Token | , |
Token ID | 11 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
azz | +0.258 |
pillar | +0.240 |
och | +0.231 |
affe | +0.229 |
pp | +0.228 |
Neg logit contributions | |
the | -0.124 |
not | -0.123 |
an | -0.117 |
making | -0.114 |
many | -0.106 |
Token | they |
Token ID | 484 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
not | +0.060 |
an | +0.059 |
making | +0.059 |
the | +0.059 |
many | +0.056 |
Neg logit contributions | |
azz | -0.167 |
pillar | -0.156 |
och | -0.152 |
hell | -0.151 |
enny | -0.149 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +7.14e+00 |
Pos logit contributions | |
azz | +0.144 |
pillar | +0.131 |
och | +0.126 |
enny | +0.124 |
pp | +0.124 |
Neg logit contributions | |
not | -0.137 |
the | -0.136 |
an | -0.133 |
making | -0.132 |
many | -0.126 |
Token | 't |
Token ID | 470 |
Feature activation | +7.95e+00 |
Pos logit contributions | |
not | +0.122 |
the | +0.120 |
an | +0.112 |
making | +0.109 |
going | +0.099 |
Neg logit contributions | |
azz | -0.390 |
pillar | -0.364 |
och | -0.358 |
pp | -0.357 |
affe | -0.353 |
Token | have |
Token ID | 423 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
not | +0.157 |
the | +0.154 |
an | +0.147 |
making | +0.147 |
many | +0.134 |
Neg logit contributions | |
azz | -0.409 |
pillar | -0.383 |
och | -0.374 |
enny | -0.368 |
pp | -0.368 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.48e+01 |
Pos logit contributions | |
making | +0.155 |
not | +0.151 |
adding | +0.141 |
saving | +0.137 |
climbing | +0.137 |
Neg logit contributions | |
azz | -0.518 |
hell | -0.500 |
pillar | -0.483 |
y | -0.479 |
enny | -0.477 |
Token | broken |
Token ID | 5445 |
Feature activation | +4.96e+00 |
Pos logit contributions | |
not | +0.282 |
making | +0.269 |
an | +0.258 |
the | +0.255 |
adding | +0.252 |
Neg logit contributions | |
azz | -0.689 |
hell | -0.642 |
pillar | -0.641 |
y | -0.640 |
och | -0.638 |
Token | . |
Token ID | 13 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
the | +0.128 |
not | +0.126 |
an | +0.122 |
making | +0.121 |
going | +0.113 |
Neg logit contributions | |
azz | -0.231 |
pillar | -0.214 |
och | -0.208 |
affe | -0.205 |
enny | -0.204 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.31e+01 |
Pos logit contributions | |
azz | +0.218 |
pp | +0.208 |
pillar | +0.206 |
hell | +0.206 |
ppy | +0.202 |
Neg logit contributions | |
the | -0.096 |
not | -0.091 |
an | -0.089 |
many | -0.080 |
making | -0.079 |
Token | my |
Token ID | 1820 |
Feature activation | -9.12e+00 |
Pos logit contributions | |
azz | +0.370 |
och | +0.344 |
pillar | +0.319 |
hell | +0.311 |
pp | +0.307 |
Neg logit contributions | |
the | -0.584 |
not | -0.577 |
an | -0.546 |
making | -0.538 |
his | -0.525 |
Token | realized |
Token ID | 6939 |
Feature activation | +6.37e+00 |
Pos logit contributions | |
azz | +0.377 |
pillar | +0.349 |
och | +0.333 |
enny | +0.332 |
affe | +0.329 |
Neg logit contributions | |
not | -0.263 |
the | -0.261 |
making | -0.256 |
an | -0.256 |
going | -0.242 |
Token | so |
Token ID | 523 |
Feature activation | +8.59e+00 |
Pos logit contributions | |
making | +0.065 |
climbing | +0.061 |
adding | +0.060 |
digging | +0.058 |
going | +0.056 |
Neg logit contributions | |
azz | -0.304 |
y | -0.287 |
hell | -0.280 |
pillar | -0.279 |
enny | -0.278 |
Token | they |
Token ID | 484 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
making | +0.162 |
not | +0.155 |
an | +0.150 |
the | +0.146 |
going | +0.145 |
Neg logit contributions | |
azz | -0.427 |
pillar | -0.396 |
och | -0.389 |
hell | -0.384 |
enny | -0.384 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +6.95e+00 |
Pos logit contributions | |
azz | +0.131 |
pillar | +0.121 |
och | +0.117 |
enny | +0.115 |
pp | +0.114 |
Neg logit contributions | |
not | -0.103 |
the | -0.102 |
an | -0.097 |
making | -0.097 |
going | -0.093 |
Token | 't |
Token ID | 470 |
Feature activation | +7.81e+00 |
Pos logit contributions | |
not | +0.127 |
the | +0.125 |
an | +0.118 |
making | +0.117 |
going | +0.106 |
Neg logit contributions | |
azz | -0.368 |
pillar | -0.345 |
och | -0.337 |
pp | -0.333 |
enny | -0.332 |
Token | have |
Token ID | 423 |
Feature activation | +1.03e+01 |
Pos logit contributions | |
not | +0.154 |
the | +0.149 |
making | +0.141 |
an | +0.139 |
going | +0.129 |
Neg logit contributions | |
azz | -0.408 |
pillar | -0.383 |
och | -0.375 |
enny | -0.370 |
pp | -0.369 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.46e+01 |
Pos logit contributions | |
making | +0.203 |
adding | +0.201 |
guessing | +0.192 |
digging | +0.190 |
testing | +0.190 |
Neg logit contributions | |
azz | -0.408 |
ird | -0.407 |
hell | -0.396 |
y | -0.375 |
pillar | -0.373 |
Token | in |
Token ID | 287 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
not | +0.267 |
adding | +0.263 |
making | +0.254 |
many | +0.247 |
conquering | +0.246 |
Neg logit contributions | |
azz | -0.620 |
y | -0.612 |
och | -0.582 |
hell | -0.582 |
pillar | -0.579 |
Token | trouble |
Token ID | 5876 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
not | +0.134 |
making | +0.115 |
adding | +0.108 |
going | +0.104 |
an | +0.103 |
Neg logit contributions | |
azz | -0.555 |
pillar | -0.525 |
hell | -0.525 |
y | -0.516 |
och | -0.514 |
Token | . |
Token ID | 13 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
not | +0.077 |
the | +0.074 |
an | +0.074 |
making | +0.072 |
many | +0.070 |
Neg logit contributions | |
azz | -0.117 |
pillar | -0.109 |
hell | -0.106 |
och | -0.106 |
enny | -0.103 |
Token | From |
Token ID | 3574 |
Feature activation | +5.63e+00 |
Pos logit contributions | |
azz | +0.199 |
pillar | +0.187 |
pp | +0.183 |
och | +0.182 |
hell | +0.181 |
Neg logit contributions | |
the | -0.095 |
not | -0.095 |
an | -0.091 |
making | -0.087 |
many | -0.084 |
Token | then |
Token ID | 788 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
not | +0.104 |
making | +0.102 |
an | +0.096 |
adding | +0.094 |
the | +0.094 |
Neg logit contributions | |
azz | -0.274 |
pillar | -0.259 |
och | -0.256 |
itty | -0.248 |
affe | -0.246 |
Token | dark |
Token ID | 3223 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
not | +0.005 |
the | +0.005 |
making | +0.005 |
an | +0.005 |
going | +0.005 |
Neg logit contributions | |
azz | -0.007 |
pillar | -0.006 |
hell | -0.006 |
enny | -0.006 |
och | -0.006 |
Token | tunnel |
Token ID | 13275 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
azz | +0.221 |
och | +0.205 |
pp | +0.203 |
pillar | +0.203 |
affe | +0.198 |
Neg logit contributions | |
the | -0.121 |
not | -0.114 |
an | -0.113 |
making | -0.111 |
many | -0.107 |
Token | and |
Token ID | 290 |
Feature activation | +5.91e+00 |
Pos logit contributions | |
azz | +0.458 |
pillar | +0.425 |
och | +0.410 |
hell | +0.408 |
enny | +0.408 |
Neg logit contributions | |
not | -0.255 |
an | -0.246 |
the | -0.243 |
making | -0.240 |
going | -0.226 |
Token | hadn |
Token ID | 8020 |
Feature activation | +8.19e+00 |
Pos logit contributions | |
not | +0.139 |
an | +0.137 |
adding | +0.137 |
example | +0.136 |
making | +0.131 |
Neg logit contributions | |
azz | -0.217 |
pillar | -0.202 |
enny | -0.196 |
y | -0.191 |
pp | -0.189 |
Token | 't |
Token ID | 470 |
Feature activation | +8.37e+00 |
Pos logit contributions | |
not | +0.160 |
the | +0.156 |
making | +0.151 |
an | +0.150 |
many | +0.138 |
Neg logit contributions | |
azz | -0.420 |
pillar | -0.393 |
och | -0.382 |
enny | -0.379 |
hell | -0.375 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.45e+01 |
Pos logit contributions | |
not | +0.242 |
making | +0.236 |
adding | +0.231 |
an | +0.225 |
helping | +0.223 |
Neg logit contributions | |
azz | -0.311 |
pillar | -0.287 |
hell | -0.275 |
ppy | -0.271 |
enny | -0.271 |
Token | lost |
Token ID | 2626 |
Feature activation | +5.40e+00 |
Pos logit contributions | |
adding | +0.367 |
example | +0.345 |
warnings | +0.343 |
making | +0.332 |
measuring | +0.332 |
Neg logit contributions | |
azz | -0.513 |
y | -0.511 |
ppy | -0.494 |
enny | -0.482 |
ety | -0.468 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
not | +0.139 |
making | +0.134 |
the | +0.131 |
an | +0.129 |
adding | +0.125 |
Neg logit contributions | |
azz | -0.230 |
pillar | -0.214 |
hell | -0.210 |
och | -0.205 |
enny | -0.205 |
Token | He |
Token ID | 679 |
Feature activation | -5.72e+00 |
Pos logit contributions | |
azz | +0.201 |
pillar | +0.186 |
enny | +0.181 |
och | +0.178 |
hell | +0.176 |
Neg logit contributions | |
not | -0.118 |
making | -0.116 |
the | -0.115 |
an | -0.111 |
going | -0.107 |
Token | had |
Token ID | 550 |
Feature activation | +8.35e+00 |
Pos logit contributions | |
azz | +0.214 |
pillar | +0.202 |
ppy | +0.188 |
enny | +0.188 |
hell | +0.186 |
Neg logit contributions | |
making | -0.181 |
not | -0.179 |
an | -0.175 |
the | -0.173 |
adding | -0.169 |
Token | succeeded |
Token ID | 14131 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
adding | +0.199 |
making | +0.191 |
measuring | +0.187 |
testing | +0.183 |
counting | +0.179 |
Neg logit contributions | |
azz | -0.331 |
pillar | -0.311 |
enny | -0.308 |
ppy | -0.302 |
hell | -0.296 |
Token | , |
Token ID | 11 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
the | +0.056 |
not | +0.056 |
an | +0.053 |
making | +0.053 |
going | +0.049 |
Neg logit contributions | |
azz | -0.115 |
pillar | -0.106 |
och | -0.103 |
pp | -0.103 |
hell | -0.102 |
Token | she |
Token ID | 673 |
Feature activation | -5.89e+00 |
Pos logit contributions | |
not | +0.070 |
making | +0.069 |
an | +0.066 |
the | +0.066 |
saving | +0.065 |
Neg logit contributions | |
azz | -0.130 |
pillar | -0.122 |
enny | -0.117 |
och | -0.117 |
hell | -0.115 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +6.59e+00 |
Pos logit contributions | |
azz | +0.241 |
pillar | +0.219 |
och | +0.216 |
pp | +0.210 |
enny | +0.209 |
Neg logit contributions | |
not | -0.183 |
the | -0.180 |
an | -0.173 |
making | -0.172 |
many | -0.163 |
Token | 't |
Token ID | 470 |
Feature activation | +7.37e+00 |
Pos logit contributions | |
not | +0.118 |
the | +0.117 |
an | +0.110 |
making | +0.108 |
going | +0.098 |
Neg logit contributions | |
azz | -0.352 |
pillar | -0.329 |
och | -0.322 |
pp | -0.319 |
hell | -0.317 |
Token | have |
Token ID | 423 |
Feature activation | +1.00e+01 |
Pos logit contributions | |
not | +0.151 |
the | +0.150 |
an | +0.146 |
making | +0.144 |
many | +0.133 |
Neg logit contributions | |
azz | -0.369 |
pillar | -0.345 |
och | -0.336 |
enny | -0.332 |
hell | -0.331 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.44e+01 |
Pos logit contributions | |
making | +0.148 |
not | +0.146 |
adding | +0.144 |
saving | +0.138 |
testing | +0.135 |
Neg logit contributions | |
azz | -0.492 |
hell | -0.467 |
enny | -0.466 |
pillar | -0.465 |
ppy | -0.456 |
Token | so |
Token ID | 523 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
not | +0.284 |
making | +0.270 |
adding | +0.265 |
many | +0.252 |
an | +0.250 |
Neg logit contributions | |
azz | -0.645 |
pillar | -0.611 |
och | -0.606 |
enny | -0.604 |
y | -0.604 |
Token | dirty |
Token ID | 11841 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
making | +0.223 |
not | +0.217 |
the | +0.213 |
an | +0.206 |
adding | +0.201 |
Neg logit contributions | |
azz | -0.475 |
pillar | -0.453 |
och | -0.439 |
enny | -0.438 |
hell | -0.437 |
Token | and |
Token ID | 290 |
Feature activation | +5.96e+00 |
Pos logit contributions | |
the | +0.058 |
not | +0.057 |
making | +0.054 |
an | +0.054 |
going | +0.051 |
Neg logit contributions | |
azz | -0.115 |
pillar | -0.106 |
och | -0.103 |
affe | -0.103 |
pp | -0.102 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
an | +0.125 |
adding | +0.118 |
not | +0.117 |
making | +0.116 |
the | +0.116 |
Neg logit contributions | |
azz | -0.248 |
pillar | -0.240 |
och | -0.235 |
y | -0.233 |
enny | -0.232 |
Token | 't |
Token ID | 470 |
Feature activation | +7.93e+00 |
Pos logit contributions | |
not | +0.132 |
the | +0.130 |
an | +0.124 |
making | +0.123 |
many | +0.111 |
Neg logit contributions | |
azz | -0.381 |
pillar | -0.357 |
och | -0.349 |
pp | -0.344 |
enny | -0.344 |
Token | was |
Token ID | 373 |
Feature activation | +9.98e+00 |
Pos logit contributions | |
azz | +0.248 |
pillar | +0.229 |
och | +0.220 |
enny | +0.218 |
pp | +0.216 |
Neg logit contributions | |
not | -0.181 |
the | -0.179 |
making | -0.176 |
an | -0.176 |
many | -0.166 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
many | +0.173 |
the | +0.171 |
making | +0.170 |
an | +0.169 |
help | +0.166 |
Neg logit contributions | |
azz | -0.428 |
ird | -0.425 |
och | -0.405 |
pillar | -0.398 |
hell | -0.397 |
Token | she |
Token ID | 673 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
an | +0.046 |
not | +0.045 |
many | +0.043 |
the | +0.042 |
making | +0.042 |
Neg logit contributions | |
azz | -0.129 |
pillar | -0.122 |
enny | -0.119 |
y | -0.117 |
och | -0.117 |
Token | hadn |
Token ID | 8020 |
Feature activation | +7.70e+00 |
Pos logit contributions | |
azz | +0.233 |
pillar | +0.211 |
och | +0.205 |
hell | +0.202 |
affe | +0.202 |
Neg logit contributions | |
not | -0.184 |
the | -0.183 |
making | -0.177 |
an | -0.175 |
going | -0.167 |
Token | 't |
Token ID | 470 |
Feature activation | +8.14e+00 |
Pos logit contributions | |
not | +0.140 |
the | +0.139 |
an | +0.132 |
making | +0.127 |
going | +0.117 |
Neg logit contributions | |
azz | -0.411 |
pillar | -0.385 |
och | -0.376 |
pp | -0.375 |
affe | -0.373 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.44e+01 |
Pos logit contributions | |
not | +0.226 |
the | +0.216 |
an | +0.216 |
making | +0.214 |
many | +0.212 |
Neg logit contributions | |
azz | -0.319 |
pillar | -0.296 |
enny | -0.286 |
ird | -0.286 |
hell | -0.285 |
Token | into |
Token ID | 656 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
not | +0.219 |
warnings | +0.203 |
making | +0.199 |
adding | +0.194 |
conquering | +0.193 |
Neg logit contributions | |
azz | -0.643 |
y | -0.638 |
och | -0.629 |
pillar | -0.619 |
enny | -0.614 |
Token | trouble |
Token ID | 5876 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
not | +0.163 |
making | +0.140 |
going | +0.133 |
adding | +0.130 |
climbing | +0.127 |
Neg logit contributions | |
azz | -0.537 |
pillar | -0.529 |
hell | -0.525 |
ird | -0.521 |
och | -0.515 |
Token | for |
Token ID | 329 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
not | +0.058 |
the | +0.057 |
an | +0.057 |
making | +0.054 |
many | +0.053 |
Neg logit contributions | |
azz | -0.097 |
pillar | -0.091 |
hell | -0.089 |
och | -0.089 |
enny | -0.087 |
Token | playing |
Token ID | 2712 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
biggest | +0.105 |
up | +0.104 |
not | +0.100 |
same | +0.094 |
Listen | +0.093 |
Neg logit contributions | |
ird | -0.542 |
hell | -0.505 |
och | -0.500 |
azz | -0.499 |
ppy | -0.486 |
Token | with |
Token ID | 351 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
not | +0.094 |
the | +0.093 |
an | +0.089 |
making | +0.088 |
many | +0.081 |
Neg logit contributions | |
azz | -0.242 |
pillar | -0.227 |
och | -0.221 |
enny | -0.218 |
pp | -0.218 |
Token | best |
Token ID | 1266 |
Feature activation | -6.04e-01 |
Pos logit contributions | |
not | +0.070 |
making | +0.066 |
an | +0.065 |
the | +0.065 |
many | +0.060 |
Neg logit contributions | |
azz | -0.204 |
pillar | -0.190 |
och | -0.187 |
pp | -0.183 |
enny | -0.182 |
Token | hug |
Token ID | 16225 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
azz | +0.027 |
pillar | +0.025 |
affe | +0.024 |
hell | +0.024 |
och | +0.024 |
Neg logit contributions | |
not | -0.015 |
the | -0.015 |
an | -0.014 |
making | -0.014 |
many | -0.014 |
Token | I |
Token ID | 314 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
the | +0.015 |
not | +0.015 |
an | +0.014 |
making | +0.014 |
many | +0.013 |
Neg logit contributions | |
azz | -0.034 |
pillar | -0.032 |
och | -0.031 |
enny | -0.031 |
pp | -0.031 |
Token | 've |
Token ID | 1053 |
Feature activation | +7.53e+00 |
Pos logit contributions | |
not | +0.001 |
the | +0.001 |
making | +0.001 |
an | +0.001 |
going | +0.001 |
Neg logit contributions | |
azz | -0.002 |
pillar | -0.002 |
och | -0.002 |
enny | -0.002 |
hell | -0.002 |
Token | ever |
Token ID | 1683 |
Feature activation | +7.97e+00 |
Pos logit contributions | |
not | +0.193 |
the | +0.188 |
an | +0.186 |
making | +0.184 |
many | +0.178 |
Neg logit contributions | |
azz | -0.323 |
pillar | -0.305 |
pp | -0.292 |
och | -0.290 |
hell | -0.289 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.42e+01 |
Pos logit contributions | |
the | +0.249 |
not | +0.248 |
an | +0.241 |
making | +0.240 |
many | +0.228 |
Neg logit contributions | |
azz | -0.319 |
pillar | -0.291 |
och | -0.285 |
enny | -0.278 |
affe | -0.275 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
not | +0.244 |
making | +0.224 |
an | +0.206 |
adding | +0.205 |
the | +0.204 |
Neg logit contributions | |
azz | -0.712 |
pillar | -0.681 |
och | -0.660 |
hell | -0.658 |
y | -0.656 |
Token |
|
Token ID | 198 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
azz | +0.182 |
pillar | +0.173 |
och | +0.164 |
enny | +0.163 |
affe | +0.161 |
Neg logit contributions | |
not | -0.132 |
an | -0.125 |
making | -0.125 |
the | -0.122 |
going | -0.122 |
Token |
|
Token ID | 198 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
azz | +0.168 |
pillar | +0.159 |
enny | +0.152 |
och | +0.151 |
affe | +0.148 |
Neg logit contributions | |
not | -0.105 |
making | -0.102 |
an | -0.101 |
the | -0.099 |
going | -0.097 |
Token | Bear |
Token ID | 36352 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
azz | +0.134 |
pillar | +0.126 |
och | +0.118 |
enny | +0.117 |
affe | +0.114 |
Neg logit contributions | |
not | -0.160 |
the | -0.157 |
an | -0.157 |
making | -0.156 |
going | -0.151 |
Token | y |
Token ID | 88 |
Feature activation | -8.11e+00 |
Pos logit contributions | |
azz | +0.324 |
och | +0.295 |
ird | +0.271 |
pillar | +0.263 |
hell | +0.260 |
Neg logit contributions | |
the | -0.673 |
not | -0.671 |
an | -0.634 |
making | -0.628 |
his | -0.622 |
Token | the |
Token ID | 262 |
Feature activation | +6.72e-01 |
Pos logit contributions | |
not | +0.080 |
making | +0.073 |
going | +0.063 |
adding | +0.063 |
climbing | +0.056 |
Neg logit contributions | |
azz | -0.545 |
pillar | -0.523 |
hell | -0.501 |
och | -0.500 |
y | -0.498 |
Token | jewelry |
Token ID | 22634 |
Feature activation | -1.06e+01 |
Pos logit contributions | |
not | +0.018 |
the | +0.017 |
making | +0.017 |
an | +0.016 |
going | +0.016 |
Neg logit contributions | |
pillar | -0.028 |
azz | -0.028 |
itty | -0.026 |
och | -0.025 |
affe | -0.025 |
Token | box |
Token ID | 3091 |
Feature activation | -8.85e+00 |
Pos logit contributions | |
azz | +0.527 |
pillar | +0.483 |
och | +0.471 |
enny | +0.468 |
hell | +0.464 |
Neg logit contributions | |
not | -0.240 |
the | -0.238 |
making | -0.229 |
an | -0.225 |
many | -0.208 |
Token | and |
Token ID | 290 |
Feature activation | +6.56e+00 |
Pos logit contributions | |
azz | +0.416 |
pillar | +0.387 |
och | +0.378 |
enny | +0.372 |
pp | +0.371 |
Neg logit contributions | |
not | -0.205 |
the | -0.203 |
an | -0.197 |
making | -0.195 |
many | -0.182 |
Token | had |
Token ID | 550 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
not | +0.160 |
an | +0.159 |
the | +0.157 |
making | +0.154 |
many | +0.148 |
Neg logit contributions | |
azz | -0.271 |
pillar | -0.256 |
och | -0.244 |
enny | -0.242 |
hell | -0.242 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.42e+01 |
Pos logit contributions | |
making | +0.146 |
adding | +0.140 |
not | +0.134 |
testing | +0.133 |
helping | +0.130 |
Neg logit contributions | |
azz | -0.498 |
pillar | -0.471 |
hell | -0.469 |
lets | -0.466 |
ppy | -0.455 |
Token | in |
Token ID | 287 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
not | +0.233 |
making | +0.220 |
adding | +0.208 |
many | +0.199 |
an | +0.197 |
Neg logit contributions | |
azz | -0.681 |
och | -0.646 |
pp | -0.633 |
pillar | -0.632 |
y | -0.629 |
Token | through |
Token ID | 832 |
Feature activation | +8.66e+00 |
Pos logit contributions | |
not | +0.112 |
making | +0.102 |
adding | +0.091 |
going | +0.090 |
many | +0.086 |
Neg logit contributions | |
azz | -0.562 |
pillar | -0.532 |
hell | -0.524 |
och | -0.521 |
y | -0.515 |
Token | a |
Token ID | 257 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
not | +0.086 |
making | +0.071 |
going | +0.064 |
an | +0.061 |
the | +0.056 |
Neg logit contributions | |
azz | -0.498 |
pillar | -0.484 |
hell | -0.467 |
och | -0.461 |
enny | -0.461 |
Token | tiny |
Token ID | 7009 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
the | +0.058 |
not | +0.056 |
making | +0.054 |
an | +0.054 |
many | +0.053 |
Neg logit contributions | |
azz | -0.118 |
pp | -0.107 |
pillar | -0.105 |
och | -0.105 |
enny | -0.105 |
Token | little |
Token ID | 1310 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
azz | +0.097 |
och | +0.088 |
pp | +0.088 |
enny | +0.085 |
ppy | +0.084 |
Neg logit contributions | |
the | -0.046 |
an | -0.040 |
making | -0.039 |
many | -0.039 |
to | -0.038 |
Token | list |
Token ID | 1351 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
not | +0.053 |
the | +0.052 |
an | +0.051 |
making | +0.050 |
many | +0.047 |
Neg logit contributions | |
azz | -0.099 |
pillar | -0.093 |
och | -0.090 |
hell | -0.089 |
enny | -0.088 |
Token | of |
Token ID | 286 |
Feature activation | +9.66e+00 |
Pos logit contributions | |
azz | +0.154 |
pillar | +0.146 |
pp | +0.137 |
affe | +0.136 |
och | +0.136 |
Neg logit contributions | |
the | -0.076 |
not | -0.074 |
making | -0.070 |
an | -0.068 |
to | -0.068 |
Token | children |
Token ID | 1751 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
not | +0.183 |
an | +0.171 |
making | +0.171 |
the | +0.163 |
adding | +0.157 |
Neg logit contributions | |
azz | -0.437 |
hell | -0.429 |
pillar | -0.423 |
och | -0.412 |
pp | -0.406 |
Token | who |
Token ID | 508 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
azz | +0.565 |
pillar | +0.531 |
och | +0.519 |
enny | +0.512 |
pp | +0.510 |
Neg logit contributions | |
not | -0.161 |
the | -0.158 |
an | -0.149 |
making | -0.148 |
many | -0.132 |
Token | had |
Token ID | 550 |
Feature activation | +9.99e+00 |
