Token | and |
Token ID | 290 |
Feature activation | +7.54e-01 |
Pos logit contributions | |
mine | +0.041 |
Grac | +0.037 |
berry | +0.037 |
ease | +0.036 |
oft | +0.036 |
Neg logit contributions | |
site | -0.027 |
office | -0.026 |
her | -0.026 |
mistake | -0.023 |
ia | -0.023 |
Token | that |
Token ID | 326 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
mine | +0.059 |
berry | +0.055 |
berries | +0.051 |
oft | +0.048 |
Grac | +0.048 |
Neg logit contributions | |
site | -0.056 |
office | -0.053 |
her | -0.051 |
ia | -0.049 |
wasteful | -0.048 |
Token | was |
Token ID | 373 |
Feature activation | -7.83e-01 |
Pos logit contributions | |
site | +0.183 |
mistake | +0.178 |
not | +0.171 |
men | +0.170 |
office | +0.169 |
Neg logit contributions | |
mine | -0.280 |
berry | -0.243 |
amb | -0.232 |
| | -0.230 |
oft | -0.229 |
Token | all |
Token ID | 477 |
Feature activation | +6.62e+00 |
Pos logit contributions | |
site | +0.058 |
her | +0.053 |
office | +0.052 |
ia | +0.050 |
wasteful | +0.046 |
Neg logit contributions | |
mine | -0.061 |
berry | -0.056 |
're | -0.055 |
wherever | -0.053 |
Benny | -0.053 |
Token | that |
Token ID | 326 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
mine | +0.607 |
berry | +0.558 |
berries | +0.532 |
're | +0.515 |
Grac | +0.500 |
Neg logit contributions | |
site | -0.412 |
her | -0.403 |
office | -0.390 |
wasteful | -0.356 |
old | -0.342 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.205 |
her | +0.190 |
not | +0.188 |
men | +0.186 |
wrong | +0.183 |
Neg logit contributions | |
mine | -0.247 |
berry | -0.231 |
amb | -0.209 |
superstar | -0.208 |
berries | -0.207 |
Token | . |
Token ID | 13 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
mine | +1.032 |
berry | +0.922 |
superstar | +0.921 |
oft | +0.911 |
g | +0.881 |
Neg logit contributions | |
site | -0.606 |
her | -0.584 |
office | -0.524 |
yourselves | -0.506 |
ia | -0.497 |
Token | The |
Token ID | 383 |
Feature activation | +9.14e-01 |
Pos logit contributions | |
mine | +0.011 |
berry | +0.009 |
g | +0.009 |
ease | +0.009 |
mattered | +0.009 |
Neg logit contributions | |
site | -0.008 |
her | -0.008 |
office | -0.008 |
ia | -0.007 |
mistake | -0.006 |
Token | moral |
Token ID | 6573 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
mine | +0.081 |
berry | +0.076 |
're | +0.075 |
Benny | +0.075 |
wherever | +0.074 |
Neg logit contributions | |
site | -0.057 |
her | -0.050 |
office | -0.049 |
ia | -0.048 |
wasteful | -0.042 |
Token | of |
Token ID | 286 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
mine | +0.014 |
Benny | +0.013 |
're | +0.013 |
berry | +0.013 |
wherever | +0.013 |
Neg logit contributions | |
site | -0.031 |
her | -0.028 |
office | -0.028 |
ia | -0.028 |
wasteful | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
mine | +0.090 |
're | +0.084 |
berry | +0.083 |
mattered | +0.080 |
ease | +0.079 |
Neg logit contributions | |
site | -0.087 |
her | -0.081 |
office | -0.080 |
ia | -0.075 |
mistake | -0.071 |
Token | the |
Token ID | 262 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
site | +0.200 |
office | +0.181 |
her | +0.176 |
mistake | +0.175 |
ia | +0.163 |
Neg logit contributions | |
mine | -0.231 |
berry | -0.211 |
're | -0.208 |
oft | -0.202 |
berries | -0.197 |
Token | mall |
Token ID | 17374 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
site | +0.012 |
her | +0.011 |
office | +0.011 |
mistake | +0.010 |
ia | +0.010 |
Neg logit contributions | |
mine | -0.014 |
Grac | -0.012 |
berry | -0.012 |
Will | -0.012 |
're | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +6.83e-02 |
Pos logit contributions | |
mine | +0.090 |
berry | +0.083 |
Benny | +0.082 |
're | +0.081 |
ease | +0.080 |
Neg logit contributions | |
site | -0.106 |
her | -0.097 |
office | -0.095 |
ia | -0.091 |
men | -0.084 |
Token | All |
Token ID | 1439 |
Feature activation | +6.00e+00 |
Pos logit contributions | |
mine | +0.005 |
berry | +0.004 |
g | +0.004 |
're | +0.004 |
berries | +0.004 |
Neg logit contributions | |
site | -0.006 |
her | -0.005 |
office | -0.005 |
ia | -0.005 |
wasteful | -0.004 |
Token | that |
Token ID | 326 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
mine | +0.452 |
berry | +0.423 |
're | +0.402 |
berries | +0.398 |
Grac | +0.393 |
Neg logit contributions | |
site | -0.428 |
office | -0.400 |
her | -0.393 |
ia | -0.388 |
men | -0.356 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.251 |
office | +0.238 |
mistake | +0.223 |
her | +0.223 |
ia | +0.217 |
Neg logit contributions | |
mine | -0.213 |
berry | -0.197 |
tie | -0.187 |
berries | -0.186 |
amb | -0.180 |
Token | was |
Token ID | 373 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
mine | +0.882 |
berry | +0.793 |
ease | +0.762 |
Grac | +0.754 |
oft | +0.754 |
Neg logit contributions | |
site | -0.775 |
her | -0.720 |
office | -0.704 |
ia | -0.670 |
mistake | -0.621 |
Token | that |
Token ID | 326 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
site | +0.058 |
her | +0.056 |
office | +0.053 |
ia | +0.050 |
wasteful | +0.046 |
Neg logit contributions | |
mine | -0.060 |
berry | -0.055 |
berries | -0.053 |
're | -0.052 |
oft | -0.051 |
Token | they |
Token ID | 484 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
site | +0.244 |
her | +0.223 |
office | +0.223 |
ia | +0.216 |
mistake | +0.210 |
Neg logit contributions | |
mine | -0.215 |
berry | -0.209 |
berries | -0.193 |
oft | -0.191 |
superstar | -0.181 |
Token | were |
Token ID | 547 |
Feature activation | +5.02e+00 |
Pos logit contributions | |
her | +0.002 |
site | +0.002 |
wasteful | +0.002 |
not | +0.002 |
ia | +0.002 |
Neg logit contributions | |
berry | -0.002 |
berries | -0.002 |
Nice | -0.002 |
mine | -0.002 |
bird | -0.002 |
Token | together |
Token ID | 1978 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
mine | +0.375 |
berry | +0.349 |
berries | +0.335 |
're | +0.333 |
Benny | +0.331 |
Neg logit contributions | |
site | -0.377 |
her | -0.360 |
office | -0.348 |
ia | -0.321 |
men | -0.308 |
Token | They |
Token ID | 1119 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
mine | +0.000 |
're | +0.000 |
berry | +0.000 |
ease | +0.000 |
wherever | +0.000 |
Neg logit contributions | |
site | -0.000 |
ia | -0.000 |
her | -0.000 |
office | -0.000 |
urse | -0.000 |
Token | both |
Token ID | 1111 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
ia | +0.128 |
her | +0.124 |
not | +0.106 |
office | +0.106 |
site | +0.105 |
Neg logit contributions | |
berries | -0.154 |
berry | -0.154 |
Nice | -0.151 |
mine | -0.139 |
Emmy | -0.135 |
Token | learned |
Token ID | 4499 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
berries | +0.296 |
berry | +0.288 |
mine | +0.272 |
Grac | +0.263 |
Emmy | +0.261 |
Neg logit contributions | |
ia | -0.279 |
her | -0.260 |
site | -0.245 |
office | -0.236 |
ena | -0.235 |
Token | that |
Token ID | 326 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
mine | +0.204 |
berry | +0.190 |
Benny | +0.189 |
're | +0.188 |
wherever | +0.188 |
Neg logit contributions | |
site | -0.160 |
her | -0.142 |
office | -0.139 |
ia | -0.136 |
wasteful | -0.120 |
Token | what |
Token ID | 644 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
site | +0.221 |
office | +0.203 |
mistake | +0.200 |
her | +0.193 |
men | +0.183 |
Neg logit contributions | |
mine | -0.283 |
berry | -0.244 |
're | -0.235 |
oft | -0.233 |
tie | -0.229 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.277 |
office | +0.259 |
ia | +0.256 |
her | +0.251 |
mistake | +0.245 |
Neg logit contributions | |
mine | -0.378 |
berry | -0.335 |
ease | -0.324 |
oft | -0.307 |
berries | -0.305 |
Token | was |
Token ID | 373 |
Feature activation | -8.69e-01 |
Pos logit contributions | |
mine | +0.986 |
berry | +0.910 |
berries | +0.855 |
ease | +0.853 |
superstar | +0.843 |
Neg logit contributions | |
site | -0.662 |
her | -0.648 |
office | -0.604 |
ia | -0.585 |
mistake | -0.555 |
Token | being |
Token ID | 852 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
site | +0.055 |
her | +0.055 |
office | +0.051 |
ia | +0.049 |
men | +0.045 |
Neg logit contributions | |
mine | -0.078 |
berry | -0.073 |
berries | -0.069 |
g | -0.067 |
oft | -0.065 |
Token | nice |
Token ID | 3621 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
her | +0.058 |
site | +0.057 |
office | +0.051 |
wasteful | +0.049 |
ia | +0.047 |
Neg logit contributions | |
mine | -0.097 |
berry | -0.091 |
berries | -0.087 |
ease | -0.085 |
oft | -0.083 |
Token | and |
Token ID | 290 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
site | +0.093 |
her | +0.089 |
office | +0.089 |
ia | +0.086 |
mistake | +0.077 |
Neg logit contributions | |
mine | -0.119 |
berry | -0.105 |
ease | -0.104 |
Grac | -0.102 |
g | -0.102 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
mine | +0.058 |
berry | +0.055 |
berries | +0.051 |
ease | +0.048 |
're | +0.047 |
Neg logit contributions | |
her | -0.035 |
site | -0.034 |
office | -0.031 |
wasteful | -0.031 |
yourselves | -0.028 |
Token | organized |
Token ID | 8389 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
site | +0.058 |
her | +0.052 |
office | +0.052 |
ia | +0.051 |
men | +0.046 |
Neg logit contributions | |
mine | -0.052 |
Benny | -0.048 |
wherever | -0.048 |
're | -0.048 |
berry | -0.048 |
Token | and |
Token ID | 290 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
site | +0.063 |
her | +0.059 |
office | +0.058 |
ia | +0.055 |
mistake | +0.051 |
Neg logit contributions | |
mine | -0.061 |
berry | -0.057 |
're | -0.055 |
Benny | -0.054 |
ease | -0.054 |
Token | that |
Token ID | 326 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
mine | +0.041 |
berry | +0.038 |
're | +0.036 |
Benny | +0.036 |
wherever | +0.035 |
Neg logit contributions | |
site | -0.058 |
her | -0.054 |
office | -0.053 |
ia | -0.052 |
wasteful | -0.048 |
Token | 's |
Token ID | 338 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
site | +0.245 |
office | +0.232 |
her | +0.230 |
ia | +0.218 |
mistake | +0.212 |
Neg logit contributions | |
mine | -0.221 |
berry | -0.205 |
Benny | -0.191 |
berries | -0.190 |
oft | -0.190 |
Token | what |
Token ID | 644 |
Feature activation | -4.28e+00 |
Pos logit contributions | |
mine | +0.104 |
wherever | +0.099 |
Benny | +0.097 |
're | +0.097 |
mattered | +0.096 |
Neg logit contributions | |
site | -0.104 |
her | -0.090 |
office | -0.088 |
ia | -0.088 |
men | -0.081 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.317 |
office | +0.303 |
her | +0.296 |
ia | +0.293 |
mistake | +0.275 |
Neg logit contributions | |
mine | -0.312 |
berry | -0.289 |
ease | -0.270 |
Benny | -0.268 |
oft | -0.266 |
Token | most |
Token ID | 749 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
mine | +0.873 |
berry | +0.816 |
ease | +0.773 |
g | +0.771 |
superstar | +0.768 |
Neg logit contributions | |
site | -0.755 |
her | -0.729 |
office | -0.704 |
ia | -0.662 |
mistake | -0.608 |
Token | . |
Token ID | 13 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
mine | +0.279 |
're | +0.249 |
berry | +0.248 |
Benny | +0.243 |
Grac | +0.241 |
Neg logit contributions | |
site | -0.176 |
her | -0.162 |
office | -0.157 |
ia | -0.151 |
mistake | -0.141 |
Token | Since |
Token ID | 4619 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
mine | +0.006 |
berry | +0.005 |
ease | +0.005 |
wherever | +0.005 |
're | +0.005 |
Neg logit contributions | |
site | -0.006 |
her | -0.006 |
ia | -0.006 |
office | -0.006 |
yourselves | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
mine | +0.152 |
berry | +0.141 |
Benny | +0.139 |
're | +0.139 |
wherever | +0.138 |
Neg logit contributions | |
site | -0.132 |
her | -0.118 |
office | -0.115 |
ia | -0.113 |
wasteful | -0.101 |
Token | couldn |
Token ID | 3521 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
mine | +0.092 |
berry | +0.085 |
Benny | +0.083 |
wherever | +0.083 |
're | +0.083 |
Neg logit contributions | |
site | -0.090 |
her | -0.083 |
office | -0.081 |
ia | -0.079 |
wasteful | -0.071 |
Token | and |
Token ID | 290 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
mine | +0.163 |
're | +0.145 |
ease | +0.145 |
berry | +0.145 |
Benny | +0.142 |
Neg logit contributions | |
site | -0.162 |
office | -0.151 |
her | -0.144 |
ia | -0.143 |
mistake | -0.138 |
Token | that |
Token ID | 326 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
mine | +0.047 |
berry | +0.044 |
berries | +0.041 |
're | +0.039 |
Benny | +0.039 |
Neg logit contributions | |
site | -0.054 |
her | -0.052 |
office | -0.051 |
ia | -0.049 |
wasteful | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
office | +0.244 |
site | +0.244 |
her | +0.228 |
ia | +0.220 |
mistake | +0.216 |
Neg logit contributions | |
mine | -0.230 |
berry | -0.207 |
tie | -0.196 |
oft | -0.195 |
berries | -0.195 |
Token | all |
Token ID | 477 |
Feature activation | +6.57e+00 |
Pos logit contributions | |
site | +0.061 |
her | +0.057 |
office | +0.055 |
ia | +0.054 |
wasteful | +0.049 |
Neg logit contributions | |
mine | -0.057 |
berry | -0.053 |
're | -0.052 |
Benny | -0.051 |
wherever | -0.050 |
Token | that |
Token ID | 326 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
mine | +0.560 |
berry | +0.522 |
berries | +0.498 |
're | +0.483 |
Benny | +0.480 |
Neg logit contributions | |
site | -0.425 |
her | -0.418 |
office | -0.414 |
wasteful | -0.361 |
mistake | -0.357 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.240 |
office | +0.232 |
her | +0.228 |
ia | +0.218 |
mistake | +0.213 |
Neg logit contributions | |
mine | -0.226 |
berry | -0.212 |
berries | -0.198 |
oft | -0.197 |
tie | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | +5.62e-02 |
Pos logit contributions | |
mine | +0.970 |
berry | +0.880 |
oft | +0.860 |
ease | +0.857 |
superstar | +0.842 |
Neg logit contributions | |
site | -0.660 |
her | -0.635 |
office | -0.622 |
ia | -0.571 |
yourselves | -0.546 |
Token | And |
Token ID | 843 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
mine | +0.004 |
stuffed | +0.004 |
berry | +0.004 |
're | +0.004 |
mattered | +0.004 |
Neg logit contributions | |
site | -0.004 |
office | -0.004 |
her | -0.004 |
ia | -0.004 |
urse | -0.003 |
Token | so |
Token ID | 523 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
site | +0.015 |
office | +0.014 |
her | +0.014 |
ia | +0.014 |
mistake | +0.013 |
Neg logit contributions | |
mine | -0.013 |
berry | -0.013 |
berries | -0.012 |
ease | -0.012 |
're | -0.012 |
Token | she |
Token ID | 673 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
site | +0.094 |
her | +0.087 |
office | +0.087 |
ia | +0.086 |
men | +0.077 |
Neg logit contributions | |
mine | -0.078 |
berry | -0.072 |
're | -0.069 |
ease | -0.069 |
Benny | -0.068 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.03e+00 |
Pos logit contributions | |
mine | +0.085 |
berry | +0.080 |
berries | +0.080 |
ease | +0.079 |
wherever | +0.079 |
Neg logit contributions | |
site | -0.085 |
ia | -0.084 |
her | -0.084 |
office | -0.081 |
mistake | -0.073 |
Token | And |
Token ID | 843 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
mine | +0.016 |
berry | +0.014 |
're | +0.014 |
ease | +0.014 |
g | +0.014 |
Neg logit contributions | |
site | -0.017 |
her | -0.017 |
office | -0.016 |
ia | -0.016 |
wasteful | -0.014 |
Token | that |
Token ID | 326 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
site | +0.021 |
office | +0.020 |
her | +0.020 |
ia | +0.020 |
mistake | +0.018 |
Neg logit contributions | |
mine | -0.019 |
berry | -0.018 |
're | -0.017 |
ease | -0.017 |
berries | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
site | +0.209 |
office | +0.209 |
her | +0.204 |
ia | +0.192 |
mistake | +0.191 |
Neg logit contributions | |
mine | -0.253 |
berry | -0.229 |
ease | -0.222 |
berries | -0.221 |
tie | -0.215 |
Token | what |
Token ID | 644 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
mine | +0.095 |
Benny | +0.089 |
wherever | +0.089 |
berry | +0.088 |
're | +0.087 |
Neg logit contributions | |
site | -0.106 |
her | -0.093 |
office | -0.091 |
ia | -0.091 |
men | -0.084 |
Token | really |
Token ID | 1107 |
Feature activation | -5.57e-01 |
Pos logit contributions | |
office | +0.275 |
site | +0.260 |
ia | +0.258 |
her | +0.254 |
mistake | +0.243 |
Neg logit contributions | |
mine | -0.357 |
berry | -0.326 |
ease | -0.305 |
hello | -0.302 |
berries | -0.302 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.040 |
her | +0.037 |
office | +0.036 |
ia | +0.035 |
wasteful | +0.032 |
Neg logit contributions | |
mine | -0.043 |
berry | -0.040 |
're | -0.039 |
wherever | -0.039 |
Benny | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
mine | +0.911 |
berry | +0.842 |
ease | +0.810 |
oft | +0.800 |
berries | +0.796 |
Neg logit contributions | |
site | -0.717 |
her | -0.696 |
office | -0.692 |
ia | -0.633 |
mistake | -0.593 |
Token | The |
Token ID | 383 |
Feature activation | +8.05e-01 |
Pos logit contributions | |
mine | +0.020 |
're | +0.018 |
berry | +0.017 |
ease | +0.017 |
g | +0.017 |
Neg logit contributions | |
site | -0.018 |
her | -0.017 |
ia | -0.017 |
office | -0.017 |
urse | -0.014 |
Token | end |
Token ID | 886 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
mine | +0.079 |
berry | +0.073 |
're | +0.072 |
Benny | +0.072 |
wherever | +0.071 |
Neg logit contributions | |
site | -0.043 |
office | -0.039 |
her | -0.038 |
ia | -0.036 |
mistake | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
site | +0.190 |
her | +0.173 |
office | +0.169 |
ia | +0.166 |
wasteful | +0.153 |
Neg logit contributions | |
mine | -0.157 |
berry | -0.146 |
Benny | -0.143 |
're | -0.142 |
wherever | -0.142 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
mine | +0.017 |
berry | +0.015 |
're | +0.015 |
ease | +0.015 |
g | +0.015 |
Neg logit contributions | |
site | -0.015 |
her | -0.015 |
office | -0.014 |
ia | -0.014 |
urse | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
site | +0.222 |
her | +0.206 |
office | +0.201 |
ia | +0.197 |
wasteful | +0.185 |
Neg logit contributions | |
mine | -0.133 |
Benny | -0.127 |
berry | -0.125 |
wherever | -0.121 |
're | -0.120 |
Token | that |
Token ID | 326 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
site | +0.025 |
office | +0.022 |
her | +0.022 |
men | +0.020 |
ia | +0.020 |
Neg logit contributions | |
mine | -0.025 |
berry | -0.023 |
berries | -0.021 |
oft | -0.021 |
're | -0.021 |
Token | 's |
Token ID | 338 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
site | +0.222 |
office | +0.215 |
her | +0.207 |
mistake | +0.201 |
ia | +0.180 |
Neg logit contributions | |
mine | -0.253 |
berry | -0.232 |
berries | -0.226 |
tie | -0.219 |
Grac | -0.217 |
Token | all |
Token ID | 477 |
Feature activation | +6.58e+00 |
Pos logit contributions | |
mine | +0.099 |
wherever | +0.093 |
Benny | +0.093 |
berry | +0.091 |
're | +0.091 |
Neg logit contributions | |
site | -0.104 |
her | -0.092 |
office | -0.089 |
ia | -0.089 |
men | -0.081 |
Token | that |
Token ID | 326 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
mine | +0.510 |
berry | +0.495 |
berries | +0.480 |
Grac | +0.448 |
ease | +0.446 |
Neg logit contributions | |
site | -0.458 |
office | -0.449 |
her | -0.430 |
wasteful | -0.396 |
ia | -0.393 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.254 |
office | +0.232 |
her | +0.222 |
mistake | +0.221 |
men | +0.213 |
Neg logit contributions | |
mine | -0.229 |
berry | -0.218 |
berries | -0.201 |
tie | -0.195 |
amb | -0.192 |
Token | . |
Token ID | 13 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
mine | +0.955 |
berry | +0.866 |
berries | +0.840 |
Grac | +0.812 |
oft | +0.807 |
Neg logit contributions | |
site | -0.698 |
her | -0.657 |
office | -0.648 |
ia | -0.574 |
mistake | -0.573 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
mine | +0.016 |
berry | +0.014 |
superstar | +0.014 |
ease | +0.013 |
tie | +0.013 |
Neg logit contributions | |
site | -0.012 |
her | -0.011 |
office | -0.011 |
ia | -0.010 |
wasteful | -0.009 |
Token | Once |
Token ID | 7454 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
site | +0.121 |
her | +0.111 |
ia | +0.106 |
office | +0.106 |
men | +0.093 |
Neg logit contributions | |
mine | -0.111 |
're | -0.101 |
berry | -0.101 |
g | -0.097 |
mattered | -0.097 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
mine | +0.311 |
berry | +0.281 |
're | +0.274 |
Benny | +0.272 |
wherever | +0.270 |
Neg logit contributions | |
site | -0.450 |
her | -0.415 |
office | -0.408 |
ia | -0.406 |
men | -0.368 |
Token | a |
Token ID | 257 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
mine | +0.113 |
berry | +0.106 |
Grac | +0.103 |
're | +0.102 |
berries | +0.101 |
Neg logit contributions | |
site | -0.078 |
her | -0.074 |
ia | -0.069 |
office | -0.069 |
x | -0.060 |
Token | peaceful |
Token ID | 12309 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
site | +0.098 |
her | +0.090 |
office | +0.088 |
ia | +0.087 |
wasteful | +0.080 |
Neg logit contributions | |
mine | -0.074 |
berry | -0.067 |
Benny | -0.065 |
're | -0.065 |
wherever | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
mine | +0.073 |
berry | +0.067 |
're | +0.067 |
ease | +0.065 |
Benny | +0.065 |
Neg logit contributions | |
site | -0.076 |
her | -0.069 |
office | -0.068 |
ia | -0.065 |
men | -0.060 |
Token | like |
Token ID | 588 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
site | +0.028 |
her | +0.026 |
office | +0.025 |
ia | +0.023 |
urgent | +0.023 |
Neg logit contributions | |
mine | -0.028 |
berry | -0.026 |
're | -0.024 |
berries | -0.024 |
oft | -0.024 |
Token | nothing |
Token ID | 2147 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
mine | +0.260 |
berry | +0.231 |
're | +0.227 |
oft | +0.217 |
berries | +0.216 |
Neg logit contributions | |
site | -0.247 |
office | -0.224 |
her | -0.220 |
ia | -0.210 |
men | -0.203 |
Token | else |
Token ID | 2073 |
Feature activation | -6.53e-01 |
Pos logit contributions | |
site | +0.010 |
her | +0.009 |
ia | +0.009 |
office | +0.009 |
men | +0.008 |
Neg logit contributions | |
mine | -0.010 |
berry | -0.010 |
berries | -0.009 |
ease | -0.009 |
're | -0.009 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.046 |
her | +0.043 |
office | +0.042 |
ia | +0.041 |
men | +0.038 |
Neg logit contributions | |
mine | -0.052 |
berry | -0.046 |
ease | -0.045 |
Grac | -0.044 |
berries | -0.043 |
Token | except |
