Token | Ms |
Token ID | 6997 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
� | +0.010 |
Transparency | +0.009 |
Millions | +0.009 |
Reports | +0.009 |
Thousands | +0.009 |
Neg logit contributions | |
groom | -0.008 |
hander | -0.008 |
cha | -0.008 |
ladies | -0.007 |
gentlemen | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
groom | +0.053 |
hander | +0.051 |
cha | +0.050 |
ladies | +0.049 |
iquette | +0.048 |
Neg logit contributions | |
� | -0.054 |
Transparency | -0.053 |
Millions | -0.052 |
Reports | -0.051 |
Thousands | -0.050 |
Token | McCain |
Token ID | 14264 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
� | +0.012 |
Transparency | +0.011 |
Millions | +0.011 |
Reports | +0.011 |
Thousands | +0.010 |
Neg logit contributions | |
groom | -0.022 |
hander | -0.021 |
cha | -0.021 |
ladies | -0.020 |
iquette | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
� | +0.118 |
Transparency | +0.115 |
Millions | +0.113 |
Reports | +0.112 |
Thousands | +0.109 |
Neg logit contributions | |
groom | -0.078 |
hander | -0.073 |
cha | -0.072 |
ladies | -0.069 |
iquette | -0.068 |
Token | his |
Token ID | 465 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
groom | +0.042 |
hander | +0.040 |
cha | +0.039 |
ladies | +0.037 |
iquette | +0.037 |
Neg logit contributions | |
� | -0.063 |
Transparency | -0.062 |
Millions | -0.060 |
Reports | -0.060 |
Thousands | -0.058 |
Token | spouse |
Token ID | 16731 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
groom | +0.058 |
hander | +0.052 |
cha | +0.049 |
ladies | +0.045 |
iquette | +0.045 |
Neg logit contributions | |
� | -0.255 |
Transparency | -0.252 |
Millions | -0.249 |
Reports | -0.246 |
Thousands | -0.242 |
Token | , |
Token ID | 11 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
groom | +0.571 |
hander | +0.535 |
cha | +0.529 |
ladies | +0.511 |
iquette | +0.502 |
Neg logit contributions | |
� | -0.846 |
Transparency | -0.830 |
Millions | -0.814 |
Reports | -0.806 |
Thousands | -0.785 |
Token | is |
Token ID | 318 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
groom | +0.037 |
hander | +0.035 |
cha | +0.034 |
ladies | +0.033 |
iquette | +0.033 |
Neg logit contributions | |
� | -0.052 |
Transparency | -0.052 |
Millions | -0.050 |
Reports | -0.050 |
Thousands | -0.049 |
Token | president |
Token ID | 1893 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
groom | +0.061 |
hander | +0.057 |
cha | +0.056 |
ladies | +0.054 |
iquette | +0.054 |
Neg logit contributions | |
� | -0.086 |
Transparency | -0.084 |
Millions | -0.082 |
Reports | -0.081 |
Thousands | -0.079 |
Token | of |
Token ID | 286 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
groom | +0.174 |
hander | +0.164 |
cha | +0.163 |
ladies | +0.157 |
iquette | +0.155 |
Neg logit contributions | |
� | -0.214 |
Transparency | -0.210 |
Millions | -0.206 |
Reports | -0.203 |
Thousands | -0.198 |
Token | the |
Token ID | 262 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
� | +0.023 |
Transparency | +0.022 |
Millions | +0.021 |
Reports | +0.021 |
Thousands | +0.021 |
Neg logit contributions | |
groom | -0.021 |
hander | -0.019 |
cha | -0.019 |
ladies | -0.019 |
iquette | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
groom | +0.078 |
hander | +0.073 |
cha | +0.073 |
iquette | +0.072 |
concess | +0.072 |
Neg logit contributions | |
� | -0.074 |
Transparency | -0.074 |
Millions | -0.072 |
Reports | -0.071 |
Thousands | -0.069 |
Token | " |
Token ID | 1 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
groom | +0.064 |
hander | +0.061 |
cha | +0.060 |
ladies | +0.058 |
iquette | +0.058 |
Neg logit contributions | |
� | -0.069 |
Transparency | -0.068 |
Millions | -0.067 |
Reports | -0.066 |
Thousands | -0.064 |
Token | Senior |
Token ID | 31224 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
groom | +0.059 |
hander | +0.057 |
cha | +0.056 |
ladies | +0.055 |
iquette | +0.054 |
Neg logit contributions | |
� | -0.054 |
Transparency | -0.052 |
Millions | -0.051 |
Reports | -0.050 |
Thousands | -0.049 |
Token | public |
Token ID | 1171 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
groom | +0.296 |
hander | +0.280 |
cha | +0.279 |
ladies | +0.271 |
iquette | +0.268 |
Neg logit contributions | |
� | -0.298 |
Transparency | -0.291 |
Millions | -0.284 |
Reports | -0.280 |
Thousands | -0.271 |
Token | servants |
Token ID | 17523 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
groom | +0.028 |
hander | +0.026 |
cha | +0.026 |
ladies | +0.025 |
iquette | +0.024 |
Neg logit contributions | |
� | -0.061 |
Transparency | -0.060 |
Millions | -0.059 |
Reports | -0.059 |
Thousands | -0.057 |
Token | superv |
Token ID | 29745 |
Feature activation | +5.58e+00 |
Pos logit contributions | |
groom | +0.170 |
hander | +0.165 |
cha | +0.157 |
iquette | +0.153 |
ladies | +0.150 |
Neg logit contributions | |
� | -0.315 |
Transparency | -0.313 |
Millions | -0.307 |
Reports | -0.304 |
Thousands | -0.296 |
Token | ise |
Token ID | 786 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
groom | +0.425 |
hander | +0.396 |
cha | +0.385 |
ladies | +0.371 |
iquette | +0.364 |
Neg logit contributions | |
� | -0.972 |
Transparency | -0.965 |
Millions | -0.950 |
Reports | -0.943 |
Thousands | -0.920 |
Token | ang |
Token ID | 3550 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
groom | +0.172 |
hander | +0.164 |
cha | +0.160 |
ladies | +0.155 |
iquette | +0.155 |
Neg logit contributions | |
� | -0.231 |
Transparency | -0.228 |
Millions | -0.223 |
Reports | -0.220 |
Thousands | -0.215 |
Token | l |
Token ID | 75 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
groom | +0.069 |
hander | +0.065 |
cha | +0.064 |
ladies | +0.062 |
iquette | +0.061 |
Neg logit contributions | |
� | -0.099 |
Transparency | -0.097 |
Millions | -0.095 |
Reports | -0.094 |
Thousands | -0.091 |
Token | ophone |
Token ID | 38656 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
groom | +0.013 |
hander | +0.012 |
cha | +0.012 |
ladies | +0.011 |
iquette | +0.011 |
Neg logit contributions | |
� | -0.031 |
Transparency | -0.030 |
Millions | -0.030 |
Reports | -0.029 |
Thousands | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
groom | +0.265 |
hander | +0.251 |
cha | +0.248 |
ladies | +0.240 |
iquette | +0.238 |
Neg logit contributions | |
� | -0.331 |
Transparency | -0.325 |
Millions | -0.318 |
Reports | -0.314 |
Thousands | -0.305 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
groom | +0.038 |
hander | +0.036 |
cha | +0.035 |
ladies | +0.034 |
iquette | +0.034 |
Neg logit contributions | |
� | -0.058 |
Transparency | -0.057 |
Millions | -0.056 |
Reports | -0.055 |
Thousands | -0.054 |
Token | make |
Token ID | 787 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
groom | +0.011 |
hander | +0.010 |
cha | +0.010 |
ladies | +0.010 |
iquette | +0.010 |
Neg logit contributions | |
� | -0.017 |
Transparency | -0.017 |
Millions | -0.017 |
Reports | -0.016 |
Thousands | -0.016 |
Token | it |
Token ID | 340 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
groom | +0.009 |
hander | +0.009 |
cha | +0.009 |
ladies | +0.008 |
iquette | +0.008 |
Neg logit contributions | |
� | -0.014 |
Transparency | -0.013 |
Millions | -0.013 |
Reports | -0.013 |
Thousands | -0.013 |
Token | easier |
Token ID | 4577 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
� | +0.071 |
Transparency | +0.070 |
Millions | +0.068 |
Reports | +0.068 |
Thousands | +0.066 |
Neg logit contributions | |
groom | -0.055 |
hander | -0.052 |
cha | -0.051 |
ladies | -0.050 |
iquette | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
groom | +0.087 |
hander | +0.081 |
cha | +0.080 |
ladies | +0.078 |
iquette | +0.077 |
Neg logit contributions | |
� | -0.124 |
Transparency | -0.121 |
Millions | -0.119 |
Reports | -0.118 |
Thousands | -0.115 |
Token | superv |
Token ID | 29745 |
Feature activation | +5.54e+00 |
Pos logit contributions | |
� | +0.029 |
Transparency | +0.029 |
Millions | +0.028 |
Reports | +0.028 |
Thousands | +0.027 |
Neg logit contributions | |
groom | -0.019 |
hander | -0.018 |
cha | -0.017 |
ladies | -0.017 |
iquette | -0.016 |
Token | ise |
Token ID | 786 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
groom | +0.407 |
hander | +0.375 |
cha | +0.371 |
ladies | +0.353 |
iquette | +0.349 |
Neg logit contributions | |
� | -0.980 |
Transparency | -0.973 |
Millions | -0.956 |
Reports | -0.949 |
Thousands | -0.925 |
Token | setup |
Token ID | 9058 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
groom | +0.153 |
hander | +0.145 |
cha | +0.142 |
ladies | +0.137 |
iquette | +0.136 |
Neg logit contributions | |
� | -0.234 |
Transparency | -0.231 |
Millions | -0.227 |
Reports | -0.224 |
Thousands | -0.218 |
Token | periods |
Token ID | 9574 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
groom | +0.243 |
hander | +0.229 |
cha | +0.227 |
ladies | +0.221 |
iquette | +0.218 |
Neg logit contributions | |
� | -0.294 |
Transparency | -0.288 |
Millions | -0.282 |
Reports | -0.278 |
Thousands | -0.270 |
Token | . |
Token ID | 13 |
Feature activation | -9.09e-02 |
Pos logit contributions | |
groom | +0.038 |
hander | +0.036 |
cha | +0.036 |
ladies | +0.035 |
iquette | +0.035 |
Neg logit contributions | |
� | -0.045 |
Transparency | -0.044 |
Millions | -0.043 |
Reports | -0.043 |
Thousands | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +5.56e-01 |
Pos logit contributions | |
� | +0.011 |
Transparency | +0.011 |
Millions | +0.011 |
Reports | +0.010 |
Thousands | +0.010 |
Neg logit contributions | |
groom | -0.012 |
hander | -0.011 |
cha | -0.011 |
ladies | -0.011 |
iquette | -0.011 |
Token | younger |
Token ID | 7099 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
groom | +0.034 |
hander | +0.032 |
cha | +0.031 |
iquette | +0.030 |
ladies | +0.030 |
Neg logit contributions | |
� | -0.067 |
Transparency | -0.065 |
Millions | -0.064 |
Reports | -0.064 |
Thousands | -0.062 |
Token | or |
Token ID | 393 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
groom | +0.112 |
hander | +0.106 |
cha | +0.103 |
iquette | +0.099 |
ladies | +0.098 |
Neg logit contributions | |
� | -0.215 |
Transparency | -0.211 |
Millions | -0.208 |
Reports | -0.205 |
Thousands | -0.200 |
Token | older |
Token ID | 4697 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
groom | +0.171 |
hander | +0.165 |
cha | +0.162 |
iquette | +0.159 |
ladies | +0.154 |
Neg logit contributions | |
� | -0.274 |
Transparency | -0.266 |
Millions | -0.263 |
Reports | -0.259 |
Trend | -0.253 |
Token | than |
Token ID | 621 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
groom | +0.033 |
hander | +0.031 |
cha | +0.031 |
iquette | +0.030 |
ladies | +0.029 |
Neg logit contributions | |
� | -0.039 |
Transparency | -0.039 |
Millions | -0.038 |
Reports | -0.037 |
Thousands | -0.036 |
Token | her |
Token ID | 607 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
� | +0.025 |
Transparency | +0.025 |
Millions | +0.024 |
Reports | +0.024 |
Thousands | +0.023 |
Neg logit contributions | |
groom | -0.013 |
hander | -0.013 |
cha | -0.012 |
iquette | -0.012 |
ladies | -0.012 |
Token | spouse |
Token ID | 16731 |
Feature activation | +5.40e+00 |
Pos logit contributions | |
groom | +0.048 |
hander | +0.043 |
cha | +0.039 |
iquette | +0.036 |
ladies | +0.034 |
Neg logit contributions | |
� | -0.323 |
Transparency | -0.319 |
Millions | -0.316 |
Reports | -0.313 |
Thousands | -0.307 |
Token | . |
Token ID | 13 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
groom | +0.623 |
hander | +0.591 |
cha | +0.583 |
ladies | +0.565 |
iquette | +0.560 |
Neg logit contributions | |
� | -0.746 |
Transparency | -0.732 |
Millions | -0.716 |
Reports | -0.707 |
Thousands | -0.687 |
Token |
|
Token ID | 198 |
Feature activation | +3.13e-01 |
Pos logit contributions | |
� | +0.038 |
Transparency | +0.037 |
Millions | +0.036 |
Reports | +0.035 |
Thousands | +0.034 |
Neg logit contributions | |
groom | -0.041 |
hander | -0.039 |
cha | -0.039 |
ladies | -0.038 |
gentlemen | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
groom | +0.039 |
hander | +0.037 |
cha | +0.037 |
iquette | +0.036 |
ladies | +0.035 |
Neg logit contributions | |
� | -0.040 |
Transparency | -0.040 |
Millions | -0.039 |
Reports | -0.039 |
Thousands | -0.037 |
Token | Related |
Token ID | 9819 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
groom | +0.036 |
hander | +0.034 |
cha | +0.034 |
ladies | +0.033 |
iquette | +0.033 |
Neg logit contributions | |
� | -0.038 |
Transparency | -0.037 |
Millions | -0.036 |
Reports | -0.036 |
Thousands | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
groom | +0.004 |
hander | +0.003 |
cha | +0.003 |
ladies | +0.003 |
iquette | +0.003 |
Neg logit contributions | |
� | -0.003 |
Transparency | -0.003 |
Millions | -0.003 |
Reports | -0.003 |
Thousands | -0.003 |
Token | Ms |
Token ID | 10128 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
groom | +0.040 |
hander | +0.038 |
cha | +0.038 |
ladies | +0.037 |
iquette | +0.036 |
Neg logit contributions | |
� | -0.043 |
Transparency | -0.042 |
Millions | -0.041 |
Reports | -0.041 |
Thousands | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
groom | +0.051 |
hander | +0.049 |
cha | +0.048 |
ladies | +0.047 |
iquette | +0.047 |
Neg logit contributions | |
� | -0.051 |
Transparency | -0.050 |
Millions | -0.049 |
Reports | -0.049 |
Thousands | -0.047 |
Token | McCain |
Token ID | 14264 |
Feature activation | -9.62e-01 |
Pos logit contributions | |
� | +0.026 |
Transparency | +0.026 |
Millions | +0.024 |
Reports | +0.024 |
Thousands | +0.023 |
Neg logit contributions | |
groom | -0.052 |
hander | -0.050 |
cha | -0.050 |
ladies | -0.049 |
iquette | -0.048 |
Token | , |
Token ID | 11 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
� | +0.148 |
Transparency | +0.145 |
Millions | +0.142 |
Reports | +0.140 |
Thousands | +0.137 |
Neg logit contributions | |
groom | -0.096 |
hander | -0.091 |
cha | -0.089 |
ladies | -0.086 |
iquette | -0.085 |
Token | his |
Token ID | 465 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
groom | +0.017 |
hander | +0.016 |
cha | +0.016 |
iquette | +0.015 |
ladies | +0.015 |
Neg logit contributions | |
� | -0.025 |
Transparency | -0.025 |
Millions | -0.025 |
Reports | -0.024 |
Thousands | -0.024 |
Token | spouse |
Token ID | 16731 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
groom | +0.055 |
hander | +0.049 |
cha | +0.047 |
ladies | +0.043 |
iquette | +0.043 |
Neg logit contributions | |
� | -0.211 |
Transparency | -0.209 |
Millions | -0.206 |
Reports | -0.204 |
Thousands | -0.200 |
Token | , |
Token ID | 11 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
groom | +0.538 |
hander | +0.502 |
cha | +0.497 |
ladies | +0.480 |
iquette | +0.470 |
Neg logit contributions | |
� | -0.822 |
Transparency | -0.807 |
Millions | -0.791 |
Reports | -0.782 |
Thousands | -0.763 |
Token | is |
Token ID | 318 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
groom | +0.016 |
hander | +0.016 |
cha | +0.015 |
ladies | +0.015 |
iquette | +0.015 |
Neg logit contributions | |
� | -0.023 |
Transparency | -0.023 |
Millions | -0.022 |
Reports | -0.022 |
Thousands | -0.021 |
Token | also |
Token ID | 635 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
groom | +0.039 |
hander | +0.037 |
cha | +0.036 |
ladies | +0.035 |
iquette | +0.035 |
Neg logit contributions | |
� | -0.056 |
Transparency | -0.055 |
Millions | -0.054 |
Reports | -0.053 |
Thousands | -0.052 |
Token | sole |
Token ID | 6195 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
groom | +0.071 |
hander | +0.067 |
cha | +0.066 |
ladies | +0.063 |
iquette | +0.063 |
Neg logit contributions | |
� | -0.098 |
Transparency | -0.097 |
Millions | -0.094 |
Reports | -0.093 |
Thousands | -0.091 |
Token | owner |
Token ID | 4870 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
groom | +0.114 |
hander | +0.106 |
cha | +0.103 |
ladies | +0.099 |
iquette | +0.099 |
Neg logit contributions | |
� | -0.226 |
Transparency | -0.222 |
Millions | -0.219 |
Reports | -0.216 |
Thousands | -0.211 |
Token | company |
Token ID | 1664 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
� | +0.201 |
Transparency | +0.197 |
Millions | +0.193 |
Reports | +0.191 |
Thousands | +0.185 |
Neg logit contributions | |
groom | -0.152 |
hander | -0.143 |
cha | -0.141 |
ladies | -0.136 |
iquette | -0.136 |
Token | generates |
Token ID | 18616 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
groom | +0.055 |
hander | +0.052 |
cha | +0.051 |
ladies | +0.049 |
iquette | +0.049 |
Neg logit contributions | |
� | -0.086 |
Transparency | -0.085 |
Millions | -0.083 |
Reports | -0.082 |
Thousands | -0.080 |
Token | income |
Token ID | 3739 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
groom | +0.027 |
hander | +0.025 |
cha | +0.025 |
ladies | +0.024 |
gentlemen | +0.024 |
Neg logit contributions | |
� | -0.028 |
Transparency | -0.027 |
Millions | -0.026 |
Reports | -0.026 |
Thousands | -0.025 |
Token | for |
Token ID | 329 |
Feature activation | +4.04e-01 |
Pos logit contributions | |
� | +0.025 |
Transparency | +0.025 |
Millions | +0.024 |
Reports | +0.024 |
Thousands | +0.023 |
Neg logit contributions | |
groom | -0.020 |
hander | -0.019 |
cha | -0.019 |
ladies | -0.018 |
gentlemen | -0.018 |
Token | his |
Token ID | 465 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
groom | +0.045 |
hander | +0.042 |
cha | +0.042 |
ladies | +0.040 |
iquette | +0.040 |
Neg logit contributions | |
� | -0.058 |
Transparency | -0.057 |
Millions | -0.056 |
Reports | -0.055 |
Thousands | -0.053 |
Token | spouse |
Token ID | 16731 |
Feature activation | +5.32e+00 |
Pos logit contributions | |
groom | +0.100 |
hander | +0.093 |
cha | +0.092 |
ladies | +0.089 |
iquette | +0.088 |
Neg logit contributions | |
� | -0.151 |
Transparency | -0.148 |
Millions | -0.145 |
Reports | -0.143 |
Thousands | -0.140 |
Token | , |
Token ID | 11 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
groom | +0.546 |
hander | +0.510 |
cha | +0.508 |
ladies | +0.489 |
gentlemen | +0.478 |
Neg logit contributions | |
� | -0.805 |
Transparency | -0.790 |
Millions | -0.772 |
Reports | -0.765 |
Thousands | -0.745 |
Token | according |
Token ID | 1864 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
groom | +0.010 |
hander | +0.010 |
cha | +0.010 |
ladies | +0.009 |
iquette | +0.009 |
Neg logit contributions | |
� | -0.014 |
Transparency | -0.014 |
Millions | -0.014 |
Reports | -0.014 |
Thousands | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | -5.47e-01 |
Pos logit contributions | |
groom | +0.134 |
hander | +0.128 |
cha | +0.126 |
iquette | +0.123 |
ladies | +0.122 |
Neg logit contributions | |
� | -0.121 |
Transparency | -0.120 |
Millions | -0.117 |
Reports | -0.115 |
Thousands | -0.111 |
Token | his |
Token ID | 465 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
� | +0.074 |
Transparency | +0.071 |
Millions | +0.070 |
Reports | +0.069 |
Thousands | +0.067 |
Neg logit contributions | |
groom | -0.066 |
hander | -0.062 |
cha | -0.062 |
ladies | -0.060 |
iquette | -0.059 |
Token | disclosure |
Token ID | 13019 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
groom | +0.122 |
hander | +0.116 |
cha | +0.115 |
ladies | +0.112 |
iquette | +0.111 |
Neg logit contributions | |
� | -0.110 |
Transparency | -0.107 |
Millions | -0.105 |
Reports | -0.103 |
Thousands | -0.100 |
Token |
|
Token ID | 198 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
groom | +0.055 |
hander | +0.052 |
cha | +0.051 |
ladies | +0.050 |
iquette | +0.049 |
Neg logit contributions | |
� | -0.069 |
Transparency | -0.068 |
Millions | -0.067 |
Reports | -0.066 |
Thousands | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | +3.29e-01 |
Pos logit contributions | |
groom | +0.038 |
hander | +0.036 |
cha | +0.036 |
iquette | +0.035 |
concess | +0.035 |
Neg logit contributions | |
Transparency | -0.036 |
� | -0.036 |
Millions | -0.035 |
Reports | -0.035 |
Thousands | -0.034 |
Token | Senior |
Token ID | 31224 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
groom | +0.042 |
hander | +0.040 |
cha | +0.040 |
ladies | +0.038 |
iquette | +0.038 |
Neg logit contributions | |
� | -0.041 |
Transparency | -0.040 |
Millions | -0.039 |
Reports | -0.039 |
Thousands | -0.038 |
Token | public |
Token ID | 1171 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
groom | +0.268 |
cha | +0.253 |
hander | +0.253 |
ladies | +0.246 |
iquette | +0.242 |
Neg logit contributions | |
� | -0.269 |
Transparency | -0.262 |
Millions | -0.257 |
Reports | -0.253 |
Thousands | -0.245 |
Token | servants |
Token ID | 17523 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
groom | +0.008 |
cha | +0.007 |
hander | +0.007 |
ladies | +0.007 |
gentlemen | +0.006 |
Neg logit contributions | |
� | -0.018 |
Transparency | -0.018 |
Millions | -0.018 |
Reports | -0.018 |
Thousands | -0.017 |
Token | superv |
Token ID | 29745 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
groom | +0.152 |
hander | +0.145 |
cha | +0.140 |
