Token | use |
Token ID | 779 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
Max | +0.075 |
She | +0.074 |
Jake | +0.073 |
Un | +0.073 |
They | +0.073 |
Neg logit contributions | |
aches | -0.099 |
le | -0.094 |
ered | -0.090 |
away | -0.088 |
up | -0.087 |
Token | it |
Token ID | 340 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
le | +0.262 |
aches | +0.256 |
ling | +0.244 |
an | +0.230 |
ered | +0.229 |
Neg logit contributions | |
She | -0.378 |
They | -0.372 |
He | -0.350 |
Un | -0.346 |
Not | -0.344 |
Token | to |
Token ID | 284 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
le | +0.328 |
aches | +0.315 |
ered | +0.292 |
ck | +0.288 |
ling | +0.284 |
Neg logit contributions | |
She | -0.492 |
They | -0.479 |
He | -0.451 |
Max | -0.440 |
Did | -0.440 |
Token | build |
Token ID | 1382 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
le | +0.192 |
ling | +0.180 |
aches | +0.176 |
- | +0.171 |
les | +0.168 |
Neg logit contributions | |
She | -0.261 |
They | -0.257 |
Un | -0.253 |
Max | -0.248 |
Jake | -0.242 |
Token | our |
Token ID | 674 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
le | +0.292 |
aches | +0.291 |
ling | +0.271 |
ered | +0.261 |
an | +0.254 |
Neg logit contributions | |
They | -0.405 |
She | -0.397 |
He | -0.372 |
Max | -0.367 |
Something | -0.366 |
Token | nests |
Token ID | 44382 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
le | +0.278 |
aches | +0.262 |
ered | +0.257 |
ling | +0.236 |
ck | +0.234 |
Neg logit contributions | |
They | -0.346 |
She | -0.334 |
Deal | -0.327 |
Max | -0.325 |
Un | -0.322 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
ered | +0.458 |
le | +0.440 |
aches | +0.434 |
ck | +0.419 |
crown | +0.404 |
Neg logit contributions | |
They | -0.681 |
She | -0.670 |
He | -0.629 |
Un | -0.591 |
Max | -0.582 |
Token | â |
Token ID | 22940 |
Feature activation | +6.41e-02 |
Pos logit contributions | |
Un | +0.110 |
Jake | +0.101 |
Lisa | +0.098 |
Don | +0.096 |
Enough | +0.094 |
Neg logit contributions | |
up | -0.218 |
inside | -0.215 |
pe | -0.208 |
away | -0.206 |
in | -0.206 |
Token | € |
Token ID | 26391 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
le | +0.141 |
aches | +0.139 |
away | +0.135 |
ered | +0.134 |
up | +0.129 |
Neg logit contributions | |
She | -0.093 |
Un | -0.091 |
They | -0.088 |
Did | -0.088 |
Max | -0.086 |
Token | |
Token ID | 220 |
Feature activation | -4.56e-03 |
Pos logit contributions | |
aches | +0.086 |
- | +0.081 |
up | +0.078 |
le | +0.078 |
ered | +0.077 |
Neg logit contributions | |
Deal | -0.056 |
Un | -0.054 |
She | -0.053 |
Oh | -0.053 |
Well | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | -8.46e-02 |
Pos logit contributions | |
Un | +0.006 |
Don | +0.006 |
Oh | +0.006 |
Really | +0.006 |
Well | +0.006 |
Neg logit contributions | |
up | -0.010 |
aches | -0.009 |
- | -0.009 |
inside | -0.009 |
ered | -0.009 |
Token | his |
Token ID | 465 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
She | +0.004 |
They | +0.004 |
Not | +0.004 |
Something | +0.004 |
Don | +0.004 |
Neg logit contributions | |
aches | -0.002 |
le | -0.001 |
ered | -0.001 |
ig | -0.001 |
island | -0.001 |
Token | head |
Token ID | 1182 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
le | +0.255 |
aches | +0.247 |
ered | +0.235 |
ling | +0.218 |
away | +0.217 |
Neg logit contributions | |
She | -0.323 |
They | -0.316 |
He | -0.305 |
Deal | -0.305 |
Un | -0.304 |
Token | out |
Token ID | 503 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
aches | +0.306 |
le | +0.279 |
ered | +0.274 |
ck | +0.259 |
away | +0.243 |
Neg logit contributions | |
She | -0.570 |
They | -0.543 |
He | -0.529 |
Un | -0.527 |
Max | -0.518 |
Token | of |
Token ID | 286 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
aches | +0.252 |
le | +0.226 |
ish | +0.213 |
ered | +0.204 |
ck | +0.203 |
Neg logit contributions | |
She | -0.426 |
They | -0.410 |
He | -0.403 |
Un | -0.380 |
Jake | -0.373 |
Token | his |
Token ID | 465 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
le | +0.212 |
aches | +0.210 |
ling | +0.196 |
ish | +0.193 |
an | +0.187 |
Neg logit contributions | |
She | -0.231 |
They | -0.222 |
Jake | -0.218 |
He | -0.216 |
Un | -0.215 |
Token | bur |
Token ID | 4356 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
aches | +0.233 |
le | +0.233 |
ered | +0.219 |
ck | +0.207 |
ling | +0.199 |
Neg logit contributions | |
She | -0.282 |
They | -0.269 |
Deal | -0.265 |
Un | -0.265 |
He | -0.262 |
Token | row |
Token ID | 808 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
aches | +0.416 |
ring | +0.383 |
crown | +0.380 |
prize | +0.366 |
coins | +0.361 |
Neg logit contributions | |
She | -0.608 |
He | -0.589 |
Un | -0.560 |
They | -0.544 |
Deal | -0.533 |
Token | and |
Token ID | 290 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
aches | +0.347 |
ck | +0.333 |
le | +0.330 |
ered | +0.305 |
pe | +0.286 |
Neg logit contributions | |
She | -0.628 |
He | -0.597 |
They | -0.595 |
Max | -0.558 |
Un | -0.553 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
le | +0.096 |
aches | +0.093 |
ling | +0.084 |
ish | +0.084 |
ig | +0.084 |
Neg logit contributions | |
She | -0.106 |
They | -0.103 |
Un | -0.102 |
He | -0.100 |
Jake | -0.098 |
Token | something |
Token ID | 1223 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
aches | +0.190 |
le | +0.186 |
away | +0.170 |
ig | +0.168 |
ered | +0.166 |
Neg logit contributions | |
She | -0.227 |
They | -0.219 |
He | -0.215 |
Un | -0.211 |
Jake | -0.210 |
Token | new |
Token ID | 649 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
aches | +0.247 |
le | +0.242 |
ered | +0.233 |
away | +0.223 |
ig | +0.222 |
Neg logit contributions | |
She | -0.296 |
Un | -0.284 |
They | -0.284 |
He | -0.280 |
Don | -0.276 |
Token | too |
Token ID | 1165 |
Feature activation | +9.95e-02 |
Pos logit contributions | |
le | +0.181 |
aches | +0.178 |
ered | +0.158 |
les | +0.157 |
ish | +0.154 |
Neg logit contributions | |
She | -0.211 |
They | -0.202 |
He | -0.193 |
Un | -0.193 |
Deal | -0.192 |
Token | expensive |
Token ID | 5789 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
aches | +0.193 |
le | +0.191 |
ered | +0.183 |
away | +0.181 |
crown | +0.173 |
Neg logit contributions | |
She | -0.171 |
Un | -0.168 |
They | -0.164 |
Max | -0.160 |
Jake | -0.157 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
aches | +0.332 |
le | +0.314 |
ered | +0.302 |
ck | +0.295 |
crown | +0.272 |
Neg logit contributions | |
She | -0.487 |
They | -0.480 |
He | -0.457 |
Un | -0.439 |
Why | -0.421 |
Token | get |
Token ID | 651 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
le | +0.192 |
ling | +0.178 |
aches | +0.177 |
- | +0.173 |
away | +0.171 |
Neg logit contributions | |
She | -0.207 |
They | -0.203 |
Un | -0.195 |
Max | -0.191 |
He | -0.188 |
Token | the |
Token ID | 262 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
le | +0.222 |
aches | +0.202 |
ig | +0.200 |
an | +0.184 |
ered | +0.182 |
Neg logit contributions | |
They | -0.315 |
She | -0.314 |
He | -0.299 |
Don | -0.298 |
Un | -0.291 |
Token | penny |
Token ID | 26369 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
le | +0.202 |
ered | +0.199 |
aches | +0.194 |
ck | +0.185 |
ling | +0.169 |
Neg logit contributions | |
Oh | -0.248 |
She | -0.247 |
They | -0.247 |
Max | -0.242 |
Un | -0.238 |
Token | back |
Token ID | 736 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
le | +0.382 |
aches | +0.375 |
ered | +0.372 |
ck | +0.349 |
crown | +0.313 |
Neg logit contributions | |
She | -0.716 |
They | -0.677 |
He | -0.664 |
Max | -0.641 |
Un | -0.639 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
le | +0.215 |
aches | +0.207 |
ck | +0.188 |
ered | +0.187 |
ling | +0.172 |
Neg logit contributions | |
She | -0.380 |
They | -0.366 |
He | -0.350 |
Un | -0.327 |
Did | -0.326 |
Token | So |
Token ID | 1406 |
Feature activation | -6.32e-03 |
Pos logit contributions | |
Un | +0.126 |
Deal | +0.115 |
Jake | +0.114 |
Lisa | +0.112 |
Max | +0.107 |
Neg logit contributions | |
up | -0.216 |
away | -0.210 |
pe | -0.207 |
inside | -0.206 |
in | -0.205 |
Token | Helen |
Token ID | 22695 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
She | +0.015 |
They | +0.014 |
Un | +0.014 |
Max | +0.014 |
Jake | +0.014 |
Neg logit contributions | |
aches | -0.008 |
le | -0.008 |
away | -0.008 |
ered | -0.008 |
ling | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
aches | +0.244 |
le | +0.224 |
crown | +0.216 |
ling | +0.213 |
away | +0.213 |
Neg logit contributions | |
She | -0.246 |
They | -0.234 |
Max | -0.229 |
Un | -0.225 |
Well | -0.222 |
Token | children |
Token ID | 1751 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
She | +0.147 |
Un | +0.144 |
They | +0.142 |
Max | +0.141 |
Jake | +0.139 |
Neg logit contributions | |
aches | -0.137 |
le | -0.135 |
away | -0.129 |
ered | -0.126 |
up | -0.124 |
Token | put |
Token ID | 1234 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
aches | +0.264 |
le | +0.254 |
away | +0.231 |
ered | +0.230 |
coins | +0.227 |
Neg logit contributions | |
She | -0.378 |
They | -0.370 |
Did | -0.341 |
He | -0.339 |
Jake | -0.336 |
Token | on |
Token ID | 319 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
aches | +0.253 |
le | +0.237 |
ered | +0.231 |
ck | +0.219 |
ling | +0.204 |
Neg logit contributions | |
She | -0.426 |
They | -0.424 |
Max | -0.395 |
He | -0.391 |
Un | -0.386 |
Token | their |
Token ID | 511 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
aches | +0.256 |
le | +0.254 |
ered | +0.236 |
away | +0.231 |
ling | +0.228 |
Neg logit contributions | |
She | -0.295 |
They | -0.290 |
Un | -0.280 |
Max | -0.278 |
Jake | -0.274 |
Token | bath |
Token ID | 7837 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
le | +0.224 |
aches | +0.219 |
ered | +0.207 |
ling | +0.192 |
ck | +0.192 |
Neg logit contributions | |
They | -0.293 |
She | -0.292 |
Un | -0.284 |
Max | -0.281 |
Deal | -0.280 |
Token | ro |
Token ID | 305 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
aches | +0.523 |
le | +0.502 |
ered | +0.494 |
ck | +0.486 |
pe | +0.468 |
Neg logit contributions | |
She | -0.392 |
They | -0.391 |
Max | -0.364 |
Un | -0.358 |
Jake | -0.358 |
Token | bes |
Token ID | 12636 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
le | +0.573 |
aches | +0.572 |
ered | +0.559 |
ish | +0.548 |
ie | +0.531 |
Neg logit contributions | |
They | -0.465 |
She | -0.455 |
Un | -0.443 |
Max | -0.442 |
Jake | -0.423 |
Token | and |
Token ID | 290 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
aches | +0.321 |
ered | +0.290 |
le | +0.278 |
ck | +0.276 |
crown | +0.252 |
Neg logit contributions | |
She | -0.580 |
They | -0.566 |
Un | -0.538 |
Max | -0.538 |
He | -0.531 |
Token | ran |
Token ID | 4966 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
aches | +0.088 |
le | +0.086 |
ered | +0.081 |
ling | +0.078 |
away | +0.078 |
Neg logit contributions | |
She | -0.099 |
They | -0.098 |
Un | -0.096 |
Max | -0.094 |
Jake | -0.092 |
Token | around |
Token ID | 1088 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
aches | +0.192 |
le | +0.161 |
ered | +0.158 |
ck | +0.150 |
ish | +0.143 |
Neg logit contributions | |
She | -0.329 |
They | -0.324 |
Max | -0.309 |
Un | -0.305 |
He | -0.303 |
Token | in |
Token ID | 287 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
aches | +0.229 |
le | +0.208 |
ered | +0.199 |
ish | +0.181 |
ck | +0.180 |
Neg logit contributions | |
She | -0.374 |
They | -0.368 |
Max | -0.353 |
Un | -0.352 |
Jake | -0.350 |
Token | he |
Token ID | 339 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
le | +0.236 |
aches | +0.230 |
ered | +0.219 |
away | +0.218 |
ling | +0.208 |
Neg logit contributions | |
She | -0.270 |
They | -0.260 |
Un | -0.252 |
Did | -0.249 |
Max | -0.247 |
Token | could |
Token ID | 714 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
aches | +0.241 |
le | +0.236 |
away | +0.228 |
ered | +0.221 |
up | +0.218 |
Neg logit contributions | |
She | -0.200 |
They | -0.191 |
Max | -0.184 |
He | -0.184 |
Did | -0.182 |
Token | finally |
Token ID | 3443 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
le | +0.156 |
ling | +0.148 |
les | +0.146 |
aches | +0.146 |
away | +0.142 |
Neg logit contributions | |
She | -0.191 |
They | -0.179 |
He | -0.175 |
Not | -0.166 |
Something | -0.162 |
Token | create |
Token ID | 2251 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
le | +0.209 |
ling | +0.201 |
- | +0.193 |
an | +0.191 |
les | +0.191 |
Neg logit contributions | |
She | -0.237 |
They | -0.222 |
He | -0.219 |
Not | -0.210 |
Something | -0.206 |
Token | his |
Token ID | 465 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
aches | +0.259 |
le | +0.243 |
ered | +0.225 |
ck | +0.225 |
ling | +0.223 |
Neg logit contributions | |
She | -0.363 |
They | -0.358 |
He | -0.348 |
Jake | -0.341 |
Max | -0.333 |
Token | pearl |
Token ID | 43836 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
le | +0.213 |
aches | +0.204 |
ered | +0.191 |
ling | +0.184 |
ck | +0.183 |
Neg logit contributions | |
She | -0.252 |
Deal | -0.251 |
They | -0.250 |
He | -0.241 |
Jake | -0.239 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
aches | +0.376 |
ered | +0.374 |
le | +0.370 |
ck | +0.368 |
ling | +0.313 |
Neg logit contributions | |
She | -0.641 |
He | -0.624 |
They | -0.612 |
Something | -0.550 |
Maybe | -0.538 |
Token | He |
Token ID | 679 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
Un | +0.146 |
Lisa | +0.129 |
Grand | +0.128 |
Don | +0.127 |
Jake | +0.126 |
Neg logit contributions | |
up | -0.227 |
inside | -0.217 |
away | -0.214 |
home | -0.214 |
pe | -0.213 |
Token | was |
Token ID | 373 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
aches | +0.087 |
le | +0.086 |
away | +0.082 |
ling | +0.078 |
up | +0.077 |
Neg logit contributions | |
She | -0.090 |
They | -0.088 |
Jake | -0.085 |
Max | -0.084 |
Well | -0.084 |
Token | so |
Token ID | 523 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
le | +0.110 |
aches | +0.102 |
les | +0.097 |
ling | +0.094 |
ish | +0.094 |
Neg logit contributions | |
She | -0.110 |
They | -0.102 |
Un | -0.102 |
Jake | -0.101 |
He | -0.101 |
Token | proud |
Token ID | 6613 |
Feature activation | +6.19e-02 |
Pos logit contributions | |
le | +0.039 |
aches | +0.038 |
away | +0.036 |
ered | +0.035 |
up | +0.033 |
Neg logit contributions | |
She | -0.041 |
Un | -0.039 |
They | -0.039 |
Jake | -0.038 |
Deal | -0.038 |
Token | pretend |
Token ID | 16614 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
She | +0.197 |
Un | +0.197 |
Max | +0.191 |
They | +0.190 |
Jake | +0.188 |
Neg logit contributions | |
aches | -0.214 |
le | -0.212 |
away | -0.204 |
ered | -0.200 |
up | -0.199 |
Token | she |
Token ID | 673 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
le | +0.337 |
aches | +0.324 |
ered | +0.319 |
away | +0.312 |
ling | +0.304 |
Neg logit contributions | |
She | -0.471 |
They | -0.465 |
He | -0.436 |
Un | -0.433 |
Jake | -0.426 |
Token | was |
Token ID | 373 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
aches | +0.234 |
le | +0.229 |
away | +0.221 |
ered | +0.217 |
up | +0.211 |
Neg logit contributions | |
She | -0.211 |
They | -0.202 |
Un | -0.200 |
Max | -0.197 |
Jake | -0.195 |
Token | on |
Token ID | 319 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
le | +0.162 |
aches | +0.161 |
ered | +0.147 |
away | +0.146 |
ling | +0.139 |
Neg logit contributions | |
She | -0.236 |
They | -0.226 |
Un | -0.225 |
Jake | -0.222 |
Max | -0.222 |
Token | an |
Token ID | 281 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
aches | +0.176 |
le | +0.174 |
away | +0.163 |
ered | +0.160 |
up | +0.154 |
Neg logit contributions | |
She | -0.264 |
Un | -0.258 |
They | -0.256 |
Max | -0.254 |
Jake | -0.251 |
Token | island |
Token ID | 7022 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
aches | +0.168 |
le | +0.158 |
away | +0.154 |
ling | +0.151 |
ered | +0.150 |
Neg logit contributions | |
She | -0.234 |
Un | -0.232 |
Grand | -0.226 |
Jake | -0.221 |
Max | -0.218 |
Token | . |
Token ID | 13 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
le | +0.404 |
aches | +0.395 |
ck | +0.383 |
ered | +0.379 |
ish | +0.360 |
Neg logit contributions | |
She | -0.693 |
They | -0.658 |
He | -0.629 |
Un | -0.602 |
Lisa | -0.590 |
Token | One |
Token ID | 1881 |
Feature activation | -2.43e-03 |
Pos logit contributions | |
Un | +0.102 |
Deal | +0.091 |
Jake | +0.091 |
Max | +0.086 |
Don | +0.084 |
Neg logit contributions | |
up | -0.235 |
inside | -0.234 |
pe | -0.226 |
in | -0.225 |
home | -0.222 |
Token | day |
Token ID | 1110 |
Feature activation | +9.84e-02 |
Pos logit contributions | |
She | +0.005 |
They | +0.005 |
Un | +0.004 |
Jake | +0.004 |
Max | +0.004 |
Neg logit contributions | |
aches | -0.004 |
le | -0.004 |
away | -0.004 |
ered | -0.004 |
up | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +6.28e-03 |
Pos logit contributions | |
aches | +0.102 |
le | +0.100 |
away | +0.092 |
ered | +0.089 |
up | +0.085 |
Neg logit contributions | |
She | -0.255 |
Un | -0.251 |
They | -0.248 |
Max | -0.247 |
Jake | -0.245 |
Token | she |
Token ID | 673 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
aches | +0.012 |
le | +0.012 |
away | +0.012 |
up | +0.012 |
pe | +0.012 |
Neg logit contributions | |
Un | -0.010 |
Deal | -0.010 |
She | -0.010 |
Max | -0.009 |
They | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
le | +0.281 |
aches | +0.278 |
ered | +0.260 |
ck | +0.249 |
ish | +0.244 |
Neg logit contributions | |
She | -0.475 |
They | -0.460 |
Un | -0.434 |
He | -0.432 |
Grand | -0.415 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
aches | +0.045 |
le | +0.044 |
away | +0.043 |
ered | +0.042 |
up | +0.042 |
Neg logit contributions | |
She | -0.041 |
Un | -0.040 |
Max | -0.039 |
They | -0.039 |
Jake | -0.039 |
Token | my |
Token ID | 1820 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
aches | +0.255 |
le | +0.246 |
ered | +0.235 |
away | +0.228 |
coins | +0.225 |
Neg logit contributions | |
She | -0.212 |
They | -0.200 |
Max | -0.199 |
Un | -0.199 |
He | -0.193 |
Token | put |
Token ID | 1234 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
aches | +0.200 |
le | +0.199 |
ered | +0.182 |
away | +0.182 |
coins | +0.174 |
Neg logit contributions | |
She | -0.226 |
They | -0.219 |
He | -0.209 |
Un | -0.206 |
Max | -0.203 |
Token | the |
Token ID | 262 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
le | +0.230 |
aches | +0.228 |
ered | +0.208 |
ck | +0.200 |
ling | +0.197 |
Neg logit contributions | |
She | -0.319 |
They | -0.309 |
He | -0.307 |
Max | -0.299 |
Un | -0.289 |
Token | seas |
Token ID | 21547 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
le | +0.168 |
ered | +0.167 |
aches | +0.159 |
ck | +0.152 |
ering | +0.140 |
Neg logit contributions | |
Max | -0.210 |
Oh | -0.208 |
She | -0.204 |
Un | -0.202 |
Don | -0.201 |
Token | hell |
Token ID | 12758 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
ered | +0.496 |
pe | +0.466 |
ish | +0.463 |
egg | +0.460 |
island | +0.454 |
Neg logit contributions | |
Max | -0.479 |
She | -0.476 |
He | -0.452 |
They | -0.450 |
Lisa | -0.432 |
Token | in |
Token ID | 287 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
aches | +0.286 |
ered | +0.284 |
le | +0.268 |
ck | +0.245 |
egg | +0.221 |
Neg logit contributions | |
She | -0.563 |
They | -0.523 |
He | -0.521 |
Max | -0.508 |
Un | -0.483 |
Token | his |
Token ID | 465 |
Feature activation | +9.86e-02 |
Pos logit contributions | |
aches | +0.129 |
le | +0.127 |
away | +0.123 |
ered | +0.118 |
up | +0.116 |
Neg logit contributions | |
She | -0.189 |
Un | -0.187 |
They | -0.183 |
Max | -0.183 |
Jake | -0.181 |
Token | pocket |
Token ID | 10000 |
Feature activation | +2.37e-01 |
Pos logit contributions | |
le | +0.163 |
aches | +0.154 |
ered | +0.150 |
ling | +0.141 |
ck | +0.139 |
Neg logit contributions | |
She | -0.202 |
They | -0.199 |
Un | -0.196 |
He | -0.193 |
Max | -0.189 |
Token | and |
Token ID | 290 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
le | +0.366 |
aches | +0.356 |
ered | +0.352 |
ck | +0.341 |
ish | +0.294 |
Neg logit contributions | |
She | -0.509 |
They | -0.490 |
He | -0.479 |
Un | -0.442 |
Grand | -0.432 |
Token | man |
Token ID | 582 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
She | +0.226 |
They | +0.221 |
Max | +0.217 |
Un | +0.214 |
Jake | +0.214 |
Neg logit contributions | |
aches | -0.085 |
le | -0.081 |
away | -0.070 |
ered | -0.067 |
ling | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +8.07e-02 |
Pos logit contributions | |
aches | +0.383 |
le | +0.352 |
away | +0.326 |
ered | +0.322 |
crown | +0.316 |
Neg logit contributions | |
She | -0.502 |
They | -0.483 |
He | -0.474 |
Did | -0.458 |
Un | -0.457 |
Token | carefully |
Token ID | 7773 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
le | +0.147 |
aches | +0.147 |
away | +0.138 |
ered | +0.137 |
up | +0.133 |
Neg logit contributions | |
She | -0.146 |
Un | -0.141 |
They | -0.140 |
Max | -0.137 |
Jake | -0.135 |
Token | brought |
Token ID | 3181 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
aches | +0.205 |
le | +0.205 |
an | +0.195 |
ish | +0.192 |
ling | +0.192 |
Neg logit contributions | |
She | -0.275 |
They | -0.270 |
Un | -0.264 |
He | -0.252 |
Jake | -0.249 |
Token | the |
Token ID | 262 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
aches | +0.229 |
le | +0.221 |
ling | +0.202 |
ered | +0.200 |
ig | +0.199 |
Neg logit contributions | |
She | -0.286 |
They | -0.279 |
Max | -0.263 |
He | -0.262 |
