Token | dodge |
Token ID | 24885 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
ies | +0.381 |
ishing | +0.368 |
ig | +0.363 |
ater | +0.339 |
- | +0.333 |
Neg logit contributions | |
Don | -0.371 |
My | -0.366 |
It | -0.327 |
Would | -0.325 |
That | -0.322 |
Token | the |
Token ID | 262 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
ies | +0.168 |
ig | +0.153 |
eye | +0.148 |
- | +0.145 |
ish | +0.145 |
Neg logit contributions | |
Don | -0.141 |
My | -0.132 |
It | -0.123 |
Would | -0.121 |
That | -0.121 |
Token | lion |
Token ID | 18744 |
Feature activation | -1.38e+01 |
Pos logit contributions | |
Don | +0.461 |
My | +0.438 |
Would | +0.433 |
Here | +0.425 |
There | +0.419 |
Neg logit contributions | |
ies | -0.352 |
- | -0.316 |
ig | -0.306 |
ishing | -0.305 |
ish | -0.285 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e+01 |
Pos logit contributions | |
ies | +0.429 |
ishing | +0.417 |
ig | +0.384 |
choice | +0.345 |
ish | +0.343 |
Neg logit contributions | |
Don | -0.560 |
My | -0.554 |
It | -0.526 |
It | -0.522 |
That | -0.507 |
Token | " |
Token ID | 366 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
, | +0.501 |
. | +0.481 |
window | +0.436 |
- | +0.431 |
house | +0.416 |
Neg logit contributions | |
Ah | -0.563 |
Looks | -0.562 |
Take | -0.554 |
Ta | -0.537 |
Don | -0.537 |
Token | My |
Token ID | 3666 |
Feature activation | +3.80e+01 |
Pos logit contributions | |
Everything | +0.260 |
Did | +0.255 |
From | +0.240 |
Would | +0.226 |
Thanks | +0.207 |
Neg logit contributions | |
- | -0.583 |
ater | -0.577 |
an | -0.551 |
ies | -0.520 |
times | -0.497 |
Token | gir |
Token ID | 37370 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
My | +1.506 |
Don | +1.471 |
It | +1.444 |
Who | +1.387 |
Would | +1.386 |
Neg logit contributions | |
ies | -1.250 |
ishing | -1.133 |
ig | -1.025 |
ie | -0.994 |
- | -0.980 |
Token | affe |
Token ID | 21223 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
ies | +0.528 |
ishing | +0.464 |
- | +0.434 |
decision | +0.433 |
ig | +0.429 |
Neg logit contributions | |
Don | -0.505 |
Well | -0.488 |
My | -0.486 |
It | -0.466 |
It | -0.464 |
Token | is |
Token ID | 318 |
Feature activation | +8.20e+00 |
Pos logit contributions | |
ies | +0.449 |
- | +0.388 |
ishing | +0.384 |
eye | +0.383 |
ig | +0.379 |
Neg logit contributions | |
Don | -0.343 |
My | -0.324 |
It | -0.304 |
I | -0.295 |
Well | -0.295 |
Token | the |
Token ID | 262 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
Don | +0.337 |
My | +0.316 |
Would | +0.304 |
It | +0.297 |
There | +0.296 |
Neg logit contributions | |
ies | -0.323 |
ig | -0.285 |
- | -0.284 |
ish | -0.278 |
an | -0.275 |
Token | tallest |
Token ID | 38760 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
My | +0.448 |
Don | +0.445 |
Would | +0.414 |
I | +0.409 |
There | +0.400 |
Neg logit contributions | |
ies | -0.396 |
- | -0.344 |
ishing | -0.339 |
ig | -0.331 |
ish | -0.310 |
Token | TV |
Token ID | 3195 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
Don | +0.499 |
My | +0.488 |
Would | +0.476 |
It | +0.469 |
I | +0.455 |
Neg logit contributions | |
ies | -0.302 |
ishing | -0.268 |
- | -0.266 |
ig | -0.265 |
ish | -0.261 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
ies | +0.255 |
ishing | +0.247 |
ig | +0.241 |
ish | +0.228 |
- | +0.212 |
Neg logit contributions | |
Don | -0.533 |
My | -0.523 |
Would | -0.522 |
She | -0.494 |
It | -0.491 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e+01 |
Pos logit contributions | |
. | +0.498 |
, | +0.483 |
- | +0.470 |
ies | +0.438 |
and | +0.424 |
Neg logit contributions | |
Looks | -0.602 |
Don | -0.585 |
Ah | -0.558 |
Take | -0.554 |
Here | -0.553 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e+01 |
Pos logit contributions | |
Don | +0.475 |
My | +0.416 |
Well | +0.410 |
Would | +0.402 |
Did | +0.383 |
Neg logit contributions | |
ies | -0.950 |
- | -0.941 |
ig | -0.835 |
an | -0.834 |
. | -0.826 |
Token | " |
Token ID | 1 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
Don | +0.687 |
Would | +0.644 |
My | +0.644 |
Well | +0.637 |
Did | +0.604 |
Neg logit contributions | |
ies | -0.741 |
- | -0.705 |
an | -0.626 |
ig | -0.620 |
ish | -0.599 |
Token | My |
Token ID | 3666 |
Feature activation | +3.80e+01 |
Pos logit contributions | |
Everything | +0.081 |
Sadly | +0.078 |
Did | +0.076 |
From | +0.075 |
Eventually | +0.074 |
Neg logit contributions | |
- | -0.170 |
ies | -0.161 |
song | -0.157 |
ater | -0.155 |
an | -0.154 |
Token | name |
Token ID | 1438 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
My | +1.136 |
Don | +1.104 |
It | +1.048 |
Would | +1.045 |
It | +0.997 |
Neg logit contributions | |
ies | -1.633 |
ishing | -1.547 |
ig | -1.517 |
- | -1.466 |
ish | -1.422 |
Token | is |
Token ID | 318 |
Feature activation | +7.87e+00 |
Pos logit contributions | |
ies | +0.628 |
- | +0.550 |
ig | +0.541 |
ish | +0.536 |
ishing | +0.533 |
Neg logit contributions | |
Don | -0.548 |
My | -0.529 |
Would | -0.496 |
It | -0.489 |
Well | -0.483 |
Token | Ben |
Token ID | 3932 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
Don | +0.258 |
Would | +0.224 |
Well | +0.222 |
My | +0.219 |
Here | +0.214 |
Neg logit contributions | |
ies | -0.377 |
- | -0.364 |
ig | -0.326 |
an | -0.324 |
ish | -0.324 |
Token | ," |
Token ID | 553 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
My | +0.358 |
Don | +0.356 |
It | +0.336 |
She | +0.335 |
I | +0.332 |
Neg logit contributions | |
ishing | -0.193 |
ies | -0.189 |
ig | -0.189 |
decision | -0.178 |
eye | -0.172 |
Token | Ben |
Token ID | 3932 |
Feature activation | +7.86e+00 |
Pos logit contributions | |
ies | +0.609 |
- | +0.552 |
ig | +0.521 |
eye | +0.514 |
ish | +0.513 |
Neg logit contributions | |
Don | -0.470 |
Would | -0.423 |
My | -0.418 |
Well | -0.416 |
Here | -0.391 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.02e+01 |
Pos logit contributions | |
It | +0.413 |
Don | +0.409 |
My | +0.407 |
It | +0.392 |
That | +0.382 |
Neg logit contributions | |
ies | -0.286 |
an | -0.263 |
ish | -0.252 |
ig | -0.251 |
ishing | -0.249 |
Token | Ben |
Token ID | 3932 |
Feature activation | +8.07e+00 |
Pos logit contributions | |
ies | +0.834 |
- | +0.768 |
ig | +0.720 |
eye | +0.715 |
. | +0.685 |
Neg logit contributions | |
Don | -0.747 |
Would | -0.724 |
My | -0.665 |
Here | -0.664 |
Well | -0.652 |
Token | says |
Token ID | 1139 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
Don | +0.290 |
My | +0.271 |
Would | +0.259 |
Well | +0.254 |
It | +0.243 |
Neg logit contributions | |
ies | -0.354 |
- | -0.324 |
ig | -0.313 |
an | -0.303 |
ishing | -0.302 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
ies | +0.051 |
island | +0.049 |
an | +0.048 |
ig | +0.048 |
ish | +0.047 |
Neg logit contributions | |
Don | -0.093 |
My | -0.089 |
It | -0.086 |
Would | -0.084 |
It | -0.083 |
Token | " |
Token ID | 366 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
- | +0.616 |
ies | +0.590 |
. | +0.570 |
, | +0.549 |
head | +0.500 |
Neg logit contributions | |
Don | -0.511 |
Here | -0.510 |
My | -0.492 |
There | -0.458 |
Our | -0.457 |
Token | My |
Token ID | 3666 |
Feature activation | +3.80e+01 |
Pos logit contributions | |
Did | +0.210 |
Everything | +0.203 |
Would | +0.197 |
From | +0.189 |
Don | +0.176 |
Neg logit contributions | |
- | -0.674 |
ater | -0.640 |
an | -0.628 |
ies | -0.623 |
. | -0.566 |
Token | car |
Token ID | 1097 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
My | +1.529 |
It | +1.469 |
Don | +1.418 |
It | +1.409 |
I | +1.364 |
Neg logit contributions | |
ies | -1.297 |
ishing | -1.111 |
ie | -1.095 |
- | -1.060 |
ater | -1.040 |
Token | is |
Token ID | 318 |
Feature activation | +8.11e+00 |
Pos logit contributions | |
ies | +0.470 |
ishing | +0.408 |
- | +0.403 |
eye | +0.403 |
ig | +0.401 |
Neg logit contributions | |
Don | -0.354 |
My | -0.343 |
It | -0.319 |
I | -0.311 |
Well | -0.308 |
Token | cooler |
Token ID | 19346 |
Feature activation | -1.63e+01 |
Pos logit contributions | |
Don | +0.320 |
My | +0.310 |
It | +0.290 |
Would | +0.288 |
There | +0.280 |
Neg logit contributions | |
ies | -0.325 |
ish | -0.285 |
an | -0.284 |
- | -0.282 |
ig | -0.281 |
Token | than |
Token ID | 621 |
Feature activation | -1.55e+01 |
Pos logit contributions | |
ies | +0.480 |
ig | +0.465 |
ishing | +0.438 |
eye | +0.389 |
ish | +0.384 |
Neg logit contributions | |
I | -0.670 |
It | -0.644 |
Don | -0.635 |
That | -0.633 |
My | -0.603 |
Token | yours |
Token ID | 12431 |
Feature activation | -8.70e-02 |
Pos logit contributions | |
ies | +0.659 |
- | +0.612 |
ig | +0.568 |
ish | +0.561 |
an | +0.555 |
Neg logit contributions | |
Don | -0.594 |
My | -0.549 |
Would | -0.547 |
Well | -0.525 |
There | -0.508 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e+01 |
Pos logit contributions | |
ies | +0.335 |
ig | +0.324 |
decision | +0.319 |
an | +0.303 |
choice | +0.301 |
Neg logit contributions | |
Don | -0.633 |
It | -0.605 |
She | -0.600 |
My | -0.597 |
I | -0.594 |
Token | She |
Token ID | 1375 |
Feature activation | +2.60e+01 |
Pos logit contributions | |
- | +0.743 |
ies | +0.739 |
eye | +0.734 |
fort | +0.694 |
house | +0.684 |
Neg logit contributions | |
Don | -0.356 |
My | -0.320 |
Would | -0.312 |
Well | -0.304 |
Looks | -0.304 |
Token | said |
Token ID | 531 |
Feature activation | +5.83e+00 |
Pos logit contributions | |
Don | +1.016 |
My | +0.939 |
Would | +0.938 |
Well | +0.903 |
Here | +0.873 |
Neg logit contributions | |
ies | -1.089 |
- | -1.015 |
ig | -0.930 |
ish | -0.920 |
an | -0.909 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
Don | +0.346 |
My | +0.335 |
Would | +0.329 |
Well | +0.317 |
It | +0.317 |
Neg logit contributions | |
ies | -0.113 |
- | -0.089 |
an | -0.088 |
eye | -0.080 |
ish | -0.080 |
Token | " |
Token ID | 366 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
- | +0.633 |
ies | +0.624 |
, | +0.543 |
. | +0.528 |
ig | +0.518 |
Neg logit contributions | |
Don | -0.553 |
My | -0.527 |
Here | -0.524 |
Looks | -0.513 |
Ah | -0.506 |
Token | My |
Token ID | 3666 |
Feature activation | +3.80e+01 |
Pos logit contributions | |
Don | +0.213 |
From | +0.190 |
Would | +0.184 |
Did | +0.181 |
My | +0.179 |
Neg logit contributions | |
ies | -0.641 |
- | -0.636 |
ater | -0.622 |
eye | -0.609 |
an | -0.599 |
Token | corn |
Token ID | 11676 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
My | +1.540 |
It | +1.528 |
Don | +1.521 |
It | +1.493 |
I | +1.469 |
Neg logit contributions | |
ies | -1.102 |
ishing | -0.970 |
ig | -0.961 |
- | -0.945 |
yt | -0.942 |
Token | needs |
Token ID | 2476 |
Feature activation | +7.26e+00 |
Pos logit contributions | |
ies | +0.522 |
- | +0.473 |
ishing | +0.463 |
ig | +0.462 |
an | +0.461 |
Neg logit contributions | |
Don | -0.408 |
My | -0.391 |
It | -0.378 |
I | -0.372 |
Well | -0.366 |
Token | water |
Token ID | 1660 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
Don | +0.308 |
My | +0.293 |
Would | +0.285 |
There | +0.281 |
It | +0.276 |
Neg logit contributions | |
ies | -0.273 |
an | -0.237 |
- | -0.233 |
ig | -0.231 |
ish | -0.231 |
Token | to |
Token ID | 284 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
ies | +0.412 |
decision | +0.404 |
ig | +0.393 |
an | +0.388 |
choice | +0.366 |
Neg logit contributions | |
Don | -0.556 |
There | -0.529 |
I | -0.526 |
My | -0.517 |
It | -0.514 |
Token | grow |
Token ID | 1663 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
ies | +0.035 |
- | +0.032 |
ig | +0.030 |
ish | +0.029 |
ishing | +0.029 |
Neg logit contributions | |
Don | -0.035 |
My | -0.035 |
Would | -0.033 |
It | -0.033 |
There | -0.032 |
Token | L |
Token ID | 406 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
ies | +0.832 |
- | +0.750 |
eye | +0.675 |
ig | +0.674 |
. | +0.658 |
Neg logit contributions | |
Don | -0.765 |
Would | -0.716 |
My | -0.682 |
Here | -0.673 |
Did | -0.654 |
Token | ila |
Token ID | 10102 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Don | +0.268 |
Would | +0.247 |
My | +0.246 |
Well | +0.233 |
Here | +0.232 |
Neg logit contributions | |
ies | -0.303 |
- | -0.282 |
ig | -0.263 |
an | -0.257 |
ish | -0.256 |
Token | says |
Token ID | 1139 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
Don | +0.485 |
My | +0.465 |
Would | +0.445 |
Well | +0.436 |
It | +0.416 |
Neg logit contributions | |
ies | -0.535 |
- | -0.485 |
ig | -0.458 |
an | -0.454 |
decision | -0.453 |
Token | . |
Token ID | 13 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
ies | +0.045 |
an | +0.044 |
island | +0.040 |
ish | +0.040 |
ig | +0.039 |
Neg logit contributions | |
Don | -0.096 |
My | -0.095 |
Would | -0.089 |
It | -0.089 |
It | -0.088 |
Token | " |
Token ID | 366 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
- | +0.685 |
ies | +0.679 |
, | +0.647 |
. | +0.646 |
eye | +0.601 |
Neg logit contributions | |
Don | -0.396 |
Here | -0.368 |
Looks | -0.364 |
Would | -0.356 |
There | -0.338 |
Token | My |
Token ID | 3666 |
Feature activation | +3.80e+01 |
Pos logit contributions | |
Don | +0.207 |
Did | +0.186 |
Would | +0.175 |
Here | +0.171 |
Everything | +0.171 |
Neg logit contributions | |
- | -0.698 |
ies | -0.651 |
an | -0.628 |
ater | -0.625 |
. | -0.599 |
Token | car |
Token ID | 1097 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
My | +1.533 |
It | +1.482 |
Don | +1.429 |
It | +1.404 |
I | +1.359 |
Neg logit contributions | |
ies | -1.224 |
ishing | -1.078 |
ie | -1.049 |
- | -1.039 |
ater | -1.003 |
Token | is |
Token ID | 318 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
ies | +0.482 |
ishing | +0.420 |
- | +0.419 |
ish | +0.409 |
ig | +0.408 |
Neg logit contributions | |
Don | -0.341 |
My | -0.337 |
It | -0.309 |
I | -0.302 |
Well | -0.301 |
Token | faster |
Token ID | 5443 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
Don | +0.323 |
My | +0.313 |
Would | +0.297 |
It | +0.292 |
There | +0.287 |
Neg logit contributions | |
ies | -0.327 |
- | -0.293 |
ish | -0.288 |
ig | -0.284 |
an | -0.282 |
Token | than |
Token ID | 621 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
ish | +0.137 |
ies | +0.135 |
decision | +0.131 |
ig | +0.127 |
ishing | +0.127 |
Neg logit contributions | |
Don | -0.167 |
It | -0.164 |
I | -0.159 |
My | -0.158 |
It | -0.155 |
Token | yours |
Token ID | 12431 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
ies | +0.682 |
- | +0.634 |
ig | +0.592 |
ish | +0.586 |
an | +0.582 |
Neg logit contributions | |
Don | -0.560 |
My | -0.519 |
Would | -0.513 |
Well | -0.492 |
There | -0.476 |
Token | L |
Token ID | 43 |
Feature activation | +1.67e+01 |
Pos logit contributions | |
Don | +0.516 |
Did | +0.515 |
Would | +0.480 |
My | +0.476 |
Well | +0.474 |
Neg logit contributions | |
ies | -0.880 |
- | -0.852 |
an | -0.768 |
ig | -0.766 |
fort | -0.730 |
Token | ily |
Token ID | 813 |
Feature activation | +1.79e+01 |
Pos logit contributions | |
Don | +0.507 |
My | +0.462 |
Would | +0.456 |
Well | +0.433 |
There | +0.415 |
Neg logit contributions | |
ies | -0.844 |
- | -0.792 |
ig | -0.746 |
ish | -0.740 |
an | -0.735 |
Token | said |
Token ID | 531 |
Feature activation | +5.40e+00 |
Pos logit contributions | |
Don | +0.770 |
My | +0.722 |
Would | +0.715 |
Well | +0.691 |
There | +0.671 |
Neg logit contributions | |
ies | -0.670 |
- | -0.614 |
ig | -0.566 |
ish | -0.559 |
an | -0.554 |
Token | , |
Token ID | 11 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
Don | +0.295 |
My | +0.290 |
Would | +0.283 |
It | +0.276 |
Did | +0.272 |
Neg logit contributions | |
ies | -0.119 |
an | -0.101 |
eye | -0.096 |
decision | -0.092 |
ishing | -0.091 |
Token | " |
Token ID | 366 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
- | +0.705 |
ies | +0.696 |
ig | +0.606 |
ish | +0.593 |
, | +0.593 |
Neg logit contributions | |
Don | -0.504 |
My | -0.460 |
Here | -0.456 |
Ah | -0.445 |
Well | -0.444 |
Token | My |
Token ID | 3666 |
Feature activation | +3.80e+01 |
Pos logit contributions | |
Everything | +0.182 |
Sadly | +0.157 |
Don | +0.155 |
Did | +0.155 |
Higher | +0.152 |
Neg logit contributions | |
ies | -0.605 |
an | -0.603 |
- | -0.602 |
ater | -0.560 |
hat | -0.543 |
Token | name |
Token ID | 1438 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
Don | +1.265 |
My | +1.229 |
Would | +1.179 |
It | +1.149 |
Something | +1.126 |
Neg logit contributions | |
ies | -1.513 |
- | -1.394 |
ig | -1.388 |
ishing | -1.361 |
island | -1.339 |
Token | is |
Token ID | 318 |
Feature activation | +7.97e+00 |
Pos logit contributions | |
ies | +0.573 |
- | +0.520 |
ig | +0.488 |
ish | +0.482 |
an | +0.477 |
Neg logit contributions | |
Don | -0.612 |
My | -0.579 |
Would | -0.567 |
Well | -0.547 |
There | -0.533 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.21e+01 |
Pos logit contributions | |
Don | +0.257 |
Would | +0.225 |
Well | +0.221 |
My | +0.219 |
There | +0.212 |
Neg logit contributions | |
ies | -0.389 |
- | -0.372 |
ig | -0.339 |
ish | -0.338 |
an | -0.336 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
My | +0.637 |
Don | +0.626 |
Would | +0.592 |
I | +0.583 |
It | +0.582 |
Neg logit contributions | |
ies | -0.276 |
decision | -0.248 |
ig | -0.237 |
ishing | -0.235 |
- | -0.230 |
Token | What |
Token ID | 1867 |
Feature activation | +1.64e+01 |
Pos logit contributions | |
ies | +0.685 |
- | +0.654 |
eye | +0.635 |
, | +0.622 |
. | +0.613 |
Neg logit contributions | |
Don | -0.439 |
Well | -0.358 |
Here | -0.353 |
There | -0.351 |
Would | -0.350 |
Token | � |
Token ID | 6184 |
Feature activation | +5.15e+00 |
Pos logit contributions | |
- | +0.594 |
, | +0.587 |
. | +0.570 |
ies | +0.568 |
! | +0.556 |
Neg logit contributions | |
Don | -0.437 |
My | -0.408 |
Here | -0.390 |
Our | -0.380 |
Take | -0.373 |
Token | � |
Token ID | 95 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
Don | +0.210 |
My | +0.194 |
Would | +0.194 |
Well | +0.185 |
Here | +0.179 |
Neg logit contributions | |
ies | -0.199 |
- | -0.193 |
ig | -0.179 |
an | -0.175 |
ish | -0.174 |
Token | € |
Token ID | 26391 |
Feature activation | +8.14e+00 |
Pos logit contributions | |
Don | +0.023 |
Would | +0.020 |
My | +0.020 |
Well | +0.019 |
There | +0.018 |
Neg logit contributions | |
- | -0.031 |
ies | -0.030 |
an | -0.029 |
ig | -0.028 |
eye | -0.028 |
Token | � |
Token ID | 129 |
Feature activation | +5.37e+00 |
Pos logit contributions | |
Don | +0.258 |
Would | +0.239 |
My | +0.236 |
Here | +0.222 |
Well | +0.221 |
Neg logit contributions | |
ies | -0.391 |
- | -0.383 |
ig | -0.349 |
an | -0.340 |
ish | -0.338 |
Token | � |
Token ID | 241 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
Don | +0.211 |
My | +0.197 |
Would | +0.195 |
Well | +0.189 |
There | +0.183 |
Neg logit contributions | |
ies | -0.219 |
- | -0.206 |
an | -0.189 |
ig | -0.189 |
ish | -0.187 |
Token | My |
Token ID | 3666 |
Feature activation | +3.80e+01 |
Pos logit contributions | |
Don | +0.007 |
Here | +0.006 |
Would | +0.006 |
My | +0.006 |
Grab | +0.006 |
Neg logit contributions | |
- | -0.019 |
ies | -0.018 |
. | -0.017 |
ater | -0.017 |
! | -0.017 |
Token | name |
Token ID | 1438 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
Don | +1.332 |
My | +1.314 |
Would | +1.225 |
It | +1.186 |
It | +1.183 |
Neg logit contributions | |
ies | -1.595 |
ig | -1.384 |
ishing | -1.372 |
- | -1.368 |
ie | -1.329 |
Token | is |
Token ID | 318 |
Feature activation | +8.32e+00 |
Pos logit contributions | |
ies | +0.629 |
- | +0.610 |
ig | +0.545 |
ish | +0.540 |
an | +0.539 |
Neg logit contributions | |
Don | -0.565 |
Would | -0.524 |
My | -0.513 |
Well | -0.498 |
There | -0.485 |
Token | Tim |
Token ID | 5045 |
Feature activation | +8.04e+00 |
Pos logit contributions | |
Don | +0.296 |
Would | +0.255 |
Here | +0.252 |
Well | +0.247 |
There | +0.244 |
Neg logit contributions | |
- | -0.364 |
ies | -0.354 |
ish | -0.310 |
ig | -0.309 |
an | -0.306 |
Token | ! |
Token ID | 0 |
Feature activation | -1.84e+01 |
Pos logit contributions | |
Don | +0.419 |
My | +0.410 |
Would | +0.390 |
It | +0.384 |
That | +0.379 |
Neg logit contributions | |
ies | -0.208 |
- | -0.168 |
ig | -0.165 |
decision | -0.162 |
eye | -0.161 |
Token | â |
Token ID | 22940 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
- | +0.765 |
. | +0.705 |
ies | +0.695 |
, | +0.684 |
!" | +0.681 |
Neg logit contributions | |
Don | -0.498 |
Over | -0.414 |
My | -0.412 |
Here | -0.410 |
Take | -0.398 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
Don | +0.027 |
My | +0.026 |
Would | +0.025 |
It | +0.025 |
Well | +0.025 |
Neg logit contributions | |
ies | -0.015 |
an | -0.014 |
eye | -0.012 |
ish | -0.012 |
- | -0.012 |
Token | Bob |
Token ID | 5811 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
ies | +0.674 |
- | +0.633 |
decision | +0.565 |
ish | +0.560 |
choice | +0.555 |
Neg logit contributions | |
Don | -0.570 |
Would | -0.550 |
My | -0.515 |
Here | -0.503 |
Well | -0.500 |
Token | said |
Token ID | 531 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
Don | +0.100 |
My | +0.093 |
Would | +0.091 |
Well | +0.087 |
There | +0.084 |
Neg logit contributions | |
ies | -0.133 |
- | -0.124 |
ig | -0.116 |
ish | -0.115 |
an | -0.115 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
Don | +0.331 |
My | +0.317 |
Would | +0.315 |
Well | +0.306 |
It | +0.301 |
Neg logit contributions | |
ies | -0.118 |
- | -0.094 |
an | -0.086 |
ig | -0.085 |
ish | -0.084 |
Token | " |
Token ID | 366 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
- | +0.662 |
ies | +0.636 |
, | +0.556 |
. | +0.549 |
ig | +0.524 |
Neg logit contributions | |
Don | -0.540 |
Here | -0.525 |
My | -0.507 |
There | -0.503 |
Would | -0.482 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Their | +0.217 |
Did | +0.206 |
Don | +0.188 |
Sadly | +0.187 |