iquette | +0.135 |
ladies | +0.135 |
Neg logit contributions | |
� | -0.277 |
Transparency | -0.274 |
Millions | -0.268 |
Reports | -0.266 |
Thousands | -0.259 |
Token | ise |
Token ID | 786 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
groom | +0.428 |
hander | +0.393 |
ladies | +0.377 |
cha | +0.376 |
gentlemen | +0.366 |
Neg logit contributions | |
� | -0.895 |
Transparency | -0.893 |
Millions | -0.883 |
Reports | -0.875 |
Thousands | -0.851 |
Token | ang |
Token ID | 3550 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
groom | +0.128 |
cha | +0.119 |
hander | +0.119 |
ladies | +0.113 |
iquette | +0.112 |
Neg logit contributions | |
� | -0.209 |
Transparency | -0.206 |
Millions | -0.202 |
Reports | -0.200 |
Thousands | -0.194 |
Token | l |
Token ID | 75 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
groom | +0.043 |
cha | +0.041 |
hander | +0.040 |
iquette | +0.039 |
ladies | +0.039 |
Neg logit contributions | |
� | -0.050 |
Transparency | -0.048 |
Millions | -0.047 |
Reports | -0.046 |
Thousands | -0.045 |
Token | ophone |
Token ID | 38656 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
� | +0.014 |
Transparency | +0.013 |
Millions | +0.013 |
Reports | +0.013 |
Thousands | +0.013 |
Neg logit contributions | |
groom | -0.007 |
hander | -0.007 |
cha | -0.007 |
concess | -0.007 |
ladies | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
groom | +0.209 |
cha | +0.194 |
hander | +0.194 |
ladies | +0.187 |
iquette | +0.184 |
Neg logit contributions | |
� | -0.321 |
Transparency | -0.316 |
Millions | -0.310 |
Reports | -0.306 |
Thousands | -0.298 |
Token | message |
Token ID | 3275 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
� | +0.099 |
Transparency | +0.097 |
Millions | +0.095 |
Reports | +0.093 |
Thousands | +0.091 |
Neg logit contributions | |
groom | -0.075 |
hander | -0.071 |
cha | -0.070 |
ladies | -0.068 |
iquette | -0.067 |
Token | for |
Token ID | 329 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
� | +0.237 |
Transparency | +0.232 |
Millions | +0.228 |
Reports | +0.225 |
Thousands | +0.219 |
Neg logit contributions | |
groom | -0.172 |
hander | -0.162 |
cha | -0.160 |
ladies | -0.154 |
iquette | -0.153 |
Token | the |
Token ID | 262 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
groom | +0.052 |
hander | +0.049 |
cha | +0.049 |
ladies | +0.047 |
iquette | +0.047 |
Neg logit contributions | |
� | -0.060 |
Transparency | -0.058 |
Millions | -0.057 |
Reports | -0.056 |
Thousands | -0.055 |
Token | trailing |
Token ID | 25462 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
groom | +0.035 |
hander | +0.033 |
cha | +0.033 |
ladies | +0.031 |
iquette | +0.031 |
Neg logit contributions | |
� | -0.046 |
Transparency | -0.045 |
Millions | -0.044 |
Reports | -0.043 |
Thousands | -0.042 |
Token | " |
Token ID | 366 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
� | +0.159 |
Transparency | +0.156 |
Millions | +0.152 |
Reports | +0.150 |
Thousands | +0.146 |
Neg logit contributions | |
groom | -0.130 |
cha | -0.123 |
hander | -0.122 |
ladies | -0.119 |
gentlemen | -0.116 |
Token | */ |
Token ID | 16208 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
� | +0.339 |
Transparency | +0.332 |
Millions | +0.325 |
Reports | +0.320 |
Thousands | +0.312 |
Neg logit contributions | |
groom | -0.255 |
hander | -0.243 |
cha | -0.239 |
ladies | -0.231 |
iquette | -0.229 |
Token | ". |
Token ID | 1911 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
groom | +0.527 |
hander | +0.497 |
cha | +0.493 |
ladies | +0.474 |
iquette | +0.468 |
Neg logit contributions | |
� | -0.781 |
Transparency | -0.768 |
Millions | -0.751 |
Reports | -0.741 |
Thousands | -0.724 |
Token | This |
Token ID | 770 |
Feature activation | -3.24e-01 |
Pos logit contributions | |
groom | +0.126 |
hander | +0.120 |
cha | +0.119 |
ladies | +0.116 |
iquette | +0.116 |
Neg logit contributions | |
� | -0.092 |
Transparency | -0.089 |
Millions | -0.086 |
Reports | -0.085 |
Thousands | -0.082 |
Token | line |
Token ID | 1627 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
� | +0.057 |
Transparency | +0.056 |
Millions | +0.055 |
Reports | +0.055 |
Thousands | +0.053 |
Neg logit contributions | |
groom | -0.025 |
hander | -0.023 |
cha | -0.023 |
ladies | -0.021 |
iquette | -0.021 |
Token | is |
Token ID | 318 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
groom | +0.035 |
hander | +0.033 |
cha | +0.032 |
ladies | +0.030 |
iquette | +0.030 |
Neg logit contributions | |
� | -0.084 |
Transparency | -0.083 |
Millions | -0.082 |
Reports | -0.081 |
Thousands | -0.079 |
Token | OK |
Token ID | 7477 |
Feature activation | -9.39e-02 |
Pos logit contributions | |
groom | +0.036 |
hander | +0.034 |
cha | +0.034 |
ladies | +0.032 |
iquette | +0.032 |
Neg logit contributions | |
� | -0.044 |
Transparency | -0.043 |
Millions | -0.042 |
Reports | -0.042 |
Thousands | -0.040 |
Token | he |
Token ID | 339 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
groom | +0.063 |
hander | +0.060 |
cha | +0.059 |
ladies | +0.056 |
iquette | +0.056 |
Neg logit contributions | |
� | -0.100 |
Transparency | -0.098 |
Millions | -0.096 |
Reports | -0.095 |
Thousands | -0.093 |
Token | is |
Token ID | 318 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
� | +0.002 |
Transparency | +0.002 |
Millions | +0.002 |
Reports | +0.002 |
Thousands | +0.002 |
Neg logit contributions | |
groom | -0.001 |
hander | -0.001 |
cha | -0.001 |
iquette | -0.001 |
ladies | -0.001 |
Token | younger |
Token ID | 7099 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
groom | +0.022 |
hander | +0.021 |
cha | +0.020 |
iquette | +0.020 |
ladies | +0.020 |
Neg logit contributions | |
� | -0.025 |
Transparency | -0.024 |
Millions | -0.024 |
Reports | -0.023 |
Thousands | -0.023 |
Token | than |
Token ID | 621 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
groom | +0.125 |
hander | +0.118 |
cha | +0.117 |
ladies | +0.113 |
iquette | +0.112 |
Neg logit contributions | |
� | -0.152 |
Transparency | -0.149 |
Millions | -0.146 |
Reports | -0.144 |
Thousands | -0.140 |
Token | his |
Token ID | 465 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
� | +0.067 |
Transparency | +0.065 |
Millions | +0.065 |
Reports | +0.064 |
Thousands | +0.062 |
Neg logit contributions | |
groom | -0.039 |
hander | -0.037 |
cha | -0.036 |
iquette | -0.035 |
ladies | -0.034 |
Token | spouse |
Token ID | 16731 |
Feature activation | +5.12e+00 |
Pos logit contributions | |
groom | +0.021 |
hander | +0.019 |
cha | +0.016 |
iquette | +0.014 |
ladies | +0.013 |
Neg logit contributions | |
� | -0.182 |
Transparency | -0.179 |
Millions | -0.177 |
Reports | -0.175 |
Thousands | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
groom | +0.569 |
hander | +0.538 |
cha | +0.531 |
ladies | +0.513 |
iquette | +0.509 |
Neg logit contributions | |
� | -0.730 |
Transparency | -0.717 |
Millions | -0.701 |
Reports | -0.692 |
Thousands | -0.673 |
Token |
|
Token ID | 198 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
� | +0.071 |
Transparency | +0.069 |
Millions | +0.066 |
Reports | +0.065 |
Thousands | +0.063 |
Neg logit contributions | |
groom | -0.090 |
hander | -0.086 |
cha | -0.085 |
ladies | -0.084 |
gentlemen | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | +9.79e-03 |
Pos logit contributions | |
groom | +0.000 |
hander | +0.000 |
cha | +0.000 |
iquette | +0.000 |
concess | +0.000 |
Neg logit contributions | |
Transparency | -0.000 |
� | -0.000 |
Millions | -0.000 |
Reports | -0.000 |
Thousands | -0.000 |
Token | For |
Token ID | 1890 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
groom | +0.001 |
hander | +0.001 |
cha | +0.001 |
ladies | +0.001 |
iquette | +0.001 |
Neg logit contributions | |
� | -0.001 |
Transparency | -0.001 |
Millions | -0.001 |
Reports | -0.001 |
Thousands | -0.001 |
Token | years |
Token ID | 812 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
groom | +0.010 |
hander | +0.009 |
cha | +0.009 |
ladies | +0.009 |
iquette | +0.009 |
Neg logit contributions | |
� | -0.017 |
Transparency | -0.016 |
Millions | -0.016 |
Reports | -0.016 |
Thousands | -0.016 |
Token | ] |
Token ID | 60 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
� | +0.063 |
Transparency | +0.062 |
Millions | +0.061 |
Reports | +0.060 |
Thousands | +0.058 |
Neg logit contributions | |
groom | -0.049 |
hander | -0.046 |
cha | -0.046 |
ladies | -0.044 |
gentlemen | -0.043 |
Token | which |
Token ID | 543 |
Feature activation | +9.29e-02 |
Pos logit contributions | |
groom | +0.038 |
hander | +0.036 |
cha | +0.036 |
ladies | +0.034 |
gentlemen | +0.034 |
Neg logit contributions | |
� | -0.053 |
Transparency | -0.052 |
Millions | -0.051 |
Reports | -0.050 |
Thousands | -0.049 |
Token | Obama |
Token ID | 2486 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
groom | +0.011 |
hander | +0.010 |
cha | +0.010 |
ladies | +0.010 |
iquette | +0.010 |
Neg logit contributions | |
� | -0.013 |
Transparency | -0.012 |
Millions | -0.012 |
Reports | -0.012 |
Thousands | -0.012 |
Token | came |
Token ID | 1625 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
� | +0.446 |
Transparency | +0.438 |
Millions | +0.429 |
Reports | +0.424 |
Thousands | +0.412 |
Neg logit contributions | |
groom | -0.332 |
hander | -0.316 |
cha | -0.310 |
ladies | -0.299 |
iquette | -0.298 |
Token | to |
Token ID | 284 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
� | +0.040 |
Transparency | +0.040 |
Millions | +0.039 |
Reports | +0.038 |
Thousands | +0.037 |
Neg logit contributions | |
groom | -0.037 |
hander | -0.035 |
cha | -0.034 |
ladies | -0.033 |
iquette | -0.033 |
Token | chair |
Token ID | 5118 |
Feature activation | +5.04e+00 |
Pos logit contributions | |
� | +0.022 |
Transparency | +0.021 |
Millions | +0.021 |
Reports | +0.021 |
Thousands | +0.020 |
Neg logit contributions | |
groom | -0.016 |
hander | -0.015 |
cha | -0.015 |
ladies | -0.014 |
iquette | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
groom | +0.528 |
hander | +0.496 |
cha | +0.492 |
ladies | +0.475 |
iquette | +0.465 |
Neg logit contributions | |
� | -0.749 |
Transparency | -0.738 |
Millions | -0.720 |
Reports | -0.713 |
Thousands | -0.693 |
Token | also |
Token ID | 635 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
groom | +0.026 |
hander | +0.025 |
cha | +0.024 |
ladies | +0.023 |
iquette | +0.023 |
Neg logit contributions | |
� | -0.037 |
Transparency | -0.036 |
Millions | -0.036 |
Reports | -0.035 |
Thousands | -0.034 |
Token | counted |
Token ID | 14789 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
groom | +0.109 |
hander | +0.103 |
cha | +0.102 |
ladies | +0.098 |
iquette | +0.097 |
Neg logit contributions | |
� | -0.151 |
Transparency | -0.148 |
Millions | -0.145 |
Reports | -0.143 |
Thousands | -0.139 |
Token | the |
Token ID | 262 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
� | +0.344 |
Transparency | +0.338 |
Millions | +0.331 |
Reports | +0.327 |
Thousands | +0.319 |
Neg logit contributions | |
groom | -0.241 |
hander | -0.227 |
cha | -0.224 |
ladies | -0.216 |
iquette | -0.214 |
Token | involvement |
Token ID | 9750 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
groom | +0.079 |
hander | +0.075 |
cha | +0.075 |
ladies | +0.072 |
gentlemen | +0.071 |
Neg logit contributions | |
� | -0.086 |
Transparency | -0.084 |
Millions | -0.082 |
Reports | -0.081 |
Thousands | -0.078 |
Token | the |
Token ID | 262 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
� | +0.062 |
Transparency | +0.060 |
Millions | +0.058 |
Reports | +0.058 |
Thousands | +0.056 |
Neg logit contributions | |
groom | -0.044 |
cha | -0.041 |
hander | -0.041 |
ladies | -0.040 |
gentlemen | -0.039 |
Token | conference |
Token ID | 4495 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
groom | +0.077 |
cha | +0.073 |
hander | +0.073 |
ladies | +0.071 |
gentlemen | +0.070 |
Neg logit contributions | |
� | -0.076 |
Transparency | -0.073 |
Millions | -0.071 |
Reports | -0.070 |
Thousands | -0.068 |
Token | website |
Token ID | 3052 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
� | +0.049 |
Transparency | +0.048 |
Millions | +0.047 |
Reports | +0.046 |
Thousands | +0.045 |
Neg logit contributions | |
groom | -0.042 |
hander | -0.039 |
cha | -0.039 |
ladies | -0.038 |
iquette | -0.037 |
Token | by |
Token ID | 416 |
Feature activation | -4.61e-04 |
Pos logit contributions | |
groom | +0.026 |
cha | +0.024 |
hander | +0.024 |
ladies | +0.023 |
gentlemen | +0.023 |
Neg logit contributions | |
� | -0.035 |
Transparency | -0.034 |
Millions | -0.034 |
Reports | -0.033 |
Thousands | -0.032 |
Token | two |
Token ID | 734 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
� | +0.000 |
Transparency | +0.000 |
Millions | +0.000 |
Reports | +0.000 |
Thousands | +0.000 |
Neg logit contributions | |
groom | -0.000 |
cha | -0.000 |
hander | -0.000 |
ladies | -0.000 |
gentlemen | -0.000 |
Token | chair |
Token ID | 5118 |
Feature activation | +5.00e+00 |
Pos logit contributions | |
groom | +0.064 |
cha | +0.060 |
hander | +0.060 |
ladies | +0.058 |
gentlemen | +0.058 |
Neg logit contributions | |
� | -0.066 |
Transparency | -0.064 |
Millions | -0.062 |
Reports | -0.061 |
Thousands | -0.059 |
Token | men |
Token ID | 3653 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
groom | +0.435 |
hander | +0.402 |
cha | +0.400 |
ladies | +0.386 |
gentlemen | +0.376 |
Neg logit contributions | |
� | -0.835 |
Transparency | -0.819 |
Millions | -0.805 |
Reports | -0.796 |
Thousands | -0.779 |
Token | because |
Token ID | 780 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
groom | +0.032 |
hander | +0.030 |
cha | +0.030 |
ladies | +0.029 |
gentlemen | +0.029 |
Neg logit contributions | |
� | -0.043 |
Transparency | -0.042 |
Millions | -0.042 |
Reports | -0.041 |
Thousands | -0.040 |
Token | they |
Token ID | 484 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
� | +0.098 |
Transparency | +0.095 |
Millions | +0.094 |
Reports | +0.092 |
Thousands | +0.090 |
Neg logit contributions | |
groom | -0.082 |
cha | -0.077 |
hander | -0.076 |
ladies | -0.075 |
gentlemen | -0.074 |
Token | could |
Token ID | 714 |
Feature activation | +5.40e-01 |
Pos logit contributions | |
� | +0.115 |
Transparency | +0.112 |
Millions | +0.110 |
Reports | +0.108 |
Thousands | +0.105 |
Neg logit contributions | |
groom | -0.107 |
hander | -0.101 |
cha | -0.101 |
ladies | -0.097 |
iquette | -0.096 |
Token | not |
Token ID | 407 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
groom | +0.081 |
hander | +0.077 |
cha | +0.076 |
ladies | +0.074 |
gentlemen | +0.073 |
Neg logit contributions | |
� | -0.058 |
Transparency | -0.056 |
Millions | -0.054 |
Reports | -0.053 |
Thousands | -0.051 |
Token | all |
Token ID | 477 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
groom | +0.046 |
hander | +0.043 |
cha | +0.042 |
ladies | +0.041 |
iquette | +0.041 |
Neg logit contributions | |
� | -0.065 |
Transparency | -0.064 |
Millions | -0.063 |
Reports | -0.062 |
Thousands | -0.060 |
Token | sparkling |
Token ID | 39072 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
� | +0.104 |
Transparency | +0.102 |
Millions | +0.100 |
Reports | +0.099 |
Thousands | +0.096 |
Neg logit contributions | |
groom | -0.068 |
hander | -0.063 |
cha | -0.063 |
ladies | -0.060 |
iquette | -0.060 |
Token | green |
Token ID | 4077 |
Feature activation | +5.59e-01 |
Pos logit contributions | |
groom | +0.070 |
hander | +0.066 |
cha | +0.065 |
ladies | +0.063 |
iquette | +0.062 |
Neg logit contributions | |
� | -0.102 |
Transparency | -0.100 |
Millions | -0.098 |
Reports | -0.097 |
Thousands | -0.094 |
Token | eyes |
Token ID | 2951 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
groom | +0.056 |
hander | +0.052 |
cha | +0.052 |
ladies | +0.050 |
iquette | +0.049 |
Neg logit contributions | |
� | -0.086 |
Transparency | -0.084 |
Millions | -0.083 |
Reports | -0.082 |
Thousands | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
� | +0.112 |
Transparency | +0.110 |
Millions | +0.108 |
Reports | +0.107 |
Thousands | +0.104 |
Neg logit contributions | |
groom | -0.078 |
hander | -0.073 |
cha | -0.072 |
ladies | -0.070 |
iquette | -0.069 |
Token | boun |
Token ID | 31283 |
Feature activation | +4.94e+00 |
Pos logit contributions | |
groom | +0.017 |
hander | +0.015 |
cha | +0.015 |
ladies | +0.015 |
iquette | +0.014 |
Neg logit contributions | |
� | -0.026 |
Transparency | -0.025 |
Millions | -0.025 |
Reports | -0.024 |
Thousands | -0.024 |
Token | cy |
Token ID | 948 |
Feature activation | +9.12e-01 |
Pos logit contributions | |
groom | +0.394 |
hander | +0.366 |
cha | +0.357 |
ladies | +0.340 |
iquette | +0.337 |
Neg logit contributions | |
� | -0.858 |
Transparency | -0.845 |
Millions | -0.830 |
Reports | -0.820 |
Thousands | -0.803 |
Token | blond |
Token ID | 32749 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
groom | +0.091 |
hander | +0.086 |
cha | +0.084 |
ladies | +0.081 |
iquette | +0.080 |
Neg logit contributions | |
� | -0.140 |
Transparency | -0.138 |
Millions | -0.135 |
Reports | -0.133 |
Thousands | -0.130 |
Token | hair |
Token ID | 4190 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
groom | +0.057 |
hander | +0.053 |
cha | +0.052 |
ladies | +0.050 |
iquette | +0.049 |
Neg logit contributions | |
� | -0.117 |
Transparency | -0.116 |
Millions | -0.114 |
Reports | -0.112 |
Thousands | -0.110 |
Token | , |
Token ID | 11 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
groom | +0.076 |
hander | +0.071 |
cha | +0.071 |
ladies | +0.068 |
iquette | +0.067 |
Neg logit contributions | |
� | -0.113 |
Transparency | -0.111 |
Millions | -0.109 |
Reports | -0.107 |
Thousands | -0.105 |
Token | and |
Token ID | 290 |
Feature activation | -9.85e-02 |
Pos logit contributions | |
groom | +0.016 |
hander | +0.015 |
cha | +0.015 |
ladies | +0.014 |
iquette | +0.014 |
Neg logit contributions | |
� | -0.024 |
Transparency | -0.023 |
Millions | -0.023 |
Reports | -0.023 |
Thousands | -0.022 |
Token | orders |
Token ID | 6266 |
Feature activation | -6.77e-01 |
Pos logit contributions | |
groom | +0.053 |
hander | +0.050 |
cha | +0.050 |
ladies | +0.048 |
iquette | +0.047 |
Neg logit contributions | |
� | -0.073 |
Transparency | -0.071 |
Millions | -0.070 |
Reports | -0.069 |
Thousands | -0.067 |
Token | of |
Token ID | 286 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
� | +0.099 |
Transparency | +0.097 |
Millions | +0.096 |
Reports | +0.094 |
Thousands | +0.092 |
Neg logit contributions | |
groom | -0.072 |
hander | -0.068 |
cha | -0.068 |
ladies | -0.066 |
gentlemen | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
� | +0.018 |
Transparency | +0.017 |
Millions | +0.017 |
Reports | +0.017 |
Thousands | +0.016 |
Neg logit contributions | |
groom | -0.014 |
hander | -0.013 |
cha | -0.013 |
ladies | -0.013 |
gentlemen | -0.013 |
Token | committee |
Token ID | 5583 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
groom | +0.063 |
hander | +0.060 |
cha | +0.060 |
ladies | +0.058 |
iquette | +0.057 |
Neg logit contributions | |
� | -0.069 |
Transparency | -0.067 |
Millions | -0.066 |
Reports | -0.065 |
Thousands | -0.063 |
Token | 's |
Token ID | 338 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
groom | +0.254 |
cha | +0.237 |
hander | +0.237 |
ladies | +0.227 |
iquette | +0.222 |
Neg logit contributions | |
� | -0.446 |
Transparency | -0.438 |
Millions | -0.431 |
Reports | -0.425 |
Thousands | -0.416 |
Token | chair |
Token ID | 5118 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
groom | +0.025 |
hander | +0.023 |
cha | +0.023 |
ladies | +0.022 |
iquette | +0.022 |
Neg logit contributions | |
� | -0.035 |
Transparency | -0.034 |
Millions | -0.033 |
Reports | -0.033 |
Thousands | -0.032 |
Token | woman |
Token ID | 8580 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
groom | +0.434 |
hander | +0.403 |
cha | +0.401 |
ladies | +0.385 |
iquette | +0.377 |
Neg logit contributions | |
� | -0.814 |
Transparency | -0.800 |
Millions | -0.787 |
Reports | -0.778 |
Thousands | -0.760 |
Token | , |
Token ID | 11 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
groom | +0.107 |
hander | +0.100 |
cha | +0.100 |
ladies | +0.096 |
iquette | +0.094 |
Neg logit contributions | |
� | -0.165 |
Transparency | -0.162 |
Millions | -0.159 |
Reports | -0.157 |
Thousands | -0.153 |
Token | Sen |
Token ID | 2311 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
groom | +0.046 |
hander | +0.044 |
cha | +0.044 |
ladies | +0.043 |
gentlemen | +0.042 |
Neg logit contributions | |
� | -0.046 |
Transparency | -0.044 |
Millions | -0.044 |
Reports | -0.043 |
Thousands | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
groom | +0.211 |
hander | +0.199 |
cha | +0.197 |
iquette | +0.191 |
ladies | +0.191 |
Neg logit contributions | |
� | -0.203 |
Transparency | -0.201 |
Millions | -0.196 |
Reports | -0.193 |
Thousands | -0.186 |
Token | Joan |
Token ID | 26640 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
� | +0.019 |