Un | -0.262 |
Token | salt |
Token ID | 8268 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
ered | +0.186 |
aches | +0.180 |
le | +0.177 |
ck | +0.168 |
eking | +0.151 |
Neg logit contributions | |
Max | -0.222 |
She | -0.221 |
Oh | -0.217 |
Un | -0.216 |
They | -0.215 |
Token | back |
Token ID | 736 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ered | +0.400 |
aches | +0.377 |
ck | +0.371 |
le | +0.350 |
eking | +0.340 |
Neg logit contributions | |
They | -0.658 |
She | -0.652 |
He | -0.621 |
Un | -0.591 |
Max | -0.555 |
Token | to |
Token ID | 284 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
le | +0.229 |
aches | +0.213 |
ered | +0.202 |
ck | +0.202 |
ling | +0.187 |
Neg logit contributions | |
She | -0.353 |
They | -0.347 |
He | -0.325 |
Un | -0.314 |
Sure | -0.302 |
Token | her |
Token ID | 607 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
le | +0.177 |
aches | +0.176 |
away | +0.164 |
ered | +0.163 |
ling | +0.158 |
Neg logit contributions | |
She | -0.236 |
Un | -0.232 |
They | -0.229 |
Max | -0.224 |
Jake | -0.221 |
Token | mother |
Token ID | 2802 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
le | +0.165 |
aches | +0.162 |
ered | +0.155 |
ling | +0.144 |
ck | +0.141 |
Neg logit contributions | |
She | -0.222 |
They | -0.218 |
Un | -0.209 |
He | -0.205 |
Max | -0.203 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
aches | +0.235 |
ered | +0.213 |
le | +0.211 |
ck | +0.199 |
coins | +0.189 |
Neg logit contributions | |
She | -0.417 |
They | -0.399 |
He | -0.382 |
Un | -0.373 |
Deal | -0.357 |
Token | ready |
Token ID | 3492 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
She | +0.103 |
Un | +0.101 |
They | +0.099 |
Max | +0.097 |
Don | +0.096 |
Neg logit contributions | |
aches | -0.079 |
le | -0.079 |
ered | -0.072 |
away | -0.071 |
ling | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
le | +0.211 |
aches | +0.210 |
ered | +0.186 |
ck | +0.184 |
ish | +0.182 |
Neg logit contributions | |
She | -0.359 |
They | -0.350 |
He | -0.332 |
Un | -0.328 |
Jake | -0.320 |
Token | he |
Token ID | 339 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
aches | +0.078 |
le | +0.078 |
away | +0.073 |
ered | +0.073 |
up | +0.069 |
Neg logit contributions | |
She | -0.088 |
They | -0.085 |
Un | -0.084 |
Max | -0.083 |
Jake | -0.082 |
Token | left |
Token ID | 1364 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
aches | +0.169 |
le | +0.165 |
away | +0.156 |
ered | +0.152 |
ling | +0.147 |
Neg logit contributions | |
She | -0.182 |
They | -0.175 |
Un | -0.169 |
He | -0.168 |
Jake | -0.167 |
Token | his |
Token ID | 465 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
le | +0.218 |
aches | +0.211 |
ish | +0.195 |
ck | +0.193 |
ered | +0.188 |
Neg logit contributions | |
She | -0.287 |
He | -0.274 |
They | -0.271 |
Jake | -0.260 |
Jake | -0.252 |
Token | wallet |
Token ID | 13008 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
le | +0.230 |
aches | +0.217 |
ered | +0.212 |
ck | +0.195 |
ling | +0.192 |
Neg logit contributions | |
She | -0.261 |
Un | -0.254 |
Deal | -0.253 |
Max | -0.253 |
They | -0.252 |
Token | on |
Token ID | 319 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
aches | +0.389 |
le | +0.372 |
ck | +0.362 |
ered | +0.356 |
pe | +0.295 |
Neg logit contributions | |
She | -0.689 |
He | -0.665 |
Max | -0.663 |
Un | -0.631 |
Grand | -0.622 |
Token | the |
Token ID | 262 |
Feature activation | +9.46e-02 |
Pos logit contributions | |
aches | +0.186 |
le | +0.179 |
ered | +0.166 |
ling | +0.161 |
away | +0.160 |
Neg logit contributions | |
She | -0.274 |
They | -0.265 |
Un | -0.264 |
Max | -0.260 |
Jake | -0.258 |
Token | table |
Token ID | 3084 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
le | +0.158 |
aches | +0.153 |
ered | +0.149 |
ck | +0.138 |
ling | +0.135 |
Neg logit contributions | |
She | -0.188 |
Un | -0.187 |
Max | -0.183 |
They | -0.181 |
He | -0.177 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
aches | +0.313 |
le | +0.302 |
ered | +0.289 |
ck | +0.286 |
ling | +0.257 |
Neg logit contributions | |
She | -0.563 |
They | -0.542 |
He | -0.528 |
Un | -0.514 |
Jake | -0.502 |
Token | But |
Token ID | 887 |
Feature activation | -4.03e-03 |
Pos logit contributions | |
Un | +0.131 |
Lisa | +0.120 |
Jake | +0.118 |
Max | +0.114 |
Don | +0.114 |
Neg logit contributions | |
inside | -0.232 |
up | -0.232 |
pe | -0.226 |
home | -0.219 |
in | -0.218 |
Token | skills |
Token ID | 4678 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
aches | +0.335 |
le | +0.318 |
ered | +0.305 |
away | +0.295 |
up | +0.289 |
Neg logit contributions | |
She | -0.534 |
They | -0.518 |
He | -0.498 |
Max | -0.498 |
Un | -0.494 |
Token | to |
Token ID | 284 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
aches | +0.252 |
le | +0.250 |
ered | +0.223 |
away | +0.216 |
pe | +0.212 |
Neg logit contributions | |
She | -0.549 |
They | -0.528 |
Max | -0.514 |
Un | -0.513 |
He | -0.512 |
Token | climb |
Token ID | 12080 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
le | +0.198 |
aches | +0.183 |
ling | +0.182 |
ig | +0.178 |
- | +0.178 |
Neg logit contributions | |
She | -0.255 |
They | -0.246 |
Un | -0.239 |
Max | -0.234 |
He | -0.234 |
Token | up |
Token ID | 510 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
aches | +0.268 |
le | +0.257 |
ered | +0.236 |
ck | +0.225 |
ling | +0.209 |
Neg logit contributions | |
She | -0.578 |
They | -0.556 |
He | -0.541 |
Max | -0.531 |
Un | -0.527 |
Token | the |
Token ID | 262 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
le | +0.243 |
aches | +0.242 |
ered | +0.216 |
ck | +0.212 |
ish | +0.210 |
Neg logit contributions | |
She | -0.438 |
They | -0.419 |
He | -0.407 |
Un | -0.392 |
Max | -0.387 |
Token | mast |
Token ID | 17288 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
le | +0.177 |
aches | +0.174 |
ered | +0.171 |
ck | +0.163 |
away | +0.148 |
Neg logit contributions | |
She | -0.265 |
They | -0.249 |
He | -0.248 |
Max | -0.247 |
Un | -0.242 |
Token | and |
Token ID | 290 |
Feature activation | +6.27e-02 |
Pos logit contributions | |
aches | +0.389 |
le | +0.374 |
ered | +0.373 |
ck | +0.359 |
pe | +0.337 |
Neg logit contributions | |
She | -0.680 |
They | -0.667 |
He | -0.649 |
Un | -0.609 |
Max | -0.582 |
Token | jump |
Token ID | 4391 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
le | +0.104 |
aches | +0.103 |
- | +0.093 |
ling | +0.093 |
ered | +0.091 |
Neg logit contributions | |
She | -0.127 |
They | -0.125 |
He | -0.121 |
Un | -0.116 |
Don | -0.115 |
Token | onto |
Token ID | 4291 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
aches | +0.195 |
le | +0.170 |
ered | +0.156 |
ck | +0.153 |
ish | +0.144 |
Neg logit contributions | |
She | -0.486 |
They | -0.481 |
He | -0.468 |
Max | -0.453 |
Jake | -0.436 |
Token | the |
Token ID | 262 |
Feature activation | +9.67e-02 |
Pos logit contributions | |
aches | +0.188 |
le | +0.184 |
ered | +0.169 |
ling | +0.168 |
ish | +0.159 |
Neg logit contributions | |
She | -0.276 |
They | -0.266 |
Un | -0.258 |
Max | -0.257 |
He | -0.256 |
Token | pirate |
Token ID | 25868 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
ered | +0.157 |
aches | +0.153 |
le | +0.149 |
ck | +0.137 |
ie | +0.130 |
Neg logit contributions | |
Max | -0.183 |
He | -0.181 |
They | -0.181 |
Un | -0.179 |
Grand | -0.177 |
Token | to |
Token ID | 284 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
She | +0.039 |
Max | +0.039 |
Un | +0.039 |
Jake | +0.038 |
They | +0.038 |
Neg logit contributions | |
aches | -0.032 |
le | -0.030 |
away | -0.028 |
ered | -0.028 |
up | -0.027 |
Token | work |
Token ID | 670 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
le | +0.244 |
ling | +0.231 |
- | +0.222 |
ig | +0.222 |
aches | +0.221 |
Neg logit contributions | |
She | -0.288 |
They | -0.280 |
Un | -0.262 |
Max | -0.260 |
He | -0.259 |
Token | gathering |
Token ID | 11228 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
aches | +0.271 |
le | +0.264 |
ered | +0.235 |
ck | +0.231 |
pe | +0.228 |
Neg logit contributions | |
She | -0.533 |
They | -0.522 |
Jake | -0.504 |
Un | -0.503 |
Max | -0.503 |
Token | up |
Token ID | 510 |
Feature activation | +1.89e-01 |
Pos logit contributions | |
aches | +0.286 |
ck | +0.274 |
le | +0.271 |
ered | +0.253 |
an | +0.250 |
Neg logit contributions | |
They | -0.501 |
She | -0.491 |
He | -0.463 |
Something | -0.440 |
Max | -0.425 |
Token | the |
Token ID | 262 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
le | +0.269 |
aches | +0.264 |
ered | +0.248 |
ck | +0.247 |
an | +0.241 |
Neg logit contributions | |
She | -0.412 |
They | -0.410 |
He | -0.382 |
Un | -0.370 |
Jake | -0.362 |
Token | sand |
Token ID | 6450 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
ered | +0.212 |
ck | +0.212 |
aches | +0.212 |
le | +0.207 |
away | +0.175 |
Neg logit contributions | |
She | -0.251 |
They | -0.243 |
Max | -0.236 |
Un | -0.236 |
Oh | -0.235 |
Token | and |
Token ID | 290 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
ck | +0.393 |
le | +0.387 |
ered | +0.382 |
aches | +0.379 |
away | +0.322 |
Neg logit contributions | |
They | -0.665 |
She | -0.664 |
He | -0.632 |
Un | -0.595 |
Jake | -0.573 |
Token | carrying |
Token ID | 6872 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
le | +0.105 |
aches | +0.100 |
- | +0.090 |
ered | +0.090 |
ling | +0.090 |
Neg logit contributions | |
She | -0.139 |
They | -0.137 |
Un | -0.131 |
He | -0.129 |
Jake | -0.129 |
Token | it |
Token ID | 340 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
aches | +0.242 |
le | +0.235 |
ling | +0.225 |
ck | +0.221 |
an | +0.219 |
Neg logit contributions | |
They | -0.323 |
She | -0.323 |
He | -0.300 |
Something | -0.299 |
Jake | -0.299 |
Token | to |
Token ID | 284 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
le | +0.258 |
aches | +0.240 |
ck | +0.218 |
les | +0.207 |
ered | +0.205 |
Neg logit contributions | |
She | -0.423 |
They | -0.413 |
He | -0.393 |
Max | -0.378 |
Something | -0.372 |
Token | the |
Token ID | 262 |
Feature activation | +9.38e-02 |
Pos logit contributions | |
aches | +0.160 |
le | +0.159 |
away | +0.149 |
ered | +0.147 |
up | +0.143 |
Neg logit contributions | |
She | -0.209 |
Un | -0.205 |
They | -0.203 |
Max | -0.200 |
Jake | -0.198 |
Token | the |
Token ID | 262 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
aches | +0.098 |
ered | +0.086 |
ck | +0.086 |
ish | +0.086 |
le | +0.084 |
Neg logit contributions | |
She | -0.142 |
They | -0.132 |
Un | -0.125 |
He | -0.124 |
Max | -0.124 |
Token | lake |
Token ID | 13546 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
le | +0.210 |
aches | +0.205 |
ered | +0.201 |
away | +0.184 |
ck | +0.181 |
Neg logit contributions | |
She | -0.282 |
Un | -0.275 |
They | -0.275 |
Max | -0.273 |
Deal | -0.266 |
Token | was |
Token ID | 373 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
aches | +0.391 |
le | +0.370 |
ered | +0.349 |
ck | +0.344 |
crown | +0.321 |
Neg logit contributions | |
They | -0.659 |
She | -0.651 |
He | -0.618 |
Did | -0.575 |
Don | -0.569 |
Token | a |
Token ID | 257 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
le | +0.179 |
aches | +0.176 |
ish | +0.160 |
ig | +0.158 |
ered | +0.158 |
Neg logit contributions | |
She | -0.188 |
They | -0.187 |
Un | -0.183 |
He | -0.181 |
Don | -0.178 |
Token | tiny |
Token ID | 7009 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
ered | +0.177 |
aches | +0.177 |
le | +0.170 |
ck | +0.151 |
les | +0.146 |
Neg logit contributions | |
Un | -0.185 |
They | -0.179 |
Max | -0.177 |
She | -0.176 |
Grand | -0.172 |
Token | island |
Token ID | 7022 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
aches | +0.353 |
ered | +0.328 |
eking | +0.288 |
ck | +0.280 |
le | +0.257 |
Neg logit contributions | |
They | -0.375 |
She | -0.360 |
He | -0.356 |
Today | -0.317 |
Thank | -0.314 |
Token | and |
Token ID | 290 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
aches | +0.461 |
ered | +0.455 |
le | +0.440 |
ck | +0.434 |
ig | +0.400 |
Neg logit contributions | |
She | -0.616 |
They | -0.611 |
He | -0.581 |
Un | -0.548 |
Don | -0.512 |
Token | M |
Token ID | 337 |
Feature activation | +8.62e-02 |
Pos logit contributions | |
le | +0.102 |
aches | +0.100 |
away | +0.093 |
ered | +0.091 |
ling | +0.090 |
Neg logit contributions | |
She | -0.113 |
They | -0.110 |
Un | -0.105 |
Don | -0.104 |
He | -0.103 |
Token | illy |
Token ID | 6548 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
le | +0.179 |
aches | +0.173 |
ered | +0.165 |
ie | +0.164 |
- | +0.164 |
Neg logit contributions | |
Un | -0.137 |
Max | -0.135 |
Don | -0.133 |
She | -0.130 |
They | -0.128 |
Token | wanted |
Token ID | 2227 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
aches | +0.280 |
le | +0.269 |
away | +0.265 |
ered | +0.253 |
up | +0.244 |
Neg logit contributions | |
She | -0.268 |
They | -0.265 |
He | -0.247 |
Max | -0.240 |
Un | -0.239 |
Token | to |
Token ID | 284 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
le | +0.158 |
aches | +0.150 |
ling | +0.147 |
les | +0.139 |
ish | +0.137 |
Neg logit contributions | |
She | -0.186 |
They | -0.176 |
Un | -0.170 |
He | -0.165 |
Max | -0.164 |
Token | must |
Token ID | 1276 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
aches | +0.196 |
le | +0.194 |
ling | +0.184 |
- | +0.184 |
ered | +0.181 |
Neg logit contributions | |
They | -0.168 |
She | -0.168 |
Did | -0.154 |
Not | -0.152 |
Deal | -0.151 |
Token | not |
Token ID | 407 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
le | +0.294 |
ling | +0.289 |
les | +0.277 |
ish | +0.269 |
ishing | +0.267 |
Neg logit contributions | |
They | -0.294 |
She | -0.287 |
Not | -0.265 |
Something | -0.259 |
Deal | -0.258 |
Token | rush |
Token ID | 10484 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
le | +0.173 |
les | +0.169 |
ig | +0.166 |
aches | +0.165 |
ling | +0.164 |
Neg logit contributions | |
They | -0.195 |
She | -0.190 |
Un | -0.188 |
Max | -0.185 |
He | -0.177 |
Token | into |
Token ID | 656 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
aches | +0.234 |
le | +0.214 |
ered | +0.199 |
crown | +0.199 |
ck | +0.189 |
Neg logit contributions | |
She | -0.454 |
They | -0.451 |
Max | -0.429 |
He | -0.423 |
Un | -0.423 |
Token | the |
Token ID | 262 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
aches | +0.238 |
le | +0.233 |
ish | +0.216 |
ling | +0.212 |
ered | +0.212 |
Neg logit contributions | |
She | -0.261 |
They | -0.261 |
Un | -0.251 |
Jake | -0.250 |
Max | -0.247 |
Token | wreck |
Token ID | 15228 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
aches | +0.198 |
le | +0.193 |
ered | +0.187 |
ck | +0.177 |
ling | +0.165 |
Neg logit contributions | |
She | -0.220 |
They | -0.214 |
Un | -0.213 |
Max | -0.208 |
He | -0.201 |
Token | ." |
Token ID | 526 |
Feature activation | -9.03e-02 |
Pos logit contributions | |
aches | +0.416 |
ered | +0.389 |
ck | +0.386 |
le | +0.382 |
crown | +0.370 |
Neg logit contributions | |
She | -0.638 |
They | -0.634 |
He | -0.614 |
Un | -0.581 |
Jake | -0.573 |
Token | They |
Token ID | 1119 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
Un | +0.129 |
Deal | +0.123 |
Jake | +0.114 |
Max | +0.113 |
Grand | +0.112 |
Neg logit contributions | |
away | -0.185 |
up | -0.178 |
pe | -0.172 |
aches | -0.172 |
le | -0.170 |
Token | stopped |
Token ID | 5025 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
le | +0.058 |
aches | +0.058 |
- | +0.055 |
ling | +0.053 |
ig | +0.052 |
Neg logit contributions | |
She | -0.075 |
They | -0.073 |
Un | -0.069 |
He | -0.069 |
Max | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
aches | +0.135 |
le | +0.134 |
crown | +0.113 |
ered | +0.112 |
les | +0.112 |
Neg logit contributions | |
They | -0.169 |
She | -0.166 |
He | -0.157 |
Un | -0.156 |
Don | -0.150 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
aches | +0.044 |
le | +0.042 |
ling | +0.039 |
ig | +0.038 |
- | +0.038 |
Neg logit contributions | |
She | -0.042 |
They | -0.041 |
Un | -0.040 |
Don | -0.039 |
Max | -0.039 |
Token | � |
Token ID | 241 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
aches | +0.256 |
le | +0.244 |
ered | +0.241 |
up | +0.240 |
away | +0.238 |
Neg logit contributions | |
Did | -0.287 |
Un | -0.285 |
She | -0.284 |
Deal | -0.283 |
Max | -0.280 |
Token | I |
Token ID | 40 |
Feature activation | +7.79e-02 |
Pos logit contributions | |
aches | +0.229 |
up | +0.225 |
le | +0.224 |
- | +0.215 |
pe | +0.214 |
Neg logit contributions | |
Deal | -0.107 |
Un | -0.100 |
Max | -0.094 |
She | -0.092 |
Jake | -0.092 |
Token | can |
Token ID | 460 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
aches | +0.172 |
le | +0.165 |
coins | +0.161 |
crown | +0.158 |
ered | +0.158 |
Neg logit contributions | |
She | -0.113 |
They | -0.105 |
Did | -0.102 |
Don | -0.099 |
He | -0.099 |
Token | keep |
Token ID | 1394 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
aches | +0.204 |
le | +0.204 |
away | +0.202 |
ling | +0.190 |
coins | +0.187 |
Neg logit contributions | |
She | -0.235 |
They | -0.220 |
He | -0.210 |
Not | -0.208 |
Max | -0.207 |
Token | your |
Token ID | 534 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
le | +0.254 |
aches | +0.241 |
ling | +0.219 |
an | +0.219 |
ered | +0.218 |
Neg logit contributions | |
She | -0.255 |
Un | -0.246 |
Max | -0.244 |
They | -0.243 |
Tyler | -0.240 |
Token | pond |
Token ID | 16723 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
le | +0.267 |
aches | +0.261 |
ered | +0.251 |
ck | +0.236 |
ling | +0.226 |
Neg logit contributions | |
Deal | -0.309 |
She | -0.302 |
They | -0.299 |
He | -0.286 |
Max | -0.283 |
Token | if |
Token ID | 611 |
Feature activation | +7.92e-02 |
Pos logit contributions | |
ered | +0.382 |
le | +0.381 |
aches | +0.377 |
ck | +0.370 |
pe | +0.343 |
Neg logit contributions | |
She | -0.622 |
They | -0.618 |
He | -0.590 |
Max | -0.587 |
Un | -0.587 |
Token | you |
Token ID | 345 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
le | +0.173 |
aches | +0.162 |
ling | +0.159 |
away | +0.155 |
ered | +0.154 |
Neg logit contributions | |
She | -0.121 |
They | -0.115 |
Un | -0.114 |
Max | -0.111 |
He | -0.109 |
Token | give |
Token ID | 1577 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
aches | +0.243 |
le | +0.231 |
ling | +0.229 |
ered | +0.228 |
crown | +0.226 |
Neg logit contributions | |
She | -0.181 |
They | -0.181 |
He | -0.166 |
Not | -0.165 |
Did | -0.161 |
Token | me |
Token ID | 502 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
aches | +0.239 |
le | +0.220 |
ered | +0.217 |
ig | +0.213 |
ling | +0.211 |
Neg logit contributions | |
She | -0.277 |
They | -0.265 |
Max | -0.261 |
Un | -0.250 |
Don | -0.247 |
Token | something |
Token ID | 1223 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
le | +0.243 |
ered | +0.232 |
aches | +0.230 |
ck | +0.222 |
ling | +0.218 |
Neg logit contributions | |
She | -0.312 |
They | -0.303 |
Un | -0.295 |
Max | -0.287 |
Deal | -0.286 |
Token | the |
Token ID | 262 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
aches | +0.188 |
le | +0.186 |
away | +0.169 |
ered | +0.168 |
ling | +0.163 |
Neg logit contributions | |
She | -0.267 |
They | -0.258 |
Un | -0.255 |
Max | -0.252 |
Jake | -0.250 |
Token | corner |
Token ID | 5228 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
aches | +0.200 |
le | +0.199 |
ered | +0.190 |
ck | +0.178 |
away | +0.172 |
Neg logit contributions | |
She | -0.237 |
They | -0.227 |
Un | -0.226 |
Max | -0.224 |
Jake | -0.219 |
Token | was |
Token ID | 373 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
aches | +0.258 |
le | +0.256 |
away | +0.230 |
ered | +0.223 |
ling | +0.219 |
Neg logit contributions | |
She | -0.504 |
They | -0.485 |
Un | -0.467 |
Jake | -0.465 |
Lisa | -0.461 |
Token | a |
Token ID | 257 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
le | +0.176 |
aches | +0.174 |
ered | +0.157 |
away | +0.154 |
ish | +0.153 |
Neg logit contributions | |
She | -0.221 |
Un | -0.213 |
They | -0.209 |
Max | -0.207 |
Jake | -0.206 |
Token | big |
Token ID | 1263 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
aches | +0.167 |
le | +0.164 |
ered | +0.161 |
away | +0.146 |
ck | +0.144 |
Neg logit contributions | |
She | -0.201 |
Un | -0.200 |
Max | -0.196 |
They | -0.193 |
Jake | -0.190 |
Token | freezer |
Token ID | 30967 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
aches | +0.243 |
ered | +0.225 |
le | +0.219 |
ck | +0.206 |
ling | +0.194 |
Neg logit contributions | |
She | -0.343 |
They | -0.327 |
He | -0.318 |
Un | -0.309 |
Max | -0.308 |
Token | - |
Token ID | 532 |
Feature activation | +8.90e-03 |
Pos logit contributions | |
aches | +0.428 |
le | +0.412 |
ered | +0.396 |
ck | +0.379 |
les | +0.351 |
Neg logit contributions | |
She | -0.682 |
They | -0.647 |
He | -0.638 |
Un | -0.629 |
Grand | -0.601 |
Token | filled |
Token ID | 5901 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
le | +0.014 |
aches | +0.014 |
away | +0.013 |
ered | +0.012 |
ling | +0.012 |
Neg logit contributions | |
She | -0.019 |
They | -0.019 |
Un | -0.018 |
He | -0.018 |
Max | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +9.40e-02 |
Pos logit contributions | |
aches | +0.195 |
le | +0.194 |
ered | +0.171 |
away | +0.168 |
ig | +0.160 |
Neg logit contributions | |
She | -0.432 |
They | -0.423 |
Un | -0.416 |
Max | -0.412 |
Jake | -0.411 |
Token | ice |
Token ID | 4771 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
le | +0.149 |
aches | +0.144 |
ered | +0.132 |
ig | +0.130 |
away | +0.129 |
