Token | looked |
Token ID | 3114 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
led | +0.262 |
ed | +0.220 |
ier | +0.209 |
med | +0.200 |
ked | +0.199 |
Neg logit contributions | |
great | -0.174 |
think | -0.171 |
pleasure | -0.169 |
plan | -0.166 |
learn | -0.165 |
Token | around |
Token ID | 1088 |
Feature activation | -9.00e+00 |
Pos logit contributions | |
led | +0.102 |
ed | +0.084 |
awed | +0.069 |
ied | +0.069 |
ked | +0.068 |
Neg logit contributions | |
pleasure | -0.103 |
welcome | -0.091 |
Deal | -0.091 |
great | -0.091 |
learn | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
led | +0.595 |
ed | +0.495 |
ied | +0.420 |
y | +0.402 |
ted | +0.400 |
Neg logit contributions | |
pleasure | -0.516 |
Deal | -0.473 |
welcome | -0.435 |
deal | -0.421 |
learn | -0.419 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
pleasure | +0.201 |
Deal | +0.178 |
welcome | +0.175 |
learn | +0.172 |
great | +0.171 |
Neg logit contributions | |
led | -0.239 |
ed | -0.197 |
ied | -0.168 |
ked | -0.163 |
awed | -0.163 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
led | +0.080 |
ed | +0.067 |
ied | +0.057 |
y | +0.053 |
ing | +0.053 |
Neg logit contributions | |
pleasure | -0.084 |
Deal | -0.078 |
welcome | -0.073 |
deal | -0.068 |
meantime | -0.067 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.315 |
ed | +0.262 |
ling | +0.238 |
ly | +0.236 |
ied | +0.235 |
Neg logit contributions | |
pleasure | -0.209 |
Deal | -0.174 |
Today | -0.148 |
Sometimes | -0.144 |
welcome | -0.142 |
Token | of |
Token ID | 286 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
pleasure | +1.256 |
Deal | +1.101 |
welcome | +1.097 |
great | +1.081 |
learn | +1.066 |
Neg logit contributions | |
led | -1.550 |
ed | -1.279 |
awed | -1.094 |
ked | -1.089 |
ied | -1.052 |
Token | something |
Token ID | 1223 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
led | +0.823 |
ed | +0.699 |
ied | +0.629 |
med | +0.605 |
ted | +0.602 |
Neg logit contributions | |
Deal | -0.564 |
pleasure | -0.530 |
welcome | -0.468 |
Today | -0.460 |
deal | -0.447 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.533 |
great | +0.458 |
welcome | +0.456 |
think | +0.424 |
fun | +0.421 |
Neg logit contributions | |
led | -0.940 |
ed | -0.799 |
ked | -0.706 |
awed | -0.690 |
ied | -0.689 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +4.62e+00 |
Pos logit contributions | |
led | +0.369 |
ed | +0.365 |
ted | +0.288 |
y | +0.278 |
awed | +0.277 |
Neg logit contributions | |
pleasure | -0.294 |
Deal | -0.275 |
welcome | -0.256 |
deal | -0.255 |
great | -0.253 |
Token | she |
Token ID | 673 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
pleasure | +0.158 |
welcome | +0.127 |
Deal | +0.125 |
great | +0.124 |
learn | +0.120 |
Neg logit contributions | |
led | -0.407 |
ed | -0.349 |
ied | -0.316 |
awed | -0.312 |
ked | -0.310 |
Token | a |
Token ID | 257 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
pleasure | +0.188 |
Deal | +0.165 |
learn | +0.155 |
know | +0.146 |
think | +0.145 |
Neg logit contributions | |
led | -0.368 |
ed | -0.314 |
ied | -0.276 |
y | -0.270 |
med | -0.265 |
Token | surprise |
Token ID | 5975 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
pleasure | +0.138 |
Deal | +0.137 |
learn | +0.124 |
know | +0.123 |
think | +0.113 |
Neg logit contributions | |
led | -0.207 |
ed | -0.178 |
y | -0.172 |
ling | -0.161 |
lers | -0.159 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.626 |
welcome | +0.574 |
Deal | +0.525 |
great | +0.512 |
deal | +0.507 |
Neg logit contributions | |
led | -1.011 |
ed | -0.829 |
awed | -0.751 |
ked | -0.737 |
ied | -0.727 |
Token | But |
Token ID | 887 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
led | +0.424 |
ed | +0.397 |
ted | +0.338 |
ked | +0.328 |
med | +0.326 |
Neg logit contributions | |
pleasure | -0.226 |
plan | -0.221 |
great | -0.215 |
deal | -0.204 |
learn | -0.203 |
Token | I |
Token ID | 314 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
pleasure | +0.438 |
Deal | +0.412 |
learn | +0.353 |
meantime | +0.345 |
know | +0.342 |
Neg logit contributions | |
led | -0.887 |
ed | -0.771 |
ied | -0.689 |
ing | -0.664 |
we | -0.660 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.109 |
Deal | +0.098 |
welcome | +0.079 |
meantime | +0.074 |
great | +0.071 |
Neg logit contributions | |
led | -0.260 |
ed | -0.230 |
y | -0.203 |
ling | -0.190 |
ave | -0.190 |
Token | you |
Token ID | 345 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
pleasure | +0.905 |
Deal | +0.790 |
welcome | +0.731 |
learn | +0.705 |
think | +0.681 |
Neg logit contributions | |
led | -1.909 |
ed | -1.649 |
ied | -1.458 |
med | -1.404 |
ked | -1.403 |
Token | will |
Token ID | 481 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
pleasure | +0.355 |
Deal | +0.318 |
welcome | +0.274 |
great | +0.248 |
meantime | +0.241 |
Neg logit contributions | |
led | -0.862 |
ed | -0.748 |
ied | -0.647 |
y | -0.640 |
ked | -0.640 |
Token | like |
Token ID | 588 |
Feature activation | +4.80e+00 |
Pos logit contributions | |
pleasure | +0.195 |
Deal | +0.178 |
meantime | +0.135 |
Promise | +0.131 |
Sometimes | +0.129 |
Neg logit contributions | |
led | -0.284 |
ed | -0.258 |
y | -0.247 |
ier | -0.239 |
red | -0.220 |
Token | it |
Token ID | 340 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
pleasure | +0.216 |
Deal | +0.195 |
welcome | +0.178 |
learn | +0.167 |
think | +0.163 |
Neg logit contributions | |
led | -0.373 |
ed | -0.319 |
ied | -0.280 |
ted | -0.272 |
med | -0.272 |
Token | . |
Token ID | 13 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
pleasure | +0.294 |
Deal | +0.260 |
welcome | +0.259 |
great | +0.249 |
learn | +0.248 |
Neg logit contributions | |
led | -0.394 |
ed | -0.327 |
ied | -0.284 |
awed | -0.279 |
ked | -0.277 |
Token | picture |
Token ID | 4286 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
pleasure | +0.170 |
Deal | +0.156 |
welcome | +0.134 |
think | +0.132 |
know | +0.130 |
Neg logit contributions | |
led | -0.213 |
ed | -0.179 |
ling | -0.159 |
y | -0.158 |
lers | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.243 |
welcome | +0.224 |
Deal | +0.214 |
learn | +0.212 |
great | +0.210 |
Neg logit contributions | |
led | -0.352 |
ed | -0.289 |
ked | -0.257 |
awed | -0.255 |
ied | -0.254 |
Token | What |
Token ID | 1867 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
ed | +0.294 |
led | +0.289 |
med | +0.254 |
als | +0.252 |
es | +0.250 |
Neg logit contributions | |
plan | -0.266 |
great | -0.231 |
pleasure | -0.229 |
deal | -0.227 |
chores | -0.223 |
Token | do |
Token ID | 466 |
Feature activation | +7.66e+00 |
Pos logit contributions | |
pleasure | +0.222 |
Deal | +0.205 |
welcome | +0.171 |
meantime | +0.164 |
think | +0.158 |
Neg logit contributions | |
led | -0.415 |
ed | -0.354 |
ied | -0.313 |
y | -0.307 |
es | -0.305 |
Token | you |
Token ID | 345 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
pleasure | +0.306 |
Deal | +0.259 |
welcome | +0.222 |
meantime | +0.210 |
know | +0.209 |
Neg logit contributions | |
led | -0.637 |
ed | -0.534 |
we | -0.487 |
ied | -0.479 |
es | -0.467 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.439 |
Deal | +0.372 |
Sometimes | +0.311 |
meantime | +0.299 |
welcome | +0.296 |
Neg logit contributions | |
led | -0.719 |
ed | -0.622 |
y | -0.556 |
ave | -0.525 |
ier | -0.521 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.09e+00 |
Pos logit contributions | |
pleasure | +1.088 |
Deal | +0.964 |
welcome | +0.888 |
learn | +0.855 |
think | +0.838 |
Neg logit contributions | |
led | -1.753 |
ed | -1.487 |
ied | -1.306 |
med | -1.250 |
ted | -1.235 |
Token |
|
Token ID | 198 |
Feature activation | +9.27e+00 |
Pos logit contributions | |
pleasure | +0.288 |
Deal | +0.240 |
great | +0.229 |
deal | +0.227 |
plan | +0.218 |
Neg logit contributions | |
led | -0.533 |
ed | -0.493 |
ted | -0.428 |
awed | -0.427 |
es | -0.420 |
Token |
|
Token ID | 198 |
Feature activation | +9.25e+00 |
Pos logit contributions | |
pleasure | +0.357 |
Deal | +0.284 |
deal | +0.255 |
great | +0.247 |
welcome | +0.242 |
Neg logit contributions | |
led | -0.768 |
ed | -0.681 |
awed | -0.608 |
ied | -0.602 |
med | -0.597 |
Token | Anna |
Token ID | 31160 |
Feature activation | -7.19e+00 |
Pos logit contributions | |
pleasure | +0.491 |
Deal | +0.425 |
welcome | +0.401 |
great | +0.389 |
learn | +0.381 |
Neg logit contributions | |
led | -0.653 |
ed | -0.553 |
ied | -0.474 |
awed | -0.468 |
med | -0.462 |
Token | and |
Token ID | 290 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
led | +0.481 |
ed | +0.397 |
ied | +0.342 |
ked | +0.328 |
awed | +0.326 |
Neg logit contributions | |
pleasure | -0.400 |
Deal | -0.361 |
welcome | -0.348 |
learn | -0.340 |
think | -0.337 |
Token | kind |
Token ID | 1611 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
pleasure | +0.181 |
Deal | +0.153 |
welcome | +0.139 |
learn | +0.132 |
great | +0.129 |
Neg logit contributions | |
led | -0.466 |
ed | -0.405 |
ied | -0.360 |
ked | -0.351 |
awed | -0.350 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
led | +0.117 |
ed | +0.093 |
awed | +0.087 |
ied | +0.086 |
ked | +0.084 |
Neg logit contributions | |
pleasure | -0.073 |
welcome | -0.066 |
learn | -0.061 |
Deal | -0.061 |
think | -0.059 |
Token | responsible |
Token ID | 4497 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
pleasure | +0.145 |
Deal | +0.134 |
meantime | +0.116 |
Today | +0.109 |
welcome | +0.108 |
Neg logit contributions | |
led | -0.287 |
ed | -0.246 |
ked | -0.215 |
ied | -0.208 |
ted | -0.207 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
led | +0.033 |
ed | +0.027 |
awed | +0.025 |
ked | +0.024 |
ied | +0.024 |
Neg logit contributions | |
pleasure | -0.025 |
welcome | -0.022 |
Deal | -0.022 |
learn | -0.021 |
great | -0.021 |
Token | We |
Token ID | 775 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
led | +0.380 |
ed | +0.363 |
ted | +0.317 |
ked | +0.304 |
med | +0.303 |
Neg logit contributions | |
plan | -0.240 |
pleasure | -0.227 |
great | -0.219 |
deal | -0.218 |
learn | -0.197 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.147 |
Deal | +0.137 |
welcome | +0.110 |
meantime | +0.105 |
great | +0.104 |
Neg logit contributions | |
led | -0.321 |
ed | -0.287 |
y | -0.247 |
ked | -0.244 |
ied | -0.241 |
Token | you |
Token ID | 345 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
pleasure | +0.963 |
Deal | +0.854 |
welcome | +0.782 |
learn | +0.753 |
think | +0.736 |
Neg logit contributions | |
led | -1.851 |
ed | -1.589 |
ied | -1.400 |
ked | -1.348 |
ted | -1.345 |
Token | are |
Token ID | 389 |
Feature activation | +8.87e+00 |
Pos logit contributions | |
pleasure | +0.369 |
Deal | +0.348 |
welcome | +0.287 |
meantime | +0.268 |
Today | +0.258 |
Neg logit contributions | |
led | -0.835 |
ed | -0.729 |
y | -0.623 |
ied | -0.617 |
ked | -0.615 |
Token | ready |
Token ID | 3492 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
pleasure | +0.340 |
Deal | +0.300 |
welcome | +0.260 |
learn | +0.258 |
think | +0.250 |
Neg logit contributions | |
led | -0.743 |
ed | -0.638 |
ied | -0.561 |
ked | -0.552 |
med | -0.547 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
pleasure | +0.190 |
Deal | +0.174 |
welcome | +0.162 |
learn | +0.156 |
great | +0.155 |
Neg logit contributions | |
led | -0.235 |
ed | -0.197 |
ied | -0.167 |
ked | -0.159 |
y | -0.159 |
Token | have |
Token ID | 423 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
led | +0.281 |
ed | +0.246 |
y | +0.229 |
med | +0.225 |
ted | +0.224 |
Neg logit contributions | |
pleasure | -0.175 |
Deal | -0.165 |
Today | -0.122 |
meantime | -0.119 |
Sometimes | -0.118 |
Token | cones |
Token ID | 47314 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
pleasure | +0.013 |
Deal | +0.012 |
welcome | +0.011 |
learn | +0.011 |
think | +0.011 |
Neg logit contributions | |
led | -0.019 |
ed | -0.016 |
ied | -0.013 |
ked | -0.013 |
awed | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
led | +0.324 |
ed | +0.270 |
awed | +0.235 |
ked | +0.234 |
ied | +0.231 |
Neg logit contributions | |
pleasure | -0.235 |
learn | -0.221 |
welcome | -0.216 |
think | -0.210 |
have | -0.204 |
Token | smile |
Token ID | 8212 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
pleasure | +0.152 |
Deal | +0.132 |
welcome | +0.122 |
great | +0.116 |
learn | +0.111 |
Neg logit contributions | |
led | -0.289 |
ed | -0.247 |
ked | -0.211 |
ted | -0.209 |
med | -0.207 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.466 |
great | +0.426 |
learn | +0.423 |
welcome | +0.414 |
OK | +0.411 |
Neg logit contributions | |
led | -0.713 |
ied | -0.577 |
ed | -0.554 |
awed | -0.535 |
web | -0.504 |
Token | They |
Token ID | 1119 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
led | +0.436 |
ed | +0.408 |
ted | +0.337 |
awed | +0.331 |
ked | +0.331 |
Neg logit contributions | |
pleasure | -0.251 |
Deal | -0.226 |
deal | -0.215 |
meantime | -0.206 |
great | -0.205 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.041 |
Deal | +0.035 |
welcome | +0.028 |
Today | +0.027 |
meantime | +0.027 |
Neg logit contributions | |
led | -0.109 |
ed | -0.095 |
ked | -0.084 |
ted | -0.082 |
y | -0.082 |
Token | it |
Token ID | 340 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
pleasure | +0.987 |
Deal | +0.835 |
welcome | +0.811 |
great | +0.783 |
learn | +0.774 |
Neg logit contributions | |
led | -1.841 |
ed | -1.586 |
ied | -1.369 |
ked | -1.365 |
awed | -1.361 |
Token | is |
Token ID | 318 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
pleasure | +0.191 |
Deal | +0.158 |
welcome | +0.151 |
great | +0.148 |
learn | +0.146 |
Neg logit contributions | |
led | -0.490 |
ed | -0.426 |
ied | -0.386 |
ked | -0.379 |
awed | -0.374 |
Token | the |
Token ID | 262 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
pleasure | +0.162 |
Deal | +0.143 |
learn | +0.124 |
welcome | +0.122 |
think | +0.114 |
Neg logit contributions | |
led | -0.405 |
ed | -0.346 |
ied | -0.307 |
ked | -0.299 |
y | -0.299 |
Token | best |
Token ID | 1266 |
Feature activation | +5.40e+00 |
Pos logit contributions | |
led | +0.142 |
ed | +0.124 |
ier | +0.111 |
y | +0.109 |
ling | +0.108 |
Neg logit contributions | |
pleasure | -0.071 |
Deal | -0.068 |
know | -0.062 |
learn | -0.061 |
Not | -0.058 |
Token | ice |
Token ID | 4771 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
pleasure | +0.227 |
Deal | +0.207 |
welcome | +0.184 |
learn | +0.181 |
great | +0.175 |
Neg logit contributions | |
led | -0.427 |
ed | -0.372 |
ied | -0.325 |
y | -0.315 |
awed | -0.314 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
led | +0.037 |
ed | +0.030 |
ied | +0.025 |
ked | +0.024 |
awed | +0.024 |
Neg logit contributions | |
pleasure | -0.033 |
Deal | -0.030 |
welcome | -0.029 |
learn | -0.028 |
great | -0.028 |
Token | tell |
Token ID | 1560 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
led | +0.304 |
ling | +0.252 |
ed | +0.250 |
y | +0.244 |
ied | +0.238 |
Neg logit contributions | |
pleasure | -0.166 |
Deal | -0.148 |
Today | -0.130 |
meantime | -0.119 |
Anything | -0.106 |
Token | someone |
Token ID | 2130 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
pleasure | +0.446 |
Deal | +0.406 |
welcome | +0.350 |
meantime | +0.331 |
learn | +0.331 |
Neg logit contributions | |
led | -0.873 |
ed | -0.754 |
ied | -0.670 |
ted | -0.641 |
ked | -0.634 |
Token | if |
Token ID | 611 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
pleasure | +0.276 |
welcome | +0.238 |
great | +0.231 |
learn | +0.229 |
Deal | +0.226 |
Neg logit contributions | |
led | -0.408 |
ed | -0.344 |
awed | -0.300 |
ked | -0.299 |
ied | -0.295 |
Token | we |
Token ID | 356 |
Feature activation | +5.29e+00 |
Pos logit contributions | |
pleasure | +0.150 |
Deal | +0.131 |
welcome | +0.115 |
learn | +0.108 |
great | +0.106 |
Neg logit contributions | |
led | -0.385 |
ed | -0.336 |
ied | -0.298 |
ked | -0.289 |
ted | -0.287 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.178 |
Deal | +0.152 |
welcome | +0.131 |
great | +0.124 |
learn | +0.118 |
Neg logit contributions | |
led | -0.471 |
ed | -0.413 |
ied | -0.363 |
ked | -0.358 |
ted | -0.352 |
Token | there |
Token ID | 612 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
pleasure | +1.210 |
Deal | +1.072 |
welcome | +1.042 |
learn | +1.004 |
great | +0.998 |
Neg logit contributions | |
led | -1.607 |
ed | -1.347 |
ied | -1.147 |
awed | -1.124 |
ked | -1.122 |
Token | 's |
Token ID | 338 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
led | +0.220 |
ed | +0.191 |
ied | +0.167 |
y | +0.161 |
ling | +0.160 |
Neg logit contributions | |
pleasure | -0.110 |
Deal | -0.100 |
welcome | -0.086 |
learn | -0.081 |
great | -0.080 |
Token | a |
Token ID | 257 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
led | +0.215 |
ed | +0.185 |
ied | +0.158 |
ier | +0.156 |
ling | +0.156 |
Neg logit contributions | |
pleasure | -0.112 |
Deal | -0.103 |
welcome | -0.085 |
learn | -0.082 |
deal | -0.081 |
Token | monster |
Token ID | 9234 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
Deal | +0.121 |
pleasure | +0.116 |
Today | +0.092 |
Not | +0.088 |
learn | +0.087 |
Neg logit contributions | |
led | -0.235 |
ed | -0.204 |
ling | -0.196 |
y | -0.193 |
ier | -0.187 |
Token | in |
Token ID | 287 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
led | +0.391 |
ed | +0.317 |
ied | +0.288 |
awed | +0.284 |
ked | +0.273 |
Neg logit contributions | |
pleasure | -0.280 |
welcome | -0.256 |
learn | -0.248 |
think | -0.246 |
great | -0.243 |
Token | about |
Token ID | 546 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
pleasure | +0.213 |
welcome | +0.187 |
Deal | +0.182 |
learn | +0.177 |
great | +0.176 |
Neg logit contributions | |
led | -0.281 |
ed | -0.231 |
awed | -0.201 |
ked | -0.200 |
ied | -0.198 |
Token | that |
Token ID | 326 |
Feature activation | -9.92e-01 |
Pos logit contributions | |
led | +0.148 |
ed | +0.127 |
ied | +0.112 |
y | +0.108 |
med | +0.106 |
Neg logit contributions | |
pleasure | -0.083 |
Deal | -0.082 |
welcome | -0.067 |
learn | -0.066 |
deal | -0.064 |
Token | garden |
Token ID | 11376 |
Feature activation | +4.71e+00 |
Pos logit contributions | |
led | +0.081 |
ed | +0.070 |
ied | +0.061 |
y | +0.059 |
ling | +0.059 |
Neg logit contributions | |
pleasure | -0.040 |
Deal | -0.037 |
welcome | -0.031 |
learn | -0.031 |
great | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.224 |
welcome | +0.192 |
Deal | +0.190 |
great | +0.183 |
learn | +0.182 |
Neg logit contributions | |
led | -0.355 |
ed | -0.302 |
ied | -0.261 |
ked | -0.261 |
awed | -0.260 |
Token | I |
Token ID | 314 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
led | +0.345 |
ed | +0.345 |
med | +0.293 |
ted | +0.293 |
als | +0.284 |
Neg logit contributions | |
great | -0.246 |
plan | -0.244 |
deal | -0.236 |
pleasure | -0.235 |
fun | -0.224 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.101 |
Deal | +0.088 |
welcome | +0.074 |
great | +0.067 |
deal | +0.064 |
Neg logit contributions | |
led | -0.280 |
ed | -0.249 |
y | -0.214 |
ied | -0.213 |
ked | -0.209 |
Token | you |
Token ID | 345 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
pleasure | +0.961 |
Deal | +0.885 |
welcome | +0.784 |
learn | +0.753 |
great | +0.733 |
Neg logit contributions | |
led | -1.828 |
ed | -1.577 |
ied | -1.391 |
med | -1.321 |
ked | -1.315 |
Token | 'll |
Token ID | 1183 |
Feature activation | +6.61e+00 |
Pos logit contributions | |
pleasure | +0.405 |
Deal | +0.367 |
welcome | +0.327 |
great | +0.304 |
learn | +0.298 |
Neg logit contributions | |
led | -0.808 |
ed | -0.702 |
ied | -0.603 |
ked | -0.592 |
y | -0.589 |
Token | love |
Token ID | 1842 |
Feature activation | +1.49e+01 |
Pos logit contributions | |
pleasure | +0.237 |
Deal | +0.233 |
Today | +0.159 |
meantime | +0.158 |
Sometimes | +0.156 |
Neg logit contributions | |
led | -0.457 |
ed | -0.414 |
y | -0.382 |
ier | -0.380 |
ied | -0.361 |
Token | it |
Token ID | 340 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
pleasure | +0.698 |
Deal | +0.655 |
welcome | +0.576 |
learn | +0.548 |
deal | +0.541 |
Neg logit contributions | |
led | -1.127 |
ed | -0.965 |
ied | -0.840 |
med | -0.820 |
y | -0.819 |
Token | ." |
Token ID | 526 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
pleasure | +0.261 |
Deal | +0.237 |
welcome | +0.215 |
learn | +0.210 |
think | +0.205 |
Neg logit contributions | |
led | -0.425 |
ed | -0.365 |
ied | -0.311 |
y | -0.303 |
ier | -0.299 |
Token | busy |
Token ID | 8179 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
pleasure | +0.396 |
Deal | +0.338 |
welcome | +0.321 |
learn | +0.304 |
great | +0.301 |
Neg logit contributions | |
led | -0.843 |
ed | -0.729 |
ied | -0.642 |
ked | -0.628 |
awed | -0.627 |
Token | that |
Token ID | 326 |
Feature activation | -9.82e-01 |
Pos logit contributions | |
pleasure | +0.152 |
welcome | +0.129 |
great | +0.127 |
learn | +0.127 |
think | +0.126 |
Neg logit contributions | |
led | -0.288 |
ed | -0.243 |
ied | -0.217 |
ked | -0.214 |
awed | -0.210 |
Token | he |
Token ID | 339 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
led | +0.059 |
ed | +0.048 |
ied | +0.039 |
ling | +0.037 |
ted | +0.037 |
Neg logit contributions | |
pleasure | -0.062 |
Deal | -0.058 |
welcome | -0.053 |
learn | -0.051 |
think | -0.051 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
led | +0.415 |
ed | +0.355 |
ied | +0.317 |
awed | +0.313 |
ier | +0.311 |
Neg logit contributions | |
pleasure | -0.261 |
great | -0.258 |
think | -0.249 |
learn | -0.243 |
have | -0.240 |
Token | 't |
Token ID | 470 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
led | +0.285 |
ed | +0.255 |
ied | +0.223 |
med | +0.219 |
ted | +0.217 |
Neg logit contributions | |
pleasure | -0.116 |
Deal | -0.092 |
welcome | -0.084 |
great | -0.076 |
think | -0.075 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.089 |
Deal | +0.073 |
welcome | +0.061 |
great | +0.055 |
meantime | +0.049 |
Neg logit contributions | |
led | -0.286 |
ed | -0.254 |
ked | -0.224 |
ied | -0.223 |
y | -0.220 |
Token | straight |
Token ID | 3892 |
Feature activation | -8.09e+00 |
Pos logit contributions | |
pleasure | +1.247 |
welcome | +1.093 |
Deal | +1.091 |
great | +1.063 |
learn | +1.056 |
Neg logit contributions | |
led | -1.577 |
ed | -1.309 |
awed | -1.109 |
ked | -1.105 |
ied | -1.089 |
Token | ! |
Token ID | 0 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
