Token |
|
Token ID | 198 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
itched | +0.182 |
tal | +0.172 |
fin | +0.152 |
eg | +0.152 |
ades | +0.144 |
Neg logit contributions | |
She | -0.136 |
she | -0.135 |
day | -0.128 |
weekend | -0.126 |
week | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
itched | +0.086 |
tal | +0.082 |
eg | +0.078 |
izza | +0.072 |
orse | +0.071 |
Neg logit contributions | |
She | -0.066 |
she | -0.064 |
day | -0.064 |
always | -0.057 |
sometimes | -0.056 |
Token | And |
Token ID | 1870 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
itched | +0.098 |
tal | +0.094 |
eg | +0.089 |
orse | +0.085 |
izza | +0.083 |
Neg logit contributions | |
She | -0.062 |
day | -0.060 |
she | -0.060 |
like | -0.053 |
always | -0.053 |
Token | from |
Token ID | 422 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
itched | +0.112 |
tal | +0.108 |
eg | +0.103 |
orse | +0.098 |
izza | +0.096 |
Neg logit contributions | |
She | -0.067 |
day | -0.066 |
she | -0.065 |
like | -0.058 |
especially | -0.057 |
Token | that |
Token ID | 326 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
She | +0.173 |
she | +0.164 |
day | +0.163 |
especially | +0.151 |
my | +0.150 |
Neg logit contributions | |
itched | -0.224 |
tal | -0.216 |
eg | -0.200 |
orse | -0.193 |
fin | -0.188 |
Token | day |
Token ID | 1110 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
She | +0.022 |
she | +0.015 |
especially | +0.010 |
She | +0.009 |
Hi | +0.007 |
Neg logit contributions | |
tal | -0.150 |
itched | -0.149 |
fin | -0.138 |
orse | -0.138 |
eg | -0.137 |
Token | on |
Token ID | 319 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
She | +0.813 |
day | +0.798 |
she | +0.793 |
like | +0.716 |
especially | +0.715 |
Neg logit contributions | |
itched | -1.080 |
tal | -1.029 |
eg | -0.970 |
orse | -0.916 |
izza | -0.895 |
Token | , |
Token ID | 11 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
She | +0.061 |
day | +0.060 |
she | +0.058 |
especially | +0.056 |
like | +0.056 |
Neg logit contributions | |
itched | -0.069 |
tal | -0.066 |
eg | -0.060 |
orse | -0.056 |
ves | -0.055 |
Token | Sally |
Token ID | 25737 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
She | +0.021 |
day | +0.020 |
she | +0.019 |
weekend | +0.019 |
my | +0.019 |
Neg logit contributions | |
tal | -0.028 |
itched | -0.028 |
eg | -0.026 |
orse | -0.025 |
fin | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
She | +0.284 |
day | +0.278 |
she | +0.277 |
like | +0.242 |
especially | +0.241 |
Neg logit contributions | |
itched | -0.493 |
tal | -0.474 |
eg | -0.449 |
orse | -0.427 |
izza | -0.418 |
Token | Tom |
Token ID | 4186 |
Feature activation | +4.99e-03 |
Pos logit contributions | |
She | +0.216 |
day | +0.212 |
she | +0.204 |
especially | +0.201 |
weekend | +0.197 |
Neg logit contributions | |
tal | -0.262 |
itched | -0.260 |
eg | -0.239 |
orse | -0.227 |
fin | -0.221 |
Token |
|
Token ID | 198 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
She | +0.223 |
day | +0.220 |
she | +0.216 |
like | +0.198 |
always | +0.194 |
Neg logit contributions | |
itched | -0.472 |
tal | -0.453 |
eg | -0.432 |
orse | -0.409 |
izza | -0.403 |
Token |
|
Token ID | 198 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
itched | +0.086 |
tal | +0.082 |
eg | +0.078 |
izza | +0.072 |
orse | +0.071 |
Neg logit contributions | |
She | -0.066 |
she | -0.064 |
day | -0.064 |
always | -0.057 |
sometimes | -0.056 |
Token | And |
Token ID | 1870 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
itched | +0.098 |
tal | +0.094 |
eg | +0.089 |
orse | +0.085 |
izza | +0.083 |
Neg logit contributions | |
She | -0.062 |
day | -0.060 |
she | -0.060 |
like | -0.053 |
always | -0.053 |
Token | from |
Token ID | 422 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
itched | +0.112 |
tal | +0.108 |
eg | +0.103 |
orse | +0.098 |
izza | +0.096 |
Neg logit contributions | |
She | -0.067 |
day | -0.066 |
she | -0.065 |
like | -0.058 |
especially | -0.057 |
Token | that |
Token ID | 326 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
She | +0.173 |
she | +0.164 |
day | +0.163 |
especially | +0.151 |
my | +0.150 |
Neg logit contributions | |
itched | -0.224 |
tal | -0.216 |
eg | -0.200 |
orse | -0.193 |
fin | -0.188 |
Token | day |
Token ID | 1110 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
She | +0.022 |
she | +0.015 |
especially | +0.010 |
She | +0.009 |
Hi | +0.007 |
Neg logit contributions | |
tal | -0.150 |
itched | -0.149 |
fin | -0.138 |
orse | -0.138 |
eg | -0.137 |
Token | on |
Token ID | 319 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
She | +0.813 |
day | +0.798 |
she | +0.793 |
like | +0.716 |
especially | +0.715 |
Neg logit contributions | |
itched | -1.080 |
tal | -1.029 |
eg | -0.970 |
orse | -0.916 |
izza | -0.895 |
Token | , |
Token ID | 11 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
She | +0.061 |
day | +0.060 |
she | +0.058 |
especially | +0.056 |
like | +0.056 |
Neg logit contributions | |
itched | -0.069 |
tal | -0.066 |
eg | -0.060 |
orse | -0.056 |
ves | -0.055 |
Token | Laura |
Token ID | 16753 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
She | +0.021 |
day | +0.020 |
she | +0.019 |
weekend | +0.019 |
my | +0.019 |
Neg logit contributions | |
tal | -0.028 |
itched | -0.028 |
eg | -0.026 |
orse | -0.025 |
fin | -0.024 |
Token | learned |
Token ID | 4499 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
itched | +0.034 |
tal | +0.033 |
eg | +0.031 |
orse | +0.030 |
izza | +0.029 |
Neg logit contributions | |
She | -0.020 |
day | -0.019 |
she | -0.019 |
like | -0.017 |
sometimes | -0.017 |
Token | an |
Token ID | 281 |
Feature activation | -9.75e-02 |
Pos logit contributions | |
itched | +0.052 |
tal | +0.050 |
eg | +0.047 |
orse | +0.045 |
izza | +0.044 |
Neg logit contributions | |
She | -0.040 |
day | -0.039 |
she | -0.039 |
like | -0.035 |
always | -0.034 |
Token | something |
Token ID | 1223 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
she | +0.555 |
She | +0.553 |
day | +0.540 |
always | +0.503 |
sometimes | +0.502 |
Neg logit contributions | |
tal | -0.982 |
itched | -0.976 |
eg | -0.907 |
izza | -0.864 |
orse | -0.852 |
Token | else |
Token ID | 2073 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
itched | +0.135 |
tal | +0.134 |
orse | +0.122 |
eg | +0.121 |
fin | +0.119 |
Neg logit contributions | |
She | -0.096 |
she | -0.083 |
day | -0.083 |
my | -0.082 |
sometimes | -0.078 |
Token | with |
Token ID | 351 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
itched | +0.114 |
tal | +0.110 |
eg | +0.102 |
orse | +0.099 |
fin | +0.096 |
Neg logit contributions | |
She | -0.078 |
day | -0.074 |
she | -0.073 |
always | -0.066 |
sometimes | -0.066 |
Token | paper |
Token ID | 3348 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
She | +0.049 |
day | +0.048 |
she | +0.048 |
always | +0.042 |
like | +0.042 |
Neg logit contributions | |
itched | -0.070 |
tal | -0.068 |
eg | -0.064 |
orse | -0.060 |
izza | -0.059 |
Token | another |
Token ID | 1194 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
itched | +0.111 |
tal | +0.107 |
eg | +0.101 |
orse | +0.096 |
izza | +0.094 |
Neg logit contributions | |
She | -0.079 |
she | -0.077 |
day | -0.077 |
like | -0.069 |
sometimes | -0.068 |
Token | day |
Token ID | 1110 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
itched | +0.235 |
tal | +0.224 |
eg | +0.211 |
orse | +0.197 |
izza | +0.194 |
Neg logit contributions | |
She | -0.208 |
day | -0.207 |
she | -0.206 |
like | -0.186 |
always | -0.185 |
Token | ," |
Token ID | 553 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
She | +0.823 |
she | +0.813 |
day | +0.809 |
like | +0.728 |
always | +0.720 |
Neg logit contributions | |
itched | -1.066 |
tal | -1.024 |
eg | -0.970 |
orse | -0.905 |
izza | -0.893 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
itched | +0.374 |
tal | +0.360 |
eg | +0.344 |
orse | +0.323 |
izza | +0.322 |
Neg logit contributions | |
She | -0.205 |
she | -0.203 |
day | -0.194 |
always | -0.170 |
like | -0.169 |
Token | said |
Token ID | 531 |
Feature activation | -8.66e-02 |
Pos logit contributions | |
She | +0.445 |
day | +0.445 |
she | +0.437 |
weekend | +0.402 |
like | +0.401 |
Neg logit contributions | |
itched | -0.408 |
tal | -0.388 |
eg | -0.360 |
orse | -0.338 |
izza | -0.330 |
Token | . |
Token ID | 13 |
Feature activation | +6.54e-04 |
Pos logit contributions | |
tal | +0.237 |
eg | +0.233 |
rods | +0.223 |
itched | +0.218 |
engu | +0.215 |
Neg logit contributions | |
she | -0.128 |
She | -0.116 |
her | -0.105 |
day | -0.102 |
like | -0.099 |
Token |
|
Token ID | 198 |
Feature activation | -8.36e-02 |
Pos logit contributions | |
She | +0.001 |
she | +0.001 |
day | +0.001 |
my | +0.001 |
like | +0.001 |
Neg logit contributions | |
tal | -0.002 |
itched | -0.002 |
eg | -0.001 |
urger | -0.001 |
orse | -0.001 |
Token | !" |
Token ID | 2474 |
Feature activation | -9.86e-02 |
Pos logit contributions | |
itched | +0.145 |
tal | +0.131 |
eg | +0.129 |
rods | +0.129 |
cannon | +0.128 |
Neg logit contributions | |
She | -0.122 |
she | -0.113 |
especially | -0.104 |
like | -0.103 |
heart | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
itched | +0.229 |
tal | +0.221 |
eg | +0.211 |
izza | +0.194 |
orse | +0.193 |
Neg logit contributions | |
She | -0.165 |
she | -0.162 |
day | -0.153 |
like | -0.138 |
sometimes | -0.137 |
Token |
|
Token ID | 198 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
itched | +0.098 |
tal | +0.096 |
eg | +0.093 |
urger | +0.091 |
cannon | +0.086 |
Neg logit contributions | |
She | -0.072 |
she | -0.066 |
day | -0.063 |
She | -0.059 |
my | -0.057 |
Token | From |
Token ID | 4863 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
itched | +0.120 |
tal | +0.117 |
eg | +0.110 |
izza | +0.103 |
urger | +0.102 |
Neg logit contributions | |
She | -0.072 |
she | -0.069 |
day | -0.067 |
always | -0.059 |
especially | -0.059 |
Token | that |
Token ID | 326 |
Feature activation | +4.25e-02 |
Pos logit contributions | |
She | +0.021 |
day | +0.020 |
she | +0.020 |
like | +0.018 |
always | +0.018 |
Neg logit contributions | |
itched | -0.030 |
tal | -0.029 |
eg | -0.027 |
orse | -0.026 |
izza | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | +4.85e-01 |
Pos logit contributions | |
She | +0.016 |
she | +0.011 |
especially | +0.006 |
like | +0.005 |
She | +0.004 |
Neg logit contributions | |
itched | -0.147 |
tal | -0.146 |
eg | -0.137 |
orse | -0.136 |
fin | -0.133 |
Token | on |
Token ID | 319 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
She | +0.725 |
she | +0.717 |
day | +0.717 |
always | +0.621 |
like | +0.621 |
Neg logit contributions | |
itched | -1.166 |
tal | -1.137 |
eg | -1.076 |
orse | -1.026 |
izza | -1.023 |
Token | , |
Token ID | 11 |
Feature activation | +9.59e-03 |
Pos logit contributions | |
She | +0.061 |
day | +0.059 |
she | +0.058 |
especially | +0.056 |
like | +0.055 |
Neg logit contributions | |
itched | -0.067 |
tal | -0.064 |
eg | -0.057 |
orse | -0.054 |
ves | -0.053 |
Token | Tiny |
Token ID | 20443 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
She | +0.014 |
day | +0.013 |
she | +0.013 |
weekend | +0.012 |
especially | +0.012 |
Neg logit contributions | |
itched | -0.021 |
tal | -0.021 |
eg | -0.020 |
orse | -0.019 |
izza | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
itched | +0.294 |
tal | +0.283 |
eg | +0.266 |
orse | +0.256 |
fin | +0.247 |
Neg logit contributions | |
She | -0.176 |
day | -0.170 |
she | -0.170 |
like | -0.149 |
weekend | -0.148 |
Token | Max |
Token ID | 5436 |
Feature activation | -9.95e-03 |
Pos logit contributions | |
She | +0.229 |
day | +0.213 |
weekend | +0.209 |
she | +0.208 |
especially | +0.204 |
Neg logit contributions | |
tal | -0.251 |
itched | -0.238 |
eg | -0.219 |
orse | -0.215 |
fin | -0.207 |
Token | and |
Token ID | 290 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
itched | +0.370 |
tal | +0.353 |
eg | +0.343 |
orse | +0.324 |
izza | +0.320 |
Neg logit contributions | |
She | -0.255 |
she | -0.244 |
day | -0.233 |
like | -0.215 |
sometimes | -0.213 |
Token | ' |
Token ID | 705 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
She | +0.280 |
day | +0.273 |
she | +0.271 |
like | +0.243 |
always | +0.242 |
Neg logit contributions | |
itched | -0.387 |
tal | -0.373 |
eg | -0.350 |
orse | -0.333 |
izza | -0.326 |
Token | no |
Token ID | 3919 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
itched | +0.536 |
tal | +0.518 |
eg | +0.492 |
orse | +0.472 |
izza | +0.466 |
Neg logit contributions | |
She | -0.236 |
she | -0.229 |
day | -0.228 |
like | -0.194 |
always | -0.193 |
Token | ' |
Token ID | 6 |
Feature activation | -7.27e-02 |
Pos logit contributions | |
day | +0.092 |
she | +0.089 |
She | +0.088 |
always | +0.083 |
sometimes | +0.082 |
Neg logit contributions | |
itched | -0.096 |
eg | -0.092 |
tal | -0.091 |
cannon | -0.082 |
fin | -0.081 |
Token | all |
Token ID | 477 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
itched | +0.162 |
tal | +0.156 |
eg | +0.148 |
orse | +0.141 |
fin | +0.136 |
Neg logit contributions | |
She | -0.123 |
day | -0.121 |
she | -0.121 |
always | -0.109 |
like | -0.109 |
Token | day |
Token ID | 1110 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
She | +0.292 |
she | +0.284 |
day | +0.282 |
especially | +0.256 |
like | +0.255 |
Neg logit contributions | |
itched | -0.363 |
tal | -0.350 |
eg | -0.322 |
orse | -0.307 |
izza | -0.303 |
Token | . |
Token ID | 13 |
Feature activation | +8.29e-03 |
Pos logit contributions | |
she | +0.894 |
day | +0.885 |
She | +0.880 |
like | +0.797 |
always | +0.795 |
Neg logit contributions | |
itched | -1.005 |
tal | -0.957 |
eg | -0.926 |
orse | -0.849 |
fin | -0.831 |
Token | But |
Token ID | 887 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
She | +0.013 |
she | +0.012 |
day | +0.012 |
sometimes | +0.011 |
always | +0.011 |
Neg logit contributions | |
itched | -0.020 |
tal | -0.019 |
eg | -0.018 |
orse | -0.017 |
fin | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
She | +0.237 |
day | +0.223 |
she | +0.221 |
especially | +0.197 |
like | +0.197 |
Neg logit contributions | |
itched | -0.376 |
tal | -0.363 |
eg | -0.339 |
orse | -0.329 |
izza | -0.323 |
Token | child |
Token ID | 1200 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
She | +0.028 |
she | +0.027 |
day | +0.027 |
like | +0.024 |
sometimes | +0.024 |
Neg logit contributions | |
itched | -0.041 |
tal | -0.040 |
eg | -0.038 |
orse | -0.036 |
izza | -0.035 |
Token | was |
Token ID | 373 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
itched | +0.043 |
tal | +0.041 |
eg | +0.039 |
orse | +0.037 |
izza | +0.036 |
Neg logit contributions | |
She | -0.027 |
day | -0.026 |
she | -0.026 |
like | -0.023 |
sometimes | -0.023 |
Token | tick |
Token ID | 4378 |
Feature activation | +7.01e-02 |
Pos logit contributions | |
itched | +0.071 |
tal | +0.065 |
eg | +0.059 |
fin | +0.059 |
cannon | +0.057 |
Neg logit contributions | |
She | -0.066 |
day | -0.063 |
she | -0.063 |
weekend | -0.059 |
week | -0.059 |
Token | - |
Token ID | 12 |
Feature activation | -9.25e-02 |
Pos logit contributions | |
She | +0.132 |
day | +0.129 |
she | +0.129 |
like | +0.117 |
always | +0.117 |
Neg logit contributions | |
itched | -0.142 |
tal | -0.135 |
eg | -0.126 |
orse | -0.119 |
izza | -0.116 |
Token | t |
Token ID | 83 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
itched | +0.179 |
tal | +0.170 |
eg | +0.158 |
orse | +0.150 |
ves | +0.145 |
Neg logit contributions | |
She | -0.179 |
day | -0.176 |
she | -0.174 |
like | -0.160 |
especially | -0.160 |
Token | ocked |
Token ID | 3543 |
Feature activation | -8.87e-03 |
Pos logit contributions | |
She | +0.177 |
day | +0.161 |
she | +0.159 |
weekend | +0.153 |
girl | +0.146 |
Neg logit contributions | |
itched | -0.350 |
tal | -0.343 |
orse | -0.313 |
eg | -0.307 |
fin | -0.301 |
Token | all |
Token ID | 477 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
itched | +0.020 |
tal | +0.019 |
eg | +0.018 |
orse | +0.017 |
fin | +0.017 |
Neg logit contributions | |
She | -0.014 |
day | -0.013 |
she | -0.013 |
weekend | -0.012 |
especially | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
She | +0.253 |
especially | +0.235 |
she | +0.232 |
She | +0.220 |
weekend | +0.219 |
Neg logit contributions | |
tal | -0.352 |
itched | -0.347 |
fin | -0.284 |
cannon | -0.280 |
eg | -0.279 |
Token | long |
Token ID | 890 |
Feature activation | +3.55e-02 |
Pos logit contributions | |
she | +0.778 |
She | +0.758 |
day | +0.748 |
like | +0.672 |
always | +0.662 |
Neg logit contributions | |
itched | -1.122 |
tal | -1.094 |
eg | -1.065 |
izza | -0.991 |
orse | -0.971 |
Token | . |
Token ID | 13 |
Feature activation | +5.16e-03 |
Pos logit contributions | |
She | +0.059 |
day | +0.058 |
she | +0.058 |
like | +0.052 |
always | +0.051 |
Neg logit contributions | |
itched | -0.079 |
tal | -0.076 |
eg | -0.071 |
orse | -0.067 |
izza | -0.066 |
Token | One |
Token ID | 1881 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
She | +0.009 |
she | +0.008 |
day | +0.008 |
sometimes | +0.007 |
always | +0.007 |
Neg logit contributions | |
itched | -0.012 |
tal | -0.011 |
eg | -0.011 |
orse | -0.010 |
izza | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
She | +0.094 |
She | +0.055 |
she | +0.052 |
You | +0.045 |
like | +0.042 |
Neg logit contributions | |
tal | -0.517 |
itched | -0.515 |
orse | -0.507 |
fin | -0.488 |
eg | -0.469 |
Token | , |
Token ID | 11 |
Feature activation | +3.86e-03 |
Pos logit contributions | |
she | +0.744 |
She | +0.712 |
day | +0.682 |
like | +0.613 |
always | +0.593 |
Neg logit contributions | |
itched | -1.144 |
tal | -1.121 |
eg | -1.063 |
orse | -1.003 |
izza | -1.001 |
Token | ." |
Token ID | 526 |
Feature activation | -9.92e-03 |
Pos logit contributions | |
itched | +0.072 |
tal | +0.069 |
eg | +0.065 |
orse | +0.061 |
izza | +0.060 |
Neg logit contributions | |
She | -0.055 |
she | -0.054 |
day | -0.054 |
always | -0.048 |
sometimes | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
itched | +0.022 |
tal | +0.021 |
eg | +0.020 |
orse | +0.019 |
izza | +0.019 |
Neg logit contributions | |
She | -0.017 |
she | -0.016 |
day | -0.016 |
like | -0.014 |
sometimes | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -5.22e-02 |
Pos logit contributions | |
itched | +0.108 |
tal | +0.104 |
eg | +0.100 |
urger | +0.095 |
izza | +0.092 |
Neg logit contributions | |
She | -0.086 |
she | -0.080 |
day | -0.079 |
my | -0.071 |
She | -0.070 |
Token | From |
Token ID | 4863 |
Feature activation | +9.81e-03 |
Pos logit contributions | |
itched | +0.128 |
tal | +0.124 |
eg | +0.117 |
orse | +0.111 |
izza | +0.110 |
Neg logit contributions | |
She | -0.075 |
she | -0.073 |
day | -0.073 |
like | -0.064 |
always | -0.064 |
Token | that |
Token ID | 326 |
Feature activation | +3.93e-02 |
Pos logit contributions | |
She | +0.016 |
day | +0.016 |
she | +0.016 |
especially | +0.014 |
always | +0.014 |
Neg logit contributions | |
itched | -0.022 |
tal | -0.021 |
eg | -0.020 |
orse | -0.019 |
fin | -0.018 |
Token | day |
Token ID | 1110 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
She | +0.019 |
she | +0.013 |
especially | +0.009 |
like | +0.008 |
She | +0.008 |
Neg logit contributions | |
tal | -0.131 |
itched | -0.130 |
eg | -0.122 |
orse | -0.122 |
fin | -0.119 |
Token | on |
Token ID | 319 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
She | +0.740 |
day | +0.729 |
she | +0.727 |
like | +0.637 |
always | +0.636 |
Neg logit contributions | |
itched | -1.146 |
tal | -1.111 |
eg | -1.050 |
orse | -1.001 |
izza | -0.990 |
Token | , |
Token ID | 11 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
She | +0.059 |
day | +0.057 |
especially | +0.056 |
she | +0.055 |
like | +0.055 |
Neg logit contributions | |
itched | -0.059 |
tal | -0.056 |
eg | -0.050 |
orse | -0.047 |
ves | -0.047 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
She | +0.010 |
day | +0.010 |
she | +0.009 |
weekend | +0.009 |
my | +0.009 |
Neg logit contributions | |
itched | -0.016 |
tal | -0.016 |
eg | -0.015 |
orse | -0.015 |
izza | -0.014 |
Token | always |
Token ID | 1464 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
She | +0.322 |
she | +0.306 |
day | +0.304 |
weekend | +0.287 |
week | +0.278 |
Neg logit contributions | |
itched | -0.502 |
tal | -0.472 |
eg | -0.445 |
orse | -0.435 |
fin | -0.415 |
Token | cleaned |
Token ID | 20750 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
She | +0.700 |
she | +0.679 |
day | +0.676 |
like | +0.610 |
especially | +0.608 |
Neg logit contributions | |
itched | -0.845 |
tal | -0.812 |
eg | -0.752 |
orse | -0.718 |
fin | -0.700 |
Token | and |
Token ID | 290 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
She | +0.341 |
she | +0.335 |
day | +0.334 |
like | +0.299 |
sometimes | +0.298 |
Neg logit contributions | |
itched | -0.445 |
tal | -0.428 |
eg | -0.403 |
orse | -0.379 |
izza | -0.373 |
Token | it |
Token ID | 340 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
She | +0.280 |
day | +0.273 |
she | +0.271 |
like | +0.243 |
always | +0.242 |
Neg logit contributions | |
itched | -0.387 |
tal | -0.373 |
eg | -0.350 |
orse | -0.333 |
izza | -0.326 |
Token | was |
Token ID | 373 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
She | +0.031 |
she | +0.030 |
day | +0.030 |
always | +0.027 |
like | +0.027 |
Neg logit contributions | |
itched | -0.049 |
tal | -0.047 |
eg | -0.045 |
orse | -0.042 |
izza | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
day | +0.076 |
she | +0.074 |
She | +0.073 |
always | +0.067 |
like | +0.067 |
Neg logit contributions | |
itched | -0.137 |
tal | -0.132 |
eg | -0.129 |
orse | -0.121 |
izza | -0.120 |
Token | wonderful |
Token ID | 7932 |
Feature activation | +9.82e-02 |
Pos logit contributions | |
itched | +0.027 |
tal | +0.026 |