Something | +0.183 |
Neg logit contributions | |
- | -0.573 |
est | -0.573 |
ies | -0.564 |
ater | -0.558 |
an | -0.550 |
Token | prints |
Token ID | 20842 |
Feature activation | -1.74e+01 |
Pos logit contributions | |
Don | +1.566 |
My | +1.530 |
It | +1.502 |
Something | +1.486 |
It | +1.469 |
Neg logit contributions | |
ies | -1.129 |
- | -0.992 |
ie | -0.981 |
ig | -0.963 |
island | -0.961 |
Token | look |
Token ID | 804 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
ies | +0.706 |
- | +0.641 |
ig | +0.612 |
an | +0.609 |
ish | +0.601 |
Neg logit contributions | |
Don | -0.668 |
My | -0.638 |
It | -0.599 |
Would | -0.597 |
Well | -0.597 |
Token | great |
Token ID | 1049 |
Feature activation | +6.03e+00 |
Pos logit contributions | |
ies | +0.186 |
ig | +0.167 |
an | +0.167 |
- | +0.164 |
ish | +0.160 |
Neg logit contributions | |
Don | -0.168 |
My | -0.160 |
Would | -0.154 |
It | -0.151 |
There | -0.147 |
Token | ! |
Token ID | 0 |
Feature activation | -1.84e+01 |
Pos logit contributions | |
Don | +0.258 |
My | +0.258 |
It | +0.255 |
I | +0.249 |
It | +0.246 |
Neg logit contributions | |
an | -0.162 |
ies | -0.156 |
decision | -0.140 |
ig | -0.139 |
eye | -0.136 |
Token | I |
Token ID | 314 |
Feature activation | +1.56e+01 |
Pos logit contributions | |
- | +0.690 |
ish | +0.645 |
ig | +0.627 |
eye | +0.624 |
ies | +0.615 |
Neg logit contributions | |
Don | -0.620 |
Would | -0.502 |
My | -0.501 |
There | -0.496 |
Here | -0.494 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.01e+01 |
Pos logit contributions | |
ies | +0.536 |
ishing | +0.470 |
ig | +0.465 |
ish | +0.457 |
an | +0.440 |
Neg logit contributions | |
Don | -0.720 |
My | -0.687 |
I | -0.681 |
It | -0.669 |
There | -0.651 |
Token | Anna |
Token ID | 11735 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
ies | +0.689 |
- | +0.684 |
mess | +0.619 |
. | +0.613 |
choice | +0.610 |
Neg logit contributions | |
Don | -0.790 |
Would | -0.757 |
My | -0.737 |
Did | -0.705 |
Here | -0.702 |
Token | says |
Token ID | 1139 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
Don | +0.530 |
My | +0.496 |
Would | +0.484 |
Well | +0.469 |
It | +0.450 |
Neg logit contributions | |
ies | -0.554 |
- | -0.505 |
ig | -0.477 |
an | -0.469 |
ish | -0.467 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
ies | +0.046 |
island | +0.045 |
an | +0.044 |
ig | +0.043 |
ish | +0.041 |
Neg logit contributions | |
Don | -0.086 |
My | -0.083 |
It | -0.078 |
She | -0.077 |
It | -0.076 |
Token | " |
Token ID | 366 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
- | +0.544 |
ies | +0.530 |
, | +0.526 |
. | +0.524 |
window | +0.502 |
Neg logit contributions | |
Don | -0.484 |
Looks | -0.483 |
Ah | -0.478 |
Here | -0.471 |
Would | -0.462 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Everything | +0.211 |
Did | +0.202 |
From | +0.190 |
Would | +0.186 |
Well | +0.165 |
Neg logit contributions | |
- | -0.671 |
ater | -0.662 |
an | -0.621 |
ies | -0.614 |
mess | -0.558 |
Token | doll |
Token ID | 3654 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
It | +1.475 |
My | +1.466 |
Don | +1.375 |
It | +1.369 |
Me | +1.358 |
Neg logit contributions | |
ies | -1.282 |
ishing | -1.145 |
ie | -1.087 |
yt | -1.053 |
ater | -1.034 |
Token | is |
Token ID | 318 |
Feature activation | +8.11e+00 |
Pos logit contributions | |
ies | +0.694 |
ishing | +0.609 |
an | +0.601 |
- | +0.601 |
eye | +0.597 |
Neg logit contributions | |
Don | -0.494 |
My | -0.488 |
It | -0.442 |
I | -0.434 |
There | -0.426 |
Token | having |
Token ID | 1719 |
Feature activation | +8.78e+00 |
Pos logit contributions | |
Don | +0.314 |
My | +0.302 |
It | +0.279 |
Would | +0.279 |
There | +0.275 |
Neg logit contributions | |
ies | -0.333 |
ig | -0.296 |
ish | -0.294 |
an | -0.292 |
- | -0.289 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
Don | +0.349 |
My | +0.327 |
Would | +0.320 |
Well | +0.308 |
There | +0.305 |
Neg logit contributions | |
ies | -0.359 |
- | -0.327 |
ig | -0.309 |
ish | -0.308 |
an | -0.307 |
Token | tea |
Token ID | 8887 |
Feature activation | -1.62e+01 |
Pos logit contributions | |
It | +0.460 |
Don | +0.451 |
My | +0.435 |
See | +0.432 |
Would | +0.425 |
Neg logit contributions | |
ies | -0.554 |
ishing | -0.467 |
an | -0.462 |
ie | -0.453 |
ish | -0.436 |
Token | smiled |
Token ID | 13541 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
ies | +0.481 |
- | +0.465 |
ig | +0.417 |
an | +0.415 |
ish | +0.414 |
Neg logit contributions | |
Don | -0.394 |
Would | -0.364 |
My | -0.361 |
Here | -0.346 |
There | -0.344 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
Don | +0.142 |
My | +0.134 |
Would | +0.131 |
It | +0.129 |
Well | +0.128 |
Neg logit contributions | |
ies | -0.068 |
ig | -0.056 |
- | -0.054 |
an | -0.054 |
ish | -0.051 |
Token | said |
Token ID | 531 |
Feature activation | +5.54e+00 |
Pos logit contributions | |
ies | +0.414 |
- | +0.399 |
ig | +0.341 |
ish | +0.341 |
eye | +0.331 |
Neg logit contributions | |
Don | -0.386 |
Would | -0.373 |
Here | -0.362 |
My | -0.361 |
There | -0.352 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
Don | +0.349 |
My | +0.333 |
Would | +0.332 |
Well | +0.325 |
There | +0.320 |
Neg logit contributions | |
ies | -0.097 |
- | -0.085 |
ig | -0.065 |
ish | -0.062 |
an | -0.059 |
Token | " |
Token ID | 366 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
- | +0.680 |
ies | +0.646 |
. | +0.573 |
, | +0.573 |
ig | +0.549 |
Neg logit contributions | |
Don | -0.498 |
Here | -0.487 |
My | -0.485 |
There | -0.465 |
Would | -0.459 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Did | +0.198 |
Their | +0.196 |
Everything | +0.179 |
Higher | +0.168 |
Something | +0.167 |
Neg logit contributions | |
- | -0.607 |
an | -0.596 |
ies | -0.593 |
ater | -0.577 |
est | -0.549 |
Token | friend |
Token ID | 1545 |
Feature activation | -2.21e+01 |
Pos logit contributions | |
Don | +1.472 |
My | +1.428 |
It | +1.363 |
Something | +1.332 |
Would | +1.315 |
Neg logit contributions | |
ies | -1.373 |
ig | -1.219 |
- | -1.193 |
ishing | -1.192 |
ie | -1.127 |
Token | is |
Token ID | 318 |
Feature activation | +8.32e+00 |
Pos logit contributions | |
ies | +0.875 |
- | +0.799 |
ig | +0.754 |
an | +0.738 |
ish | +0.736 |
Neg logit contributions | |
Don | -0.893 |
My | -0.839 |
Would | -0.819 |
Well | -0.793 |
It | -0.770 |
Token | just |
Token ID | 655 |
Feature activation | +8.03e-01 |
Pos logit contributions | |
Don | +0.337 |
My | +0.321 |
Would | +0.311 |
It | +0.297 |
Well | +0.293 |
Neg logit contributions | |
ies | -0.333 |
- | -0.296 |
ig | -0.287 |
an | -0.282 |
ish | -0.282 |
Token | hiding |
Token ID | 11816 |
Feature activation | -1.09e+01 |
Pos logit contributions | |
Don | +0.033 |
My | +0.031 |
Would | +0.030 |
Well | +0.028 |
It | +0.028 |
Neg logit contributions | |
ies | -0.031 |
- | -0.028 |
ig | -0.028 |
an | -0.027 |
ish | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
ies | +0.333 |
ig | +0.282 |
decision | +0.277 |
ish | +0.275 |
eye | +0.274 |
Neg logit contributions | |
Don | -0.499 |
My | -0.472 |
It | -0.454 |
There | -0.447 |
Would | -0.447 |
Token | big |
Token ID | 1263 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
Don | +1.379 |
My | +1.301 |
Would | +1.299 |
Well | +1.250 |
There | +1.236 |
Neg logit contributions | |
ies | -1.099 |
- | -0.969 |
ig | -0.960 |
ish | -0.922 |
ishing | -0.903 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.09e+01 |
Pos logit contributions | |
Don | +0.125 |
He | +0.120 |
The | +0.119 |
It | +0.117 |
Hi | +0.117 |
Neg logit contributions | |
ig | -0.202 |
ishing | -0.195 |
ies | -0.192 |
an | -0.187 |
ater | -0.186 |
Token | says |
Token ID | 1139 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
ies | +0.461 |
- | +0.424 |
ig | +0.396 |
ish | +0.390 |
an | +0.390 |
Neg logit contributions | |
Don | -0.413 |
My | -0.388 |
Would | -0.384 |
Well | -0.368 |
It | -0.355 |
Token | , |
Token ID | 11 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
ies | +0.050 |
an | +0.042 |
eye | +0.041 |
ishing | +0.041 |
ater | +0.040 |
Neg logit contributions | |
Don | -0.109 |
My | -0.107 |
Would | -0.106 |
It | -0.104 |
It | -0.103 |
Token | " |
Token ID | 366 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
- | +0.700 |
ies | +0.685 |
ig | +0.601 |
, | +0.587 |
castle | +0.582 |
Neg logit contributions | |
Don | -0.510 |
My | -0.480 |
Here | -0.461 |
There | -0.449 |
Ah | -0.446 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Everything | +0.195 |
Higher | +0.185 |
Finally | +0.178 |
Sadly | +0.176 |
Unfortunately | +0.168 |
Neg logit contributions | |
- | -0.603 |
an | -0.569 |
ies | -0.555 |
ater | -0.535 |
. | -0.527 |
Token | name |
Token ID | 1438 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
Don | +1.226 |
My | +1.204 |
Would | +1.137 |
It | +1.116 |
Something | +1.092 |
Neg logit contributions | |
ies | -1.620 |
ig | -1.520 |
ishing | -1.484 |
- | -1.457 |
ish | -1.417 |
Token | is |
Token ID | 318 |
Feature activation | +7.93e+00 |
Pos logit contributions | |
ies | +0.606 |
- | +0.537 |
ig | +0.518 |
ish | +0.513 |
an | +0.507 |
Neg logit contributions | |
Don | -0.583 |
My | -0.558 |
Would | -0.536 |
Well | -0.519 |
It | -0.514 |
Token | Ben |
Token ID | 3932 |
Feature activation | +7.87e+00 |
Pos logit contributions | |
Don | +0.259 |
Would | +0.227 |
Well | +0.223 |
My | +0.218 |
There | +0.214 |
Neg logit contributions | |
ies | -0.378 |
- | -0.369 |
ig | -0.328 |
ish | -0.327 |
an | -0.327 |
Token | ." |
Token ID | 526 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
My | +0.392 |
Don | +0.390 |
Would | +0.365 |
I | +0.363 |
It | +0.363 |
Neg logit contributions | |
ies | -0.192 |
ig | -0.173 |
ishing | -0.167 |
decision | -0.167 |
eye | -0.157 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e+01 |
Pos logit contributions | |
ies | +0.585 |
- | +0.578 |
eye | +0.545 |
. | +0.538 |
fort | +0.527 |
Neg logit contributions | |
Don | -0.308 |
My | -0.287 |
Would | -0.262 |
Well | -0.241 |
That | -0.241 |
Token | The |
Token ID | 383 |
Feature activation | +2.72e+01 |
Pos logit contributions | |
- | +0.692 |
ies | +0.639 |
. | +0.626 |
, | +0.623 |
eye | +0.610 |
Neg logit contributions | |
Don | -0.428 |
My | -0.386 |
Would | -0.378 |
Looks | -0.355 |
Well | -0.355 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
Don | +1.131 |
My | +1.047 |
Would | +1.021 |
It | +1.001 |
There | +0.995 |
Neg logit contributions | |
ies | -1.067 |
- | -0.971 |
ig | -0.953 |
ish | -0.897 |
an | -0.895 |
Token | says |
Token ID | 1139 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
ies | +0.431 |
- | +0.399 |
ish | +0.366 |
ig | +0.366 |
an | +0.359 |
Neg logit contributions | |
Don | -0.442 |
My | -0.410 |
Would | -0.408 |
Well | -0.392 |
There | -0.385 |
Token | , |
Token ID | 11 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
ies | +0.034 |
- | +0.029 |
ig | +0.023 |
ish | +0.022 |
an | +0.021 |
Neg logit contributions | |
Don | -0.118 |
My | -0.112 |
Would | -0.112 |
Well | -0.109 |
There | -0.107 |
Token | " |
Token ID | 366 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
- | +0.648 |
ies | +0.621 |
, | +0.569 |
. | +0.549 |
ig | +0.531 |
Neg logit contributions | |
Don | -0.535 |
My | -0.507 |
Here | -0.502 |
Well | -0.484 |
Looks | -0.484 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Don | +0.191 |
Something | +0.190 |
Everything | +0.188 |
Did | +0.178 |
From | +0.172 |
Neg logit contributions | |
- | -0.657 |
an | -0.610 |
ies | -0.607 |
ater | -0.592 |
ish | -0.586 |
Token | bike |
Token ID | 7161 |
Feature activation | -2.15e+01 |
Pos logit contributions | |
My | +1.643 |
Don | +1.613 |
It | +1.562 |
It | +1.562 |
I | +1.512 |
Neg logit contributions | |
ies | -1.192 |
ie | -1.054 |
an | -0.973 |
- | -0.965 |
ishing | -0.959 |
Token | is |
Token ID | 318 |
Feature activation | +8.12e+00 |
Pos logit contributions | |
ies | +0.939 |
- | +0.823 |
ig | +0.815 |
an | +0.808 |
ishing | +0.801 |
Neg logit contributions | |
Don | -0.756 |
My | -0.729 |
Would | -0.668 |
Well | -0.664 |
It | -0.662 |
Token | better |
Token ID | 1365 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
Don | +0.337 |
My | +0.327 |
Would | +0.307 |
It | +0.304 |
There | +0.301 |
Neg logit contributions | |
ies | -0.311 |
- | -0.271 |
an | -0.269 |
ish | -0.265 |
ig | -0.264 |
Token | than |
Token ID | 621 |
Feature activation | -1.55e+01 |
Pos logit contributions | |
Don | +0.028 |
My | +0.027 |
I | +0.026 |
It | +0.025 |
You | +0.025 |
Neg logit contributions | |
an | -0.022 |
ish | -0.021 |
ies | -0.021 |
ig | -0.021 |
decision | -0.019 |
Token | yours |
Token ID | 12431 |
Feature activation | -7.51e-02 |
Pos logit contributions | |
ies | +0.661 |
- | +0.607 |
an | +0.577 |
ig | +0.576 |
ish | +0.571 |
Neg logit contributions | |
Don | -0.583 |
My | -0.552 |
Would | -0.536 |
Well | -0.512 |
There | -0.508 |
Token | , |
Token ID | 11 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
Don | +0.279 |
My | +0.274 |
She | +0.256 |
It | +0.256 |
It | +0.248 |
Neg logit contributions | |
ies | -0.143 |
an | -0.143 |
island | -0.138 |
ig | -0.133 |
fort | -0.126 |
Token | feeling |
Token ID | 4203 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
ies | +0.596 |
- | +0.542 |
ig | +0.510 |
ish | +0.501 |
an | +0.496 |
Neg logit contributions | |
Don | -0.682 |
My | -0.640 |
Would | -0.632 |
Well | -0.608 |
There | -0.590 |
Token | angry |
Token ID | 7954 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ies | +0.088 |
- | +0.084 |
eye | +0.072 |
ish | +0.072 |
ig | +0.070 |
Neg logit contributions | |
Don | -0.123 |
Would | -0.117 |
My | -0.117 |
Well | -0.115 |
There | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
Don | +0.073 |
Why | +0.068 |
She | +0.067 |
My | +0.066 |
It | +0.066 |
Neg logit contributions | |
ig | -0.047 |
ies | -0.045 |
an | -0.045 |
island | -0.043 |
ish | -0.043 |
Token | " |
Token ID | 366 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
- | +0.533 |
ies | +0.522 |
, | +0.510 |
. | +0.494 |
window | +0.464 |
Neg logit contributions | |
My | -0.505 |
Here | -0.503 |
Ah | -0.503 |
Don | -0.501 |
Take | -0.495 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Everything | +0.215 |
From | +0.209 |
Did | +0.209 |
Would | +0.180 |
Finally | +0.166 |
Neg logit contributions | |
ater | -0.641 |
- | -0.620 |
ies | -0.595 |
an | -0.586 |
times | -0.534 |
Token | castle |
Token ID | 16669 |
Feature activation | -2.42e+01 |
Pos logit contributions | |
It | +1.591 |
It | +1.558 |
My | +1.552 |
Something | +1.545 |
I | +1.521 |
Neg logit contributions | |
ies | -1.011 |
ishing | -0.885 |
- | -0.874 |
ie | -0.869 |
ig | -0.837 |
Token | has |
Token ID | 468 |
Feature activation | +1.52e+01 |
Pos logit contributions | |
ies | +1.076 |
- | +0.961 |
ishing | +0.951 |
eye | +0.940 |
ig | +0.940 |
Neg logit contributions | |
Don | -0.758 |
My | -0.701 |
It | -0.674 |
I | -0.672 |
Well | -0.665 |
Token | a |
Token ID | 257 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
Don | +0.565 |
My | +0.535 |
Would | +0.509 |
There | +0.499 |
Well | +0.492 |
Neg logit contributions | |
ies | -0.645 |
- | -0.589 |
ig | -0.564 |
ish | -0.558 |
an | -0.554 |
Token | mo |
Token ID | 6941 |
Feature activation | -1.99e+01 |
Pos logit contributions | |
Don | +0.537 |
It | +0.533 |
See | +0.516 |
Ah | +0.504 |
My | +0.504 |
Neg logit contributions | |
ies | -0.451 |
ishing | -0.391 |
ie | -0.368 |
an | -0.361 |
- | -0.361 |
Token | at |
Token ID | 265 |
Feature activation | -7.15e+00 |
Pos logit contributions | |
ies | +0.630 |
- | +0.528 |
decision | +0.516 |
fort | +0.514 |
choice | +0.510 |
Neg logit contributions | |
Don | -0.925 |
My | -0.891 |
Well | -0.867 |
Would | -0.861 |
There | -0.852 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.04e+01 |
Pos logit contributions | |
It | +0.021 |
Don | +0.020 |
That | +0.020 |
There | +0.020 |
I | +0.020 |
Neg logit contributions | |
eye | -0.015 |
outfit | -0.015 |
ishing | -0.015 |
an | -0.014 |
island | -0.014 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
ies | +0.796 |
- | +0.731 |
, | +0.694 |
. | +0.687 |
eye | +0.673 |
Neg logit contributions | |
Don | -0.753 |
Would | -0.749 |
Well | -0.691 |
My | -0.685 |
That | -0.682 |
Token | cries |
Token ID | 24691 |
Feature activation | -8.29e+00 |
Pos logit contributions | |
Don | +0.459 |
My | +0.426 |
Would | +0.399 |
Well | +0.394 |
It | +0.393 |
Neg logit contributions | |
ies | -0.496 |
- | -0.455 |
ig | -0.446 |
ishing | -0.440 |
ish | -0.433 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
ies | +0.203 |
decision | +0.193 |
an | +0.192 |
ig | +0.192 |
collection | +0.191 |
Neg logit contributions | |
Don | -0.434 |
My | -0.404 |
It | -0.395 |
She | -0.394 |
Come | -0.392 |
Token | " |
Token ID | 366 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
- | +0.622 |
ies | +0.605 |
. | +0.582 |
, | +0.572 |
eye | +0.512 |
Neg logit contributions | |
Looks | -0.502 |
Here | -0.496 |
Would | -0.485 |
Don | -0.480 |
My | -0.463 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Would | +0.205 |
Everything | +0.166 |
Did | +0.166 |
From | +0.158 |
Would | +0.156 |
Neg logit contributions | |
ies | -0.674 |
- | -0.668 |
ater | -0.655 |
an | -0.630 |
mess | -0.622 |
Token | knee |
Token ID | 10329 |
Feature activation | -2.54e+01 |
Pos logit contributions | |
My | +1.675 |
Don | +1.633 |
It | +1.628 |
I | +1.594 |
There | +1.503 |
Neg logit contributions | |
ies | -1.068 |
ishing | -0.961 |
ie | -0.952 |
- | -0.916 |
an | -0.915 |
Token | hurts |
Token ID | 20406 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ies | +1.053 |
island | +0.958 |
ishing | +0.949 |
ig | +0.936 |
- | +0.906 |
Neg logit contributions | |
Don | -0.928 |
My | -0.846 |
It | -0.842 |
I | -0.822 |
That | -0.801 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.04e+01 |
Pos logit contributions | |
Don | +0.095 |
It | +0.093 |
I | +0.089 |
My | +0.089 |
That | +0.089 |
Neg logit contributions | |
ies | -0.046 |
ig | -0.041 |
ishing | -0.040 |
decision | -0.038 |
ish | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +1.79e+01 |
Pos logit contributions | |
- | +0.932 |
. | +0.927 |
ies | +0.919 |
, | +0.897 |
eye | +0.840 |
Neg logit contributions | |
Would | -0.539 |
Don | -0.522 |
Looks | -0.500 |
My | -0.490 |
Well | -0.487 |
Token |
|
Token ID | 198 |
Feature activation | +1.78e+01 |
Pos logit contributions | |
Don | +0.458 |
Would | +0.434 |
Well | +0.403 |
My | +0.402 |
Looks | +0.387 |
Neg logit contributions | |
- | -0.934 |
ies | -0.926 |
ig | -0.837 |
. | -0.833 |
an | -0.811 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e+01 |
Pos logit contributions | |
ies | +0.433 |
an | +0.376 |
ishing | +0.361 |
eye | +0.357 |
ig | +0.351 |
Neg logit contributions | |
My | -0.812 |
Don | -0.812 |
It | -0.753 |
Would | -0.752 |
That | -0.730 |
Token | She |
Token ID | 1375 |
Feature activation | +2.60e+01 |
Pos logit contributions | |
- | +0.634 |
ies | +0.622 |
eye | +0.615 |
ish | +0.615 |
. | +0.607 |
Neg logit contributions | |
Don | -0.396 |
Ah | -0.348 |
Listen | -0.346 |
Did | -0.344 |
My | -0.340 |
Token | said |
Token ID | 531 |
Feature activation | +5.66e+00 |
Pos logit contributions | |
Don | +0.893 |
My | +0.824 |
Would | +0.815 |
Well | +0.776 |
There | +0.755 |
Neg logit contributions | |
ies | -1.205 |
- | -1.127 |
ig | -1.053 |
ish | -1.041 |
an | -1.033 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
Don | +0.336 |
My | +0.321 |
Would | +0.319 |
Well | +0.311 |
There | +0.305 |
Neg logit contributions | |
ies | -0.120 |
- | -0.101 |
ig | -0.087 |
ish | -0.085 |
an | -0.084 |
Token | " |
Token ID | 366 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
- | +0.677 |
ies | +0.654 |
ig | +0.564 |
, | +0.557 |
ish | +0.551 |
Neg logit contributions | |
Don | -0.517 |
My | -0.486 |
Here | -0.477 |
There | -0.470 |
Looks | -0.460 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Did | +0.209 |
Something | +0.201 |
Their | +0.193 |
Don | +0.187 |
Everything | +0.174 |
Neg logit contributions | |
- | -0.633 |
ies | -0.619 |
ater | -0.589 |
an | -0.582 |
ie | -0.573 |
Token | house |
Token ID | 2156 |
Feature activation | -2.55e+01 |
Pos logit contributions | |
Don | +1.574 |
My | +1.472 |
Would | +1.437 |
It | +1.423 |
It | +1.411 |
Neg logit contributions | |
ies | -1.307 |
ishing | -1.111 |
ie | -1.097 |
ig | -1.095 |
- | -1.076 |
Token | will |
Token ID | 481 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
ies | +1.079 |
- | +0.971 |
ig | +0.945 |
an | +0.934 |
ish | +0.926 |
Neg logit contributions | |
Don | -0.929 |
My | -0.873 |
Would | -0.824 |
Well | -0.821 |
It | -0.814 |
Token | be |
Token ID | 307 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
Don | +0.098 |
My | +0.091 |
Would | +0.090 |
Well | +0.087 |
There | +0.085 |
Neg logit contributions | |
ies | -0.096 |
- | -0.089 |
ig | -0.083 |
ish | -0.082 |
an | -0.081 |
Token | the |
Token ID | 262 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
Don | +0.146 |
My | +0.134 |
Would | +0.132 |
There | +0.130 |
It | +0.127 |
Neg logit contributions | |
ies | -0.142 |
- | -0.129 |
an | -0.128 |
ig | -0.127 |
ish | -0.126 |
Token | perfect |
Token ID | 2818 |
Feature activation | +7.74e+00 |
Pos logit contributions | |
Don | +0.471 |
My | +0.437 |
Would | +0.428 |
There | +0.418 |
Here | +0.417 |
Neg logit contributions | |
ies | -0.417 |
- | -0.369 |
ig | -0.357 |
ishing | -0.338 |
ish | -0.327 |
Token | bird |
Token ID | 6512 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
Don | +0.475 |
Would | +0.463 |
My | +0.459 |
It | +0.444 |
Here | +0.435 |
Neg logit contributions | |
ies | -0.315 |
ig | -0.298 |
ishing | -0.291 |
- | -0.289 |