Transparency | +0.018 |
Millions | +0.018 |
Reports | +0.018 |
Thousands | +0.017 |
Neg logit contributions | |
groom | -0.016 |
hander | -0.015 |
cha | -0.015 |
ladies | -0.014 |
iquette | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
� | +0.183 |
Transparency | +0.179 |
Millions | +0.175 |
Reports | +0.173 |
Thousands | +0.168 |
Neg logit contributions | |
groom | -0.150 |
hander | -0.142 |
cha | -0.140 |
ladies | -0.136 |
iquette | -0.134 |
Token |
|
Token ID | 198 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
groom | +0.004 |
hander | +0.004 |
cha | +0.004 |
ladies | +0.003 |
iquette | +0.003 |
Neg logit contributions | |
� | -0.004 |
Transparency | -0.004 |
Millions | -0.004 |
Reports | -0.004 |
Trend | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
groom | +0.081 |
hander | +0.076 |
cha | +0.075 |
iquette | +0.074 |
concess | +0.073 |
Neg logit contributions | |
� | -0.078 |
Transparency | -0.078 |
Millions | -0.077 |
Reports | -0.076 |
Thousands | -0.073 |
Token | Despite |
Token ID | 8332 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
groom | +0.079 |
hander | +0.076 |
cha | +0.075 |
ladies | +0.073 |
iquette | +0.072 |
Neg logit contributions | |
� | -0.080 |
Transparency | -0.078 |
Millions | -0.076 |
Reports | -0.075 |
Thousands | -0.073 |
Token | her |
Token ID | 607 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
� | +0.070 |
Transparency | +0.068 |
Millions | +0.067 |
Reports | +0.066 |
Thousands | +0.064 |
Neg logit contributions | |
groom | -0.061 |
cha | -0.057 |
hander | -0.057 |
ladies | -0.056 |
gentlemen | -0.054 |
Token | conc |
Token ID | 1673 |
Feature activation | +4.89e+00 |
Pos logit contributions | |
groom | +0.232 |
cha | +0.219 |
hander | +0.218 |
ladies | +0.213 |
gentlemen | +0.209 |
Neg logit contributions | |
� | -0.232 |
Transparency | -0.225 |
Millions | -0.220 |
Reports | -0.217 |
Thousands | -0.211 |
Token | ili |
Token ID | 2403 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
groom | +0.523 |
hander | +0.494 |
cha | +0.487 |
ladies | +0.470 |
iquette | +0.467 |
Neg logit contributions | |
� | -0.717 |
Transparency | -0.704 |
Millions | -0.689 |
Reports | -0.680 |
Thousands | -0.663 |
Token | atory |
Token ID | 2870 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
groom | +0.029 |
hander | +0.027 |
cha | +0.026 |
ladies | +0.026 |
iquette | +0.026 |
Neg logit contributions | |
� | -0.035 |
Transparency | -0.035 |
Millions | -0.034 |
Reports | -0.033 |
Thousands | -0.032 |
Token | tone |
Token ID | 8216 |
Feature activation | -8.22e-01 |
Pos logit contributions | |
groom | +0.098 |
hander | +0.093 |
cha | +0.092 |
ladies | +0.090 |
iquette | +0.089 |
Neg logit contributions | |
� | -0.100 |
Transparency | -0.098 |
Millions | -0.095 |
Reports | -0.094 |
Thousands | -0.091 |
Token | , |
Token ID | 11 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
� | +0.128 |
Transparency | +0.126 |
Millions | +0.124 |
Reports | +0.122 |
Thousands | +0.119 |
Neg logit contributions | |
groom | -0.080 |
hander | -0.075 |
cha | -0.074 |
ladies | -0.072 |
iquette | -0.071 |
Token | Merkel |
Token ID | 18263 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
groom | +0.076 |
hander | +0.073 |
cha | +0.072 |
ladies | +0.071 |
gentlemen | +0.070 |
Neg logit contributions | |
� | -0.056 |
Transparency | -0.054 |
Millions | -0.053 |
Reports | -0.052 |
Thousands | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +5.09e-02 |
Pos logit contributions | |
groom | +0.123 |
hander | +0.117 |
cha | +0.115 |
ladies | +0.111 |
iquette | +0.110 |
Neg logit contributions | |
� | -0.165 |
Transparency | -0.162 |
Millions | -0.159 |
Reports | -0.157 |
Thousands | -0.153 |
Token | recently |
Token ID | 2904 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
groom | +0.006 |
hander | +0.006 |
cha | +0.006 |
ladies | +0.006 |
iquette | +0.006 |
Neg logit contributions | |
� | -0.007 |
Transparency | -0.007 |
Millions | -0.006 |
Reports | -0.006 |
Thousands | -0.006 |
Token | appointed |
Token ID | 9899 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
groom | +0.068 |
hander | +0.063 |
cha | +0.063 |
ladies | +0.061 |
gentlemen | +0.059 |
Neg logit contributions | |
� | -0.100 |
Transparency | -0.098 |
Millions | -0.097 |
Reports | -0.095 |
Thousands | -0.093 |
Token | Hydro |
Token ID | 32116 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
groom | +0.181 |
hander | +0.170 |
cha | +0.170 |
ladies | +0.164 |
gentlemen | +0.162 |
Neg logit contributions | |
� | -0.194 |
Transparency | -0.188 |
Millions | -0.185 |
Reports | -0.183 |
Thousands | -0.177 |
Token | board |
Token ID | 3096 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
� | +0.265 |
Transparency | +0.259 |
Millions | +0.254 |
Reports | +0.251 |
Thousands | +0.244 |
Neg logit contributions | |
groom | -0.217 |
hander | -0.205 |
cha | -0.203 |
ladies | -0.196 |
iquette | -0.194 |
Token | chair |
Token ID | 5118 |
Feature activation | +4.87e+00 |
Pos logit contributions | |
groom | +0.108 |
hander | +0.101 |
cha | +0.099 |
ladies | +0.094 |
iquette | +0.093 |
Neg logit contributions | |
� | -0.220 |
Transparency | -0.217 |
Millions | -0.213 |
Reports | -0.211 |
Thousands | -0.206 |
Token | Sandy |
Token ID | 17581 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
groom | +0.561 |
hander | +0.531 |
cha | +0.524 |
ladies | +0.508 |
iquette | +0.503 |
Neg logit contributions | |
� | -0.674 |
Transparency | -0.660 |
Millions | -0.647 |
Reports | -0.639 |
Thousands | -0.621 |
Token | Riley |
Token ID | 22637 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
� | +0.059 |
Transparency | +0.057 |
Millions | +0.056 |
Reports | +0.055 |
Thousands | +0.053 |
Neg logit contributions | |
groom | -0.066 |
hander | -0.063 |
cha | -0.062 |
ladies | -0.060 |
iquette | -0.060 |
Token | briefed |
Token ID | 32664 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
� | +0.047 |
Transparency | +0.046 |
Millions | +0.045 |
Reports | +0.044 |
Thousands | +0.043 |
Neg logit contributions | |
groom | -0.037 |
hander | -0.035 |
cha | -0.035 |
ladies | -0.034 |
iquette | -0.033 |
Token | media |
Token ID | 2056 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
groom | +0.006 |
hander | +0.006 |
cha | +0.006 |
ladies | +0.006 |
iquette | +0.006 |
Neg logit contributions | |
� | -0.008 |
Transparency | -0.007 |
Millions | -0.007 |
Reports | -0.007 |
Thousands | -0.007 |
Token | on |
Token ID | 319 |
Feature activation | +9.88e-01 |
Pos logit contributions | |
� | +0.404 |
Transparency | +0.395 |
Millions | +0.388 |
Reports | +0.383 |
Thousands | +0.374 |
Neg logit contributions | |
groom | -0.270 |
hander | -0.252 |
cha | -0.251 |
ladies | -0.242 |
gentlemen | -0.238 |
Token | who |
Token ID | 508 |
Feature activation | +5.62e-01 |
Pos logit contributions | |
� | +0.054 |
Transparency | +0.053 |
Millions | +0.052 |
Reports | +0.051 |
Thousands | +0.050 |
Neg logit contributions | |
groom | -0.044 |
hander | -0.041 |
cha | -0.041 |
ladies | -0.040 |
iquette | -0.039 |
Token | would |
Token ID | 561 |
Feature activation | +9.64e-01 |
Pos logit contributions | |
groom | +0.054 |
hander | +0.051 |
cha | +0.050 |
ladies | +0.048 |
iquette | +0.048 |
Neg logit contributions | |
� | -0.088 |
Transparency | -0.087 |
Millions | -0.085 |
Reports | -0.084 |
Thousands | -0.082 |
Token | select |
Token ID | 2922 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
groom | +0.094 |
hander | +0.089 |
cha | +0.087 |
ladies | +0.084 |
iquette | +0.083 |
Neg logit contributions | |
� | -0.150 |
Transparency | -0.147 |
Millions | -0.145 |
Reports | -0.143 |
Thousands | -0.139 |
Token | a |
Token ID | 257 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
groom | +0.277 |
hander | +0.261 |
cha | +0.258 |
ladies | +0.250 |
iquette | +0.247 |
Neg logit contributions | |
� | -0.350 |
Transparency | -0.343 |
Millions | -0.336 |
Reports | -0.331 |
Thousands | -0.322 |
Token | rotating |
Token ID | 24012 |
Feature activation | +3.80e-01 |
Pos logit contributions | |
groom | +0.016 |
hander | +0.015 |
cha | +0.015 |
ladies | +0.014 |
iquette | +0.014 |
Neg logit contributions | |
� | -0.016 |
Transparency | -0.016 |
Millions | -0.015 |
Reports | -0.015 |
Thousands | -0.015 |
Token | chair |
Token ID | 5118 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
groom | +0.047 |
hander | +0.044 |
cha | +0.044 |
ladies | +0.042 |
iquette | +0.042 |
Neg logit contributions | |
� | -0.050 |
Transparency | -0.049 |
Millions | -0.048 |
Reports | -0.047 |
Thousands | -0.046 |
Token | to |
Token ID | 284 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
groom | +0.467 |
hander | +0.438 |
cha | +0.432 |
ladies | +0.415 |
iquette | +0.410 |
Neg logit contributions | |
� | -0.762 |
Transparency | -0.749 |
Millions | -0.735 |
Reports | -0.727 |
Thousands | -0.709 |
Token | serve |
Token ID | 4691 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
� | +0.055 |
Transparency | +0.055 |
Millions | +0.053 |
Reports | +0.053 |
Thousands | +0.052 |
Neg logit contributions | |
groom | -0.031 |
hander | -0.030 |
cha | -0.029 |
iquette | -0.028 |
ladies | -0.028 |
Token | as |
Token ID | 355 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
groom | +0.330 |
hander | +0.309 |
cha | +0.308 |
ladies | +0.297 |
gentlemen | +0.291 |
Neg logit contributions | |
� | -0.459 |
Transparency | -0.449 |
Millions | -0.440 |
Reports | -0.435 |
Thousands | -0.423 |
Token | head |
Token ID | 1182 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
� | +0.020 |
Transparency | +0.020 |
Millions | +0.020 |
Reports | +0.019 |
Thousands | +0.019 |
Neg logit contributions | |
groom | -0.017 |
hander | -0.016 |
cha | -0.015 |
ladies | -0.015 |
iquette | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | -2.62e-01 |
Pos logit contributions | |
groom | +0.067 |
hander | +0.064 |
cha | +0.063 |
ladies | +0.061 |
iquette | +0.061 |
Neg logit contributions | |
� | -0.069 |
Transparency | -0.068 |
Millions | -0.066 |
Reports | -0.065 |
Thousands | -0.063 |
Token | with |
Token ID | 351 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
groom | +0.141 |
hander | +0.132 |
cha | +0.131 |
ladies | +0.126 |
gentlemen | +0.124 |
Neg logit contributions | |
� | -0.198 |
Transparency | -0.195 |
Millions | -0.190 |
Reports | -0.189 |
Thousands | -0.184 |
Token | two |
Token ID | 734 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
groom | +0.093 |
cha | +0.087 |
hander | +0.087 |
ladies | +0.085 |
gentlemen | +0.083 |
Neg logit contributions | |
� | -0.107 |
Transparency | -0.104 |
Millions | -0.102 |
Reports | -0.101 |
Thousands | -0.098 |
Token | homes |
Token ID | 5682 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
groom | +0.062 |
hander | +0.058 |
cha | +0.058 |
ladies | +0.056 |
gentlemen | +0.055 |
Neg logit contributions | |
� | -0.071 |
Transparency | -0.070 |
Millions | -0.068 |
Reports | -0.067 |
Thousands | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
groom | +0.043 |
hander | +0.041 |
cha | +0.040 |
ladies | +0.039 |
iquette | +0.039 |
Neg logit contributions | |
� | -0.060 |
Transparency | -0.059 |
Millions | -0.057 |
Reports | -0.057 |
Thousands | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
groom | +0.027 |
hander | +0.025 |
cha | +0.025 |
ladies | +0.025 |
iquette | +0.024 |
Neg logit contributions | |
� | -0.032 |
Transparency | -0.032 |
Millions | -0.031 |
Reports | -0.031 |
Thousands | -0.030 |
Token | wife |
Token ID | 3656 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
groom | +0.020 |
hander | +0.019 |
cha | +0.019 |
ladies | +0.018 |
iquette | +0.018 |
Neg logit contributions | |
� | -0.030 |
Transparency | -0.029 |
Millions | -0.028 |
Reports | -0.028 |
Thousands | -0.027 |
Token | with |
Token ID | 351 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
groom | +0.509 |
hander | +0.476 |
cha | +0.471 |
ladies | +0.456 |
iquette | +0.447 |
Neg logit contributions | |
� | -0.721 |
Transparency | -0.711 |
Millions | -0.693 |
Reports | -0.688 |
Thousands | -0.668 |
Token | a |
Token ID | 257 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
groom | +0.088 |
hander | +0.083 |
cha | +0.082 |
ladies | +0.080 |
gentlemen | +0.079 |
Neg logit contributions | |
� | -0.101 |
Transparency | -0.099 |
Millions | -0.097 |
Reports | -0.096 |
Thousands | -0.093 |
Token | federal |
Token ID | 2717 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
groom | +0.025 |
hander | +0.023 |
cha | +0.023 |
ladies | +0.022 |
iquette | +0.022 |
Neg logit contributions | |
� | -0.040 |
Transparency | -0.039 |
Millions | -0.038 |
Reports | -0.038 |
Thousands | -0.037 |
Token | bank |
Token ID | 3331 |
Feature activation | +6.45e-01 |
Pos logit contributions | |
� | +0.027 |
Transparency | +0.026 |
Millions | +0.026 |
Reports | +0.025 |
Thousands | +0.025 |
Neg logit contributions | |
groom | -0.019 |
hander | -0.018 |
cha | -0.018 |
ladies | -0.017 |
iquette | -0.017 |
Token | fraud |
Token ID | 7394 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
groom | +0.074 |
hander | +0.070 |
cha | +0.070 |
ladies | +0.067 |
iquette | +0.067 |
Neg logit contributions | |
� | -0.089 |
Transparency | -0.088 |
Millions | -0.086 |
Reports | -0.085 |
Thousands | -0.082 |
Token | k |
Token ID | 74 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
groom | +0.106 |
cha | +0.099 |
hander | +0.097 |
ladies | +0.096 |
gentlemen | +0.095 |
Neg logit contributions | |
� | -0.133 |
Transparency | -0.128 |
Millions | -0.127 |
Reports | -0.125 |
Thousands | -0.122 |
Token | end |
Token ID | 437 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
groom | +0.191 |
cha | +0.182 |
hander | +0.178 |
ladies | +0.176 |
iquette | +0.175 |
Neg logit contributions | |
� | -0.129 |
Transparency | -0.123 |
Millions | -0.120 |
Reports | -0.117 |
Trend | -0.113 |
Token | all |
Token ID | 439 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
� | +0.076 |
Transparency | +0.072 |
Millions | +0.071 |
Reports | +0.069 |
Thousands | +0.067 |
Neg logit contributions | |
groom | -0.096 |
cha | -0.091 |
hander | -0.091 |
iquette | -0.089 |
ladies | -0.088 |
Token | became |
Token ID | 2627 |
Feature activation | -6.72e-01 |
Pos logit contributions | |
� | +0.166 |
Transparency | +0.164 |
Millions | +0.160 |
Reports | +0.158 |
Thousands | +0.153 |
Neg logit contributions | |
groom | -0.142 |
hander | -0.134 |
cha | -0.133 |
ladies | -0.128 |
iquette | -0.127 |
Token | a |
Token ID | 257 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
� | +0.099 |
Transparency | +0.097 |
Millions | +0.095 |
Reports | +0.094 |
Thousands | +0.092 |
Neg logit contributions | |
groom | -0.072 |
hander | -0.067 |
cha | -0.067 |
ladies | -0.064 |
iquette | -0.063 |
Token | veterinarian |
Token ID | 41404 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
groom | +0.017 |
hander | +0.016 |
cha | +0.016 |
ladies | +0.015 |
iquette | +0.015 |
Neg logit contributions | |
� | -0.022 |
Transparency | -0.021 |
Millions | -0.021 |
Reports | -0.021 |
Thousands | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
groom | +0.495 |
hander | +0.466 |
cha | +0.459 |
ladies | +0.442 |
iquette | +0.439 |
Neg logit contributions | |
� | -0.734 |
Transparency | -0.721 |
Millions | -0.707 |
Reports | -0.698 |
Thousands | -0.680 |
Token |
|
Token ID | 198 |
Feature activation | +4.62e-01 |
Pos logit contributions | |
� | +0.019 |
Transparency | +0.018 |
Millions | +0.018 |
Reports | +0.017 |
Thousands | +0.017 |
Neg logit contributions | |
groom | -0.017 |
hander | -0.016 |
cha | -0.016 |
ladies | -0.016 |
iquette | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
groom | +0.060 |
hander | +0.056 |
cha | +0.056 |
iquette | +0.055 |
concess | +0.055 |
Neg logit contributions | |
� | -0.057 |
Transparency | -0.057 |
Millions | -0.056 |
Reports | -0.055 |
Thousands | -0.054 |
Token | Information |
Token ID | 21918 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
groom | +0.058 |
hander | +0.056 |
cha | +0.055 |
ladies | +0.053 |
iquette | +0.053 |
Neg logit contributions | |
� | -0.057 |
Transparency | -0.056 |
Millions | -0.055 |
Reports | -0.054 |
Thousands | -0.052 |
Token | about |
Token ID | 546 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
� | +0.351 |
Transparency | +0.344 |
Millions | +0.337 |
Reports | +0.333 |
Thousands | +0.324 |
Neg logit contributions | |
groom | -0.285 |
hander | -0.270 |
cha | -0.266 |
ladies | -0.257 |
iquette | -0.255 |
Token | closely |
Token ID | 7173 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
� | +0.171 |
Transparency | +0.166 |
Millions | +0.164 |
Reports | +0.161 |
Thousands | +0.157 |
Neg logit contributions | |
groom | -0.146 |
hander | -0.137 |
cha | -0.136 |
ladies | -0.132 |
gentlemen | -0.129 |
Token | linked |
Token ID | 6692 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
groom | +0.129 |
hander | +0.123 |
cha | +0.121 |
ladies | +0.117 |
iquette | +0.116 |
Neg logit contributions | |
� | -0.153 |
Transparency | -0.151 |
Millions | -0.147 |
Reports | -0.145 |
Thousands | -0.141 |
Token | to |
Token ID | 284 |
Feature activation | -3.78e-01 |
Pos logit contributions | |
� | +0.252 |
Transparency | +0.248 |
Millions | +0.241 |
Reports | +0.238 |
Thousands | +0.232 |
Neg logit contributions | |
groom | -0.209 |
hander | -0.197 |
cha | -0.196 |
ladies | -0.189 |
gentlemen | -0.186 |
Token | Obama |
Token ID | 2486 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
� | +0.053 |
Transparency | +0.051 |
Millions | +0.050 |
Reports | +0.050 |
Thousands | +0.048 |
Neg logit contributions | |
groom | -0.044 |
hander | -0.041 |
cha | -0.041 |
ladies | -0.040 |
iquette | -0.039 |
Token | 's |
Token ID | 338 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
� | +0.511 |
Transparency | +0.501 |
Millions | +0.491 |
Reports | +0.486 |
Thousands | +0.475 |
Neg logit contributions | |
groom | -0.318 |
hander | -0.296 |
cha | -0.294 |
ladies | -0.282 |
iquette | -0.278 |
Token | chair |
Token ID | 5118 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
groom | +0.016 |
hander | +0.015 |
cha | +0.015 |
ladies | +0.014 |
iquette | +0.014 |
Neg logit contributions | |
� | -0.017 |
Transparency | -0.017 |
Millions | -0.017 |
Reports | -0.016 |
Thousands | -0.016 |
Token | manship |
Token ID | 25428 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
groom | +0.528 |
hander | +0.499 |
cha | +0.494 |
ladies | +0.477 |
iquette | +0.472 |
Neg logit contributions | |
� | -0.698 |
Transparency | -0.685 |
Millions | -0.671 |
Reports | -0.664 |
Thousands | -0.645 |
Token | of |
Token ID | 286 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
groom | +0.007 |
hander | +0.007 |
cha | +0.007 |
ladies | +0.006 |
iquette | +0.006 |
Neg logit contributions | |
� | -0.008 |
Transparency | -0.008 |
Millions | -0.008 |
Reports | -0.008 |
Thousands | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +4.35e-01 |
Pos logit contributions | |
� | +0.038 |
Transparency | +0.037 |
Millions | +0.037 |
Reports | +0.036 |
Thousands | +0.035 |
Neg logit contributions | |
groom | -0.026 |
cha | -0.024 |
hander | -0.024 |
ladies | -0.023 |
iquette | -0.023 |
Token | Chicago |
Token ID | 4842 |
Feature activation | +9.88e-01 |
Pos logit contributions | |
groom | +0.061 |
hander | +0.059 |
cha | +0.059 |
ladies | +0.057 |
iquette | +0.056 |
Neg logit contributions | |
� | -0.049 |
Transparency | -0.047 |
Millions | -0.046 |
Reports | -0.046 |
Thousands | -0.044 |
Token | Ann |
Token ID | 5506 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
groom | +0.115 |
hander | +0.109 |
cha | +0.108 |
ladies | +0.105 |
iquette | +0.103 |
Neg logit contributions | |
� | -0.135 |
Transparency | -0.132 |
Millions | -0.129 |
Reports | -0.128 |
Thousands | -0.124 |
Token | 18 |
Token ID | 1248 |
Feature activation | +3.66e-01 |
Pos logit contributions | |
groom | +0.098 |
hander | +0.094 |
cha | +0.093 |
iquette | +0.091 |
ladies | +0.090 |
Neg logit contributions | |
� | -0.088 |
Transparency | -0.086 |
Millions | -0.084 |
Reports | -0.083 |
Thousands | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
groom | +0.037 |
hander | +0.034 |
cha | +0.034 |
ladies | +0.033 |
iquette | +0.032 |
Neg logit contributions | |
� | -0.056 |
Transparency | -0.055 |
Millions | -0.054 |
Reports | -0.053 |
Thousands | -0.052 |
Token | when |
Token ID | 618 |
Feature activation | +8.76e-01 |
Pos logit contributions | |
groom | +0.027 |
hander | +0.026 |
cha | +0.026 |
ladies | +0.025 |
iquette | +0.025 |
Neg logit contributions | |
� | -0.030 |
Transparency | -0.029 |
Millions | -0.028 |
Reports | -0.028 |
Thousands | -0.027 |
Token | Premier |
Token ID | 9952 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
groom | +0.105 |
hander | +0.100 |
cha | +0.098 |
ladies | +0.096 |
iquette | +0.095 |
Neg logit contributions | |
� | -0.117 |
Transparency | -0.115 |
Millions | -0.112 |
Reports | -0.111 |
Thousands | -0.107 |
Token | League |
Token ID | 4041 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