eg | +0.024 |
orse | +0.023 |
izza | +0.023 |
Neg logit contributions | |
day | -0.017 |
She | -0.017 |
she | -0.017 |
like | -0.015 |
always | -0.015 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.121 |
she | +0.117 |
day | +0.117 |
like | +0.100 |
always | +0.100 |
Neg logit contributions | |
itched | -0.262 |
tal | -0.253 |
eg | -0.240 |
orse | -0.230 |
izza | -0.226 |
Token | . |
Token ID | 13 |
Feature activation | +4.49e-03 |
Pos logit contributions | |
day | +0.751 |
she | +0.750 |
She | +0.737 |
like | +0.667 |
always | +0.651 |
Neg logit contributions | |
itched | -1.135 |
tal | -1.106 |
eg | -1.083 |
izza | -0.988 |
orse | -0.967 |
Token | |
Token ID | 220 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
She | +0.008 |
day | +0.008 |
she | +0.008 |
like | +0.007 |
always | +0.007 |
Neg logit contributions | |
itched | -0.009 |
tal | -0.009 |
eg | -0.008 |
orse | -0.008 |
izza | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -7.78e-02 |
Pos logit contributions | |
itched | +0.405 |
tal | +0.402 |
eg | +0.378 |
fin | +0.375 |
urger | +0.368 |
Neg logit contributions | |
She | -0.291 |
she | -0.254 |
day | -0.238 |
my | -0.224 |
always | -0.206 |
Token |
|
Token ID | 198 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
itched | +0.169 |
tal | +0.163 |
eg | +0.155 |
fin | +0.143 |
urger | +0.142 |
Neg logit contributions | |
She | -0.143 |
day | -0.137 |
she | -0.137 |
especially | -0.122 |
my | -0.121 |
Token | In |
Token ID | 818 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
itched | +0.180 |
tal | +0.172 |
eg | +0.162 |
orse | +0.157 |
izza | +0.152 |
Neg logit contributions | |
She | -0.124 |
day | -0.122 |
she | -0.122 |
like | -0.111 |
sometimes | -0.108 |
Token | ummy |
Token ID | 13513 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
She | +0.022 |
she | +0.022 |
day | +0.021 |
like | +0.019 |
always | +0.019 |
Neg logit contributions | |
itched | -0.036 |
tal | -0.035 |
eg | -0.033 |
orse | -0.031 |
izza | -0.031 |
Token | lunch |
Token ID | 9965 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
She | +0.305 |
day | +0.294 |
she | +0.293 |
especially | +0.265 |
sometimes | +0.264 |
Neg logit contributions | |
itched | -0.407 |
tal | -0.396 |
eg | -0.364 |
orse | -0.352 |
izza | -0.346 |
Token | . |
Token ID | 13 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
itched | +0.128 |
tal | +0.122 |
eg | +0.114 |
orse | +0.111 |
izza | +0.109 |
Neg logit contributions | |
She | -0.098 |
day | -0.093 |
she | -0.092 |
matter | -0.085 |
sometimes | -0.085 |
Token | From |
Token ID | 3574 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
She | +0.045 |
day | +0.045 |
she | +0.045 |
like | +0.040 |
always | +0.040 |
Neg logit contributions | |
itched | -0.070 |
tal | -0.067 |
eg | -0.063 |
orse | -0.060 |
izza | -0.059 |
Token | that |
Token ID | 326 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
itched | +0.345 |
tal | +0.334 |
eg | +0.305 |
orse | +0.294 |
fin | +0.289 |
Neg logit contributions | |
She | -0.254 |
she | -0.240 |
day | -0.239 |
especially | -0.218 |
like | -0.214 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.021 |
she | +0.014 |
She | +0.010 |
like | +0.008 |
especially | +0.008 |
Neg logit contributions | |
tal | -0.136 |
itched | -0.135 |
orse | -0.128 |
eg | -0.127 |
fin | -0.126 |
Token | on |
Token ID | 319 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
She | +0.749 |
day | +0.736 |
she | +0.735 |
like | +0.648 |
always | +0.646 |
Neg logit contributions | |
itched | -1.137 |
tal | -1.098 |
eg | -1.038 |
orse | -0.986 |
izza | -0.972 |
Token | , |
Token ID | 11 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
She | +0.070 |
day | +0.068 |
she | +0.066 |
especially | +0.065 |
like | +0.065 |
Neg logit contributions | |
itched | -0.074 |
tal | -0.070 |
eg | -0.062 |
orse | -0.059 |
ves | -0.059 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
She | +0.021 |
day | +0.020 |
weekend | +0.019 |
ock | +0.019 |
my | +0.019 |
Neg logit contributions | |
tal | -0.031 |
itched | -0.030 |
eg | -0.028 |
orse | -0.028 |
izza | -0.027 |
Token | my |
Token ID | 1820 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
She | +0.044 |
she | +0.040 |
She | +0.028 |
weekend | +0.028 |
Welcome | +0.027 |
Neg logit contributions | |
itched | -0.153 |
tal | -0.150 |
eg | -0.143 |
orse | -0.141 |
urger | -0.138 |
Token | always |
Token ID | 1464 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
She | +0.587 |
she | +0.561 |
day | +0.543 |
weekend | +0.494 |
like | +0.488 |
Neg logit contributions | |
itched | -0.997 |
tal | -0.952 |
eg | -0.902 |
orse | -0.878 |
fin | -0.845 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
itched | +0.628 |
tal | +0.592 |
eg | +0.554 |
orse | +0.539 |
fin | +0.524 |
Neg logit contributions | |
She | -0.411 |
day | -0.395 |
she | -0.381 |
weekend | -0.357 |
especially | -0.351 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
She | +0.043 |
she | +0.042 |
day | +0.042 |
like | +0.037 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.072 |
eg | -0.069 |
orse | -0.065 |
izza | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
itched | +0.089 |
tal | +0.085 |
eg | +0.081 |
urger | +0.075 |
fin | +0.074 |
Neg logit contributions | |
She | -0.081 |
day | -0.077 |
she | -0.076 |
my | -0.069 |
week | -0.068 |
Token | From |
Token ID | 4863 |
Feature activation | +6.65e-03 |
Pos logit contributions | |
itched | +0.114 |
tal | +0.109 |
eg | +0.103 |
orse | +0.099 |
izza | +0.096 |
Neg logit contributions | |
She | -0.073 |
day | -0.072 |
she | -0.072 |
like | -0.064 |
always | -0.063 |
Token | that |
Token ID | 326 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
She | +0.011 |
day | +0.011 |
she | +0.011 |
especially | +0.010 |
my | +0.010 |
Neg logit contributions | |
itched | -0.015 |
tal | -0.014 |
eg | -0.013 |
orse | -0.012 |
fin | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.020 |
she | +0.013 |
She | +0.010 |
especially | +0.010 |
like | +0.009 |
Neg logit contributions | |
tal | -0.132 |
itched | -0.132 |
eg | -0.123 |
orse | -0.123 |
fin | -0.122 |
Token | on |
Token ID | 319 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
She | +0.775 |
day | +0.757 |
she | +0.756 |
like | +0.675 |
always | +0.672 |
Neg logit contributions | |
itched | -1.111 |
tal | -1.065 |
eg | -1.005 |
orse | -0.951 |
izza | -0.933 |
Token | , |
Token ID | 11 |
Feature activation | +9.42e-03 |
Pos logit contributions | |
She | +0.065 |
day | +0.063 |
especially | +0.062 |
she | +0.061 |
like | +0.061 |
Neg logit contributions | |
itched | -0.066 |
tal | -0.062 |
eg | -0.055 |
ves | -0.052 |
orse | -0.052 |
Token | Owen |
Token ID | 22605 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
She | +0.014 |
day | +0.013 |
she | +0.012 |
weekend | +0.012 |
my | +0.012 |
Neg logit contributions | |
itched | -0.020 |
tal | -0.020 |
eg | -0.019 |
orse | -0.018 |
ves | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
itched | +0.497 |
tal | +0.473 |
eg | +0.452 |
orse | +0.434 |
izza | +0.417 |
Neg logit contributions | |
She | -0.280 |
she | -0.270 |
day | -0.267 |
weekend | -0.236 |
week | -0.233 |
Token | more |
Token ID | 517 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
She | +0.033 |
day | +0.032 |
she | +0.032 |
like | +0.028 |
sometimes | +0.028 |
Neg logit contributions | |
itched | -0.065 |
tal | -0.063 |
eg | -0.060 |
orse | -0.057 |
izza | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
itched | +0.235 |
cks | +0.219 |
tal | +0.215 |
eg | +0.212 |
ves | +0.208 |
Neg logit contributions | |
day | -0.133 |
she | -0.129 |
She | -0.121 |
girl | -0.114 |
like | -0.113 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
She | +0.043 |
she | +0.042 |
day | +0.042 |
like | +0.037 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.072 |
eg | -0.069 |
orse | -0.065 |
izza | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
itched | +0.089 |
tal | +0.085 |
eg | +0.081 |
urger | +0.075 |
fin | +0.074 |
Neg logit contributions | |
She | -0.081 |
day | -0.077 |
she | -0.076 |
my | -0.069 |
week | -0.068 |
Token | From |
Token ID | 4863 |
Feature activation | +6.65e-03 |
Pos logit contributions | |
itched | +0.114 |
tal | +0.109 |
eg | +0.103 |
orse | +0.099 |
izza | +0.096 |
Neg logit contributions | |
She | -0.073 |
day | -0.072 |
she | -0.072 |
like | -0.064 |
always | -0.063 |
Token | that |
Token ID | 326 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
She | +0.011 |
day | +0.011 |
she | +0.011 |
especially | +0.010 |
my | +0.010 |
Neg logit contributions | |
itched | -0.015 |
tal | -0.014 |
eg | -0.013 |
orse | -0.012 |
fin | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.020 |
she | +0.013 |
She | +0.010 |
especially | +0.010 |
like | +0.009 |
Neg logit contributions | |
tal | -0.132 |
itched | -0.132 |
eg | -0.123 |
orse | -0.123 |
fin | -0.122 |
Token | on |
Token ID | 319 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
She | +0.775 |
day | +0.757 |
she | +0.756 |
like | +0.675 |
always | +0.672 |
Neg logit contributions | |
itched | -1.111 |
tal | -1.065 |
eg | -1.005 |
orse | -0.951 |
izza | -0.933 |
Token | , |
Token ID | 11 |
Feature activation | +9.42e-03 |
Pos logit contributions | |
She | +0.065 |
day | +0.063 |
especially | +0.062 |
she | +0.061 |
like | +0.061 |
Neg logit contributions | |
itched | -0.066 |
tal | -0.062 |
eg | -0.055 |
ves | -0.052 |
orse | -0.052 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
She | +0.014 |
day | +0.013 |
she | +0.012 |
weekend | +0.012 |
my | +0.012 |
Neg logit contributions | |
itched | -0.020 |
tal | -0.020 |
eg | -0.019 |
orse | -0.018 |
ves | -0.017 |
Token | my |
Token ID | 1820 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
She | +0.041 |
she | +0.038 |
weekend | +0.027 |
day | +0.025 |
She | +0.023 |
Neg logit contributions | |
itched | -0.150 |
tal | -0.146 |
eg | -0.140 |
orse | -0.135 |
fin | -0.132 |
Token | always |
Token ID | 1464 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
She | +0.572 |
she | +0.551 |
day | +0.548 |
weekend | +0.481 |
like | +0.477 |
Neg logit contributions | |
itched | -1.017 |
tal | -0.972 |
eg | -0.921 |
orse | -0.886 |
fin | -0.857 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
itched | +0.077 |
tal | +0.075 |
eg | +0.067 |
fin | +0.066 |
orse | +0.065 |
Neg logit contributions | |
She | -0.079 |
she | -0.075 |
day | -0.072 |
my | -0.069 |
like | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
She | +0.043 |
she | +0.042 |
day | +0.042 |
like | +0.037 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.072 |
eg | -0.069 |
orse | -0.065 |
izza | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
itched | +0.089 |
tal | +0.085 |
eg | +0.081 |
urger | +0.075 |
fin | +0.074 |
Neg logit contributions | |
She | -0.081 |
day | -0.077 |
she | -0.076 |
my | -0.069 |
week | -0.068 |
Token | From |
Token ID | 4863 |
Feature activation | +6.65e-03 |
Pos logit contributions | |
itched | +0.114 |
tal | +0.109 |
eg | +0.103 |
orse | +0.099 |
izza | +0.096 |
Neg logit contributions | |
She | -0.073 |
day | -0.072 |
she | -0.072 |
like | -0.064 |
always | -0.063 |
Token | that |
Token ID | 326 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
She | +0.011 |
day | +0.011 |
she | +0.011 |
especially | +0.010 |
my | +0.010 |
Neg logit contributions | |
itched | -0.015 |
tal | -0.014 |
eg | -0.013 |
orse | -0.012 |
fin | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.020 |
she | +0.013 |
She | +0.010 |
especially | +0.010 |
like | +0.009 |
Neg logit contributions | |
tal | -0.132 |
itched | -0.132 |
eg | -0.123 |
orse | -0.123 |
fin | -0.122 |
Token | on |
Token ID | 319 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
She | +0.775 |
day | +0.757 |
she | +0.756 |
like | +0.675 |
always | +0.672 |
Neg logit contributions | |
itched | -1.111 |
tal | -1.065 |
eg | -1.005 |
orse | -0.951 |
izza | -0.933 |
Token | , |
Token ID | 11 |
Feature activation | +9.42e-03 |
Pos logit contributions | |
She | +0.065 |
day | +0.063 |
especially | +0.062 |
she | +0.061 |
like | +0.061 |
Neg logit contributions | |
itched | -0.066 |
tal | -0.062 |
eg | -0.055 |
ves | -0.052 |
orse | -0.052 |
Token | they |
Token ID | 484 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
She | +0.014 |
day | +0.013 |
she | +0.012 |
weekend | +0.012 |
my | +0.012 |
Neg logit contributions | |
itched | -0.020 |
tal | -0.020 |
eg | -0.019 |
orse | -0.018 |
ves | -0.017 |
Token | learned |
Token ID | 4499 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
She | +0.059 |
day | +0.056 |
she | +0.056 |
weekend | +0.052 |
week | +0.050 |
Neg logit contributions | |
itched | -0.119 |
tal | -0.114 |
eg | -0.111 |
orse | -0.106 |
fin | -0.103 |
Token | that |
Token ID | 326 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
itched | +0.064 |
tal | +0.062 |
eg | +0.060 |
orse | +0.057 |
izza | +0.055 |
Neg logit contributions | |
she | -0.040 |
She | -0.039 |
day | -0.039 |
sometimes | -0.035 |
like | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
itched | +0.371 |
tal | +0.356 |
eg | +0.332 |
orse | +0.321 |
fin | +0.319 |
Neg logit contributions | |
She | -0.261 |
she | -0.246 |
day | -0.239 |
like | -0.224 |
especially | -0.222 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
She | +0.043 |
she | +0.042 |
day | +0.042 |
like | +0.037 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.072 |
eg | -0.069 |
orse | -0.065 |
izza | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
itched | +0.089 |
tal | +0.085 |
eg | +0.081 |
urger | +0.075 |
fin | +0.074 |
Neg logit contributions | |
She | -0.081 |
day | -0.077 |
she | -0.076 |
my | -0.069 |
week | -0.068 |
Token | From |
Token ID | 4863 |
Feature activation | +6.65e-03 |
Pos logit contributions | |
itched | +0.114 |
tal | +0.109 |
eg | +0.103 |
orse | +0.099 |
izza | +0.096 |
Neg logit contributions | |
She | -0.073 |
day | -0.072 |
she | -0.072 |
like | -0.064 |
always | -0.063 |
Token | that |
Token ID | 326 |
Feature activation | +4.02e-02 |
Pos logit contributions | |
She | +0.011 |
day | +0.011 |
she | +0.011 |
especially | +0.010 |
my | +0.010 |
Neg logit contributions | |
itched | -0.015 |
tal | -0.014 |
eg | -0.013 |
orse | -0.012 |
fin | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.020 |
she | +0.013 |
She | +0.010 |
especially | +0.010 |
like | +0.009 |
Neg logit contributions | |
tal | -0.132 |
itched | -0.132 |
eg | -0.123 |
orse | -0.123 |
fin | -0.122 |
Token | on |
Token ID | 319 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
She | +0.775 |
day | +0.757 |
she | +0.756 |
like | +0.675 |
always | +0.672 |
Neg logit contributions | |
itched | -1.111 |
tal | -1.065 |
eg | -1.005 |
orse | -0.951 |
izza | -0.933 |
Token | , |
Token ID | 11 |
Feature activation | +9.42e-03 |
Pos logit contributions | |
She | +0.065 |
day | +0.063 |
especially | +0.062 |
she | +0.061 |
like | +0.061 |
Neg logit contributions | |
itched | -0.066 |
tal | -0.062 |
eg | -0.055 |
ves | -0.052 |
orse | -0.052 |
Token | Amy |
Token ID | 14235 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
She | +0.014 |
day | +0.013 |
she | +0.012 |
weekend | +0.012 |
my | +0.012 |
Neg logit contributions | |
itched | -0.020 |
tal | -0.020 |
eg | -0.019 |
orse | -0.018 |
ves | -0.017 |
Token | learned |
Token ID | 4499 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
She | +0.085 |
day | +0.082 |
she | +0.081 |
weekend | +0.075 |
week | +0.073 |
Neg logit contributions | |
itched | -0.159 |
tal | -0.151 |
eg | -0.142 |
orse | -0.137 |
fin | -0.133 |
Token | to |
Token ID | 284 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
itched | +0.060 |
tal | +0.058 |
eg | +0.054 |
orse | +0.050 |
izza | +0.050 |
Neg logit contributions | |
She | -0.047 |
day | -0.047 |
she | -0.046 |
especially | -0.042 |
like | -0.041 |
Token | opl |
Token ID | 20106 |
Feature activation | +5.66e-02 |
Pos logit contributions | |
She | +0.133 |
she | +0.118 |
my | +0.118 |
especially | +0.114 |
day | +0.111 |
Neg logit contributions | |
tal | -0.170 |
itched | -0.168 |
fin | -0.152 |
eg | -0.150 |
orse | -0.149 |
Token | ane |
Token ID | 1531 |
Feature activation | +8.48e-03 |
Pos logit contributions | |
She | +0.105 |
day | +0.104 |
she | +0.103 |
like | +0.094 |
always | +0.093 |
Neg logit contributions | |
itched | -0.114 |
tal | -0.110 |
eg | -0.102 |
orse | -0.095 |
izza | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
day | +0.013 |
she | +0.013 |
She | +0.013 |
always | +0.012 |
like | +0.012 |
Neg logit contributions | |
itched | -0.020 |
tal | -0.019 |
eg | -0.018 |
izza | -0.017 |
orse | -0.017 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
She | +0.057 |
day | +0.056 |
she | +0.056 |
like | +0.049 |
always | +0.049 |
Neg logit contributions | |
itched | -0.102 |
tal | -0.098 |
eg | -0.093 |
orse | -0.089 |
izza | -0.087 |
Token | One |
Token ID | 3198 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
itched | +0.444 |
tal | +0.429 |
eg | +0.407 |
urger | +0.391 |
izza | +0.387 |
Neg logit contributions | |
She | -0.343 |
day | -0.334 |
she | -0.318 |
my | -0.290 |
like | -0.284 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.132 |
she | +0.090 |
Why | +0.071 |
You | +0.068 |
especially | +0.068 |
Neg logit contributions | |
itched | -0.743 |
tal | -0.737 |
orse | -0.714 |
fin | -0.674 |
eg | -0.672 |
Token | , |
Token ID | 11 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
she | +0.729 |
day | +0.719 |
She | +0.701 |
week | +0.601 |
like | +0.599 |
Neg logit contributions | |
tal | -1.185 |
itched | -1.160 |
eg | -1.096 |
wounds | -1.062 |
izza | -1.040 |
Token | a |
Token ID | 257 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
She | +0.028 |
she | +0.027 |
day | +0.027 |
like | +0.024 |
always | +0.024 |
Neg logit contributions | |
itched | -0.035 |
tal | -0.034 |
eg | -0.032 |
orse | -0.030 |
izza | -0.029 |
Token | turkey |
Token ID | 26876 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
itched | +0.050 |
tal | +0.047 |
eg | +0.046 |
orse | +0.043 |
fin | +0.042 |
Neg logit contributions | |
She | -0.037 |
she | -0.035 |
day | -0.035 |
sometimes | -0.032 |
like | -0.032 |
Token | had |
Token ID | 550 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
itched | +0.078 |
tal | +0.074 |
eg | +0.069 |
orse | +0.067 |
fin | +0.064 |
Neg logit contributions | |
She | -0.061 |
she | -0.059 |
day | -0.059 |
weekend | -0.053 |
especially | -0.053 |
Token | an |
Token ID | 281 |
Feature activation | -9.60e-02 |
Pos logit contributions | |
She | +0.256 |
she | +0.250 |
day | +0.248 |
always | +0.224 |
like | +0.219 |
Neg logit contributions | |
itched | -0.350 |
tal | -0.335 |
eg | -0.323 |
orse | -0.302 |
izza | -0.302 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
She | +0.154 |
she | +0.151 |
day | +0.149 |
weekend | +0.137 |
always | +0.137 |
Neg logit contributions | |
itched | -0.170 |
tal | -0.162 |
eg | -0.149 |
fin | -0.139 |
orse | -0.139 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
She | +0.043 |
she | +0.042 |
day | +0.042 |
like | +0.037 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.072 |
eg | -0.069 |
orse | -0.065 |
izza | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
itched | +0.089 |
tal | +0.085 |
eg | +0.081 |
urger | +0.075 |
fin | +0.074 |
Neg logit contributions | |
She | -0.081 |
day | -0.077 |
she | -0.076 |
my | -0.069 |
week | -0.068 |
Token | Later |
Token ID | 18602 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
itched | +0.114 |
tal | +0.109 |
eg | +0.103 |
orse | +0.099 |
izza | +0.096 |
Neg logit contributions | |
She | -0.073 |
day | -0.072 |
she | -0.072 |
like | -0.064 |
always | -0.063 |
Token | that |
Token ID | 326 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
She | +0.003 |
day | +0.003 |
she | +0.003 |
especially | +0.003 |
like | +0.003 |
Neg logit contributions | |
itched | -0.004 |
tal | -0.004 |
eg | -0.003 |
orse | -0.003 |
izza | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.012 |
she | +0.008 |
especially | +0.006 |
like | +0.005 |
sometimes | +0.003 |
Neg logit contributions | |
itched | -0.134 |
tal | -0.130 |
eg | -0.125 |
orse | -0.124 |
fin | -0.121 |
Token | , |
Token ID | 11 |
Feature activation | +6.32e-03 |
Pos logit contributions | |
she | +0.773 |
She | +0.760 |
day | +0.749 |
always | +0.639 |
week | +0.638 |
Neg logit contributions | |
itched | -1.134 |
tal | -1.117 |
eg | -1.047 |
orse | -0.997 |
izza | -0.987 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
She | +0.008 |
day | +0.008 |
she | +0.008 |
especially | +0.007 |
sometimes | +0.007 |
Neg logit contributions | |
itched | -0.016 |
tal | -0.015 |
eg | -0.014 |
orse | -0.014 |
izza | -0.014 |
Token | knew |
Token ID | 2993 |
Feature activation | +5.06e-02 |
Pos logit contributions | |
She | +0.199 |
day | +0.196 |
she | +0.195 |
like | +0.173 |
weekend | +0.172 |
Neg logit contributions | |
itched | -0.295 |
tal | -0.282 |
eg | -0.265 |
orse | -0.253 |
izza | -0.245 |
Token | something |
Token ID | 1223 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
day | +0.043 |
She | +0.043 |
my | +0.036 |
weekend | +0.035 |
especially | +0.035 |
Neg logit contributions | |
itched | -0.146 |
tal | -0.142 |
eg | -0.130 |
orse | -0.130 |
ves | -0.129 |
Token | was |
Token ID | 373 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
tal | +0.218 |
itched | +0.215 |
ves | +0.197 |
eg | +0.187 |
ggy | +0.186 |
Neg logit contributions | |
She | -0.186 |
my | -0.169 |
day | -0.169 |
weekend | -0.162 |
she | -0.161 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
itched | +0.319 |
tal | +0.311 |
eg | +0.294 |
fin | +0.278 |
urger | +0.278 |
Neg logit contributions | |
She | -0.251 |
she | -0.241 |
day | -0.235 |
my | -0.209 |
week | -0.206 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
She | +0.043 |
she | +0.042 |
day | +0.042 |
