | -0.056 |
Majesty | -0.055 |
Token | stopped |
Token ID | 5025 |
Feature activation | -3.56e+01 |
Pos logit contributions | |
Jones | +0.028 |
needed | +0.028 |
whispered | +0.027 |
meant | +0.027 |
aisle | +0.026 |
Neg logit contributions | |
n | -0.051 |
disl | -0.048 |
pieces | -0.047 |
Majesty | -0.047 |
arrest | -0.046 |
Token | roaring |
Token ID | 43170 |
Feature activation | -2.98e+01 |
Pos logit contributions | |
needed | +0.043 |
whispered | +0.042 |
meant | +0.040 |
Jones | +0.038 |
said | +0.036 |
Neg logit contributions | |
n | -0.056 |
disl | -0.055 |
pieces | -0.054 |
Majesty | -0.053 |
iu | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | -4.72e+01 |
Pos logit contributions | |
needed | +0.046 |
Fuel | +0.043 |
whispered | +0.042 |
Ancient | +0.041 |
meant | +0.040 |
Neg logit contributions | |
victory | -0.036 |
n | -0.034 |
pause | -0.031 |
pieces | -0.031 |
how | -0.031 |
Token | his |
Token ID | 465 |
Feature activation | -3.87e+01 |
Pos logit contributions | |
needed | +0.044 |
whispered | +0.043 |
meant | +0.042 |
Jones | +0.039 |
faded | +0.036 |
Neg logit contributions | |
n | -0.085 |
disl | -0.083 |
pieces | -0.081 |
Majesty | -0.080 |
arrest | -0.080 |
Token | anger |
Token ID | 8993 |
Feature activation | -3.78e+01 |
Pos logit contributions | |
whispered | +0.058 |
meant | +0.057 |
needed | +0.057 |
Jones | +0.053 |
spoke | +0.051 |
Neg logit contributions | |
victory | -0.049 |
n | -0.047 |
Majesty | -0.047 |
disl | -0.047 |
interests | -0.046 |
Token | suddenly |
Token ID | 6451 |
Feature activation | -4.48e+01 |
Pos logit contributions | |
Jones | +0.047 |
aisle | +0.044 |
Ancient | +0.043 |
Fuel | +0.043 |
whispered | +0.043 |
Neg logit contributions | |
n | -0.057 |
victory | -0.053 |
pause | -0.052 |
blast | -0.052 |
arrest | -0.051 |
Token | vanished |
Token ID | 23717 |
Feature activation | -4.55e+01 |
Pos logit contributions | |
aisle | +0.045 |
Fuel | +0.043 |
Jones | +0.043 |
Ancient | +0.043 |
White | +0.041 |
Neg logit contributions | |
Majesty | -0.071 |
n | -0.070 |
disl | -0.070 |
arrest | -0.069 |
victory | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -6.68e+01 |
Pos logit contributions | |
whispered | +0.062 |
needed | +0.062 |
meant | +0.058 |
Jones | +0.055 |
said | +0.054 |
Neg logit contributions | |
n | -0.060 |
victory | -0.059 |
disl | -0.058 |
pieces | -0.058 |
Majesty | -0.058 |
Token | for |
Token ID | 329 |
Feature activation | -1.71e+02 |
Pos logit contributions | |
needed | +0.197 |
whispered | +0.193 |
meant | +0.186 |
Jones | +0.176 |
said | +0.167 |
Neg logit contributions | |
n | -0.257 |
disl | -0.254 |
pieces | -0.252 |
interests | -0.245 |
Majesty | -0.243 |
Token | animals |
Token ID | 4695 |
Feature activation | -1.67e+02 |
Pos logit contributions | |
whispered | +0.207 |
needed | +0.201 |
meant | +0.199 |
dark | +0.197 |
said | +0.190 |
Neg logit contributions | |
n | -0.274 |
iu | -0.258 |
ieved | -0.255 |
disl | -0.255 |
Majesty | -0.255 |
Token | and |