ish | -0.280 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
ies | +0.447 |
ishing | +0.435 |
ig | +0.421 |
ater | +0.397 |
- | +0.392 |
Neg logit contributions | |
My | -0.952 |
Don | -0.931 |
Would | -0.920 |
It | -0.847 |
It | -0.845 |
Token |
|
Token ID | 198 |
Feature activation | +1.79e+01 |
Pos logit contributions | |
. | +0.621 |
- | +0.607 |
, | +0.604 |
ies | +0.597 |
eye | +0.565 |
Neg logit contributions | |
Don | -0.458 |
Looks | -0.447 |
Well | -0.424 |
My | -0.423 |
Here | -0.422 |
Token |
|
Token ID | 198 |
Feature activation | +1.79e+01 |
Pos logit contributions | |
Don | +0.544 |
My | +0.479 |
Looks | +0.467 |
Well | +0.467 |
Did | +0.454 |
Neg logit contributions | |
- | -0.848 |
ies | -0.840 |
. | -0.767 |
an | -0.748 |
, | -0.726 |
Token | " |
Token ID | 1 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
Don | +0.714 |
My | +0.680 |
Well | +0.673 |
Would | +0.671 |
Did | +0.659 |
Neg logit contributions | |
ies | -0.671 |
- | -0.650 |
an | -0.571 |
ater | -0.540 |
ig | -0.522 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Everything | +0.090 |
From | +0.090 |
Sadly | +0.086 |
Unfortunately | +0.086 |
Eventually | +0.084 |
Neg logit contributions | |
- | -0.165 |
ies | -0.162 |
an | -0.151 |
song | -0.148 |
hat | -0.145 |
Token | name |
Token ID | 1438 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
My | +1.168 |
Don | +1.125 |
It | +1.061 |
Would | +1.051 |
It | +1.029 |
Neg logit contributions | |
ies | -1.592 |
ishing | -1.537 |
ig | -1.469 |
- | -1.438 |
ie | -1.396 |
Token | is |
Token ID | 318 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
ies | +0.617 |
- | +0.545 |
ig | +0.532 |
ish | +0.524 |
ishing | +0.521 |
Neg logit contributions | |
Don | -0.562 |
My | -0.540 |
Would | -0.512 |
It | -0.499 |
Well | -0.496 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.21e+01 |
Pos logit contributions | |
Don | +0.265 |
Would | +0.233 |
Well | +0.230 |
My | +0.230 |
Here | +0.221 |
Neg logit contributions | |
ies | -0.369 |
- | -0.354 |
an | -0.320 |
ish | -0.318 |
ig | -0.318 |
Token | ," |
Token ID | 553 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
My | +0.534 |
Don | +0.526 |
She | +0.505 |
It | +0.496 |
It | +0.494 |
Neg logit contributions | |
ishing | -0.319 |
ig | -0.311 |
ies | -0.309 |
decision | -0.308 |
ish | -0.278 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.21e+01 |
Pos logit contributions | |
ies | +0.581 |
- | +0.519 |
eye | +0.483 |
ish | +0.471 |
ig | +0.467 |
Neg logit contributions | |
Don | -0.497 |
Would | -0.449 |
My | -0.448 |
Well | -0.445 |
Even | -0.424 |
Token | name |
Token ID | 1438 |
Feature activation | -1.49e+01 |
Pos logit contributions | |
Don | +0.392 |
My | +0.384 |
Would | +0.372 |
It | +0.351 |
It | +0.349 |
Neg logit contributions | |
ies | -0.463 |
- | -0.413 |
ish | -0.401 |
ishing | -0.400 |
an | -0.391 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
ies | +0.460 |
- | +0.389 |
ig | +0.374 |
ish | +0.366 |
an | +0.365 |
Neg logit contributions | |
Don | -0.734 |
My | -0.710 |
Would | -0.694 |
Well | -0.667 |
It | -0.665 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
- | +0.499 |
ies | +0.477 |
. | +0.465 |
, | +0.460 |
eye | +0.429 |
Neg logit contributions | |
Don | -0.277 |
Looks | -0.275 |
Ah | -0.253 |
Would | -0.245 |
Take | -0.243 |
Token |
|
Token ID | 198 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
Don | +0.551 |
Looks | +0.519 |
Would | +0.486 |
Well | +0.479 |
Did | +0.467 |
Neg logit contributions | |
- | -0.909 |
ies | -0.861 |
. | -0.766 |
ig | -0.762 |
an | -0.755 |
Token | " |
Token ID | 1 |
Feature activation | +4.21e+00 |
Pos logit contributions | |
Don | +0.469 |
Did | +0.464 |
Would | +0.453 |
Well | +0.442 |
Looks | +0.408 |
Neg logit contributions | |
- | -0.905 |
ies | -0.886 |
an | -0.800 |
ater | -0.755 |
ig | -0.751 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Everything | +0.079 |
Sadly | +0.077 |
Eventually | +0.071 |
Unfortunately | +0.069 |
Did | +0.069 |
Neg logit contributions | |
- | -0.174 |
ies | -0.168 |
song | -0.160 |
game | -0.159 |
is | -0.158 |
Token | name |
Token ID | 1438 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
Don | +1.268 |
My | +1.251 |
Would | +1.165 |
It | +1.153 |
It | +1.142 |
Neg logit contributions | |
ies | -1.589 |
ishing | -1.445 |
- | -1.445 |
ig | -1.425 |
ish | -1.423 |
Token | is |
Token ID | 318 |
Feature activation | +8.23e+00 |
Pos logit contributions | |
ies | +0.582 |
- | +0.530 |
ig | +0.497 |
ish | +0.491 |
an | +0.486 |
Neg logit contributions | |
Don | -0.611 |
My | -0.575 |
Would | -0.564 |
Well | -0.546 |
There | -0.531 |
Token | Jack |
Token ID | 3619 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
Don | +0.279 |
Would | +0.240 |
Well | +0.236 |
Here | +0.230 |
My | +0.227 |
Neg logit contributions | |
ies | -0.384 |
- | -0.375 |
ig | -0.334 |
ish | -0.331 |
ie | -0.327 |
Token | ," |
Token ID | 553 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
My | +0.190 |
Don | +0.188 |
It | +0.176 |
She | +0.175 |
Would | +0.174 |
Neg logit contributions | |
ig | -0.121 |
decision | -0.117 |
eye | -0.116 |
ies | -0.114 |
ishing | -0.114 |
Token | the |
Token ID | 262 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
ies | +0.490 |
- | +0.466 |
ie | +0.422 |
, | +0.394 |
choice | +0.394 |
Neg logit contributions | |
Don | -0.518 |
Would | -0.486 |
My | -0.471 |
There | -0.466 |
Come | -0.464 |
Token | The |
Token ID | 464 |
Feature activation | +3.04e+01 |
Pos logit contributions | |
Don | +0.370 |
Did | +0.347 |
Would | +0.329 |
Well | +0.310 |
Looks | +0.294 |
Neg logit contributions | |
- | -0.980 |
ies | -0.901 |
ater | -0.884 |
an | -0.849 |
mess | -0.825 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
Don | +1.307 |
My | +1.256 |
Would | +1.234 |
I | +1.174 |
Well | +1.169 |
Neg logit contributions | |
ies | -1.149 |
ig | -0.994 |
- | -0.988 |
ish | -0.963 |
an | -0.941 |
Token | replied |
Token ID | 8712 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ies | +0.456 |
- | +0.444 |
ish | +0.389 |
ig | +0.387 |
eye | +0.381 |
Neg logit contributions | |
Don | -0.424 |
Would | -0.392 |
My | -0.381 |
There | -0.380 |
Here | -0.380 |
Token | : |
Token ID | 25 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
ies | +0.061 |
- | +0.053 |
ig | +0.052 |
ish | +0.051 |
an | +0.050 |
Neg logit contributions | |
Don | -0.073 |
My | -0.069 |
Would | -0.068 |
Well | -0.066 |
It | -0.064 |
Token | " |
Token ID | 366 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
- | +0.025 |
ies | +0.023 |
, | +0.022 |
. | +0.022 |
castle | +0.021 |
Neg logit contributions | |
Don | -0.022 |
There | -0.020 |
Well | -0.020 |
Here | -0.019 |
My | -0.019 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Their | +0.192 |
Everything | +0.189 |
Don | +0.173 |
Sadly | +0.166 |
From | +0.162 |
Neg logit contributions | |
- | -0.618 |
ater | -0.604 |
ies | -0.582 |
an | -0.569 |
ish | -0.544 |
Token | mum |
Token ID | 25682 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
My | +1.815 |
Don | +1.760 |
Would | +1.676 |
It | +1.672 |
I | +1.657 |
Neg logit contributions | |
ies | -1.097 |
an | -0.900 |
- | -0.896 |
ie | -0.880 |
island | -0.868 |
Token | gave |
Token ID | 2921 |
Feature activation | +6.00e+00 |
Pos logit contributions | |
ies | +0.555 |
- | +0.489 |
an | +0.479 |
ig | +0.474 |
fort | +0.467 |
Neg logit contributions | |
Don | -0.672 |
My | -0.654 |
Would | -0.624 |
Well | -0.621 |
It | -0.610 |
Token | it |
Token ID | 340 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
Don | +0.238 |
Would | +0.221 |
My | +0.220 |
Well | +0.213 |
There | +0.205 |
Neg logit contributions | |
ies | -0.246 |
- | -0.230 |
ig | -0.211 |
ish | -0.210 |
an | -0.207 |
Token | to |
Token ID | 284 |
Feature activation | -9.75e-01 |
Pos logit contributions | |
Don | +0.043 |
My | +0.042 |
Would | +0.040 |
It | +0.039 |
There | +0.039 |
Neg logit contributions | |
ies | -0.041 |
ig | -0.036 |
- | -0.035 |
an | -0.035 |
ish | -0.034 |
Token | me |
Token ID | 502 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
ies | +0.036 |
- | +0.033 |
ig | +0.030 |
ish | +0.030 |
an | +0.029 |
Neg logit contributions | |
Don | -0.043 |
My | -0.040 |
Would | -0.040 |
Well | -0.038 |
There | -0.037 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.03e+01 |
Pos logit contributions | |
Don | +0.323 |
It | +0.322 |
That | +0.313 |
There | +0.309 |
My | +0.305 |
Neg logit contributions | |
ies | -0.145 |
eye | -0.133 |
an | -0.127 |
decision | -0.123 |
ishing | -0.123 |
Token | Anna |
Token ID | 11735 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
ies | +0.746 |
- | +0.738 |
, | +0.720 |
. | +0.711 |
eye | +0.655 |
Neg logit contributions | |
Would | -0.763 |
Don | -0.751 |
Great | -0.677 |
Looks | -0.673 |
Here | -0.671 |
Token | cried |
Token ID | 16896 |
Feature activation | +5.62e+00 |
Pos logit contributions | |
Don | +0.426 |
My | +0.390 |
Would | +0.382 |
Well | +0.365 |
There | +0.351 |
Neg logit contributions | |
ies | -0.640 |
- | -0.596 |
ig | -0.566 |
ish | -0.560 |
an | -0.556 |
Token | . |
Token ID | 13 |
Feature activation | -1.53e+01 |
Pos logit contributions | |
Don | +0.307 |
My | +0.292 |
It | +0.278 |
She | +0.277 |
Would | +0.276 |
Neg logit contributions | |
ies | -0.132 |
an | -0.123 |
ig | -0.121 |
island | -0.115 |
decision | -0.115 |
Token | " |
Token ID | 366 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
- | +0.618 |
, | +0.607 |
. | +0.603 |
ies | +0.588 |
window | +0.533 |
Neg logit contributions | |
Looks | -0.498 |
Here | -0.469 |
Would | -0.459 |
Don | -0.446 |
Take | -0.446 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Would | +0.215 |
Everything | +0.174 |
Did | +0.172 |
Would | +0.165 |
Finally | +0.161 |
Neg logit contributions | |
ies | -0.656 |
- | -0.649 |
ater | -0.634 |
mess | -0.628 |
an | -0.612 |
Token | eye |
Token ID | 4151 |
Feature activation | -2.79e+01 |
Pos logit contributions | |
Don | +1.763 |
My | +1.747 |
It | +1.675 |
I | +1.637 |
Well | +1.600 |
Neg logit contributions | |
ies | -0.961 |
ishing | -0.891 |
ish | -0.828 |
- | -0.771 |
ie | -0.751 |
Token | ! |
Token ID | 0 |
Feature activation | -1.86e+01 |
Pos logit contributions | |
ies | +1.119 |
ishing | +0.947 |
- | +0.940 |
island | +0.936 |
decision | +0.934 |
Neg logit contributions | |
Don | -1.069 |
My | -1.049 |
It | -0.971 |
I | -0.946 |
That | -0.928 |
Token | My |
Token ID | 2011 |
Feature activation | +2.55e+01 |
Pos logit contributions | |
- | +0.861 |
eye | +0.794 |
ies | +0.790 |
knee | +0.751 |
ish | +0.744 |
Neg logit contributions | |
Would | -0.495 |
Don | -0.486 |
Looks | -0.445 |
Well | -0.441 |
Here | -0.419 |
Token | eye |
Token ID | 4151 |
Feature activation | -2.79e+01 |
Pos logit contributions | |
Don | +1.251 |
My | +1.220 |
It | +1.172 |
I | +1.133 |
Ah | +1.123 |
Neg logit contributions | |
ies | -0.582 |
ie | -0.500 |
ishing | -0.499 |
ish | -0.483 |
island | -0.460 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.03e+01 |
Pos logit contributions | |
ies | +1.142 |
- | +0.968 |
ishing | +0.959 |
decision | +0.953 |
island | +0.944 |
Neg logit contributions | |
Don | -1.060 |
My | -1.037 |
It | -0.958 |
I | -0.928 |
Well | -0.915 |
Token | L |
Token ID | 43 |
Feature activation | +1.66e+01 |
Pos logit contributions | |
Don | +0.564 |
Would | +0.543 |
Well | +0.525 |
My | +0.524 |
Here | +0.509 |
Neg logit contributions | |
- | -0.889 |
ies | -0.875 |
an | -0.745 |
ig | -0.718 |
ie | -0.717 |
Token | ily |
Token ID | 813 |
Feature activation | +1.77e+01 |
Pos logit contributions | |
Don | +0.521 |
My | +0.475 |
Would | +0.469 |
Well | +0.446 |
There | +0.429 |
Neg logit contributions | |
ies | -0.828 |
- | -0.779 |
ig | -0.731 |
ish | -0.724 |
an | -0.721 |
Token | replied |
Token ID | 8712 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Don | +0.674 |
My | +0.625 |
Would | +0.624 |
Well | +0.592 |
Here | +0.588 |
Neg logit contributions | |
ies | -0.762 |
- | -0.713 |
ish | -0.651 |
ig | -0.650 |
an | -0.643 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
ies | +0.033 |
an | +0.028 |
eye | +0.027 |
ig | +0.025 |
island | +0.025 |
Neg logit contributions | |
Don | -0.086 |
My | -0.083 |
Would | -0.080 |
It | -0.079 |
That | -0.078 |
Token | " |
Token ID | 366 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
ies | +0.748 |
- | +0.741 |
ish | +0.639 |
ig | +0.625 |
an | +0.617 |
Neg logit contributions | |
Don | -0.478 |
My | -0.435 |
Would | -0.431 |
Here | -0.429 |
Well | -0.428 |
Token | My |
Token ID | 3666 |
Feature activation | +3.79e+01 |
Pos logit contributions | |
Would | +0.200 |
Everything | +0.190 |
From | +0.152 |
known | +0.151 |
Well | +0.145 |
Neg logit contributions | |
- | -0.616 |
ies | -0.603 |
an | -0.578 |
ater | -0.547 |
ish | -0.543 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
Don | +1.502 |
My | +1.501 |
It | +1.456 |
Me | +1.426 |
It | +1.413 |
Neg logit contributions | |
ies | -1.281 |
ishing | -1.149 |
ish | -1.092 |
- | -1.086 |
ig | -1.069 |
Token | my |
Token ID | 1820 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
ies | +0.553 |
- | +0.507 |
an | +0.499 |
ig | +0.498 |
decision | +0.486 |
Neg logit contributions | |
Don | -0.482 |
My | -0.459 |
Would | -0.432 |
Well | -0.431 |
It | -0.426 |
Token | gave |
Token ID | 2921 |
Feature activation | +6.16e+00 |
Pos logit contributions | |
ies | +0.151 |
- | +0.138 |
ig | +0.136 |
an | +0.134 |
fort | +0.130 |
Neg logit contributions | |
Don | -0.121 |
My | -0.114 |
Would | -0.106 |
Well | -0.106 |
It | -0.105 |
Token | it |
Token ID | 340 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
Don | +0.223 |
Would | +0.206 |
My | +0.204 |
Well | +0.198 |
There | +0.188 |
Neg logit contributions | |
ies | -0.277 |
- | -0.261 |
ish | -0.240 |
ig | -0.239 |
an | -0.235 |
Token | to |
Token ID | 284 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
Don | +0.042 |
My | +0.041 |
There | +0.038 |
It | +0.037 |
Would | +0.037 |
Neg logit contributions | |
ies | -0.045 |
ig | -0.041 |
decision | -0.039 |
an | -0.039 |
ish | -0.038 |
Token | Tim |
Token ID | 5045 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
Don | +0.044 |
My | +0.041 |
Would | +0.041 |
Well | +0.039 |
Here | +0.038 |
Neg logit contributions | |
ies | -0.049 |
- | -0.045 |
ig | -0.042 |
ish | -0.041 |
an | -0.041 |
Token | my |
Token ID | 1820 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
Don | +0.016 |
My | +0.014 |
Would | +0.014 |
Well | +0.014 |
I | +0.013 |
Neg logit contributions | |
ies | -0.018 |
- | -0.016 |
ig | -0.016 |
an | -0.015 |
ishing | -0.015 |
Token | took |
Token ID | 1718 |
Feature activation | +5.37e+00 |
Pos logit contributions | |
ies | +0.508 |
- | +0.470 |
ig | +0.455 |
an | +0.446 |
ish | +0.442 |
Neg logit contributions | |
Don | -0.338 |
My | -0.310 |
Would | -0.299 |
Well | -0.290 |
There | -0.274 |
Token | a |
Token ID | 257 |
Feature activation | +6.59e+00 |
Pos logit contributions | |
Don | +0.183 |
My | +0.168 |
Would | +0.166 |
Well | +0.159 |
There | +0.153 |
Neg logit contributions | |
ies | -0.249 |
- | -0.233 |
ig | -0.218 |
ish | -0.216 |
an | -0.214 |
Token | deep |
Token ID | 2769 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
Don | +0.296 |
My | +0.283 |
Would | +0.274 |
It | +0.273 |
There | +0.273 |
Neg logit contributions | |
ies | -0.220 |
an | -0.186 |
ishing | -0.182 |
- | -0.177 |
ig | -0.168 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.81e+01 |
Pos logit contributions | |
Don | +0.121 |
There | +0.116 |
It | +0.114 |
It | +0.114 |
Be | +0.111 |
Neg logit contributions | |
ies | -0.058 |
decision | -0.049 |
ys | -0.048 |
uddle | -0.042 |
- | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
ies | +1.170 |
decision | +0.969 |
an | +0.962 |
choice | +0.951 |
ish | +0.925 |
Neg logit contributions | |
Don | -1.944 |
My | -1.841 |
It | -1.763 |
Well | -1.749 |
There | -1.738 |
Token | opened |
Token ID | 4721 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
ies | +0.669 |
- | +0.630 |
ig | +0.570 |
eye | +0.559 |
ish | +0.559 |
Neg logit contributions | |
Don | -0.590 |
Would | -0.561 |
My | -0.550 |
Well | -0.538 |
Here | -0.521 |
Token | the |
Token ID | 262 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
Don | +0.140 |
My | +0.130 |
Would | +0.126 |
There | +0.119 |
Well | +0.119 |
Neg logit contributions | |
ies | -0.218 |
- | -0.201 |
ig | -0.192 |
an | -0.191 |
ish | -0.190 |
Token | rocket |
Token ID | 10701 |
Feature activation | -1.98e+01 |
Pos logit contributions | |
Don | +0.260 |
It | +0.251 |
My | +0.250 |
Here | +0.246 |
Would | +0.242 |
Neg logit contributions | |
ies | -0.163 |
- | -0.132 |
ishing | -0.128 |
ie | -0.118 |
ig | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e+01 |
Pos logit contributions | |
ies | +0.613 |
choice | +0.492 |
ishing | +0.484 |
ig | +0.483 |
- | +0.475 |
Neg logit contributions | |
My | -0.880 |
It | -0.858 |
Don | -0.853 |
It | -0.838 |
There | -0.822 |
Token | she |
Token ID | 673 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
ies | +0.459 |
- | +0.431 |
ig | +0.408 |
ish | +0.405 |
an | +0.405 |
Neg logit contributions | |
Don | -0.266 |
My | -0.242 |
Would | -0.237 |
Well | -0.226 |
There | -0.218 |
Token | was |
Token ID | 373 |
Feature activation | +9.48e+00 |
Pos logit contributions | |
Don | +0.368 |
My | +0.339 |
Would | +0.335 |
Well | +0.320 |
There | +0.308 |
Neg logit contributions | |
ies | -0.519 |
- | -0.485 |
ig | -0.455 |
ish | -0.451 |
an | -0.448 |
Token | almost |
Token ID | 2048 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
Don | +0.355 |
My | +0.346 |
Would | +0.327 |
It | +0.325 |
There | +0.324 |
Neg logit contributions | |
ies | -0.383 |
- | -0.342 |
an | -0.341 |
ish | -0.337 |
ig | -0.331 |
Token | out |
Token ID | 503 |
Feature activation | -7.18e+00 |
Pos logit contributions | |
Don | +0.170 |
My | +0.159 |
Would | +0.154 |
There | +0.152 |
It | +0.150 |
Neg logit contributions | |
ies | -0.160 |
an | -0.150 |
- | -0.149 |
ig | -0.147 |
ish | -0.143 |
Token | of |
Token ID | 286 |
Feature activation | -4.28e-01 |
Pos logit contributions | |
an | +0.213 |
ies | +0.212 |
ish | +0.207 |
ig | +0.199 |
- | +0.177 |
Neg logit contributions | |
Don | -0.336 |
My | -0.319 |
Would | -0.314 |
There | -0.308 |
Well | -0.304 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.79e+01 |
Pos logit contributions | |
ies | +0.019 |
an | +0.018 |
- | +0.017 |
ish | +0.017 |
ig | +0.017 |
Neg logit contributions | |
Don | -0.015 |
Would | -0.014 |
My | -0.014 |
There | -0.014 |
Here | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.00e+01 |
Pos logit contributions | |
ies | +0.912 |
an | +0.754 |
ig | +0.748 |
ish | +0.739 |
- | +0.712 |
Neg logit contributions | |
Don | -2.131 |
My | -2.042 |
Would | -1.964 |
Well | -1.956 |
It | -1.948 |
Token | Finally |
Token ID | 9461 |
Feature activation | +1.30e+01 |
Pos logit contributions | |
- | +0.848 |
ies | +0.844 |
. | +0.828 |
, | +0.807 |
house | +0.800 |
Neg logit contributions | |
Don | -0.518 |
Take | -0.433 |
My | -0.433 |
Listen | -0.425 |
Over | -0.417 |
Token | she |
Token ID | 673 |
Feature activation | +1.30e+01 |
Pos logit contributions | |
Don | +0.770 |
My | +0.737 |
Would | +0.727 |
Well | +0.716 |
There | +0.706 |
Neg logit contributions | |
ies | -0.266 |
- | -0.220 |
an | -0.200 |
ish | -0.196 |
ig | -0.188 |
Token | stopped |
Token ID | 5025 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
Don | +0.433 |
My | +0.398 |
Would | +0.396 |
Well | +0.374 |
Here | +0.367 |
Neg logit contributions | |
ies | -0.612 |
- | -0.581 |
ig | -0.538 |
ish | -0.528 |
an | -0.525 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
ies | +0.036 |
an | +0.033 |
ig | +0.031 |
ish | +0.031 |
- | +0.031 |
Neg logit contributions | |
Don | -0.055 |
My | -0.052 |
Would | -0.049 |
It | -0.048 |
Well | -0.047 |
Token | L |
Token ID | 43 |
Feature activation | +9.36e+00 |
Pos logit contributions | |
Don | +0.332 |
Did | +0.324 |
Would | +0.302 |
My | +0.295 |
Well | +0.294 |
Neg logit contributions | |
- | -0.535 |
ies | -0.524 |
ig | -0.493 |
an | -0.480 |
home | -0.466 |
Token | ily |
Token ID | 813 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
Don | +0.304 |
My | +0.277 |
Would | +0.275 |
Well | +0.260 |
There | +0.254 |
Neg logit contributions | |
ies | -0.454 |
- | -0.425 |
ig | -0.400 |
an | -0.393 |
ish | -0.390 |
Token | took |
Token ID | 1718 |
Feature activation | +5.41e+00 |
Pos logit contributions | |
Don | +0.451 |
My | +0.421 |
Would | +0.415 |
Well | +0.403 |
It | +0.386 |
Neg logit contributions | |
ies | -0.421 |
- | -0.385 |
ig | -0.360 |
ish | -0.354 |
an | -0.353 |
Token | a |
Token ID | 257 |
Feature activation | +6.48e+00 |
Pos logit contributions | |
Don | +0.181 |
My | +0.169 |
Would | +0.164 |
Well | +0.157 |
There | +0.154 |
Neg logit contributions | |
ies | -0.254 |
- | -0.233 |
ig | -0.221 |
an | -0.221 |
ish | -0.219 |
Token | deep |
Token ID | 2769 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
Don | +0.296 |
My | +0.283 |
Would | +0.272 |
It | +0.270 |
There | +0.267 |
Neg logit contributions | |
ies | -0.220 |
- | -0.186 |
an | -0.185 |
ishing | -0.181 |
ish | -0.175 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.78e+01 |
Pos logit contributions | |
Don | +0.140 |
There | +0.131 |
My | +0.129 |
It | +0.129 |
Well | +0.125 |
Neg logit contributions | |
ies | -0.051 |
decision | -0.037 |
ish | -0.036 |
- | -0.035 |
ys | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
ies | +1.074 |
an | +0.877 |
decision | +0.866 |
ish | +0.857 |
choice | +0.844 |
Neg logit contributions | |
Don | -1.987 |
My | -1.889 |
Well | -1.804 |
It | -1.796 |
Would | -1.785 |
Token | started |