led | +0.562 |
ed | +0.466 |
ied | +0.404 |
awed | +0.397 |
ked | +0.391 |
Neg logit contributions | |
pleasure | -0.433 |
welcome | -0.376 |
Deal | -0.376 |
great | -0.366 |
learn | -0.362 |
Token |
|
Token ID | 198 |
Feature activation | +9.25e+00 |
Pos logit contributions | |
led | +0.158 |
ed | +0.151 |
awed | +0.126 |
ted | +0.124 |
ied | +0.122 |
Neg logit contributions | |
pleasure | -0.078 |
Deal | -0.065 |
deal | -0.062 |
learn | -0.062 |
welcome | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | +9.27e+00 |
Pos logit contributions | |
pleasure | +0.394 |
Deal | +0.333 |
deal | +0.285 |
welcome | +0.283 |
learn | +0.256 |
Neg logit contributions | |
led | -0.743 |
ed | -0.664 |
ied | -0.587 |
y | -0.581 |
awed | -0.569 |
Token | The |
Token ID | 464 |
Feature activation | -5.59e+00 |
Pos logit contributions | |
pleasure | +0.379 |
Deal | +0.316 |
welcome | +0.300 |
learn | +0.275 |
great | +0.273 |
Neg logit contributions | |
led | -0.781 |
ed | -0.675 |
ied | -0.597 |
awed | -0.580 |
ked | -0.578 |
Token | of |
Token ID | 286 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
pleasure | +0.564 |
great | +0.508 |
welcome | +0.496 |
learn | +0.472 |
have | +0.466 |
Neg logit contributions | |
led | -0.858 |
ed | -0.688 |
awed | -0.643 |
ked | -0.642 |
ied | -0.604 |
Token | it |
Token ID | 340 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
led | +0.907 |
ed | +0.770 |
ied | +0.678 |
med | +0.663 |
ted | +0.653 |
Neg logit contributions | |
pleasure | -0.595 |
Deal | -0.548 |
welcome | -0.491 |
learn | -0.470 |
deal | -0.467 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.276 |
great | +0.254 |
welcome | +0.251 |
learn | +0.245 |
think | +0.244 |
Neg logit contributions | |
led | -0.387 |
ed | -0.322 |
ked | -0.292 |
ied | -0.290 |
awed | -0.287 |
Token | Molly |
Token ID | 30236 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
led | +0.424 |
ed | +0.387 |
ied | +0.323 |
y | +0.323 |
ted | +0.318 |
Neg logit contributions | |
pleasure | -0.240 |
welcome | -0.208 |
Deal | -0.205 |
deal | -0.204 |
great | -0.203 |
Token | would |
Token ID | 561 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
led | +0.185 |
ed | +0.153 |
ied | +0.134 |
ked | +0.130 |
awed | +0.130 |
Neg logit contributions | |
pleasure | -0.141 |
learn | -0.124 |
think | -0.123 |
great | -0.122 |
welcome | -0.121 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.405 |
ed | +0.354 |
ied | +0.318 |
y | +0.317 |
med | +0.310 |
Neg logit contributions | |
pleasure | -0.180 |
Deal | -0.168 |
welcome | -0.133 |
great | -0.126 |
OK | -0.126 |
Token | a |
Token ID | 257 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
pleasure | +1.138 |
Deal | +0.976 |
welcome | +0.964 |
great | +0.950 |
learn | +0.939 |
Neg logit contributions | |
led | -1.683 |
ed | -1.423 |
ked | -1.216 |
awed | -1.210 |
ied | -1.204 |
Token | lot |
Token ID | 1256 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
Deal | +0.100 |
pleasure | +0.094 |
welcome | +0.071 |
OK | +0.068 |
learn | +0.066 |
Neg logit contributions | |
led | -0.269 |
ed | -0.236 |
y | -0.227 |
ling | -0.220 |
mer | -0.214 |
Token | about |
Token ID | 546 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
pleasure | +0.584 |
welcome | +0.525 |
great | +0.506 |
learn | +0.498 |
think | +0.480 |
Neg logit contributions | |
led | -0.844 |
ed | -0.705 |
awed | -0.630 |
ked | -0.628 |
ied | -0.588 |
Token | her |
Token ID | 607 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
led | +0.151 |
ed | +0.131 |
ied | +0.113 |
med | +0.110 |
y | +0.110 |
Neg logit contributions | |
pleasure | -0.087 |
Deal | -0.082 |
welcome | -0.072 |
learn | -0.067 |
deal | -0.066 |
Token | team |
Token ID | 1074 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
led | +0.195 |
ed | +0.171 |
ied | +0.148 |
y | +0.146 |
ling | +0.145 |
Neg logit contributions | |
pleasure | -0.081 |
Deal | -0.076 |
welcome | -0.066 |
learn | -0.061 |
know | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
pleasure | +0.692 |
welcome | +0.606 |
Deal | +0.583 |
learn | +0.569 |
great | +0.568 |
Neg logit contributions | |
led | -1.046 |
ed | -0.876 |
awed | -0.761 |
ked | -0.759 |
ied | -0.755 |
Token | do |
Token ID | 466 |
Feature activation | +7.67e+00 |
Pos logit contributions | |
led | +0.310 |
ed | +0.261 |
ied | +0.235 |
ling | +0.233 |
ier | +0.232 |
Neg logit contributions | |
pleasure | -0.193 |
Deal | -0.175 |
Today | -0.143 |
meantime | -0.133 |
Okay | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.375 |
Deal | +0.331 |
welcome | +0.318 |
learn | +0.305 |
great | +0.303 |
Neg logit contributions | |
led | -0.570 |
ed | -0.483 |
ied | -0.417 |
ked | -0.406 |
awed | -0.405 |
Token | Do |
Token ID | 2141 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
led | +0.372 |
ed | +0.361 |
ted | +0.309 |
med | +0.303 |
ked | +0.303 |
Neg logit contributions | |
pleasure | -0.237 |
great | -0.235 |
plan | -0.232 |
deal | -0.216 |
meantime | -0.209 |
Token | you |
Token ID | 345 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
pleasure | +0.214 |
Deal | +0.171 |
welcome | +0.157 |
meantime | +0.151 |
Today | +0.147 |
Neg logit contributions | |
led | -0.348 |
ed | -0.294 |
ied | -0.277 |
es | -0.267 |
ing | -0.264 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.327 |
Deal | +0.280 |
meantime | +0.215 |
great | +0.210 |
welcome | +0.206 |
Neg logit contributions | |
led | -0.832 |
ed | -0.722 |
y | -0.646 |
ling | -0.640 |
ier | -0.639 |
Token | you |
Token ID | 345 |
Feature activation | +9.88e+00 |
Pos logit contributions | |
pleasure | +1.162 |
Deal | +1.056 |
welcome | +0.965 |
learn | +0.932 |
great | +0.917 |
Neg logit contributions | |
led | -1.648 |
ed | -1.390 |
ied | -1.205 |
med | -1.154 |
ted | -1.146 |
Token | will |
Token ID | 481 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
pleasure | +0.435 |
Deal | +0.402 |
welcome | +0.342 |
meantime | +0.327 |
great | +0.320 |
Neg logit contributions | |
led | -0.763 |
ed | -0.662 |
y | -0.562 |
ied | -0.558 |
ked | -0.552 |
Token | succeed |
Token ID | 6758 |
Feature activation | +7.81e+00 |
Pos logit contributions | |
pleasure | +0.180 |
Deal | +0.166 |
meantime | +0.116 |
Today | +0.111 |
Sometimes | +0.110 |
Neg logit contributions | |
led | -0.312 |
ed | -0.287 |
ier | -0.270 |
y | -0.269 |
es | -0.254 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.09e+00 |
Pos logit contributions | |
pleasure | +0.329 |
Deal | +0.287 |
welcome | +0.260 |
learn | +0.246 |
great | +0.242 |
Neg logit contributions | |
led | -0.639 |
ed | -0.547 |
ied | -0.478 |
ling | -0.468 |
y | -0.467 |
Token | |
Token ID | 220 |
Feature activation | +1.55e+01 |
Pos logit contributions | |
pleasure | +0.212 |
Deal | +0.173 |
great | +0.156 |
welcome | +0.149 |
deal | +0.149 |
Neg logit contributions | |
led | -0.642 |
ed | -0.577 |
ted | -0.503 |
awed | -0.499 |
med | -0.499 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
led | +0.484 |
ed | +0.395 |
ied | +0.392 |
awed | +0.378 |
we | +0.372 |
Neg logit contributions | |
pleasure | -0.245 |
welcome | -0.205 |
great | -0.204 |
Deal | -0.195 |
OK | -0.189 |
Token | said |
Token ID | 531 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
pleasure | +0.174 |
welcome | +0.150 |
great | +0.148 |
think | +0.148 |
learn | +0.143 |
Neg logit contributions | |
led | -0.256 |
ed | -0.215 |
ied | -0.194 |
ier | -0.188 |
awed | -0.185 |
Token | , |
Token ID | 11 |
Feature activation | +5.37e+00 |
Pos logit contributions | |
led | +0.130 |
ed | +0.111 |
awed | +0.101 |
ied | +0.101 |
ked | +0.099 |
Neg logit contributions | |
pleasure | -0.050 |
welcome | -0.042 |
great | -0.041 |
Deal | -0.040 |
learn | -0.038 |
Token | " |
Token ID | 366 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
pleasure | +0.214 |
Deal | +0.183 |
welcome | +0.173 |
learn | +0.164 |
great | +0.162 |
Neg logit contributions | |
led | -0.447 |
ed | -0.388 |
ied | -0.339 |
ked | -0.334 |
awed | -0.331 |
Token | I |
Token ID | 40 |
Feature activation | -7.58e+00 |
Pos logit contributions | |
pleasure | +0.010 |
Deal | +0.009 |
welcome | +0.009 |
great | +0.008 |
learn | +0.008 |
Neg logit contributions | |
led | -0.017 |
ed | -0.015 |
ied | -0.013 |
ked | -0.013 |
awed | -0.013 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.615 |
ed | +0.569 |
y | +0.495 |
ave | +0.464 |
ier | +0.461 |
Neg logit contributions | |
Deal | -0.245 |
pleasure | -0.240 |
meantime | -0.171 |
welcome | -0.169 |
compromise | -0.158 |
Token | there |
Token ID | 612 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
pleasure | +0.869 |
Deal | +0.737 |
welcome | +0.697 |
learn | +0.659 |
great | +0.653 |
Neg logit contributions | |
led | -1.952 |
ed | -1.691 |
ied | -1.492 |
ked | -1.461 |
awed | -1.459 |
Token | is |
Token ID | 318 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
led | +0.215 |
ed | +0.185 |
ied | +0.161 |
y | +0.156 |
awed | +0.154 |
Neg logit contributions | |
pleasure | -0.116 |
Deal | -0.105 |
welcome | -0.094 |
learn | -0.089 |
great | -0.088 |
Token | some |
Token ID | 617 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
pleasure | +0.168 |
Deal | +0.139 |
welcome | +0.134 |
great | +0.126 |
learn | +0.125 |
Neg logit contributions | |
led | -0.406 |
ed | -0.351 |
ied | -0.312 |
awed | -0.307 |
ked | -0.306 |
Token | wool |
Token ID | 25749 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
pleasure | +0.197 |
welcome | +0.171 |
great | +0.165 |
Deal | +0.164 |
learn | +0.158 |
Neg logit contributions | |
led | -0.322 |
ed | -0.277 |
ked | -0.240 |
awed | -0.239 |
ied | -0.237 |
Token | in |
Token ID | 287 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
led | +0.138 |
ed | +0.112 |
awed | +0.105 |
ied | +0.099 |
ked | +0.097 |
Neg logit contributions | |
pleasure | -0.108 |
welcome | -0.100 |
learn | -0.095 |
great | -0.094 |
think | -0.093 |
Token | ice |
Token ID | 4771 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
led | +0.147 |
ed | +0.125 |
ied | +0.108 |
awed | +0.105 |
ked | +0.105 |
Neg logit contributions | |
pleasure | -0.093 |
Deal | -0.082 |
welcome | -0.079 |
learn | -0.076 |
great | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
pleasure | +0.153 |
welcome | +0.149 |
learn | +0.148 |
deal | +0.145 |
think | +0.141 |
Neg logit contributions | |
led | -0.234 |
ed | -0.198 |
ked | -0.179 |
med | -0.177 |
awed | -0.177 |
Token | Anna |
Token ID | 11735 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
led | +0.397 |
ed | +0.365 |
ked | +0.308 |
ted | +0.307 |
med | +0.295 |
Neg logit contributions | |
pleasure | -0.270 |
Deal | -0.246 |
plan | -0.237 |
deal | -0.235 |
promise | -0.231 |
Token | and |
Token ID | 290 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
led | +0.243 |
ed | +0.203 |
ied | +0.182 |
ked | +0.174 |
awed | +0.173 |
Neg logit contributions | |
pleasure | -0.157 |
Deal | -0.138 |
welcome | -0.136 |
learn | -0.133 |
think | -0.131 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
pleasure | +0.129 |
Deal | +0.108 |
welcome | +0.101 |
learn | +0.095 |
great | +0.094 |
Neg logit contributions | |
led | -0.317 |
ed | -0.276 |
ied | -0.244 |
ked | -0.240 |
awed | -0.239 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.190 |
ed | +0.166 |
ked | +0.146 |
ling | +0.144 |
ted | +0.142 |
Neg logit contributions | |
pleasure | -0.072 |
Deal | -0.061 |
welcome | -0.050 |
great | -0.048 |
meantime | -0.046 |
Token | they |
Token ID | 484 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
pleasure | +0.898 |
welcome | +0.748 |
great | +0.728 |
learn | +0.707 |
Deal | +0.680 |
Neg logit contributions | |
led | -1.921 |
ed | -1.667 |
ked | -1.481 |
awed | -1.475 |
ied | -1.436 |
Token | are |
Token ID | 389 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
led | +0.184 |
ed | +0.160 |
ling | +0.139 |
y | +0.138 |
lers | +0.138 |
Neg logit contributions | |
pleasure | -0.067 |
Deal | -0.058 |
welcome | -0.048 |
meantime | -0.041 |
great | -0.040 |
Token | funny |
Token ID | 8258 |
Feature activation | +4.90e+00 |
Pos logit contributions | |
pleasure | +0.328 |
welcome | +0.282 |
Deal | +0.273 |
great | +0.264 |
learn | +0.253 |
Neg logit contributions | |
led | -0.769 |
ed | -0.671 |
ied | -0.598 |
awed | -0.585 |
ked | -0.578 |
Token | . |
Token ID | 13 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
pleasure | +0.212 |
welcome | +0.199 |
fun | +0.191 |
love | +0.184 |
OK | +0.184 |
Neg logit contributions | |
led | -0.369 |
ed | -0.292 |
ied | -0.282 |
ked | -0.272 |
awed | -0.269 |
Token | They |
Token ID | 1119 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
led | +0.410 |
ed | +0.374 |
ted | +0.307 |
y | +0.304 |
ked | +0.298 |
Neg logit contributions | |
pleasure | -0.285 |
Deal | -0.243 |
deal | -0.238 |
plan | -0.233 |
learn | -0.230 |
Token | cat |
Token ID | 3797 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
led | +0.357 |
ed | +0.298 |
ked | +0.247 |
awed | +0.245 |
ted | +0.245 |
Neg logit contributions | |
pleasure | -0.327 |
Deal | -0.290 |
welcome | -0.258 |
think | -0.252 |
know | -0.252 |
Token | said |
Token ID | 531 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
led | +0.359 |
ed | +0.296 |
ied | +0.280 |
ier | +0.273 |
ked | +0.263 |
Neg logit contributions | |
pleasure | -0.205 |
think | -0.201 |
learn | -0.198 |
great | -0.190 |
know | -0.189 |
Token | , |
Token ID | 11 |
Feature activation | +5.37e+00 |
Pos logit contributions | |
led | +0.129 |
ed | +0.111 |
ied | +0.099 |
awed | +0.098 |
ked | +0.097 |
Neg logit contributions | |
pleasure | -0.052 |
welcome | -0.044 |
Deal | -0.043 |
great | -0.041 |
learn | -0.040 |
Token | " |
Token ID | 366 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
pleasure | +0.219 |
Deal | +0.188 |
welcome | +0.176 |
learn | +0.168 |
great | +0.166 |
Neg logit contributions | |
led | -0.443 |
ed | -0.385 |
ied | -0.336 |
ked | -0.332 |
awed | -0.327 |
Token | I |
Token ID | 40 |
Feature activation | -7.58e+00 |
Pos logit contributions | |
pleasure | +0.011 |
Deal | +0.010 |
welcome | +0.009 |
learn | +0.009 |
great | +0.009 |
Neg logit contributions | |
led | -0.017 |
ed | -0.014 |
ied | -0.012 |
ked | -0.012 |
awed | -0.012 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.639 |
ed | +0.576 |
y | +0.506 |
ked | +0.479 |
ied | +0.477 |
Neg logit contributions | |
pleasure | -0.262 |
Deal | -0.243 |
welcome | -0.183 |
meantime | -0.179 |
great | -0.169 |
Token | we |
Token ID | 356 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
pleasure | +0.964 |
Deal | +0.867 |
welcome | +0.774 |
learn | +0.753 |
think | +0.730 |
Neg logit contributions | |
led | -1.844 |
ed | -1.587 |
ied | -1.389 |
ked | -1.347 |
med | -1.344 |
Token | should |
Token ID | 815 |
Feature activation | -5.81e+00 |
Pos logit contributions | |
pleasure | +0.201 |
Deal | +0.176 |
welcome | +0.156 |
great | +0.144 |
learn | +0.143 |
Neg logit contributions | |
led | -0.447 |
ed | -0.391 |
ied | -0.338 |
ked | -0.336 |
y | -0.332 |
Token | cut |
Token ID | 2005 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
led | +0.418 |
ed | +0.365 |
y | +0.340 |
ly | +0.331 |
ling | +0.324 |
Neg logit contributions | |
pleasure | -0.219 |
Deal | -0.194 |
meantime | -0.140 |
Today | -0.140 |
great | -0.136 |
Token | the |
Token ID | 262 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
led | +0.305 |
ed | +0.262 |
ing | +0.222 |
ied | +0.221 |
ted | +0.217 |
Neg logit contributions | |
pleasure | -0.191 |
Deal | -0.172 |
welcome | -0.150 |
meantime | -0.144 |
learn | -0.143 |
Token | pepper |
Token ID | 13385 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
led | +0.117 |
ed | +0.096 |
ling | +0.076 |
ked | +0.075 |
ied | +0.075 |
Neg logit contributions | |
pleasure | -0.125 |
Deal | -0.117 |
welcome | -0.106 |
learn | -0.104 |
think | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
led | +0.163 |
ed | +0.129 |
ied | +0.117 |
awed | +0.115 |
attered | +0.114 |
Neg logit contributions | |
pleasure | -0.099 |
Deal | -0.095 |
welcome | -0.094 |
learn | -0.089 |
OK | -0.089 |
Token |
|
Token ID | 198 |
Feature activation | +9.27e+00 |
Pos logit contributions | |
led | +0.452 |
ed | +0.412 |
ked | +0.350 |
ted | +0.345 |
awed | +0.344 |
Neg logit contributions | |
pleasure | -0.261 |
Deal | -0.211 |
welcome | -0.198 |
great | -0.198 |
deal | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | +9.25e+00 |
Pos logit contributions | |
pleasure | +0.396 |
Deal | +0.294 |
deal | +0.278 |
welcome | +0.263 |
great | +0.261 |
Neg logit contributions | |
led | -0.734 |
ed | -0.673 |
awed | -0.586 |
ied | -0.583 |
med | -0.577 |
Token | " |
Token ID | 1 |
Feature activation | -3.34e-01 |
Pos logit contributions | |
pleasure | +0.474 |
Deal | +0.391 |
welcome | +0.372 |
great | +0.357 |
deal | +0.347 |
Neg logit contributions | |
led | -0.682 |
ed | -0.591 |
ied | -0.501 |
awed | -0.497 |
med | -0.493 |
Token | I |
Token ID | 40 |
Feature activation | -7.58e+00 |
Pos logit contributions | |
led | +0.028 |
ed | +0.025 |
ied | +0.022 |
ked | +0.021 |
awed | +0.021 |
Neg logit contributions | |
pleasure | -0.013 |
Deal | -0.011 |
welcome | -0.010 |
learn | -0.010 |
great | -0.010 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.653 |
ed | +0.587 |
y | +0.505 |
ied | +0.484 |
ked | +0.484 |
Neg logit contributions | |
pleasure | -0.268 |
Deal | -0.243 |
welcome | -0.199 |
meantime | -0.182 |
learn | -0.171 |
Token | so |
Token ID | 523 |
Feature activation | +9.96e+00 |
Pos logit contributions | |
pleasure | +0.860 |
Deal | +0.739 |
welcome | +0.680 |
learn | +0.646 |
great | +0.628 |
Neg logit contributions | |
led | -1.960 |
ed | -1.702 |
ied | -1.505 |
ked | -1.460 |
awed | -1.454 |
Token | , |
Token ID | 11 |
Feature activation | +5.38e+00 |
Pos logit contributions | |
pleasure | +0.576 |
Deal | +0.512 |
meantime | +0.428 |
discuss | +0.426 |
learn | +0.422 |
Neg logit contributions | |
led | -0.618 |
ed | -0.504 |
y | -0.470 |
es | -0.436 |
ing | -0.436 |
Token | Lily |
Token ID | 20037 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
pleasure | +0.256 |
Deal | +0.213 |
learn | +0.202 |
welcome | +0.195 |
meantime | +0.195 |
Neg logit contributions | |
led | -0.349 |
ed | -0.296 |
ly | -0.269 |
ier | -0.257 |
ted | -0.255 |
Token | ," |
Token ID | 553 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
led | +0.005 |
ed | +0.004 |
ked | +0.004 |
ied | +0.004 |
awed | +0.003 |
Neg logit contributions | |
pleasure | -0.003 |
welcome | -0.003 |
think | -0.002 |
have | -0.002 |
great | -0.002 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
pleasure | +0.209 |
Deal | +0.174 |
welcome | +0.162 |
great | +0.160 |
compromise | +0.153 |
Neg logit contributions | |
led | -0.231 |
ed | -0.199 |
ted | -0.182 |
y | -0.169 |
ked | -0.168 |
Token | away |
Token ID | 1497 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
led | +0.288 |
ed | +0.245 |
awed | +0.222 |
ied | +0.219 |
med | +0.218 |
Neg logit contributions | |
pleasure | -0.207 |
welcome | -0.193 |
great | -0.187 |
think | -0.184 |
try | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
led | +0.075 |
ed | +0.061 |
awed | +0.056 |
ked | +0.056 |
med | +0.056 |
Neg logit contributions | |
pleasure | -0.053 |
welcome | -0.049 |
great | -0.047 |
think | -0.045 |
learn | -0.045 |
Token | She |
Token ID | 1375 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
led | +0.437 |
ed | +0.398 |
ied | +0.334 |
awed | +0.332 |
ted | +0.328 |
Neg logit contributions | |
pleasure | -0.259 |
Deal | -0.230 |
welcome | -0.217 |
deal | -0.213 |
great | -0.212 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
led | +0.245 |
ed | +0.203 |
ied | +0.177 |
awed | +0.173 |
ier | +0.173 |
Neg logit contributions | |
pleasure | -0.208 |
think | -0.191 |
learn | -0.186 |
great | -0.185 |
know | -0.183 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
led | +0.074 |
ed | +0.063 |
ied | +0.051 |
ing | +0.049 |
y | +0.048 |
Neg logit contributions | |
pleasure | -0.087 |
Deal | -0.081 |
welcome | -0.075 |
deal | -0.072 |
great | -0.071 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.316 |
ed | +0.265 |
ling | +0.247 |
ked | +0.240 |
ried | +0.231 |
Neg logit contributions | |
pleasure | -0.198 |
Deal | -0.177 |
Today | -0.158 |
meantime | -0.153 |
Not | -0.148 |
Token | of |
Token ID | 286 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
pleasure | +1.216 |
Deal | +1.062 |
welcome | +1.053 |
great | +1.026 |
learn | +1.017 |
Neg logit contributions | |
led | -1.589 |
ed | -1.320 |
awed | -1.121 |
ked | -1.119 |
ied | -1.104 |
Token | a |
Token ID | 257 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
led | +0.844 |
ed | +0.718 |
ied | +0.639 |
ted | +0.630 |
ing | +0.607 |
Neg logit contributions | |
Deal | -0.554 |
pleasure | -0.514 |
Today | -0.459 |
welcome | -0.452 |
deal | -0.442 |
Token | way |
Token ID | 835 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
Deal | +0.129 |
pleasure | +0.125 |
learn | +0.105 |
welcome | +0.099 |
Not | +0.098 |
Neg logit contributions | |
led | -0.254 |
ed | -0.219 |
ling | -0.201 |
y | -0.200 |
ied | -0.192 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
pleasure | +0.548 |
welcome | +0.483 |
Deal | +0.463 |
learn | +0.455 |
great | +0.450 |
Neg logit contributions | |
led | -0.790 |
ed | -0.645 |
awed | -0.564 |
ked | -0.554 |
ied | -0.547 |
Token | help |
Token ID | 1037 |
Feature activation | +7.33e+00 |
Pos logit contributions | |
led | +0.344 |
ed | +0.291 |
ling | +0.262 |
ked | +0.260 |
ied | +0.260 |
Neg logit contributions | |
pleasure | -0.174 |
Deal | -0.168 |
Today | -0.136 |
meantime | -0.126 |
welcome | -0.124 |
Token | 't |
Token ID | 470 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
pleasure | +0.186 |
Deal | +0.136 |
welcome | +0.135 |
deal | +0.125 |
compromise | +0.119 |
Neg logit contributions | |
led | -0.343 |
ed | -0.305 |
ied | -0.270 |
med | -0.267 |
ier | -0.259 |
Token | want |
Token ID | 765 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
pleasure | +0.097 |
Deal | +0.080 |
welcome | +0.065 |
great | +0.062 |
learn | +0.058 |
Neg logit contributions | |
led | -0.281 |
ed | -0.247 |
ied | -0.218 |
y | -0.216 |
ked | -0.216 |
Token | his |
Token ID | 465 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
pleasure | +0.632 |
Deal | +0.572 |
welcome | +0.549 |
learn | +0.532 |
great | +0.528 |
Neg logit contributions | |
led | -0.701 |
ed | -0.579 |