Token ID | 290 |
Feature activation | -1.77e+02 |
Pos logit contributions | |
needed | +0.193 |
whispered | +0.181 |
meant | +0.176 |
faded | +0.171 |
Jones | +0.171 |
Neg logit contributions | |
n | -0.265 |
pieces | -0.255 |
interests | -0.248 |
victory | -0.248 |
disl | -0.247 |
Token | bugs |
Token ID | 11316 |
Feature activation | -1.69e+02 |
Pos logit contributions | |
needed | +0.223 |
whispered | +0.208 |
said | +0.201 |
dark | +0.194 |
meant | +0.190 |
Neg logit contributions | |
ieved | -0.337 |
han | -0.310 |
iu | -0.310 |
ext | -0.300 |
immers | -0.296 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e+02 |
Pos logit contributions | |
needed | +0.194 |
meant | +0.181 |
whispered | +0.179 |
Jones | +0.175 |
White | +0.164 |
Neg logit contributions | |
n | -0.275 |
pieces | -0.263 |
disl | -0.255 |
arrest | -0.253 |
interests | -0.249 |
Token | They |
Token ID | 1119 |
Feature activation | -1.82e+02 |
Pos logit contributions | |
needed | +0.165 |
whispered | +0.157 |
meant | +0.150 |
Jones | +0.138 |
said | +0.133 |
Neg logit contributions | |
n | -0.351 |
disl | -0.350 |
pieces | -0.346 |
Majesty | -0.341 |
arrest | -0.339 |
Token | had |
Token ID | 550 |
Feature activation | -1.86e+02 |
Pos logit contributions | |
elly | +0.254 |
oky | +0.250 |
Jones | +0.235 |
late | +0.231 |
wicked | +0.225 |
Neg logit contributions | |
spend | -0.255 |
n | -0.231 |
pause | -0.225 |
explore | -0.219 |
stay | -0.217 |
Token | a |
Token ID | 257 |
Feature activation | -1.87e+02 |
Pos logit contributions | |
needed | +0.247 |
whispered | +0.243 |
meant | +0.234 |
Jones | +0.224 |
said | +0.213 |
Neg logit contributions | |
n | -0.266 |
disl | -0.261 |
pieces | -0.257 |
interests | -0.250 |
Majesty | -0.248 |
Token | big |
Token ID | 1263 |
Feature activation | -1.80e+02 |
Pos logit contributions | |
needed | +0.240 |
whispered | +0.236 |
meant | +0.229 |
Jones | +0.216 |
said | +0.207 |
Neg logit contributions | |
n | -0.272 |
disl | -0.268 |
pieces | -0.264 |
interests | -0.258 |
Majesty | -0.258 |
Token | book |
Token ID | 1492 |
Feature activation | -1.80e+02 |
Pos logit contributions | |
needed | +0.235 |
whispered | +0.229 |
meant | +0.220 |
Jones | +0.210 |
said | +0.203 |
Neg logit contributions | |
n | -0.265 |
disl | -0.262 |
pieces | -0.258 |
iu | -0.249 |
interests | -0.246 |
Token | with |
Token ID | 351 |
Feature activation | -1.90e+02 |
Pos logit contributions | |
needed | +0.306 |
whispered | +0.297 |
Jones | +0.290 |
meant | +0.288 |
White | +0.284 |
Neg logit contributions | |
n | -0.179 |
interests | -0.167 |
victory | -0.165 |
pieces | -0.165 |
blast | -0.162 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
needed | +0.026 |
whispered | +0.025 |
meant | +0.024 |
Jones | +0.023 |
said | +0.023 |
Neg logit contributions | |
n | -0.022 |
pieces | -0.022 |
disl | -0.022 |
iu | -0.021 |
Majesty | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -3.95e+01 |
Pos logit contributions | |
needed | +0.024 |