groom | +0.209 |
hander | +0.201 |
cha | +0.200 |
iquette | +0.195 |
ladies | +0.193 |
Neg logit contributions | |
� | -0.119 |
Transparency | -0.117 |
Millions | -0.113 |
Reports | -0.111 |
Thousands | -0.105 |
Token | chair |
Token ID | 5118 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
� | +0.200 |
Transparency | +0.196 |
Millions | +0.192 |
Reports | +0.189 |
Thousands | +0.184 |
Neg logit contributions | |
groom | -0.157 |
hander | -0.148 |
cha | -0.146 |
ladies | -0.142 |
iquette | -0.140 |
Token | men |
Token ID | 3653 |
Feature activation | +6.40e-02 |
Pos logit contributions | |
groom | +0.537 |
hander | +0.506 |
cha | +0.501 |
ladies | +0.489 |
iquette | +0.484 |
Neg logit contributions | |
� | -0.674 |
Transparency | -0.659 |
Millions | -0.646 |
Reports | -0.638 |
Thousands | -0.622 |
Token | met |
Token ID | 1138 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
groom | +0.007 |
hander | +0.007 |
cha | +0.007 |
ladies | +0.006 |
iquette | +0.006 |
Neg logit contributions | |
� | -0.009 |
Transparency | -0.009 |
Millions | -0.009 |
Reports | -0.009 |
Thousands | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
groom | +0.016 |
hander | +0.015 |
cha | +0.015 |
ladies | +0.014 |
iquette | +0.014 |
Neg logit contributions | |
� | -0.021 |
Transparency | -0.021 |
Millions | -0.020 |
Reports | -0.020 |
Thousands | -0.020 |
Token | discuss |
Token ID | 2112 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
� | +0.070 |
Transparency | +0.070 |
Millions | +0.068 |
Reports | +0.068 |
Thousands | +0.066 |
Neg logit contributions | |
groom | -0.036 |
hander | -0.034 |
cha | -0.033 |
iquette | -0.031 |
ladies | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +3.44e-01 |
Pos logit contributions | |
groom | +0.155 |
hander | +0.146 |
cha | +0.144 |
ladies | +0.139 |
iquette | +0.138 |
Neg logit contributions | |
� | -0.212 |
Transparency | -0.209 |
Millions | -0.204 |
Reports | -0.202 |
Thousands | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | -4.44e-01 |
Pos logit contributions | |
� | +0.221 |
Transparency | +0.217 |
Millions | +0.213 |
Reports | +0.210 |
Thousands | +0.205 |
Neg logit contributions | |
groom | -0.169 |
hander | -0.160 |
cha | -0.157 |
ladies | -0.152 |
iquette | -0.151 |
Token |
|
Token ID | 198 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
� | +0.050 |
Transparency | +0.048 |
Millions | +0.047 |
Reports | +0.046 |
Thousands | +0.045 |
Neg logit contributions | |
groom | -0.062 |
hander | -0.059 |
cha | -0.059 |
ladies | -0.058 |
iquette | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
groom | +0.025 |
hander | +0.024 |
cha | +0.024 |
ladies | +0.023 |
iquette | +0.023 |
Neg logit contributions | |
� | -0.031 |
Transparency | -0.030 |
Millions | -0.029 |
Reports | -0.029 |
Thousands | -0.028 |
Token | Is |
Token ID | 3792 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
groom | +0.026 |
hander | +0.025 |
cha | +0.024 |
ladies | +0.024 |
iquette | +0.024 |
Neg logit contributions | |
� | -0.025 |
Transparency | -0.025 |
Millions | -0.024 |
Reports | -0.024 |
Thousands | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
� | +0.031 |
Transparency | +0.030 |
Millions | +0.030 |
Reports | +0.030 |
Thousands | +0.029 |
Neg logit contributions | |
groom | -0.025 |
hander | -0.024 |
cha | -0.024 |
ladies | -0.023 |
iquette | -0.023 |
Token | Trend |
Token ID | 22836 |
Feature activation | -5.94e+00 |
Pos logit contributions | |
groom | +0.027 |
cha | +0.026 |
hander | +0.026 |
ladies | +0.025 |
gentlemen | +0.025 |
Neg logit contributions | |
� | -0.025 |
Transparency | -0.024 |
Millions | -0.024 |
Reports | -0.024 |
Thousands | -0.023 |
Token | y |
Token ID | 88 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
� | +0.731 |
Transparency | +0.712 |
Millions | +0.694 |
Reports | +0.681 |
Thousands | +0.661 |
Neg logit contributions | |
groom | -0.776 |
hander | -0.742 |
cha | -0.733 |
ladies | -0.713 |
iquette | -0.708 |
Token | Internet |
Token ID | 4455 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
groom | +0.003 |
hander | +0.003 |
cha | +0.003 |
ladies | +0.003 |
iquette | +0.003 |
Neg logit contributions | |
� | -0.002 |
Transparency | -0.002 |
Millions | -0.002 |
Reports | -0.002 |
Thousands | -0.002 |
Token | Drug |
Token ID | 11095 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
� | +0.352 |
Transparency | +0.341 |
Millions | +0.333 |
Reports | +0.327 |
Thousands | +0.317 |
Neg logit contributions | |
groom | -0.389 |
hander | -0.369 |
cha | -0.367 |
ladies | -0.357 |
iquette | -0.355 |
Token | K |
Token ID | 509 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
� | +0.014 |
Transparency | +0.013 |
Millions | +0.013 |
Reports | +0.012 |
Thousands | +0.012 |
Neg logit contributions | |
groom | -0.021 |
hander | -0.020 |
cha | -0.020 |
iquette | -0.020 |
ladies | -0.020 |
Token | ratom |
Token ID | 44616 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
� | +0.033 |
Transparency | +0.032 |
Millions | +0.031 |
Reports | +0.031 |
Thousands | +0.030 |
Neg logit contributions | |
groom | -0.031 |
hander | -0.030 |
cha | -0.029 |
ladies | -0.028 |
iquette | -0.028 |
Token | filmed |
Token ID | 18976 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
� | +0.259 |
Transparency | +0.254 |
Millions | +0.249 |
Reports | +0.246 |
Thousands | +0.239 |
Neg logit contributions | |
groom | -0.185 |
hander | -0.174 |
cha | -0.172 |
ladies | -0.166 |
iquette | -0.164 |
Token | in |
Token ID | 287 |
Feature activation | -5.00e-01 |
Pos logit contributions | |
� | +0.048 |
Transparency | +0.047 |
Millions | +0.046 |
Reports | +0.046 |
Thousands | +0.045 |
Neg logit contributions | |
groom | -0.031 |
hander | -0.029 |
cha | -0.029 |
ladies | -0.028 |
iquette | -0.028 |
Token | a |
Token ID | 257 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
� | +0.067 |
Transparency | +0.065 |
Millions | +0.064 |
Reports | +0.063 |
Thousands | +0.061 |
Neg logit contributions | |
groom | -0.060 |
hander | -0.057 |
cha | -0.056 |
ladies | -0.055 |
gentlemen | -0.054 |
Token | studio |
Token ID | 8034 |
Feature activation | +8.96e-01 |
Pos logit contributions | |
� | +0.063 |
Transparency | +0.062 |
Millions | +0.061 |
Reports | +0.060 |
Thousands | +0.058 |
Neg logit contributions | |
groom | -0.043 |
hander | -0.041 |
cha | -0.040 |
ladies | -0.039 |
iquette | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
groom | +0.091 |
hander | +0.085 |
cha | +0.084 |
ladies | +0.081 |
iquette | +0.080 |
Neg logit contributions | |
� | -0.137 |
Transparency | -0.134 |
Millions | -0.132 |
Reports | -0.130 |
Thousands | -0.127 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.82e+00 |
Pos logit contributions | |
� | +0.075 |
Transparency | +0.072 |
Millions | +0.069 |
Reports | +0.068 |
Trend | +0.065 |
Neg logit contributions | |
groom | -0.105 |
hander | -0.100 |
cha | -0.100 |
ladies | -0.098 |
gentlemen | -0.096 |
Token | courtesy |
Token ID | 12537 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
� | +0.927 |
Transparency | +0.911 |
Millions | +0.894 |
Reports | +0.883 |
Thousands | +0.862 |
Neg logit contributions | |
groom | -0.549 |
hander | -0.510 |
cha | -0.507 |
ladies | -0.487 |
iquette | -0.479 |
Token | of |
Token ID | 286 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
groom | +0.197 |
hander | +0.188 |
cha | +0.187 |
ladies | +0.181 |
gentlemen | +0.178 |
Neg logit contributions | |
� | -0.208 |
Transparency | -0.205 |
Millions | -0.200 |
Reports | -0.195 |
Thousands | -0.190 |
Token | Columbia |
Token ID | 9309 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
� | +0.077 |
Transparency | +0.075 |
Millions | +0.073 |
Reports | +0.071 |
Thousands | +0.069 |
Neg logit contributions | |
groom | -0.102 |
hander | -0.098 |
cha | -0.097 |
ladies | -0.095 |
gentlemen | -0.093 |
Token | Pictures |
Token ID | 20051 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
� | +0.140 |
Transparency | +0.135 |
Millions | +0.129 |
Reports | +0.126 |
Thousands | +0.121 |
Neg logit contributions | |
groom | -0.242 |
cha | -0.233 |
hander | -0.232 |
ladies | -0.227 |
iquette | -0.225 |
Token | The |
Token ID | 383 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
� | +0.278 |
Transparency | +0.270 |
Millions | +0.260 |
Reports | +0.256 |
Thousands | +0.246 |
Neg logit contributions | |
groom | -0.380 |
hander | -0.364 |
cha | -0.361 |
ladies | -0.353 |
iquette | -0.348 |
Token | � |
Token ID | 251 |
Feature activation | +4.93e-01 |
Pos logit contributions | |
groom | +0.017 |
hander | +0.016 |
cha | +0.016 |
ladies | +0.016 |
iquette | +0.016 |
Neg logit contributions | |
� | -0.016 |
Transparency | -0.016 |
Millions | -0.015 |
Reports | -0.015 |
Thousands | -0.015 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
groom | +0.056 |
hander | +0.053 |
cha | +0.053 |
ladies | +0.052 |
iquette | +0.050 |
Neg logit contributions | |
� | -0.068 |
Transparency | -0.066 |
Millions | -0.065 |
Reports | -0.064 |
Thousands | -0.062 |
Token | Image |
Token ID | 5159 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
groom | +0.082 |
hander | +0.079 |
cha | +0.078 |
ladies | +0.076 |
iquette | +0.075 |
Neg logit contributions | |
� | -0.078 |
Transparency | -0.076 |
Millions | -0.074 |
Reports | -0.073 |
Thousands | -0.071 |
Token | copyright |
Token ID | 6634 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
� | +0.532 |
Transparency | +0.519 |
Millions | +0.507 |
Reports | +0.497 |
Thousands | +0.481 |
Neg logit contributions | |
groom | -0.616 |
hander | -0.586 |
cha | -0.582 |
ladies | -0.566 |
iquette | -0.561 |
Token | publicity |
Token ID | 19403 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
� | +0.048 |
Transparency | +0.045 |
Millions | +0.044 |
Reports | +0.043 |
Thousands | +0.041 |
Neg logit contributions | |
groom | -0.084 |
cha | -0.080 |
hander | -0.080 |
ladies | -0.078 |
iquette | -0.077 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.79e+00 |
Pos logit contributions | |
� | +0.262 |
Transparency | +0.256 |
Millions | +0.248 |
Reports | +0.244 |
Thousands | +0.236 |
Neg logit contributions | |
groom | -0.322 |
hander | -0.308 |
cha | -0.305 |
ladies | -0.297 |
iquette | -0.293 |
Token | caption |
Token ID | 8305 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
� | +0.646 |
Transparency | +0.640 |
Millions | +0.615 |
Reports | +0.605 |
Thousands | +0.587 |
Neg logit contributions | |
groom | -0.810 |
hander | -0.785 |
cha | -0.765 |
iquette | -0.762 |
ladies | -0.740 |
Token | The |
Token ID | 383 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
groom | +0.005 |
hander | +0.005 |
cha | +0.005 |
ladies | +0.005 |
iquette | +0.005 |
Neg logit contributions | |
� | -0.004 |
Transparency | -0.004 |
Millions | -0.004 |
Reports | -0.004 |
Thousands | -0.003 |
Token | Greeks |
Token ID | 25059 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
� | +0.156 |
Transparency | +0.152 |
Millions | +0.148 |
Reports | +0.146 |
Thousands | +0.141 |
Neg logit contributions | |
groom | -0.146 |
hander | -0.137 |
cha | -0.137 |
ladies | -0.133 |
gentlemen | -0.131 |
Token | were |
Token ID | 547 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
� | +0.248 |
Transparency | +0.243 |
Millions | +0.238 |
Reports | +0.235 |
Thousands | +0.228 |
Neg logit contributions | |
groom | -0.186 |
hander | -0.175 |
cha | -0.173 |
ladies | -0.167 |
iquette | -0.165 |
Token | totally |
Token ID | 6635 |
Feature activation | -8.27e-01 |
Pos logit contributions | |
� | +0.239 |
Transparency | +0.235 |
Millions | +0.229 |
Reports | +0.227 |
Thousands | +0.220 |
Neg logit contributions | |
groom | -0.193 |
hander | -0.181 |
cha | -0.180 |
ladies | -0.174 |
gentlemen | -0.171 |
Token | warrant |
Token ID | 8084 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
groom | +0.006 |
hander | +0.005 |
cha | +0.005 |
ladies | +0.005 |
iquette | +0.005 |
Neg logit contributions | |
� | -0.011 |
Transparency | -0.011 |
Millions | -0.010 |
Reports | -0.010 |
Thousands | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
groom | +0.135 |
hander | +0.127 |
cha | +0.127 |
ladies | +0.122 |
gentlemen | +0.120 |
Neg logit contributions | |
� | -0.170 |
Transparency | -0.167 |
Millions | -0.164 |
Reports | -0.161 |
Thousands | -0.157 |
Token |
|
Token ID | 198 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
groom | +0.003 |
hander | +0.003 |
cha | +0.003 |
ladies | +0.003 |
iquette | +0.003 |
Neg logit contributions | |
� | -0.005 |
Transparency | -0.004 |
Millions | -0.004 |
Reports | -0.004 |
Trend | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
groom | +0.081 |
hander | +0.076 |
cha | +0.076 |
iquette | +0.075 |
concess | +0.074 |
Neg logit contributions | |
� | -0.079 |
Transparency | -0.079 |
Millions | -0.078 |
Reports | -0.077 |
Thousands | -0.074 |
Token | Wiki |
Token ID | 32603 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
groom | +0.083 |
hander | +0.081 |
cha | +0.079 |
ladies | +0.077 |
iquette | +0.077 |
Neg logit contributions | |
� | -0.080 |
Transparency | -0.078 |
Millions | -0.076 |
Reports | -0.075 |
Thousands | -0.072 |
Token | Leaks |
Token ID | 17874 |
Feature activation | -5.60e+00 |
Pos logit contributions | |
� | +0.049 |
Transparency | +0.045 |
Millions | +0.036 |
Reports | +0.033 |
Thousands | +0.023 |
Neg logit contributions | |
groom | -0.596 |
hander | -0.578 |
cha | -0.577 |
ladies | -0.570 |
iquette | -0.566 |
Token | pr |
Token ID | 778 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
� | +0.843 |
Transparency | +0.826 |
Millions | +0.811 |
Reports | +0.798 |
Thousands | +0.779 |
Neg logit contributions | |
groom | -0.581 |
hander | -0.546 |
cha | -0.540 |
ladies | -0.520 |
iquette | -0.513 |
Token | ides |
Token ID | 1460 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
groom | +0.320 |
hander | +0.305 |
cha | +0.302 |
ladies | +0.294 |
iquette | +0.293 |
Neg logit contributions | |
� | -0.308 |
Transparency | -0.301 |
Millions | -0.293 |
Reports | -0.289 |
Thousands | -0.280 |
Token | itself |
Token ID | 2346 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
� | +0.010 |
Transparency | +0.010 |
Millions | +0.010 |
Reports | +0.010 |
Thousands | +0.009 |
Neg logit contributions | |
groom | -0.008 |
cha | -0.008 |
hander | -0.008 |
ladies | -0.008 |
gentlemen | -0.007 |
Token | as |
Token ID | 355 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
groom | +0.051 |
hander | +0.047 |
cha | +0.047 |
ladies | +0.045 |
gentlemen | +0.044 |
Neg logit contributions | |
� | -0.094 |
Transparency | -0.092 |
Millions | -0.091 |
Reports | -0.090 |
Thousands | -0.088 |
Token | a |
Token ID | 257 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
groom | +0.015 |
cha | +0.014 |
hander | +0.014 |
ladies | +0.014 |
gentlemen | +0.014 |
Neg logit contributions | |
� | -0.020 |
Transparency | -0.019 |
Millions | -0.019 |
Reports | -0.019 |
Thousands | -0.018 |
Token | Not |
Token ID | 1892 |
Feature activation | -3.92e+00 |
Pos logit contributions | |
� | +0.077 |
Transparency | +0.074 |
Millions | +0.071 |
Reports | +0.070 |
Thousands | +0.067 |
Neg logit contributions | |
groom | -0.115 |
hander | -0.109 |
cha | -0.109 |
ladies | -0.107 |
gentlemen | -0.106 |
Token | sure |
Token ID | 1654 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
� | +0.583 |
Transparency | +0.572 |
Millions | +0.560 |
Reports | +0.553 |
Thousands | +0.539 |
Neg logit contributions | |
groom | -0.411 |
hander | -0.386 |
cha | -0.383 |
ladies | -0.370 |
iquette | -0.363 |
Token | . |
Token ID | 13 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
groom | +0.141 |
hander | +0.133 |
cha | +0.132 |
ladies | +0.127 |
gentlemen | +0.125 |
Neg logit contributions | |
� | -0.190 |
Transparency | -0.186 |
Millions | -0.182 |
Reports | -0.180 |
Thousands | -0.175 |
Token | Sketch |
Token ID | 17001 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
� | +0.080 |
Transparency | +0.077 |
Millions | +0.074 |
Reports | +0.072 |
Thousands | +0.070 |
Neg logit contributions | |
groom | -0.111 |
hander | -0.106 |
cha | -0.106 |
ladies | -0.104 |
gentlemen | -0.102 |
Token | in |
Token ID | 287 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
� | +0.067 |
Transparency | +0.066 |
Millions | +0.064 |
Reports | +0.064 |
Thousands | +0.062 |
Neg logit contributions | |
groom | -0.052 |
hander | -0.050 |
cha | -0.049 |
ladies | -0.047 |
iquette | -0.047 |
Token | IM |
Token ID | 8959 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
� | +0.068 |
Transparency | +0.067 |
Millions | +0.065 |
Reports | +0.064 |
Thousands | +0.063 |
Neg logit contributions | |
groom | -0.055 |
hander | -0.052 |
cha | -0.051 |
ladies | -0.050 |
iquette | -0.049 |
Token | – |
Token ID | 1906 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
� | +0.724 |
Transparency | +0.710 |
Millions | +0.693 |
Reports | +0.683 |
Thousands | +0.663 |
Neg logit contributions | |
groom | -0.692 |
hander | -0.659 |
cha | -0.650 |
ladies | -0.632 |
iquette | -0.627 |
Token | for |
Token ID | 329 |
Feature activation | +9.10e-02 |
Pos logit contributions | |
groom | +0.091 |
hander | +0.087 |
cha | +0.086 |
ladies | +0.083 |
iquette | +0.083 |
Neg logit contributions | |
� | -0.082 |
Transparency | -0.080 |
Millions | -0.078 |
Reports | -0.077 |
Thousands | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | -5.10e-02 |
Pos logit contributions | |
groom | +0.011 |
hander | +0.010 |
cha | +0.010 |
ladies | +0.010 |
gentlemen | +0.010 |
Neg logit contributions | |
� | -0.012 |
Transparency | -0.012 |
Millions | -0.012 |
Reports | -0.012 |
Thousands | -0.011 |
Token | sake |
Token ID | 11060 |
Feature activation | +1.29e-03 |
Pos logit contributions | |
� | +0.007 |
Transparency | +0.007 |
Millions | +0.007 |
Reports | +0.007 |
Thousands | +0.006 |
Neg logit contributions | |
groom | -0.006 |
hander | -0.006 |
cha | -0.006 |
ladies | -0.005 |
gentlemen | -0.005 |
Token | of |
Token ID | 286 |
Feature activation | -7.25e-01 |
Pos logit contributions | |
groom | +0.000 |
hander | +0.000 |
cha | +0.000 |
iquette | +0.000 |
ladies | +0.000 |
Neg logit contributions | |
� | -0.000 |
Transparency | -0.000 |
Millions | -0.000 |
Reports | -0.000 |
Thousands | -0.000 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.85e+00 |
Pos logit contributions | |
� | +0.016 |
Transparency | +0.015 |
Millions | +0.015 |
Reports | +0.015 |
Thousands | +0.014 |
Neg logit contributions | |
groom | -0.011 |
hander | -0.011 |
cha | -0.011 |
ladies | -0.010 |
gentlemen | -0.010 |
Token | : |
Token ID | 25 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
� | +0.795 |
Transparency | +0.780 |
Millions | +0.762 |
Reports | +0.752 |
Thousands | +0.730 |
Neg logit contributions | |
groom | -0.689 |
hander | -0.654 |
cha | -0.646 |
ladies | -0.625 |
iquette | -0.621 |
Token | Se |
Token ID | 1001 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
groom | +0.147 |
hander | +0.140 |
cha | +0.139 |
ladies | +0.135 |
iquette | +0.133 |
Neg logit contributions | |
� | -0.124 |
Transparency | -0.120 |
Millions | -0.117 |
Reports | -0.115 |
Thousands | -0.111 |
Token | ill |
Token ID | 359 |
Feature activation | +8.53e-01 |
Pos logit contributions | |
� | +0.102 |
Transparency | +0.100 |
Millions | +0.098 |
Reports | +0.097 |
Thousands | +0.095 |
Neg logit contributions | |
groom | -0.061 |
hander | -0.057 |
cha | -0.056 |
ladies | -0.054 |
iquette | -0.053 |
Token | ons |
Token ID | 684 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
groom | +0.096 |
hander | +0.091 |
cha | +0.090 |
ladies | +0.087 |
iquette | +0.087 |
Neg logit contributions | |
� | -0.120 |
Transparency | -0.118 |
Millions | -0.115 |
Reports | -0.114 |
Thousands | -0.111 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.58e+00 |
Pos logit contributions | |
groom | +0.132 |
hander | +0.126 |
cha | +0.125 |
ladies | +0.121 |
iquette | +0.120 |
Neg logit contributions | |
� | -0.125 |
Transparency | -0.122 |
Millions | -0.119 |
Reports | -0.117 |
Thousands | -0.114 |
Token | : |
Token ID | 25 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
� | +0.865 |
Transparency | +0.849 |
Millions | +0.833 |
Reports | +0.823 |
Thousands | +0.804 |
Neg logit contributions | |
groom | -0.549 |
hander | -0.516 |
cha | -0.509 |
ladies | -0.489 |
iquette | -0.484 |
Token | Wild |
Token ID | 6183 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
groom | +0.087 |
hander | +0.083 |
cha | +0.083 |
ladies | +0.081 |
gentlemen | +0.080 |
Neg logit contributions | |
� | -0.066 |
Transparency | -0.063 |
Millions | -0.062 |
Reports | -0.060 |
Thousands | -0.058 |
Token | fires |
Token ID | 27312 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
� | +0.203 |
Transparency | +0.199 |
Millions | +0.193 |
Reports | +0.190 |
Thousands | +0.184 |
Neg logit contributions | |
groom | -0.188 |
hander | -0.180 |
cha | -0.176 |
ladies | -0.171 |
iquette | -0.170 |
Token | also |
Token ID | 635 |
Feature activation | +6.02e-01 |
Pos logit contributions | |