ied | -0.486 |
ked | -0.469 |
awed | -0.467 |
Token | friends |
Token ID | 2460 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
led | +0.248 |
ed | +0.217 |
ied | +0.188 |
ked | +0.181 |
y | +0.181 |
Neg logit contributions | |
pleasure | -0.132 |
Deal | -0.123 |
welcome | -0.105 |
meantime | -0.099 |
learn | -0.099 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
pleasure | +0.199 |
welcome | +0.176 |
Deal | +0.176 |
learn | +0.170 |
great | +0.168 |
Neg logit contributions | |
led | -0.242 |
ed | -0.199 |
ied | -0.168 |
awed | -0.165 |
ked | -0.164 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.340 |
ed | +0.298 |
y | +0.280 |
ling | +0.278 |
ier | +0.270 |
Neg logit contributions | |
pleasure | -0.177 |
Deal | -0.158 |
meantime | -0.112 |
great | -0.108 |
Not | -0.108 |
Token | he |
Token ID | 339 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
pleasure | +1.186 |
welcome | +1.047 |
Deal | +1.025 |
great | +1.023 |
learn | +1.008 |
Neg logit contributions | |
led | -1.607 |
ed | -1.355 |
ked | -1.165 |
awed | -1.154 |
ied | -1.120 |
Token | was |
Token ID | 373 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
led | +0.442 |
ed | +0.362 |
ied | +0.335 |
med | +0.330 |
ked | +0.325 |
Neg logit contributions | |
pleasure | -0.267 |
great | -0.242 |
think | -0.238 |
know | -0.233 |
have | -0.230 |
Token | bad |
Token ID | 2089 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
led | +0.120 |
ed | +0.102 |
ied | +0.089 |
ked | +0.087 |
awed | +0.087 |
Neg logit contributions | |
pleasure | -0.070 |
Deal | -0.061 |
welcome | -0.058 |
learn | -0.056 |
great | -0.056 |
Token | at |
Token ID | 379 |
Feature activation | -7.67e+00 |
Pos logit contributions | |
pleasure | +0.119 |
welcome | +0.107 |
think | +0.102 |
great | +0.101 |
learn | +0.101 |
Neg logit contributions | |
led | -0.165 |
ed | -0.133 |
awed | -0.122 |
ied | -0.120 |
med | -0.120 |
Token | jumping |
Token ID | 14284 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
led | +0.553 |
ed | +0.465 |
ied | +0.421 |
ked | +0.393 |
ling | +0.390 |
Neg logit contributions | |
pleasure | -0.376 |
Deal | -0.345 |
welcome | -0.299 |
learn | -0.293 |
meantime | -0.292 |
Token | bed |
Token ID | 3996 |
Feature activation | +8.42e+00 |
Pos logit contributions | |
led | +0.153 |
ed | +0.131 |
ied | +0.114 |
ling | +0.111 |
ted | +0.109 |
Neg logit contributions | |
pleasure | -0.086 |
Deal | -0.074 |
welcome | -0.068 |
know | -0.067 |
learn | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.422 |
welcome | +0.375 |
great | +0.364 |
learn | +0.363 |
Deal | +0.356 |
Neg logit contributions | |
led | -0.592 |
ed | -0.487 |
awed | -0.437 |
ied | -0.433 |
ked | -0.432 |
Token | They |
Token ID | 1119 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
led | +0.371 |
ed | +0.357 |
ked | +0.285 |
ted | +0.283 |
es | +0.279 |
Neg logit contributions | |
pleasure | -0.265 |
plan | -0.244 |
deal | -0.240 |
learn | -0.231 |
promise | -0.229 |
Token | did |
Token ID | 750 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
pleasure | +0.057 |
think | +0.053 |
learn | +0.052 |
know | +0.050 |
Deal | +0.050 |
Neg logit contributions | |
led | -0.094 |
ed | -0.080 |
ied | -0.071 |
ing | -0.070 |
med | -0.070 |
Token | not |
Token ID | 407 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
led | +0.135 |
ed | +0.115 |
ied | +0.104 |
ling | +0.100 |
ked | +0.100 |
Neg logit contributions | |
pleasure | -0.059 |
Deal | -0.049 |
welcome | -0.044 |
meantime | -0.037 |
Today | -0.035 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.305 |
Deal | +0.250 |
welcome | +0.199 |
great | +0.185 |
meantime | +0.175 |
Neg logit contributions | |
led | -0.926 |
ed | -0.815 |
y | -0.722 |
ied | -0.715 |
ling | -0.711 |
Token | they |
Token ID | 484 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
pleasure | +0.972 |
Deal | +0.828 |
welcome | +0.804 |
great | +0.774 |
learn | +0.768 |
Neg logit contributions | |
led | -1.851 |
ed | -1.597 |
ied | -1.381 |
ked | -1.377 |
awed | -1.374 |
Token | would |
Token ID | 561 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
led | +0.155 |
ed | +0.132 |
med | +0.119 |
ied | +0.118 |
awed | +0.118 |
Neg logit contributions | |
pleasure | -0.083 |
think | -0.082 |
learn | -0.081 |
have | -0.081 |
know | -0.079 |
Token | sleep |
Token ID | 3993 |
Feature activation | +8.13e+00 |
Pos logit contributions | |
led | +0.387 |
ed | +0.331 |
y | +0.314 |
ling | +0.299 |
ier | +0.298 |
Neg logit contributions | |
pleasure | -0.189 |
Deal | -0.163 |
welcome | -0.124 |
meantime | -0.118 |
Today | -0.118 |
Token | well |
Token ID | 880 |
Feature activation | -5.06e+00 |
Pos logit contributions | |
pleasure | +0.443 |
great | +0.410 |
welcome | +0.407 |
Deal | +0.397 |
learn | +0.395 |
Neg logit contributions | |
led | -0.513 |
ed | -0.430 |
awed | -0.379 |
ied | -0.379 |
ked | -0.368 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
led | +0.347 |
ed | +0.293 |
ied | +0.265 |
awed | +0.264 |
ked | +0.258 |
Neg logit contributions | |
pleasure | -0.246 |
great | -0.224 |
welcome | -0.221 |
think | -0.217 |
learn | -0.215 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
pleasure | +0.244 |
welcome | +0.209 |
Deal | +0.202 |
great | +0.199 |
think | +0.197 |
Neg logit contributions | |
led | -0.366 |
ed | -0.302 |
ied | -0.274 |
awed | -0.272 |
ked | -0.267 |
Token | take |
Token ID | 1011 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
pleasure | +0.121 |
Deal | +0.110 |
welcome | +0.087 |
meantime | +0.084 |
great | +0.081 |
Neg logit contributions | |
led | -0.308 |
ed | -0.267 |
ied | -0.234 |
ling | -0.228 |
ked | -0.227 |
Token | some |
Token ID | 617 |
Feature activation | +4.21e+00 |
Pos logit contributions | |
pleasure | +0.108 |
Deal | +0.097 |
welcome | +0.086 |
learn | +0.082 |
think | +0.080 |
Neg logit contributions | |
led | -0.228 |
ed | -0.200 |
ied | -0.173 |
med | -0.169 |
ted | -0.167 |
Token | time |
Token ID | 640 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
pleasure | +0.157 |
Deal | +0.132 |
welcome | +0.125 |
learn | +0.118 |
great | +0.117 |
Neg logit contributions | |
led | -0.364 |
ed | -0.316 |
ied | -0.279 |
ked | -0.274 |
awed | -0.273 |
Token | to |
Token ID | 284 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
pleasure | +0.568 |
welcome | +0.498 |
learn | +0.488 |
great | +0.480 |
think | +0.473 |
Neg logit contributions | |
led | -0.727 |
ed | -0.591 |
awed | -0.521 |
ked | -0.513 |
ied | -0.504 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
led | +0.319 |
ed | +0.275 |
med | +0.252 |
ied | +0.252 |
ling | +0.250 |
Neg logit contributions | |
pleasure | -0.166 |
Deal | -0.156 |
Today | -0.123 |
Not | -0.118 |
meantime | -0.111 |
Token | before |
Token ID | 878 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
pleasure | +1.273 |
welcome | +1.116 |
Deal | +1.103 |
great | +1.101 |
learn | +1.083 |
Neg logit contributions | |
led | -1.526 |
ed | -1.264 |
awed | -1.090 |
ked | -1.081 |
ied | -1.033 |
Token | he |
Token ID | 339 |
Feature activation | -5.95e+00 |
Pos logit contributions | |
led | +0.206 |
ed | +0.174 |
ied | +0.149 |
med | +0.146 |
ted | +0.146 |
Neg logit contributions | |
pleasure | -0.148 |
Deal | -0.135 |
welcome | -0.124 |
learn | -0.116 |
deal | -0.114 |
Token | did |
Token ID | 750 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
led | +0.385 |
ed | +0.312 |
ied | +0.276 |
awed | +0.267 |
ked | +0.264 |
Neg logit contributions | |
pleasure | -0.341 |
Deal | -0.299 |
learn | -0.299 |
welcome | -0.299 |
think | -0.299 |
Token | anything |
Token ID | 1997 |
Feature activation | +6.16e+00 |
Pos logit contributions | |
led | +0.127 |
ed | +0.109 |
ied | +0.095 |
ked | +0.092 |
awed | +0.091 |
Neg logit contributions | |
pleasure | -0.069 |
Deal | -0.062 |
welcome | -0.057 |
learn | -0.053 |
great | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.320 |
great | +0.279 |
welcome | +0.278 |
Deal | +0.271 |
think | +0.267 |
Neg logit contributions | |
led | -0.435 |
ed | -0.370 |
ked | -0.318 |
ied | -0.317 |
awed | -0.317 |
Token | , |
Token ID | 11 |
Feature activation | +5.38e+00 |
Pos logit contributions | |
led | +0.359 |
ed | +0.298 |
ied | +0.285 |
med | +0.268 |
ked | +0.266 |
Neg logit contributions | |
pleasure | -0.158 |
welcome | -0.131 |
great | -0.127 |
learn | -0.122 |
Deal | -0.120 |
Token | so |
Token ID | 523 |
Feature activation | +9.97e+00 |
Pos logit contributions | |
pleasure | +0.222 |
Deal | +0.202 |
welcome | +0.180 |
learn | +0.171 |
great | +0.167 |
Neg logit contributions | |
led | -0.435 |
ed | -0.376 |
ied | -0.331 |
awed | -0.322 |
ked | -0.321 |
Token | she |
Token ID | 673 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
pleasure | +0.582 |
Deal | +0.535 |
welcome | +0.504 |
learn | +0.490 |
think | +0.479 |
Neg logit contributions | |
led | -0.649 |
ed | -0.538 |
ied | -0.450 |
awed | -0.427 |
ked | -0.426 |
Token | couldn |
Token ID | 3521 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
led | +0.299 |
ed | +0.245 |
ier | +0.229 |
ked | +0.227 |
awed | +0.224 |
Neg logit contributions | |
pleasure | -0.211 |
great | -0.210 |
think | -0.203 |
have | -0.201 |
know | -0.195 |
Token | 't |
Token ID | 470 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
led | +0.278 |
ed | +0.252 |
ied | +0.220 |
med | +0.216 |
y | +0.214 |
Neg logit contributions | |
pleasure | -0.119 |
Deal | -0.096 |
welcome | -0.086 |
deal | -0.079 |
great | -0.077 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.107 |
Deal | +0.090 |
welcome | +0.081 |
great | +0.071 |
deal | +0.067 |
Neg logit contributions | |
led | -0.266 |
ed | -0.235 |
ied | -0.207 |
ked | -0.204 |
y | -0.203 |
Token | of |
Token ID | 286 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
pleasure | +1.247 |
welcome | +1.092 |
Deal | +1.090 |
great | +1.077 |
learn | +1.073 |
Neg logit contributions | |
led | -1.569 |
ed | -1.297 |
ked | -1.109 |
awed | -1.104 |
ied | -1.057 |
Token | a |
Token ID | 257 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
led | +0.823 |
ed | +0.714 |
ied | +0.627 |
we | +0.607 |
med | +0.606 |
Neg logit contributions | |
Deal | -0.574 |
pleasure | -0.554 |
welcome | -0.470 |
Today | -0.463 |
deal | -0.462 |
Token | good |
Token ID | 922 |
Feature activation | +5.67e+00 |
Pos logit contributions | |
pleasure | +0.151 |
Deal | +0.146 |
welcome | +0.123 |
learn | +0.122 |
know | +0.118 |
Neg logit contributions | |
led | -0.241 |
ed | -0.208 |
y | -0.181 |
ied | -0.180 |
ling | -0.176 |
Token | idea |
Token ID | 2126 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
pleasure | +0.160 |
Deal | +0.147 |
welcome | +0.114 |
learn | +0.108 |
great | +0.101 |
Neg logit contributions | |
led | -0.524 |
ed | -0.467 |
ied | -0.416 |
y | -0.408 |
ling | -0.402 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
pleasure | +0.580 |
welcome | +0.565 |
learn | +0.534 |
think | +0.520 |
have | +0.516 |
Neg logit contributions | |
led | -1.014 |
ed | -0.792 |
awed | -0.756 |
ked | -0.741 |
ling | -0.709 |
Token | a |
Token ID | 257 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
pleasure | +0.454 |
Deal | +0.407 |
welcome | +0.380 |
learn | +0.360 |
great | +0.351 |
Neg logit contributions | |
led | -0.780 |
ed | -0.669 |
ied | -0.579 |
ked | -0.563 |
awed | -0.560 |
Token | diary |
Token ID | 26339 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
Deal | +0.170 |
pleasure | +0.168 |
welcome | +0.144 |
learn | +0.137 |
meantime | +0.136 |
Neg logit contributions | |
led | -0.201 |
ed | -0.174 |
y | -0.149 |
ly | -0.148 |
lers | -0.141 |
Token | . |
Token ID | 13 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
led | +0.172 |
ed | +0.140 |
ied | +0.126 |
awed | +0.125 |
ked | +0.121 |
Neg logit contributions | |
pleasure | -0.109 |
welcome | -0.095 |
Deal | -0.092 |
learn | -0.092 |
great | -0.090 |
Token | Do |
Token ID | 2141 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
led | +0.398 |
ed | +0.378 |
ked | +0.315 |
ted | +0.313 |
med | +0.308 |
Neg logit contributions | |
pleasure | -0.240 |
plan | -0.225 |
great | -0.220 |
deal | -0.206 |
meantime | -0.203 |
Token | you |
Token ID | 345 |
Feature activation | +9.89e+00 |
Pos logit contributions | |
pleasure | +0.200 |
Deal | +0.154 |
welcome | +0.150 |
meantime | +0.131 |
deal | +0.129 |
Neg logit contributions | |
led | -0.385 |
ed | -0.330 |
ied | -0.301 |
ted | -0.291 |
we | -0.291 |
Token | think |
Token ID | 892 |
Feature activation | +2.29e+01 |
Pos logit contributions | |
pleasure | +0.389 |
Deal | +0.342 |
welcome | +0.275 |
great | +0.269 |
meantime | +0.267 |
Neg logit contributions | |
led | -0.775 |
ed | -0.668 |
y | -0.607 |
ier | -0.597 |
ly | -0.581 |
Token | I |
Token ID | 314 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
pleasure | +1.033 |
Deal | +0.919 |
welcome | +0.838 |
learn | +0.798 |
think | +0.789 |
Neg logit contributions | |
led | -1.774 |
ed | -1.524 |
ied | -1.341 |
med | -1.292 |
ted | -1.272 |
Token | should |
Token ID | 815 |
Feature activation | -5.81e+00 |
Pos logit contributions | |
pleasure | +0.132 |
Deal | +0.118 |
welcome | +0.106 |
great | +0.100 |
learn | +0.098 |
Neg logit contributions | |
led | -0.249 |
ed | -0.216 |
ied | -0.186 |
ked | -0.182 |
y | -0.181 |
Token | get |
Token ID | 651 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
led | +0.402 |
ed | +0.345 |
ly | +0.321 |
y | +0.306 |
ied | +0.305 |
Neg logit contributions | |
pleasure | -0.246 |
Deal | -0.218 |
welcome | -0.174 |
meantime | -0.174 |
Today | -0.168 |
Token | one |
Token ID | 530 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
pleasure | +0.426 |
Deal | +0.369 |
welcome | +0.340 |
learn | +0.318 |
great | +0.315 |
Neg logit contributions | |
led | -0.923 |
ed | -0.802 |
ied | -0.702 |
ked | -0.684 |
med | -0.680 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.08e+00 |
Pos logit contributions | |
pleasure | +0.375 |
Deal | +0.331 |
welcome | +0.287 |
learn | +0.283 |
great | +0.281 |
Neg logit contributions | |
led | -0.683 |
ed | -0.593 |
ied | -0.515 |
y | -0.509 |
ier | -0.496 |
Token | bed |
Token ID | 3996 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
led | +0.126 |
ed | +0.107 |
ied | +0.092 |
ted | +0.087 |
med | +0.087 |
Neg logit contributions | |
pleasure | -0.095 |
Deal | -0.090 |
welcome | -0.080 |
learn | -0.076 |
great | -0.074 |
Token | that |
Token ID | 326 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
pleasure | +0.172 |
learn | +0.155 |
welcome | +0.154 |
great | +0.149 |
think | +0.149 |
Neg logit contributions | |
led | -0.257 |
ed | -0.214 |
ked | -0.200 |
awed | -0.193 |
ied | -0.191 |
Token | he |
Token ID | 339 |
Feature activation | -8.94e+00 |
Pos logit contributions | |
led | +0.350 |
ed | +0.297 |
ied | +0.257 |
ked | +0.247 |
awed | +0.247 |
Neg logit contributions | |
pleasure | -0.222 |
Deal | -0.197 |
welcome | -0.186 |
learn | -0.178 |
great | -0.176 |
Token | sn |
Token ID | 3013 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
led | +0.595 |
ed | +0.499 |
ied | +0.429 |
ked | +0.425 |
awed | +0.419 |
Neg logit contributions | |
pleasure | -0.491 |
Deal | -0.438 |
learn | -0.427 |
welcome | -0.426 |
think | -0.424 |
Token | ugg |
Token ID | 6837 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
led | +0.447 |
ed | +0.346 |
ked | +0.291 |
ied | +0.286 |
med | +0.278 |
Neg logit contributions | |
pleasure | -0.501 |
Deal | -0.462 |
welcome | -0.453 |
learn | -0.434 |
great | -0.423 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.042 |
ked | -0.047 |
ed | -0.057 |
awed | -0.102 |
ied | -0.110 |
Neg logit contributions | |
pleasure | -1.487 |
welcome | -1.431 |
great | -1.419 |
Deal | -1.418 |
learn | -1.388 |
Token | in |
Token ID | 287 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
led | +1.667 |
ed | +1.372 |
ied | +1.165 |
awed | +1.146 |
ked | +1.142 |
Neg logit contributions | |
pleasure | -1.546 |
Deal | -1.371 |
welcome | -1.355 |
great | -1.320 |
learn | -1.306 |
Token | . |
Token ID | 13 |
Feature activation | -7.40e+00 |
Pos logit contributions | |
led | +0.990 |
ed | +0.852 |
ied | +0.726 |
ted | +0.725 |
med | +0.725 |
Neg logit contributions | |
pleasure | -0.606 |
Deal | -0.558 |
learn | -0.487 |
welcome | -0.484 |
compromise | -0.482 |
Token | But |
Token ID | 887 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
led | +0.501 |
ed | +0.463 |
ted | +0.396 |
med | +0.381 |
awed | +0.378 |
Neg logit contributions | |
pleasure | -0.304 |
deal | -0.280 |
great | -0.279 |
plan | -0.278 |
learn | -0.272 |
Token | one |
Token ID | 530 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
pleasure | +0.517 |
Deal | +0.470 |
welcome | +0.440 |
learn | +0.425 |
great | +0.419 |
Neg logit contributions | |
led | -0.719 |
ed | -0.608 |
ied | -0.517 |
ked | -0.500 |
ted | -0.498 |
Token | night |
Token ID | 1755 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
pleasure | +0.232 |
welcome | +0.181 |
Deal | +0.172 |
great | +0.166 |
learn | +0.160 |
Neg logit contributions | |
led | -0.694 |
ed | -0.609 |
ked | -0.550 |
awed | -0.543 |
ied | -0.540 |
Token | The |
Token ID | 383 |
Feature activation | -7.93e+00 |
Pos logit contributions | |
led | +0.174 |
ed | +0.149 |
awed | +0.127 |
ied | +0.126 |
ked | +0.126 |
Neg logit contributions | |
pleasure | -0.121 |
welcome | -0.108 |
great | -0.103 |
Deal | -0.101 |
OK | -0.098 |
Token | cater |
Token ID | 20825 |
Feature activation | -1.16e+01 |
Pos logit contributions | |
led | +0.575 |
ed | +0.485 |
ied | +0.417 |
ling | +0.412 |
y | +0.408 |
Neg logit contributions | |
pleasure | -0.377 |
Deal | -0.337 |
welcome | -0.307 |
learn | -0.300 |
know | -0.296 |
Token | pillar |
Token ID | 41643 |
Feature activation | -1.54e+01 |
Pos logit contributions | |
led | +0.620 |
ed | +0.481 |
ied | +0.415 |
ked | +0.414 |
awed | +0.371 |
Neg logit contributions | |
pleasure | -0.764 |
welcome | -0.707 |
Deal | -0.703 |
learn | -0.681 |
think | -0.681 |
Token | w |
Token ID | 266 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
led | +1.034 |
ed | +0.866 |
ied | +0.756 |
ked | +0.742 |
ier | +0.715 |
Neg logit contributions | |
pleasure | -0.817 |
Deal | -0.722 |
welcome | -0.721 |
learn | -0.716 |
think | -0.715 |
Token | igg |
Token ID | 6950 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
led | +0.184 |
ed | +0.151 |
ied | +0.127 |
ked | +0.126 |
med | +0.124 |
Neg logit contributions | |
pleasure | -0.170 |
Deal | -0.152 |
welcome | -0.149 |
great | -0.142 |
learn | -0.141 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.077 |
med | +0.023 |
we | +0.020 |
ed | +0.005 |
awed | +0.002 |
Neg logit contributions | |
pleasure | -0.918 |
think | -0.896 |
great | -0.894 |
welcome | -0.889 |
know | -0.877 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
led | +1.962 |
ed | +1.661 |
ied | +1.439 |
awed | +1.402 |
ked | +1.401 |
Neg logit contributions | |
pleasure | -1.273 |
Deal | -1.115 |
welcome | -1.081 |
learn | -1.031 |
great | -1.030 |
Token | and |
Token ID | 290 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
led | +0.826 |
ed | +0.699 |
ied | +0.601 |
ked | +0.595 |
awed | +0.590 |
Neg logit contributions | |
pleasure | -0.534 |
Deal | -0.465 |
welcome | -0.458 |
learn | -0.437 |
great | -0.433 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
pleasure | +0.111 |
think | +0.099 |
welcome | +0.098 |
learn | +0.098 |
great | +0.096 |
Neg logit contributions | |
led | -0.132 |
ed | -0.108 |
ied | -0.097 |
ier | -0.095 |
awed | -0.094 |
Token | safe |
Token ID | 3338 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
led | +0.282 |
ed | +0.244 |
ied | +0.213 |
ling | +0.207 |
ked | +0.207 |
Neg logit contributions | |
pleasure | -0.117 |
Deal | -0.104 |
welcome | -0.090 |
learn | -0.088 |
deal | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -7.05e+00 |
Pos logit contributions | |
pleasure | +0.117 |
welcome | +0.105 |
think | +0.094 |
great | +0.094 |
be | +0.093 |
Neg logit contributions | |
led | -0.179 |
ed | -0.143 |
ied | -0.138 |
awed | -0.137 |
ked | -0.134 |
Token | she |
Token ID | 673 |
Feature activation | -9.20e+00 |
Pos logit contributions | |
led | +0.805 |
ed | +0.685 |
awed | +0.629 |
ked | +0.612 |
ied | +0.611 |
Neg logit contributions | |
pleasure | -0.392 |
welcome | -0.337 |
great | -0.336 |
learn | -0.313 |
Deal | -0.309 |
Token | did |
Token ID | 750 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
led | +0.587 |
ed | +0.480 |
ied | +0.412 |
ked | +0.409 |
awed | +0.404 |
Neg logit contributions | |
pleasure | -0.534 |
welcome | -0.473 |
Deal | -0.471 |
learn | -0.461 |
think | -0.458 |
Token | , |
Token ID | 11 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
led | +0.413 |
ed | +0.358 |
ied | +0.314 |
med | +0.303 |
y | +0.298 |
Neg logit contributions | |
pleasure | -0.226 |
Deal | -0.201 |
welcome | -0.179 |
deal | -0.160 |
great | -0.159 |
Token | it |
Token ID | 340 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
pleasure | +0.116 |
Deal | +0.104 |
welcome | +0.097 |
great | +0.093 |
learn | +0.093 |
Neg logit contributions | |
led | -0.177 |
ed | -0.151 |
ied | -0.129 |
ked | -0.127 |
awed | -0.125 |
Token | spark |
Token ID | 9009 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
pleasure | +0.143 |
think | +0.133 |
great | +0.132 |
learn | +0.131 |
welcome | +0.128 |
Neg logit contributions | |
led | -0.207 |
ed | -0.174 |
ied | -0.158 |
ked | -0.158 |
awed | -0.154 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.050 |
year | +0.017 |
ied | +0.011 |
awed | -0.000 |
ed | -0.001 |
Neg logit contributions | |
great | -0.606 |
pleasure | -0.602 |
fun | -0.573 |
think | -0.572 |
be | -0.572 |
Token | with |
Token ID | 351 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
led | +1.883 |
ed | +1.579 |
ied | +1.394 |
awed | +1.380 |
ked | +1.364 |
Neg logit contributions | |
pleasure | -1.352 |
welcome | -1.182 |
great | -1.150 |
Deal | -1.120 |
think | -1.079 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
led | +0.309 |
ed | +0.270 |
ied | +0.237 |
ked | +0.232 |
awed | +0.231 |
Neg logit contributions | |
pleasure | -0.127 |
Deal | -0.112 |
welcome | -0.101 |
learn | -0.097 |
think | -0.093 |
Token | most |
Token ID | 749 |
Feature activation | +6.71e+00 |
Pos logit contributions | |
led | +0.260 |
ed | +0.220 |
ied | +0.190 |
ked | +0.184 |
awed | +0.183 |
Neg logit contributions | |
pleasure | -0.166 |
Deal | -0.150 |