like | +0.037 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.072 |
eg | -0.069 |
orse | -0.065 |
izza | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
itched | +0.089 |
tal | +0.085 |
eg | +0.081 |
urger | +0.075 |
fin | +0.074 |
Neg logit contributions | |
She | -0.081 |
day | -0.077 |
she | -0.076 |
my | -0.069 |
week | -0.068 |
Token | Later |
Token ID | 18602 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
itched | +0.114 |
tal | +0.109 |
eg | +0.103 |
orse | +0.099 |
izza | +0.096 |
Neg logit contributions | |
She | -0.073 |
day | -0.072 |
she | -0.072 |
like | -0.064 |
always | -0.063 |
Token | that |
Token ID | 326 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
She | +0.003 |
day | +0.003 |
she | +0.003 |
especially | +0.003 |
like | +0.003 |
Neg logit contributions | |
itched | -0.004 |
tal | -0.004 |
eg | -0.003 |
orse | -0.003 |
izza | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.012 |
she | +0.008 |
especially | +0.006 |
like | +0.005 |
sometimes | +0.003 |
Neg logit contributions | |
itched | -0.134 |
tal | -0.130 |
eg | -0.125 |
orse | -0.124 |
fin | -0.121 |
Token | , |
Token ID | 11 |
Feature activation | +6.32e-03 |
Pos logit contributions | |
she | +0.773 |
She | +0.760 |
day | +0.749 |
always | +0.639 |
week | +0.638 |
Neg logit contributions | |
itched | -1.134 |
tal | -1.117 |
eg | -1.047 |
orse | -0.997 |
izza | -0.987 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.22e-02 |
Pos logit contributions | |
She | +0.008 |
day | +0.008 |
she | +0.008 |
especially | +0.007 |
sometimes | +0.007 |
Neg logit contributions | |
itched | -0.016 |
tal | -0.015 |
eg | -0.014 |
orse | -0.014 |
izza | -0.014 |
Token | my |
Token ID | 1820 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
She | +0.030 |
she | +0.029 |
day | +0.024 |
weekend | +0.018 |
especially | +0.018 |
Neg logit contributions | |
itched | -0.167 |
tal | -0.163 |
eg | -0.157 |
orse | -0.152 |
fin | -0.149 |
Token | and |
Token ID | 290 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
She | +0.597 |
she | +0.591 |
day | +0.581 |
always | +0.534 |
my | +0.525 |
Neg logit contributions | |
itched | -1.014 |
tal | -0.979 |
eg | -0.928 |
izza | -0.886 |
ades | -0.875 |
Token | his |
Token ID | 465 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
She | +0.233 |
day | +0.228 |
she | +0.228 |
like | +0.207 |
always | +0.206 |
Neg logit contributions | |
itched | -0.253 |
tal | -0.242 |
eg | -0.226 |
orse | -0.213 |
izza | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
itched | +0.113 |
tal | +0.108 |
eg | +0.103 |
izza | +0.095 |
orse | +0.094 |
Neg logit contributions | |
day | -0.105 |
She | -0.103 |
she | -0.101 |
weekend | -0.096 |
week | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
She | +0.043 |
she | +0.042 |
day | +0.042 |
like | +0.037 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.072 |
eg | -0.069 |
orse | -0.065 |
izza | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
itched | +0.089 |
tal | +0.085 |
eg | +0.081 |
urger | +0.075 |
fin | +0.074 |
Neg logit contributions | |
She | -0.081 |
day | -0.077 |
she | -0.076 |
my | -0.069 |
week | -0.068 |
Token | Later |
Token ID | 18602 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
itched | +0.114 |
tal | +0.109 |
eg | +0.103 |
orse | +0.099 |
izza | +0.096 |
Neg logit contributions | |
She | -0.073 |
day | -0.072 |
she | -0.072 |
like | -0.064 |
always | -0.063 |
Token | that |
Token ID | 326 |
Feature activation | +3.84e-02 |
Pos logit contributions | |
She | +0.003 |
day | +0.003 |
she | +0.003 |
especially | +0.003 |
like | +0.003 |
Neg logit contributions | |
itched | -0.004 |
tal | -0.004 |
eg | -0.003 |
orse | -0.003 |
izza | -0.003 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.012 |
she | +0.008 |
especially | +0.006 |
like | +0.005 |
sometimes | +0.003 |
Neg logit contributions | |
itched | -0.134 |
tal | -0.130 |
eg | -0.125 |
orse | -0.124 |
fin | -0.121 |
Token | , |
Token ID | 11 |
Feature activation | +6.32e-03 |
Pos logit contributions | |
she | +0.773 |
She | +0.760 |
day | +0.749 |
always | +0.639 |
week | +0.638 |
Neg logit contributions | |
itched | -1.134 |
tal | -1.117 |
eg | -1.047 |
orse | -0.997 |
izza | -0.987 |
Token | when |
Token ID | 618 |
Feature activation | +2.62e-01 |
Pos logit contributions | |
She | +0.008 |
day | +0.008 |
she | +0.008 |
especially | +0.007 |
sometimes | +0.007 |
Neg logit contributions | |
itched | -0.016 |
tal | -0.015 |
eg | -0.014 |
orse | -0.014 |
izza | -0.014 |
Token | her |
Token ID | 607 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
She | +0.383 |
day | +0.361 |
she | +0.346 |
especially | +0.336 |
my | +0.334 |
Neg logit contributions | |
itched | -0.608 |
tal | -0.586 |
eg | -0.547 |
orse | -0.533 |
fin | -0.515 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
day | +0.709 |
she | +0.675 |
She | +0.642 |
always | +0.610 |
girl | +0.608 |
Neg logit contributions | |
itched | -0.757 |
tal | -0.737 |
eg | -0.665 |
cks | -0.651 |
orse | -0.642 |
Token | my |
Token ID | 1820 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
itched | +0.470 |
tal | +0.441 |
eg | +0.430 |
izza | +0.408 |
orse | +0.407 |
Neg logit contributions | |
day | -0.157 |
She | -0.150 |
she | -0.146 |
like | -0.123 |
weekend | -0.123 |
Token | answer |
Token ID | 3280 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
tal | +0.071 |
eg | +0.065 |
itched | +0.063 |
fin | +0.063 |
rods | +0.061 |
Neg logit contributions | |
she | -0.066 |
especially | -0.064 |
She | -0.064 |
weekend | -0.062 |
sometimes | -0.062 |
Token | questions |
Token ID | 2683 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
She | +0.245 |
day | +0.239 |
she | +0.237 |
always | +0.212 |
especially | +0.211 |
Neg logit contributions | |
itched | -0.347 |
tal | -0.334 |
eg | -0.311 |
orse | -0.297 |
izza | -0.291 |
Token | . |
Token ID | 13 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
She | +0.053 |
day | +0.052 |
she | +0.051 |
especially | +0.046 |
weekend | +0.046 |
Neg logit contributions | |
itched | -0.070 |
tal | -0.066 |
eg | -0.062 |
orse | -0.059 |
izza | -0.059 |
Token | From |
Token ID | 3574 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
She | +0.046 |
she | +0.044 |
day | +0.044 |
like | +0.038 |
sometimes | +0.038 |
Neg logit contributions | |
itched | -0.081 |
tal | -0.078 |
eg | -0.074 |
orse | -0.070 |
izza | -0.069 |
Token | that |
Token ID | 326 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
itched | +0.357 |
tal | +0.345 |
eg | +0.318 |
orse | +0.306 |
fin | +0.299 |
Neg logit contributions | |
She | -0.245 |
day | -0.235 |
she | -0.233 |
especially | -0.212 |
always | -0.208 |
Token | day |
Token ID | 1110 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
She | +0.017 |
she | +0.011 |
especially | +0.008 |
like | +0.006 |
She | +0.006 |
Neg logit contributions | |
itched | -0.153 |
tal | -0.151 |
eg | -0.143 |
orse | -0.141 |
izza | -0.140 |
Token | on |
Token ID | 319 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
She | +0.756 |
she | +0.741 |
day | +0.741 |
like | +0.655 |
always | +0.653 |
Neg logit contributions | |
itched | -1.128 |
tal | -1.088 |
eg | -1.028 |
orse | -0.976 |
izza | -0.959 |
Token | , |
Token ID | 11 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
She | +0.068 |
day | +0.066 |
she | +0.064 |
especially | +0.062 |
like | +0.062 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.071 |
eg | -0.064 |
orse | -0.061 |
ves | -0.061 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
She | +0.021 |
day | +0.020 |
weekend | +0.019 |
my | +0.018 |
ock | +0.018 |
Neg logit contributions | |
itched | -0.032 |
tal | -0.032 |
eg | -0.029 |
orse | -0.029 |
ves | -0.028 |
Token | my |
Token ID | 1820 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
She | +0.048 |
she | +0.045 |
weekend | +0.035 |
She | +0.033 |
day | +0.033 |
Neg logit contributions | |
itched | -0.140 |
tal | -0.137 |
eg | -0.130 |
orse | -0.126 |
urger | -0.121 |
Token | and |
Token ID | 290 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
She | +0.627 |
she | +0.604 |
day | +0.598 |
weekend | +0.553 |
week | +0.540 |
Neg logit contributions | |
itched | -0.941 |
tal | -0.897 |
eg | -0.840 |
orse | -0.812 |
fin | -0.778 |
Token | them |
Token ID | 606 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
day | +0.093 |
She | +0.093 |
she | +0.089 |
weekend | +0.085 |
week | +0.083 |
Neg logit contributions | |
itched | -0.130 |
tal | -0.123 |
eg | -0.120 |
orse | -0.112 |
izza | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
day | +0.079 |
she | +0.078 |
She | +0.076 |
like | +0.073 |
sometimes | +0.069 |
Neg logit contributions | |
itched | -0.133 |
tal | -0.128 |
eg | -0.123 |
orse | -0.114 |
izza | -0.113 |
Token |
|
Token ID | 198 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
She | +0.058 |
she | +0.055 |
day | +0.054 |
sometimes | +0.048 |
like | +0.048 |
Neg logit contributions | |
itched | -0.097 |
tal | -0.094 |
eg | -0.089 |
orse | -0.083 |
izza | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
itched | +0.106 |
tal | +0.102 |
eg | +0.097 |
urger | +0.092 |
fin | +0.090 |
Neg logit contributions | |
She | -0.084 |
day | -0.078 |
she | -0.078 |
my | -0.070 |
week | -0.069 |
Token | One |
Token ID | 3198 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
itched | +0.135 |
tal | +0.130 |
eg | +0.123 |
orse | +0.117 |
izza | +0.115 |
Neg logit contributions | |
She | -0.084 |
day | -0.083 |
she | -0.082 |
like | -0.073 |
always | -0.073 |
Token | day |
Token ID | 1110 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
She | +0.104 |
she | +0.068 |
my | +0.040 |
She | +0.040 |
especially | +0.039 |
Neg logit contributions | |
itched | -0.759 |
tal | -0.746 |
orse | -0.724 |
fin | -0.714 |
eg | -0.708 |
Token | , |
Token ID | 11 |
Feature activation | +9.63e-03 |
Pos logit contributions | |
she | +0.748 |
She | +0.703 |
day | +0.679 |
like | +0.611 |
her | +0.577 |
Neg logit contributions | |
itched | -1.219 |
tal | -1.194 |
eg | -1.122 |
orse | -1.058 |
wounds | -1.055 |
Token | they |
Token ID | 484 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
She | +0.015 |
day | +0.015 |
she | +0.014 |
like | +0.013 |
always | +0.013 |
Neg logit contributions | |
itched | -0.022 |
tal | -0.021 |
eg | -0.020 |
orse | -0.019 |
izza | -0.019 |
Token | see |
Token ID | 766 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
She | +0.075 |
day | +0.074 |
she | +0.073 |
weekend | +0.064 |
week | +0.064 |
Neg logit contributions | |
itched | -0.122 |
tal | -0.118 |
eg | -0.113 |
orse | -0.107 |
fin | -0.106 |
Token | a |
Token ID | 257 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
she | +0.029 |
day | +0.027 |
She | +0.027 |
like | +0.027 |
sometimes | +0.025 |
Neg logit contributions | |
itched | -0.055 |
eg | -0.053 |
tal | -0.052 |
urger | -0.048 |
orse | -0.048 |
Token | big |
Token ID | 1263 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
itched | +0.071 |
tal | +0.068 |
eg | +0.062 |
orse | +0.060 |
cks | +0.060 |
Neg logit contributions | |
day | -0.043 |
she | -0.041 |
She | -0.039 |
girl | -0.037 |
week | -0.036 |
Token | them |
Token ID | 606 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
She | +0.198 |
day | +0.190 |
she | +0.186 |
weekend | +0.175 |
especially | +0.172 |
Neg logit contributions | |
itched | -0.235 |
tal | -0.220 |
fin | -0.196 |
eg | -0.196 |
cannon | -0.191 |
Token | . |
Token ID | 13 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
day | +0.079 |
she | +0.078 |
She | +0.076 |
like | +0.073 |
sometimes | +0.069 |
Neg logit contributions | |
itched | -0.133 |
tal | -0.128 |
eg | -0.123 |
orse | -0.114 |
izza | -0.113 |
Token |
|
Token ID | 198 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
She | +0.058 |
she | +0.055 |
day | +0.054 |
sometimes | +0.048 |
like | +0.048 |
Neg logit contributions | |
itched | -0.097 |
tal | -0.094 |
eg | -0.089 |
orse | -0.083 |
izza | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
itched | +0.106 |
tal | +0.102 |
eg | +0.097 |
urger | +0.092 |
fin | +0.090 |
Neg logit contributions | |
She | -0.084 |
day | -0.078 |
she | -0.078 |
my | -0.070 |
week | -0.069 |
Token | One |
Token ID | 3198 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
itched | +0.135 |
tal | +0.130 |
eg | +0.123 |
orse | +0.117 |
izza | +0.115 |
Neg logit contributions | |
She | -0.084 |
day | -0.083 |
she | -0.082 |
like | -0.073 |
always | -0.073 |
Token | day |
Token ID | 1110 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
She | +0.104 |
she | +0.068 |
my | +0.040 |
She | +0.040 |
especially | +0.039 |
Neg logit contributions | |
itched | -0.759 |
tal | -0.746 |
orse | -0.724 |
fin | -0.714 |
eg | -0.708 |
Token | , |
Token ID | 11 |
Feature activation | +9.63e-03 |
Pos logit contributions | |
she | +0.748 |
She | +0.703 |
day | +0.679 |
like | +0.611 |
her | +0.577 |
Neg logit contributions | |
itched | -1.219 |
tal | -1.194 |
eg | -1.122 |
orse | -1.058 |
wounds | -1.055 |
Token | they |
Token ID | 484 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
She | +0.015 |
day | +0.015 |
she | +0.014 |
like | +0.013 |
always | +0.013 |
Neg logit contributions | |
itched | -0.022 |
tal | -0.021 |
eg | -0.020 |
orse | -0.019 |
izza | -0.019 |
Token | found |
Token ID | 1043 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
She | +0.075 |
day | +0.074 |
she | +0.073 |
weekend | +0.064 |
week | +0.064 |
Neg logit contributions | |
itched | -0.122 |
tal | -0.118 |
eg | -0.113 |
orse | -0.107 |
fin | -0.106 |
Token | something |
Token ID | 1223 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
she | +0.035 |
day | +0.034 |
She | +0.034 |
like | +0.032 |
her | +0.031 |
Neg logit contributions | |
itched | -0.072 |
eg | -0.067 |
tal | -0.065 |
orse | -0.065 |
anas | -0.063 |
Token | different |
Token ID | 1180 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
itched | +0.262 |
tal | +0.252 |
eg | +0.240 |
orse | +0.227 |
izza | +0.224 |
Neg logit contributions | |
she | -0.159 |
She | -0.159 |
day | -0.158 |
like | -0.141 |
always | -0.138 |
Token | and |
Token ID | 290 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
itched | +0.734 |
tal | +0.702 |
eg | +0.661 |
orse | +0.627 |
izza | +0.614 |
Neg logit contributions | |
She | -0.523 |
day | -0.511 |
she | -0.509 |
always | -0.453 |
like | -0.452 |
Token | his |
Token ID | 465 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
She | +0.117 |
day | +0.114 |
she | +0.113 |
like | +0.102 |
always | +0.101 |
Neg logit contributions | |
itched | -0.161 |
tal | -0.155 |
eg | -0.145 |
orse | -0.138 |
izza | -0.135 |
Token | dark |
Token ID | 3223 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
itched | +0.228 |
tal | +0.218 |
orse | +0.216 |
engu | +0.215 |
cks | +0.212 |
Neg logit contributions | |
day | -0.276 |
she | -0.241 |
She | -0.236 |
gloomy | -0.234 |
wind | -0.232 |
Token | mustache |
Token ID | 49303 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
day | +0.392 |
She | +0.390 |
she | +0.385 |
my | +0.363 |
always | +0.358 |
Neg logit contributions | |
itched | -0.287 |
tal | -0.274 |
eg | -0.272 |
izza | -0.267 |
engu | -0.250 |
Token | tw |
Token ID | 665 |
Feature activation | +6.80e-03 |
Pos logit contributions | |
tal | +0.497 |
itched | +0.485 |
eg | +0.470 |
izza | +0.454 |
orse | +0.445 |
Neg logit contributions | |
she | -0.309 |
She | -0.297 |
day | -0.295 |
always | -0.279 |
like | -0.264 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.79e-01 |
Pos logit contributions | |
She | +0.016 |
always | +0.015 |
she | +0.015 |
especially | +0.015 |
day | +0.015 |
Neg logit contributions | |
tal | -0.008 |
izza | -0.008 |
cannon | -0.008 |
rods | -0.007 |
eg | -0.007 |
Token | with |
Token ID | 351 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
itched | +1.523 |
tal | +1.475 |
eg | +1.401 |
orse | +1.338 |
izza | +1.307 |
Neg logit contributions | |
She | -1.119 |
she | -1.094 |
day | -1.093 |
like | -0.996 |
always | -0.978 |
Token | joy |
Token ID | 8716 |
Feature activation | -8.82e-02 |
Pos logit contributions | |
itched | +0.134 |
tal | +0.127 |
eg | +0.123 |
orse | +0.116 |
izza | +0.115 |
Neg logit contributions | |
She | -0.113 |
she | -0.112 |
day | -0.111 |
like | -0.102 |
always | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
itched | +0.198 |
tal | +0.195 |
eg | +0.184 |
orse | +0.174 |
izza | +0.172 |
Neg logit contributions | |
she | -0.145 |
She | -0.145 |
day | -0.141 |
like | -0.130 |
always | -0.127 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
itched | +0.119 |
tal | +0.114 |
eg | +0.107 |
orse | +0.101 |
izza | +0.099 |
Neg logit contributions | |
She | -0.100 |
she | -0.098 |
day | -0.098 |
always | -0.088 |
like | -0.087 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
tal | +0.278 |
itched | +0.274 |
jaws | +0.262 |
urger | +0.257 |
fins | +0.257 |
Neg logit contributions | |
day | -0.267 |
She | -0.266 |
she | -0.253 |
She | -0.227 |
always | -0.225 |
Token | She |
Token ID | 1375 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
itched | +0.167 |
tal | +0.158 |
eg | +0.150 |
izza | +0.148 |
rolled | +0.145 |
Neg logit contributions | |
day | -0.057 |
she | -0.049 |
especially | -0.049 |
weekend | -0.047 |
matter | -0.047 |
Token | st |
Token ID | 336 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
she | +0.622 |
day | +0.603 |
She | +0.592 |
always | +0.579 |
sometimes | +0.565 |
Neg logit contributions | |
itched | -0.959 |
tal | -0.943 |
eg | -0.917 |
izza | -0.866 |
urger | -0.831 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
tal | +0.135 |
orse | +0.129 |
fin | +0.118 |
urger | +0.118 |
cannon | +0.116 |
Neg logit contributions | |
She | -0.325 |
she | -0.281 |
my | -0.275 |
She | -0.272 |
her | -0.261 |
Token | and |
Token ID | 290 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
itched | +1.681 |
tal | +1.657 |
eg | +1.616 |
orse | +1.543 |
izza | +1.511 |
Neg logit contributions | |
She | -0.968 |
she | -0.958 |
day | -0.931 |
like | -0.847 |
sometimes | -0.834 |
Token | st |
Token ID | 336 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
she | +0.123 |
day | +0.120 |
She | +0.117 |
sometimes | +0.110 |
always | +0.109 |
Neg logit contributions | |
itched | -0.157 |
eg | -0.152 |
tal | -0.152 |
izza | -0.140 |
ades | -0.137 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
orse | +0.124 |
tal | +0.123 |
fin | +0.115 |
gru | +0.111 |
urger | +0.110 |
Neg logit contributions | |
She | -0.333 |
she | -0.292 |
like | -0.284 |
She | -0.280 |
her | -0.276 |
Token | until |
Token ID | 1566 |
Feature activation | +3.78e-03 |
Pos logit contributions | |
itched | +1.684 |
tal | +1.656 |
eg | +1.607 |
orse | +1.535 |
izza | +1.502 |
Neg logit contributions | |
She | -0.969 |
she | -0.955 |
day | -0.927 |
like | -0.850 |
sometimes | -0.825 |
Token | she |
Token ID | 673 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
She | +0.005 |
day | +0.004 |
my | +0.004 |
especially | +0.004 |
like | +0.004 |
Neg logit contributions | |
itched | -0.010 |
tal | -0.009 |
eg | -0.009 |
orse | -0.009 |
ves | -0.008 |
Token | had |
Token ID | 550 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
she | +0.573 |
day | +0.561 |
She | +0.551 |
always | +0.524 |
like | +0.510 |
Neg logit contributions | |
itched | -0.985 |
tal | -0.970 |
eg | -0.919 |
izza | -0.873 |
ades | -0.845 |
Token | a |
Token ID | 257 |
Feature activation | -8.95e-02 |
Pos logit contributions | |
She | +0.148 |
she | +0.148 |
day | +0.143 |
always | +0.135 |
like | +0.133 |
Neg logit contributions | |
itched | -0.212 |
tal | -0.200 |
eg | -0.196 |
orse | -0.185 |
izza | -0.179 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
itched | +0.221 |
tal | +0.212 |
eg | +0.204 |
orse | +0.195 |
izza | +0.192 |
Neg logit contributions | |
day | -0.132 |
she | -0.130 |
She | -0.128 |
always | -0.116 |
like | -0.115 |
Token | to |
Token ID | 284 |
Feature activation | -6.90e-02 |
Pos logit contributions | |
tal | +0.265 |
eg | +0.258 |
itched | +0.252 |
izza | +0.242 |
orse | +0.238 |
Neg logit contributions | |
she | -0.151 |
She | -0.144 |
day | -0.137 |
her | -0.128 |
always | -0.122 |
Token | stitch |
Token ID | 24695 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
itched | +0.155 |
tal | +0.149 |
eg | +0.141 |
orse | +0.133 |
izza | +0.131 |
Neg logit contributions | |
She | -0.115 |
she | -0.112 |
day | -0.112 |
like | -0.101 |
always | -0.100 |
Token | . |
Token ID | 13 |
Feature activation | -5.92e-02 |
Pos logit contributions | |
tal | +0.377 |
itched | +0.373 |
eg | +0.372 |
orse | +0.354 |
izza | +0.352 |
Neg logit contributions | |
she | -0.217 |
She | -0.201 |
day | -0.192 |
her | -0.187 |
sometimes | -0.183 |
Token | She |
Token ID | 1375 |
Feature activation | +4.05e-01 |
Pos logit contributions | |
itched | +0.167 |
tal | +0.158 |
eg | +0.150 |
izza | +0.148 |
rolled | +0.145 |
Neg logit contributions | |
day | -0.057 |
she | -0.049 |
especially | -0.049 |
weekend | -0.047 |
matter | -0.047 |
Token | st |
Token ID | 336 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
she | +0.622 |
day | +0.603 |
She | +0.592 |
always | +0.579 |
sometimes | +0.565 |
Neg logit contributions | |
itched | -0.959 |
tal | -0.943 |
eg | -0.917 |
izza | -0.866 |
urger | -0.831 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
tal | +0.135 |
orse | +0.129 |
fin | +0.118 |
urger | +0.118 |
cannon | +0.116 |
Neg logit contributions | |
She | -0.325 |
she | -0.281 |
my | -0.275 |
She | -0.272 |
her | -0.261 |
Token | and |
Token ID | 290 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
itched | +1.681 |
tal | +1.657 |
eg | +1.616 |
orse | +1.543 |
izza | +1.511 |
Neg logit contributions | |
She | -0.968 |
she | -0.958 |
day | -0.931 |
like | -0.847 |
sometimes | -0.834 |
Token | st |