� | +0.142 |
Transparency | +0.134 |
Millions | +0.125 |
Reports | +0.120 |
Thousands | +0.110 |
Neg logit contributions | |
groom | -0.529 |
hander | -0.511 |
cha | -0.508 |
ladies | -0.499 |
iquette | -0.495 |
Token | raged |
Token ID | 49000 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
groom | +0.069 |
hander | +0.065 |
cha | +0.064 |
ladies | +0.062 |
iquette | +0.061 |
Neg logit contributions | |
� | -0.084 |
Transparency | -0.083 |
Millions | -0.081 |
Reports | -0.079 |
Thousands | -0.077 |
Token | Mediterranean |
Token ID | 19517 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
groom | +0.062 |
hander | +0.058 |
cha | +0.058 |
ladies | +0.056 |
gentlemen | +0.055 |
Neg logit contributions | |
� | -0.065 |
Transparency | -0.063 |
Millions | -0.062 |
Reports | -0.061 |
Thousands | -0.059 |
Token | island |
Token ID | 7022 |
Feature activation | +3.61e-03 |
Pos logit contributions | |
� | +0.137 |
Transparency | +0.134 |
Millions | +0.131 |
Reports | +0.128 |
Thousands | +0.124 |
Neg logit contributions | |
groom | -0.153 |
hander | -0.146 |
cha | -0.145 |
ladies | -0.141 |
gentlemen | -0.139 |
Token | of |
Token ID | 286 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
groom | +0.000 |
hander | +0.000 |
cha | +0.000 |
ladies | +0.000 |
gentlemen | +0.000 |
Neg logit contributions | |
� | -0.000 |
Transparency | -0.000 |
Millions | -0.000 |
Reports | -0.000 |
Thousands | -0.000 |
Token | Cors |
Token ID | 26978 |
Feature activation | +4.40e-02 |
Pos logit contributions | |
� | +0.048 |
Transparency | +0.046 |
Millions | +0.045 |
Reports | +0.044 |
Thousands | +0.043 |
Neg logit contributions | |
groom | -0.054 |
cha | -0.051 |
hander | -0.051 |
ladies | -0.050 |
gentlemen | -0.049 |
Token | ica |
Token ID | 3970 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
groom | +0.005 |
hander | +0.004 |
cha | +0.004 |
ladies | +0.004 |
gentlemen | +0.004 |
Neg logit contributions | |
� | -0.007 |
Transparency | -0.007 |
Millions | -0.006 |
Reports | -0.006 |
Thousands | -0.006 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.55e+00 |
Pos logit contributions | |
groom | +0.179 |
hander | +0.167 |
cha | +0.166 |
ladies | +0.160 |
gentlemen | +0.157 |
Neg logit contributions | |
� | -0.268 |
Transparency | -0.263 |
Millions | -0.256 |
Reports | -0.253 |
Thousands | -0.246 |
Token | : |
Token ID | 25 |
Feature activation | +6.51e-01 |
Pos logit contributions | |
� | +0.809 |
Transparency | +0.807 |
Millions | +0.792 |
Reports | +0.785 |
Thousands | +0.763 |
Neg logit contributions | |
groom | -0.597 |
hander | -0.558 |
cha | -0.552 |
iquette | -0.538 |
ladies | -0.529 |
Token | Flames |
Token ID | 25814 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
groom | +0.098 |
hander | +0.093 |
cha | +0.093 |
ladies | +0.091 |
gentlemen | +0.089 |
Neg logit contributions | |
� | -0.068 |
Transparency | -0.064 |
Millions | -0.062 |
Reports | -0.061 |
Thousands | -0.058 |
Token | threatened |
Token ID | 8556 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
� | +0.221 |
Transparency | +0.218 |
Millions | +0.211 |
Reports | +0.209 |
Thousands | +0.203 |
Neg logit contributions | |
groom | -0.183 |
hander | -0.171 |
cha | -0.171 |
ladies | -0.165 |
iquette | -0.162 |
Token | homes |
Token ID | 5682 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
� | +0.049 |
Transparency | +0.048 |
Millions | +0.047 |
Reports | +0.046 |
Thousands | +0.045 |
Neg logit contributions | |
groom | -0.042 |
cha | -0.040 |
hander | -0.039 |
ladies | -0.038 |
gentlemen | -0.038 |
Token | in |
Token ID | 287 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
� | +0.005 |
Transparency | +0.005 |
Millions | +0.005 |
Reports | +0.005 |
Thousands | +0.005 |
Neg logit contributions | |
groom | -0.003 |
hander | -0.003 |
cha | -0.003 |
ladies | -0.003 |
iquette | -0.003 |
Token | one |
Token ID | 505 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
groom | +0.034 |
hander | +0.032 |
cha | +0.032 |
ladies | +0.031 |
iquette | +0.030 |
Neg logit contributions | |
� | -0.038 |
Transparency | -0.038 |
Millions | -0.037 |
Reports | -0.036 |
Thousands | -0.035 |
Token | thousand |
Token ID | 7319 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
� | +0.113 |
Transparency | +0.111 |
Millions | +0.108 |
Reports | +0.107 |
Thousands | +0.104 |
Neg logit contributions | |
groom | -0.100 |
hander | -0.095 |
cha | -0.094 |
ladies | -0.091 |
iquette | -0.090 |
Token | � |
Token ID | 447 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
� | +0.292 |
Transparency | +0.288 |
Millions | +0.283 |
Reports | +0.280 |
Thousands | +0.274 |
Neg logit contributions | |
groom | -0.135 |
hander | -0.125 |
cha | -0.123 |
ladies | -0.117 |
iquette | -0.115 |
Token | � |
Token ID | 247 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
groom | +0.032 |
hander | +0.030 |
cha | +0.030 |
ladies | +0.030 |
iquette | +0.030 |
Neg logit contributions | |
Transparency | -0.007 |
Millions | -0.006 |
� | -0.006 |
Reports | -0.006 |
Thousands | -0.005 |
Token | ). |
Token ID | 737 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
� | +0.328 |
Transparency | +0.323 |
Millions | +0.318 |
Reports | +0.315 |
Thousands | +0.309 |
Neg logit contributions | |
groom | -0.111 |
hander | -0.101 |
cha | -0.098 |
ladies | -0.092 |
iquette | -0.091 |
Token | IM |
Token ID | 8959 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
groom | +0.263 |
hander | +0.251 |
cha | +0.250 |
ladies | +0.245 |
gentlemen | +0.241 |
Neg logit contributions | |
� | -0.184 |
Transparency | -0.177 |
Millions | -0.171 |
Reports | -0.168 |
Thousands | -0.161 |
Token | M |
Token ID | 44 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
� | +0.892 |
Transparency | +0.878 |
Millions | +0.861 |
Reports | +0.851 |
Thousands | +0.831 |
Neg logit contributions | |
groom | -0.514 |
hander | -0.480 |
cha | -0.473 |
ladies | -0.455 |
iquette | -0.450 |
Token | OR |
Token ID | 1581 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
� | +0.018 |
Transparency | +0.018 |
Millions | +0.017 |
Reports | +0.017 |
Thousands | +0.017 |
Neg logit contributions | |
groom | -0.010 |
hander | -0.010 |
cha | -0.009 |
ladies | -0.009 |
iquette | -0.009 |
Token | AL |
Token ID | 1847 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
� | +0.129 |
Transparency | +0.127 |
Millions | +0.123 |
Reports | +0.121 |
Thousands | +0.118 |
Neg logit contributions | |
groom | -0.141 |
hander | -0.134 |
cha | -0.133 |
ladies | -0.129 |
iquette | -0.128 |
Token | ? |
Token ID | 30 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
� | +0.256 |
Transparency | +0.252 |
Millions | +0.246 |
Reports | +0.243 |
Thousands | +0.236 |
Neg logit contributions | |
groom | -0.197 |
hander | -0.186 |
cha | -0.185 |
ladies | -0.179 |
iquette | -0.176 |
Token | IMP |
Token ID | 30023 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
groom | +0.197 |
cha | +0.188 |
hander | +0.188 |
ladies | +0.184 |
gentlemen | +0.181 |
Neg logit contributions | |
� | -0.130 |
Transparency | -0.125 |
Millions | -0.120 |
Reports | -0.118 |
Thousands | -0.113 |
Token | town |
Token ID | 3240 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
groom | +0.064 |
cha | +0.060 |
hander | +0.060 |
ladies | +0.058 |
gentlemen | +0.057 |
Neg logit contributions | |
� | -0.071 |
Transparency | -0.070 |
Millions | -0.068 |
Reports | -0.067 |
Thousands | -0.065 |
Token | of |
Token ID | 286 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
groom | +0.213 |
hander | +0.201 |
cha | +0.201 |
ladies | +0.195 |
iquette | +0.192 |
Neg logit contributions | |
� | -0.227 |
Transparency | -0.223 |
Millions | -0.217 |
Reports | -0.213 |
Thousands | -0.207 |
Token | Big |
Token ID | 4403 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
� | +0.057 |
Transparency | +0.055 |
Millions | +0.053 |
Reports | +0.052 |
Thousands | +0.051 |
Neg logit contributions | |
groom | -0.067 |
hander | -0.064 |
cha | -0.063 |
ladies | -0.062 |
iquette | -0.061 |
Token | ug |
Token ID | 1018 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
� | +0.234 |
Transparency | +0.230 |
Millions | +0.224 |
Reports | +0.221 |
Thousands | +0.215 |
Neg logit contributions | |
groom | -0.179 |
hander | -0.171 |
cha | -0.168 |
iquette | -0.162 |
ladies | -0.162 |
Token | lia |
Token ID | 24660 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
� | +0.062 |
Transparency | +0.061 |
Millions | +0.060 |
Reports | +0.059 |
Thousands | +0.057 |
Neg logit contributions | |
groom | -0.045 |
hander | -0.042 |
cha | -0.042 |
ladies | -0.040 |
iquette | -0.040 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.53e+00 |
Pos logit contributions | |
groom | +0.129 |
hander | +0.124 |
cha | +0.123 |
ladies | +0.120 |
iquette | +0.118 |
Neg logit contributions | |
� | -0.103 |
Transparency | -0.101 |
Millions | -0.097 |
Reports | -0.095 |
Thousands | -0.092 |
Token | : |
Token ID | 25 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
� | +0.793 |
Transparency | +0.793 |
Millions | +0.779 |
Reports | +0.773 |
Thousands | +0.751 |
Neg logit contributions | |
groom | -0.605 |
hander | -0.565 |
cha | -0.561 |
iquette | -0.548 |
concess | -0.537 |
Token | Around |
Token ID | 16824 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
groom | +0.103 |
hander | +0.098 |
cha | +0.097 |
ladies | +0.096 |
gentlemen | +0.094 |
Neg logit contributions | |
� | -0.064 |
Transparency | -0.060 |
Millions | -0.057 |
Reports | -0.056 |
Pastebin | -0.053 |
Token | 1 |
Token ID | 352 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
� | +0.013 |
Transparency | +0.013 |
Millions | +0.012 |
Reports | +0.012 |
Thousands | +0.012 |
Neg logit contributions | |
groom | -0.015 |
hander | -0.015 |
cha | -0.015 |
ladies | -0.014 |
iquette | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
� | +0.083 |
Transparency | +0.083 |
Millions | +0.080 |
Reports | +0.080 |
Thousands | +0.077 |
Neg logit contributions | |
groom | -0.054 |
hander | -0.050 |
cha | -0.050 |
ladies | -0.048 |
iquette | -0.047 |
Token | 800 |
Token ID | 7410 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
groom | +0.013 |
hander | +0.013 |
cha | +0.012 |
ladies | +0.012 |
iquette | +0.012 |
Neg logit contributions | |
� | -0.012 |
Transparency | -0.012 |
Millions | -0.012 |
Reports | -0.011 |
Thousands | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
� | +0.124 |
Transparency | +0.122 |
Millions | +0.119 |
Reports | +0.118 |
Thousands | +0.115 |
Neg logit contributions | |
groom | -0.077 |
hander | -0.072 |
cha | -0.071 |
ladies | -0.068 |
iquette | -0.067 |
Token | very |
Token ID | 845 |
Feature activation | +6.26e-01 |
Pos logit contributions | |
� | +0.019 |
Transparency | +0.019 |
Millions | +0.019 |
Reports | +0.019 |
Thousands | +0.018 |
Neg logit contributions | |
groom | -0.015 |
hander | -0.014 |
cha | -0.014 |
ladies | -0.013 |
iquette | -0.013 |
Token | small |
Token ID | 1402 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
groom | +0.061 |
hander | +0.058 |
cha | +0.056 |
iquette | +0.054 |
ladies | +0.054 |
Neg logit contributions | |
� | -0.097 |
Transparency | -0.096 |
Millions | -0.094 |
Reports | -0.093 |
Thousands | -0.091 |
Token | lens |
Token ID | 10317 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
groom | +0.056 |
hander | +0.053 |
cha | +0.053 |
ladies | +0.051 |
gentlemen | +0.050 |
Neg logit contributions | |
� | -0.071 |
Transparency | -0.070 |
Millions | -0.068 |
Reports | -0.068 |
Thousands | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -4.02e-01 |
Pos logit contributions | |
� | +0.073 |
Transparency | +0.072 |
Millions | +0.070 |
Reports | +0.069 |
Thousands | +0.067 |
Neg logit contributions | |
groom | -0.059 |
hander | -0.055 |
cha | -0.055 |
ladies | -0.053 |
iquette | -0.052 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
� | +0.049 |
Transparency | +0.047 |
Millions | +0.045 |
Reports | +0.045 |
Thousands | +0.043 |
Neg logit contributions | |
groom | -0.053 |
hander | -0.050 |
cha | -0.050 |
ladies | -0.049 |
iquette | -0.048 |
Token | via |
Token ID | 2884 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
� | +0.772 |
Transparency | +0.757 |
Millions | +0.741 |
Reports | +0.731 |
Thousands | +0.711 |
Neg logit contributions | |
groom | -0.614 |
hander | -0.581 |
cha | -0.573 |
ladies | -0.554 |
iquette | -0.549 |
Token | Pan |
Token ID | 5961 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
groom | +0.181 |
hander | +0.173 |
cha | +0.172 |
ladies | +0.168 |
gentlemen | +0.166 |
Neg logit contributions | |
� | -0.134 |
Transparency | -0.130 |
Millions | -0.126 |
Reports | -0.124 |
Thousands | -0.119 |
Token | or |
Token ID | 273 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
� | +0.148 |
Transparency | +0.145 |
Millions | +0.142 |
Reports | +0.140 |
Thousands | +0.137 |
Neg logit contributions | |
groom | -0.118 |
hander | -0.112 |
cha | -0.111 |
ladies | -0.107 |
iquette | -0.106 |
Token | am |
Token ID | 321 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
groom | +0.220 |
hander | +0.210 |
cha | +0.207 |
ladies | +0.200 |
iquette | +0.198 |
Neg logit contributions | |
� | -0.269 |
Transparency | -0.264 |
Millions | -0.257 |
Reports | -0.254 |
Thousands | -0.248 |
Token | io |
Token ID | 952 |
Feature activation | +9.30e-01 |
Pos logit contributions | |
groom | +0.027 |
hander | +0.025 |
cha | +0.025 |
ladies | +0.024 |
iquette | +0.024 |
Neg logit contributions | |
� | -0.030 |
Transparency | -0.029 |
Millions | -0.028 |
Reports | -0.028 |
Thousands | -0.027 |
Token | ) |
Token ID | 8 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
groom | +0.076 |
hander | +0.070 |
cha | +0.069 |
ladies | +0.065 |
iquette | +0.064 |
Neg logit contributions | |
� | -0.206 |
Transparency | -0.204 |
Millions | -0.200 |
Reports | -0.198 |
Thousands | -0.194 |
Token | Must |
Token ID | 12039 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
groom | +0.305 |
hander | +0.291 |
cha | +0.291 |
ladies | +0.285 |
gentlemen | +0.281 |
Neg logit contributions | |
� | -0.218 |
Transparency | -0.210 |
Millions | -0.203 |
Reports | -0.199 |
Thousands | -0.192 |
Token | be |
Token ID | 307 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
groom | +0.131 |
hander | +0.123 |
cha | +0.123 |
ladies | +0.118 |
iquette | +0.116 |
Neg logit contributions | |
� | -0.168 |
Transparency | -0.165 |
Millions | -0.161 |
Reports | -0.160 |
Thousands | -0.155 |
Token | something |
Token ID | 1223 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
� | +0.204 |
Transparency | +0.199 |
Millions | +0.195 |
Reports | +0.193 |
Thousands | +0.187 |
Neg logit contributions | |
groom | -0.164 |
hander | -0.154 |
cha | -0.153 |
ladies | -0.149 |
gentlemen | -0.146 |
Token | after |
Token ID | 706 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
� | +0.156 |
Transparency | +0.153 |
Millions | +0.150 |
Reports | +0.148 |
Thousands | +0.144 |
Neg logit contributions | |
groom | -0.111 |
hander | -0.105 |
cha | -0.104 |
ladies | -0.100 |
gentlemen | -0.098 |
Token | IM |
Token ID | 8959 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
groom | +0.020 |
hander | +0.018 |
cha | +0.018 |
ladies | +0.018 |
gentlemen | +0.018 |
Neg logit contributions | |
� | -0.021 |
Transparency | -0.021 |
Millions | -0.020 |
Reports | -0.020 |
Thousands | -0.019 |
Token | (= |
Token ID | 46121 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
� | +0.699 |
Transparency | +0.688 |
Millions | +0.669 |
Reports | +0.659 |
Thousands | +0.641 |
Neg logit contributions | |
groom | -0.680 |
hander | -0.647 |
cha | -0.641 |
ladies | -0.623 |
iquette | -0.616 |
Token | � |
Token ID | 564 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
groom | +0.045 |
hander | +0.043 |
cha | +0.042 |
ladies | +0.041 |
iquette | +0.041 |
Neg logit contributions | |
� | -0.049 |
Transparency | -0.048 |
Millions | -0.046 |
Reports | -0.046 |
Thousands | -0.045 |
Token | � |
Token ID | 246 |
Feature activation | +2.84e-01 |
Pos logit contributions | |
� | +0.317 |
Transparency | +0.315 |
Millions | +0.304 |
Reports | +0.299 |
Thousands | +0.285 |
Neg logit contributions | |
groom | -0.530 |
hander | -0.508 |
cha | -0.504 |
ladies | -0.493 |
iquette | -0.490 |
Token | one |
Token ID | 505 |
Feature activation | -8.42e-01 |
Pos logit contributions | |
groom | +0.034 |
hander | +0.032 |
cha | +0.032 |
ladies | +0.031 |
iquette | +0.030 |
Neg logit contributions | |
� | -0.038 |
Transparency | -0.038 |
Millions | -0.037 |
Reports | -0.036 |
Thousands | -0.035 |
Token | thousand |
Token ID | 7319 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
� | +0.113 |
Transparency | +0.111 |
Millions | +0.108 |
Reports | +0.107 |
Thousands | +0.104 |
Neg logit contributions | |
groom | -0.100 |
hander | -0.095 |
cha | -0.094 |
ladies | -0.091 |
iquette | -0.090 |
Token | hang |
Token ID | 8181 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
� | +0.127 |
Transparency | +0.125 |
Millions | +0.123 |
Reports | +0.121 |
Thousands | +0.118 |
Neg logit contributions | |
groom | -0.081 |
hander | -0.077 |
cha | -0.075 |
iquette | -0.073 |
ladies | -0.072 |
Token | out |
Token ID | 503 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
� | +0.223 |
Transparency | +0.218 |
Millions | +0.213 |
Reports | +0.210 |
Thousands | +0.204 |
Neg logit contributions | |
groom | -0.211 |
hander | -0.201 |
cha | -0.198 |
ladies | -0.192 |
iquette | -0.191 |
Token | at |
Token ID | 379 |
Feature activation | +8.04e-01 |
Pos logit contributions | |
� | +0.183 |
Transparency | +0.179 |
Millions | +0.176 |
Reports | +0.174 |
Thousands | +0.169 |
Neg logit contributions | |
groom | -0.126 |
hander | -0.120 |
cha | -0.117 |
iquette | -0.113 |
ladies | -0.112 |
Token | places |
Token ID | 4113 |
Feature activation | -5.56e-01 |
Pos logit contributions | |
groom | +0.089 |
hander | +0.084 |
cha | +0.083 |
ladies | +0.080 |
iquette | +0.079 |
Neg logit contributions | |
� | -0.115 |
Transparency | -0.113 |
Millions | -0.111 |
Reports | -0.110 |
Thousands | -0.106 |
Token | like |
Token ID | 588 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
� | +0.091 |
Transparency | +0.089 |
Millions | +0.087 |
Reports | +0.086 |
Thousands | +0.084 |
Neg logit contributions | |
groom | -0.050 |
hander | -0.047 |
cha | -0.046 |
ladies | -0.045 |
iquette | -0.044 |
Token | Numbers |
Token ID | 27797 |
Feature activation | -5.40e+00 |
Pos logit contributions | |
� | +0.008 |
Transparency | +0.008 |
Millions | +0.007 |
Reports | +0.007 |
Thousands | +0.007 |
Neg logit contributions | |
groom | -0.008 |
hander | -0.007 |
cha | -0.007 |
ladies | -0.007 |
iquette | -0.007 |
Token | as |
Token ID | 355 |
Feature activation | -6.42e-01 |
Pos logit contributions | |
� | +0.714 |
Transparency | +0.700 |
Millions | +0.684 |
Reports | +0.675 |
Thousands | +0.655 |
Neg logit contributions | |
groom | -0.654 |
hander | -0.622 |
cha | -0.614 |
ladies | -0.596 |
iquette | -0.592 |
Token | a |
Token ID | 257 |
Feature activation | -3.90e-01 |
Pos logit contributions | |
� | +0.096 |
Transparency | +0.094 |
Millions | +0.092 |
Reports | +0.091 |
Thousands | +0.089 |
Neg logit contributions | |
groom | -0.067 |
hander | -0.063 |
cha | -0.062 |
ladies | -0.060 |
iquette | -0.060 |
Token | compromise |
Token ID | 13110 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
� | +0.062 |
Transparency | +0.061 |
Millions | +0.060 |
Reports | +0.059 |
Thousands | +0.058 |
Neg logit contributions | |
groom | -0.037 |
hander | -0.035 |
cha | -0.034 |
iquette | -0.033 |
ladies | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
� | +0.131 |
Transparency | +0.129 |
Millions | +0.126 |
Reports | +0.125 |
Thousands | +0.121 |
Neg logit contributions | |
groom | -0.098 |
hander | -0.092 |
cha | -0.091 |
ladies | -0.088 |
iquette | -0.087 |
Token | promise |
Token ID | 6991 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
� | +0.068 |
Transparency | +0.066 |
Millions | +0.065 |
Reports | +0.064 |
Thousands | +0.063 |
Neg logit contributions | |
groom | -0.049 |
hander | -0.046 |
cha | -0.045 |
ladies | -0.044 |
iquette | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
� | +0.027 |
Transparency | +0.025 |
Millions | +0.025 |
Reports | +0.024 |
Pastebin | +0.024 |
Neg logit contributions | |
groom | -0.022 |
hander | -0.021 |
cha | -0.021 |
ladies | -0.020 |
iquette | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
groom | +0.054 |
hander | +0.051 |
cha | +0.051 |
iquette | +0.050 |
concess | +0.049 |
Neg logit contributions | |
� | -0.053 |
Transparency | -0.053 |
Millions | -0.052 |
Reports | -0.051 |
Thousands | -0.050 |
Token | Image |
Token ID | 5159 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
groom | +0.059 |
hander | +0.057 |
cha | +0.056 |
ladies | +0.054 |
iquette | +0.054 |
Neg logit contributions | |
� | -0.052 |
Transparency | -0.051 |
Millions | -0.049 |
Reports | -0.048 |
Thousands | -0.047 |
Token | copyright |
Token ID | 6634 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