whispered | +0.023 |
meant | +0.022 |
Jones | +0.022 |
faded | +0.021 |
Neg logit contributions | |
n | -0.029 |
pieces | -0.028 |
interests | -0.027 |
arrest | -0.026 |
disl | -0.026 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.04e+01 |
Pos logit contributions | |
needed | +0.039 |
whispered | +0.038 |
meant | +0.036 |
Jones | +0.034 |
faded | +0.032 |
Neg logit contributions | |
n | -0.069 |
disl | -0.069 |
pieces | -0.067 |
interests | -0.066 |
arrest | -0.065 |
Token | L |
Token ID | 43 |
Feature activation | -1.82e+01 |
Pos logit contributions | |
whispered | +0.041 |
needed | +0.041 |
meant | +0.038 |
faded | +0.036 |
chy | +0.035 |
Neg logit contributions | |
n | -0.070 |
blast | -0.065 |
victory | -0.063 |
arrest | -0.063 |
ric | -0.063 |
Token | ily |
Token ID | 813 |
Feature activation | -2.40e+01 |
Pos logit contributions | |
needed | +0.020 |
whispered | +0.020 |
meant | +0.019 |
said | +0.018 |
Jones | +0.017 |
Neg logit contributions | |
disl | -0.032 |
n | -0.030 |
Majesty | -0.030 |
pieces | -0.030 |
interests | -0.029 |
Token | liked |
Token ID | 8288 |
Feature activation | -4.02e+01 |
Pos logit contributions | |
Pr | +0.030 |
lessness | +0.030 |
Jones | +0.029 |
Fuel | +0.029 |
White | +0.029 |
Neg logit contributions | |
arrest | -0.032 |
n | -0.031 |
disl | -0.030 |
interests | -0.029 |
blast | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | -4.15e+01 |
Pos logit contributions | |
needed | +0.056 |
whispered | +0.054 |
meant | +0.053 |
Jones | +0.050 |
said | +0.049 |
Neg logit contributions | |
n | -0.055 |
disl | -0.054 |
pieces | -0.054 |
Majesty | -0.052 |
interests | -0.052 |
Token | play |
Token ID | 711 |
Feature activation | -4.23e+01 |
Pos logit contributions | |
needed | +0.063 |
whispered | +0.062 |
said | +0.059 |
meant | +0.057 |
spoke | +0.056 |
Neg logit contributions | |
ieved | -0.053 |
hey | -0.052 |
iu | -0.052 |
Majesty | -0.051 |
ric | -0.050 |
Token | with |
Token ID | 351 |
Feature activation | -4.66e+01 |
Pos logit contributions | |
whispered | +0.047 |
needed | +0.045 |
said | +0.043 |
meant | +0.042 |
Jones | +0.040 |
Neg logit contributions | |
disl | -0.070 |
Majesty | -0.069 |
ric | -0.068 |
n | -0.068 |
iu | -0.068 |
Token | clay |
Token ID | 21558 |
Feature activation | -3.49e+01 |
Pos logit contributions | |
needed | +0.059 |
whispered | +0.057 |
meant | +0.055 |
Jones | +0.052 |
faded | +0.050 |
Neg logit contributions | |
disl | -0.070 |
n | -0.070 |
pieces | -0.068 |
arrest | -0.067 |
interests | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -6.15e+01 |
Pos logit contributions | |
whispered | +0.042 |
needed | +0.041 |
meant | +0.039 |
Jones | +0.038 |
faded | +0.038 |
Neg logit contributions | |
pieces | -0.050 |
n | -0.050 |
disl | -0.049 |
victory | -0.049 |
arrest | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -1.14e+02 |
Pos logit contributions | |
needed | +0.137 |
whispered | +0.131 |
meant | +0.131 |
faded | +0.125 |
Jones | +0.120 |
Neg logit contributions | |