Neg logit contributions | |
She | -0.200 |
They | -0.196 |
He | -0.186 |
Un | -0.183 |
Don | -0.180 |
Token | cream |
Token ID | 8566 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
ered | +0.285 |
le | +0.273 |
ck | +0.262 |
aches | +0.252 |
ieve | +0.252 |
Neg logit contributions | |
They | -0.567 |
She | -0.554 |
He | -0.541 |
Cookies | -0.528 |
Un | -0.527 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
She | +0.066 |
They | +0.065 |
Un | +0.064 |
He | +0.061 |
Not | +0.061 |
Neg logit contributions | |
le | -0.048 |
aches | -0.046 |
ig | -0.044 |
away | -0.044 |
ered | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
le | +0.220 |
aches | +0.210 |
ling | +0.202 |
ered | +0.194 |
away | +0.191 |
Neg logit contributions | |
She | -0.284 |
They | -0.277 |
Un | -0.269 |
He | -0.264 |
Max | -0.262 |
Token | pick |
Token ID | 2298 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
le | +0.211 |
ling | +0.204 |
les | +0.189 |
aches | +0.189 |
- | +0.188 |
Neg logit contributions | |
She | -0.258 |
They | -0.250 |
Un | -0.239 |
Max | -0.236 |
Something | -0.234 |
Token | up |
Token ID | 510 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
aches | +0.309 |
ered | +0.279 |
le | +0.274 |
eking | +0.273 |
ck | +0.266 |
Neg logit contributions | |
She | -0.492 |
They | -0.462 |
He | -0.444 |
Max | -0.428 |
Oh | -0.414 |
Token | the |
Token ID | 262 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
le | +0.291 |
aches | +0.276 |
ered | +0.267 |
ling | +0.261 |
ck | +0.255 |
Neg logit contributions | |
She | -0.386 |
They | -0.372 |
He | -0.353 |
Un | -0.343 |
Max | -0.337 |
Token | crab |
Token ID | 32202 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
ered | +0.198 |
le | +0.197 |
ck | +0.189 |
aches | +0.187 |
ering | +0.164 |
Neg logit contributions | |
She | -0.263 |
Oh | -0.251 |
Max | -0.248 |
They | -0.245 |
Don | -0.239 |
Token | and |
Token ID | 290 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
ered | +0.444 |
aches | +0.410 |
ck | +0.396 |
ts | +0.394 |
le | +0.390 |
Neg logit contributions | |
She | -0.621 |
They | -0.572 |
He | -0.554 |
Maybe | -0.493 |
Un | -0.493 |
Token | have |
Token ID | 423 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
le | +0.105 |
ling | +0.098 |
aches | +0.096 |
ig | +0.093 |
- | +0.092 |
Neg logit contributions | |
She | -0.129 |
They | -0.126 |
He | -0.120 |
Something | -0.119 |
Un | -0.118 |
Token | a |
Token ID | 257 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
le | +0.156 |
ling | +0.145 |
aches | +0.137 |
les | +0.133 |
ish | +0.132 |
Neg logit contributions | |
She | -0.152 |
They | -0.146 |
Un | -0.144 |
Something | -0.140 |
Don | -0.138 |
Token | closer |
Token ID | 5699 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
le | +0.129 |
ered | +0.118 |
aches | +0.115 |
ling | +0.114 |
les | +0.109 |
Neg logit contributions | |
Un | -0.148 |
She | -0.146 |
Max | -0.144 |
Something | -0.144 |
They | -0.143 |
Token | look |
Token ID | 804 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
le | +0.246 |
aches | +0.243 |
ling | +0.234 |
ered | +0.226 |
ts | +0.222 |
Neg logit contributions | |
She | -0.387 |
They | -0.374 |
He | -0.371 |
Today | -0.342 |
Un | -0.328 |
Token | to |
Token ID | 284 |
Feature activation | +9.85e-02 |
Pos logit contributions | |
le | +0.131 |
aches | +0.118 |
ling | +0.116 |
ish | +0.113 |
up | +0.110 |
Neg logit contributions | |
She | -0.148 |
Un | -0.139 |
They | -0.138 |
Don | -0.137 |
Max | -0.135 |
Token | put |
Token ID | 1234 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
le | +0.151 |
ling | +0.151 |
les | +0.143 |
- | +0.141 |
aches | +0.140 |
Neg logit contributions | |
She | -0.203 |
They | -0.194 |
Deal | -0.184 |
Un | -0.184 |
Something | -0.183 |
Token | on |
Token ID | 319 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
aches | +0.218 |
le | +0.217 |
ered | +0.195 |
ck | +0.194 |
ling | +0.193 |
Neg logit contributions | |
She | -0.382 |
They | -0.360 |
Max | -0.346 |
He | -0.343 |
Un | -0.337 |
Token | her |
Token ID | 607 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
le | +0.225 |
aches | +0.218 |
ered | +0.200 |
ling | +0.200 |
ck | +0.195 |
Neg logit contributions | |
She | -0.271 |
They | -0.254 |
Un | -0.250 |
Max | -0.249 |
Jake | -0.244 |
Token | tooth |
Token ID | 16162 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
le | +0.173 |
aches | +0.163 |
ered | +0.162 |
ling | +0.157 |
ck | +0.152 |
Neg logit contributions | |
She | -0.220 |
Max | -0.209 |
Something | -0.207 |
They | -0.206 |
Deal | -0.205 |
Token | brush |
Token ID | 32680 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
ered | +0.463 |
aches | +0.450 |
ck | +0.450 |
le | +0.419 |
up | +0.402 |
Neg logit contributions | |
She | -0.466 |
Max | -0.417 |
He | -0.414 |
They | -0.398 |
Un | -0.397 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
le | +0.395 |
aches | +0.377 |
ered | +0.375 |
ck | +0.365 |
pe | +0.356 |
Neg logit contributions | |
She | -0.680 |
They | -0.631 |
Max | -0.609 |
He | -0.608 |
Un | -0.604 |
Token | |
Token ID | 220 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
Un | +0.117 |
Jake | +0.112 |
Deal | +0.103 |
Max | +0.102 |
Lisa | +0.100 |
Neg logit contributions | |
up | -0.258 |
inside | -0.247 |
pe | -0.244 |
home | -0.240 |
away | -0.237 |
Token |
|
Token ID | 198 |
Feature activation | -8.76e-02 |
Pos logit contributions | |
Un | +0.016 |
Deal | +0.015 |
Oh | +0.015 |
Jake | +0.014 |
Really | +0.014 |
Neg logit contributions | |
up | -0.024 |
aches | -0.024 |
inside | -0.022 |
ered | -0.022 |
le | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -9.15e-02 |
Pos logit contributions | |
Deal | +0.131 |
Did | +0.127 |
Sorry | +0.125 |
Un | +0.121 |
Cookies | +0.119 |
Neg logit contributions | |
aches | -0.168 |
up | -0.157 |
le | -0.156 |
- | -0.153 |
pe | -0.152 |
Token | One |
Token ID | 3198 |
Feature activation | -9.83e-03 |
Pos logit contributions | |
Did | +0.137 |
Deal | +0.134 |
They | +0.131 |
Cookies | +0.128 |
ses | +0.124 |
Neg logit contributions | |
aches | -0.181 |
le | -0.161 |
- | -0.160 |
away | -0.159 |
ered | -0.158 |
Token | the |
Token ID | 262 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
aches | +0.253 |
le | +0.240 |
ling | +0.227 |
ck | +0.221 |
an | +0.220 |
Neg logit contributions | |
She | -0.331 |
They | -0.324 |
Max | -0.308 |
He | -0.303 |
Jake | -0.302 |
Token | key |
Token ID | 1994 |
Feature activation | +2.68e-01 |
Pos logit contributions | |
aches | +0.200 |
ered | +0.199 |
le | +0.198 |
ck | +0.187 |
ling | +0.171 |
Neg logit contributions | |
She | -0.247 |
Max | -0.246 |
Un | -0.239 |
They | -0.239 |
Oh | -0.238 |
Token | to |
Token ID | 284 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
aches | +0.376 |
le | +0.353 |
ered | +0.351 |
ck | +0.344 |
ling | +0.316 |
Neg logit contributions | |
She | -0.610 |
They | -0.568 |
He | -0.556 |
Un | -0.534 |
Max | -0.533 |
Token | unlock |
Token ID | 12116 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
le | +0.171 |
aches | +0.170 |
away | +0.157 |
ered | +0.155 |
ling | +0.153 |
Neg logit contributions | |
She | -0.251 |
Un | -0.245 |
They | -0.243 |
Max | -0.238 |
Jake | -0.236 |
Token | the |
Token ID | 262 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
aches | +0.306 |
le | +0.280 |
ling | +0.268 |
ish | +0.261 |
ck | +0.260 |
Neg logit contributions | |
She | -0.435 |
They | -0.420 |
He | -0.405 |
Max | -0.395 |
Jake | -0.393 |
Token | fountain |
Token ID | 29420 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
le | +0.187 |
aches | +0.181 |
ered | +0.171 |
ck | +0.169 |
ling | +0.161 |
Neg logit contributions | |
She | -0.222 |
Un | -0.212 |
Max | -0.209 |
Deal | -0.209 |
They | -0.208 |
Token | ! |
Token ID | 0 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
aches | +0.397 |
ered | +0.371 |
le | +0.368 |
ck | +0.367 |
ig | +0.324 |
Neg logit contributions | |
She | -0.682 |
They | -0.670 |
He | -0.629 |
Un | -0.622 |
Max | -0.594 |
Token | |
Token ID | 220 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Un | +0.138 |
Deal | +0.128 |
Jake | +0.128 |
Don | +0.123 |
Lisa | +0.123 |
Neg logit contributions | |
up | -0.201 |
away | -0.196 |
inside | -0.194 |
le | -0.193 |
home | -0.191 |
Token |
|
Token ID | 198 |
Feature activation | -8.80e-02 |
Pos logit contributions | |
Un | +0.019 |
Oh | +0.017 |
Deal | +0.017 |
Don | +0.017 |
Jake | +0.017 |
Neg logit contributions | |
up | -0.032 |
aches | -0.032 |
ered | -0.030 |
le | -0.030 |
inside | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -9.27e-02 |
Pos logit contributions | |
Did | +0.127 |
Deal | +0.126 |
Un | +0.120 |
Sorry | +0.118 |
She | +0.115 |
Neg logit contributions | |
aches | -0.176 |
le | -0.165 |
ered | -0.163 |
up | -0.162 |
- | -0.162 |
Token | Joe |
Token ID | 19585 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Did | +0.132 |
Deal | +0.128 |
She | +0.122 |
They | +0.122 |
Cookies | +0.121 |
Neg logit contributions | |
aches | -0.191 |
le | -0.176 |
ered | -0.174 |
- | -0.174 |
ck | -0.172 |
Token | started |
Token ID | 2067 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
aches | +0.271 |
le | +0.267 |
away | +0.250 |
ered | +0.244 |
up | +0.236 |
Neg logit contributions | |
She | -0.484 |
Un | -0.473 |
They | -0.469 |
Max | -0.466 |
Jake | -0.461 |
Token | to |
Token ID | 284 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
le | +0.179 |
aches | +0.170 |
ered | +0.153 |
ling | +0.149 |
away | +0.147 |
Neg logit contributions | |
She | -0.247 |
They | -0.236 |
He | -0.229 |
Un | -0.226 |
Jake | -0.224 |
Token | run |
Token ID | 1057 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
le | +0.166 |
aches | +0.164 |
- | +0.150 |
ling | +0.145 |
ered | +0.145 |
Neg logit contributions | |
She | -0.240 |
They | -0.228 |
Un | -0.226 |
Jake | -0.223 |
Max | -0.222 |
Token | around |
Token ID | 1088 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
aches | +0.231 |
le | +0.196 |
ered | +0.191 |
crown | +0.175 |
ck | +0.175 |
Neg logit contributions | |
She | -0.453 |
They | -0.441 |
Max | -0.427 |
He | -0.424 |
Un | -0.416 |
Token | the |
Token ID | 262 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
aches | +0.288 |
le | +0.269 |
ered | +0.254 |
ish | +0.246 |
ck | +0.236 |
Neg logit contributions | |
She | -0.386 |
They | -0.376 |
He | -0.362 |
Max | -0.359 |
Jake | -0.357 |
Token | pit |
Token ID | 6028 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
aches | +0.180 |
le | +0.179 |
ered | +0.173 |
ck | +0.160 |
away | +0.157 |
Neg logit contributions | |
She | -0.198 |
They | -0.192 |
Max | -0.191 |
Un | -0.189 |
Jake | -0.187 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
aches | +0.431 |
le | +0.411 |
ered | +0.402 |
ck | +0.388 |
crown | +0.342 |
Neg logit contributions | |
She | -0.686 |
They | -0.678 |
He | -0.650 |
Un | -0.589 |
Maybe | -0.588 |
Token | It |
Token ID | 632 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
Un | +0.125 |
Deal | +0.122 |
Lisa | +0.120 |
Jake | +0.115 |
Grand | +0.114 |
Neg logit contributions | |
up | -0.255 |
inside | -0.234 |
pe | -0.231 |
away | -0.227 |
in | -0.222 |
Token | was |
Token ID | 373 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
aches | +0.233 |
le | +0.221 |
coins | +0.205 |
ered | +0.205 |
crown | +0.200 |
Neg logit contributions | |
She | -0.217 |
Max | -0.211 |
They | -0.206 |
Don | -0.206 |
He | -0.204 |
Token | big |
Token ID | 1263 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
aches | +0.090 |
le | +0.088 |
away | +0.083 |
ered | +0.081 |
up | +0.079 |
Neg logit contributions | |
She | -0.142 |
Un | -0.140 |
They | -0.138 |
Max | -0.137 |
Jake | -0.135 |
Token | but |
Token ID | 475 |
Feature activation | +3.12e-02 |
Pos logit contributions | |
aches | +0.173 |
ered | +0.149 |
le | +0.141 |
ck | +0.136 |
away | +0.124 |
Neg logit contributions | |
She | -0.232 |
They | -0.224 |
He | -0.218 |
Max | -0.209 |
Did | -0.202 |
Token | was |
Token ID | 373 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
aches | +0.347 |
le | +0.324 |
ered | +0.297 |
away | +0.296 |
coins | +0.293 |
Neg logit contributions | |
She | -0.432 |
They | -0.420 |
He | -0.402 |
Jake | -0.396 |
Did | -0.396 |
Token | trying |
Token ID | 2111 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
le | +0.174 |
aches | +0.171 |
ered | +0.154 |
les | +0.150 |
away | +0.148 |
Neg logit contributions | |
She | -0.208 |
Un | -0.199 |
They | -0.197 |
Jake | -0.195 |
Max | -0.193 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
le | +0.239 |
aches | +0.225 |
ered | +0.211 |
ck | +0.205 |
an | +0.205 |
Neg logit contributions | |
She | -0.284 |
They | -0.275 |
Un | -0.264 |
He | -0.262 |
Not | -0.257 |
Token | take |
Token ID | 1011 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
le | +0.178 |
ling | +0.177 |
ig | +0.174 |
les | +0.173 |
lers | +0.172 |
Neg logit contributions | |
She | -0.206 |
Max | -0.198 |
Un | -0.198 |
They | -0.197 |
Something | -0.196 |
Token | the |
Token ID | 262 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
aches | +0.236 |
le | +0.224 |
ling | +0.213 |
ered | +0.205 |
ck | +0.202 |
Neg logit contributions | |
She | -0.276 |
They | -0.265 |
Max | -0.261 |
Un | -0.255 |
He | -0.255 |
Token | coin |
Token ID | 10752 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
ered | +0.196 |
aches | +0.191 |
le | +0.191 |
ck | +0.179 |
ling | +0.165 |
Neg logit contributions | |
Max | -0.217 |
She | -0.208 |
Un | -0.204 |
Oh | -0.204 |
Deal | -0.203 |
Token | from |
Token ID | 422 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
ered | +0.418 |
aches | +0.394 |
ck | +0.368 |
le | +0.362 |
eking | +0.349 |
Neg logit contributions | |
She | -0.633 |
They | -0.607 |
He | -0.601 |
Un | -0.577 |
Max | -0.566 |
Token | the |
Token ID | 262 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
aches | +0.191 |
le | +0.188 |
ered | +0.174 |
away | +0.173 |
ling | +0.169 |
Neg logit contributions | |
She | -0.225 |
Un | -0.220 |
They | -0.217 |
Max | -0.216 |
Jake | -0.213 |
Token | man |
Token ID | 582 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
le | +0.155 |
aches | +0.151 |
ered | +0.150 |
ck | +0.134 |
ling | +0.132 |
Neg logit contributions | |
She | -0.197 |
Un | -0.195 |
Max | -0.194 |
They | -0.190 |
Jake | -0.187 |
Token | 's |
Token ID | 338 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
aches | +0.349 |
ered | +0.325 |
le | +0.296 |
crown | +0.281 |
ck | +0.273 |
Neg logit contributions | |
She | -0.512 |
They | -0.501 |
Un | -0.481 |
He | -0.475 |
Max | -0.466 |
Token | pocket |
Token ID | 10000 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
le | +0.181 |
aches | +0.170 |
ered | +0.162 |
ling | +0.155 |
ck | +0.154 |
Neg logit contributions | |
She | -0.221 |
They | -0.213 |
He | -0.210 |
Un | -0.208 |
Deal | -0.205 |
Token | . |
Token ID | 13 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
They | +0.063 |
She | +0.063 |
Un | +0.062 |
Max | +0.058 |
Grand | +0.058 |
Neg logit contributions | |
aches | -0.036 |
ered | -0.033 |
le | -0.033 |
away | -0.029 |
crown | -0.029 |
Token | They |
Token ID | 1119 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
Johnny | +0.312 |
Kim | +0.307 |
Un | +0.291 |
Jake | +0.287 |
Tyler | +0.282 |
Neg logit contributions | |
inside | -0.569 |
up | -0.560 |
in | -0.552 |
away | -0.535 |
again | -0.516 |
Token | smiled |
Token ID | 13541 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
She | +0.151 |
Un | +0.147 |
They | +0.146 |
Max | +0.145 |
Jake | +0.141 |
Neg logit contributions | |
aches | -0.156 |
le | -0.155 |
away | -0.146 |
ered | -0.145 |
up | -0.144 |
Token | and |
Token ID | 290 |
Feature activation | -9.31e-02 |
Pos logit contributions | |
She | +0.137 |
Un | +0.136 |
Max | +0.134 |
Jake | +0.133 |
They | +0.133 |
Neg logit contributions | |
le | -0.043 |
aches | -0.042 |
away | -0.040 |
up | -0.037 |
ered | -0.036 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
She | +0.180 |
Un | +0.177 |
They | +0.175 |
Max | +0.173 |
Jake | +0.169 |
Neg logit contributions | |
aches | -0.159 |
le | -0.157 |
away | -0.146 |
ered | -0.146 |
up | -0.142 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
She | +0.067 |
They | +0.065 |
Un | +0.064 |
Max | +0.064 |
He | +0.063 |
Neg logit contributions | |
aches | -0.028 |
le | -0.026 |
ered | -0.024 |
away | -0.023 |
pe | -0.022 |
Token | They |
Token ID | 1119 |
Feature activation | -9.12e-02 |
Pos logit contributions | |
Kim | +0.333 |
Johnny | +0.331 |
Jake | +0.307 |
Un | +0.303 |
Tyler | +0.302 |
Neg logit contributions | |
inside | -0.563 |
in | -0.549 |
up | -0.543 |
again | -0.517 |
away | -0.510 |
Token | had |
Token ID | 550 |
Feature activation | -9.02e-02 |
Pos logit contributions | |
She | +0.163 |
Un | +0.160 |
They | +0.157 |
Max | +0.157 |
Jake | +0.154 |
Neg logit contributions | |
aches | -0.166 |
le | -0.164 |
away | -0.157 |
ered | -0.154 |
up | -0.151 |
Token | a |
Token ID | 257 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
She | +0.177 |
Un | +0.175 |
They | +0.173 |
Max | +0.171 |
Jake | +0.168 |
Neg logit contributions | |
le | -0.150 |
aches | -0.149 |
away | -0.139 |
ered | -0.138 |
up | -0.135 |
Token | great |
Token ID | 1049 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
She | +0.165 |
Un | +0.163 |
They | +0.160 |
Max | +0.160 |
Jake | +0.157 |
Neg logit contributions | |
aches | -0.123 |
le | -0.122 |
away | -0.115 |
ered | -0.114 |
up | -0.110 |
Token | adventure |
Token ID | 8855 |
Feature activation | +4.70e-02 |
Pos logit contributions | |
She | +0.027 |
They | +0.027 |
Un | +0.027 |
Max | +0.027 |
Jake | +0.026 |
Neg logit contributions | |
aches | -0.025 |
le | -0.024 |
ered | -0.023 |
away | -0.022 |
ling | -0.022 |
Token | joy |
Token ID | 8716 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
She | +0.138 |
They | +0.138 |
Un | +0.129 |
Don | +0.126 |
Not | +0.125 |
Neg logit contributions | |
le | -0.120 |
aches | -0.115 |
ered | -0.108 |
ling | -0.106 |
away | -0.104 |
Token | ." |
Token ID | 526 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
aches | +0.038 |
le | +0.035 |
ered | +0.034 |
ck | +0.032 |
pe | +0.032 |
Neg logit contributions | |
She | -0.071 |
They | -0.069 |
He | -0.064 |
Max | -0.064 |
Un | -0.063 |
Token | The |
Token ID | 383 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Un | +0.339 |
Deal | +0.332 |
Jake | +0.316 |
Did | +0.308 |
Max | +0.304 |
Neg logit contributions | |
up | -0.410 |
pe | -0.406 |
inside | -0.404 |
away | -0.402 |
again | -0.399 |
Token | tree |
Token ID | 5509 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
She | +0.307 |
Max | +0.288 |
Un | +0.287 |
They | +0.285 |
He | +0.282 |
Neg logit contributions | |
le | -0.136 |
aches | -0.124 |
ered | -0.120 |
away | -0.114 |
ck | -0.108 |
Token | says |
Token ID | 1139 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
aches | +0.095 |
le | +0.092 |
away | +0.090 |
ered | +0.089 |
ishing | +0.085 |
Neg logit contributions | |
She | -0.084 |
They | -0.077 |
He | -0.074 |
Lisa | -0.069 |
Max | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
She | +0.200 |
They | +0.200 |
He | +0.186 |
Their | +0.173 |
Did | +0.172 |
Neg logit contributions | |
aches | -0.140 |
le | -0.122 |
pe | -0.115 |
island | -0.114 |
ered | -0.113 |
Token |
|
Token ID | 198 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
Deal | +0.387 |
Un | +0.377 |
Jake | +0.368 |
Don | +0.352 |
Brother | +0.351 |
Neg logit contributions | |
inside | -0.424 |
again | -0.420 |
pe | -0.417 |
aches | -0.413 |
coins | -0.409 |
Token |
|
Token ID | 198 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
Did | +0.369 |
Deal | +0.364 |
Un | +0.348 |
Well | +0.331 |
Don | +0.328 |
Neg logit contributions | |
up | -0.442 |
aches | -0.442 |
- | -0.433 |
inside | -0.418 |
away | -0.417 |
Token | The |
Token ID | 464 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
Did | +0.348 |
Deal | +0.331 |
She | +0.315 |
Something | +0.308 |
Nice | +0.307 |
Neg logit contributions | |
aches | -0.490 |
away | -0.485 |
- | -0.474 |
le | -0.457 |
ck | -0.452 |
Token | tree |
Token ID | 5509 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
She | +0.316 |
Max | +0.291 |
They | +0.288 |
He | +0.288 |
Un | +0.286 |
Neg logit contributions | |
le | -0.167 |
ered | -0.148 |
aches | -0.145 |
ck | -0.137 |
away | -0.136 |
Token | shakes |
Token ID | 29959 |
Feature activation | -6.64e-03 |
Pos logit contributions | |
aches | +0.072 |
le | +0.071 |
away | +0.068 |
ered | +0.067 |
ling | +0.064 |
Neg logit contributions | |
She | -0.072 |
They | -0.067 |
He | -0.064 |
Lisa | -0.062 |
Well | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
ered | +0.121 |
aches | +0.116 |
le | +0.114 |
ck | +0.110 |
eking | +0.110 |
Neg logit contributions | |
She | -0.175 |
They | -0.172 |
He | -0.164 |
Un | -0.154 |
Max | -0.150 |
Token | fish |
Token ID | 5916 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
She | +0.238 |
Un | +0.232 |
They | +0.231 |
Max | +0.227 |
Jake | +0.223 |
Neg logit contributions | |
le | -0.161 |
aches | -0.160 |
away | -0.150 |
ered | -0.150 |
up | -0.143 |
Token | , |
Token ID | 11 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
ered | +0.059 |
le | +0.056 |
aches | +0.055 |
eking | +0.052 |
ering | +0.049 |
Neg logit contributions | |
They | -0.104 |
She | -0.102 |
He | -0.096 |
Cookies | -0.089 |
Max | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | -9.14e-02 |
Pos logit contributions | |
She | +0.251 |
They | +0.248 |
Un | +0.238 |
He | +0.234 |
Max | +0.233 |