Token ID | 336 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
she | +0.123 |
day | +0.120 |
She | +0.117 |
sometimes | +0.110 |
always | +0.109 |
Neg logit contributions | |
itched | -0.157 |
eg | -0.152 |
tal | -0.152 |
izza | -0.140 |
ades | -0.137 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
orse | +0.124 |
tal | +0.123 |
fin | +0.115 |
gru | +0.111 |
urger | +0.110 |
Neg logit contributions | |
She | -0.333 |
she | -0.292 |
like | -0.284 |
She | -0.280 |
her | -0.276 |
Token | until |
Token ID | 1566 |
Feature activation | +3.78e-03 |
Pos logit contributions | |
itched | +1.684 |
tal | +1.656 |
eg | +1.607 |
orse | +1.535 |
izza | +1.502 |
Neg logit contributions | |
She | -0.969 |
she | -0.955 |
day | -0.927 |
like | -0.850 |
sometimes | -0.825 |
Token | she |
Token ID | 673 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
She | +0.005 |
day | +0.004 |
my | +0.004 |
especially | +0.004 |
like | +0.004 |
Neg logit contributions | |
itched | -0.010 |
tal | -0.009 |
eg | -0.009 |
orse | -0.009 |
ves | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
itched | +0.210 |
tal | +0.202 |
eg | +0.196 |
orse | +0.186 |
izza | +0.185 |
Neg logit contributions | |
She | -0.113 |
she | -0.112 |
day | -0.110 |
like | -0.098 |
always | -0.096 |
Token | thread |
Token ID | 4704 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
she | +0.122 |
She | +0.118 |
day | +0.116 |
sometimes | +0.104 |
always | +0.103 |
Neg logit contributions | |
itched | -0.166 |
tal | -0.157 |
eg | -0.154 |
orse | -0.143 |
izza | -0.142 |
Token | and |
Token ID | 290 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
itched | +0.046 |
tal | +0.045 |
eg | +0.044 |
orse | +0.041 |
izza | +0.041 |
Neg logit contributions | |
she | -0.031 |
She | -0.031 |
day | -0.030 |
like | -0.027 |
always | -0.027 |
Token | carefully |
Token ID | 7773 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
She | +0.122 |
she | +0.120 |
day | +0.120 |
like | +0.107 |
always | +0.107 |
Neg logit contributions | |
itched | -0.162 |
tal | -0.155 |
eg | -0.147 |
orse | -0.138 |
izza | -0.136 |
Token | st |
Token ID | 336 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
eg | +0.361 |
tal | +0.353 |
itched | +0.344 |
izza | +0.327 |
orse | +0.311 |
Neg logit contributions | |
she | -0.329 |
She | -0.301 |
like | -0.291 |
day | -0.289 |
her | -0.284 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
orse | +0.121 |
eg | +0.115 |
tal | +0.112 |
cannon | +0.111 |
urger | +0.108 |
Neg logit contributions | |
She | -0.319 |
she | -0.293 |
her | -0.272 |
She | -0.272 |
like | -0.263 |
Token | it |
Token ID | 340 |
Feature activation | -7.62e-02 |
Pos logit contributions | |
itched | +1.679 |
tal | +1.642 |
eg | +1.618 |
orse | +1.555 |
izza | +1.519 |
Neg logit contributions | |
She | -0.966 |
she | -0.959 |
day | -0.911 |
like | -0.848 |
sometimes | -0.805 |
Token | to |
Token ID | 284 |
Feature activation | -6.75e-02 |
Pos logit contributions | |
itched | +0.165 |
tal | +0.160 |
eg | +0.151 |
orse | +0.143 |
izza | +0.140 |
Neg logit contributions | |
She | -0.131 |
she | -0.129 |
day | -0.128 |
like | -0.116 |
always | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
itched | +0.161 |
tal | +0.154 |
eg | +0.146 |
orse | +0.139 |
izza | +0.137 |
Neg logit contributions | |
She | -0.103 |
she | -0.101 |
day | -0.100 |
like | -0.089 |
always | -0.089 |
Token | fabric |
Token ID | 9664 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
itched | +0.076 |
tal | +0.074 |
eg | +0.067 |
fin | +0.066 |
wounds | +0.063 |
Neg logit contributions | |
She | -0.052 |
she | -0.050 |
day | -0.049 |
especially | -0.046 |
my | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -5.87e-02 |
Pos logit contributions | |
She | +0.097 |
day | +0.095 |
she | +0.094 |
especially | +0.086 |
always | +0.085 |
Neg logit contributions | |
itched | -0.117 |
tal | -0.111 |
eg | -0.102 |
orse | -0.098 |
fin | -0.095 |
Token | off |
Token ID | 572 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
itched | +0.016 |
tal | +0.016 |
eg | +0.015 |
orse | +0.015 |
fin | +0.014 |
Neg logit contributions | |
she | -0.011 |
She | -0.011 |
day | -0.011 |
like | -0.010 |
her | -0.009 |
Token | some |
Token ID | 617 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
itched | +0.480 |
tal | +0.473 |
eg | +0.457 |
orse | +0.443 |
fin | +0.429 |
Neg logit contributions | |
she | -0.270 |
She | -0.264 |
day | -0.252 |
her | -0.228 |
like | -0.223 |
Token | fabric |
Token ID | 9664 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
itched | +0.261 |
tal | +0.251 |
eg | +0.242 |
orse | +0.228 |
fin | +0.219 |
Neg logit contributions | |
day | -0.210 |
she | -0.209 |
She | -0.207 |
like | -0.189 |
always | -0.183 |
Token | and |
Token ID | 290 |
Feature activation | +7.54e-02 |
Pos logit contributions | |
she | +0.073 |
She | +0.072 |
day | +0.070 |
like | +0.062 |
sometimes | +0.060 |
Neg logit contributions | |
itched | -0.155 |
tal | -0.151 |
eg | -0.144 |
orse | -0.138 |
izza | -0.136 |
Token | st |
Token ID | 336 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
she | +0.130 |
She | +0.126 |
day | +0.125 |
always | +0.113 |
like | +0.113 |
Neg logit contributions | |
itched | -0.169 |
tal | -0.162 |
eg | -0.156 |
orse | -0.144 |
fin | -0.142 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
tal | +0.139 |
fin | +0.137 |
orse | +0.131 |
gru | +0.131 |
rifle | +0.130 |
Neg logit contributions | |
She | -0.305 |
she | -0.274 |
She | -0.251 |
her | -0.249 |
like | -0.248 |
Token | it |
Token ID | 340 |
Feature activation | -7.40e-02 |
Pos logit contributions | |
eg | +1.736 |
tal | +1.730 |
itched | +1.702 |
orse | +1.670 |
urger | +1.589 |
Neg logit contributions | |
she | -0.950 |
She | -0.934 |
day | -0.879 |
her | -0.859 |
like | -0.844 |
Token | around |
Token ID | 1088 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
itched | +0.167 |
tal | +0.164 |
eg | +0.155 |
orse | +0.149 |
izza | +0.144 |
Neg logit contributions | |
She | -0.122 |
she | -0.120 |
day | -0.119 |
like | -0.108 |
always | -0.106 |
Token | the |
Token ID | 262 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
She | +0.141 |
day | +0.138 |
she | +0.138 |
like | +0.122 |
always | +0.121 |
Neg logit contributions | |
itched | -0.223 |
tal | -0.214 |
eg | -0.203 |
orse | -0.193 |
izza | -0.189 |
Token | waist |
Token ID | 16139 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
itched | +0.066 |
tal | +0.065 |
eg | +0.061 |
fin | +0.056 |
orse | +0.055 |
Neg logit contributions | |
She | -0.058 |
she | -0.056 |
day | -0.054 |
sometimes | -0.052 |
especially | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
itched | +0.297 |
tal | +0.279 |
eg | +0.263 |
orse | +0.249 |
izza | +0.242 |
Neg logit contributions | |
She | -0.252 |
day | -0.247 |
she | -0.244 |
sometimes | -0.222 |
always | -0.221 |
Token | needle |
Token ID | 17598 |
Feature activation | -8.12e-02 |
Pos logit contributions | |
itched | +0.173 |
tal | +0.161 |
eg | +0.156 |
orse | +0.150 |
izza | +0.149 |
Neg logit contributions | |
day | -0.141 |
she | -0.139 |
She | -0.137 |
girl | -0.128 |
my | -0.126 |
Token | and |
Token ID | 290 |
Feature activation | +8.04e-02 |
Pos logit contributions | |
itched | +0.201 |
tal | +0.192 |
eg | +0.179 |
orse | +0.169 |
fin | +0.169 |
Neg logit contributions | |
She | -0.115 |
day | -0.114 |
she | -0.110 |
weekend | -0.101 |
especially | -0.099 |
Token | thread |
Token ID | 4704 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
she | +0.149 |
She | +0.135 |
day | +0.131 |
sometimes | +0.125 |
her | +0.124 |
Neg logit contributions | |
itched | -0.182 |
eg | -0.174 |
tal | -0.169 |
izza | -0.159 |
ades | -0.158 |
Token | and |
Token ID | 290 |
Feature activation | +8.00e-02 |
Pos logit contributions | |
itched | +0.033 |
tal | +0.033 |
eg | +0.032 |
orse | +0.030 |
izza | +0.030 |
Neg logit contributions | |
she | -0.022 |
She | -0.022 |
day | -0.021 |
like | -0.019 |
sometimes | -0.019 |
Token | st |
Token ID | 336 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
she | +0.139 |
She | +0.137 |
day | +0.134 |
sometimes | +0.123 |
always | +0.122 |
Neg logit contributions | |
itched | -0.177 |
tal | -0.169 |
eg | -0.162 |
orse | -0.150 |
izza | -0.149 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
orse | +0.128 |
crew | +0.127 |
tal | +0.121 |
cannon | +0.118 |
fin | +0.114 |
Neg logit contributions | |
She | -0.289 |
she | -0.253 |
She | -0.246 |
her | -0.239 |
my | -0.236 |
Token | up |
Token ID | 510 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
itched | +1.840 |
tal | +1.820 |
eg | +1.794 |
orse | +1.748 |
izza | +1.705 |
Neg logit contributions | |
She | -0.825 |
she | -0.810 |
day | -0.734 |
like | -0.700 |
her | -0.680 |
Token | her |
Token ID | 607 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
itched | +0.088 |
tal | +0.085 |
eg | +0.082 |
orse | +0.079 |
izza | +0.078 |
Neg logit contributions | |
She | -0.051 |
she | -0.050 |
day | -0.048 |
like | -0.044 |
sometimes | -0.043 |
Token | knee |
Token ID | 10329 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
day | +0.622 |
she | +0.594 |
She | +0.567 |
like | +0.533 |
always | +0.521 |
Neg logit contributions | |
itched | -0.645 |
tal | -0.612 |
eg | -0.591 |
izza | -0.576 |
orse | -0.557 |
Token | . |
Token ID | 13 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
itched | +0.586 |
tal | +0.577 |
eg | +0.550 |
orse | +0.527 |
izza | +0.525 |
Neg logit contributions | |
She | -0.409 |
she | -0.409 |
day | -0.398 |
like | -0.369 |
always | -0.360 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
itched | +0.140 |
tal | +0.136 |
eg | +0.129 |
orse | +0.124 |
fin | +0.121 |
Neg logit contributions | |
She | -0.058 |
she | -0.055 |
day | -0.053 |
always | -0.045 |
like | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
itched | +0.391 |
tal | +0.380 |
eg | +0.352 |
fin | +0.336 |
orse | +0.333 |
Neg logit contributions | |
She | -0.220 |
day | -0.214 |
she | -0.212 |
weekend | -0.198 |
always | -0.194 |
Token | and |
Token ID | 290 |
Feature activation | +7.96e-02 |
Pos logit contributions | |
itched | +0.112 |
tal | +0.108 |
eg | +0.102 |
orse | +0.099 |
fin | +0.096 |
Neg logit contributions | |
She | -0.053 |
day | -0.052 |
she | -0.051 |
sometimes | -0.044 |
weekend | -0.044 |
Token | his |
Token ID | 465 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
she | +0.126 |
She | +0.112 |
day | +0.110 |
like | +0.107 |
always | +0.104 |
Neg logit contributions | |
itched | -0.204 |
eg | -0.193 |
tal | -0.193 |
ades | -0.180 |
izza | -0.175 |
Token | nose |
Token ID | 9686 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
itched | +0.206 |
cks | +0.190 |
ades | +0.186 |
ves | +0.183 |
ues | +0.180 |
Neg logit contributions | |
day | -0.279 |
She | -0.240 |
she | -0.238 |
night | -0.237 |
morning | -0.237 |
Token | tw |
Token ID | 665 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
tal | +0.727 |
itched | +0.714 |
eg | +0.699 |
izza | +0.662 |
orse | +0.652 |
Neg logit contributions | |
she | -0.463 |
She | -0.456 |
day | -0.424 |
like | -0.411 |
always | -0.405 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
She | +0.036 |
she | +0.035 |
always | +0.034 |
especially | +0.034 |
like | +0.034 |
Neg logit contributions | |
crew | -0.011 |
cannon | -0.010 |
tal | -0.010 |
izza | -0.010 |
eg | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
eg | +1.654 |
tal | +1.627 |
itched | +1.617 |
orse | +1.590 |
izza | +1.562 |
Neg logit contributions | |
like | -0.982 |
She | -0.962 |
she | -0.920 |
day | -0.916 |
her | -0.875 |
Token | He |
Token ID | 679 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
itched | +0.132 |
tal | +0.129 |
ades | +0.124 |
eg | +0.122 |
wounds | +0.121 |
Neg logit contributions | |
She | -0.071 |
she | -0.056 |
day | -0.051 |
always | -0.044 |
morning | -0.043 |
Token | had |
Token ID | 550 |
Feature activation | +9.75e-02 |
Pos logit contributions | |
she | +0.066 |
She | +0.062 |
always | +0.062 |
day | +0.061 |
especially | +0.060 |
Neg logit contributions | |
itched | -0.111 |
tal | -0.108 |
eg | -0.104 |
izza | -0.098 |
ades | -0.097 |
Token | never |
Token ID | 1239 |
Feature activation | +5.50e-02 |
Pos logit contributions | |
She | +0.146 |
she | +0.146 |
day | +0.143 |
always | +0.137 |
like | +0.136 |
Neg logit contributions | |
itched | -0.240 |
eg | -0.223 |
tal | -0.222 |
orse | -0.207 |
izza | -0.205 |
Token | tasted |
Token ID | 29187 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
She | +0.098 |
she | +0.094 |
day | +0.094 |
sometimes | +0.084 |
always | +0.084 |
Neg logit contributions | |
itched | -0.115 |
tal | -0.112 |
eg | -0.102 |
orse | -0.099 |
izza | -0.095 |
Token | icked |
Token ID | 9484 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
itched | +0.044 |
tal | +0.043 |
eg | +0.041 |
izza | +0.034 |
orse | +0.033 |
Neg logit contributions | |
She | -0.136 |
she | -0.132 |
day | -0.127 |
like | -0.125 |
always | -0.123 |
Token | his |
Token ID | 465 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
itched | +1.262 |
tal | +1.200 |
eg | +1.150 |
orse | +1.082 |
fin | +1.074 |
Neg logit contributions | |
day | -0.615 |
She | -0.598 |
she | -0.582 |
weekend | -0.531 |
always | -0.521 |
Token | lips |
Token ID | 11914 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
itched | +0.204 |
izza | +0.184 |
uers | +0.181 |
plug | +0.180 |
eg | +0.180 |
Neg logit contributions | |
day | -0.246 |
She | -0.243 |
she | -0.240 |
girl | -0.238 |
wind | -0.236 |
Token | and |
Token ID | 290 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
itched | +1.054 |
tal | +0.979 |
eg | +0.932 |
orse | +0.895 |
fin | +0.879 |
Neg logit contributions | |
day | -0.633 |
She | -0.632 |
she | -0.614 |
weekend | -0.574 |
week | -0.554 |
Token | tw |
Token ID | 665 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
she | +0.147 |
She | +0.136 |
day | +0.130 |
like | +0.128 |
always | +0.126 |
Neg logit contributions | |
itched | -0.169 |
eg | -0.166 |
tal | -0.161 |
izza | -0.155 |
ades | -0.152 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
She | +0.029 |
she | +0.029 |
always | +0.028 |
day | +0.028 |
her | +0.028 |
Neg logit contributions | |
izza | -0.014 |
eg | -0.014 |
tal | -0.014 |
cannon | -0.013 |
rods | -0.012 |
Token | his |
Token ID | 465 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
tal | +1.588 |
itched | +1.571 |
eg | +1.556 |
izza | +1.493 |
orse | +1.486 |
Neg logit contributions | |
She | -1.065 |
she | -1.032 |
like | -0.974 |
her | -0.972 |
day | -0.955 |
Token | tail |
Token ID | 7894 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
itched | +0.214 |
izza | +0.202 |
ades | +0.196 |
eg | +0.196 |
tal | +0.195 |
Neg logit contributions | |
day | -0.246 |
She | -0.239 |
she | -0.234 |
like | -0.228 |
wind | -0.225 |
Token | . |
Token ID | 13 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
itched | +0.574 |
tal | +0.564 |
eg | +0.540 |
orse | +0.508 |
izza | +0.505 |
Neg logit contributions | |
She | -0.392 |
she | -0.386 |
day | -0.377 |
like | -0.355 |
sometimes | -0.342 |
Token | He |
Token ID | 679 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
itched | +0.129 |
tal | +0.126 |
eg | +0.118 |
orse | +0.113 |
izza | +0.110 |
Neg logit contributions | |
She | -0.074 |
she | -0.069 |
day | -0.066 |
like | -0.059 |
always | -0.058 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
she | +0.069 |
She | +0.066 |
day | +0.064 |
always | +0.063 |
especially | +0.062 |
Neg logit contributions | |
itched | -0.106 |
tal | -0.103 |
eg | -0.098 |
izza | -0.094 |
ades | -0.089 |
Token | them |
Token ID | 606 |
Feature activation | -4.41e-02 |
Pos logit contributions | |
She | +0.047 |
day | +0.047 |
she | +0.046 |
always | +0.042 |
especially | +0.042 |
Neg logit contributions | |
itched | -0.071 |
tal | -0.068 |
eg | -0.063 |
orse | -0.060 |
fin | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
itched | +0.106 |
tal | +0.105 |
eg | +0.099 |
orse | +0.093 |
izza | +0.092 |
Neg logit contributions | |
She | -0.069 |
she | -0.067 |
day | -0.065 |
like | -0.060 |
sometimes | -0.059 |
Token | Then |
Token ID | 3244 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
itched | +0.117 |
tal | +0.112 |
eg | +0.107 |
orse | +0.102 |
ves | +0.101 |
Neg logit contributions | |
She | -0.079 |
she | -0.071 |
day | -0.066 |
like | -0.062 |
sometimes | -0.060 |
Token | it |
Token ID | 340 |
Feature activation | -7.50e-02 |
Pos logit contributions | |
itched | +0.412 |
tal | +0.397 |
eg | +0.375 |
orse | +0.358 |
izza | +0.351 |
Neg logit contributions | |
She | -0.245 |
day | -0.240 |
she | -0.237 |
like | -0.210 |
always | -0.210 |
Token | tw |
Token ID | 665 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
itched | +0.178 |
tal | +0.173 |
eg | +0.172 |
izza | +0.165 |
ades | +0.155 |
Neg logit contributions | |
she | -0.125 |
day | -0.116 |
She | -0.115 |
always | -0.111 |
sometimes | -0.107 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
She | +0.026 |
she | +0.025 |
day | +0.024 |
always | +0.024 |
her | +0.024 |
Neg logit contributions | |
izza | -0.015 |
eg | -0.015 |
tal | -0.014 |
cannon | -0.014 |
urger | -0.014 |
Token | its |
Token ID | 663 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
itched | +1.511 |
tal | +1.476 |
eg | +1.434 |
orse | +1.378 |
izza | +1.348 |
Neg logit contributions | |
She | -1.132 |
she | -1.096 |
day | -1.080 |
like | -1.027 |
sometimes | -0.978 |
Token | nose |
Token ID | 9686 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
itched | +0.441 |
eg | +0.409 |
izza | +0.400 |
ades | +0.399 |
tal | +0.394 |
Neg logit contributions | |
day | -0.348 |
She | -0.333 |
she | -0.319 |
like | -0.312 |
always | -0.309 |
Token | and |
Token ID | 290 |
Feature activation | +7.82e-02 |
Pos logit contributions | |
itched | +0.725 |
tal | +0.708 |
eg | +0.673 |
orse | +0.636 |
izza | +0.628 |
Neg logit contributions | |
She | -0.472 |
she | -0.464 |
day | -0.456 |
like | -0.415 |
sometimes | -0.406 |
Token | it |
Token ID | 340 |
Feature activation | -7.57e-02 |
Pos logit contributions | |
she | +0.142 |
She | +0.134 |
day | +0.133 |
always | +0.120 |
like | +0.120 |
Neg logit contributions | |
itched | -0.173 |
tal | -0.162 |
eg | -0.160 |
izza | -0.151 |
ades | -0.147 |
Token | seemed |
Token ID | 3947 |
Feature activation | -5.25e-02 |
Pos logit contributions | |
itched | +0.180 |
tal | +0.175 |
eg | +0.171 |
izza | +0.164 |
orse | +0.155 |
Neg logit contributions | |
she | -0.122 |
day | -0.115 |
She | -0.114 |
always | -0.108 |
sometimes | -0.106 |
Token | hungry |
Token ID | 14720 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
itched | +0.150 |
tal | +0.144 |
fin | +0.131 |
orse | +0.125 |
fins | +0.123 |
Neg logit contributions | |
She | -0.109 |
day | -0.108 |
weekend | -0.106 |
she | -0.103 |
You | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
tal | +0.030 |
itched | +0.030 |
fin | +0.027 |
orse | +0.025 |
eg | +0.025 |
Neg logit contributions | |
She | -0.036 |
day | -0.035 |
weekend | -0.034 |
she | -0.033 |
my | -0.033 |
Token | His |
Token ID | 2399 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
itched | +0.135 |
tal | +0.129 |
eg | +0.123 |
orse | +0.119 |
ades | +0.116 |
Neg logit contributions | |
She | -0.070 |
she | -0.064 |
day | -0.062 |
like | -0.055 |
always | -0.055 |
Token | nose |
Token ID | 9686 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
itched | +0.551 |
tal | +0.518 |
eg | +0.496 |
ades | +0.475 |
orse | +0.475 |
Neg logit contributions | |
day | -0.572 |
she | -0.527 |
She | -0.527 |
always | -0.495 |
morning | -0.483 |
Token | tw |
Token ID | 665 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
itched | +0.697 |
tal | +0.674 |
eg | +0.640 |
orse | +0.602 |
izza | +0.595 |
Neg logit contributions | |
She | -0.492 |
she | -0.485 |
day | -0.478 |
always | -0.430 |
like | -0.429 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
She | +0.036 |
she | +0.034 |
day | +0.033 |
especially | +0.033 |
always | +0.033 |
Neg logit contributions | |
eg | -0.013 |
tal | -0.013 |
izza | -0.012 |
cannon | -0.012 |
itched | -0.011 |
Token | as |
Token ID | 355 |
Feature activation | +1.49e-03 |
Pos logit contributions | |
itched | +1.660 |
tal | +1.608 |
eg | +1.543 |
orse | +1.473 |
izza | +1.442 |
Neg logit contributions | |
She | -0.957 |
day | -0.933 |
she | -0.931 |
like | -0.840 |
always | -0.821 |
Token | he |
Token ID | 339 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
She | +0.002 |
day | +0.002 |
she | +0.002 |
my | +0.002 |
weekend | +0.002 |
Neg logit contributions | |
itched | -0.003 |
tal | -0.003 |
eg | -0.003 |
orse | -0.003 |
ves | -0.003 |
Token | smelled |
Token ID | 37609 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
she | +0.051 |
She | +0.048 |
day | +0.047 |
always | +0.046 |
especially | +0.045 |
Neg logit contributions | |
itched | -0.081 |
tal | -0.078 |
eg | -0.078 |
izza | -0.073 |
ades | -0.071 |
Token | something |
Token ID | 1223 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
itched | +0.028 |
tal | +0.027 |
eg | +0.025 |
orse | +0.024 |
izza | +0.024 |
Neg logit contributions | |
She | -0.018 |
day | -0.017 |
she | -0.017 |
always | -0.015 |
sometimes | -0.015 |
Token | in |
Token ID | 287 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
itched | +0.362 |
tal | +0.347 |
eg | +0.322 |
orse | +0.307 |
fin | +0.301 |
Neg logit contributions | |
She | -0.267 |
day | -0.260 |
she | -0.255 |
sometimes | -0.232 |
weekend | -0.232 |
Token | loud |
Token ID | 7812 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
itched | +0.205 |
tal | +0.195 |
uffed | +0.181 |
izza | +0.178 |
rolled | +0.178 |
Neg logit contributions | |
day | -0.114 |
girl | -0.102 |
She | -0.099 |
she | -0.098 |
gloomy | -0.095 |
Token | , |
Token ID | 11 |
Feature activation | -4.16e-02 |
Pos logit contributions | |