� | +0.476 |
Transparency | +0.461 |
Millions | +0.453 |
Reports | +0.441 |
Thousands | +0.427 |
Neg logit contributions | |
groom | -0.545 |
hander | -0.519 |
cha | -0.516 |
ladies | -0.503 |
iquette | -0.498 |
Token | AFP |
Token ID | 8916 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
� | +0.004 |
Transparency | +0.004 |
Millions | +0.004 |
Reports | +0.004 |
Thousands | +0.003 |
Neg logit contributions | |
groom | -0.009 |
cha | -0.009 |
hander | -0.009 |
ladies | -0.009 |
gentlemen | -0.009 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.33e+00 |
Pos logit contributions | |
� | +0.111 |
Transparency | +0.104 |
Millions | +0.094 |
Reports | +0.086 |
Trend | +0.076 |
Neg logit contributions | |
groom | -0.594 |
hander | -0.574 |
cha | -0.574 |
ladies | -0.566 |
iquette | -0.560 |
Token | caption |
Token ID | 8305 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
� | +0.611 |
Transparency | +0.607 |
Millions | +0.586 |
Reports | +0.579 |
Thousands | +0.560 |
Neg logit contributions | |
groom | -0.735 |
hander | -0.704 |
cha | -0.693 |
iquette | -0.683 |
ladies | -0.669 |
Token | The |
Token ID | 383 |
Feature activation | -8.33e-01 |
Pos logit contributions | |
groom | +0.074 |
hander | +0.071 |
cha | +0.070 |
ladies | +0.070 |
gentlemen | +0.068 |
Neg logit contributions | |
� | -0.047 |
Transparency | -0.043 |
Millions | -0.041 |
Reports | -0.040 |
Trend | -0.039 |
Token | Kurds |
Token ID | 23880 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
� | +0.089 |
Transparency | +0.085 |
Millions | +0.082 |
Reports | +0.081 |
Thousands | +0.078 |
Neg logit contributions | |
groom | -0.125 |
cha | -0.117 |
hander | -0.117 |
ladies | -0.116 |
gentlemen | -0.114 |
Token | ' |
Token ID | 6 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
� | +0.275 |
Transparency | +0.270 |
Millions | +0.264 |
Reports | +0.261 |
Thousands | +0.254 |
Neg logit contributions | |
groom | -0.187 |
hander | -0.175 |
cha | -0.173 |
ladies | -0.167 |
iquette | -0.165 |
Token | push |
Token ID | 4574 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
groom | +0.030 |
cha | +0.028 |
hander | +0.028 |
ladies | +0.027 |
gentlemen | +0.027 |
Neg logit contributions | |
� | -0.030 |
Transparency | -0.029 |
Millions | -0.028 |
Reports | -0.028 |
Thousands | -0.027 |
Token | erville |
Token ID | 33487 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
� | +0.165 |
Transparency | +0.162 |
Millions | +0.159 |
Reports | +0.157 |
Thousands | +0.154 |
Neg logit contributions | |
groom | -0.077 |
hander | -0.072 |
cha | -0.070 |
ladies | -0.066 |
iquette | -0.066 |
Token | 's |
Token ID | 338 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
groom | +0.360 |
hander | +0.338 |
cha | +0.336 |
ladies | +0.325 |
iquette | +0.321 |
Neg logit contributions | |
� | -0.450 |
Transparency | -0.443 |
Millions | -0.431 |
Reports | -0.425 |
Thousands | -0.414 |
Token | Union |
Token ID | 4479 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
groom | +0.011 |
hander | +0.010 |
cha | +0.010 |
ladies | +0.010 |
iquette | +0.010 |
Neg logit contributions | |
� | -0.010 |
Transparency | -0.009 |
Millions | -0.009 |
Reports | -0.009 |
Thousands | -0.009 |
Token | Square |
Token ID | 9276 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
� | +0.123 |
Transparency | +0.120 |
Millions | +0.117 |
Reports | +0.115 |
Thousands | +0.111 |
Neg logit contributions | |
groom | -0.144 |
hander | -0.138 |
cha | -0.137 |
ladies | -0.133 |
iquette | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
� | +0.044 |
Transparency | +0.043 |
Millions | +0.042 |
Reports | +0.042 |
Thousands | +0.041 |
Neg logit contributions | |
groom | -0.035 |
hander | -0.033 |
cha | -0.033 |
ladies | -0.031 |
iquette | -0.031 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.32e+00 |
Pos logit contributions | |
� | +0.035 |
Transparency | +0.033 |
Millions | +0.032 |
Reports | +0.032 |
Thousands | +0.031 |
Neg logit contributions | |
groom | -0.040 |
hander | -0.038 |
cha | -0.037 |
ladies | -0.037 |
gentlemen | -0.036 |
Token | by |
Token ID | 416 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
� | +0.839 |
Transparency | +0.824 |
Millions | +0.809 |
Reports | +0.798 |
Thousands | +0.779 |
Neg logit contributions | |
groom | -0.512 |
hander | -0.476 |
cha | -0.473 |
ladies | -0.455 |
iquette | -0.448 |
Token | Emmanuel |
Token ID | 32390 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
� | +0.020 |
Transparency | +0.019 |
Millions | +0.019 |
Reports | +0.018 |
Thousands | +0.018 |
Neg logit contributions | |
groom | -0.022 |
hander | -0.021 |
cha | -0.021 |
ladies | -0.020 |
iquette | -0.020 |
Token | College |
Token ID | 5535 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
groom | +0.135 |
hander | +0.127 |
cha | +0.126 |
ladies | +0.122 |
iquette | +0.121 |
Neg logit contributions | |
� | -0.167 |
Transparency | -0.164 |
Millions | -0.160 |
Reports | -0.158 |
Thousands | -0.154 |
Token | The |
Token ID | 383 |
Feature activation | -7.35e-01 |
Pos logit contributions | |
groom | +0.256 |
hander | +0.243 |
cha | +0.241 |
ladies | +0.235 |
gentlemen | +0.231 |
Neg logit contributions | |
� | -0.244 |
Transparency | -0.237 |
Millions | -0.230 |
Reports | -0.227 |
Thousands | -0.219 |
Token | main |
Token ID | 1388 |
Feature activation | +9.28e-01 |
Pos logit contributions | |
� | +0.116 |
Transparency | +0.114 |
Millions | +0.111 |
Reports | +0.110 |
Thousands | +0.107 |
Neg logit contributions | |
groom | -0.071 |
hander | -0.066 |
cha | -0.066 |
ladies | -0.063 |
iquette | -0.062 |
Token | agent |
Token ID | 5797 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
groom | +0.075 |
hander | +0.068 |
cha | +0.068 |
ladies | +0.064 |
iquette | +0.062 |
Neg logit contributions | |
� | -0.172 |
Transparency | -0.170 |
Millions | -0.167 |
Reports | -0.165 |
Thousands | -0.162 |
Token | Bruno |
Token ID | 31045 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
groom | +0.212 |
hander | +0.200 |
cha | +0.198 |
ladies | +0.191 |
iquette | +0.189 |
Neg logit contributions | |
� | -0.260 |
Transparency | -0.255 |
Millions | -0.249 |
Reports | -0.246 |
Thousands | -0.239 |
Token | Sat |
Token ID | 7031 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
groom | +0.162 |
hander | +0.154 |
cha | +0.153 |
ladies | +0.149 |
iquette | +0.147 |
Neg logit contributions | |
� | -0.142 |
Transparency | -0.139 |
Millions | -0.135 |
Reports | -0.133 |
Thousands | -0.129 |
Token | in |
Token ID | 259 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
groom | +0.105 |
hander | +0.098 |
cha | +0.097 |
ladies | +0.093 |
iquette | +0.092 |
Neg logit contributions | |
� | -0.165 |
Transparency | -0.162 |
Millions | -0.159 |
Reports | -0.157 |
Thousands | -0.153 |
Token | at |
Token ID | 379 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
groom | +0.050 |
hander | +0.048 |
cha | +0.047 |
ladies | +0.045 |
iquette | +0.045 |
Neg logit contributions | |
� | -0.066 |
Transparency | -0.065 |
Millions | -0.063 |
Reports | -0.062 |
Thousands | -0.061 |
Token | IM |
Token ID | 8959 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
groom | +0.123 |
hander | +0.116 |
cha | +0.115 |
ladies | +0.111 |
iquette | +0.110 |
Neg logit contributions | |
� | -0.149 |
Transparency | -0.146 |
Millions | -0.143 |
Reports | -0.141 |
Thousands | -0.138 |
Token | G |
Token ID | 38 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
� | +0.610 |
Transparency | +0.596 |
Millions | +0.580 |
Reports | +0.571 |
Thousands | +0.552 |
Neg logit contributions | |
groom | -0.727 |
hander | -0.695 |
cha | -0.688 |
ladies | -0.670 |
iquette | -0.665 |
Token | handled |
Token ID | 12118 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
� | +0.061 |
Transparency | +0.059 |
Millions | +0.058 |
Reports | +0.057 |
Thousands | +0.056 |
Neg logit contributions | |
groom | -0.059 |
cha | -0.056 |
hander | -0.056 |
ladies | -0.054 |
gentlemen | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
groom | +0.024 |
hander | +0.022 |
cha | +0.022 |
ladies | +0.021 |
iquette | +0.021 |
Neg logit contributions | |
� | -0.034 |
Transparency | -0.033 |
Millions | -0.032 |
Reports | -0.032 |
Thousands | -0.031 |
Token | deal |
Token ID | 1730 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
groom | +0.030 |
hander | +0.029 |
cha | +0.028 |
ladies | +0.028 |
iquette | +0.027 |
Neg logit contributions | |
� | -0.037 |
Transparency | -0.036 |
Millions | -0.036 |
Reports | -0.035 |
Thousands | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
groom | +0.013 |
hander | +0.013 |
cha | +0.012 |
ladies | +0.012 |
iquette | +0.012 |
Neg logit contributions | |
� | -0.017 |
Transparency | -0.017 |
Millions | -0.017 |
Reports | -0.017 |
Thousands | -0.016 |
Token | til |
Token ID | 47163 |
Feature activation | +4.66e-01 |
Pos logit contributions | |
� | +0.209 |
Transparency | +0.206 |
Millions | +0.201 |
Reports | +0.199 |
Thousands | +0.193 |
Neg logit contributions | |
groom | -0.150 |
hander | -0.142 |
cha | -0.140 |
ladies | -0.135 |
iquette | -0.134 |
Token | I |
Token ID | 314 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
groom | +0.059 |
hander | +0.056 |
cha | +0.055 |
ladies | +0.054 |
iquette | +0.053 |
Neg logit contributions | |
� | -0.059 |
Transparency | -0.058 |
Millions | -0.057 |
Reports | -0.056 |
Thousands | -0.054 |
Token | get |
Token ID | 651 |
Feature activation | -8.65e-01 |
Pos logit contributions | |
� | +0.256 |
Transparency | +0.252 |
Millions | +0.246 |
Reports | +0.244 |
Thousands | +0.237 |
Neg logit contributions | |
groom | -0.175 |
hander | -0.165 |
cha | -0.162 |
ladies | -0.156 |
iquette | -0.156 |
Token | my |
Token ID | 616 |
Feature activation | +5.86e-01 |
Pos logit contributions | |
� | +0.128 |
Transparency | +0.126 |
Millions | +0.123 |
Reports | +0.122 |
Thousands | +0.119 |
Neg logit contributions | |
groom | -0.091 |
hander | -0.086 |
cha | -0.085 |
ladies | -0.082 |
iquette | -0.081 |
Token | here |
Token ID | 994 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
groom | +0.064 |
hander | +0.060 |
cha | +0.060 |
ladies | +0.058 |
iquette | +0.057 |
Neg logit contributions | |
� | -0.085 |
Transparency | -0.083 |
Millions | -0.081 |
Reports | -0.081 |
Thousands | -0.078 |
Token | IM |
Token ID | 8959 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
groom | +0.139 |
hander | +0.131 |
cha | +0.130 |
ladies | +0.125 |
iquette | +0.124 |
Neg logit contributions | |
� | -0.188 |
Transparency | -0.185 |
Millions | -0.181 |
Reports | -0.179 |
Thousands | -0.174 |
Token | M |
Token ID | 44 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
� | +0.578 |
Transparency | +0.563 |
Millions | +0.548 |
Reports | +0.539 |
Thousands | +0.519 |
Neg logit contributions | |
groom | -0.759 |
hander | -0.727 |
cha | -0.718 |
ladies | -0.701 |
iquette | -0.698 |
Token | IG |
Token ID | 3528 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
groom | +0.019 |
hander | +0.018 |
cha | +0.018 |
ladies | +0.018 |
iquette | +0.018 |
Neg logit contributions | |
� | -0.013 |
Transparency | -0.013 |
Millions | -0.013 |
Reports | -0.013 |
Thousands | -0.012 |
Token | RANT |
Token ID | 32506 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
� | +0.208 |
Transparency | +0.205 |
Millions | +0.198 |
Reports | +0.194 |
Thousands | +0.187 |
Neg logit contributions | |
groom | -0.327 |
hander | -0.312 |
cha | -0.311 |
ladies | -0.304 |
gentlemen | -0.300 |
Token | food |
Token ID | 2057 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
� | +0.203 |
Transparency | +0.199 |
Millions | +0.194 |
Reports | +0.191 |
Thousands | +0.186 |
Neg logit contributions | |
groom | -0.206 |
hander | -0.196 |
cha | -0.195 |
ladies | -0.189 |
iquette | -0.187 |
Token | from |
Token ID | 422 |
Feature activation | -7.97e-01 |
Pos logit contributions | |
groom | +0.032 |
hander | +0.030 |
cha | +0.029 |
ladies | +0.028 |
iquette | +0.028 |
Neg logit contributions | |
� | -0.041 |
Transparency | -0.041 |
Millions | -0.040 |
Reports | -0.039 |
Thousands | -0.038 |
Token | in |
Token ID | 287 |
Feature activation | -3.57e-01 |
Pos logit contributions | |
groom | +0.097 |
cha | +0.091 |
hander | +0.091 |
ladies | +0.088 |
gentlemen | +0.087 |
Neg logit contributions | |
� | -0.104 |
Transparency | -0.102 |
Millions | -0.100 |
Reports | -0.098 |
Thousands | -0.095 |
Token | the |
Token ID | 262 |
Feature activation | +4.79e-02 |
Pos logit contributions | |
� | +0.056 |
Transparency | +0.055 |
Millions | +0.054 |
Reports | +0.053 |
Thousands | +0.052 |
Neg logit contributions | |
groom | -0.035 |
hander | -0.033 |
cha | -0.032 |
ladies | -0.031 |
iquette | -0.031 |
Token | CG |
Token ID | 29925 |
Feature activation | -4.10e-01 |
Pos logit contributions | |
groom | +0.005 |
hander | +0.005 |
cha | +0.005 |
ladies | +0.005 |
iquette | +0.005 |
Neg logit contributions | |
� | -0.007 |
Transparency | -0.007 |
Millions | -0.007 |
Reports | -0.007 |
Thousands | -0.006 |
Token | PR |
Token ID | 4805 |
Feature activation | -6.27e-01 |
Pos logit contributions | |
� | +0.053 |
Transparency | +0.051 |
Millions | +0.051 |
Reports | +0.050 |
Thousands | +0.048 |
Neg logit contributions | |
groom | -0.051 |
hander | -0.048 |
cha | -0.048 |
ladies | -0.047 |
iquette | -0.046 |
Token | OG |
Token ID | 7730 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
� | +0.078 |
Transparency | +0.077 |
Millions | +0.076 |
Reports | +0.074 |
Thousands | +0.072 |
Neg logit contributions | |
groom | -0.080 |
hander | -0.076 |
cha | -0.075 |
ladies | -0.073 |
iquette | -0.073 |
Token | RAM |
Token ID | 24115 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
� | +0.133 |
Transparency | +0.130 |
Millions | +0.126 |
Reports | +0.123 |
Thousands | +0.118 |
Neg logit contributions | |
groom | -0.232 |
hander | -0.222 |
cha | -0.221 |
ladies | -0.216 |
iquette | -0.214 |
Token | / |
Token ID | 14 |
Feature activation | +8.51e-01 |
Pos logit contributions | |
� | +0.761 |
Transparency | +0.748 |
Millions | +0.733 |
Reports | +0.723 |
Thousands | +0.705 |
Neg logit contributions | |
groom | -0.572 |
hander | -0.540 |
cha | -0.534 |
ladies | -0.515 |
iquette | -0.510 |
Token | END |
Token ID | 23578 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
groom | +0.105 |
hander | +0.100 |
cha | +0.100 |
ladies | +0.096 |
iquette | +0.095 |
Neg logit contributions | |
� | -0.110 |
Transparency | -0.107 |
Millions | -0.106 |
Reports | -0.104 |
Thousands | -0.101 |
Token | CG |
Token ID | 39816 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
� | +0.212 |
Transparency | +0.208 |
Millions | +0.204 |
Reports | +0.201 |
Thousands | +0.194 |
Neg logit contributions | |
groom | -0.214 |
hander | -0.203 |
cha | -0.202 |
ladies | -0.196 |
gentlemen | -0.193 |
Token | section |
Token ID | 2665 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
� | +0.026 |
Transparency | +0.025 |
Millions | +0.025 |
Reports | +0.024 |
Thousands | +0.024 |
Neg logit contributions | |
groom | -0.022 |
hander | -0.021 |
cha | -0.020 |
ladies | -0.020 |
iquette | -0.020 |
Token | ! |
Token ID | 0 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
groom | +0.177 |
hander | +0.165 |
cha | +0.162 |
ladies | +0.155 |
iquette | +0.154 |
Neg logit contributions | |
� | -0.313 |
Transparency | -0.308 |
Millions | -0.302 |
Reports | -0.299 |
Thousands | -0.291 |
Token | Opt |
Token ID | 13123 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Transparency | +0.103 |
� | +0.102 |
Millions | +0.099 |
Reports | +0.097 |
Thousands | +0.093 |
Neg logit contributions | |
groom | -0.163 |
hander | -0.157 |
cha | -0.156 |
iquette | -0.153 |
ladies | -0.152 |
Token | ronics |
Token ID | 20844 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
Transparency | +0.222 |
� | +0.221 |
Millions | +0.217 |
Reports | +0.216 |
Thousands | +0.207 |
Neg logit contributions | |
groom | -0.159 |
hander | -0.147 |
cha | -0.146 |
iquette | -0.140 |
ladies | -0.139 |
Token | AH |
Token ID | 28159 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
groom | +0.058 |
cha | +0.056 |
hander | +0.055 |
ladies | +0.054 |
iquette | +0.053 |
Neg logit contributions | |
� | -0.050 |
Transparency | -0.048 |
Millions | -0.048 |
Reports | -0.047 |
Thousands | -0.045 |
Token | VA |
Token ID | 11731 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
groom | +0.085 |
hander | +0.080 |
cha | +0.079 |
iquette | +0.078 |
ladies | +0.077 |
Neg logit contributions | |
� | -0.053 |
Transparency | -0.052 |
Millions | -0.050 |
Reports | -0.049 |
Thousands | -0.047 |
Token | ( |
Token ID | 357 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Transparency | +0.138 |
� | +0.137 |
Millions | +0.134 |
Reports | +0.132 |
Thousands | +0.127 |
Neg logit contributions | |
groom | -0.166 |
hander | -0.157 |
cha | -0.155 |
iquette | -0.152 |
ladies | -0.151 |
Token | IPS |
Token ID | 47643 |
Feature activation | -5.25e+00 |
Pos logit contributions | |
� | +0.043 |
Transparency | +0.039 |
Millions | +0.036 |
Reports | +0.033 |
Thousands | +0.029 |
Neg logit contributions | |
groom | -0.288 |
hander | -0.281 |
cha | -0.278 |
ladies | -0.274 |
iquette | -0.273 |
Token | - |
Token ID | 12 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
Transparency | +0.709 |
� | +0.700 |
Millions | +0.688 |
Reports | +0.682 |
Thousands | +0.661 |
Neg logit contributions | |
groom | -0.625 |
hander | -0.576 |
cha | -0.575 |
iquette | -0.566 |
concess | -0.554 |
Token | type |
Token ID | 4906 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
� | +0.045 |
Transparency | +0.043 |
Millions | +0.043 |
Reports | +0.042 |
Thousands | +0.041 |
Neg logit contributions | |
groom | -0.049 |
cha | -0.046 |
hander | -0.044 |
ladies | -0.044 |
iquette | -0.044 |
Token | ) |
Token ID | 8 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
Transparency | +0.227 |
� | +0.227 |
Millions | +0.224 |
Reports | +0.222 |
Thousands | +0.217 |
Neg logit contributions | |
groom | -0.094 |
hander | -0.084 |
cha | -0.083 |
concess | -0.081 |
iquette | -0.081 |
Token | @ |
Token ID | 2488 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
groom | +0.259 |
hander | +0.246 |
cha | +0.243 |
iquette | +0.238 |
ladies | +0.236 |
Neg logit contributions | |
� | -0.220 |
Transparency | -0.219 |
Millions | -0.213 |
Reports | -0.210 |
Thousands | -0.203 |
Token | 144 |
Token ID | 20224 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
� | +0.019 |
Transparency | +0.018 |
Millions | +0.018 |
Reports | +0.018 |
Thousands | +0.017 |
Neg logit contributions | |
groom | -0.020 |
hander | -0.019 |
cha | -0.019 |
ladies | -0.019 |
iquette | -0.018 |
Token | than |
Token ID | 621 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
� | +0.026 |
Transparency | +0.026 |
Millions | +0.025 |
Reports | +0.025 |
Thousands | +0.024 |
Neg logit contributions | |
groom | -0.021 |
hander | -0.020 |
cha | -0.020 |
ladies | -0.019 |
iquette | -0.019 |
Token | an |
Token ID | 281 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
� | +0.003 |
Transparency | +0.003 |
Millions | +0.003 |
Reports | +0.003 |
Thousands | +0.003 |
Neg logit contributions | |
groom | -0.002 |
hander | -0.002 |
cha | -0.002 |
ladies | -0.002 |
iquette | -0.002 |
Token | alien |
Token ID | 8756 |
Feature activation | -6.36e-01 |
Pos logit contributions | |
� | +0.056 |
Transparency | +0.055 |
Millions | +0.053 |
Reports | +0.053 |
Thousands | +0.051 |
Neg logit contributions | |
groom | -0.043 |
hander | -0.041 |
cha | -0.041 |
ladies | -0.039 |
iquette | -0.039 |
Token | egg |
Token ID | 5935 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
� | +0.091 |
Transparency | +0.089 |
Millions | +0.087 |
Reports | +0.086 |
Thousands | +0.084 |
Neg logit contributions | |
groom | -0.071 |
hander | -0.067 |
cha | -0.066 |
ladies | -0.064 |
iquette | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
groom | +0.004 |
hander | +0.004 |
cha | +0.004 |
ladies | +0.004 |
iquette | +0.004 |
Neg logit contributions | |
� | -0.005 |
Transparency | -0.005 |
Millions | -0.005 |
Reports | -0.005 |
Thousands | -0.005 |
Token | Image |
Token ID | 7412 |
Feature activation | -5.22e+00 |
Pos logit contributions | |
� | +0.018 |
Transparency | +0.018 |
Millions | +0.017 |
Reports | +0.017 |
Thousands | +0.016 |
Neg logit contributions | |
groom | -0.021 |
hander | -0.020 |
cha | -0.020 |
ladies | -0.020 |
iquette | -0.019 |
Token | courtesy |
Token ID | 12537 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
� | +0.753 |
Transparency | +0.739 |
Millions | +0.723 |
Reports | +0.713 |
Thousands | +0.694 |
Neg logit contributions | |
groom | -0.572 |
hander | -0.536 |
cha | -0.534 |
ladies | -0.516 |
iquette | -0.509 |
Token | of |
Token ID | 286 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
groom | +0.274 |
hander | +0.260 |
cha | +0.258 |
ladies | +0.252 |
gentlemen | +0.247 |
Neg logit contributions | |
� | -0.275 |
Transparency | -0.271 |
Millions | -0.263 |
Reports | -0.259 |