Neg logit contributions | |
le | -0.155 |
ered | -0.150 |
aches | -0.147 |
away | -0.138 |
ling | -0.137 |
Token | flowers |
Token ID | 12734 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
They | +0.196 |
She | +0.196 |
Un | +0.189 |
He | +0.183 |
Max | +0.181 |
Neg logit contributions | |
le | -0.142 |
aches | -0.134 |
ered | -0.129 |
ling | -0.127 |
away | -0.125 |
Token | . |
Token ID | 13 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
ered | +0.079 |
le | +0.075 |
aches | +0.071 |
ering | +0.070 |
ck | +0.070 |
Neg logit contributions | |
They | -0.106 |
She | -0.106 |
He | -0.099 |
Cookies | -0.088 |
Un | -0.085 |
Token | The |
Token ID | 383 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Uh | +0.327 |
liv | +0.319 |
Un | +0.315 |
Jake | +0.314 |
Lyn | +0.309 |
Neg logit contributions | |
inside | -0.509 |
in | -0.490 |
up | -0.481 |
. | -0.475 |
coins | -0.467 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
She | +0.261 |
Un | +0.258 |
Max | +0.252 |
Don | +0.251 |
Sam | +0.248 |
Neg logit contributions | |
le | -0.189 |
ered | -0.182 |
aches | -0.179 |
away | -0.174 |
up | -0.165 |
Token | had |
Token ID | 550 |
Feature activation | -8.61e-02 |
Pos logit contributions | |
aches | +0.058 |
le | +0.058 |
away | +0.057 |
ered | +0.056 |
up | +0.054 |
Neg logit contributions | |
She | -0.054 |
They | -0.053 |
He | -0.050 |
Max | -0.049 |
Un | -0.049 |
Token | more |
Token ID | 517 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
She | +0.175 |
They | +0.171 |
Un | +0.170 |
Max | +0.166 |
Jake | +0.162 |
Neg logit contributions | |
le | -0.139 |
aches | -0.135 |
ered | -0.127 |
away | -0.126 |
ling | -0.124 |
Token | food |
Token ID | 2057 |
Feature activation | +6.68e-02 |
Pos logit contributions | |
She | +0.054 |
They | +0.051 |
He | +0.050 |
And | +0.045 |
Next | +0.045 |
Neg logit contributions | |
ered | -0.034 |
le | -0.034 |
away | -0.033 |
ering | -0.032 |
eking | -0.032 |
Token | hugged |
Token ID | 41130 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
She | +0.004 |
They | +0.004 |
Max | +0.004 |
Well | +0.004 |
Jake | +0.003 |
Neg logit contributions | |
aches | -0.006 |
le | -0.005 |
up | -0.005 |
ig | -0.005 |
away | -0.005 |
Token | them |
Token ID | 606 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
Un | +0.094 |
She | +0.093 |
Jake | +0.092 |
Did | +0.092 |
Max | +0.091 |
Neg logit contributions | |
le | -0.033 |
away | -0.031 |
aches | -0.030 |
up | -0.028 |
- | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
aches | +0.042 |
le | +0.038 |
ered | +0.037 |
ck | +0.033 |
pe | +0.032 |
Neg logit contributions | |
She | -0.088 |
They | -0.087 |
Max | -0.084 |
Un | -0.083 |
He | -0.082 |
Token | kissed |
Token ID | 27946 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
She | +0.139 |
They | +0.135 |
Un | +0.135 |
Max | +0.131 |
Jake | +0.128 |
Neg logit contributions | |
aches | -0.144 |
le | -0.144 |
away | -0.135 |
ered | -0.134 |
ling | -0.131 |
Token | them |
Token ID | 606 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
She | +0.089 |
Un | +0.088 |
They | +0.087 |
Max | +0.086 |
Jake | +0.086 |
Neg logit contributions | |
aches | -0.037 |
le | -0.036 |
away | -0.034 |
ered | -0.032 |
up | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -2.50e-01 |
Pos logit contributions | |
aches | +0.050 |
ered | +0.045 |
le | +0.045 |
pe | +0.040 |
ck | +0.039 |
Neg logit contributions | |
She | -0.094 |
They | -0.093 |
He | -0.087 |
Max | -0.087 |
Un | -0.085 |
Token | They |
Token ID | 1119 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
Kim | +0.340 |
Johnny | +0.323 |
Jake | +0.320 |
Ol | +0.293 |
Max | +0.284 |
Neg logit contributions | |
up | -0.575 |
inside | -0.560 |
more | -0.540 |
away | -0.538 |
again | -0.536 |
Token | said |
Token ID | 531 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
She | +0.176 |
Un | +0.173 |
They | +0.170 |
Max | +0.169 |
Jake | +0.166 |
Neg logit contributions | |
aches | -0.162 |
le | -0.160 |
away | -0.153 |
ered | -0.150 |
up | -0.147 |
Token | they |
Token ID | 484 |
Feature activation | -7.31e-02 |
Pos logit contributions | |
She | +0.137 |
They | +0.131 |
Un | +0.120 |
Max | +0.118 |
Don | +0.114 |
Neg logit contributions | |
aches | -0.237 |
le | -0.234 |
ered | -0.226 |
away | -0.224 |
ish | -0.218 |
Token | were |
Token ID | 547 |
Feature activation | -6.67e-02 |
Pos logit contributions | |
She | +0.123 |
Un | +0.121 |
They | +0.117 |
Max | +0.117 |
Jake | +0.116 |
Neg logit contributions | |
aches | -0.142 |
le | -0.141 |
away | -0.135 |
ered | -0.132 |
up | -0.130 |
Token | proud |
Token ID | 6613 |
Feature activation | -7.68e-02 |
Pos logit contributions | |
She | +0.125 |
They | +0.121 |
Un | +0.121 |
Max | +0.117 |
Deal | +0.117 |
Neg logit contributions | |
le | -0.120 |
aches | -0.117 |
ered | -0.110 |
away | -0.108 |
ling | -0.105 |
Token | listened |
Token ID | 16399 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
She | +0.148 |
They | +0.143 |
Un | +0.143 |
Max | +0.140 |
Well | +0.138 |
Neg logit contributions | |
le | -0.153 |
aches | -0.152 |
away | -0.143 |
up | -0.142 |
- | -0.142 |
Token | and |
Token ID | 290 |
Feature activation | -9.36e-02 |
Pos logit contributions | |
She | +0.089 |
They | +0.089 |
Un | +0.084 |
He | +0.083 |
Max | +0.080 |
Neg logit contributions | |
aches | -0.068 |
le | -0.061 |
ered | -0.059 |
ig | -0.056 |
ish | -0.056 |
Token | asked |
Token ID | 1965 |
Feature activation | -8.22e-02 |
Pos logit contributions | |
She | +0.159 |
They | +0.158 |
Un | +0.154 |
Don | +0.151 |
Max | +0.149 |
Neg logit contributions | |
aches | -0.187 |
le | -0.180 |
ling | -0.167 |
les | -0.165 |
- | -0.163 |
Token | and |
Token ID | 290 |
Feature activation | -8.53e-02 |
Pos logit contributions | |
She | +0.156 |
They | +0.152 |
Un | +0.151 |
Jake | +0.143 |
Max | +0.143 |
Neg logit contributions | |
aches | -0.144 |
le | -0.142 |
away | -0.134 |
ered | -0.133 |
ling | -0.128 |
Token | learned |
Token ID | 4499 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
She | +0.164 |
They | +0.162 |
Un | +0.161 |
Jake | +0.155 |
Lisa | +0.154 |
Neg logit contributions | |
le | -0.152 |
aches | -0.145 |
ling | -0.141 |
les | -0.140 |
ish | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
They | +0.094 |
She | +0.092 |
He | +0.085 |
Un | +0.085 |
Their | +0.084 |
Neg logit contributions | |
le | -0.093 |
aches | -0.088 |
ling | -0.088 |
ig | -0.085 |
ered | -0.083 |
Token | They |
Token ID | 1119 |
Feature activation | -9.16e-02 |
Pos logit contributions | |
Ol | +0.302 |
Max | +0.296 |
Jake | +0.276 |
Emily | +0.271 |
Tyler | +0.268 |
Neg logit contributions | |
inside | -0.551 |
up | -0.550 |
coins | -0.519 |
home | -0.518 |
again | -0.509 |
Token | painted |
Token ID | 13055 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
She | +0.166 |
Un | +0.162 |
They | +0.160 |
Max | +0.157 |
Well | +0.154 |
Neg logit contributions | |
le | -0.165 |
aches | -0.164 |
away | -0.156 |
up | -0.153 |
ered | -0.153 |
Token | their |
Token ID | 511 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
They | +0.035 |
She | +0.035 |
He | +0.032 |
Un | +0.031 |
Lisa | +0.031 |
Neg logit contributions | |
aches | -0.025 |
le | -0.024 |
ered | -0.023 |
ck | -0.022 |
ling | -0.022 |
Token | own |
Token ID | 898 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
They | +0.060 |
She | +0.059 |
Lisa | +0.056 |
Un | +0.056 |
Deal | +0.055 |
Neg logit contributions | |
le | -0.057 |
aches | -0.053 |
ered | -0.051 |
ling | -0.049 |
away | -0.048 |
Token | paintings |
Token ID | 21641 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
le | +0.019 |
aches | +0.016 |
ling | +0.016 |
away | +0.016 |
ered | +0.015 |
Neg logit contributions | |
They | -0.021 |
She | -0.021 |
Lisa | -0.018 |
Not | -0.018 |
Un | -0.018 |
Token | them |
Token ID | 606 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
Un | +0.034 |
She | +0.034 |
Deal | +0.033 |
Did | +0.033 |
Well | +0.033 |
Neg logit contributions | |
le | -0.024 |
away | -0.023 |
aches | -0.023 |
up | -0.023 |
ered | -0.022 |
Token | up |
Token ID | 510 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
aches | +0.044 |
le | +0.039 |
les | +0.034 |
crown | +0.033 |
pearl | +0.033 |
Neg logit contributions | |
She | -0.097 |
They | -0.094 |
He | -0.088 |
Max | -0.083 |
Maybe | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | -7.87e-02 |
Pos logit contributions | |
aches | +0.030 |
le | +0.029 |
ered | +0.026 |
ish | +0.025 |
les | +0.025 |
Neg logit contributions | |
She | -0.045 |
They | -0.043 |
He | -0.041 |
Un | -0.041 |
Max | -0.039 |
Token | hugged |
Token ID | 41130 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
She | +0.133 |
Don | +0.131 |
Un | +0.126 |
They | +0.125 |
Max | +0.124 |
Neg logit contributions | |
aches | -0.161 |
le | -0.154 |
ig | -0.140 |
ered | -0.139 |
ish | -0.137 |
Token | them |
Token ID | 606 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
Deal | +0.083 |
Un | +0.081 |
Did | +0.081 |
She | +0.080 |
Well | +0.080 |
Neg logit contributions | |
le | -0.050 |
away | -0.049 |
aches | -0.046 |
up | -0.046 |
- | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
aches | +0.053 |
ered | +0.049 |
le | +0.048 |
ck | +0.043 |
pe | +0.042 |
Neg logit contributions | |
She | -0.088 |
They | -0.086 |
He | -0.080 |
Max | -0.080 |
Un | -0.077 |
Token | She |
Token ID | 1375 |
Feature activation | -9.80e-02 |
Pos logit contributions | |
Uh | +0.329 |
Jake | +0.321 |
Great | +0.320 |
Thanks | +0.311 |
Em | +0.308 |
Neg logit contributions | |
inside | -0.521 |
, | -0.514 |
up | -0.507 |
in | -0.490 |
out | -0.487 |
Token | was |
Token ID | 373 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
She | +0.181 |
Un | +0.178 |
Max | +0.175 |
They | +0.174 |
Jake | +0.171 |
Neg logit contributions | |
aches | -0.174 |
le | -0.172 |
away | -0.167 |
ered | -0.160 |
up | -0.160 |
Token | angry |
Token ID | 7954 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
She | +0.130 |
Un | +0.122 |
They | +0.117 |
Max | +0.116 |
Grand | +0.116 |
Neg logit contributions | |
le | -0.143 |
aches | -0.132 |
ish | -0.132 |
les | -0.130 |
ig | -0.129 |
Token | and |
Token ID | 290 |
Feature activation | -8.63e-02 |
Pos logit contributions | |
She | +0.060 |
They | +0.058 |
Un | +0.057 |
Max | +0.056 |
He | +0.056 |
Neg logit contributions | |
aches | -0.028 |
le | -0.025 |
ered | -0.024 |
away | -0.023 |
crown | -0.021 |
Token | worried |
Token ID | 7960 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
She | +0.181 |
Un | +0.174 |
They | +0.174 |
Don | +0.170 |
Max | +0.169 |
Neg logit contributions | |
le | -0.137 |
aches | -0.135 |
ered | -0.124 |
away | -0.122 |
up | -0.117 |
Token | with |
Token ID | 351 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
aches | +0.058 |
le | +0.053 |
ered | +0.053 |
away | +0.049 |
ish | +0.049 |
Neg logit contributions | |
She | -0.111 |
They | -0.110 |
Un | -0.108 |
Lisa | -0.106 |
Max | -0.106 |
Token | pupp |
Token ID | 13595 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
She | +0.105 |
They | +0.103 |
Un | +0.103 |
Max | +0.101 |
Jake | +0.100 |
Neg logit contributions | |
aches | -0.078 |
le | -0.077 |
away | -0.072 |
ered | -0.072 |
up | -0.069 |
Token | ets |
Token ID | 1039 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
aches | +0.070 |
le | +0.065 |
away | +0.065 |
ered | +0.064 |
up | +0.063 |
Neg logit contributions | |
She | -0.087 |
They | -0.086 |
Un | -0.083 |
Deal | -0.082 |
Max | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | -6.75e-02 |
Pos logit contributions | |
aches | +0.022 |
le | +0.021 |
ish | +0.019 |
ered | +0.019 |
ck | +0.018 |
Neg logit contributions | |
She | -0.041 |
They | -0.040 |
He | -0.038 |
Un | -0.037 |
Max | -0.037 |
Token | songs |
Token ID | 7259 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
They | +0.134 |
She | +0.133 |
Un | +0.130 |
Max | +0.126 |
Jake | +0.125 |
Neg logit contributions | |
aches | -0.111 |
le | -0.111 |
ered | -0.104 |
away | -0.102 |
ling | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
aches | +0.035 |
le | +0.031 |
ered | +0.031 |
ck | +0.028 |
away | +0.027 |
Neg logit contributions | |
She | -0.068 |
They | -0.067 |
Un | -0.064 |
Max | -0.064 |
He | -0.064 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
Un | +0.352 |
Jake | +0.341 |
liv | +0.340 |
Kim | +0.330 |
Max | +0.329 |
Neg logit contributions | |
inside | -0.478 |
up | -0.473 |
more | -0.470 |
again | -0.466 |
in | -0.460 |
Token | and |
Token ID | 290 |
Feature activation | -9.24e-02 |
Pos logit contributions | |
She | +0.071 |
They | +0.070 |
Max | +0.068 |
Un | +0.067 |
Jake | +0.067 |
Neg logit contributions | |
aches | -0.041 |
le | -0.036 |
ered | -0.036 |
away | -0.034 |
up | -0.032 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
She | +0.096 |
They | +0.091 |
Un | +0.090 |
Max | +0.087 |
Jake | +0.084 |
Neg logit contributions | |
aches | -0.242 |
le | -0.239 |
away | -0.230 |
ered | -0.229 |
up | -0.225 |
Token | laughed |
Token ID | 13818 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
She | +0.060 |
They | +0.059 |
Max | +0.055 |
Well | +0.054 |
Jake | +0.053 |
Neg logit contributions | |
aches | -0.073 |
le | -0.069 |
ered | -0.067 |
away | -0.065 |
up | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | -9.40e-02 |
Pos logit contributions | |
She | +0.068 |
They | +0.067 |
Un | +0.066 |
Max | +0.065 |
Jake | +0.064 |
Neg logit contributions | |
aches | -0.029 |
le | -0.027 |
ered | -0.025 |
away | -0.024 |
pe | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
Deal | +0.355 |
Un | +0.338 |
Jake | +0.327 |
Chris | +0.316 |
Great | +0.313 |
Neg logit contributions | |
again | -0.310 |
pe | -0.307 |
up | -0.307 |
inside | -0.304 |
in | -0.302 |
Token |
|
Token ID | 198 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
Deal | +0.321 |
Un | +0.311 |
Did | +0.309 |
Well | +0.301 |
Jake | +0.287 |
Neg logit contributions | |
- | -0.448 |
up | -0.441 |
away | -0.429 |
le | -0.429 |
aches | -0.428 |
Token | L |
Token ID | 43 |
Feature activation | -9.50e-02 |
Pos logit contributions | |
Deal | +0.320 |
Did | +0.302 |
Not | +0.293 |
Nice | +0.278 |
She | +0.275 |
Neg logit contributions | |
- | -0.493 |
away | -0.486 |
aches | -0.452 |
le | -0.450 |
inside | -0.442 |
Token | ily |
Token ID | 813 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
She | +0.153 |
Un | +0.150 |
They | +0.147 |
Max | +0.146 |
Jake | +0.144 |
Neg logit contributions | |
aches | -0.191 |
le | -0.189 |
away | -0.182 |
ered | -0.179 |
up | -0.176 |
Token | nodded |
Token ID | 14464 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
She | +0.078 |
Un | +0.075 |
Max | +0.074 |
They | +0.073 |
Jake | +0.072 |
Neg logit contributions | |
aches | -0.089 |
le | -0.086 |
away | -0.083 |
ered | -0.081 |
up | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
She | +0.053 |
They | +0.050 |
Un | +0.050 |
He | +0.048 |
Max | +0.048 |
Neg logit contributions | |
aches | -0.029 |
ered | -0.024 |
le | -0.024 |
coins | -0.023 |
away | -0.022 |
Token | She |
Token ID | 1375 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Jake | +0.278 |
Ted | +0.251 |
Un | +0.251 |
Uh | +0.237 |
Grand | +0.237 |
Neg logit contributions | |
up | -0.598 |
inside | -0.582 |
again | -0.562 |
, | -0.546 |
in | -0.540 |
Token | was |
Token ID | 373 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
She | +0.190 |
Un | +0.186 |
Max | +0.183 |
They | +0.182 |
Jake | +0.178 |
Neg logit contributions | |
aches | -0.182 |
le | -0.178 |
away | -0.172 |
ered | -0.168 |
up | -0.164 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
She | +0.120 |
Deal | +0.109 |
Un | +0.108 |
They | +0.105 |
He | +0.104 |
Neg logit contributions | |
le | -0.153 |
aches | -0.140 |
les | -0.140 |
ie | -0.137 |
ish | -0.137 |
Token | , |
Token ID | 11 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
She | +0.042 |
They | +0.040 |
Un | +0.039 |
He | +0.039 |
Max | +0.039 |
Neg logit contributions | |
aches | -0.020 |
le | -0.018 |
ered | -0.017 |
away | -0.016 |
up | -0.015 |
Token | but |
Token ID | 475 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Un | +0.244 |
She | +0.240 |
Jake | +0.239 |
Max | +0.238 |
Deal | +0.237 |
Neg logit contributions | |
aches | -0.181 |
le | -0.179 |
away | -0.177 |
up | -0.169 |
pe | -0.167 |
Token | them |
Token ID | 606 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
She | +0.069 |
They | +0.067 |
Un | +0.065 |
Max | +0.063 |
He | +0.060 |
Neg logit contributions | |
aches | -0.073 |
le | -0.067 |
ered | -0.067 |
ling | -0.065 |
crown | -0.064 |
Token | back |
Token ID | 736 |
Feature activation | -1.58e-03 |
Pos logit contributions | |
aches | +0.038 |
le | +0.036 |
ered | +0.033 |
ling | +0.032 |
ck | +0.030 |
Neg logit contributions | |
She | -0.063 |
Max | -0.063 |
They | -0.062 |
Un | -0.059 |
Sam | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
She | +0.003 |
They | +0.003 |
Un | +0.003 |
He | +0.003 |
Max | +0.003 |
Neg logit contributions | |
aches | -0.003 |
le | -0.003 |
ling | -0.003 |
eking | -0.003 |
les | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -4.19e-02 |
Pos logit contributions | |
Max | +0.068 |
Well | +0.068 |
Did | +0.068 |
Jake | +0.068 |
Deal | +0.067 |
Neg logit contributions | |
away | -0.028 |
aches | -0.027 |
le | -0.026 |
up | -0.026 |
pe | -0.026 |
Token | garden |
Token ID | 11376 |
Feature activation | +9.06e-02 |
Pos logit contributions | |
She | +0.087 |
Un | +0.086 |
They | +0.084 |
Max | +0.084 |
Jake | +0.081 |
Neg logit contributions | |
le | -0.067 |
aches | -0.066 |
ered | -0.061 |
away | -0.060 |
ling | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
aches | +0.144 |
le | +0.140 |
ered | +0.139 |
ck | +0.133 |
ling | +0.118 |
Neg logit contributions | |
She | -0.208 |
They | -0.201 |
He | -0.181 |
Un | -0.177 |
Maybe | -0.171 |
Token | She |
Token ID | 1375 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
Jake | +0.303 |
Max | +0.297 |
Chris | +0.270 |
Ted | +0.265 |
Johnny | +0.253 |
Neg logit contributions | |
inside | -0.594 |
pe | -0.586 |
up | -0.575 |
out | -0.538 |
home | -0.537 |
Token | closes |
Token ID | 20612 |
Feature activation | -7.68e-03 |
Pos logit contributions | |
She | +0.168 |
They | +0.163 |
Un | +0.154 |
Deal | +0.150 |
Max | +0.148 |
Neg logit contributions | |
le | -0.195 |
aches | -0.193 |
away | -0.187 |
ling | -0.184 |
ered | -0.182 |
Token | the |
Token ID | 262 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
She | +0.015 |
They | +0.014 |
Un | +0.014 |
Max | +0.014 |
He | +0.013 |
Neg logit contributions | |
aches | -0.014 |
le | -0.013 |
ered | -0.012 |
ling | -0.012 |
ish | -0.012 |
Token | gate |
Token ID | 8946 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
She | +0.113 |
They | +0.110 |
Un | +0.109 |
Deal | +0.106 |
Max | +0.105 |
Neg logit contributions | |
le | -0.087 |
aches | -0.081 |
ered | -0.077 |
ling | -0.072 |
ck | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
aches | +0.140 |
le | +0.132 |
ish | +0.117 |
ered | +0.115 |
ling | +0.114 |
Neg logit contributions | |
She | -0.229 |
They | -0.217 |
He | -0.204 |
Un | -0.194 |
Maybe | -0.194 |
Token | the |
Token ID | 262 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
aches | +0.009 |
le | +0.009 |
ling | +0.008 |
ered | +0.008 |
crown | +0.008 |
Neg logit contributions | |
She | -0.013 |
They | -0.013 |
He | -0.012 |
Max | -0.012 |
Un | -0.012 |
Token | ladder |
Token ID | 18002 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
They | +0.052 |
She | +0.051 |
He | +0.051 |
Un | +0.048 |
Max | +0.048 |
Neg logit contributions | |
le | -0.050 |
ie | -0.042 |
ered | -0.042 |
aches | -0.041 |
ck | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
aches | +0.167 |
le | +0.159 |
ered | +0.150 |
ck | +0.138 |
les | +0.135 |
Neg logit contributions | |
She | -0.247 |
They | -0.242 |
He | -0.236 |
Un | -0.224 |
Max | -0.213 |
Token | slide |
Token ID | 10649 |
Feature activation | +7.35e-02 |
Pos logit contributions | |
They | +0.138 |
She | +0.137 |
He | +0.132 |
Un | +0.131 |
Something | +0.125 |
Neg logit contributions | |
le | -0.128 |
ling | -0.119 |
an | -0.112 |
ie | -0.112 |
ig | -0.109 |
Token | down |
Token ID | 866 |
Feature activation | -7.77e-03 |
Pos logit contributions | |
aches | +0.095 |
le | +0.095 |
ered | +0.087 |
ck | +0.077 |
crown | +0.077 |
Neg logit contributions | |
She | -0.178 |
They | -0.176 |
He | -0.172 |
Un | -0.168 |
Max | -0.164 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
She | +0.018 |
They | +0.018 |
He | +0.018 |
Un | +0.017 |
Max | +0.016 |
Neg logit contributions | |
aches | -0.011 |
le | -0.011 |
ered | -0.010 |
ish | -0.009 |
ie | -0.009 |
Token | Lily |
Token ID | 20037 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
Un | +0.318 |
Deal | +0.286 |
Jake | +0.283 |
Don | +0.265 |
Kelly | +0.262 |
Neg logit contributions | |
inside | -0.565 |
coins | -0.550 |
up | -0.548 |
in | -0.540 |
away | -0.532 |
Token | and |
Token ID | 290 |
Feature activation | -9.65e-02 |
Pos logit contributions | |
She | +0.101 |
Max | +0.098 |
They | +0.097 |
Jake | +0.097 |
Un | +0.096 |
Neg logit contributions | |
aches | -0.058 |
le | -0.056 |
ered | -0.052 |
away | -0.052 |
ling | -0.048 |
Token | Sam |
Token ID | 3409 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
Deal | +0.088 |
She | +0.086 |
Did | +0.085 |
Well | +0.085 |
Un | +0.084 |
Neg logit contributions | |
away | -0.252 |
aches | -0.250 |
le | -0.247 |
up | -0.246 |
pe | -0.246 |
Token | slid |
Token ID | 27803 |