Token ID | 2067 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ies | +0.666 |
- | +0.630 |
ig | +0.568 |
eye | +0.556 |
ish | +0.546 |
Neg logit contributions | |
Don | -0.586 |
Would | -0.565 |
My | -0.548 |
Well | -0.545 |
Here | -0.529 |
Token | to |
Token ID | 284 |
Feature activation | -5.90e+00 |
Pos logit contributions | |
ies | +0.083 |
an | +0.074 |
- | +0.074 |
ish | +0.073 |
ig | +0.072 |
Neg logit contributions | |
Don | -0.094 |
My | -0.088 |
Would | -0.086 |
Well | -0.081 |
It | -0.081 |
Token | climb |
Token ID | 12080 |
Feature activation | -1.22e+01 |
Pos logit contributions | |
ies | +0.234 |
- | +0.216 |
an | +0.201 |
ig | +0.196 |
ish | +0.195 |
Neg logit contributions | |
Don | -0.241 |
My | -0.231 |
Would | -0.215 |
It | -0.207 |
There | -0.205 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e+01 |
Pos logit contributions | |
ies | +0.442 |
- | +0.384 |
an | +0.375 |
ish | +0.371 |
ig | +0.368 |
Neg logit contributions | |
Don | -0.550 |
My | -0.517 |
Would | -0.503 |
Well | -0.488 |
There | -0.485 |
Token | , |
Token ID | 11 |
Feature activation | -2.07e+01 |
Pos logit contributions | |
Don | +1.009 |
My | +0.963 |
Would | +0.949 |
There | +0.937 |
Well | +0.925 |
Neg logit contributions | |
ies | -0.297 |
- | -0.242 |
an | -0.231 |
ish | -0.214 |
ig | -0.211 |
Token | Lily |
Token ID | 20037 |
Feature activation | +5.63e+00 |
Pos logit contributions | |
ies | +0.919 |
- | +0.852 |
ig | +0.782 |
ish | +0.780 |
an | +0.771 |
Neg logit contributions | |
Don | -0.738 |
My | -0.686 |
Would | -0.679 |
Well | -0.657 |
Here | -0.633 |
Token | took |
Token ID | 1718 |
Feature activation | +5.84e+00 |
Pos logit contributions | |
Don | +0.266 |
My | +0.252 |
Would | +0.243 |
Well | +0.235 |
It | +0.232 |
Neg logit contributions | |
ies | -0.174 |
- | -0.158 |
an | -0.149 |
ig | -0.149 |
decision | -0.147 |
Token | a |
Token ID | 257 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
Don | +0.197 |
My | +0.183 |
Would | +0.178 |
Well | +0.170 |
There | +0.167 |
Neg logit contributions | |
ies | -0.273 |
- | -0.251 |
ig | -0.239 |
an | -0.238 |
ish | -0.236 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
Don | +0.319 |
My | +0.308 |
It | +0.300 |
Would | +0.297 |
Here | +0.296 |
Neg logit contributions | |
ies | -0.244 |
an | -0.211 |
ishing | -0.208 |
- | -0.197 |
ish | -0.190 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.78e+01 |
Pos logit contributions | |
Don | +0.153 |
There | +0.148 |
It | +0.144 |
My | +0.140 |
It | +0.139 |
Neg logit contributions | |
ies | -0.065 |
decision | -0.053 |
ys | -0.050 |
ish | -0.048 |
- | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | -1.53e+01 |
Pos logit contributions | |
ies | +1.066 |
decision | +0.880 |
an | +0.863 |
choice | +0.858 |
ish | +0.846 |
Neg logit contributions | |
Don | -1.976 |
My | -1.895 |
It | -1.809 |
Well | -1.797 |
There | -1.781 |
Token | went |
Token ID | 1816 |
Feature activation | -7.61e-01 |
Pos logit contributions | |
ies | +0.644 |
- | +0.608 |
ig | +0.548 |
eye | +0.541 |
ish | +0.531 |
Neg logit contributions | |
Don | -0.577 |
Would | -0.550 |
My | -0.540 |
Well | -0.530 |
Here | -0.509 |
Token | into |
Token ID | 656 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
ies | +0.026 |
ish | +0.024 |
an | +0.023 |
decision | +0.023 |
ater | +0.023 |
Neg logit contributions | |
Don | -0.033 |
My | -0.032 |
Would | -0.031 |
There | -0.030 |
It | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +5.56e+00 |
Pos logit contributions | |
ies | +0.157 |
- | +0.145 |
ig | +0.137 |
ish | +0.137 |
an | +0.136 |
Neg logit contributions | |
Don | -0.118 |
My | -0.110 |
Would | -0.108 |
Well | -0.103 |
There | -0.100 |
Token | woods |
Token ID | 16479 |
Feature activation | -9.82e+00 |
Pos logit contributions | |
Don | +0.288 |
My | +0.275 |
Would | +0.270 |
Here | +0.262 |
There | +0.262 |
Neg logit contributions | |
ies | -0.165 |
- | -0.141 |
ig | -0.126 |
ish | -0.124 |
an | -0.123 |
Token | a |
Token ID | 257 |
Feature activation | +6.38e+00 |
Pos logit contributions | |
ies | +0.339 |
- | +0.311 |
ig | +0.297 |
ish | +0.297 |
an | +0.294 |
Neg logit contributions | |
Don | -0.263 |
My | -0.245 |
Would | -0.240 |
Well | -0.227 |
There | -0.225 |
Token | cave |
Token ID | 11527 |
Feature activation | -2.87e+01 |
Pos logit contributions | |
Don | +0.297 |
Would | +0.285 |
My | +0.283 |
It | +0.281 |
There | +0.274 |
Neg logit contributions | |
ies | -0.212 |
ishing | -0.180 |
- | -0.173 |
ish | -0.171 |
an | -0.166 |
Token | with |
Token ID | 351 |
Feature activation | -9.62e+00 |
Pos logit contributions | |
ies | +0.721 |
ish | +0.643 |
ig | +0.639 |
an | +0.601 |
- | +0.585 |
Neg logit contributions | |
Don | -1.467 |
My | -1.387 |
It | -1.364 |
Would | -1.363 |
There | -1.347 |
Token | my |
Token ID | 616 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
ies | +0.399 |
- | +0.370 |
eye | +0.329 |
an | +0.326 |
ig | +0.326 |
Neg logit contributions | |
Don | -0.376 |
Would | -0.347 |
My | -0.342 |
Well | -0.341 |
That | -0.320 |
Token | fire |
Token ID | 2046 |
Feature activation | -1.70e+01 |
Pos logit contributions | |
Don | +0.158 |
My | +0.156 |
Would | +0.149 |
Here | +0.147 |
I | +0.143 |
Neg logit contributions | |
ies | -0.097 |
- | -0.086 |
ishing | -0.085 |
ig | -0.082 |
ish | -0.077 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.77e+01 |
Pos logit contributions | |
ig | +0.398 |
ies | +0.398 |
decision | +0.393 |
ishing | +0.384 |
ater | +0.346 |
Neg logit contributions | |
Don | -0.734 |
Did | -0.733 |
It | -0.725 |
I | -0.724 |
Why | -0.720 |
Token | ." |
Token ID | 526 |
Feature activation | -1.64e+01 |
Pos logit contributions | |
ies | +0.859 |
- | +0.708 |
ig | +0.691 |
ish | +0.654 |
an | +0.651 |
Neg logit contributions | |
Don | -2.147 |
My | -2.055 |
Would | -2.003 |
Well | -1.975 |
It | -1.954 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
ies | +0.771 |
- | +0.760 |
. | +0.705 |
, | +0.689 |
home | +0.668 |
Neg logit contributions | |
Don | -0.437 |
Would | -0.385 |
Take | -0.373 |
My | -0.368 |
There | -0.367 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e+01 |
Pos logit contributions | |
Don | +0.372 |
Did | +0.362 |
My | +0.341 |
Well | +0.327 |
Looks | +0.324 |
Neg logit contributions | |
- | -0.615 |
ies | -0.578 |
an | -0.531 |
. | -0.530 |
ie | -0.517 |
Token | They |
Token ID | 2990 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
Did | +0.386 |
Don | +0.356 |
Would | +0.334 |
Well | +0.333 |
Everything | +0.328 |
Neg logit contributions | |
- | -0.572 |
ies | -0.549 |
an | -0.507 |
ater | -0.503 |
ig | -0.486 |
Token | go |
Token ID | 467 |
Feature activation | -9.30e+00 |
Pos logit contributions | |
Don | +0.392 |
My | +0.364 |
Would | +0.362 |
Well | +0.348 |
There | +0.336 |
Neg logit contributions | |
ies | -0.432 |
- | -0.399 |
ig | -0.372 |
ish | -0.368 |
an | -0.364 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e+01 |
Pos logit contributions | |
ies | +0.796 |
ish | +0.672 |
ig | +0.665 |
an | +0.649 |
ishing | +0.638 |
Neg logit contributions | |
Don | -1.813 |
My | -1.730 |
Would | -1.682 |
Well | -1.673 |
It | -1.655 |
Token | He |
Token ID | 679 |
Feature activation | +1.72e+01 |
Pos logit contributions | |
ies | +0.912 |
- | +0.892 |
. | +0.873 |
, | +0.845 |
house | +0.832 |
Neg logit contributions | |
Don | -0.519 |
My | -0.438 |
Take | -0.420 |
Here | -0.420 |
Listen | -0.409 |
Token | took |
Token ID | 1718 |
Feature activation | +5.70e+00 |
Pos logit contributions | |
Don | +0.597 |
My | +0.552 |
Would | +0.544 |
Well | +0.521 |
There | +0.506 |
Neg logit contributions | |
ies | -0.784 |
- | -0.733 |
ig | -0.683 |
ish | -0.678 |
an | -0.672 |
Token | a |
Token ID | 257 |
Feature activation | +7.16e+00 |
Pos logit contributions | |
Don | +0.210 |
My | +0.195 |
Would | +0.193 |
Well | +0.185 |
There | +0.179 |
Neg logit contributions | |
ies | -0.249 |
- | -0.231 |
ig | -0.215 |
ish | -0.213 |
an | -0.212 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
Don | +0.314 |
My | +0.298 |
It | +0.290 |
Would | +0.288 |
Here | +0.285 |
Neg logit contributions | |
ies | -0.255 |
ishing | -0.213 |
an | -0.212 |
- | -0.208 |
ish | -0.203 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.76e+01 |
Pos logit contributions | |
Don | +0.155 |
It | +0.146 |
There | +0.145 |
Be | +0.143 |
Ah | +0.142 |
Neg logit contributions | |
ies | -0.074 |
ys | -0.054 |
ish | -0.053 |
- | -0.050 |
an | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | -1.53e+01 |
Pos logit contributions | |
ies | +1.046 |
an | +0.828 |
decision | +0.814 |
ish | +0.814 |
- | +0.805 |
Neg logit contributions | |
Don | -2.005 |
My | -1.899 |
It | -1.825 |
Well | -1.824 |
Would | -1.815 |
Token | started |
Token ID | 2067 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
ies | +0.646 |
- | +0.616 |
ig | +0.554 |
eye | +0.546 |
ish | +0.539 |
Neg logit contributions | |
Don | -0.564 |
Would | -0.536 |
My | -0.532 |
Well | -0.515 |
Here | -0.495 |
Token | to |
Token ID | 284 |
Feature activation | -5.64e+00 |
Pos logit contributions | |
ies | +0.073 |
- | +0.065 |
an | +0.064 |
ish | +0.063 |
ig | +0.063 |
Neg logit contributions | |
Don | -0.083 |
My | -0.078 |
Would | -0.077 |
Well | -0.073 |
There | -0.072 |
Token | yell |
Token ID | 28946 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
ies | +0.219 |
- | +0.200 |
an | +0.186 |
ish | +0.185 |
ig | +0.184 |
Neg logit contributions | |
Don | -0.235 |
My | -0.223 |
Would | -0.213 |
It | -0.204 |
There | -0.201 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
ies | +0.391 |
ig | +0.333 |
an | +0.326 |
ish | +0.317 |
ishing | +0.310 |
Neg logit contributions | |
Don | -0.711 |
My | -0.675 |
Would | -0.653 |
It | -0.652 |
It | -0.645 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.99e+01 |
Pos logit contributions | |
Don | +0.049 |
My | +0.048 |
Would | +0.045 |
Here | +0.045 |
It | +0.045 |
Neg logit contributions | |
ies | -0.048 |
- | -0.042 |
ishing | -0.041 |
ish | -0.040 |
an | -0.040 |
Token | my |
Token ID | 1820 |
Feature activation | -9.62e+00 |
Pos logit contributions | |
ies | +0.896 |
- | +0.833 |
ig | +0.781 |
ish | +0.773 |
an | +0.769 |
Neg logit contributions | |
Don | -0.712 |
My | -0.658 |
Would | -0.651 |
Well | -0.623 |
There | -0.601 |
Token | took |
Token ID | 1718 |
Feature activation | +5.95e+00 |
Pos logit contributions | |
ies | +0.437 |
- | +0.406 |
ig | +0.386 |
an | +0.381 |
ish | +0.379 |
Neg logit contributions | |
Don | -0.333 |
My | -0.307 |
Would | -0.304 |
Well | -0.290 |
There | -0.278 |
Token | a |
Token ID | 257 |
Feature activation | +6.91e+00 |
Pos logit contributions | |
Don | +0.202 |
My | +0.186 |
Would | +0.184 |
Well | +0.177 |
There | +0.169 |
Neg logit contributions | |
ies | -0.276 |
- | -0.259 |
ig | -0.241 |
ish | -0.240 |
an | -0.237 |
Token | deep |
Token ID | 2769 |
Feature activation | +2.98e+00 |
Pos logit contributions | |
Don | +0.296 |
My | +0.286 |
It | +0.279 |
Would | +0.278 |
Here | +0.274 |
Neg logit contributions | |
ies | -0.245 |
an | -0.211 |
ishing | -0.206 |
ish | -0.197 |
ig | -0.195 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.75e+01 |
Pos logit contributions | |
Don | +0.149 |
There | +0.143 |
My | +0.139 |
It | +0.138 |
It | +0.137 |
Neg logit contributions | |
ies | -0.066 |
ys | -0.051 |
decision | -0.050 |
ish | -0.049 |
an | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | -1.55e+01 |
Pos logit contributions | |
ies | +1.027 |
an | +0.847 |
decision | +0.830 |
ish | +0.816 |
ig | +0.812 |
Neg logit contributions | |
Don | -1.986 |
My | -1.891 |
Well | -1.821 |
It | -1.808 |
Would | -1.797 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
ies | +0.649 |
- | +0.609 |
ig | +0.541 |
ish | +0.527 |
eye | +0.525 |
Neg logit contributions | |
Don | -0.588 |
Would | -0.555 |
My | -0.553 |
Well | -0.544 |
Here | -0.523 |
Token | . |
Token ID | 13 |
Feature activation | -1.99e+01 |
Pos logit contributions | |
ies | +0.052 |
ig | +0.052 |
an | +0.048 |
decision | +0.045 |
ish | +0.044 |
Neg logit contributions | |
Don | -0.117 |
My | -0.111 |
Would | -0.107 |
It | -0.106 |
That | -0.102 |
Token | She |
Token ID | 1375 |
Feature activation | +2.12e+01 |
Pos logit contributions | |
ies | +0.769 |
window | +0.754 |
. | +0.743 |
, | +0.738 |
- | +0.737 |
Neg logit contributions | |
Don | -0.575 |
My | -0.564 |
Take | -0.554 |
Ah | -0.550 |
Our | -0.519 |
Token | said |
Token ID | 531 |
Feature activation | +8.49e-01 |
Pos logit contributions | |
Don | +0.813 |
My | +0.757 |
Would | +0.755 |
Here | +0.734 |
There | +0.720 |
Neg logit contributions | |
ies | -0.900 |
- | -0.867 |
ig | -0.766 |
an | -0.742 |
ish | -0.742 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e+01 |
Pos logit contributions | |
ies | +0.667 |
- | +0.631 |
decision | +0.545 |
choice | +0.537 |
home | +0.534 |
Neg logit contributions | |
Don | -0.585 |
Would | -0.549 |
My | -0.535 |
Well | -0.528 |
Did | -0.503 |
Token | She |
Token ID | 1375 |
Feature activation | +1.95e+01 |
Pos logit contributions | |
ies | +0.906 |
- | +0.894 |
. | +0.853 |
, | +0.845 |
house | +0.820 |
Neg logit contributions | |
Don | -0.457 |
My | -0.380 |
Take | -0.377 |
Well | -0.371 |
Here | -0.370 |
Token | took |
Token ID | 1718 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
Don | +0.693 |
My | +0.640 |
Would | +0.634 |
Well | +0.606 |
There | +0.592 |
Neg logit contributions | |
ies | -0.876 |
- | -0.818 |
ig | -0.758 |
ish | -0.751 |
an | -0.743 |
Token | a |
Token ID | 257 |
Feature activation | +7.26e+00 |
Pos logit contributions | |
Don | +0.221 |
My | +0.204 |
Would | +0.204 |
Well | +0.196 |
There | +0.187 |
Neg logit contributions | |
ies | -0.235 |
- | -0.220 |
ig | -0.202 |
ish | -0.200 |
an | -0.197 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
Don | +0.319 |
My | +0.308 |
It | +0.298 |
Would | +0.295 |
Here | +0.292 |
Neg logit contributions | |
ies | -0.252 |
an | -0.215 |
ishing | -0.211 |
- | -0.205 |
ish | -0.201 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.75e+01 |
Pos logit contributions | |
Don | +0.153 |
There | +0.146 |
It | +0.146 |
It | +0.139 |
Be | +0.139 |
Neg logit contributions | |
ies | -0.070 |
ys | -0.052 |
ish | -0.051 |
decision | -0.050 |
an | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | -1.55e+01 |
Pos logit contributions | |
ies | +1.036 |
an | +0.835 |
decision | +0.829 |
ish | +0.817 |
choice | +0.811 |
Neg logit contributions | |
Don | -2.000 |
My | -1.908 |
It | -1.827 |
Well | -1.819 |
Would | -1.811 |
Token | shouted |
Token ID | 17293 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
ies | +0.639 |
- | +0.611 |
ig | +0.547 |
eye | +0.541 |
ish | +0.525 |
Neg logit contributions | |
Don | -0.578 |
Would | -0.559 |
My | -0.545 |
Well | -0.535 |
Here | -0.516 |
Token | " |
Token ID | 366 |
Feature activation | +7.40e+00 |
Pos logit contributions | |
ies | +0.018 |
an | +0.014 |
ig | +0.014 |
ish | +0.014 |
decision | +0.013 |
Neg logit contributions | |
Don | -0.032 |
My | -0.031 |
Would | -0.030 |
It | -0.029 |
It | -0.029 |
Token | ST |
Token ID | 2257 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
Don | +0.149 |
Would | +0.140 |
My | +0.133 |
Did | +0.131 |
Well | +0.125 |
Neg logit contributions | |
- | -0.398 |
ies | -0.397 |
ig | -0.363 |
an | -0.361 |
ish | -0.359 |
Token | OP |
Token ID | 3185 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
Don | +0.130 |
My | +0.118 |
Would | +0.116 |
Well | +0.110 |
There | +0.106 |
Neg logit contributions | |
ies | -0.200 |
- | -0.192 |
ig | -0.182 |
ish | -0.180 |
an | -0.178 |
Token | torch |
Token ID | 28034 |
Feature activation | -2.12e+01 |
Pos logit contributions | |
Don | +0.210 |
My | +0.208 |
Here | +0.203 |
Would | +0.196 |
I | +0.191 |
Neg logit contributions | |
ies | -0.135 |
- | -0.107 |
ishing | -0.105 |
ig | -0.102 |
ie | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | -2.13e+01 |
Pos logit contributions | |
ies | +0.617 |
an | +0.532 |
ig | +0.526 |
decision | +0.522 |
- | +0.518 |
Neg logit contributions | |
Don | -1.028 |
My | -0.979 |
It | -0.934 |
There | -0.928 |
Well | -0.920 |
Token | took |
Token ID | 1718 |
Feature activation | +6.60e+00 |
Pos logit contributions | |
ies | +0.743 |
- | +0.683 |
ig | +0.610 |
eye | +0.605 |
ish | +0.599 |
Neg logit contributions | |
Don | -0.952 |
Would | -0.903 |
My | -0.897 |
Well | -0.881 |
Here | -0.851 |
Token | a |
Token ID | 257 |
Feature activation | +7.82e+00 |
Pos logit contributions | |
Don | +0.259 |
My | +0.242 |
Would | +0.238 |
Well | +0.230 |
There | +0.224 |
Neg logit contributions | |
ies | -0.273 |
- | -0.250 |
ig | -0.233 |
ish | -0.231 |
an | -0.230 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.70e+00 |
Pos logit contributions | |
Don | +0.307 |
It | +0.300 |
My | +0.296 |
There | +0.283 |
Would | +0.281 |
Neg logit contributions | |
ies | -0.302 |
ishing | -0.265 |
an | -0.257 |
ish | -0.247 |
- | -0.240 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.74e+01 |
Pos logit contributions | |
Don | +0.158 |
It | +0.157 |
It | +0.155 |
There | +0.152 |
Be | +0.150 |
Neg logit contributions | |
ies | -0.097 |
ys | -0.076 |
ish | -0.074 |
ishing | -0.070 |
an | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
ies | +0.967 |
an | +0.778 |
ish | +0.773 |
decision | +0.765 |
choice | +0.747 |
Neg logit contributions | |
Don | -2.011 |
My | -1.928 |
It | -1.854 |
Well | -1.840 |
There | -1.823 |
Token | started |
Token ID | 2067 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ies | +0.615 |
- | +0.580 |
ig | +0.526 |
eye | +0.520 |
ish | +0.514 |
Neg logit contributions | |
Don | -0.564 |
Would | -0.531 |
My | -0.525 |
Well | -0.509 |
Here | -0.489 |
Token | on |
Token ID | 319 |
Feature activation | -5.03e+00 |
Pos logit contributions | |
ies | +0.066 |
an | +0.058 |
- | +0.058 |
ish | +0.058 |
ig | +0.057 |
Neg logit contributions | |
Don | -0.073 |
My | -0.068 |
Would | -0.066 |
Well | -0.063 |
There | -0.062 |
Token | her |
Token ID | 607 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
ies | +0.225 |
an | +0.203 |
ish | +0.203 |
- | +0.198 |
ig | +0.194 |
Neg logit contributions | |
Don | -0.183 |
My | -0.170 |
Would | -0.164 |
It | -0.156 |
There | -0.155 |
Token | adventure |
Token ID | 8855 |
Feature activation | -3.10e+01 |
Pos logit contributions | |
My | +0.504 |
Don | +0.475 |
There | +0.453 |
It | +0.437 |
I | +0.433 |
Neg logit contributions | |
ies | -0.377 |
ishing | -0.308 |
ish | -0.300 |
ater | -0.287 |
an | -0.285 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
Don | +0.353 |
My | +0.336 |
Did | +0.330 |
Here | +0.312 |
Our | +0.310 |
Neg logit contributions | |
- | -0.576 |
ies | -0.531 |
. | -0.511 |
ig | -0.495 |
an | -0.479 |
Token | Tom |
Token ID | 13787 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
Did | +0.410 |
Don | +0.395 |
Would | +0.387 |
My | +0.369 |
Here | +0.364 |
Neg logit contributions | |
- | -0.498 |
ies | -0.456 |
ig | -0.425 |
an | -0.415 |
ater | -0.405 |
Token | took |
Token ID | 1718 |
Feature activation | +6.23e+00 |
Pos logit contributions | |
Don | +0.089 |
My | +0.083 |
Would | +0.082 |
Well | +0.079 |
There | +0.076 |
Neg logit contributions | |
ies | -0.102 |
- | -0.095 |
ig | -0.088 |
ish | -0.087 |
an | -0.086 |
Token | a |
Token ID | 257 |
Feature activation | +7.45e+00 |
Pos logit contributions | |
Don | +0.226 |
My | +0.210 |
Would | +0.207 |
Well | +0.199 |
There | +0.193 |
Neg logit contributions | |
ies | -0.276 |
- | -0.255 |
ig | -0.239 |
ish | -0.237 |
an | -0.235 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
Don | +0.323 |
My | +0.302 |
Would | +0.291 |
It | +0.291 |
There | +0.281 |
Neg logit contributions | |
ies | -0.273 |
- | -0.234 |
ishing | -0.232 |
an | -0.229 |
ish | -0.226 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.74e+01 |
Pos logit contributions | |
Don | +0.173 |
It | +0.159 |
There | +0.157 |
My | +0.157 |
Be | +0.155 |
Neg logit contributions | |
ies | -0.072 |
ish | -0.051 |
ishing | -0.047 |
ys | -0.047 |
- | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | -1.52e+01 |
Pos logit contributions | |
ies | +1.078 |
an | +0.853 |
ish | +0.853 |
decision | +0.846 |
choice | +0.836 |
Neg logit contributions | |
Don | -1.960 |
My | -1.854 |
It | -1.770 |
Well | -1.764 |
Would | -1.751 |
Token | said |
Token ID | 531 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
ies | +0.611 |
- | +0.597 |
eye | +0.522 |
ig | +0.521 |
an | +0.504 |
Neg logit contributions | |
Don | -0.561 |
Would | -0.554 |
My | -0.535 |
Well | -0.534 |
Here | -0.521 |
Token | , |
Token ID | 11 |
Feature activation | -2.05e+01 |
Pos logit contributions | |
Don | +0.042 |
My | +0.041 |
Would | +0.040 |
Well | +0.039 |
There | +0.039 |
Neg logit contributions | |
ies | -0.010 |
- | -0.009 |
ig | -0.007 |
ish | -0.006 |
an | -0.006 |
Token | " |
Token ID | 366 |
Feature activation | +7.78e+00 |
Pos logit contributions | |
ies | +0.898 |
- | +0.894 |
ig | +0.772 |
ish | +0.764 |
, | +0.760 |
Neg logit contributions | |
Don | -0.636 |
My | -0.615 |
Here | -0.614 |
There | -0.589 |
Would | -0.586 |
Token | I |
Token ID | 40 |
Feature activation | +2.19e+01 |
Pos logit contributions | |
Did | +0.119 |
From | +0.110 |
Their | +0.108 |
Higher | +0.099 |
Would | +0.097 |
Neg logit contributions | |
an | -0.409 |
- | -0.408 |
ies | -0.396 |
ater | -0.390 |
le | -0.381 |
Token |
|
Token ID | 198 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
Don | +0.353 |
My | +0.348 |
Did | +0.337 |
Well | +0.314 |
Our | +0.302 |
Neg logit contributions | |
- | -0.521 |
ies | -0.497 |
. | -0.472 |
home | -0.448 |
an | -0.444 |
Token | Mark |
Token ID | 9704 |
Feature activation | +9.13e+00 |