Pos logit contributions | |
not | +0.051 |
the | +0.050 |
an | +0.049 |
making | +0.048 |
many | +0.046 |
Neg logit contributions | |
azz | -0.076 |
pillar | -0.070 |
och | -0.068 |
enny | -0.067 |
pp | -0.067 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.41e+01 |
Pos logit contributions | |
making | +0.167 |
climbing | +0.162 |
not | +0.162 |
adding | +0.161 |
testing | +0.158 |
Neg logit contributions | |
azz | -0.450 |
hell | -0.436 |
pillar | -0.426 |
ird | -0.423 |
y | -0.419 |
Token | sick |
Token ID | 6639 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
not | +0.221 |
making | +0.212 |
climbing | +0.199 |
adding | +0.197 |
testing | +0.190 |
Neg logit contributions | |
azz | -0.692 |
y | -0.659 |
hell | -0.655 |
och | -0.653 |
pp | -0.653 |
Token | from |
Token ID | 422 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
the | +0.061 |
not | +0.057 |
making | +0.056 |
an | +0.055 |
to | +0.053 |
Neg logit contributions | |
azz | -0.131 |
pillar | -0.118 |
pp | -0.118 |
och | -0.116 |
enny | -0.115 |
Token | eating |
Token ID | 6600 |
Feature activation | +7.89e+00 |
Pos logit contributions | |
not | +0.153 |
an | +0.140 |
making | +0.132 |
many | +0.129 |
facing | +0.123 |
Neg logit contributions | |
azz | -0.519 |
y | -0.494 |
pillar | -0.491 |
och | -0.487 |
ppy | -0.483 |
Token | too |
Token ID | 1165 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
not | +0.174 |
the | +0.164 |
making | +0.164 |
an | +0.162 |
going | +0.151 |
Neg logit contributions | |
azz | -0.372 |
pillar | -0.351 |
och | -0.343 |
hell | -0.342 |
pp | -0.338 |
Token | much |
Token ID | 881 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
not | +0.057 |
making | +0.057 |
the | +0.056 |
an | +0.055 |
going | +0.051 |
Neg logit contributions | |
azz | -0.125 |
pillar | -0.117 |
hell | -0.115 |
och | -0.114 |
enny | -0.113 |
Token | 't |
Token ID | 470 |
Feature activation | +7.15e+00 |
Pos logit contributions | |
making | +0.168 |
the | +0.168 |
not | +0.163 |
an | +0.160 |
saving | +0.159 |
Neg logit contributions | |
azz | -0.366 |
pillar | -0.339 |
hell | -0.330 |
enny | -0.327 |
och | -0.324 |
Token | notice |
Token ID | 4003 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
not | +0.162 |
the | +0.155 |
making | +0.142 |
an | +0.140 |
going | +0.136 |
Neg logit contributions | |
azz | -0.360 |
pillar | -0.347 |
och | -0.339 |
pp | -0.335 |
itty | -0.333 |
Token | her |
Token ID | 607 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
not | +0.179 |
the | +0.170 |
making | +0.167 |
an | +0.167 |
many | +0.152 |
Neg logit contributions | |
azz | -0.448 |
pillar | -0.418 |
och | -0.406 |
enny | -0.401 |
hell | -0.400 |
Token | dress |
Token ID | 6576 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
azz | +0.019 |
pillar | +0.017 |
och | +0.017 |
enny | +0.017 |
hell | +0.017 |
Neg logit contributions | |
not | -0.008 |
the | -0.008 |
an | -0.008 |
making | -0.008 |
many | -0.007 |
Token | had |
Token ID | 550 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
azz | +0.194 |
pillar | +0.185 |
och | +0.179 |
affe | +0.177 |
enny | +0.176 |
Neg logit contributions | |
not | -0.096 |
the | -0.092 |
making | -0.088 |
an | -0.087 |
going | -0.082 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.41e+01 |
Pos logit contributions | |
not | +0.239 |
the | +0.236 |
making | +0.235 |
an | +0.226 |
helping | +0.218 |
Neg logit contributions | |
azz | -0.481 |
pillar | -0.440 |
hell | -0.436 |
enny | -0.425 |
och | -0.424 |
Token | caught |
Token ID | 4978 |
Feature activation | +7.48e+00 |
Pos logit contributions | |
not | +0.313 |
the | +0.304 |
making | +0.296 |
an | +0.294 |
many | +0.272 |
Neg logit contributions | |
azz | -0.677 |
pillar | -0.626 |
och | -0.616 |
hell | -0.608 |
pp | -0.605 |
Token | on |
Token ID | 319 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
the | +0.123 |
not | +0.117 |
an | +0.114 |
making | +0.110 |
going | +0.100 |
Neg logit contributions | |
azz | -0.423 |
pillar | -0.394 |
och | -0.387 |
enny | -0.384 |
affe | -0.382 |
Token | a |
Token ID | 257 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
not | +0.156 |
making | +0.142 |
an | +0.138 |
the | +0.137 |
going | +0.125 |
Neg logit contributions | |
azz | -0.581 |
pillar | -0.552 |
hell | -0.537 |
och | -0.532 |
y | -0.526 |
Token | branch |
Token ID | 8478 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
the | +0.082 |
not | +0.080 |
making | +0.076 |
an | +0.074 |
many | +0.072 |
Neg logit contributions | |
azz | -0.154 |
enny | -0.140 |
pp | -0.139 |
och | -0.136 |
ppy | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
azz | +0.207 |
pillar | +0.192 |
enny | +0.189 |
och | +0.188 |
pp | +0.186 |
Neg logit contributions | |
not | -0.099 |
the | -0.099 |
making | -0.095 |
an | -0.092 |
going | -0.087 |
Token | poisonous |
Token ID | 32199 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
an | +0.115 |
the | +0.113 |
making | +0.107 |
not | +0.106 |
climbing | +0.101 |
Neg logit contributions | |
azz | -0.478 |
hell | -0.452 |
ppy | -0.450 |
pillar | -0.449 |
och | -0.444 |
Token | ! |
Token ID | 0 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
azz | +0.030 |
och | +0.027 |
pillar | +0.027 |
enny | +0.027 |
pp | +0.026 |
Neg logit contributions | |
the | -0.016 |
not | -0.015 |
making | -0.015 |
an | -0.015 |
many | -0.013 |
Token | You |
Token ID | 921 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
azz | +0.085 |
pp | +0.081 |
pillar | +0.080 |
hell | +0.080 |
och | +0.079 |
Neg logit contributions | |
not | -0.052 |
the | -0.051 |
an | -0.049 |
making | -0.047 |
many | -0.045 |
Token | could |
Token ID | 714 |
Feature activation | +8.53e+00 |
Pos logit contributions | |
azz | +0.073 |
pillar | +0.066 |
och | +0.065 |
enny | +0.064 |
pp | +0.063 |
Neg logit contributions | |
not | -0.059 |
the | -0.058 |
an | -0.055 |
making | -0.054 |
many | -0.052 |
Token | have |
Token ID | 423 |
Feature activation | +9.32e+00 |
Pos logit contributions | |
an | +0.206 |
the | +0.204 |
making | +0.200 |
many | +0.199 |
not | +0.193 |
Neg logit contributions | |
azz | -0.354 |
pillar | -0.339 |
hell | -0.324 |
ppy | -0.320 |
itty | -0.318 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.41e+01 |
Pos logit contributions | |
making | +0.189 |
hearing | +0.186 |
testing | +0.186 |
conquering | +0.185 |
exploring | +0.183 |
Neg logit contributions | |
hell | -0.368 |
azz | -0.363 |
pillar | -0.358 |
ird | -0.354 |
y | -0.354 |
Token | very |
Token ID | 845 |
Feature activation | +9.07e+00 |
Pos logit contributions | |
not | +0.246 |
making | +0.241 |
hearing | +0.233 |
adding | +0.231 |
conquering | +0.231 |
Neg logit contributions | |
azz | -0.628 |
y | -0.614 |
pillar | -0.611 |
hell | -0.609 |
pp | -0.603 |
Token | sick |
Token ID | 6639 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
making | +0.192 |
adding | +0.177 |
hearing | +0.176 |
exploring | +0.173 |
counting | +0.173 |
Neg logit contributions | |
hell | -0.392 |
y | -0.379 |
azz | -0.376 |
pillar | -0.371 |
pp | -0.370 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
the | +0.062 |
not | +0.061 |
making | +0.059 |
an | +0.058 |
many | +0.054 |
Neg logit contributions | |
azz | -0.124 |
pillar | -0.113 |
och | -0.112 |
pp | -0.111 |
enny | -0.110 |
Token |
|
Token ID | 198 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
azz | +0.223 |
pillar | +0.208 |
och | +0.202 |
pp | +0.201 |
hell | +0.199 |
Neg logit contributions | |
the | -0.111 |
not | -0.110 |
an | -0.105 |
making | -0.104 |
many | -0.098 |
Token |
|
Token ID | 198 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
azz | +0.203 |
pillar | +0.191 |
pp | +0.189 |
och | +0.187 |
affe | +0.184 |
Neg logit contributions | |
the | -0.089 |
not | -0.084 |
an | -0.080 |
many | -0.078 |
making | -0.076 |
Token | found |
Token ID | 1043 |
Feature activation | +7.45e+00 |
Pos logit contributions | |
not | +0.195 |
making | +0.190 |
the | +0.186 |
an | +0.185 |
going | +0.180 |
Neg logit contributions | |
azz | -0.268 |
pillar | -0.247 |
enny | -0.239 |
och | -0.238 |
pp | -0.230 |
Token | a |
Token ID | 257 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
not | +0.074 |
making | +0.069 |
going | +0.056 |
adding | +0.053 |
taking | +0.051 |
Neg logit contributions | |
azz | -0.433 |
pillar | -0.413 |
och | -0.403 |
hell | -0.400 |
enny | -0.400 |
Token | baby |
Token ID | 5156 |
Feature activation | -1.13e+01 |
Pos logit contributions | |
not | +0.044 |
the | +0.043 |
an | +0.042 |
making | +0.042 |
many | +0.038 |
Neg logit contributions | |
azz | -0.107 |
pillar | -0.100 |
och | -0.097 |
enny | -0.096 |
pp | -0.096 |
Token | bunny |
Token ID | 44915 |
Feature activation | -1.10e+01 |
Pos logit contributions | |
azz | +0.477 |
och | +0.412 |
enny | +0.412 |
pp | +0.412 |
pillar | +0.406 |
Neg logit contributions | |
the | -0.360 |
making | -0.346 |
not | -0.345 |
an | -0.340 |
going | -0.324 |
Token | had |
Token ID | 550 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
azz | +0.550 |
och | +0.492 |
pillar | +0.489 |
pp | +0.488 |
enny | +0.485 |
Neg logit contributions | |
the | -0.257 |
not | -0.245 |
making | -0.239 |
an | -0.237 |
going | -0.220 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
not | +0.244 |
the | +0.241 |
making | +0.241 |
an | +0.230 |
many | +0.222 |
Neg logit contributions | |
azz | -0.446 |
pillar | -0.422 |
hell | -0.407 |
enny | -0.394 |
och | -0.394 |
Token | stuck |
Token ID | 7819 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
not | +0.256 |
making | +0.231 |
the | +0.214 |
an | +0.213 |
many | +0.213 |
Neg logit contributions | |
azz | -0.677 |
pillar | -0.642 |
och | -0.636 |
hell | -0.635 |
y | -0.626 |
Token | inside |
Token ID | 2641 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
not | +0.055 |
the | +0.052 |
making | +0.050 |
an | +0.050 |
many | +0.047 |
Neg logit contributions | |
azz | -0.140 |
pillar | -0.131 |
och | -0.128 |
enny | -0.127 |
pp | -0.127 |
Token | ! |
Token ID | 0 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
not | +0.079 |
making | +0.072 |
an | +0.072 |
the | +0.069 |
many | +0.068 |
Neg logit contributions | |
azz | -0.137 |
pillar | -0.129 |
hell | -0.126 |
och | -0.126 |
y | -0.123 |
Token | The |
Token ID | 383 |
Feature activation | -2.86e+00 |
Pos logit contributions | |
azz | +0.163 |
pillar | +0.152 |
och | +0.148 |
enny | +0.146 |
affe | +0.145 |
Neg logit contributions | |
not | -0.070 |
the | -0.069 |
making | -0.066 |
an | -0.066 |
many | -0.061 |
Token | farmer |
Token ID | 18739 |
Feature activation | -1.58e+01 |
Pos logit contributions | |
azz | +0.113 |
pillar | +0.109 |
affe | +0.102 |
och | +0.101 |
itty | +0.099 |
Neg logit contributions | |
not | -0.087 |
the | -0.083 |
making | -0.083 |
an | -0.083 |
going | -0.078 |
Token | her |
Token ID | 607 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
making | +0.094 |
adding | +0.090 |
not | +0.089 |
climbing | +0.089 |
many | +0.088 |
Neg logit contributions | |
azz | -0.118 |
pillar | -0.107 |
och | -0.107 |
enny | -0.105 |
pp | -0.103 |
Token | t |
Token ID | 256 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
azz | +0.090 |
pillar | +0.085 |
och | +0.082 |
affe | +0.081 |
enny | +0.081 |
Neg logit contributions | |
not | -0.038 |
making | -0.037 |
the | -0.037 |
an | -0.035 |
going | -0.033 |
Token | eddy |
Token ID | 21874 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
azz | +0.147 |
hell | +0.147 |
pillar | +0.140 |
suit | +0.134 |
affe | +0.126 |
Neg logit contributions | |
to | -0.185 |
not | -0.183 |
his | -0.179 |
next | -0.177 |
an | -0.177 |
Token | bear |
Token ID | 6842 |
Feature activation | -1.33e+01 |
Pos logit contributions | |
azz | +0.538 |
pillar | +0.486 |
och | +0.482 |
hell | +0.467 |
enny | +0.467 |
Neg logit contributions | |
the | -0.435 |
not | -0.420 |
an | -0.414 |
making | -0.401 |
his | -0.378 |
Token | had |
Token ID | 550 |
Feature activation | +9.42e+00 |
Pos logit contributions | |
azz | +0.571 |
pillar | +0.524 |
och | +0.511 |
enny | +0.502 |
pp | +0.500 |
Neg logit contributions | |
not | -0.380 |
the | -0.376 |
an | -0.364 |
making | -0.361 |
many | -0.340 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
making | +0.205 |
not | +0.201 |
adding | +0.190 |
guessing | +0.189 |
next | +0.188 |
Neg logit contributions | |
azz | -0.402 |
hell | -0.383 |
ird | -0.379 |
pillar | -0.372 |
ppy | -0.367 |
Token | heavier |
Token ID | 20140 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
not | +0.243 |
making | +0.229 |
adding | +0.210 |
many | +0.206 |
hearing | +0.202 |
Neg logit contributions | |
azz | -0.657 |
och | -0.637 |
hell | -0.629 |
pillar | -0.624 |
y | -0.623 |
Token | , |
Token ID | 11 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
the | +0.054 |
not | +0.053 |
an | +0.051 |
making | +0.051 |
many | +0.048 |
Neg logit contributions | |
azz | -0.085 |
pillar | -0.079 |
och | -0.076 |
enny | -0.075 |
affe | -0.074 |
Token | but |
Token ID | 475 |
Feature activation | +4.20e+00 |
Pos logit contributions | |
not | +0.080 |
the | +0.079 |
making | +0.078 |
an | +0.077 |
many | +0.077 |
Neg logit contributions | |
azz | -0.112 |
pillar | -0.102 |
och | -0.099 |
hell | -0.097 |
enny | -0.096 |
Token | her |
Token ID | 607 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
making | +0.073 |
not | +0.073 |
an | +0.071 |
the | +0.069 |
many | +0.068 |
Neg logit contributions | |
azz | -0.211 |
och | -0.194 |
pillar | -0.194 |
enny | -0.192 |
pp | -0.189 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.11e+01 |
Pos logit contributions | |
azz | +0.059 |
pillar | +0.056 |
och | +0.054 |
enny | +0.054 |
hell | +0.054 |
Neg logit contributions | |
not | -0.018 |
the | -0.018 |
an | -0.017 |
making | -0.017 |
many | -0.015 |
Token | looking |
Token ID | 2045 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
not | +0.065 |
an | +0.064 |
making | +0.062 |
the | +0.059 |
many | +0.059 |
Neg logit contributions | |
azz | -0.193 |
pillar | -0.184 |
och | -0.178 |
ird | -0.177 |
enny | -0.174 |
Token | for |
Token ID | 329 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
not | +0.089 |
making | +0.085 |
an | +0.083 |
the | +0.082 |
adding | +0.078 |
Neg logit contributions | |
azz | -0.192 |
pillar | -0.180 |
hell | -0.175 |
och | -0.175 |
enny | -0.172 |
Token | dessert |
Token ID | 23084 |
Feature activation | -7.66e+00 |
Pos logit contributions | |
not | +0.165 |
making | +0.159 |
adding | +0.139 |
going | +0.139 |
an | +0.138 |
Neg logit contributions | |
azz | -0.585 |
hell | -0.565 |
pillar | -0.564 |
ird | -0.551 |
och | -0.547 |
Token | and |
Token ID | 290 |
Feature activation | +6.51e+00 |
Pos logit contributions | |
azz | +0.365 |
pillar | +0.338 |
och | +0.328 |
enny | +0.322 |
affe | +0.321 |
Neg logit contributions | |
not | -0.187 |
the | -0.185 |
making | -0.176 |
an | -0.176 |
many | -0.164 |
Token | always |
Token ID | 1464 |
Feature activation | +9.46e+00 |
Pos logit contributions | |
an | +0.126 |
not | +0.125 |
the | +0.121 |
making | +0.118 |
many | +0.112 |
Neg logit contributions | |
azz | -0.295 |
pillar | -0.280 |
hell | -0.276 |
och | -0.275 |
pp | -0.271 |
Token | kept |
Token ID | 4030 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
an | +0.230 |
not | +0.223 |
the | +0.222 |
many | +0.215 |
making | +0.215 |
Neg logit contributions | |
azz | -0.379 |
ird | -0.358 |
y | -0.356 |
pp | -0.354 |
pillar | -0.351 |
Token | an |
Token ID | 281 |
Feature activation | +5.06e+00 |
Pos logit contributions | |
not | +0.179 |
many | +0.165 |
an | +0.162 |
making | +0.156 |
conquering | +0.153 |
Neg logit contributions | |
azz | -0.618 |
ird | -0.612 |
och | -0.611 |
y | -0.606 |
pp | -0.601 |
Token | eye |
Token ID | 4151 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
not | +0.137 |
making | +0.136 |
the | +0.134 |
an | +0.128 |
saving | +0.121 |
Neg logit contributions | |
pillar | -0.199 |
hell | -0.199 |
azz | -0.198 |
affe | -0.196 |
ird | -0.190 |
Token | out |
Token ID | 503 |
Feature activation | +4.28e+00 |
Pos logit contributions | |
not | +0.001 |
the | +0.001 |
an | +0.001 |
making | +0.001 |
many | +0.001 |
Neg logit contributions | |
azz | -0.002 |
pillar | -0.002 |
och | -0.002 |
enny | -0.002 |
pp | -0.002 |
Token | for |
Token ID | 329 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
not | +0.074 |
making | +0.070 |
adding | +0.065 |
an | +0.065 |
going | +0.061 |
Neg logit contributions | |
azz | -0.207 |
och | -0.196 |
pillar | -0.194 |
hell | -0.192 |
enny | -0.190 |
Token | any |
Token ID | 597 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
not | +0.166 |
making | +0.161 |
an | +0.150 |
going | +0.140 |
the | +0.139 |
Neg logit contributions | |
azz | -0.599 |
hell | -0.591 |
ird | -0.587 |
och | -0.583 |
pillar | -0.582 |
Token | jealous |
Token ID | 19354 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
saving | +0.178 |
making | +0.174 |
climbing | +0.167 |
hearing | +0.165 |
conquering | +0.165 |
Neg logit contributions | |
azz | -0.301 |
y | -0.293 |
as | -0.288 |
hell | -0.285 |
ite | -0.281 |
Token | . |
Token ID | 13 |
Feature activation | -4.75e+00 |
Pos logit contributions | |
the | +0.053 |
not | +0.052 |
making | +0.051 |
an | +0.050 |
going | +0.047 |
Neg logit contributions | |
azz | -0.099 |
pillar | -0.092 |
och | -0.090 |
pp | -0.090 |
affe | -0.088 |
Token | They |
Token ID | 1119 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
azz | +0.194 |
pillar | +0.178 |
och | +0.172 |
enny | +0.170 |
pp | +0.168 |
Neg logit contributions | |
not | -0.144 |
the | -0.142 |
making | -0.138 |
an | -0.138 |
many | -0.130 |
Token | hadn |
Token ID | 8020 |
Feature activation | +7.14e+00 |
Pos logit contributions | |
azz | +0.174 |
pillar | +0.162 |
och | +0.157 |
enny | +0.154 |
pp | +0.154 |
Neg logit contributions | |
not | -0.109 |
the | -0.108 |
making | -0.104 |
an | -0.103 |
many | -0.097 |
Token | 't |
Token ID | 470 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
not | +0.125 |
the | +0.123 |
an | +0.117 |
making | +0.113 |
going | +0.104 |
Neg logit contributions | |
azz | -0.384 |
pillar | -0.362 |
och | -0.354 |
pp | -0.351 |
affe | -0.351 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
not | +0.225 |
the | +0.220 |
making | +0.215 |
an | +0.215 |
many | +0.204 |
Neg logit contributions | |
azz | -0.327 |
pillar | -0.299 |
och | -0.288 |
enny | -0.287 |
hell | -0.287 |
Token | something |
Token ID | 1223 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
not | +0.223 |
making | +0.209 |
next | +0.195 |
adding | +0.192 |
many | +0.188 |
Neg logit contributions | |
azz | -0.675 |
y | -0.636 |
och | -0.621 |
pp | -0.617 |
hell | -0.615 |
Token | as |
Token ID | 355 |
Feature activation | +7.83e+00 |
Pos logit contributions | |
an | +0.004 |
the | +0.004 |
not | +0.004 |
climbing | +0.004 |
making | +0.004 |
Neg logit contributions | |
azz | -0.013 |
hell | -0.012 |
och | -0.012 |
pillar | -0.012 |
enny | -0.012 |
Token | special |
Token ID | 2041 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
making | +0.114 |
not | +0.107 |
saving | +0.106 |
climbing | +0.105 |
adding | +0.097 |
Neg logit contributions | |
azz | -0.395 |
hell | -0.369 |
enny | -0.361 |
pillar | -0.358 |
och | -0.352 |
Token | as |
Token ID | 355 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
not | +0.059 |
the | +0.057 |
an | +0.056 |
making | +0.055 |
many | +0.053 |
Neg logit contributions | |
azz | -0.112 |
pillar | -0.104 |
och | -0.102 |
enny | -0.100 |
hell | -0.100 |
Token | Sammy |
Token ID | 44067 |
Feature activation | -1.15e+01 |
Pos logit contributions | |
making | +0.115 |
not | +0.115 |
saving | +0.106 |
climbing | +0.102 |
preventing | +0.101 |
Neg logit contributions | |
azz | -0.364 |
enny | -0.331 |
pillar | -0.328 |
hell | -0.326 |
och | -0.321 |
Token | knew |
Token ID | 2993 |
Feature activation | +5.46e+00 |
Pos logit contributions | |
azz | +0.240 |
pillar | +0.223 |
enny | +0.212 |
och | +0.210 |
hell | +0.208 |
Neg logit contributions | |
not | -0.174 |
the | -0.171 |
making | -0.170 |
an | -0.169 |
going | -0.160 |
Token | his |
Token ID | 465 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
making | +0.106 |
adding | +0.096 |
not | +0.091 |
guessing | +0.090 |
saving | +0.090 |
Neg logit contributions | |
azz | -0.238 |
pillar | -0.224 |
enny | -0.224 |
och | -0.223 |
hell | -0.217 |
Token | mum |
Token ID | 25682 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
pillar | +0.028 |
hell | +0.027 |
enny | +0.027 |
azz | +0.027 |
y | +0.026 |
Neg logit contributions | |
saving | -0.019 |
making | -0.019 |
preventing | -0.017 |
causing | -0.017 |
an | -0.017 |
Token | must |
Token ID | 1276 |
Feature activation | +8.32e+00 |
Pos logit contributions | |
azz | +0.540 |
pillar | +0.487 |
och | +0.481 |
pp | +0.475 |
enny | +0.473 |
Neg logit contributions | |
the | -0.341 |
not | -0.336 |
an | -0.314 |
making | -0.313 |
going | -0.298 |
Token | 've |
Token ID | 1053 |
Feature activation | +8.27e+00 |
Pos logit contributions | |
not | +0.156 |
the | +0.154 |
an | +0.148 |
making | +0.147 |
many | +0.134 |
Neg logit contributions | |
azz | -0.432 |
pillar | -0.405 |
och | -0.395 |
enny | -0.389 |
pp | -0.388 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
making | +0.184 |
many | +0.181 |
an | +0.181 |
the | +0.179 |
not | +0.179 |
Neg logit contributions | |
azz | -0.360 |
pillar | -0.336 |
ird | -0.329 |
hell | -0.326 |
och | -0.325 |
Token | it |
Token ID | 340 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
not | +0.199 |
adding | +0.186 |
warnings | +0.181 |
conquering | +0.179 |
making | +0.176 |
Neg logit contributions | |
azz | -0.610 |
och | -0.608 |
y | -0.605 |
enny | -0.580 |
pp | -0.578 |
Token | down |
Token ID | 866 |
Feature activation | +5.98e+00 |
Pos logit contributions | |
azz | +0.130 |
pillar | +0.121 |
och | +0.118 |
hell | +0.118 |
enny | +0.118 |
Neg logit contributions | |
not | -0.068 |
the | -0.066 |
making | -0.066 |
an | -0.064 |
many | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +6.76e+00 |
Pos logit contributions | |
not | +0.142 |
making | +0.134 |
adding | +0.131 |
many | +0.129 |
an | +0.128 |
Neg logit contributions | |
azz | -0.248 |
och | -0.232 |