Token ID | 2845 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
mine | +0.988 |
berry | +0.896 |
Grac | +0.882 |
oft | +0.876 |
ease | +0.874 |
Neg logit contributions | |
site | -0.645 |
her | -0.584 |
office | -0.563 |
ia | -0.548 |
yourselves | -0.488 |
Token | the |
Token ID | 262 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
mine | +0.124 |
berry | +0.109 |
oft | +0.106 |
're | +0.106 |
berries | +0.104 |
Neg logit contributions | |
site | -0.099 |
office | -0.093 |
her | -0.085 |
mistake | -0.082 |
ia | -0.081 |
Token | present |
Token ID | 1944 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
site | +0.022 |
her | +0.020 |
office | +0.020 |
ia | +0.019 |
men | +0.018 |
Neg logit contributions | |
mine | -0.027 |
berry | -0.023 |
Benny | -0.023 |
Grac | -0.023 |
oft | -0.023 |
Token | moment |
Token ID | 2589 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
mine | +0.237 |
berry | +0.217 |
Benny | +0.211 |
're | +0.210 |
ease | +0.209 |
Neg logit contributions | |
site | -0.281 |
her | -0.259 |
office | -0.255 |
ia | -0.249 |
men | -0.227 |
Token | . |
Token ID | 13 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
mine | +0.165 |
berry | +0.150 |
're | +0.145 |
ease | +0.142 |
Grac | +0.142 |
Neg logit contributions | |
site | -0.171 |
her | -0.160 |
ia | -0.153 |
office | -0.152 |
men | -0.137 |
Token | was |
Token ID | 373 |
Feature activation | -9.59e-01 |
Pos logit contributions | |
mine | +0.232 |
berry | +0.218 |
Benny | +0.207 |
wherever | +0.206 |
Grac | +0.206 |
Neg logit contributions | |
site | -0.271 |
her | -0.255 |
office | -0.244 |
ia | -0.240 |
men | -0.223 |
Token | the |
Token ID | 262 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
site | +0.066 |
her | +0.059 |
office | +0.058 |
ia | +0.057 |
men | +0.050 |
Neg logit contributions | |
mine | -0.077 |
Benny | -0.072 |
berry | -0.071 |
're | -0.071 |
wherever | -0.071 |
Token | only |
Token ID | 691 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
site | +0.022 |
office | +0.020 |
her | +0.020 |
mistake | +0.018 |
ia | +0.018 |
Neg logit contributions | |
mine | -0.027 |
berry | -0.024 |
're | -0.024 |
wherever | -0.023 |
Benny | -0.023 |
Token | one |
Token ID | 530 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
site | +0.053 |
her | +0.048 |
office | +0.047 |
ia | +0.046 |
men | +0.042 |
Neg logit contributions | |
mine | -0.049 |
berry | -0.045 |
're | -0.044 |
Benny | -0.043 |
wherever | -0.043 |
Token | that |
Token ID | 326 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
mine | +0.333 |
berry | +0.297 |
ease | +0.294 |
're | +0.294 |
Grac | +0.291 |
Neg logit contributions | |
site | -0.322 |
her | -0.297 |
office | -0.293 |
ia | -0.284 |
men | -0.259 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.10e+01 |
Pos logit contributions | |
site | +0.258 |
her | +0.240 |
office | +0.231 |
ia | +0.223 |
men | +0.219 |
Neg logit contributions | |
mine | -0.239 |
berry | -0.224 |
berries | -0.206 |
oft | -0.204 |
Grac | -0.200 |
Token | . |
Token ID | 13 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
mine | +0.936 |
berry | +0.866 |
oft | +0.830 |
Grac | +0.819 |
superstar | +0.817 |
Neg logit contributions | |
site | -0.704 |
her | -0.662 |
office | -0.631 |
ia | -0.600 |
mistake | -0.565 |
Token | He |
Token ID | 679 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
site | +0.002 |
her | +0.002 |
office | +0.002 |
ia | +0.002 |
wasteful | +0.002 |
Neg logit contributions | |
mine | -0.001 |
berry | -0.001 |
're | -0.001 |
wherever | -0.001 |
ease | -0.001 |
Token | demanded |
Token ID | 12284 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
site | +0.124 |
her | +0.118 |
ia | +0.113 |
office | +0.111 |
mistake | +0.103 |
Neg logit contributions | |
mine | -0.113 |
berry | -0.104 |
wherever | -0.103 |
're | -0.101 |
ease | -0.100 |
Token | that |
Token ID | 326 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
site | +0.242 |
her | +0.216 |
office | +0.212 |
ia | +0.207 |
wasteful | +0.185 |
Neg logit contributions | |
mine | -0.257 |
Benny | -0.241 |
berry | -0.236 |
're | -0.236 |
wherever | -0.235 |
Token | everyone |
Token ID | 2506 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
site | +0.246 |
office | +0.221 |
mistake | +0.209 |
her | +0.205 |
men | +0.204 |
Neg logit contributions | |
mine | -0.249 |
berry | -0.230 |
oft | -0.230 |
're | -0.228 |
berries | -0.221 |
Token | and |
Token ID | 290 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
her | +0.035 |
site | +0.033 |
office | +0.032 |
ia | +0.030 |
mistake | +0.028 |
Neg logit contributions | |
mine | -0.048 |
berry | -0.045 |
oft | -0.045 |
ease | -0.044 |
hello | -0.044 |
Token | that |
Token ID | 326 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
mine | +0.046 |
berry | +0.043 |
berries | +0.039 |
're | +0.038 |
oft | +0.038 |
Neg logit contributions | |
site | -0.044 |
office | -0.043 |
her | -0.040 |
mistake | -0.039 |
wasteful | -0.038 |
Token | 's |
Token ID | 338 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
office | +0.213 |
mistake | +0.202 |
site | +0.201 |
her | +0.194 |
not | +0.189 |
Neg logit contributions | |
mine | -0.261 |
berry | -0.240 |
amb | -0.234 |
tie | -0.232 |
oft | -0.225 |
Token | all |
Token ID | 477 |
Feature activation | +6.54e+00 |
Pos logit contributions | |
mine | +0.108 |
Benny | +0.100 |
berry | +0.100 |
wherever | +0.100 |
're | +0.099 |
Neg logit contributions | |
site | -0.091 |
her | -0.080 |
office | -0.078 |
ia | -0.077 |
men | -0.069 |
Token | that |
Token ID | 326 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
mine | +0.574 |
berry | +0.547 |
berries | +0.517 |
're | +0.495 |
ease | +0.488 |
Neg logit contributions | |
her | -0.400 |
office | -0.384 |
site | -0.380 |
wasteful | -0.358 |
wrong | -0.345 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
site | +0.208 |
her | +0.207 |
office | +0.205 |
not | +0.199 |
wasteful | +0.188 |
Neg logit contributions | |
mine | -0.242 |
berry | -0.238 |
amb | -0.227 |
tie | -0.218 |
oft | -0.215 |
Token | . |
Token ID | 13 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
mine | +0.995 |
berry | +0.917 |
ease | +0.887 |
oft | +0.883 |
superstar | +0.873 |
Neg logit contributions | |
her | -0.598 |
site | -0.572 |
office | -0.551 |
ia | -0.522 |
yourselves | -0.505 |
Token | From |
Token ID | 3574 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
site | +0.002 |
her | +0.002 |
ia | +0.002 |
office | +0.002 |
urse | +0.002 |
Neg logit contributions | |
mine | -0.002 |
berry | -0.002 |
're | -0.002 |
ease | -0.002 |
g | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
Benny | +0.087 |
mine | +0.084 |
wherever | +0.083 |
Emmy | +0.079 |
Ros | +0.075 |
Neg logit contributions | |
site | -0.115 |
office | -0.098 |
ia | -0.096 |
her | -0.095 |
men | -0.092 |
Token | day |
Token ID | 1110 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
office | +0.183 |
mistake | +0.175 |
site | +0.173 |
her | +0.164 |
ia | +0.156 |
Neg logit contributions | |
mine | -0.277 |
berries | -0.262 |
berry | -0.261 |
als | -0.260 |
ease | -0.256 |
Token | on |
Token ID | 319 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
mine | +0.045 |
berry | +0.041 |
Benny | +0.040 |
're | +0.040 |
wherever | +0.039 |
Neg logit contributions | |
site | -0.066 |
her | -0.061 |
office | -0.060 |
ia | -0.059 |
men | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
mine | +0.208 |
're | +0.183 |
ease | +0.180 |
berry | +0.179 |
oft | +0.178 |
Neg logit contributions | |
site | -0.139 |
office | -0.138 |
mistake | -0.129 |
her | -0.128 |
ia | -0.126 |
Token | that |
Token ID | 326 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
mine | +0.059 |
berry | +0.056 |
berries | +0.051 |
oft | +0.048 |
superstar | +0.047 |
Neg logit contributions | |
site | -0.045 |
office | -0.044 |
her | -0.043 |
wasteful | -0.041 |
mistake | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
office | +0.212 |
site | +0.212 |
her | +0.204 |
mistake | +0.194 |
wasteful | +0.190 |
Neg logit contributions | |
mine | -0.262 |
berry | -0.245 |
oft | -0.222 |
amb | -0.221 |
berries | -0.221 |
Token | all |
Token ID | 477 |
Feature activation | +6.61e+00 |
Pos logit contributions | |
site | +0.053 |
her | +0.051 |
office | +0.048 |
ia | +0.046 |
wasteful | +0.043 |
Neg logit contributions | |
mine | -0.068 |
berry | -0.063 |
're | -0.061 |
Benny | -0.060 |
berries | -0.060 |
Token | that |
Token ID | 326 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
mine | +0.614 |
berry | +0.575 |
berries | +0.543 |
're | +0.533 |
ease | +0.519 |
Neg logit contributions | |
her | -0.389 |
site | -0.376 |
office | -0.375 |
wasteful | -0.346 |
old | -0.335 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
her | +0.222 |
site | +0.221 |
office | +0.212 |
mistake | +0.203 |
wasteful | +0.202 |
Neg logit contributions | |
mine | -0.239 |
berry | -0.233 |
berries | -0.208 |
amb | -0.205 |
oft | -0.199 |
Token | . |
Token ID | 13 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
mine | +0.992 |
berry | +0.917 |
oft | +0.870 |
ease | +0.869 |
superstar | +0.865 |
Neg logit contributions | |
her | -0.611 |
site | -0.606 |
office | -0.568 |
ia | -0.539 |
yourselves | -0.530 |
Token | From |
Token ID | 3574 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
mine | +0.003 |
're | +0.002 |
berry | +0.002 |
g | +0.002 |
ease | +0.002 |
Neg logit contributions | |
site | -0.003 |
her | -0.003 |
ia | -0.003 |
office | -0.003 |
wasteful | -0.002 |
Token | that |
Token ID | 326 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
Benny | +0.090 |
mine | +0.089 |
wherever | +0.087 |
Emmy | +0.082 |
berry | +0.078 |
Neg logit contributions | |
site | -0.113 |
office | -0.097 |
ia | -0.096 |
her | -0.095 |
urse | -0.090 |
Token | day |
Token ID | 1110 |
Feature activation | +7.46e-01 |
Pos logit contributions | |
office | +0.183 |
site | +0.178 |
her | +0.166 |
mistake | +0.165 |
ia | +0.158 |
Neg logit contributions | |
mine | -0.276 |
berries | -0.263 |
berry | -0.260 |
ease | -0.258 |
als | -0.255 |
Token | on |
Token ID | 319 |
Feature activation | +8.57e-01 |
Pos logit contributions | |
mine | +0.045 |
berry | +0.042 |
Benny | +0.040 |
're | +0.040 |
wherever | +0.040 |
Neg logit contributions | |
site | -0.066 |
her | -0.061 |
office | -0.060 |
ia | -0.059 |
men | -0.054 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
mine | +0.041 |
berry | +0.037 |
Benny | +0.036 |
're | +0.036 |
wherever | +0.036 |
Neg logit contributions | |
site | -0.059 |
her | -0.055 |
office | -0.054 |
ia | -0.053 |
wasteful | -0.049 |
Token | 't |
Token ID | 470 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
mine | +0.059 |
Benny | +0.052 |
're | +0.051 |
berry | +0.050 |
ease | +0.049 |
Neg logit contributions | |
site | -0.118 |
office | -0.109 |
her | -0.107 |
ia | -0.106 |
wasteful | -0.101 |
Token | understand |
Token ID | 1833 |
Feature activation | +2.40e+00 |
Pos logit contributions | |
mine | +0.173 |
berry | +0.167 |
wherever | +0.159 |
berries | +0.159 |
ease | +0.153 |
Neg logit contributions | |
site | -0.208 |
her | -0.196 |
office | -0.193 |
ia | -0.192 |
mistake | -0.185 |
Token | why |
Token ID | 1521 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
mine | +0.198 |
berry | +0.185 |
wherever | +0.183 |
Benny | +0.183 |
're | +0.181 |
Neg logit contributions | |
site | -0.162 |
her | -0.146 |
office | -0.139 |
ia | -0.138 |
wasteful | -0.122 |
Token | it |
Token ID | 340 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
site | +0.400 |
her | +0.365 |
office | +0.364 |
ia | +0.356 |
men | +0.329 |
Neg logit contributions | |
mine | -0.291 |
berry | -0.263 |
're | -0.259 |
wherever | -0.249 |
ease | -0.248 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
mine | +0.050 |
berry | +0.046 |
wherever | +0.045 |
ease | +0.044 |
berries | +0.043 |
Neg logit contributions | |
site | -0.052 |
her | -0.049 |
ia | -0.047 |
office | -0.046 |
mistake | -0.042 |
Token | so |
Token ID | 523 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
mine | +0.967 |
berry | +0.886 |
superstar | +0.865 |
're | +0.843 |
oft | +0.842 |
Neg logit contributions | |
site | -0.664 |
her | -0.619 |
office | -0.607 |
ia | -0.552 |
mistake | -0.535 |
Token | much |
Token ID | 881 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
site | +0.094 |
her | +0.086 |
office | +0.084 |
ia | +0.084 |
men | +0.076 |
Neg logit contributions | |
mine | -0.087 |
berry | -0.081 |
're | -0.079 |
wherever | -0.077 |
Benny | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
site | +0.123 |
her | +0.112 |
office | +0.110 |
ia | +0.108 |
men | +0.098 |
Neg logit contributions | |
mine | -0.123 |
berry | -0.113 |
're | -0.112 |
Benny | -0.109 |
wherever | -0.108 |
Token | She |
Token ID | 1375 |
Feature activation | -8.24e-01 |
Pos logit contributions | |
mine | +0.005 |
berry | +0.004 |
're | +0.004 |
wherever | +0.004 |
Benny | +0.004 |
Neg logit contributions | |
site | -0.006 |
her | -0.005 |
office | -0.005 |
ia | -0.005 |
men | -0.005 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
site | +0.070 |
her | +0.066 |
office | +0.064 |
ia | +0.063 |
mistake | +0.057 |
Neg logit contributions | |
mine | -0.053 |
berry | -0.048 |
wherever | -0.048 |
're | -0.047 |
Benny | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
site | +0.228 |
her | +0.217 |
office | +0.202 |
ia | +0.196 |
men | +0.183 |
Neg logit contributions | |
mine | -0.251 |
berry | -0.235 |
berries | -0.228 |
ease | -0.222 |
're | -0.219 |
Token | that |
Token ID | 326 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
mine | +0.047 |
berry | +0.044 |
berries | +0.041 |
're | +0.040 |
Grac | +0.039 |
Neg logit contributions | |
site | -0.048 |
office | -0.044 |
her | -0.044 |
ia | -0.041 |
wasteful | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | -9.01e-01 |
Pos logit contributions | |
site | +0.243 |
office | +0.228 |
her | +0.209 |
men | +0.208 |
mistake | +0.204 |
Neg logit contributions | |
mine | -0.266 |
berry | -0.235 |
berries | -0.218 |
oft | -0.212 |
're | -0.211 |
Token | all |
Token ID | 477 |
Feature activation | +6.56e+00 |
Pos logit contributions | |
site | +0.069 |
her | +0.062 |
office | +0.061 |
ia | +0.060 |
men | +0.054 |
Neg logit contributions | |
mine | -0.066 |
berry | -0.061 |
Benny | -0.061 |
wherever | -0.061 |
're | -0.061 |
Token | that |
Token ID | 326 |
Feature activation | -3.28e+00 |
Pos logit contributions | |
mine | +0.553 |
berry | +0.519 |
berries | +0.494 |
're | +0.476 |
Grac | +0.461 |
Neg logit contributions | |
site | -0.446 |
her | -0.426 |
office | -0.426 |
urgent | -0.369 |
ia | -0.369 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
site | +0.242 |
office | +0.217 |
her | +0.217 |
men | +0.204 |
ia | +0.201 |
Neg logit contributions | |
mine | -0.258 |
berry | -0.238 |
berries | -0.220 |
superstar | -0.212 |
oft | -0.209 |
Token | . |
Token ID | 13 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
mine | +0.971 |
berry | +0.897 |
berries | +0.854 |
superstar | +0.845 |
Grac | +0.833 |
Neg logit contributions | |
site | -0.664 |
her | -0.624 |
office | -0.606 |
ia | -0.569 |
yourselves | -0.542 |
Token | From |
Token ID | 3574 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
mine | +0.003 |
berry | +0.003 |
're | +0.003 |
ease | +0.003 |
wherever | +0.003 |
Neg logit contributions | |
site | -0.004 |
her | -0.004 |
office | -0.003 |
ia | -0.003 |
men | -0.003 |
Token | that |
Token ID | 326 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
Benny | +0.087 |
wherever | +0.082 |
mine | +0.080 |
Hop | +0.074 |
Emmy | +0.074 |
Neg logit contributions | |
site | -0.114 |
office | -0.101 |
ia | -0.099 |
her | -0.098 |
urse | -0.097 |
Token | day |
Token ID | 1110 |
Feature activation | +7.37e-01 |
Pos logit contributions | |
site | +0.201 |
office | +0.197 |
mistake | +0.180 |
her | +0.179 |
ia | +0.172 |
Neg logit contributions | |
mine | -0.272 |
berry | -0.256 |
berries | -0.252 |
ease | -0.242 |
Grac | -0.236 |
Token | on |
Token ID | 319 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
mine | +0.044 |
berry | +0.040 |
're | +0.039 |
Benny | +0.039 |
wherever | +0.038 |
Neg logit contributions | |
site | -0.067 |
her | -0.062 |
office | -0.060 |
ia | -0.059 |
wasteful | -0.055 |
Token | her |
Token ID | 607 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
mine | +0.006 |
berry | +0.006 |
're | +0.005 |
Benny | +0.005 |
wherever | +0.005 |
Neg logit contributions | |
site | -0.008 |
her | -0.007 |
office | -0.007 |
ia | -0.007 |
mistake | -0.007 |
Token | or |
Token ID | 393 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
mine | +0.183 |
're | +0.159 |
ease | +0.159 |
berry | +0.158 |
berries | +0.154 |
Neg logit contributions | |
site | -0.151 |
office | -0.149 |
ia | -0.145 |
her | -0.143 |
mistake | -0.136 |
Token | not |
Token ID | 407 |
Feature activation | -4.31e+00 |
Pos logit contributions | |
site | +0.103 |
her | +0.098 |
office | +0.097 |
ia | +0.091 |
wasteful | +0.088 |
Neg logit contributions | |
mine | -0.107 |
berry | -0.100 |
're | -0.092 |
berries | -0.091 |
ease | -0.090 |
Token | , |
Token ID | 11 |
Feature activation | -3.98e-01 |
Pos logit contributions | |
site | +0.315 |
her | +0.304 |
office | +0.297 |
ia | +0.280 |
mistake | +0.274 |
Neg logit contributions | |
mine | -0.320 |
berry | -0.295 |
berries | -0.278 |
Benny | -0.277 |
ease | -0.274 |
Token | what |
Token ID | 644 |
Feature activation | -4.33e+00 |
Pos logit contributions | |
site | +0.030 |
her | +0.029 |
office | +0.029 |
ia | +0.027 |
wasteful | +0.026 |
Neg logit contributions | |
mine | -0.028 |
berry | -0.026 |
ease | -0.025 |
oft | -0.024 |
're | -0.024 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
office | +0.283 |
her | +0.280 |
site | +0.277 |
ia | +0.267 |
wasteful | +0.258 |
Neg logit contributions | |
mine | -0.357 |
berry | -0.343 |
ease | -0.317 |
berries | -0.311 |
oft | -0.301 |
Token | was |
Token ID | 373 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
mine | +0.882 |
berry | +0.819 |
ease | +0.785 |
g | +0.779 |
superstar | +0.763 |
Neg logit contributions | |
her | -0.719 |
site | -0.708 |
office | -0.679 |
ia | -0.646 |
yourselves | -0.619 |
Token | being |
Token ID | 852 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
her | +0.059 |
site | +0.059 |
office | +0.055 |
ia | +0.051 |
wasteful | +0.048 |
Neg logit contributions | |
mine | -0.070 |
berry | -0.066 |
berries | -0.062 |
g | -0.061 |
're | -0.060 |
Token | comfortable |
Token ID | 6792 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
her | +0.075 |
site | +0.070 |
office | +0.066 |
wasteful | +0.063 |
yourselves | +0.060 |
Neg logit contributions | |
mine | -0.097 |
berry | -0.093 |
berries | -0.088 |
ease | -0.086 |
're | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
mine | +0.110 |
berry | +0.101 |
ease | +0.100 |
g | +0.099 |
're | +0.098 |
Neg logit contributions | |
site | -0.094 |
her | -0.089 |
office | -0.088 |
ia | -0.080 |
mistake | -0.075 |
Token | having |
Token ID | 1719 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
mine | +0.057 |
berry | +0.055 |
berries | +0.051 |
're | +0.048 |
bird | +0.048 |
Neg logit contributions | |
her | -0.038 |
site | -0.038 |
office | -0.036 |
wasteful | -0.032 |
mistake | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
mine | +0.413 |
berry | +0.384 |
ease | +0.376 |
berries | +0.366 |
g | +0.363 |
Neg logit contributions | |
site | -0.350 |
office | -0.335 |
her | -0.335 |
ia | -0.310 |
mistake | -0.278 |
Token | that |
Token ID | 326 |
Feature activation | -3.24e+00 |
Pos logit contributions | |
mine | +0.058 |
berry | +0.055 |
berries | +0.050 |
're | +0.049 |
ease | +0.048 |
Neg logit contributions | |
site | -0.055 |
her | -0.052 |
office | -0.051 |
ia | -0.048 |
wasteful | -0.048 |
Token | 's |
Token ID | 338 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
site | +0.216 |
office | +0.215 |
her | +0.204 |
mistake | +0.197 |
wasteful | +0.192 |
Neg logit contributions | |
mine | -0.262 |
berry | -0.245 |
amb | -0.223 |
oft | -0.222 |
berries | -0.220 |
Token | all |
Token ID | 477 |
Feature activation | +6.58e+00 |
Pos logit contributions | |
mine | +0.115 |
berry | +0.108 |
're | +0.106 |
Benny | +0.105 |
wherever | +0.104 |
Neg logit contributions | |
site | -0.086 |
her | -0.078 |
office | -0.075 |
ia | -0.073 |
wasteful | -0.065 |
Token | that |
Token ID | 326 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
mine | +0.587 |
berry | +0.577 |
berries | +0.538 |
're | +0.503 |
amb | +0.490 |
Neg logit contributions | |
her | -0.396 |
site | -0.391 |
office | -0.377 |
wasteful | -0.358 |
old | -0.344 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
site | +0.226 |
her | +0.219 |
office | +0.218 |
mistake | +0.199 |
wasteful | +0.198 |
Neg logit contributions | |
mine | -0.247 |
berry | -0.240 |
berries | -0.214 |
oft | -0.212 |
amb | -0.212 |
Token | . |
Token ID | 13 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
mine | +0.977 |
berry | +0.908 |
berries | +0.850 |
ease | +0.847 |
oft | +0.846 |
Neg logit contributions | |
site | -0.643 |
her | -0.636 |
office | -0.599 |
ia | -0.552 |
yourselves | -0.548 |
Token | From |
Token ID | 3574 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
mine | +0.019 |
ease | +0.017 |
're | +0.017 |
berry | +0.017 |
wherever | +0.016 |
Neg logit contributions | |
site | -0.019 |
her | -0.019 |
office | -0.018 |
ia | -0.018 |
mistake | -0.016 |
Token | then |
Token ID | 788 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
Benny | +0.094 |
mine | +0.092 |
wherever | +0.088 |
Emmy | +0.086 |
berry | +0.082 |
Neg logit contributions | |
site | -0.111 |
office | -0.095 |
ia | -0.094 |
her | -0.092 |
men | -0.087 |
Token | on |
Token ID | 319 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
site | +0.401 |
her | +0.371 |
office | +0.366 |
ia | +0.357 |
men | +0.328 |
Neg logit contributions | |
mine | -0.284 |
berry | -0.259 |
're | -0.252 |
Benny | -0.251 |
ease | -0.247 |
Token | , |
Token ID | 11 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
mine | +0.062 |
berry | +0.057 |
're | +0.057 |
ease | +0.056 |
wherever | +0.055 |
Neg logit contributions | |
site | -0.063 |
her | -0.058 |
office | -0.057 |
ia | -0.057 |
mistake | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