Pos logit contributions | |
Did | +0.406 |
Don | +0.382 |
Would | +0.370 |
My | +0.366 |
Well | +0.361 |
Neg logit contributions | |
- | -0.457 |
ies | -0.433 |
ater | -0.395 |
ig | -0.394 |
an | -0.392 |
Token | took |
Token ID | 1718 |
Feature activation | +6.38e+00 |
Pos logit contributions | |
Don | +0.394 |
My | +0.371 |
Would | +0.363 |
Well | +0.355 |
It | +0.341 |
Neg logit contributions | |
ies | -0.331 |
- | -0.295 |
ig | -0.279 |
ish | -0.274 |
an | -0.271 |
Token | a |
Token ID | 257 |
Feature activation | +7.46e+00 |
Pos logit contributions | |
Don | +0.217 |
My | +0.203 |
Would | +0.198 |
Well | +0.189 |
There | +0.186 |
Neg logit contributions | |
ies | -0.294 |
- | -0.267 |
ig | -0.257 |
ish | -0.254 |
an | -0.253 |
Token | big |
Token ID | 1263 |
Feature activation | -6.90e-01 |
Pos logit contributions | |
Don | +0.328 |
My | +0.318 |
It | +0.307 |
Would | +0.303 |
There | +0.296 |
Neg logit contributions | |
ies | -0.246 |
ishing | -0.211 |
an | -0.209 |
- | -0.203 |
ish | -0.196 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
ies | +0.023 |
ishing | +0.020 |
an | +0.019 |
ys | +0.018 |
surroundings | +0.018 |
Neg logit contributions | |
It | -0.026 |
Grab | -0.026 |
Don | -0.026 |
Be | -0.025 |
Here | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
ies | +1.143 |
an | +0.926 |
ish | +0.911 |
decision | +0.911 |
choice | +0.905 |
Neg logit contributions | |
Don | -1.876 |
My | -1.794 |
It | -1.720 |
Well | -1.693 |
That | -1.675 |
Token | dove |
Token ID | 38862 |
Feature activation | -1.78e+01 |
Pos logit contributions | |
ies | +0.615 |
- | +0.600 |
ig | +0.521 |
eye | +0.516 |
ish | +0.504 |
Neg logit contributions | |
Don | -0.556 |
Would | -0.539 |
My | -0.518 |
Well | -0.515 |
Here | -0.502 |
Token | down |
Token ID | 866 |
Feature activation | -8.62e+00 |
Pos logit contributions | |
ies | +0.602 |
decision | +0.542 |
choice | +0.526 |
mess | +0.505 |
ish | +0.501 |
Neg logit contributions | |
Don | -0.810 |
My | -0.789 |
Would | -0.746 |
It | -0.712 |
Come | -0.711 |
Token | to |
Token ID | 284 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
ies | +0.307 |
- | +0.260 |
ish | +0.260 |
ig | +0.258 |
an | +0.254 |
Neg logit contributions | |
Don | -0.399 |
My | -0.375 |
Would | -0.369 |
Well | -0.355 |
There | -0.353 |
Token | the |
Token ID | 262 |
Feature activation | +5.92e+00 |
Pos logit contributions | |
ies | +0.226 |
- | +0.216 |
ig | +0.198 |
eye | +0.197 |
an | +0.195 |
Neg logit contributions | |
Don | -0.212 |
Would | -0.193 |
My | -0.193 |
Well | -0.190 |
There | -0.183 |
Token | so |
Token ID | 523 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ies | +0.812 |
- | +0.749 |
ig | +0.680 |
ish | +0.675 |
an | +0.668 |
Neg logit contributions | |
Don | -0.885 |
My | -0.833 |
Would | -0.825 |
Well | -0.798 |
Here | -0.777 |
Token | he |
Token ID | 339 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
ies | +0.081 |
- | +0.076 |
ig | +0.070 |
ish | +0.069 |
an | +0.068 |
Neg logit contributions | |
Don | -0.052 |
Would | -0.048 |
My | -0.047 |
Well | -0.045 |
Here | -0.044 |
Token | took |
Token ID | 1718 |
Feature activation | +6.81e+00 |
Pos logit contributions | |
Don | +0.370 |
My | +0.342 |
Would | +0.338 |
Well | +0.323 |
There | +0.312 |
Neg logit contributions | |
ies | -0.503 |
- | -0.470 |
ig | -0.438 |
ish | -0.435 |
an | -0.432 |
Token | a |
Token ID | 257 |
Feature activation | +8.06e+00 |
Pos logit contributions | |
Don | +0.243 |
My | +0.227 |
Would | +0.222 |
Well | +0.213 |
There | +0.209 |
Neg logit contributions | |
ies | -0.305 |
- | -0.279 |
ig | -0.264 |
ish | -0.262 |
an | -0.261 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
Don | +0.311 |
It | +0.296 |
My | +0.293 |
There | +0.283 |
Would | +0.281 |
Neg logit contributions | |
ies | -0.314 |
ishing | -0.278 |
an | -0.267 |
- | -0.259 |
ish | -0.256 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
Don | +0.157 |
It | +0.151 |
It | +0.150 |
There | +0.150 |
He | +0.147 |
Neg logit contributions | |
ies | -0.088 |
ys | -0.071 |
ish | -0.064 |
an | -0.064 |
uddle | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | -1.46e+01 |
Pos logit contributions | |
ies | +0.995 |
an | +0.781 |
ish | +0.776 |
ig | +0.773 |
decision | +0.753 |
Neg logit contributions | |
Don | -2.023 |
My | -1.922 |
It | -1.845 |
Well | -1.842 |
Would | -1.838 |
Token | kn |
Token ID | 638 |
Feature activation | -5.39e+00 |
Pos logit contributions | |
ies | +0.621 |
- | +0.588 |
ig | +0.531 |
eye | +0.523 |
ish | +0.522 |
Neg logit contributions | |
Don | -0.538 |
Would | -0.507 |
My | -0.505 |
Well | -0.485 |
Here | -0.470 |
Token | elt |
Token ID | 2120 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
ies | +0.283 |
- | +0.271 |
ig | +0.253 |
ish | +0.252 |
an | +0.247 |
Neg logit contributions | |
Don | -0.151 |
Would | -0.137 |
My | -0.137 |
Well | -0.127 |
Here | -0.122 |
Token | down |
Token ID | 866 |
Feature activation | -8.52e+00 |
Pos logit contributions | |
Don | +0.205 |
My | +0.188 |
Would | +0.183 |
Well | +0.180 |
It | +0.180 |
Neg logit contributions | |
ies | -0.144 |
- | -0.127 |
ig | -0.125 |
an | -0.122 |
ish | -0.119 |
Token | on |
Token ID | 319 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
ies | +0.255 |
ig | +0.212 |
- | +0.211 |
ish | +0.209 |
an | +0.207 |
Neg logit contributions | |
Don | -0.439 |
My | -0.411 |
Would | -0.406 |
Well | -0.395 |
There | -0.392 |
Token | little |
Token ID | 1310 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
Don | +0.233 |
My | +0.219 |
Would | +0.219 |
Well | +0.213 |
It | +0.210 |
Neg logit contributions | |
ies | -0.167 |
- | -0.142 |
ig | -0.138 |
ishing | -0.138 |
ish | -0.134 |
Token | pirate |
Token ID | 25868 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
ies | +0.186 |
- | +0.162 |
ishing | +0.157 |
ig | +0.157 |
an | +0.154 |
Neg logit contributions | |
Don | -0.243 |
My | -0.228 |
Would | -0.222 |
It | -0.220 |
I | -0.219 |
Token | took |
Token ID | 1718 |
Feature activation | +6.87e+00 |
Pos logit contributions | |
ies | +0.928 |
- | +0.859 |
ig | +0.840 |
ish | +0.820 |
eye | +0.807 |
Neg logit contributions | |
Don | -0.682 |
My | -0.626 |
Would | -0.615 |
Well | -0.589 |
There | -0.564 |
Token | a |
Token ID | 257 |
Feature activation | +7.90e+00 |
Pos logit contributions | |
Don | +0.260 |
My | +0.241 |
Would | +0.239 |
Well | +0.229 |
There | +0.221 |
Neg logit contributions | |
ies | -0.294 |
- | -0.276 |
ig | -0.255 |
ish | -0.252 |
an | -0.250 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
Don | +0.317 |
My | +0.296 |
It | +0.295 |
Would | +0.286 |
Ah | +0.281 |
Neg logit contributions | |
ies | -0.311 |
ishing | -0.273 |
an | -0.259 |
- | -0.256 |
ish | -0.256 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
Don | +0.175 |
It | +0.164 |
It | +0.163 |
There | +0.161 |
Ah | +0.159 |
Neg logit contributions | |
ies | -0.095 |
ys | -0.069 |
ish | -0.065 |
ater | -0.063 |
- | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
ies | +1.018 |
an | +0.798 |
- | +0.793 |
ish | +0.784 |
ig | +0.782 |
Neg logit contributions | |
Don | -2.003 |
My | -1.894 |
Well | -1.818 |
It | -1.818 |
Would | -1.810 |
Token | he |
Token ID | 339 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
ies | +0.623 |
- | +0.598 |
ig | +0.534 |
eye | +0.529 |
an | +0.524 |
Neg logit contributions | |
Don | -0.539 |
Would | -0.513 |
My | -0.510 |
Well | -0.490 |
Here | -0.479 |
Token | stood |
Token ID | 6204 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
Don | +0.410 |
My | +0.380 |
Would | +0.377 |
Well | +0.361 |
Here | +0.353 |
Neg logit contributions | |
ies | -0.532 |
- | -0.502 |
ig | -0.464 |
ish | -0.458 |
an | -0.457 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
Don | +0.141 |
My | +0.132 |
Would | +0.129 |
It | +0.127 |
There | +0.124 |
Neg logit contributions | |
ies | -0.109 |
- | -0.094 |
ig | -0.091 |
ish | -0.090 |
an | -0.088 |
Token | He |
Token ID | 679 |
Feature activation | +1.92e+01 |
Pos logit contributions | |
. | +0.761 |
- | +0.756 |
eye | +0.748 |
home | +0.739 |
, | +0.723 |
Neg logit contributions | |
Don | -0.540 |
My | -0.479 |
Did | -0.430 |
Looks | -0.420 |
Listen | -0.420 |
Token |
|
Token ID | 198 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
Don | +0.347 |
Did | +0.331 |
My | +0.312 |
Would | +0.305 |
Well | +0.304 |
Neg logit contributions | |
- | -0.600 |
ies | -0.573 |
. | -0.530 |
an | -0.520 |
ig | -0.508 |
Token | Jack |
Token ID | 14295 |
Feature activation | +6.51e+00 |
Pos logit contributions | |
Did | +0.370 |
Don | +0.349 |
Would | +0.333 |
Everything | +0.323 |
Well | +0.319 |
Neg logit contributions | |
- | -0.552 |
ies | -0.535 |
an | -0.485 |
ater | -0.475 |
ig | -0.473 |
Token | took |
Token ID | 1718 |
Feature activation | +6.67e+00 |
Pos logit contributions | |
Don | +0.233 |
My | +0.216 |
Would | +0.212 |
Well | +0.205 |
There | +0.196 |
Neg logit contributions | |
ies | -0.289 |
- | -0.267 |
ig | -0.252 |
ish | -0.248 |
an | -0.247 |
Token | a |
Token ID | 257 |
Feature activation | +7.72e+00 |
Pos logit contributions | |
Don | +0.213 |
My | +0.199 |
Would | +0.193 |
Well | +0.184 |
There | +0.182 |
Neg logit contributions | |
ies | -0.321 |
- | -0.292 |
ig | -0.281 |
an | -0.278 |
ish | -0.277 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
Don | +0.358 |
My | +0.341 |
Would | +0.328 |
It | +0.325 |
Here | +0.319 |
Neg logit contributions | |
ies | -0.256 |
- | -0.216 |
an | -0.212 |
ig | -0.207 |
ishing | -0.207 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
Don | +0.198 |
It | +0.185 |
My | +0.185 |
There | +0.181 |
Well | +0.179 |
Neg logit contributions | |
ies | -0.075 |
ish | -0.051 |
an | -0.047 |
ig | -0.047 |
ys | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -1.46e+01 |
Pos logit contributions | |
ies | +1.134 |
an | +0.928 |
ish | +0.905 |
ig | +0.891 |
fort | +0.881 |
Neg logit contributions | |
Don | -1.907 |
My | -1.805 |
It | -1.738 |
Well | -1.719 |
Would | -1.702 |
Token | was |
Token ID | 373 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
ies | +0.606 |
- | +0.590 |
ig | +0.517 |
ish | +0.504 |
eye | +0.503 |
Neg logit contributions | |
Don | -0.535 |
Would | -0.517 |
My | -0.504 |
Well | -0.489 |
Here | -0.484 |
Token | relieved |
Token ID | 22598 |
Feature activation | +6.97e+00 |
Pos logit contributions | |
Don | +0.449 |
My | +0.418 |
Would | +0.415 |
Well | +0.400 |
There | +0.388 |
Neg logit contributions | |
ies | -0.463 |
- | -0.430 |
ig | -0.397 |
ish | -0.392 |
an | -0.389 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
Don | +0.356 |
My | +0.329 |
It | +0.322 |
Would | +0.322 |
Well | +0.318 |
Neg logit contributions | |
ies | -0.204 |
ig | -0.179 |
an | -0.177 |
ish | -0.174 |
- | -0.163 |
Token | They |
Token ID | 1119 |
Feature activation | +1.47e+01 |
Pos logit contributions | |
- | +0.830 |
ies | +0.788 |
. | +0.775 |
eye | +0.762 |
, | +0.756 |
Neg logit contributions | |
Don | -0.522 |
My | -0.474 |
Here | -0.430 |
Our | -0.428 |
Would | -0.422 |
Token | . |
Token ID | 13 |
Feature activation | -2.03e+01 |
Pos logit contributions | |
ies | +0.588 |
- | +0.519 |
ig | +0.491 |
an | +0.488 |
ish | +0.484 |
Neg logit contributions | |
Don | -0.990 |
My | -0.955 |
Would | -0.927 |
Well | -0.900 |
There | -0.895 |
Token | He |
Token ID | 679 |
Feature activation | +1.77e+01 |
Pos logit contributions | |
ies | +0.851 |
- | +0.844 |
. | +0.829 |
, | +0.809 |
house | +0.804 |
Neg logit contributions | |
Don | -0.517 |
Listen | -0.430 |
My | -0.425 |
Take | -0.424 |
Over | -0.418 |
Token | took |
Token ID | 1718 |
Feature activation | +6.29e+00 |
Pos logit contributions | |
Don | +0.652 |
My | +0.602 |
Would | +0.599 |
Well | +0.572 |
There | +0.563 |
Neg logit contributions | |
ies | -0.768 |
- | -0.723 |
ig | -0.664 |
ish | -0.658 |
an | -0.650 |
Token | a |
Token ID | 257 |
Feature activation | +7.73e+00 |
Pos logit contributions | |
Don | +0.246 |
My | +0.230 |
Would | +0.227 |
Well | +0.219 |
There | +0.213 |
Neg logit contributions | |
ies | -0.260 |
- | -0.239 |
ig | -0.223 |
ish | -0.221 |
an | -0.220 |
Token | big |
Token ID | 1263 |
Feature activation | -4.37e-01 |
Pos logit contributions | |
Don | +0.319 |
My | +0.312 |
It | +0.305 |
Would | +0.294 |
There | +0.294 |
Neg logit contributions | |
ies | -0.281 |
an | -0.242 |
ishing | -0.238 |
- | -0.226 |
ish | -0.226 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
ies | +0.016 |
an | +0.014 |
ishing | +0.013 |
ys | +0.013 |
surroundings | +0.012 |
Neg logit contributions | |
It | -0.016 |
Grab | -0.015 |
Higher | -0.015 |
Don | -0.015 |
Be | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -1.49e+01 |
Pos logit contributions | |
ies | +1.076 |
an | +0.865 |
decision | +0.856 |
ish | +0.846 |
choice | +0.845 |
Neg logit contributions | |
Don | -1.922 |
My | -1.843 |
It | -1.775 |
Well | -1.756 |
Would | -1.739 |
Token | then |
Token ID | 788 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
ies | +0.620 |
- | +0.605 |
eye | +0.529 |
ig | +0.525 |
ish | +0.510 |
Neg logit contributions | |
Don | -0.549 |
Would | -0.530 |
My | -0.513 |
Well | -0.504 |
Here | -0.485 |
Token | carefully |
Token ID | 7773 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
ies | +0.176 |
- | +0.164 |
ig | +0.153 |
ish | +0.152 |
an | +0.151 |
Neg logit contributions | |
Don | -0.132 |
My | -0.122 |
Would | -0.121 |
Well | -0.115 |
There | -0.111 |
Token | walked |
Token ID | 6807 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
ies | +0.017 |
an | +0.015 |
ish | +0.015 |
decision | +0.015 |
- | +0.015 |
Neg logit contributions | |
Don | -0.012 |
My | -0.011 |
Would | -0.010 |
There | -0.010 |
Well | -0.010 |
Token | up |
Token ID | 510 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
Don | +0.174 |
My | +0.162 |
Would | +0.161 |
It | +0.154 |
There | +0.154 |
Neg logit contributions | |
ies | -0.117 |
ig | -0.101 |
ish | -0.099 |
- | -0.098 |
an | -0.095 |
Token | So |
Token ID | 1406 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
- | +0.869 |
ies | +0.859 |
. | +0.841 |
, | +0.823 |
house | +0.778 |
Neg logit contributions | |
Don | -0.501 |
Over | -0.432 |
Well | -0.424 |
My | -0.422 |
Listen | -0.418 |
Token | he |
Token ID | 339 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
ies | +0.120 |
- | +0.110 |
ig | +0.101 |
ish | +0.099 |
an | +0.098 |
Neg logit contributions | |
Don | -0.141 |
My | -0.132 |
Would | -0.132 |
Well | -0.127 |
Here | -0.123 |
Token | took |
Token ID | 1718 |
Feature activation | +6.69e+00 |
Pos logit contributions | |
Don | +0.403 |
My | +0.375 |
Would | +0.370 |
Well | +0.356 |
There | +0.345 |
Neg logit contributions | |
ies | -0.453 |
- | -0.420 |
ig | -0.390 |
ish | -0.386 |
an | -0.383 |
Token | a |
Token ID | 257 |
Feature activation | +7.81e+00 |
Pos logit contributions | |
Don | +0.238 |
My | +0.220 |
Would | +0.217 |
Well | +0.208 |
There | +0.201 |
Neg logit contributions | |
ies | -0.301 |
- | -0.280 |
ig | -0.262 |
ish | -0.260 |
an | -0.258 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
Don | +0.331 |
My | +0.316 |
It | +0.311 |
Would | +0.302 |
There | +0.300 |
Neg logit contributions | |
ies | -0.278 |
ishing | -0.244 |
an | -0.236 |
ish | -0.227 |
ig | -0.220 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
Don | +0.175 |
It | +0.166 |
There | +0.163 |
It | +0.163 |
Be | +0.161 |
Neg logit contributions | |
ies | -0.091 |
ys | -0.072 |
decision | -0.067 |
ish | -0.065 |
castle | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
ies | +1.060 |
an | +0.840 |
decision | +0.833 |
ish | +0.826 |
ig | +0.824 |
Neg logit contributions | |
Don | -1.972 |
My | -1.882 |
It | -1.807 |
Well | -1.782 |
Would | -1.782 |
Token | whist |
Token ID | 32262 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
ies | +0.613 |
- | +0.585 |
ig | +0.518 |
eye | +0.511 |
ish | +0.506 |
Neg logit contributions | |
Don | -0.555 |
Would | -0.531 |
My | -0.523 |
Well | -0.513 |
Here | -0.495 |
Token | led |
Token ID | 992 |
Feature activation | +8.70e+00 |
Pos logit contributions | |
ies | +0.398 |
- | +0.356 |
ig | +0.323 |
ish | +0.318 |
an | +0.315 |
Neg logit contributions | |
Don | -0.638 |
My | -0.603 |
Would | -0.596 |
Well | -0.580 |
There | -0.566 |
Token | as |
Token ID | 355 |
Feature activation | -1.59e+01 |
Pos logit contributions | |
Don | +0.393 |
My | +0.365 |
Would | +0.358 |
It | +0.342 |
There | +0.341 |
Neg logit contributions | |
ies | -0.311 |
ig | -0.271 |
an | -0.265 |
ish | -0.264 |
- | -0.260 |
Token | loud |
Token ID | 7812 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
ies | +0.737 |
- | +0.701 |
ig | +0.626 |
an | +0.607 |
ish | +0.606 |
Neg logit contributions | |
Don | -0.523 |
Would | -0.498 |
My | -0.487 |
Well | -0.472 |
Here | -0.458 |
Token |
|
Token ID | 198 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
Don | +0.378 |
My | +0.364 |
Did | +0.353 |
Well | +0.329 |
Looks | +0.324 |
Neg logit contributions | |
- | -0.529 |
ies | -0.492 |
. | -0.479 |
an | -0.463 |
ig | -0.455 |
Token | He |
Token ID | 1544 |
Feature activation | +1.67e+01 |
Pos logit contributions | |
Did | +0.357 |
Don | +0.335 |
My | +0.310 |
Would | +0.310 |
Everything | +0.303 |
Neg logit contributions | |
- | -0.523 |
ies | -0.485 |
an | -0.470 |
ater | -0.465 |
ig | -0.461 |
Token | took |
Token ID | 1718 |
Feature activation | +6.45e+00 |
Pos logit contributions | |
Don | +0.612 |
My | +0.567 |
Would | +0.559 |
Well | +0.537 |
There | +0.526 |
Neg logit contributions | |
ies | -0.734 |
- | -0.692 |
ig | -0.639 |
ish | -0.631 |
an | -0.626 |
Token | a |
Token ID | 257 |
Feature activation | +7.61e+00 |
Pos logit contributions | |
Don | +0.234 |
My | +0.217 |
Would | +0.214 |
Well | +0.206 |
There | +0.199 |
Neg logit contributions | |
ies | -0.286 |
- | -0.265 |
ig | -0.248 |
ish | -0.245 |
an | -0.244 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
Don | +0.324 |
My | +0.309 |
It | +0.301 |
Would | +0.300 |
There | +0.291 |
Neg logit contributions | |
ies | -0.274 |
ishing | -0.235 |
an | -0.228 |
- | -0.225 |
ish | -0.223 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
Don | +0.169 |
It | +0.160 |
There | +0.160 |
Be | +0.155 |
It | +0.155 |
Neg logit contributions | |
ies | -0.083 |
ish | -0.061 |
ys | -0.060 |
an | -0.056 |
ishing | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e+01 |
Pos logit contributions | |
ies | +1.055 |
an | +0.835 |
ish | +0.829 |
decision | +0.821 |
choice | +0.812 |
Neg logit contributions | |
Don | -1.968 |
My | -1.872 |
It | -1.788 |
Well | -1.785 |
Would | -1.779 |
Token | He |
Token ID | 679 |
Feature activation | +1.88e+01 |
Pos logit contributions | |
- | +0.951 |
eye | +0.904 |
ies | +0.896 |
. | +0.862 |
ig | +0.859 |
Neg logit contributions | |
Don | -0.511 |
My | -0.458 |
Would | -0.413 |
Here | -0.405 |
Our | -0.405 |
Token | said |
Token ID | 531 |
Feature activation | +9.89e-01 |
Pos logit contributions | |
Don | +0.685 |
My | +0.637 |
Would | +0.636 |
Well | +0.605 |
Here | +0.597 |
Neg logit contributions | |
ies | -0.825 |
- | -0.782 |
ig | -0.718 |
ish | -0.704 |
an | -0.700 |
Token | , |
Token ID | 11 |
Feature activation | -2.03e+01 |
Pos logit contributions | |
Don | +0.057 |
My | +0.055 |
Would | +0.054 |
Well | +0.053 |
There | +0.052 |
Neg logit contributions | |
ies | -0.022 |
- | -0.018 |
ig | -0.016 |
ish | -0.016 |
an | -0.016 |
Token | " |
Token ID | 366 |
Feature activation | +8.00e+00 |
Pos logit contributions | |
ies | +0.915 |
- | +0.907 |
ig | +0.792 |
eye | +0.787 |
ish | +0.780 |
Neg logit contributions | |
Don | -0.625 |
My | -0.602 |
Here | -0.581 |
Well | -0.567 |
Would | -0.564 |
Token | didn |
Token ID | 1422 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
Don | +0.313 |
Would | +0.261 |
My | +0.257 |
Here | +0.255 |
Did | +0.241 |
Neg logit contributions | |
ies | -0.547 |
- | -0.539 |
ig | -0.506 |
. | -0.495 |
ish | -0.489 |
Token | 't |
Token ID | 470 |
Feature activation | +1.89e+01 |
Pos logit contributions | |
Don | +0.089 |
Would | +0.086 |
Here | +0.078 |
My | +0.075 |
Well | +0.072 |
Neg logit contributions | |
- | -0.182 |
ies | -0.179 |
ig | -0.164 |
decision | -0.161 |
choice | -0.160 |
Token | need |
Token ID | 761 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
Don | +0.677 |
My | +0.621 |
Would | +0.596 |
Well | +0.571 |
It | +0.569 |
Neg logit contributions | |
ies | -0.821 |
ig | -0.763 |
- | -0.762 |
ish | -0.740 |
ishing | -0.738 |
Token | magical |
Token ID | 10883 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
ies | +0.101 |
an | +0.090 |
- | +0.089 |
ish | +0.087 |
ig | +0.087 |
Neg logit contributions | |
Don | -0.083 |
My | -0.078 |
Would | -0.075 |
There | -0.073 |
Here | -0.071 |
Token | fair |
Token ID | 3148 |
Feature activation | -9.05e+00 |
Pos logit contributions | |
an | +0.085 |
ies | +0.084 |
ish | +0.080 |
ig | +0.076 |
ater | +0.073 |
Neg logit contributions | |
Don | -0.131 |
My | -0.124 |
Would | -0.123 |
There | -0.120 |
It | -0.118 |
Token | ies |
Token ID | 444 |
Feature activation | -3.72e+01 |
Pos logit contributions | |
ig | +0.211 |
an | +0.211 |
ish | +0.188 |
- | +0.173 |
fort | +0.173 |
Neg logit contributions | |
Don | -0.400 |
My | -0.381 |
I | -0.376 |
There | -0.370 |
Well | -0.365 |
Token | to |
Token ID | 284 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
ies | +1.145 |
an | +1.059 |
ig | +1.033 |
- | +1.031 |
ish | +0.981 |
Neg logit contributions | |
Don | -1.750 |
My | -1.673 |
Would | -1.583 |
There | -1.570 |
I | -1.538 |
Token | help |
Token ID | 1037 |
Feature activation | -9.83e+00 |
Pos logit contributions | |
ies | +0.254 |
- | +0.235 |
ig | +0.220 |
ish | +0.217 |
an | +0.215 |
Neg logit contributions | |
Don | -0.264 |