y | -0.232 |
pillar | -0.227 |
enny | -0.225 |
Token | he |
Token ID | 339 |
Feature activation | -4.67e+00 |
Pos logit contributions | |
an | +0.156 |
not | +0.155 |
making | +0.150 |
many | +0.149 |
the | +0.145 |
Neg logit contributions | |
azz | -0.283 |
pillar | -0.263 |
och | -0.260 |
ird | -0.256 |
enny | -0.254 |
Token | couldn |
Token ID | 3521 |
Feature activation | +9.71e+00 |
Pos logit contributions | |
azz | +0.199 |
pillar | +0.183 |
och | +0.178 |
enny | +0.175 |
pp | +0.174 |
Neg logit contributions | |
not | -0.133 |
the | -0.131 |
an | -0.127 |
making | -0.127 |
many | -0.119 |
Token | Sam |
Token ID | 3409 |
Feature activation | -1.37e+01 |
Pos logit contributions | |
not | +0.127 |
the | +0.123 |
an | +0.122 |
making | +0.121 |
many | +0.112 |
Neg logit contributions | |
azz | -0.266 |
pillar | -0.247 |
och | -0.240 |
enny | -0.238 |
pp | -0.237 |
Token | was |
Token ID | 373 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
azz | +0.578 |
pillar | +0.532 |
och | +0.520 |
enny | +0.509 |
pp | +0.506 |
Neg logit contributions | |
the | -0.403 |
not | -0.402 |
an | -0.386 |
making | -0.384 |
many | -0.363 |
Token | glad |
Token ID | 9675 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
an | +0.153 |
not | +0.153 |
making | +0.152 |
the | +0.146 |
climbing | +0.146 |
Neg logit contributions | |
azz | -0.510 |
y | -0.474 |
ird | -0.473 |
pillar | -0.472 |
hell | -0.465 |
Token | he |
Token ID | 339 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
not | +0.071 |
an | +0.069 |
making | +0.068 |
the | +0.065 |
many | +0.064 |
Neg logit contributions | |
azz | -0.214 |
pillar | -0.199 |
enny | -0.192 |
affe | -0.191 |
hell | -0.190 |
Token | had |
Token ID | 550 |
Feature activation | +8.79e+00 |
Pos logit contributions | |
azz | +0.248 |
pillar | +0.224 |
och | +0.222 |
pp | +0.214 |
hell | +0.214 |
Neg logit contributions | |
the | -0.180 |
not | -0.178 |
making | -0.171 |
an | -0.170 |
going | -0.162 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
not | +0.197 |
making | +0.191 |
an | +0.187 |
the | +0.186 |
adding | +0.181 |
Neg logit contributions | |
azz | -0.378 |
pillar | -0.347 |
ird | -0.343 |
hell | -0.342 |
enny | -0.341 |
Token | the |
Token ID | 262 |
Feature activation | +8.20e-01 |
Pos logit contributions | |
not | +0.172 |
making | +0.156 |
climbing | +0.141 |
adding | +0.138 |
many | +0.130 |
Neg logit contributions | |
azz | -0.727 |
enny | -0.681 |
y | -0.680 |
och | -0.679 |
hell | -0.677 |
Token | invitation |
Token ID | 17023 |
Feature activation | -6.15e+00 |
Pos logit contributions | |
not | +0.019 |
the | +0.019 |
an | +0.018 |
making | +0.018 |
many | +0.017 |
Neg logit contributions | |
azz | -0.039 |
pillar | -0.036 |
och | -0.035 |
enny | -0.035 |
pp | -0.034 |
Token | in |
Token ID | 287 |
Feature activation | +9.20e+00 |
Pos logit contributions | |
azz | +0.314 |
pillar | +0.294 |
och | +0.285 |
enny | +0.282 |
pp | +0.281 |
Neg logit contributions | |
not | -0.127 |
the | -0.126 |
an | -0.119 |
making | -0.118 |
going | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
not | +0.068 |
making | +0.055 |
an | +0.047 |
going | +0.046 |
climbing | +0.044 |
Neg logit contributions | |
azz | -0.551 |
pillar | -0.515 |
och | -0.512 |
enny | -0.508 |
hell | -0.504 |
Token | mail |
Token ID | 6920 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
azz | +0.001 |
pillar | +0.001 |
och | +0.001 |
hell | +0.001 |
enny | +0.001 |
Neg logit contributions | |
not | -0.001 |
the | -0.001 |
an | -0.001 |
making | -0.001 |
going | -0.001 |
Token | stamp |
Token ID | 17977 |
Feature activation | -1.04e+01 |
Pos logit contributions | |
azz | +0.054 |
pillar | +0.051 |
och | +0.049 |
hell | +0.049 |
affe | +0.048 |
Neg logit contributions | |
not | -0.028 |
the | -0.027 |
an | -0.027 |
making | -0.026 |
many | -0.025 |
Token | because |
Token ID | 780 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
azz | +0.509 |
pillar | +0.466 |
och | +0.457 |
enny | +0.453 |
pp | +0.449 |
Neg logit contributions | |
the | -0.247 |
not | -0.244 |
making | -0.234 |
an | -0.227 |
going | -0.213 |
Token | she |
Token ID | 673 |
Feature activation | -5.37e+00 |
Pos logit contributions | |
not | +0.043 |
making | +0.042 |
an | +0.039 |
the | +0.038 |
saving | +0.038 |
Neg logit contributions | |
azz | -0.119 |
pillar | -0.112 |
och | -0.110 |
enny | -0.108 |
hell | -0.107 |
Token | had |
Token ID | 550 |
Feature activation | +9.29e+00 |
Pos logit contributions | |
azz | +0.221 |
pillar | +0.201 |
och | +0.196 |
pp | +0.193 |
hell | +0.191 |
Neg logit contributions | |
not | -0.164 |
the | -0.162 |
an | -0.157 |
making | -0.157 |
many | -0.148 |
Token | never |
Token ID | 1239 |
Feature activation | +9.14e+00 |
Pos logit contributions | |
not | +0.180 |
making | +0.178 |
saving | +0.168 |
the | +0.168 |
helping | +0.167 |
Neg logit contributions | |
azz | -0.418 |
pillar | -0.399 |
hell | -0.391 |
ird | -0.382 |
enny | -0.378 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
not | +0.271 |
the | +0.258 |
an | +0.254 |
making | +0.252 |
many | +0.245 |
Neg logit contributions | |
azz | -0.360 |
pillar | -0.334 |
enny | -0.321 |
hell | -0.321 |
ppy | -0.319 |
Token | one |
Token ID | 530 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
not | +0.210 |
making | +0.197 |
saving | +0.173 |
climbing | +0.170 |
the | +0.170 |
Neg logit contributions | |
azz | -0.730 |
pillar | -0.691 |
hell | -0.678 |
och | -0.677 |
y | -0.676 |
Token | before |
Token ID | 878 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
azz | +0.093 |
pillar | +0.087 |
och | +0.085 |
enny | +0.083 |
hell | +0.083 |
Neg logit contributions | |
not | -0.042 |
the | -0.041 |
an | -0.039 |
making | -0.039 |
many | -0.036 |
Token | ! |
Token ID | 0 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
not | +0.114 |
an | +0.109 |
saving | +0.108 |
the | +0.107 |
making | +0.106 |
Neg logit contributions | |
azz | -0.136 |
pillar | -0.131 |
y | -0.125 |
och | -0.123 |
hell | -0.122 |
Token | |
Token ID | 220 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
azz | +0.151 |
pp | +0.142 |
pillar | +0.141 |
hell | +0.139 |
ppy | +0.138 |
Neg logit contributions | |
the | -0.081 |
an | -0.076 |
not | -0.076 |
many | -0.071 |
making | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
azz | +0.225 |
pillar | +0.215 |
pp | +0.208 |
affe | +0.208 |
och | +0.207 |
Neg logit contributions | |
the | -0.099 |
not | -0.099 |
an | -0.095 |
many | -0.088 |
making | -0.087 |
Token | ! |
Token ID | 0 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
not | +0.001 |
the | +0.001 |
making | +0.001 |
an | +0.001 |
going | +0.001 |
Neg logit contributions | |
azz | -0.002 |
pillar | -0.002 |
och | -0.002 |
enny | -0.002 |
hell | -0.002 |
Token | I |
Token ID | 314 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
azz | +0.072 |
pillar | +0.067 |
hell | +0.065 |
och | +0.065 |
pp | +0.065 |
Neg logit contributions | |
not | -0.041 |
the | -0.041 |
an | -0.040 |
making | -0.039 |
many | -0.036 |
Token | wish |
Token ID | 4601 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
not | +0.005 |
the | +0.005 |
an | +0.005 |
making | +0.004 |
to | +0.004 |
Neg logit contributions | |
azz | -0.008 |
pillar | -0.007 |
och | -0.007 |
affe | -0.007 |
pp | -0.007 |
Token | I |
Token ID | 314 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
an | +0.030 |
making | +0.029 |
adding | +0.028 |
not | +0.028 |
the | +0.028 |
Neg logit contributions | |
azz | -0.062 |
pillar | -0.057 |
enny | -0.056 |
och | -0.055 |
hell | -0.055 |
Token | had |
Token ID | 550 |
Feature activation | +9.53e+00 |
Pos logit contributions | |
not | +0.026 |
the | +0.026 |
an | +0.025 |
making | +0.025 |
going | +0.023 |
Neg logit contributions | |
azz | -0.031 |
pillar | -0.028 |
och | -0.027 |
pp | -0.026 |
enny | -0.026 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
making | +0.163 |
adding | +0.158 |
digging | +0.152 |
not | +0.149 |
shore | +0.146 |
Neg logit contributions | |
azz | -0.443 |
pillar | -0.415 |
hell | -0.412 |
ird | -0.411 |
enny | -0.403 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
making | +0.183 |
not | +0.180 |
adding | +0.173 |
climbing | +0.159 |
conquering | +0.156 |
Neg logit contributions | |
azz | -0.691 |
pillar | -0.663 |
och | -0.663 |
enny | -0.648 |
hell | -0.643 |
Token | real |
Token ID | 1103 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
not | +0.057 |
making | +0.056 |
the | +0.055 |
an | +0.054 |
saving | +0.052 |
Neg logit contributions | |
azz | -0.047 |
pillar | -0.046 |
hell | -0.044 |
och | -0.043 |
affe | -0.042 |
Token | diamond |
Token ID | 15291 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
azz | +0.054 |
pillar | +0.050 |
och | +0.048 |
hell | +0.047 |
affe | +0.047 |
Neg logit contributions | |
not | -0.055 |
the | -0.054 |
making | -0.053 |
an | -0.053 |
many | -0.050 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
azz | +0.501 |
pillar | +0.463 |
och | +0.451 |
enny | +0.448 |
hell | +0.440 |
Neg logit contributions | |
not | -0.226 |
the | -0.225 |
making | -0.212 |
an | -0.212 |
going | -0.196 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.43e+01 |
Pos logit contributions | |
azz | +0.208 |
pillar | +0.194 |
enny | +0.187 |
och | +0.186 |
hell | +0.182 |
Neg logit contributions | |
not | -0.127 |
the | -0.122 |
making | -0.121 |
an | -0.120 |
going | -0.114 |
Token | long |
Token ID | 890 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
not | +0.042 |
the | +0.042 |
an | +0.040 |
making | +0.040 |
many | +0.036 |
Neg logit contributions | |
azz | -0.121 |
pillar | -0.114 |
och | -0.111 |
enny | -0.109 |
pp | -0.109 |
Token | time |
Token ID | 640 |
Feature activation | +7.74e-01 |
Pos logit contributions | |
not | +0.001 |
the | +0.001 |
making | +0.001 |
an | +0.001 |
going | +0.001 |
Neg logit contributions | |
azz | -0.002 |
pillar | -0.002 |
och | -0.002 |
enny | -0.002 |
pp | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +6.02e+00 |
Pos logit contributions | |
not | +0.021 |
the | +0.021 |
an | +0.020 |
making | +0.020 |
many | +0.019 |
Neg logit contributions | |
azz | -0.034 |
pillar | -0.031 |
och | -0.030 |
enny | -0.030 |
pp | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | -5.16e+00 |
Pos logit contributions | |
not | +0.107 |
making | +0.104 |
an | +0.102 |
the | +0.100 |
adding | +0.095 |
Neg logit contributions | |
azz | -0.293 |
pillar | -0.272 |
och | -0.271 |
enny | -0.266 |
hell | -0.264 |
Token | had |
Token ID | 550 |
Feature activation | +9.98e+00 |
Pos logit contributions | |
azz | +0.218 |
pillar | +0.201 |
och | +0.195 |
enny | +0.191 |
pp | +0.191 |
Neg logit contributions | |
not | -0.152 |
the | -0.150 |
an | -0.145 |
making | -0.144 |
many | -0.136 |
Token | gotten |
Token ID | 7891 |
Feature activation | +1.40e+01 |
Pos logit contributions | |
making | +0.207 |
not | +0.202 |
the | +0.197 |
adding | +0.195 |
helping | +0.195 |
Neg logit contributions | |
azz | -0.452 |
pillar | -0.408 |
hell | -0.407 |
pp | -0.397 |
y | -0.394 |
Token | wet |
Token ID | 9583 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
not | +0.308 |
making | +0.297 |
the | +0.281 |
adding | +0.281 |
an | +0.278 |
Neg logit contributions | |
azz | -0.626 |
och | -0.578 |
pp | -0.574 |
y | -0.569 |
hell | -0.565 |
Token | from |
Token ID | 422 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
the | +0.026 |
an | +0.024 |
making | +0.024 |
not | +0.024 |
going | +0.023 |
Neg logit contributions | |
azz | -0.060 |
pillar | -0.055 |
enny | -0.054 |
och | -0.054 |
affe | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
not | +0.125 |
making | +0.102 |
an | +0.100 |
helping | +0.088 |
saving | +0.087 |
Neg logit contributions | |
azz | -0.589 |
pillar | -0.555 |
och | -0.554 |
y | -0.553 |
pp | -0.545 |
Token | rain |
Token ID | 6290 |
Feature activation | -6.92e+00 |
Pos logit contributions | |
not | +0.012 |
the | +0.012 |
an | +0.012 |
making | +0.012 |
his | +0.011 |
Neg logit contributions | |
azz | -0.016 |
pillar | -0.016 |
affe | -0.015 |
och | -0.015 |
hell | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
azz | +0.359 |
pillar | +0.338 |
enny | +0.328 |
och | +0.327 |
affe | +0.325 |
Neg logit contributions | |
the | -0.152 |
not | -0.150 |
making | -0.142 |
to | -0.139 |
going | -0.133 |
Token | smell |
Token ID | 8508 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
not | +0.211 |
an | +0.204 |
the | +0.202 |
making | +0.199 |
many | +0.187 |
Neg logit contributions | |
azz | -0.547 |
pillar | -0.507 |
och | -0.498 |
hell | -0.491 |
pp | -0.490 |
Token | the |
Token ID | 262 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
not | +0.058 |
making | +0.053 |
an | +0.050 |
the | +0.048 |
climbing | +0.047 |
Neg logit contributions | |
azz | -0.213 |
pillar | -0.201 |
hell | -0.198 |
och | -0.195 |
enny | -0.191 |
Token | flowers |
Token ID | 12734 |
Feature activation | -8.24e+00 |
Pos logit contributions | |
not | +0.021 |
the | +0.020 |
an | +0.020 |
making | +0.020 |
many | +0.019 |
Neg logit contributions | |
azz | -0.042 |
pillar | -0.039 |
och | -0.038 |
hell | -0.037 |
pp | -0.037 |
Token | instead |
Token ID | 2427 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
azz | +0.390 |
pillar | +0.363 |
och | +0.353 |
enny | +0.348 |
pp | +0.347 |
Neg logit contributions | |
not | -0.194 |
the | -0.192 |
an | -0.185 |
making | -0.184 |
many | -0.171 |
Token | of |
Token ID | 286 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
not | +0.115 |
the | +0.115 |
making | +0.109 |
an | +0.109 |
many | +0.103 |
Neg logit contributions | |
azz | -0.177 |
pillar | -0.164 |
och | -0.159 |
pp | -0.157 |
enny | -0.156 |
Token | causing |
Token ID | 6666 |
Feature activation | +1.39e+01 |
Pos logit contributions | |
not | +0.154 |
an | +0.142 |
the | +0.129 |
making | +0.127 |
many | +0.126 |
Neg logit contributions | |
azz | -0.515 |
pillar | -0.489 |
hell | -0.472 |
och | -0.472 |
y | -0.466 |
Token | trouble |
Token ID | 5876 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
not | +0.198 |
making | +0.179 |
saving | +0.169 |
an | +0.157 |
adding | +0.154 |
Neg logit contributions | |
azz | -0.732 |
pillar | -0.702 |
hell | -0.675 |
y | -0.674 |
ppy | -0.653 |
Token | . |
Token ID | 13 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
not | +0.060 |
the | +0.059 |
an | +0.058 |
making | +0.057 |
many | +0.054 |
Neg logit contributions | |
azz | -0.102 |
pillar | -0.095 |
och | -0.092 |
hell | -0.091 |
enny | -0.090 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.30e+00 |
Pos logit contributions | |
azz | +0.193 |
pillar | +0.180 |
och | +0.175 |
enny | +0.172 |
pp | +0.171 |
Neg logit contributions | |
not | -0.101 |
the | -0.099 |
an | -0.096 |
making | -0.096 |
many | -0.089 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
azz | +0.424 |
pillar | +0.397 |
och | +0.393 |
pp | +0.392 |
affe | +0.388 |
Neg logit contributions | |
the | -0.184 |
not | -0.170 |
an | -0.169 |
making | -0.159 |
many | -0.156 |
Token | upon |
Token ID | 2402 |
Feature activation | +3.00e+00 |
Pos logit contributions | |
the | +0.006 |
not | +0.006 |
an | +0.006 |
making | +0.006 |
many | +0.006 |
Neg logit contributions | |
azz | -0.007 |
pillar | -0.007 |
och | -0.007 |
pp | -0.007 |
affe | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
azz | +0.082 |
och | +0.073 |
pillar | +0.073 |
hell | +0.072 |
enny | +0.072 |
Neg logit contributions | |
not | -0.056 |
making | -0.056 |
an | -0.055 |
the | -0.055 |
going | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
azz | +0.226 |
pillar | +0.208 |
och | +0.202 |
enny | +0.200 |
affe | +0.197 |
Neg logit contributions | |
not | -0.125 |
making | -0.122 |
the | -0.120 |
an | -0.119 |
going | -0.116 |
Token |
|
Token ID | 198 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
azz | +0.203 |
pillar | +0.191 |
och | +0.185 |
pp | +0.185 |
affe | +0.182 |
Neg logit contributions | |
the | -0.094 |
not | -0.093 |
an | -0.088 |
making | -0.084 |
many | -0.084 |
Token | The |
Token ID | 464 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
azz | +0.167 |
pillar | +0.151 |
pp | +0.147 |
och | +0.145 |
affe | +0.145 |
Neg logit contributions | |
the | -0.149 |
not | -0.146 |
many | -0.139 |
an | -0.137 |
next | -0.136 |
Token | great |
Token ID | 1049 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
azz | +0.103 |
pillar | +0.095 |
och | +0.088 |
hell | +0.087 |
enny | +0.087 |
Neg logit contributions | |
not | -0.120 |
the | -0.117 |
an | -0.117 |
making | -0.116 |
going | -0.112 |
Token | Zen |
Token ID | 14760 |
Feature activation | -2.13e+01 |
Pos logit contributions | |
azz | +0.020 |
pillar | +0.018 |
enny | +0.017 |
och | +0.017 |
hell | +0.017 |
Neg logit contributions | |
not | -0.018 |
the | -0.018 |
an | -0.017 |
making | -0.017 |
going | -0.017 |
Token | o |
Token ID | 78 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
azz | +0.650 |
pillar | +0.619 |
och | +0.583 |
hell | +0.580 |
enny | +0.553 |
Neg logit contributions | |
not | -0.873 |
the | -0.870 |
to | -0.823 |
an | -0.810 |
making | -0.807 |
Token | hugs |
Token ID | 40657 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
azz | +0.563 |
pillar | +0.521 |
och | +0.506 |
enny | +0.498 |
pp | +0.496 |
Neg logit contributions | |
not | -0.342 |
the | -0.339 |
an | -0.328 |
making | -0.327 |
many | -0.307 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.40e+01 |
Pos logit contributions | |
the | +0.036 |
an | +0.029 |
not | +0.028 |
his | +0.026 |
making | +0.025 |
Neg logit contributions | |
azz | -0.126 |
pillar | -0.119 |
pp | -0.119 |
och | -0.119 |
ird | -0.116 |
Token | and |
Token ID | 290 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
azz | +0.692 |
pillar | +0.643 |
och | +0.637 |
enny | +0.620 |
pp | +0.620 |
Neg logit contributions | |
the | -0.339 |
not | -0.316 |
an | -0.302 |
making | -0.286 |
many | -0.286 |
Token | says |
Token ID | 1139 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
not | +0.122 |
making | +0.118 |
an | +0.115 |
the | +0.112 |
saving | +0.110 |
Neg logit contributions | |
azz | -0.188 |
pillar | -0.170 |
affe | -0.160 |
och | -0.160 |
enny | -0.158 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.84e+00 |
Pos logit contributions | |
azz | +0.366 |
pillar | +0.339 |
enny | +0.326 |
och | +0.319 |
pp | +0.316 |
Neg logit contributions | |
not | -0.299 |
the | -0.292 |
an | -0.289 |
making | -0.288 |
many | -0.273 |
Token | a |
Token ID | 257 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
facing | +0.085 |
not | +0.085 |
making | +0.081 |
adding | +0.079 |
climbing | +0.078 |
Neg logit contributions | |
azz | -0.407 |
pillar | -0.381 |
hell | -0.380 |
enny | -0.373 |
y | -0.371 |
Token | big |
Token ID | 1263 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
not | +0.020 |
the | +0.019 |
making | +0.019 |
an | +0.018 |
many | +0.017 |
Neg logit contributions | |
azz | -0.062 |
pillar | -0.058 |
och | -0.057 |
ppy | -0.056 |
enny | -0.056 |
Token | box |
Token ID | 3091 |
Feature activation | -9.51e+00 |
Pos logit contributions | |
azz | +0.063 |
pillar | +0.058 |
och | +0.056 |
enny | +0.056 |
pp | +0.055 |
Neg logit contributions | |
the | -0.040 |
not | -0.039 |
making | -0.038 |
an | -0.038 |
going | -0.036 |
Token | of |
Token ID | 286 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
azz | +0.444 |
pillar | +0.413 |
och | +0.406 |
enny | +0.397 |
pp | +0.397 |
Neg logit contributions | |
not | -0.220 |
the | -0.218 |
an | -0.213 |
making | -0.207 |
many | -0.197 |
Token | Leg |
Token ID | 3564 |
Feature activation | -1.99e+01 |
Pos logit contributions | |
not | +0.211 |
the | +0.205 |
an | +0.199 |
making | +0.199 |
many | +0.185 |
Neg logit contributions | |
azz | -0.384 |
pillar | -0.357 |
och | -0.348 |
hell | -0.341 |
enny | -0.341 |
Token | os |
Token ID | 418 |
Feature activation | -7.21e+00 |
Pos logit contributions | |
azz | +0.585 |
pillar | +0.580 |
affe | +0.545 |
hell | +0.516 |
pp | +0.479 |
Neg logit contributions | |
making | -0.840 |
not | -0.832 |
the | -0.822 |
going | -0.805 |
to | -0.786 |
Token | that |
Token ID | 326 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
azz | +0.366 |
pillar | +0.338 |
enny | +0.337 |
affe | +0.327 |
och | +0.323 |
Neg logit contributions | |
the | -0.174 |
making | -0.163 |
going | -0.153 |
not | -0.144 |
an | -0.140 |
Token | he |
Token ID | 339 |
Feature activation | -5.63e+00 |
Pos logit contributions | |
not | +0.043 |
making | +0.042 |
an | +0.042 |
the | +0.042 |
climbing | +0.040 |
Neg logit contributions | |
azz | -0.068 |
och | -0.061 |
pillar | -0.061 |
enny | -0.060 |
hell | -0.059 |
Token | really |
Token ID | 1107 |
Feature activation | +5.78e+00 |
Pos logit contributions | |
azz | +0.232 |
pillar | +0.213 |
och | +0.207 |
enny | +0.203 |
pp | +0.202 |
Neg logit contributions | |
not | -0.169 |
the | -0.167 |
an | -0.163 |
making | -0.162 |
many | -0.153 |
Token | wanted |
Token ID | 2227 |
Feature activation | +9.01e+00 |
Pos logit contributions | |
not | +0.178 |
the | +0.174 |
an | +0.169 |
making | +0.167 |
many | +0.161 |
Neg logit contributions | |
azz | -0.240 |
pillar | -0.221 |
och | -0.216 |
enny | -0.210 |
pp | -0.209 |
Token | Yes |
Token ID | 5297 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
azz | +0.087 |
pillar | +0.077 |
hell | +0.077 |
ppy | +0.076 |
och | +0.076 |
Neg logit contributions | |
the | -0.049 |
not | -0.048 |
an | -0.048 |
making | -0.046 |
next | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
azz | +0.088 |
pillar | +0.082 |
och | +0.080 |
pp | +0.079 |
enny | +0.077 |
Neg logit contributions | |
the | -0.046 |
not | -0.046 |
an | -0.043 |
making | -0.042 |
going | -0.039 |
Token | please |
Token ID | 3387 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
not | +0.086 |
the | +0.085 |
making | +0.083 |
an | +0.083 |
many | +0.078 |
Neg logit contributions | |
azz | -0.131 |
pillar | -0.121 |
och | -0.117 |
enny | -0.115 |
affe | -0.115 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
not | +0.065 |
the | +0.065 |
an | +0.064 |
making | +0.063 |
many | +0.062 |
Neg logit contributions | |
azz | -0.079 |
pillar | -0.073 |
och | -0.070 |
pp | -0.070 |
affe | -0.069 |