site | +0.320 |
her | +0.297 |
office | +0.290 |
ia | +0.276 |
mistake | +0.254 |
Neg logit contributions | |
mine | -0.380 |
berry | -0.339 |
Benny | -0.331 |
Grac | -0.329 |
superstar | -0.326 |
Token | and |
Token ID | 290 |
Feature activation | +5.43e-01 |
Pos logit contributions | |
site | +0.033 |
office | +0.031 |
her | +0.031 |
men | +0.027 |
ia | +0.027 |
Neg logit contributions | |
mine | -0.038 |
berry | -0.033 |
oft | -0.032 |
berries | -0.032 |
're | -0.031 |
Token | that |
Token ID | 326 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
mine | +0.041 |
berry | +0.038 |
berries | +0.036 |
oft | +0.034 |
're | +0.034 |
Neg logit contributions | |
site | -0.041 |
office | -0.037 |
her | -0.036 |
ia | -0.035 |
men | -0.035 |
Token | 's |
Token ID | 338 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
site | +0.242 |
office | +0.220 |
her | +0.213 |
men | +0.204 |
ia | +0.202 |
Neg logit contributions | |
mine | -0.253 |
berry | -0.226 |
berries | -0.213 |
tie | -0.209 |
oft | -0.205 |
Token | what |
Token ID | 644 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
mine | +0.096 |
wherever | +0.090 |
Benny | +0.090 |
're | +0.089 |
berry | +0.089 |
Neg logit contributions | |
site | -0.095 |
her | -0.084 |
office | -0.083 |
ia | -0.082 |
men | -0.074 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
site | +0.304 |
office | +0.272 |
men | +0.265 |
her | +0.264 |
ia | +0.261 |
Neg logit contributions | |
mine | -0.372 |
berry | -0.336 |
berries | -0.311 |
ease | -0.308 |
oft | -0.306 |
Token | . |
Token ID | 13 |
Feature activation | +4.91e-02 |
Pos logit contributions | |
mine | +0.899 |
berry | +0.812 |
ease | +0.781 |
g | +0.777 |
superstar | +0.773 |
Neg logit contributions | |
site | -0.740 |
her | -0.692 |
office | -0.653 |
ia | -0.629 |
yourselves | -0.581 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
mine | +0.004 |
stuffed | +0.004 |
g | +0.004 |
ease | +0.003 |
mattered | +0.003 |
Neg logit contributions | |
site | -0.003 |
ia | -0.003 |
office | -0.003 |
her | -0.003 |
urse | -0.002 |
Token | One |
Token ID | 3198 |
Feature activation | -4.43e-01 |
Pos logit contributions | |
site | +0.113 |
her | +0.102 |
ia | +0.099 |
office | +0.097 |
men | +0.084 |
Neg logit contributions | |
mine | -0.134 |
're | -0.124 |
g | -0.122 |
berry | -0.120 |
superstar | -0.120 |
Token | day |
Token ID | 1110 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
site | +0.040 |
her | +0.037 |
office | +0.036 |
ia | +0.036 |
men | +0.033 |
Neg logit contributions | |
mine | -0.027 |
berry | -0.024 |
Benny | -0.023 |
're | -0.023 |
wherever | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
mine | +0.052 |
berry | +0.048 |
ease | +0.046 |
're | +0.046 |
wherever | +0.046 |
Neg logit contributions | |
site | -0.053 |
her | -0.048 |
ia | -0.047 |
office | -0.046 |
men | -0.043 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
mine | +0.042 |
berry | +0.039 |
berries | +0.036 |
Benny | +0.036 |
're | +0.035 |
Neg logit contributions | |
site | -0.048 |
her | -0.044 |
ia | -0.043 |
office | -0.042 |
urgent | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
mine | +0.177 |
berry | +0.159 |
're | +0.154 |
ease | +0.154 |
Grac | +0.153 |
Neg logit contributions | |
site | -0.143 |
her | -0.124 |
ia | -0.123 |
office | -0.123 |
men | -0.109 |
Token | like |
Token ID | 588 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
site | +0.031 |
her | +0.028 |
office | +0.027 |
ia | +0.025 |
mistake | +0.025 |
Neg logit contributions | |
mine | -0.034 |
berry | -0.032 |
berries | -0.030 |
're | -0.030 |
Grac | -0.030 |
Token | nothing |
Token ID | 2147 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
mine | +0.278 |
berry | +0.250 |
're | +0.244 |
berries | +0.239 |
oft | +0.235 |
Neg logit contributions | |
site | -0.218 |
office | -0.198 |
her | -0.191 |
ia | -0.185 |
men | -0.174 |
Token | else |
Token ID | 2073 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
site | +0.016 |
her | +0.015 |
ia | +0.015 |
office | +0.014 |
urgent | +0.013 |
Neg logit contributions | |
mine | -0.019 |
berry | -0.017 |
berries | -0.016 |
ease | -0.016 |
're | -0.016 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
site | +0.050 |
her | +0.047 |
ia | +0.046 |
office | +0.045 |
men | +0.041 |
Neg logit contributions | |
mine | -0.062 |
berry | -0.056 |
Grac | -0.053 |
ease | -0.053 |
Emmy | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
mine | +1.003 |
berry | +0.930 |
Grac | +0.919 |
Emmy | +0.911 |
Benny | +0.890 |
Neg logit contributions | |
site | -0.611 |
her | -0.564 |
ia | -0.551 |
office | -0.529 |
yourselves | -0.472 |
Token | |
Token ID | 220 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
site | +0.002 |
her | +0.002 |
office | +0.002 |
ia | +0.002 |
men | +0.002 |
Neg logit contributions | |
mine | -0.002 |
berry | -0.001 |
're | -0.001 |
wherever | -0.001 |
Benny | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
mine | +0.250 |
Benny | +0.225 |
're | +0.224 |
wherever | +0.224 |
berry | +0.222 |
Neg logit contributions | |
site | -0.425 |
office | -0.387 |
her | -0.386 |
ia | -0.374 |
wasteful | -0.359 |
Token |
|
Token ID | 198 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
mine | +0.009 |
Benny | +0.008 |
wherever | +0.008 |
berry | +0.008 |
're | +0.008 |
Neg logit contributions | |
site | -0.015 |
office | -0.014 |
her | -0.014 |
ia | -0.013 |
wasteful | -0.013 |
Token | The |
Token ID | 464 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
mine | +0.011 |
wherever | +0.010 |
Benny | +0.010 |
're | +0.010 |
ease | +0.010 |
Neg logit contributions | |
site | -0.011 |
office | -0.010 |
ia | -0.009 |
her | -0.009 |
men | -0.009 |
Token | their |
Token ID | 511 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
site | +0.244 |
office | +0.221 |
her | +0.215 |
ia | +0.203 |
mistake | +0.202 |
Neg logit contributions | |
mine | -0.256 |
berry | -0.224 |
oft | -0.219 |
're | -0.211 |
tie | -0.210 |
Token | l |
Token ID | 300 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
site | +0.091 |
her | +0.084 |
office | +0.083 |
ia | +0.081 |
men | +0.075 |
Neg logit contributions | |
mine | -0.067 |
berry | -0.060 |
're | -0.059 |
Benny | -0.059 |
wherever | -0.059 |
Token | oll |
Token ID | 692 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
mine | +0.268 |
berry | +0.248 |
Benny | +0.248 |
wherever | +0.247 |
're | +0.246 |
Neg logit contributions | |
site | -0.210 |
her | -0.196 |
ia | -0.191 |
office | -0.188 |
wasteful | -0.162 |
Token | ip |
Token ID | 541 |
Feature activation | -7.79e-01 |
Pos logit contributions | |
site | +0.222 |
ia | +0.218 |
her | +0.211 |
office | +0.210 |
wasteful | +0.188 |
Neg logit contributions | |
mine | -0.162 |
g | -0.146 |
ease | -0.145 |
wherever | -0.144 |
're | -0.143 |
Token | ops |
Token ID | 2840 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
site | +0.061 |
her | +0.060 |
ia | +0.059 |
office | +0.057 |
ups | +0.050 |
Neg logit contributions | |
mine | -0.050 |
wherever | -0.045 |
berry | -0.044 |
ease | -0.044 |
superstar | -0.044 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
mine | +0.116 |
berry | +0.109 |
Grac | +0.106 |
superstar | +0.105 |
Benny | +0.104 |
Neg logit contributions | |
site | -0.144 |
her | -0.142 |
office | -0.134 |
ia | -0.132 |
men | -0.119 |
Token | . |
Token ID | 13 |
Feature activation | +6.61e-03 |
Pos logit contributions | |
mine | +0.938 |
berry | +0.859 |
oft | +0.840 |
superstar | +0.824 |
ease | +0.822 |
Neg logit contributions | |
site | -0.674 |
her | -0.647 |
office | -0.600 |
ia | -0.580 |
yourselves | -0.541 |
Token | By |
Token ID | 2750 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
mine | +0.000 |
berry | +0.000 |
g | +0.000 |
wherever | +0.000 |
're | +0.000 |
Neg logit contributions | |
site | -0.001 |
her | -0.000 |
office | -0.000 |
ia | -0.000 |
men | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
site | +0.278 |
her | +0.253 |
office | +0.247 |
ia | +0.237 |
mistake | +0.222 |
Neg logit contributions | |
mine | -0.261 |
berry | -0.242 |
're | -0.240 |
wherever | -0.232 |
Benny | -0.232 |
Token | end |
Token ID | 886 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
site | +0.014 |
her | +0.013 |
office | +0.013 |
ia | +0.012 |
wasteful | +0.011 |
Neg logit contributions | |
mine | -0.019 |
berry | -0.018 |
're | -0.018 |
Benny | -0.018 |
wherever | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
site | +0.229 |
her | +0.212 |
office | +0.209 |
ia | +0.205 |
men | +0.190 |
Neg logit contributions | |
mine | -0.141 |
berry | -0.126 |
're | -0.123 |
Benny | -0.123 |
wherever | -0.122 |
Token | folder |
Token ID | 9483 |
Feature activation | +4.09e+00 |
Pos logit contributions | |
mine | +0.155 |
berry | +0.139 |
're | +0.131 |
ease | +0.130 |
berries | +0.127 |
Neg logit contributions | |
site | -0.174 |
office | -0.171 |
her | -0.165 |
ia | -0.162 |
mistake | -0.154 |
Token | back |
Token ID | 736 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
mine | +0.317 |
berry | +0.298 |
Will | +0.286 |
berries | +0.284 |
Benny | +0.280 |
Neg logit contributions | |
office | -0.288 |
site | -0.280 |
her | -0.277 |
ia | -0.256 |
mistake | -0.226 |
Token | , |
Token ID | 11 |
Feature activation | -4.75e-01 |
Pos logit contributions | |
mine | +0.101 |
berry | +0.093 |
're | +0.090 |
Benny | +0.090 |
Grac | +0.089 |
Neg logit contributions | |
site | -0.089 |
office | -0.085 |
her | -0.083 |
ia | -0.078 |
mistake | -0.072 |
Token | nothing |
Token ID | 2147 |
Feature activation | -2.31e-01 |
Pos logit contributions | |
office | +0.029 |
site | +0.028 |
her | +0.028 |
mistake | +0.026 |
ia | +0.025 |
Neg logit contributions | |
mine | -0.040 |
berry | -0.037 |
oft | -0.035 |
berries | -0.035 |
're | -0.035 |
Token | else |
Token ID | 2073 |
Feature activation | -7.57e-01 |
Pos logit contributions | |
site | +0.016 |
her | +0.016 |
ia | +0.015 |
office | +0.015 |
wasteful | +0.014 |
Neg logit contributions | |
mine | -0.018 |
berry | -0.017 |
berries | -0.016 |
ease | -0.015 |
're | -0.015 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.09e+01 |
Pos logit contributions | |
site | +0.056 |
her | +0.055 |
office | +0.054 |
ia | +0.053 |
mistake | +0.048 |
Neg logit contributions | |
mine | -0.055 |
berry | -0.051 |
ease | -0.048 |
berries | -0.047 |
Grac | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +4.53e-03 |
Pos logit contributions | |
mine | +0.968 |
berry | +0.907 |
're | +0.877 |
oft | +0.866 |
ease | +0.865 |
Neg logit contributions | |
site | -0.625 |
her | -0.598 |
office | -0.597 |
ia | -0.553 |
yourselves | -0.512 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
mine | +0.000 |
berry | +0.000 |
're | +0.000 |
ease | +0.000 |
g | +0.000 |
Neg logit contributions | |
site | -0.000 |
office | -0.000 |
her | -0.000 |
ia | -0.000 |
urse | -0.000 |
Token | One |
Token ID | 3198 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
site | +0.097 |
her | +0.086 |
ia | +0.084 |
office | +0.083 |
men | +0.069 |
Neg logit contributions | |
mine | -0.148 |
're | -0.139 |
ease | -0.135 |
g | -0.135 |
mattered | -0.134 |
Token | day |
Token ID | 1110 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
site | +0.042 |
her | +0.039 |
office | +0.039 |
ia | +0.038 |
men | +0.035 |
Neg logit contributions | |
mine | -0.028 |
berry | -0.025 |
Benny | -0.024 |
're | -0.024 |
wherever | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
mine | +0.049 |
berry | +0.045 |
ease | +0.044 |
're | +0.043 |
wherever | +0.043 |
Neg logit contributions | |
site | -0.049 |
her | -0.045 |
ia | -0.045 |
office | -0.044 |
men | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
mine | +0.003 |
Benny | +0.002 |
berry | +0.002 |
wherever | +0.002 |
're | +0.002 |
Neg logit contributions | |
site | -0.003 |
her | -0.002 |
ia | -0.002 |
office | -0.002 |
men | -0.002 |
Token | meeting |
Token ID | 3249 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
site | +0.015 |
ia | +0.015 |
her | +0.014 |
office | +0.013 |
ena | +0.013 |
Neg logit contributions | |
mine | -0.013 |
're | -0.013 |
wherever | -0.012 |
g | -0.012 |
berry | -0.012 |
Token | hadn |
Token ID | 8020 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
site | +0.173 |
her | +0.158 |
office | +0.156 |
ia | +0.150 |
men | +0.139 |
Neg logit contributions | |
mine | -0.153 |
berry | -0.142 |
Benny | -0.138 |
're | -0.137 |
wherever | -0.135 |
Token | 't |
Token ID | 470 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
mine | +0.061 |
berry | +0.056 |
Benny | +0.055 |
're | +0.054 |
wherever | +0.054 |
Neg logit contributions | |
site | -0.092 |
her | -0.086 |
office | -0.084 |
ia | -0.082 |
men | -0.075 |
Token | really |
Token ID | 1107 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
mine | +0.174 |
're | +0.160 |
berry | +0.159 |
wherever | +0.158 |
Benny | +0.158 |
Neg logit contributions | |
site | -0.208 |
her | -0.189 |
office | -0.185 |
ia | -0.184 |
wasteful | -0.166 |
Token | mattered |
Token ID | 33343 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
site | +0.044 |
her | +0.040 |
office | +0.039 |
ia | +0.038 |
men | +0.034 |
Neg logit contributions | |
mine | -0.042 |
're | -0.039 |
berry | -0.039 |
wherever | -0.038 |
Benny | -0.038 |
Token | anyway |
Token ID | 6949 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
mine | +0.945 |
berry | +0.886 |
Grac | +0.854 |
're | +0.849 |
oft | +0.845 |
Neg logit contributions | |
site | -0.696 |
her | -0.621 |
office | -0.613 |
ia | -0.565 |
yourselves | -0.534 |
Token | . |
Token ID | 13 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
mine | +0.219 |
berry | +0.201 |
're | +0.198 |
Benny | +0.198 |
wherever | +0.197 |
Neg logit contributions | |
site | -0.263 |
her | -0.241 |
office | -0.235 |
ia | -0.231 |
wasteful | -0.211 |
Token | They |
Token ID | 1119 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
mine | +0.010 |
berry | +0.008 |
're | +0.008 |
wherever | +0.008 |
ease | +0.008 |
Neg logit contributions | |
site | -0.011 |
her | -0.010 |
office | -0.010 |
ia | -0.010 |
men | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | +7.33e-01 |
Pos logit contributions | |
her | +0.154 |
site | +0.151 |
office | +0.141 |
ia | +0.139 |
men | +0.132 |
Neg logit contributions | |
berry | -0.140 |
berries | -0.137 |
Nice | -0.134 |
mine | -0.130 |
wherever | -0.130 |
Token | still |
Token ID | 991 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
mine | +0.054 |
berry | +0.051 |
berries | +0.048 |
Grac | +0.047 |
wherever | +0.047 |
Neg logit contributions | |
site | -0.057 |
her | -0.053 |
office | -0.051 |
ia | -0.048 |
mistake | -0.046 |
Token | cream |
Token ID | 8566 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
mine | +0.071 |
berry | +0.070 |
g | +0.068 |
wherever | +0.067 |
ease | +0.066 |
Neg logit contributions | |
site | -0.054 |
her | -0.046 |
ia | -0.043 |
office | -0.039 |
wasteful | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | -9.00e-01 |
Pos logit contributions | |
site | +0.237 |
her | +0.214 |
office | +0.211 |
ia | +0.204 |
men | +0.184 |
Neg logit contributions | |
mine | -0.289 |
're | -0.270 |
wherever | -0.270 |
Benny | -0.268 |
berry | -0.268 |
Token | available |
Token ID | 1695 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
site | +0.073 |
ia | +0.063 |
office | +0.063 |
her | +0.059 |
men | +0.058 |
Neg logit contributions | |
mine | -0.061 |
Benny | -0.060 |
wherever | -0.059 |
ease | -0.058 |
g | -0.058 |
Token | at |
Token ID | 379 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
site | +0.159 |
her | +0.148 |
office | +0.146 |
ia | +0.137 |
mistake | +0.128 |
Neg logit contributions | |
mine | -0.152 |
berry | -0.140 |
're | -0.135 |
ease | -0.133 |
Benny | -0.131 |
Token | the |
Token ID | 262 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
mine | +0.093 |
Benny | +0.086 |
berry | +0.084 |
wherever | +0.084 |
ease | +0.084 |
Neg logit contributions | |
site | -0.097 |
her | -0.087 |
ia | -0.086 |
office | -0.085 |
men | -0.080 |
Token | pharmacy |
Token ID | 32763 |
Feature activation | -1.34e+01 |
Pos logit contributions | |
site | +0.012 |
ia | +0.011 |
her | +0.011 |
office | +0.009 |
urgent | +0.009 |
Neg logit contributions | |
mine | -0.015 |
berry | -0.015 |
wherever | -0.014 |
g | -0.014 |
're | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -4.39e-01 |
Pos logit contributions | |
site | +1.138 |
her | +1.045 |
office | +1.018 |
ia | +1.012 |
wasteful | +0.928 |
Neg logit contributions | |
mine | -0.869 |
berry | -0.790 |
're | -0.780 |
Benny | -0.778 |
wherever | -0.777 |
Token | so |
Token ID | 523 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
site | +0.035 |
her | +0.033 |
office | +0.033 |
ia | +0.031 |
mistake | +0.029 |
Neg logit contributions | |
mine | -0.030 |
berry | -0.027 |
're | -0.027 |
ease | -0.026 |
mattered | -0.026 |
Token | they |
Token ID | 484 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
site | +0.103 |
her | +0.095 |
office | +0.093 |
ia | +0.092 |
wasteful | +0.085 |
Neg logit contributions | |
mine | -0.077 |
berry | -0.069 |
're | -0.068 |
ease | -0.066 |
wherever | -0.066 |
Token | went |
Token ID | 1816 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
her | +0.005 |
site | +0.005 |
office | +0.005 |
ep | +0.004 |
ia | +0.004 |
Neg logit contributions | |
Nice | -0.005 |
berry | -0.005 |
berries | -0.005 |
mine | -0.005 |
wherever | -0.005 |
Token | there |
Token ID | 612 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
site | +0.248 |
her | +0.229 |
office | +0.226 |
ia | +0.219 |
men | +0.203 |
Neg logit contributions | |
mine | -0.178 |
berry | -0.162 |
're | -0.160 |
Benny | -0.158 |
wherever | -0.156 |
Token | introduce |
Token ID | 10400 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
mine | +0.144 |
berry | +0.138 |
berries | +0.132 |
Benny | +0.132 |
wherever | +0.131 |
Neg logit contributions | |
site | -0.141 |
her | -0.130 |
office | -0.127 |
ia | -0.122 |
mistake | -0.111 |
Token | him |
Token ID | 683 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
mine | +0.166 |
berry | +0.158 |
're | +0.152 |
superstar | +0.148 |
berries | +0.146 |
Neg logit contributions | |
site | -0.163 |
her | -0.150 |
ia | -0.145 |
office | -0.144 |
men | -0.133 |
Token | to |
Token ID | 284 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
mine | +0.017 |
berry | +0.016 |
're | +0.015 |
superstar | +0.015 |
Grac | +0.015 |
Neg logit contributions | |
site | -0.015 |
her | -0.014 |
office | -0.014 |
ia | -0.014 |
mistake | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
site | +0.009 |
office | +0.008 |
her | +0.008 |
ia | +0.008 |
men | +0.007 |
Neg logit contributions | |
mine | -0.009 |
berry | -0.008 |
're | -0.008 |
berries | -0.008 |
Benny | -0.008 |
Token | passport |
Token ID | 17981 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
site | +0.024 |
her | +0.022 |
ia | +0.021 |
office | +0.021 |
wasteful | +0.019 |
Neg logit contributions | |
mine | -0.025 |
berry | -0.023 |
Benny | -0.023 |
wherever | -0.023 |
're | -0.023 |
Token | office |
Token ID | 2607 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
site | +0.152 |
her | +0.138 |
office | +0.136 |
ia | +0.132 |
men | +0.120 |
Neg logit contributions | |
mine | -0.135 |
berry | -0.124 |
Benny | -0.122 |
're | -0.121 |
ease | -0.119 |
Token | workers |
Token ID | 3259 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
site | +1.043 |
her | +0.954 |
office | +0.934 |
ia | +0.915 |
men | +0.834 |
Neg logit contributions | |
mine | -0.888 |
berry | -0.815 |
're | -0.800 |
Benny | -0.799 |
wherever | -0.792 |
Token | and |
Token ID | 290 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
site | +0.503 |
her | +0.465 |
office | +0.464 |
ia | +0.419 |
men | +0.397 |
Neg logit contributions | |
mine | -0.453 |
Benny | -0.429 |
berry | -0.428 |
ease | -0.417 |
Grac | -0.414 |
Token | get |
Token ID | 651 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
mine | +0.028 |
berry | +0.026 |
Benny | +0.024 |
're | +0.024 |
wherever | +0.024 |
Neg logit contributions | |
site | -0.032 |
office | -0.029 |
her | -0.029 |
ia | -0.027 |
men | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
site | +0.230 |
her | +0.213 |
office | +0.209 |
ia | +0.198 |
men | +0.189 |
Neg logit contributions | |
mine | -0.191 |
berry | -0.176 |
're | -0.169 |
Benny | -0.168 |
wherever | -0.165 |
Token | passport |
Token ID | 17981 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
site | +0.088 |
her | +0.080 |
office | +0.078 |
ia | +0.076 |
men | +0.069 |
Neg logit contributions | |
mine | -0.095 |
berry | -0.088 |
Benny | -0.086 |
're | -0.086 |
wherever | -0.085 |
Token | going |
Token ID | 1016 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
mine | +0.340 |
berry | +0.312 |
're | +0.305 |
Benny | +0.304 |
wherever | +0.302 |
Neg logit contributions | |
site | -0.375 |
her | -0.343 |
office | -0.337 |
ia | -0.327 |
men | -0.300 |
Token | to |
Token ID | 284 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
site | +0.247 |
her | +0.228 |
office | +0.227 |
ia | +0.213 |
men | +0.200 |
Neg logit contributions | |
mine | -0.191 |
berry | -0.173 |
Benny | -0.168 |
're | -0.166 |
ease | -0.164 |
Token | the |
Token ID | 262 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
site | +0.002 |
office | +0.002 |
her | +0.002 |
ia | +0.002 |
mistake | +0.002 |
Neg logit contributions | |
mine | -0.002 |
berry | -0.001 |
berries | -0.001 |
Benny | -0.001 |
wherever | -0.001 |
Token | doctor |
Token ID | 6253 |
Feature activation | -4.77e+00 |
Pos logit contributions | |
site | +0.021 |
ia | +0.020 |
her | +0.019 |
office | +0.018 |
wasteful | +0.018 |
Neg logit contributions | |
mine | -0.021 |
berry | -0.020 |
wherever | -0.020 |
Benny | -0.019 |
're | -0.019 |
Token | 's |
Token ID | 338 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
site | +0.399 |
her | +0.369 |
office | +0.366 |
ia | +0.349 |
men | +0.321 |
Neg logit contributions | |
mine | -0.313 |
berry | -0.286 |
Benny | -0.280 |
're | -0.278 |
ease | -0.276 |
Token | office |
Token ID | 2607 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
berry | +0.102 |
g | +0.101 |
wherever | +0.100 |
mine | +0.099 |
Will | +0.098 |
Neg logit contributions | |
site | -0.055 |
ia | -0.051 |
her | -0.049 |