itched | +0.225 |
tal | +0.224 |
eg | +0.215 |
izza | +0.208 |
rolled | +0.201 |
Neg logit contributions | |
She | -0.135 |
day | -0.133 |
she | -0.130 |
like | -0.121 |
always | -0.114 |
Token | high |
Token ID | 1029 |
Feature activation | +5.54e-03 |
Pos logit contributions | |
itched | +0.101 |
tal | +0.096 |
eg | +0.092 |
orse | +0.086 |
izza | +0.086 |
Neg logit contributions | |
She | -0.062 |
she | -0.061 |
day | -0.061 |
like | -0.055 |
always | -0.053 |
Token | - |
Token ID | 12 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
day | +0.010 |
like | +0.008 |
She | +0.008 |
she | +0.008 |
sometimes | +0.008 |
Neg logit contributions | |
itched | -0.012 |
tal | -0.012 |
izza | -0.011 |
eg | -0.011 |
orse | -0.011 |
Token | p |
Token ID | 79 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
itched | +0.372 |
tal | +0.358 |
eg | +0.330 |
orse | +0.308 |
izza | +0.305 |
Neg logit contributions | |
She | -0.336 |
day | -0.332 |
she | -0.332 |
like | -0.303 |
sometimes | -0.298 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
tal | +0.319 |
eg | +0.299 |
plug | +0.292 |
izza | +0.292 |
itched | +0.279 |
Neg logit contributions | |
she | -0.197 |
like | -0.196 |
She | -0.195 |
day | -0.184 |
always | -0.175 |
Token | noise |
Token ID | 7838 |
Feature activation | -6.52e-03 |
Pos logit contributions | |
tal | +1.470 |
eg | +1.439 |
itched | +1.412 |
izza | +1.400 |
orse | +1.341 |
Neg logit contributions | |
She | -1.110 |
like | -1.077 |
day | -1.075 |
she | -1.044 |
her | -0.992 |
Token | back |
Token ID | 736 |
Feature activation | -7.61e-02 |
Pos logit contributions | |
itched | +0.016 |
tal | +0.016 |
eg | +0.015 |
izza | +0.014 |
orse | +0.014 |
Neg logit contributions | |
She | -0.010 |
she | -0.010 |
day | -0.009 |
like | -0.009 |
sometimes | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
itched | +0.187 |
tal | +0.185 |
eg | +0.178 |
orse | +0.170 |
izza | +0.169 |
Neg logit contributions | |
She | -0.112 |
she | -0.111 |
day | -0.108 |
like | -0.101 |
sometimes | -0.097 |
Token |
|
Token ID | 198 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
itched | +0.133 |
tal | +0.131 |
eg | +0.123 |
urger | +0.121 |
rolled | +0.116 |
Neg logit contributions | |
She | -0.080 |
she | -0.069 |
like | -0.060 |
day | -0.057 |
sometimes | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
tal | +0.377 |
itched | +0.369 |
urger | +0.368 |
eg | +0.351 |
rifle | +0.345 |
Neg logit contributions | |
She | -0.172 |
she | -0.145 |
day | -0.127 |
She | -0.125 |
sometimes | -0.114 |
Token | forest |
Token ID | 8222 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
tal | +0.055 |
itched | +0.053 |
fin | +0.048 |
eg | +0.048 |
fins | +0.045 |
Neg logit contributions | |
she | -0.032 |
like | -0.030 |
She | -0.030 |
especially | -0.029 |
sometimes | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
she | +0.261 |
She | +0.261 |
day | +0.259 |
like | +0.231 |
always | +0.228 |
Neg logit contributions | |
itched | -0.382 |
tal | -0.378 |
eg | -0.366 |
izza | -0.340 |
orse | -0.328 |
Token | The |
Token ID | 383 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
itched | +0.109 |
tal | +0.105 |
eg | +0.098 |
orse | +0.095 |
izza | +0.093 |
Neg logit contributions | |
day | -0.065 |
she | -0.063 |
sometimes | -0.059 |
She | -0.059 |
especially | -0.058 |
Token | bunny |
Token ID | 44915 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
fin | +0.246 |
itched | +0.242 |
tal | +0.240 |
eg | +0.225 |
fins | +0.223 |
Neg logit contributions | |
She | -0.270 |
she | -0.265 |
like | -0.231 |
weekend | -0.229 |
reating | -0.229 |
Token | tw |
Token ID | 665 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
She | +0.069 |
she | +0.066 |
day | +0.066 |
like | +0.059 |
weekend | +0.058 |
Neg logit contributions | |
itched | -0.113 |
tal | -0.109 |
eg | -0.102 |
orse | -0.100 |
fin | -0.096 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
She | +0.049 |
day | +0.048 |
she | +0.048 |
always | +0.045 |
especially | +0.045 |
Neg logit contributions | |
itched | -0.020 |
tal | -0.020 |
eg | -0.019 |
izza | -0.017 |
cannon | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
itched | +1.431 |
tal | +1.384 |
eg | +1.377 |
izza | +1.313 |
orse | +1.293 |
Neg logit contributions | |
She | -1.183 |
day | -1.167 |
she | -1.151 |
like | -1.072 |
always | -1.057 |
Token | nose |
Token ID | 9686 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
itched | +0.227 |
tal | +0.207 |
eg | +0.201 |
izza | +0.198 |
orse | +0.191 |
Neg logit contributions | |
day | -0.232 |
She | -0.218 |
she | -0.216 |
always | -0.205 |
like | -0.203 |
Token | , |
Token ID | 11 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
itched | +0.758 |
tal | +0.737 |
eg | +0.701 |
orse | +0.665 |
izza | +0.659 |
Neg logit contributions | |
She | -0.418 |
she | -0.409 |
day | -0.406 |
like | -0.358 |
always | -0.357 |
Token | making |
Token ID | 1642 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
itched | +0.083 |
tal | +0.080 |
fin | +0.073 |
orse | +0.072 |
fins | +0.069 |
Neg logit contributions | |
She | -0.068 |
day | -0.064 |
she | -0.062 |
weekend | -0.061 |
week | -0.059 |
Token | tiny |
Token ID | 7009 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
itched | +0.034 |
eg | +0.033 |
tal | +0.033 |
izza | +0.031 |
orse | +0.030 |
Neg logit contributions | |
she | -0.022 |
day | -0.022 |
She | -0.022 |
like | -0.020 |
always | -0.020 |
Token | big |
Token ID | 1263 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
itched | +0.099 |
eg | +0.097 |
tal | +0.094 |
izza | +0.092 |
orse | +0.091 |
Neg logit contributions | |
She | -0.088 |
she | -0.086 |
like | -0.080 |
day | -0.079 |
her | -0.076 |
Token | eyes |
Token ID | 2951 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
itched | +0.042 |
plug | +0.042 |
izza | +0.042 |
ades | +0.042 |
eg | +0.041 |
Neg logit contributions | |
She | -0.077 |
she | -0.072 |
day | -0.071 |
like | -0.071 |
her | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
tal | +0.754 |
itched | +0.748 |
eg | +0.714 |
izza | +0.666 |
orse | +0.664 |
Neg logit contributions | |
She | -0.525 |
she | -0.518 |
day | -0.488 |
like | -0.470 |
sometimes | -0.457 |
Token | It |
Token ID | 632 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
itched | +0.105 |
tal | +0.102 |
eg | +0.096 |
orse | +0.092 |
izza | +0.090 |
Neg logit contributions | |
She | -0.066 |
she | -0.064 |
day | -0.062 |
like | -0.056 |
sometimes | -0.055 |
Token | tw |
Token ID | 665 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
itched | +0.110 |
tal | +0.107 |
eg | +0.103 |
izza | +0.100 |
ades | +0.095 |
Neg logit contributions | |
she | -0.068 |
She | -0.068 |
day | -0.065 |
like | -0.062 |
sometimes | -0.061 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.69e-01 |
Pos logit contributions | |
She | +0.049 |
she | +0.048 |
like | +0.046 |
especially | +0.045 |
always | +0.045 |
Neg logit contributions | |
izza | -0.019 |
tal | -0.019 |
eg | -0.018 |
cannon | -0.016 |
crew | -0.016 |
Token | its |
Token ID | 663 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
itched | +1.433 |
tal | +1.385 |
eg | +1.312 |
orse | +1.238 |
izza | +1.214 |
Neg logit contributions | |
She | -1.183 |
she | -1.155 |
day | -1.141 |
like | -1.049 |
sometimes | -1.035 |
Token | nose |
Token ID | 9686 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
itched | +0.340 |
eg | +0.312 |
izza | +0.310 |
tal | +0.307 |
ades | +0.305 |
Neg logit contributions | |
She | -0.410 |
day | -0.406 |
like | -0.390 |
she | -0.390 |
always | -0.377 |
Token | . |
Token ID | 13 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
itched | +0.694 |
tal | +0.671 |
eg | +0.634 |
orse | +0.601 |
izza | +0.592 |
Neg logit contributions | |
She | -0.467 |
she | -0.457 |
day | -0.454 |
like | -0.408 |
sometimes | -0.403 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
itched | +0.109 |
tal | +0.106 |
eg | +0.101 |
orse | +0.095 |
izza | +0.093 |
Neg logit contributions | |
She | -0.067 |
she | -0.064 |
day | -0.061 |
like | -0.056 |
sometimes | -0.054 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
she | +0.269 |
She | +0.265 |
always | +0.245 |
like | +0.245 |
sometimes | +0.239 |
Neg logit contributions | |
tal | -0.404 |
itched | -0.394 |
eg | -0.389 |
izza | -0.381 |
ades | -0.361 |
Token | of |
Token ID | 286 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
itched | +0.099 |
tal | +0.088 |
eg | +0.085 |
ves | +0.082 |
ades | +0.081 |
Neg logit contributions | |
day | -0.056 |
She | -0.056 |
weekend | -0.052 |
she | -0.052 |
week | -0.049 |
Token | red |
Token ID | 2266 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
She | +0.245 |
day | +0.239 |
she | +0.238 |
always | +0.212 |
like | +0.212 |
Neg logit contributions | |
itched | -0.341 |
tal | -0.327 |
eg | -0.306 |
orse | -0.289 |
izza | -0.284 |
Token | cloth |
Token ID | 16270 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
She | +0.099 |
day | +0.098 |
she | +0.097 |
weekend | +0.094 |
especially | +0.093 |
Neg logit contributions | |
itched | -0.119 |
tal | -0.110 |
eg | -0.100 |
fin | -0.093 |
cannon | -0.091 |
Token | and |
Token ID | 290 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
itched | +0.134 |
tal | +0.122 |
eg | +0.117 |
fin | +0.108 |
ves | +0.107 |
Neg logit contributions | |
She | -0.092 |
day | -0.091 |
she | -0.087 |
weekend | -0.083 |
especially | -0.082 |
Token | st |
Token ID | 336 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
She | +0.152 |
day | +0.149 |
she | +0.148 |
like | +0.134 |
always | +0.133 |
Neg logit contributions | |
itched | -0.192 |
tal | -0.184 |
eg | -0.173 |
orse | -0.163 |
izza | -0.160 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
tal | +0.138 |
orse | +0.132 |
fin | +0.122 |
gru | +0.118 |
crew | +0.116 |
Neg logit contributions | |
She | -0.300 |
she | -0.264 |
She | -0.249 |
like | -0.248 |
her | -0.244 |
Token | her |
Token ID | 607 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
itched | +1.628 |
tal | +1.621 |
eg | +1.561 |
orse | +1.518 |
izza | +1.465 |
Neg logit contributions | |
She | -0.999 |
she | -0.978 |
day | -0.937 |
like | -0.893 |
sometimes | -0.853 |
Token | name |
Token ID | 1438 |
Feature activation | -1.91e-01 |
Pos logit contributions | |
day | +0.663 |
She | +0.639 |
she | +0.637 |
like | +0.581 |
always | +0.577 |
Neg logit contributions | |
itched | -0.608 |
tal | -0.577 |
eg | -0.561 |
orse | -0.553 |
izza | -0.542 |
Token | on |
Token ID | 319 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
itched | +0.403 |
tal | +0.379 |
eg | +0.359 |
orse | +0.334 |
izza | +0.332 |
Neg logit contributions | |
She | -0.333 |
day | -0.329 |
she | -0.326 |
especially | -0.296 |
always | -0.295 |
Token | it |
Token ID | 340 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
itched | +0.046 |
tal | +0.045 |
eg | +0.042 |
orse | +0.040 |
izza | +0.040 |
Neg logit contributions | |
She | -0.027 |
she | -0.027 |
day | -0.026 |
like | -0.023 |
sometimes | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
itched | +0.146 |
tal | +0.138 |
eg | +0.130 |
orse | +0.122 |
izza | +0.120 |
Neg logit contributions | |
She | -0.120 |
day | -0.118 |
she | -0.118 |
always | -0.107 |
like | -0.106 |
Token | to |
Token ID | 284 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
itched | +0.009 |
tal | +0.008 |
eg | +0.008 |
orse | +0.008 |
izza | +0.008 |
Neg logit contributions | |
she | -0.005 |
She | -0.005 |
day | -0.005 |
like | -0.004 |
sometimes | -0.004 |
Token | work |
Token ID | 670 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
itched | +0.151 |
eg | +0.146 |
tal | +0.141 |
ades | +0.138 |
izza | +0.138 |
Neg logit contributions | |
she | -0.082 |
She | -0.079 |
day | -0.079 |
her | -0.072 |
always | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
She | +0.023 |
day | +0.022 |
she | +0.022 |
like | +0.019 |
always | +0.019 |
Neg logit contributions | |
itched | -0.039 |
tal | -0.038 |
eg | -0.036 |
orse | -0.034 |
izza | -0.033 |
Token | carefully |
Token ID | 7773 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
she | +0.132 |
day | +0.126 |
She | +0.122 |
always | +0.115 |
sometimes | +0.114 |
Neg logit contributions | |
itched | -0.197 |
tal | -0.186 |
eg | -0.185 |
ades | -0.170 |
izza | -0.169 |
Token | st |
Token ID | 336 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
eg | +0.354 |
tal | +0.349 |
itched | +0.348 |
izza | +0.326 |
ades | +0.317 |
Neg logit contributions | |
she | -0.300 |
She | -0.275 |
day | -0.269 |
like | -0.263 |
always | -0.259 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
orse | +0.128 |
urger | +0.125 |
cannon | +0.122 |
tal | +0.119 |
izza | +0.118 |
Neg logit contributions | |
She | -0.283 |
she | -0.259 |
She | -0.238 |
her | -0.237 |
always | -0.226 |
Token | the |
Token ID | 262 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
eg | +1.665 |
itched | +1.646 |
orse | +1.610 |
tal | +1.592 |
izza | +1.553 |
Neg logit contributions | |
She | -0.957 |
she | -0.948 |
day | -0.907 |
her | -0.888 |
like | -0.888 |
Token | coat |
Token ID | 13209 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
itched | +0.055 |
tal | +0.053 |
eg | +0.050 |
orse | +0.047 |
izza | +0.046 |
Neg logit contributions | |
She | -0.043 |
day | -0.042 |
she | -0.042 |
like | -0.038 |
always | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
itched | +0.293 |
tal | +0.283 |
eg | +0.267 |
orse | +0.253 |
izza | +0.248 |
Neg logit contributions | |
She | -0.210 |
she | -0.206 |
day | -0.205 |
like | -0.184 |
always | -0.183 |
Token |
|
Token ID | 198 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
itched | +0.118 |
tal | +0.113 |
eg | +0.109 |
urger | +0.104 |
fin | +0.103 |
Neg logit contributions | |
She | -0.090 |
she | -0.080 |
day | -0.076 |
always | -0.068 |
like | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
tal | +0.320 |
itched | +0.319 |
urger | +0.310 |
eg | +0.305 |
jaws | +0.301 |
Neg logit contributions | |
She | -0.187 |
she | -0.166 |
day | -0.165 |
She | -0.147 |
always | -0.145 |
Token | was |
Token ID | 373 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
itched | +0.109 |
eg | +0.106 |
tal | +0.105 |
izza | +0.104 |
rods | +0.099 |
Neg logit contributions | |
she | -0.055 |
She | -0.053 |
day | -0.051 |
always | -0.050 |
sometimes | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
itched | +0.047 |
eg | +0.046 |
tal | +0.046 |
izza | +0.044 |
ves | +0.043 |
Neg logit contributions | |
she | -0.027 |
day | -0.027 |
She | -0.027 |
like | -0.026 |
always | -0.026 |
Token | high |
Token ID | 1029 |
Feature activation | +9.86e-03 |
Pos logit contributions | |
itched | +0.186 |
tal | +0.175 |
eg | +0.165 |
cks | +0.160 |
ves | +0.159 |
Neg logit contributions | |
day | -0.116 |
she | -0.109 |
She | -0.106 |
always | -0.098 |
girl | -0.097 |
Token | - |
Token ID | 12 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
day | +0.018 |
She | +0.018 |
she | +0.018 |
like | +0.016 |
sometimes | +0.016 |
Neg logit contributions | |
itched | -0.020 |
tal | -0.019 |
eg | -0.018 |
orse | -0.017 |
izza | -0.017 |
Token | p |
Token ID | 79 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
itched | +0.384 |
tal | +0.366 |
eg | +0.346 |
orse | +0.331 |
izza | +0.322 |
Neg logit contributions | |
She | -0.292 |
day | -0.283 |
she | -0.282 |
always | -0.257 |
especially | -0.256 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
tal | +0.254 |
itched | +0.241 |
eg | +0.240 |
izza | +0.229 |
rods | +0.219 |
Neg logit contributions | |
she | -0.244 |
day | -0.241 |
She | -0.237 |
like | -0.226 |
always | -0.224 |
Token | , |
Token ID | 11 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
itched | +1.586 |
tal | +1.542 |
eg | +1.483 |
orse | +1.410 |
izza | +1.400 |
Neg logit contributions | |
day | -1.007 |
She | -0.992 |
she | -0.961 |
like | -0.895 |
always | -0.883 |
Token | sque |
Token ID | 35776 |
Feature activation | -2.17e-01 |
Pos logit contributions | |
itched | +0.091 |
tal | +0.087 |
eg | +0.082 |
orse | +0.079 |
fin | +0.077 |
Neg logit contributions | |
She | -0.057 |
day | -0.055 |
she | -0.055 |
sometimes | -0.048 |
always | -0.048 |
Token | aky |
Token ID | 15492 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
itched | +0.390 |
tal | +0.359 |
eg | +0.318 |
ades | +0.309 |
fin | +0.307 |
Neg logit contributions | |
she | -0.481 |
day | -0.465 |
She | -0.463 |
weekend | -0.427 |
sometimes | -0.423 |
Token | noise |
Token ID | 7838 |
Feature activation | -3.09e-03 |
Pos logit contributions | |
day | +0.028 |
She | +0.027 |
she | +0.027 |
like | +0.025 |
always | +0.024 |
Neg logit contributions | |
itched | -0.035 |
tal | -0.033 |
eg | -0.032 |
izza | -0.030 |
rods | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
itched | +0.007 |
tal | +0.007 |
eg | +0.007 |
orse | +0.006 |
izza | +0.006 |
Neg logit contributions | |
She | -0.005 |
she | -0.005 |
day | -0.005 |
like | -0.004 |
always | -0.004 |
Token | loud |
Token ID | 7812 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
she | +0.029 |
day | +0.028 |
her | +0.028 |
She | +0.026 |
like | +0.025 |
Neg logit contributions | |
itched | -0.055 |
tal | -0.055 |
eg | -0.053 |
izza | -0.050 |
orse | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
itched | +0.172 |
tal | +0.168 |
eg | +0.162 |
izza | +0.147 |
orse | +0.146 |
Neg logit contributions | |
She | -0.178 |
she | -0.174 |
day | -0.173 |
like | -0.170 |
sometimes | -0.164 |
Token | high |
Token ID | 1029 |
Feature activation | +8.88e-03 |
Pos logit contributions | |
itched | +0.097 |
tal | +0.094 |
eg | +0.089 |
orse | +0.085 |
izza | +0.084 |
Neg logit contributions | |
She | -0.049 |
day | -0.048 |
she | -0.047 |
always | -0.041 |
sometimes | -0.041 |
Token | - |
Token ID | 12 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
day | +0.014 |
She | +0.013 |
she | +0.013 |
like | +0.012 |
sometimes | +0.012 |
Neg logit contributions | |
itched | -0.021 |
tal | -0.021 |
eg | -0.020 |
orse | -0.019 |
izza | -0.019 |
Token | p |
Token ID | 79 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
itched | +0.348 |
tal | +0.335 |
eg | +0.306 |
orse | +0.282 |
izza | +0.281 |
Neg logit contributions | |
She | -0.347 |
she | -0.343 |
day | -0.343 |
like | -0.316 |
sometimes | -0.311 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
tal | +0.313 |
eg | +0.301 |
plug | +0.286 |
izza | +0.284 |
itched | +0.281 |
Neg logit contributions | |
she | -0.189 |
She | -0.188 |
like | -0.185 |
day | -0.172 |
always | -0.162 |
Token | noises |
Token ID | 26782 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
tal | +1.536 |
eg | +1.507 |
itched | +1.483 |
orse | +1.439 |
izza | +1.428 |
Neg logit contributions | |
She | -1.067 |
like | -1.031 |
she | -1.024 |
day | -0.997 |
sometimes | -0.944 |
Token | . |
Token ID | 13 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
itched | +0.506 |
tal | +0.496 |
eg | +0.468 |
orse | +0.438 |
izza | +0.437 |
Neg logit contributions | |
She | -0.331 |
she | -0.328 |
day | -0.324 |
like | -0.301 |
sometimes | -0.291 |
Token | They |
Token ID | 1119 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
tal | +0.130 |
itched | +0.129 |
eg | +0.122 |
urger | +0.118 |
ves | +0.114 |
Neg logit contributions | |
She | -0.066 |
she | -0.054 |
day | -0.049 |
like | -0.047 |
sometimes | -0.045 |
Token | sounded |
Token ID | 14846 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
she | +0.099 |
She | +0.097 |
day | +0.096 |
always | +0.094 |
like | +0.094 |
Neg logit contributions | |
itched | -0.176 |
tal | -0.170 |
eg | -0.160 |
izza | -0.154 |
orse | -0.148 |
Token | like |
Token ID | 588 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
She | +0.009 |
day | +0.008 |
she | +0.008 |
like | +0.005 |
always | +0.004 |
Neg logit contributions | |
itched | -0.092 |
tal | -0.090 |
eg | -0.087 |
orse | -0.084 |
izza | -0.083 |
Token | She |
Token ID | 1375 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
itched | +0.139 |
tal | +0.134 |
eg | +0.129 |
izza | +0.124 |
orse | +0.122 |
Neg logit contributions | |
day | -0.029 |
she | -0.025 |
She | -0.023 |
especially | -0.021 |
sometimes | -0.021 |
Token | se |
Token ID | 384 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
she | +0.660 |
day | +0.627 |
always | +0.618 |
She | +0.615 |
especially | +0.608 |
Neg logit contributions | |
itched | -0.989 |
tal | -0.974 |
eg | -0.932 |
izza | -0.868 |
cannon | -0.858 |
Token | wed |
Token ID | 19103 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
itched | +0.269 |
tal | +0.254 |
eg | +0.234 |
orse | +0.215 |
izza | +0.212 |
Neg logit contributions | |
She | -0.327 |
she | -0.325 |
day | -0.319 |
sometimes | -0.293 |
especially | -0.292 |
Token | , |
Token ID | 11 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
itched | +0.338 |
tal | +0.324 |
eg | +0.306 |
orse | +0.289 |
izza | +0.284 |
Neg logit contributions | |
She | -0.224 |
day | -0.219 |
she | -0.218 |
always | -0.194 |
like | -0.193 |
Token | st |
Token ID | 336 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
itched | +0.077 |
tal | +0.072 |
eg | +0.067 |
orse | +0.066 |
fin | +0.062 |
Neg logit contributions | |
She | -0.064 |
day | -0.063 |
she | -0.060 |
weekend | -0.056 |
week | -0.056 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
tal | +0.138 |
orse | +0.121 |
fin | +0.110 |
cannon | +0.105 |
wounds | +0.105 |
Neg logit contributions | |
She | -0.297 |
she | -0.267 |
my | -0.249 |
like | -0.249 |
She | -0.248 |
Token | , |
Token ID | 11 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
itched | +1.601 |
tal | +1.578 |
eg | +1.509 |
orse | +1.448 |
izza | +1.423 |
Neg logit contributions | |
She | -0.990 |
she | -0.984 |
day | -0.952 |
like | -0.883 |
sometimes | -0.850 |
Token | and |
Token ID | 290 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