Token | The |
Token ID | 383 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
azz | +0.210 |
pillar | +0.194 |
och | +0.189 |
enny | +0.187 |
pp | +0.186 |
Neg logit contributions | |
not | -0.118 |
the | -0.115 |
making | -0.112 |
an | -0.112 |
going | -0.106 |
Token | camel |
Token ID | 41021 |
Feature activation | -1.96e+01 |
Pos logit contributions | |
azz | +0.121 |
pillar | +0.110 |
affe | +0.104 |
hell | +0.102 |
och | +0.101 |
Neg logit contributions | |
not | -0.138 |
making | -0.138 |
the | -0.137 |
an | -0.136 |
adding | -0.130 |
Token | gave |
Token ID | 2921 |
Feature activation | +9.07e+00 |
Pos logit contributions | |
azz | +0.828 |
och | +0.762 |
pillar | +0.756 |
enny | +0.734 |
pp | +0.716 |
Neg logit contributions | |
not | -0.609 |
the | -0.607 |
making | -0.561 |
an | -0.558 |
to | -0.531 |
Token | him |
Token ID | 683 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
the | +0.187 |
an | +0.172 |
not | +0.168 |
making | +0.155 |
many | +0.150 |
Neg logit contributions | |
azz | -0.488 |
pillar | -0.459 |
och | -0.450 |
enny | -0.439 |
pp | -0.438 |
Token | a |
Token ID | 257 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
azz | +0.267 |
pillar | +0.255 |
hell | +0.250 |
enny | +0.247 |
ppy | +0.246 |
Neg logit contributions | |
not | -0.039 |
making | -0.038 |
going | -0.029 |
the | -0.028 |
adding | -0.027 |
Token | drink |
Token ID | 4144 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
not | +0.029 |
the | +0.028 |
making | +0.028 |
an | +0.027 |
going | +0.025 |
Neg logit contributions | |
azz | -0.079 |
pillar | -0.074 |
och | -0.072 |
hell | -0.071 |
enny | -0.071 |
Token | of |
Token ID | 286 |
Feature activation | +9.85e+00 |
Pos logit contributions | |
not | +0.063 |
the | +0.062 |
an | +0.061 |
making | +0.061 |
going | +0.057 |
Neg logit contributions | |
azz | -0.100 |
pillar | -0.093 |
och | -0.090 |
hell | -0.089 |
enny | -0.088 |
Token | finish |
Token ID | 5461 |
Feature activation | +9.74e+00 |
Pos logit contributions | |
not | +0.181 |
an | +0.180 |
the | +0.174 |
making | +0.173 |
many | +0.168 |
Neg logit contributions | |
azz | -0.222 |
pillar | -0.202 |
pp | -0.197 |
enny | -0.194 |
ppy | -0.194 |
Token | your |
Token ID | 534 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
not | +0.183 |
an | +0.174 |
saving | +0.165 |
many | +0.160 |
preventing | +0.158 |
Neg logit contributions | |
azz | -0.411 |
y | -0.405 |
pillar | -0.394 |
hell | -0.389 |
ird | -0.386 |
Token | homework |
Token ID | 26131 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
not | +0.061 |
making | +0.058 |
the | +0.058 |
an | +0.056 |
going | +0.055 |
Neg logit contributions | |
azz | -0.108 |
pillar | -0.105 |
hell | -0.100 |
affe | -0.099 |
ird | -0.099 |
Token | ." |
Token ID | 526 |
Feature activation | -6.60e+00 |
Pos logit contributions | |
azz | +0.210 |
pillar | +0.195 |
och | +0.190 |
enny | +0.187 |
hell | +0.187 |
Neg logit contributions | |
not | -0.107 |
the | -0.106 |
an | -0.103 |
making | -0.102 |
many | -0.095 |
Token | The |
Token ID | 383 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
azz | +0.317 |
pillar | +0.300 |
pp | +0.291 |
hell | +0.289 |
och | +0.289 |
Neg logit contributions | |
the | -0.157 |
not | -0.153 |
an | -0.147 |
making | -0.142 |
many | -0.135 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.94e+01 |
Pos logit contributions | |
azz | +0.195 |
pillar | +0.180 |
och | +0.175 |
enny | +0.171 |
pp | +0.171 |
Neg logit contributions | |
not | -0.125 |
the | -0.123 |
an | -0.120 |
making | -0.119 |
many | -0.112 |
Token | realized |
Token ID | 6939 |
Feature activation | +5.42e+00 |
Pos logit contributions | |
azz | +0.807 |
pillar | +0.743 |
och | +0.725 |
enny | +0.709 |
pp | +0.707 |
Neg logit contributions | |
not | -0.576 |
the | -0.571 |
an | -0.551 |
making | -0.550 |
many | -0.519 |
Token | that |
Token ID | 326 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
not | +0.081 |
the | +0.079 |
an | +0.075 |
making | +0.074 |
many | +0.065 |
Neg logit contributions | |
azz | -0.305 |
pillar | -0.288 |
och | -0.281 |
enny | -0.277 |
pp | -0.277 |
Token | he |
Token ID | 339 |
Feature activation | -6.73e+00 |
Pos logit contributions | |
not | +0.008 |
an | +0.008 |
making | +0.008 |
the | +0.007 |
many | +0.007 |
Neg logit contributions | |
azz | -0.022 |
pillar | -0.020 |
enny | -0.020 |
och | -0.020 |
hell | -0.019 |
Token | needed |
Token ID | 2622 |
Feature activation | +8.31e+00 |
Pos logit contributions | |
azz | +0.278 |
och | +0.252 |
pillar | +0.251 |
pp | +0.249 |
enny | +0.241 |
Neg logit contributions | |
not | -0.212 |
the | -0.207 |
an | -0.199 |
making | -0.196 |
going | -0.186 |
Token | to |
Token ID | 284 |
Feature activation | +9.98e+00 |
Pos logit contributions | |
adding | +0.097 |
making | +0.093 |
not | +0.091 |
going | +0.091 |
facing | +0.087 |
Neg logit contributions | |
azz | -0.406 |
enny | -0.392 |
och | -0.382 |
ird | -0.376 |
pillar | -0.376 |
Token | I |
Token ID | 314 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
making | +0.116 |
adding | +0.111 |
saving | +0.109 |
many | +0.106 |
exploring | +0.105 |
Neg logit contributions | |
azz | -0.400 |
pillar | -0.369 |
och | -0.364 |
enny | -0.359 |
affe | -0.353 |
Token | so |
Token ID | 523 |
Feature activation | +7.64e+00 |
Pos logit contributions | |
azz | +0.001 |
pillar | +0.001 |
enny | +0.001 |
och | +0.001 |
ppy | +0.001 |
Neg logit contributions | |
not | -0.000 |
the | -0.000 |
an | -0.000 |
making | -0.000 |
many | -0.000 |
Token | sad |
Token ID | 6507 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
making | +0.158 |
the | +0.157 |
not | +0.154 |
an | +0.152 |
saving | +0.143 |
Neg logit contributions | |
azz | -0.364 |
pillar | -0.347 |
hell | -0.333 |
och | -0.330 |
enny | -0.329 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
not | +0.039 |
the | +0.038 |
making | +0.037 |
an | +0.037 |
many | +0.035 |
Neg logit contributions | |
azz | -0.071 |
pillar | -0.066 |
och | -0.063 |
enny | -0.063 |
affe | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
azz | +0.185 |
pillar | +0.171 |
och | +0.165 |
enny | +0.164 |
affe | +0.160 |
Neg logit contributions | |
not | -0.093 |
making | -0.092 |
going | -0.087 |
an | -0.086 |
saving | -0.086 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.94e+01 |
Pos logit contributions | |
azz | +0.145 |
pillar | +0.136 |
och | +0.130 |
affe | +0.128 |
enny | +0.128 |
Neg logit contributions | |
not | -0.091 |
the | -0.089 |
an | -0.087 |
making | -0.087 |
going | -0.082 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
azz | +0.891 |
pillar | +0.833 |
och | +0.815 |
enny | +0.800 |
hell | +0.799 |
Neg logit contributions | |
not | -0.493 |
the | -0.489 |
an | -0.463 |
making | -0.463 |
going | -0.429 |
Token | himself |
Token ID | 2241 |
Feature activation | -5.64e+00 |
Pos logit contributions | |
not | +0.036 |
the | +0.035 |
making | +0.035 |
an | +0.034 |
many | +0.032 |
Neg logit contributions | |
azz | -0.064 |
pillar | -0.059 |
och | -0.058 |
enny | -0.057 |
pp | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | -5.42e+00 |
Pos logit contributions | |
azz | +0.252 |
pillar | +0.233 |
och | +0.227 |
enny | +0.223 |
pp | +0.223 |
Neg logit contributions | |
not | -0.146 |
the | -0.144 |
an | -0.140 |
making | -0.140 |
many | -0.131 |
Token | He |
Token ID | 679 |
Feature activation | -6.94e+00 |
Pos logit contributions | |
azz | +0.256 |
pillar | +0.239 |
och | +0.232 |
pp | +0.229 |
hell | +0.228 |
Neg logit contributions | |
not | -0.128 |
the | -0.128 |
an | -0.123 |
making | -0.121 |
many | -0.113 |
Token | looked |
Token ID | 3114 |
Feature activation | +5.73e+00 |
Pos logit contributions | |
azz | +0.283 |
pillar | +0.260 |
och | +0.251 |
enny | +0.247 |
pp | +0.246 |
Neg logit contributions | |
not | -0.208 |
the | -0.206 |
an | -0.201 |
making | -0.201 |
many | -0.190 |
Token | ." |
Token ID | 526 |
Feature activation | -6.74e+00 |
Pos logit contributions | |
azz | +0.010 |
pillar | +0.009 |
och | +0.009 |
hell | +0.009 |
pp | +0.009 |
Neg logit contributions | |
not | -0.004 |
the | -0.003 |
an | -0.003 |
making | -0.003 |
many | -0.003 |
Token | |
Token ID | 220 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
azz | +0.322 |
pillar | +0.298 |
och | +0.292 |
pp | +0.290 |
hell | +0.287 |
Neg logit contributions | |
the | -0.164 |
not | -0.156 |
an | -0.154 |
making | -0.149 |
many | -0.143 |
Token |
|
Token ID | 198 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
azz | +0.269 |
pillar | +0.251 |
och | +0.244 |
pp | +0.240 |
enny | +0.240 |
Neg logit contributions | |
not | -0.139 |
the | -0.138 |
an | -0.133 |
making | -0.132 |
many | -0.124 |
Token |
|
Token ID | 198 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
azz | +0.205 |
pillar | +0.190 |
och | +0.185 |
pp | +0.182 |
enny | +0.182 |
Neg logit contributions | |
not | -0.106 |
the | -0.105 |
an | -0.101 |
making | -0.100 |
many | -0.094 |
Token | The |
Token ID | 464 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
azz | +0.221 |
pillar | +0.203 |
och | +0.202 |
pp | +0.201 |
enny | +0.197 |
Neg logit contributions | |
the | -0.104 |
not | -0.103 |
an | -0.098 |
making | -0.094 |
many | -0.093 |
Token | camel |
Token ID | 41021 |
Feature activation | -1.94e+01 |
Pos logit contributions | |
azz | +0.120 |
pillar | +0.116 |
affe | +0.108 |
ppy | +0.106 |
pp | +0.104 |
Neg logit contributions | |
not | -0.112 |
making | -0.108 |
an | -0.107 |
the | -0.107 |
going | -0.103 |
Token | laughed |
Token ID | 13818 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
azz | +0.796 |
pillar | +0.728 |
och | +0.712 |
enny | +0.696 |
pp | +0.693 |
Neg logit contributions | |
not | -0.588 |
the | -0.584 |
an | -0.561 |
making | -0.558 |
many | -0.530 |
Token | and |
Token ID | 290 |
Feature activation | +4.65e+00 |
Pos logit contributions | |
not | +0.051 |
the | +0.050 |
making | +0.050 |
an | +0.049 |
going | +0.046 |
Neg logit contributions | |
azz | -0.081 |
pillar | -0.075 |
och | -0.072 |
enny | -0.071 |
hell | -0.071 |
Token | nodded |
Token ID | 14464 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
not | +0.144 |
the | +0.139 |
an | +0.139 |
making | +0.139 |
many | +0.131 |
Neg logit contributions | |
azz | -0.182 |
pillar | -0.167 |
och | -0.160 |
pp | -0.158 |
enny | -0.157 |
Token | his |
Token ID | 465 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
not | +0.036 |
making | +0.035 |
many | +0.033 |
an | +0.033 |
going | +0.033 |
Neg logit contributions | |
azz | -0.071 |
y | -0.065 |
pillar | -0.065 |
ppy | -0.063 |
hell | -0.063 |
Token | head |
Token ID | 1182 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
azz | +0.055 |
affe | +0.055 |
pillar | +0.054 |
hell | +0.052 |
suit | +0.051 |
Neg logit contributions | |
making | -0.040 |
the | -0.039 |
an | -0.039 |
not | -0.037 |
going | -0.037 |
Token | fight |
Token ID | 1907 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
the | +0.141 |
making | +0.138 |
an | +0.137 |
not | +0.135 |
many | +0.126 |
Neg logit contributions | |
azz | -0.435 |
pillar | -0.406 |
och | -0.397 |
hell | -0.396 |
pp | -0.394 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.91e+00 |
Pos logit contributions | |
not | +0.027 |
making | +0.027 |
an | +0.026 |
the | +0.026 |
going | +0.025 |
Neg logit contributions | |
azz | -0.080 |
pillar | -0.074 |
hell | -0.072 |
affe | -0.072 |
och | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | -4.71e+00 |
Pos logit contributions | |
azz | +0.224 |
pillar | +0.207 |
och | +0.202 |
enny | +0.199 |
affe | +0.197 |
Neg logit contributions | |
not | -0.122 |
the | -0.119 |
making | -0.118 |
an | -0.116 |
going | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
azz | +0.232 |
pillar | +0.218 |
pp | +0.216 |
och | +0.213 |
hell | +0.210 |
Neg logit contributions | |
the | -0.106 |
not | -0.104 |
an | -0.099 |
making | -0.091 |
many | -0.091 |
Token | The |
Token ID | 464 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
azz | +0.192 |
pp | +0.184 |
pillar | +0.176 |
hell | +0.174 |
och | +0.172 |
Neg logit contributions | |
the | -0.151 |
not | -0.151 |
an | -0.145 |
many | -0.136 |
making | -0.130 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.93e+01 |
Pos logit contributions | |
azz | +0.173 |
pillar | +0.157 |
och | +0.154 |
enny | +0.152 |
pp | +0.152 |
Neg logit contributions | |
not | -0.118 |
the | -0.118 |
an | -0.114 |
making | -0.113 |
many | -0.108 |
Token | was |
Token ID | 373 |
Feature activation | +8.77e+00 |
Pos logit contributions | |
azz | +0.808 |
pillar | +0.742 |
och | +0.732 |
enny | +0.714 |
pp | +0.711 |
Neg logit contributions | |
not | -0.570 |
the | -0.564 |
an | -0.541 |
making | -0.539 |
many | -0.506 |
Token | surprised |
Token ID | 6655 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
making | +0.167 |
the | +0.164 |
adding | +0.163 |
an | +0.160 |
many | +0.156 |
Neg logit contributions | |
azz | -0.391 |
hell | -0.359 |
pillar | -0.357 |
ird | -0.353 |
itty | -0.350 |
Token | . |
Token ID | 13 |
Feature activation | -5.77e+00 |
Pos logit contributions | |
the | +0.110 |
not | +0.110 |
an | +0.103 |
making | +0.103 |
going | +0.095 |
Neg logit contributions | |
azz | -0.205 |
pillar | -0.192 |
och | -0.186 |
pp | -0.184 |
enny | -0.182 |
Token | He |
Token ID | 679 |
Feature activation | -6.98e+00 |
Pos logit contributions | |
azz | +0.249 |
pillar | +0.231 |
och | +0.224 |
pp | +0.221 |
enny | +0.220 |
Neg logit contributions | |
not | -0.160 |
the | -0.159 |
an | -0.154 |
making | -0.152 |
many | -0.144 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
azz | +0.278 |
pillar | +0.256 |
och | +0.245 |
affe | +0.242 |
enny | +0.241 |
Neg logit contributions | |
not | -0.212 |
the | -0.210 |
making | -0.206 |
an | -0.206 |
many | -0.195 |
Token | ." |
Token ID | 526 |
Feature activation | -6.99e+00 |
Pos logit contributions | |
not | +0.063 |
the | +0.062 |
an | +0.060 |
making | +0.059 |
many | +0.055 |
Neg logit contributions | |
azz | -0.131 |
pillar | -0.122 |
och | -0.119 |
enny | -0.117 |
pp | -0.117 |
Token | |
Token ID | 220 |
Feature activation | -5.90e+00 |
Pos logit contributions | |
azz | +0.355 |
pillar | +0.334 |
och | +0.326 |
pp | +0.323 |
ppy | +0.322 |
Neg logit contributions | |
the | -0.160 |
not | -0.146 |
an | -0.144 |
making | -0.139 |
many | -0.132 |
Token |
|
Token ID | 198 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
azz | +0.295 |
pillar | +0.283 |
och | +0.273 |
pp | +0.272 |
affe | +0.268 |
Neg logit contributions | |
not | -0.128 |
the | -0.127 |
an | -0.119 |
many | -0.112 |
making | -0.112 |
Token |
|
Token ID | 198 |
Feature activation | -4.70e+00 |
Pos logit contributions | |
azz | +0.212 |
pillar | +0.198 |
och | +0.193 |
pp | +0.190 |
enny | +0.189 |
Neg logit contributions | |
not | -0.107 |
the | -0.106 |
an | -0.101 |
making | -0.100 |
many | -0.095 |
Token | The |
Token ID | 464 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
azz | +0.204 |
pillar | +0.188 |
och | +0.185 |
pp | +0.182 |
enny | +0.180 |
Neg logit contributions | |
not | -0.136 |
the | -0.133 |
an | -0.128 |
making | -0.126 |
many | -0.123 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.93e+01 |
Pos logit contributions | |
azz | +0.147 |
pillar | +0.138 |
hell | +0.130 |
affe | +0.130 |
och | +0.128 |
Neg logit contributions | |
not | -0.115 |
making | -0.114 |
the | -0.113 |
an | -0.112 |
adding | -0.104 |
Token | went |
Token ID | 1816 |
Feature activation | +5.49e+00 |
Pos logit contributions | |
azz | +0.846 |
pillar | +0.779 |
och | +0.767 |
enny | +0.745 |
pp | +0.744 |
Neg logit contributions | |
not | -0.530 |
the | -0.525 |
an | -0.505 |
making | -0.501 |
many | -0.473 |
Token | to |
Token ID | 284 |
Feature activation | +9.00e+00 |
Pos logit contributions | |
not | +0.069 |
an | +0.064 |
making | +0.063 |
adding | +0.062 |
preventing | +0.060 |
Neg logit contributions | |
azz | -0.296 |
och | -0.266 |
y | -0.266 |
pp | -0.265 |
enny | -0.262 |
Token | the |
Token ID | 262 |
Feature activation | -4.99e-01 |
Pos logit contributions | |
facing | +0.095 |
making | +0.095 |
not | +0.093 |
preventing | +0.093 |
adding | +0.090 |
Neg logit contributions | |
azz | -0.444 |
hel | -0.392 |
ird | -0.389 |
of | -0.387 |
enny | -0.385 |
Token | battlefield |
Token ID | 13480 |
Feature activation | -8.81e+00 |
Pos logit contributions | |
azz | +0.018 |
pillar | +0.018 |
affe | +0.016 |
y | +0.016 |
ari | +0.015 |
Neg logit contributions | |
not | -0.013 |
making | -0.013 |
an | -0.013 |
preventing | -0.013 |
adding | -0.013 |
Token | with |
Token ID | 351 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
azz | +0.378 |
pillar | +0.348 |
enny | +0.338 |
y | +0.336 |
hell | +0.335 |
Neg logit contributions | |
the | -0.218 |
not | -0.218 |
an | -0.215 |
making | -0.215 |
going | -0.209 |
Token | real |
Token ID | 1103 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
not | +0.090 |
the | +0.089 |
an | +0.086 |
making | +0.086 |
many | +0.082 |
Neg logit contributions | |
azz | -0.106 |
pillar | -0.097 |
och | -0.094 |
enny | -0.092 |
pp | -0.091 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
azz | +0.049 |
pillar | +0.044 |
och | +0.044 |
enny | +0.043 |
pp | +0.042 |
Neg logit contributions | |
not | -0.044 |
the | -0.044 |
an | -0.042 |
making | -0.042 |
many | -0.040 |
Token | anymore |
Token ID | 7471 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
azz | +0.780 |
och | +0.719 |
enny | +0.719 |
pillar | +0.717 |
oped | +0.705 |
Neg logit contributions | |
not | -0.305 |
the | -0.301 |
to | -0.284 |
making | -0.262 |
an | -0.255 |
Token | ," |
Token ID | 553 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
azz | +0.103 |
pillar | +0.095 |
och | +0.093 |
enny | +0.091 |
pp | +0.091 |
Neg logit contributions | |
not | -0.061 |
the | -0.061 |
an | -0.059 |
making | -0.058 |
many | -0.055 |
Token | the |
Token ID | 262 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
azz | +0.223 |
pillar | +0.216 |
affe | +0.212 |
och | +0.202 |
itty | +0.202 |
Neg logit contributions | |
not | -0.044 |
going | -0.039 |
making | -0.039 |
saving | -0.037 |
exploring | -0.034 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.93e+01 |
Pos logit contributions | |
azz | +0.125 |
pillar | +0.112 |
och | +0.109 |
enny | +0.107 |
pp | +0.106 |
Neg logit contributions | |
not | -0.132 |
the | -0.131 |
an | -0.128 |
making | -0.127 |
many | -0.122 |
Token | said |
Token ID | 531 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
azz | +0.830 |
och | +0.796 |
pillar | +0.790 |
enny | +0.751 |
itty | +0.747 |
Neg logit contributions | |
not | -0.566 |
the | -0.557 |
an | -0.522 |
making | -0.496 |
many | -0.485 |
Token | . |
Token ID | 13 |
Feature activation | -5.94e+00 |
Pos logit contributions | |
azz | +0.066 |
hell | +0.059 |
y | +0.059 |
pillar | +0.059 |
ppy | +0.058 |
Neg logit contributions | |
not | -0.052 |
making | -0.052 |
an | -0.051 |
climbing | -0.050 |
going | -0.050 |
Token |
|
Token ID | 198 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
azz | +0.245 |
pillar | +0.221 |
affe | +0.210 |
enny | +0.210 |
och | +0.208 |
Neg logit contributions | |
not | -0.174 |
making | -0.171 |
going | -0.168 |
an | -0.163 |
adding | -0.160 |
Token |
|
Token ID | 198 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
azz | +0.196 |
pillar | +0.181 |
och | +0.176 |
enny | +0.174 |
hell | +0.173 |
Neg logit contributions | |
not | -0.106 |
the | -0.104 |
making | -0.102 |
an | -0.102 |
going | -0.095 |
Token | " |
Token ID | 1 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
azz | +0.186 |
pillar | +0.172 |
och | +0.164 |
affe | +0.164 |
enny | +0.163 |
Neg logit contributions | |
not | -0.129 |
making | -0.127 |
the | -0.127 |
an | -0.125 |
going | -0.121 |
Token | delight |
Token ID | 10974 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
not | +0.179 |
the | +0.172 |
making | +0.168 |
an | +0.168 |
many | +0.155 |
Neg logit contributions | |
azz | -0.419 |
pillar | -0.391 |
och | -0.379 |
hell | -0.375 |
enny | -0.375 |
Token | . |
Token ID | 13 |
Feature activation | -4.95e+00 |
Pos logit contributions | |
azz | +0.069 |
pillar | +0.064 |
och | +0.062 |
enny | +0.061 |
itty | +0.061 |
Neg logit contributions | |
not | -0.034 |
the | -0.034 |
making | -0.032 |
an | -0.032 |
many | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -3.99e+00 |
Pos logit contributions | |
azz | +0.238 |
pillar | +0.226 |
pp | +0.221 |
hell | +0.220 |
och | +0.219 |
Neg logit contributions | |
the | -0.113 |
not | -0.110 |
an | -0.107 |
making | -0.102 |
many | -0.098 |
Token |
|
Token ID | 198 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
azz | +0.201 |
pp | +0.192 |
pillar | +0.190 |
och | +0.186 |
affe | +0.185 |
Neg logit contributions | |
the | -0.084 |
not | -0.081 |
an | -0.078 |
many | -0.073 |
making | -0.069 |
Token | The |
Token ID | 464 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
azz | +0.181 |
pp | +0.170 |
pillar | +0.164 |
hell | +0.164 |
och | +0.164 |
Neg logit contributions | |
not | -0.126 |
the | -0.124 |
an | -0.119 |
many | -0.113 |
making | -0.110 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
azz | +0.155 |
pillar | +0.141 |
och | +0.139 |
pp | +0.137 |
enny | +0.137 |
Neg logit contributions | |
the | -0.088 |
not | -0.088 |
an | -0.084 |
making | -0.084 |
many | -0.079 |
Token | laughed |
Token ID | 13818 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
azz | +0.827 |
och | +0.758 |
pillar | +0.752 |
enny | +0.729 |
pp | +0.728 |
Neg logit contributions | |
the | -0.558 |
not | -0.556 |
making | -0.529 |
an | -0.521 |
many | -0.488 |
Token | and |
Token ID | 290 |
Feature activation | +4.87e+00 |
Pos logit contributions | |
the | +0.036 |
not | +0.036 |
an | +0.034 |
making | +0.034 |
many | +0.031 |
Neg logit contributions | |
azz | -0.069 |
pillar | -0.064 |
och | -0.063 |
pp | -0.061 |
enny | -0.061 |
Token | cl |
Token ID | 537 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
not | +0.145 |
the | +0.142 |
an | +0.140 |
making | +0.139 |
many | +0.132 |
Neg logit contributions | |
azz | -0.199 |
pillar | -0.183 |
och | -0.176 |
enny | -0.173 |
pp | -0.173 |
Token | apped |
Token ID | 6320 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