Thousands | -0.251 |
Token | Columbia |
Token ID | 9309 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
� | +0.017 |
Transparency | +0.017 |
Millions | +0.017 |
Reports | +0.016 |
Thousands | +0.016 |
Neg logit contributions | |
groom | -0.019 |
hander | -0.018 |
cha | -0.018 |
ladies | -0.018 |
iquette | -0.017 |
Token | Pictures |
Token ID | 20051 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
� | +0.096 |
Transparency | +0.093 |
Millions | +0.090 |
Reports | +0.088 |
Thousands | +0.085 |
Neg logit contributions | |
groom | -0.156 |
hander | -0.150 |
cha | -0.149 |
ladies | -0.146 |
iquette | -0.144 |
Token | With |
Token ID | 2080 |
Feature activation | -4.37e-02 |
Pos logit contributions | |
� | +0.296 |
Transparency | +0.290 |
Millions | +0.282 |
Reports | +0.278 |
Thousands | +0.271 |
Neg logit contributions | |
groom | -0.245 |
hander | -0.232 |
cha | -0.231 |
ladies | -0.223 |
gentlemen | -0.219 |
Token | eder |
Token ID | 5702 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
� | +0.048 |
Transparency | +0.047 |
Millions | +0.047 |
Reports | +0.046 |
Thousands | +0.045 |
Neg logit contributions | |
groom | -0.029 |
cha | -0.028 |
hander | -0.027 |
ladies | -0.026 |
gentlemen | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
groom | +0.235 |
hander | +0.222 |
cha | +0.221 |
ladies | +0.213 |
iquette | +0.210 |
Neg logit contributions | |
� | -0.266 |
Transparency | -0.262 |
Millions | -0.256 |
Reports | -0.252 |
Thousands | -0.246 |
Token | have |
Token ID | 423 |
Feature activation | -6.33e-01 |
Pos logit contributions | |
� | +0.103 |
Transparency | +0.102 |
Millions | +0.100 |
Reports | +0.099 |
Thousands | +0.096 |
Neg logit contributions | |
groom | -0.062 |
hander | -0.059 |
cha | -0.057 |
ladies | -0.055 |
iquette | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
� | +0.093 |
Transparency | +0.091 |
Millions | +0.089 |
Reports | +0.088 |
Thousands | +0.086 |
Neg logit contributions | |
groom | -0.068 |
hander | -0.064 |
cha | -0.064 |
ladies | -0.061 |
iquette | -0.061 |
Token | SC |
Token ID | 6374 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
� | +0.030 |
Transparency | +0.030 |
Millions | +0.029 |
Reports | +0.029 |
Thousands | +0.028 |
Neg logit contributions | |
groom | -0.024 |
hander | -0.022 |
cha | -0.022 |
ladies | -0.021 |
iquette | -0.021 |
Token | RAM |
Token ID | 24115 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
� | +0.036 |
Transparency | +0.036 |
Millions | +0.035 |
Reports | +0.034 |
Thousands | +0.033 |
Neg logit contributions | |
groom | -0.045 |
hander | -0.043 |
cha | -0.042 |
ladies | -0.041 |
iquette | -0.041 |
Token | / |
Token ID | 14 |
Feature activation | +9.31e-01 |
Pos logit contributions | |
� | +0.751 |
Transparency | +0.737 |
Millions | +0.721 |
Reports | +0.712 |
Thousands | +0.693 |
Neg logit contributions | |
groom | -0.572 |
hander | -0.541 |
cha | -0.534 |
ladies | -0.515 |
iquette | -0.511 |
Token | alcohol |
Token ID | 5548 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
groom | +0.120 |
hander | +0.115 |
cha | +0.114 |
ladies | +0.110 |
iquette | +0.110 |
Neg logit contributions | |
� | -0.115 |
Transparency | -0.113 |
Millions | -0.110 |
Reports | -0.108 |
Thousands | -0.105 |
Token | monitoring |
Token ID | 9904 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
groom | +0.066 |
hander | +0.062 |
cha | +0.062 |
ladies | +0.060 |
iquette | +0.059 |
Neg logit contributions | |
� | -0.092 |
Transparency | -0.090 |
Millions | -0.089 |
Reports | -0.088 |
Thousands | -0.085 |
Token | device |
Token ID | 3335 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
� | +0.212 |
Transparency | +0.210 |
Millions | +0.205 |
Reports | +0.203 |
Thousands | +0.197 |
Neg logit contributions | |
groom | -0.145 |
hander | -0.139 |
cha | -0.134 |
iquette | -0.131 |
ladies | -0.130 |
Token | and |
Token ID | 290 |
Feature activation | -1.87e-01 |
Pos logit contributions | |
groom | +0.082 |
hander | +0.078 |
cha | +0.076 |
ladies | +0.074 |
iquette | +0.073 |
Neg logit contributions | |
� | -0.111 |
Transparency | -0.109 |
Millions | -0.107 |
Reports | -0.105 |
Thousands | -0.103 |
Token | on |
Token ID | 319 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
� | +0.036 |
Transparency | +0.035 |
Millions | +0.035 |
Reports | +0.034 |
Thousands | +0.033 |
Neg logit contributions | |
groom | -0.027 |
hander | -0.025 |
cha | -0.025 |
ladies | -0.024 |
iquette | -0.024 |
Token | several |
Token ID | 1811 |
Feature activation | +5.77e-01 |
Pos logit contributions | |
groom | +0.111 |
hander | +0.105 |
cha | +0.104 |
ladies | +0.100 |
iquette | +0.099 |
Neg logit contributions | |
� | -0.140 |
Transparency | -0.136 |
Millions | -0.134 |
Reports | -0.132 |
Thousands | -0.128 |
Token | fronts |
Token ID | 29324 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
groom | +0.068 |
hander | +0.064 |
cha | +0.064 |
ladies | +0.062 |
gentlemen | +0.061 |
Neg logit contributions | |
� | -0.078 |
Transparency | -0.077 |
Millions | -0.075 |
Reports | -0.074 |
Thousands | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
� | +0.218 |
Transparency | +0.214 |
Millions | +0.209 |
Reports | +0.207 |
Thousands | +0.201 |
Neg logit contributions | |
groom | -0.161 |
hander | -0.152 |
cha | -0.150 |
ladies | -0.145 |
iquette | -0.143 |
Token | The |
Token ID | 383 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
� | +0.027 |
Transparency | +0.026 |
Millions | +0.026 |
Reports | +0.025 |
Thousands | +0.024 |
Neg logit contributions | |
groom | -0.046 |
hander | -0.044 |
cha | -0.044 |
ladies | -0.043 |
iquette | -0.042 |
Token | pian |
Token ID | 43923 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
� | +0.114 |
Transparency | +0.112 |
Millions | +0.110 |
Reports | +0.108 |
Thousands | +0.106 |
Neg logit contributions | |
groom | -0.072 |
hander | -0.068 |
cha | -0.066 |
ladies | -0.064 |
iquette | -0.063 |
Token | ist |
Token ID | 396 |
Feature activation | -8.31e-01 |
Pos logit contributions | |
groom | +0.472 |
cha | +0.447 |
hander | +0.442 |
iquette | +0.429 |
ladies | +0.428 |
Neg logit contributions | |
� | -0.443 |
Transparency | -0.429 |
Millions | -0.423 |
Reports | -0.417 |
Thousands | -0.403 |
Token | found |
Token ID | 1043 |
Feature activation | -6.22e-01 |
Pos logit contributions | |
� | +0.132 |
Transparency | +0.130 |
Millions | +0.128 |
Reports | +0.126 |
Thousands | +0.123 |
Neg logit contributions | |
groom | -0.078 |
hander | -0.074 |
cha | -0.072 |
ladies | -0.069 |
iquette | -0.069 |
Token | it |
Token ID | 340 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
� | +0.090 |
Transparency | +0.088 |
Millions | +0.086 |
Reports | +0.085 |
Thousands | +0.083 |
Neg logit contributions | |
groom | -0.068 |
hander | -0.064 |
cha | -0.064 |
ladies | -0.062 |
iquette | -0.061 |
Token | to |
Token ID | 284 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
� | +0.094 |
Transparency | +0.092 |
Millions | +0.091 |
Reports | +0.090 |
Thousands | +0.087 |
Neg logit contributions | |
groom | -0.062 |
hander | -0.059 |
cha | -0.058 |
ladies | -0.056 |
iquette | -0.055 |
Token | be |
Token ID | 307 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
� | +0.049 |
Transparency | +0.048 |
Millions | +0.047 |
Reports | +0.046 |
Thousands | +0.045 |
Neg logit contributions | |
groom | -0.044 |
hander | -0.041 |
cha | -0.041 |
ladies | -0.040 |
iquette | -0.039 |
Token | zi |
Token ID | 17027 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
groom | +0.105 |
hander | +0.099 |
cha | +0.098 |
ladies | +0.095 |
iquette | +0.094 |
Neg logit contributions | |
� | -0.142 |
Transparency | -0.139 |
Millions | -0.136 |
Reports | -0.135 |
Thousands | -0.131 |
Token | ak |
Token ID | 461 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
� | +0.271 |
Transparency | +0.263 |
Millions | +0.258 |
Reports | +0.256 |
Thousands | +0.250 |
Neg logit contributions | |
groom | -0.150 |
hander | -0.141 |
cha | -0.138 |
ladies | -0.133 |
iquette | -0.132 |
Token | ( |
Token ID | 357 |
Feature activation | -7.31e-01 |
Pos logit contributions | |
groom | +0.162 |
hander | +0.153 |
cha | +0.152 |
ladies | +0.146 |
iquette | +0.144 |
Neg logit contributions | |
� | -0.225 |
Transparency | -0.221 |
Millions | -0.216 |
Reports | -0.214 |
Thousands | -0.208 |
Token | 4 |
Token ID | 19 |
Feature activation | +8.37e-01 |
Pos logit contributions | |
� | +0.108 |
Transparency | +0.107 |
Millions | +0.104 |
Reports | +0.103 |
Thousands | +0.100 |
Neg logit contributions | |
groom | -0.077 |
hander | -0.073 |
cha | -0.072 |
ladies | -0.069 |
iquette | -0.068 |
Token | : |
Token ID | 25 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
groom | +0.081 |
hander | +0.076 |
cha | +0.075 |
ladies | +0.072 |
iquette | +0.071 |
Neg logit contributions | |
� | -0.131 |
Transparency | -0.129 |
Millions | -0.127 |
Reports | -0.125 |
Thousands | -0.122 |
Token | 45 |
Token ID | 2231 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
groom | +0.102 |
cha | +0.095 |
hander | +0.095 |
iquette | +0.091 |
ladies | +0.091 |
Neg logit contributions | |
� | -0.135 |
Transparency | -0.133 |
Millions | -0.131 |
Reports | -0.130 |
Thousands | -0.126 |
Token | ), |
Token ID | 828 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
groom | +0.228 |
hander | +0.208 |
cha | +0.203 |
ladies | +0.192 |
iquette | +0.187 |
Neg logit contributions | |
� | -0.652 |
Transparency | -0.645 |
Millions | -0.635 |
Reports | -0.628 |
Thousands | -0.615 |
Token | Col |
Token ID | 1623 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
groom | +0.307 |
hander | +0.290 |
cha | +0.287 |
ladies | +0.278 |
iquette | +0.275 |
Neg logit contributions | |
� | -0.344 |
Transparency | -0.338 |
Millions | -0.330 |
Reports | -0.325 |
Thousands | -0.316 |
Token | ton |
Token ID | 1122 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
groom | +0.118 |
hander | +0.110 |
cha | +0.108 |
ladies | +0.102 |
iquette | +0.102 |
Neg logit contributions | |
� | -0.245 |
Transparency | -0.242 |
Millions | -0.237 |
Reports | -0.235 |
Thousands | -0.230 |
Token | Par |
Token ID | 2547 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
groom | +0.023 |
hander | +0.022 |
cha | +0.022 |
ladies | +0.021 |
iquette | +0.021 |
Neg logit contributions | |
� | -0.018 |
Transparency | -0.018 |
Millions | -0.017 |
Reports | -0.017 |
Thousands | -0.016 |
Token | ay |
Token ID | 323 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
groom | +0.128 |
hander | +0.119 |
cha | +0.116 |
iquette | +0.111 |
ladies | +0.110 |
Neg logit contributions | |
� | -0.283 |
Transparency | -0.280 |
Millions | -0.274 |
Reports | -0.272 |
Thousands | -0.266 |
Token | extras |
Token ID | 33849 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
groom | +0.024 |
hander | +0.023 |
cha | +0.023 |
ladies | +0.022 |
iquette | +0.022 |
Neg logit contributions | |
� | -0.029 |
Transparency | -0.029 |
Millions | -0.028 |
Reports | -0.028 |
Thousands | -0.027 |
Token | which |
Token ID | 543 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
groom | +0.034 |
hander | +0.032 |
cha | +0.032 |
ladies | +0.031 |
iquette | +0.030 |
Neg logit contributions | |
� | -0.045 |
Transparency | -0.044 |
Millions | -0.043 |
Reports | -0.043 |
Thousands | -0.041 |
Token | make |
Token ID | 787 |
Feature activation | +9.22e-02 |
Pos logit contributions | |
groom | +0.010 |
hander | +0.009 |
cha | +0.009 |
ladies | +0.009 |
iquette | +0.009 |
Neg logit contributions | |
� | -0.019 |
Transparency | -0.018 |
Millions | -0.018 |
Reports | -0.018 |
Thousands | -0.017 |
Token | them |
Token ID | 606 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
groom | +0.009 |
hander | +0.008 |
cha | +0.008 |
ladies | +0.008 |
iquette | +0.008 |
Neg logit contributions | |
� | -0.014 |
Transparency | -0.014 |
Millions | -0.014 |
Reports | -0.014 |
Thousands | -0.013 |
Token | more |
Token ID | 517 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
groom | +0.019 |
hander | +0.018 |
cha | +0.018 |
ladies | +0.017 |
iquette | +0.017 |
Neg logit contributions | |
� | -0.030 |
Transparency | -0.029 |
Millions | -0.029 |
Reports | -0.029 |
Thousands | -0.028 |
Token | suitable |
Token ID | 11080 |
Feature activation | +4.51e+00 |
Pos logit contributions | |
� | +0.213 |
Transparency | +0.210 |
Millions | +0.206 |
Reports | +0.204 |
Thousands | +0.199 |
Neg logit contributions | |
groom | -0.118 |
hander | -0.111 |
cha | -0.109 |
iquette | -0.104 |
ladies | -0.104 |
Token | overall |
Token ID | 4045 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
groom | +0.453 |
hander | +0.422 |
cha | +0.421 |
ladies | +0.405 |
gentlemen | +0.397 |
Neg logit contributions | |
� | -0.695 |
Transparency | -0.681 |
Millions | -0.668 |
Reports | -0.658 |
Thousands | -0.643 |
Token | for |
Token ID | 329 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
groom | +0.030 |
hander | +0.029 |
cha | +0.028 |
ladies | +0.027 |
iquette | +0.027 |
Neg logit contributions | |
� | -0.043 |
Transparency | -0.042 |
Millions | -0.041 |
Reports | -0.040 |
Thousands | -0.040 |
Token | gaming |
Token ID | 7776 |
Feature activation | +4.42e-01 |
Pos logit contributions | |
groom | +0.073 |
hander | +0.069 |
cha | +0.069 |
ladies | +0.066 |
iquette | +0.066 |
Neg logit contributions | |
� | -0.091 |
Transparency | -0.090 |
Millions | -0.088 |
Reports | -0.087 |
Thousands | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
groom | +0.050 |
hander | +0.048 |
cha | +0.047 |
ladies | +0.045 |
iquette | +0.045 |
Neg logit contributions | |
� | -0.062 |
Transparency | -0.061 |
Millions | -0.060 |
Reports | -0.059 |
Thousands | -0.057 |
Token | Firstly |
Token ID | 41039 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
� | +0.017 |
Transparency | +0.016 |
Millions | +0.016 |
Reports | +0.016 |
Thousands | +0.015 |
Neg logit contributions | |
groom | -0.027 |
hander | -0.026 |
cha | -0.026 |
ladies | -0.025 |
gentlemen | -0.025 |
Token | area |
Token ID | 1989 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
� | +0.095 |
Transparency | +0.092 |
Millions | +0.090 |
Reports | +0.089 |
Thousands | +0.086 |
Neg logit contributions | |
groom | -0.107 |
cha | -0.101 |
hander | -0.101 |
ladies | -0.098 |
gentlemen | -0.096 |
Token | : |
Token ID | 25 |
Feature activation | +7.49e-01 |
Pos logit contributions | |
groom | +0.078 |
hander | +0.073 |
cha | +0.073 |
ladies | +0.070 |
iquette | +0.069 |
Neg logit contributions | |
� | -0.101 |
Transparency | -0.098 |
Millions | -0.097 |
Reports | -0.096 |
Thousands | -0.093 |
Token | your |
Token ID | 534 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
groom | +0.085 |
hander | +0.080 |
cha | +0.080 |
ladies | +0.077 |
gentlemen | +0.076 |
Neg logit contributions | |
� | -0.105 |
Transparency | -0.102 |
Millions | -0.100 |
Reports | -0.099 |
Thousands | -0.096 |
Token | raw |
Token ID | 8246 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
groom | +0.422 |
cha | +0.397 |
hander | +0.396 |
ladies | +0.384 |
gentlemen | +0.377 |
Neg logit contributions | |
� | -0.463 |
Transparency | -0.448 |
Millions | -0.441 |
Reports | -0.436 |
Thousands | -0.422 |
Token | file |
Token ID | 2393 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
� | +0.256 |
Transparency | +0.247 |
Millions | +0.244 |
Reports | +0.241 |
Thousands | +0.233 |
Neg logit contributions | |
groom | -0.253 |
hander | -0.239 |
cha | -0.238 |
ladies | -0.231 |
gentlemen | -0.227 |
Token | will |
Token ID | 481 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
groom | +0.028 |
hander | +0.026 |
cha | +0.026 |
ladies | +0.025 |
iquette | +0.024 |
Neg logit contributions | |
� | -0.062 |
Transparency | -0.061 |
Millions | -0.060 |
Reports | -0.060 |
Thousands | -0.058 |
Token | contain |
Token ID | 3994 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
groom | +0.324 |
cha | +0.304 |
hander | +0.302 |
ladies | +0.291 |
iquette | +0.285 |
Neg logit contributions | |
� | -0.400 |
Transparency | -0.389 |
Millions | -0.381 |
Reports | -0.378 |
Thousands | -0.368 |
Token | only |
Token ID | 691 |
Feature activation | -8.91e-01 |
Pos logit contributions | |
groom | +0.104 |
cha | +0.098 |
hander | +0.097 |
ladies | +0.094 |
gentlemen | +0.092 |
Neg logit contributions | |
� | -0.117 |
Transparency | -0.113 |
Millions | -0.111 |
Reports | -0.110 |
Thousands | -0.106 |
Token | the |
Token ID | 262 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
� | +0.122 |
Transparency | +0.119 |
Millions | +0.117 |
Reports | +0.115 |
Thousands | +0.112 |
Neg logit contributions | |
groom | -0.104 |
cha | -0.098 |
hander | -0.098 |
ladies | -0.095 |
gentlemen | -0.093 |
Token | selected |
Token ID | 6163 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
groom | +0.049 |
cha | +0.046 |
hander | +0.046 |
ladies | +0.045 |
gentlemen | +0.044 |
Neg logit contributions | |
� | -0.056 |
Transparency | -0.054 |
Millions | -0.053 |
Reports | -0.052 |
Thousands | -0.051 |
Token | aspect |
Token ID | 4843 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
groom | +0.172 |
cha | +0.162 |
hander | +0.160 |
ladies | +0.156 |
gentlemen | +0.153 |
Neg logit contributions | |
� | -0.201 |
Transparency | -0.194 |
Millions | -0.191 |
Reports | -0.189 |
Thousands | -0.183 |
Token | long |
Token ID | 890 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
� | +0.201 |
Transparency | +0.196 |
Millions | +0.191 |
Reports | +0.190 |
Thousands | +0.184 |
Neg logit contributions | |
groom | -0.201 |
cha | -0.187 |
hander | -0.186 |
ladies | -0.182 |
gentlemen | -0.179 |
Token | been |
Token ID | 587 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
groom | +0.032 |
cha | +0.030 |
hander | +0.030 |
ladies | +0.029 |
gentlemen | +0.029 |
Neg logit contributions | |
� | -0.024 |
Transparency | -0.023 |
Millions | -0.023 |
Reports | -0.022 |
Thousands | -0.022 |
Token | excluded |
Token ID | 15009 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
� | +0.353 |
Transparency | +0.346 |
Millions | +0.339 |
Reports | +0.334 |
Thousands | +0.325 |
Neg logit contributions | |
groom | -0.283 |
hander | -0.266 |
cha | -0.264 |
ladies | -0.256 |
iquette | -0.252 |
Token | from |
Token ID | 422 |
Feature activation | -8.90e-01 |
Pos logit contributions | |
� | +0.004 |
Transparency | +0.004 |
Millions | +0.004 |
Reports | +0.004 |
Thousands | +0.004 |
Neg logit contributions | |
groom | -0.003 |
hander | -0.003 |
cha | -0.003 |
ladies | -0.003 |
gentlemen | -0.003 |
Token | practicing |
Token ID | 18207 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
� | +0.141 |
Transparency | +0.139 |
Millions | +0.136 |
Reports | +0.135 |
Thousands | +0.131 |
Neg logit contributions | |
groom | -0.085 |
hander | -0.079 |
cha | -0.078 |
ladies | -0.075 |
iquette | -0.074 |
Token | medicine |
Token ID | 9007 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
groom | +0.180 |
hander | +0.165 |
cha | +0.160 |
ladies | +0.151 |
iquette | +0.150 |
Neg logit contributions | |
� | -0.485 |
Transparency | -0.478 |
Millions | -0.470 |
Reports | -0.465 |
Thousands | -0.456 |
Token | on |
Token ID | 319 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
groom | +0.333 |
hander | +0.312 |
cha | +0.311 |
ladies | +0.300 |
iquette | +0.295 |
Neg logit contributions | |
� | -0.436 |
Transparency | -0.428 |
Millions | -0.418 |
Reports | -0.414 |
Thousands | -0.402 |
Token | the |
Token ID | 262 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
groom | +0.074 |
hander | +0.069 |
cha | +0.069 |
ladies | +0.066 |
iquette | +0.065 |
Neg logit contributions | |
� | -0.098 |
Transparency | -0.096 |
Millions | -0.094 |
Reports | -0.093 |
Thousands | -0.091 |
Token | grounds |
Token ID | 9384 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
groom | +0.018 |
cha | +0.017 |
hander | +0.017 |
ladies | +0.016 |
gentlemen | +0.016 |
Neg logit contributions | |
� | -0.024 |
Transparency | -0.023 |
Millions | -0.023 |
Reports | -0.022 |
Thousands | -0.022 |
Token | that |
Token ID | 326 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
groom | +0.151 |
cha | +0.142 |
hander | +0.142 |
ladies | +0.139 |
gentlemen | +0.136 |
Neg logit contributions | |
� | -0.170 |
Transparency | -0.168 |
Millions | -0.166 |
Reports | -0.163 |
Thousands | -0.159 |
Token | medicine |
Token ID | 9007 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
� | +0.089 |
Transparency | +0.087 |
Millions | +0.085 |
Reports | +0.084 |
Thousands | +0.082 |
Neg logit contributions | |
groom | -0.074 |
hander | -0.070 |
cha | -0.070 |
ladies | -0.067 |
gentlemen | -0.066 |
Token | mos |
Token ID | 16785 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
groom | +0.135 |
hander | +0.126 |
cha | +0.124 |
ladies | +0.120 |
iquette | +0.119 |
Neg logit contributions | |
� | -0.206 |
Transparency | -0.202 |
Millions | -0.198 |
Reports | -0.196 |
Thousands | -0.190 |
Token | cos |
Token ID | 8615 |
Feature activation | +3.95e+00 |
Pos logit contributions | |
groom | +0.040 |
hander | +0.038 |