itched | +0.088 |
tal | +0.084 |
eg | +0.079 |
orse | +0.076 |
fin | +0.074 |
Neg logit contributions | |
She | -0.054 |
day | -0.053 |
she | -0.052 |
sometimes | -0.046 |
like | -0.046 |
Token | pinned |
Token ID | 25711 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
She | +0.139 |
day | +0.134 |
she | +0.132 |
like | +0.119 |
always | +0.118 |
Neg logit contributions | |
itched | -0.193 |
tal | -0.185 |
eg | -0.173 |
orse | -0.166 |
izza | -0.161 |
Token | her |
Token ID | 607 |
Feature activation | +3.24e-01 |
Pos logit contributions | |
itched | +0.619 |
tal | +0.589 |
eg | +0.548 |
fin | +0.505 |
izza | +0.505 |
Neg logit contributions | |
She | -0.446 |
day | -0.437 |
she | -0.417 |
weekend | -0.398 |
always | -0.398 |
Token | masterpiece |
Token ID | 30669 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
day | +0.597 |
she | +0.583 |
She | +0.574 |
like | +0.526 |
my | +0.525 |
Neg logit contributions | |
itched | -0.684 |
tal | -0.647 |
eg | -0.602 |
orse | -0.593 |
izza | -0.577 |
Token | its |
Token ID | 663 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
she | +0.150 |
day | +0.146 |
She | +0.141 |
like | +0.137 |
always | +0.137 |
Neg logit contributions | |
itched | -0.225 |
eg | -0.214 |
tal | -0.209 |
izza | -0.199 |
orse | -0.192 |
Token | mother |
Token ID | 2802 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
itched | +0.426 |
tal | +0.414 |
eg | +0.384 |
orse | +0.363 |
fin | +0.357 |
Neg logit contributions | |
She | -0.310 |
she | -0.300 |
day | -0.295 |
especially | -0.264 |
sometimes | -0.264 |
Token | who |
Token ID | 508 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
itched | +0.313 |
tal | +0.302 |
eg | +0.280 |
orse | +0.268 |
izza | +0.260 |
Neg logit contributions | |
She | -0.257 |
day | -0.248 |
she | -0.248 |
always | -0.222 |
sometimes | -0.222 |
Token | had |
Token ID | 550 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
itched | +0.100 |
tal | +0.096 |
eg | +0.089 |
orse | +0.087 |
izza | +0.085 |
Neg logit contributions | |
She | -0.064 |
day | -0.062 |
she | -0.059 |
weekend | -0.055 |
especially | -0.055 |
Token | st |
Token ID | 336 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
she | +0.162 |
day | +0.157 |
She | +0.149 |
always | +0.147 |
like | +0.143 |
Neg logit contributions | |
itched | -0.255 |
eg | -0.246 |
tal | -0.241 |
izza | -0.226 |
orse | -0.221 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
tal | +0.118 |
cannon | +0.116 |
rods | +0.116 |
eg | +0.114 |
orse | +0.112 |
Neg logit contributions | |
She | -0.281 |
she | -0.252 |
my | -0.248 |
always | -0.245 |
like | -0.244 |
Token | the |
Token ID | 262 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
itched | +1.514 |
tal | +1.457 |
eg | +1.363 |
orse | +1.290 |
izza | +1.268 |
Neg logit contributions | |
She | -1.065 |
day | -1.042 |
she | -1.040 |
always | -0.923 |
like | -0.922 |
Token | egg |
Token ID | 5935 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
itched | +0.049 |
tal | +0.047 |
eg | +0.043 |
orse | +0.040 |
fin | +0.040 |
Neg logit contributions | |
She | -0.051 |
she | -0.050 |
day | -0.049 |
like | -0.046 |
especially | -0.045 |
Token | together |
Token ID | 1978 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
itched | +0.254 |
tal | +0.244 |
eg | +0.227 |
orse | +0.216 |
izza | +0.210 |
Neg logit contributions | |
She | -0.202 |
day | -0.197 |
she | -0.195 |
especially | -0.176 |
always | -0.176 |
Token | . |
Token ID | 13 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
itched | +0.344 |
tal | +0.329 |
eg | +0.301 |
orse | +0.289 |
fin | +0.281 |
Neg logit contributions | |
She | -0.296 |
day | -0.286 |
she | -0.282 |
weekend | -0.257 |
especially | -0.257 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.70e-01 |
Pos logit contributions | |
itched | +0.108 |
tal | +0.103 |
eg | +0.097 |
orse | +0.092 |
izza | +0.090 |
Neg logit contributions | |
She | -0.092 |
she | -0.090 |
day | -0.089 |
like | -0.080 |
always | -0.080 |
Token | sweet |
Token ID | 6029 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
itched | +0.177 |
tal | +0.166 |
eg | +0.161 |
orse | +0.152 |
izza | +0.151 |
Neg logit contributions | |
day | -0.108 |
She | -0.106 |
she | -0.105 |
like | -0.093 |
always | -0.092 |
Token | , |
Token ID | 11 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
She | +0.135 |
she | +0.132 |
day | +0.131 |
sometimes | +0.120 |
like | +0.120 |
Neg logit contributions | |
itched | -0.141 |
tal | -0.135 |
eg | -0.126 |
orse | -0.118 |
izza | -0.115 |
Token | high |
Token ID | 1029 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
itched | +0.071 |
tal | +0.068 |
eg | +0.064 |
orse | +0.061 |
izza | +0.060 |
Neg logit contributions | |
She | -0.052 |
day | -0.051 |
she | -0.050 |
sometimes | -0.045 |
always | -0.045 |
Token | - |
Token ID | 12 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
She | +0.025 |
she | +0.025 |
day | +0.025 |
like | +0.022 |
always | +0.022 |
Neg logit contributions | |
itched | -0.031 |
tal | -0.029 |
eg | -0.028 |
orse | -0.026 |
izza | -0.025 |
Token | p |
Token ID | 79 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
itched | +0.354 |
tal | +0.339 |
eg | +0.322 |
orse | +0.307 |
izza | +0.297 |
Neg logit contributions | |
She | -0.301 |
day | -0.293 |
she | -0.292 |
sometimes | -0.270 |
always | -0.268 |
Token | itched |
Token ID | 10981 |
Feature activation | -6.62e-01 |
Pos logit contributions | |
tal | +0.254 |
itched | +0.253 |
eg | +0.247 |
izza | +0.230 |
rods | +0.223 |
Neg logit contributions | |
she | -0.223 |
She | -0.222 |
day | -0.216 |
like | -0.200 |
always | -0.197 |
Token | voice |
Token ID | 3809 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
itched | +1.400 |
tal | +1.338 |
eg | +1.284 |
orse | +1.212 |
izza | +1.196 |
Neg logit contributions | |
She | -1.180 |
day | -1.155 |
she | -1.153 |
like | -1.053 |
always | -1.034 |
Token | . |
Token ID | 13 |
Feature activation | -3.80e-02 |
Pos logit contributions | |
itched | +0.411 |
tal | +0.397 |
eg | +0.378 |
orse | +0.355 |
izza | +0.353 |
Neg logit contributions | |
She | -0.307 |
she | -0.305 |
day | -0.296 |
like | -0.277 |
always | -0.266 |
Token | � |
Token ID | 6184 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
itched | +0.093 |
tal | +0.090 |
eg | +0.085 |
orse | +0.081 |
izza | +0.079 |
Neg logit contributions | |
She | -0.057 |
she | -0.055 |
day | -0.053 |
like | -0.048 |
always | -0.047 |
Token | � |
Token ID | 95 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
itched | +0.281 |
tal | +0.274 |
eg | +0.248 |
fin | +0.232 |
orse | +0.231 |
Neg logit contributions | |
She | -0.324 |
she | -0.314 |
day | -0.307 |
like | -0.290 |
always | -0.285 |
Token | € |
Token ID | 26391 |
Feature activation | +6.28e-02 |
Pos logit contributions | |
itched | +0.363 |
tal | +0.345 |
eg | +0.324 |
orse | +0.306 |
fin | +0.303 |
Neg logit contributions | |
She | -0.291 |
day | -0.278 |
she | -0.274 |
like | -0.254 |
always | -0.253 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-02 |
Pos logit contributions | |
She | +0.503 |
day | +0.501 |
she | +0.500 |
always | +0.439 |
like | +0.438 |
Neg logit contributions | |
itched | -0.699 |
tal | -0.674 |
eg | -0.637 |
orse | -0.605 |
izza | -0.596 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
She | +0.043 |
she | +0.042 |
day | +0.042 |
like | +0.037 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.075 |
tal | -0.072 |
eg | -0.069 |
orse | -0.065 |
izza | -0.064 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
itched | +0.439 |
tal | +0.426 |
eg | +0.403 |
urger | +0.393 |
izza | +0.385 |
Neg logit contributions | |
She | -0.329 |
day | -0.319 |
she | -0.305 |
my | -0.276 |
week | -0.266 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
She | +0.251 |
she | +0.246 |
day | +0.246 |
like | +0.208 |
always | +0.208 |
Neg logit contributions | |
itched | -0.531 |
tal | -0.515 |
eg | -0.489 |
orse | -0.465 |
izza | -0.460 |
Token | a |
Token ID | 257 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
She | +0.230 |
day | +0.227 |
she | +0.224 |
weekend | +0.210 |
week | +0.208 |
Neg logit contributions | |
itched | -0.205 |
tal | -0.195 |
orse | -0.175 |
eg | -0.175 |
fin | -0.166 |
Token | time |
Token ID | 640 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
itched | +0.052 |
tal | +0.050 |
eg | +0.048 |
orse | +0.047 |
izza | +0.045 |
Neg logit contributions | |
She | -0.020 |
she | -0.020 |
day | -0.017 |
sometimes | -0.016 |
like | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
She | +0.466 |
she | +0.456 |
day | +0.456 |
like | +0.409 |
always | +0.407 |
Neg logit contributions | |
itched | -0.598 |
tal | -0.574 |
eg | -0.540 |
orse | -0.509 |
izza | -0.500 |
Token | there |
Token ID | 612 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
She | +0.015 |
she | +0.014 |
day | +0.014 |
like | +0.013 |
especially | +0.013 |
Neg logit contributions | |
itched | -0.025 |
tal | -0.023 |
eg | -0.022 |
orse | -0.021 |
izza | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
She | +0.133 |
she | +0.129 |
day | +0.129 |
like | +0.116 |
especially | +0.116 |
Neg logit contributions | |
itched | -0.177 |
tal | -0.168 |
eg | -0.158 |
orse | -0.150 |
izza | -0.145 |
Token | a |
Token ID | 257 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
She | +0.065 |
she | +0.063 |
day | +0.062 |
matter | +0.061 |
my | +0.059 |
Neg logit contributions | |
itched | -0.054 |
tal | -0.051 |
orse | -0.045 |
eg | -0.045 |
izza | -0.042 |
Token | little |
Token ID | 1310 |
Feature activation | +3.81e-02 |
Pos logit contributions | |
itched | +0.064 |
tal | +0.061 |
eg | +0.059 |
orse | +0.056 |
fin | +0.054 |
Neg logit contributions | |
She | -0.046 |
she | -0.044 |
day | -0.044 |
sometimes | -0.040 |
like | -0.039 |
Token | said |
Token ID | 531 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
itched | +0.360 |
tal | +0.340 |
eg | +0.326 |
ades | +0.312 |
izza | +0.310 |
Neg logit contributions | |
She | -0.225 |
she | -0.223 |
my | -0.201 |
always | -0.194 |
sometimes | -0.191 |
Token | . |
Token ID | 13 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
tal | +0.364 |
itched | +0.344 |
eg | +0.336 |
izza | +0.316 |
orse | +0.313 |
Neg logit contributions | |
she | -0.199 |
She | -0.193 |
day | -0.166 |
her | -0.155 |
like | -0.153 |
Token |
|
Token ID | 198 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
itched | +0.111 |
tal | +0.108 |
eg | +0.101 |
urger | +0.098 |
izza | +0.097 |
Neg logit contributions | |
She | -0.074 |
she | -0.068 |
day | -0.060 |
sometimes | -0.056 |
like | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
tal | +0.345 |
itched | +0.338 |
urger | +0.333 |
eg | +0.326 |
jaws | +0.318 |
Neg logit contributions | |
She | -0.189 |
she | -0.169 |
day | -0.156 |
She | -0.148 |
my | -0.144 |
Token | " |
Token ID | 1 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
itched | +0.279 |
tal | +0.268 |
eg | +0.251 |
orse | +0.235 |
izza | +0.232 |
Neg logit contributions | |
She | -0.229 |
she | -0.223 |
day | -0.221 |
sometimes | -0.201 |
like | -0.200 |
Token | You |
Token ID | 1639 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
itched | +0.322 |
orse | +0.300 |
eg | +0.292 |
ves | +0.290 |
ades | +0.289 |
Neg logit contributions | |
day | -0.108 |
She | -0.106 |
my | -0.105 |
she | -0.094 |
week | -0.092 |
Token | are |
Token ID | 389 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
She | +0.349 |
she | +0.339 |
day | +0.339 |
sometimes | +0.280 |
especially | +0.280 |
Neg logit contributions | |
itched | -0.826 |
tal | -0.799 |
eg | -0.767 |
orse | -0.730 |
izza | -0.722 |
Token | welcome |
Token ID | 7062 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
She | +0.187 |
she | +0.184 |
day | +0.184 |
weekend | +0.162 |
week | +0.160 |
Neg logit contributions | |
itched | -0.321 |
tal | -0.311 |
eg | -0.298 |
orse | -0.279 |
izza | -0.277 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
She | +0.278 |
she | +0.272 |
day | +0.260 |
like | +0.244 |
sometimes | +0.236 |
Neg logit contributions | |
itched | -0.392 |
tal | -0.374 |
eg | -0.355 |
orse | -0.343 |
izza | -0.334 |
Token | We |
Token ID | 775 |
Feature activation | -9.54e-02 |
Pos logit contributions | |
itched | +0.111 |
tal | +0.109 |
eg | +0.104 |
orse | +0.100 |
izza | +0.099 |
Neg logit contributions | |
day | -0.057 |
she | -0.056 |
She | -0.053 |
especially | -0.050 |
week | -0.049 |
Token | are |
Token ID | 389 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
itched | +0.231 |
tal | +0.223 |
eg | +0.221 |
orse | +0.205 |
izza | +0.203 |
Neg logit contributions | |
She | -0.128 |
she | -0.127 |
day | -0.113 |
weekend | -0.106 |
week | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
itched | +0.394 |
tal | +0.380 |
eg | +0.360 |
orse | +0.342 |
izza | +0.337 |
Neg logit contributions | |
She | -0.229 |
she | -0.223 |
day | -0.223 |
like | -0.196 |
always | -0.195 |
Token | thanked |
Token ID | 26280 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
she | +0.292 |
She | +0.277 |
day | +0.273 |
always | +0.250 |
like | +0.243 |
Neg logit contributions | |
itched | -0.353 |
tal | -0.334 |
eg | -0.331 |
izza | -0.308 |
ades | -0.304 |
Token | Miss |
Token ID | 4544 |
Feature activation | +7.17e-02 |
Pos logit contributions | |
itched | +0.358 |
tal | +0.343 |
eg | +0.334 |
izza | +0.316 |
orse | +0.314 |
Neg logit contributions | |
she | -0.174 |
She | -0.173 |
day | -0.166 |
always | -0.143 |
like | -0.141 |
Token | Rose |
Token ID | 8049 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
She | +0.118 |
day | +0.116 |
she | +0.115 |
like | +0.103 |
always | +0.103 |
Neg logit contributions | |
itched | -0.159 |
tal | -0.153 |
eg | -0.144 |
orse | -0.136 |
izza | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
She | +0.043 |
she | +0.043 |
day | +0.043 |
like | +0.038 |
sometimes | +0.038 |
Neg logit contributions | |
itched | -0.068 |
tal | -0.066 |
eg | -0.062 |
orse | -0.059 |
izza | -0.058 |
Token | She |
Token ID | 1375 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
She | +0.014 |
she | +0.013 |
day | +0.012 |
sometimes | +0.011 |
always | +0.011 |
Neg logit contributions | |
itched | -0.027 |
tal | -0.026 |
eg | -0.025 |
orse | -0.024 |
fin | -0.024 |
Token | felt |
Token ID | 2936 |
Feature activation | -1.72e-03 |
Pos logit contributions | |
she | +0.767 |
She | +0.736 |
always | +0.711 |
day | +0.696 |
especially | +0.666 |
Neg logit contributions | |
itched | -1.102 |
tal | -1.037 |
eg | -1.027 |
izza | -0.963 |
ades | -0.946 |
Token | good |
Token ID | 922 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
itched | +0.004 |
tal | +0.004 |
eg | +0.003 |
orse | +0.003 |
fin | +0.003 |
Neg logit contributions | |
She | -0.003 |
day | -0.003 |
she | -0.003 |
weekend | -0.002 |
my | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
She | +0.163 |
day | +0.154 |
she | +0.152 |
especially | +0.141 |
weekend | +0.139 |
Neg logit contributions | |
itched | -0.264 |
tal | -0.254 |
eg | -0.235 |
orse | -0.228 |
fin | -0.223 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
She | +0.212 |
day | +0.207 |
she | +0.196 |
weekend | +0.193 |
week | +0.190 |
Neg logit contributions | |
itched | -0.253 |
tal | -0.244 |
eg | -0.220 |
orse | -0.216 |
fin | -0.206 |
Token | to |
Token ID | 284 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
She | +0.291 |
day | +0.284 |
she | +0.283 |
like | +0.255 |
always | +0.253 |
Neg logit contributions | |
itched | -0.366 |
tal | -0.351 |
eg | -0.329 |
orse | -0.311 |
izza | -0.305 |
Token | Ben |
Token ID | 3932 |
Feature activation | -6.83e-03 |
Pos logit contributions | |
She | +0.139 |
she | +0.137 |
day | +0.134 |
sometimes | +0.119 |
like | +0.119 |
Neg logit contributions | |
itched | -0.218 |
tal | -0.210 |
eg | -0.198 |
orse | -0.188 |
izza | -0.185 |
Token | learned |
Token ID | 4499 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
itched | +0.019 |
tal | +0.019 |
eg | +0.018 |
izza | +0.018 |
alks | +0.017 |
Neg logit contributions | |
like | -0.008 |
she | -0.008 |
She | -0.008 |
always | -0.007 |
sometimes | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
itched | +0.171 |
tal | +0.164 |
eg | +0.161 |
orse | +0.159 |
urger | +0.150 |
Neg logit contributions | |
she | -0.105 |
She | -0.102 |
her | -0.093 |
day | -0.091 |
sometimes | -0.090 |
Token | lesson |
Token ID | 11483 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
itched | +0.170 |
tal | +0.163 |
eg | +0.152 |
orse | +0.149 |
izza | +0.144 |
Neg logit contributions | |
day | -0.101 |
She | -0.100 |
she | -0.098 |
always | -0.087 |
week | -0.086 |
Token | that |
Token ID | 326 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
itched | +0.080 |
tal | +0.076 |
eg | +0.071 |
orse | +0.068 |
izza | +0.066 |
Neg logit contributions | |
She | -0.071 |
day | -0.069 |
she | -0.068 |
especially | -0.063 |
always | -0.063 |
Token | day |
Token ID | 1110 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
orse | +0.033 |
itched | +0.032 |
fin | +0.032 |
tal | +0.032 |
cannon | +0.032 |
Neg logit contributions | |
She | -0.007 |
She | -0.005 |
heart | -0.005 |
she | -0.004 |
especially | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
She | +0.679 |
she | +0.673 |
day | +0.654 |
sometimes | +0.593 |
like | +0.592 |
Neg logit contributions | |
itched | -1.011 |
tal | -1.003 |
eg | -0.957 |
orse | -0.895 |
izza | -0.893 |
Token | They |
Token ID | 1119 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
itched | +0.099 |
tal | +0.095 |
eg | +0.091 |
orse | +0.086 |
izza | +0.086 |
Neg logit contributions | |
day | -0.053 |
she | -0.052 |
She | -0.051 |
especially | -0.046 |
always | -0.046 |
Token | learned |
Token ID | 4499 |
Feature activation | -6.88e-02 |
Pos logit contributions | |
She | +0.117 |
day | +0.116 |
she | +0.114 |
weekend | +0.103 |
week | +0.102 |
Neg logit contributions | |
itched | -0.168 |
tal | -0.165 |
eg | -0.156 |
orse | -0.147 |
fin | -0.143 |
Token | that |
Token ID | 326 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
itched | +0.179 |
tal | +0.171 |
eg | +0.165 |
orse | +0.162 |
izza | +0.155 |
Neg logit contributions | |
she | -0.100 |
She | -0.099 |
day | -0.093 |
sometimes | -0.089 |
like | -0.089 |
Token | not |
Token ID | 407 |
Feature activation | -9.95e-02 |
Pos logit contributions | |
itched | +0.030 |
tal | +0.029 |
orse | +0.027 |
eg | +0.027 |
fin | +0.027 |
Neg logit contributions | |
She | -0.017 |
day | -0.015 |
she | -0.015 |
my | -0.014 |
like | -0.014 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
itched | +0.041 |
tal | +0.040 |
eg | +0.038 |
orse | +0.036 |
izza | +0.035 |
Neg logit contributions | |
She | -0.027 |
day | -0.027 |
she | -0.027 |
like | -0.024 |
always | -0.024 |
Token | S |
Token ID | 50 |
Feature activation | +8.77e-02 |
Pos logit contributions | |
itched | +0.504 |
tal | +0.486 |
eg | +0.454 |
orse | +0.429 |
izza | +0.419 |
Neg logit contributions | |
She | -0.453 |
day | -0.442 |
she | -0.436 |
like | -0.394 |
sometimes | -0.392 |
Token | ara |
Token ID | 3301 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
She | +0.138 |
day | +0.137 |
she | +0.135 |
my | +0.130 |
week | +0.121 |
Neg logit contributions | |
itched | -0.204 |
tal | -0.202 |
fin | -0.182 |
eg | -0.179 |
orse | -0.177 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
She | +0.080 |
she | +0.079 |
day | +0.077 |
especially | +0.070 |
weekend | +0.070 |
Neg logit contributions | |
itched | -0.133 |
tal | -0.127 |
eg | -0.121 |
orse | -0.115 |
izza | -0.111 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
day | +0.172 |
She | +0.162 |
matter | +0.161 |
weekend | +0.160 |
especially | +0.159 |
Neg logit contributions | |
itched | -0.233 |
tal | -0.232 |
orse | -0.210 |
eg | -0.208 |
fin | -0.197 |
Token | like |
Token ID | 588 |
Feature activation | +3.88e-01 |
Pos logit contributions | |
day | +0.015 |
she | +0.015 |
She | +0.015 |
weekend | +0.014 |
week | +0.014 |
Neg logit contributions | |
itched | -0.026 |
tal | -0.025 |
eg | -0.024 |
orse | -0.023 |
izza | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
She | +0.664 |
day | +0.650 |
she | +0.650 |
like | +0.584 |
always | +0.582 |
Neg logit contributions | |
itched | -0.844 |
tal | -0.809 |
eg | -0.761 |
orse | -0.718 |
izza | -0.704 |
Token | play |
Token ID | 711 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
itched | +0.097 |
tal | +0.093 |
eg | +0.088 |
orse | +0.085 |
fin | +0.082 |
Neg logit contributions | |
She | -0.050 |
day | -0.049 |
she | -0.049 |
sometimes | -0.043 |
always | -0.042 |
Token | in |
Token ID | 287 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
day | +0.157 |
She | +0.155 |
she | +0.155 |
like | +0.145 |
sometimes | +0.138 |
Neg logit contributions | |
tal | -0.295 |
itched | -0.293 |
eg | -0.279 |
izza | -0.265 |
orse | -0.263 |
Token | the |
Token ID | 262 |
Feature activation | -3.96e-03 |
Pos logit contributions | |
She | +0.128 |
she | +0.124 |
day | +0.124 |
always | +0.114 |
especially | +0.114 |
Neg logit contributions | |
itched | -0.158 |
tal | -0.150 |
eg | -0.139 |
orse | -0.133 |
fin | -0.130 |
Token | park |
Token ID | 3952 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
itched | +0.008 |
tal | +0.008 |
eg | +0.008 |
rods | +0.008 |
fin | +0.008 |
Neg logit contributions | |
She | -0.006 |
she | -0.006 |
matter | -0.006 |
especially | -0.005 |
like | -0.005 |
Token | go |
Token ID | 467 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
itched | +0.047 |
tal | +0.045 |
eg | +0.044 |
izza | +0.041 |
orse | +0.040 |
Neg logit contributions | |
she | -0.028 |
day | -0.028 |
She | -0.028 |
like | -0.026 |
always | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
She | +0.292 |
day | +0.285 |
she | +0.281 |
especially | +0.258 |
always | +0.257 |
Neg logit contributions | |