the | +0.016 |
an | +0.016 |
not | +0.016 |
his | +0.015 |
making | +0.015 |
Neg logit contributions | |
azz | -0.022 |
pillar | -0.021 |
hell | -0.020 |
och | -0.020 |
affe | -0.020 |
Token | his |
Token ID | 465 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
not | +0.051 |
the | +0.049 |
making | +0.048 |
an | +0.048 |
going | +0.043 |
Neg logit contributions | |
azz | -0.170 |
pillar | -0.159 |
och | -0.155 |
hell | -0.154 |
enny | -0.154 |
Token | sunlight |
Token ID | 19606 |
Feature activation | -6.11e+00 |
Pos logit contributions | |
azz | +0.035 |
pillar | +0.034 |
och | +0.032 |
hell | +0.032 |
affe | +0.032 |
Neg logit contributions | |
not | -0.022 |
making | -0.021 |
the | -0.020 |
an | -0.020 |
going | -0.019 |
Token | ! |
Token ID | 0 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
azz | +0.295 |
pillar | +0.270 |
enny | +0.263 |
och | +0.263 |
affe | +0.261 |
Neg logit contributions | |
the | -0.154 |
not | -0.149 |
making | -0.147 |
an | -0.142 |
going | -0.134 |
Token |
|
Token ID | 198 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
azz | +0.153 |
pillar | +0.142 |
och | +0.139 |
pp | +0.138 |
hell | +0.136 |
Neg logit contributions | |
the | -0.083 |
not | -0.082 |
an | -0.080 |
making | -0.077 |
many | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
azz | +0.176 |
pillar | +0.164 |
pp | +0.163 |
och | +0.162 |
affe | +0.158 |
Neg logit contributions | |
the | -0.082 |
not | -0.081 |
an | -0.078 |
making | -0.073 |
many | -0.073 |
Token | The |
Token ID | 464 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
azz | +0.180 |
pp | +0.171 |
och | +0.165 |
pillar | +0.161 |
hell | +0.159 |
Neg logit contributions | |
the | -0.111 |
not | -0.108 |
an | -0.108 |
many | -0.101 |
making | -0.096 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
azz | +0.152 |
pillar | +0.139 |
och | +0.135 |
enny | +0.133 |
pp | +0.133 |
Neg logit contributions | |
not | -0.110 |
the | -0.109 |
an | -0.106 |
making | -0.105 |
many | -0.100 |
Token | was |
Token ID | 373 |
Feature activation | +8.77e+00 |
Pos logit contributions | |
azz | +0.837 |
pillar | +0.768 |
och | +0.758 |
enny | +0.737 |
pp | +0.735 |
Neg logit contributions | |
not | -0.536 |
the | -0.534 |
an | -0.513 |
making | -0.508 |
many | -0.480 |
Token | so |
Token ID | 523 |
Feature activation | +6.92e+00 |
Pos logit contributions | |
help | +0.162 |
making | +0.157 |
the | +0.156 |
your | +0.156 |
warnings | +0.154 |
Neg logit contributions | |
azz | -0.343 |
y | -0.330 |
ird | -0.329 |
pillar | -0.327 |
itty | -0.320 |
Token | excited |
Token ID | 6568 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
making | +0.166 |
an | +0.152 |
saving | +0.152 |
the | +0.152 |
not | +0.149 |
Neg logit contributions | |
azz | -0.286 |
pillar | -0.271 |
y | -0.268 |
hell | -0.263 |
enny | -0.261 |
Token | that |
Token ID | 326 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
not | +0.064 |
the | +0.063 |
an | +0.061 |
making | +0.060 |
many | +0.057 |
Neg logit contributions | |
azz | -0.139 |
pillar | -0.130 |
och | -0.125 |
enny | -0.124 |
hell | -0.124 |
Token | he |
Token ID | 339 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
making | +0.024 |
saving | +0.024 |
not | +0.024 |
preventing | +0.023 |
adding | +0.022 |
Neg logit contributions | |
azz | -0.032 |
pillar | -0.030 |
hell | -0.030 |
y | -0.029 |
och | -0.029 |
Token | go |
Token ID | 467 |
Feature activation | +6.69e+00 |
Pos logit contributions | |
not | +0.071 |
the | +0.070 |
making | +0.068 |
an | +0.068 |
many | +0.064 |
Neg logit contributions | |
azz | -0.179 |
pillar | -0.168 |
och | -0.163 |
enny | -0.161 |
pp | -0.160 |
Token | explore |
Token ID | 7301 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
not | +0.132 |
the | +0.123 |
making | +0.123 |
an | +0.121 |
many | +0.117 |
Neg logit contributions | |
azz | -0.333 |
pillar | -0.306 |
och | -0.301 |
pp | -0.301 |
enny | -0.300 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
not | +0.104 |
making | +0.095 |
an | +0.085 |
adding | +0.084 |
the | +0.084 |
Neg logit contributions | |
azz | -0.312 |
pillar | -0.296 |
hell | -0.286 |
och | -0.283 |
enny | -0.282 |
Token |
|
Token ID | 198 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
azz | +0.207 |
pillar | +0.199 |
pp | +0.194 |
hell | +0.194 |
och | +0.192 |
Neg logit contributions | |
the | -0.108 |
not | -0.097 |
an | -0.096 |
making | -0.089 |
to | -0.089 |
Token | The |
Token ID | 464 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
azz | +0.212 |
pp | +0.207 |
pillar | +0.204 |
och | +0.200 |
ppy | +0.197 |
Neg logit contributions | |
the | -0.082 |
not | -0.075 |
an | -0.073 |
many | -0.067 |
his | -0.065 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.92e+01 |
Pos logit contributions | |
azz | +0.179 |
pillar | +0.163 |
och | +0.161 |
pp | +0.160 |
enny | +0.160 |
Neg logit contributions | |
the | -0.101 |
not | -0.099 |
an | -0.095 |
making | -0.094 |
many | -0.090 |
Token | was |
Token ID | 373 |
Feature activation | +9.03e+00 |
Pos logit contributions | |
azz | +0.843 |
och | +0.788 |
pillar | +0.776 |
hell | +0.749 |
pp | +0.746 |
Neg logit contributions | |
the | -0.541 |
not | -0.536 |
an | -0.508 |
making | -0.499 |
many | -0.471 |
Token | over |
Token ID | 625 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
making | +0.152 |
the | +0.150 |
many | +0.148 |
not | +0.148 |
help | +0.147 |
Neg logit contributions | |
azz | -0.416 |
pillar | -0.382 |
ird | -0.379 |
y | -0.374 |
itty | -0.371 |
Token | joy |
Token ID | 2633 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
not | +0.240 |
making | +0.214 |
saving | +0.206 |
going | +0.204 |
adding | +0.203 |
Neg logit contributions | |
azz | -0.239 |
pillar | -0.216 |
y | -0.212 |
enny | -0.207 |
itty | -0.198 |
Token | ed |
Token ID | 276 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
azz | +0.121 |
och | +0.109 |
pp | +0.106 |
pillar | +0.105 |
ppy | +0.105 |
Neg logit contributions | |
the | -0.120 |
to | -0.119 |
an | -0.114 |
not | -0.108 |
making | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
not | +0.128 |
the | +0.123 |
an | +0.123 |
making | +0.123 |
many | +0.117 |
Neg logit contributions | |
azz | -0.193 |
pillar | -0.179 |
och | -0.174 |
enny | -0.170 |
pp | -0.170 |
Token | It |
Token ID | 632 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
azz | +0.228 |
pillar | +0.214 |
hell | +0.207 |
och | +0.207 |
pp | +0.206 |
Neg logit contributions | |
the | -0.128 |
not | -0.124 |
an | -0.122 |
making | -0.115 |
many | -0.111 |
Token | might |
Token ID | 1244 |
Feature activation | +5.33e+00 |
Pos logit contributions | |
azz | +0.234 |
pillar | +0.217 |
hell | +0.204 |
affe | +0.203 |
ppy | +0.201 |
Neg logit contributions | |
the | -0.169 |
not | -0.169 |
an | -0.167 |
making | -0.167 |
many | -0.162 |
Token | be |
Token ID | 307 |
Feature activation | +9.46e+00 |
Pos logit contributions | |
making | +0.096 |
the | +0.094 |
many | +0.094 |
an | +0.093 |
adding | +0.092 |
Neg logit contributions | |
azz | -0.258 |
pillar | -0.238 |
hell | -0.236 |
ppy | -0.232 |
affe | -0.232 |
Token | spicy |
Token ID | 26880 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
making | +0.159 |
not | +0.157 |
adding | +0.157 |
an | +0.151 |
the | +0.150 |
Neg logit contributions | |
azz | -0.456 |
ppy | -0.430 |
hell | -0.422 |
y | -0.411 |
pp | -0.407 |
Token | ." |
Token ID | 526 |
Feature activation | -6.58e+00 |
Pos logit contributions | |
azz | +0.023 |
pillar | +0.022 |
och | +0.021 |
hell | +0.021 |
affe | +0.020 |
Neg logit contributions | |
not | -0.013 |
the | -0.013 |
an | -0.012 |
making | -0.012 |
many | -0.012 |
Token | Nem |
Token ID | 22547 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
azz | +0.294 |
pillar | +0.273 |
och | +0.265 |
pp | +0.261 |
hell | +0.261 |
Neg logit contributions | |
the | -0.175 |
not | -0.174 |
an | -0.168 |
making | -0.164 |
many | -0.156 |
Token | o |
Token ID | 78 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
och | +0.632 |
azz | +0.616 |
ird | +0.612 |
hell | +0.600 |
pillar | +0.592 |
Neg logit contributions | |
not | -0.705 |
the | -0.691 |
to | -0.685 |
his | -0.644 |
an | -0.633 |
Token | thought |
Token ID | 1807 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
azz | +0.485 |
pillar | +0.443 |
och | +0.423 |
enny | +0.419 |
pp | +0.417 |
Neg logit contributions | |
not | -0.401 |
the | -0.395 |
making | -0.389 |
an | -0.388 |
going | -0.368 |
Token | for |
Token ID | 329 |
Feature activation | +9.91e+00 |
Pos logit contributions | |
not | +0.010 |
making | +0.009 |
an | +0.009 |
saving | +0.009 |
adding | +0.009 |
Neg logit contributions | |
azz | -0.021 |
hell | -0.019 |
pillar | -0.019 |
enny | -0.019 |
affe | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
making | +0.089 |
not | +0.088 |
adding | +0.080 |
climbing | +0.080 |
going | +0.072 |
Neg logit contributions | |
azz | -0.569 |
pillar | -0.535 |
hell | -0.533 |
ppy | -0.529 |
enny | -0.526 |
Token | moment |
Token ID | 2589 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
not | +0.018 |
the | +0.018 |
an | +0.017 |
making | +0.017 |
many | +0.016 |
Neg logit contributions | |
azz | -0.029 |
pillar | -0.027 |
och | -0.026 |
pp | -0.026 |
enny | -0.025 |
Token | yours |
Token ID | 12431 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
not | +0.162 |
making | +0.157 |
an | +0.154 |
the | +0.151 |
many | +0.146 |
Neg logit contributions | |
azz | -0.469 |
pp | -0.430 |
hell | -0.430 |
ppy | -0.429 |
pillar | -0.427 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
azz | +0.137 |
pillar | +0.127 |
och | +0.123 |
enny | +0.122 |
pp | +0.121 |
Neg logit contributions | |
not | -0.073 |
the | -0.073 |
an | -0.070 |
making | -0.070 |
many | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
azz | +0.224 |
pillar | +0.210 |
pp | +0.205 |
och | +0.205 |
hell | +0.204 |
Neg logit contributions | |
the | -0.111 |
not | -0.104 |
an | -0.101 |
making | -0.097 |
many | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | -4.33e+00 |
Pos logit contributions | |
azz | +0.209 |
pp | +0.201 |
pillar | +0.195 |
och | +0.195 |
ppy | +0.192 |
Neg logit contributions | |
the | -0.087 |
not | -0.081 |
an | -0.078 |
many | -0.074 |
making | -0.067 |
Token | The |
Token ID | 464 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
azz | +0.173 |
pp | +0.165 |
och | +0.158 |
suit | +0.156 |
ppy | +0.153 |
Neg logit contributions | |
not | -0.135 |
the | -0.134 |
an | -0.128 |
many | -0.124 |
next | -0.118 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
azz | +0.163 |
pillar | +0.150 |
och | +0.146 |
enny | +0.144 |
pp | +0.144 |
Neg logit contributions | |
not | -0.095 |
the | -0.095 |
an | -0.091 |
making | -0.091 |
many | -0.085 |
Token | was |
Token ID | 373 |
Feature activation | +8.99e+00 |
Pos logit contributions | |
azz | +0.833 |
och | +0.763 |
pillar | +0.761 |
enny | +0.734 |
pp | +0.733 |
Neg logit contributions | |
not | -0.545 |
the | -0.544 |
an | -0.514 |
making | -0.512 |
many | -0.481 |
Token | very |
Token ID | 845 |
Feature activation | +6.25e+00 |
Pos logit contributions | |
making | +0.156 |
help | +0.153 |
the | +0.149 |
an | +0.147 |
adding | +0.147 |
Neg logit contributions | |
azz | -0.399 |
ird | -0.379 |
hell | -0.370 |
pillar | -0.368 |
y | -0.361 |
Token | impatient |
Token ID | 33440 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
making | +0.138 |
an | +0.137 |
the | +0.132 |
saving | +0.129 |
going | +0.126 |
Neg logit contributions | |
azz | -0.273 |
hell | -0.253 |
pillar | -0.251 |
och | -0.249 |
ird | -0.248 |
Token | and |
Token ID | 290 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
the | +0.030 |
not | +0.029 |
making | +0.028 |
an | +0.028 |
going | +0.026 |
Neg logit contributions | |
azz | -0.058 |
pillar | -0.054 |
och | -0.053 |
pp | -0.052 |
enny | -0.051 |
Token | he |
Token ID | 339 |
Feature activation | -5.94e+00 |
Pos logit contributions | |
making | +0.146 |
an | +0.144 |
not | +0.141 |
the | +0.140 |
many | +0.137 |
Neg logit contributions | |
azz | -0.201 |
pillar | -0.184 |
affe | -0.178 |
och | -0.177 |
enny | -0.176 |
Token | walls |
Token ID | 7714 |
Feature activation | -6.16e+00 |
Pos logit contributions | |
not | +0.005 |
making | +0.004 |
an | +0.004 |
the | +0.004 |
going | +0.004 |
Neg logit contributions | |
azz | -0.009 |
pillar | -0.008 |
affe | -0.008 |
hell | -0.008 |
y | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
azz | +0.267 |
pillar | +0.247 |
och | +0.240 |
hell | +0.237 |
pp | +0.236 |
Neg logit contributions | |
not | -0.165 |
the | -0.163 |
an | -0.160 |
making | -0.157 |
many | -0.148 |
Token |
|
Token ID | 198 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
azz | +0.192 |
pillar | +0.179 |
och | +0.175 |
pp | +0.172 |
enny | +0.171 |
Neg logit contributions | |
the | -0.109 |
not | -0.108 |
an | -0.104 |
making | -0.102 |
many | -0.098 |
Token |
|
Token ID | 198 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
azz | +0.150 |
pillar | +0.141 |
och | +0.139 |
pp | +0.138 |
ppy | +0.135 |
Neg logit contributions | |
the | -0.070 |
not | -0.069 |
an | -0.066 |
making | -0.063 |
many | -0.062 |
Token | The |
Token ID | 464 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
azz | +0.156 |
pillar | +0.143 |
och | +0.141 |
pp | +0.140 |
enny | +0.137 |
Neg logit contributions | |
the | -0.106 |
not | -0.106 |
an | -0.103 |
making | -0.100 |
many | -0.098 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
azz | +0.104 |
pillar | +0.094 |
affe | +0.088 |
och | +0.087 |
hell | +0.087 |
Neg logit contributions | |
not | -0.134 |
the | -0.132 |
making | -0.132 |
an | -0.131 |
adding | -0.125 |
Token | was |
Token ID | 373 |
Feature activation | +8.77e+00 |
Pos logit contributions | |
azz | +0.754 |
pillar | +0.693 |
och | +0.659 |
pp | +0.655 |
affe | +0.655 |
Neg logit contributions | |
not | -0.589 |
the | -0.582 |
an | -0.573 |
making | -0.572 |
going | -0.540 |
Token | very |
Token ID | 845 |
Feature activation | +5.92e+00 |
Pos logit contributions | |
instructions | +0.188 |
advice | +0.186 |
safety | +0.176 |
warnings | +0.176 |
help | +0.172 |
Neg logit contributions | |
azz | -0.321 |
y | -0.310 |
ppy | -0.299 |
pp | -0.292 |
itty | -0.292 |
Token | brave |
Token ID | 14802 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
an | +0.126 |
making | +0.126 |
adding | +0.117 |
the | +0.115 |
saving | +0.114 |
Neg logit contributions | |
azz | -0.262 |
y | -0.240 |
pp | -0.238 |
pillar | -0.237 |
och | -0.236 |
Token | and |
Token ID | 290 |
Feature activation | +4.42e+00 |
Pos logit contributions | |
azz | +0.040 |
pillar | +0.037 |
och | +0.037 |
pp | +0.036 |
enny | +0.036 |
Neg logit contributions | |
the | -0.023 |
not | -0.023 |
an | -0.022 |
making | -0.022 |
many | -0.021 |
Token | looked |
Token ID | 3114 |
Feature activation | +6.30e+00 |
Pos logit contributions | |
adding | +0.114 |
making | +0.113 |
an | +0.108 |
saving | +0.105 |
climbing | +0.105 |
Neg logit contributions | |
azz | -0.166 |
affe | -0.156 |
pillar | -0.153 |
ppy | -0.148 |
y | -0.146 |
Token | you |
Token ID | 345 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
an | +0.171 |
not | +0.168 |
making | +0.162 |
saving | +0.154 |
exploring | +0.150 |
Neg logit contributions | |
azz | -0.589 |
pillar | -0.564 |
hell | -0.546 |
affe | -0.539 |
y | -0.529 |
Token | ." |
Token ID | 526 |
Feature activation | -6.64e+00 |
Pos logit contributions | |
azz | +0.082 |
pillar | +0.076 |
och | +0.074 |
pp | +0.073 |
hell | +0.073 |
Neg logit contributions | |
not | -0.046 |
the | -0.046 |
an | -0.045 |
making | -0.044 |
many | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
azz | +0.312 |
pillar | +0.290 |
och | +0.282 |
pp | +0.278 |
enny | +0.277 |
Neg logit contributions | |
not | -0.161 |
the | -0.161 |
an | -0.154 |
making | -0.152 |
many | -0.143 |
Token |
|
Token ID | 198 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
azz | +0.212 |
pillar | +0.201 |
pp | +0.199 |
och | +0.197 |
ppy | +0.192 |
Neg logit contributions | |
the | -0.093 |
not | -0.089 |
an | -0.083 |
many | -0.079 |
making | -0.077 |
Token | The |
Token ID | 464 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
azz | +0.150 |
pp | +0.145 |
pillar | +0.136 |
hell | +0.135 |
suit | +0.133 |
Neg logit contributions | |
the | -0.162 |
not | -0.161 |
next | -0.149 |
an | -0.149 |
many | -0.149 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
azz | +0.162 |
pillar | +0.150 |
och | +0.146 |
enny | +0.144 |
pp | +0.143 |
Neg logit contributions | |
not | -0.100 |
the | -0.100 |
an | -0.096 |
making | -0.096 |
many | -0.090 |
Token | was |
Token ID | 373 |
Feature activation | +8.77e+00 |
Pos logit contributions | |
azz | +0.796 |
pillar | +0.732 |
och | +0.714 |
enny | +0.700 |
pp | +0.696 |
Neg logit contributions | |
not | -0.565 |
the | -0.559 |
an | -0.541 |
making | -0.538 |
many | -0.508 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
adding | +0.182 |
exploring | +0.182 |
making | +0.182 |
shore | +0.179 |
help | +0.177 |
Neg logit contributions | |
azz | -0.342 |
hell | -0.323 |
ird | -0.310 |
itty | -0.307 |
och | -0.302 |
Token | . |
Token ID | 13 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
the | +0.068 |
not | +0.066 |
making | +0.062 |
an | +0.062 |
many | +0.059 |
Neg logit contributions | |
azz | -0.127 |
pillar | -0.119 |
pp | -0.117 |
och | -0.116 |
enny | -0.114 |
Token | He |
Token ID | 679 |
Feature activation | -6.44e+00 |
Pos logit contributions | |
azz | +0.205 |
pillar | +0.189 |
och | +0.183 |
enny | +0.180 |
hell | +0.179 |
Neg logit contributions | |
not | -0.145 |
the | -0.143 |
making | -0.139 |
an | -0.139 |
many | -0.131 |
Token | had |
Token ID | 550 |
Feature activation | +8.32e+00 |
Pos logit contributions | |
azz | +0.244 |
pillar | +0.223 |
affe | +0.212 |
och | +0.211 |
hell | +0.210 |
Neg logit contributions | |
the | -0.204 |
not | -0.204 |
making | -0.201 |
an | -0.200 |
many | -0.191 |
Token | it |
Token ID | 340 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
the | +0.096 |
not | +0.095 |
an | +0.087 |
making | +0.085 |
many | +0.074 |
Neg logit contributions | |
azz | -0.490 |
pillar | -0.462 |
och | -0.452 |
enny | -0.448 |
pp | -0.446 |
Token | in |
Token ID | 287 |
Feature activation | +9.58e+00 |
Pos logit contributions | |
azz | +0.202 |
hell | +0.191 |
pillar | +0.188 |
ppy | +0.188 |
y | +0.188 |
Neg logit contributions | |
not | -0.079 |
making | -0.077 |
the | -0.077 |
an | -0.076 |
adding | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
not | +0.067 |
making | +0.058 |
going | +0.048 |
an | +0.045 |
the | +0.044 |
Neg logit contributions | |
azz | -0.586 |
pillar | -0.553 |
och | -0.546 |
hell | -0.543 |
enny | -0.537 |
Token | trash |
Token ID | 13913 |
Feature activation | -6.83e+00 |
Pos logit contributions | |
not | +0.008 |
making | +0.007 |
the | +0.007 |
an | +0.007 |
going | +0.007 |
Neg logit contributions | |
azz | -0.012 |
pillar | -0.011 |
och | -0.011 |
hell | -0.011 |
affe | -0.011 |
Token | ." |
Token ID | 526 |
Feature activation | -6.64e+00 |
Pos logit contributions | |
azz | +0.341 |
pillar | +0.312 |
pp | +0.308 |
affe | +0.306 |
hell | +0.305 |
Neg logit contributions | |
the | -0.163 |
not | -0.144 |
making | -0.143 |
an | -0.143 |
to | -0.137 |
Token | Nem |
Token ID | 22547 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
azz | +0.321 |
pillar | +0.300 |
pp | +0.295 |
och | +0.293 |
hell | +0.291 |
Neg logit contributions | |
the | -0.157 |
not | -0.153 |
an | -0.147 |
making | -0.141 |
many | -0.135 |
Token | o |
Token ID | 78 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
och | +0.594 |
azz | +0.586 |
pillar | +0.572 |
hell | +0.567 |
ird | +0.545 |
Neg logit contributions | |
not | -0.754 |
the | -0.741 |
to | -0.724 |
his | -0.693 |
an | -0.679 |
Token | and |
Token ID | 290 |
Feature activation | +3.22e+00 |
Pos logit contributions | |
azz | +0.510 |
pillar | +0.466 |
och | +0.448 |
enny | +0.442 |
pp | +0.440 |
Neg logit contributions | |
not | -0.425 |
the | -0.420 |
an | -0.411 |
making | -0.411 |
many | -0.389 |
Token | his |
Token ID | 465 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
the | +0.080 |
not | +0.078 |
an | +0.075 |
making | +0.073 |
many | +0.068 |
Neg logit contributions | |
azz | -0.153 |
pillar | -0.144 |
och | -0.141 |
hell | -0.138 |
pp | -0.138 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
azz | +0.165 |
pillar | +0.154 |
och | +0.151 |
pp | +0.149 |
enny | +0.149 |
Neg logit contributions | |
not | -0.055 |
the | -0.054 |
an | -0.052 |
making | -0.051 |
many | -0.047 |
Token | worked |
Token ID | 3111 |
Feature activation | +5.49e+00 |
Pos logit contributions | |
azz | +0.446 |
pillar | +0.416 |
och | +0.403 |
affe | +0.391 |
pp | +0.390 |
Neg logit contributions | |
the | -0.303 |
not | -0.300 |
an | -0.287 |
making | -0.284 |
many | -0.269 |
Token | . |
Token ID | 13 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
azz | +0.087 |
pillar | +0.081 |
och | +0.079 |
enny | +0.077 |
pp | +0.077 |
Neg logit contributions | |
not | -0.046 |
the | -0.046 |
an | -0.044 |
making | -0.044 |
many | -0.041 |
Token | |
Token ID | 220 |
Feature activation | -4.72e+00 |
Pos logit contributions | |
azz | +0.218 |
pillar | +0.204 |
pp | +0.200 |
och | +0.199 |
hell | +0.197 |
Neg logit contributions | |
the | -0.118 |
not | -0.117 |
an | -0.114 |
making | -0.109 |
many | -0.105 |
Token |
|
Token ID | 198 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
azz | +0.236 |
pillar | +0.228 |
pp | +0.221 |
och | +0.219 |
affe | +0.218 |
Neg logit contributions | |
the | -0.099 |
not | -0.099 |
an | -0.094 |
many | -0.088 |
making | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
azz | +0.173 |
pp | +0.165 |
pillar | +0.164 |
och | +0.160 |
affe | +0.159 |
Neg logit contributions | |
the | -0.077 |
not | -0.075 |
an | -0.072 |
many | -0.069 |
making | -0.064 |
Token | The |
Token ID | 464 |
Feature activation | -3.54e+00 |
Pos logit contributions | |
azz | +0.163 |
pp | +0.156 |
pillar | +0.150 |
och | +0.148 |
hell | +0.146 |
Neg logit contributions | |
not | -0.115 |
the | -0.114 |
an | -0.111 |
many | -0.106 |
next | -0.103 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