welcome | -0.140 |
learn | -0.136 |
great | -0.132 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
pleasure | +0.259 |
welcome | +0.203 |
Deal | +0.202 |
great | +0.199 |
learn | +0.188 |
Neg logit contributions | |
led | -0.574 |
ed | -0.501 |
ied | -0.443 |
ked | -0.440 |
awed | -0.433 |
Token | light |
Token ID | 1657 |
Feature activation | +5.06e-01 |
Pos logit contributions | |
pleasure | +0.079 |
welcome | +0.066 |
great | +0.064 |
Deal | +0.064 |
learn | +0.062 |
Neg logit contributions | |
led | -0.124 |
ed | -0.105 |
ied | -0.092 |
ked | -0.092 |
awed | -0.090 |
Token | , |
Token ID | 11 |
Feature activation | +7.14e-01 |
Pos logit contributions | |
pleasure | +0.199 |
great | +0.184 |
learn | +0.168 |
think | +0.167 |
have | +0.167 |
Neg logit contributions | |
led | -0.349 |
awed | -0.282 |
ed | -0.281 |
ked | -0.274 |
ied | -0.263 |
Token | Bob |
Token ID | 5811 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
pleasure | +0.033 |
Deal | +0.029 |
welcome | +0.026 |
learn | +0.026 |
great | +0.025 |
Neg logit contributions | |
led | -0.054 |
ed | -0.046 |
ied | -0.040 |
ked | -0.039 |
y | -0.039 |
Token | 's |
Token ID | 338 |
Feature activation | -5.78e+00 |
Pos logit contributions | |
led | +0.197 |
ed | +0.161 |
ied | +0.143 |
ked | +0.136 |
awed | +0.135 |
Neg logit contributions | |
pleasure | -0.164 |
welcome | -0.149 |
Deal | -0.143 |
think | -0.142 |
learn | -0.142 |
Token | eyes |
Token ID | 2951 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
led | +0.457 |
ed | +0.398 |
ied | +0.352 |
ked | +0.341 |
y | +0.339 |
Neg logit contributions | |
pleasure | -0.215 |
Deal | -0.198 |
learn | -0.177 |
welcome | -0.174 |
think | -0.169 |
Token | spark |
Token ID | 9009 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
pleasure | +0.040 |
Deal | +0.035 |
welcome | +0.035 |
great | +0.035 |
learn | +0.034 |
Neg logit contributions | |
led | -0.055 |
ed | -0.046 |
ied | -0.040 |
awed | -0.039 |
ked | -0.039 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.165 |
year | +0.154 |
ied | +0.149 |
awed | +0.144 |
ase | +0.134 |
Neg logit contributions | |
great | -0.465 |
pleasure | -0.455 |
fun | -0.426 |
look | -0.419 |
love | -0.417 |
Token | with |
Token ID | 351 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
led | +1.722 |
ed | +1.430 |
awed | +1.276 |
ied | +1.269 |
ked | +1.216 |
Neg logit contributions | |
pleasure | -1.480 |
welcome | -1.326 |
great | -1.300 |
Deal | -1.194 |
fun | -1.193 |
Token | joy |
Token ID | 8716 |
Feature activation | +1.25e+01 |
Pos logit contributions | |
led | +0.314 |
ed | +0.274 |
ied | +0.241 |
ked | +0.237 |
ted | +0.236 |
Neg logit contributions | |
pleasure | -0.119 |
Deal | -0.109 |
learn | -0.093 |
welcome | -0.093 |
think | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | -7.26e+00 |
Pos logit contributions | |
pleasure | +0.559 |
welcome | +0.469 |
love | +0.463 |
great | +0.461 |
be | +0.444 |
Neg logit contributions | |
led | -0.838 |
ed | -0.696 |
awed | -0.663 |
ied | -0.639 |
attered | -0.626 |
Token | His |
Token ID | 2399 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
led | +0.487 |
ed | +0.446 |
ted | +0.383 |
awed | +0.375 |
y | +0.370 |
Neg logit contributions | |
pleasure | -0.304 |
learn | -0.286 |
great | -0.284 |
deal | -0.276 |
Deal | -0.274 |
Token | sister |
Token ID | 6621 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
led | +0.429 |
ed | +0.372 |
ied | +0.328 |
ked | +0.317 |
lers | +0.315 |
Neg logit contributions | |
pleasure | -0.184 |
Deal | -0.180 |
welcome | -0.159 |
learn | -0.153 |
meantime | -0.146 |
Token | so |
Token ID | 523 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
pleasure | +0.104 |
Deal | +0.082 |
welcome | +0.075 |
learn | +0.067 |
great | +0.066 |
Neg logit contributions | |
led | -0.318 |
ed | -0.281 |
ied | -0.252 |
ked | -0.244 |
awed | -0.244 |
Token | happy |
Token ID | 3772 |
Feature activation | +7.43e+00 |
Pos logit contributions | |
pleasure | +0.218 |
Deal | +0.198 |
welcome | +0.172 |
learn | +0.169 |
deal | +0.164 |
Neg logit contributions | |
led | -0.437 |
ed | -0.373 |
ied | -0.329 |
ling | -0.315 |
awed | -0.312 |
Token | that |
Token ID | 326 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
pleasure | +0.367 |
welcome | +0.322 |
think | +0.303 |
learn | +0.303 |
great | +0.299 |
Neg logit contributions | |
led | -0.534 |
ed | -0.440 |
ked | -0.397 |
ied | -0.393 |
awed | -0.384 |
Token | she |
Token ID | 673 |
Feature activation | -7.11e+00 |
Pos logit contributions | |
led | +0.274 |
ed | +0.229 |
ied | +0.195 |
awed | +0.185 |
ked | +0.185 |
Neg logit contributions | |
pleasure | -0.212 |
Deal | -0.194 |
welcome | -0.180 |
learn | -0.176 |
great | -0.173 |
Token | whist |
Token ID | 32262 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
led | +0.448 |
ed | +0.365 |
ied | +0.323 |
ked | +0.321 |
awed | +0.320 |
Neg logit contributions | |
pleasure | -0.410 |
welcome | -0.367 |
great | -0.360 |
learn | -0.358 |
think | -0.358 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.029 |
ed | -0.015 |
ked | -0.035 |
awed | -0.039 |
ied | -0.043 |
Neg logit contributions | |
pleasure | -0.561 |
great | -0.550 |
welcome | -0.537 |
learn | -0.529 |
Deal | -0.528 |
Token | all |
Token ID | 477 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
led | +1.897 |
ed | +1.600 |
ied | +1.370 |
ked | +1.333 |
awed | +1.328 |
Neg logit contributions | |
pleasure | -1.333 |
Deal | -1.187 |
welcome | -1.135 |
learn | -1.093 |
great | -1.083 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
led | +0.046 |
ed | +0.040 |
ked | +0.036 |
awed | +0.036 |
ied | +0.035 |
Neg logit contributions | |
pleasure | -0.022 |
welcome | -0.020 |
great | -0.018 |
Deal | -0.018 |
learn | -0.017 |
Token | way |
Token ID | 835 |
Feature activation | +9.36e+00 |
Pos logit contributions | |
led | +0.311 |
ed | +0.268 |
ied | +0.239 |
ling | +0.235 |
ked | +0.230 |
Neg logit contributions | |
pleasure | -0.118 |
Deal | -0.104 |
welcome | -0.090 |
learn | -0.087 |
know | -0.083 |
Token | home |
Token ID | 1363 |
Feature activation | +6.58e+00 |
Pos logit contributions | |
pleasure | +0.521 |
welcome | +0.461 |
great | +0.438 |
learn | +0.438 |
Deal | +0.419 |
Neg logit contributions | |
led | -0.628 |
ed | -0.501 |
awed | -0.446 |
ked | -0.435 |
ied | -0.417 |
Token | . |
Token ID | 13 |
Feature activation | -7.05e+00 |
Pos logit contributions | |
pleasure | +0.346 |
welcome | +0.305 |
learn | +0.302 |
great | +0.296 |
think | +0.284 |
Neg logit contributions | |
led | -0.456 |
ed | -0.364 |
awed | -0.336 |
ked | -0.327 |
ied | -0.323 |
Token | around |
Token ID | 1088 |
Feature activation | -1.37e+01 |
Pos logit contributions | |
led | +0.324 |
ed | +0.262 |
awed | +0.240 |
ked | +0.237 |
ied | +0.230 |
Neg logit contributions | |
pleasure | -0.191 |
welcome | -0.182 |
great | -0.165 |
OK | -0.162 |
think | -0.162 |
Token | . |
Token ID | 13 |
Feature activation | -9.52e+00 |
Pos logit contributions | |
led | +1.012 |
ed | +0.857 |
ied | +0.737 |
awed | +0.714 |
ked | +0.712 |
Neg logit contributions | |
pleasure | -0.672 |
Deal | -0.602 |
welcome | -0.564 |
learn | -0.540 |
great | -0.537 |
Token | She |
Token ID | 1375 |
Feature activation | -6.79e+00 |
Pos logit contributions | |
led | +0.767 |
ed | +0.694 |
ted | +0.592 |
ied | +0.590 |
med | +0.589 |
Neg logit contributions | |
pleasure | -0.355 |
Deal | -0.301 |
welcome | -0.282 |
great | -0.280 |
deal | -0.279 |
Token | w |
Token ID | 266 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
led | +0.438 |
ed | +0.360 |
ied | +0.308 |
ked | +0.301 |
awed | +0.297 |
Neg logit contributions | |
pleasure | -0.388 |
Deal | -0.346 |
welcome | -0.340 |
learn | -0.339 |
think | -0.337 |
Token | igg |
Token ID | 6950 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
led | +0.160 |
ed | +0.131 |
ked | +0.104 |
ied | +0.102 |
ted | +0.101 |
Neg logit contributions | |
pleasure | -0.191 |
Deal | -0.176 |
welcome | -0.169 |
great | -0.164 |
learn | -0.163 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.083 |
ed | +0.038 |
ked | +0.036 |
we | +0.029 |
awed | +0.024 |
Neg logit contributions | |
think | -0.863 |
great | -0.861 |
pleasure | -0.856 |
learn | -0.849 |
know | -0.846 |
Token | it |
Token ID | 340 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
led | +2.015 |
ed | +1.707 |
ied | +1.497 |
awed | +1.469 |
ked | +1.458 |
Neg logit contributions | |
pleasure | -1.220 |
Deal | -1.046 |
welcome | -1.025 |
great | -0.986 |
learn | -0.975 |
Token | and |
Token ID | 290 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
pleasure | +0.181 |
great | +0.163 |
think | +0.161 |
welcome | +0.158 |
learn | +0.157 |
Neg logit contributions | |
led | -0.264 |
ed | -0.216 |
ied | -0.202 |
awed | -0.199 |
ked | -0.191 |
Token | squ |
Token ID | 2809 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
pleasure | +0.111 |
think | +0.099 |
learn | +0.098 |
great | +0.097 |
welcome | +0.097 |
Neg logit contributions | |
led | -0.133 |
ed | -0.110 |
ied | -0.099 |
ier | -0.095 |
awed | -0.094 |
Token | igg |
Token ID | 6950 |
Feature activation | -9.25e+00 |
Pos logit contributions | |
led | +0.088 |
ed | +0.067 |
ied | +0.059 |
we | +0.055 |
web | +0.054 |
Neg logit contributions | |
pleasure | -0.138 |
Deal | -0.126 |
welcome | -0.123 |
great | -0.120 |
think | -0.120 |
Token | led |
Token ID | 992 |
Feature activation | -2.54e+01 |
Pos logit contributions | |
led | +0.052 |
year | +0.030 |
web | +0.026 |
ked | +0.019 |
awed | +0.015 |
Neg logit contributions | |
great | -0.776 |
pleasure | -0.756 |
good | -0.743 |
think | -0.743 |
know | -0.724 |
Token | graceful |
Token ID | 44363 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
led | +0.253 |
ed | +0.210 |
ied | +0.169 |
ling | +0.168 |
y | +0.163 |
Neg logit contributions | |
pleasure | -0.246 |
Deal | -0.232 |
welcome | -0.212 |
learn | -0.210 |
know | -0.206 |
Token | car |
Token ID | 1097 |
Feature activation | -4.86e+00 |
Pos logit contributions | |
led | +0.426 |
ed | +0.357 |
ied | +0.311 |
awed | +0.305 |
ked | +0.305 |
Neg logit contributions | |
pleasure | -0.297 |
welcome | -0.254 |
Deal | -0.251 |
learn | -0.241 |
great | -0.241 |
Token | - |
Token ID | 532 |
Feature activation | -4.65e+00 |
Pos logit contributions | |
led | +0.335 |
ed | +0.267 |
ied | +0.246 |
awed | +0.246 |
ked | +0.235 |
Neg logit contributions | |
pleasure | -0.243 |
welcome | -0.214 |
learn | -0.209 |
think | -0.204 |
great | -0.204 |
Token | it |
Token ID | 340 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
led | +0.355 |
ed | +0.310 |
ied | +0.266 |
ked | +0.262 |
y | +0.261 |
Neg logit contributions | |
pleasure | -0.195 |
Deal | -0.183 |
welcome | -0.158 |
learn | -0.151 |
meantime | -0.148 |
Token | spark |
Token ID | 9009 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
pleasure | +0.153 |
welcome | +0.133 |
great | +0.130 |
Deal | +0.129 |
think | +0.129 |
Neg logit contributions | |
led | -0.213 |
ed | -0.179 |
ied | -0.156 |
ked | -0.155 |
awed | -0.154 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.091 |
ed | +0.040 |
ied | +0.028 |
awed | +0.027 |
ked | +0.023 |
Neg logit contributions | |
pleasure | -0.617 |
great | -0.594 |
think | -0.562 |
welcome | -0.561 |
fun | -0.552 |
Token | in |
Token ID | 287 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
led | +1.785 |
ed | +1.487 |
awed | +1.284 |
ied | +1.284 |
ked | +1.256 |
Neg logit contributions | |
pleasure | -1.462 |
welcome | -1.276 |
great | -1.240 |
Deal | -1.194 |
think | -1.164 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
led | +0.938 |
ed | +0.800 |
med | +0.700 |
ied | +0.696 |
y | +0.683 |
Neg logit contributions | |
pleasure | -0.643 |
Deal | -0.616 |
learn | -0.540 |
deal | -0.533 |
welcome | -0.521 |
Token | sunshine |
Token ID | 34488 |
Feature activation | +7.20e+00 |
Pos logit contributions | |
led | +0.268 |
ed | +0.228 |
ied | +0.197 |
ling | +0.191 |
ked | +0.191 |
Neg logit contributions | |
pleasure | -0.159 |
Deal | -0.143 |
welcome | -0.131 |
learn | -0.128 |
think | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | -7.13e+00 |
Pos logit contributions | |
pleasure | +0.380 |
welcome | +0.333 |
learn | +0.314 |
great | +0.313 |
think | +0.307 |
Neg logit contributions | |
led | -0.499 |
ed | -0.408 |
ied | -0.359 |
awed | -0.356 |
ked | -0.356 |
Token | One |
Token ID | 1881 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
led | +0.570 |
ed | +0.521 |
ted | +0.449 |
med | +0.447 |
ied | +0.442 |
Neg logit contributions | |
pleasure | -0.247 |
Deal | -0.214 |
great | -0.205 |
deal | -0.205 |
learn | -0.201 |
Token |
|
Token ID | 198 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
pleasure | +0.203 |
Deal | +0.176 |
welcome | +0.170 |
learn | +0.162 |
great | +0.161 |
Neg logit contributions | |
led | -0.357 |
ed | -0.305 |
ied | -0.266 |
awed | -0.258 |
ked | -0.258 |
Token |
|
Token ID | 198 |
Feature activation | +5.61e+00 |
Pos logit contributions | |
pleasure | +0.209 |
deal | +0.164 |
Deal | +0.160 |
great | +0.153 |
welcome | +0.151 |
Neg logit contributions | |
led | -0.340 |
ed | -0.298 |
ied | -0.264 |
med | -0.263 |
y | -0.261 |
Token | S |
Token ID | 50 |
Feature activation | -6.61e+00 |
Pos logit contributions | |
pleasure | +0.312 |
Deal | +0.253 |
welcome | +0.245 |
deal | +0.244 |
great | +0.239 |
Neg logit contributions | |
led | -0.359 |
ed | -0.306 |
ied | -0.269 |
med | -0.267 |
y | -0.265 |
Token | ara |
Token ID | 3301 |
Feature activation | -7.76e+00 |
Pos logit contributions | |
led | +0.501 |
ed | +0.427 |
ied | +0.369 |
ked | +0.360 |
med | +0.356 |
Neg logit contributions | |
pleasure | -0.311 |
Deal | -0.280 |
welcome | -0.267 |
learn | -0.254 |
great | -0.250 |
Token | gigg |
Token ID | 30442 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
led | +0.528 |
ed | +0.435 |
ied | +0.386 |
ked | +0.369 |
ier | +0.362 |
Neg logit contributions | |
pleasure | -0.411 |
Deal | -0.369 |
welcome | -0.361 |
learn | -0.357 |
think | -0.356 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.002 |
we | -0.008 |
stones | -0.011 |
ed | -0.011 |
web | -0.011 |
Neg logit contributions | |
Deal | -0.248 |
pleasure | -0.242 |
great | -0.233 |
welcome | -0.233 |
OK | -0.230 |
Token | . |
Token ID | 13 |
Feature activation | -7.57e+00 |
Pos logit contributions | |
led | +1.794 |
ed | +1.499 |
ied | +1.259 |
ked | +1.229 |
awed | +1.219 |
Neg logit contributions | |
pleasure | -1.442 |
Deal | -1.297 |
welcome | -1.238 |
learn | -1.201 |
great | -1.189 |
Token | She |
Token ID | 1375 |
Feature activation | -5.34e+00 |
Pos logit contributions | |
led | +0.501 |
ed | +0.457 |
ked | +0.372 |
ted | +0.371 |
awed | +0.368 |
Neg logit contributions | |
pleasure | -0.374 |
Deal | -0.339 |
great | -0.324 |
learn | -0.321 |
deal | -0.316 |
Token | picked |
Token ID | 6497 |
Feature activation | -1.14e+01 |
Pos logit contributions | |
led | +0.347 |
ed | +0.287 |
ied | +0.245 |
ked | +0.235 |
awed | +0.232 |
Neg logit contributions | |
pleasure | -0.308 |
Deal | -0.275 |
welcome | -0.269 |
great | -0.263 |
learn | -0.263 |
Token | up |
Token ID | 510 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
led | +0.824 |
ed | +0.698 |
ied | +0.594 |
ked | +0.578 |
awed | +0.575 |
Neg logit contributions | |
pleasure | -0.588 |
Deal | -0.527 |
welcome | -0.500 |
learn | -0.482 |
great | -0.476 |
Token | her |
Token ID | 607 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
led | +0.243 |
ed | +0.209 |
ied | +0.186 |
awed | +0.181 |
ked | +0.181 |
Neg logit contributions | |
pleasure | -0.107 |
Deal | -0.090 |
welcome | -0.087 |
great | -0.083 |
learn | -0.083 |
Token | Mom |
Token ID | 11254 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
led | +0.403 |
ed | +0.376 |
y | +0.334 |
red | +0.334 |
ged | +0.333 |
Neg logit contributions | |
pleasure | -0.183 |
Deal | -0.143 |
Sometimes | -0.089 |
deal | -0.088 |
Not | -0.087 |
Token | smiled |
Token ID | 13541 |
Feature activation | -9.67e+00 |
Pos logit contributions | |
pleasure | +0.136 |
Deal | +0.122 |
welcome | +0.118 |
learn | +0.114 |
think | +0.112 |
Neg logit contributions | |
led | -0.182 |
ed | -0.152 |
ied | -0.131 |
ked | -0.128 |
awed | -0.127 |
Token | , |
Token ID | 11 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
led | +0.736 |
ed | +0.614 |
ied | +0.565 |
awed | +0.542 |
ted | +0.531 |
Neg logit contributions | |
pleasure | -0.452 |
welcome | -0.389 |
Deal | -0.378 |
great | -0.360 |
learn | -0.352 |
Token | t |
Token ID | 256 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
pleasure | +0.111 |
Deal | +0.099 |
welcome | +0.092 |
learn | +0.089 |
great | +0.088 |
Neg logit contributions | |
led | -0.182 |
ed | -0.155 |
ied | -0.133 |
ked | -0.132 |
awed | -0.131 |
Token | uss |
Token ID | 1046 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
led | +0.122 |
lers | +0.083 |
ed | +0.079 |
ils | +0.076 |
ked | +0.073 |
Neg logit contributions | |
pleasure | -0.163 |
Deal | -0.157 |
welcome | -0.147 |
love | -0.146 |
Mom | -0.142 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.224 |
attered | +0.185 |
ed | +0.180 |
ted | +0.168 |
ked | +0.164 |
Neg logit contributions | |
be | -0.214 |
| -0.211 |
think | -0.210 |
have | -0.209 |
pleasure | -0.207 |
Token | his |
Token ID | 465 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
led | +1.975 |
ed | +1.655 |
ied | +1.485 |
awed | +1.468 |
ier | +1.445 |
Neg logit contributions | |
pleasure | -1.254 |
welcome | -1.081 |
great | -1.024 |
Deal | -1.024 |
think | -0.987 |
Token | hair |
Token ID | 4190 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
led | +0.368 |
ed | +0.313 |
ied | +0.269 |
ked | +0.265 |
awed | +0.262 |
Neg logit contributions | |
pleasure | -0.222 |
Deal | -0.199 |
welcome | -0.188 |
learn | -0.184 |
great | -0.177 |
Token | and |
Token ID | 290 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
pleasure | +0.174 |
be | +0.156 |
learn | +0.153 |
great | +0.151 |
rest | +0.149 |
Neg logit contributions | |
led | -0.306 |
ed | -0.255 |
ied | -0.235 |
year | -0.227 |
ted | -0.220 |
Token | replied |
Token ID | 8712 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
pleasure | +0.120 |
welcome | +0.105 |
Deal | +0.103 |
great | +0.102 |
think | +0.101 |
Neg logit contributions | |
led | -0.148 |
ed | -0.121 |
ied | -0.108 |
ked | -0.103 |
awed | -0.103 |
Token | , |
Token ID | 11 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
led | +0.024 |
ed | +0.020 |
ied | +0.018 |
awed | +0.018 |
ked | +0.018 |
Neg logit contributions | |
pleasure | -0.010 |
Deal | -0.008 |
welcome | -0.008 |
great | -0.008 |
learn | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
be | +0.121 |
have | +0.120 |
great | +0.119 |
think | +0.116 |
try | +0.114 |
Neg logit contributions | |
led | -0.188 |
ier | -0.169 |
ed | -0.160 |
ing | -0.157 |
ied | -0.154 |
Token | shiny |
Token ID | 22441 |
Feature activation | -5.93e+00 |
Pos logit contributions | |
led | +0.095 |
ed | +0.081 |
y | +0.075 |
ling | +0.073 |
ied | +0.070 |
Neg logit contributions | |
pleasure | -0.053 |
Deal | -0.052 |
learn | -0.044 |
know | -0.042 |
welcome | -0.041 |
Token | rock |
Token ID | 3881 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
led | +0.443 |
ed | +0.376 |
ied | +0.325 |
awed | +0.318 |
ked | +0.316 |
Neg logit contributions | |
pleasure | -0.289 |
welcome | -0.248 |
Deal | -0.243 |
great | -0.238 |
learn | -0.234 |
Token | that |
Token ID | 326 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
led | +0.294 |
ed | +0.236 |
ied | +0.220 |
awed | +0.219 |
med | +0.213 |
Neg logit contributions | |
pleasure | -0.171 |
have | -0.160 |
be | -0.158 |
welcome | -0.158 |
think | -0.154 |
Token | spark |
Token ID | 9009 |
Feature activation | -6.70e+00 |
Pos logit contributions | |
led | +0.241 |
ed | +0.199 |
ked | +0.171 |
ied | +0.170 |
awed | +0.169 |
Neg logit contributions | |
pleasure | -0.197 |
welcome | -0.174 |
Deal | -0.168 |
great | -0.168 |
learn | -0.165 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.112 |
awed | +0.084 |
ied | +0.064 |
ainted | +0.063 |
year | +0.061 |
Neg logit contributions | |
pleasure | -0.543 |
great | -0.540 |
have | -0.516 |
be | -0.515 |
fun | -0.511 |
Token | in |
Token ID | 287 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
led | +1.750 |
ed | +1.438 |
awed | +1.303 |
ied | +1.258 |
ked | +1.234 |
Neg logit contributions | |
pleasure | -1.479 |
welcome | -1.324 |
great | -1.316 |
think | -1.203 |
fun | -1.199 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
led | +1.021 |
ed | +0.877 |
ied | +0.762 |
med | +0.748 |
y | +0.745 |
Neg logit contributions | |
pleasure | -0.614 |
Deal | -0.572 |
welcome | -0.497 |
learn | -0.491 |
deal | -0.476 |
Token | sun |
Token ID | 4252 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
led | +0.279 |
ed | +0.239 |
ied | +0.209 |
ked | +0.203 |
awed | +0.203 |
Neg logit contributions | |
pleasure | -0.152 |
Deal | -0.132 |
welcome | -0.124 |
learn | -0.119 |
great | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | -7.13e+00 |
Pos logit contributions | |
pleasure | +0.169 |
welcome | +0.147 |
be | +0.140 |
learn | +0.140 |
have | +0.138 |
Neg logit contributions | |
led | -0.220 |
ed | -0.178 |
ied | -0.159 |
awed | -0.157 |
ked | -0.151 |
Token | She |
Token ID | 1375 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
led | +0.553 |
ed | +0.501 |
ted | +0.431 |
ied | +0.426 |
med | +0.426 |
Neg logit contributions | |
pleasure | -0.271 |
Deal | -0.238 |
great | -0.225 |
learn | -0.221 |
welcome | -0.221 |
Token |
|
Token ID | 198 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
pleasure | +0.114 |
Deal | +0.106 |
welcome | +0.090 |
meantime | +0.089 |
great | +0.088 |
Neg logit contributions | |
led | -0.215 |
ed | -0.187 |
ked | -0.160 |
ted | -0.158 |
ied | -0.157 |
Token |
|
Token ID | 198 |
Feature activation | +5.61e+00 |
Pos logit contributions | |
pleasure | +0.209 |
deal | +0.164 |
Deal | +0.160 |
great | +0.153 |
welcome | +0.151 |
Neg logit contributions | |
led | -0.340 |
ed | -0.298 |
ied | -0.264 |
med | -0.263 |
y | -0.261 |
Token | As |
Token ID | 1722 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
pleasure | +0.312 |
Deal | +0.253 |
welcome | +0.245 |
deal | +0.244 |
great | +0.239 |
Neg logit contributions | |
led | -0.359 |
ed | -0.306 |
ied | -0.269 |
med | -0.267 |
y | -0.265 |
Token | he |
Token ID | 339 |