wasteful | -0.047 |
cks | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
site | +1.066 |
her | +0.973 |
ia | +0.953 |
office | +0.945 |
wasteful | +0.877 |
Neg logit contributions | |
mine | -0.856 |
berry | -0.775 |
wherever | -0.769 |
're | -0.769 |
Benny | -0.759 |
Token | Molly |
Token ID | 30236 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
site | +0.010 |
her | +0.009 |
office | +0.009 |
ia | +0.008 |
men | +0.008 |
Neg logit contributions | |
mine | -0.007 |
berry | -0.006 |
wherever | -0.006 |
're | -0.006 |
ease | -0.006 |
Token | said |
Token ID | 531 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
mine | +0.235 |
berry | +0.214 |
're | +0.210 |
wherever | +0.210 |
Benny | +0.209 |
Neg logit contributions | |
site | -0.301 |
her | -0.278 |
office | -0.272 |
ia | -0.266 |
men | -0.243 |
Token | to |
Token ID | 284 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
site | +0.184 |
office | +0.166 |
her | +0.165 |
ia | +0.164 |
men | +0.155 |
Neg logit contributions | |
mine | -0.165 |
berry | -0.151 |
ease | -0.149 |
're | -0.146 |
wherever | -0.144 |
Token | Joe |
Token ID | 5689 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
mine | +0.002 |
berry | +0.002 |
Benny | +0.002 |
're | +0.002 |
wherever | +0.002 |
Neg logit contributions | |
site | -0.003 |
her | -0.002 |
office | -0.002 |
ia | -0.002 |
wasteful | -0.002 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
site | +0.368 |
her | +0.343 |
office | +0.336 |
ia | +0.330 |
men | +0.303 |
Neg logit contributions | |
mine | -0.303 |
berry | -0.278 |
're | -0.272 |
wherever | -0.268 |
Benny | -0.267 |
Token | back |
Token ID | 736 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
site | +0.168 |
her | +0.157 |
office | +0.152 |
ia | +0.148 |
men | +0.137 |
Neg logit contributions | |
mine | -0.133 |
berry | -0.123 |
're | -0.122 |
Benny | -0.117 |
ease | -0.116 |
Token | out |
Token ID | 503 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
mine | +0.085 |
berry | +0.077 |
're | +0.077 |
Benny | +0.074 |
ease | +0.073 |
Neg logit contributions | |
site | -0.106 |
her | -0.098 |
office | -0.097 |
ia | -0.093 |
men | -0.085 |
Token | of |
Token ID | 286 |
Feature activation | +8.93e-01 |
Pos logit contributions | |
site | +0.088 |
her | +0.081 |
office | +0.080 |
ia | +0.079 |
mistake | +0.071 |
Neg logit contributions | |
mine | -0.081 |
're | -0.075 |
berry | -0.073 |
Benny | -0.071 |
mattered | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
mine | +0.065 |
're | +0.063 |
berry | +0.059 |
mattered | +0.058 |
superstar | +0.057 |
Neg logit contributions | |
site | -0.069 |
office | -0.062 |
her | -0.062 |
ia | -0.059 |
mistake | -0.056 |
Token | office |
Token ID | 2607 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
site | +0.021 |
her | +0.020 |
ia | +0.020 |
office | +0.018 |
wasteful | +0.017 |
Neg logit contributions | |
mine | -0.027 |
berry | -0.026 |
Benny | -0.026 |
wherever | -0.025 |
g | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +1.20e-03 |
Pos logit contributions | |
site | +1.046 |
her | +0.959 |
office | +0.932 |
ia | +0.927 |
wasteful | +0.854 |
Neg logit contributions | |
mine | -0.879 |
berry | -0.800 |
're | -0.784 |
wherever | -0.784 |
Benny | -0.782 |
Token | She |
Token ID | 1375 |
Feature activation | -9.64e-01 |
Pos logit contributions | |
mine | +0.000 |
're | +0.000 |
berry | +0.000 |
mattered | +0.000 |
ease | +0.000 |
Neg logit contributions | |
site | -0.000 |
her | -0.000 |
office | -0.000 |
ia | -0.000 |
urse | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | -8.61e-01 |
Pos logit contributions | |
site | +0.082 |
her | +0.076 |
office | +0.074 |
ia | +0.073 |
men | +0.066 |
Neg logit contributions | |
mine | -0.063 |
berry | -0.057 |
wherever | -0.056 |
're | -0.056 |
Benny | -0.056 |
Token | so |
Token ID | 523 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
site | +0.074 |
her | +0.067 |
office | +0.066 |
ia | +0.066 |
wasteful | +0.060 |
Neg logit contributions | |
mine | -0.055 |
berry | -0.050 |
're | -0.049 |
Benny | -0.049 |
wherever | -0.049 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
site | +0.105 |
her | +0.097 |
office | +0.095 |
ia | +0.093 |
wasteful | +0.085 |
Neg logit contributions | |
mine | -0.084 |
berry | -0.077 |
Benny | -0.075 |
're | -0.075 |
wherever | -0.074 |
Token | want |
Token ID | 765 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
berry | +0.160 |
mine | +0.159 |
berries | +0.154 |
Benny | +0.147 |
wherever | +0.146 |
Neg logit contributions | |
site | -0.202 |
her | -0.193 |
office | -0.189 |
ia | -0.184 |
wasteful | -0.173 |
Token | to |
Token ID | 284 |
Feature activation | +6.95e-02 |
Pos logit contributions | |
mine | +0.149 |
berry | +0.138 |
're | +0.131 |
berries | +0.126 |
oft | +0.126 |
Neg logit contributions | |
site | -0.164 |
her | -0.154 |
office | -0.152 |
ia | -0.145 |
x | -0.136 |
Token | leave |
Token ID | 2666 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
mine | +0.005 |
berry | +0.004 |
Benny | +0.004 |
're | +0.004 |
wherever | +0.004 |
Neg logit contributions | |
site | -0.006 |
her | -0.005 |
office | -0.005 |
ia | -0.005 |
men | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
site | +0.229 |
her | +0.211 |
office | +0.209 |
ia | +0.197 |
men | +0.185 |
Neg logit contributions | |
mine | -0.190 |
berry | -0.178 |
're | -0.174 |
Benny | -0.165 |
oft | -0.164 |
Token | construction |
Token ID | 5103 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
site | +0.015 |
ia | +0.014 |
her | +0.013 |
wasteful | +0.012 |
yourselves | +0.012 |
Neg logit contributions | |
mine | -0.016 |
wherever | -0.016 |
berry | -0.015 |
Benny | -0.015 |
're | -0.015 |
Token | site |
Token ID | 2524 |
Feature activation | -1.29e+01 |
Pos logit contributions | |
ia | +0.234 |
her | +0.195 |
site | +0.184 |
wasteful | +0.181 |
urse | +0.169 |
Neg logit contributions | |
mine | -0.512 |
wherever | -0.477 |
whenever | -0.458 |
berry | -0.456 |
g | -0.456 |
Token | . |
Token ID | 13 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
site | +1.031 |
her | +0.946 |
office | +0.929 |
ia | +0.891 |
men | +0.817 |
Neg logit contributions | |
mine | -0.894 |
berry | -0.829 |
're | -0.810 |
Benny | -0.808 |
ease | -0.786 |
Token | Daddy |
Token ID | 26926 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
mine | +0.004 |
berry | +0.004 |
're | +0.004 |
wherever | +0.004 |
Benny | +0.004 |
Neg logit contributions | |
site | -0.006 |
her | -0.005 |
office | -0.005 |
ia | -0.005 |
wasteful | -0.005 |
Token | noticed |
Token ID | 6810 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
mine | +0.144 |
berry | +0.131 |
Benny | +0.130 |
're | +0.128 |
wherever | +0.125 |
Neg logit contributions | |
site | -0.183 |
her | -0.166 |
office | -0.163 |
ia | -0.162 |
wasteful | -0.150 |
Token | and |
Token ID | 290 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
site | +0.024 |
her | +0.022 |
office | +0.021 |
ia | +0.021 |
wasteful | +0.018 |
Neg logit contributions | |
mine | -0.031 |
berry | -0.029 |
're | -0.029 |
wherever | -0.028 |
ease | -0.028 |
Token | asked |
Token ID | 1965 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
mine | +0.040 |
Benny | +0.039 |
Grac | +0.036 |
Will | +0.036 |
berry | +0.036 |
Neg logit contributions | |
site | -0.061 |
her | -0.056 |
office | -0.056 |
ia | -0.055 |
yourselves | -0.051 |
Token | place |
Token ID | 1295 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
mine | +0.093 |
berry | +0.088 |
Benny | +0.086 |
wherever | +0.085 |
're | +0.084 |
Neg logit contributions | |
site | -0.094 |
her | -0.086 |
ia | -0.083 |
office | -0.082 |
wasteful | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +8.03e-02 |
Pos logit contributions | |
site | +0.127 |
her | +0.116 |
office | +0.113 |
ia | +0.111 |
wasteful | +0.101 |
Neg logit contributions | |
mine | -0.111 |
berry | -0.102 |
're | -0.100 |
Benny | -0.100 |
wherever | -0.099 |
Token | It |
Token ID | 632 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
mine | +0.005 |
berry | +0.004 |
're | +0.004 |
wherever | +0.004 |
Benny | +0.004 |
Neg logit contributions | |
site | -0.007 |
her | -0.007 |
office | -0.007 |
ia | -0.006 |
men | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
mine | +0.048 |
berry | +0.044 |
wherever | +0.042 |
ease | +0.042 |
're | +0.042 |
Neg logit contributions | |
site | -0.052 |
her | -0.048 |
office | -0.046 |
ia | -0.045 |
men | -0.041 |
Token | an |
Token ID | 281 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
site | +0.070 |
ia | +0.062 |
office | +0.061 |
her | +0.061 |
wasteful | +0.056 |
Neg logit contributions | |
mine | -0.054 |
Benny | -0.054 |
wherever | -0.053 |
ease | -0.051 |
're | -0.051 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
mine | +0.082 |
Benny | +0.077 |
berry | +0.077 |
g | +0.076 |
're | +0.076 |
Neg logit contributions | |
site | -0.061 |
ia | -0.054 |
her | -0.053 |
office | -0.050 |
wasteful | -0.048 |
Token | ! |
Token ID | 0 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
site | +1.120 |
her | +1.031 |
ia | +1.008 |
office | +1.004 |
wasteful | +0.933 |
Neg logit contributions | |
mine | -0.794 |
berry | -0.714 |
wherever | -0.706 |
're | -0.703 |
Benny | -0.700 |
Token | |
Token ID | 220 |
Feature activation | +4.46e+00 |
Pos logit contributions | |
mine | +0.068 |
berry | +0.061 |
're | +0.059 |
Benny | +0.059 |
wherever | +0.058 |
Neg logit contributions | |
site | -0.127 |
her | -0.118 |
office | -0.116 |
ia | -0.114 |
men | -0.106 |
Token |
|
Token ID | 198 |
Feature activation | +1.84e-01 |
Pos logit contributions | |
mine | +0.234 |
Benny | +0.214 |
're | +0.213 |
wherever | +0.212 |
ease | +0.206 |
Neg logit contributions | |
site | -0.432 |
her | -0.386 |
office | -0.386 |
ia | -0.378 |
wasteful | -0.367 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
mine | +0.011 |
berry | +0.010 |
Benny | +0.010 |
're | +0.010 |
wherever | +0.010 |
Neg logit contributions | |
site | -0.017 |
her | -0.015 |
office | -0.015 |
ia | -0.015 |
wasteful | -0.014 |
Token | Sarah |
Token ID | 29284 |
Feature activation | +9.43e-02 |
Pos logit contributions | |
mine | +0.011 |
wherever | +0.010 |
're | +0.010 |
Benny | +0.010 |
berry | +0.010 |
Neg logit contributions | |
site | -0.015 |
ia | -0.013 |
her | -0.013 |
office | -0.013 |
men | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -1.73e-01 |
Pos logit contributions | |
site | +0.637 |
office | +0.586 |
her | +0.585 |
ia | +0.559 |
men | +0.527 |
Neg logit contributions | |
mine | -0.527 |
berry | -0.479 |
're | -0.466 |
Benny | -0.464 |
Grac | -0.451 |
Token | stay |
Token ID | 2652 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
site | +0.015 |
her | +0.014 |
office | +0.014 |
ia | +0.013 |
mistake | +0.012 |
Neg logit contributions | |
mine | -0.011 |
berry | -0.010 |
Benny | -0.010 |
're | -0.010 |
wherever | -0.009 |
Token | at |
Token ID | 379 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
site | +0.196 |
her | +0.179 |
office | +0.175 |
ia | +0.174 |
wasteful | +0.159 |
Neg logit contributions | |
mine | -0.150 |
berry | -0.136 |
Benny | -0.134 |
wherever | -0.134 |
're | -0.134 |
Token | the |
Token ID | 262 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
mine | +0.067 |
berry | +0.062 |
're | +0.060 |
Benny | +0.058 |
wherever | +0.058 |
Neg logit contributions | |
site | -0.085 |
her | -0.078 |
office | -0.078 |
ia | -0.074 |
mistake | -0.069 |
Token | post |
Token ID | 1281 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
site | +0.036 |
her | +0.032 |
office | +0.032 |
ia | +0.031 |
wasteful | +0.028 |
Neg logit contributions | |
mine | -0.039 |
berry | -0.036 |
're | -0.035 |
Benny | -0.035 |
wherever | -0.035 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.368 |
her | +0.326 |
office | +0.319 |
ia | +0.308 |
men | +0.270 |
Neg logit contributions | |
mine | -0.534 |
berry | -0.500 |
're | -0.494 |
Benny | -0.493 |
wherever | -0.489 |
Token | and |
Token ID | 290 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
site | +1.061 |
her | +0.973 |
office | +0.954 |
ia | +0.933 |
men | +0.853 |
Neg logit contributions | |
mine | -0.858 |
berry | -0.787 |
're | -0.772 |
Benny | -0.771 |
wherever | -0.762 |
Token | pick |
Token ID | 2298 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
mine | +0.023 |
berry | +0.022 |
Benny | +0.020 |
berries | +0.020 |
're | +0.020 |
Neg logit contributions | |
site | -0.027 |
office | -0.025 |
her | -0.024 |
ia | -0.023 |
mistake | -0.022 |
Token | up |
Token ID | 510 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
mine | +0.087 |
berry | +0.080 |
're | +0.079 |
Benny | +0.077 |
wherever | +0.076 |
Neg logit contributions | |
site | -0.087 |
her | -0.082 |
office | -0.080 |
ia | -0.076 |
men | -0.071 |
Token | her |
Token ID | 607 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
site | +0.142 |
office | +0.133 |
her | +0.132 |
ia | +0.128 |
men | +0.122 |
Neg logit contributions | |
mine | -0.126 |
berry | -0.115 |
're | -0.113 |
oft | -0.110 |
superstar | -0.109 |
Token | paper |
Token ID | 3348 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
mine | +0.148 |
berry | +0.132 |
Benny | +0.131 |
're | +0.131 |
ease | +0.127 |
Neg logit contributions | |
site | -0.145 |
office | -0.137 |
her | -0.133 |
ia | -0.126 |
mistake | -0.119 |
Token | to |
Token ID | 284 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
site | +0.182 |
office | +0.170 |
her | +0.169 |
ia | +0.162 |
men | +0.155 |
Neg logit contributions | |
mine | -0.228 |
berry | -0.204 |
superstar | -0.193 |
Grac | -0.189 |
berries | -0.189 |
Token | stir |
Token ID | 11240 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
site | +0.001 |
office | +0.001 |
her | +0.001 |
ia | +0.001 |
mistake | +0.001 |
Neg logit contributions | |
mine | -0.001 |
berry | -0.001 |
berries | -0.001 |
oft | -0.001 |
Grac | -0.001 |
Token | his |
Token ID | 465 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
site | +0.147 |
her | +0.136 |
ia | +0.130 |
office | +0.128 |
yourselves | +0.118 |
Neg logit contributions | |
mine | -0.139 |
're | -0.129 |
berry | -0.127 |
superstar | -0.124 |
Benny | -0.124 |
Token | dad |
Token ID | 9955 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
mine | +0.149 |
Benny | +0.132 |
're | +0.132 |
berry | +0.130 |
wherever | +0.129 |
Neg logit contributions | |
site | -0.149 |
office | -0.137 |
her | -0.135 |
ia | -0.131 |
mistake | -0.119 |
Token | 's |
Token ID | 338 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
site | +0.244 |
her | +0.240 |
office | +0.238 |
ia | +0.210 |
x | +0.198 |
Neg logit contributions | |
mine | -0.236 |
ease | -0.221 |
Benny | -0.221 |
Grac | -0.219 |
superstar | -0.219 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
mine | +0.098 |
're | +0.089 |
berry | +0.089 |
Benny | +0.088 |
wherever | +0.087 |
Neg logit contributions | |
site | -0.096 |
office | -0.087 |
her | -0.087 |
ia | -0.082 |
mistake | -0.076 |
Token | ! |
Token ID | 0 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
site | +1.005 |
her | +0.924 |
office | +0.916 |
ia | +0.861 |
men | +0.791 |
Neg logit contributions | |
mine | -0.912 |
berry | -0.847 |
Benny | -0.833 |
're | -0.829 |
ease | -0.809 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
mine | +0.093 |
mattered | +0.082 |
're | +0.082 |
ease | +0.081 |
g | +0.081 |
Neg logit contributions | |
site | -0.095 |
ia | -0.088 |
her | -0.088 |
office | -0.088 |
men | -0.075 |
Token | Once |
Token ID | 7454 |
Feature activation | +5.03e+00 |
Pos logit contributions | |
site | +0.113 |
her | +0.102 |
ia | +0.100 |
office | +0.099 |
men | +0.085 |
Neg logit contributions | |
mine | -0.130 |
're | -0.118 |
berry | -0.116 |
g | -0.116 |
ease | -0.115 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
mine | +0.298 |
berry | +0.270 |
're | +0.264 |
Benny | +0.263 |
wherever | +0.260 |
Neg logit contributions | |
site | -0.454 |
her | -0.420 |
office | -0.412 |
ia | -0.407 |
wasteful | -0.373 |
Token | a |
Token ID | 257 |
Feature activation | -9.82e-01 |
Pos logit contributions | |
mine | +0.103 |
berry | +0.096 |
're | +0.093 |
Grac | +0.092 |
Benny | +0.092 |
Neg logit contributions | |
site | -0.074 |
her | -0.070 |
office | -0.066 |
ia | -0.066 |
x | -0.057 |
Token | coming |
Token ID | 2406 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
site | +0.168 |
her | +0.154 |
office | +0.150 |
ia | +0.149 |
wasteful | +0.136 |
Neg logit contributions | |
mine | -0.154 |
berry | -0.142 |
're | -0.140 |
wherever | -0.139 |
Benny | -0.139 |
Token | from |
Token ID | 422 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
site | +0.540 |
office | +0.485 |
her | +0.483 |
ia | +0.477 |
wasteful | +0.438 |
Neg logit contributions | |
mine | -0.435 |
wherever | -0.402 |
Benny | -0.401 |
're | -0.394 |
berry | -0.391 |
Token | a |
Token ID | 257 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
site | +0.120 |
ia | +0.116 |
office | +0.109 |
her | +0.109 |
ups | +0.107 |
Neg logit contributions | |
wherever | -0.073 |
mine | -0.073 |
Benny | -0.072 |
ease | -0.070 |
Emmy | -0.063 |
Token | nearby |
Token ID | 6716 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
site | +0.072 |
her | +0.066 |
ia | +0.064 |
office | +0.063 |
wasteful | +0.057 |
Neg logit contributions | |
mine | -0.072 |
berry | -0.068 |
Benny | -0.067 |
wherever | -0.067 |
're | -0.066 |
Token | construction |
Token ID | 5103 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
site | +0.095 |
her | +0.092 |
ia | +0.089 |
wasteful | +0.082 |
office | +0.080 |
Neg logit contributions | |
mine | -0.089 |
berry | -0.089 |
Benny | -0.087 |
g | -0.084 |
wherever | -0.083 |
Token | site |
Token ID | 2524 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
ia | +0.172 |
wasteful | +0.128 |
inger | +0.118 |
urse | +0.118 |
her | +0.115 |
Neg logit contributions | |
mine | -0.521 |
wherever | -0.492 |
g | -0.492 |
berry | -0.484 |
were | -0.474 |
Token | . |
Token ID | 13 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
site | +1.069 |
ia | +0.995 |
her | +0.990 |
office | +0.966 |
wasteful | +0.908 |
Neg logit contributions | |
mine | -0.808 |
wherever | -0.748 |
're | -0.725 |
berry | -0.725 |
Benny | -0.711 |
Token | The |
Token ID | 383 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
mine | +0.001 |
ease | +0.001 |
're | +0.001 |
berry | +0.001 |
wherever | +0.001 |
Neg logit contributions | |
site | -0.003 |
her | -0.003 |
ia | -0.003 |
office | -0.003 |
men | -0.002 |
Token | workers |
Token ID | 3259 |
Feature activation | -6.24e+00 |
Pos logit contributions | |
g | +0.070 |
berry | +0.066 |
mine | +0.065 |
're | +0.065 |
stuffed | +0.063 |
Neg logit contributions | |
site | -0.047 |
her | -0.046 |
ia | -0.046 |
yourselves | -0.041 |
inger | -0.040 |
Token | had |
Token ID | 550 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
site | +0.640 |
her | +0.595 |
office | +0.589 |
ia | +0.562 |
men | +0.552 |
Neg logit contributions | |
mine | -0.275 |
berry | -0.261 |
Benny | -0.260 |
wherever | -0.254 |
Grac | -0.244 |
Token | accidentally |
Token ID | 14716 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
mine | +0.046 |
're | +0.042 |
Benny | +0.041 |
berry | +0.041 |
wherever | +0.041 |
Neg logit contributions | |
site | -0.061 |
her | -0.056 |
office | -0.055 |
ia | -0.054 |
wasteful | -0.050 |
Token | going |
Token ID | 1016 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
mine | +0.087 |
berry | +0.081 |
're | +0.076 |
berries | +0.075 |
ease | +0.075 |
Neg logit contributions | |
site | -0.098 |
her | -0.092 |
office | -0.087 |
ia | -0.087 |
wasteful | -0.082 |
Token | to |
Token ID | 284 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
site | +0.192 |
her | +0.177 |
office | +0.175 |
ia | +0.165 |
men | +0.153 |
Neg logit contributions | |
mine | -0.158 |
berry | -0.146 |
're | -0.140 |
Benny | -0.140 |
g | -0.138 |
Token | be |
Token ID | 307 |
Feature activation | -5.55e+00 |
Pos logit contributions | |
mine | +0.013 |
berry | +0.012 |
're | +0.012 |
Benny | +0.012 |
wherever | +0.012 |
Neg logit contributions | |
site | -0.019 |
her | -0.017 |
office | -0.017 |
ia | -0.016 |
mistake | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
site | +0.417 |
her | +0.376 |
office | +0.367 |
ia | +0.363 |
wasteful | +0.327 |
Neg logit contributions | |
mine | -0.414 |
berry | -0.382 |
Benny | -0.379 |
're | -0.377 |
wherever | -0.376 |
Token | new |
Token ID | 649 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
site | +0.083 |
office | +0.075 |
her | +0.075 |
ia | +0.070 |
mistake | +0.066 |
Neg logit contributions | |
mine | -0.091 |
berry | -0.081 |
're | -0.081 |
ease | -0.079 |
berries | -0.079 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.057 |
office | +0.054 |
her | +0.050 |
ia | +0.050 |
mistake | +0.047 |
Neg logit contributions | |
mine | -0.054 |
're | -0.048 |
ease | -0.048 |
berry | -0.047 |
Benny | -0.046 |
Token | for |
Token ID | 329 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
site | +1.012 |
her | +0.926 |
office | +0.914 |
ia | +0.874 |
men | +0.803 |
Neg logit contributions | |
mine | -0.905 |
berry | -0.840 |
Benny | -0.824 |
're | -0.822 |
ease | -0.811 |
Token | people |
Token ID | 661 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
site | +0.343 |
her | +0.316 |
ia | +0.308 |
office | +0.306 |
wasteful | +0.281 |
Neg logit contributions | |
mine | -0.278 |
Benny | -0.256 |
wherever | -0.253 |
berry | -0.252 |
're | -0.248 |
Token | to |
Token ID | 284 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
site | +0.196 |
office | +0.190 |
her | +0.185 |
ia | +0.165 |
men | +0.157 |
Neg logit contributions | |
mine | -0.215 |
berry | -0.204 |
Benny | -0.201 |
ease | -0.201 |
Grac | -0.195 |
Token | work |
Token ID | 670 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
site | +0.010 |
her | +0.009 |
office | +0.009 |
ia | +0.009 |
men | +0.008 |
Neg logit contributions | |
mine | -0.007 |
berry | -0.006 |
Benny | -0.006 |
're | -0.006 |
wherever | -0.006 |
Token | in |
Token ID | 287 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
site | +0.360 |
her | +0.333 |
office | +0.322 |
ia | +0.310 |
men | +0.285 |
Neg logit contributions | |
mine | -0.306 |
berry | -0.287 |
're | -0.278 |
Benny | -0.275 |
wherever | -0.268 |
Token |