itched | -0.330 |
tal | -0.321 |
eg | -0.290 |
orse | -0.279 |
fin | -0.272 |
Token | put |
Token ID | 1234 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
She | +0.173 |
she | +0.172 |
day | +0.167 |
like | +0.150 |
sometimes | +0.150 |
Neg logit contributions | |
itched | -0.288 |
tal | -0.276 |
eg | -0.264 |
orse | -0.247 |
izza | -0.245 |
Token | something |
Token ID | 1223 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
itched | +0.050 |
tal | +0.047 |
eg | +0.046 |
orse | +0.044 |
izza | +0.043 |
Neg logit contributions | |
she | -0.034 |
day | -0.034 |
She | -0.034 |
like | -0.031 |
sometimes | -0.030 |
Token | great |
Token ID | 1049 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
itched | +0.302 |
tal | +0.285 |
eg | +0.281 |
orse | +0.262 |
izza | +0.258 |
Neg logit contributions | |
she | -0.182 |
day | -0.177 |
She | -0.170 |
like | -0.164 |
sometimes | -0.156 |
Token | there |
Token ID | 612 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
She | +0.052 |
she | +0.052 |
day | +0.052 |
like | +0.047 |
sometimes | +0.046 |
Neg logit contributions | |
itched | -0.070 |
tal | -0.067 |
eg | -0.064 |
orse | -0.060 |
izza | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
She | +0.109 |
day | +0.106 |
she | +0.106 |
especially | +0.096 |
always | +0.096 |
Neg logit contributions | |
itched | -0.124 |
tal | -0.119 |
eg | -0.111 |
orse | -0.105 |
izza | -0.103 |
Token | help |
Token ID | 1037 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
itched | +0.059 |
tal | +0.057 |
eg | +0.054 |
izza | +0.051 |
orse | +0.051 |
Neg logit contributions | |
She | -0.037 |
she | -0.037 |
day | -0.037 |
like | -0.033 |
always | -0.033 |
Token | it |
Token ID | 340 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
itched | +0.122 |
tal | +0.118 |
eg | +0.111 |
orse | +0.106 |
izza | +0.104 |
Neg logit contributions | |
She | -0.068 |
day | -0.066 |
she | -0.066 |
always | -0.057 |
like | -0.057 |
Token | ." |
Token ID | 526 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
itched | +0.080 |
tal | +0.077 |
eg | +0.074 |
orse | +0.069 |
izza | +0.068 |
Neg logit contributions | |
She | -0.056 |
she | -0.055 |
day | -0.054 |
like | -0.050 |
always | -0.049 |
Token | |
Token ID | 220 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
itched | +0.117 |
tal | +0.111 |
eg | +0.107 |
ades | +0.100 |
urger | +0.100 |
Neg logit contributions | |
She | -0.098 |
she | -0.094 |
day | -0.089 |
my | -0.079 |
always | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
day | +0.595 |
She | +0.591 |
she | +0.576 |
like | +0.533 |
always | +0.529 |
Neg logit contributions | |
itched | -0.844 |
tal | -0.817 |
eg | -0.771 |
izza | -0.734 |
ades | -0.720 |
Token | Tim |
Token ID | 5045 |
Feature activation | +9.36e-03 |
Pos logit contributions | |
itched | +0.068 |
tal | +0.059 |
eg | +0.059 |
izza | +0.056 |
orse | +0.054 |
Neg logit contributions | |
she | -0.066 |
day | -0.063 |
girl | -0.060 |
weekend | -0.060 |
always | -0.059 |
Token | my |
Token ID | 1820 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
she | +0.008 |
She | +0.007 |
weekend | +0.005 |
day | +0.005 |
week | +0.004 |
Neg logit contributions | |
itched | -0.028 |
tal | -0.027 |
eg | -0.026 |
orse | -0.025 |
urger | -0.025 |
Token | loved |
Token ID | 6151 |
Feature activation | +8.56e-02 |
Pos logit contributions | |
She | +0.486 |
she | +0.477 |
day | +0.452 |
weekend | +0.414 |
like | +0.404 |
Neg logit contributions | |
itched | -0.922 |
tal | -0.881 |
eg | -0.852 |
orse | -0.821 |
fin | -0.786 |
Token | to |
Token ID | 284 |
Feature activation | -5.23e-02 |
Pos logit contributions | |
She | +0.148 |
she | +0.145 |
day | +0.144 |
like | +0.131 |
sometimes | +0.130 |
Neg logit contributions | |
itched | -0.184 |
tal | -0.177 |
eg | -0.166 |
orse | -0.157 |
izza | -0.154 |
Token | play |
Token ID | 711 |
Feature activation | +9.79e-02 |
Pos logit contributions | |
itched | +0.142 |
tal | +0.138 |
eg | +0.130 |
orse | +0.124 |
izza | +0.123 |
Neg logit contributions | |
She | -0.063 |
she | -0.062 |
day | -0.062 |
like | -0.053 |
always | -0.053 |
Token | games |
Token ID | 1830 |
Feature activation | +8.16e-02 |
Pos logit contributions | |
She | +0.115 |
she | +0.113 |
day | +0.113 |
like | +0.101 |
always | +0.098 |
Neg logit contributions | |
tal | -0.269 |
itched | -0.267 |
eg | -0.250 |
izza | -0.241 |
orse | -0.240 |
Token | with |
Token ID | 351 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
She | +0.162 |
day | +0.153 |
matter | +0.151 |
she | +0.149 |
weekend | +0.148 |
Neg logit contributions | |
itched | -0.143 |
tal | -0.123 |
eg | -0.121 |
ves | -0.120 |
orse | -0.117 |
Token | his |
Token ID | 465 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
itched | +0.100 |
tal | +0.100 |
eg | +0.090 |
orse | +0.087 |
izza | +0.086 |
Neg logit contributions | |
day | -0.079 |
She | -0.078 |
she | -0.078 |
especially | -0.071 |
sometimes | -0.070 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
itched | +0.267 |
tal | +0.264 |
orse | +0.235 |
eg | +0.232 |
izza | +0.229 |
Neg logit contributions | |
day | -0.192 |
she | -0.177 |
She | -0.172 |
girl | -0.156 |
always | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
itched | +0.287 |
tal | +0.267 |
eg | +0.255 |
orse | +0.237 |
fin | +0.233 |
Neg logit contributions | |
She | -0.265 |
she | -0.260 |
day | -0.254 |
weekend | -0.231 |
especially | -0.230 |
Token | with |
Token ID | 351 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
itched | +0.138 |
tal | +0.138 |
eg | +0.126 |
orse | +0.120 |
fin | +0.120 |
Neg logit contributions | |
She | -0.120 |
she | -0.113 |
day | -0.112 |
always | -0.102 |
weekend | -0.101 |
Token | you |
Token ID | 345 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
itched | +0.103 |
tal | +0.099 |
eg | +0.095 |
orse | +0.091 |
izza | +0.089 |
Neg logit contributions | |
She | -0.038 |
day | -0.038 |
she | -0.037 |
like | -0.032 |
always | -0.031 |
Token | ." |
Token ID | 526 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
day | +0.340 |
She | +0.335 |
she | +0.334 |
like | +0.312 |
sometimes | +0.303 |
Neg logit contributions | |
itched | -0.473 |
tal | -0.455 |
eg | -0.432 |
orse | -0.401 |
ades | -0.393 |
Token |
|
Token ID | 198 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
itched | +0.158 |
tal | +0.145 |
eg | +0.141 |
ves | +0.141 |
cks | +0.138 |
Neg logit contributions | |
She | -0.114 |
she | -0.100 |
day | -0.095 |
my | -0.088 |
always | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
jaws | +0.291 |
itched | +0.289 |
tal | +0.283 |
sed | +0.282 |
fins | +0.281 |
Neg logit contributions | |
She | -0.144 |
day | -0.130 |
she | -0.128 |
She | -0.118 |
my | -0.114 |
Token | The |
Token ID | 464 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
itched | +0.260 |
tal | +0.252 |
eg | +0.237 |
fin | +0.220 |
cannon | +0.219 |
Neg logit contributions | |
She | -0.170 |
she | -0.166 |
day | -0.162 |
especially | -0.150 |
always | -0.149 |
Token | little |
Token ID | 1310 |
Feature activation | +9.15e-03 |
Pos logit contributions | |
itched | +0.105 |
tal | +0.101 |
eg | +0.095 |
orse | +0.090 |
izza | +0.089 |
Neg logit contributions | |
day | -0.064 |
She | -0.063 |
she | -0.062 |
always | -0.055 |
sometimes | -0.055 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
She | +0.010 |
she | +0.010 |
day | +0.009 |
especially | +0.008 |
always | +0.008 |
Neg logit contributions | |
itched | -0.025 |
tal | -0.025 |
eg | -0.023 |
orse | -0.022 |
fin | -0.022 |
Token | smiled |
Token ID | 13541 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
she | +0.263 |
always | +0.259 |
She | +0.258 |
day | +0.252 |
especially | +0.237 |
Neg logit contributions | |
itched | -0.395 |
tal | -0.384 |
eg | -0.363 |
cks | -0.347 |
ades | -0.345 |
Token | and |
Token ID | 290 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
itched | +0.564 |
tal | +0.553 |
eg | +0.536 |
orse | +0.512 |
izza | +0.510 |
Neg logit contributions | |
She | -0.268 |
she | -0.256 |
day | -0.245 |
like | -0.230 |
sometimes | -0.217 |
Token | was |
Token ID | 373 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
she | +0.181 |
She | +0.168 |
day | +0.163 |
always | +0.153 |
her | +0.144 |
Neg logit contributions | |
itched | -0.220 |
eg | -0.206 |
tal | -0.203 |
ades | -0.198 |
izza | -0.194 |
Token | they |
Token ID | 484 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
She | +0.167 |
day | +0.163 |
she | +0.161 |
like | +0.146 |
sometimes | +0.145 |
Neg logit contributions | |
itched | -0.234 |
tal | -0.224 |
eg | -0.210 |
orse | -0.201 |
izza | -0.196 |
Token | played |
Token ID | 2826 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
She | +0.031 |
day | +0.030 |
she | +0.030 |
weekend | +0.027 |
week | +0.027 |
Neg logit contributions | |
itched | -0.044 |
tal | -0.043 |
eg | -0.040 |
orse | -0.038 |
izza | -0.037 |
Token | together |
Token ID | 1978 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
She | +0.032 |
she | +0.032 |
day | +0.032 |
like | +0.030 |
always | +0.029 |
Neg logit contributions | |
itched | -0.040 |
tal | -0.039 |
eg | -0.037 |
orse | -0.034 |
izza | -0.034 |
Token | every |
Token ID | 790 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
itched | +0.354 |
tal | +0.340 |
eg | +0.322 |
orse | +0.307 |
izza | +0.301 |
Neg logit contributions | |
She | -0.205 |
she | -0.198 |
day | -0.198 |
like | -0.173 |
always | -0.173 |
Token | day |
Token ID | 1110 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
She | +0.049 |
she | +0.033 |
like | +0.014 |
especially | +0.010 |
always | +0.010 |
Neg logit contributions | |
itched | -0.388 |
tal | -0.383 |
orse | -0.366 |
eg | -0.358 |
fin | -0.357 |
Token | . |
Token ID | 13 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
She | +0.761 |
she | +0.752 |
day | +0.748 |
like | +0.673 |
sometimes | +0.667 |
Neg logit contributions | |
itched | -0.973 |
tal | -0.938 |
eg | -0.887 |
orse | -0.827 |
izza | -0.819 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
itched | +0.059 |
tal | +0.057 |
eg | +0.053 |
orse | +0.050 |
izza | +0.049 |
Neg logit contributions | |
She | -0.052 |
she | -0.052 |
day | -0.052 |
like | -0.047 |
especially | -0.047 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
itched | +0.591 |
tal | +0.572 |
eg | +0.540 |
urger | +0.519 |
orse | +0.514 |
Neg logit contributions | |
She | -0.430 |
day | -0.420 |
she | -0.400 |
like | -0.362 |
my | -0.355 |
Token | upon |
Token ID | 2402 |
Feature activation | +6.70e-02 |
Pos logit contributions | |
She | +0.194 |
day | +0.188 |
she | +0.188 |
like | +0.163 |
always | +0.162 |
Neg logit contributions | |
itched | -0.379 |
tal | -0.366 |
eg | -0.347 |
orse | -0.332 |
izza | -0.326 |
Token | a |
Token ID | 257 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
She | +0.130 |
day | +0.128 |
she | +0.127 |
weekend | +0.118 |
week | +0.116 |
Neg logit contributions | |
itched | -0.124 |
tal | -0.116 |
eg | -0.106 |
orse | -0.105 |
fin | -0.100 |
Token | time |
Token ID | 640 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
itched | +0.162 |
tal | +0.157 |
eg | +0.148 |
orse | +0.146 |
izza | +0.141 |
Neg logit contributions | |
She | -0.060 |
she | -0.059 |
day | -0.052 |
sometimes | -0.048 |
like | -0.044 |
Token | The |
Token ID | 383 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
itched | +0.109 |
tal | +0.104 |
eg | +0.100 |
orse | +0.094 |
fin | +0.092 |
Neg logit contributions | |
She | -0.085 |
she | -0.082 |
day | -0.079 |
always | -0.071 |
sometimes | -0.071 |
Token | little |
Token ID | 1310 |
Feature activation | -6.33e-03 |
Pos logit contributions | |
itched | +0.299 |
tal | +0.280 |
eg | +0.259 |
fin | +0.256 |
orse | +0.253 |
Neg logit contributions | |
She | -0.264 |
she | -0.249 |
day | -0.235 |
like | -0.231 |
always | -0.231 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
tal | +0.016 |
itched | +0.015 |
fin | +0.015 |
fist | +0.014 |
fins | +0.014 |
Neg logit contributions | |
She | -0.006 |
always | -0.006 |
especially | -0.006 |
weekend | -0.006 |
ock | -0.005 |
Token | noticed |
Token ID | 6810 |
Feature activation | -6.43e-02 |
Pos logit contributions | |
she | +0.453 |
She | +0.446 |
always | +0.436 |
day | +0.424 |
especially | +0.398 |
Neg logit contributions | |
itched | -0.886 |
tal | -0.864 |
eg | -0.843 |
izza | -0.806 |
ades | -0.801 |
Token | too |
Token ID | 1165 |
Feature activation | +9.58e-02 |
Pos logit contributions | |
itched | +0.150 |
tal | +0.146 |
eg | +0.142 |
orse | +0.131 |
izza | +0.130 |
Neg logit contributions | |
she | -0.103 |
She | -0.102 |
day | -0.099 |
sometimes | -0.090 |
like | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
She | +0.151 |
day | +0.149 |
she | +0.147 |
always | +0.132 |
weekend | +0.131 |
Neg logit contributions | |
itched | -0.221 |
tal | -0.211 |
eg | -0.198 |
orse | -0.189 |
izza | -0.184 |
Token | said |
Token ID | 531 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
she | +0.124 |
She | +0.120 |
day | +0.108 |
always | +0.104 |
sometimes | +0.102 |
Neg logit contributions | |
itched | -0.202 |
tal | -0.198 |
eg | -0.193 |
izza | -0.178 |
ades | -0.175 |
Token | , |
Token ID | 11 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
tal | +0.336 |
itched | +0.331 |
eg | +0.325 |
urger | +0.303 |
orse | +0.301 |
Neg logit contributions | |
she | -0.216 |
She | -0.205 |
day | -0.178 |
like | -0.172 |
sometimes | -0.168 |
Token | " |
Token ID | 366 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
tal | +0.096 |
itched | +0.095 |
eg | +0.092 |
ades | +0.084 |
izza | +0.083 |
Neg logit contributions | |
She | -0.077 |
she | -0.077 |
day | -0.068 |
always | -0.064 |
like | -0.063 |
Token | Oh |
Token ID | 5812 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
itched | +0.340 |
tal | +0.332 |
fin | +0.302 |
pri | +0.302 |
orse | +0.302 |
Neg logit contributions | |
day | -0.189 |
she | -0.180 |
She | -0.173 |
weekend | -0.165 |
week | -0.162 |
Token | no |
Token ID | 645 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
she | +0.249 |
She | +0.231 |
day | +0.227 |
like | +0.210 |
always | +0.210 |
Neg logit contributions | |
itched | -0.436 |
eg | -0.415 |
tal | -0.409 |
ades | -0.399 |
rods | -0.390 |
Token | fly |
Token ID | 6129 |
Feature activation | +1.90e-03 |
Pos logit contributions | |
itched | +0.090 |
tal | +0.087 |
eg | +0.080 |
orse | +0.076 |
fin | +0.075 |
Neg logit contributions | |
She | -0.076 |
day | -0.074 |
she | -0.073 |
weekend | -0.067 |
week | -0.067 |
Token | away |
Token ID | 1497 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
She | +0.003 |
day | +0.003 |
she | +0.003 |
weekend | +0.003 |
week | +0.003 |
Neg logit contributions | |
itched | -0.004 |
tal | -0.004 |
eg | -0.004 |
fin | -0.003 |
orse | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
itched | +0.035 |
tal | +0.034 |
eg | +0.032 |
orse | +0.030 |
izza | +0.030 |
Neg logit contributions | |
She | -0.025 |
she | -0.025 |
day | -0.025 |
like | -0.022 |
always | -0.022 |
Token | but |
Token ID | 475 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
itched | +0.071 |
tal | +0.070 |
eg | +0.065 |
orse | +0.064 |
fin | +0.063 |
Neg logit contributions | |
day | -0.027 |
She | -0.027 |
she | -0.025 |
weekend | -0.023 |
week | -0.022 |
Token | he |
Token ID | 339 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
She | +0.309 |
day | +0.298 |
she | +0.295 |
like | +0.267 |
especially | +0.265 |
Neg logit contributions | |
itched | -0.420 |
tal | -0.407 |
eg | -0.379 |
orse | -0.363 |
fin | -0.352 |
Token | hurt |
Token ID | 5938 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
she | +0.087 |
day | +0.086 |
She | +0.083 |
always | +0.080 |
especially | +0.077 |
Neg logit contributions | |
itched | -0.119 |
tal | -0.115 |
eg | -0.110 |
izza | -0.107 |
ades | -0.104 |
Token | his |
Token ID | 465 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
itched | +0.193 |
tal | +0.184 |
eg | +0.172 |
orse | +0.161 |
izza | +0.157 |
Neg logit contributions | |
She | -0.200 |
she | -0.196 |
day | -0.196 |
like | -0.179 |
always | -0.179 |
Token | wing |
Token ID | 8539 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
itched | +0.184 |
tal | +0.166 |
eg | +0.164 |
izza | +0.162 |
ades | +0.159 |
Neg logit contributions | |
day | -0.229 |
She | -0.215 |
she | -0.213 |
always | -0.207 |
like | -0.205 |
Token | . |
Token ID | 13 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
itched | +0.640 |
tal | +0.613 |
eg | +0.566 |
orse | +0.531 |
fin | +0.527 |
Neg logit contributions | |
She | -0.479 |
day | -0.469 |
she | -0.467 |
weekend | -0.427 |
week | -0.417 |
Token | Ups |
Token ID | 35949 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
itched | +0.089 |
tal | +0.085 |
eg | +0.083 |
urger | +0.080 |
orse | +0.080 |
Neg logit contributions | |
She | -0.067 |
she | -0.058 |
day | -0.055 |
always | -0.051 |
sometimes | -0.048 |
Token | et |
Token ID | 316 |
Feature activation | +3.63e-02 |
Pos logit contributions | |
tal | +0.282 |
itched | +0.275 |
cks | +0.246 |
eg | +0.245 |
orse | +0.236 |
Neg logit contributions | |
She | -0.258 |
she | -0.258 |
day | -0.242 |
I | -0.226 |
weekend | -0.223 |
Token | for |
Token ID | 329 |
Feature activation | +9.84e-02 |
Pos logit contributions | |
itched | +0.094 |
tal | +0.090 |
eg | +0.085 |
orse | +0.081 |
fin | +0.079 |
Neg logit contributions | |
She | -0.070 |
day | -0.068 |
she | -0.067 |
my | -0.061 |
always | -0.061 |
Token | Spot |
Token ID | 15899 |
Feature activation | -9.65e-02 |
Pos logit contributions | |
she | +0.075 |
She | +0.072 |
day | +0.068 |
always | +0.053 |
sometimes | +0.050 |
Neg logit contributions | |
itched | -0.312 |
tal | -0.307 |
eg | -0.298 |
izza | -0.285 |
orse | -0.283 |
Token | . |
Token ID | 13 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
itched | +0.208 |
tal | +0.199 |
eg | +0.187 |
orse | +0.177 |
izza | +0.173 |
Neg logit contributions | |
She | -0.167 |
day | -0.163 |
she | -0.163 |
like | -0.146 |
always | -0.146 |
Token | She |
Token ID | 1375 |
Feature activation | +4.26e-01 |
Pos logit contributions | |
itched | +0.091 |
tal | +0.087 |
eg | +0.083 |
orse | +0.079 |
izza | +0.078 |
Neg logit contributions | |
She | -0.049 |
day | -0.048 |
she | -0.048 |
like | -0.042 |
always | -0.042 |
Token | used |
Token ID | 973 |
Feature activation | -7.17e-02 |
Pos logit contributions | |
day | +0.606 |
she | +0.600 |
always | +0.597 |
She | +0.581 |
especially | +0.555 |
Neg logit contributions | |
itched | -1.053 |
tal | -1.043 |
eg | -0.967 |
izza | -0.914 |
eed | -0.890 |
Token | some |
Token ID | 617 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
itched | +0.171 |
tal | +0.166 |
eg | +0.159 |
orse | +0.151 |
izza | +0.148 |
Neg logit contributions | |
She | -0.109 |
she | -0.107 |
day | -0.106 |
like | -0.094 |
always | -0.093 |
Token | paper |
Token ID | 3348 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
itched | +0.213 |
tal | +0.204 |
eg | +0.191 |
orse | +0.180 |
izza | +0.177 |
Neg logit contributions | |
She | -0.181 |
she | -0.177 |
day | -0.177 |
like | -0.159 |
always | -0.159 |
Token | and |
Token ID | 290 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
itched | +0.244 |
tal | +0.238 |
eg | +0.224 |
orse | +0.215 |
izza | +0.209 |
Neg logit contributions | |
She | -0.161 |
she | -0.160 |
day | -0.155 |
like | -0.141 |
sometimes | -0.140 |
Token | cr |
Token ID | 1067 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
she | +0.172 |
She | +0.168 |
day | +0.165 |
sometimes | +0.153 |
like | +0.150 |
Neg logit contributions | |
itched | -0.195 |
tal | -0.185 |
eg | -0.179 |
orse | -0.165 |
izza | -0.162 |
Token | ay |
Token ID | 323 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
itched | +0.412 |
tal | +0.400 |
eg | +0.375 |
orse | +0.361 |
fin | +0.350 |
Neg logit contributions | |
She | -0.288 |
she | -0.276 |
day | -0.274 |
like | -0.248 |
always | -0.245 |
Token | ons |
Token ID | 684 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
tal | +0.042 |
itched | +0.042 |
eg | +0.040 |
fin | +0.038 |
fist | +0.036 |
Neg logit contributions | |
she | -0.053 |
She | -0.052 |
day | -0.052 |
my | -0.049 |
like | -0.048 |
Token | never |
Token ID | 1239 |
Feature activation | +5.71e-02 |
Pos logit contributions | |
She | +0.293 |
day | +0.286 |
she | +0.286 |
like | +0.249 |
sometimes | +0.248 |
Neg logit contributions | |
itched | -0.529 |
tal | -0.509 |
eg | -0.483 |
orse | -0.460 |
izza | -0.452 |
Token | go |
Token ID | 467 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
She | +0.077 |
she | +0.074 |
day | +0.074 |
weekend | +0.064 |
sometimes | +0.063 |
Neg logit contributions | |
itched | -0.142 |
tal | -0.140 |
eg | -0.130 |
orse | -0.128 |
fin | -0.124 |
Token | near |
Token ID | 1474 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
She | +0.209 |
day | +0.202 |
she | +0.195 |
weekend | +0.183 |
week | +0.182 |
Neg logit contributions | |
itched | -0.241 |
tal | -0.234 |
eg | -0.207 |
orse | -0.206 |
fin | -0.201 |
Token | the |
Token ID | 262 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
She | +0.099 |
she | +0.099 |
day | +0.097 |
like | +0.087 |
sometimes | +0.086 |
Neg logit contributions | |
itched | -0.154 |
tal | -0.147 |
eg | -0.143 |
orse | -0.134 |
izza | -0.133 |
Token | river |
Token ID | 7850 |
Feature activation | -3.84e-03 |
Pos logit contributions | |
itched | +0.081 |
tal | +0.077 |
eg | +0.073 |
orse | +0.069 |
izza | +0.068 |
Neg logit contributions | |
She | -0.052 |
day | -0.052 |
she | -0.051 |
like | -0.046 |
always | -0.046 |
Token | again |
Token ID | 757 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
itched | +0.008 |
tal | +0.007 |
eg | +0.007 |
orse | +0.007 |
izza | +0.006 |
Neg logit contributions | |
She | -0.007 |
she | -0.007 |
day | -0.007 |
like | -0.006 |
always | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -5.92e-02 |
Pos logit contributions | |
itched | +0.304 |
tal | +0.296 |
eg | +0.262 |
orse | +0.259 |
fin | +0.250 |
Neg logit contributions | |
She | -0.257 |
day | -0.249 |
she | -0.245 |
always | -0.224 |
weekend | -0.223 |
Token | She |
Token ID | 1375 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
itched | +0.157 |
tal | +0.153 |
eg | +0.144 |
izza | +0.137 |
orse | +0.137 |
Neg logit contributions | |
day | -0.069 |
she | -0.067 |
She | -0.065 |
like | -0.060 |
especially | -0.060 |
Token | was |
Token ID | 373 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
day | +0.563 |
she | +0.555 |
always | +0.541 |
She | +0.535 |
especially | +0.513 |
Neg logit contributions | |
itched | -1.044 |
tal | -0.971 |
eg | -0.967 |
izza | -0.918 |
ades | -0.896 |
Token | too |
Token ID | 1165 |
Feature activation | +8.43e-02 |
Pos logit contributions | |
itched | +0.078 |
tal | +0.075 |
eg | +0.072 |
orse | +0.068 |
izza | +0.068 |
Neg logit contributions | |
She | -0.040 |
she | -0.039 |
day | -0.039 |
always | -0.035 |
like | -0.035 |
Token | scared |
Token ID | 12008 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
She | +0.119 |
she | +0.117 |
day | +0.115 |
like | +0.107 |
always | +0.105 |
Neg logit contributions | |
itched | -0.212 |
tal | -0.205 |
eg | -0.200 |
izza | -0.189 |
ves | -0.185 |
Token | called |
Token ID | 1444 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
she | +0.595 |
She | +0.595 |
day | +0.564 |
always | +0.561 |
like | +0.534 |
Neg logit contributions | |
itched | -1.009 |
tal | -0.969 |
eg | -0.922 |
izza | -0.896 |
ades | -0.868 |
Token | his |
Token ID | 465 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
itched | +0.070 |
tal | +0.067 |
eg | +0.064 |
orse | +0.061 |
izza | +0.060 |
Neg logit contributions | |
She | -0.039 |
she | -0.038 |
day | -0.038 |
like | -0.033 |
always | -0.033 |
Token | name |
Token ID | 1438 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
itched | +0.282 |
tal | +0.263 |
orse | +0.259 |
ves | +0.239 |
eg | +0.238 |
Neg logit contributions | |
day | -0.238 |
She | -0.223 |
she | -0.214 |
like | -0.211 |
girl | -0.206 |
Token | . |
Token ID | 13 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
itched | +0.471 |
tal | +0.446 |
eg | +0.423 |
orse | +0.397 |
izza | +0.394 |
Neg logit contributions | |
She | -0.305 |
day | -0.303 |
she | -0.297 |
especially | -0.268 |
always | -0.266 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
tal | +0.120 |
itched | +0.119 |
eg | +0.110 |
orse | +0.105 |
urger | +0.103 |
Neg logit contributions | |
She | -0.094 |
she | -0.088 |
day | -0.081 |
like | -0.075 |
always | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
tal | +0.357 |
itched | +0.349 |
urger | +0.344 |
eg | +0.333 |
jaws | +0.328 |
Neg logit contributions | |
She | -0.205 |
she | -0.186 |
day | -0.174 |
She | -0.152 |
week | -0.146 |
Token | " |
Token ID | 1 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
itched | +0.291 |
tal | +0.279 |
eg | +0.262 |
orse | +0.246 |
izza | +0.242 |
Neg logit contributions | |
She | -0.238 |
she | -0.234 |
day | -0.233 |
like | -0.210 |
always | -0.209 |
Token | Ben |
Token ID | 11696 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
itched | +0.278 |
tal | +0.265 |
orse | +0.256 |
ves | +0.251 |
eg | +0.251 |
Neg logit contributions | |
day | -0.187 |
She | -0.185 |
she | -0.172 |
like | -0.169 |
always | -0.163 |
Token | , |
Token ID | 11 |
Feature activation | -4.82e-02 |
Pos logit contributions | |
She | +0.019 |
she | +0.019 |
day | +0.018 |
my | +0.017 |
always | +0.017 |
Neg logit contributions | |
itched | -0.031 |
tal | -0.030 |
eg | -0.029 |
ades | -0.027 |
orse | -0.027 |
Token | wait |
Token ID | 4043 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
itched | +0.104 |
tal | +0.104 |
eg | +0.099 |
orse | +0.098 |
fin | +0.096 |
Neg logit contributions | |
She | -0.072 |
day | -0.068 |
she | -0.063 |
weekend | -0.063 |
week | -0.062 |
Token | for |
Token ID | 329 |
Feature activation | +7.43e-02 |
Pos logit contributions | |
She | +0.174 |
day | +0.172 |
she | +0.172 |
like | +0.157 |
always | +0.153 |
Neg logit contributions | |
itched | -0.234 |
tal | -0.224 |
eg | -0.214 |
orse | -0.199 |
izza | -0.196 |
Token | " |
Token ID | 1 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
itched | +0.319 |
tal | +0.312 |
eg | +0.291 |
orse | +0.274 |
fin | +0.271 |
Neg logit contributions | |
She | -0.215 |
she | -0.210 |
day | -0.207 |
like | -0.185 |
sometimes | -0.185 |
Token | Tom |
Token ID | 13787 |
Feature activation | -4.16e-03 |
Pos logit contributions | |
itched | +0.269 |
tal | +0.260 |
eg | +0.250 |
rods | +0.247 |
ades | +0.243 |
Neg logit contributions | |
She | -0.188 |
she | -0.186 |
day | -0.184 |
always | -0.166 |
especially | -0.161 |
Token | ! |
Token ID | 0 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
itched | +0.011 |
tal | +0.011 |
eg | +0.010 |
orse | +0.010 |
izza | +0.010 |
Neg logit contributions | |
She | -0.005 |
day | -0.005 |
she | -0.005 |
like | -0.004 |
my | -0.004 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
itched | +0.289 |
tal | +0.280 |
eg | +0.259 |
ves | +0.249 |
orse | +0.248 |
Neg logit contributions | |
She | -0.189 |
she | -0.175 |
day | -0.174 |
like | -0.154 |
sometimes | -0.151 |
Token | ! |
Token ID | 0 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
She | +0.298 |
she | +0.292 |
day | +0.292 |
like | +0.264 |
always | +0.263 |
Neg logit contributions | |
itched | -0.350 |
tal | -0.336 |
eg | -0.315 |
orse | -0.297 |
izza | -0.291 |
Token | What |
Token ID | 1867 |
Feature activation | -9.71e-02 |
Pos logit contributions | |
itched | +0.286 |
tal | +0.275 |
eg | +0.260 |
orse | +0.247 |
izza | +0.242 |
Neg logit contributions | |
She | -0.182 |
she | -0.177 |
day | -0.177 |
like | -0.157 |
sometimes | -0.155 |
Token | are |
Token ID | 389 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
itched | +0.250 |
tal | +0.241 |
eg | +0.230 |
orse | +0.219 |
izza | +0.216 |
Neg logit contributions | |
She | -0.128 |
day | -0.124 |
she | -0.124 |
like | -0.107 |
sometimes | -0.106 |
Token | you |
Token ID | 345 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
She | +0.136 |
day | +0.130 |
my | +0.119 |
she | +0.113 |
especially | +0.111 |
Neg logit contributions | |
itched | -0.342 |
tal | -0.326 |
eg | -0.315 |
ves | -0.307 |
orse | -0.306 |
Token | doing |
Token ID | 1804 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
day | +0.177 |
she | +0.171 |
She | +0.170 |
always | +0.150 |
like | +0.147 |
Neg logit contributions | |
itched | -0.528 |
tal | -0.523 |
eg | -0.501 |
izza | -0.478 |
orse | -0.471 |
Token | ?" |
Token ID | 1701 |
Feature activation | -6.80e-02 |
Pos logit contributions | |
She | +0.183 |
day | +0.179 |
she | +0.179 |
like | +0.160 |
always | +0.159 |
Neg logit contributions | |
itched | -0.248 |
tal | -0.238 |
eg | -0.224 |
orse | -0.212 |
izza | -0.208 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
itched | +0.127 |
tal | +0.119 |
eg | +0.111 |
orse | +0.106 |
izza | +0.101 |
Neg logit contributions | |
day | -0.134 |
She | -0.133 |
she | -0.129 |
like | -0.123 |
always | -0.122 |
Token | said |
Token ID | 531 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
itched | +0.232 |
eg | +0.227 |
tal | +0.226 |
cannon | +0.209 |
izza | +0.208 |
Neg logit contributions | |
she | -0.150 |
She | -0.137 |
day | -0.134 |
always | -0.130 |
sometimes | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | -5.79e-02 |
Pos logit contributions | |
tal | +0.392 |
eg | +0.380 |
itched | +0.367 |
rods | +0.354 |
ades | +0.352 |
Neg logit contributions | |
she | -0.189 |
She | -0.173 |
like | -0.147 |
day | -0.146 |
her | -0.140 |
Token |
|
Token ID | 198 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
itched | +0.137 |
tal | +0.131 |
eg | +0.129 |
ves | +0.124 |
wounds | +0.123 |
Neg logit contributions | |
She | -0.106 |
she | -0.093 |
day | -0.082 |
always | -0.075 |
sometimes | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
tal | +0.360 |
urger | +0.356 |
itched | +0.353 |
fist | +0.345 |
eg | +0.345 |
Neg logit contributions | |
She | -0.173 |
day | -0.154 |
she | -0.148 |
She | -0.132 |
always | -0.123 |
Token | P |
Token ID | 47 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
itched | +0.329 |
tal | +0.319 |
eg | +0.304 |
izza | +0.281 |
orse | +0.278 |
Neg logit contributions | |
She | -0.196 |
she | -0.188 |
day | -0.187 |
sometimes | -0.170 |
always | -0.169 |
Token | ete |
Token ID | 14471 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
She | +0.060 |
day | +0.059 |
she | +0.059 |
always | +0.056 |
week | +0.055 |
Neg logit contributions | |
itched | -0.039 |
tal | -0.035 |
wounds | -0.033 |
eg | -0.032 |
fin | -0.032 |
Token | nodded |
Token ID | 14464 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
tal | +1.166 |
itched | +1.164 |
eg | +1.113 |
cks | +1.062 |
ades | +1.042 |
Neg logit contributions | |
always | -0.533 |
She | -0.533 |
she | -0.532 |
day | -0.506 |
sometimes | -0.493 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
tal | +0.603 |
eg | +0.593 |
itched | +0.588 |
izza | +0.566 |
engu | +0.548 |
Neg logit contributions | |
she | -0.295 |
She | -0.293 |
day | -0.277 |
like | -0.271 |
sometimes | -0.258 |
Token | Together |
Token ID | 17083 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
itched | +0.141 |
tal | +0.137 |
eg | +0.129 |
ves | +0.126 |
orse | +0.124 |
Neg logit contributions | |
She | -0.103 |
she | -0.089 |
day | -0.085 |
always | -0.077 |
sometimes | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
itched | +0.450 |
tal | +0.447 |
eg | +0.435 |
ades | +0.405 |
cks | +0.401 |
Neg logit contributions | |
she | -0.256 |
day | -0.253 |
She | -0.245 |
sometimes | -0.222 |
always | -0.206 |
Token | they |
Token ID | 484 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
itched | +0.137 |
tal | +0.133 |
eg | +0.132 |
urger | +0.121 |
ades | +0.121 |
Neg logit contributions | |
she | -0.069 |
day | -0.065 |
She | -0.064 |
sometimes | -0.057 |
like | -0.054 |
Token | special |
Token ID | 2041 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
itched | +0.407 |
tal | +0.392 |
eg | +0.371 |
orse | +0.352 |
izza | +0.346 |
Neg logit contributions | |
She | -0.253 |
day | -0.247 |
she | -0.247 |
like | -0.218 |
always | -0.217 |
Token | with |
Token ID | 351 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
itched | +0.062 |
tal | +0.060 |
eg | +0.056 |
orse | +0.053 |
fin | +0.052 |
Neg logit contributions | |
She | -0.051 |
she | -0.048 |
day | -0.048 |
especially | -0.044 |
my | -0.044 |
Token | him |
Token ID | 683 |
Feature activation | +4.22e-03 |
Pos logit contributions | |
itched | +0.142 |
tal | +0.137 |
eg | +0.129 |
orse | +0.123 |
izza | +0.119 |
Neg logit contributions | |
She | -0.107 |
day | -0.107 |
she | -0.106 |
like | -0.098 |
always | -0.095 |
Token | - |
Token ID | 532 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
she | +0.007 |
day | +0.007 |
She | +0.007 |
like | +0.006 |
always | +0.006 |
Neg logit contributions | |
itched | -0.010 |
tal | -0.010 |
eg | -0.009 |
orse | -0.009 |
izza | -0.008 |
Token | a |
Token ID | 257 |
Feature activation | -9.03e-02 |
Pos logit contributions | |
itched | +0.171 |
tal | +0.163 |
eg | +0.152 |
fin | +0.147 |
orse | +0.146 |
Neg logit contributions | |
She | -0.178 |
day | -0.175 |
she | -0.166 |
week | -0.161 |
weekend | -0.160 |
Token | hook |
Token ID | 8011 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
itched | +0.205 |
tal | +0.196 |
eg | +0.184 |
orse | +0.175 |
izza | +0.172 |
Neg logit contributions | |
She | -0.147 |
day | -0.146 |
she | -0.146 |
like | -0.129 |
always | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
itched | +0.898 |
tal | +0.861 |
eg | +0.809 |
orse | +0.764 |
izza | +0.749 |
Neg logit contributions | |
She | -0.712 |
day | -0.698 |
she | -0.698 |
like | -0.628 |
always | -0.625 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
itched | +0.150 |
tal | +0.146 |
urger | +0.136 |
wounds | +0.136 |
eg | +0.135 |
Neg logit contributions | |
She | -0.099 |
she | -0.081 |
day | -0.080 |
my | -0.075 |
like | -0.072 |
Token |
|
Token ID | 198 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
itched | +0.343 |
tal | +0.339 |
eg | +0.318 |
urger | +0.318 |
jaws | +0.310 |
Neg logit contributions | |
She | -0.210 |
day | -0.197 |
she | -0.191 |
my | -0.175 |
week | -0.165 |
Token | Sam |
Token ID | 16305 |
Feature activation | -7.93e-02 |
Pos logit contributions | |
itched | +0.322 |
tal | +0.312 |
eg | +0.290 |
izza | +0.275 |
orse | +0.274 |
Neg logit contributions | |
She | -0.219 |
day | -0.211 |
she | -0.208 |
like | -0.188 |
my | -0.186 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
itched | +0.266 |
tal | +0.260 |
eg | +0.248 |
orse | +0.241 |
izza | +0.235 |
Neg logit contributions | |
She | -0.036 |
she | -0.034 |
day | -0.033 |
like | -0.022 |
weekend | -0.021 |
Token | into |
Token ID | 656 |
Feature activation | -9.91e-02 |
Pos logit contributions | |
itched | +0.137 |
tal | +0.134 |
eg | +0.128 |
orse | +0.124 |
izza | +0.121 |
Neg logit contributions | |
She | -0.077 |
she | -0.076 |
day | -0.075 |
like | -0.072 |
sometimes | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
itched | +0.260 |
tal | +0.248 |
eg | +0.246 |
orse | +0.233 |
izza | +0.231 |
Neg logit contributions | |
she | -0.133 |
day | -0.132 |
She | -0.131 |
like | -0.116 |
her | -0.116 |
Token | pond |
Token ID | 16723 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
She | +0.041 |
she | +0.040 |
day | +0.039 |
always | +0.036 |
sometimes | +0.036 |
Neg logit contributions | |
itched | -0.054 |
tal | -0.053 |
eg | -0.050 |
orse | -0.046 |
fin | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
itched | +0.036 |
tal | +0.035 |
eg | +0.034 |
izza | +0.031 |
orse | +0.031 |
Neg logit contributions | |
day | -0.017 |
She | -0.017 |
she | -0.017 |
like | -0.015 |
sometimes | -0.015 |
Token | sw |
Token ID | 1509 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
she | +0.225 |
day | +0.217 |
She | +0.213 |
like | +0.202 |
sometimes | +0.198 |
Neg logit contributions | |
itched | -0.309 |
tal | -0.290 |
eg | -0.282 |
izza | -0.262 |
ades | -0.261 |
Token | am |
Token ID | 321 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
tal | +0.236 |
izza | +0.226 |
itched | +0.226 |
eg | +0.219 |
orse | +0.199 |
Neg logit contributions | |
day | -0.373 |
like | -0.363 |
She | -0.361 |
always | -0.361 |
ie | -0.360 |
Token | together |
Token ID | 1978 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
itched | +0.995 |
tal | +0.958 |
eg | +0.915 |
izza | +0.861 |
orse | +0.856 |
Neg logit contributions | |
She | -0.616 |
day | -0.615 |
she | -0.610 |
like | -0.556 |
sometimes | -0.535 |
Token | . |
Token ID | 13 |
Feature activation | +1.90e-03 |
Pos logit contributions | |
itched | +0.308 |
tal | +0.301 |
eg | +0.298 |
izza | +0.277 |
ades | +0.268 |
Neg logit contributions | |
day | -0.185 |
she | -0.184 |
She | -0.175 |
like | -0.170 |
sometimes | -0.156 |
Token | They |
Token ID | 1119 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
She | +0.003 |
she | +0.003 |
day | +0.003 |
sometimes | +0.002 |
like | +0.002 |
Neg logit contributions | |
itched | -0.005 |
tal | -0.005 |
eg | -0.004 |
urger | -0.004 |
ades | -0.004 |
Token | became |
Token ID | 2627 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
day | +0.160 |
she | +0.159 |
like | +0.154 |
always | +0.154 |
especially | +0.151 |
Neg logit contributions | |
itched | -0.282 |
tal | -0.268 |
eg | -0.253 |
izza | -0.247 |
ades | -0.238 |
Token | best |
Token ID | 1266 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
She | +0.054 |
she | +0.052 |
day | +0.052 |
sometimes | +0.046 |
always | +0.046 |
Neg logit contributions | |
itched | -0.081 |
tal | -0.078 |
eg | -0.073 |
orse | -0.070 |
fin | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
itched | +0.188 |
tal | +0.180 |
eg | +0.176 |
izza | +0.167 |
orse | +0.162 |
Neg logit contributions | |
day | -0.122 |
She | -0.118 |
she | -0.117 |
like | -0.107 |
sometimes | -0.103 |
Token | Together |
Token ID | 17083 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
itched | +0.122 |
tal | +0.117 |
eg | +0.112 |
ades | +0.105 |
fin | +0.105 |
Neg logit contributions | |
She | -0.095 |
she | -0.088 |
day | -0.080 |
always | -0.073 |
like | -0.071 |
Token | , |
Token ID | 11 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
itched | +0.418 |
tal | +0.411 |
eg | +0.390 |
orse | +0.369 |
izza | +0.362 |
Neg logit contributions | |
she | -0.274 |
She | -0.272 |
day | -0.271 |
sometimes | -0.237 |
always | -0.231 |
Token | they |
Token ID | 484 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
itched | +0.111 |
tal | +0.108 |
eg | +0.103 |
orse | +0.096 |
izza | +0.095 |
Neg logit contributions | |
she | -0.065 |
She | -0.064 |
day | -0.063 |
sometimes | -0.055 |
like | -0.055 |
Token | sw |
Token ID | 1509 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
itched | +0.031 |
tal | +0.030 |
eg | +0.027 |
izza | +0.027 |
wounds | +0.025 |
Neg logit contributions | |
day | -0.023 |
always | -0.022 |
like | -0.022 |
sometimes | -0.022 |
especially | -0.022 |
Token | am |
Token ID | 321 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
tal | +0.298 |
itched | +0.278 |
eg | +0.277 |
izza | +0.266 |
orse | +0.254 |
Neg logit contributions | |
day | -0.514 |
always | -0.500 |
She | -0.482 |
afternoon | -0.477 |
ie | -0.474 |
Token | off |
Token ID | 572 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
itched | +1.177 |
tal | +1.134 |
eg | +1.079 |
orse | +1.025 |
izza | +1.018 |
Neg logit contributions | |
She | -0.627 |
day | -0.625 |
she | -0.615 |
like | -0.551 |
sometimes | -0.538 |
Token | to |
Token ID | 284 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
itched | +0.501 |
tal | +0.491 |
eg | +0.484 |
urger | +0.471 |
izza | +0.468 |
Neg logit contributions | |
day | -0.254 |
she | -0.254 |
She | -0.246 |
like | -0.221 |
her | -0.207 |
Token | the |
Token ID | 262 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
itched | +0.178 |
eg | +0.165 |
tal | +0.164 |
izza | +0.160 |
ades | +0.159 |
Neg logit contributions | |
day | -0.093 |
she | -0.085 |
She | -0.083 |
like | -0.081 |
always | -0.078 |
Token | sea |
Token ID | 5417 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
itched | +0.068 |
tal | +0.064 |
eg | +0.059 |
izza | +0.056 |
ves | +0.054 |
Neg logit contributions | |
day | -0.061 |
she | -0.056 |
She | -0.054 |
sometimes | -0.053 |
like | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
day | +0.032 |
she | +0.029 |
like | +0.027 |
always | +0.026 |
sometimes | +0.025 |
Neg logit contributions | |
itched | -0.073 |
tal | -0.070 |
eg | -0.069 |
plug | -0.064 |
izza | -0.064 |
Token | their |
Token ID | 511 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
itched | +0.027 |
tal | +0.026 |
eg | +0.025 |
orse | +0.023 |
izza | +0.023 |
Neg logit contributions | |
She | -0.021 |
she | -0.021 |
day | -0.021 |
like | -0.019 |
always | -0.019 |
Token | dad |
Token ID | 9955 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
itched | +0.140 |
tal | +0.130 |
eg | +0.124 |
ves | +0.116 |
orse | +0.116 |
Neg logit contributions | |
day | -0.127 |
she | -0.112 |
She | -0.109 |
summer | -0.106 |
morning | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | -8.06e-03 |
Pos logit contributions | |
itched | +0.436 |
tal | +0.418 |
eg | +0.393 |
orse | +0.371 |
izza | +0.362 |
Neg logit contributions | |
She | -0.324 |
day | -0.323 |
she | -0.318 |
like | -0.289 |
sometimes | -0.286 |
Token | They |
Token ID | 1119 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
itched | +0.020 |
tal | +0.019 |
eg | +0.018 |
orse | +0.017 |
izza | +0.017 |
Neg logit contributions | |
she | -0.011 |
day | -0.011 |
She | -0.010 |
always | -0.009 |
my | -0.009 |
Token | have |
Token ID | 423 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
She | +0.125 |
she | +0.118 |
day | +0.111 |
weekend | +0.103 |
week | +0.101 |
Neg logit contributions | |
itched | -0.270 |
tal | -0.264 |
orse | -0.258 |
eg | -0.257 |
fin | -0.249 |
Token | rods |
Token ID | 32858 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
day | +0.173 |
She | +0.167 |
she | +0.167 |
always | +0.157 |
like | +0.154 |
Neg logit contributions | |
itched | -0.248 |
tal | -0.233 |
eg | -0.226 |
izza | -0.208 |
orse | -0.205 |
Token | and |
Token ID | 290 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
itched | +1.245 |
tal | +1.213 |
eg | +1.144 |
orse | +1.085 |
izza | +1.059 |
Neg logit contributions | |
day | -0.650 |
She | -0.647 |
she | -0.643 |
like | -0.588 |
sometimes | -0.569 |
Token | hooks |
Token ID | 26569 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
She | +0.256 |
day | +0.255 |
she | +0.255 |
like | +0.236 |
sometimes | +0.235 |
Neg logit contributions | |
itched | -0.221 |
tal | -0.206 |
eg | -0.192 |
orse | -0.175 |
izza | -0.172 |
Token | and |
Token ID | 290 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
itched | +0.984 |
tal | +0.973 |
eg | +0.930 |
orse | +0.867 |
urger | +0.857 |
Neg logit contributions | |
day | -0.498 |
she | -0.496 |
like | -0.494 |
She | -0.475 |
sometimes | -0.461 |
Token | worms |
Token ID | 29174 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
day | +0.234 |
she | +0.233 |
She | +0.230 |
like | +0.221 |
sometimes | +0.217 |
Neg logit contributions | |
itched | -0.244 |
tal | -0.224 |
eg | -0.212 |
ades | -0.192 |
orse | -0.191 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-02 |
Pos logit contributions | |
itched | +0.124 |
tal | +0.122 |
eg | +0.117 |
izza | +0.108 |
orse | +0.108 |
Neg logit contributions | |
she | -0.076 |
She | -0.074 |
day | -0.074 |
like | -0.069 |
sometimes | -0.067 |