Feature activation | +2.46e-03 |
Pos logit contributions | |
She | +0.020 |
They | +0.019 |
Max | +0.019 |
Well | +0.019 |
Un | +0.018 |
Neg logit contributions | |
aches | -0.024 |
le | -0.024 |
- | -0.022 |
ered | -0.022 |
up | -0.022 |
Token | down |
Token ID | 866 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
le | +0.003 |
aches | +0.003 |
ered | +0.003 |
ck | +0.003 |
ling | +0.003 |
Neg logit contributions | |
She | -0.006 |
They | -0.006 |
He | -0.006 |
Max | -0.005 |
Un | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
Did | +0.362 |
Un | +0.349 |
Chris | +0.332 |
Deal | +0.332 |
Well | +0.323 |
Neg logit contributions | |
up | -0.425 |
away | -0.425 |
inside | -0.408 |
le | -0.406 |
in | -0.395 |
Token | Anna |
Token ID | 31160 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
Did | +0.326 |
Nice | +0.294 |
Sure | +0.289 |
Tyler | +0.284 |
Chris | +0.282 |
Neg logit contributions | |
away | -0.503 |
le | -0.487 |
aches | -0.470 |
ck | -0.455 |
ling | -0.455 |
Token | cries |
Token ID | 24691 |
Feature activation | -5.42e-02 |
Pos logit contributions | |
She | +0.082 |
They | +0.078 |
Max | +0.075 |
Un | +0.075 |
Deal | +0.074 |
Neg logit contributions | |
aches | -0.070 |
le | -0.066 |
away | -0.062 |
ered | -0.061 |
ling | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
She | +0.143 |
They | +0.138 |
Un | +0.135 |
Max | +0.134 |
He | +0.132 |
Neg logit contributions | |
aches | -0.064 |
le | -0.056 |
ered | -0.054 |
away | -0.049 |
crown | -0.047 |
Token | cries |
Token ID | 24691 |
Feature activation | -5.04e-02 |
Pos logit contributions | |
She | +0.169 |
They | +0.161 |
Un | +0.161 |
Max | +0.157 |
Deal | +0.155 |
Neg logit contributions | |
aches | -0.173 |
le | -0.172 |
away | -0.160 |
ered | -0.158 |
ling | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
She | +0.134 |
They | +0.130 |
Un | +0.128 |
Max | +0.127 |
Did | +0.125 |
Neg logit contributions | |
aches | -0.056 |
le | -0.050 |
ered | -0.047 |
away | -0.044 |
pe | -0.041 |
Token | She |
Token ID | 1375 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
Johnny | +0.298 |
David | +0.287 |
Jake | +0.272 |
Chris | +0.264 |
Kim | +0.255 |
Neg logit contributions | |
up | -0.641 |
in | -0.608 |
inside | -0.598 |
away | -0.579 |
, | -0.576 |
Token | misses |
Token ID | 18297 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
She | +0.193 |
Un | +0.193 |
Max | +0.188 |
They | +0.187 |
Jake | +0.187 |
Neg logit contributions | |
aches | -0.167 |
le | -0.167 |
away | -0.160 |
ered | -0.156 |
up | -0.154 |
Token | Fl |
Token ID | 1610 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
She | +0.065 |
Un | +0.063 |
They | +0.062 |
Max | +0.061 |
Jake | +0.060 |
Neg logit contributions | |
aches | -0.070 |
le | -0.069 |
away | -0.066 |
ered | -0.066 |
up | -0.063 |
Token | uffy |
Token ID | 15352 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
Un | +0.043 |
She | +0.043 |
Don | +0.042 |
Well | +0.042 |
Did | +0.041 |
Neg logit contributions | |
le | -0.055 |
- | -0.052 |
aches | -0.052 |
up | -0.050 |
away | -0.050 |
Token | so |
Token ID | 523 |
Feature activation | -9.72e-02 |
Pos logit contributions | |
aches | +0.038 |
le | +0.035 |
ered | +0.033 |
ck | +0.031 |
pe | +0.030 |
Neg logit contributions | |
She | -0.076 |
They | -0.073 |
Max | -0.072 |
He | -0.070 |
Un | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
She | +0.017 |
They | +0.017 |
Un | +0.017 |
Max | +0.016 |
Jake | +0.016 |
Neg logit contributions | |
aches | -0.008 |
le | -0.007 |
ered | -0.007 |
crown | -0.006 |
away | -0.006 |
Token | They |
Token ID | 1119 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
Un | +0.366 |
Deal | +0.354 |
Tyler | +0.317 |
Chris | +0.317 |
Great | +0.310 |
Neg logit contributions | |
in | -0.485 |
inside | -0.479 |
away | -0.466 |
up | -0.457 |
, | -0.439 |
Token | cried |
Token ID | 16896 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
She | +0.163 |
Un | +0.160 |
They | +0.157 |
Max | +0.157 |
Jake | +0.154 |
Neg logit contributions | |
aches | -0.135 |
le | -0.134 |
away | -0.127 |
ered | -0.125 |
up | -0.122 |
Token | and |
Token ID | 290 |
Feature activation | -8.91e-02 |
Pos logit contributions | |
She | +0.168 |
They | +0.164 |
Un | +0.163 |
Max | +0.161 |
Jake | +0.157 |
Neg logit contributions | |
aches | -0.072 |
le | -0.067 |
ered | -0.062 |
away | -0.060 |
up | -0.056 |
Token | screamed |
Token ID | 25421 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
She | +0.169 |
Un | +0.166 |
They | +0.163 |
Max | +0.162 |
Jake | +0.159 |
Neg logit contributions | |
aches | -0.154 |
le | -0.153 |
away | -0.144 |
ered | -0.142 |
up | -0.140 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
She | +0.126 |
They | +0.124 |
Un | +0.121 |
Max | +0.121 |
Deal | +0.118 |
Neg logit contributions | |
aches | -0.055 |
le | -0.050 |
ered | -0.046 |
away | -0.043 |
pe | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
Deal | +0.447 |
Un | +0.435 |
concession | +0.413 |
Tyler | +0.411 |
Ta | +0.405 |
Neg logit contributions | |
in | -0.413 |
inside | -0.398 |
away | -0.391 |
up | -0.374 |
, | -0.363 |
Token |
|
Token ID | 198 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
Deal | +0.405 |
Okay | +0.401 |
Really | +0.398 |
Did | +0.398 |
Great | +0.396 |
Neg logit contributions | |
inside | -0.389 |
up | -0.376 |
in | -0.369 |
- | -0.357 |
ck | -0.350 |
Token | But |
Token ID | 1537 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
Nice | +0.332 |
Did | +0.324 |
Deal | +0.314 |
She | +0.311 |
Well | +0.291 |
Neg logit contributions | |
away | -0.488 |
inside | -0.486 |
- | -0.479 |
aches | -0.475 |
ck | -0.471 |
Token | their |
Token ID | 511 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
She | +0.304 |
Un | +0.299 |
They | +0.293 |
Max | +0.293 |
Jake | +0.289 |
Neg logit contributions | |
aches | -0.227 |
le | -0.224 |
away | -0.213 |
ered | -0.208 |
up | -0.204 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
She | +0.075 |
They | +0.075 |
Max | +0.073 |
Un | +0.073 |
Jake | +0.071 |
Neg logit contributions | |
aches | -0.114 |
le | -0.114 |
ered | -0.107 |
away | -0.105 |
ling | -0.102 |
Token | hugged |
Token ID | 41130 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
She | +0.176 |
Un | +0.172 |
They | +0.170 |
Max | +0.169 |
Well | +0.166 |
Neg logit contributions | |
aches | -0.162 |
le | -0.160 |
away | -0.153 |
ered | -0.150 |
up | -0.145 |
Token | them |
Token ID | 606 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
She | +0.103 |
Un | +0.101 |
They | +0.100 |
Max | +0.099 |
Jake | +0.098 |
Neg logit contributions | |
aches | -0.054 |
le | -0.053 |
away | -0.049 |
ered | -0.048 |
up | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
aches | +0.032 |
le | +0.030 |
ered | +0.028 |
away | +0.026 |
pe | +0.025 |
Neg logit contributions | |
She | -0.089 |
They | -0.087 |
Un | -0.086 |
Max | -0.086 |
Deal | -0.084 |
Token | kissed |
Token ID | 27946 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
She | +0.137 |
They | +0.134 |
Un | +0.131 |
Max | +0.128 |
Well | +0.125 |
Neg logit contributions | |
le | -0.148 |
aches | -0.147 |
ered | -0.137 |
away | -0.136 |
ling | -0.135 |
Token | them |
Token ID | 606 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
She | +0.099 |
They | +0.096 |
Un | +0.096 |
Max | +0.094 |
Jake | +0.093 |
Neg logit contributions | |
aches | -0.046 |
le | -0.044 |
ered | -0.041 |
away | -0.041 |
up | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
aches | +0.052 |
ered | +0.047 |
le | +0.047 |
pe | +0.043 |
ck | +0.042 |
Neg logit contributions | |
She | -0.100 |
They | -0.098 |
Max | -0.093 |
He | -0.092 |
Un | -0.089 |
Token | She |
Token ID | 1375 |
Feature activation | -9.91e-02 |
Pos logit contributions | |
Jake | +0.305 |
Ted | +0.300 |
Max | +0.285 |
Chris | +0.279 |
Mark | +0.265 |
Neg logit contributions | |
inside | -0.604 |
up | -0.583 |
pe | -0.574 |
in | -0.565 |
more | -0.556 |
Token | told |
Token ID | 1297 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
She | +0.191 |
Un | +0.187 |
They | +0.184 |
Max | +0.183 |
Jake | +0.180 |
Neg logit contributions | |
aches | -0.168 |
le | -0.166 |
away | -0.159 |
ered | -0.155 |
up | -0.152 |
Token | them |
Token ID | 606 |
Feature activation | +5.46e-03 |
Pos logit contributions | |
She | +0.223 |
They | +0.215 |
Un | +0.210 |
Deal | +0.204 |
Well | +0.203 |
Neg logit contributions | |
aches | -0.121 |
le | -0.119 |
away | -0.115 |
ered | -0.110 |
ish | -0.108 |
Token | she |
Token ID | 673 |
Feature activation | -8.06e-02 |
Pos logit contributions | |
le | +0.009 |
aches | +0.009 |
away | +0.009 |
ered | +0.008 |
ck | +0.008 |
Neg logit contributions | |
She | -0.011 |
They | -0.011 |
Max | -0.010 |
Did | -0.010 |
Not | -0.010 |
Token | loved |
Token ID | 6151 |
Feature activation | -3.89e-02 |
Pos logit contributions | |
She | +0.143 |
They | +0.138 |
Un | +0.137 |
Max | +0.136 |
Did | +0.133 |
Neg logit contributions | |
aches | -0.150 |
le | -0.147 |
away | -0.143 |
ered | -0.139 |
up | -0.135 |
Token | They |
Token ID | 2990 |
Feature activation | -8.69e-02 |
Pos logit contributions | |
Did | +0.325 |
Deal | +0.306 |
Not | +0.290 |
She | +0.283 |
Right | +0.275 |
Neg logit contributions | |
aches | -0.529 |
ck | -0.514 |
- | -0.510 |
le | -0.487 |
pe | -0.482 |
Token | got |
Token ID | 1392 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
She | +0.170 |
Un | +0.166 |
They | +0.164 |
Max | +0.163 |
Jake | +0.160 |
Neg logit contributions | |
aches | -0.144 |
le | -0.143 |
away | -0.135 |
ered | -0.133 |
up | -0.131 |
Token | in |
Token ID | 287 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
They | +0.074 |
She | +0.070 |
Don | +0.065 |
Their | +0.065 |
He | +0.064 |
Neg logit contributions | |
aches | -0.043 |
le | -0.043 |
ig | -0.038 |
ered | -0.036 |
eking | -0.036 |
Token | the |
Token ID | 262 |
Feature activation | -5.95e-02 |
Pos logit contributions | |
She | +0.113 |
Un | +0.111 |
They | +0.109 |
Max | +0.109 |
Jake | +0.107 |
Neg logit contributions | |
aches | -0.079 |
le | -0.078 |
away | -0.074 |
ered | -0.073 |
up | -0.071 |
Token | car |
Token ID | 1097 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
Un | +0.143 |
She | +0.143 |
Max | +0.141 |
Jake | +0.141 |
Deal | +0.139 |
Neg logit contributions | |
aches | -0.069 |
away | -0.068 |
le | -0.067 |
up | -0.063 |
pe | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
aches | +0.093 |
le | +0.084 |
ered | +0.082 |
ck | +0.077 |
eking | +0.074 |
Neg logit contributions | |
They | -0.122 |
She | -0.122 |
He | -0.116 |
Un | -0.109 |
Max | -0.107 |
Token | They |
Token ID | 1119 |
Feature activation | -9.68e-02 |
Pos logit contributions | |
Un | +0.274 |
Ol | +0.261 |
Jake | +0.257 |
Brother | +0.251 |
Deal | +0.250 |
Neg logit contributions | |
inside | -0.600 |
in | -0.563 |
up | -0.559 |
away | -0.557 |
home | -0.537 |
Token | hugged |
Token ID | 41130 |
Feature activation | -5.21e-02 |
Pos logit contributions | |
She | +0.176 |
Un | +0.175 |
They | +0.172 |
Max | +0.171 |
Jake | +0.166 |
Neg logit contributions | |
le | -0.172 |
aches | -0.171 |
away | -0.160 |
ered | -0.160 |
up | -0.159 |
Token | their |
Token ID | 511 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
She | +0.101 |
They | +0.101 |
Un | +0.096 |
Max | +0.095 |
He | +0.091 |
Neg logit contributions | |
aches | -0.096 |
ered | -0.086 |
le | -0.085 |
away | -0.080 |
ling | -0.079 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.07e-03 |
Pos logit contributions | |
They | +0.041 |
She | +0.040 |
Un | +0.039 |
Max | +0.039 |
Deal | +0.037 |
Neg logit contributions | |
le | -0.070 |
aches | -0.068 |
ered | -0.067 |
away | -0.063 |
ling | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | -7.64e-02 |
Pos logit contributions | |
aches | +0.009 |
ered | +0.009 |
ck | +0.008 |
eking | +0.008 |
le | +0.007 |
Neg logit contributions | |
She | -0.011 |
They | -0.011 |
He | -0.010 |
Max | -0.009 |
Un | -0.009 |
Token | some |
Token ID | 617 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
They | +0.052 |
She | +0.049 |
Max | +0.046 |
He | +0.044 |
Their | +0.044 |
Neg logit contributions | |
aches | -0.054 |
eking | -0.049 |
ig | -0.046 |
ering | -0.046 |
ling | -0.045 |
Token | blocks |
Token ID | 7021 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
aches | +0.023 |
ered | +0.022 |
le | +0.020 |
eking | +0.020 |
ling | +0.020 |
Neg logit contributions | |
They | -0.017 |
Max | -0.017 |
She | -0.017 |
Jake | -0.017 |
Lisa | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -7.42e-02 |
Pos logit contributions | |
aches | +0.103 |
ered | +0.099 |
le | +0.097 |
ling | +0.086 |
eking | +0.085 |
Neg logit contributions | |
She | -0.163 |
They | -0.160 |
Un | -0.153 |
Max | -0.151 |
He | -0.150 |
Token | joined |
Token ID | 5399 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
They | +0.144 |
She | +0.135 |
Max | +0.132 |
Un | +0.130 |
Lisa | +0.128 |
Neg logit contributions | |
le | -0.130 |
ling | -0.127 |
aches | -0.125 |
ering | -0.123 |
an | -0.120 |
Token | them |
Token ID | 606 |
Feature activation | +3.65e-02 |
Pos logit contributions | |
They | +0.076 |
She | +0.073 |
Max | +0.071 |
Jake | +0.069 |
Kids | +0.068 |
Neg logit contributions | |
aches | -0.056 |
crown | -0.045 |
ig | -0.045 |
ered | -0.044 |
le | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
aches | +0.056 |
ered | +0.054 |
le | +0.054 |
ering | +0.047 |
eking | +0.047 |
Neg logit contributions | |
They | -0.083 |
She | -0.082 |
He | -0.075 |
Max | -0.073 |
Not | -0.070 |
Token | They |
Token ID | 1119 |
Feature activation | -9.08e-02 |
Pos logit contributions | |
Jake | +0.283 |
Johnny | +0.283 |
Kim | +0.275 |
Chris | +0.264 |
Ol | +0.262 |
Neg logit contributions | |
inside | -0.598 |
up | -0.583 |
in | -0.569 |
away | -0.547 |
again | -0.537 |
Token | helped |
Token ID | 4193 |
Feature activation | -7.20e-02 |
Pos logit contributions | |
She | +0.170 |
They | +0.168 |
Max | +0.165 |
Un | +0.165 |
Jake | +0.160 |
Neg logit contributions | |
le | -0.157 |
aches | -0.154 |
ered | -0.147 |
away | -0.147 |
up | -0.143 |
Token | them |
Token ID | 606 |
Feature activation | +1.91e-02 |
Pos logit contributions | |
They | +0.145 |
She | +0.141 |
Un | +0.133 |
Max | +0.131 |
Lisa | +0.128 |
Neg logit contributions | |
ig | -0.124 |
aches | -0.118 |
ish | -0.112 |
eking | -0.108 |
ered | -0.108 |
Token | make |
Token ID | 787 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
le | +0.035 |
ering | +0.032 |
pearl | +0.032 |
eking | +0.032 |
ling | +0.032 |
Neg logit contributions | |
They | -0.037 |
She | -0.037 |
He | -0.033 |
Max | -0.030 |
Something | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
They | +0.072 |
She | +0.067 |
Lisa | +0.063 |
Jake | +0.063 |
Jake | +0.062 |
Neg logit contributions | |
aches | -0.059 |
ig | -0.059 |
le | -0.058 |
ered | -0.054 |
ling | -0.053 |
Token | Y |
Token ID | 575 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
Well | +0.230 |
Lisa | +0.229 |
Jake | +0.224 |
Chris | +0.223 |
Un | +0.218 |
Neg logit contributions | |
inside | -0.522 |
up | -0.514 |
le | -0.504 |
in | -0.503 |
- | -0.495 |
Token | uck |
Token ID | 1347 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
Un | +0.085 |
She | +0.084 |
Jake | +0.083 |
Lisa | +0.083 |
Max | +0.083 |
Neg logit contributions | |
le | -0.061 |
aches | -0.059 |
away | -0.057 |
ered | -0.056 |
- | -0.056 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
aches | +0.069 |
le | +0.064 |
ered | +0.060 |
crown | +0.058 |
ck | +0.057 |
Neg logit contributions | |
She | -0.138 |
They | -0.135 |
He | -0.129 |
Max | -0.129 |
Un | -0.129 |
Token | she |
Token ID | 673 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
Deal | +0.380 |
Un | +0.375 |
Great | +0.348 |
Thanks | +0.341 |
Nice | +0.330 |
Neg logit contributions | |
away | -0.336 |
up | -0.321 |
in | -0.318 |
pe | -0.317 |
inside | -0.315 |
Token | said |
Token ID | 531 |
Feature activation | -8.19e-02 |
Pos logit contributions | |
She | +0.154 |
Un | +0.152 |
They | +0.149 |
Max | +0.148 |
Jake | +0.146 |
Neg logit contributions | |
aches | -0.137 |
le | -0.136 |
away | -0.129 |
ered | -0.127 |
up | -0.124 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
She | +0.199 |
They | +0.190 |
He | +0.183 |
Max | +0.175 |
Un | +0.174 |
Neg logit contributions | |
aches | -0.125 |
le | -0.107 |
ered | -0.102 |
island | -0.101 |
coins | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
liv | +0.473 |
Deal | +0.470 |
Okay | +0.466 |
arge | +0.456 |
Looks | +0.455 |
Neg logit contributions | |
in | -0.370 |
inside | -0.368 |
away | -0.350 |
on | -0.348 |
out | -0.348 |
Token |
|
Token ID | 198 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
Deal | +0.406 |
Did | +0.394 |
Really | +0.383 |
Un | +0.381 |
Well | +0.380 |
Neg logit contributions | |
up | -0.385 |
inside | -0.379 |
away | -0.366 |
- | -0.364 |
in | -0.360 |
Token | Max |
Token ID | 11518 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
Nice | +0.238 |
Did | +0.233 |
Deal | +0.226 |
Today | +0.221 |
Lem | +0.207 |
Neg logit contributions | |
away | -0.593 |
- | -0.551 |
inside | -0.540 |
le | -0.534 |
ck | -0.533 |
Token | looked |
Token ID | 3114 |
Feature activation | -8.87e-03 |
Pos logit contributions | |
She | +0.041 |
Un | +0.041 |
They | +0.040 |
Max | +0.040 |
Deal | +0.040 |
Neg logit contributions | |
le | -0.033 |
aches | -0.033 |
away | -0.031 |
ered | -0.030 |
up | -0.030 |
Token | at |
Token ID | 379 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
She | +0.020 |
He | +0.019 |
They | +0.019 |
Don | +0.018 |
Why | +0.017 |
Neg logit contributions | |
aches | -0.015 |
ered | -0.013 |
le | -0.012 |
island | -0.012 |
ck | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
Un | +0.418 |
Deal | +0.398 |
Jake | +0.379 |
Chris | +0.370 |
Grand | +0.365 |
Neg logit contributions | |
again | -0.373 |
up | -0.369 |
away | -0.365 |
- | -0.362 |
inside | -0.361 |
Token |
|
Token ID | 198 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
Deal | +0.353 |
Did | +0.352 |
Un | +0.338 |
Not | +0.323 |
Well | +0.318 |
Neg logit contributions | |
up | -0.445 |
- | -0.445 |
inside | -0.423 |
away | -0.423 |
le | -0.418 |
Token | L |
Token ID | 43 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
Did | +0.297 |
Deal | +0.284 |
Not | +0.269 |
Whe | +0.258 |
Nice | +0.253 |
Neg logit contributions | |
- | -0.528 |
away | -0.519 |
aches | -0.513 |
le | -0.506 |
more | -0.498 |
Token | ily |
Token ID | 813 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
She | +0.154 |
Un | +0.149 |
They | +0.146 |
Max | +0.146 |
Jake | +0.143 |
Neg logit contributions | |
aches | -0.201 |
le | -0.199 |
away | -0.192 |
ered | -0.188 |
up | -0.185 |
Token | smiled |
Token ID | 13541 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
She | +0.087 |
They | +0.084 |
Max | +0.083 |
Jake | +0.080 |
Well | +0.080 |
Neg logit contributions | |
aches | -0.090 |
le | -0.086 |
ered | -0.082 |
away | -0.081 |
ling | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
She | +0.134 |
They | +0.129 |
Un | +0.129 |
Max | +0.126 |
He | +0.125 |
Neg logit contributions | |
aches | -0.056 |
le | -0.051 |
ered | -0.049 |
away | -0.046 |
pe | -0.043 |
Token | She |
Token ID | 1375 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Jake | +0.290 |
Un | +0.267 |
Deal | +0.262 |
Grand | +0.250 |
Tyler | +0.248 |
Neg logit contributions | |
inside | -0.606 |
up | -0.598 |
, | -0.549 |
in | -0.546 |
again | -0.541 |
Token | said |
Token ID | 531 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
She | +0.201 |
Un | +0.198 |
They | +0.194 |
Max | +0.194 |
Jake | +0.191 |
Neg logit contributions | |
aches | -0.178 |
le | -0.176 |
away | -0.168 |
ered | -0.164 |
up | -0.161 |
Token | , |
Token ID | 11 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
She | +0.263 |
They | +0.245 |
He | +0.235 |
Did | +0.233 |
Max | +0.224 |
Neg logit contributions | |
le | -0.134 |
aches | -0.129 |
ered | -0.118 |
ck | -0.114 |
away | -0.109 |
Token | " |
Token ID | 366 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
Jake | +0.163 |
Deal | +0.161 |
Lisa | +0.160 |
Un | +0.159 |
Max | +0.154 |
Neg logit contributions | |
aches | -0.291 |
inside | -0.289 |
up | -0.287 |
pe | -0.287 |
away | -0.286 |
Token | OK |
Token ID | 11380 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
Deal | +0.198 |
Tyler | +0.173 |
Jake | +0.170 |
Today | +0.161 |
Their | +0.159 |
Neg logit contributions | |
le | -0.309 |
away | -0.303 |
pe | -0.299 |
- | -0.297 |
ring | -0.296 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
aches | +0.133 |
le | +0.120 |
ered | +0.112 |
ck | +0.104 |
crown | +0.103 |
Neg logit contributions | |
She | -0.255 |
They | -0.247 |
He | -0.239 |
Un | -0.231 |
Max | -0.231 |
Token | She |
Token ID | 1375 |
Feature activation | -9.71e-02 |
Pos logit contributions | |
Un | +0.355 |
Grand | +0.351 |
Jake | +0.333 |
Mark | +0.328 |
Kim | +0.325 |
Neg logit contributions | |
up | -0.466 |
inside | -0.445 |
out | -0.442 |
in | -0.433 |
home | -0.429 |
Token | smiles |
Token ID | 21845 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
She | +0.176 |
They | +0.168 |
He | +0.166 |
Max | +0.160 |
Well | +0.158 |
Neg logit contributions | |
aches | -0.186 |
le | -0.180 |
ered | -0.166 |
ling | -0.166 |
away | -0.166 |
Token | and |
Token ID | 290 |
Feature activation | -8.89e-02 |
Pos logit contributions | |
She | +0.085 |
Un | +0.084 |
Max | +0.083 |
They | +0.083 |
Jake | +0.082 |
Neg logit contributions | |
le | -0.028 |
aches | -0.027 |