My | -0.252 |
Would | -0.247 |
Well | -0.235 |
There | -0.232 |
Token | her |
Token ID | 607 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
ies | +0.431 |
ig | +0.416 |
ish | +0.402 |
- | +0.402 |
an | +0.402 |
Neg logit contributions | |
Don | -0.318 |
My | -0.311 |
Would | -0.295 |
It | -0.277 |
I | -0.270 |
Token | . |
Token ID | 13 |
Feature activation | -1.97e+01 |
Pos logit contributions | |
My | +0.463 |
Don | +0.446 |
Would | +0.431 |
I | +0.416 |
It | +0.408 |
Neg logit contributions | |
ies | -0.271 |
ig | -0.262 |
- | -0.257 |
an | -0.252 |
ishing | -0.250 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.36e+00 |
Pos logit contributions | |
ies | +0.799 |
. | +0.790 |
house | +0.780 |
- | +0.780 |
home | +0.773 |
Neg logit contributions | |
Don | -0.519 |
Take | -0.485 |
Over | -0.460 |
My | -0.458 |
Ah | -0.457 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
ies | +0.111 |
- | +0.096 |
ig | +0.091 |
an | +0.091 |
ish | +0.084 |
Neg logit contributions | |
Don | -0.286 |
My | -0.276 |
Would | -0.267 |
It | -0.262 |
Well | -0.261 |
Token | She |
Token ID | 1375 |
Feature activation | +2.00e+01 |
Pos logit contributions | |
ies | +0.907 |
- | +0.902 |
. | +0.874 |
, | +0.856 |
house | +0.828 |
Neg logit contributions | |
Don | -0.515 |
Take | -0.425 |
My | -0.422 |
Listen | -0.421 |
Well | -0.409 |
Token | took |
Token ID | 1718 |
Feature activation | +6.09e+00 |
Pos logit contributions | |
Don | +0.713 |
My | +0.655 |
Would | +0.647 |
Well | +0.622 |
There | +0.615 |
Neg logit contributions | |
ies | -0.902 |
- | -0.845 |
ig | -0.775 |
ish | -0.773 |
an | -0.758 |
Token | a |
Token ID | 257 |
Feature activation | +7.49e+00 |
Pos logit contributions | |
Don | +0.235 |
My | +0.214 |
Would | +0.214 |
Well | +0.209 |
That | +0.196 |
Neg logit contributions | |
ies | -0.255 |
- | -0.244 |
ish | -0.221 |
ig | -0.219 |
eye | -0.213 |
Token | deep |
Token ID | 2769 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
Don | +0.325 |
My | +0.316 |
It | +0.306 |
Would | +0.304 |
There | +0.300 |
Neg logit contributions | |
ies | -0.259 |
an | -0.221 |
ishing | -0.216 |
- | -0.214 |
ig | -0.211 |
Token | breath |
Token ID | 8033 |
Feature activation | -3.71e+01 |
Pos logit contributions | |
Don | +0.173 |
It | +0.165 |
There | +0.163 |
My | +0.160 |
It | +0.160 |
Neg logit contributions | |
ies | -0.073 |
ys | -0.053 |
decision | -0.050 |
ish | -0.050 |
an | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
ies | +1.068 |
an | +0.875 |
ish | +0.852 |
decision | +0.849 |
choice | +0.839 |
Neg logit contributions | |
Don | -1.946 |
My | -1.858 |
It | -1.784 |
Would | -1.769 |
Well | -1.766 |
Token | and |
Token ID | 290 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
ies | +0.594 |
- | +0.550 |
eye | +0.468 |
ig | +0.464 |
ish | +0.450 |
Neg logit contributions | |
Don | -1.010 |
Would | -0.973 |
My | -0.960 |
Well | -0.953 |
Here | -0.933 |
Token | slowly |
Token ID | 6364 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
ies | +0.618 |
- | +0.590 |
ig | +0.526 |
eye | +0.519 |
ish | +0.515 |
Neg logit contributions | |
Don | -0.540 |
Would | -0.510 |
My | -0.502 |
Well | -0.495 |
Here | -0.474 |
Token | stepped |
Token ID | 10764 |
Feature activation | +6.56e-01 |
Pos logit contributions | |
ies | +0.101 |
an | +0.091 |
- | +0.090 |
ish | +0.090 |
decision | +0.090 |
Neg logit contributions | |
Don | -0.065 |
My | -0.059 |
Would | -0.057 |
There | -0.055 |
Well | -0.054 |
Token | closer |
Token ID | 5699 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
Don | +0.030 |
My | +0.028 |
Would | +0.028 |
There | +0.027 |
It | +0.027 |
Neg logit contributions | |
ies | -0.025 |
decision | -0.020 |
- | -0.020 |
choice | -0.020 |
an | -0.020 |
Token | named |
Token ID | 3706 |
Feature activation | -1.43e+01 |
Pos logit contributions | |
ies | +0.648 |
- | +0.586 |
an | +0.555 |
ig | +0.552 |
ish | +0.548 |
Neg logit contributions | |
Don | -0.668 |
My | -0.634 |
Would | -0.622 |
Well | -0.593 |
It | -0.576 |
Token | Lily |
Token ID | 20037 |
Feature activation | +5.63e+00 |
Pos logit contributions | |
ies | +0.692 |
- | +0.642 |
ie | +0.596 |
ish | +0.581 |
ig | +0.572 |
Neg logit contributions | |
Don | -0.459 |
Well | -0.410 |
My | -0.408 |
Would | -0.406 |
Here | -0.390 |
Token | who |
Token ID | 508 |
Feature activation | -5.16e+00 |
Pos logit contributions | |
Don | +0.355 |
My | +0.340 |
Would | +0.337 |
Well | +0.330 |
There | +0.324 |
Neg logit contributions | |
ies | -0.099 |
- | -0.082 |
ig | -0.066 |
ish | -0.064 |
an | -0.063 |
Token | loved |
Token ID | 6151 |
Feature activation | +5.64e+00 |
Pos logit contributions | |
ies | +0.224 |
- | +0.223 |
ie | +0.191 |
ish | +0.185 |
ig | +0.182 |
Neg logit contributions | |
Don | -0.184 |
Would | -0.173 |
There | -0.164 |
My | -0.163 |
Here | -0.162 |
Token | fair |
Token ID | 3148 |
Feature activation | -8.84e+00 |
Pos logit contributions | |
Don | +0.241 |
My | +0.228 |
Would | +0.225 |
Well | +0.212 |
Here | +0.210 |
Neg logit contributions | |
ies | -0.209 |
- | -0.188 |
ig | -0.188 |
an | -0.184 |
ish | -0.181 |
Token | ies |
Token ID | 444 |
Feature activation | -3.71e+01 |
Pos logit contributions | |
ig | +0.238 |
an | +0.221 |
et | +0.203 |
ish | +0.203 |
decision | +0.195 |
Neg logit contributions | |
My | -0.377 |
Don | -0.353 |
She | -0.352 |
I | -0.351 |
Well | -0.327 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e+01 |
Pos logit contributions | |
ies | +1.161 |
an | +1.118 |
decision | +1.112 |
ishing | +1.057 |
ig | +1.032 |
Neg logit contributions | |
My | -1.759 |
Don | -1.659 |
Would | -1.577 |
I | -1.573 |
It | -1.528 |
Token | She |
Token ID | 1375 |
Feature activation | +2.06e+01 |
Pos logit contributions | |
ies | +0.925 |
ish | +0.873 |
fort | +0.865 |
- | +0.859 |
castle | +0.855 |
Neg logit contributions | |
Don | -0.593 |
Ah | -0.501 |
My | -0.477 |
Well | -0.467 |
Take | -0.454 |
Token | would |
Token ID | 561 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
Don | +0.666 |
My | +0.602 |
Would | +0.599 |
Well | +0.572 |
Here | +0.560 |
Neg logit contributions | |
ies | -1.009 |
- | -0.944 |
ig | -0.881 |
ish | -0.869 |
an | -0.863 |
Token | go |
Token ID | 467 |
Feature activation | -9.77e+00 |
Pos logit contributions | |
Don | +0.180 |
My | +0.170 |
Would | +0.167 |
Well | +0.160 |
There | +0.155 |
Neg logit contributions | |
ies | -0.183 |
- | -0.169 |
ig | -0.158 |
ish | -0.156 |
an | -0.154 |
Token | to |
Token ID | 284 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
ies | +0.331 |
- | +0.294 |
ig | +0.282 |
an | +0.281 |
ish | +0.280 |
Neg logit contributions | |
Don | -0.441 |
My | -0.422 |
Would | -0.422 |
Well | -0.396 |
There | -0.394 |
Token | sweet |
Token ID | 6029 |
Feature activation | -7.12e+00 |
Pos logit contributions | |
Don | +0.640 |
My | +0.616 |
Would | +0.587 |
There | +0.578 |
Well | +0.578 |
Neg logit contributions | |
an | -0.431 |
ies | -0.424 |
ig | -0.396 |
ish | -0.382 |
eye | -0.365 |
Token | cand |
Token ID | 2658 |
Feature activation | -2.06e+01 |
Pos logit contributions | |
ig | +0.234 |
an | +0.234 |
ship | +0.215 |
island | +0.210 |
fort | +0.209 |
Neg logit contributions | |
Don | -0.294 |
Come | -0.271 |
Here | -0.268 |
My | -0.265 |
There | -0.265 |
Token | ies |
Token ID | 444 |
Feature activation | -3.43e+01 |
Pos logit contributions | |
decision | +0.121 |
eye | +0.112 |
island | +0.108 |
ig | +0.107 |
an | +0.105 |
Neg logit contributions | |
Don | -1.289 |
My | -1.273 |
There | -1.252 |
It | -1.246 |
I | -1.197 |
Token | earlier |
Token ID | 2961 |
Feature activation | -4.65e+00 |
Pos logit contributions | |
ies | +0.970 |
an | +0.866 |
ig | +0.830 |
- | +0.788 |
decision | +0.776 |
Neg logit contributions | |
Don | -1.771 |
My | -1.694 |
There | -1.593 |
Would | -1.588 |
It | -1.580 |
Token | . |
Token ID | 13 |
Feature activation | -1.81e+01 |
Pos logit contributions | |
ies | +0.100 |
an | +0.076 |
ish | +0.075 |
ig | +0.075 |
- | +0.074 |
Neg logit contributions | |
Don | -0.276 |
My | -0.264 |
Would | -0.258 |
Well | -0.254 |
It | -0.250 |
Token | She |
Token ID | 1375 |
Feature activation | +2.30e+01 |
Pos logit contributions | |
. | +0.736 |
- | +0.733 |
ies | +0.720 |
, | +0.716 |
house | +0.703 |
Neg logit contributions | |
Ah | -0.502 |
Don | -0.485 |
Take | -0.483 |
Over | -0.467 |
Our | -0.461 |
Token | told |
Token ID | 1297 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Don | +0.823 |
Would | +0.770 |
My | +0.761 |
There | +0.735 |
Well | +0.731 |
Neg logit contributions | |
ies | -1.023 |
- | -0.974 |
ish | -0.874 |
eye | -0.872 |
ig | -0.867 |
Token | Max |
Token ID | 5436 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
Don | +0.041 |
My | +0.037 |
Would | +0.037 |
Well | +0.035 |
There | +0.034 |
Neg logit contributions | |
ies | -0.061 |
- | -0.057 |
ig | -0.054 |
ish | -0.053 |
an | -0.053 |
Token | that |
Token ID | 326 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
Don | +0.214 |
Would | +0.203 |
My | +0.202 |
Well | +0.197 |
Here | +0.194 |
Neg logit contributions | |
ies | -0.155 |
- | -0.143 |
ig | -0.123 |
ish | -0.122 |
an | -0.119 |
Token | she |
Token ID | 673 |
Feature activation | +1.48e+01 |
Pos logit contributions | |
ies | +0.204 |
- | +0.188 |
ish | +0.169 |
ig | +0.168 |
ie | +0.167 |
Neg logit contributions | |
Don | -0.149 |
Would | -0.136 |
My | -0.136 |
Well | -0.133 |
Here | -0.128 |
Token | needed |
Token ID | 2622 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
Don | +0.479 |
Would | +0.439 |
My | +0.436 |
Well | +0.415 |
Here | +0.407 |
Neg logit contributions | |
ies | -0.720 |
- | -0.682 |
ish | -0.632 |
ig | -0.616 |
an | -0.612 |
Token | borrow |
Token ID | 8804 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ies | +0.051 |
- | +0.044 |
ig | +0.043 |
an | +0.043 |
ish | +0.043 |
Neg logit contributions | |
Don | -0.049 |
My | -0.048 |
Would | -0.046 |
Here | -0.043 |
There | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e+01 |
Pos logit contributions | |
Don | +0.011 |
My | +0.011 |
Would | +0.011 |
It | +0.010 |
It | +0.010 |
Neg logit contributions | |
ies | -0.007 |
ig | -0.005 |
an | -0.005 |
ish | -0.005 |
ishing | -0.005 |
Token | |
Token ID | 220 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
- | +0.661 |
ies | +0.640 |
. | +0.634 |
fort | +0.616 |
eye | +0.614 |
Neg logit contributions | |
Don | -0.475 |
Here | -0.411 |
My | -0.409 |
Take | -0.396 |
Would | -0.388 |
Token |
|
Token ID | 198 |
Feature activation | +1.81e+01 |
Pos logit contributions | |
Don | +0.372 |
Would | +0.327 |
Here | +0.324 |
Looks | +0.322 |
Get | +0.320 |
Neg logit contributions | |
- | -0.519 |
ies | -0.519 |
. | -0.487 |
home | -0.450 |
ig | -0.446 |
Token |
|
Token ID | 198 |
Feature activation | +1.81e+01 |
Pos logit contributions | |
Don | +0.534 |
Here | +0.455 |
My | +0.447 |
Would | +0.433 |
Take | +0.433 |
Neg logit contributions | |
- | -0.891 |
ies | -0.853 |
. | -0.777 |
ig | -0.762 |
ie | -0.757 |
Token | He |
Token ID | 1544 |
Feature activation | +2.21e+01 |
Pos logit contributions | |
Don | +0.466 |
Here | +0.419 |
Would | +0.418 |
My | +0.409 |
Well | +0.391 |
Neg logit contributions | |
- | -0.905 |
ies | -0.875 |
fort | -0.806 |
ig | -0.796 |
castle | -0.778 |
Token | went |
Token ID | 1816 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
Don | +0.807 |
My | +0.744 |
Would | +0.731 |
Well | +0.720 |
Here | +0.701 |
Neg logit contributions | |
ies | -0.963 |
- | -0.939 |
ig | -0.840 |
ish | -0.824 |
an | -0.818 |
Token | to |
Token ID | 284 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
Don | +0.157 |
Would | +0.150 |
My | +0.149 |
There | +0.138 |
It | +0.135 |
Neg logit contributions | |
ies | -0.169 |
an | -0.153 |
ish | -0.152 |
ig | -0.147 |
decision | -0.147 |
Token | the |
Token ID | 262 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
ies | +0.048 |
- | +0.045 |
eye | +0.041 |
ig | +0.040 |
decision | +0.040 |
Neg logit contributions | |
Don | -0.046 |
Well | -0.041 |
Would | -0.041 |
My | -0.041 |
There | -0.039 |
Token | park |
Token ID | 3952 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
Don | +0.491 |
My | +0.476 |
Would | +0.463 |
Here | +0.441 |
There | +0.439 |
Neg logit contributions | |
ies | -0.358 |
- | -0.312 |
ig | -0.289 |
ish | -0.283 |
an | -0.280 |
Token | and |
Token ID | 290 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
ies | +0.424 |
ig | +0.373 |
ish | +0.366 |
an | +0.361 |
- | +0.351 |
Neg logit contributions | |
Don | -0.542 |
My | -0.516 |
Would | -0.509 |
Well | -0.483 |
It | -0.475 |
Token | , |
Token ID | 11 |
Feature activation | -1.76e+01 |
Pos logit contributions | |
ies | +0.275 |
an | +0.247 |
- | +0.240 |
ig | +0.232 |
ish | +0.231 |
Neg logit contributions | |
Don | -0.358 |
My | -0.338 |
Would | -0.334 |
Well | -0.321 |
Here | -0.320 |
Token | dad |
Token ID | 9955 |
Feature activation | -1.79e+01 |
Pos logit contributions | |
ies | +0.790 |
- | +0.715 |
decision | +0.636 |
ie | +0.631 |
eye | +0.630 |
Neg logit contributions | |
Don | -0.647 |
Would | -0.608 |
My | -0.591 |
Here | -0.574 |
Well | -0.566 |
Token | said |
Token ID | 531 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
ies | +0.767 |
- | +0.710 |
ig | +0.665 |
ish | +0.658 |
an | +0.653 |
Neg logit contributions | |
Don | -0.669 |
My | -0.622 |
Would | -0.615 |
Well | -0.591 |
There | -0.571 |
Token | : |
Token ID | 25 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
Don | +0.176 |
My | +0.167 |
Would | +0.164 |
Well | +0.160 |
There | +0.157 |
Neg logit contributions | |
ies | -0.126 |
- | -0.119 |
ig | -0.101 |
ish | -0.100 |
an | -0.097 |
Token | " |
Token ID | 366 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
- | +0.089 |
ies | +0.083 |
, | +0.079 |
. | +0.077 |
! | +0.072 |
Neg logit contributions | |
Don | -0.075 |
My | -0.069 |
There | -0.067 |
Well | -0.065 |
Here | -0.065 |
Token | You |
Token ID | 1639 |
Feature activation | +2.13e+01 |
Pos logit contributions | |
Did | +0.160 |
Higher | +0.158 |
Don | +0.145 |
Their | +0.140 |
Everything | +0.139 |
Neg logit contributions | |
ies | -0.557 |
- | -0.544 |
ie | -0.508 |
ater | -0.504 |
ishing | -0.496 |
Token | were |
Token ID | 547 |
Feature activation | +1.91e+01 |
Pos logit contributions | |
Don | +0.767 |
Would | +0.714 |
My | +0.705 |
Well | +0.668 |
Here | +0.652 |
Neg logit contributions | |
ies | -0.952 |
- | -0.886 |
ish | -0.834 |
ig | -0.822 |
eye | -0.818 |
Token | very |
Token ID | 845 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
Don | +0.743 |
My | +0.695 |
Would | +0.685 |
Well | +0.658 |
There | +0.642 |
Neg logit contributions | |
ies | -0.799 |
- | -0.732 |
ig | -0.686 |
ish | -0.677 |
an | -0.677 |
Token | brave |
Token ID | 14802 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
Don | +0.131 |
My | +0.123 |
Would | +0.122 |
Well | +0.118 |
There | +0.115 |
Neg logit contributions | |
ies | -0.102 |
- | -0.093 |
ig | -0.085 |
ish | -0.084 |
an | -0.083 |
Token | trying |
Token ID | 2111 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
Don | +0.043 |
Would | +0.040 |
Well | +0.040 |
There | +0.039 |
My | +0.039 |
Neg logit contributions | |
ies | -0.023 |
an | -0.022 |
ig | -0.022 |
ish | -0.020 |
eye | -0.019 |
Token | all |
Token ID | 477 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
ies | +0.235 |
an | +0.220 |
ig | +0.205 |
ish | +0.205 |
- | +0.205 |
Neg logit contributions | |
Don | -0.219 |
My | -0.208 |
It | -0.204 |
Would | -0.199 |
There | -0.192 |
Token | me |
Token ID | 502 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
ies | +0.017 |
- | +0.016 |
ig | +0.014 |
ish | +0.014 |
an | +0.014 |
Neg logit contributions | |
Don | -0.020 |
My | -0.019 |
Would | -0.019 |
Well | -0.018 |
There | -0.017 |
Token | ." |
Token ID | 526 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
ies | +0.090 |
decision | +0.079 |
an | +0.077 |
ishing | +0.077 |
choice | +0.077 |
Neg logit contributions | |
Don | -0.135 |
My | -0.133 |
It | -0.128 |
I | -0.122 |
There | -0.121 |
Token | |
Token ID | 220 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
- | +0.587 |
ies | +0.578 |
eye | +0.538 |
. | +0.537 |
home | +0.531 |
Neg logit contributions | |
Don | -0.344 |
Looks | -0.317 |
Would | -0.315 |
Ah | -0.306 |
Take | -0.305 |
Token |
|
Token ID | 198 |
Feature activation | +1.75e+01 |
Pos logit contributions | |
Don | +0.344 |
Would | +0.314 |
Looks | +0.299 |
Well | +0.294 |
My | +0.294 |
Neg logit contributions | |
ies | -0.502 |
- | -0.490 |
. | -0.461 |
home | -0.433 |
ig | -0.428 |
Token |
|
Token ID | 198 |
Feature activation | +1.74e+01 |
Pos logit contributions | |
Don | +0.566 |
Looks | +0.565 |
Take | +0.522 |
Would | +0.520 |
Well | +0.518 |
Neg logit contributions | |
- | -0.718 |
ies | -0.686 |
. | -0.629 |
, | -0.606 |
home | -0.584 |
Token | Her |
Token ID | 9360 |
Feature activation | +2.50e+01 |
Pos logit contributions | |
Don | +0.481 |
Would | +0.472 |
Did | +0.460 |
Well | +0.440 |
Ah | +0.421 |
Neg logit contributions | |
- | -0.832 |
ies | -0.814 |
ig | -0.709 |
an | -0.704 |
ater | -0.690 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.43e+01 |
Pos logit contributions | |
My | +0.637 |
Don | +0.617 |
Me | +0.597 |
Well | +0.560 |
Would | +0.554 |
Neg logit contributions | |
ishing | -1.156 |
ies | -1.150 |
an | -1.127 |
ish | -1.120 |
ater | -1.108 |
Token | my |
Token ID | 1820 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
ies | +0.540 |
- | +0.494 |
ig | +0.457 |
ish | +0.452 |
an | +0.448 |
Neg logit contributions | |
Don | -0.610 |
My | -0.571 |
Would | -0.565 |
Well | -0.546 |
There | -0.530 |
Token | replied |
Token ID | 8712 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
ies | +0.195 |
- | +0.180 |
ig | +0.171 |
ish | +0.169 |
an | +0.168 |
Neg logit contributions | |
Don | -0.142 |
My | -0.130 |
Would | -0.128 |
Well | -0.123 |
It | -0.118 |
Token | , |
Token ID | 11 |
Feature activation | -1.68e+01 |
Pos logit contributions | |
ies | +0.048 |
an | +0.041 |
ig | +0.038 |
ish | +0.037 |
eye | +0.037 |
Neg logit contributions | |
Don | -0.136 |
My | -0.129 |
Would | -0.126 |
It | -0.124 |
Well | -0.123 |
Token | " |
Token ID | 366 |
Feature activation | +1.16e+01 |
Pos logit contributions | |
- | +0.740 |
ies | +0.733 |
, | +0.618 |
. | +0.601 |
ish | +0.592 |
Neg logit contributions | |
Don | -0.548 |
Here | -0.525 |
My | -0.520 |
Would | -0.511 |
There | -0.508 |
Token | ara |
Token ID | 3301 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Don | +0.223 |
My | +0.208 |
Would | +0.207 |
Well | +0.200 |
There | +0.193 |
Neg logit contributions | |
ies | -0.203 |
- | -0.187 |
ig | -0.173 |
ish | -0.171 |
an | -0.169 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
Don | +0.066 |
My | +0.063 |
Would | +0.061 |
Well | +0.060 |
Come | +0.059 |
Neg logit contributions | |
ies | -0.049 |
- | -0.042 |
decision | -0.041 |
ig | -0.040 |
ish | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
Don | +0.582 |
My | +0.555 |
Would | +0.543 |
Well | +0.517 |
There | +0.515 |
Neg logit contributions | |
ies | -0.516 |
- | -0.462 |
ig | -0.450 |
an | -0.445 |
ish | -0.437 |
Token | smile |
Token ID | 8212 |
Feature activation | -2.26e+01 |
Pos logit contributions | |
ies | +0.034 |
- | +0.031 |
ig | +0.029 |
an | +0.028 |
ish | +0.028 |
Neg logit contributions | |
Don | -0.042 |
My | -0.040 |
Would | -0.039 |
Well | -0.038 |
There | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
ig | +0.554 |
ies | +0.554 |
an | +0.541 |
decision | +0.539 |
choice | +0.511 |
Neg logit contributions | |
Don | -1.153 |
My | -1.134 |
It | -1.067 |
Would | -1.052 |
There | -1.046 |
Token | She |
Token ID | 1375 |
Feature activation | +2.61e+01 |
Pos logit contributions | |
ies | +0.786 |
- | +0.769 |
eye | +0.737 |
ish | +0.711 |
. | +0.708 |
Neg logit contributions | |
Don | -0.342 |
Well | -0.274 |
My | -0.265 |
Would | -0.264 |
Ah | -0.250 |
Token | smiled |
Token ID | 13541 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
Don | +0.875 |
Would | +0.805 |
My | +0.786 |
Well | +0.759 |
Here | +0.745 |
Neg logit contributions | |
ies | -1.225 |
- | -1.172 |
ig | -1.067 |
an | -1.042 |
eye | -1.042 |
Token | when |
Token ID | 618 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
Don | +0.128 |
My | +0.124 |
Would | +0.115 |
It | +0.115 |
There | +0.113 |
Neg logit contributions | |
ies | -0.079 |
ig | -0.077 |
an | -0.074 |
ish | -0.070 |
- | -0.069 |
Token | she |
Token ID | 673 |
Feature activation | +1.76e+01 |
Pos logit contributions | |
ies | +0.169 |
- | +0.155 |
ie | +0.136 |
ig | +0.136 |
castle | +0.134 |
Neg logit contributions | |
Don | -0.119 |
Would | -0.110 |
Well | -0.106 |
My | -0.103 |
Here | -0.102 |
Token | played |
Token ID | 2826 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
Don | +0.607 |
My | +0.555 |
Would | +0.551 |
Well | +0.526 |
There | +0.509 |
Neg logit contributions | |
ies | -0.807 |
- | -0.761 |
ig | -0.709 |
ish | -0.696 |
an | -0.690 |
Token | with |
Token ID | 351 |
Feature activation | -3.91e+00 |
Pos logit contributions | |
Don | +0.085 |
My | +0.085 |
It | +0.082 |
Would | +0.081 |
There | +0.079 |
Neg logit contributions | |
ies | -0.086 |
an | -0.077 |
ish | -0.076 |
decision | -0.074 |
ig | -0.073 |
Token | it |
Token ID | 340 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
- | +0.101 |
ies | +0.101 |
ig | +0.086 |
ie | +0.085 |
eye | +0.083 |
Neg logit contributions | |
Don | -0.080 |
Would | -0.072 |
Well | -0.070 |
There | -0.068 |
Here | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | +1.57e+01 |
Pos logit contributions | |
Don | +0.041 |
Would | +0.039 |
Here | +0.038 |
There | +0.038 |
My | +0.038 |
Neg logit contributions | |