Feature activation | -8.53e+00 |
Pos logit contributions | |
pleasure | +0.200 |
Deal | +0.183 |
welcome | +0.169 |
learn | +0.164 |
think | +0.160 |
Neg logit contributions | |
led | -0.274 |
ed | -0.231 |
ied | -0.198 |
ted | -0.190 |
ked | -0.190 |
Token | padd |
Token ID | 14098 |
Feature activation | -7.07e+00 |
Pos logit contributions | |
led | +0.562 |
ed | +0.462 |
ied | +0.401 |
ked | +0.393 |
ier | +0.384 |
Neg logit contributions | |
pleasure | -0.471 |
welcome | -0.413 |
Deal | -0.412 |
great | -0.412 |
learn | -0.412 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.014 |
we | -0.004 |
ed | -0.028 |
ked | -0.054 |
awed | -0.055 |
Neg logit contributions | |
pleasure | -0.683 |
welcome | -0.676 |
great | -0.660 |
know | -0.659 |
think | -0.659 |
Token | along |
Token ID | 1863 |
Feature activation | -1.08e+01 |
Pos logit contributions | |
led | +1.887 |
ed | +1.589 |
ied | +1.360 |
ked | +1.323 |
awed | +1.318 |
Neg logit contributions | |
pleasure | -1.346 |
Deal | -1.199 |
welcome | -1.147 |
learn | -1.104 |
great | -1.093 |
Token | , |
Token ID | 11 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
led | +0.894 |
ed | +0.767 |
ied | +0.678 |
awed | +0.667 |
ked | +0.662 |
Neg logit contributions | |
pleasure | -0.438 |
Deal | -0.367 |
welcome | -0.362 |
great | -0.342 |
learn | -0.342 |
Token | he |
Token ID | 339 |
Feature activation | -7.34e+00 |
Pos logit contributions | |
pleasure | +0.205 |
Deal | +0.183 |
welcome | +0.176 |
learn | +0.169 |
great | +0.168 |
Neg logit contributions | |
led | -0.270 |
ed | -0.226 |
ied | -0.192 |
ked | -0.187 |
awed | -0.186 |
Token | saw |
Token ID | 2497 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
led | +0.463 |
ed | +0.377 |
ied | +0.325 |
awed | +0.314 |
ked | +0.310 |
Neg logit contributions | |
pleasure | -0.427 |
Deal | -0.378 |
learn | -0.377 |
welcome | -0.374 |
great | -0.373 |
Token | an |
Token ID | 281 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
led | +0.293 |
ed | +0.253 |
ied | +0.231 |
awed | +0.227 |
ked | +0.227 |
Neg logit contributions | |
pleasure | -0.094 |
welcome | -0.070 |
great | -0.070 |
learn | -0.066 |
Deal | -0.065 |
Token | bed |
Token ID | 3996 |
Feature activation | +3.76e+00 |
Pos logit contributions | |
pleasure | +0.048 |
Deal | +0.043 |
welcome | +0.040 |
great | +0.038 |
learn | +0.037 |
Neg logit contributions | |
led | -0.065 |
ed | -0.055 |
ied | -0.047 |
ked | -0.046 |
ling | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
pleasure | +0.172 |
learn | +0.155 |
welcome | +0.154 |
great | +0.149 |
think | +0.149 |
Neg logit contributions | |
led | -0.257 |
ed | -0.214 |
ked | -0.200 |
awed | -0.193 |
ied | -0.191 |
Token | he |
Token ID | 339 |
Feature activation | -8.94e+00 |
Pos logit contributions | |
pleasure | +0.084 |
Deal | +0.074 |
welcome | +0.071 |
learn | +0.068 |
great | +0.067 |
Neg logit contributions | |
led | -0.129 |
ed | -0.109 |
ied | -0.094 |
ked | -0.092 |
awed | -0.091 |
Token | sn |
Token ID | 3013 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
led | +0.597 |
ed | +0.496 |
ied | +0.424 |
ked | +0.414 |
awed | +0.410 |
Neg logit contributions | |
pleasure | -0.497 |
Deal | -0.447 |
welcome | -0.435 |
learn | -0.426 |
great | -0.421 |
Token | ugg |
Token ID | 6837 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
led | +0.464 |
ed | +0.365 |
ked | +0.305 |
ied | +0.301 |
med | +0.291 |
Neg logit contributions | |
pleasure | -0.492 |
Deal | -0.457 |
welcome | -0.448 |
learn | -0.428 |
great | -0.417 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.024 |
ked | -0.060 |
ed | -0.066 |
ied | -0.118 |
awed | -0.123 |
Neg logit contributions | |
pleasure | -1.490 |
welcome | -1.446 |
great | -1.435 |
Deal | -1.434 |
learn | -1.389 |
Token | under |
Token ID | 739 |
Feature activation | -1.76e+01 |
Pos logit contributions | |
led | +1.602 |
ed | +1.311 |
ied | +1.098 |
ked | +1.077 |
awed | +1.077 |
Neg logit contributions | |
pleasure | -1.611 |
Deal | -1.438 |
welcome | -1.419 |
great | -1.382 |
learn | -1.367 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
led | +1.401 |
ed | +1.208 |
ted | +1.041 |
ied | +1.038 |
med | +1.027 |
Neg logit contributions | |
pleasure | -0.740 |
Deal | -0.664 |
welcome | -0.605 |
learn | -0.579 |
deal | -0.560 |
Token | covers |
Token ID | 8698 |
Feature activation | -8.55e+00 |
Pos logit contributions | |
led | +0.316 |
ed | +0.276 |
ied | +0.246 |
ked | +0.242 |
awed | +0.241 |
Neg logit contributions | |
pleasure | -0.116 |
Deal | -0.095 |
welcome | -0.091 |
learn | -0.085 |
great | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
led | +0.591 |
ed | +0.488 |
awed | +0.456 |
ked | +0.454 |
attered | +0.418 |
Neg logit contributions | |
pleasure | -0.404 |
welcome | -0.372 |
great | -0.342 |
learn | -0.340 |
rest | -0.337 |
Token | fell |
Token ID | 3214 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
pleasure | +0.127 |
Deal | +0.113 |
welcome | +0.110 |
learn | +0.108 |
great | +0.107 |
Neg logit contributions | |
led | -0.157 |
ed | -0.131 |
ied | -0.114 |
awed | -0.110 |
ked | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
led | +0.089 |
ed | +0.073 |
ied | +0.069 |
ked | +0.069 |
awed | +0.067 |
Neg logit contributions | |
pleasure | -0.033 |
welcome | -0.028 |
fun | -0.028 |
great | -0.028 |
learn | -0.026 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
led | +0.833 |
ed | +0.751 |
ted | +0.642 |
y | +0.639 |
ied | +0.637 |
Neg logit contributions | |
pleasure | -0.406 |
Deal | -0.342 |
great | -0.324 |
deal | -0.315 |
welcome | -0.314 |
Token | my |
Token ID | 1820 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
led | +0.120 |
ed | +0.101 |
ied | +0.088 |
ked | +0.086 |
awed | +0.085 |
Neg logit contributions | |
pleasure | -0.084 |
Deal | -0.074 |
welcome | -0.073 |
learn | -0.071 |
think | -0.070 |
Token | sn |
Token ID | 3013 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
pleasure | +0.098 |
Deal | +0.087 |
learn | +0.086 |
welcome | +0.086 |
think | +0.085 |
Neg logit contributions | |
led | -0.123 |
ed | -0.103 |
ied | -0.089 |
ked | -0.087 |
awed | -0.085 |
Token | ugg |
Token ID | 6837 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
led | +0.438 |
ed | +0.338 |
ied | +0.277 |
ked | +0.273 |
med | +0.266 |
Neg logit contributions | |
pleasure | -0.528 |
Deal | -0.482 |
welcome | -0.476 |
learn | -0.453 |
great | -0.444 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.034 |
ed | -0.061 |
ked | -0.069 |
ied | -0.114 |
year | -0.122 |
Neg logit contributions | |
pleasure | -1.474 |
welcome | -1.451 |
great | -1.436 |
Deal | -1.434 |
learn | -1.403 |
Token | up |
Token ID | 510 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
led | +1.744 |
ed | +1.447 |
ied | +1.225 |
ked | +1.187 |
awed | +1.186 |
Neg logit contributions | |
pleasure | -1.489 |
Deal | -1.336 |
welcome | -1.292 |
learn | -1.248 |
great | -1.245 |
Token | to |
Token ID | 284 |
Feature activation | -5.96e+00 |
Pos logit contributions | |
led | +0.218 |
ed | +0.181 |
ied | +0.159 |
ked | +0.155 |
awed | +0.153 |
Neg logit contributions | |
pleasure | -0.174 |
welcome | -0.157 |
Deal | -0.152 |
think | -0.152 |
great | -0.151 |
Token | his |
Token ID | 465 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
led | +0.397 |
ed | +0.332 |
ted | +0.303 |
med | +0.285 |
ting | +0.283 |
Neg logit contributions | |
pleasure | -0.285 |
Deal | -0.273 |
compromise | -0.225 |
meantime | -0.224 |
deal | -0.219 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.07e+00 |
Pos logit contributions | |
led | +0.374 |
ed | +0.321 |
ied | +0.281 |
ling | +0.276 |
ked | +0.276 |
Neg logit contributions | |
pleasure | -0.172 |
Deal | -0.154 |
welcome | -0.142 |
learn | -0.134 |
think | -0.131 |
Token | and |
Token ID | 290 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
pleasure | +0.229 |
welcome | +0.202 |
Deal | +0.197 |
learn | +0.195 |
think | +0.190 |
Neg logit contributions | |
led | -0.391 |
ed | -0.328 |
ied | -0.292 |
awed | -0.290 |
ked | -0.285 |
Token | hedge |
Token ID | 19859 |
Feature activation | -7.20e+00 |
Pos logit contributions | |
pleasure | +0.556 |
Deal | +0.499 |
welcome | +0.485 |
learn | +0.469 |
great | +0.467 |
Neg logit contributions | |
led | -0.646 |
ed | -0.533 |
ied | -0.448 |
ked | -0.437 |
awed | -0.435 |
Token | . |
Token ID | 13 |
Feature activation | -9.52e+00 |
Pos logit contributions | |
led | +0.482 |
ed | +0.386 |
ked | +0.355 |
med | +0.353 |
ied | +0.353 |
Neg logit contributions | |
pleasure | -0.339 |
welcome | -0.312 |
learn | -0.308 |
think | -0.308 |
have | -0.302 |
Token | He |
Token ID | 679 |
Feature activation | -7.17e+00 |
Pos logit contributions | |
led | +0.745 |
ed | +0.675 |
ted | +0.578 |
y | +0.575 |
med | +0.574 |
Neg logit contributions | |
pleasure | -0.368 |
Deal | -0.314 |
great | -0.295 |
deal | -0.290 |
welcome | -0.286 |
Token | w |
Token ID | 266 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
led | +0.469 |
ed | +0.387 |
ied | +0.332 |
ked | +0.324 |
awed | +0.321 |
Neg logit contributions | |
pleasure | -0.402 |
Deal | -0.358 |
welcome | -0.353 |
learn | -0.352 |
think | -0.351 |
Token | igg |
Token ID | 6950 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
led | +0.173 |
ed | +0.143 |
ied | +0.116 |
ked | +0.115 |
med | +0.113 |
Neg logit contributions | |
pleasure | -0.179 |
Deal | -0.163 |
welcome | -0.158 |
great | -0.152 |
learn | -0.151 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.066 |
ed | +0.020 |
we | +0.016 |
med | +0.012 |
awed | +0.006 |
Neg logit contributions | |
think | -0.894 |
pleasure | -0.893 |
great | -0.888 |
know | -0.880 |
learn | -0.873 |
Token | and |
Token ID | 290 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
led | +1.965 |
ed | +1.662 |
ied | +1.454 |
awed | +1.428 |
ked | +1.414 |
Neg logit contributions | |
pleasure | -1.262 |
Deal | -1.082 |
welcome | -1.072 |
great | -1.031 |
learn | -1.016 |
Token | squ |
Token ID | 2809 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
pleasure | +0.102 |
think | +0.091 |
welcome | +0.090 |
great | +0.089 |
learn | +0.089 |
Neg logit contributions | |
led | -0.123 |
ed | -0.102 |
ied | -0.092 |
ier | -0.089 |
awed | -0.086 |
Token | irmed |
Token ID | 15491 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
led | +0.092 |
ed | +0.069 |
ied | +0.062 |
we | +0.060 |
web | +0.059 |
Neg logit contributions | |
pleasure | -0.147 |
Deal | -0.135 |
welcome | -0.133 |
great | -0.129 |
think | -0.129 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
led | +0.780 |
ed | +0.637 |
awed | +0.616 |
ied | +0.594 |
med | +0.584 |
Neg logit contributions | |
pleasure | -0.412 |
welcome | -0.369 |
great | -0.364 |
think | -0.334 |
Deal | -0.327 |
Token | soon |
Token ID | 2582 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
pleasure | +0.127 |
Deal | +0.113 |
welcome | +0.111 |
great | +0.108 |
learn | +0.107 |
Neg logit contributions | |
led | -0.158 |
ed | -0.132 |
ied | -0.115 |
awed | -0.109 |
ked | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
led | +0.094 |
ed | +0.075 |
awed | +0.067 |
ier | +0.067 |
ked | +0.067 |
Neg logit contributions | |
pleasure | -0.079 |
great | -0.076 |
welcome | -0.075 |
learn | -0.074 |
think | -0.072 |
Token | spikes |
Token ID | 27198 |
Feature activation | -7.37e+00 |
Pos logit contributions | |
led | +0.079 |
ed | +0.067 |
ied | +0.058 |
awed | +0.056 |
ked | +0.056 |
Neg logit contributions | |
pleasure | -0.054 |
Deal | -0.047 |
welcome | -0.046 |
learn | -0.044 |
great | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -8.86e+00 |
Pos logit contributions | |
led | +0.452 |
ed | +0.367 |
awed | +0.349 |
year | +0.345 |
ied | +0.342 |
Neg logit contributions | |
pleasure | -0.337 |
have | -0.334 |
know | -0.331 |
think | -0.329 |
welcome | -0.320 |
Token | They |
Token ID | 1119 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
led | +0.654 |
ed | +0.600 |
ted | +0.508 |
ked | +0.500 |
med | +0.499 |
Neg logit contributions | |
pleasure | -0.363 |
deal | -0.295 |
great | -0.295 |
Deal | -0.294 |
plan | -0.289 |
Token | grow |
Token ID | 1663 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
led | +0.112 |
ed | +0.097 |
ied | +0.086 |
awed | +0.084 |
ked | +0.083 |
Neg logit contributions | |
pleasure | -0.051 |
Deal | -0.043 |
welcome | -0.041 |
learn | -0.039 |
great | -0.039 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
great | +0.012 |
pleasure | +0.011 |
welcome | +0.011 |
learn | +0.011 |
good | +0.011 |
Neg logit contributions | |
led | -0.005 |
ed | -0.004 |
awed | -0.003 |
ked | -0.003 |
ied | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
led | +1.863 |
ed | +1.558 |
ied | +1.357 |
awed | +1.332 |
ked | +1.309 |
Neg logit contributions | |
pleasure | -1.374 |
welcome | -1.186 |
Deal | -1.179 |
great | -1.145 |
learn | -1.122 |
Token | roared |
Token ID | 45615 |
Feature activation | -1.23e+01 |
Pos logit contributions | |
pleasure | +0.093 |
welcome | +0.082 |
think | +0.081 |
Deal | +0.081 |
learn | +0.081 |
Neg logit contributions | |
led | -0.134 |
ed | -0.115 |
ied | -0.103 |
awed | -0.098 |
ier | -0.098 |
Token | and |
Token ID | 290 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
led | +0.879 |
ed | +0.707 |
ied | +0.674 |
awed | +0.659 |
ted | +0.627 |
Neg logit contributions | |
pleasure | -0.582 |
welcome | -0.528 |
great | -0.527 |
OK | -0.511 |
Deal | -0.505 |
Token | his |
Token ID | 465 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
pleasure | +0.115 |
welcome | +0.104 |
think | +0.104 |
Deal | +0.103 |
learn | +0.101 |
Neg logit contributions | |
led | -0.148 |
ed | -0.124 |
ied | -0.113 |
awed | -0.105 |
ier | -0.105 |
Token | sed |
Token ID | 36622 |
Feature activation | -6.01e+00 |
Pos logit contributions | |
led | +0.323 |
ed | +0.273 |
ied | +0.235 |
ked | +0.229 |
awed | +0.228 |
Neg logit contributions | |
pleasure | -0.218 |
Deal | -0.194 |
welcome | -0.185 |
learn | -0.178 |
great | -0.176 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -5.18e-02 |
Pos logit contributions | |
pleasure | +0.134 |
welcome | +0.123 |
think | +0.123 |
learn | +0.122 |
Deal | +0.120 |
Neg logit contributions | |
led | -0.188 |
ed | -0.163 |
ier | -0.141 |
med | -0.139 |
awed | -0.138 |
Token | , |
Token ID | 11 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
led | +0.004 |
ed | +0.004 |
ied | +0.003 |
awed | +0.003 |
ked | +0.003 |
Neg logit contributions | |
pleasure | -0.002 |
welcome | -0.002 |
Deal | -0.002 |
great | -0.002 |
learn | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | -6.12e+00 |
Pos logit contributions | |
pleasure | +0.078 |
Deal | +0.068 |
welcome | +0.065 |
learn | +0.062 |
great | +0.062 |
Neg logit contributions | |
led | -0.135 |
ed | -0.115 |
ied | -0.100 |
awed | -0.098 |
ked | -0.098 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
led | +0.542 |
ed | +0.473 |
ied | +0.420 |
ked | +0.411 |
awed | +0.410 |
Neg logit contributions | |
pleasure | -0.210 |
Deal | -0.176 |
welcome | -0.164 |
learn | -0.154 |
great | -0.151 |
Token | tick |
Token ID | 4378 |
Feature activation | -9.91e+00 |
Pos logit contributions | |
pleasure | +0.179 |
Deal | +0.166 |
welcome | +0.165 |
learn | +0.164 |
know | +0.162 |
Neg logit contributions | |
led | -0.263 |
ed | -0.212 |
ied | -0.188 |
ked | -0.182 |
awed | -0.180 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.106 |
ed | +0.039 |
ied | +0.001 |
ier | -0.013 |
awed | -0.016 |
Neg logit contributions | |
pleasure | -0.899 |
great | -0.890 |
Deal | -0.888 |
practice | -0.866 |
welcome | -0.863 |
Token | him |
Token ID | 683 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
led | +2.029 |
ed | +1.725 |
ied | +1.511 |
awed | +1.487 |
ked | +1.465 |
Neg logit contributions | |
pleasure | -1.204 |
Deal | -1.051 |
welcome | -1.025 |
great | -0.984 |
learn | -0.964 |
Token | and |
Token ID | 290 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
pleasure | +0.003 |
welcome | +0.003 |
think | +0.003 |
Deal | +0.003 |
great | +0.003 |
Neg logit contributions | |
led | -0.004 |
ed | -0.003 |
awed | -0.003 |
ied | -0.003 |
ted | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | -7.34e+00 |
Pos logit contributions | |
pleasure | +0.104 |
Deal | +0.093 |
think | +0.092 |
welcome | +0.092 |
great | +0.090 |
Neg logit contributions | |
led | -0.142 |
ed | -0.116 |
ied | -0.106 |
awed | -0.102 |
ier | -0.099 |
Token | laughed |
Token ID | 13818 |
Feature activation | -8.04e+00 |
Pos logit contributions | |
led | +0.462 |
ed | +0.372 |
ied | +0.333 |
awed | +0.318 |
ier | +0.313 |
Neg logit contributions | |
pleasure | -0.412 |
think | -0.371 |
Deal | -0.367 |
great | -0.365 |
learn | -0.364 |
Token | and |
Token ID | 290 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
led | +0.586 |
ed | +0.473 |
ied | +0.449 |
awed | +0.444 |
lers | +0.415 |
Neg logit contributions | |
pleasure | -0.378 |
great | -0.340 |
welcome | -0.327 |
OK | -0.323 |
Deal | -0.316 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e+01 |
Pos logit contributions | |
led | +0.214 |
ed | +0.182 |
ling | +0.164 |
y | +0.159 |
ing | +0.158 |
Neg logit contributions | |
Deal | -0.129 |
pleasure | -0.129 |
learn | -0.104 |
OK | -0.103 |
know | -0.101 |
Token | But |
Token ID | 887 |
Feature activation | +7.78e+00 |
Pos logit contributions | |
led | +0.833 |
ed | +0.751 |
ted | +0.642 |
y | +0.639 |
ied | +0.637 |
Neg logit contributions | |
pleasure | -0.406 |
Deal | -0.342 |
great | -0.324 |
deal | -0.315 |
welcome | -0.314 |
Token | it |
Token ID | 340 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
pleasure | +0.360 |
Deal | +0.322 |
welcome | +0.296 |
learn | +0.284 |
great | +0.282 |
Neg logit contributions | |
led | -0.590 |
ed | -0.505 |
ied | -0.435 |
ked | -0.425 |
ted | -0.422 |
Token | tw |
Token ID | 665 |
Feature activation | -7.54e+00 |
Pos logit contributions | |
pleasure | +0.119 |
think | +0.105 |
great | +0.104 |
learn | +0.102 |
welcome | +0.102 |
Neg logit contributions | |
led | -0.190 |
ed | -0.162 |
ied | -0.144 |
ked | -0.143 |
awed | -0.139 |
Token | ink |
Token ID | 676 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
led | +0.510 |
ed | +0.418 |
ied | +0.353 |
awed | +0.347 |
ked | +0.338 |
Neg logit contributions | |
pleasure | -0.430 |
Deal | -0.389 |
welcome | -0.379 |
learn | -0.366 |
great | -0.357 |
Token | led |
Token ID | 992 |
Feature activation | -2.62e+01 |
Pos logit contributions | |
led | +0.407 |
ed | +0.297 |
ked | +0.238 |
ied | +0.229 |
awed | +0.217 |
Neg logit contributions | |
pleasure | -1.015 |
great | -0.969 |
welcome | -0.961 |
think | -0.947 |
learn | -0.943 |
Token | in |
Token ID | 287 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
led | +1.871 |
ed | +1.576 |
ied | +1.357 |
awed | +1.329 |
ked | +1.320 |
Neg logit contributions | |
pleasure | -1.369 |
Deal | -1.194 |
welcome | -1.177 |
great | -1.139 |
learn | -1.109 |
Token | the |
Token ID | 262 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
led | +0.926 |
ed | +0.789 |
y | +0.699 |
med | +0.689 |
ied | +0.684 |
Neg logit contributions | |
pleasure | -0.637 |
Deal | -0.598 |
learn | -0.533 |
deal | -0.513 |
welcome | -0.508 |
Token | sunlight |
Token ID | 19606 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
led | +0.270 |
ed | +0.230 |
ied | +0.199 |
awed | +0.193 |
ked | +0.193 |
Neg logit contributions | |
pleasure | -0.159 |
Deal | -0.141 |
welcome | -0.130 |
learn | -0.126 |
great | -0.124 |
Token | and |
Token ID | 290 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
led | +0.096 |
ed | +0.079 |
ied | +0.070 |
ked | +0.070 |
awed | +0.069 |
Neg logit contributions | |
pleasure | -0.067 |
welcome | -0.059 |
learn | -0.055 |
great | -0.055 |
think | -0.054 |
Token | danced |
Token ID | 39480 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
pleasure | +0.119 |
welcome | +0.102 |
Deal | +0.100 |
think | +0.100 |
learn | +0.099 |
Neg logit contributions | |
led | -0.163 |
ed | -0.136 |
ied | -0.121 |
awed | -0.117 |
ked | -0.116 |
Token | many |
Token ID | 867 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
led | +0.517 |
ed | +0.442 |
ied | +0.387 |
ted | +0.373 |
ked | +0.372 |
Neg logit contributions | |
pleasure | -0.279 |
Deal | -0.251 |
welcome | -0.226 |
learn | -0.223 |
great | -0.214 |
Token | colors |
Token ID | 7577 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
led | +0.289 |
ed | +0.250 |
ied | +0.217 |
ked | +0.216 |
awed | +0.213 |
Neg logit contributions | |
pleasure | -0.162 |
Deal | -0.137 |
welcome | -0.137 |
great | -0.132 |
learn | -0.130 |
Token | . |
Token ID | 13 |
Feature activation | -8.45e+00 |
Pos logit contributions | |
led | +0.101 |
ed | +0.084 |
ked | +0.077 |
awed | +0.077 |
ted | +0.073 |
Neg logit contributions | |
pleasure | -0.069 |
welcome | -0.062 |
have | -0.060 |
great | -0.059 |
learn | -0.059 |
Token | It |
Token ID | 632 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
led | +0.602 |
ed | +0.556 |
y | +0.476 |
ted | +0.471 |
med | +0.471 |
Neg logit contributions | |
pleasure | -0.344 |
deal | -0.288 |
great | -0.283 |
Deal | -0.277 |
plan | -0.277 |
Token | spark |
Token ID | 9009 |
Feature activation | -6.44e+00 |
Pos logit contributions | |
pleasure | +0.062 |
welcome | +0.053 |
Deal | +0.053 |
learn | +0.053 |
think | +0.053 |
Neg logit contributions | |
led | -0.122 |
ed | -0.105 |
ied | -0.094 |
ked | -0.092 |
awed | -0.092 |
Token | led |
Token ID | 992 |
Feature activation | -2.60e+01 |
Pos logit contributions | |
led | +0.118 |
ed | +0.063 |
ied | +0.049 |
awed | +0.038 |
ked | +0.036 |
Neg logit contributions | |
pleasure | -0.570 |
great | -0.563 |
think | -0.537 |
welcome | -0.531 |
learn | -0.530 |
Token | in |
Token ID | 287 |
Feature activation | -1.35e+01 |
Pos logit contributions | |
led | +1.745 |
ed | +1.453 |
ied | +1.250 |
awed | +1.246 |
ked | +1.218 |
Neg logit contributions | |
pleasure | -1.454 |
welcome | -1.288 |
great | -1.260 |
Deal | -1.218 |
think | -1.193 |
Token | the |
Token ID | 262 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
led | +0.919 |
ed | +0.769 |
med | +0.685 |
ied | +0.683 |
y | +0.679 |
Neg logit contributions | |
pleasure | -0.650 |
Deal | -0.633 |
learn | -0.534 |
deal | -0.532 |
welcome | -0.530 |
Token | sun |
Token ID | 4252 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
led | +0.262 |
ed | +0.224 |
ied | +0.194 |
ked | +0.189 |
awed | +0.188 |
Neg logit contributions | |