azz | +0.158 |
pillar | +0.145 |
och | +0.142 |
enny | +0.140 |
pp | +0.140 |
Neg logit contributions | |
not | -0.094 |
the | -0.093 |
an | -0.090 |
making | -0.090 |
many | -0.084 |
Token | had |
Token ID | 550 |
Feature activation | +7.99e+00 |
Pos logit contributions | |
azz | +0.818 |
pillar | +0.747 |
och | +0.741 |
enny | +0.720 |
hell | +0.718 |
Neg logit contributions | |
the | -0.547 |
not | -0.546 |
an | -0.522 |
making | -0.521 |
many | -0.489 |
Token | a |
Token ID | 257 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
making | +0.148 |
adding | +0.140 |
saving | +0.139 |
not | +0.136 |
helping | +0.135 |
Neg logit contributions | |
azz | -0.370 |
pillar | -0.346 |
hell | -0.330 |
enny | -0.328 |
ird | -0.322 |
Token | very |
Token ID | 845 |
Feature activation | +5.82e+00 |
Pos logit contributions | |
not | +0.030 |
the | +0.030 |
making | +0.028 |
an | +0.028 |
many | +0.026 |
Neg logit contributions | |
azz | -0.086 |
pillar | -0.081 |
och | -0.079 |
hell | -0.078 |
enny | -0.078 |
Token | successful |
Token ID | 4388 |
Feature activation | +7.94e-01 |
Pos logit contributions | |
making | +0.107 |
the | +0.104 |
an | +0.099 |
saving | +0.099 |
not | +0.095 |
Neg logit contributions | |
azz | -0.286 |
pillar | -0.276 |
hell | -0.273 |
affe | -0.269 |
enny | -0.261 |
Token | adventure |
Token ID | 8855 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
not | +0.018 |
the | +0.018 |
an | +0.017 |
making | +0.017 |
many | +0.016 |
Neg logit contributions | |
azz | -0.038 |
pillar | -0.035 |
och | -0.035 |
enny | -0.034 |
pp | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
azz | +0.154 |
pillar | +0.144 |
och | +0.141 |
pp | +0.139 |
enny | +0.137 |
Neg logit contributions | |
not | -0.077 |
the | -0.077 |
an | -0.074 |
making | -0.072 |
many | -0.069 |
Token | One |
Token ID | 3198 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
azz | +0.164 |
pillar | +0.152 |
och | +0.151 |
pp | +0.150 |
enny | +0.146 |
Neg logit contributions | |
not | -0.106 |
the | -0.106 |
an | -0.103 |
many | -0.099 |
making | -0.098 |
Token | day |
Token ID | 1110 |
Feature activation | -3.75e+00 |
Pos logit contributions | |
azz | +0.119 |
pillar | +0.112 |
och | +0.109 |
enny | +0.108 |
hell | +0.107 |
Neg logit contributions | |
not | -0.045 |
the | -0.044 |
making | -0.042 |
an | -0.042 |
many | -0.038 |
Token | , |
Token ID | 11 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
azz | +0.146 |
pillar | +0.136 |
enny | +0.130 |
y | +0.130 |
och | +0.129 |
Neg logit contributions | |
not | -0.112 |
making | -0.105 |
an | -0.101 |
going | -0.101 |
the | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
not | +0.014 |
making | +0.012 |
going | +0.012 |
climbing | +0.012 |
adding | +0.011 |
Neg logit contributions | |
azz | -0.073 |
pillar | -0.069 |
affe | -0.068 |
y | -0.068 |
enny | -0.067 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
azz | +0.071 |
pillar | +0.064 |
och | +0.060 |
hell | +0.060 |
affe | +0.060 |
Neg logit contributions | |
not | -0.094 |
making | -0.093 |
the | -0.092 |
an | -0.092 |
adding | -0.088 |
Token | decided |
Token ID | 3066 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
azz | +0.837 |
och | +0.783 |
pillar | +0.766 |
hell | +0.756 |
pp | +0.744 |
Neg logit contributions | |
the | -0.533 |
not | -0.530 |
an | -0.509 |
making | -0.497 |
many | -0.472 |
Token | to |
Token ID | 284 |
Feature activation | +9.19e+00 |
Pos logit contributions | |
adding | +0.071 |
making | +0.066 |
example | +0.061 |
accomplishment | +0.061 |
saving | +0.058 |
Neg logit contributions | |
azz | -0.198 |
pillar | -0.179 |
affe | -0.177 |
y | -0.175 |
enny | -0.173 |
Token | go |
Token ID | 467 |
Feature activation | +6.66e+00 |
Pos logit contributions | |
making | +0.195 |
not | +0.187 |
the | +0.182 |
an | +0.180 |
many | +0.178 |
Neg logit contributions | |
azz | -0.431 |
pillar | -0.401 |
affe | -0.384 |
ird | -0.383 |
enny | -0.379 |
Token | for |
Token ID | 329 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
not | +0.120 |
adding | +0.113 |
preventing | +0.113 |
nothing | +0.109 |
making | +0.105 |
Neg logit contributions | |
azz | -0.313 |
pp | -0.291 |
och | -0.280 |
itty | -0.277 |
y | -0.276 |
Token | a |
Token ID | 257 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
adding | +0.176 |
making | +0.171 |
preventing | +0.165 |
not | +0.163 |
up | +0.148 |
Neg logit contributions | |
azz | -0.500 |
ird | -0.467 |
hell | -0.462 |
ppy | -0.443 |
pillar | -0.441 |
Token | garden |
Token ID | 11376 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
azz | +0.040 |
pillar | +0.037 |
och | +0.036 |
enny | +0.036 |
pp | +0.036 |
Neg logit contributions | |
the | -0.020 |
not | -0.019 |
an | -0.019 |
making | -0.019 |
many | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
azz | +0.277 |
pillar | +0.256 |
och | +0.249 |
enny | +0.247 |
pp | +0.245 |
Neg logit contributions | |
not | -0.139 |
the | -0.139 |
making | -0.133 |
an | -0.132 |
many | -0.123 |
Token |
|
Token ID | 198 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
azz | +0.243 |
pillar | +0.227 |
pp | +0.227 |
hell | +0.221 |
och | +0.221 |
Neg logit contributions | |
the | -0.128 |
an | -0.123 |
not | -0.123 |
making | -0.115 |
many | -0.113 |
Token |
|
Token ID | 198 |
Feature activation | -4.28e+00 |
Pos logit contributions | |
azz | +0.190 |
pillar | +0.178 |
pp | +0.177 |
och | +0.175 |
affe | +0.172 |
Neg logit contributions | |
the | -0.089 |
not | -0.087 |
an | -0.084 |
many | -0.079 |
making | -0.079 |
Token | The |
Token ID | 464 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
azz | +0.174 |
pp | +0.168 |
pillar | +0.158 |
och | +0.157 |
affe | +0.156 |
Neg logit contributions | |
the | -0.128 |
not | -0.125 |
an | -0.124 |
many | -0.119 |
his | -0.112 |
Token | prince |
Token ID | 19716 |
Feature activation | -1.91e+01 |
Pos logit contributions | |
azz | +0.149 |
pillar | +0.140 |
och | +0.135 |
enny | +0.132 |
hell | +0.132 |
Neg logit contributions | |
not | -0.096 |
the | -0.094 |
an | -0.092 |
making | -0.092 |
going | -0.086 |
Token | had |
Token ID | 550 |
Feature activation | +7.78e+00 |
Pos logit contributions | |
azz | +0.837 |
pillar | +0.764 |
och | +0.763 |
hell | +0.738 |
pp | +0.737 |
Neg logit contributions | |
the | -0.535 |
not | -0.531 |
an | -0.505 |
making | -0.505 |
many | -0.474 |
Token | solved |
Token ID | 16019 |
Feature activation | +5.68e+00 |
Pos logit contributions | |
making | +0.159 |
not | +0.150 |
adding | +0.150 |
saving | +0.147 |
helping | +0.146 |
Neg logit contributions | |
azz | -0.351 |
pillar | -0.329 |
enny | -0.314 |
hell | -0.313 |
ird | -0.306 |
Token | the |
Token ID | 262 |
Feature activation | +3.96e-01 |
Pos logit contributions | |
not | +0.037 |
making | +0.030 |
going | +0.023 |
help | +0.021 |
helping | +0.020 |
Neg logit contributions | |
azz | -0.339 |
pillar | -0.318 |
hell | -0.318 |
y | -0.318 |
och | -0.312 |
Token | puzzle |
Token ID | 15027 |
Feature activation | -7.67e+00 |
Pos logit contributions | |
not | +0.011 |
the | +0.010 |
making | +0.010 |
an | +0.010 |
his | +0.010 |
Neg logit contributions | |
pillar | -0.016 |
azz | -0.016 |
hell | -0.015 |
y | -0.014 |
och | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +9.00e+00 |
Pos logit contributions | |
azz | +0.342 |
pillar | +0.319 |
och | +0.309 |
enny | +0.304 |
hell | +0.304 |
Neg logit contributions | |
not | -0.197 |
the | -0.193 |
an | -0.189 |
making | -0.187 |
going | -0.175 |
Token | it |
Token ID | 340 |
Feature activation | -4.80e+00 |
Pos logit contributions | |
not | +0.160 |
making | +0.142 |
an | +0.134 |
adding | +0.133 |
many | +0.132 |
Neg logit contributions | |
azz | -0.536 |
pillar | -0.508 |
hell | -0.496 |
affe | -0.492 |
och | -0.486 |
Token | . |
Token ID | 13 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
azz | +0.205 |
pillar | +0.189 |
och | +0.182 |
hell | +0.182 |
pp | +0.181 |
Neg logit contributions | |
not | -0.127 |
the | -0.124 |
an | -0.122 |
making | -0.122 |
many | -0.116 |
Token | He |
Token ID | 679 |
Feature activation | -6.22e+00 |
Pos logit contributions | |
azz | +0.178 |
pillar | +0.164 |
affe | +0.160 |
enny | +0.156 |
och | +0.153 |
Neg logit contributions | |
not | -0.144 |
making | -0.142 |
going | -0.138 |
an | -0.137 |
the | -0.136 |
Token | looked |
Token ID | 3114 |
Feature activation | +5.88e+00 |
Pos logit contributions | |
azz | +0.269 |
pillar | +0.249 |
och | +0.239 |
enny | +0.237 |
pp | +0.236 |
Neg logit contributions | |
not | -0.172 |
the | -0.170 |
making | -0.165 |
an | -0.165 |
many | -0.155 |
Token | and |
Token ID | 290 |
Feature activation | +6.33e+00 |
Pos logit contributions | |
not | +0.149 |
making | +0.140 |
saving | +0.138 |
adding | +0.134 |
an | +0.132 |
Neg logit contributions | |
azz | -0.249 |
pillar | -0.231 |
y | -0.230 |
hell | -0.224 |
ppy | -0.219 |
Token | looked |
Token ID | 3114 |
Feature activation | +6.56e+00 |
Pos logit contributions | |
not | +0.208 |
an | +0.198 |
making | +0.197 |
many | +0.196 |
saving | +0.194 |
Neg logit contributions | |
azz | -0.225 |
pillar | -0.201 |
ppy | -0.194 |
pp | -0.192 |
y | -0.192 |
Token | until |
Token ID | 1566 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
not | +0.173 |
making | +0.162 |
an | +0.161 |
saving | +0.159 |
the | +0.155 |
Neg logit contributions | |
azz | -0.268 |
pillar | -0.248 |
hell | -0.248 |
y | -0.245 |
pp | -0.240 |
Token | he |
Token ID | 339 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
making | +0.123 |
warnings | +0.122 |
warning | +0.121 |
saving | +0.121 |
going | +0.120 |
Neg logit contributions | |
azz | -0.164 |
enny | -0.156 |
pillar | -0.154 |
och | -0.153 |
y | -0.150 |
Token | finally |
Token ID | 3443 |
Feature activation | +5.60e+00 |
Pos logit contributions | |
azz | +0.175 |
pillar | +0.161 |
och | +0.154 |
enny | +0.154 |
pp | +0.151 |
Neg logit contributions | |
not | -0.134 |
the | -0.133 |
making | -0.130 |
an | -0.130 |
many | -0.123 |
Token | found |
Token ID | 1043 |
Feature activation | +7.71e+00 |
Pos logit contributions | |
not | +0.169 |
the | +0.167 |
making | +0.164 |
an | +0.163 |
many | +0.155 |
Neg logit contributions | |
azz | -0.227 |
pillar | -0.209 |
och | -0.201 |
enny | -0.199 |
hell | -0.198 |
Token | a |
Token ID | 257 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
crossing | +0.131 |
not | +0.131 |
making | +0.130 |
facing | +0.127 |
going | +0.124 |
Neg logit contributions | |
azz | -0.333 |
pillar | -0.329 |
hell | -0.323 |
affe | -0.312 |
y | -0.312 |
Token | and |
Token ID | 290 |
Feature activation | +5.09e+00 |
Pos logit contributions | |
an | +0.051 |
not | +0.050 |
the | +0.050 |
making | +0.050 |
adding | +0.048 |
Neg logit contributions | |
azz | -0.075 |
enny | -0.067 |
pillar | -0.067 |
y | -0.066 |
och | -0.065 |
Token | bored |
Token ID | 17533 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
an | +0.154 |
making | +0.151 |
adding | +0.151 |
not | +0.149 |
the | +0.148 |
Neg logit contributions | |
azz | -0.186 |
pillar | -0.165 |
ppy | -0.164 |
y | -0.160 |
och | -0.159 |
Token | . |
Token ID | 13 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
azz | +0.015 |
pillar | +0.014 |
och | +0.014 |
pp | +0.013 |
enny | +0.013 |
Neg logit contributions | |
not | -0.008 |
the | -0.008 |
an | -0.008 |
making | -0.008 |
many | -0.007 |
Token | He |
Token ID | 679 |
Feature activation | -6.26e+00 |
Pos logit contributions | |
azz | +0.191 |
pillar | +0.175 |
och | +0.170 |
enny | +0.168 |
hell | +0.166 |
Neg logit contributions | |
not | -0.122 |
the | -0.120 |
making | -0.118 |
an | -0.117 |
going | -0.110 |
Token | looked |
Token ID | 3114 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
azz | +0.242 |
pillar | +0.220 |
ppy | +0.209 |
enny | +0.208 |
affe | +0.207 |
Neg logit contributions | |
the | -0.193 |
making | -0.193 |
not | -0.192 |
an | -0.189 |
adding | -0.180 |
Token | at |
Token ID | 379 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
not | +0.153 |
making | +0.143 |
adding | +0.143 |
saving | +0.137 |
climbing | +0.134 |
Neg logit contributions | |
azz | -0.249 |
y | -0.226 |
pillar | -0.225 |
ppy | -0.218 |
enny | -0.218 |
Token | Spot |
Token ID | 15899 |
Feature activation | -1.17e+01 |
Pos logit contributions | |
not | +0.255 |
making | +0.246 |
the | +0.240 |
an | +0.239 |
going | +0.227 |
Neg logit contributions | |
azz | -0.477 |
pillar | -0.438 |
och | -0.426 |
enny | -0.421 |
affe | -0.418 |
Token | and |
Token ID | 290 |
Feature activation | +5.61e+00 |
Pos logit contributions | |
azz | +0.485 |
pillar | +0.452 |
ppy | +0.432 |
och | +0.431 |
y | +0.431 |
Neg logit contributions | |
not | -0.315 |
an | -0.310 |
the | -0.306 |
making | -0.300 |
many | -0.288 |
Token | said |
Token ID | 531 |
Feature activation | +6.48e-01 |
Pos logit contributions | |
not | +0.164 |
the | +0.160 |
an | +0.160 |
making | +0.159 |
many | +0.151 |
Neg logit contributions | |
azz | -0.227 |
pillar | -0.207 |
och | -0.200 |
enny | -0.196 |
pp | -0.196 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
making | +0.016 |
not | +0.016 |
the | +0.015 |
an | +0.015 |
going | +0.014 |
Neg logit contributions | |
azz | -0.028 |
pillar | -0.026 |
hell | -0.026 |
och | -0.026 |
enny | -0.025 |
Token | " |
Token ID | 366 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
an | +0.050 |
the | +0.048 |
not | +0.048 |
many | +0.048 |
making | +0.047 |
Neg logit contributions | |
azz | -0.106 |
pillar | -0.097 |
och | -0.096 |
affe | -0.093 |
hell | -0.092 |
Token | around |
Token ID | 1088 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
not | +0.069 |
an | +0.065 |
making | +0.060 |
the | +0.058 |
many | +0.058 |
Neg logit contributions | |
azz | -0.154 |
pillar | -0.145 |
och | -0.142 |
hell | -0.142 |
y | -0.141 |
Token | . |
Token ID | 13 |
Feature activation | -4.70e+00 |
Pos logit contributions | |
not | +0.141 |
preventing | +0.133 |
saving | +0.131 |
climbing | +0.128 |
conquering | +0.128 |
Neg logit contributions | |
y | -0.108 |
azz | -0.102 |
hell | -0.097 |
pillar | -0.094 |
hel | -0.089 |
Token | He |
Token ID | 679 |
Feature activation | -6.32e+00 |
Pos logit contributions | |
azz | +0.205 |
pillar | +0.189 |
och | +0.182 |
enny | +0.182 |
hell | +0.180 |
Neg logit contributions | |
not | -0.127 |
making | -0.122 |
the | -0.120 |
an | -0.119 |
going | -0.115 |
Token | peek |
Token ID | 27185 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
azz | +0.234 |
pillar | +0.219 |
ppy | +0.207 |
enny | +0.206 |
affe | +0.204 |
Neg logit contributions | |
not | -0.197 |
an | -0.194 |
making | -0.194 |
the | -0.192 |
going | -0.186 |
Token | ed |
Token ID | 276 |
Feature activation | +6.45e+00 |
Pos logit contributions | |
the | +0.017 |
not | +0.017 |
an | +0.016 |
making | +0.016 |
many | +0.016 |
Neg logit contributions | |
azz | -0.012 |
pillar | -0.011 |
och | -0.010 |
pp | -0.010 |
affe | -0.010 |
Token | at |
Token ID | 379 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
not | +0.156 |
making | +0.141 |
going | +0.132 |
adding | +0.131 |
helping | +0.129 |
Neg logit contributions | |
azz | -0.273 |
pillar | -0.260 |
y | -0.256 |
hell | -0.253 |
och | -0.244 |
Token | the |
Token ID | 262 |
Feature activation | -2.21e-01 |
Pos logit contributions | |
not | +0.089 |
making | +0.084 |
going | +0.078 |
facing | +0.074 |
helping | +0.073 |
Neg logit contributions | |
azz | -0.567 |
pillar | -0.558 |
hell | -0.546 |
y | -0.540 |
affe | -0.540 |
Token | shelves |
Token ID | 20726 |
Feature activation | -6.57e+00 |
Pos logit contributions | |
pillar | +0.009 |
azz | +0.009 |
affe | +0.008 |
hell | +0.008 |
och | +0.008 |
Neg logit contributions | |
not | -0.006 |
an | -0.006 |
the | -0.006 |
making | -0.006 |
exploring | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | +8.33e+00 |
Pos logit contributions | |
azz | +0.217 |
y | +0.213 |
och | +0.202 |
hel | +0.201 |
enny | +0.200 |
Neg logit contributions | |
an | -0.188 |
your | -0.184 |
not | -0.181 |
our | -0.181 |
the | -0.179 |
Token | spices |
Token ID | 27952 |
Feature activation | -7.24e+00 |
Pos logit contributions | |
not | +0.164 |
an | +0.141 |
going | +0.141 |
making | +0.138 |
facing | +0.138 |
Neg logit contributions | |
azz | -0.355 |
pillar | -0.350 |
hell | -0.349 |
och | -0.342 |
affe | -0.334 |
Token | and |
Token ID | 290 |
Feature activation | +5.92e+00 |
Pos logit contributions | |
azz | +0.294 |
pillar | +0.275 |
och | +0.268 |
hell | +0.266 |
pp | +0.265 |
Neg logit contributions | |
not | -0.202 |
an | -0.196 |
the | -0.193 |
making | -0.188 |
many | -0.180 |
Token | wrong |
Token ID | 2642 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
exploring | +0.156 |
warnings | +0.142 |
help | +0.130 |
testing | +0.130 |
climbing | +0.129 |
Neg logit contributions | |
azz | -0.485 |
hell | -0.463 |
y | -0.458 |
ppy | -0.456 |
och | -0.452 |
Token | . |
Token ID | 13 |
Feature activation | -4.57e+00 |
Pos logit contributions | |
not | +0.045 |
the | +0.044 |
an | +0.043 |
making | +0.043 |
many | +0.041 |
Neg logit contributions | |
azz | -0.071 |
pillar | -0.065 |
och | -0.063 |
hell | -0.063 |
enny | -0.063 |
Token | The |
Token ID | 383 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
azz | +0.226 |
pillar | +0.212 |
och | +0.207 |
pp | +0.205 |
enny | +0.204 |
Neg logit contributions | |
the | -0.099 |
not | -0.099 |
an | -0.095 |
making | -0.092 |
many | -0.087 |
Token | caul |
Token ID | 46087 |
Feature activation | -7.45e+00 |
Pos logit contributions | |
pillar | +0.114 |
azz | +0.107 |
affe | +0.106 |
hell | +0.104 |
och | +0.104 |
Neg logit contributions | |
not | -0.088 |
an | -0.085 |
the | -0.082 |
making | -0.081 |
saving | -0.081 |
Token | iflower |
Token ID | 42642 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
azz | +0.291 |
pillar | +0.265 |
och | +0.257 |
enny | +0.252 |
pp | +0.251 |
Neg logit contributions | |
not | -0.239 |
the | -0.237 |
an | -0.230 |
making | -0.230 |
many | -0.218 |
Token | did |
Token ID | 750 |
Feature activation | +9.34e+00 |
Pos logit contributions | |
azz | +0.471 |
pillar | +0.453 |
hell | +0.426 |
och | +0.420 |
affe | +0.419 |
Neg logit contributions | |
an | -0.371 |
the | -0.365 |
not | -0.364 |
making | -0.353 |
many | -0.351 |
Token | not |
Token ID | 407 |
Feature activation | +8.48e+00 |
Pos logit contributions | |
exploring | +0.116 |
testing | +0.088 |
climbing | +0.082 |
helping | +0.082 |
shore | +0.080 |
Neg logit contributions | |
hell | -0.487 |
azz | -0.469 |
y | -0.428 |
hel | -0.423 |
ppy | -0.421 |
Token | taste |
Token ID | 6938 |
Feature activation | +5.20e+00 |
Pos logit contributions | |
not | +0.192 |
the | +0.190 |
an | +0.183 |
making | +0.182 |
many | +0.169 |
Neg logit contributions | |
azz | -0.412 |
pillar | -0.384 |
och | -0.374 |
enny | -0.368 |
pp | -0.367 |
Token | good |
Token ID | 922 |
Feature activation | +4.75e+00 |
Pos logit contributions | |
making | +0.084 |
saving | +0.082 |
not | +0.080 |
exploring | +0.078 |
helping | +0.077 |
Neg logit contributions | |
azz | -0.256 |
hell | -0.256 |
pillar | -0.245 |
y | -0.244 |
ppy | -0.240 |
Token | . |
Token ID | 13 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
not | +0.128 |
an | +0.123 |
the | +0.123 |
making | +0.119 |
many | +0.115 |
Neg logit contributions | |
azz | -0.202 |
pillar | -0.189 |
hell | -0.183 |
och | -0.182 |
enny | -0.180 |
Token | It |
Token ID | 632 |
Feature activation | -4.75e+00 |
Pos logit contributions | |
azz | +0.155 |
pillar | +0.142 |
och | +0.137 |
enny | +0.134 |
hell | +0.133 |
Neg logit contributions | |
not | -0.125 |
the | -0.124 |
making | -0.121 |
an | -0.121 |
many | -0.114 |
Token | day |
Token ID | 1110 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
azz | +0.090 |
pillar | +0.084 |
och | +0.082 |
pp | +0.081 |
enny | +0.081 |
Neg logit contributions | |
not | -0.032 |
the | -0.032 |
an | -0.030 |
making | -0.030 |
many | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
azz | +0.162 |
pillar | +0.150 |
och | +0.146 |
enny | +0.145 |
hell | +0.143 |
Neg logit contributions | |
not | -0.097 |
the | -0.094 |
making | -0.093 |
an | -0.093 |
going | -0.087 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
not | +0.030 |
making | +0.030 |
an | +0.029 |
the | +0.029 |
saving | +0.028 |
Neg logit contributions | |
azz | -0.065 |
pillar | -0.058 |
och | -0.058 |
hell | -0.057 |
enny | -0.057 |
Token | was |
Token ID | 373 |
Feature activation | +8.99e+00 |
Pos logit contributions | |
azz | +0.601 |
och | +0.554 |
pillar | +0.552 |
pp | +0.537 |
hell | +0.535 |
Neg logit contributions | |
the | -0.381 |
not | -0.378 |
an | -0.362 |
making | -0.356 |
many | -0.337 |
Token | playing |
Token ID | 2712 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
instructions | +0.191 |
accomplishment | +0.186 |
directions | +0.186 |
advice | +0.185 |
potential | +0.181 |
Neg logit contributions | |
itty | -0.339 |
azz | -0.326 |
y | -0.320 |
coat | -0.317 |
ppy | -0.315 |
Token | on |
Token ID | 319 |
Feature activation | +9.05e+00 |
Pos logit contributions | |
not | +0.075 |
going | +0.069 |
adding | +0.069 |
the | +0.068 |
an | +0.067 |
Neg logit contributions | |
azz | -0.148 |
pillar | -0.146 |
itty | -0.138 |
hell | -0.137 |
y | -0.136 |
Token | a |
Token ID | 257 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
not | +0.093 |
warning | +0.086 |
preventing | +0.079 |
warnings | +0.072 |
facing | +0.071 |
Neg logit contributions | |
azz | -0.465 |
pillar | -0.451 |
y | -0.439 |
itty | -0.430 |
hel | -0.429 |
Token | big |
Token ID | 1263 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
not | +0.035 |
the | +0.034 |
an | +0.034 |
making | +0.034 |
many | +0.031 |
Neg logit contributions | |
azz | -0.067 |
pillar | -0.063 |
hell | -0.060 |
och | -0.060 |
affe | -0.060 |
Token | green |
Token ID | 4077 |
Feature activation | -3.35e+00 |
Pos logit contributions | |
azz | +0.044 |
pillar | +0.041 |
och | +0.039 |
affe | +0.039 |
hell | +0.039 |
Neg logit contributions | |
not | -0.029 |
the | -0.028 |
an | -0.028 |
making | -0.027 |
many | -0.026 |
Token | hill |
Token ID | 12788 |
Feature activation | -8.02e+00 |
Pos logit contributions | |
azz | +0.156 |
och | +0.141 |
pillar | +0.139 |
enny | +0.139 |
pp | +0.138 |