ies | -0.047 |
- | -0.044 |
ie | -0.039 |
ish | -0.038 |
eye | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
My | +0.735 |
Don | +0.732 |
Would | +0.684 |
Did | +0.684 |
She | +0.683 |
Neg logit contributions | |
ig | -0.418 |
ies | -0.379 |
an | -0.362 |
ishing | -0.362 |
breath | -0.358 |
Token | Annie |
Token ID | 25860 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
ies | +0.415 |
- | +0.378 |
ie | +0.371 |
, | +0.318 |
. | +0.307 |
Neg logit contributions | |
Don | -0.394 |
Here | -0.380 |
There | -0.371 |
Well | -0.357 |
Would | -0.356 |
Token | said |
Token ID | 531 |
Feature activation | +5.61e+00 |
Pos logit contributions | |
Don | +0.362 |
My | +0.333 |
Would | +0.330 |
Well | +0.316 |
There | +0.309 |
Neg logit contributions | |
ies | -0.454 |
- | -0.424 |
ig | -0.393 |
ish | -0.390 |
an | -0.386 |
Token | it |
Token ID | 340 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Don | +0.224 |
My | +0.208 |
Would | +0.207 |
Well | +0.199 |
There | +0.194 |
Neg logit contributions | |
ies | -0.227 |
- | -0.212 |
ig | -0.194 |
ish | -0.192 |
an | -0.189 |
Token | was |
Token ID | 373 |
Feature activation | +1.57e+01 |
Pos logit contributions | |
Don | +0.038 |
Would | +0.035 |
My | +0.034 |
Well | +0.033 |
There | +0.033 |
Neg logit contributions | |
ies | -0.049 |
- | -0.047 |
ish | -0.042 |
an | -0.041 |
ig | -0.041 |
Token | called |
Token ID | 1444 |
Feature activation | -5.56e+00 |
Pos logit contributions | |
Don | +0.498 |
My | +0.475 |
Would | +0.451 |
It | +0.415 |
Well | +0.409 |
Neg logit contributions | |
ies | -0.737 |
- | -0.689 |
ig | -0.677 |
ish | -0.668 |
an | -0.668 |
Token | � |
Token ID | 6184 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
ies | +0.271 |
- | +0.249 |
ig | +0.240 |
ish | +0.238 |
an | +0.236 |
Neg logit contributions | |
Don | -0.175 |
My | -0.163 |
Would | -0.160 |
Well | -0.147 |
It | -0.146 |
Token | � |
Token ID | 95 |
Feature activation | +6.65e-01 |
Pos logit contributions | |
Don | +0.192 |
My | +0.179 |
Would | +0.179 |
Well | +0.170 |
It | +0.164 |
Neg logit contributions | |
ies | -0.230 |
- | -0.204 |
ish | -0.193 |
ig | -0.192 |
an | -0.191 |
Token | € |
Token ID | 26391 |
Feature activation | +8.00e+00 |
Pos logit contributions | |
Don | +0.024 |
My | +0.022 |
Would | +0.021 |
Well | +0.021 |
There | +0.020 |
Neg logit contributions | |
- | -0.028 |
ies | -0.027 |
an | -0.025 |
ig | -0.025 |
eye | -0.024 |
Token | big |
Token ID | 1263 |
Feature activation | -8.00e-01 |
Pos logit contributions | |
Don | +0.468 |
Here | +0.440 |
My | +0.434 |
It | +0.428 |
I | +0.421 |
Neg logit contributions | |
ig | -0.336 |
ies | -0.336 |
- | -0.316 |
ie | -0.294 |
ishing | -0.293 |
Token | t |
Token ID | 256 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
ies | +0.023 |
ishing | +0.021 |
ig | +0.021 |
- | +0.020 |
an | +0.019 |
Neg logit contributions | |
Don | -0.035 |
It | -0.033 |
I | -0.033 |
It | -0.033 |
Look | -0.032 |
Token | arp |
Token ID | 5117 |
Feature activation | -2.33e+01 |
Pos logit contributions | |
ies | +0.450 |
decision | +0.442 |
ig | +0.442 |
problem | +0.412 |
choice | +0.411 |
Neg logit contributions | |
My | -0.604 |
I | -0.580 |
Don | -0.572 |
It | -0.559 |
Well | -0.544 |
Token | . |
Token ID | 13 |
Feature activation | -2.09e+01 |
Pos logit contributions | |
ies | +0.673 |
- | +0.610 |
ig | +0.588 |
decision | +0.574 |
choice | +0.554 |
Neg logit contributions | |
Don | -1.116 |
My | -1.082 |
Would | -1.025 |
Well | -1.020 |
It | -0.998 |
Token | They |
Token ID | 1119 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
ies | +0.948 |
- | +0.942 |
. | +0.895 |
, | +0.887 |
eye | +0.879 |
Neg logit contributions | |
Don | -0.501 |
Well | -0.421 |
My | -0.407 |
Listen | -0.393 |
Take | -0.391 |
Token | dragged |
Token ID | 17901 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
Don | +0.473 |
My | +0.442 |
Would | +0.432 |
Well | +0.416 |
There | +0.404 |
Neg logit contributions | |
ies | -0.547 |
- | -0.510 |
ig | -0.477 |
ish | -0.469 |
an | -0.469 |
Token | it |
Token ID | 340 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
Don | +0.096 |
My | +0.088 |
Would | +0.086 |
Well | +0.081 |
There | +0.081 |
Neg logit contributions | |
ies | -0.143 |
- | -0.130 |
ig | -0.127 |
an | -0.125 |
ish | -0.125 |
Token | to |
Token ID | 284 |
Feature activation | -6.24e+00 |
Pos logit contributions | |
ies | +0.169 |
an | +0.149 |
ish | +0.144 |
decision | +0.143 |
ater | +0.143 |
Neg logit contributions | |
Don | -0.164 |
My | -0.150 |
There | -0.148 |
It | -0.140 |
Would | -0.138 |
Token | the |
Token ID | 262 |
Feature activation | +5.19e+00 |
Pos logit contributions | |
ies | +0.254 |
- | +0.239 |
eye | +0.220 |
decision | +0.212 |
choice | +0.211 |
Neg logit contributions | |
Don | -0.241 |
Well | -0.221 |
My | -0.221 |
Would | -0.221 |
Did | -0.207 |
Token | sandbox |
Token ID | 35204 |
Feature activation | -2.56e+01 |
Pos logit contributions | |
Don | +0.253 |
My | +0.246 |
Would | +0.239 |
Here | +0.237 |
I | +0.231 |
Neg logit contributions | |
ies | -0.169 |
- | -0.144 |
ig | -0.139 |
ishing | -0.134 |
ish | -0.127 |
Token | and |
Token ID | 290 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
ies | +0.722 |
ig | +0.721 |
ish | +0.679 |
ishing | +0.640 |
an | +0.639 |
Neg logit contributions | |
Don | -1.159 |
My | -1.113 |
It | -1.084 |
Would | -1.079 |
There | -1.056 |
Token | she |
Token ID | 673 |
Feature activation | +1.69e+01 |
Pos logit contributions | |
Don | +0.525 |
Would | +0.478 |
My | +0.472 |
Well | +0.449 |
Here | +0.436 |
Neg logit contributions | |
ies | -0.784 |
- | -0.733 |
ig | -0.687 |
ish | -0.658 |
eye | -0.654 |
Token | got |
Token ID | 1392 |
Feature activation | +5.82e+00 |
Pos logit contributions | |
Don | +0.597 |
My | +0.545 |
Would | +0.543 |
Well | +0.518 |
Here | +0.505 |
Neg logit contributions | |
ies | -0.769 |
- | -0.728 |
ig | -0.676 |
ish | -0.664 |
eye | -0.656 |
Token | to |
Token ID | 284 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
Don | +0.294 |
My | +0.280 |
Would | +0.276 |
Well | +0.268 |
There | +0.265 |
Neg logit contributions | |
ies | -0.175 |
- | -0.155 |
an | -0.142 |
ig | -0.142 |
ish | -0.140 |
Token | school |
Token ID | 1524 |
Feature activation | -9.47e+00 |
Pos logit contributions | |
ies | +0.059 |
- | +0.053 |
ig | +0.048 |
ish | +0.048 |
an | +0.047 |
Neg logit contributions | |
Don | -0.082 |
My | -0.077 |
Would | -0.077 |
Well | -0.074 |
There | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | -1.64e+01 |
Pos logit contributions | |
ies | +0.195 |
an | +0.158 |
ish | +0.150 |
- | +0.148 |
ig | +0.141 |
Neg logit contributions | |
Don | -0.557 |
My | -0.541 |
Would | -0.533 |
Well | -0.520 |
There | -0.510 |
Token | she |
Token ID | 673 |
Feature activation | +1.70e+01 |
Pos logit contributions | |
ies | +0.714 |
- | +0.654 |
ie | +0.596 |
ig | +0.591 |
ish | +0.575 |
Neg logit contributions | |
Don | -0.559 |
Would | -0.527 |
Here | -0.508 |
Well | -0.506 |
My | -0.494 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.25e+00 |
Pos logit contributions | |
Don | +0.598 |
Would | +0.544 |
My | +0.541 |
Here | +0.519 |
Well | +0.518 |
Neg logit contributions | |
ies | -0.780 |
- | -0.740 |
ig | -0.680 |
ish | -0.666 |
eye | -0.658 |
Token | lots |
Token ID | 6041 |
Feature activation | +9.92e+00 |
Pos logit contributions | |
Don | +0.283 |
Would | +0.260 |
My | +0.259 |
Well | +0.250 |
There | +0.242 |
Neg logit contributions | |
ies | -0.301 |
- | -0.283 |
ig | -0.257 |
ish | -0.253 |
eye | -0.252 |
Token | of |
Token ID | 286 |
Feature activation | +4.41e+00 |
Pos logit contributions | |
Don | +0.333 |
My | +0.329 |
Would | +0.315 |
Well | +0.308 |
There | +0.299 |
Neg logit contributions | |
ies | -0.420 |
an | -0.408 |
ish | -0.404 |
ig | -0.398 |
fort | -0.382 |
Token | other |
Token ID | 584 |
Feature activation | +8.61e+00 |
Pos logit contributions | |
Don | +0.207 |
My | +0.200 |
Would | +0.196 |
Well | +0.187 |
Here | +0.184 |
Neg logit contributions | |
ies | -0.144 |
ig | -0.129 |
ish | -0.127 |
an | -0.127 |
- | -0.125 |
Token | kids |
Token ID | 3988 |
Feature activation | -1.42e+01 |
Pos logit contributions | |
My | +0.341 |
Don | +0.338 |
Would | +0.321 |
Well | +0.312 |
Come | +0.305 |
Neg logit contributions | |
an | -0.292 |
ies | -0.289 |
ig | -0.285 |
ish | -0.282 |
egg | -0.267 |
Token | You |
Token ID | 921 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
eye | +0.648 |
- | +0.647 |
ish | +0.624 |
ies | +0.621 |
an | +0.590 |
Neg logit contributions | |
Don | -0.482 |
Would | -0.422 |
Over | -0.410 |
Well | -0.404 |
My | -0.402 |
Token | will |
Token ID | 481 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
Don | +0.497 |
Would | +0.462 |
My | +0.449 |
Well | +0.433 |
Here | +0.418 |
Neg logit contributions | |
ies | -0.676 |
- | -0.632 |
ish | -0.594 |
ig | -0.587 |
eye | -0.578 |
Token | get |
Token ID | 651 |
Feature activation | -5.76e-03 |
Pos logit contributions | |
Don | +0.053 |
My | +0.049 |
Would | +0.048 |
Well | +0.045 |
There | +0.045 |
Neg logit contributions | |
ies | -0.075 |
- | -0.070 |
ig | -0.067 |
ish | -0.066 |
an | -0.065 |
Token | sick |
Token ID | 6639 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
ies | +0.000 |
- | +0.000 |
ig | +0.000 |
an | +0.000 |
ish | +0.000 |
Neg logit contributions | |
Don | -0.000 |
My | -0.000 |
Would | -0.000 |
Well | -0.000 |
There | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e+01 |
Pos logit contributions | |
Don | +0.121 |
I | +0.114 |
There | +0.112 |
Why | +0.111 |
It | +0.109 |
Neg logit contributions | |
ig | -0.099 |
ies | -0.096 |
an | -0.093 |
ish | -0.086 |
decision | -0.085 |
Token | And |
Token ID | 843 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
eye | +0.638 |
- | +0.634 |
ies | +0.599 |
ish | +0.591 |
ig | +0.565 |
Neg logit contributions | |
Don | -0.471 |
Would | -0.420 |
Over | -0.414 |
Well | -0.403 |
My | -0.395 |
Token | you |
Token ID | 345 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
Don | +0.395 |
My | +0.365 |
Would | +0.361 |
Well | +0.346 |
There | +0.334 |
Neg logit contributions | |
ies | -0.497 |
- | -0.463 |
ig | -0.433 |
ish | -0.429 |
an | -0.425 |
Token | have |
Token ID | 423 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
Don | +0.296 |
My | +0.272 |
Would | +0.268 |
Well | +0.256 |
There | +0.248 |
Neg logit contributions | |
ies | -0.432 |
- | -0.403 |
ig | -0.380 |
ish | -0.375 |
an | -0.374 |
Token | to |
Token ID | 284 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
Don | +0.539 |
My | +0.507 |
Would | +0.505 |
Well | +0.489 |
There | +0.474 |
Neg logit contributions | |
ies | -0.388 |
- | -0.355 |
ig | -0.319 |
ish | -0.317 |
an | -0.310 |
Token | say |
Token ID | 910 |
Feature activation | -6.41e+00 |
Pos logit contributions | |
ies | +0.061 |
- | +0.056 |
ig | +0.055 |
ishing | +0.054 |
ish | +0.053 |
Neg logit contributions | |
Don | -0.044 |
My | -0.042 |
It | -0.040 |
Would | -0.039 |
That | -0.039 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
ies | +0.300 |
- | +0.269 |
ig | +0.261 |
an | +0.261 |
ish | +0.259 |
Neg logit contributions | |
Don | -0.214 |
My | -0.203 |
Would | -0.196 |
It | -0.184 |
Well | -0.184 |
Token | 's |
Token ID | 338 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
My | +0.119 |
Don | +0.115 |
There | +0.106 |
Would | +0.104 |
It | +0.100 |
Neg logit contributions | |
ies | -0.200 |
an | -0.179 |
choice | -0.177 |
decision | -0.177 |
- | -0.175 |
Token | bed |
Token ID | 3996 |
Feature activation | -2.12e+01 |
Pos logit contributions | |
My | +0.329 |
Don | +0.306 |
It | +0.303 |
She | +0.302 |
It | +0.299 |
Neg logit contributions | |
ies | -0.223 |
ishing | -0.195 |
- | -0.189 |
ig | -0.181 |
ish | -0.178 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e+01 |
Pos logit contributions | |
ies | +0.540 |
ish | +0.512 |
an | +0.483 |
ig | +0.477 |
ishing | +0.431 |
Neg logit contributions | |
Don | -1.072 |
My | -1.020 |
There | -1.016 |
It | -1.003 |
Well | -0.986 |
Token | Sue |
Token ID | 26089 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
- | +0.624 |
ies | +0.604 |
eye | +0.555 |
fort | +0.537 |
decision | +0.535 |
Neg logit contributions | |
My | -0.504 |
Don | -0.502 |
Would | -0.476 |
Did | -0.451 |
Ah | -0.442 |
Token | was |
Token ID | 373 |
Feature activation | +1.54e+01 |
Pos logit contributions | |
Don | +0.146 |
Would | +0.135 |
My | +0.134 |
Well | +0.129 |
Here | +0.125 |
Neg logit contributions | |
ies | -0.183 |
- | -0.172 |
ig | -0.159 |
ish | -0.157 |
an | -0.156 |
Token | so |
Token ID | 523 |
Feature activation | +3.49e+00 |
Pos logit contributions | |
Don | +0.622 |
My | +0.584 |
Would | +0.575 |
Well | +0.552 |
There | +0.539 |
Neg logit contributions | |
ies | -0.625 |
- | -0.573 |
ig | -0.536 |
ish | -0.530 |
an | -0.526 |
Token | happy |
Token ID | 3772 |
Feature activation | +5.29e+00 |
Pos logit contributions | |
Don | +0.167 |
Would | +0.161 |
Well | +0.156 |
My | +0.156 |
Here | +0.152 |
Neg logit contributions | |
ies | -0.117 |
- | -0.105 |
eye | -0.090 |
ig | -0.087 |
an | -0.087 |
Token | and |
Token ID | 290 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
Don | +0.225 |
My | +0.205 |
It | +0.202 |
Would | +0.197 |
Well | +0.195 |
Neg logit contributions | |
an | -0.185 |
ig | -0.182 |
ish | -0.181 |
ies | -0.179 |
eye | -0.167 |
Token | stopped |
Token ID | 5025 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
ies | +0.459 |
- | +0.427 |
eye | +0.379 |
ish | +0.366 |
an | +0.365 |
Neg logit contributions | |
Don | -0.368 |
Would | -0.354 |
My | -0.340 |
Well | -0.336 |
Here | -0.333 |
Token | weeping |
Token ID | 45705 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
Don | +0.138 |
My | +0.131 |
Would | +0.127 |
Well | +0.119 |
It | +0.118 |
Neg logit contributions | |
ies | -0.123 |
- | -0.111 |
an | -0.109 |
ig | -0.108 |
ish | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
an | +0.066 |
ig | +0.066 |
ies | +0.066 |
ish | +0.060 |
ishing | +0.056 |
Neg logit contributions | |
Don | -0.109 |
My | -0.102 |
She | -0.102 |
It | -0.098 |
It | -0.097 |
Token | close |
Token ID | 1969 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
ies | +0.268 |
ig | +0.267 |
decision | +0.262 |
an | +0.261 |
ater | +0.253 |
Neg logit contributions | |
Don | -0.276 |
My | -0.269 |
There | -0.263 |
Would | -0.257 |
I | -0.254 |
Token | to |
Token ID | 284 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
ies | +0.109 |
an | +0.105 |
ish | +0.103 |
ig | +0.102 |
- | +0.100 |
Neg logit contributions | |
Don | -0.098 |
My | -0.090 |
There | -0.089 |
Would | -0.087 |
It | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | +7.70e+00 |
Pos logit contributions | |
ies | +0.172 |
- | +0.159 |
ig | +0.155 |
an | +0.152 |
ish | +0.152 |
Neg logit contributions | |
Don | -0.108 |
My | -0.105 |
Would | -0.100 |
It | -0.091 |
I | -0.089 |
Token | white |
Token ID | 2330 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
Don | +0.341 |
Would | +0.322 |
Here | +0.322 |
My | +0.317 |
I | +0.316 |
Neg logit contributions | |
ies | -0.285 |
ishing | -0.248 |
- | -0.246 |
ig | -0.242 |
ish | -0.232 |
Token | dog |
Token ID | 3290 |
Feature activation | -1.75e+01 |
Pos logit contributions | |
ies | +0.125 |
ishing | +0.117 |
decision | +0.110 |
ig | +0.109 |
ising | +0.101 |
Neg logit contributions | |
It | -0.137 |
I | -0.136 |
Would | -0.133 |
Let | -0.133 |
Look | -0.132 |
Token | again |
Token ID | 757 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
ies | +0.570 |
decision | +0.477 |
choice | +0.475 |
ishing | +0.471 |
an | +0.461 |
Neg logit contributions | |
Don | -0.802 |
My | -0.802 |
Would | -0.784 |
I | -0.726 |
Do | -0.724 |
Token | . |
Token ID | 13 |
Feature activation | -1.75e+01 |
Pos logit contributions | |
ies | +0.309 |
an | +0.299 |
ish | +0.289 |
ig | +0.279 |
decision | +0.267 |
Neg logit contributions | |
Don | -0.582 |
My | -0.560 |
It | -0.543 |
I | -0.537 |
There | -0.536 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
. | +0.659 |
eye | +0.643 |
- | +0.597 |
ies | +0.588 |
, | +0.582 |
Neg logit contributions | |
Don | -0.539 |
Looks | -0.501 |
My | -0.497 |
Would | -0.495 |
Take | -0.476 |
Token | L |
Token ID | 43 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
- | +0.132 |
ies | +0.131 |
ig | +0.121 |
ish | +0.116 |
an | +0.115 |
Neg logit contributions | |
Don | -0.085 |
My | -0.073 |
Would | -0.073 |
Well | -0.071 |
Looks | -0.065 |
Token | ila |
Token ID | 10102 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
Don | +0.487 |
My | +0.466 |
Would | +0.450 |
Well | +0.429 |
There | +0.427 |
Neg logit contributions | |
ies | -0.634 |
- | -0.589 |
ish | -0.573 |
ig | -0.561 |
ishing | -0.553 |
Token | and |
Token ID | 290 |
Feature activation | -1.22e+01 |
Pos logit contributions | |
Don | +0.585 |
My | +0.558 |
Would | +0.547 |
Well | +0.536 |
There | +0.527 |
Neg logit contributions | |
ies | -0.272 |
- | -0.231 |
ig | -0.211 |
ish | -0.207 |
an | -0.202 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.32e+01 |
Pos logit contributions | |
ies | +0.136 |
- | +0.121 |
an | +0.117 |
ig | +0.117 |
ish | +0.117 |
Neg logit contributions | |
Don | -0.176 |
My | -0.166 |
Would | -0.163 |
Well | -0.160 |
It | -0.158 |
Token |
|
Token ID | 198 |
Feature activation | +1.53e+01 |
Pos logit contributions | |
- | +0.976 |
. | +0.960 |
ies | +0.956 |
, | +0.905 |
home | +0.813 |
Neg logit contributions | |
Don | -0.685 |
Would | -0.682 |
Looks | -0.636 |
Take | -0.615 |
Over | -0.610 |
Token |
|
Token ID | 198 |
Feature activation | +1.53e+01 |
Pos logit contributions | |
Don | +0.481 |
Did | +0.475 |
Looks | +0.452 |
Well | +0.433 |
Would | +0.432 |
Neg logit contributions | |
- | -0.674 |
ies | -0.633 |
. | -0.617 |
an | -0.557 |
, | -0.537 |
Token | S |
Token ID | 50 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
Did | +0.573 |
Don | +0.530 |
Would | +0.528 |
Everything | +0.516 |
Higher | +0.494 |
Neg logit contributions | |
- | -0.606 |
ies | -0.594 |
an | -0.522 |
ig | -0.502 |
ater | -0.481 |
Token | ara |
Token ID | 3301 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Don | +0.083 |
My | +0.077 |
Would | +0.075 |
Well | +0.074 |
There | +0.073 |
Neg logit contributions | |
ies | -0.095 |
- | -0.086 |
ish | -0.082 |
an | -0.081 |
ig | -0.080 |
Token | and |
Token ID | 290 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
ies | +0.046 |
- | +0.040 |
ig | +0.037 |
ish | +0.036 |
an | +0.036 |
Neg logit contributions | |
Don | -0.080 |
My | -0.076 |
Would | -0.075 |
Well | -0.073 |
It | -0.071 |
Token | Tom |
Token ID | 4186 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ies | +0.632 |
- | +0.592 |
ish | +0.552 |
eye | +0.547 |
ig | +0.542 |
Neg logit contributions | |
Don | -0.413 |
Would | -0.385 |
My | -0.371 |
Well | -0.363 |
There | -0.360 |
Token | look |
Token ID | 804 |
Feature activation | -7.42e+00 |
Pos logit contributions | |
Don | +0.178 |
My | +0.169 |
Would | +0.161 |
Well | +0.157 |
There | +0.154 |
Neg logit contributions | |
ies | -0.192 |
- | -0.178 |
ig | -0.167 |
an | -0.163 |
ish | -0.160 |
Token | at |
Token ID | 379 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
ies | +0.262 |
an | +0.258 |
ig | +0.257 |
island | +0.240 |
ater | +0.239 |
Neg logit contributions | |
Don | -0.280 |
It | -0.253 |
My | -0.253 |
Would | -0.251 |
From | -0.241 |
Token | Mom |
Token ID | 11254 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
ies | +0.257 |
- | +0.240 |
ig | +0.225 |
ish | +0.222 |
an | +0.221 |
Neg logit contributions | |
Don | -0.208 |
My | -0.194 |
Would | -0.190 |
Well | -0.183 |
There | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | -1.74e+01 |
Pos logit contributions | |
ies | +0.115 |
ig | +0.107 |
an | +0.104 |
island | +0.096 |
ater | +0.095 |
Neg logit contributions | |
My | -0.222 |
Don | -0.222 |
Would | -0.214 |
That | -0.209 |
It | -0.206 |
Token |
|
Token ID | 198 |
Feature activation | +1.81e+01 |
Pos logit contributions | |
- | +0.652 |
. | +0.613 |
, | +0.593 |
! | +0.582 |
eye | +0.579 |
Neg logit contributions | |
Don | -0.459 |
My | -0.422 |
Our | -0.389 |
Did | -0.378 |
Would | -0.377 |
Token |
|
Token ID | 198 |
Feature activation | +1.81e+01 |
Pos logit contributions | |
Don | +0.576 |
Looks | +0.517 |
My | +0.510 |
Did | +0.496 |
Well | +0.493 |
Neg logit contributions | |
- | -0.842 |
ies | -0.775 |
. | -0.720 |
home | -0.711 |
an | -0.709 |
Token | Back |
Token ID | 7282 |
Feature activation | +1.91e+01 |
Pos logit contributions | |
Did | +0.474 |
Don | +0.429 |
My | +0.421 |
Would | +0.411 |
Well | +0.401 |
Neg logit contributions | |
- | -0.859 |
ies | -0.753 |
an | -0.716 |
mess | -0.685 |
! | -0.683 |
Token | at |
Token ID | 379 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
Don | +0.965 |
My | +0.919 |
Well | +0.894 |
There | +0.882 |
Would | +0.879 |
Neg logit contributions | |
ies | -0.536 |
- | -0.440 |
an | -0.434 |
decision | -0.434 |
ig | -0.421 |
Token | home |
Token ID | 1363 |
Feature activation | -2.07e+01 |
Pos logit contributions | |
ies | +0.102 |
- | +0.091 |
ig | +0.082 |
ish | +0.081 |
an | +0.080 |
Neg logit contributions | |
Don | -0.177 |
My | -0.167 |
Would | -0.166 |
Well | -0.161 |
There | -0.158 |
Token | , |
Token ID | 11 |
Feature activation | -1.60e+01 |
Pos logit contributions | |
ies | +0.384 |
- | +0.304 |
an | +0.304 |
ish | +0.300 |
ig | +0.292 |
Neg logit contributions | |
Don | -1.223 |
My | -1.191 |
Would | -1.161 |
Well | -1.143 |
There | -1.129 |
Token | Tim |
Token ID | 5045 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
ies | +0.739 |
- | +0.699 |
ish | +0.626 |
decision | +0.615 |
eye | +0.610 |
Neg logit contributions | |