away | -0.026 |
up | -0.024 |
ered | -0.024 |
Token | sings |
Token ID | 33041 |
Feature activation | -2.55e-02 |
Pos logit contributions | |
She | +0.171 |
Un | +0.169 |
Max | +0.166 |
They | +0.165 |
Jake | +0.163 |
Neg logit contributions | |
aches | -0.149 |
le | -0.147 |
away | -0.141 |
ered | -0.138 |
up | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
She | +0.060 |
They | +0.057 |
Un | +0.057 |
Max | +0.057 |
Jake | +0.055 |
Neg logit contributions | |
aches | -0.036 |
le | -0.033 |
ered | -0.032 |
away | -0.030 |
pe | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
Jake | +0.397 |
Uh | +0.394 |
bek | +0.390 |
Mark | +0.388 |
Kim | +0.383 |
Neg logit contributions | |
up | -0.450 |
inside | -0.428 |
out | -0.404 |
home | -0.399 |
in | -0.393 |
Token |
|
Token ID | 198 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
Deal | +0.389 |
Did | +0.378 |
Un | +0.370 |
Really | +0.369 |
Sorry | +0.362 |
Neg logit contributions | |
up | -0.414 |
inside | -0.402 |
aches | -0.386 |
le | -0.380 |
away | -0.375 |
Token | Ben |
Token ID | 11696 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
Did | +0.332 |
Deal | +0.302 |
Nice | +0.301 |
She | +0.296 |
They | +0.293 |
Neg logit contributions | |
aches | -0.482 |
away | -0.476 |
le | -0.467 |
ling | -0.451 |
- | -0.441 |
Token | does |
Token ID | 857 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
She | +0.057 |
They | +0.055 |
Max | +0.054 |
He | +0.053 |
Jake | +0.052 |
Neg logit contributions | |
aches | -0.055 |
le | -0.052 |
ered | -0.050 |
away | -0.048 |
ling | -0.046 |
Token | not |
Token ID | 407 |
Feature activation | -6.95e-02 |
Pos logit contributions | |
She | +0.111 |
They | +0.107 |
Un | +0.106 |
Not | +0.104 |
Max | +0.101 |
Neg logit contributions | |
le | -0.116 |
aches | -0.111 |
ish | -0.107 |
away | -0.105 |
ling | -0.105 |
Token | tomorrow |
Token ID | 9439 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
ered | +0.097 |
aches | +0.094 |
le | +0.085 |
les | +0.081 |
eking | +0.081 |
Neg logit contributions | |
She | -0.165 |
They | -0.162 |
He | -0.154 |
Un | -0.152 |
Max | -0.147 |
Token | . |
Token ID | 13 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
She | +0.025 |
They | +0.025 |
Un | +0.024 |
He | +0.023 |
Max | +0.023 |
Neg logit contributions | |
aches | -0.014 |
ered | -0.014 |
le | -0.013 |
ck | -0.012 |
ling | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | -9.54e-02 |
Pos logit contributions | |
Johnny | +0.396 |
Kim | +0.394 |
Chris | +0.393 |
Jake | +0.376 |
Matt | +0.365 |
Neg logit contributions | |
up | -0.435 |
inside | -0.418 |
in | -0.411 |
, | -0.401 |
," | -0.392 |
Token | kisses |
Token ID | 36544 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
She | +0.155 |
They | +0.151 |
Deal | +0.144 |
Max | +0.139 |
Un | +0.139 |
Neg logit contributions | |
aches | -0.193 |
le | -0.189 |
away | -0.185 |
ered | -0.183 |
ling | -0.180 |
Token | them |
Token ID | 606 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
She | +0.074 |
Un | +0.074 |
They | +0.072 |
Max | +0.072 |
Jake | +0.072 |
Neg logit contributions | |
aches | -0.030 |
le | -0.030 |
away | -0.028 |
ered | -0.027 |
up | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
aches | +0.038 |
ered | +0.035 |
le | +0.034 |
ering | +0.030 |
ck | +0.030 |
Neg logit contributions | |
She | -0.071 |
They | -0.070 |
Max | -0.066 |
He | -0.065 |
Un | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
Kim | +0.426 |
Johnny | +0.420 |
bek | +0.416 |
Chris | +0.411 |
Uh | +0.409 |
Neg logit contributions | |
up | -0.402 |
inside | -0.400 |
in | -0.389 |
, | -0.374 |
out | -0.366 |
Token |
|
Token ID | 198 |
Feature activation | -2.34e-01 |
Pos logit contributions | |
Did | +0.373 |
Un | +0.371 |
Deal | +0.368 |
Really | +0.362 |
Uh | +0.356 |
Neg logit contributions | |
up | -0.444 |
inside | -0.429 |
aches | -0.422 |
le | -0.407 |
away | -0.406 |
Token | Anna |
Token ID | 31160 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
Did | +0.345 |
Nice | +0.318 |
Something | +0.300 |
Sure | +0.297 |
Maybe | +0.291 |
Neg logit contributions | |
away | -0.507 |
aches | -0.493 |
le | -0.493 |
pe | -0.464 |
ling | -0.463 |
Token | and |
Token ID | 290 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
She | +0.116 |
They | +0.113 |
Max | +0.107 |
He | +0.106 |
Un | +0.105 |
Neg logit contributions | |
aches | -0.068 |
ered | -0.060 |
le | -0.060 |
away | -0.056 |
ck | -0.053 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
Well | +0.113 |
Did | +0.112 |
Un | +0.108 |
Don | +0.106 |
Grand | +0.102 |
Neg logit contributions | |
up | -0.249 |
le | -0.248 |
away | -0.247 |
aches | -0.240 |
pe | -0.236 |
Token | ." |
Token ID | 526 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
They | +0.095 |
She | +0.094 |
Un | +0.086 |
Max | +0.084 |
He | +0.084 |
Neg logit contributions | |
aches | -0.065 |
le | -0.062 |
ered | -0.058 |
crown | -0.057 |
ling | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
Un | +0.480 |
Chris | +0.471 |
Thanks | +0.463 |
istence | +0.463 |
Kim | +0.457 |
Neg logit contributions | |
in | -0.325 |
up | -0.322 |
again | -0.316 |
inside | -0.315 |
, | -0.313 |
Token |
|
Token ID | 198 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
Did | +0.378 |
Un | +0.358 |
Deal | +0.341 |
Chris | +0.336 |
Not | +0.332 |
Neg logit contributions | |
up | -0.418 |
inside | -0.410 |
le | -0.403 |
away | -0.401 |
- | -0.398 |
Token | Anna |
Token ID | 31160 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
Did | +0.404 |
Nice | +0.372 |
Not | +0.364 |
Everywhere | +0.363 |
Chris | +0.359 |
Neg logit contributions | |
away | -0.407 |
le | -0.398 |
more | -0.396 |
aches | -0.384 |
in | -0.376 |
Token | smiled |
Token ID | 13541 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
She | +0.079 |
They | +0.076 |
Max | +0.075 |
Un | +0.074 |
Jake | +0.074 |
Neg logit contributions | |
aches | -0.082 |
le | -0.078 |
away | -0.075 |
ered | -0.074 |
up | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
She | +0.142 |
Un | +0.140 |
They | +0.138 |
Max | +0.137 |
Jake | +0.136 |
Neg logit contributions | |
aches | -0.050 |
le | -0.048 |
away | -0.044 |
ered | -0.043 |
up | -0.040 |
Token | She |
Token ID | 1375 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
Chris | +0.269 |
Jake | +0.267 |
Un | +0.265 |
Johnny | +0.262 |
Kim | +0.250 |
Neg logit contributions | |
inside | -0.617 |
up | -0.612 |
in | -0.578 |
again | -0.568 |
, | -0.555 |
Token | said |
Token ID | 531 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
She | +0.209 |
Un | +0.207 |
Deal | +0.206 |
They | +0.202 |
Jake | +0.201 |
Neg logit contributions | |
le | -0.163 |
aches | -0.162 |
away | -0.155 |
pe | -0.150 |
ered | -0.150 |
Token | , |
Token ID | 11 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
She | +0.272 |
They | +0.257 |
Did | +0.245 |
He | +0.244 |
Not | +0.239 |
Neg logit contributions | |
le | -0.128 |
aches | -0.119 |
ered | -0.107 |
away | -0.106 |
ck | -0.105 |
Token | " |
Token ID | 366 |
Feature activation | -1.49e-01 |
Pos logit contributions | |
Jake | +0.171 |
Un | +0.170 |
Lisa | +0.167 |
Max | +0.163 |
Deal | +0.160 |
Neg logit contributions | |
inside | -0.289 |
up | -0.284 |
aches | -0.282 |
away | -0.281 |
pe | -0.281 |
Token | Thank |
Token ID | 10449 |
Feature activation | -8.36e-02 |
Pos logit contributions | |
Tyler | +0.177 |
Deal | +0.170 |
Their | +0.166 |
uddy | +0.163 |
Jake | +0.159 |
Neg logit contributions | |
away | -0.314 |
le | -0.313 |
- | -0.307 |
pe | -0.294 |
ring | -0.285 |
Token | . |
Token ID | 13 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
aches | +0.190 |
ered | +0.180 |
le | +0.164 |
ck | +0.160 |
ig | +0.147 |
Neg logit contributions | |
She | -0.262 |
They | -0.260 |
He | -0.246 |
Why | -0.235 |
Maybe | -0.234 |
Token | And |
Token ID | 843 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
Un | +0.142 |
Jake | +0.137 |
Chris | +0.126 |
Max | +0.126 |
Lisa | +0.125 |
Neg logit contributions | |
le | -0.246 |
in | -0.235 |
pe | -0.234 |
inside | -0.233 |
aches | -0.233 |
Token | you |
Token ID | 345 |
Feature activation | +6.16e-02 |
Pos logit contributions | |
She | +0.022 |
They | +0.021 |
Un | +0.021 |
Max | +0.020 |
Did | +0.020 |
Neg logit contributions | |
le | -0.028 |
aches | -0.027 |
away | -0.026 |
ling | -0.026 |
ered | -0.026 |
Token | have |
Token ID | 423 |
Feature activation | +5.11e-02 |
Pos logit contributions | |
aches | +0.127 |
le | +0.125 |
ling | +0.121 |
ering | +0.121 |
ered | +0.119 |
Neg logit contributions | |
They | -0.097 |
She | -0.093 |
Max | -0.090 |
Not | -0.089 |
Something | -0.088 |
Token | no |
Token ID | 645 |
Feature activation | +6.73e-02 |
Pos logit contributions | |
le | +0.091 |
aches | +0.084 |
ling | +0.080 |
ered | +0.080 |
up | +0.074 |
Neg logit contributions | |
She | -0.101 |
Un | -0.099 |
They | -0.099 |
Max | -0.097 |
Not | -0.093 |
Token | stitch |
Token ID | 24695 |
Feature activation | +2.05e-01 |
Pos logit contributions | |
le | +0.100 |
ered | +0.095 |
ling | +0.095 |
aches | +0.094 |
ering | +0.086 |
Neg logit contributions | |
They | -0.143 |
She | -0.140 |
Un | -0.134 |
He | -0.133 |
Deal | -0.133 |
Token | and |
Token ID | 290 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
aches | +0.251 |
ered | +0.244 |
le | +0.230 |
pe | +0.210 |
ck | +0.210 |
Neg logit contributions | |
She | -0.485 |
They | -0.473 |
Max | -0.445 |
He | -0.442 |
Un | -0.441 |
Token | no |
Token ID | 645 |
Feature activation | +6.43e-02 |
Pos logit contributions | |
le | +0.046 |
aches | +0.045 |
ered | +0.042 |
away | +0.042 |
ling | +0.041 |
Neg logit contributions | |
She | -0.054 |
Un | -0.053 |
They | -0.052 |
Max | -0.052 |
Did | -0.050 |
Token | band |
Token ID | 4097 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
le | +0.094 |
aches | +0.092 |
ling | +0.091 |
ered | +0.091 |
ering | +0.083 |
Neg logit contributions | |
She | -0.138 |
They | -0.134 |
Max | -0.129 |
He | -0.129 |
Un | -0.128 |
Token | age |
Token ID | 496 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
aches | +0.227 |
le | +0.226 |
ered | +0.212 |
pe | +0.209 |
island | +0.202 |
Neg logit contributions | |
She | -0.410 |
They | -0.407 |
Un | -0.392 |
Did | -0.389 |
Max | -0.386 |
Token | . |
Token ID | 13 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
aches | +0.261 |
le | +0.252 |
ered | +0.246 |
pe | +0.232 |
ck | +0.227 |
Neg logit contributions | |
She | -0.467 |
They | -0.450 |
Un | -0.434 |
He | -0.430 |
Max | -0.426 |
Token | The |
Token ID | 383 |
Feature activation | -2.85e-03 |
Pos logit contributions | |
Un | +0.133 |
Deal | +0.128 |
Max | +0.121 |
Jake | +0.120 |
Lisa | +0.120 |
Neg logit contributions | |
away | -0.191 |
up | -0.190 |
pe | -0.187 |
aches | -0.187 |
le | -0.185 |
Token | family |
Token ID | 1641 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
She | +0.006 |
Un | +0.006 |
They | +0.006 |
Max | +0.006 |
Jake | +0.006 |
Neg logit contributions | |
aches | -0.004 |
le | -0.004 |
away | -0.004 |
ered | -0.004 |
up | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
aches | +0.182 |
le | +0.180 |
away | +0.171 |
ered | +0.164 |
ling | +0.160 |
Neg logit contributions | |
She | -0.204 |
They | -0.198 |
Jake | -0.185 |
He | -0.184 |
Well | -0.184 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | +6.87e-02 |
Pos logit contributions | |
aches | +0.038 |
le | +0.038 |
away | +0.036 |
ered | +0.035 |
up | +0.034 |
Neg logit contributions | |
She | -0.057 |
Un | -0.056 |
Max | -0.055 |
They | -0.055 |
Jake | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +6.10e-02 |
Pos logit contributions | |
le | +0.108 |
aches | +0.105 |
away | +0.098 |
ered | +0.098 |
ling | +0.094 |
Neg logit contributions | |
She | -0.144 |
They | -0.141 |
Un | -0.140 |
Max | -0.136 |
Jake | -0.135 |
Token | picnic |
Token ID | 35715 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
aches | +0.098 |
le | +0.096 |
ered | +0.094 |
away | +0.091 |
ck | +0.088 |
Neg logit contributions | |
She | -0.121 |
They | -0.119 |
Max | -0.116 |
Un | -0.114 |
Deal | -0.113 |
Token | together |
Token ID | 1978 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
ered | +0.298 |
aches | +0.296 |
ck | +0.288 |
le | +0.284 |
eking | +0.251 |
Neg logit contributions | |
She | -0.487 |
They | -0.485 |
He | -0.448 |
Un | -0.431 |
Max | -0.428 |
Token | , |
Token ID | 11 |
Feature activation | -7.47e-03 |
Pos logit contributions | |
aches | +0.117 |
ered | +0.113 |
le | +0.111 |
ck | +0.105 |
an | +0.104 |
Neg logit contributions | |
She | -0.170 |
They | -0.166 |
He | -0.155 |
Un | -0.154 |
Max | -0.144 |
Token | and |
Token ID | 290 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
She | +0.017 |
They | +0.017 |
Un | +0.017 |
Max | +0.016 |
Jake | +0.016 |
Neg logit contributions | |
le | -0.010 |
aches | -0.010 |
ered | -0.009 |
away | -0.009 |
ling | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
aches | +0.024 |
le | +0.024 |
ered | +0.022 |
away | +0.022 |
ling | +0.021 |
Neg logit contributions | |
She | -0.024 |
They | -0.023 |
Un | -0.023 |
Max | -0.022 |
Jake | -0.022 |
Token | memory |
Token ID | 4088 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
aches | +0.058 |
le | +0.057 |
ered | +0.054 |
away | +0.053 |
ck | +0.050 |
Neg logit contributions | |
She | -0.062 |
Un | -0.061 |
They | -0.060 |
Max | -0.060 |
Well | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | +9.43e-02 |
Pos logit contributions | |
aches | +0.182 |
le | +0.180 |
away | +0.169 |
ered | +0.166 |
up | +0.159 |
Neg logit contributions | |
She | -0.245 |
Un | -0.239 |
They | -0.236 |
Max | -0.235 |
Jake | -0.231 |
Token | big |
Token ID | 1263 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
aches | +0.148 |
le | +0.147 |
ered | +0.142 |
away | +0.134 |
up | +0.127 |
Neg logit contributions | |
Un | -0.196 |
She | -0.196 |
Max | -0.192 |
They | -0.189 |
Jake | -0.186 |
Token | tree |
Token ID | 5509 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
aches | +0.248 |
ered | +0.242 |
le | +0.234 |
ck | +0.216 |
eking | +0.207 |
Neg logit contributions | |
She | -0.320 |
They | -0.308 |
He | -0.299 |
Un | -0.289 |
Max | -0.287 |
Token | in |
Token ID | 287 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
aches | +0.351 |
ered | +0.338 |
le | +0.338 |
les | +0.299 |
away | +0.297 |
Neg logit contributions | |
She | -0.573 |
They | -0.544 |
He | -0.528 |
Un | -0.500 |
Why | -0.474 |
Token | a |
Token ID | 257 |
Feature activation | +7.86e-02 |
Pos logit contributions | |
aches | +0.109 |
le | +0.106 |
away | +0.103 |
ered | +0.099 |
up | +0.097 |
Neg logit contributions | |
Un | -0.230 |
She | -0.230 |
Max | -0.227 |
Deal | -0.226 |
Did | -0.224 |
Token | park |
Token ID | 3952 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
le | +0.122 |
aches | +0.118 |
ered | +0.116 |
ck | +0.106 |
ling | +0.099 |
Neg logit contributions | |
She | -0.177 |
Un | -0.172 |
Max | -0.166 |
They | -0.164 |
Lisa | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
le | +0.281 |
aches | +0.280 |
ered | +0.250 |
ck | +0.238 |
away | +0.234 |
Neg logit contributions | |
She | -0.536 |
They | -0.520 |
He | -0.495 |
Un | -0.481 |
Max | -0.472 |
Token | Many |
Token ID | 4650 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
Un | +0.113 |
Jake | +0.107 |
Deal | +0.104 |
Lisa | +0.102 |
Kids | +0.101 |
Neg logit contributions | |
pe | -0.216 |
up | -0.216 |
in | -0.212 |
inside | -0.212 |
aches | -0.202 |
Token | animals |
Token ID | 4695 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
away | +0.164 |
ered | +0.160 |
aches | +0.158 |
le | +0.156 |
up | +0.146 |
Neg logit contributions | |
She | -0.214 |
They | -0.199 |
Well | -0.194 |
Un | -0.193 |
Sam | -0.193 |
Token | lived |
Token ID | 5615 |
Feature activation | +8.11e-02 |
Pos logit contributions | |
aches | +0.332 |
le | +0.329 |
ig | +0.314 |
egg | +0.309 |
an | +0.306 |
Neg logit contributions | |
She | -0.257 |
They | -0.252 |
He | -0.217 |
Did | -0.214 |
Don | -0.213 |
Token | in |
Token ID | 287 |
Feature activation | +7.19e-02 |
Pos logit contributions | |
le | +0.087 |
aches | +0.082 |
ered | +0.070 |
ck | +0.068 |
les | +0.066 |
Neg logit contributions | |
She | -0.222 |
They | -0.213 |
He | -0.203 |
Un | -0.203 |
Max | -0.203 |
Token | When |
Token ID | 1649 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
Un | +0.090 |
Jake | +0.083 |
Max | +0.081 |
Lisa | +0.081 |
Chris | +0.081 |
Neg logit contributions | |
le | -0.140 |
pe | -0.136 |
away | -0.136 |
up | -0.135 |
inside | -0.135 |
Token | I |
Token ID | 314 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
le | +0.035 |
aches | +0.034 |
ling | +0.032 |
away | +0.032 |
ered | +0.032 |
Neg logit contributions | |
She | -0.033 |
They | -0.032 |
Un | -0.032 |
Max | -0.031 |
Jake | -0.030 |
Token | was |
Token ID | 373 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
aches | +0.115 |
le | +0.111 |
ling | +0.108 |
up | +0.107 |
away | +0.105 |
Neg logit contributions | |
She | -0.118 |
They | -0.113 |
Max | -0.108 |
Did | -0.107 |
Something | -0.107 |
Token | a |
Token ID | 257 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
le | +0.136 |
an | +0.134 |
aches | +0.127 |
ish | +0.125 |
ling | +0.123 |
Neg logit contributions | |
She | -0.161 |
Jake | -0.156 |
Jake | -0.153 |
Deal | -0.150 |
Not | -0.150 |
Token | young |
Token ID | 1862 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
aches | +0.132 |
ered | +0.128 |
le | +0.126 |
ck | +0.113 |
away | +0.112 |
Neg logit contributions | |
Un | -0.157 |
She | -0.156 |
Max | -0.152 |
They | -0.151 |
Jake | -0.149 |
Token | soldier |
Token ID | 10686 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
aches | +0.399 |
ered | +0.385 |
ck | +0.348 |
le | +0.345 |
eking | +0.338 |
Neg logit contributions | |
She | -0.262 |
They | -0.242 |
He | -0.233 |
Un | -0.217 |
Max | -0.217 |
Token | , |
Token ID | 11 |
Feature activation | +3.51e-02 |
Pos logit contributions | |
aches | +0.298 |
ered | +0.266 |
le | +0.260 |
ck | +0.239 |
away | +0.234 |
Neg logit contributions | |
She | -0.564 |
They | -0.547 |
Max | -0.533 |
Un | -0.531 |
He | -0.527 |
Token | I |
Token ID | 314 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
aches | +0.071 |
le | +0.071 |
away | +0.068 |
ered | +0.067 |
up | +0.065 |
Neg logit contributions | |
She | -0.056 |
Un | -0.054 |
They | -0.054 |
Max | -0.053 |
Jake | -0.052 |
Token | traveled |
Token ID | 14113 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
aches | +0.103 |
le | +0.098 |
up | +0.095 |
away | +0.094 |
ered | +0.093 |
Neg logit contributions | |
She | -0.083 |
They | -0.081 |
Did | -0.077 |
Max | -0.076 |
He | -0.076 |
Token | to |
Token ID | 284 |
Feature activation | +9.77e-02 |
Pos logit contributions | |
aches | +0.253 |
le | +0.225 |
ered | +0.218 |
ck | +0.211 |
ish | +0.203 |
Neg logit contributions | |
She | -0.405 |
They | -0.393 |
He | -0.383 |
Max | -0.380 |
Un | -0.375 |
Token | many |
Token ID | 867 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
aches | +0.165 |
le | +0.162 |
ered | +0.153 |
ling | +0.152 |
ish | +0.151 |
Neg logit contributions | |
They | -0.192 |
She | -0.192 |
Un | -0.192 |
Max | -0.182 |
He | -0.181 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
aches | +0.410 |
ered | +0.404 |
le | +0.376 |
ck | +0.357 |
pe | +0.335 |
Neg logit contributions | |
They | -0.608 |
She | -0.599 |
Max | -0.590 |
He | -0.574 |
Un | -0.565 |
Token | It |
Token ID | 632 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
Un | +0.149 |
Jake | +0.141 |
Chris | +0.140 |
Lisa | +0.137 |
Kelly | +0.133 |
Neg logit contributions | |
away | -0.191 |
up | -0.188 |
le | -0.187 |
inside | -0.187 |
in | -0.183 |
Token | was |
Token ID | 373 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
aches | +0.228 |
le | +0.211 |
ered | +0.201 |
crown | +0.192 |
ering | +0.191 |
Neg logit contributions | |
They | -0.213 |
She | -0.211 |
He | -0.200 |
Don | -0.200 |
Max | -0.197 |
Token | our |
Token ID | 674 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
le | +0.091 |
aches | +0.088 |
ered | +0.082 |
les | +0.080 |
ling | +0.079 |
Neg logit contributions | |
They | -0.096 |
She | -0.095 |
Un | -0.094 |
Max | -0.092 |
Deal | -0.091 |
Token | favorite |
Token ID | 4004 |
Feature activation | +1.87e-01 |
Pos logit contributions | |
le | +0.182 |
aches | +0.181 |
ered | +0.180 |
ck | +0.147 |
ling | +0.146 |
Neg logit contributions | |
They | -0.257 |
She | -0.238 |
Max | -0.237 |
Deal | -0.237 |
He | -0.234 |
Token | toy |
Token ID | 13373 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
aches | +0.295 |
ered | +0.288 |
eking | +0.271 |
ck | +0.265 |
ering | +0.252 |
Neg logit contributions | |
They | -0.380 |
He | -0.337 |
Kids | -0.321 |
She | -0.320 |
Max | -0.318 |
Token | ," |
Token ID | 553 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
aches | +0.276 |
le | +0.258 |
ered | +0.248 |
ck | +0.226 |
away | +0.226 |
Neg logit contributions | |
She | -0.509 |
They | -0.508 |
He | -0.483 |
Max | -0.482 |
Did | -0.479 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.37e-02 |
Pos logit contributions | |
Deal | +0.123 |
Un | +0.122 |
Max | +0.112 |
Did | +0.110 |
Chris | +0.109 |
Neg logit contributions | |
pe | -0.127 |
away | -0.126 |
aches | -0.120 |
in | -0.119 |
up | -0.117 |
Token | said |
Token ID | 531 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
le | +0.109 |
away | +0.105 |
aches | +0.105 |
up | +0.104 |
pe | +0.100 |
Neg logit contributions | |
Un | -0.158 |
Max | -0.153 |