pleasure | -0.149 |
Deal | -0.133 |
welcome | -0.122 |
learn | -0.118 |
think | -0.116 |
Token | , |
Token ID | 11 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
pleasure | +0.170 |
welcome | +0.147 |
Deal | +0.143 |
learn | +0.140 |
great | +0.137 |
Neg logit contributions | |
led | -0.242 |
ed | -0.201 |
ied | -0.177 |
awed | -0.173 |
ked | -0.171 |
Token | and |
Token ID | 290 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
pleasure | +0.211 |
Deal | +0.187 |
welcome | +0.180 |
learn | +0.173 |
great | +0.171 |
Neg logit contributions | |
led | -0.308 |
ed | -0.260 |
ied | -0.224 |
ked | -0.218 |
awed | -0.217 |
Token | It |
Token ID | 632 |
Feature activation | +8.21e-02 |
Pos logit contributions | |
led | +0.833 |
ed | +0.751 |
ted | +0.642 |
y | +0.639 |
ied | +0.637 |
Neg logit contributions | |
pleasure | -0.406 |
Deal | -0.342 |
great | -0.324 |
deal | -0.315 |
welcome | -0.314 |
Token | bark |
Token ID | 21405 |
Feature activation | -6.02e+00 |
Pos logit contributions | |
pleasure | +0.004 |
think | +0.003 |
learn | +0.003 |
great | +0.003 |
have | +0.003 |
Neg logit contributions | |
led | -0.006 |
ed | -0.005 |
ied | -0.005 |
ked | -0.004 |
awed | -0.004 |
Token | ed |
Token ID | 276 |
Feature activation | -2.30e+01 |
Pos logit contributions | |
led | +0.149 |
ied | +0.069 |
ed | +0.052 |
awed | +0.046 |
we | +0.038 |
Neg logit contributions | |
pleasure | -0.525 |
great | -0.510 |
welcome | -0.505 |
OK | -0.494 |
Deal | -0.491 |
Token | and |
Token ID | 290 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
led | +1.652 |
ed | +1.381 |
ied | +1.204 |
awed | +1.173 |
ked | +1.156 |
Neg logit contributions | |
pleasure | -1.190 |
Deal | -1.044 |
welcome | -1.025 |
great | -0.988 |
learn | -0.975 |
Token | grow |
Token ID | 1663 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
pleasure | +0.074 |
Deal | +0.066 |
welcome | +0.064 |
learn | +0.062 |
great | +0.062 |
Neg logit contributions | |
led | -0.095 |
ed | -0.079 |
ied | -0.069 |
awed | -0.066 |
ked | -0.065 |
Token | led |
Token ID | 992 |
Feature activation | -2.60e+01 |
Pos logit contributions | |
great | +0.040 |
welcome | +0.039 |
good | +0.038 |
fun | +0.038 |
pleasure | +0.038 |
Neg logit contributions | |
led | -0.000 |
ked | +0.001 |
attered | +0.001 |
ied | +0.002 |
awed | +0.002 |
Token | at |
Token ID | 379 |
Feature activation | -9.20e+00 |
Pos logit contributions | |
led | +1.756 |
ed | +1.443 |
ied | +1.271 |
awed | +1.246 |
ked | +1.209 |
Neg logit contributions | |
pleasure | -1.462 |
welcome | -1.287 |
Deal | -1.263 |
great | -1.242 |
think | -1.208 |
Token | them |
Token ID | 606 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
led | +0.720 |
ed | +0.621 |
ied | +0.543 |
ted | +0.525 |
ked | +0.522 |
Neg logit contributions | |
pleasure | -0.402 |
Deal | -0.368 |
welcome | -0.325 |
learn | -0.321 |
great | -0.311 |
Token | , |
Token ID | 11 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
pleasure | +0.187 |
welcome | +0.168 |
learn | +0.168 |
think | +0.164 |
Deal | +0.162 |
Neg logit contributions | |
led | -0.229 |
ed | -0.186 |
ied | -0.159 |
awed | -0.158 |
ked | -0.154 |
Token | and |
Token ID | 290 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
pleasure | +0.199 |
Deal | +0.176 |
welcome | +0.167 |
learn | +0.160 |
great | +0.159 |
Neg logit contributions | |
led | -0.308 |
ed | -0.262 |
ied | -0.225 |
ked | -0.221 |
awed | -0.220 |
Token | they |
Token ID | 484 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
pleasure | +0.117 |
welcome | +0.101 |
think | +0.100 |
Deal | +0.100 |
learn | +0.099 |
Neg logit contributions | |
led | -0.173 |
ed | -0.146 |
ied | -0.131 |
awed | -0.125 |
ier | -0.125 |
Token | looked |
Token ID | 3114 |
Feature activation | -5.33e+00 |
Pos logit contributions | |
led | +0.068 |
ed | +0.057 |
ied | +0.050 |
ked | +0.048 |
awed | +0.048 |
Neg logit contributions | |
pleasure | -0.046 |
Deal | -0.041 |
welcome | -0.040 |
learn | -0.038 |
great | -0.038 |
Token | at |
Token ID | 379 |
Feature activation | -1.07e+01 |
Pos logit contributions | |
led | +0.439 |
ed | +0.373 |
ied | +0.327 |
ling | +0.320 |
ked | +0.316 |
Neg logit contributions | |
pleasure | -0.216 |
Deal | -0.196 |
welcome | -0.173 |
learn | -0.172 |
think | -0.161 |
Token | them |
Token ID | 606 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
led | +0.811 |
ed | +0.701 |
ied | +0.612 |
ted | +0.599 |
awed | +0.587 |
Neg logit contributions | |
pleasure | -0.486 |
Deal | -0.446 |
welcome | -0.391 |
learn | -0.376 |
deal | -0.371 |
Token | and |
Token ID | 290 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
pleasure | +0.127 |
Deal | +0.113 |
welcome | +0.111 |
learn | +0.108 |
great | +0.107 |
Neg logit contributions | |
led | -0.145 |
ed | -0.120 |
ied | -0.101 |
awed | -0.099 |
ked | -0.098 |
Token | grow |
Token ID | 1663 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
pleasure | +0.074 |
welcome | +0.065 |
think | +0.065 |
Deal | +0.065 |
learn | +0.064 |
Neg logit contributions | |
led | -0.092 |
ed | -0.076 |
ied | -0.068 |
awed | -0.064 |
ier | -0.063 |
Token | led |
Token ID | 992 |
Feature activation | -2.60e+01 |
Pos logit contributions | |
great | +0.038 |
welcome | +0.036 |
good | +0.035 |
pleasure | +0.035 |
okay | +0.035 |
Neg logit contributions | |
led | -0.004 |
ked | -0.002 |
awed | -0.002 |
ed | -0.002 |
ted | -0.001 |
Token | again |
Token ID | 757 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
led | +1.742 |
ed | +1.430 |
ied | +1.253 |
awed | +1.236 |
ked | +1.199 |
Neg logit contributions | |
pleasure | -1.473 |
welcome | -1.309 |
great | -1.265 |
Deal | -1.250 |
think | -1.230 |
Token | . |
Token ID | 13 |
Feature activation | -7.26e+00 |
Pos logit contributions | |
led | +0.136 |
ed | +0.111 |
ied | +0.094 |
ked | +0.091 |
awed | +0.090 |
Neg logit contributions | |
pleasure | -0.130 |
welcome | -0.116 |
Deal | -0.116 |
learn | -0.112 |
great | -0.112 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
led | +0.540 |
ed | +0.499 |
ted | +0.419 |
ked | +0.414 |
awed | +0.413 |
Neg logit contributions | |
pleasure | -0.292 |
Deal | -0.245 |
great | -0.242 |
deal | -0.240 |
plan | -0.239 |
Token | and |
Token ID | 290 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
pleasure | +0.005 |
think | +0.005 |
welcome | +0.005 |
learn | +0.005 |
Deal | +0.004 |
Neg logit contributions | |
led | -0.009 |
ed | -0.007 |
ied | -0.007 |
ier | -0.006 |
ked | -0.006 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
pleasure | +0.081 |
Deal | +0.067 |
welcome | +0.067 |
learn | +0.062 |
think | +0.061 |
Neg logit contributions | |
led | -0.217 |
ed | -0.187 |
ied | -0.169 |
awed | -0.165 |
ked | -0.163 |
Token | more |
Token ID | 517 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
led | +0.744 |
ed | +0.645 |
ied | +0.562 |
ked | +0.551 |
awed | +0.542 |
Neg logit contributions | |
pleasure | -0.385 |
Deal | -0.339 |
welcome | -0.315 |
learn | -0.297 |
great | -0.293 |
Token | clothes |
Token ID | 8242 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
pleasure | +0.159 |
welcome | +0.142 |
learn | +0.141 |
great | +0.140 |
think | +0.140 |
Neg logit contributions | |
led | -0.259 |
ed | -0.222 |
ked | -0.196 |
awed | -0.196 |
ied | -0.188 |
Token | and |
Token ID | 290 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
pleasure | +0.145 |
learn | +0.133 |
welcome | +0.127 |
see | +0.127 |
think | +0.126 |
Neg logit contributions | |
led | -0.325 |
ed | -0.266 |
awed | -0.254 |
ied | -0.240 |
ted | -0.236 |
Token | toys |
Token ID | 14958 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
pleasure | +0.166 |
learn | +0.166 |
think | +0.159 |
see | +0.156 |
have | +0.153 |
Neg logit contributions | |
led | -0.235 |
ed | -0.198 |
awed | -0.182 |
ied | -0.180 |
y | -0.174 |
Token | , |
Token ID | 11 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
pleasure | +0.109 |
have | +0.102 |
learn | +0.102 |
think | +0.101 |
see | +0.100 |
Neg logit contributions | |
led | -0.204 |
ed | -0.161 |
awed | -0.156 |
attered | -0.148 |
ling | -0.148 |
Token | but |
Token ID | 475 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
pleasure | +0.223 |
Deal | +0.192 |
welcome | +0.186 |
learn | +0.182 |
think | +0.177 |
Neg logit contributions | |
led | -0.429 |
ed | -0.370 |
ied | -0.324 |
awed | -0.319 |
ked | -0.314 |
Token | they |
Token ID | 484 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
pleasure | +0.503 |
Deal | +0.440 |
welcome | +0.421 |
learn | +0.404 |
great | +0.401 |
Neg logit contributions | |
led | -0.830 |
ed | -0.706 |
ied | -0.612 |
awed | -0.598 |
ked | -0.597 |
Token | did |
Token ID | 750 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
led | +0.152 |
ed | +0.129 |
med | +0.114 |
awed | +0.112 |
ied | +0.111 |
Neg logit contributions | |
pleasure | -0.091 |
think | -0.089 |
learn | -0.086 |
know | -0.086 |
have | -0.085 |
Token | not |
Token ID | 407 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
led | +0.162 |
ed | +0.143 |
ied | +0.128 |
ked | +0.126 |
awed | +0.125 |
Neg logit contributions | |
pleasure | -0.042 |
Deal | -0.033 |
welcome | -0.029 |
learn | -0.024 |
great | -0.024 |
Token | find |
Token ID | 1064 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
pleasure | +0.241 |
Deal | +0.186 |
welcome | +0.144 |
great | +0.131 |
learn | +0.119 |
Neg logit contributions | |
led | -0.990 |
ed | -0.880 |
ied | -0.786 |
ked | -0.779 |
y | -0.776 |
Token | the |
Token ID | 262 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
pleasure | +0.449 |
welcome | +0.365 |
Deal | +0.358 |
great | +0.349 |
learn | +0.335 |
Neg logit contributions | |
led | -0.959 |
ed | -0.822 |
ied | -0.728 |
awed | -0.723 |
ked | -0.710 |
Token | learned |
Token ID | 4499 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
led | +0.006 |
ed | +0.005 |
ied | +0.005 |
ked | +0.005 |
awed | +0.005 |
Neg logit contributions | |
pleasure | -0.004 |
learn | -0.003 |
welcome | -0.003 |
great | -0.003 |
think | -0.003 |
Token | that |
Token ID | 326 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
pleasure | +0.003 |
Deal | +0.002 |
welcome | +0.002 |
learn | +0.002 |
great | +0.002 |
Neg logit contributions | |
led | -0.004 |
ed | -0.003 |
ied | -0.003 |
ked | -0.003 |
awed | -0.003 |
Token | sometimes |
Token ID | 3360 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
led | +0.072 |
ed | +0.061 |
ied | +0.054 |
ted | +0.051 |
ing | +0.051 |
Neg logit contributions | |
pleasure | -0.050 |
Deal | -0.048 |
welcome | -0.042 |
great | -0.039 |
deal | -0.039 |
Token | things |
Token ID | 1243 |
Feature activation | +7.98e+00 |
Pos logit contributions | |
Deal | +0.061 |
pleasure | +0.060 |
welcome | +0.048 |
meantime | +0.046 |
learn | +0.045 |
Neg logit contributions | |
led | -0.090 |
ed | -0.074 |
ling | -0.064 |
ing | -0.064 |
lers | -0.063 |
Token | can |
Token ID | 460 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
pleasure | +0.301 |
Deal | +0.267 |
welcome | +0.234 |
learn | +0.218 |
great | +0.214 |
Neg logit contributions | |
led | -0.683 |
ed | -0.588 |
ied | -0.523 |
ked | -0.505 |
ted | -0.501 |
Token | look |
Token ID | 804 |
Feature activation | +1.53e+01 |
Pos logit contributions | |
pleasure | +0.180 |
Deal | +0.155 |
welcome | +0.127 |
great | +0.110 |
meantime | +0.110 |
Neg logit contributions | |
led | -0.381 |
ed | -0.326 |
ling | -0.304 |
y | -0.300 |
ier | -0.292 |
Token | scary |
Token ID | 14343 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
pleasure | +0.607 |
Deal | +0.527 |
welcome | +0.489 |
learn | +0.465 |
great | +0.451 |
Neg logit contributions | |
led | -1.277 |
ed | -1.098 |
ied | -0.966 |
ked | -0.943 |
awed | -0.936 |
Token | , |
Token ID | 11 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
led | +0.054 |
ed | +0.045 |
ied | +0.040 |
awed | +0.040 |
ked | +0.039 |
Neg logit contributions | |
pleasure | -0.033 |
welcome | -0.028 |
Deal | -0.028 |
great | -0.027 |
learn | -0.026 |
Token | but |
Token ID | 475 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
pleasure | +0.247 |
Deal | +0.238 |
welcome | +0.208 |
meantime | +0.192 |
learn | +0.191 |
Neg logit contributions | |
led | -0.400 |
ed | -0.339 |
ied | -0.294 |
y | -0.285 |
we | -0.284 |
Token | it |
Token ID | 340 |
Feature activation | +5.52e+00 |
Pos logit contributions | |
pleasure | +0.520 |
Deal | +0.510 |
welcome | +0.430 |
meantime | +0.425 |
learn | +0.401 |
Neg logit contributions | |
led | -0.755 |
ed | -0.640 |
ied | -0.572 |
ing | -0.547 |
y | -0.534 |
Token | 's |
Token ID | 338 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
pleasure | +0.288 |
Deal | +0.257 |
welcome | +0.244 |
learn | +0.232 |
great | +0.231 |
Neg logit contributions | |
led | -0.397 |
ed | -0.332 |
ied | -0.283 |
ked | -0.273 |
ling | -0.272 |
Token | . |
Token ID | 13 |
Feature activation | -5.92e+00 |
Pos logit contributions | |
pleasure | +0.520 |
great | +0.469 |
welcome | +0.464 |
Deal | +0.462 |
think | +0.459 |
Neg logit contributions | |
led | -0.673 |
ed | -0.553 |
ked | -0.493 |
ied | -0.484 |
awed | -0.471 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
led | +0.373 |
ed | +0.358 |
med | +0.300 |
ted | +0.298 |
es | +0.295 |
Neg logit contributions | |
deal | -0.242 |
pleasure | -0.241 |
plan | -0.236 |
great | -0.212 |
learn | -0.211 |
Token | Once |
Token ID | 7454 |
Feature activation | +7.19e+00 |
Pos logit contributions | |
pleasure | +0.060 |
Deal | +0.049 |
welcome | +0.042 |
learn | +0.041 |
great | +0.039 |
Neg logit contributions | |
led | -0.200 |
ed | -0.179 |
ied | -0.157 |
awed | -0.155 |
ted | -0.154 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
pleasure | +0.378 |
Deal | +0.342 |
welcome | +0.319 |
learn | +0.313 |
great | +0.305 |
Neg logit contributions | |
led | -0.509 |
ed | -0.427 |
ied | -0.372 |
y | -0.352 |
awed | -0.351 |
Token | a |
Token ID | 257 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
led | +0.121 |
ed | +0.104 |
y | +0.086 |
med | +0.086 |
awed | +0.086 |
Neg logit contributions | |
pleasure | -0.075 |
Deal | -0.068 |
welcome | -0.059 |
deal | -0.056 |
meantime | -0.056 |
Token | time |
Token ID | 640 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
pleasure | +0.124 |
Deal | +0.104 |
learn | +0.102 |
welcome | +0.098 |
Anything | +0.092 |
Neg logit contributions | |
led | -0.265 |
y | -0.216 |
ed | -0.215 |
ling | -0.206 |
awed | -0.203 |
Token | , |
Token ID | 11 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
pleasure | +0.556 |
Deal | +0.514 |
learn | +0.473 |
deal | +0.461 |
compromise | +0.445 |
Neg logit contributions | |
led | -0.725 |
ed | -0.646 |
y | -0.539 |
ied | -0.534 |
ing | -0.521 |
Token | there |
Token ID | 612 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
pleasure | +0.307 |
Deal | +0.276 |
welcome | +0.265 |
learn | +0.257 |
great | +0.254 |
Neg logit contributions | |
led | -0.357 |
ed | -0.296 |
ied | -0.249 |
ked | -0.241 |
awed | -0.241 |
Token | was |
Token ID | 373 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
led | +0.154 |
ed | +0.128 |
ied | +0.102 |
y | +0.100 |
awed | +0.097 |
Neg logit contributions | |
pleasure | -0.181 |
Deal | -0.175 |
learn | -0.156 |
welcome | -0.155 |
think | -0.150 |
Token | a |
Token ID | 257 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
led | +0.119 |
ed | +0.104 |
med | +0.088 |
ied | +0.087 |
awed | +0.086 |
Neg logit contributions | |
pleasure | -0.067 |
Deal | -0.064 |
deal | -0.057 |
learn | -0.055 |
welcome | -0.055 |
Token | little |
Token ID | 1310 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
pleasure | +0.144 |
Deal | +0.133 |
learn | +0.112 |
welcome | +0.109 |
know | +0.108 |
Neg logit contributions | |
led | -0.257 |
ed | -0.218 |
y | -0.196 |
ling | -0.195 |
ked | -0.188 |
Token | . |
Token ID | 13 |
Feature activation | -5.92e+00 |
Pos logit contributions | |
led | +0.083 |
ed | +0.069 |
awed | +0.063 |
ied | +0.063 |
ked | +0.061 |
Neg logit contributions | |
pleasure | -0.049 |
great | -0.043 |
have | -0.042 |
welcome | -0.042 |
be | -0.042 |
Token | No |
Token ID | 1400 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
led | +0.406 |
ed | +0.362 |
ted | +0.299 |
med | +0.297 |
ked | +0.292 |
Neg logit contributions | |
pleasure | -0.293 |
welcome | -0.265 |
Deal | -0.260 |
think | -0.257 |
learn | -0.254 |
Token | one |
Token ID | 530 |
Feature activation | +8.65e+00 |
Pos logit contributions | |
pleasure | +0.077 |
Deal | +0.064 |
welcome | +0.060 |
learn | +0.056 |
great | +0.056 |
Neg logit contributions | |
led | -0.195 |
ed | -0.169 |
ied | -0.150 |
ked | -0.147 |
awed | -0.147 |
Token | would |
Token ID | 561 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
pleasure | +0.446 |
think | +0.389 |
welcome | +0.388 |
know | +0.383 |
great | +0.374 |
Neg logit contributions | |
led | -0.618 |
ed | -0.499 |
ked | -0.468 |
awed | -0.449 |
ier | -0.437 |
Token | ever |
Token ID | 1683 |
Feature activation | -6.72e+00 |
Pos logit contributions | |
led | +0.409 |
ed | +0.358 |
ked | +0.324 |
y | +0.322 |
ier | +0.319 |
Neg logit contributions | |
pleasure | -0.169 |
Deal | -0.164 |
welcome | -0.126 |
meantime | -0.118 |
Today | -0.117 |
Token | guess |
Token ID | 4724 |
Feature activation | +1.73e+01 |
Pos logit contributions | |
led | +0.633 |
ed | +0.561 |
ked | +0.496 |
y | +0.492 |
ied | +0.492 |
Neg logit contributions | |
pleasure | -0.180 |
Deal | -0.159 |
welcome | -0.123 |
great | -0.111 |
meantime | -0.102 |
Token | it |
Token ID | 340 |
Feature activation | +5.53e+00 |
Pos logit contributions | |
pleasure | +0.738 |
Deal | +0.618 |
welcome | +0.609 |
great | +0.588 |
learn | +0.583 |
Neg logit contributions | |
led | -1.398 |
ed | -1.195 |
ied | -1.049 |
awed | -1.035 |
ked | -1.028 |
Token | was |
Token ID | 373 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
pleasure | +0.285 |
great | +0.254 |
welcome | +0.253 |
think | +0.249 |
learn | +0.248 |
Neg logit contributions | |
led | -0.385 |
ed | -0.310 |
ied | -0.278 |
awed | -0.276 |
ked | -0.273 |
Token | a |
Token ID | 257 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
led | +0.127 |
ed | +0.108 |
ied | +0.094 |
awed | +0.092 |
ked | +0.092 |
Neg logit contributions | |
pleasure | -0.073 |
welcome | -0.062 |
Deal | -0.062 |
great | -0.059 |
learn | -0.058 |
Token | lung |
Token ID | 12317 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
Deal | +0.136 |
pleasure | +0.132 |
learn | +0.115 |
welcome | +0.113 |
know | +0.108 |
Neg logit contributions | |
led | -0.241 |
ed | -0.210 |
y | -0.184 |
ling | -0.179 |
ked | -0.178 |
Token | . |
Token ID | 13 |
Feature activation | -5.96e+00 |
Pos logit contributions | |
led | +0.169 |
ed | +0.141 |
ied | +0.124 |
awed | +0.122 |
ked | +0.120 |
Neg logit contributions | |
pleasure | -0.112 |
welcome | -0.097 |
great | -0.093 |
learn | -0.090 |
think | -0.090 |
Token | She |
Token ID | 1375 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
led | +0.470 |
ed | +0.444 |
ted | +0.379 |
ked | +0.366 |
awed | +0.365 |
Neg logit contributions | |
pleasure | -0.215 |
deal | -0.188 |
plan | -0.180 |
Deal | -0.179 |
great | -0.178 |
Token | knew |
Token ID | 2993 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
led | +0.251 |
ed | +0.206 |
ier | +0.182 |
ied | +0.181 |
ing | +0.174 |
Neg logit contributions | |
pleasure | -0.204 |
think | -0.185 |
learn | -0.181 |
great | -0.181 |
know | -0.178 |
Token | she |
Token ID | 673 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
led | +0.177 |
ed | +0.149 |
ied | +0.129 |
awed | +0.129 |
ked | +0.127 |
Neg logit contributions | |
pleasure | -0.106 |
Deal | -0.089 |
welcome | -0.089 |
great | -0.086 |
learn | -0.084 |
Token | had |
Token ID | 550 |
Feature activation | -7.56e+00 |
Pos logit contributions | |
led | +0.335 |
ed | +0.272 |
awed | +0.253 |
med | +0.252 |
ked | +0.251 |
Neg logit contributions | |
pleasure | -0.203 |
have | -0.189 |
great | -0.185 |
think | -0.184 |
know | -0.183 |
Token | to |
Token ID | 284 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
led | +0.552 |
ed | +0.458 |
ied | +0.392 |
ked | +0.389 |
ling | +0.386 |
Neg logit contributions | |
pleasure | -0.389 |
welcome | -0.335 |
great | -0.330 |
Deal | -0.329 |
learn | -0.327 |
Token | tell |
Token ID | 1560 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
led | +0.320 |
ed | +0.274 |
ly | +0.248 |
ling | +0.244 |
ied | +0.243 |
Neg logit contributions | |
pleasure | -0.168 |
Deal | -0.151 |
Today | -0.123 |
Sometimes | -0.122 |
Not | -0.117 |
Token | her |
Token ID | 607 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
pleasure | +0.539 |
Deal | +0.485 |
welcome | +0.452 |
learn | +0.428 |
great | +0.426 |
Neg logit contributions | |
led | -0.796 |
ed | -0.677 |
ied | -0.586 |
ked | -0.564 |
ted | -0.563 |
Token | dad |
Token ID | 9955 |
Feature activation | +7.02e+00 |
Pos logit contributions | |
led | +0.211 |
ed | +0.183 |
ied | +0.162 |
awed | +0.161 |
ked | +0.161 |
Neg logit contributions | |
pleasure | -0.079 |
Deal | -0.064 |
welcome | -0.062 |
learn | -0.059 |
great | -0.059 |
Token | about |
Token ID | 546 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
pleasure | +0.291 |
welcome | +0.249 |
learn | +0.243 |
great | +0.237 |
Deal | +0.237 |
Neg logit contributions | |
led | -0.569 |
ed | -0.486 |
awed | -0.429 |
ked | -0.426 |
ied | -0.418 |
Token | the |
Token ID | 262 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
led | +0.157 |
ed | +0.135 |
ied | +0.119 |
ked | +0.115 |
awed | +0.115 |
Neg logit contributions | |
pleasure | -0.085 |
Deal | -0.075 |
welcome | -0.070 |
learn | -0.066 |
great | -0.065 |
Token | bug |
Token ID | 5434 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
led | +0.155 |
ed | +0.132 |
ied | +0.116 |
awed | +0.112 |
ked | +0.112 |
Neg logit contributions | |
pleasure | -0.086 |
Deal | -0.076 |
welcome | -0.071 |
learn | -0.068 |
great | -0.066 |
Token | handed |
Token ID | 10158 |
Feature activation | -7.39e+00 |
Pos logit contributions | |
pleasure | +0.133 |
welcome | +0.117 |
learn | +0.115 |
have | +0.112 |
think | +0.112 |
Neg logit contributions | |
led | -0.240 |
ed | -0.199 |
awed | -0.182 |
ied | -0.180 |
ked | -0.178 |
Token | the |
Token ID | 262 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
led | +0.641 |
ed | +0.546 |
awed | +0.491 |
ied | +0.486 |
ked | +0.479 |
Neg logit contributions | |
pleasure | -0.263 |
welcome | -0.219 |