n | -0.187 |
blast | -0.163 |
interests | -0.163 |
disl | -0.163 |
pieces | -0.162 |
Token | One |
Token ID | 3198 |
Feature activation | -1.15e+02 |
Pos logit contributions | |
needed | +0.201 |
whispered | +0.198 |
meant | +0.195 |
Jones | +0.191 |
faded | +0.182 |
Neg logit contributions | |
n | -0.125 |
disl | -0.112 |
pieces | -0.110 |
blast | -0.108 |
arrest | -0.108 |
Token | little |
Token ID | 1310 |
Feature activation | -1.06e+02 |
Pos logit contributions | |
whispered | +0.149 |
needed | +0.143 |
Jones | +0.142 |
meant | +0.138 |
spoke | +0.129 |
Neg logit contributions | |
n | -0.167 |
victory | -0.163 |
blast | -0.159 |
interests | -0.156 |
pieces | -0.156 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.02e+02 |
Pos logit contributions | |
needed | +0.109 |
whispered | +0.108 |
meant | +0.103 |
chy | +0.097 |
Jones | +0.096 |
Neg logit contributions | |
n | -0.174 |
pieces | -0.174 |
victory | -0.172 |
disl | -0.172 |
interests | -0.171 |
Token | , |
Token ID | 11 |
Feature activation | -1.20e+02 |
Pos logit contributions | |
chy | +0.120 |
Jones | +0.117 |
blank | +0.115 |
meant | +0.114 |
faded | +0.112 |
Neg logit contributions | |
n | -0.155 |
pieces | -0.139 |
arrest | -0.138 |
spend | -0.138 |
blast | -0.137 |
Token | named |
Token ID | 3706 |
Feature activation | -1.16e+02 |
Pos logit contributions | |
chy | +0.226 |
faded | +0.208 |
needed | +0.201 |
meant | +0.200 |
crooked | +0.197 |
Neg logit contributions | |
n | -0.112 |
hey | -0.109 |
join | -0.106 |
how | -0.105 |
our | -0.102 |
Token | Linda |
Token ID | 21776 |
Feature activation | -1.08e+02 |
Pos logit contributions | |
needed | +0.119 |
meant | +0.116 |
whispered | +0.111 |
said | +0.111 |
Jones | +0.110 |
Neg logit contributions | |
iu | -0.201 |
interests | -0.200 |
blast | -0.200 |
ric | -0.198 |
earned | -0.197 |
Token | , |
Token ID | 11 |
Feature activation | -1.25e+02 |
Pos logit contributions | |
meant | +0.141 |
needed | +0.137 |
faded | +0.137 |
dark | +0.136 |
whispered | +0.135 |
Neg logit contributions | |
n | -0.155 |
pieces | -0.136 |
disl | -0.136 |
arrest | -0.136 |
spend | -0.132 |
Token | came |
Token ID | 1625 |
Feature activation | -1.24e+02 |
Pos logit contributions | |
chy | +0.218 |
crooked | +0.209 |
dusty | +0.201 |
stiff | +0.200 |
oky | +0.197 |
Neg logit contributions | |
how | -0.141 |
join | -0.128 |
spend | -0.127 |
n | -0.125 |
hey | -0.117 |
Token | up |
Token ID | 510 |
Feature activation | -1.20e+02 |
Pos logit contributions | |
needed | +0.166 |
whispered | +0.162 |
meant | +0.156 |
Jones | +0.151 |
said | +0.144 |
Neg logit contributions | |
disl | -0.179 |
n | -0.176 |
pieces | -0.170 |
iu | -0.167 |
Majesty | -0.166 |
Token | to |
Token ID | 284 |
Feature activation | -1.28e+02 |
Pos logit contributions | |
whispered | +0.210 |
needed | +0.207 |
meant | +0.203 |
chy | +0.194 |
Lover | +0.193 |
Neg logit contributions | |
victory | -0.119 |
join | -0.107 |
ap | -0.103 |
arrest | -0.102 |
n | -0.100 |