Jake | -0.150 |
Deal | -0.149 |
Well | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
aches | +0.041 |
le | +0.038 |
ered | +0.035 |
island | +0.035 |
coins | +0.033 |
Neg logit contributions | |
She | -0.070 |
They | -0.067 |
He | -0.064 |
Un | -0.062 |
Max | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
liv | +0.253 |
Un | +0.246 |
Jake | +0.239 |
Ta | +0.239 |
Josh | +0.238 |
Neg logit contributions | |
, | -0.213 |
up | -0.208 |
again | -0.207 |
in | -0.207 |
inside | -0.202 |
Token | I |
Token ID | 40 |
Feature activation | -7.70e-02 |
Pos logit contributions | |
Deal | +0.178 |
Their | +0.148 |
Tyler | +0.142 |
Lena | +0.140 |
uddy | +0.140 |
Neg logit contributions | |
le | -0.292 |
pe | -0.285 |
b | -0.279 |
- | -0.275 |
aches | -0.272 |
Token | want |
Token ID | 765 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
She | +0.124 |
They | +0.118 |
Did | +0.115 |
Un | +0.113 |
Max | +0.112 |
Neg logit contributions | |
aches | -0.156 |
le | -0.149 |
away | -0.145 |
ered | -0.144 |
up | -0.143 |
Token | to |
Token ID | 284 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
She | +0.042 |
They | +0.040 |
Un | +0.039 |
He | +0.039 |
Not | +0.039 |
Neg logit contributions | |
aches | -0.035 |
le | -0.034 |
ling | -0.032 |
ered | -0.032 |
away | -0.030 |
Token | help |
Token ID | 1037 |
Feature activation | -6.06e-03 |
Pos logit contributions | |
She | +0.033 |
They | +0.030 |
He | +0.030 |
Not | +0.029 |
Deal | +0.029 |
Neg logit contributions | |
le | -0.026 |
ling | -0.026 |
aches | -0.025 |
ering | -0.025 |
away | -0.024 |
Token | polish |
Token ID | 25245 |
Feature activation | +7.23e-02 |
Pos logit contributions | |
She | +0.012 |
They | +0.011 |
He | +0.011 |
Not | +0.011 |
Un | +0.011 |
Neg logit contributions | |
aches | -0.010 |
ered | -0.009 |
le | -0.009 |
crown | -0.009 |
away | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | +6.37e-02 |
Pos logit contributions | |
aches | +0.108 |
ered | +0.104 |
le | +0.104 |
ck | +0.099 |
ling | +0.096 |
Neg logit contributions | |
She | -0.150 |
They | -0.145 |
He | -0.141 |
Un | -0.139 |
Max | -0.136 |
Token | too |
Token ID | 1165 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
aches | +0.088 |
ck | +0.081 |
le | +0.080 |
ered | +0.078 |
ling | +0.073 |
Neg logit contributions | |
She | -0.152 |
They | -0.144 |
He | -0.142 |
Maybe | -0.130 |
Max | -0.130 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
She | +0.054 |
They | +0.052 |
He | +0.051 |
Un | +0.050 |
Max | +0.050 |
Neg logit contributions | |
aches | -0.028 |
le | -0.026 |
ered | -0.024 |
away | -0.023 |
ck | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
Un | +0.216 |
Deal | +0.216 |
Jake | +0.205 |
Max | +0.203 |
Lisa | +0.200 |
Neg logit contributions | |
away | -0.486 |
up | -0.481 |
pe | -0.477 |
le | -0.476 |
aches | -0.474 |
Token |
|
Token ID | 198 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
Deal | +0.275 |
Did | +0.272 |
She | +0.268 |
Un | +0.263 |
Don | +0.255 |
Neg logit contributions | |
aches | -0.461 |
up | -0.442 |
- | -0.441 |
le | -0.435 |
away | -0.431 |
Token | The |
Token ID | 464 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Deal | +0.321 |
Did | +0.319 |
She | +0.310 |
Cookies | +0.299 |
They | +0.297 |
Neg logit contributions | |
aches | -0.454 |
- | -0.413 |
le | -0.410 |
away | -0.410 |
pe | -0.409 |
Token | him |
Token ID | 683 |
Feature activation | +7.95e-02 |
Pos logit contributions | |
aches | +0.184 |
ered | +0.152 |
le | +0.152 |
crown | +0.151 |
ig | +0.148 |
Neg logit contributions | |
She | -0.233 |
They | -0.229 |
He | -0.224 |
Max | -0.218 |
Un | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
aches | +0.113 |
le | +0.107 |
ered | +0.100 |
les | +0.098 |
ck | +0.096 |
Neg logit contributions | |
She | -0.178 |
He | -0.169 |
They | -0.168 |
Un | -0.152 |
Max | -0.151 |
Token | The |
Token ID | 383 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
Un | +0.301 |
Grand | +0.274 |
Lisa | +0.274 |
Kids | +0.263 |
Deal | +0.263 |
Neg logit contributions | |
up | -0.340 |
pe | -0.326 |
away | -0.324 |
inside | -0.318 |
in | -0.310 |
Token | little |
Token ID | 1310 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
Max | +0.110 |
She | +0.106 |
Oh | +0.106 |
Jake | +0.105 |
He | +0.104 |
Neg logit contributions | |
aches | -0.108 |
ered | -0.098 |
le | -0.097 |
away | -0.088 |
ck | -0.088 |
Token | blue |
Token ID | 4171 |
Feature activation | +5.88e-02 |
Pos logit contributions | |
aches | +0.030 |
ered | +0.027 |
le | +0.025 |
eking | +0.024 |
ck | +0.024 |
Neg logit contributions | |
She | -0.026 |
He | -0.025 |
They | -0.024 |
Wow | -0.023 |
Oh | -0.023 |
Token | bird |
Token ID | 6512 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
aches | +0.111 |
ck | +0.096 |
ered | +0.095 |
le | +0.091 |
eking | +0.090 |
Neg logit contributions | |
She | -0.118 |
He | -0.114 |
They | -0.108 |
Max | -0.105 |
Jake | -0.104 |
Token | was |
Token ID | 373 |
Feature activation | -1.76e-03 |
Pos logit contributions | |
aches | +0.222 |
coins | +0.197 |
penn | +0.197 |
le | +0.186 |
pump | +0.185 |
Neg logit contributions | |
Don | -0.166 |
Max | -0.164 |
She | -0.162 |
He | -0.154 |
Well | -0.154 |
Token | happy |
Token ID | 3772 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
She | +0.003 |
He | +0.003 |
They | +0.003 |
Don | +0.003 |
Un | +0.003 |
Neg logit contributions | |
le | -0.004 |
aches | -0.003 |
ish | -0.003 |
les | -0.003 |
ie | -0.003 |
Token | again |
Token ID | 757 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
aches | +0.042 |
le | +0.039 |
ered | +0.037 |
away | +0.034 |
ck | +0.033 |
Neg logit contributions | |
She | -0.072 |
They | -0.069 |
He | -0.068 |
Un | -0.067 |
Max | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
aches | +0.054 |
ered | +0.049 |
le | +0.048 |
crown | +0.045 |
les | +0.044 |
Neg logit contributions | |
She | -0.074 |
They | -0.072 |
Un | -0.070 |
He | -0.069 |
Not | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
Un | +0.367 |
Grand | +0.350 |
Lisa | +0.335 |
Kids | +0.330 |
Jake | +0.329 |
Neg logit contributions | |
up | -0.279 |
pe | -0.255 |
home | -0.252 |
inside | -0.247 |
in | -0.244 |
Token | icked |
Token ID | 9484 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
aches | +0.142 |
le | +0.139 |
away | +0.133 |
up | +0.131 |
ered | +0.128 |
Neg logit contributions | |
She | -0.142 |
Un | -0.135 |
Max | -0.135 |
They | -0.134 |
Jake | -0.132 |
Token | her |
Token ID | 607 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
away | +0.086 |
le | +0.082 |
up | +0.082 |
inside | +0.075 |
- | +0.074 |
Neg logit contributions | |
Un | -0.112 |
Well | -0.106 |
Deal | -0.104 |
Did | -0.104 |
Grand | -0.104 |
Token | knee |
Token ID | 10329 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
aches | +0.062 |
le | +0.057 |
ered | +0.056 |
island | +0.052 |
coins | +0.052 |
Neg logit contributions | |
She | -0.092 |
They | -0.087 |
He | -0.087 |
Deal | -0.086 |
Max | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
aches | +0.278 |
le | +0.253 |
ered | +0.248 |
island | +0.237 |
coins | +0.225 |
Neg logit contributions | |
She | -0.407 |
They | -0.377 |
He | -0.375 |
Max | -0.363 |
Un | -0.358 |
Token | make |
Token ID | 787 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
aches | +0.068 |
le | +0.066 |
ered | +0.062 |
coins | +0.061 |
away | +0.061 |
Neg logit contributions | |
She | -0.088 |
Max | -0.084 |
Un | -0.083 |
They | -0.083 |
Deal | -0.081 |
Token | it |
Token ID | 340 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
aches | +0.086 |
le | +0.083 |
ered | +0.078 |
away | +0.078 |
pe | +0.075 |
Neg logit contributions | |
She | -0.095 |
They | -0.091 |
Max | -0.089 |
Un | -0.088 |
Jake | -0.087 |
Token | feel |
Token ID | 1254 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
le | +0.198 |
aches | +0.197 |
coins | +0.179 |
ered | +0.177 |
island | +0.175 |
Neg logit contributions | |
She | -0.286 |
They | -0.264 |
He | -0.262 |
Max | -0.246 |
Un | -0.245 |
Token | better |
Token ID | 1365 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
aches | +0.122 |
le | +0.118 |
ered | +0.108 |
coins | +0.103 |
ig | +0.102 |
Neg logit contributions | |
She | -0.200 |
Don | -0.193 |
They | -0.189 |
He | -0.188 |
Un | -0.183 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
aches | +0.103 |
ered | +0.093 |
le | +0.087 |
crown | +0.083 |
les | +0.078 |
Neg logit contributions | |
She | -0.148 |
They | -0.142 |
Un | -0.137 |
He | -0.137 |
Max | -0.132 |
Token | |
Token ID | 220 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
Un | +0.308 |
Deal | +0.283 |
Jake | +0.281 |
Chris | +0.272 |
Grand | +0.272 |
Neg logit contributions | |
up | -0.309 |
away | -0.306 |
inside | -0.279 |
- | -0.270 |
again | -0.267 |
Token |
|
Token ID | 198 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Un | +0.119 |
Jake | +0.110 |
Deal | +0.110 |
Uh | +0.110 |
Really | +0.109 |
Neg logit contributions | |
up | -0.156 |
aches | -0.147 |
inside | -0.141 |
away | -0.139 |
- | -0.139 |
Token | garden |
Token ID | 11376 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
aches | +0.033 |
le | +0.033 |
ered | +0.031 |
away | +0.029 |
ck | +0.028 |
Neg logit contributions | |
She | -0.042 |
Un | -0.041 |
Max | -0.041 |
They | -0.041 |
Jake | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
aches | +0.261 |
ered | +0.248 |
le | +0.235 |
ck | +0.233 |
ling | +0.203 |
Neg logit contributions | |
They | -0.354 |
She | -0.353 |
He | -0.319 |
Maybe | -0.307 |
Did | -0.305 |
Token | It |
Token ID | 632 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
Un | +0.187 |
Lisa | +0.179 |
Jake | +0.176 |
Well | +0.166 |
Don | +0.163 |
Neg logit contributions | |
up | -0.366 |
inside | -0.364 |
le | -0.360 |
away | -0.356 |
pe | -0.347 |
Token | 's |
Token ID | 338 |
Feature activation | -6.74e-03 |
Pos logit contributions | |
aches | +0.106 |
le | +0.095 |
penn | +0.093 |
ck | +0.092 |
- | +0.091 |
Neg logit contributions | |
She | -0.095 |
They | -0.094 |
He | -0.087 |
Maybe | -0.083 |
Not | -0.082 |
Token | a |
Token ID | 257 |
Feature activation | -6.35e-03 |
Pos logit contributions | |
She | +0.014 |
They | +0.014 |
Un | +0.013 |
Max | +0.013 |
Deal | +0.013 |
Neg logit contributions | |
le | -0.011 |
aches | -0.011 |
ered | -0.010 |
ling | -0.009 |
- | -0.009 |
Token | great |
Token ID | 1049 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
Un | +0.013 |
Max | +0.012 |
She | +0.012 |
They | +0.012 |
Jake | +0.012 |
Neg logit contributions | |
le | -0.011 |
aches | -0.010 |
ered | -0.010 |
away | -0.009 |
ling | -0.009 |
Token | game |
Token ID | 983 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
aches | +0.088 |
ered | +0.085 |
le | +0.085 |
ck | +0.080 |
eking | +0.078 |
Neg logit contributions | |
They | -0.094 |
He | -0.094 |
She | -0.090 |
Today | -0.090 |
Whe | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
aches | +0.133 |
le | +0.126 |
ered | +0.118 |
ling | +0.116 |
away | +0.111 |
Neg logit contributions | |
She | -0.240 |
They | -0.240 |
Max | -0.227 |
Did | -0.223 |
Un | -0.223 |
Token | play |
Token ID | 711 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
ling | +0.031 |
lers | +0.029 |
le | +0.028 |
ering | +0.028 |
ish | +0.028 |
Neg logit contributions | |
She | -0.036 |
They | -0.035 |
Max | -0.034 |
He | -0.033 |
Sam | -0.033 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
aches | +0.103 |
ered | +0.089 |
le | +0.087 |
ck | +0.081 |
ish | +0.080 |
Neg logit contributions | |
They | -0.177 |
Max | -0.174 |
She | -0.174 |
He | -0.166 |
Jake | -0.164 |
Token |
|
Token ID | 198 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
Deal | +0.192 |
Un | +0.190 |
Grand | +0.169 |
Jake | +0.168 |
Lisa | +0.167 |
Neg logit contributions | |
away | -0.373 |
pe | -0.368 |
le | -0.364 |
aches | -0.360 |
- | -0.360 |
Token | man |
Token ID | 582 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
She | +0.023 |
Un | +0.022 |
They | +0.022 |
Max | +0.022 |
Jake | +0.022 |
Neg logit contributions | |
aches | -0.017 |
le | -0.017 |
away | -0.016 |
ered | -0.016 |
up | -0.015 |
Token | came |
Token ID | 1625 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
aches | +0.279 |
le | +0.229 |
ered | +0.224 |
penn | +0.223 |
crown | +0.222 |
Neg logit contributions | |
She | -0.207 |
Un | -0.206 |
Max | -0.205 |
Did | -0.197 |
Jake | -0.197 |
Token | and |
Token ID | 290 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
aches | +0.096 |
le | +0.080 |
crown | +0.076 |
coins | +0.074 |
ered | +0.073 |
Neg logit contributions | |
She | -0.138 |
They | -0.135 |
Un | -0.130 |
He | -0.128 |
Max | -0.127 |
Token | shouted |
Token ID | 17293 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
She | +0.034 |
Un | +0.033 |
Max | +0.033 |
They | +0.033 |
Jake | +0.032 |
Neg logit contributions | |
aches | -0.027 |
le | -0.027 |
away | -0.025 |
ered | -0.025 |
up | -0.024 |
Token | at |
Token ID | 379 |
Feature activation | -1.30e-03 |
Pos logit contributions | |
aches | +0.017 |
le | +0.017 |
away | +0.015 |
ered | +0.015 |
up | +0.014 |
Neg logit contributions | |
She | -0.047 |
Un | -0.047 |
They | -0.046 |
Max | -0.046 |
Jake | -0.045 |
Token | them |
Token ID | 606 |
Feature activation | +8.91e-02 |
Pos logit contributions | |
She | +0.002 |
Un | +0.002 |
Deal | +0.002 |
Did | +0.002 |
Max | +0.002 |
Neg logit contributions | |
aches | -0.003 |
le | -0.003 |
away | -0.002 |
ered | -0.002 |
up | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
aches | +0.100 |
le | +0.095 |
ered | +0.082 |
away | +0.081 |
crown | +0.077 |
Neg logit contributions | |
She | -0.233 |
They | -0.228 |
Un | -0.220 |
Max | -0.218 |
He | -0.218 |
Token | He |
Token ID | 679 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
Deal | +0.301 |
Un | +0.272 |
Lisa | +0.268 |
Jake | +0.257 |
Emily | +0.254 |
Neg logit contributions | |
up | -0.374 |
pe | -0.363 |
inside | -0.354 |
- | -0.350 |
away | -0.349 |
Token | was |
Token ID | 373 |
Feature activation | -2.91e-04 |
Pos logit contributions | |
She | +0.026 |
Un | +0.026 |
They | +0.025 |
Max | +0.025 |
Jake | +0.025 |
Neg logit contributions | |
aches | -0.021 |
le | -0.021 |
away | -0.020 |
ered | -0.019 |
up | -0.019 |
Token | very |
Token ID | 845 |
Feature activation | -8.38e-03 |
Pos logit contributions | |
She | +0.001 |
Un | +0.001 |
They | +0.001 |
Max | +0.001 |
Jake | +0.001 |
Neg logit contributions | |
le | -0.000 |
aches | -0.000 |
ered | -0.000 |
away | -0.000 |
ling | -0.000 |
Token | angry |
Token ID | 7954 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
She | +0.017 |
Un | +0.017 |
They | +0.016 |
Max | +0.016 |
Jake | +0.016 |
Neg logit contributions | |
le | -0.014 |
aches | -0.014 |
ered | -0.013 |
away | -0.013 |
up | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | -4.99e-02 |
Pos logit contributions | |
aches | +0.005 |
le | +0.004 |
ered | +0.004 |
away | +0.004 |
crown | +0.004 |
Neg logit contributions | |
She | -0.009 |
They | -0.008 |
Max | -0.008 |
Un | -0.008 |
Well | -0.008 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
Did | +0.061 |
Well | +0.060 |
Kim | +0.058 |
Un | +0.057 |
Don | +0.057 |
Neg logit contributions | |
le | -0.118 |
up | -0.117 |
away | -0.117 |
aches | -0.112 |
pe | -0.111 |
Token | feel |
Token ID | 1254 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
aches | +0.052 |
le | +0.050 |
ered | +0.048 |
- | +0.048 |
ie | +0.046 |
Neg logit contributions | |
She | -0.051 |
They | -0.049 |
Deal | -0.046 |
Max | -0.046 |
Something | -0.045 |
Token | sorry |
Token ID | 7926 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
aches | +0.046 |
le | +0.046 |
ered | +0.043 |
away | +0.042 |
- | +0.041 |
Neg logit contributions | |
She | -0.054 |
They | -0.053 |
Un | -0.052 |
Max | -0.051 |
Don | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
le | +0.015 |
aches | +0.015 |
away | +0.014 |
up | +0.013 |
ered | +0.013 |
Neg logit contributions | |
Un | -0.047 |
She | -0.047 |
Max | -0.046 |
Jake | -0.046 |
They | -0.046 |
Token | They |
Token ID | 1119 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
Johnny | +0.360 |
Kim | +0.350 |
Ol | +0.346 |
Chris | +0.340 |
Tyler | +0.339 |
Neg logit contributions | |
up | -0.382 |
in | -0.371 |
inside | -0.367 |
. | -0.355 |
, | -0.341 |
Token | say |
Token ID | 910 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
She | +0.090 |
Un | +0.089 |
Max | +0.087 |
They | +0.087 |
Jake | +0.086 |
Neg logit contributions | |
aches | -0.076 |
le | -0.076 |
away | -0.072 |
ered | -0.071 |
up | -0.069 |
Token | sorry |
Token ID | 7926 |
Feature activation | -3.50e-03 |
Pos logit contributions | |
She | +0.054 |
Un | +0.052 |
They | +0.051 |
Max | +0.051 |
Jake | +0.050 |
Neg logit contributions | |
aches | -0.076 |
le | -0.075 |
away | -0.072 |
ered | -0.071 |
up | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | +2.29e-03 |
Pos logit contributions | |
She | +0.008 |
They | +0.008 |
Un | +0.008 |
Max | +0.008 |
Jake | +0.008 |
Neg logit contributions | |
aches | -0.004 |
le | -0.004 |
ered | -0.004 |
away | -0.004 |
up | -0.004 |
Token | Mom |
Token ID | 11254 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
le | +0.004 |
away | +0.004 |
up | +0.004 |
aches | +0.004 |
pe | +0.004 |
Neg logit contributions | |
Did | -0.004 |
Well | -0.004 |
Don | -0.004 |
Un | -0.004 |
She | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
aches | +0.035 |
ered | +0.033 |
le | +0.032 |
ck | +0.030 |
away | +0.029 |
Neg logit contributions | |
She | -0.059 |
They | -0.057 |
He | -0.053 |
Max | -0.053 |
Un | -0.052 |
Token | big |
Token ID | 1263 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
aches | +0.001 |
le | +0.001 |
ered | +0.001 |
away | +0.001 |
ling | +0.001 |
Neg logit contributions | |
She | -0.001 |
Un | -0.001 |
They | -0.001 |
Max | -0.001 |
Jake | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -7.41e-02 |
Pos logit contributions | |
aches | +0.044 |
le | +0.040 |
ered | +0.039 |
ck | +0.037 |
ling | +0.034 |
Neg logit contributions | |
She | -0.066 |
They | -0.063 |
He | -0.061 |
Max | -0.060 |
Un | -0.060 |
Token | green |
Token ID | 4077 |
Feature activation | +6.51e-02 |
Pos logit contributions | |
She | +0.152 |
Un | +0.150 |
They | +0.148 |
Max | +0.147 |
Jake | +0.143 |
Neg logit contributions | |
aches | -0.120 |
le | -0.119 |
ered | -0.114 |
away | -0.107 |
up | -0.103 |
Token | forest |
Token ID | 8222 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
aches | +0.104 |
ered | +0.097 |
le | +0.095 |
ck | +0.094 |
away | +0.087 |
Neg logit contributions | |
She | -0.142 |
They | -0.136 |
He | -0.131 |
Max | -0.129 |
Did | -0.127 |
Token | ! |
Token ID | 0 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
aches | +0.116 |
le | +0.112 |
ered | +0.106 |
ck | +0.097 |
away | +0.095 |
Neg logit contributions | |
She | -0.217 |
They | -0.217 |
Un | -0.206 |
He | -0.203 |
Max | -0.202 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.25e-02 |
Pos logit contributions | |
Deal | +0.390 |
Jake | +0.372 |
Sorry | +0.364 |
Un | +0.361 |
Okay | +0.361 |
Neg logit contributions | |
home | -0.331 |
again | -0.318 |
up | -0.315 |
in | -0.311 |
away | -0.309 |
Token | Jack |
Token ID | 14295 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
Un | +0.101 |
Deal | +0.099 |
Jake | +0.090 |
Grand | +0.090 |
She | +0.089 |
Neg logit contributions | |
le | -0.158 |
up | -0.156 |
away | -0.155 |
inside | -0.154 |
pe | -0.148 |
Token | was |
Token ID | 373 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
aches | +0.025 |
le | +0.024 |
away | +0.023 |
ered | +0.023 |
up | +0.021 |
Neg logit contributions | |
She | -0.036 |
They | -0.034 |
Un | -0.034 |
Max | -0.034 |
Jake | -0.033 |
Token | at |
Token ID | 379 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
She | +0.065 |
They | +0.061 |
Un | +0.060 |
Max | +0.059 |
Deal | +0.058 |
Neg logit contributions | |
aches | -0.049 |
le | -0.049 |
ered | -0.044 |
away | -0.042 |
les | -0.041 |
Token | the |
Token ID | 262 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
She | +0.103 |
Un | +0.101 |
They | +0.100 |
Max | +0.099 |
Jake | +0.098 |
Neg logit contributions | |
aches | -0.053 |
le | -0.052 |
away | -0.048 |
ered | -0.047 |
up | -0.045 |
Token | beach |
Token ID | 10481 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
She | +0.107 |
Un | +0.106 |
Deal | +0.105 |
They | +0.104 |
Max | +0.104 |
Neg logit contributions | |
aches | -0.067 |
le | -0.066 |
away | -0.064 |
up | -0.062 |
pe | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
ered | +0.037 |
le | +0.037 |
away | +0.036 |
ling | +0.036 |
ering | +0.036 |
Neg logit contributions | |
She | -0.038 |
They | -0.036 |
Cookies | -0.032 |
Deal | -0.031 |
Now | -0.031 |
Token | please |
Token ID | 3387 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
She | +0.090 |
They | +0.086 |
Un | +0.086 |
Max | +0.083 |
Deal | +0.081 |
Neg logit contributions | |
aches | -0.126 |
le | -0.126 |
ered | -0.121 |
away | -0.119 |
ling | -0.116 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
le | +0.025 |
ling | +0.025 |
aches | +0.024 |
lers | +0.024 |
ered | +0.023 |
Neg logit contributions | |
She | -0.025 |
They | -0.025 |
Un | -0.024 |
Deal | -0.023 |
Did | -0.023 |
Token | The |
Token ID | 383 |
Feature activation | -8.57e-02 |
Pos logit contributions | |
Un | +0.150 |
Deal | +0.147 |
Jake | +0.139 |
Max | +0.136 |
Lisa | +0.130 |
Neg logit contributions | |
aches | -0.422 |
le | -0.417 |
away | -0.416 |
pe | -0.407 |
up | -0.405 |
Token | cow |