Don | -0.522 |
Would | -0.501 |
My | -0.473 |
Well | -0.463 |
Here | -0.461 |
Token | started |
Token ID | 2067 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
Don | +0.260 |
My | +0.244 |
Would | +0.235 |
Well | +0.223 |
There | +0.217 |
Neg logit contributions | |
ies | -0.365 |
- | -0.338 |
ig | -0.320 |
an | -0.317 |
ish | -0.317 |
Token | making |
Token ID | 1642 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
Don | +0.105 |
My | +0.100 |
Would | +0.098 |
Well | +0.092 |
It | +0.091 |
Neg logit contributions | |
ies | -0.094 |
- | -0.084 |
an | -0.083 |
ig | -0.082 |
ish | -0.080 |
Token | his |
Token ID | 465 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
ies | +0.091 |
ig | +0.082 |
- | +0.081 |
an | +0.081 |
ish | +0.078 |
Neg logit contributions | |
Don | -0.081 |
My | -0.077 |
Would | -0.075 |
It | -0.073 |
There | -0.072 |
Token | jam |
Token ID | 16853 |
Feature activation | -1.96e+01 |
Pos logit contributions | |
Don | +0.626 |
My | +0.613 |
Me | +0.607 |
Would | +0.588 |
I | +0.572 |
Neg logit contributions | |
ies | -0.470 |
- | -0.395 |
an | -0.379 |
ist | -0.375 |
ig | -0.371 |
Token | the |
Token ID | 262 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
Don | +0.148 |
My | +0.142 |
It | +0.135 |
Would | +0.131 |
That | +0.131 |
Neg logit contributions | |
ies | -0.216 |
ig | -0.191 |
- | -0.190 |
an | -0.188 |
ish | -0.186 |
Token | pastry |
Token ID | 41738 |
Feature activation | -1.89e+01 |
Pos logit contributions | |
Don | +0.465 |
Here | +0.442 |
Would | +0.437 |
That | +0.436 |
My | +0.436 |
Neg logit contributions | |
ies | -0.370 |
ig | -0.354 |
- | -0.319 |
an | -0.312 |
ish | -0.307 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
ies | +0.636 |
an | +0.526 |
ig | +0.524 |
ishing | +0.511 |
- | +0.494 |
Neg logit contributions | |
Don | -0.850 |
My | -0.830 |
I | -0.822 |
It | -0.809 |
That | -0.790 |
Token | He |
Token ID | 679 |
Feature activation | +2.31e+01 |
Pos logit contributions | |
- | +0.479 |
ies | +0.460 |
. | +0.449 |
mess | +0.434 |
, | +0.431 |
Neg logit contributions | |
Don | -0.332 |
Take | -0.306 |
Looks | -0.304 |
My | -0.303 |
Would | -0.301 |
Token | looked |
Token ID | 3114 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
Don | +0.879 |
Would | +0.834 |
My | +0.832 |
Here | +0.810 |
Well | +0.798 |
Neg logit contributions | |
ies | -0.942 |
- | -0.940 |
an | -0.773 |
eye | -0.770 |
mess | -0.768 |
Token | for |
Token ID | 329 |
Feature activation | -4.99e+00 |
Pos logit contributions | |
Don | +0.413 |
My | +0.377 |
Would | +0.371 |
Well | +0.359 |
It | +0.356 |
Neg logit contributions | |
ies | -0.260 |
ig | -0.230 |
- | -0.227 |
an | -0.220 |
decision | -0.210 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
ies | +0.220 |
- | +0.212 |
eye | +0.191 |
choice | +0.187 |
ig | +0.186 |
Neg logit contributions | |
Don | -0.177 |
Well | -0.161 |
Would | -0.161 |
My | -0.159 |
That | -0.149 |
Token | . |
Token ID | 13 |
Feature activation | -1.53e+01 |
Pos logit contributions | |
Don | +0.543 |
My | +0.536 |
There | +0.472 |
It | +0.471 |
Would | +0.461 |
Neg logit contributions | |
ies | -0.393 |
an | -0.345 |
ig | -0.329 |
ishing | -0.322 |
choice | -0.321 |
Token | He |
Token ID | 679 |
Feature activation | +2.30e+01 |
Pos logit contributions | |
- | +0.690 |
. | +0.656 |
ies | +0.645 |
eye | +0.619 |
, | +0.617 |
Neg logit contributions | |
Don | -0.406 |
Looks | -0.391 |
Take | -0.388 |
Would | -0.383 |
My | -0.376 |
Token | found |
Token ID | 1043 |
Feature activation | +7.24e+00 |
Pos logit contributions | |
Don | +0.899 |
Would | +0.874 |
My | +0.855 |
Here | +0.853 |
Well | +0.833 |
Neg logit contributions | |
ies | -0.909 |
- | -0.887 |
eye | -0.761 |
ig | -0.735 |
an | -0.733 |
Token | her |
Token ID | 607 |
Feature activation | +1.55e+01 |
Pos logit contributions | |
Don | +0.285 |
Would | +0.265 |
Well | +0.264 |
My | +0.256 |
That | +0.245 |
Neg logit contributions | |
ies | -0.293 |
- | -0.286 |
eye | -0.251 |
ig | -0.248 |
ish | -0.247 |
Token | The |
Token ID | 464 |
Feature activation | +2.65e+01 |
Pos logit contributions | |
Did | +0.399 |
Don | +0.378 |
Would | +0.361 |
Everything | +0.347 |
Well | +0.341 |
Neg logit contributions | |
- | -0.625 |
ies | -0.609 |
an | -0.553 |
ig | -0.545 |
ater | -0.542 |
Token | cat |
Token ID | 3797 |
Feature activation | -1.56e+01 |
Pos logit contributions | |
Don | +1.239 |
Would | +1.174 |
My | +1.147 |
Well | +1.127 |
Here | +1.108 |
Neg logit contributions | |
ies | -0.861 |
ishing | -0.754 |
- | -0.740 |
ish | -0.729 |
ig | -0.726 |
Token | thought |
Token ID | 1807 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
ies | +0.676 |
- | +0.627 |
ig | +0.585 |
ish | +0.580 |
an | +0.575 |
Neg logit contributions | |
Don | -0.579 |
My | -0.537 |
Would | -0.531 |
Well | -0.510 |
There | -0.493 |
Token | about |
Token ID | 546 |
Feature activation | -5.71e+00 |
Pos logit contributions | |
ies | +0.103 |
- | +0.087 |
ish | +0.086 |
an | +0.086 |
ishing | +0.082 |
Neg logit contributions | |
Don | -0.146 |
My | -0.137 |
Would | -0.135 |
Well | -0.133 |
It | -0.129 |
Token | this |
Token ID | 428 |
Feature activation | +3.00e+00 |
Pos logit contributions | |
ies | +0.242 |
- | +0.231 |
ig | +0.211 |
decision | +0.202 |
ish | +0.202 |
Neg logit contributions | |
Don | -0.212 |
My | -0.197 |
Would | -0.192 |
Well | -0.185 |
There | -0.184 |
Token | and |
Token ID | 290 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
Don | +0.136 |
Would | +0.128 |
My | +0.122 |
It | +0.121 |
Well | +0.120 |
Neg logit contributions | |
ies | -0.098 |
ishing | -0.083 |
an | -0.080 |
ig | -0.079 |
ish | -0.078 |
Token | realized |
Token ID | 6939 |
Feature activation | +5.78e+00 |
Pos logit contributions | |
ies | +0.572 |
- | +0.533 |
ie | +0.451 |
ish | +0.449 |
ig | +0.449 |
Neg logit contributions | |
Don | -0.504 |
My | -0.491 |
Here | -0.479 |
Would | -0.471 |
There | -0.467 |
Token | that |
Token ID | 326 |
Feature activation | -5.06e+00 |
Pos logit contributions | |
Don | +0.229 |
My | +0.214 |
Would | +0.210 |
Well | +0.203 |
It | +0.197 |
Neg logit contributions | |
ies | -0.236 |
- | -0.214 |
ish | -0.204 |
ig | -0.203 |
an | -0.203 |
Token | the |
Token ID | 262 |
Feature activation | +7.12e+00 |
Pos logit contributions | |
ies | +0.233 |
- | +0.218 |
ig | +0.201 |
ish | +0.198 |
eye | +0.195 |
Neg logit contributions | |
Don | -0.172 |
My | -0.158 |
Would | -0.157 |
Well | -0.151 |
There | -0.146 |
Token | bird |
Token ID | 6512 |
Feature activation | -2.20e+01 |
Pos logit contributions | |
Don | +0.343 |
Would | +0.320 |
My | +0.313 |
Here | +0.310 |
Well | +0.306 |
Neg logit contributions | |
ies | -0.239 |
- | -0.204 |
ishing | -0.202 |
ig | -0.200 |
ish | -0.191 |
Token | was |
Token ID | 373 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
ies | +1.102 |
- | +1.031 |
ig | +0.977 |
ish | +0.965 |
an | +0.961 |
Neg logit contributions | |
Don | -0.666 |
My | -0.607 |
Would | -0.594 |
Well | -0.567 |
There | -0.541 |
Token | sofa |
Token ID | 34902 |
Feature activation | -1.53e+01 |
Pos logit contributions | |
My | +0.432 |
Don | +0.422 |
Here | +0.411 |
Me | +0.407 |
That | +0.401 |
Neg logit contributions | |
ies | -0.339 |
ig | -0.309 |
ishing | -0.295 |
ish | -0.295 |
- | -0.281 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e+01 |
Pos logit contributions | |
ies | +0.494 |
ig | +0.417 |
ish | +0.416 |
ishing | +0.404 |
an | +0.395 |
Neg logit contributions | |
Don | -0.689 |
My | -0.662 |
Would | -0.656 |
It | -0.647 |
It | -0.638 |
Token | It |
Token ID | 632 |
Feature activation | +2.15e+01 |
Pos logit contributions | |
eye | +0.677 |
. | +0.655 |
- | +0.630 |
, | +0.611 |
window | +0.599 |
Neg logit contributions | |
Don | -0.403 |
My | -0.381 |
Would | -0.364 |
Well | -0.336 |
Looks | -0.335 |
Token | has |
Token ID | 468 |
Feature activation | +1.46e+01 |
Pos logit contributions | |
Don | +0.911 |
My | +0.846 |
Would | +0.840 |
Well | +0.815 |
There | +0.794 |
Neg logit contributions | |
ies | -0.818 |
- | -0.756 |
ig | -0.694 |
ish | -0.686 |
an | -0.681 |
Token | its |
Token ID | 663 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
Don | +0.619 |
My | +0.581 |
Would | +0.574 |
Well | +0.555 |
There | +0.540 |
Neg logit contributions | |
ies | -0.558 |
- | -0.513 |
ig | -0.474 |
ish | -0.468 |
an | -0.464 |
Token | wool |
Token ID | 25749 |
Feature activation | -2.05e+01 |
Pos logit contributions | |
Don | +0.602 |
My | +0.593 |
Here | +0.575 |
There | +0.558 |
It | +0.556 |
Neg logit contributions | |
ies | -0.468 |
ishing | -0.430 |
- | -0.429 |
ig | -0.428 |
ish | -0.396 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e+01 |
Pos logit contributions | |
decision | +0.578 |
choice | +0.526 |
ies | +0.520 |
ig | +0.508 |
ishing | +0.492 |
Neg logit contributions | |
My | -0.883 |
Don | -0.875 |
I | -0.832 |
It | -0.829 |
There | -0.823 |
Token | It |
Token ID | 632 |
Feature activation | +2.15e+01 |
Pos logit contributions | |
. | +0.666 |
- | +0.653 |
eye | +0.651 |
, | +0.619 |
!" | +0.584 |
Neg logit contributions | |
Don | -0.408 |
My | -0.366 |
Would | -0.353 |
Looks | -0.348 |
Well | -0.330 |
Token | does |
Token ID | 857 |
Feature activation | +5.72e+00 |
Pos logit contributions | |
Don | +0.928 |
Would | +0.857 |
My | +0.856 |
Well | +0.831 |
There | +0.810 |
Neg logit contributions | |
ies | -0.806 |
- | -0.745 |
ig | -0.679 |
ish | -0.674 |
an | -0.670 |
Token | not |
Token ID | 407 |
Feature activation | +9.37e+00 |
Pos logit contributions | |
Don | +0.170 |
My | +0.155 |
Would | +0.151 |
Well | +0.141 |
It | +0.140 |
Neg logit contributions | |
ies | -0.289 |
- | -0.270 |
ig | -0.260 |
ish | -0.259 |
an | -0.254 |
Token | quarrel |
Token ID | 41998 |
Feature activation | -1.20e+01 |
Pos logit contributions | |
Don | +0.346 |
My | +0.328 |
There | +0.307 |
It | +0.305 |
Would | +0.299 |
Neg logit contributions | |
ies | -0.369 |
ig | -0.369 |
- | -0.360 |
ishing | -0.344 |
ish | -0.342 |
Token | to |
Token ID | 284 |
Feature activation | -6.72e+00 |
Pos logit contributions | |
ies | +0.973 |
- | +0.857 |
ig | +0.807 |
decision | +0.796 |
an | +0.793 |
Neg logit contributions | |
Don | -1.346 |
My | -1.286 |
Would | -1.239 |
Well | -1.208 |
It | -1.190 |
Token | the |
Token ID | 262 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
ies | +0.266 |
- | +0.247 |
ig | +0.224 |
ish | +0.224 |
an | +0.223 |
Neg logit contributions | |
Don | -0.274 |
My | -0.252 |
Would | -0.252 |
Well | -0.245 |
There | -0.236 |
Token | old |
Token ID | 1468 |
Feature activation | -9.53e+00 |
Pos logit contributions | |
Don | +0.237 |
My | +0.227 |
Would | +0.222 |
I | +0.215 |
It | +0.214 |
Neg logit contributions | |
ies | -0.158 |
- | -0.138 |
ig | -0.135 |
ishing | -0.129 |
ish | -0.122 |
Token | tree |
Token ID | 5509 |
Feature activation | -2.64e+01 |
Pos logit contributions | |
ies | +0.278 |
ig | +0.251 |
ishing | +0.241 |
- | +0.238 |
ish | +0.233 |
Neg logit contributions | |
Don | -0.459 |
My | -0.433 |
Would | -0.424 |
It | -0.409 |
I | -0.408 |
Token | house |
Token ID | 4803 |
Feature activation | -1.73e+01 |
Pos logit contributions | |
ies | +0.631 |
- | +0.521 |
ig | +0.499 |
an | +0.487 |
ish | +0.476 |
Neg logit contributions | |
Don | -1.444 |
My | -1.386 |
Would | -1.359 |
It | -1.313 |
Well | -1.306 |
Token | . |
Token ID | 13 |
Feature activation | -1.99e+01 |
Pos logit contributions | |
ies | +0.462 |
ig | +0.385 |
- | +0.384 |
ish | +0.371 |
an | +0.367 |
Neg logit contributions | |
Don | -0.908 |
My | -0.855 |
Would | -0.841 |
Well | -0.823 |
It | -0.817 |
Token | He |
Token ID | 679 |
Feature activation | +1.83e+01 |
Pos logit contributions | |
- | +0.833 |
ies | +0.824 |
. | +0.813 |
, | +0.784 |
house | +0.778 |
Neg logit contributions | |
Don | -0.501 |
Listen | -0.431 |
My | -0.427 |
Take | -0.414 |
Over | -0.407 |
Token | placed |
Token ID | 4624 |
Feature activation | -8.35e+00 |
Pos logit contributions | |
Don | +0.636 |
My | +0.584 |
Would | +0.578 |
Well | +0.555 |
There | +0.541 |
Neg logit contributions | |
ies | -0.837 |
- | -0.786 |
ig | -0.730 |
ish | -0.723 |
an | -0.715 |
Token | it |
Token ID | 340 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
ies | +0.355 |
- | +0.336 |
ig | +0.303 |
ish | +0.303 |
an | +0.297 |
Neg logit contributions | |
Don | -0.311 |
My | -0.291 |
Would | -0.288 |
Well | -0.278 |
There | -0.264 |
Token | at |
Token ID | 379 |
Feature activation | -7.88e+00 |
Pos logit contributions | |
ies | +0.126 |
decision | +0.111 |
choice | +0.110 |
an | +0.109 |
- | +0.108 |
Neg logit contributions | |
Don | -0.158 |
My | -0.151 |
There | -0.146 |
Would | -0.140 |
It | -0.139 |
Token | the |
Token ID | 262 |
Feature activation | +6.05e+00 |
Pos logit contributions | |
ies | +0.352 |
- | +0.329 |
ig | +0.305 |
ish | +0.303 |
an | +0.300 |
Neg logit contributions | |
Don | -0.282 |
My | -0.261 |
Would | -0.258 |
Well | -0.248 |
There | -0.239 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
Don | +0.121 |
Well | +0.116 |
You | +0.113 |
My | +0.112 |
Did | +0.110 |
Neg logit contributions | |
an | -0.086 |
ig | -0.085 |
ies | -0.082 |
island | -0.077 |
ish | -0.077 |
Token | She |
Token ID | 1375 |
Feature activation | +2.51e+01 |
Pos logit contributions | |
ies | +0.840 |
- | +0.816 |
eye | +0.775 |
ish | +0.755 |
fort | +0.745 |
Neg logit contributions | |
Don | -0.364 |
My | -0.333 |
Ah | -0.319 |
Would | -0.310 |
Well | -0.298 |
Token | ran |
Token ID | 4966 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
Don | +0.886 |
My | +0.815 |
Would | +0.812 |
Well | +0.778 |
There | +0.755 |
Neg logit contributions | |
ies | -1.143 |
- | -1.068 |
ig | -0.988 |
ish | -0.980 |
an | -0.970 |
Token | around |
Token ID | 1088 |
Feature activation | -6.06e+00 |
Pos logit contributions | |
Don | +0.143 |
Would | +0.133 |
My | +0.133 |
There | +0.130 |
It | +0.128 |
Neg logit contributions | |
ies | -0.117 |
ig | -0.099 |
ish | -0.099 |
an | -0.098 |
- | -0.095 |
Token | the |
Token ID | 262 |
Feature activation | +9.57e+00 |
Pos logit contributions | |
ies | +0.270 |
an | +0.244 |
ish | +0.241 |
ig | +0.237 |
- | +0.235 |
Neg logit contributions | |
Don | -0.225 |
My | -0.210 |
Would | -0.210 |
It | -0.189 |
Well | -0.188 |
Token | garden |
Token ID | 11376 |
Feature activation | -2.10e+01 |
Pos logit contributions | |
Don | +0.468 |
My | +0.443 |
It | +0.438 |
Would | +0.435 |
I | +0.431 |
Neg logit contributions | |
ies | -0.288 |
choice | -0.243 |
ig | -0.240 |
decision | -0.240 |
- | -0.239 |
Token | and |
Token ID | 290 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
ies | +0.720 |
an | +0.694 |
ish | +0.676 |
ig | +0.668 |
choice | +0.614 |
Neg logit contributions | |
Don | -0.861 |
My | -0.833 |
Would | -0.820 |
She | -0.778 |
There | -0.753 |
Token | played |
Token ID | 2826 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ies | +0.461 |
- | +0.428 |
ig | +0.389 |
ish | +0.385 |
eye | +0.379 |
Neg logit contributions | |
Don | -0.404 |
Would | -0.376 |
My | -0.376 |
Well | -0.370 |
Here | -0.356 |
Token | with |
Token ID | 351 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
Don | +0.068 |
My | +0.064 |
Would | +0.064 |
It | +0.062 |
There | +0.061 |
Neg logit contributions | |
ies | -0.058 |
ish | -0.051 |
ig | -0.049 |
an | -0.049 |
- | -0.048 |
Token | her |
Token ID | 607 |
Feature activation | +1.50e+01 |
Pos logit contributions | |
ies | +0.215 |
- | +0.201 |
ig | +0.189 |
ish | +0.187 |
an | +0.186 |
Neg logit contributions | |
Don | -0.152 |
My | -0.141 |
Would | -0.138 |
Well | -0.132 |
There | -0.128 |
Token | toys |
Token ID | 14958 |
Feature activation | -1.83e+01 |
Pos logit contributions | |
Don | +0.676 |
My | +0.664 |
Me | +0.630 |
There | +0.614 |
It | +0.606 |
Neg logit contributions | |
ies | -0.389 |
an | -0.388 |
decision | -0.376 |
- | -0.358 |
yt | -0.354 |
Token | ! |
Token ID | 0 |
Feature activation | -2.44e+01 |
Pos logit contributions | |
ies | +0.185 |
- | +0.165 |
ig | +0.161 |
an | +0.157 |
ish | +0.156 |
Neg logit contributions | |
Don | -0.221 |
My | -0.215 |
Would | -0.201 |
It | -0.198 |
Well | -0.198 |
Token | My |
Token ID | 2011 |
Feature activation | +1.99e+01 |
Pos logit contributions | |
- | +1.010 |
eye | +0.945 |
. | +0.945 |
ig | +0.906 |
!" | +0.905 |
Neg logit contributions | |
Don | -0.645 |
Over | -0.515 |
There | -0.495 |
Would | -0.489 |
My | -0.485 |
Token | grass |
Token ID | 8701 |
Feature activation | -1.90e+01 |
Pos logit contributions | |
My | +0.895 |
Don | +0.893 |
Would | +0.859 |
It | +0.842 |
Well | +0.830 |
Neg logit contributions | |
ies | -0.597 |
- | -0.504 |
ishing | -0.484 |
ig | -0.467 |
an | -0.457 |
Token | y |
Token ID | 88 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
ies | +0.796 |
- | +0.719 |
ig | +0.691 |
ishing | +0.677 |
ish | +0.674 |
Neg logit contributions | |
Don | -0.690 |
My | -0.653 |
It | -0.616 |
Well | -0.615 |
Would | -0.610 |
Token | paradise |
Token ID | 31354 |
Feature activation | -1.27e+01 |
Pos logit contributions | |
ies | +0.219 |
- | +0.197 |
ig | +0.187 |
an | +0.182 |
ishing | +0.182 |
Neg logit contributions | |
Don | -0.278 |
My | -0.268 |
Would | -0.251 |
It | -0.251 |
I | -0.250 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.51e+01 |
Pos logit contributions | |
ies | +0.480 |
- | +0.417 |
ig | +0.414 |
ish | +0.411 |
an | +0.409 |
Neg logit contributions | |
Don | -0.508 |
My | -0.472 |
It | -0.466 |
Well | -0.461 |
I | -0.461 |
Token | But |
Token ID | 887 |
Feature activation | +7.35e+00 |
Pos logit contributions | |
- | +1.152 |
ies | +1.140 |
. | +1.076 |
, | +1.059 |
! | +1.001 |
Neg logit contributions | |
Don | -0.635 |
Would | -0.593 |
Over | -0.549 |
Even | -0.537 |
Listen | -0.537 |
Token | when |
Token ID | 618 |
Feature activation | -8.67e+00 |
Pos logit contributions | |
Don | +0.260 |
Would | +0.243 |
My | +0.234 |
Well | +0.230 |
Here | +0.226 |
Neg logit contributions | |
ies | -0.329 |
- | -0.309 |
ig | -0.286 |
ish | -0.284 |
eye | -0.275 |
Token | he |
Token ID | 339 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
ies | +0.401 |
- | +0.379 |
ig | +0.344 |
ish | +0.336 |
decision | +0.336 |
Neg logit contributions | |
Don | -0.288 |
Would | -0.272 |
My | -0.258 |
Well | -0.253 |
Here | -0.245 |
Token | arrived |
Token ID | 5284 |
Feature activation | +7.99e+00 |
Pos logit contributions | |
Don | +0.408 |
My | +0.375 |
Would | +0.373 |
Well | +0.358 |
Here | +0.353 |
Neg logit contributions | |
ies | -0.503 |
- | -0.477 |
ig | -0.442 |
ish | -0.435 |
eye | -0.426 |
Token | in |
Token ID | 287 |
Feature activation | -5.13e+00 |
Pos logit contributions | |
Don | +0.423 |
My | +0.404 |
Would | +0.391 |
Well | +0.385 |
It | +0.384 |
Neg logit contributions | |
ies | -0.210 |
- | -0.175 |
an | -0.168 |
ig | -0.165 |
ish | -0.161 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
ies | +0.605 |
ig | +0.565 |
ish | +0.535 |
an | +0.532 |
ishing | +0.512 |
Neg logit contributions | |
My | -1.274 |
Don | -1.265 |
It | -1.189 |
There | -1.187 |
I | -1.186 |
Token | Sam |
Token ID | 3409 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
, | +0.670 |
. | +0.647 |
- | +0.647 |
ies | +0.642 |
house | +0.642 |
Neg logit contributions | |
Don | -0.632 |
Take | -0.562 |
My | -0.552 |
Ah | -0.551 |
Listen | -0.546 |
Token | looked |
Token ID | 3114 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
ies | +0.048 |
- | +0.046 |
ig | +0.041 |
ish | +0.041 |
an | +0.041 |
Neg logit contributions | |
Don | -0.041 |
Would | -0.038 |
My | -0.038 |
Here | -0.036 |
Well | -0.036 |
Token | at |
Token ID | 379 |
Feature activation | -7.13e+00 |
Pos logit contributions | |
Don | +0.235 |
My | +0.221 |
Would | +0.220 |
Well | +0.209 |
It | +0.207 |
Neg logit contributions | |
ies | -0.141 |
- | -0.122 |
ig | -0.120 |
an | -0.116 |
ish | -0.112 |
Token | the |
Token ID | 262 |
Feature activation | +6.61e+00 |
Pos logit contributions | |
ies | +0.323 |
- | +0.302 |
ig | +0.280 |
ish | +0.278 |
an | +0.276 |
Neg logit contributions | |
Don | -0.254 |
My | -0.233 |
Would | -0.232 |
Well | -0.222 |
There | -0.214 |
Token | hole |
Token ID | 7604 |
Feature activation | -2.95e+01 |
Pos logit contributions | |
Don | +0.303 |
My | +0.289 |
Would | +0.280 |
Here | +0.279 |
I | +0.277 |
Neg logit contributions | |
ies | -0.227 |
ig | -0.201 |
- | -0.195 |
ishing | -0.191 |
ish | -0.179 |
Token | and |
Token ID | 290 |
Feature activation | -1.37e+01 |
Pos logit contributions | |
ies | +0.781 |
ig | +0.674 |
an | +0.673 |
- | +0.654 |
ish | +0.654 |
Neg logit contributions | |
Don | -1.482 |
My | -1.416 |
It | -1.366 |
Would | -1.358 |
Well | -1.348 |
Token | said |
Token ID | 531 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
ies | +0.560 |
- | +0.503 |
castle | +0.444 |
ish | +0.442 |
ie | +0.438 |
Neg logit contributions | |
Don | -0.538 |
My | -0.513 |
Here | -0.506 |
Would | -0.504 |
Well | -0.496 |
Token | , |
Token ID | 11 |
Feature activation | -1.93e+01 |
Pos logit contributions | |
Don | +0.129 |
My | +0.124 |
Would | +0.123 |
Well | +0.121 |
Here | +0.119 |
Neg logit contributions | |
ies | -0.032 |
- | -0.027 |
ig | -0.020 |
ish | -0.019 |
an | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | +9.17e+00 |
Pos logit contributions | |
ies | +0.810 |
- | +0.808 |
, | +0.701 |
. | +0.682 |
ig | +0.668 |
Neg logit contributions | |
Don | -0.628 |
Here | -0.613 |
My | -0.604 |
There | -0.580 |
Well | -0.575 |
Token | Wow |
Token ID | 22017 |
Feature activation | +6.36e+00 |
Pos logit contributions | |
Did | +0.145 |
Their | +0.143 |
Not | +0.137 |
From | +0.134 |
Everything | +0.134 |
Neg logit contributions | |
an | -0.438 |
ies | -0.427 |
- | -0.420 |
est | -0.410 |
eye | -0.405 |