|
Token ID | 198 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
mine | +0.001 |
Benny | +0.001 |
berry | +0.001 |
wherever | +0.001 |
ease | +0.001 |
Neg logit contributions | |
site | -0.002 |
office | -0.002 |
her | -0.002 |
ia | -0.002 |
wasteful | -0.002 |
Token | At |
Token ID | 2953 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
mine | +0.002 |
berry | +0.002 |
Benny | +0.002 |
wherever | +0.002 |
're | +0.002 |
Neg logit contributions | |
site | -0.002 |
office | -0.001 |
her | -0.001 |
ia | -0.001 |
men | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
site | +0.229 |
her | +0.215 |
ia | +0.204 |
office | +0.202 |
wasteful | +0.195 |
Neg logit contributions | |
mine | -0.169 |
wherever | -0.160 |
Benny | -0.156 |
ease | -0.154 |
berry | -0.148 |
Token | doctor |
Token ID | 6253 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
site | +0.031 |
her | +0.027 |
ia | +0.027 |
office | +0.026 |
wasteful | +0.025 |
Neg logit contributions | |
mine | -0.031 |
berry | -0.030 |
wherever | -0.029 |
Benny | -0.029 |
g | -0.029 |
Token | 's |
Token ID | 338 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
site | +0.392 |
her | +0.360 |
office | +0.351 |
ia | +0.347 |
wasteful | +0.319 |
Neg logit contributions | |
mine | -0.302 |
berry | -0.276 |
're | -0.271 |
Benny | -0.271 |
wherever | -0.268 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
g | +0.085 |
berry | +0.084 |
wherever | +0.083 |
mine | +0.081 |
ease | +0.080 |
Neg logit contributions | |
site | -0.065 |
her | -0.060 |
ia | -0.058 |
wasteful | -0.053 |
yourselves | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | -5.75e-01 |
Pos logit contributions | |
site | +1.136 |
her | +1.053 |
ia | +1.025 |
office | +1.003 |
wasteful | +0.971 |
Neg logit contributions | |
mine | -0.758 |
wherever | -0.668 |
're | -0.668 |
berry | -0.664 |
Benny | -0.658 |
Token | the |
Token ID | 262 |
Feature activation | -2.47e-01 |
Pos logit contributions | |
site | +0.049 |
her | +0.045 |
office | +0.045 |
ia | +0.044 |
men | +0.041 |
Neg logit contributions | |
mine | -0.037 |
berry | -0.033 |
're | -0.033 |
wherever | -0.032 |
ease | -0.032 |
Token | doctor |
Token ID | 6253 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
site | +0.017 |
her | +0.015 |
ia | +0.015 |
cks | +0.014 |
yourselves | +0.013 |
Neg logit contributions | |
mine | -0.018 |
berry | -0.018 |
wherever | -0.017 |
Benny | -0.017 |
g | -0.017 |
Token | had |
Token ID | 550 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
site | +0.391 |
her | +0.359 |
office | +0.351 |
ia | +0.345 |
wasteful | +0.317 |
Neg logit contributions | |
mine | -0.299 |
berry | -0.273 |
're | -0.268 |
Benny | -0.268 |
wherever | -0.265 |
Token | a |
Token ID | 257 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
mine | +0.034 |
berry | +0.031 |
're | +0.031 |
Benny | +0.031 |
wherever | +0.030 |
Neg logit contributions | |
site | -0.041 |
her | -0.038 |
office | -0.037 |
ia | -0.037 |
wasteful | -0.033 |
Token | man |
Token ID | 582 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
mine | +0.032 |
berry | +0.029 |
're | +0.029 |
wherever | +0.029 |
Benny | +0.029 |
Neg logit contributions | |
site | -0.033 |
her | -0.030 |
office | -0.030 |
ia | -0.029 |
men | -0.026 |
Token | who |
Token ID | 508 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
site | +0.312 |
her | +0.288 |
office | +0.288 |
ia | +0.271 |
men | +0.246 |
Neg logit contributions | |
mine | -0.290 |
're | -0.269 |
ease | -0.267 |
berry | -0.266 |
Benny | -0.262 |
Token | worked |
Token ID | 3111 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
site | +0.078 |
her | +0.074 |
office | +0.072 |
ia | +0.070 |
wasteful | +0.065 |
Neg logit contributions | |
mine | -0.047 |
berry | -0.043 |
're | -0.041 |
wherever | -0.041 |
Benny | -0.041 |
Token | in |
Token ID | 287 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
site | +0.155 |
ia | +0.141 |
office | +0.140 |
her | +0.138 |
men | +0.127 |
Neg logit contributions | |
mine | -0.121 |
wherever | -0.118 |
Benny | -0.114 |
're | -0.111 |
ease | -0.110 |
Token | an |
Token ID | 281 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
site | +0.007 |
her | +0.007 |
ia | +0.007 |
office | +0.007 |
wasteful | +0.006 |
Neg logit contributions | |
mine | -0.005 |
wherever | -0.005 |
Benny | -0.005 |
ease | -0.005 |
berry | -0.005 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
mine | +0.079 |
g | +0.075 |
berry | +0.074 |
Benny | +0.073 |
ease | +0.073 |
Neg logit contributions | |
site | -0.063 |
her | -0.058 |
ia | -0.057 |
urse | -0.052 |
fighters | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
site | +1.086 |
ia | +1.006 |
her | +0.998 |
office | +0.945 |
wasteful | +0.927 |
Neg logit contributions | |
mine | -0.808 |
wherever | -0.734 |
berry | -0.700 |
're | -0.699 |
Benny | -0.687 |
Token | Everyday |
Token ID | 48154 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
mine | +0.003 |
berry | +0.002 |
Benny | +0.002 |
're | +0.002 |
wherever | +0.002 |
Neg logit contributions | |
site | -0.004 |
her | -0.003 |
ia | -0.003 |
office | -0.003 |
wasteful | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
mine | +0.213 |
ease | +0.194 |
Grac | +0.191 |
're | +0.190 |
berry | +0.189 |
Neg logit contributions | |
site | -0.196 |
office | -0.177 |
ia | -0.167 |
men | -0.163 |
her | -0.162 |
Token | would |
Token ID | 561 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
mine | +0.069 |
berry | +0.061 |
Benny | +0.060 |
're | +0.060 |
wherever | +0.059 |
Neg logit contributions | |
site | -0.127 |
her | -0.119 |
office | -0.117 |
ia | -0.114 |
men | -0.106 |
Token | come |
Token ID | 1282 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
mine | +0.142 |
berry | +0.132 |
Benny | +0.129 |
're | +0.127 |
wherever | +0.127 |
Neg logit contributions | |
site | -0.181 |
her | -0.166 |
office | -0.165 |
ia | -0.157 |
men | -0.145 |
Token | to |
Token ID | 284 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
site | +0.139 |
her | +0.127 |
office | +0.124 |
ia | +0.124 |
yourselves | +0.119 |
Neg logit contributions | |
mine | -0.082 |
Benny | -0.078 |
Will | -0.074 |
're | -0.073 |
berry | -0.073 |
Token | hurry |
Token ID | 23290 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
mine | +0.008 |
berry | +0.007 |
're | +0.007 |
wherever | +0.007 |
Benny | +0.007 |
Neg logit contributions | |
site | -0.011 |
her | -0.010 |
office | -0.010 |
ia | -0.010 |
men | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -3.30e-04 |
Pos logit contributions | |
site | +0.238 |
her | +0.218 |
ia | +0.217 |
office | +0.213 |
wasteful | +0.202 |
Neg logit contributions | |
mine | -0.181 |
wherever | -0.164 |
Benny | -0.162 |
're | -0.161 |
berry | -0.161 |
Token | the |
Token ID | 262 |
Feature activation | -3.74e-01 |
Pos logit contributions | |
site | +0.000 |
her | +0.000 |
office | +0.000 |
ia | +0.000 |
men | +0.000 |
Neg logit contributions | |
mine | -0.000 |
berry | -0.000 |
're | -0.000 |
Benny | -0.000 |
wherever | -0.000 |
Token | post |
Token ID | 1281 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
site | +0.025 |
her | +0.023 |
ia | +0.023 |
office | +0.021 |
wasteful | +0.020 |
Neg logit contributions | |
mine | -0.029 |
berry | -0.028 |
Benny | -0.027 |
wherever | -0.027 |
're | -0.027 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.389 |
her | +0.352 |
ia | +0.341 |
office | +0.328 |
wasteful | +0.308 |
Neg logit contributions | |
mine | -0.503 |
berry | -0.468 |
wherever | -0.461 |
're | -0.459 |
Benny | -0.452 |
Token | to |
Token ID | 284 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
site | +1.065 |
her | +0.973 |
ia | +0.963 |
office | +0.941 |
wasteful | +0.883 |
Neg logit contributions | |
mine | -0.843 |
wherever | -0.756 |
're | -0.753 |
berry | -0.752 |
Benny | -0.736 |
Token | mail |
Token ID | 6920 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
site | +0.005 |
her | +0.005 |
ia | +0.005 |
office | +0.005 |
wasteful | +0.004 |
Neg logit contributions | |
mine | -0.004 |
Benny | -0.003 |
're | -0.003 |
berry | -0.003 |
wherever | -0.003 |
Token | this |
Token ID | 428 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
site | +0.185 |
her | +0.170 |
office | +0.166 |
ia | +0.163 |
wasteful | +0.149 |
Neg logit contributions | |
mine | -0.150 |
berry | -0.137 |
're | -0.134 |
Benny | -0.134 |
wherever | -0.133 |
Token | letter |
Token ID | 3850 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
mine | +0.015 |
berry | +0.014 |
wherever | +0.014 |
're | +0.014 |
Benny | +0.013 |
Neg logit contributions | |
site | -0.013 |
her | -0.012 |
ia | -0.012 |
office | -0.011 |
wasteful | -0.011 |
Token | ." |
Token ID | 526 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
site | +0.205 |
her | +0.188 |
office | +0.185 |
ia | +0.179 |
men | +0.163 |
Neg logit contributions | |
mine | -0.187 |
berry | -0.172 |
Benny | -0.168 |
're | -0.167 |
ease | -0.166 |
Token | would |
Token ID | 561 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
mine | +0.065 |
berry | +0.058 |
wherever | +0.057 |
Benny | +0.056 |
're | +0.056 |
Neg logit contributions | |
site | -0.121 |
her | -0.114 |
office | -0.113 |
ia | -0.111 |
men | -0.103 |
Token | go |
Token ID | 467 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
mine | +0.133 |
berry | +0.129 |
berries | +0.126 |
Grac | +0.122 |
Benny | +0.122 |
Neg logit contributions | |
site | -0.158 |
office | -0.147 |
her | -0.145 |
ia | -0.138 |
mistake | -0.131 |
Token | to |
Token ID | 284 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
site | +0.327 |
office | +0.305 |
her | +0.297 |
ia | +0.272 |
men | +0.260 |
Neg logit contributions | |
mine | -0.304 |
berry | -0.275 |
're | -0.271 |
mattered | -0.268 |
ease | -0.260 |
Token | the |
Token ID | 262 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
site | +0.004 |
her | +0.004 |
office | +0.004 |
ia | +0.004 |
men | +0.003 |
Neg logit contributions | |
mine | -0.004 |
berry | -0.003 |
're | -0.003 |
Benny | -0.003 |
wherever | -0.003 |
Token | post |
Token ID | 1281 |
Feature activation | -6.03e+00 |
Pos logit contributions | |
site | +0.027 |
ia | +0.025 |
her | +0.024 |
office | +0.022 |
wasteful | +0.021 |
Neg logit contributions | |
mine | -0.028 |
berry | -0.028 |
Benny | -0.027 |
wherever | -0.027 |
g | -0.026 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.150 |
her | +0.108 |
office | +0.101 |
ia | +0.090 |
men | +0.051 |
Neg logit contributions | |
mine | -0.753 |
berry | -0.718 |
're | -0.713 |
Benny | -0.712 |
wherever | -0.707 |
Token | and |
Token ID | 290 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
site | +1.073 |
her | +0.988 |
office | +0.961 |
ia | +0.958 |
wasteful | +0.879 |
Neg logit contributions | |
mine | -0.840 |
berry | -0.763 |
wherever | -0.750 |
're | -0.749 |
Benny | -0.747 |
Token | he |
Token ID | 339 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
mine | +0.033 |
berry | +0.031 |
berries | +0.029 |
Benny | +0.029 |
're | +0.029 |
Neg logit contributions | |
site | -0.045 |
office | -0.042 |
her | -0.041 |
ia | -0.038 |
men | -0.036 |
Token | would |
Token ID | 561 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
mine | +0.062 |
berry | +0.057 |
Benny | +0.055 |
wherever | +0.054 |
're | +0.053 |
Neg logit contributions | |
site | -0.112 |
her | -0.105 |
office | -0.105 |
ia | -0.100 |
men | -0.093 |
Token | look |
Token ID | 804 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
mine | +0.124 |
berry | +0.119 |
Benny | +0.116 |
berries | +0.113 |
mattered | +0.113 |
Neg logit contributions | |
site | -0.155 |
office | -0.145 |
her | -0.143 |
ia | -0.135 |
mistake | -0.126 |
Token | at |
Token ID | 379 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
mine | +0.003 |
berry | +0.003 |
're | +0.003 |
Benny | +0.003 |
berries | +0.003 |
Neg logit contributions | |
site | -0.003 |
her | -0.003 |
office | -0.002 |
ia | -0.002 |
men | -0.002 |
Token | taking |
Token ID | 2263 |
Feature activation | -3.60e+00 |
Pos logit contributions | |
site | +0.067 |
office | +0.058 |
ia | +0.057 |
her | +0.057 |
men | +0.051 |
Neg logit contributions | |
Benny | -0.062 |
mine | -0.061 |
wherever | -0.060 |
ease | -0.059 |
're | -0.059 |
Token | rocks |
Token ID | 12586 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
site | +0.275 |
her | +0.255 |
office | +0.245 |
ia | +0.237 |
men | +0.225 |
Neg logit contributions | |
mine | -0.265 |
berry | -0.246 |
're | -0.237 |
berries | -0.229 |
wherever | -0.226 |
Token | to |
Token ID | 284 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
mine | +0.114 |
berry | +0.108 |
Benny | +0.103 |
Grac | +0.103 |
're | +0.103 |
Neg logit contributions | |
site | -0.125 |
her | -0.119 |
office | -0.113 |
ia | -0.106 |
men | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
mine | +0.001 |
berry | +0.001 |
're | +0.001 |
Benny | +0.001 |
wherever | +0.001 |
Neg logit contributions | |
site | -0.001 |
her | -0.001 |
office | -0.001 |
ia | -0.001 |
wasteful | -0.001 |
Token | building |
Token ID | 2615 |
Feature activation | -5.76e+00 |
Pos logit contributions | |
site | +0.022 |
her | +0.020 |
office | +0.019 |
ia | +0.019 |
wasteful | +0.017 |
Neg logit contributions | |
mine | -0.025 |
berry | -0.023 |
Benny | -0.023 |
're | -0.023 |
wherever | -0.023 |
Token | site |
Token ID | 2524 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.464 |
ia | +0.434 |
her | +0.431 |
office | +0.414 |
wasteful | +0.394 |
Neg logit contributions | |
mine | -0.368 |
wherever | -0.341 |
're | -0.335 |
berry | -0.332 |
ease | -0.323 |
Token | down |
Token ID | 866 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
site | +1.052 |
her | +0.969 |
ia | +0.951 |
office | +0.946 |
wasteful | +0.865 |
Neg logit contributions | |
mine | -0.841 |
berry | -0.770 |
wherever | -0.766 |
're | -0.762 |
Benny | -0.751 |
Token | the |
Token ID | 262 |
Feature activation | -3.89e-01 |
Pos logit contributions | |
mine | +0.023 |
wherever | +0.021 |
berry | +0.021 |
Benny | +0.020 |
ease | +0.020 |
Neg logit contributions | |
site | -0.036 |
ia | -0.034 |
her | -0.033 |
office | -0.033 |
wasteful | -0.031 |
Token | road |
Token ID | 2975 |
Feature activation | -5.90e+00 |
Pos logit contributions | |
ia | +0.023 |
site | +0.022 |
her | +0.021 |
wasteful | +0.020 |
office | +0.019 |
Neg logit contributions | |
mine | -0.030 |
wherever | -0.029 |
berry | -0.028 |
Benny | -0.028 |
're | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
site | +0.486 |
ia | +0.455 |
her | +0.453 |
office | +0.439 |
wasteful | +0.419 |
Neg logit contributions | |
mine | -0.376 |
wherever | -0.341 |
're | -0.338 |
berry | -0.337 |
Benny | -0.324 |
Token | Kim |
Token ID | 6502 |
Feature activation | -5.69e+00 |
Pos logit contributions | |
mine | +0.003 |
're | +0.003 |
berry | +0.003 |
wherever | +0.003 |
ease | +0.002 |
Neg logit contributions | |
site | -0.004 |
her | -0.004 |
office | -0.004 |
ia | -0.004 |
men | -0.003 |
Token | huge |
Token ID | 3236 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
site | +0.061 |
her | +0.056 |
ia | +0.054 |
office | +0.053 |
wasteful | +0.047 |
Neg logit contributions | |
mine | -0.071 |
berry | -0.069 |
Benny | -0.068 |
wherever | -0.066 |
're | -0.066 |
Token | crane |
Token ID | 41175 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
site | +0.149 |
her | +0.142 |
ia | +0.133 |
office | +0.130 |
wasteful | +0.120 |
Neg logit contributions | |
mine | -0.151 |
berry | -0.146 |
Benny | -0.142 |
g | -0.138 |
wherever | -0.138 |
Token | at |
Token ID | 379 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
mine | +0.087 |
berry | +0.081 |
wherever | +0.079 |
're | +0.077 |
Benny | +0.077 |
Neg logit contributions | |
site | -0.110 |
her | -0.102 |
ia | -0.101 |
office | -0.097 |
wasteful | -0.092 |
Token | a |
Token ID | 257 |
Feature activation | -9.25e-01 |
Pos logit contributions | |
mine | +0.070 |
wherever | +0.064 |
ease | +0.063 |
Benny | +0.063 |
berry | +0.060 |
Neg logit contributions | |
site | -0.121 |
ia | -0.119 |
her | -0.115 |
office | -0.110 |
wasteful | -0.108 |
Token | construction |
Token ID | 5103 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
site | +0.066 |
ia | +0.064 |
her | +0.061 |
office | +0.058 |
wasteful | +0.057 |
Neg logit contributions | |
mine | -0.065 |
berry | -0.064 |
Benny | -0.063 |
wherever | -0.063 |
're | -0.060 |
Token | site |
Token ID | 2524 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
ia | +0.153 |
wasteful | +0.106 |
urse | +0.099 |
wen | +0.089 |
her | +0.088 |
Neg logit contributions | |
mine | -0.554 |
wherever | -0.529 |
berry | -0.525 |
mineral | -0.509 |
trophy | -0.505 |
Token | . |
Token ID | 13 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
ia | +1.123 |
site | +1.094 |
her | +1.032 |
wasteful | +1.009 |
office | +1.006 |
Neg logit contributions | |
mine | -0.697 |
wherever | -0.674 |
berry | -0.597 |
whenever | -0.590 |
're | -0.586 |
Token | The |
Token ID | 383 |
Feature activation | +8.58e-01 |
Pos logit contributions | |
mine | +0.017 |
berry | +0.016 |
're | +0.015 |
Benny | +0.015 |
wherever | +0.015 |
Neg logit contributions | |
site | -0.028 |
her | -0.026 |
office | -0.026 |
ia | -0.025 |
men | -0.023 |
Token | crane |
Token ID | 41175 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
're | +0.042 |
mine | +0.042 |
wherever | +0.040 |
ease | +0.039 |
berry | +0.038 |
Neg logit contributions | |
ia | -0.078 |
site | -0.074 |
her | -0.072 |
office | -0.070 |
urse | -0.069 |
Token | was |
Token ID | 373 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
mine | +0.075 |
berry | +0.068 |
're | +0.067 |
Benny | +0.066 |
wherever | +0.065 |
Neg logit contributions | |
site | -0.120 |
her | -0.112 |
office | -0.110 |
ia | -0.110 |
wasteful | -0.102 |
Token | lifting |
Token ID | 16842 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
ia | +0.064 |
site | +0.064 |
urse | +0.060 |
office | +0.058 |
yourselves | +0.056 |
Neg logit contributions | |
g | -0.039 |
wherever | -0.037 |
Benny | -0.037 |
're | -0.036 |
ease | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -5.02e-01 |
Pos logit contributions | |
office | +0.088 |
site | +0.087 |
her | +0.082 |
ia | +0.073 |
x | +0.071 |
Neg logit contributions | |
mine | -0.089 |
superstar | -0.080 |
ease | -0.078 |
're | -0.078 |
berry | -0.078 |
Token | Sophia |
Token ID | 39953 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
site | +0.041 |
her | +0.040 |
office | +0.039 |
ia | +0.037 |
men | +0.034 |
Neg logit contributions | |
mine | -0.032 |
berry | -0.029 |
're | -0.028 |
berries | -0.027 |
ease | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
site | +0.147 |
her | +0.137 |
office | +0.135 |
ia | +0.132 |
men | +0.117 |
Neg logit contributions | |
mine | -0.143 |
berry | -0.131 |
're | -0.129 |
Benny | -0.129 |
ease | -0.129 |
Token | to |
Token ID | 284 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
site | +0.034 |
her | +0.033 |
office | +0.033 |
ia | +0.030 |
men | +0.028 |
Neg logit contributions | |
mine | -0.034 |
berry | -0.031 |
ease | -0.030 |
're | -0.030 |
berries | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
mine | +0.006 |
berries | +0.006 |
berry | +0.006 |
're | +0.006 |
oft | +0.006 |
Neg logit contributions | |
site | -0.007 |
her | -0.006 |
office | -0.006 |
ia | -0.006 |
x | -0.006 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.016 |
her | +0.014 |
ia | +0.014 |
office | +0.014 |
wasteful | +0.013 |
Neg logit contributions | |
mine | -0.016 |
berry | -0.015 |
Benny | -0.015 |
wherever | -0.015 |
're | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
site | +1.089 |
her | +1.000 |
ia | +0.998 |
wasteful | +0.947 |
office | +0.940 |
Neg logit contributions | |
mine | -0.809 |
wherever | -0.721 |
berry | -0.713 |
're | -0.702 |
Benny | -0.686 |
Token | Sophia |
Token ID | 39953 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
site | +0.003 |
her | +0.003 |
office | +0.003 |
ia | +0.003 |
wasteful | +0.002 |
Neg logit contributions | |
mine | -0.002 |
berry | -0.002 |
're | -0.002 |
Benny | -0.002 |
wherever | -0.002 |
Token | couldn |
Token ID | 3521 |
Feature activation | +4.60e-01 |
Pos logit contributions | |
site | +0.174 |
her | +0.160 |
office | +0.157 |
ia | +0.154 |
wasteful | +0.144 |
Neg logit contributions | |
mine | -0.115 |
berry | -0.105 |
're | -0.103 |
Benny | -0.102 |
wherever | -0.100 |
Token | 't |
Token ID | 470 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
mine | +0.027 |
berry | +0.024 |
're | +0.024 |
Benny | +0.024 |
wherever | +0.023 |
Neg logit contributions | |
site | -0.042 |
her | -0.039 |
office | -0.038 |
ia | -0.038 |
men | -0.035 |
Token | believe |
Token ID | 1975 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
mine | +0.174 |
berry | +0.170 |
Benny | +0.165 |
wherever | +0.162 |
berries | +0.162 |
Neg logit contributions | |
site | -0.202 |
her | -0.189 |
office | -0.184 |
ia | -0.176 |
x | -0.163 |
Token | take |
Token ID | 1011 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
mine | +0.006 |
berry | +0.006 |
berries | +0.006 |
Benny | +0.006 |
mattered | +0.006 |
Neg logit contributions | |
site | -0.008 |
her | -0.007 |
office | -0.007 |
ia | -0.007 |
mistake | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
site | +0.008 |
her | +0.008 |
office | +0.008 |
ia | +0.007 |
x | +0.007 |
Neg logit contributions | |
mine | -0.007 |
berry | -0.007 |
're | -0.007 |
wherever | -0.006 |
berries | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | +8.17e-02 |
Pos logit contributions | |
mine | +0.166 |
berry | +0.154 |
ease | +0.145 |
Benny | +0.145 |
're | +0.144 |
Neg logit contributions | |
site | -0.167 |
office | -0.162 |
her | -0.158 |
x | -0.142 |
ia | -0.138 |
Token | the |
Token ID | 262 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
mine | +0.005 |
Benny | +0.005 |
wherever | +0.005 |
berry | +0.005 |
're | +0.005 |
Neg logit contributions | |
site | -0.007 |
her | -0.006 |
ia | -0.006 |
office | -0.006 |
wasteful | -0.006 |
Token | post |
Token ID | 1281 |
Feature activation | -5.95e+00 |
Pos logit contributions | |
site | +0.015 |
ia | +0.014 |
her | +0.014 |
office | +0.013 |
wasteful | +0.013 |
Neg logit contributions | |
mine | -0.015 |
berry | -0.015 |
wherever | -0.014 |
Benny | -0.014 |