great | -0.204 |
Deal | -0.204 |
learn | -0.201 |
Token | ball |
Token ID | 2613 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
led | +0.160 |
ed | +0.136 |
ked | +0.113 |
awed | +0.111 |
ied | +0.110 |
Neg logit contributions | |
pleasure | -0.134 |
Deal | -0.129 |
welcome | -0.114 |
learn | -0.113 |
know | -0.111 |
Token | to |
Token ID | 284 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
led | +0.131 |
awed | +0.102 |
ed | +0.100 |
we | +0.100 |
med | +0.098 |
Neg logit contributions | |
pleasure | -0.049 |
welcome | -0.048 |
learn | -0.046 |
try | -0.045 |
great | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
led | +0.369 |
ed | +0.311 |
ied | +0.270 |
ted | +0.269 |
ked | +0.266 |
Neg logit contributions | |
pleasure | -0.236 |
Deal | -0.215 |
welcome | -0.190 |
learn | -0.182 |
think | -0.180 |
Token | dog |
Token ID | 3290 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
led | +0.165 |
ed | +0.140 |
ling | +0.116 |
awed | +0.116 |
ked | +0.115 |
Neg logit contributions | |
pleasure | -0.135 |
Deal | -0.127 |
welcome | -0.112 |
learn | -0.110 |
know | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -6.42e+00 |
Pos logit contributions | |
led | +0.209 |
ed | +0.168 |
ied | +0.149 |
awed | +0.145 |
ier | +0.141 |
Neg logit contributions | |
pleasure | -0.154 |
learn | -0.141 |
welcome | -0.141 |
think | -0.139 |
know | -0.135 |
Token |
|
Token ID | 198 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
led | +0.523 |
ed | +0.473 |
ted | +0.403 |
awed | +0.399 |
ied | +0.396 |
Neg logit contributions | |
pleasure | -0.249 |
Deal | -0.222 |
great | -0.201 |
deal | -0.198 |
welcome | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
pleasure | +0.358 |
Deal | +0.285 |
deal | +0.249 |
meantime | +0.235 |
welcome | +0.227 |
Neg logit contributions | |
led | -0.725 |
ed | -0.654 |
ied | -0.581 |
med | -0.570 |
awed | -0.560 |
Token | The |
Token ID | 464 |
Feature activation | -6.14e+00 |
Pos logit contributions | |
pleasure | +0.419 |
Deal | +0.351 |
welcome | +0.330 |
great | +0.308 |
think | +0.307 |
Neg logit contributions | |
led | -0.669 |
ed | -0.569 |
ied | -0.501 |
med | -0.492 |
awed | -0.482 |
Token | dog |
Token ID | 3290 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
led | +0.401 |
ed | +0.331 |
ling | +0.280 |
awed | +0.271 |
ked | +0.271 |
Neg logit contributions | |
pleasure | -0.354 |
Deal | -0.320 |
welcome | -0.283 |
know | -0.278 |
learn | -0.277 |
Token | stop |
Token ID | 2245 |
Feature activation | +7.59e+00 |
Pos logit contributions | |
pleasure | +0.092 |
Deal | +0.080 |
welcome | +0.070 |
learn | +0.066 |
great | +0.065 |
Neg logit contributions | |
led | -0.214 |
ed | -0.187 |
ied | -0.165 |
ked | -0.163 |
y | -0.162 |
Token | . |
Token ID | 13 |
Feature activation | -6.45e+00 |
Pos logit contributions | |
pleasure | +0.430 |
Deal | +0.383 |
welcome | +0.379 |
great | +0.369 |
learn | +0.365 |
Neg logit contributions | |
led | -0.506 |
ed | -0.413 |
ied | -0.357 |
awed | -0.343 |
ked | -0.335 |
Token | He |
Token ID | 679 |
Feature activation | -4.79e+00 |
Pos logit contributions | |
led | +0.481 |
ed | +0.433 |
ked | +0.355 |
ied | +0.353 |
ted | +0.353 |
Neg logit contributions | |
pleasure | -0.307 |
Deal | -0.260 |
welcome | -0.256 |
great | -0.240 |
think | -0.230 |
Token | spun |
Token ID | 26843 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
led | +0.327 |
ed | +0.270 |
ied | +0.243 |
ked | +0.234 |
ier | +0.232 |
Neg logit contributions | |
pleasure | -0.235 |
great | -0.223 |
think | -0.222 |
learn | -0.218 |
know | -0.215 |
Token | around |
Token ID | 1088 |
Feature activation | -9.62e+00 |
Pos logit contributions | |
led | +0.717 |
ed | +0.596 |
ied | +0.503 |
ked | +0.488 |
awed | +0.486 |
Neg logit contributions | |
pleasure | -0.596 |
Deal | -0.536 |
welcome | -0.516 |
learn | -0.498 |
great | -0.494 |
Token | and |
Token ID | 290 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
led | +0.688 |
ed | +0.587 |
ied | +0.494 |
ked | +0.485 |
med | +0.480 |
Neg logit contributions | |
pleasure | -0.495 |
Deal | -0.448 |
welcome | -0.418 |
learn | -0.399 |
great | -0.393 |
Token | around |
Token ID | 1088 |
Feature activation | -9.63e+00 |
Pos logit contributions | |
pleasure | +0.181 |
learn | +0.164 |
great | +0.163 |
think | +0.163 |
welcome | +0.161 |
Neg logit contributions | |
led | -0.173 |
ed | -0.135 |
ied | -0.119 |
awed | -0.114 |
ier | -0.112 |
Token | until |
Token ID | 1566 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
led | +0.729 |
ed | +0.624 |
ied | +0.532 |
ked | +0.523 |
awed | +0.518 |
Neg logit contributions | |
pleasure | -0.458 |
Deal | -0.409 |
welcome | -0.383 |
learn | -0.365 |
great | -0.361 |
Token | he |
Token ID | 339 |
Feature activation | -6.62e+00 |
Pos logit contributions | |
pleasure | +0.385 |
Deal | +0.340 |
welcome | +0.333 |
great | +0.326 |
learn | +0.325 |
Neg logit contributions | |
led | -0.529 |
ed | -0.438 |
ied | -0.382 |
awed | -0.373 |
ked | -0.371 |
Token | was |
Token ID | 373 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
led | +0.442 |
ed | +0.349 |
ied | +0.314 |
ked | +0.311 |
lers | +0.299 |
Neg logit contributions | |
pleasure | -0.349 |
great | -0.321 |
welcome | -0.307 |
learn | -0.307 |
think | -0.306 |
Token | dizz |
Token ID | 29527 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
led | +0.165 |
ed | +0.142 |
ied | +0.122 |
med | +0.119 |
awed | +0.119 |
Neg logit contributions | |
pleasure | -0.104 |
Deal | -0.094 |
welcome | -0.085 |
learn | -0.083 |
deal | -0.083 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
pleasure | +0.172 |
Deal | +0.152 |
welcome | +0.149 |
learn | +0.145 |
great | +0.144 |
Neg logit contributions | |
led | -0.198 |
ed | -0.163 |
ied | -0.139 |
awed | -0.135 |
ked | -0.134 |
Token | how |
Token ID | 703 |
Feature activation | +5.69e+00 |
Pos logit contributions | |
led | +0.218 |
ed | +0.189 |
ied | +0.168 |
awed | +0.166 |
ked | +0.166 |
Neg logit contributions | |
pleasure | -0.086 |
welcome | -0.067 |
Deal | -0.066 |
great | -0.064 |
learn | -0.062 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
pleasure | +0.185 |
great | +0.153 |
welcome | +0.145 |
learn | +0.135 |
know | +0.129 |
Neg logit contributions | |
led | -0.499 |
ed | -0.442 |
ked | -0.405 |
awed | -0.398 |
ted | -0.392 |
Token | and |
Token ID | 290 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
pleasure | +0.417 |
welcome | +0.341 |
great | +0.335 |
learn | +0.325 |
think | +0.317 |
Neg logit contributions | |
led | -0.879 |
ed | -0.755 |
ied | -0.671 |
awed | -0.666 |
ked | -0.664 |
Token | safe |
Token ID | 3338 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
pleasure | +0.114 |
Deal | +0.096 |
welcome | +0.090 |
learn | +0.085 |
great | +0.084 |
Neg logit contributions | |
led | -0.271 |
ed | -0.235 |
ied | -0.208 |
ked | -0.204 |
awed | -0.203 |
Token | the |
Token ID | 262 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
pleasure | +0.166 |
welcome | +0.147 |
great | +0.142 |
learn | +0.138 |
think | +0.135 |
Neg logit contributions | |
led | -0.237 |
ed | -0.197 |
awed | -0.175 |
ked | -0.172 |
ied | -0.170 |
Token | area |
Token ID | 1989 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
led | +0.190 |
ed | +0.162 |
ied | +0.142 |
ling | +0.140 |
ked | +0.138 |
Neg logit contributions | |
pleasure | -0.098 |
Deal | -0.091 |
welcome | -0.079 |
learn | -0.077 |
know | -0.075 |
Token | was |
Token ID | 373 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
led | +0.082 |
ed | +0.068 |
awed | +0.064 |
ked | +0.063 |
ted | +0.059 |
Neg logit contributions | |
pleasure | -0.057 |
welcome | -0.052 |
great | -0.052 |
have | -0.051 |
learn | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -6.24e+00 |
Pos logit contributions | |
led | +0.135 |
ed | +0.113 |
awed | +0.097 |
ked | +0.096 |
ied | +0.096 |
Neg logit contributions | |
pleasure | -0.101 |
welcome | -0.089 |
great | -0.087 |
learn | -0.084 |
think | -0.082 |
Token | They |
Token ID | 1119 |
Feature activation | +9.08e-01 |
Pos logit contributions | |
led | +0.481 |
ed | +0.443 |
ted | +0.386 |
awed | +0.379 |
ied | +0.368 |
Neg logit contributions | |
pleasure | -0.225 |
Deal | -0.204 |
deal | -0.200 |
welcome | -0.194 |
great | -0.194 |
Token | knew |
Token ID | 2993 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
pleasure | +0.049 |
learn | +0.043 |
welcome | +0.043 |
think | +0.042 |
know | +0.042 |
Neg logit contributions | |
led | -0.062 |
ed | -0.052 |
ied | -0.044 |
awed | -0.044 |
we | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
led | +1.082 |
ed | +0.941 |
ied | +0.830 |
ked | +0.814 |
awed | +0.812 |
Neg logit contributions | |
pleasure | -0.472 |
Deal | -0.400 |
welcome | -0.375 |
learn | -0.352 |
great | -0.348 |
Token | boy |
Token ID | 2933 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
pleasure | +0.145 |
Deal | +0.124 |
welcome | +0.123 |
great | +0.118 |
learn | +0.116 |
Neg logit contributions | |
led | -0.268 |
ed | -0.229 |
ied | -0.200 |
ked | -0.196 |
awed | -0.194 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e+00 |
Pos logit contributions | |
led | +0.017 |
ed | +0.014 |
ied | +0.013 |
ier | +0.012 |
awed | +0.011 |
Neg logit contributions | |
learn | -0.010 |
think | -0.010 |
pleasure | -0.009 |
have | -0.009 |
welcome | -0.009 |
Token | made |
Token ID | 925 |
Feature activation | -5.77e+00 |
Pos logit contributions | |
think | +0.165 |
pleasure | +0.163 |
learn | +0.162 |
have | +0.162 |
know | +0.161 |
Neg logit contributions | |
led | -0.247 |
ed | -0.207 |
ied | -0.187 |
awed | -0.181 |
ier | -0.181 |
Token | him |
Token ID | 683 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
led | +0.489 |
ed | +0.424 |
ied | +0.374 |
ked | +0.365 |
awed | +0.364 |
Neg logit contributions | |
pleasure | -0.223 |
Deal | -0.190 |
welcome | -0.179 |
learn | -0.169 |
great | -0.167 |
Token | fall |
Token ID | 2121 |
Feature activation | +5.37e+00 |
Pos logit contributions | |
pleasure | +0.060 |
Deal | +0.049 |
welcome | +0.049 |
learn | +0.048 |
great | +0.046 |
Neg logit contributions | |
led | -0.154 |
ed | -0.133 |
ied | -0.119 |
awed | -0.117 |
ked | -0.116 |
Token | down |
Token ID | 866 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
pleasure | +0.237 |
great | +0.230 |
welcome | +0.227 |
OK | +0.224 |
try | +0.221 |
Neg logit contributions | |
led | -0.381 |
ed | -0.297 |
awed | -0.283 |
ied | -0.282 |
ils | -0.269 |
Token | . |
Token ID | 13 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
led | +0.242 |
ed | +0.198 |
ked | +0.180 |
awed | +0.179 |
ied | +0.176 |
Neg logit contributions | |
welcome | -0.133 |
pleasure | -0.130 |
great | -0.128 |
learn | -0.123 |
have | -0.123 |
Token | She |
Token ID | 1375 |
Feature activation | -3.83e+00 |
Pos logit contributions | |
led | +0.531 |
ed | +0.475 |
ted | +0.415 |
awed | +0.414 |
ied | +0.411 |
Neg logit contributions | |
pleasure | -0.198 |
Deal | -0.162 |
welcome | -0.141 |
great | -0.130 |
learn | -0.130 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
led | +0.274 |
ed | +0.238 |
ied | +0.215 |
ier | +0.213 |
ked | +0.208 |
Neg logit contributions | |
have | -0.153 |
think | -0.151 |
great | -0.151 |
know | -0.150 |
be | -0.148 |
Token | very |
Token ID | 845 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
led | +0.150 |
ed | +0.128 |
ied | +0.110 |
ked | +0.108 |
ted | +0.107 |
Neg logit contributions | |
pleasure | -0.086 |
Deal | -0.076 |
welcome | -0.068 |
deal | -0.065 |
learn | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
led | +0.739 |
ed | +0.633 |
ied | +0.555 |
ked | +0.538 |
awed | +0.537 |
Neg logit contributions | |
pleasure | -0.399 |
Deal | -0.351 |
welcome | -0.329 |
learn | -0.316 |
great | -0.311 |
Token | dinosaur |
Token ID | 28257 |
Feature activation | -8.42e+00 |
Pos logit contributions | |
led | +0.258 |
ed | +0.219 |
ied | +0.190 |
ked | +0.185 |
awed | +0.185 |
Neg logit contributions | |
pleasure | -0.157 |
Deal | -0.137 |
welcome | -0.132 |
great | -0.126 |
learn | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -7.05e+00 |
Pos logit contributions | |
led | +0.600 |
ed | +0.505 |
ied | +0.435 |
awed | +0.433 |
ted | +0.427 |
Neg logit contributions | |
pleasure | -0.406 |
learn | -0.387 |
deal | -0.375 |
plan | -0.369 |
think | -0.367 |
Token | He |
Token ID | 679 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
led | +0.486 |
ed | +0.435 |
ked | +0.355 |
ied | +0.353 |
awed | +0.352 |
Neg logit contributions | |
pleasure | -0.339 |
Deal | -0.309 |
learn | -0.289 |
welcome | -0.289 |
great | -0.285 |
Token | couldn |
Token ID | 3521 |
Feature activation | -4.25e+00 |
Pos logit contributions | |
led | +0.348 |
ed | +0.289 |
ied | +0.250 |
ier | +0.238 |
ked | +0.235 |
Neg logit contributions | |
pleasure | -0.293 |
Deal | -0.260 |
great | -0.255 |
learn | -0.254 |
welcome | -0.253 |
Token | 't |
Token ID | 470 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
led | +0.357 |
ed | +0.323 |
med | +0.276 |
ied | +0.275 |
ted | +0.271 |
Neg logit contributions | |
pleasure | -0.167 |
Deal | -0.131 |
welcome | -0.121 |
deal | -0.113 |
great | -0.110 |
Token | believe |
Token ID | 1975 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
pleasure | +0.064 |
Deal | +0.051 |
welcome | +0.044 |
great | +0.040 |
meantime | +0.038 |
Neg logit contributions | |
led | -0.195 |
ed | -0.173 |
ked | -0.154 |
ly | -0.151 |
ier | -0.150 |
Token | that |
Token ID | 326 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
pleasure | +0.401 |
welcome | +0.330 |
think | +0.328 |
great | +0.324 |
Deal | +0.322 |
Neg logit contributions | |
led | -0.973 |
ed | -0.855 |
awed | -0.754 |
ked | -0.733 |
ied | -0.732 |
Token | a |
Token ID | 257 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
led | +0.140 |
ed | +0.120 |
ied | +0.104 |
ked | +0.101 |
awed | +0.101 |
Neg logit contributions | |
pleasure | -0.084 |
Deal | -0.074 |
welcome | -0.070 |
learn | -0.067 |
great | -0.066 |
Token | real |
Token ID | 1103 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
pleasure | +0.106 |
Deal | +0.094 |
welcome | +0.087 |
learn | +0.084 |
think | +0.081 |
Neg logit contributions | |
led | -0.209 |
ed | -0.181 |
ied | -0.156 |
ked | -0.154 |
awed | -0.153 |
Token | , |
Token ID | 11 |
Feature activation | +4.69e+00 |
Pos logit contributions | |
led | +0.075 |
ed | +0.062 |
ied | +0.054 |
awed | +0.052 |
ked | +0.052 |
Neg logit contributions | |
pleasure | -0.056 |
Deal | -0.049 |
welcome | -0.048 |
great | -0.046 |
learn | -0.046 |
Token | nap |
Token ID | 25422 |
Feature activation | +5.49e+00 |
Pos logit contributions | |
led | +0.888 |
ed | +0.799 |
y | +0.698 |
ied | +0.693 |
ling | +0.677 |
Neg logit contributions | |
pleasure | -0.352 |
Deal | -0.292 |
great | -0.234 |
welcome | -0.223 |
meantime | -0.212 |
Token | under |
Token ID | 739 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
pleasure | +0.333 |
Deal | +0.304 |
welcome | +0.299 |
great | +0.291 |
learn | +0.288 |
Neg logit contributions | |
led | -0.334 |
ed | -0.275 |
ied | -0.233 |
ked | -0.228 |
awed | -0.226 |
Token | the |
Token ID | 262 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
led | +1.210 |
ed | +1.035 |
ted | +0.890 |
med | +0.885 |
ied | +0.869 |
Neg logit contributions | |
pleasure | -0.748 |
Deal | -0.669 |
welcome | -0.614 |
deal | -0.578 |
think | -0.569 |
Token | big |
Token ID | 1263 |
Feature activation | -7.88e-01 |
Pos logit contributions | |
led | +0.206 |
ed | +0.182 |
ied | +0.163 |
ked | +0.162 |
awed | +0.160 |
Neg logit contributions | |
pleasure | -0.059 |
Deal | -0.046 |
welcome | -0.044 |
learn | -0.040 |
great | -0.040 |
Token | tree |
Token ID | 5509 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
led | +0.068 |
ed | +0.059 |
ied | +0.052 |
ked | +0.051 |
awed | +0.051 |
Neg logit contributions | |
pleasure | -0.029 |
Deal | -0.025 |
welcome | -0.024 |
learn | -0.022 |
great | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | -1.21e+01 |
Pos logit contributions | |
pleasure | +0.010 |
learn | +0.009 |
welcome | +0.009 |
think | +0.009 |
have | +0.009 |
Neg logit contributions | |
led | -0.013 |
ed | -0.011 |
ked | -0.010 |
ied | -0.009 |
awed | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
led | +0.837 |
ed | +0.714 |
ied | +0.604 |
med | +0.596 |
y | +0.594 |
Neg logit contributions | |
pleasure | -0.615 |
Deal | -0.555 |
welcome | -0.484 |
deal | -0.484 |
learn | -0.480 |
Token | park |
Token ID | 3952 |
Feature activation | +8.76e+00 |
Pos logit contributions | |
led | +0.165 |
ed | +0.142 |
ied | +0.123 |
awed | +0.121 |
ted | +0.119 |
Neg logit contributions | |
pleasure | -0.093 |
Deal | -0.081 |
welcome | -0.075 |
learn | -0.073 |
think | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
pleasure | +0.474 |
welcome | +0.432 |
learn | +0.427 |
think | +0.413 |
have | +0.405 |
Neg logit contributions | |
led | -0.602 |
ed | -0.487 |
ied | -0.422 |
ked | -0.421 |
awed | -0.417 |
Token | Her |
Token ID | 2332 |
Feature activation | -5.01e+00 |
Pos logit contributions | |
led | +0.460 |
ed | +0.425 |
ted | +0.368 |
y | +0.361 |
awed | +0.360 |
Neg logit contributions | |
pleasure | -0.247 |
Deal | -0.218 |
welcome | -0.195 |
great | -0.189 |
deal | -0.185 |
Token | mom |
Token ID | 1995 |
Feature activation | +6.24e+00 |
Pos logit contributions | |
led | +0.423 |
ed | +0.375 |
ied | +0.326 |
awed | +0.318 |
y | +0.317 |
Neg logit contributions | |
pleasure | -0.172 |
Deal | -0.152 |
welcome | -0.134 |
learn | -0.133 |
think | -0.130 |
Token | was |
Token ID | 373 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
pleasure | +0.126 |
think | +0.111 |
great | +0.110 |
learn | +0.109 |
welcome | +0.107 |
Neg logit contributions | |
led | -0.183 |
ed | -0.150 |
awed | -0.138 |
ied | -0.135 |
ked | -0.134 |
Token | too |
Token ID | 1165 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
led | +0.362 |
ed | +0.310 |
ied | +0.282 |
y | +0.270 |
ling | +0.268 |
Neg logit contributions | |
pleasure | -0.135 |
Deal | -0.114 |
learn | -0.105 |
welcome | -0.097 |
think | -0.094 |
Token | big |
Token ID | 1263 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
pleasure | +0.217 |
Deal | +0.187 |
welcome | +0.178 |
great | +0.169 |
learn | +0.169 |
Neg logit contributions | |
led | -0.436 |
ed | -0.376 |
ied | -0.330 |
ked | -0.324 |
awed | -0.322 |
Token | to |
Token ID | 284 |
Feature activation | -6.13e+00 |
Pos logit contributions | |
led | +0.207 |
ed | +0.170 |
ied | +0.157 |
awed | +0.155 |
ked | +0.152 |
Neg logit contributions | |
pleasure | -0.110 |
welcome | -0.094 |
think | -0.089 |
learn | -0.088 |
deal | -0.087 |
Token | fit |
Token ID | 4197 |
Feature activation | -4.94e+00 |
Pos logit contributions | |
led | +0.506 |
ed | +0.432 |
y | +0.389 |
ied | +0.388 |
ling | +0.384 |
Neg logit contributions | |
pleasure | -0.233 |
Deal | -0.201 |
welcome | -0.159 |
great | -0.151 |
meantime | -0.147 |
Token | in |
Token ID | 287 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
led | +0.325 |
ed | +0.265 |
awed | +0.235 |
ked | +0.234 |
ied | +0.226 |
Neg logit contributions | |
pleasure | -0.271 |
welcome | -0.244 |
great | -0.239 |
think | -0.234 |
Deal | -0.232 |
Token | the |
Token ID | 262 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
led | +0.967 |
ed | +0.830 |
ied | +0.749 |
med | +0.722 |
y | +0.709 |
Neg logit contributions | |
pleasure | -0.598 |
Deal | -0.544 |
welcome | -0.478 |
learn | -0.464 |
deal | -0.455 |
Token | suitcase |
Token ID | 45391 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
led | +0.247 |
ed | +0.210 |
ied | +0.183 |
ling | +0.177 |
awed | +0.176 |
Neg logit contributions | |
pleasure | -0.144 |
Deal | -0.128 |
welcome | -0.117 |
learn | -0.115 |
know | -0.112 |
Token | , |
Token ID | 11 |
Feature activation | +4.30e+00 |
Pos logit contributions | |
led | +0.212 |
ed | +0.174 |
awed | +0.152 |
ked | +0.151 |
ied | +0.147 |
Neg logit contributions | |
pleasure | -0.165 |
welcome | -0.149 |
learn | -0.144 |
think | -0.140 |
great | -0.140 |
Token | and |
Token ID | 290 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
pleasure | +0.189 |
Deal | +0.171 |
welcome | +0.156 |
learn | +0.149 |
great | +0.146 |
Neg logit contributions | |
led | -0.332 |
ed | -0.285 |
ied | -0.246 |
ked | -0.243 |
awed | -0.239 |
Token | that |
Token ID | 326 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
pleasure | +0.121 |
welcome | +0.103 |
Deal | +0.102 |
learn | +0.100 |
think | +0.099 |
Neg logit contributions | |
led | -0.196 |
ed | -0.164 |
ied | -0.144 |
awed | -0.144 |
ked | -0.142 |
Token | look |
Token ID | 804 |
Feature activation | +1.48e+01 |
Pos logit contributions | |
led | +0.384 |
ed | +0.343 |
ied | +0.310 |
y | +0.304 |
ted | +0.294 |
Neg logit contributions | |
pleasure | -0.182 |
Deal | -0.169 |
Today | -0.121 |
great | -0.120 |
welcome | -0.118 |
Token | for |
Token ID | 329 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
pleasure | +0.700 |
welcome | +0.624 |
Deal | +0.611 |
great | +0.604 |
learn | +0.588 |
Neg logit contributions | |
led | -1.087 |
ed | -0.932 |
awed | -0.833 |
ked | -0.822 |
ied | -0.796 |
Token | her |
Token ID | 607 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
led | +0.083 |
ed | +0.070 |
ied | +0.060 |
ted | +0.058 |
ked | +0.058 |
Neg logit contributions | |
pleasure | -0.061 |
Deal | -0.057 |
welcome | -0.052 |
learn | -0.050 |
think | -0.049 |
Token | dolls |
Token ID | 36062 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
led | +0.213 |
ed | +0.180 |
ied | +0.156 |
ked | +0.152 |
awed | +0.152 |
Neg logit contributions | |
pleasure | -0.135 |
Deal | -0.119 |
welcome | -0.114 |
learn | -0.109 |
great | -0.109 |
Token | again |
Token ID | 757 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
led | +0.266 |
ed | +0.217 |
awed | +0.196 |
ied | +0.193 |
med | +0.187 |
Neg logit contributions | |
pleasure | -0.169 |
welcome | -0.153 |
learn | -0.153 |
think | -0.147 |
have | -0.143 |
Token | . |
Token ID | 13 |
Feature activation | -6.29e+00 |
Pos logit contributions | |
led | +0.068 |
ed | +0.055 |
ied | +0.046 |
awed | +0.044 |
ked | +0.044 |
Neg logit contributions | |
pleasure | -0.068 |
Deal | -0.062 |
welcome | -0.059 |
learn | -0.057 |
great | -0.057 |
Token | She |
Token ID | 1375 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
led | +0.398 |
ed | +0.373 |
ted | +0.294 |
ked | +0.293 |
ied | +0.293 |