Token ID | 9875 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
She | +0.177 |
Un | +0.174 |
They | +0.171 |
Max | +0.169 |
Deal | +0.164 |
Neg logit contributions | |
le | -0.138 |
aches | -0.138 |
ered | -0.131 |
away | -0.128 |
up | -0.122 |
Token | was |
Token ID | 373 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
aches | +0.171 |
le | +0.161 |
away | +0.156 |
ered | +0.155 |
coins | +0.150 |
Neg logit contributions | |
She | -0.153 |
They | -0.146 |
Max | -0.140 |
He | -0.139 |
Well | -0.135 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.22e-03 |
Pos logit contributions | |
She | +0.058 |
They | +0.056 |
Un | +0.054 |
He | +0.053 |
Deal | +0.052 |
Neg logit contributions | |
le | -0.056 |
aches | -0.052 |
les | -0.050 |
ered | -0.049 |
ig | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
aches | +0.002 |
le | +0.002 |
ered | +0.002 |
ling | +0.002 |
away | +0.002 |
Neg logit contributions | |
She | -0.003 |
They | -0.003 |
He | -0.003 |
Un | -0.002 |
Jake | -0.002 |
Token | help |
Token ID | 1037 |
Feature activation | -1.14e-03 |
Pos logit contributions | |
She | +0.024 |
They | +0.023 |
Un | +0.022 |
Deal | +0.022 |
Max | +0.021 |
Neg logit contributions | |
le | -0.020 |
aches | -0.020 |
ling | -0.020 |
up | -0.019 |
away | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
She | +0.003 |
They | +0.003 |
Un | +0.003 |
He | +0.003 |
Not | +0.002 |
Neg logit contributions | |
aches | -0.001 |
le | -0.001 |
ered | -0.001 |
crown | -0.001 |
away | -0.001 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
They | +0.092 |
She | +0.091 |
He | +0.083 |
Un | +0.080 |
Don | +0.078 |
Neg logit contributions | |
le | -0.094 |
ling | -0.087 |
ig | -0.087 |
aches | -0.086 |
eking | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
She | +0.080 |
They | +0.076 |
Max | +0.075 |
Un | +0.074 |
Jake | +0.073 |
Neg logit contributions | |
aches | -0.063 |
le | -0.059 |
ered | -0.056 |
away | -0.055 |
ling | -0.053 |
Token | But |
Token ID | 887 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
Deal | +0.334 |
Un | +0.316 |
Well | +0.303 |
Looks | +0.285 |
Oh | +0.283 |
Neg logit contributions | |
up | -0.363 |
away | -0.352 |
pe | -0.339 |
in | -0.338 |
inside | -0.331 |
Token | her |
Token ID | 607 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
She | +0.156 |
Un | +0.155 |
They | +0.151 |
Max | +0.150 |
Deal | +0.148 |
Neg logit contributions | |
aches | -0.165 |
le | -0.163 |
away | -0.159 |
ered | -0.154 |
up | -0.153 |
Token | brother |
Token ID | 3956 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
She | +0.067 |
Un | +0.067 |
They | +0.065 |
Well | +0.064 |
Max | +0.064 |
Neg logit contributions | |
le | -0.058 |
aches | -0.058 |
away | -0.057 |
up | -0.056 |
- | -0.054 |
Token | said |
Token ID | 531 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
aches | +0.080 |
le | +0.079 |
away | +0.076 |
ered | +0.074 |
up | +0.073 |
Neg logit contributions | |
She | -0.091 |
Un | -0.090 |
They | -0.088 |
Max | -0.088 |
Jake | -0.087 |
Token | no |
Token ID | 645 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
She | +0.071 |
They | +0.067 |
Un | +0.067 |
Max | +0.067 |
Did | +0.066 |
Neg logit contributions | |
le | -0.067 |
aches | -0.066 |
away | -0.063 |
ered | -0.062 |
ling | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
She | +0.027 |
They | +0.026 |
Un | +0.026 |
Max | +0.025 |
Jake | +0.025 |
Neg logit contributions | |
aches | -0.011 |
le | -0.010 |
ered | -0.010 |
away | -0.010 |
up | -0.009 |
Token | kept |
Token ID | 4030 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
She | +0.069 |
Un | +0.066 |
They | +0.066 |
Max | +0.065 |
Jake | +0.063 |
Neg logit contributions | |
aches | -0.095 |
le | -0.094 |
away | -0.088 |
ered | -0.088 |
up | -0.087 |
Token | playing |
Token ID | 2712 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
le | +0.034 |
aches | +0.034 |
ered | +0.030 |
an | +0.029 |
ish | +0.029 |
Neg logit contributions | |
She | -0.038 |
Un | -0.038 |
Max | -0.037 |
Jake | -0.037 |
They | -0.037 |
Token | alone |
Token ID | 3436 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
aches | +0.073 |
le | +0.064 |
ered | +0.063 |
ck | +0.058 |
pe | +0.056 |
Neg logit contributions | |
She | -0.138 |
Max | -0.134 |
They | -0.134 |
He | -0.131 |
Jake | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
aches | +0.057 |
le | +0.049 |
ered | +0.048 |
ck | +0.046 |
crown | +0.044 |
Neg logit contributions | |
She | -0.099 |
They | -0.094 |
He | -0.092 |
Max | -0.091 |
Un | -0.089 |
Token | things |
Token ID | 1243 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
aches | +0.073 |
le | +0.072 |
away | +0.068 |
ered | +0.067 |
pe | +0.065 |
Neg logit contributions | |
She | -0.057 |
Un | -0.054 |
They | -0.054 |
Max | -0.053 |
Jake | -0.052 |
Token | right |
Token ID | 826 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
aches | +0.155 |
le | +0.151 |
ered | +0.142 |
ig | +0.140 |
island | +0.135 |
Neg logit contributions | |
She | -0.285 |
They | -0.266 |
He | -0.262 |
Max | -0.250 |
Don | -0.249 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
aches | +0.068 |
crown | +0.054 |
coins | +0.052 |
pe | +0.052 |
le | +0.051 |
Neg logit contributions | |
She | -0.142 |
They | -0.138 |
He | -0.131 |
Max | -0.131 |
Don | -0.126 |
Token | So |
Token ID | 1406 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
Jake | +0.255 |
Lisa | +0.246 |
Grand | +0.238 |
Un | +0.219 |
Deal | +0.208 |
Neg logit contributions | |
up | -0.438 |
away | -0.436 |
home | -0.412 |
ish | -0.411 |
- | -0.410 |
Token | she |
Token ID | 673 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
She | +0.175 |
Un | +0.167 |
They | +0.167 |
Max | +0.165 |
Did | +0.161 |
Neg logit contributions | |
aches | -0.119 |
le | -0.118 |
ered | -0.107 |
away | -0.106 |
up | -0.102 |
Token | went |
Token ID | 1816 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
She | +0.039 |
They | +0.038 |
Max | +0.038 |
Sam | +0.037 |
Un | +0.036 |
Neg logit contributions | |
aches | -0.043 |
le | -0.042 |
ering | -0.039 |
eking | -0.039 |
ling | -0.039 |
Token | over |
Token ID | 625 |
Feature activation | +4.52e-02 |
Pos logit contributions | |
le | +0.033 |
aches | +0.030 |
ered | +0.027 |
ck | +0.027 |
ish | +0.026 |
Neg logit contributions | |
She | -0.048 |
They | -0.045 |
He | -0.043 |
Max | -0.042 |
Un | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
aches | +0.060 |
le | +0.058 |
ered | +0.053 |
away | +0.053 |
ling | +0.050 |
Neg logit contributions | |
She | -0.107 |
Un | -0.104 |
They | -0.103 |
Max | -0.103 |
Jake | -0.101 |
Token | Sammy |
Token ID | 44067 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
aches | +0.049 |
le | +0.049 |
away | +0.047 |
ered | +0.046 |
up | +0.046 |
Neg logit contributions | |
She | -0.029 |
Un | -0.029 |
They | -0.028 |
Max | -0.028 |
Jake | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
aches | +0.143 |
le | +0.132 |
ered | +0.130 |
crown | +0.114 |
ck | +0.113 |
Neg logit contributions | |
She | -0.184 |
They | -0.172 |
He | -0.166 |
Max | -0.160 |
Un | -0.151 |
Token | gave |
Token ID | 2921 |
Feature activation | +5.52e-03 |
Pos logit contributions | |
She | +0.067 |
Un | +0.064 |
They | +0.063 |
Max | +0.063 |
Don | +0.062 |
Neg logit contributions | |
le | -0.068 |
aches | -0.066 |
ling | -0.060 |
ered | -0.059 |
les | -0.059 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
Un | +0.067 |
Well | +0.065 |
Jake | +0.064 |
Did | +0.064 |
She | +0.063 |
Neg logit contributions | |
aches | -0.075 |
le | -0.072 |
away | -0.071 |
up | -0.071 |
pe | -0.069 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
aches | +0.116 |
ered | +0.113 |
le | +0.110 |
coins | +0.106 |
away | +0.106 |
Neg logit contributions | |
She | -0.083 |
Un | -0.075 |
They | -0.075 |
Grand | -0.075 |
Max | -0.075 |
Token | she |
Token ID | 673 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
Deal | +0.256 |
Un | +0.255 |
Jake | +0.244 |
Max | +0.239 |
Well | +0.236 |
Neg logit contributions | |
aches | -0.234 |
le | -0.228 |
away | -0.228 |
up | -0.227 |
pe | -0.221 |
Token | asked |
Token ID | 1965 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
She | +0.061 |
Un | +0.059 |
Max | +0.059 |
They | +0.058 |
Jake | +0.057 |
Neg logit contributions | |
aches | -0.050 |
le | -0.050 |
away | -0.047 |
ered | -0.046 |
up | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
She | +0.074 |
They | +0.070 |
Max | +0.069 |
Un | +0.068 |
Jake | +0.066 |
Neg logit contributions | |
aches | -0.051 |
le | -0.047 |
ered | -0.044 |
away | -0.043 |
ling | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Deal | +0.297 |
Un | +0.274 |
Jake | +0.271 |
Well | +0.265 |
Chris | +0.258 |
Neg logit contributions | |
up | -0.389 |
inside | -0.364 |
away | -0.363 |
pe | -0.362 |
- | -0.361 |
Token |
|
Token ID | 198 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
Deal | +0.225 |
Un | +0.213 |
Did | +0.203 |
Well | +0.200 |
Don | +0.200 |
Neg logit contributions | |
le | -0.362 |
- | -0.360 |
up | -0.359 |
aches | -0.354 |
away | -0.353 |
Token | " |
Token ID | 1 |
Feature activation | -9.14e-02 |
Pos logit contributions | |
Deal | +0.305 |
Did | +0.293 |
Well | +0.293 |
She | +0.287 |
They | +0.284 |
Neg logit contributions | |
aches | -0.298 |
- | -0.291 |
away | -0.287 |
le | -0.286 |
up | -0.281 |
Token | Sure |
Token ID | 19457 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
Deal | +0.121 |
They | +0.112 |
He | +0.109 |
She | +0.108 |
Jake | +0.107 |
Neg logit contributions | |
pe | -0.197 |
aches | -0.197 |
away | -0.197 |
le | -0.193 |
- | -0.190 |
Token | , |
Token ID | 11 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
ling | +0.014 |
le | +0.014 |
aches | +0.014 |
les | +0.013 |
ered | +0.012 |
Neg logit contributions | |
She | -0.025 |
They | -0.023 |
Un | -0.022 |
He | -0.022 |
Max | -0.022 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
She | +0.098 |
They | +0.094 |
Max | +0.093 |
Un | +0.092 |
Jake | +0.091 |
Neg logit contributions | |
aches | -0.098 |
le | -0.097 |
away | -0.094 |
ered | -0.092 |
ling | -0.090 |
Token | friendship |
Token ID | 14738 |
Feature activation | +5.21e-02 |
Pos logit contributions | |
She | +0.040 |
They | +0.038 |
Max | +0.038 |
Un | +0.038 |
Jake | +0.038 |
Neg logit contributions | |
aches | -0.032 |
le | -0.031 |
ered | -0.029 |
ling | -0.028 |
away | -0.028 |
Token | more |
Token ID | 517 |
Feature activation | +5.98e-03 |
Pos logit contributions | |
aches | +0.080 |
ered | +0.077 |
ck | +0.071 |
le | +0.069 |
eking | +0.069 |
Neg logit contributions | |
She | -0.116 |
They | -0.115 |
He | -0.114 |
Max | -0.108 |
Un | -0.105 |
Token | than |
Token ID | 621 |
Feature activation | -4.06e-03 |
Pos logit contributions | |
ered | +0.014 |
aches | +0.013 |
eking | +0.012 |
le | +0.012 |
ish | +0.012 |
Neg logit contributions | |
She | -0.008 |
They | -0.007 |
He | -0.007 |
And | -0.007 |
Not | -0.007 |
Token | tw |
Token ID | 665 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
He | +0.008 |
She | +0.008 |
They | +0.007 |
Not | +0.007 |
Un | +0.007 |
Neg logit contributions | |
le | -0.007 |
aches | -0.007 |
an | -0.007 |
ered | -0.007 |
ish | -0.007 |
Token | igs |
Token ID | 9235 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
away | +0.084 |
inside | +0.078 |
up | +0.075 |
le | +0.073 |
island | +0.072 |
Neg logit contributions | |
She | -0.119 |
Jake | -0.118 |
Well | -0.114 |
Un | -0.114 |
Sam | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
ered | +0.208 |
aches | +0.193 |
le | +0.178 |
ck | +0.177 |
ig | +0.173 |
Neg logit contributions | |
She | -0.244 |
They | -0.242 |
He | -0.232 |
Not | -0.200 |
Max | -0.198 |
Token | He |
Token ID | 679 |
Feature activation | -7.42e-02 |
Pos logit contributions | |
Uh | +0.359 |
Kids | +0.358 |
Un | +0.342 |
liv | +0.332 |
ksh | +0.325 |
Neg logit contributions | |
inside | -0.430 |
up | -0.430 |
in | -0.407 |
. | -0.402 |
, | -0.391 |
Token | learned |
Token ID | 4499 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
She | +0.119 |
They | +0.115 |
Un | +0.114 |
Max | +0.113 |
Well | +0.112 |
Neg logit contributions | |
aches | -0.149 |
le | -0.146 |
away | -0.139 |
ered | -0.137 |
up | -0.136 |
Token | a |
Token ID | 257 |
Feature activation | -7.64e-02 |
Pos logit contributions | |
She | +0.135 |
Un | +0.131 |
They | +0.130 |
Max | +0.128 |
Jake | +0.126 |
Neg logit contributions | |
le | -0.086 |
aches | -0.085 |
away | -0.078 |
ered | -0.077 |
ling | -0.074 |
Token | hard |
Token ID | 1327 |
Feature activation | +6.79e-03 |
Pos logit contributions | |
She | +0.149 |
Un | +0.149 |
Jake | +0.148 |
Max | +0.147 |
Deal | +0.145 |
Neg logit contributions | |
aches | -0.124 |
away | -0.120 |
le | -0.119 |
up | -0.116 |
- | -0.115 |
Token | lesson |
Token ID | 11483 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
aches | +0.012 |
le | +0.012 |
away | +0.011 |
ered | +0.011 |
up | +0.011 |
Neg logit contributions | |
She | -0.013 |
Un | -0.012 |
They | -0.012 |
Max | -0.012 |
Jake | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
She | +0.053 |
They | +0.050 |
Un | +0.049 |
Max | +0.048 |
Don | +0.048 |
Neg logit contributions | |
le | -0.046 |
aches | -0.045 |
away | -0.041 |
ered | -0.041 |
ling | -0.041 |
Token | started |
Token ID | 2067 |
Feature activation | +7.75e-03 |
Pos logit contributions | |
She | +0.026 |
Max | +0.025 |
They | +0.024 |
Jake | +0.024 |
Don | +0.023 |
Neg logit contributions | |
aches | -0.034 |
le | -0.033 |
away | -0.031 |
up | -0.030 |
coins | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
le | +0.014 |
aches | +0.014 |
ered | +0.012 |
ish | +0.012 |
an | +0.012 |
Neg logit contributions | |
She | -0.016 |
They | -0.015 |
Don | -0.014 |
Un | -0.014 |
Max | -0.014 |
Token | cry |
Token ID | 3960 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
aches | +0.044 |
le | +0.044 |
coins | +0.041 |
- | +0.040 |
ered | +0.039 |
Neg logit contributions | |
She | -0.040 |
They | -0.037 |
Max | -0.036 |
Un | -0.035 |
Don | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.93e-01 |
Pos logit contributions | |
aches | +0.046 |
le | +0.045 |
away | +0.041 |
ered | +0.039 |
up | +0.038 |
Neg logit contributions | |
She | -0.135 |
Un | -0.134 |
They | -0.132 |
Max | -0.132 |
Jake | -0.130 |
Token |
|
Token ID | 198 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
Deal | +0.344 |
Un | +0.329 |
Jake | +0.317 |
Chris | +0.304 |
Tyler | +0.299 |
Neg logit contributions | |
up | -0.355 |
inside | -0.319 |
pe | -0.318 |
away | -0.315 |
- | -0.310 |
Token |
|
Token ID | 198 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Deal | +0.266 |
Un | +0.242 |
Did | +0.241 |
Really | +0.233 |
Bye | +0.229 |
Neg logit contributions | |
- | -0.307 |
up | -0.305 |
aches | -0.299 |
ered | -0.296 |
away | -0.294 |
Token | Just |
Token ID | 5703 |
Feature activation | -7.84e-02 |
Pos logit contributions | |
Deal | +0.248 |
Did | +0.242 |
They | +0.227 |
Sure | +0.224 |
Cookies | +0.219 |
Neg logit contributions | |
aches | -0.343 |
- | -0.334 |
ered | -0.325 |
away | -0.323 |
le | -0.319 |
Token | then |
Token ID | 788 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
She | +0.121 |
Un | +0.117 |
Max | +0.115 |
Don | +0.113 |
They | +0.109 |
Neg logit contributions | |
aches | -0.170 |
ered | -0.148 |
le | -0.148 |
crown | -0.145 |
coins | -0.145 |
Token | , |
Token ID | 11 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
She | +0.074 |
Un | +0.072 |
Max | +0.071 |
They | +0.071 |
Jake | +0.070 |
Neg logit contributions | |
aches | -0.023 |
le | -0.021 |
away | -0.018 |
ered | -0.018 |
up | -0.017 |
Token | her |
Token ID | 607 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
She | +0.150 |
Un | +0.144 |
Max | +0.143 |
They | +0.143 |
Jake | +0.139 |
Neg logit contributions | |
aches | -0.121 |
le | -0.118 |
away | -0.109 |
ered | -0.109 |
ling | -0.106 |
Token | something |
Token ID | 1223 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
She | +0.111 |
Un | +0.109 |
They | +0.109 |
Max | +0.106 |
Jake | +0.104 |
Neg logit contributions | |
le | -0.108 |
aches | -0.108 |
away | -0.100 |
ered | -0.100 |
ling | -0.097 |
Token | shiny |
Token ID | 22441 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
aches | +0.063 |
le | +0.059 |
ish | +0.059 |
ig | +0.057 |
away | +0.056 |
Neg logit contributions | |
She | -0.061 |
They | -0.059 |
Don | -0.058 |
Well | -0.057 |
Un | -0.057 |
Token | caught |
Token ID | 4978 |
Feature activation | +8.32e-02 |
Pos logit contributions | |
aches | +0.228 |
le | +0.197 |
ck | +0.196 |
ered | +0.195 |
crown | +0.194 |
Neg logit contributions | |
She | -0.151 |
They | -0.150 |
He | -0.139 |
Did | -0.132 |
Max | -0.129 |
Token | their |
Token ID | 511 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
aches | +0.119 |
le | +0.104 |
crown | +0.100 |
ered | +0.098 |
ck | +0.098 |
Neg logit contributions | |
She | -0.183 |
They | -0.182 |
Un | -0.175 |
Max | -0.173 |
He | -0.168 |
Token | eye |
Token ID | 4151 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
aches | +0.060 |
le | +0.055 |
ered | +0.054 |
ck | +0.052 |
ling | +0.049 |
Neg logit contributions | |
He | -0.067 |
They | -0.067 |
She | -0.065 |
Grand | -0.064 |
Something | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
aches | +0.174 |
ck | +0.163 |
ered | +0.149 |
le | +0.144 |
ishing | +0.142 |
Neg logit contributions | |
She | -0.275 |
They | -0.251 |
He | -0.249 |
Did | -0.239 |
Max | -0.232 |
Token | They |
Token ID | 1119 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
Deal | +0.297 |
Un | +0.281 |
Jake | +0.274 |
Tyler | +0.271 |
Lisa | +0.270 |
Neg logit contributions | |
up | -0.412 |
inside | -0.396 |
pe | -0.383 |
in | -0.379 |
away | -0.372 |
Token | looked |
Token ID | 3114 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
She | +0.065 |
They | +0.062 |
Un | +0.061 |
Max | +0.061 |
Jake | +0.060 |
Neg logit contributions | |
aches | -0.066 |
le | -0.065 |
away | -0.060 |
- | -0.060 |
up | -0.060 |
Token | around |
Token ID | 1088 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
aches | +0.064 |
ered | +0.049 |
le | +0.048 |
an | +0.048 |
ck | +0.048 |
Neg logit contributions | |
She | -0.095 |
They | -0.092 |
He | -0.090 |
Did | -0.085 |
Don | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
aches | +0.064 |
le | +0.060 |
ered | +0.055 |
away | +0.054 |
up | +0.050 |
Neg logit contributions | |
She | -0.140 |
They | -0.136 |
Un | -0.136 |
Max | -0.134 |
Jake | -0.133 |
Token | found |
Token ID | 1043 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
She | +0.092 |
They | +0.088 |
Jake | +0.088 |
Max | +0.088 |
Un | +0.087 |
Neg logit contributions | |
aches | -0.081 |
le | -0.075 |
ered | -0.070 |
- | -0.068 |
away | -0.068 |
Token | be |
Token ID | 307 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
She | +0.035 |
Un | +0.034 |
They | +0.034 |
Max | +0.034 |
Jake | +0.033 |
Neg logit contributions | |
aches | -0.026 |
le | -0.026 |
away | -0.024 |
ered | -0.023 |
up | -0.023 |
Token | found |
Token ID | 1043 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
aches | +0.024 |
le | +0.023 |
ish | +0.021 |
coins | +0.021 |
ered | +0.021 |
Neg logit contributions | |
She | -0.028 |
Max | -0.028 |
Lisa | -0.027 |
Un | -0.027 |
Sam | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
aches | +0.050 |
le | +0.044 |
ck | +0.042 |
ling | +0.040 |
ered | +0.040 |
Neg logit contributions | |
She | -0.078 |
They | -0.071 |
He | -0.070 |
Don | -0.068 |
Max | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
Un | +0.419 |
Chris | +0.392 |
Deal | +0.390 |
Jake | +0.389 |
Josh | +0.388 |
Neg logit contributions | |
up | -0.353 |
inside | -0.327 |
in | -0.306 |
away | -0.299 |
again | -0.295 |
Token |
|
Token ID | 198 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
Deal | +0.289 |
Un | +0.271 |
Did | +0.253 |
Really | +0.250 |
Bye | +0.244 |
Neg logit contributions | |
up | -0.380 |
- | -0.378 |
le | -0.374 |
inside | -0.373 |
away | -0.367 |
Token | Just |
Token ID | 5703 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
Deal | +0.261 |
Did | +0.235 |
Un | +0.220 |
She | +0.219 |
Well | +0.218 |
Neg logit contributions | |
- | -0.423 |
away | -0.414 |
le | -0.412 |
aches | -0.412 |
up | -0.406 |
Token | then |
Token ID | 788 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
She | +0.198 |
Un | +0.188 |
They | +0.185 |
Max | +0.180 |
Don | +0.179 |
Neg logit contributions | |
aches | -0.191 |
le | -0.174 |
ered | -0.171 |
away | -0.163 |
ling | -0.162 |
Token | , |
Token ID | 11 |
Feature activation | -9.81e-02 |
Pos logit contributions | |
Un | +0.158 |
She | +0.156 |
Deal | +0.155 |
Jake | +0.155 |
Well | +0.155 |
Neg logit contributions | |
le | -0.028 |
aches | -0.027 |
away | -0.026 |
up | -0.024 |
- | -0.023 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
Un | +0.173 |
Deal | +0.169 |
She | +0.166 |
Jake | +0.164 |
Max | +0.162 |
Neg logit contributions | |
away | -0.180 |
le | -0.178 |
aches | -0.176 |
up | -0.174 |
inside | -0.171 |
Token | 's |
Token ID | 338 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
She | +0.029 |
Un | +0.028 |
Max | +0.028 |
They | +0.028 |
Jake | +0.028 |
Neg logit contributions | |
aches | -0.020 |
le | -0.019 |
away | -0.018 |
ered | -0.018 |
up | -0.017 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
She | +0.080 |
Max | +0.078 |
Un | +0.076 |
They | +0.075 |
Jake | +0.075 |
Neg logit contributions | |
aches | -0.078 |
le | -0.076 |
ered | -0.072 |
away | -0.069 |
ck | -0.068 |