g | -0.014 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.099 |
her | +0.065 |
ia | +0.056 |
wasteful | +0.031 |
office | +0.029 |
Neg logit contributions | |
mine | -0.780 |
berry | -0.749 |
wherever | -0.748 |
're | -0.735 |
Benny | -0.731 |
Token | . |
Token ID | 13 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
site | +1.084 |
ia | +0.988 |
her | +0.985 |
office | +0.950 |
wasteful | +0.909 |
Neg logit contributions | |
mine | -0.822 |
wherever | -0.750 |
're | -0.735 |
berry | -0.727 |
Benny | -0.724 |
Token | She |
Token ID | 1375 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
mine | +0.002 |
berry | +0.002 |
Benny | +0.001 |
're | +0.001 |
wherever | +0.001 |
Neg logit contributions | |
site | -0.003 |
her | -0.003 |
office | -0.003 |
ia | -0.003 |
wasteful | -0.003 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
site | +0.086 |
her | +0.079 |
office | +0.077 |
ia | +0.076 |
wasteful | +0.071 |
Neg logit contributions | |
mine | -0.057 |
berry | -0.052 |
're | -0.051 |
Benny | -0.051 |
wherever | -0.050 |
Token | mail |
Token ID | 6920 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
mine | +0.248 |
're | +0.225 |
berry | +0.222 |
ease | +0.210 |
mattered | +0.206 |
Neg logit contributions | |
site | -0.281 |
office | -0.262 |
her | -0.253 |
ia | -0.238 |
men | -0.231 |
Token | because |
Token ID | 780 |
Feature activation | +2.41e+00 |
Pos logit contributions | |
site | +0.164 |
her | +0.151 |
office | +0.151 |
ia | +0.143 |
men | +0.133 |
Neg logit contributions | |
mine | -0.149 |
berry | -0.137 |
're | -0.135 |
Benny | -0.132 |
ease | -0.132 |
Token | lot |
Token ID | 1256 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
site | +0.079 |
her | +0.072 |
office | +0.072 |
ia | +0.070 |
mistake | +0.065 |
Neg logit contributions | |
mine | -0.092 |
berry | -0.083 |
wherever | -0.080 |
're | -0.080 |
berries | -0.079 |
Token | of |
Token ID | 286 |
Feature activation | +9.60e-01 |
Pos logit contributions | |
mine | +0.137 |
berry | +0.122 |
're | +0.118 |
Benny | +0.118 |
wherever | +0.117 |
Neg logit contributions | |
site | -0.202 |
her | -0.191 |
office | -0.183 |
ia | -0.182 |
men | -0.167 |
Token | fun |
Token ID | 1257 |
Feature activation | -4.63e-01 |
Pos logit contributions | |
mine | +0.083 |
berry | +0.077 |
wherever | +0.075 |
're | +0.075 |
Benny | +0.075 |
Neg logit contributions | |
site | -0.062 |
her | -0.056 |
office | -0.055 |
ia | -0.053 |
men | -0.049 |
Token | in |
Token ID | 287 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
site | +0.039 |
her | +0.035 |
office | +0.035 |
ia | +0.034 |
men | +0.031 |
Neg logit contributions | |
mine | -0.031 |
berry | -0.028 |
're | -0.028 |
Benny | -0.028 |
ease | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
site | +0.005 |
office | +0.005 |
her | +0.005 |
ia | +0.005 |
mistake | +0.004 |
Neg logit contributions | |
mine | -0.006 |
berry | -0.005 |
're | -0.005 |
Benny | -0.005 |
wherever | -0.005 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.021 |
her | +0.019 |
office | +0.018 |
ia | +0.018 |
men | +0.016 |
Neg logit contributions | |
mine | -0.025 |
berry | -0.023 |
're | -0.023 |
wherever | -0.022 |
Benny | -0.022 |
Token | ! |
Token ID | 0 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
site | +1.027 |
her | +0.936 |
office | +0.926 |
ia | +0.892 |
men | +0.818 |
Neg logit contributions | |
mine | -0.889 |
berry | -0.823 |
Benny | -0.809 |
're | -0.804 |
wherever | -0.790 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
mine | +0.087 |
berry | +0.079 |
're | +0.075 |
g | +0.075 |
wherever | +0.074 |
Neg logit contributions | |
site | -0.098 |
her | -0.091 |
office | -0.088 |
ia | -0.088 |
men | -0.077 |
Token | Sam |
Token ID | 16305 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
site | +0.101 |
her | +0.090 |
ia | +0.087 |
office | +0.086 |
men | +0.071 |
Neg logit contributions | |
mine | -0.154 |
're | -0.143 |
g | -0.142 |
ease | -0.140 |
wherever | -0.139 |
Token | is |
Token ID | 318 |
Feature activation | -2.75e+00 |
Pos logit contributions | |
site | +0.068 |
her | +0.065 |
office | +0.064 |
ia | +0.064 |
mistake | +0.058 |
Neg logit contributions | |
mine | -0.033 |
berry | -0.029 |
wherever | -0.029 |
Grac | -0.028 |
Benny | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
site | +0.237 |
her | +0.218 |
office | +0.213 |
ia | +0.210 |
wasteful | +0.192 |
Neg logit contributions | |
mine | -0.177 |
berry | -0.162 |
're | -0.159 |
Benny | -0.158 |
wherever | -0.157 |
Token | 't |
Token ID | 470 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
mine | +0.027 |
berry | +0.024 |
're | +0.024 |
Benny | +0.024 |
wherever | +0.023 |
Neg logit contributions | |
site | -0.042 |
her | -0.039 |
office | -0.038 |
ia | -0.038 |
men | -0.035 |
Token | believe |
Token ID | 1975 |
Feature activation | +4.45e+00 |
Pos logit contributions | |
mine | +0.174 |
berry | +0.170 |
Benny | +0.165 |
wherever | +0.162 |
berries | +0.162 |
Neg logit contributions | |
site | -0.202 |
her | -0.189 |
office | -0.184 |
ia | -0.176 |
x | -0.163 |
Token | how |
Token ID | 703 |
Feature activation | +9.48e-02 |
Pos logit contributions | |
mine | +0.282 |
're | +0.273 |
berry | +0.266 |
ease | +0.256 |
Benny | +0.246 |
Neg logit contributions | |
site | -0.369 |
office | -0.340 |
ia | -0.339 |
her | -0.333 |
yourselves | -0.312 |
Token | impressive |
Token ID | 8036 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
mine | +0.008 |
berry | +0.007 |
're | +0.007 |
Benny | +0.007 |
wherever | +0.007 |
Neg logit contributions | |
site | -0.006 |
her | -0.006 |
office | -0.006 |
ia | -0.006 |
men | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
site | +0.101 |
her | +0.093 |
office | +0.092 |
ia | +0.091 |
men | +0.084 |
Neg logit contributions | |
mine | -0.074 |
berry | -0.067 |
're | -0.067 |
ease | -0.064 |
Benny | -0.064 |
Token | office |
Token ID | 2607 |
Feature activation | -1.28e+01 |
Pos logit contributions | |
site | +0.003 |
ia | +0.003 |
her | +0.003 |
cks | +0.003 |
yourselves | +0.002 |
Neg logit contributions | |
mine | -0.007 |
g | -0.006 |
berry | -0.006 |
wherever | -0.006 |
stuffed | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
site | +1.110 |
her | +1.023 |
office | +1.005 |
ia | +0.983 |
men | +0.904 |
Neg logit contributions | |
mine | -0.802 |
berry | -0.733 |
Benny | -0.718 |
're | -0.717 |
wherever | -0.710 |
Token | ; |
Token ID | 26 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
site | +0.070 |
her | +0.063 |
ia | +0.063 |
office | +0.062 |
wasteful | +0.058 |
Neg logit contributions | |
mine | -0.052 |
wherever | -0.048 |
're | -0.047 |
Benny | -0.047 |
berry | -0.046 |
Token | it |
Token ID | 340 |
Feature activation | +7.85e-01 |
Pos logit contributions | |
site | +0.061 |
office | +0.058 |
her | +0.056 |
ia | +0.054 |
men | +0.051 |
Neg logit contributions | |
mine | -0.050 |
berry | -0.044 |
Benny | -0.043 |
're | -0.043 |
ease | -0.042 |
Token | had |
Token ID | 550 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
mine | +0.048 |
berry | +0.044 |
're | +0.043 |
Benny | +0.043 |
wherever | +0.042 |
Neg logit contributions | |
site | -0.069 |
her | -0.064 |
office | -0.063 |
ia | -0.062 |
wasteful | -0.057 |
Token | lots |
Token ID | 6041 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
mine | +0.048 |
're | +0.044 |
wherever | +0.044 |
berry | +0.043 |
ease | +0.043 |
Neg logit contributions | |
site | -0.060 |
her | -0.054 |
office | -0.053 |
ia | -0.053 |
wasteful | -0.049 |
Token | time |
Token ID | 640 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
site | +0.075 |
office | +0.065 |
her | +0.065 |
ia | +0.058 |
men | +0.058 |
Neg logit contributions | |
mine | -0.087 |
berry | -0.082 |
're | -0.082 |
Benny | -0.082 |
wherever | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
site | +0.241 |
ia | +0.220 |
her | +0.217 |
office | +0.213 |
men | +0.188 |
Neg logit contributions | |
mine | -0.248 |
berry | -0.226 |
ease | -0.223 |
're | -0.218 |
wherever | -0.216 |
Token | there |
Token ID | 612 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
site | +0.042 |
office | +0.038 |
her | +0.038 |
ia | +0.036 |
men | +0.035 |
Neg logit contributions | |
mine | -0.029 |
Benny | -0.026 |
wherever | -0.026 |
berry | -0.026 |
're | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | -9.07e-01 |
Pos logit contributions | |
site | +0.043 |
her | +0.039 |
ia | +0.039 |
office | +0.038 |
mistake | +0.034 |
Neg logit contributions | |
mine | -0.039 |
berry | -0.035 |
're | -0.034 |
Benny | -0.034 |
wherever | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
site | +0.061 |
her | +0.057 |
office | +0.054 |
ia | +0.054 |
men | +0.047 |
Neg logit contributions | |
mine | -0.074 |
berry | -0.069 |
're | -0.068 |
wherever | -0.067 |
Benny | -0.066 |
Token | bunny |
Token ID | 44915 |
Feature activation | +7.90e+00 |
Pos logit contributions | |
site | +0.079 |
her | +0.071 |
office | +0.069 |
ia | +0.067 |
wasteful | +0.061 |
Neg logit contributions | |
mine | -0.082 |
berry | -0.078 |
Benny | -0.077 |
wherever | -0.075 |
're | -0.075 |
Token | named |
Token ID | 3706 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
mine | +0.600 |
ease | +0.568 |
're | +0.561 |
berry | +0.534 |
Grac | +0.525 |
Neg logit contributions | |
site | -0.585 |
office | -0.542 |
her | -0.540 |
ia | -0.519 |
mistake | -0.457 |
Token | Benny |
Token ID | 44275 |
Feature activation | +7.82e+00 |
Pos logit contributions | |
site | +0.013 |
her | +0.012 |
ia | +0.012 |
office | +0.012 |
x | +0.011 |
Neg logit contributions | |
mine | -0.006 |
're | -0.005 |
berry | -0.004 |
mattered | -0.004 |
ease | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.39e-02 |
Pos logit contributions | |
mine | +0.599 |
berry | +0.556 |
're | +0.546 |
ease | +0.541 |
Grac | +0.535 |
Neg logit contributions | |
site | -0.561 |
her | -0.530 |
office | -0.510 |
ia | -0.503 |
mistake | -0.448 |
Token | Benny |
Token ID | 44275 |
Feature activation | +7.86e+00 |
Pos logit contributions | |
site | +0.002 |
her | +0.002 |
ia | +0.002 |
office | +0.002 |
x | +0.002 |
Neg logit contributions | |
mine | -0.001 |
're | -0.001 |
mattered | -0.001 |
superstar | -0.001 |
ease | -0.001 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
mine | +0.426 |
berry | +0.381 |
're | +0.369 |
Benny | +0.362 |
wherever | +0.362 |
Neg logit contributions | |
site | -0.752 |
her | -0.706 |
office | -0.688 |
ia | -0.682 |
wasteful | -0.624 |
Token | bug |
Token ID | 5434 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
site | +0.021 |
ia | +0.021 |
her | +0.020 |
office | +0.020 |
mistake | +0.019 |
Neg logit contributions | |
mine | -0.019 |
Emmy | -0.017 |
Benny | -0.016 |
wherever | -0.016 |
ease | -0.016 |
Token | on |
Token ID | 319 |
Feature activation | +8.72e-01 |
Pos logit contributions | |
mine | +0.409 |
Emmy | +0.384 |
're | +0.377 |
Benny | +0.374 |
superstar | +0.373 |
Neg logit contributions | |
her | -0.348 |
site | -0.344 |
office | -0.329 |
ia | -0.328 |
ep | -0.289 |
Token | it |
Token ID | 340 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
mine | +0.062 |
're | +0.059 |
berry | +0.057 |
mattered | +0.056 |
oft | +0.055 |
Neg logit contributions | |
site | -0.068 |
her | -0.064 |
ia | -0.060 |
office | -0.060 |
mistake | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
mine | +0.054 |
berry | +0.050 |
're | +0.049 |
Benny | +0.048 |
ease | +0.047 |
Neg logit contributions | |
site | -0.048 |
her | -0.046 |
office | -0.043 |
ia | -0.042 |
urgent | -0.037 |
Token | The |
Token ID | 383 |
Feature activation | +7.24e-01 |
Pos logit contributions | |
mine | +0.003 |
're | +0.003 |
berry | +0.003 |
wherever | +0.003 |
ease | +0.003 |
Neg logit contributions | |
site | -0.004 |
her | -0.004 |
ia | -0.004 |
office | -0.004 |
urgent | -0.003 |
Token | bug |
Token ID | 5434 |
Feature activation | +5.29e+00 |
Pos logit contributions | |
mine | +0.045 |
're | +0.039 |
wherever | +0.039 |
berry | +0.038 |
ease | +0.038 |
Neg logit contributions | |
site | -0.064 |
ia | -0.061 |
her | -0.060 |
office | -0.060 |
mistake | -0.055 |
Token | stood |
Token ID | 6204 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
mine | +0.389 |
're | +0.356 |
ease | +0.348 |
wherever | +0.347 |
berry | +0.344 |
Neg logit contributions | |
site | -0.393 |
her | -0.377 |
ia | -0.376 |
office | -0.344 |
ep | -0.317 |
Token | there |
Token ID | 612 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
site | +0.106 |
her | +0.096 |
office | +0.094 |
ia | +0.092 |
wasteful | +0.084 |
Neg logit contributions | |
mine | -0.096 |
berry | -0.089 |
Benny | -0.087 |
wherever | -0.087 |
're | -0.086 |
Token | for |
Token ID | 329 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
site | +0.029 |
her | +0.027 |
office | +0.026 |
ia | +0.026 |
mistake | +0.023 |
Neg logit contributions | |
mine | -0.034 |
berry | -0.032 |
Benny | -0.031 |
Grac | -0.030 |
berries | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -9.71e-01 |
Pos logit contributions | |
site | +0.284 |
her | +0.261 |
office | +0.251 |
ia | +0.248 |
men | +0.223 |
Neg logit contributions | |
mine | -0.296 |
berry | -0.278 |
're | -0.270 |
wherever | -0.262 |
Benny | -0.261 |
Token | moment |
Token ID | 2589 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
site | +0.074 |
her | +0.067 |
ia | +0.065 |
office | +0.064 |
mistake | +0.061 |
Neg logit contributions | |
mine | -0.072 |
berry | -0.063 |
berries | -0.062 |
're | -0.062 |
ease | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
mine | +0.025 |
berry | +0.023 |
berries | +0.022 |
Grac | +0.021 |
superstar | +0.021 |
Neg logit contributions | |
office | -0.021 |
site | -0.021 |
her | -0.021 |
ia | -0.019 |
wasteful | -0.018 |
Token | mom |
Token ID | 1995 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
mine | +0.042 |
berry | +0.039 |
berries | +0.036 |
're | +0.035 |
ease | +0.035 |
Neg logit contributions | |
site | -0.041 |
office | -0.040 |
her | -0.039 |
ia | -0.036 |
wasteful | -0.036 |
Token | my |
Token ID | 1820 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
office | +0.047 |
her | +0.045 |
site | +0.043 |
ia | +0.040 |
mistake | +0.039 |
Neg logit contributions | |
mine | -0.031 |
berry | -0.029 |
ease | -0.029 |
Benny | -0.028 |
superstar | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
mine | +0.131 |
berry | +0.125 |
berries | +0.120 |
ease | +0.118 |
Grac | +0.117 |
Neg logit contributions | |
office | -0.115 |
her | -0.113 |
site | -0.112 |
ia | -0.094 |
mistake | -0.093 |
Token | hug |
Token ID | 16225 |
Feature activation | +8.13e+00 |
Pos logit contributions | |
mine | +0.001 |
berry | +0.001 |
berries | +0.001 |
superstar | +0.001 |
oft | +0.001 |
Neg logit contributions | |
site | -0.001 |
her | -0.001 |
office | -0.001 |
ia | -0.001 |
mistake | -0.001 |
Token | whenever |
Token ID | 8797 |
Feature activation | +5.63e+00 |
Pos logit contributions | |
mine | +0.455 |
're | +0.436 |
berry | +0.433 |
ease | +0.418 |
oft | +0.414 |
Neg logit contributions | |
site | -0.712 |
her | -0.681 |
office | -0.667 |
ia | -0.651 |
yourselves | -0.612 |
Token | she |
Token ID | 673 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
mine | +0.342 |
berry | +0.306 |
're | +0.301 |
ease | +0.286 |
wherever | +0.285 |
Neg logit contributions | |
site | -0.508 |
office | -0.464 |
her | -0.462 |
ia | -0.456 |
men | -0.425 |
Token | needed |
Token ID | 2622 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
mine | +0.074 |
berry | +0.066 |
wherever | +0.066 |
ease | +0.066 |
Grac | +0.066 |
Neg logit contributions | |
site | -0.084 |
her | -0.082 |
office | -0.079 |
ia | -0.077 |
wasteful | -0.071 |
Token | it |
Token ID | 340 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
site | +0.110 |
her | +0.103 |
office | +0.102 |
x | +0.093 |
ia | +0.093 |
Neg logit contributions | |
mine | -0.121 |
berry | -0.109 |
're | -0.107 |
oft | -0.105 |
superstar | -0.105 |
Token | . |
Token ID | 13 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
mine | +0.050 |
berry | +0.047 |
berries | +0.044 |
Grac | +0.043 |
superstar | +0.043 |
Neg logit contributions | |
site | -0.039 |
her | -0.038 |
office | -0.036 |
ia | -0.034 |
mistake | -0.032 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
mine | +0.013 |
're | +0.012 |
ease | +0.012 |
berry | +0.012 |
wherever | +0.011 |
Neg logit contributions | |
site | -0.015 |
her | -0.014 |
ia | -0.014 |
office | -0.014 |
mistake | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
mine | +0.047 |
berry | +0.044 |
're | +0.041 |
Benny | +0.041 |
berries | +0.041 |
Neg logit contributions | |
site | -0.059 |
her | -0.056 |
office | -0.054 |
ia | -0.052 |
mistake | -0.048 |
Token | the |
Token ID | 262 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
mine | +0.008 |
berry | +0.007 |
Benny | +0.007 |
wherever | +0.007 |
're | +0.007 |
Neg logit contributions | |
site | -0.010 |
her | -0.009 |
office | -0.009 |
ia | -0.009 |
wasteful | -0.008 |
Token | letter |
Token ID | 3850 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
site | +0.007 |
ia | +0.007 |
her | +0.006 |
cks | +0.006 |
urse | +0.006 |
Neg logit contributions | |
mine | -0.009 |
wherever | -0.008 |
're | -0.008 |
ease | -0.008 |
g | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
site | +0.186 |
her | +0.173 |
office | +0.168 |
ia | +0.163 |
men | +0.144 |
Neg logit contributions | |
mine | -0.183 |
berry | -0.172 |
're | -0.167 |
Benny | -0.165 |
berries | -0.162 |
Token | He |
Token ID | 679 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
mine | +0.001 |
berry | +0.001 |
're | +0.001 |
ease | +0.001 |
mattered | +0.001 |
Neg logit contributions | |
site | -0.001 |
ia | -0.001 |
her | -0.001 |
office | -0.001 |
men | -0.001 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.01e+00 |
Pos logit contributions | |
site | +0.112 |
her | +0.106 |
office | +0.098 |
ia | +0.098 |
ep | +0.092 |
Neg logit contributions | |
mine | -0.113 |
berry | -0.103 |
're | -0.101 |
berries | -0.101 |
wherever | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
mine | +0.501 |
berry | +0.452 |
're | +0.452 |
ease | +0.445 |
berries | +0.439 |
Neg logit contributions | |
site | -0.388 |
her | -0.375 |
office | -0.360 |
ia | -0.349 |
urgent | -0.308 |
Token | dropped |
Token ID | 5710 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
mine | +0.044 |
berry | +0.041 |
berries | +0.039 |
're | +0.037 |
ease | +0.036 |
Neg logit contributions | |
site | -0.053 |
her | -0.050 |
ia | -0.048 |
office | -0.048 |
men | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
site | +0.183 |
office | +0.160 |
her | +0.157 |
men | +0.153 |
ia | +0.151 |
Neg logit contributions | |
mine | -0.250 |
're | -0.237 |
berry | -0.226 |
oft | -0.222 |
mattered | -0.222 |
Token | letter |
Token ID | 3850 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
site | +0.005 |
her | +0.005 |
office | +0.005 |
ia | +0.005 |
wasteful | +0.004 |
Neg logit contributions | |
mine | -0.004 |
berry | -0.004 |
're | -0.004 |
wherever | -0.004 |
Benny | -0.004 |
Token | in |
Token ID | 287 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
site | +0.186 |
her | +0.175 |
office | +0.169 |
ia | +0.161 |
men | +0.146 |
Neg logit contributions | |
mine | -0.181 |
berry | -0.169 |
're | -0.166 |
Benny | -0.164 |
Grac | -0.162 |
Token | the |
Token ID | 262 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
mattered | +0.256 |
're | +0.252 |
mine | +0.238 |
berry | +0.236 |
superstar | +0.234 |
Neg logit contributions | |
site | -0.186 |
office | -0.167 |
x | -0.159 |
her | -0.155 |
ia | -0.152 |
Token | bird |
Token ID | 6512 |
Feature activation | +2.91e+00 |
Pos logit contributions | |
site | +0.019 |
her | +0.018 |
office | +0.018 |
ia | +0.018 |
men | +0.016 |
Neg logit contributions | |
mine | -0.015 |
're | -0.013 |
berry | -0.013 |
Benny | -0.013 |
wherever | -0.013 |
Token | with |
Token ID | 351 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
mine | +0.263 |
're | +0.241 |
mattered | +0.238 |
ease | +0.237 |
berry | +0.231 |
Neg logit contributions | |
site | -0.165 |
office | -0.162 |
ia | -0.156 |
her | -0.149 |
mistake | -0.132 |
Token | a |
Token ID | 257 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
mine | +0.131 |
berry | +0.127 |
berries | +0.121 |
superstar | +0.118 |
days | +0.117 |
Neg logit contributions | |
her | -0.077 |
site | -0.074 |
office | -0.070 |
ia | -0.069 |
men | -0.065 |
Token | big |
Token ID | 1263 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
site | +0.075 |
office | +0.071 |
her | +0.070 |
ia | +0.068 |
mistake | +0.068 |
Neg logit contributions | |
mine | -0.078 |
berry | -0.066 |
Benny | -0.066 |
berries | -0.065 |
Will | -0.064 |
Token | smile |
Token ID | 8212 |
Feature activation | +6.77e+00 |
Pos logit contributions | |
site | +0.192 |
office | +0.185 |
ia | +0.185 |
mistake | +0.181 |
her | +0.174 |
Neg logit contributions | |
mine | -0.084 |
mattered | -0.068 |
wherever | -0.065 |
Grac | -0.064 |
Benny | -0.064 |
Token | , |
Token ID | 11 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
mine | +0.481 |
Benny | +0.461 |
ease | +0.460 |
're | +0.454 |
berry | +0.448 |
Neg logit contributions | |
site | -0.490 |
office | -0.460 |
her | -0.458 |
ia | -0.445 |
men | -0.404 |
Token | and |
Token ID | 290 |
Feature activation | +6.80e-01 |
Pos logit contributions | |
site | +0.028 |
her | +0.027 |
office | +0.026 |
ia | +0.025 |
urgent | +0.023 |
Neg logit contributions | |
mine | -0.032 |
berry | -0.030 |
're | -0.028 |
berries | -0.028 |
mattered | -0.028 |
Token | waved |
Token ID | 26834 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
mine | +0.049 |
berry | +0.046 |
're | +0.043 |
berries | +0.043 |
ease | +0.041 |
Neg logit contributions | |
site | -0.051 |
her | -0.048 |
office | -0.047 |
ia | -0.046 |
men | -0.043 |
Token | him |
Token ID | 683 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
mine | +0.223 |
're | +0.218 |
berry | +0.215 |
berries | +0.207 |
| | +0.204 |
Neg logit contributions | |
site | -0.230 |
ia | -0.230 |
her | -0.223 |
office | -0.215 |
men | -0.213 |
Token | goodbye |
Token ID | 24829 |
Feature activation | +6.43e+00 |
Pos logit contributions | |
mine | +0.026 |
berry | +0.025 |
're | +0.023 |
Benny | +0.023 |
berries | +0.023 |
Neg logit contributions | |