Neg logit contributions | |
the | -0.088 |
not | -0.086 |
making | -0.083 |
an | -0.083 |
many | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
azz | +0.361 |
y | +0.344 |
pillar | +0.339 |
och | +0.329 |
itty | +0.327 |
Neg logit contributions | |
not | -0.172 |
an | -0.167 |
the | -0.164 |
making | -0.159 |
saving | -0.154 |
Token | One |
Token ID | 1881 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
azz | +0.214 |
pillar | +0.213 |
och | +0.207 |
pp | +0.205 |
affe | +0.204 |
Neg logit contributions | |
the | -0.090 |
an | -0.077 |
not | -0.077 |
making | -0.073 |
many | -0.073 |
Token | day |
Token ID | 1110 |
Feature activation | -3.76e+00 |
Pos logit contributions | |
azz | +0.095 |
pillar | +0.088 |
och | +0.087 |
pp | +0.086 |
enny | +0.085 |
Neg logit contributions | |
the | -0.030 |
not | -0.029 |
an | -0.027 |
making | -0.027 |
many | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
azz | +0.179 |
pillar | +0.166 |
och | +0.162 |
pp | +0.159 |
affe | +0.158 |
Neg logit contributions | |
the | -0.090 |
not | -0.090 |
an | -0.086 |
making | -0.086 |
many | -0.080 |
Token | she |
Token ID | 673 |
Feature activation | -6.64e+00 |
Pos logit contributions | |
not | +0.037 |
the | +0.037 |
an | +0.035 |
making | +0.035 |
many | +0.033 |
Neg logit contributions | |
azz | -0.071 |
pillar | -0.066 |
och | -0.064 |
enny | -0.063 |
pp | -0.063 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.70e+00 |
Pos logit contributions | |
azz | +0.294 |
pillar | +0.272 |
och | +0.271 |
affe | +0.263 |
hell | +0.263 |
Neg logit contributions | |
the | -0.184 |
not | -0.184 |
an | -0.177 |
making | -0.174 |
going | -0.164 |
Token | a |
Token ID | 257 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
not | +0.065 |
making | +0.057 |
the | +0.050 |
going | +0.043 |
an | +0.042 |
Neg logit contributions | |
azz | -0.470 |
pillar | -0.447 |
enny | -0.430 |
och | -0.429 |
hell | -0.427 |
Token | mosquito |
Token ID | 35446 |
Feature activation | -1.22e+01 |
Pos logit contributions | |
the | +0.019 |
not | +0.019 |
making | +0.018 |
an | +0.017 |
new | +0.016 |
Neg logit contributions | |
azz | -0.064 |
enny | -0.059 |
ppy | -0.059 |
och | -0.058 |
pp | -0.058 |
Token | flying |
Token ID | 7348 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
azz | +0.611 |
pillar | +0.591 |
och | +0.557 |
hell | +0.554 |
itty | +0.553 |
Neg logit contributions | |
not | -0.231 |
an | -0.223 |
the | -0.222 |
many | -0.206 |
making | -0.204 |
Token | around |
Token ID | 1088 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
not | +0.038 |
making | +0.036 |
the | +0.036 |
an | +0.035 |
adding | +0.033 |
Neg logit contributions | |
azz | -0.101 |
pillar | -0.095 |
y | -0.092 |
hell | -0.091 |
och | -0.091 |
Token | her |
Token ID | 607 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
not | +0.072 |
an | +0.062 |
making | +0.062 |
many | +0.060 |
adding | +0.059 |
Neg logit contributions | |
azz | -0.175 |
pillar | -0.170 |
y | -0.167 |
hell | -0.160 |
itty | -0.160 |
Token | . |
Token ID | 13 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
azz | +0.103 |
pillar | +0.096 |
och | +0.093 |
enny | +0.092 |
pp | +0.091 |
Neg logit contributions | |
not | -0.057 |
the | -0.056 |
an | -0.054 |
making | -0.054 |
many | -0.050 |
Token | together |
Token ID | 1978 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
not | +0.022 |
the | +0.022 |
an | +0.021 |
making | +0.020 |
many | +0.019 |
Neg logit contributions | |
azz | -0.051 |
pillar | -0.048 |
och | -0.047 |
pp | -0.046 |
enny | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
not | +0.097 |
an | +0.093 |
the | +0.092 |
making | +0.091 |
many | +0.086 |
Neg logit contributions | |
azz | -0.157 |
pillar | -0.145 |
och | -0.140 |
enny | -0.138 |
pp | -0.137 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.32e+00 |
Pos logit contributions | |
azz | +0.210 |
pillar | +0.199 |
pp | +0.197 |
och | +0.197 |
ppy | +0.193 |
Neg logit contributions | |
the | -0.087 |
not | -0.081 |
an | -0.079 |
many | -0.074 |
making | -0.073 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
azz | +0.439 |
pp | +0.412 |
pillar | +0.410 |
och | +0.410 |
affe | +0.408 |
Neg logit contributions | |
the | -0.179 |
an | -0.158 |
not | -0.154 |
many | -0.145 |
making | -0.144 |
Token | upon |
Token ID | 2402 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
the | +0.004 |
not | +0.004 |
an | +0.004 |
making | +0.003 |
many | +0.003 |
Neg logit contributions | |
azz | -0.004 |
pillar | -0.004 |
och | -0.004 |
pp | -0.004 |
affe | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
not | +0.017 |
the | +0.016 |
an | +0.014 |
making | +0.013 |
many | +0.009 |
Neg logit contributions | |
azz | -0.204 |
pillar | -0.194 |
och | -0.191 |
pp | -0.188 |
enny | -0.188 |
Token | time |
Token ID | 640 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
not | +0.001 |
the | +0.001 |
an | +0.001 |
many | +0.001 |
making | +0.001 |
Neg logit contributions | |
azz | -0.003 |
och | -0.003 |
pp | -0.003 |
pillar | -0.003 |
enny | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
azz | +0.061 |
pillar | +0.058 |
och | +0.057 |
affe | +0.056 |
pp | +0.056 |
Neg logit contributions | |
not | -0.027 |
the | -0.026 |
an | -0.025 |
making | -0.024 |
many | -0.023 |
Token | there |
Token ID | 612 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
not | +0.038 |
the | +0.038 |
an | +0.036 |
making | +0.036 |
many | +0.033 |
Neg logit contributions | |
azz | -0.107 |
pillar | -0.100 |
och | -0.098 |
enny | -0.096 |
pp | -0.096 |
Token | was |
Token ID | 373 |
Feature activation | +9.18e+00 |
Pos logit contributions | |
azz | +0.066 |
pillar | +0.061 |
och | +0.060 |
pp | +0.059 |
enny | +0.058 |
Neg logit contributions | |
the | -0.038 |
not | -0.037 |
an | -0.036 |
making | -0.035 |
many | -0.033 |
Token | a |
Token ID | 257 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
not | +0.052 |
the | +0.050 |
an | +0.042 |
making | +0.040 |
many | +0.027 |
Neg logit contributions | |
azz | -0.601 |
pillar | -0.571 |
och | -0.560 |
enny | -0.554 |
pp | -0.553 |
Token | , |
Token ID | 11 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
azz | +0.507 |
pillar | +0.469 |
och | +0.455 |
enny | +0.448 |
pp | +0.447 |
Neg logit contributions | |
not | -0.297 |
the | -0.296 |
making | -0.284 |
an | -0.284 |
many | -0.265 |
Token | but |
Token ID | 475 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
not | +0.059 |
an | +0.059 |
making | +0.059 |
the | +0.058 |
many | +0.057 |
Neg logit contributions | |
azz | -0.102 |
pillar | -0.094 |
och | -0.092 |
enny | -0.090 |
hell | -0.090 |
Token | it |
Token ID | 340 |
Feature activation | -5.06e+00 |
Pos logit contributions | |
making | +0.070 |
not | +0.069 |
an | +0.068 |
the | +0.068 |
many | +0.064 |
Neg logit contributions | |
azz | -0.183 |
pillar | -0.170 |
och | -0.170 |
enny | -0.166 |
hell | -0.164 |
Token | was |
Token ID | 373 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
azz | +0.223 |
pillar | +0.206 |
och | +0.201 |
enny | +0.197 |
pp | +0.197 |
Neg logit contributions | |
not | -0.139 |
the | -0.137 |
an | -0.132 |
making | -0.132 |
many | -0.123 |
Token | too |
Token ID | 1165 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
making | +0.246 |
many | +0.246 |
adding | +0.242 |
the | +0.242 |
an | +0.234 |
Neg logit contributions | |
azz | -0.452 |
y | -0.415 |
och | -0.409 |
ppy | -0.408 |
hell | -0.407 |
Token | late |
Token ID | 2739 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
not | +0.039 |
the | +0.039 |
making | +0.038 |
an | +0.037 |
many | +0.033 |
Neg logit contributions | |
azz | -0.121 |
pillar | -0.113 |
och | -0.111 |
enny | -0.109 |
hell | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
not | +0.051 |
the | +0.050 |
an | +0.049 |
making | +0.048 |
many | +0.046 |
Neg logit contributions | |
azz | -0.086 |
pillar | -0.079 |
och | -0.078 |
enny | -0.076 |
hell | -0.076 |
Token | His |
Token ID | 2399 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
azz | +0.183 |
pillar | +0.165 |
enny | +0.164 |
och | +0.164 |
ird | +0.161 |
Neg logit contributions | |
making | -0.146 |
not | -0.140 |
the | -0.137 |
going | -0.137 |
diving | -0.135 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.26e+01 |
Pos logit contributions | |
azz | +0.092 |
pillar | +0.085 |
enny | +0.081 |
och | +0.081 |
affe | +0.080 |
Neg logit contributions | |
the | -0.044 |
making | -0.044 |
not | -0.043 |
an | -0.043 |
saving | -0.040 |
Token | wouldn |
Token ID | 3636 |
Feature activation | +5.24e+00 |
Pos logit contributions | |
azz | +0.566 |
pillar | +0.524 |
och | +0.510 |
hell | +0.506 |
pp | +0.505 |
Neg logit contributions | |
not | -0.347 |
the | -0.343 |
making | -0.328 |
an | -0.318 |
going | -0.300 |
Token | 't |
Token ID | 470 |
Feature activation | +6.32e+00 |
Pos logit contributions | |
not | +0.094 |
the | +0.092 |
an | +0.086 |
making | +0.086 |
going | +0.078 |
Neg logit contributions | |
azz | -0.280 |
pillar | -0.263 |
och | -0.257 |
pp | -0.255 |
affe | -0.253 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.81e+00 |
Pos logit contributions | |
azz | +0.188 |
pillar | +0.175 |
och | +0.172 |
pp | +0.169 |
hell | +0.168 |
Neg logit contributions | |
the | -0.098 |
not | -0.098 |
an | -0.094 |
making | -0.092 |
many | -0.089 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
azz | +0.460 |
pillar | +0.430 |
pp | +0.430 |
och | +0.430 |
affe | +0.424 |
Neg logit contributions | |
the | -0.193 |
an | -0.175 |
not | -0.170 |
many | -0.161 |
making | -0.160 |
Token | upon |
Token ID | 2402 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
the | +0.003 |
not | +0.003 |
an | +0.003 |
making | +0.003 |
many | +0.003 |
Neg logit contributions | |
azz | -0.004 |
pillar | -0.004 |
och | -0.004 |
pp | -0.004 |
affe | -0.004 |
Token | a |
Token ID | 257 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
not | +0.016 |
the | +0.016 |
an | +0.013 |
making | +0.012 |
many | +0.008 |
Neg logit contributions | |
azz | -0.204 |
pillar | -0.194 |
och | -0.191 |
pp | -0.188 |
enny | -0.188 |
Token | time |
Token ID | 640 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
not | +0.009 |
the | +0.008 |
an | +0.008 |
many | +0.007 |
making | +0.007 |
Neg logit contributions | |
azz | -0.025 |
och | -0.024 |
pp | -0.024 |
pillar | -0.024 |
enny | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
azz | +0.060 |
pillar | +0.057 |
och | +0.056 |
affe | +0.055 |
pp | +0.055 |
Neg logit contributions | |
the | -0.025 |
not | -0.025 |
an | -0.024 |
making | -0.021 |
many | -0.021 |
Token | there |
Token ID | 612 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
not | +0.043 |
the | +0.042 |
an | +0.040 |
making | +0.040 |
many | +0.037 |
Neg logit contributions | |
azz | -0.111 |
pillar | -0.104 |
och | -0.102 |
enny | -0.100 |
pp | -0.100 |
Token | was |
Token ID | 373 |
Feature activation | +9.01e+00 |
Pos logit contributions | |
azz | +0.058 |
pillar | +0.054 |
och | +0.053 |
pp | +0.052 |
affe | +0.051 |
Neg logit contributions | |
the | -0.033 |
not | -0.032 |
an | -0.031 |
making | -0.031 |
many | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
not | +0.055 |
the | +0.051 |
making | +0.044 |
an | +0.042 |
many | +0.028 |
Neg logit contributions | |
azz | -0.582 |
pillar | -0.553 |
och | -0.541 |
enny | -0.536 |
hell | -0.535 |
Token | little |
Token ID | 1310 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
the | +0.027 |
not | +0.026 |
making | +0.024 |
an | +0.024 |
many | +0.023 |
Neg logit contributions | |
azz | -0.070 |
pillar | -0.065 |
och | -0.064 |
pp | -0.064 |
enny | -0.064 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
azz | +0.152 |
hell | +0.134 |
enny | +0.133 |
ives | +0.130 |
pp | +0.130 |
Neg logit contributions | |
the | -0.112 |
new | -0.097 |
many | -0.096 |
making | -0.095 |
an | -0.093 |
Token | of |
Token ID | 286 |
Feature activation | +6.65e+00 |
Pos logit contributions | |
azz | +0.166 |
pillar | +0.156 |
och | +0.152 |
enny | +0.149 |
affe | +0.148 |
Neg logit contributions | |
not | -0.085 |
the | -0.083 |
an | -0.081 |
making | -0.080 |
many | -0.076 |
Token | their |
Token ID | 511 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
not | +0.109 |
an | +0.102 |
making | +0.098 |
the | +0.094 |
many | +0.092 |
Neg logit contributions | |
azz | -0.328 |
pillar | -0.317 |
och | -0.306 |
affe | -0.305 |
itty | -0.303 |
Token | lives |
Token ID | 3160 |
Feature activation | +8.47e-01 |
Pos logit contributions | |
azz | +0.135 |
pillar | +0.127 |
affe | +0.122 |
och | +0.122 |
hell | +0.121 |
Neg logit contributions | |
not | -0.062 |
an | -0.059 |
making | -0.059 |
the | -0.059 |
going | -0.053 |
Token | ! |
Token ID | 0 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
not | +0.020 |
the | +0.020 |
an | +0.019 |
making | +0.019 |
many | +0.018 |
Neg logit contributions | |
azz | -0.039 |
pillar | -0.037 |
och | -0.036 |
enny | -0.035 |
pp | -0.035 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.82e+00 |
Pos logit contributions | |
azz | +0.150 |
pp | +0.149 |
hell | +0.147 |
pillar | +0.145 |
ppy | +0.142 |
Neg logit contributions | |
the | -0.065 |
an | -0.058 |
not | -0.057 |
many | -0.051 |
his | -0.050 |
Token | Once |
Token ID | 7454 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
azz | +0.465 |
pp | +0.441 |
affe | +0.437 |
och | +0.434 |
ppy | +0.432 |
Neg logit contributions | |
the | -0.195 |
an | -0.167 |
many | -0.151 |
not | -0.146 |
making | -0.144 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
azz | +0.021 |
pillar | +0.020 |
och | +0.019 |
pp | +0.019 |
affe | +0.019 |
Neg logit contributions | |
the | -0.016 |
not | -0.016 |
an | -0.015 |
many | -0.015 |
making | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
the | +0.013 |
not | +0.013 |
an | +0.011 |
making | +0.008 |
many | +0.006 |
Neg logit contributions | |
azz | -0.188 |
pillar | -0.179 |
och | -0.177 |
pp | -0.175 |
enny | -0.174 |
Token | time |
Token ID | 640 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
azz | +0.012 |
och | +0.011 |
pp | +0.011 |
pillar | +0.011 |
enny | +0.010 |
Neg logit contributions | |
not | -0.003 |
the | -0.003 |
an | -0.003 |
many | -0.003 |
making | -0.003 |
Token | there |
Token ID | 612 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
azz | +0.057 |
pillar | +0.054 |
affe | +0.052 |
och | +0.052 |
pp | +0.051 |
Neg logit contributions | |
not | -0.022 |
the | -0.022 |
an | -0.021 |
making | -0.020 |
many | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +8.87e+00 |
Pos logit contributions | |
azz | +0.089 |
och | +0.083 |
pillar | +0.082 |
pp | +0.082 |
affe | +0.081 |
Neg logit contributions | |
the | -0.041 |
not | -0.038 |
an | -0.036 |
making | -0.034 |
many | -0.033 |
Token | is |
Token ID | 318 |
Feature activation | +9.26e+00 |
Pos logit contributions | |
azz | +0.294 |
och | +0.270 |
enny | +0.267 |
pillar | +0.267 |
pp | +0.256 |
Neg logit contributions | |
not | -0.187 |
the | -0.182 |
an | -0.172 |
making | -0.172 |
going | -0.161 |
Token | mine |
Token ID | 6164 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
making | +0.172 |
many | +0.170 |
adding | +0.168 |
an | +0.166 |
saving | +0.166 |
Neg logit contributions | |
azz | -0.424 |
hell | -0.388 |
ppy | -0.381 |
pp | -0.376 |
y | -0.374 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
azz | +0.078 |
pillar | +0.072 |
och | +0.070 |
enny | +0.069 |
hell | +0.069 |
Neg logit contributions | |
not | -0.052 |
the | -0.052 |
making | -0.051 |
an | -0.051 |
many | -0.048 |
Token | Sam |
Token ID | 3409 |
Feature activation | -1.50e+01 |
Pos logit contributions | |
azz | +0.217 |
och | +0.206 |
pillar | +0.206 |
pp | +0.206 |
hell | +0.206 |
Neg logit contributions | |
the | -0.104 |
an | -0.097 |
not | -0.096 |
making | -0.090 |
many | -0.083 |
Token | said |
Token ID | 531 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
azz | +0.669 |
pillar | +0.627 |
och | +0.624 |
pp | +0.607 |
hell | +0.590 |
Neg logit contributions | |
not | -0.446 |
the | -0.443 |
an | -0.419 |
making | -0.403 |
his | -0.378 |
Token | . |
Token ID | 13 |
Feature activation | -5.33e+00 |
Pos logit contributions | |
azz | +0.074 |
enny | +0.066 |
hell | +0.066 |
pillar | +0.066 |
och | +0.066 |
Neg logit contributions | |
not | -0.050 |
the | -0.050 |
making | -0.050 |
an | -0.049 |
many | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
azz | +0.246 |
pillar | +0.229 |
och | +0.223 |
pp | +0.221 |
hell | +0.219 |
Neg logit contributions | |
not | -0.134 |
the | -0.133 |
an | -0.128 |
making | -0.127 |
many | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
azz | +0.207 |
pillar | +0.196 |
och | +0.190 |
pp | +0.190 |
hell | +0.187 |
Neg logit contributions | |
the | -0.097 |
not | -0.097 |
an | -0.091 |
making | -0.088 |
many | -0.085 |
Token | They |
Token ID | 2990 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
azz | +0.170 |
pillar | +0.158 |
pp | +0.157 |
och | +0.156 |
hell | +0.154 |
Neg logit contributions | |
the | -0.157 |
not | -0.156 |
an | -0.148 |
next | -0.144 |
many | -0.142 |
Token | both |
Token ID | 1111 |
Feature activation | -7.10e-02 |
Pos logit contributions | |
azz | +0.232 |
pillar | +0.219 |
och | +0.213 |
pp | +0.207 |
hell | +0.206 |
Neg logit contributions | |
the | -0.135 |
not | -0.134 |
an | -0.129 |
making | -0.124 |
many | -0.120 |
Token | pulled |
Token ID | 5954 |
Feature activation | +7.42e+00 |
Pos logit contributions | |
azz | +0.003 |
pillar | +0.003 |
och | +0.003 |
pp | +0.003 |
enny | +0.003 |
Neg logit contributions | |
the | -0.002 |
not | -0.002 |
an | -0.002 |
making | -0.002 |
many | -0.002 |
Token | 's |
Token ID | 338 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
azz | +0.429 |
och | +0.394 |
pillar | +0.392 |
pp | +0.382 |
hell | +0.373 |
Neg logit contributions | |
not | -0.278 |
the | -0.273 |
an | -0.260 |
making | -0.255 |
many | -0.240 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.25e+01 |
Pos logit contributions | |
not | +0.034 |
the | +0.034 |
making | +0.034 |
an | +0.033 |
going | +0.032 |
Neg logit contributions | |
azz | -0.052 |
pillar | -0.048 |
enny | -0.047 |
hell | -0.046 |
och | -0.046 |
Token | noticed |
Token ID | 6810 |
Feature activation | +6.49e+00 |
Pos logit contributions | |
azz | +0.551 |
pillar | +0.511 |
och | +0.505 |
pp | +0.495 |
hell | +0.490 |
Neg logit contributions | |
the | -0.371 |
not | -0.366 |
making | -0.354 |
an | -0.334 |
to | -0.317 |
Token | the |
Token ID | 262 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
the | +0.115 |
not | +0.113 |
an | +0.108 |
making | +0.106 |
many | +0.095 |
Neg logit contributions | |
azz | -0.351 |
pillar | -0.330 |
och | -0.322 |
pp | -0.319 |
enny | -0.318 |
Token | comm |
Token ID | 725 |
Feature activation | -4.02e+00 |
Pos logit contributions | |
azz | +0.044 |
pillar | +0.040 |
och | +0.039 |
pp | +0.039 |
enny | +0.039 |
Neg logit contributions | |
not | -0.026 |
the | -0.026 |
an | -0.025 |
making | -0.025 |
many | -0.023 |
Token | otion |
Token ID | 9650 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
azz | +0.119 |
pillar | +0.106 |
och | +0.101 |
enny | +0.099 |
pp | +0.098 |
Neg logit contributions | |
not | -0.166 |
the | -0.165 |
an | -0.162 |
making | -0.161 |
many | -0.155 |
Token | and |
Token ID | 290 |
Feature activation | +5.35e+00 |
Pos logit contributions | |
azz | +0.142 |
pillar | +0.132 |
och | +0.126 |
hell | +0.126 |
enny | +0.124 |
Neg logit contributions | |
not | -0.104 |
an | -0.101 |
the | -0.100 |
making | -0.098 |
many | -0.095 |
Token | came |
Token ID | 1625 |
Feature activation | +7.21e+00 |
Pos logit contributions | |
not | +0.156 |
the | +0.153 |
an | +0.150 |
making | +0.149 |
many | +0.142 |
Neg logit contributions | |
azz | -0.222 |
pillar | -0.204 |
och | -0.198 |
enny | -0.195 |
pp | -0.194 |
Token | outside |
Token ID | 2354 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
not | +0.132 |
making | +0.121 |
many | +0.120 |
an | +0.119 |
the | +0.115 |
Neg logit contributions | |
azz | -0.345 |
y | -0.329 |
enny | -0.318 |
pillar | -0.309 |
ppy | -0.309 |
Token | . |
Token ID | 13 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
not | +0.047 |
an | +0.046 |
many | +0.045 |
climbing | +0.042 |
your | +0.042 |
Neg logit contributions | |
azz | -0.077 |
y | -0.075 |
enny | -0.069 |
pillar | -0.069 |
och | -0.068 |
Token | She |
Token ID | 1375 |
Feature activation | -5.66e+00 |
Pos logit contributions | |
azz | +0.178 |
pillar | +0.162 |
och | +0.157 |
enny | +0.155 |
pp | +0.151 |
Neg logit contributions | |
not | -0.144 |
the | -0.141 |
making | -0.138 |
an | -0.138 |
going | -0.131 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.57e+00 |
Pos logit contributions | |
azz | +0.158 |
pillar | +0.145 |
och | +0.141 |
enny | +0.139 |
affe | +0.138 |
Neg logit contributions | |
not | -0.098 |
the | -0.096 |
making | -0.094 |
an | -0.093 |
going | -0.088 |
Token | Tom |
Token ID | 13787 |
Feature activation | -1.30e+01 |
Pos logit contributions | |
azz | +0.423 |
affe | +0.392 |
pp | +0.391 |
och | +0.388 |
hell | +0.384 |
Neg logit contributions | |
the | -0.209 |
an | -0.188 |
making | -0.177 |
many | -0.176 |
not | -0.173 |
Token | and |
Token ID | 290 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
azz | +0.581 |
pillar | +0.540 |
och | +0.528 |
hell | +0.518 |
enny | +0.515 |
Neg logit contributions | |
the | -0.352 |
not | -0.344 |
an | -0.332 |
making | -0.328 |
many | -0.310 |
Token | Sam |
Token ID | 3409 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
not | +0.055 |
making | +0.053 |
the | +0.051 |
an | +0.048 |
going | +0.043 |
Neg logit contributions | |
azz | -0.263 |
pillar | -0.247 |
och | -0.241 |
affe | -0.239 |
enny | -0.238 |
Token | were |
Token ID | 547 |
Feature activation | +8.73e+00 |
Pos logit contributions | |
azz | +0.508 |
pillar | +0.461 |
och | +0.455 |
enny | +0.446 |
pp | +0.438 |
Neg logit contributions | |
the | -0.425 |
not | -0.409 |
an | -0.402 |
making | -0.394 |
going | -0.383 |
Token | friends |
Token ID | 2460 |
Feature activation | -6.97e+00 |
Pos logit contributions | |
not | +0.161 |
making | +0.153 |
the | +0.153 |
an | +0.144 |
preventing | +0.142 |
Neg logit contributions | |