Neg logit contributions | |
pleasure | -0.328 |
Deal | -0.294 |
great | -0.284 |
welcome | -0.278 |
plan | -0.266 |
Token | opened |
Token ID | 4721 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
led | +0.300 |
ed | +0.248 |
ier | +0.231 |
ing | +0.227 |
ied | +0.224 |
Neg logit contributions | |
think | -0.188 |
pleasure | -0.188 |
learn | -0.183 |
great | -0.179 |
see | -0.179 |
Token | the |
Token ID | 262 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
led | +1.035 |
ed | +0.885 |
ied | +0.773 |
awed | +0.758 |
ked | +0.757 |
Neg logit contributions | |
pleasure | -0.550 |
Deal | -0.470 |
welcome | -0.452 |
great | -0.432 |
learn | -0.430 |
Token | case |
Token ID | 1339 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
led | +0.171 |
ed | +0.144 |
ied | +0.123 |
ked | +0.120 |
awed | +0.120 |
Neg logit contributions | |
pleasure | -0.118 |
Deal | -0.105 |
welcome | -0.101 |
learn | -0.097 |
great | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
pleasure | +0.021 |
welcome | +0.019 |
learn | +0.018 |
think | +0.017 |
Deal | +0.017 |
Neg logit contributions | |
led | -0.042 |
ed | -0.035 |
ked | -0.032 |
awed | -0.031 |
med | -0.031 |
Token | pony |
Token ID | 26902 |
Feature activation | -6.42e+00 |
Pos logit contributions | |
led | +0.204 |
ed | +0.171 |
ied | +0.145 |
ked | +0.141 |
awed | +0.140 |
Neg logit contributions | |
pleasure | -0.160 |
Deal | -0.144 |
welcome | -0.137 |
learn | -0.133 |
great | -0.131 |
Token | stopped |
Token ID | 5025 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
led | +0.410 |
ed | +0.327 |
ied | +0.303 |
ier | +0.287 |
awed | +0.285 |
Neg logit contributions | |
pleasure | -0.335 |
learn | -0.315 |
think | -0.309 |
have | -0.307 |
know | -0.307 |
Token | in |
Token ID | 287 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
led | +0.375 |
ed | +0.309 |
ied | +0.271 |
awed | +0.258 |
ier | +0.255 |
Neg logit contributions | |
pleasure | -0.269 |
welcome | -0.245 |
Deal | -0.238 |
think | -0.235 |
great | -0.231 |
Token | her |
Token ID | 607 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
led | +0.946 |
ed | +0.824 |
ied | +0.690 |
ted | +0.687 |
y | +0.685 |
Neg logit contributions | |
pleasure | -0.582 |
Deal | -0.511 |
learn | -0.466 |
welcome | -0.464 |
great | -0.456 |
Token | tracks |
Token ID | 8339 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
led | +0.242 |
ed | +0.207 |
ied | +0.179 |
awed | +0.175 |
ked | +0.175 |
Neg logit contributions | |
pleasure | -0.139 |
Deal | -0.124 |
welcome | -0.116 |
learn | -0.111 |
great | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -6.56e+00 |
Pos logit contributions | |
led | +0.152 |
ed | +0.121 |
awed | +0.113 |
we | +0.111 |
ied | +0.109 |
Neg logit contributions | |
pleasure | -0.087 |
| -0.080 |
welcome | -0.079 |
learn | -0.072 |
be | -0.072 |
Token | She |
Token ID | 1375 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
led | +0.508 |
ed | +0.460 |
ted | +0.389 |
awed | +0.383 |
med | +0.382 |
Neg logit contributions | |
pleasure | -0.264 |
Deal | -0.232 |
learn | -0.219 |
deal | -0.219 |
great | -0.216 |
Token | felt |
Token ID | 2936 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
led | +0.290 |
ed | +0.240 |
ied | +0.210 |
ier | +0.208 |
awed | +0.199 |
Neg logit contributions | |
pleasure | -0.241 |
think | -0.216 |
great | -0.214 |
learn | -0.213 |
know | -0.211 |
Token | a |
Token ID | 257 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
led | +0.201 |
ed | +0.172 |
ied | +0.147 |
ling | +0.144 |
med | +0.143 |
Neg logit contributions | |
pleasure | -0.101 |
Deal | -0.095 |
deal | -0.080 |
learn | -0.078 |
welcome | -0.078 |
Token | bit |
Token ID | 1643 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
pleasure | +0.136 |
Deal | +0.127 |
welcome | +0.117 |
learn | +0.116 |
think | +0.109 |
Neg logit contributions | |
led | -0.198 |
ed | -0.165 |
ling | -0.142 |
ked | -0.138 |
y | -0.138 |
Token | scared |
Token ID | 12008 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
led | +0.054 |
ed | +0.046 |
ied | +0.041 |
awed | +0.040 |
ked | +0.040 |
Neg logit contributions | |
pleasure | -0.031 |
welcome | -0.027 |
Deal | -0.027 |
great | -0.026 |
think | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
led | +0.396 |
ed | +0.346 |
ied | +0.278 |
ked | +0.278 |
ted | +0.275 |
Neg logit contributions | |
pleasure | -0.312 |
Deal | -0.272 |
welcome | -0.257 |
great | -0.253 |
learn | -0.252 |
Token | showed |
Token ID | 3751 |
Feature activation | -5.65e+00 |
Pos logit contributions | |
led | +0.309 |
ed | +0.263 |
ing | +0.241 |
ied | +0.241 |
ting | +0.240 |
Neg logit contributions | |
think | -0.170 |
pleasure | -0.169 |
know | -0.163 |
have | -0.162 |
learn | -0.161 |
Token | her |
Token ID | 607 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
led | +0.490 |
ed | +0.411 |
awed | +0.380 |
ked | +0.372 |
ling | +0.366 |
Neg logit contributions | |
pleasure | -0.198 |
welcome | -0.166 |
great | -0.160 |
learn | -0.158 |
Deal | -0.152 |
Token | a |
Token ID | 257 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
led | +0.189 |
ed | +0.160 |
awed | +0.142 |
ked | +0.140 |
ied | +0.139 |
Neg logit contributions | |
pleasure | -0.097 |
learn | -0.081 |
welcome | -0.081 |
great | -0.079 |
Deal | -0.079 |
Token | soft |
Token ID | 2705 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
pleasure | +0.163 |
Deal | +0.149 |
welcome | +0.133 |
learn | +0.132 |
think | +0.129 |
Neg logit contributions | |
led | -0.236 |
ed | -0.204 |
ied | -0.177 |
y | -0.173 |
ked | -0.168 |
Token | bear |
Token ID | 6842 |
Feature activation | -5.69e+00 |
Pos logit contributions | |
led | +0.081 |
ed | +0.068 |
ied | +0.059 |
ked | +0.058 |
awed | +0.058 |
Neg logit contributions | |
pleasure | -0.052 |
Deal | -0.046 |
welcome | -0.044 |
learn | -0.042 |
great | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
led | +0.480 |
ed | +0.372 |
awed | +0.357 |
ied | +0.352 |
attered | +0.334 |
Neg logit contributions | |
learn | -0.214 |
think | -0.196 |
have | -0.195 |
pleasure | -0.194 |
welcome | -0.189 |
Token | He |
Token ID | 679 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
led | +0.425 |
ed | +0.370 |
ied | +0.306 |
ked | +0.305 |
ted | +0.304 |
Neg logit contributions | |
pleasure | -0.290 |
Deal | -0.254 |
welcome | -0.237 |
great | -0.232 |
learn | -0.232 |
Token | said |
Token ID | 531 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
led | +0.314 |
ed | +0.268 |
ied | +0.251 |
ing | +0.248 |
ting | +0.247 |
Neg logit contributions | |
think | -0.162 |
pleasure | -0.155 |
have | -0.154 |
know | -0.153 |
learn | -0.153 |
Token | the |
Token ID | 262 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
led | +0.113 |
ed | +0.095 |
ied | +0.082 |
awed | +0.081 |
ked | +0.080 |
Neg logit contributions | |
pleasure | -0.073 |
Deal | -0.063 |
welcome | -0.062 |
great | -0.060 |
learn | -0.059 |
Token | bear |
Token ID | 6842 |
Feature activation | -5.70e+00 |
Pos logit contributions | |
led | +0.140 |
ed | +0.118 |
ied | +0.101 |
ked | +0.098 |
awed | +0.097 |
Neg logit contributions | |
pleasure | -0.104 |
Deal | -0.093 |
welcome | -0.088 |
learn | -0.086 |
great | -0.084 |
Token | eat |
Token ID | 4483 |
Feature activation | +6.53e+00 |
Pos logit contributions | |
led | +0.419 |
ed | +0.352 |
ted | +0.313 |
ling | +0.313 |
ied | +0.312 |
Neg logit contributions | |
pleasure | -0.202 |
Deal | -0.185 |
Today | -0.134 |
welcome | -0.133 |
great | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | -6.57e+00 |
Pos logit contributions | |
pleasure | +0.362 |
welcome | +0.332 |
great | +0.321 |
learn | +0.316 |
Deal | +0.315 |
Neg logit contributions | |
led | -0.426 |
ed | -0.355 |
ked | -0.304 |
awed | -0.296 |
ied | -0.296 |
Token |
|
Token ID | 198 |
Feature activation | +8.57e+00 |
Pos logit contributions | |
led | +0.582 |
ed | +0.522 |
awed | +0.457 |
ted | +0.454 |
ied | +0.453 |
Neg logit contributions | |
pleasure | -0.234 |
Deal | -0.200 |
welcome | -0.183 |
learn | -0.163 |
great | -0.161 |
Token |
|
Token ID | 198 |
Feature activation | +8.56e+00 |
Pos logit contributions | |
pleasure | +0.335 |
Deal | +0.279 |
welcome | +0.238 |
deal | +0.234 |
learn | +0.212 |
Neg logit contributions | |
led | -0.745 |
ed | -0.659 |
ied | -0.596 |
awed | -0.582 |
y | -0.581 |
Token | L |
Token ID | 43 |
Feature activation | -7.41e+00 |
Pos logit contributions | |
pleasure | +0.396 |
Deal | +0.345 |
welcome | +0.329 |
learn | +0.313 |
great | +0.310 |
Neg logit contributions | |
led | -0.668 |
ed | -0.569 |
ied | -0.496 |
awed | -0.482 |
ked | -0.482 |
Token | ily |
Token ID | 813 |
Feature activation | -1.73e+01 |
Pos logit contributions | |
led | +0.438 |
ed | +0.354 |
ked | +0.295 |
ted | +0.285 |
awed | +0.283 |
Neg logit contributions | |
pleasure | -0.458 |
Deal | -0.421 |
welcome | -0.416 |
learn | -0.400 |
great | -0.397 |
Token | was |
Token ID | 373 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
led | +1.192 |
ed | +0.993 |
ied | +0.848 |
ked | +0.825 |
awed | +0.816 |
Neg logit contributions | |
pleasure | -0.928 |
Deal | -0.836 |
welcome | -0.803 |
learn | -0.799 |
great | -0.785 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
led | +0.171 |
ed | +0.149 |
awed | +0.125 |
ked | +0.125 |
ied | +0.124 |
Neg logit contributions | |
pleasure | -0.102 |
Deal | -0.095 |
deal | -0.081 |
meantime | -0.078 |
welcome | -0.078 |
Token | to |
Token ID | 284 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
pleasure | +0.490 |
learn | +0.423 |
welcome | +0.422 |
think | +0.419 |
Deal | +0.404 |
Neg logit contributions | |
led | -0.784 |
ed | -0.648 |
ied | -0.574 |
ked | -0.563 |
awed | -0.550 |
Token | learn |
Token ID | 2193 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
led | +0.418 |
ed | +0.394 |
red | +0.337 |
ked | +0.337 |
ted | +0.331 |
Neg logit contributions | |
pleasure | -0.179 |
Deal | -0.145 |
great | -0.101 |
meantime | -0.096 |
Sometimes | -0.094 |
Token | about |
Token ID | 546 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
pleasure | +0.580 |
welcome | +0.499 |
learn | +0.490 |
Deal | +0.485 |
great | +0.483 |
Neg logit contributions | |
led | -1.093 |
ed | -0.935 |
ked | -0.816 |
ied | -0.803 |
awed | -0.803 |
Token | loud |
Token ID | 7812 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
led | +0.131 |
ed | +0.109 |
awed | +0.103 |
ied | +0.101 |
attered | +0.099 |
Neg logit contributions | |
welcome | -0.055 |
pleasure | -0.054 |
great | -0.054 |
Deal | -0.050 |
learn | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -6.09e+00 |
Pos logit contributions | |
pleasure | +0.026 |
welcome | +0.025 |
Deal | +0.024 |
learn | +0.024 |
think | +0.024 |
Neg logit contributions | |
led | -0.038 |
ed | -0.030 |
ied | -0.029 |
awed | -0.026 |
attered | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | +9.07e+00 |
Pos logit contributions | |
led | +0.488 |
ed | +0.453 |
awed | +0.381 |
ted | +0.378 |
med | +0.370 |
Neg logit contributions | |
pleasure | -0.249 |
Deal | -0.188 |
deal | -0.177 |
promise | -0.173 |
learn | -0.172 |
Token |
|
Token ID | 198 |
Feature activation | +9.08e+00 |
Pos logit contributions | |
pleasure | +0.395 |
Deal | +0.302 |
deal | +0.293 |
welcome | +0.264 |
meantime | +0.262 |
Neg logit contributions | |
led | -0.707 |
ed | -0.647 |
ied | -0.554 |
awed | -0.549 |
es | -0.540 |
Token | L |
Token ID | 43 |
Feature activation | -6.88e+00 |
Pos logit contributions | |
pleasure | +0.390 |
Deal | +0.338 |
welcome | +0.321 |
learn | +0.306 |
great | +0.304 |
Neg logit contributions | |
led | -0.737 |
ed | -0.633 |
ied | -0.554 |
ked | -0.541 |
awed | -0.540 |
Token | ily |
Token ID | 813 |
Feature activation | -1.68e+01 |
Pos logit contributions | |
led | +0.437 |
ed | +0.347 |
ked | +0.298 |
lers | +0.291 |
ted | +0.290 |
Neg logit contributions | |
pleasure | -0.395 |
Deal | -0.379 |
welcome | -0.375 |
learn | -0.356 |
think | -0.347 |
Token | is |
Token ID | 318 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
led | +1.341 |
ed | +1.128 |
ied | +1.035 |
ked | +0.984 |
ier | +0.970 |
Neg logit contributions | |
pleasure | -0.668 |
Deal | -0.628 |
learn | -0.604 |
welcome | -0.597 |
think | -0.585 |
Token | scared |
Token ID | 12008 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
pleasure | +0.192 |
welcome | +0.178 |
Deal | +0.175 |
great | +0.169 |
learn | +0.166 |
Neg logit contributions | |
led | -0.349 |
ed | -0.298 |
ied | -0.266 |
ted | -0.259 |
ked | -0.256 |
Token | . |
Token ID | 13 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
led | +0.242 |
ed | +0.199 |
ied | +0.172 |
ked | +0.170 |
awed | +0.168 |
Neg logit contributions | |
pleasure | -0.178 |
welcome | -0.171 |
Deal | -0.168 |
learn | -0.160 |
think | -0.158 |
Token | She |
Token ID | 1375 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
led | +0.353 |
ed | +0.328 |
awed | +0.254 |
ted | +0.248 |
y | +0.246 |
Neg logit contributions | |
pleasure | -0.379 |
Deal | -0.314 |
great | -0.307 |
deal | -0.304 |
plan | -0.298 |
Token | runs |
Token ID | 4539 |
Feature activation | -5.62e+00 |
Pos logit contributions | |
led | +0.289 |
ed | +0.243 |
ied | +0.207 |
ked | +0.203 |
awed | +0.203 |
Neg logit contributions | |
pleasure | -0.202 |
Deal | -0.176 |
welcome | -0.169 |
learn | -0.161 |
great | -0.160 |
Token | snow |
Token ID | 6729 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
pleasure | +0.064 |
Deal | +0.060 |
welcome | +0.053 |
learn | +0.052 |
think | +0.050 |
Neg logit contributions | |
led | -0.104 |
ed | -0.088 |
y | -0.076 |
ling | -0.076 |
ied | -0.075 |
Token | man |
Token ID | 805 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
pleasure | +0.087 |
welcome | +0.078 |
great | +0.076 |
deal | +0.076 |
think | +0.075 |
Neg logit contributions | |
led | -0.094 |
ed | -0.076 |
ied | -0.068 |
awed | -0.066 |
ked | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -7.26e+00 |
Pos logit contributions | |
led | +0.883 |
ed | +0.733 |
ied | +0.652 |
awed | +0.636 |
ked | +0.616 |
Neg logit contributions | |
pleasure | -0.637 |
welcome | -0.563 |
learn | -0.541 |
Deal | -0.540 |
great | -0.537 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +8.81e-01 |
Pos logit contributions | |
led | +0.465 |
ed | +0.417 |
y | +0.362 |
ted | +0.357 |
med | +0.353 |
Neg logit contributions | |
pleasure | -0.337 |
great | -0.297 |
Deal | -0.286 |
plan | -0.283 |
deal | -0.283 |
Token | L |
Token ID | 43 |
Feature activation | -7.87e+00 |
Pos logit contributions | |
pleasure | +0.034 |
Deal | +0.029 |
learn | +0.026 |
welcome | +0.025 |
deal | +0.025 |
Neg logit contributions | |
led | -0.074 |
ed | -0.066 |
ied | -0.056 |
awed | -0.056 |
y | -0.056 |
Token | ily |
Token ID | 813 |
Feature activation | -1.77e+01 |
Pos logit contributions | |
led | +0.476 |
ed | +0.386 |
ied | +0.319 |
ked | +0.306 |
awed | +0.306 |
Neg logit contributions | |
pleasure | -0.497 |
Deal | -0.451 |
welcome | -0.436 |
learn | -0.423 |
great | -0.421 |
Token | was |
Token ID | 373 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
led | +1.289 |
ed | +1.080 |
ied | +0.957 |
ked | +0.945 |
awed | +0.918 |
Neg logit contributions | |
pleasure | -0.854 |
welcome | -0.750 |
Deal | -0.734 |
learn | -0.731 |
think | -0.722 |
Token | a |
Token ID | 257 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
led | +0.176 |
ed | +0.148 |
ied | +0.125 |
awed | +0.123 |
ked | +0.120 |
Neg logit contributions | |
pleasure | -0.134 |
Deal | -0.120 |
deal | -0.113 |
learn | -0.112 |
welcome | -0.111 |
Token | creative |
Token ID | 7325 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
pleasure | +0.100 |
Deal | +0.087 |
welcome | +0.079 |
learn | +0.078 |
know | +0.074 |
Neg logit contributions | |
led | -0.195 |
ed | -0.167 |
y | -0.146 |
ling | -0.145 |
ied | -0.144 |
Token | girl |
Token ID | 2576 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
pleasure | +0.101 |
Deal | +0.087 |
welcome | +0.086 |
great | +0.082 |
learn | +0.081 |
Neg logit contributions | |
led | -0.163 |
ed | -0.139 |
ied | -0.121 |
awed | -0.118 |
ked | -0.118 |
Token | who |
Token ID | 508 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
led | +0.272 |
ed | +0.227 |
ied | +0.206 |
ked | +0.198 |
awed | +0.197 |
Neg logit contributions | |
pleasure | -0.161 |
welcome | -0.151 |
have | -0.147 |
learn | -0.145 |
think | -0.144 |
Token | loudly |
Token ID | 23112 |
Feature activation | -7.97e+00 |
Pos logit contributions | |
led | +1.564 |
ed | +1.269 |
ied | +1.177 |
ier | +1.156 |
awed | +1.149 |
Neg logit contributions | |
pleasure | -0.880 |
great | -0.831 |
welcome | -0.826 |
learn | -0.791 |
try | -0.780 |
Token | . |
Token ID | 13 |
Feature activation | -6.07e+00 |
Pos logit contributions | |
led | +0.586 |
ied | +0.466 |
ed | +0.443 |
awed | +0.431 |
attered | +0.424 |
Neg logit contributions | |
pleasure | -0.339 |
welcome | -0.313 |
great | -0.299 |
learn | -0.295 |
try | -0.293 |
Token | He |
Token ID | 679 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
led | +0.423 |
ed | +0.375 |
ked | +0.308 |
ted | +0.308 |
med | +0.303 |
Neg logit contributions | |
pleasure | -0.307 |
Deal | -0.264 |
welcome | -0.253 |
learn | -0.253 |
think | -0.248 |
Token | pe |
Token ID | 613 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
led | +0.290 |
ed | +0.246 |
ied | +0.229 |
ier | +0.220 |
ing | +0.212 |
Neg logit contributions | |
pleasure | -0.203 |
think | -0.200 |
great | -0.197 |
learn | -0.192 |
have | -0.189 |
Token | ck |
Token ID | 694 |
Feature activation | -5.90e+00 |
Pos logit contributions | |
led | +0.150 |
ed | +0.133 |
ly | +0.120 |
attered | +0.118 |
ol | +0.117 |
Neg logit contributions | |
think | -0.126 |
Deal | -0.124 |
guess | -0.121 |
pleasure | -0.120 |
know | -0.118 |
Token | ed |
Token ID | 276 |
Feature activation | -2.07e+01 |
Pos logit contributions | |
led | +0.094 |
web | +0.085 |
attered | +0.074 |
rying | +0.074 |
year | +0.073 |
Neg logit contributions | |
great | -0.437 |
see | -0.424 |
have | -0.419 |
be | -0.418 |
try | -0.418 |
Token | at |
Token ID | 379 |
Feature activation | -7.90e+00 |
Pos logit contributions | |
led | +1.523 |
ed | +1.216 |
ied | +1.195 |
ier | +1.179 |
awed | +1.126 |
Neg logit contributions | |
pleasure | -0.884 |
great | -0.798 |
welcome | -0.773 |
think | -0.735 |
Deal | -0.729 |
Token | their |
Token ID | 511 |
Feature activation | +8.16e-03 |
Pos logit contributions | |
led | +0.600 |
ed | +0.512 |
ied | +0.443 |
ked | +0.435 |
ted | +0.434 |
Neg logit contributions | |
pleasure | -0.379 |
Deal | -0.330 |
welcome | -0.309 |
learn | -0.298 |
great | -0.292 |
Token | shoes |
Token ID | 10012 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
pleasure | +0.000 |
Deal | +0.000 |
welcome | +0.000 |
learn | +0.000 |
great | +0.000 |
Neg logit contributions | |
led | -0.001 |
ed | -0.000 |
ied | -0.000 |
ked | -0.000 |
awed | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
pleasure | +0.082 |
learn | +0.076 |
think | +0.072 |
welcome | +0.072 |
have | +0.071 |
Neg logit contributions | |
led | -0.197 |
ed | -0.163 |
ied | -0.147 |
awed | -0.144 |
we | -0.140 |
Token | legs |
Token ID | 7405 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
pleasure | +0.155 |
think | +0.148 |
learn | +0.144 |
be | +0.144 |
great | +0.143 |
Neg logit contributions | |
led | -0.220 |
ed | -0.183 |
ied | -0.178 |
ier | -0.172 |
ing | -0.162 |
Token | it |
Token ID | 340 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
led | +0.236 |
ed | +0.204 |
ied | +0.180 |
ked | +0.176 |
awed | +0.176 |
Neg logit contributions | |
pleasure | -0.108 |
Deal | -0.092 |
welcome | -0.087 |
learn | -0.083 |
great | -0.082 |
Token | ." |
Token ID | 526 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
pleasure | +0.250 |
welcome | +0.218 |
Deal | +0.215 |
great | +0.209 |
learn | +0.205 |
Neg logit contributions | |
led | -0.363 |
ed | -0.303 |
ied | -0.265 |
ked | -0.265 |
awed | -0.264 |
Token |
|
Token ID | 198 |
Feature activation | +8.74e+00 |
Pos logit contributions | |
pleasure | +0.055 |
Deal | +0.045 |
great | +0.040 |
welcome | +0.038 |
deal | +0.036 |
Neg logit contributions | |
led | -0.165 |
ed | -0.148 |
ted | -0.129 |
awed | -0.129 |
ied | -0.128 |
Token |
|
Token ID | 198 |
Feature activation | +8.77e+00 |
Pos logit contributions | |
pleasure | +0.318 |
Deal | +0.258 |
great | +0.222 |
deal | +0.222 |
welcome | +0.221 |
Neg logit contributions | |
led | -0.749 |
ed | -0.659 |
ied | -0.595 |
awed | -0.583 |
med | -0.575 |
Token | L |
Token ID | 43 |
Feature activation | -7.17e+00 |
Pos logit contributions | |
pleasure | +0.453 |
Deal | +0.388 |
welcome | +0.365 |
great | +0.353 |
deal | +0.345 |
Neg logit contributions | |
led | -0.621 |
ed | -0.527 |
ied | -0.464 |
med | -0.449 |
awed | -0.449 |
Token | ily |
Token ID | 813 |
Feature activation | -1.71e+01 |
Pos logit contributions | |
led | +0.437 |
ed | +0.349 |
ked | +0.291 |
ted | +0.282 |
awed | +0.282 |
Neg logit contributions | |
pleasure | -0.445 |
Deal | -0.414 |
welcome | -0.407 |
learn | -0.391 |
think | -0.384 |
Token | felt |
Token ID | 2936 |
Feature activation | -2.29e+00 |
Pos logit contributions | |
led | +1.119 |
ed | +0.913 |
ied | +0.789 |
ked | +0.766 |
awed | +0.754 |
Neg logit contributions | |
pleasure | -0.963 |
Deal | -0.863 |
welcome | -0.850 |
learn | -0.849 |
think | -0.834 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
led | +0.195 |
ed | +0.169 |
ied | +0.144 |
ling | +0.142 |
ier | +0.140 |
Neg logit contributions | |
pleasure | -0.086 |
Deal | -0.078 |
deal | -0.064 |
welcome | -0.063 |
learn | -0.062 |
Token | that |
Token ID | 326 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
pleasure | +0.560 |
Deal | +0.488 |
welcome | +0.479 |
learn | +0.464 |
great | +0.457 |
Neg logit contributions | |
led | -0.769 |
ed | -0.643 |
ied | -0.555 |
ked | -0.540 |
awed | -0.536 |
Token | she |
Token ID | 673 |
Feature activation | -4.90e+00 |
Pos logit contributions | |
led | +0.083 |
ed | +0.070 |
ied | +0.058 |
y | +0.054 |
ted | +0.053 |
Neg logit contributions | |
pleasure | -0.082 |
Deal | -0.076 |
welcome | -0.070 |
great | -0.068 |
learn | -0.067 |
Token | was |
Token ID | 373 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
led | +0.315 |
ied | +0.252 |
ed | +0.247 |
ked | +0.234 |
awed | +0.234 |
Neg logit contributions | |
pleasure | -0.248 |
have | -0.233 |
welcome | -0.230 |
think | -0.229 |
see | -0.225 |