site | -0.029 |
her | -0.027 |
office | -0.026 |
ia | -0.026 |
men | -0.023 |
Token | ?" |
Token ID | 1701 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
Benny | +0.072 |
berry | +0.070 |
're | +0.069 |
wherever | +0.068 |
g | +0.068 |
Neg logit contributions | |
site | -0.099 |
her | -0.089 |
office | -0.086 |
ia | -0.084 |
wounds | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
site | +0.026 |
her | +0.024 |
office | +0.024 |
ia | +0.024 |
men | +0.021 |
Neg logit contributions | |
mine | -0.020 |
berry | -0.019 |
're | -0.018 |
wherever | -0.018 |
Benny | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
mine | +0.012 |
Benny | +0.011 |
berry | +0.011 |
wherever | +0.011 |
ease | +0.010 |
Neg logit contributions | |
site | -0.025 |
office | -0.023 |
her | -0.023 |
wasteful | -0.022 |
ia | -0.021 |
Token | " |
Token ID | 1 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
mine | +0.017 |
berry | +0.016 |
Benny | +0.016 |
wherever | +0.016 |
g | +0.015 |
Neg logit contributions | |
site | -0.016 |
office | -0.014 |
her | -0.014 |
ia | -0.013 |
wasteful | -0.013 |
Token | Yes |
Token ID | 5297 |
Feature activation | +6.45e+00 |
Pos logit contributions | |
her | +0.244 |
site | +0.238 |
ia | +0.226 |
office | +0.216 |
wasteful | +0.210 |
Neg logit contributions | |
mine | -0.127 |
ease | -0.117 |
Benny | -0.116 |
're | -0.114 |
berry | -0.113 |
Token | , |
Token ID | 11 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
mine | +0.398 |
berry | +0.362 |
're | +0.356 |
Benny | +0.356 |
wherever | +0.345 |
Neg logit contributions | |
site | -0.585 |
her | -0.544 |
office | -0.530 |
ia | -0.499 |
wasteful | -0.481 |
Token | please |
Token ID | 3387 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
site | +0.026 |
her | +0.024 |
office | +0.024 |
ia | +0.024 |
men | +0.021 |
Neg logit contributions | |
mine | -0.025 |
berry | -0.023 |
ease | -0.022 |
're | -0.022 |
wherever | -0.022 |
Token | ," |
Token ID | 553 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
site | +0.150 |
her | +0.137 |
office | +0.134 |
ia | +0.127 |
wasteful | +0.123 |
Neg logit contributions | |
mine | -0.104 |
're | -0.100 |
berry | -0.098 |
Benny | -0.098 |
wherever | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | -2.40e-01 |
Pos logit contributions | |
site | +0.169 |
ia | +0.152 |
office | +0.150 |
her | +0.147 |
men | +0.134 |
Neg logit contributions | |
mine | -0.283 |
're | -0.271 |
berry | -0.268 |
wherever | -0.266 |
mattered | -0.265 |
Token | lady |
Token ID | 10846 |
Feature activation | -2.41e+00 |
Pos logit contributions | |
site | +0.017 |
her | +0.015 |
ia | +0.013 |
wasteful | +0.013 |
urgent | +0.013 |
Neg logit contributions | |
berry | -0.018 |
mine | -0.018 |
g | -0.017 |
bird | -0.016 |
bunny | -0.016 |
Token | said |
Token ID | 531 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
site | +0.157 |
her | +0.146 |
ia | +0.142 |
office | +0.139 |
men | +0.122 |
Neg logit contributions | |
mine | -0.196 |
wherever | -0.187 |
ease | -0.187 |
're | -0.186 |
berry | -0.182 |
Token | off |
Token ID | 572 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
site | +0.570 |
her | +0.523 |
ia | +0.511 |
office | +0.508 |
wasteful | +0.479 |
Neg logit contributions | |
mine | -0.363 |
wherever | -0.329 |
berry | -0.323 |
Benny | -0.322 |
're | -0.314 |
Token | on |
Token ID | 319 |
Feature activation | +7.29e-01 |
Pos logit contributions | |
site | +0.115 |
her | +0.105 |
office | +0.105 |
ia | +0.095 |
mistake | +0.090 |
Neg logit contributions | |
mine | -0.110 |
're | -0.104 |
berry | -0.103 |
mattered | -0.097 |
Benny | -0.097 |
Token | her |
Token ID | 607 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
mine | +0.048 |
're | +0.045 |
berry | +0.044 |
Benny | +0.041 |
mattered | +0.041 |
Neg logit contributions | |
site | -0.062 |
office | -0.057 |
her | -0.057 |
ia | -0.054 |
mistake | -0.051 |
Token | next |
Token ID | 1306 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
mine | +0.144 |
berry | +0.133 |
wherever | +0.133 |
Benny | +0.132 |
're | +0.131 |
Neg logit contributions | |
site | -0.160 |
her | -0.148 |
ia | -0.142 |
office | -0.142 |
wasteful | -0.131 |
Token | adventure |
Token ID | 8855 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
site | +0.230 |
her | +0.218 |
ia | +0.210 |
office | +0.202 |
wasteful | +0.201 |
Neg logit contributions | |
mine | -0.182 |
berry | -0.170 |
wherever | -0.168 |
Benny | -0.166 |
g | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
site | +0.081 |
her | +0.074 |
office | +0.072 |
ia | +0.071 |
wasteful | +0.065 |
Neg logit contributions | |
mine | -0.069 |
berry | -0.063 |
Benny | -0.062 |
wherever | -0.062 |
're | -0.062 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
site | +0.006 |
her | +0.006 |
office | +0.006 |
ia | +0.005 |
urse | +0.005 |
Neg logit contributions | |
mine | -0.006 |
're | -0.005 |
berry | -0.005 |
mattered | -0.005 |
g | -0.005 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
site | +0.101 |
her | +0.088 |
office | +0.086 |
ia | +0.086 |
men | +0.071 |
Neg logit contributions | |
mine | -0.154 |
're | -0.143 |
g | -0.142 |
ease | -0.140 |
mattered | -0.139 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
mine | +0.291 |
berry | +0.263 |
're | +0.257 |
Benny | +0.256 |
wherever | +0.254 |
Neg logit contributions | |
site | -0.444 |
her | -0.410 |
office | -0.403 |
ia | -0.397 |
men | -0.365 |
Token | a |
Token ID | 257 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
mine | +0.094 |
berry | +0.088 |
're | +0.085 |
Benny | +0.084 |
wherever | +0.084 |
Neg logit contributions | |
site | -0.071 |
her | -0.066 |
office | -0.063 |
ia | -0.062 |
x | -0.054 |
Token | time |
Token ID | 640 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
site | +0.072 |
her | +0.063 |
office | +0.063 |
ia | +0.057 |
wasteful | +0.055 |
Neg logit contributions | |
mine | -0.087 |
berry | -0.082 |
're | -0.081 |
Benny | -0.081 |
wherever | -0.080 |
Token | her |
Token ID | 607 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
site | +0.039 |
her | +0.039 |
office | +0.037 |
ia | +0.036 |
wasteful | +0.033 |
Neg logit contributions | |
mine | -0.040 |
berry | -0.037 |
're | -0.036 |
mattered | -0.035 |
ease | -0.035 |
Token | legs |
Token ID | 7405 |
Feature activation | -5.88e-03 |
Pos logit contributions | |
mine | +0.157 |
berry | +0.141 |
Benny | +0.140 |
're | +0.140 |
ease | +0.137 |
Neg logit contributions | |
site | -0.157 |
her | -0.144 |
office | -0.143 |
ia | -0.140 |
mistake | -0.127 |
Token | and |
Token ID | 290 |
Feature activation | +5.11e-01 |
Pos logit contributions | |
site | +0.000 |
her | +0.000 |
office | +0.000 |
ia | +0.000 |
men | +0.000 |
Neg logit contributions | |
mine | -0.000 |
berry | -0.000 |
Benny | -0.000 |
superstar | -0.000 |
mattered | -0.000 |
Token | her |
Token ID | 607 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
mine | +0.036 |
berry | +0.033 |
're | +0.031 |
oft | +0.031 |
Benny | +0.031 |
Neg logit contributions | |
site | -0.039 |
her | -0.037 |
office | -0.036 |
ia | -0.036 |
men | -0.034 |
Token | body |
Token ID | 1767 |
Feature activation | -6.25e-02 |
Pos logit contributions | |
mine | +0.167 |
berry | +0.151 |
Benny | +0.150 |
're | +0.148 |
ease | +0.145 |
Neg logit contributions | |
site | -0.146 |
her | -0.134 |
office | -0.134 |
ia | -0.133 |
mistake | -0.120 |
Token | with |
Token ID | 351 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
site | +0.005 |
her | +0.005 |
office | +0.004 |
ia | +0.004 |
men | +0.004 |
Neg logit contributions | |
mine | -0.004 |
berry | -0.004 |
Benny | -0.004 |
're | -0.004 |
ease | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
mine | +0.110 |
berry | +0.099 |
superstar | +0.098 |
| | +0.093 |
oft | +0.092 |
Neg logit contributions | |
her | -0.100 |
site | -0.095 |
ia | -0.093 |
office | -0.089 |
men | -0.088 |
Token | water |
Token ID | 1660 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
site | +0.034 |
her | +0.032 |
office | +0.031 |
ia | +0.031 |
men | +0.028 |
Neg logit contributions | |
mine | -0.037 |
Benny | -0.033 |
berry | -0.033 |
wherever | -0.033 |
're | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +8.45e-03 |
Pos logit contributions | |
mine | +0.027 |
Benny | +0.024 |
berry | +0.024 |
're | +0.024 |
ease | +0.024 |
Neg logit contributions | |
site | -0.025 |
her | -0.024 |
office | -0.023 |
ia | -0.022 |
men | -0.020 |
Token | It |
Token ID | 632 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
mine | +0.001 |
're | +0.001 |
berry | +0.001 |
mattered | +0.001 |
g | +0.001 |
Neg logit contributions | |
site | -0.001 |
her | -0.001 |
ia | -0.001 |
office | -0.001 |
urse | -0.001 |
Token | felt |
Token ID | 2936 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
mine | +0.056 |
wherever | +0.052 |
berry | +0.052 |
Benny | +0.050 |
ease | +0.050 |
Neg logit contributions | |
site | -0.056 |
her | -0.054 |
ia | -0.052 |
office | -0.048 |
mistake | -0.045 |
Token | new |
Token ID | 649 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
office | +0.053 |
site | +0.052 |
her | +0.048 |
ia | +0.047 |
mistake | +0.046 |
Neg logit contributions | |
mine | -0.088 |
berry | -0.075 |
berries | -0.075 |
're | -0.074 |
ease | -0.074 |
Token | friend |
Token ID | 1545 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
office | +0.024 |
site | +0.024 |
ia | +0.021 |
mistake | +0.021 |
her | +0.021 |
Neg logit contributions | |
mine | -0.028 |
're | -0.024 |
ease | -0.024 |
Benny | -0.024 |
hing | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
mine | +0.236 |
ease | +0.232 |
're | +0.219 |
Grac | +0.216 |
oft | +0.215 |
Neg logit contributions | |
office | -0.153 |
site | -0.139 |
her | -0.135 |
ia | -0.124 |
ena | -0.121 |
Token | draw |
Token ID | 3197 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
mine | +0.008 |
berry | +0.008 |
berries | +0.007 |
mattered | +0.007 |
oft | +0.007 |
Neg logit contributions | |
site | -0.012 |
her | -0.012 |
office | -0.012 |
ia | -0.012 |
x | -0.011 |
Token | with |
Token ID | 351 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
mine | +0.281 |
're | +0.274 |
berry | +0.261 |
oft | +0.253 |
ease | +0.253 |
Neg logit contributions | |
site | -0.272 |
her | -0.257 |
office | -0.248 |
ia | -0.236 |
men | -0.216 |
Token | . |
Token ID | 13 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
mine | +0.115 |
're | +0.113 |
oft | +0.110 |
berry | +0.109 |
superstar | +0.104 |
Neg logit contributions | |
her | -0.097 |
site | -0.094 |
office | -0.084 |
ia | -0.083 |
men | -0.076 |
Token | And |
Token ID | 843 |
Feature activation | -2.81e-01 |
Pos logit contributions | |
mine | +0.001 |
berry | +0.001 |
're | +0.001 |
ease | +0.001 |
wherever | +0.001 |
Neg logit contributions | |
site | -0.002 |
her | -0.001 |
office | -0.001 |
ia | -0.001 |
men | -0.001 |
Token | they |
Token ID | 484 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
site | +0.020 |
office | +0.018 |
her | +0.018 |
ia | +0.018 |
men | +0.016 |
Neg logit contributions | |
mine | -0.022 |
berry | -0.020 |
ease | -0.019 |
're | -0.019 |
berries | -0.019 |
Token | spent |
Token ID | 3377 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
her | +0.008 |
ia | +0.008 |
ep | +0.007 |
site | +0.007 |
ena | +0.007 |
Neg logit contributions | |
Nice | -0.009 |
berry | -0.008 |
berries | -0.008 |
hello | -0.007 |
Benny | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
mine | +0.082 |
berry | +0.075 |
're | +0.074 |
Benny | +0.072 |
wherever | +0.071 |
Neg logit contributions | |
site | -0.092 |
her | -0.085 |
office | -0.083 |
ia | -0.081 |
mistake | -0.073 |
Token | rest |
Token ID | 1334 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
site | +0.020 |
her | +0.019 |
office | +0.019 |
ia | +0.018 |
ep | +0.016 |
Neg logit contributions | |
mine | -0.017 |
berry | -0.017 |
Grac | -0.016 |
Benny | -0.016 |
berries | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -9.05e-01 |
Pos logit contributions | |
site | +0.024 |
her | +0.022 |
office | +0.022 |
ia | +0.021 |
men | +0.019 |
Neg logit contributions | |
mine | -0.027 |
berry | -0.025 |
're | -0.024 |
ease | -0.024 |
wherever | -0.024 |
Token | strong |
Token ID | 1913 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
site | +0.065 |
her | +0.060 |
office | +0.057 |
ia | +0.057 |
wasteful | +0.051 |
Neg logit contributions | |
mine | -0.069 |
berry | -0.065 |
Benny | -0.065 |
're | -0.064 |
wherever | -0.063 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
mine | +0.010 |
berry | +0.009 |
Benny | +0.009 |
g | +0.009 |
're | +0.008 |
Neg logit contributions | |
site | -0.011 |
her | -0.011 |
office | -0.010 |
ia | -0.010 |
yourselves | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
mine | +0.316 |
ease | +0.305 |
're | +0.291 |
Grac | +0.281 |
wherever | +0.280 |
Neg logit contributions | |
site | -0.264 |
office | -0.250 |
her | -0.237 |
ia | -0.236 |
men | -0.212 |
Token | to |
Token ID | 284 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
site | +0.029 |
her | +0.026 |
office | +0.026 |
ia | +0.026 |
wasteful | +0.023 |
Neg logit contributions | |
mine | -0.022 |
Benny | -0.020 |
berry | -0.020 |
're | -0.020 |
wherever | -0.020 |
Token | help |
Token ID | 1037 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
mine | +0.009 |
berry | +0.008 |
're | +0.008 |
wherever | +0.007 |
berries | +0.007 |
Neg logit contributions | |
site | -0.017 |
her | -0.016 |
office | -0.016 |
ia | -0.015 |
x | -0.015 |
Token | him |
Token ID | 683 |
Feature activation | +4.21e-01 |
Pos logit contributions | |
mine | +0.140 |
Benny | +0.120 |
berry | +0.118 |
wherever | +0.116 |
're | +0.116 |
Neg logit contributions | |
site | -0.343 |
her | -0.320 |
office | -0.317 |
ia | -0.315 |
wasteful | -0.294 |
Token | . |
Token ID | 13 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
mine | +0.029 |
're | +0.026 |
wherever | +0.026 |
berry | +0.026 |
Benny | +0.026 |
Neg logit contributions | |
site | -0.034 |
her | -0.031 |
ia | -0.030 |
office | -0.030 |
wasteful | -0.028 |
Token | The |
Token ID | 383 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
mine | +0.006 |
Benny | +0.006 |
berry | +0.006 |
're | +0.006 |
wherever | +0.006 |
Neg logit contributions | |
site | -0.011 |
her | -0.010 |
office | -0.010 |
ia | -0.010 |
mistake | -0.009 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
mine | +0.044 |
berry | +0.038 |
're | +0.038 |
wherever | +0.037 |
Benny | +0.036 |
Neg logit contributions | |
site | -0.080 |
office | -0.074 |
her | -0.074 |
ia | -0.072 |
men | -0.068 |
Token | held |
Token ID | 2714 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
mine | +0.281 |
berry | +0.257 |
're | +0.253 |
wherever | +0.252 |
Benny | +0.251 |
Neg logit contributions | |
site | -0.311 |
her | -0.285 |
office | -0.278 |
ia | -0.274 |
men | -0.248 |
Token | OK |
Token ID | 11380 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
site | +0.234 |
office | +0.216 |
her | +0.214 |
ia | +0.208 |
mistake | +0.193 |
Neg logit contributions | |
mine | -0.151 |
berry | -0.136 |
're | -0.134 |
wherever | -0.132 |
Benny | -0.131 |
Token | , |
Token ID | 11 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
site | +0.098 |
her | +0.093 |
office | +0.089 |
ia | +0.086 |
wasteful | +0.084 |
Neg logit contributions | |
mine | -0.059 |
Benny | -0.059 |
wherever | -0.057 |
're | -0.056 |
berry | -0.055 |
Token | but |
Token ID | 475 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
site | +0.035 |
her | +0.032 |
office | +0.032 |
ia | +0.031 |
wasteful | +0.029 |
Neg logit contributions | |
mine | -0.022 |
berry | -0.020 |
're | -0.019 |
Benny | -0.019 |
wherever | -0.019 |
Token | be |
Token ID | 307 |
Feature activation | -5.28e+00 |
Pos logit contributions | |
mine | +0.091 |
're | +0.085 |
Benny | +0.085 |
wherever | +0.084 |
berry | +0.084 |
Neg logit contributions | |
site | -0.104 |
her | -0.096 |
office | -0.093 |
ia | -0.091 |
wasteful | -0.083 |
Token | careful |
Token ID | 8161 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
site | +0.438 |
ia | +0.390 |
her | +0.355 |
wen | +0.338 |
office | +0.334 |
Neg logit contributions | |
g | -0.326 |
wherever | -0.323 |
Benny | -0.298 |
join | -0.290 |
compliment | -0.277 |
Token | ," |
Token ID | 553 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
site | +0.258 |
her | +0.236 |
office | +0.233 |
wasteful | +0.230 |
ia | +0.225 |
Neg logit contributions | |
mine | -0.171 |
wherever | -0.154 |
're | -0.150 |
g | -0.149 |
Benny | -0.145 |
Token | Sue |
Token ID | 26089 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
site | +0.247 |
her | +0.230 |
office | +0.221 |
ia | +0.203 |
wasteful | +0.196 |
Neg logit contributions | |
mine | -0.225 |
Benny | -0.214 |
berry | -0.210 |
g | -0.208 |
're | -0.206 |
Token | says |
Token ID | 1139 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
site | +0.124 |
her | +0.106 |
office | +0.103 |
ia | +0.095 |
wasteful | +0.088 |
Neg logit contributions | |
mine | -0.215 |
berry | -0.202 |
're | -0.201 |
Benny | -0.200 |
g | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | +1.28e-02 |
Pos logit contributions | |
site | +0.406 |
her | +0.373 |
office | +0.364 |
ia | +0.357 |
men | +0.324 |
Neg logit contributions | |
mine | -0.383 |
berry | -0.356 |
're | -0.348 |
Benny | -0.347 |
wherever | -0.343 |
Token |
|
Token ID | 198 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
mine | +0.001 |
berry | +0.001 |
Benny | +0.001 |
're | +0.001 |
wherever | +0.001 |
Neg logit contributions | |
site | -0.001 |
her | -0.001 |
office | -0.001 |
ia | -0.001 |
wasteful | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
mine | +0.010 |
Benny | +0.010 |
wherever | +0.009 |
berry | +0.009 |
're | +0.008 |
Neg logit contributions | |
site | -0.033 |
office | -0.032 |
wasteful | -0.032 |
her | -0.030 |
disrespectful | -0.030 |
Token | out |
Token ID | 503 |
Feature activation | -9.54e-01 |
Pos logit contributions | |
site | +0.147 |
her | +0.137 |
office | +0.132 |
ia | +0.128 |
men | +0.120 |
Neg logit contributions | |
mine | -0.113 |
berry | -0.103 |
're | -0.101 |
Benny | -0.099 |
wherever | -0.098 |
Token | and |
Token ID | 290 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
site | +0.074 |
her | +0.067 |
office | +0.065 |
ia | +0.065 |
men | +0.059 |
Neg logit contributions | |
mine | -0.069 |
berry | -0.063 |
're | -0.063 |
wherever | -0.061 |
ease | -0.061 |
Token | rolled |
Token ID | 11686 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
mine | +0.046 |
berry | +0.043 |
're | +0.041 |
wherever | +0.040 |
berries | +0.039 |
Neg logit contributions | |
site | -0.060 |
her | -0.056 |
office | -0.054 |
ia | -0.052 |
wasteful | -0.050 |
Token | across |
Token ID | 1973 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
mine | +0.183 |
're | +0.169 |
berry | +0.169 |
Benny | +0.159 |
wherever | +0.159 |
Neg logit contributions | |
site | -0.226 |
her | -0.207 |
office | -0.197 |
ia | -0.194 |
men | -0.185 |
Token | the |
Token ID | 262 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
site | +0.118 |
her | +0.106 |
office | +0.102 |
ia | +0.098 |
mistake | +0.090 |
Neg logit contributions | |
mine | -0.115 |
're | -0.108 |
berry | -0.104 |
wherever | -0.101 |
mattered | -0.101 |
Token | floor |
Token ID | 4314 |
Feature activation | -6.21e+00 |
Pos logit contributions | |
site | +0.016 |
ia | +0.014 |
her | +0.014 |
office | +0.013 |
wasteful | +0.012 |
Neg logit contributions | |
mine | -0.024 |
berry | -0.023 |
wherever | -0.022 |
Benny | -0.022 |
g | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
site | +0.499 |
her | +0.456 |
office | +0.446 |
ia | +0.437 |
men | +0.398 |
Neg logit contributions | |
mine | -0.431 |
berry | -0.396 |
're | -0.389 |
Benny | -0.388 |
wherever | -0.385 |
Token | |
Token ID | 220 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
mine | +0.009 |
berry | +0.008 |
wherever | +0.008 |
superstar | +0.008 |
're | +0.008 |
Neg logit contributions | |
site | -0.015 |
her | -0.014 |
office | -0.013 |
ia | -0.013 |
men | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
mine | +0.247 |
're | +0.227 |
Benny | +0.227 |
wherever | +0.221 |
berry | +0.220 |
Neg logit contributions | |
site | -0.439 |
office | -0.402 |
her | -0.396 |
ia | -0.391 |
wasteful | -0.370 |
Token |
|
Token ID | 198 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
mine | +0.021 |
berry | +0.019 |
're | +0.018 |
Benny | +0.018 |
wherever | +0.018 |
Neg logit contributions | |
site | -0.028 |
her | -0.026 |
office | -0.025 |
ia | -0.025 |
wasteful | -0.023 |
Token | Tim |
Token ID | 14967 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
mine | +0.022 |
berry | +0.021 |
Benny | +0.020 |
're | +0.020 |
wherever | +0.020 |
Neg logit contributions | |
site | -0.025 |
her | -0.022 |
office | -0.022 |
ia | -0.022 |
men | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
site | +0.187 |
her | +0.165 |
office | +0.165 |
ia | +0.155 |
men | +0.145 |
Neg logit contributions | |
mine | -0.189 |
're | -0.182 |
berry | -0.174 |
oft | -0.173 |
mattered | -0.172 |
Token | shady |
Token ID | 36135 |
Feature activation | -2.54e+00 |
Pos logit contributions | |
site | +0.021 |
her | +0.019 |
office | +0.019 |
ia | +0.018 |
wasteful | +0.017 |
Neg logit contributions | |
mine | -0.020 |
berry | -0.018 |
're | -0.018 |
Benny | -0.018 |
wherever | -0.018 |
Token | tree |
Token ID | 5509 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
site | +0.206 |
her | +0.197 |
office | +0.185 |
ia | +0.183 |
wasteful | +0.183 |
Neg logit contributions | |
mine | -0.166 |
berry | -0.158 |
Benny | -0.152 |
wherever | -0.150 |
g | -0.144 |
Token | . |
Token ID | 13 |
Feature activation | +6.85e-04 |
Pos logit contributions | |
mine | +0.072 |
berry | +0.066 |
're | +0.065 |
Benny | +0.065 |
wherever | +0.064 |
Neg logit contributions | |
site | -0.081 |
her | -0.074 |
office | -0.073 |
ia | -0.071 |
men | -0.064 |
Token | Dad |
Token ID | 17415 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
mine | +0.000 |
berry | +0.000 |
Benny | +0.000 |
're | +0.000 |
wherever | +0.000 |
Neg logit contributions | |
site | -0.000 |
her | -0.000 |
office | -0.000 |
ia | -0.000 |
wasteful | -0.000 |
Token | said |
Token ID | 531 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