azz | -0.427 |
pillar | -0.386 |
enny | -0.386 |
itty | -0.384 |
och | -0.382 |
Token | . |
Token ID | 13 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
azz | +0.331 |
pillar | +0.310 |
och | +0.299 |
enny | +0.295 |
pp | +0.295 |
Neg logit contributions | |
the | -0.172 |
not | -0.171 |
an | -0.162 |
making | -0.161 |
many | -0.150 |
Token | They |
Token ID | 1119 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
pillar | +0.162 |
pp | +0.160 |
och | +0.156 |
azz | +0.156 |
hell | +0.155 |
Neg logit contributions | |
the | -0.073 |
an | -0.067 |
not | -0.063 |
many | -0.061 |
his | -0.058 |
Token | liked |
Token ID | 8288 |
Feature activation | +9.55e+00 |
Pos logit contributions | |
azz | +0.158 |
pillar | +0.148 |
och | +0.144 |
enny | +0.140 |
hell | +0.140 |
Neg logit contributions | |
the | -0.114 |
not | -0.112 |
an | -0.107 |
making | -0.106 |
many | -0.102 |
Token | to |
Token ID | 284 |
Feature activation | +9.40e+00 |
Pos logit contributions | |
not | +0.150 |
preventing | +0.137 |
zed | +0.137 |
facing | +0.127 |
conquering | +0.123 |
Neg logit contributions | |
azz | -0.439 |
y | -0.411 |
pillar | -0.401 |
lets | -0.393 |
affe | -0.388 |
Token | play |
Token ID | 711 |
Feature activation | +4.19e+00 |
Pos logit contributions | |
the | +0.190 |
not | +0.187 |
an | +0.180 |
making | +0.174 |
going | +0.160 |
Neg logit contributions | |
azz | -0.485 |
pillar | -0.458 |
och | -0.451 |
enny | -0.441 |
pp | -0.441 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.46e+01 |
Pos logit contributions | |
the | +0.014 |
not | +0.014 |
making | +0.013 |
an | +0.013 |
many | +0.011 |
Neg logit contributions | |
azz | -0.046 |
pillar | -0.043 |
och | -0.042 |
pp | -0.042 |
enny | -0.042 |
Token | with |
Token ID | 351 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
azz | +0.759 |
pillar | +0.710 |
och | +0.694 |
enny | +0.684 |
hell | +0.680 |
Neg logit contributions | |
not | -0.273 |
the | -0.266 |
an | -0.257 |
making | -0.251 |
many | -0.235 |
Token | a |
Token ID | 257 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
not | +0.096 |
making | +0.084 |
the | +0.074 |
going | +0.066 |
adding | +0.063 |
Neg logit contributions | |
azz | -0.625 |
pillar | -0.588 |
hell | -0.571 |
och | -0.570 |
enny | -0.565 |
Token | gray |
Token ID | 12768 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
not | +0.025 |
the | +0.024 |
an | +0.023 |
making | +0.023 |
many | +0.021 |
Neg logit contributions | |
azz | -0.071 |
pillar | -0.066 |
och | -0.065 |
pp | -0.065 |
enny | -0.065 |
Token | bike |
Token ID | 7161 |
Feature activation | -9.86e+00 |
Pos logit contributions | |
azz | +0.138 |
pp | +0.124 |
ppy | +0.123 |
enny | +0.122 |
och | +0.122 |
Neg logit contributions | |
the | -0.112 |
an | -0.110 |
not | -0.108 |
making | -0.106 |
his | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
azz | +0.503 |
pillar | +0.470 |
enny | +0.459 |
och | +0.454 |
pp | +0.453 |
Neg logit contributions | |
not | -0.205 |
the | -0.205 |
making | -0.199 |
an | -0.191 |
going | -0.184 |
Token | The |
Token ID | 383 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
azz | +0.188 |
pillar | +0.173 |
affe | +0.167 |
enny | +0.167 |
y | +0.160 |
Neg logit contributions | |
not | -0.136 |
making | -0.132 |
going | -0.128 |
saving | -0.127 |
climbing | -0.125 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.68e+01 |
Pos logit contributions | |
azz | +0.106 |
pillar | +0.096 |
och | +0.093 |
pp | +0.091 |
enny | +0.090 |
Neg logit contributions | |
the | -0.120 |
not | -0.119 |
an | -0.116 |
making | -0.115 |
many | -0.111 |
Token | 's |
Token ID | 338 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
azz | +0.711 |
pillar | +0.657 |
och | +0.642 |
pp | +0.630 |
enny | +0.626 |
Neg logit contributions | |
not | -0.493 |
the | -0.491 |
an | -0.470 |
making | -0.468 |
many | -0.442 |
Token | bike |
Token ID | 7161 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
not | +0.034 |
the | +0.034 |
making | +0.033 |
an | +0.033 |
many | +0.031 |
Neg logit contributions | |
azz | -0.031 |
pillar | -0.028 |
och | -0.027 |
affe | -0.026 |
enny | -0.026 |
Token | was |
Token ID | 373 |
Feature activation | +9.85e+00 |
Pos logit contributions | |
azz | +0.478 |
pillar | +0.441 |
och | +0.428 |
enny | +0.425 |
pp | +0.423 |
Neg logit contributions | |
not | -0.326 |
the | -0.322 |
making | -0.312 |
an | -0.308 |
going | -0.294 |
Token | Sara |
Token ID | 24799 |
Feature activation | -1.18e+01 |
Pos logit contributions | |
azz | +0.151 |
pillar | +0.138 |
och | +0.134 |
enny | +0.133 |
hell | +0.132 |
Neg logit contributions | |
not | -0.093 |
the | -0.090 |
making | -0.089 |
an | -0.089 |
going | -0.083 |
Token | had |
Token ID | 550 |
Feature activation | +8.23e+00 |
Pos logit contributions | |
azz | +0.472 |
pillar | +0.432 |
och | +0.421 |
hell | +0.412 |
pp | +0.411 |
Neg logit contributions | |
not | -0.371 |
the | -0.369 |
making | -0.357 |
an | -0.355 |
many | -0.337 |
Token | a |
Token ID | 257 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
warnings | +0.093 |
preventing | +0.091 |
warning | +0.089 |
helping | +0.088 |
adding | +0.086 |
Neg logit contributions | |
azz | -0.415 |
ird | -0.391 |
y | -0.381 |
pillar | -0.380 |
itty | -0.377 |
Token | doll |
Token ID | 3654 |
Feature activation | -1.18e+01 |
Pos logit contributions | |
not | +0.052 |
the | +0.051 |
an | +0.049 |
making | +0.048 |
many | +0.045 |
Neg logit contributions | |
azz | -0.123 |
pillar | -0.114 |
och | -0.112 |
enny | -0.111 |
pp | -0.110 |
Token | and |
Token ID | 290 |
Feature activation | +6.40e+00 |
Pos logit contributions | |
azz | +0.639 |
pillar | +0.596 |
enny | +0.592 |
hell | +0.590 |
och | +0.584 |
Neg logit contributions | |
not | -0.237 |
the | -0.230 |
making | -0.222 |
an | -0.200 |
to | -0.200 |
Token | Tom |
Token ID | 4186 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
not | +0.161 |
the | +0.160 |
an | +0.155 |
making | +0.144 |
many | +0.141 |
Neg logit contributions | |
azz | -0.307 |
pillar | -0.288 |
och | -0.284 |
pp | -0.283 |
hell | -0.275 |
Token | had |
Token ID | 550 |
Feature activation | +9.32e+00 |
Pos logit contributions | |
azz | +0.469 |
pillar | +0.435 |
och | +0.417 |
enny | +0.413 |
affe | +0.410 |
Neg logit contributions | |
the | -0.353 |
not | -0.345 |
making | -0.334 |
an | -0.330 |
going | -0.317 |
Token | a |
Token ID | 257 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
not | +0.081 |
making | +0.075 |
the | +0.066 |
an | +0.060 |
going | +0.059 |
Neg logit contributions | |
azz | -0.551 |
pillar | -0.513 |
hell | -0.503 |
och | -0.503 |
ird | -0.497 |
Token | car |
Token ID | 1097 |
Feature activation | -8.26e+00 |
Pos logit contributions | |
not | +0.072 |
the | +0.070 |
an | +0.067 |
making | +0.067 |
many | +0.064 |
Neg logit contributions | |
azz | -0.151 |
pillar | -0.141 |
ppy | -0.139 |
enny | -0.138 |
och | -0.137 |
Token | . |
Token ID | 13 |
Feature activation | -3.68e+00 |
Pos logit contributions | |
azz | +0.410 |
pillar | +0.388 |
enny | +0.374 |
och | +0.374 |
pp | +0.371 |
Neg logit contributions | |
not | -0.185 |
the | -0.183 |
making | -0.176 |
an | -0.172 |
many | -0.159 |
Token | They |
Token ID | 1119 |
Feature activation | -3.73e+00 |
Pos logit contributions | |
azz | +0.169 |
pillar | +0.158 |
och | +0.154 |
pp | +0.152 |
hell | +0.151 |
Neg logit contributions | |
the | -0.095 |
not | -0.093 |
an | -0.090 |
making | -0.088 |
many | -0.084 |
Token | ability |
Token ID | 2694 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
azz | +0.045 |
pillar | +0.041 |
affe | +0.040 |
och | +0.039 |
hell | +0.039 |
Neg logit contributions | |
not | -0.026 |
an | -0.025 |
making | -0.025 |
the | -0.025 |
going | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +1.04e+01 |
Pos logit contributions | |
an | +0.053 |
not | +0.053 |
making | +0.052 |
the | +0.052 |
adding | +0.049 |
Neg logit contributions | |
azz | -0.270 |
pillar | -0.246 |
y | -0.243 |
och | -0.242 |
hell | -0.242 |
Token | create |
Token ID | 2251 |
Feature activation | +8.36e+00 |
Pos logit contributions | |
not | +0.214 |
an | +0.211 |
the | +0.210 |
making | +0.208 |
many | +0.197 |
Neg logit contributions | |
azz | -0.503 |
pillar | -0.462 |
och | -0.451 |
hell | -0.448 |
pp | -0.448 |
Token | . |
Token ID | 13 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
not | +0.125 |
making | +0.118 |
climbing | +0.110 |
going | +0.108 |
an | +0.107 |
Neg logit contributions | |
azz | -0.437 |
hell | -0.406 |
pillar | -0.400 |
affe | -0.385 |
och | -0.383 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.22e+00 |
Pos logit contributions | |
azz | +0.184 |
pillar | +0.178 |
pp | +0.175 |
hell | +0.172 |
och | +0.171 |
Neg logit contributions | |
the | -0.095 |
not | -0.094 |
an | -0.091 |
many | -0.085 |
making | -0.084 |
Token | Tim |
Token ID | 14967 |
Feature activation | -1.36e+01 |
Pos logit contributions | |
azz | +0.400 |
pp | +0.370 |
affe | +0.369 |
och | +0.368 |
pillar | +0.367 |
Neg logit contributions | |
the | -0.204 |
an | -0.186 |
not | -0.176 |
making | -0.175 |
many | -0.174 |
Token | and |
Token ID | 290 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
azz | +0.446 |
pillar | +0.400 |
och | +0.392 |
hell | +0.380 |
pp | +0.375 |
Neg logit contributions | |
the | -0.531 |
not | -0.520 |
an | -0.508 |
making | -0.506 |
many | -0.485 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.21e+01 |
Pos logit contributions | |
not | +0.046 |
the | +0.044 |
making | +0.043 |
an | +0.041 |
many | +0.036 |
Neg logit contributions | |
azz | -0.240 |
pillar | -0.226 |
och | -0.221 |
enny | -0.219 |
affe | -0.218 |
Token | are |
Token ID | 389 |
Feature activation | +8.11e+00 |
Pos logit contributions | |
azz | +0.566 |
och | +0.518 |
pillar | +0.517 |
enny | +0.507 |
hell | +0.502 |
Neg logit contributions | |
the | -0.355 |
an | -0.318 |
not | -0.311 |
going | -0.307 |
making | -0.298 |
Token | friends |
Token ID | 2460 |
Feature activation | -7.29e+00 |
Pos logit contributions | |
not | +0.152 |
the | +0.150 |
an | +0.143 |
preventing | +0.141 |
making | +0.139 |
Neg logit contributions | |
azz | -0.373 |
enny | -0.347 |
ird | -0.343 |
y | -0.339 |
itty | -0.338 |
Token | . |
Token ID | 13 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
azz | +0.335 |
pillar | +0.312 |
och | +0.302 |
enny | +0.297 |
pp | +0.297 |
Neg logit contributions | |
not | -0.189 |
the | -0.189 |
an | -0.180 |
making | -0.179 |
many | -0.167 |
Token | friends |
Token ID | 2460 |
Feature activation | -8.31e+00 |
Pos logit contributions | |
azz | +0.009 |
pillar | +0.008 |
och | +0.008 |
enny | +0.008 |
pp | +0.008 |
Neg logit contributions | |
not | -0.003 |
the | -0.003 |
an | -0.003 |
making | -0.003 |
many | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
azz | +0.390 |
pillar | +0.362 |
och | +0.351 |
enny | +0.346 |
hell | +0.345 |
Neg logit contributions | |
the | -0.205 |
not | -0.205 |
an | -0.197 |
making | -0.197 |
many | -0.183 |
Token | One |
Token ID | 1881 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
azz | +0.182 |
pillar | +0.176 |
och | +0.169 |
pp | +0.167 |
ppy | +0.164 |
Neg logit contributions | |
the | -0.104 |
not | -0.102 |
an | -0.098 |
making | -0.096 |
many | -0.091 |
Token | day |
Token ID | 1110 |
Feature activation | -3.55e+00 |
Pos logit contributions | |
azz | +0.087 |
pillar | +0.081 |
och | +0.080 |
pp | +0.079 |
enny | +0.078 |
Neg logit contributions | |
the | -0.028 |
not | -0.028 |
an | -0.027 |
making | -0.026 |
many | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
azz | +0.155 |
pillar | +0.143 |
och | +0.139 |
enny | +0.138 |
hell | +0.137 |
Neg logit contributions | |
not | -0.093 |
the | -0.090 |
making | -0.089 |
an | -0.089 |
going | -0.083 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
not | +0.043 |
making | +0.041 |
an | +0.041 |
saving | +0.040 |
the | +0.040 |
Neg logit contributions | |
azz | -0.066 |
pillar | -0.059 |
och | -0.059 |
y | -0.058 |
enny | -0.058 |
Token | my |
Token ID | 1820 |
Feature activation | -9.48e+00 |
Pos logit contributions | |
azz | +0.371 |
och | +0.352 |
pillar | +0.334 |
hell | +0.317 |
affe | +0.316 |
Neg logit contributions | |
the | -0.589 |
not | -0.579 |
an | -0.562 |
making | -0.558 |
going | -0.549 |
Token | 's |
Token ID | 338 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
azz | +0.406 |
och | +0.376 |
pillar | +0.372 |
pp | +0.362 |
hell | +0.361 |
Neg logit contributions | |
the | -0.277 |
not | -0.275 |
an | -0.263 |
making | -0.260 |
many | -0.246 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.21e+01 |
Pos logit contributions | |
not | +0.021 |
the | +0.021 |
an | +0.020 |
making | +0.020 |
many | +0.018 |
Neg logit contributions | |
azz | -0.045 |
pillar | -0.042 |
och | -0.041 |
pp | -0.041 |
hell | -0.040 |
Token | told |
Token ID | 1297 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
azz | +0.546 |
pillar | +0.512 |
och | +0.500 |
hell | +0.493 |
affe | +0.487 |
Neg logit contributions | |
not | -0.334 |
the | -0.332 |
making | -0.319 |
an | -0.310 |
going | -0.300 |
Token | him |
Token ID | 683 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
not | +0.116 |
making | +0.115 |
saving | +0.107 |
an | +0.106 |
many | +0.105 |
Neg logit contributions | |
azz | -0.238 |
hell | -0.216 |
pillar | -0.214 |
enny | -0.213 |
affe | -0.209 |
Token | on |
Token ID | 319 |
Feature activation | +1.05e+01 |
Pos logit contributions | |
azz | +0.200 |
pillar | +0.185 |
och | +0.179 |
hell | +0.179 |
enny | +0.177 |
Neg logit contributions | |
not | -0.090 |
making | -0.086 |
the | -0.086 |
an | -0.084 |
many | -0.078 |
Token | properly |
Token ID | 6105 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
not | +0.226 |
making | +0.215 |
the | +0.209 |
an | +0.208 |
many | +0.197 |
Neg logit contributions | |
azz | -0.501 |
pillar | -0.469 |
och | -0.456 |
hell | -0.452 |
enny | -0.446 |
Token | . |
Token ID | 13 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
not | +0.045 |
the | +0.044 |
an | +0.043 |
making | +0.043 |
many | +0.040 |
Neg logit contributions | |
azz | -0.079 |
pillar | -0.073 |
och | -0.071 |
enny | -0.070 |
hell | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
azz | +0.228 |
pp | +0.217 |
pillar | +0.213 |
hell | +0.212 |
och | +0.210 |
Neg logit contributions | |
the | -0.104 |
an | -0.097 |
not | -0.097 |
many | -0.089 |
making | -0.086 |
Token |
|
Token ID | 198 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
azz | +0.184 |
pp | +0.174 |
pillar | +0.171 |
och | +0.170 |
ppy | +0.167 |
Neg logit contributions | |
the | -0.076 |
not | -0.075 |
an | -0.071 |
many | -0.067 |
making | -0.063 |
Token | Am |
Token ID | 5840 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
azz | +0.180 |
pp | +0.173 |
och | +0.165 |
ppy | +0.160 |
hell | +0.160 |
Neg logit contributions | |
not | -0.107 |
the | -0.106 |
an | -0.102 |
many | -0.098 |
next | -0.096 |
Token | el |
Token ID | 417 |
Feature activation | -6.01e+00 |
Pos logit contributions | |
azz | +0.355 |
hell | +0.306 |
pillar | +0.306 |
och | +0.304 |
pp | +0.285 |
Neg logit contributions | |
the | -0.538 |
not | -0.520 |
an | -0.510 |
making | -0.496 |
many | -0.492 |
Token | ie |
Token ID | 494 |
Feature activation | -1.24e+01 |
Pos logit contributions | |
azz | +0.235 |
pillar | +0.214 |
och | +0.212 |
pp | +0.204 |
enny | +0.204 |
Neg logit contributions | |
the | -0.195 |
not | -0.193 |
an | -0.187 |
making | -0.184 |
many | -0.178 |
Token | was |
Token ID | 373 |
Feature activation | +9.38e+00 |
Pos logit contributions | |
azz | +0.500 |
pillar | +0.467 |
enny | +0.440 |
affe | +0.435 |
och | +0.434 |
Neg logit contributions | |
not | -0.362 |
making | -0.354 |
the | -0.351 |
an | -0.351 |
going | -0.333 |
Token | so |
Token ID | 523 |
Feature activation | +7.49e+00 |
Pos logit contributions | |
help | +0.176 |
making | +0.170 |
saving | +0.167 |
not | +0.165 |
many | +0.162 |
Neg logit contributions | |
azz | -0.398 |
pillar | -0.376 |
y | -0.369 |
ird | -0.363 |
ppy | -0.358 |
Token | proud |
Token ID | 6613 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
making | +0.166 |
not | +0.156 |
saving | +0.153 |
an | +0.153 |
the | +0.153 |
Neg logit contributions | |
azz | -0.338 |
pillar | -0.318 |
enny | -0.303 |
y | -0.303 |
och | -0.301 |
Token |
|
Token ID | 198 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
azz | +0.179 |
pillar | +0.167 |
pp | +0.162 |
och | +0.161 |
hell | +0.160 |
Neg logit contributions | |
not | -0.115 |
the | -0.115 |
an | -0.111 |
making | -0.108 |
many | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | -3.78e+00 |
Pos logit contributions | |
azz | +0.163 |
pillar | +0.154 |
pp | +0.150 |
och | +0.149 |
affe | +0.148 |
Neg logit contributions | |
the | -0.076 |
not | -0.075 |
an | -0.070 |
making | -0.068 |
many | -0.067 |
Token | " |
Token ID | 1 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
azz | +0.158 |
pillar | +0.146 |
pp | +0.145 |
affe | +0.143 |
hell | +0.142 |
Neg logit contributions | |
the | -0.118 |
not | -0.113 |
many | -0.107 |
an | -0.106 |
next | -0.102 |
Token | Anna |
Token ID | 31160 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
azz | +0.206 |
affe | +0.194 |
hell | +0.194 |
pillar | +0.193 |
pp | +0.192 |
Neg logit contributions | |
the | -0.118 |
not | -0.110 |
an | -0.108 |
making | -0.105 |
next | -0.104 |
Token | and |
Token ID | 290 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
azz | +0.648 |
pillar | +0.593 |
och | +0.592 |
hell | +0.584 |
enny | +0.578 |
Neg logit contributions | |
not | -0.295 |
the | -0.291 |
making | -0.275 |
an | -0.264 |
going | -0.245 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.32e+01 |
Pos logit contributions | |
the | +0.094 |
not | +0.092 |
an | +0.087 |
making | +0.084 |
many | +0.078 |
Neg logit contributions | |
azz | -0.209 |
pillar | -0.199 |
och | -0.193 |
hell | -0.192 |
pp | -0.192 |
Token | , |
Token ID | 11 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
azz | +0.640 |
pillar | +0.591 |
och | +0.580 |
enny | +0.572 |
pp | +0.567 |
Neg logit contributions | |
not | -0.315 |
the | -0.315 |
making | -0.299 |
an | -0.294 |
going | -0.275 |
Token | why |
Token ID | 1521 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
not | +0.046 |
the | +0.045 |
an | +0.044 |
making | +0.043 |
many | +0.040 |
Neg logit contributions | |
azz | -0.123 |
pillar | -0.115 |
och | -0.112 |
enny | -0.111 |
hell | -0.110 |
Token | are |
Token ID | 389 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
not | +0.069 |
the | +0.068 |
making | +0.065 |
an | +0.065 |
going | +0.062 |
Neg logit contributions | |
azz | -0.076 |
pillar | -0.071 |
och | -0.068 |
pp | -0.067 |
affe | -0.067 |
Token | you |
Token ID | 345 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
making | +0.188 |
an | +0.184 |
adding | +0.184 |
climbing | +0.180 |
preventing | +0.179 |
Neg logit contributions | |
azz | -0.354 |
och | -0.322 |
enny | -0.315 |
pillar | -0.309 |
ppy | -0.308 |
Token | fighting |
Token ID | 4330 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
azz | +0.095 |
y | +0.087 |
ird | +0.087 |
ppy | +0.086 |
enny | +0.086 |
Neg logit contributions | |
many | -0.054 |
the | -0.053 |
an | -0.052 |
his | -0.049 |
zed | -0.048 |
Token | poor |
Token ID | 3595 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
an | +0.135 |
the | +0.128 |
climbing | +0.128 |
making | +0.128 |
adding | +0.119 |
Neg logit contributions | |
azz | -0.534 |
hell | -0.497 |
pp | -0.495 |
och | -0.493 |
ppy | -0.492 |
Token | ." |
Token ID | 526 |
Feature activation | -6.62e+00 |
Pos logit contributions | |
not | +0.019 |
the | +0.018 |
an | +0.018 |
making | +0.018 |
many | +0.017 |
Neg logit contributions | |
azz | -0.034 |
pillar | -0.031 |
och | -0.030 |
hell | -0.030 |
enny | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
azz | +0.325 |
pillar | +0.308 |
pp | +0.301 |
hell | +0.299 |
och | +0.299 |
Neg logit contributions | |
the | -0.153 |
not | -0.145 |
an | -0.140 |
making | -0.133 |
many | -0.128 |
Token |
|
Token ID | 198 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
azz | +0.198 |
pillar | +0.188 |
pp | +0.186 |
och | +0.184 |
ppy | +0.180 |
Neg logit contributions | |
the | -0.086 |
not | -0.083 |
an | -0.078 |
many | -0.074 |
making | -0.073 |
Token | S |
Token ID | 50 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
azz | +0.155 |
pp | +0.149 |
hell | +0.140 |
pillar | +0.140 |
och | +0.139 |
Neg logit contributions | |
not | -0.142 |
the | -0.141 |
an | -0.132 |
many | -0.127 |
next | -0.125 |
Token | ara |
Token ID | 3301 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
azz | +0.257 |
suit | +0.235 |
och | +0.230 |
ird | +0.223 |
oled | +0.221 |
Neg logit contributions | |
the | -0.414 |
next | -0.409 |
first | -0.387 |
an | -0.387 |
not | -0.385 |
Token | felt |
Token ID | 2936 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
azz | +0.601 |
och | +0.584 |
pillar | +0.559 |
hell | +0.547 |
pp | +0.543 |
Neg logit contributions | |
not | -0.336 |
the | -0.334 |
an | -0.302 |
making | -0.294 |
to | -0.284 |
Token | sad |
Token ID | 6507 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
making | +0.162 |
climbing | +0.160 |
adding | +0.159 |
counting | +0.158 |
exploring | +0.153 |
Neg logit contributions | |
azz | -0.296 |
hell | -0.292 |
pillar | -0.290 |
och | -0.273 |
affe | -0.264 |
Token | . |
Token ID | 13 |
Feature activation | -4.88e+00 |
Pos logit contributions | |
not | +0.030 |
the | +0.030 |
an | +0.029 |
making | +0.029 |
many | +0.027 |
Neg logit contributions | |
azz | -0.051 |
pillar | -0.048 |
och | -0.046 |
enny | -0.046 |
hell | -0.045 |
Token | She |
Token ID | 1375 |
Feature activation | -6.11e+00 |
Pos logit contributions | |
azz | +0.178 |
pillar | +0.162 |
och | +0.156 |
enny | +0.153 |
hell | +0.151 |
Neg logit contributions | |
not | -0.170 |
the | -0.168 |
making | -0.164 |
an | -0.164 |
going | -0.156 |
Token | wanted |
Token ID | 2227 |
Feature activation | +8.17e+00 |
Pos logit contributions | |
azz | +0.251 |
pillar | +0.231 |
och | +0.226 |
pp | +0.221 |
enny | +0.221 |
Neg logit contributions | |
not | -0.188 |
the | -0.184 |
an | -0.179 |
making | -0.177 |
many | -0.168 |