site | +0.135 |
her | +0.127 |
office | +0.123 |
ia | +0.122 |
men | +0.108 |
Neg logit contributions | |
mine | -0.140 |
wherever | -0.132 |
berry | -0.130 |
ease | -0.130 |
're | -0.129 |
Token | yes |
Token ID | 3763 |
Feature activation | +6.86e+00 |
Pos logit contributions | |
site | +0.229 |
her | +0.215 |
office | +0.211 |
ia | +0.210 |
men | +0.199 |
Neg logit contributions | |
mine | -0.091 |
berry | -0.081 |
're | -0.077 |
ease | -0.076 |
berries | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
mine | +0.367 |
're | +0.329 |
Benny | +0.327 |
wherever | +0.320 |
g | +0.307 |
Neg logit contributions | |
site | -0.664 |
office | -0.604 |
her | -0.602 |
ia | -0.586 |
wasteful | -0.586 |
Token | the |
Token ID | 262 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
mine | +0.044 |
berry | +0.040 |
're | +0.039 |
wherever | +0.038 |
ease | +0.038 |
Neg logit contributions | |
site | -0.052 |
her | -0.048 |
office | -0.047 |
ia | -0.046 |
men | -0.043 |
Token | boy |
Token ID | 2933 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
site | +0.014 |
office | +0.012 |
her | +0.012 |
ia | +0.012 |
men | +0.011 |
Neg logit contributions | |
mine | -0.019 |
berry | -0.017 |
're | -0.017 |
wherever | -0.017 |
ease | -0.017 |
Token | pl |
Token ID | 458 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
site | +0.288 |
her | +0.274 |
ia | +0.272 |
office | +0.269 |
men | +0.237 |
Neg logit contributions | |
mine | -0.239 |
berry | -0.218 |
wherever | -0.216 |
Benny | -0.216 |
're | -0.213 |
Token | the |
Token ID | 262 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
mine | +0.160 |
're | +0.144 |
berry | +0.143 |
Benny | +0.142 |
Emmy | +0.138 |
Neg logit contributions | |
site | -0.147 |
office | -0.135 |
ia | -0.135 |
her | -0.132 |
mistake | -0.119 |
Token | card |
Token ID | 2657 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
site | +0.037 |
her | +0.034 |
office | +0.033 |
ia | +0.032 |
men | +0.030 |
Neg logit contributions | |
mine | -0.041 |
berry | -0.037 |
're | -0.036 |
Benny | -0.036 |
wherever | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
site | +0.264 |
her | +0.242 |
office | +0.237 |
ia | +0.234 |
men | +0.211 |
Neg logit contributions | |
mine | -0.241 |
berry | -0.221 |
're | -0.218 |
Benny | -0.218 |
wherever | -0.214 |
Token | " |
Token ID | 366 |
Feature activation | -4.93e-01 |
Pos logit contributions | |
site | +0.006 |
her | +0.005 |
office | +0.005 |
ia | +0.005 |
wasteful | +0.005 |
Neg logit contributions | |
mine | -0.004 |
Benny | -0.004 |
berry | -0.004 |
wherever | -0.003 |
're | -0.003 |
Token | Ta |
Token ID | 38586 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
her | +0.036 |
site | +0.036 |
office | +0.033 |
ia | +0.033 |
yourselves | +0.031 |
Neg logit contributions | |
mine | -0.033 |
Benny | -0.033 |
're | -0.033 |
wherever | -0.032 |
berry | -0.032 |
Token | - |
Token ID | 12 |
Feature activation | -5.38e+00 |
Pos logit contributions | |
mine | +0.180 |
berry | +0.169 |
Benny | +0.166 |
wherever | +0.164 |
're | +0.163 |
Neg logit contributions | |
site | -0.127 |
her | -0.116 |
office | -0.111 |
ia | -0.100 |
wasteful | -0.100 |
Token | da |
Token ID | 6814 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
site | +0.440 |
her | +0.417 |
ia | +0.400 |
office | +0.395 |
men | +0.349 |
Neg logit contributions | |
mine | -0.346 |
're | -0.330 |
berry | -0.328 |
Benny | -0.325 |
ease | -0.324 |
Token | ! |
Token ID | 0 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
site | +0.075 |
office | +0.068 |
her | +0.068 |
wasteful | +0.064 |
ia | +0.063 |
Neg logit contributions | |
mine | -0.050 |
berry | -0.044 |
wherever | -0.043 |
Benny | -0.042 |
're | -0.042 |
Token | We |
Token ID | 775 |
Feature activation | -5.85e-01 |
Pos logit contributions | |
mine | +0.075 |
ease | +0.068 |
wherever | +0.067 |
berry | +0.067 |
Benny | +0.065 |
Neg logit contributions | |
site | -0.097 |
ia | -0.091 |
her | -0.091 |
office | -0.091 |
yourselves | -0.081 |
Token | made |
Token ID | 925 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
her | +0.043 |
site | +0.042 |
ia | +0.040 |
ep | +0.040 |
office | +0.039 |
Neg logit contributions | |
Nice | -0.037 |
berries | -0.035 |
Benny | -0.034 |
Emmy | -0.034 |
berry | -0.034 |
Token | this |
Token ID | 428 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
mine | +0.038 |
Benny | +0.036 |
berry | +0.036 |
wherever | +0.035 |
g | +0.035 |
Neg logit contributions | |
site | -0.051 |
her | -0.047 |
office | -0.046 |
ia | -0.044 |
wasteful | -0.043 |
Token | in |
Token ID | 287 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
mine | +0.089 |
berry | +0.081 |
're | +0.077 |
ease | +0.075 |
wherever | +0.074 |
Neg logit contributions | |
site | -0.103 |
her | -0.094 |
office | -0.093 |
ia | -0.091 |
men | -0.085 |
Token | his |
Token ID | 465 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
mine | +0.002 |
berry | +0.001 |
're | +0.001 |
Benny | +0.001 |
wherever | +0.001 |
Neg logit contributions | |
site | -0.002 |
her | -0.002 |
office | -0.002 |
ia | -0.001 |
men | -0.001 |
Token | heart |
Token ID | 2612 |
Feature activation | +4.02e+00 |
Pos logit contributions | |
mine | +0.143 |
berry | +0.131 |
're | +0.129 |
Benny | +0.129 |
wherever | +0.129 |
Neg logit contributions | |
site | -0.171 |
her | -0.158 |
office | -0.152 |
ia | -0.151 |
wasteful | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
mine | +0.275 |
berry | +0.252 |
Benny | +0.246 |
're | +0.244 |
wherever | +0.242 |
Neg logit contributions | |
site | -0.328 |
her | -0.302 |
office | -0.296 |
ia | -0.290 |
men | -0.265 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
site | +0.001 |
her | +0.001 |
office | +0.001 |
ia | +0.001 |
wasteful | +0.001 |
Neg logit contributions | |
mine | -0.001 |
berry | -0.001 |
're | -0.001 |
Benny | -0.001 |
wherever | -0.001 |
Token | L |
Token ID | 43 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
site | +0.084 |
her | +0.074 |
office | +0.070 |
ia | +0.069 |
men | +0.057 |
Neg logit contributions | |
mine | -0.151 |
're | -0.141 |
g | -0.140 |
ease | -0.139 |
Grac | -0.137 |
Token | ily |
Token ID | 813 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
site | +0.145 |
her | +0.130 |
wasteful | +0.125 |
office | +0.124 |
urgent | +0.122 |
Neg logit contributions | |
mine | -0.122 |
berry | -0.108 |
Benny | -0.104 |
're | -0.101 |
ease | -0.101 |
Token | and |
Token ID | 290 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
site | +0.274 |
her | +0.260 |
office | +0.257 |
ia | +0.248 |
men | +0.222 |
Neg logit contributions | |
mine | -0.243 |
ease | -0.220 |
berry | -0.220 |
wherever | -0.220 |
're | -0.219 |
Token | Ben |
Token ID | 3932 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
mine | +0.059 |
berry | +0.058 |
Grac | +0.055 |
superstar | +0.055 |
berries | +0.055 |
Neg logit contributions | |
site | -0.041 |
office | -0.039 |
her | -0.039 |
ia | -0.037 |
men | -0.035 |
Token | love |
Token ID | 1842 |
Feature activation | +9.00e+00 |
Pos logit contributions | |
site | +0.166 |
her | +0.165 |
office | +0.156 |
ia | +0.155 |
men | +0.151 |
Neg logit contributions | |
mine | -0.117 |
Grac | -0.113 |
berry | -0.112 |
superstar | -0.111 |
wherever | -0.110 |
Token | to |
Token ID | 284 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
mine | +0.664 |
're | +0.601 |
ease | +0.583 |
Grac | +0.573 |
berry | +0.567 |
Neg logit contributions | |
site | -0.711 |
office | -0.665 |
her | -0.636 |
ia | -0.582 |
men | -0.578 |
Token | heavy |
Token ID | 4334 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
site | +0.081 |
her | +0.075 |
office | +0.073 |
ia | +0.072 |
mistake | +0.066 |
Neg logit contributions | |
mine | -0.081 |
berry | -0.074 |
Benny | -0.073 |
're | -0.072 |
wherever | -0.072 |
Token | sigh |
Token ID | 19680 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
site | +0.303 |
her | +0.276 |
office | +0.275 |
ia | +0.268 |
men | +0.246 |
Neg logit contributions | |
mine | -0.237 |
berry | -0.216 |
're | -0.212 |
Benny | -0.212 |
wherever | -0.210 |
Token | . |
Token ID | 13 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
site | +0.035 |
her | +0.032 |
office | +0.032 |
ia | +0.031 |
wasteful | +0.028 |
Neg logit contributions | |
mine | -0.028 |
berry | -0.025 |
're | -0.025 |
Benny | -0.025 |
wherever | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
mine | +0.006 |
berry | +0.006 |
're | +0.006 |
Benny | +0.006 |
wherever | +0.006 |
Neg logit contributions | |
site | -0.009 |
her | -0.008 |
office | -0.008 |
ia | -0.008 |
men | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
mine | +0.011 |
Benny | +0.011 |
ease | +0.010 |
wherever | +0.010 |
Will | +0.009 |
Neg logit contributions | |
site | -0.025 |
office | -0.023 |
wasteful | -0.022 |
her | -0.022 |
men | -0.022 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -7.92e+00 |
Pos logit contributions | |
mine | +0.018 |
Benny | +0.016 |
berry | +0.016 |
wherever | +0.016 |
ease | +0.015 |
Neg logit contributions | |
site | -0.016 |
office | -0.014 |
men | -0.013 |
her | -0.013 |
ia | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
site | +0.645 |
her | +0.589 |
office | +0.581 |
ia | +0.572 |
men | +0.524 |
Neg logit contributions | |
mine | -0.542 |
berry | -0.496 |
're | -0.486 |
wherever | -0.479 |
Benny | -0.479 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
site | +0.029 |
office | +0.027 |
her | +0.027 |
ia | +0.026 |
men | +0.024 |
Neg logit contributions | |
mine | -0.025 |
berry | -0.023 |
're | -0.023 |
wherever | -0.022 |
ease | -0.022 |
Token | sky |
Token ID | 6766 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
site | +0.016 |
her | +0.014 |
ia | +0.013 |
office | +0.013 |
wasteful | +0.012 |
Neg logit contributions | |
mine | -0.019 |
berry | -0.018 |
're | -0.018 |
wherever | -0.018 |
g | -0.018 |
Token | above |
Token ID | 2029 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
mine | +0.083 |
berry | +0.076 |
're | +0.074 |
g | +0.074 |
Benny | +0.070 |
Neg logit contributions | |
site | -0.153 |
office | -0.147 |
yourselves | -0.140 |
her | -0.138 |
ia | -0.137 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
mine | +0.059 |
berry | +0.053 |
're | +0.051 |
Benny | +0.051 |
wherever | +0.051 |
Neg logit contributions | |
site | -0.104 |
her | -0.096 |
office | -0.094 |
ia | -0.093 |
wasteful | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | -9.41e-03 |
Pos logit contributions | |
site | +0.317 |
her | +0.275 |
office | +0.273 |
ia | +0.270 |
men | +0.242 |
Neg logit contributions | |
mine | -0.298 |
Benny | -0.287 |
're | -0.283 |
wherever | -0.282 |
mattered | -0.273 |
Token | It |
Token ID | 632 |
Feature activation | +8.64e-01 |
Pos logit contributions | |
site | +0.001 |
her | +0.001 |
office | +0.001 |
ia | +0.001 |
x | +0.001 |
Neg logit contributions | |
mine | -0.001 |
berry | -0.001 |
ease | -0.001 |
wherever | -0.001 |
're | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -7.76e-01 |
Pos logit contributions | |
mine | +0.070 |
berry | +0.065 |
ease | +0.064 |
wherever | +0.064 |
Grac | +0.061 |
Neg logit contributions | |
site | -0.059 |
her | -0.056 |
ia | -0.053 |
office | -0.052 |
mistake | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | -9.50e-01 |
Pos logit contributions | |
site | +0.063 |
office | +0.054 |
ia | +0.052 |
her | +0.052 |
men | +0.048 |
Neg logit contributions | |
Benny | -0.053 |
wherever | -0.050 |
mine | -0.049 |
g | -0.049 |
Will | -0.049 |
Token | big |
Token ID | 1263 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
site | +0.064 |
her | +0.058 |
office | +0.057 |
ia | +0.054 |
mistake | +0.050 |
Neg logit contributions | |
mine | -0.080 |
berry | -0.073 |
're | -0.072 |
wherever | -0.071 |
Benny | -0.071 |
Token | mistake |
Token ID | 7457 |
Feature activation | -7.33e+00 |
Pos logit contributions | |
site | +0.127 |
her | +0.113 |
office | +0.113 |
ia | +0.108 |
men | +0.098 |
Neg logit contributions | |
mine | -0.150 |
berry | -0.138 |
're | -0.137 |
wherever | -0.136 |
Benny | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | -5.65e-03 |
Pos logit contributions | |
site | +0.526 |
her | +0.484 |
office | +0.476 |
ia | +0.463 |
men | +0.422 |
Neg logit contributions | |
mine | -0.566 |
ease | -0.520 |
berry | -0.520 |
're | -0.512 |
Benny | -0.505 |
Token | I |
Token ID | 314 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
site | +0.000 |
her | +0.000 |
office | +0.000 |
ia | +0.000 |
wasteful | +0.000 |
Neg logit contributions | |
mine | -0.000 |
berry | -0.000 |
're | -0.000 |
wherever | -0.000 |
ease | -0.000 |
Token | 'm |
Token ID | 1101 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
mine | +0.143 |
berry | +0.133 |
berries | +0.130 |
wherever | +0.130 |
ease | +0.130 |
Neg logit contributions | |
her | -0.167 |
site | -0.165 |
ia | -0.156 |
office | -0.151 |
wasteful | -0.148 |
Token | sorry |
Token ID | 7926 |
Feature activation | -9.18e-01 |
Pos logit contributions | |
Benny | +0.249 |
mine | +0.246 |
Will | +0.236 |
g | +0.234 |
Grac | +0.234 |
Neg logit contributions | |
site | -0.244 |
ia | -0.207 |
her | -0.198 |
office | -0.198 |
ups | -0.178 |
Token | , |
Token ID | 11 |
Feature activation | -3.93e-01 |
Pos logit contributions | |
site | +0.077 |
her | +0.071 |
office | +0.069 |
ia | +0.068 |
wasteful | +0.062 |
Neg logit contributions | |
mine | -0.061 |
berry | -0.056 |
Benny | -0.055 |
're | -0.055 |
wherever | -0.055 |
Token | the |
Token ID | 262 |
Feature activation | -5.21e-01 |
Pos logit contributions | |
site | +0.013 |
her | +0.012 |
office | +0.012 |
ia | +0.011 |
mistake | +0.010 |
Neg logit contributions | |
mine | -0.012 |
berry | -0.011 |
're | -0.011 |
Benny | -0.011 |
wherever | -0.011 |
Token | sky |
Token ID | 6766 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
site | +0.036 |
her | +0.032 |
ia | +0.032 |
office | +0.030 |
wasteful | +0.027 |
Neg logit contributions | |
mine | -0.040 |
berry | -0.038 |
Benny | -0.038 |
're | -0.038 |
g | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
mine | +0.109 |
berry | +0.100 |
Benny | +0.099 |
're | +0.098 |
wherever | +0.097 |
Neg logit contributions | |
site | -0.118 |
her | -0.108 |
office | -0.105 |
ia | -0.103 |
men | -0.092 |
Token |
|
Token ID | 198 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
mine | +0.014 |
berry | +0.013 |
're | +0.013 |
wherever | +0.012 |
ease | +0.012 |
Neg logit contributions | |
site | -0.019 |
her | -0.018 |
office | -0.017 |
ia | -0.017 |
men | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
mine | +0.007 |
Benny | +0.006 |
berry | +0.006 |
wherever | +0.006 |
ease | +0.006 |
Neg logit contributions | |
site | -0.011 |
office | -0.010 |
her | -0.010 |
ia | -0.010 |
wasteful | -0.010 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -7.95e+00 |
Pos logit contributions | |
mine | +0.005 |
berry | +0.005 |
Benny | +0.005 |
wherever | +0.005 |
're | +0.005 |
Neg logit contributions | |
site | -0.005 |
office | -0.004 |
her | -0.004 |
ia | -0.004 |
wasteful | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
site | +0.648 |
her | +0.590 |
office | +0.581 |
ia | +0.571 |
men | +0.525 |
Neg logit contributions | |
mine | -0.544 |
berry | -0.497 |
're | -0.487 |
wherever | -0.481 |
Benny | -0.479 |
Token | the |
Token ID | 262 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
site | +0.026 |
office | +0.023 |
her | +0.023 |
ia | +0.022 |
men | +0.021 |
Neg logit contributions | |
mine | -0.023 |
berry | -0.022 |
're | -0.021 |
wherever | -0.021 |
ease | -0.020 |
Token | old |
Token ID | 1468 |
Feature activation | -7.10e+00 |
Pos logit contributions | |
site | +0.042 |
her | +0.039 |
office | +0.038 |
ia | +0.037 |
men | +0.034 |
Neg logit contributions | |
mine | -0.037 |
berry | -0.034 |
're | -0.033 |
Benny | -0.033 |
wherever | -0.033 |
Token | man |
Token ID | 582 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
site | +0.422 |
her | +0.371 |
office | +0.366 |
ia | +0.349 |
men | +0.309 |
Neg logit contributions | |
mine | -0.645 |
berry | -0.601 |
're | -0.600 |
wherever | -0.592 |
Benny | -0.590 |
Token | and |
Token ID | 290 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
site | +0.351 |
her | +0.323 |
office | +0.318 |
ia | +0.312 |
wasteful | +0.290 |
Neg logit contributions | |
mine | -0.247 |
berry | -0.226 |
're | -0.221 |
Benny | -0.220 |
wherever | -0.216 |
Token | because |
Token ID | 780 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
mine | +0.117 |
're | +0.107 |
berry | +0.106 |
Benny | +0.105 |
wherever | +0.104 |
Neg logit contributions | |
site | -0.188 |
her | -0.170 |
ia | -0.163 |
office | -0.163 |
men | -0.151 |
Token | they |
Token ID | 484 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
mine | +0.152 |
berry | +0.138 |
're | +0.135 |
Benny | +0.132 |
wherever | +0.131 |
Neg logit contributions | |
site | -0.216 |
her | -0.200 |
office | -0.198 |
ia | -0.192 |
men | -0.179 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
her | +0.004 |
site | +0.004 |
office | +0.004 |
wasteful | +0.004 |
men | +0.004 |
Neg logit contributions | |
berry | -0.003 |
mine | -0.002 |
Nice | -0.002 |
berries | -0.002 |
superstar | -0.002 |
Token | 't |
Token ID | 470 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
mine | +0.059 |
're | +0.048 |
Benny | +0.048 |
berry | +0.047 |
wherever | +0.046 |
Neg logit contributions | |
site | -0.126 |
office | -0.117 |
ia | -0.116 |
her | -0.112 |
wasteful | -0.110 |
Token | have |
Token ID | 423 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
mine | +0.133 |
berry | +0.119 |
Benny | +0.115 |
're | +0.115 |
wherever | +0.114 |
Neg logit contributions | |
site | -0.258 |
her | -0.241 |
office | -0.237 |
ia | -0.233 |
wasteful | -0.217 |
Token | enough |
Token ID | 1576 |
Feature activation | -7.44e+00 |
Pos logit contributions | |
mine | +0.164 |
wherever | +0.155 |
're | +0.153 |
ease | +0.152 |
Benny | +0.151 |
Neg logit contributions | |
site | -0.163 |
her | -0.141 |
ia | -0.141 |
office | -0.138 |
urgent | -0.124 |
Token | money |
Token ID | 1637 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
site | +0.440 |
her | +0.391 |
ia | +0.373 |
office | +0.372 |
wasteful | +0.320 |
Neg logit contributions | |
mine | -0.668 |
berry | -0.631 |
're | -0.629 |
wherever | -0.626 |
Benny | -0.624 |
Token | . |
Token ID | 13 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
site | +0.378 |
her | +0.341 |
office | +0.338 |
ia | +0.337 |
wasteful | +0.304 |
Neg logit contributions | |
mine | -0.284 |
wherever | -0.257 |
're | -0.257 |
berry | -0.254 |
Benny | -0.253 |
Token |
|
Token ID | 198 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
mine | +0.012 |
berry | +0.011 |
're | +0.011 |
Benny | +0.011 |
wherever | +0.010 |
Neg logit contributions | |
site | -0.020 |
her | -0.019 |
office | -0.019 |
ia | -0.018 |
wasteful | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
mine | +0.015 |
Benny | +0.014 |
berry | +0.014 |
wherever | +0.014 |
're | +0.013 |
Neg logit contributions | |
site | -0.028 |
office | -0.025 |
her | -0.025 |
ia | -0.024 |
wasteful | -0.024 |
Token | L |
Token ID | 43 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
mine | +0.017 |
Benny | +0.016 |
Emmy | +0.015 |
wherever | +0.015 |
mattered | +0.015 |
Neg logit contributions | |
site | -0.026 |
office | -0.024 |
men | -0.022 |
ia | -0.021 |
urgent | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
site | +0.115 |
her | +0.106 |
office | +0.103 |
ia | +0.102 |
men | +0.094 |
Neg logit contributions | |
mine | -0.106 |
berry | -0.098 |
're | -0.096 |
Benny | -0.095 |
wherever | -0.093 |
Token | music |
Token ID | 2647 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
site | +0.014 |
her | +0.013 |
office | +0.012 |
ia | +0.012 |
men | +0.011 |
Neg logit contributions | |
mine | -0.015 |
berry | -0.013 |
Benny | -0.013 |
're | -0.013 |
wherever | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
mine | +0.143 |
Benny | +0.132 |
berry | +0.132 |
're | +0.129 |
ease | +0.129 |
Neg logit contributions | |
site | -0.161 |
her | -0.146 |
office | -0.143 |
ia | -0.139 |
men | -0.129 |
Token |
|
Token ID | 198 |
Feature activation | +2.76e-01 |
Pos logit contributions | |
mine | +0.009 |
berry | +0.008 |
wherever | +0.008 |
're | +0.008 |
Benny | +0.008 |
Neg logit contributions | |
site | -0.011 |
her | -0.010 |
ia | -0.010 |
office | -0.010 |
men | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
mine | +0.017 |
berry | +0.015 |
're | +0.015 |
Benny | +0.015 |
wherever | +0.015 |
Neg logit contributions | |
site | -0.024 |
her | -0.022 |
office | -0.022 |
ia | -0.022 |
wasteful | -0.020 |
Token | Then |
Token ID | 6423 |
Feature activation | -7.98e+00 |
Pos logit contributions | |
mine | +0.021 |
berry | +0.019 |
're | +0.019 |
Benny | +0.019 |
wherever | +0.019 |
Neg logit contributions | |
site | -0.018 |
her | -0.016 |
office | -0.016 |
ia | -0.015 |
wasteful | -0.014 |
Token | Maria |
Token ID | 14200 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
site | +0.671 |
office | +0.607 |
her | +0.600 |
ia | +0.599 |
men | +0.555 |
Neg logit contributions | |
mine | -0.528 |
berry | -0.485 |
're | -0.457 |
Benny | -0.451 |
wherever | -0.447 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
mine | +0.105 |
berry | +0.096 |
wherever | +0.096 |
Benny | +0.096 |
're | +0.095 |
Neg logit contributions | |
site | -0.123 |
her | -0.116 |
ia | -0.113 |
office | -0.112 |
men | -0.100 |
Token | a |
Token ID | 257 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
site | +0.150 |
her | +0.137 |
office | +0.135 |
ia | +0.130 |
men | +0.119 |
Neg logit contributions | |
mine | -0.162 |
berry | -0.151 |
're | -0.148 |
Benny | -0.144 |
wherever | -0.144 |
Token | big |
Token ID | 1263 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
site | +0.073 |
her | +0.066 |
office | +0.065 |
ia | +0.063 |
men | +0.057 |
Neg logit contributions | |
mine | -0.074 |
berry | -0.068 |
're | -0.067 |
Benny | -0.066 |
wherever | -0.066 |
Token | hill |
Token ID | 12788 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
site | +0.156 |
her | +0.142 |
office | +0.140 |
ia | +0.137 |
men | +0.126 |
Neg logit contributions | |
mine | -0.122 |
berry | -0.110 |
're | -0.109 |
Benny | -0.108 |
wherever | -0.108 |