cha | +0.037 |
ladies | +0.036 |
iquette | +0.035 |
Neg logit contributions | |
� | -0.058 |
Transparency | -0.057 |
Millions | -0.056 |
Reports | -0.055 |
Thousands | -0.054 |
Token | mon |
Token ID | 2144 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
groom | +0.331 |
hander | +0.302 |
cha | +0.300 |
ladies | +0.287 |
iquette | +0.287 |
Neg logit contributions | |
� | -0.675 |
Transparency | -0.656 |
Millions | -0.646 |
Reports | -0.640 |
Thousands | -0.625 |
Token | auts |
Token ID | 17712 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
groom | +0.245 |
hander | +0.231 |
cha | +0.228 |
ladies | +0.220 |
iquette | +0.218 |
Neg logit contributions | |
� | -0.341 |
Transparency | -0.335 |
Millions | -0.328 |
Reports | -0.324 |
Thousands | -0.315 |
Token | Sergey |
Token ID | 36106 |
Feature activation | +9.01e-01 |
Pos logit contributions | |
� | +0.079 |
Transparency | +0.078 |
Millions | +0.076 |
Reports | +0.075 |
Thousands | +0.072 |
Neg logit contributions | |
groom | -0.073 |
hander | -0.069 |
cha | -0.068 |
ladies | -0.066 |
iquette | -0.065 |
Token | Ry |
Token ID | 11089 |
Feature activation | +7.90e-01 |
Pos logit contributions | |
groom | +0.121 |
hander | +0.115 |
cha | +0.114 |
ladies | +0.111 |
iquette | +0.110 |
Neg logit contributions | |
� | -0.108 |
Transparency | -0.106 |
Millions | -0.103 |
Reports | -0.101 |
Thousands | -0.098 |
Token | zh |
Token ID | 23548 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
groom | +0.068 |
hander | +0.063 |
cha | +0.062 |
ladies | +0.059 |
iquette | +0.059 |
Neg logit contributions | |
� | -0.132 |
Transparency | -0.130 |
Millions | -0.128 |
Reports | -0.127 |
Thousands | -0.124 |
Token | ik |
Token ID | 1134 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
groom | +0.152 |
hander | +0.143 |
cha | +0.141 |
ladies | +0.136 |
iquette | +0.135 |
Neg logit contributions | |
� | -0.212 |
Transparency | -0.209 |
Millions | -0.204 |
Reports | -0.202 |
Thousands | -0.197 |
Token | ov |
Token ID | 709 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
groom | +0.161 |
hander | +0.152 |
cha | +0.150 |
ladies | +0.144 |
iquette | +0.143 |
Neg logit contributions | |
� | -0.226 |
Transparency | -0.222 |
Millions | -0.218 |
Reports | -0.215 |
Thousands | -0.209 |
Token | and |
Token ID | 290 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
groom | +0.107 |
hander | +0.100 |
cha | +0.099 |
ladies | +0.095 |
iquette | +0.094 |
Neg logit contributions | |
� | -0.163 |
Transparency | -0.161 |
Millions | -0.158 |
Reports | -0.155 |
Thousands | -0.152 |
Token | And |
Token ID | 843 |
Feature activation | -7.96e-01 |
Pos logit contributions | |
� | +0.015 |
Transparency | +0.015 |
Millions | +0.014 |
Reports | +0.014 |
Thousands | +0.014 |
Neg logit contributions | |
groom | -0.016 |
hander | -0.015 |
cha | -0.015 |
ladies | -0.014 |
iquette | -0.014 |
Token | 's |
Token ID | 338 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
� | +0.177 |
Transparency | +0.174 |
Millions | +0.171 |
Reports | +0.169 |
Thousands | +0.165 |
Neg logit contributions | |
groom | -0.101 |
hander | -0.094 |
cha | -0.093 |
ladies | -0.089 |
iquette | -0.088 |
Token | being |
Token ID | 852 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
� | +0.051 |
Transparency | +0.050 |
Millions | +0.049 |
Reports | +0.048 |
Thousands | +0.047 |
Neg logit contributions | |
groom | -0.041 |
hander | -0.038 |
cha | -0.038 |
ladies | -0.037 |
iquette | -0.036 |
Token | played |
Token ID | 2826 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
� | +0.317 |
Transparency | +0.311 |
Millions | +0.306 |
Reports | +0.302 |
Thousands | +0.294 |
Neg logit contributions | |
groom | -0.208 |
hander | -0.197 |
cha | -0.193 |
ladies | -0.187 |
iquette | -0.187 |
Token | . |
Token ID | 13 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
� | +0.020 |
Transparency | +0.020 |
Millions | +0.019 |
Reports | +0.019 |
Thousands | +0.019 |
Neg logit contributions | |
groom | -0.016 |
hander | -0.015 |
cha | -0.015 |
ladies | -0.014 |
iquette | -0.014 |
Token | It |
Token ID | 632 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
� | +0.076 |
Transparency | +0.073 |
Millions | +0.071 |
Reports | +0.070 |
Thousands | +0.067 |
Neg logit contributions | |
groom | -0.107 |
hander | -0.103 |
cha | -0.102 |
ladies | -0.100 |
iquette | -0.099 |
Token | works |
Token ID | 2499 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
� | +0.272 |
Transparency | +0.268 |
Millions | +0.263 |
Reports | +0.261 |
Thousands | +0.254 |
Neg logit contributions | |
groom | -0.145 |
hander | -0.136 |
cha | -0.133 |
ladies | -0.127 |
iquette | -0.127 |
Token | as |
Token ID | 355 |
Feature activation | -5.89e-01 |
Pos logit contributions | |
groom | +0.079 |
hander | +0.075 |
cha | +0.073 |
ladies | +0.070 |
iquette | +0.070 |
Neg logit contributions | |
� | -0.120 |
Transparency | -0.118 |
Millions | -0.116 |
Reports | -0.114 |
Thousands | -0.112 |
Token | well |
Token ID | 880 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
� | +0.095 |
Transparency | +0.093 |
Millions | +0.091 |
Reports | +0.090 |
Thousands | +0.088 |
Neg logit contributions | |
groom | -0.054 |
hander | -0.051 |
cha | -0.050 |
ladies | -0.048 |
iquette | -0.048 |
Token | when |
Token ID | 618 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
groom | +0.172 |
hander | +0.162 |
cha | +0.159 |
ladies | +0.154 |
iquette | +0.153 |
Neg logit contributions | |
� | -0.245 |
Transparency | -0.241 |
Millions | -0.236 |
Reports | -0.233 |
Thousands | -0.227 |
Token | spect |
Token ID | 5444 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
groom | +0.036 |
hander | +0.033 |
cha | +0.032 |
ladies | +0.030 |
iquette | +0.030 |
Neg logit contributions | |
� | -0.086 |
Transparency | -0.084 |
Millions | -0.083 |
Reports | -0.082 |
Thousands | -0.080 |
Token | ating |
Token ID | 803 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
� | +0.405 |
Transparency | +0.398 |
Millions | +0.391 |
Reports | +0.386 |
Thousands | +0.377 |
Neg logit contributions | |
groom | -0.242 |
hander | -0.226 |
cha | -0.223 |
ladies | -0.214 |
iquette | -0.211 |
Token | division |
Token ID | 7297 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
groom | +0.147 |
hander | +0.139 |
cha | +0.139 |
ladies | +0.134 |
iquette | +0.133 |
Neg logit contributions | |
� | -0.162 |
Transparency | -0.160 |
Millions | -0.155 |
Reports | -0.153 |
Thousands | -0.149 |
Token | against |
Token ID | 1028 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
groom | +0.016 |
hander | +0.015 |
cha | +0.015 |
ladies | +0.014 |
iquette | +0.014 |
Neg logit contributions | |
� | -0.022 |
Transparency | -0.022 |
Millions | -0.021 |
Reports | -0.021 |
Thousands | -0.021 |
Token | 41 |
Token ID | 6073 |
Feature activation | -8.47e-01 |
Pos logit contributions | |
groom | +0.033 |
hander | +0.031 |
cha | +0.031 |
ladies | +0.030 |
gentlemen | +0.030 |
Neg logit contributions | |
� | -0.030 |
Transparency | -0.030 |
Millions | -0.029 |
Reports | -0.028 |
Thousands | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
� | +0.113 |
Transparency | +0.111 |
Millions | +0.108 |
Reports | +0.107 |
Thousands | +0.104 |
Neg logit contributions | |
groom | -0.102 |
hander | -0.096 |
cha | -0.096 |
ladies | -0.093 |
gentlemen | -0.091 |
Token | 88 |
Token ID | 9193 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
� | +0.006 |
Transparency | +0.006 |
Millions | +0.006 |
Reports | +0.006 |
Thousands | +0.006 |
Neg logit contributions | |
groom | -0.006 |
hander | -0.005 |
cha | -0.005 |
ladies | -0.005 |
gentlemen | -0.005 |
Token | 11 |
Token ID | 1157 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
groom | +0.068 |
hander | +0.064 |
cha | +0.064 |
ladies | +0.062 |
iquette | +0.061 |
Neg logit contributions | |
� | -0.085 |
Transparency | -0.084 |
Millions | -0.081 |
Reports | -0.080 |
Thousands | -0.078 |
Token | tanks |
Token ID | 11657 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
groom | +0.035 |
hander | +0.034 |
cha | +0.033 |
ladies | +0.032 |
gentlemen | +0.032 |
Neg logit contributions | |
� | -0.044 |
Transparency | -0.044 |
Millions | -0.043 |
Reports | -0.042 |
Thousands | -0.041 |
Token | against |
Token ID | 1028 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
� | +0.204 |
Transparency | +0.201 |
Millions | +0.197 |
Reports | +0.195 |
Thousands | +0.191 |
Neg logit contributions | |
groom | -0.108 |
hander | -0.101 |
cha | -0.099 |
ladies | -0.095 |
iquette | -0.094 |
Token | 35 |
Token ID | 3439 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
groom | +0.017 |
hander | +0.017 |
cha | +0.016 |
ladies | +0.016 |
iquette | +0.016 |
Neg logit contributions | |
� | -0.016 |
Transparency | -0.015 |
Millions | -0.015 |
Reports | -0.015 |
Thousands | -0.014 |
Token | 12 |
Token ID | 1065 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
� | +0.103 |
Transparency | +0.101 |
Millions | +0.099 |
Reports | +0.098 |
Thousands | +0.095 |
Neg logit contributions | |
groom | -0.066 |
hander | -0.061 |
cha | -0.061 |
ladies | -0.059 |
iquette | -0.058 |
Token | ), |
Token ID | 828 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
� | +0.014 |
Transparency | +0.014 |
Millions | +0.014 |
Reports | +0.013 |
Thousands | +0.013 |
Neg logit contributions | |
groom | -0.012 |
hander | -0.012 |
cha | -0.012 |
ladies | -0.011 |
gentlemen | -0.011 |
Token | Photo |
Token ID | 5555 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
� | +0.041 |
Transparency | +0.040 |
Millions | +0.039 |
Reports | +0.038 |
Thousands | +0.036 |
Neg logit contributions | |
groom | -0.059 |
hander | -0.056 |
cha | -0.056 |
ladies | -0.055 |
iquette | -0.054 |
Token | F |
Token ID | 376 |
Feature activation | -5.46e-01 |
Pos logit contributions | |
� | +0.193 |
Transparency | +0.186 |
Millions | +0.180 |
Reports | +0.177 |
Thousands | +0.170 |
Neg logit contributions | |
groom | -0.240 |
hander | -0.229 |
cha | -0.227 |
ladies | -0.222 |
iquette | -0.219 |
Token | SU |
Token ID | 12564 |
Feature activation | -7.32e-01 |
Pos logit contributions | |
� | +0.077 |
Transparency | +0.076 |
Millions | +0.074 |
Reports | +0.073 |
Thousands | +0.071 |
Neg logit contributions | |
groom | -0.061 |
hander | -0.058 |
cha | -0.057 |
ladies | -0.055 |
iquette | -0.055 |
Token | target |
Token ID | 2496 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
� | +0.117 |
Transparency | +0.116 |
Millions | +0.113 |
Reports | +0.112 |
Thousands | +0.109 |
Neg logit contributions | |
groom | -0.068 |
hander | -0.064 |
cha | -0.063 |
ladies | -0.060 |
iquette | -0.060 |
Token | Marvin |
Token ID | 35105 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
� | +0.042 |
Transparency | +0.041 |
Millions | +0.040 |
Reports | +0.039 |
Thousands | +0.038 |
Neg logit contributions | |
groom | -0.035 |
hander | -0.033 |
cha | -0.033 |
ladies | -0.032 |
iquette | -0.031 |
Token | Wilson |
Token ID | 8127 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
groom | +0.270 |
hander | +0.257 |
cha | +0.255 |
ladies | +0.248 |
iquette | +0.245 |
Neg logit contributions | |
� | -0.257 |
Transparency | -0.251 |
Millions | -0.244 |
Reports | -0.241 |
Thousands | -0.233 |
Token | officially |
Token ID | 8720 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
groom | +0.111 |
hander | +0.105 |
cha | +0.103 |
ladies | +0.100 |
iquette | +0.099 |
Neg logit contributions | |
� | -0.161 |
Transparency | -0.159 |
Millions | -0.155 |
Reports | -0.154 |
Thousands | -0.150 |
Token | visits |
Token ID | 11864 |
Feature activation | -5.96e-02 |
Pos logit contributions | |
� | +0.035 |
Transparency | +0.034 |
Millions | +0.033 |
Reports | +0.033 |
Thousands | +0.032 |
Neg logit contributions | |
groom | -0.024 |
hander | -0.022 |
cha | -0.022 |
ladies | -0.021 |
iquette | -0.021 |
Token | F |
Token ID | 376 |
Feature activation | -4.71e-01 |
Pos logit contributions | |
� | +0.008 |
Transparency | +0.008 |
Millions | +0.008 |
Reports | +0.008 |
Thousands | +0.008 |
Neg logit contributions | |
groom | -0.007 |
hander | -0.007 |
cha | -0.006 |
ladies | -0.006 |
iquette | -0.006 |
Token | SU |
Token ID | 12564 |
Feature activation | -6.68e-01 |
Pos logit contributions | |
� | +0.048 |
Transparency | +0.047 |
Millions | +0.046 |
Reports | +0.045 |
Thousands | +0.043 |
Neg logit contributions | |
groom | -0.071 |
hander | -0.068 |
cha | -0.068 |
ladies | -0.066 |
iquette | -0.066 |
Token | for |
Token ID | 329 |
Feature activation | +8.10e-01 |
Pos logit contributions | |
� | +0.100 |
Transparency | +0.099 |
Millions | +0.097 |
Reports | +0.095 |
Thousands | +0.093 |
Neg logit contributions | |
groom | -0.069 |
hander | -0.065 |
cha | -0.064 |
ladies | -0.062 |
iquette | -0.061 |
Token | - |
Token ID | 12 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
groom | +0.120 |
hander | +0.113 |
cha | +0.111 |
ladies | +0.108 |
iquette | +0.106 |
Neg logit contributions | |
� | -0.181 |
Transparency | -0.178 |
Millions | -0.175 |
Reports | -0.173 |
Thousands | -0.169 |
Token | performance |
Token ID | 26585 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
groom | +0.019 |
hander | +0.018 |
cha | +0.018 |
ladies | +0.017 |
iquette | +0.017 |
Neg logit contributions | |
� | -0.025 |
Transparency | -0.025 |
Millions | -0.024 |
Reports | -0.024 |
Thousands | -0.023 |
Token | off |
Token ID | 572 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
groom | +0.048 |
hander | +0.045 |
cha | +0.044 |
ladies | +0.043 |
iquette | +0.042 |
Neg logit contributions | |
� | -0.067 |
Transparency | -0.065 |
Millions | -0.064 |
Reports | -0.063 |
Thousands | -0.062 |
Token | - |
Token ID | 12 |
Feature activation | +2.16e-01 |
Pos logit contributions | |
� | +0.031 |
Transparency | +0.030 |
Millions | +0.030 |
Reports | +0.029 |
Thousands | +0.029 |
Neg logit contributions | |
groom | -0.021 |
hander | -0.020 |
cha | -0.019 |
ladies | -0.019 |
iquette | -0.019 |
Token | road |
Token ID | 6344 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
groom | +0.026 |
hander | +0.024 |
cha | +0.024 |
ladies | +0.023 |
iquette | +0.023 |
Neg logit contributions | |
� | -0.029 |
Transparency | -0.028 |
Millions | -0.028 |
Reports | -0.027 |
Thousands | -0.026 |
Token | pickup |
Token ID | 19422 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
� | +0.182 |
Transparency | +0.179 |
Millions | +0.175 |
Reports | +0.173 |
Thousands | +0.169 |
Neg logit contributions | |
groom | -0.121 |
hander | -0.114 |
cha | -0.112 |
ladies | -0.108 |
iquette | -0.107 |
Token | in |
Token ID | 287 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
groom | +0.015 |
hander | +0.014 |
cha | +0.014 |
ladies | +0.013 |
iquette | +0.013 |
Neg logit contributions | |
� | -0.020 |
Transparency | -0.020 |
Millions | -0.020 |
Reports | -0.019 |
Thousands | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
groom | +0.012 |
hander | +0.011 |
cha | +0.011 |
ladies | +0.011 |
iquette | +0.011 |
Neg logit contributions | |
� | -0.016 |
Transparency | -0.015 |
Millions | -0.015 |
Reports | -0.015 |
Thousands | -0.014 |
Token | company |
Token ID | 1664 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
groom | +0.072 |
hander | +0.068 |
cha | +0.067 |
ladies | +0.066 |
iquette | +0.065 |
Neg logit contributions | |
� | -0.069 |
Transparency | -0.067 |
Millions | -0.065 |
Reports | -0.064 |
Thousands | -0.062 |
Token | 's |
Token ID | 338 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
groom | +0.080 |
hander | +0.075 |
cha | +0.074 |
ladies | +0.071 |
iquette | +0.070 |
Neg logit contributions | |
� | -0.126 |
Transparency | -0.124 |
Millions | -0.122 |
Reports | -0.120 |
Thousands | -0.117 |
Token | lineup |
Token ID | 12750 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
groom | +0.032 |
cha | +0.030 |
hander | +0.030 |
ladies | +0.030 |
iquette | +0.029 |
Neg logit contributions | |
� | -0.035 |
Transparency | -0.034 |
Millions | -0.033 |
Reports | -0.033 |
Thousands | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +7.00e-02 |
Pos logit contributions | |
� | +0.330 |
Transparency | +0.324 |
Millions | +0.316 |
Reports | +0.313 |
Thousands | +0.304 |
Neg logit contributions | |
groom | -0.230 |
hander | -0.215 |
cha | -0.214 |
ladies | -0.207 |
gentlemen | -0.203 |
Token | first |
Token ID | 717 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
groom | +0.007 |
hander | +0.006 |
cha | +0.006 |
ladies | +0.006 |
iquette | +0.006 |
Neg logit contributions | |
� | -0.011 |
Transparency | -0.011 |
Millions | -0.011 |
Reports | -0.011 |
Thousands | -0.010 |
Token | bridge |
Token ID | 7696 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
groom | +0.043 |
hander | +0.041 |
cha | +0.040 |
ladies | +0.039 |
iquette | +0.038 |
Neg logit contributions | |
� | -0.067 |
Transparency | -0.066 |
Millions | -0.064 |
Reports | -0.063 |
Thousands | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
groom | +0.182 |
hander | +0.171 |
cha | +0.170 |
ladies | +0.164 |
gentlemen | +0.160 |
Neg logit contributions | |
� | -0.264 |
Transparency | -0.259 |
Millions | -0.253 |
Reports | -0.251 |
Thousands | -0.243 |
Token | and |
Token ID | 290 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
� | +0.012 |
Transparency | +0.011 |
Millions | +0.011 |
Reports | +0.011 |
Thousands | +0.011 |
Neg logit contributions | |
groom | -0.009 |
cha | -0.008 |
hander | -0.008 |
ladies | -0.008 |
gentlemen | -0.008 |
Token | so |
Token ID | 523 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
� | +0.050 |
Transparency | +0.049 |
Millions | +0.048 |
Reports | +0.047 |
Thousands | +0.046 |
Neg logit contributions | |
groom | -0.038 |
cha | -0.035 |
hander | -0.035 |
ladies | -0.034 |
gentlemen | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +4.42e-02 |
Pos logit contributions | |
� | +0.194 |
Transparency | +0.191 |
Millions | +0.186 |
Reports | +0.184 |
Thousands | +0.178 |
Neg logit contributions | |
groom | -0.150 |
hander | -0.141 |
cha | -0.140 |
ladies | -0.135 |
gentlemen | -0.134 |
Token | king |
Token ID | 5822 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
groom | +0.005 |
hander | +0.005 |
cha | +0.005 |
ladies | +0.005 |
gentlemen | +0.005 |
Neg logit contributions | |
� | -0.006 |
Transparency | -0.006 |
Millions | -0.006 |
Reports | -0.006 |
Thousands | -0.006 |
Token | commanded |
Token ID | 22419 |
Feature activation | +7.77e-01 |
Pos logit contributions | |
groom | +0.275 |
hander | +0.258 |
cha | +0.256 |
ladies | +0.248 |
gentlemen | +0.244 |
Neg logit contributions | |
� | -0.331 |
Transparency | -0.325 |
Millions | -0.317 |
Reports | -0.314 |
Thousands | -0.303 |
Token | his |
Token ID | 465 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
groom | +0.080 |
hander | +0.074 |
cha | +0.074 |
ladies | +0.072 |
gentlemen | +0.071 |
Neg logit contributions | |
� | -0.118 |
Transparency | -0.116 |
Millions | -0.113 |
Reports | -0.112 |
Thousands | -0.109 |
Token | men |
Token ID | 1450 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
groom | +0.084 |
hander | +0.079 |
cha | +0.079 |
ladies | +0.076 |
gentlemen | +0.075 |
Neg logit contributions | |
� | -0.105 |
Transparency | -0.103 |
Millions | -0.101 |
Reports | -0.099 |
Thousands | -0.096 |
Token | Prime |
Token ID | 5537 |
Feature activation | -9.19e-01 |
Pos logit contributions | |
� | +0.020 |
Transparency | +0.020 |
Millions | +0.019 |
Reports | +0.019 |
Thousands | +0.018 |
Neg logit contributions | |
groom | -0.016 |
hander | -0.015 |
cha | -0.015 |
ladies | -0.014 |
iquette | -0.014 |
Token | Minister |
Token ID | 4139 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
� | +0.103 |
Transparency | +0.102 |
Millions | +0.100 |
Reports | +0.098 |
Thousands | +0.095 |
Neg logit contributions | |
groom | -0.129 |
hander | -0.123 |
cha | -0.122 |
iquette | -0.120 |
ladies | -0.119 |
Token | , |
Token ID | 11 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
groom | +0.028 |
hander | +0.026 |
cha | +0.026 |
ladies | +0.025 |
iquette | +0.025 |
Neg logit contributions | |
� | -0.034 |
Transparency | -0.034 |
Millions | -0.033 |
Reports | -0.033 |
Thousands | -0.032 |
Token | reports |
Token ID | 3136 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
groom | +0.029 |
hander | +0.027 |
cha | +0.027 |
ladies | +0.026 |
gentlemen | +0.026 |
Neg logit contributions | |
� | -0.037 |
Transparency | -0.036 |
Millions | -0.036 |
Reports | -0.035 |
Thousands | -0.034 |
Token | said |
Token ID | 531 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
� | +0.268 |
Transparency | +0.263 |
Millions | +0.257 |
Reports | +0.253 |
Thousands | +0.247 |
Neg logit contributions | |
groom | -0.224 |
hander | -0.212 |
cha | -0.210 |
ladies | -0.204 |
iquette | -0.201 |
Token | . |
Token ID | 13 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
groom | +0.120 |
cha | +0.113 |
hander | +0.113 |
ladies | +0.109 |
gentlemen | +0.107 |
Neg logit contributions | |
� | -0.147 |
Transparency | -0.144 |
Millions | -0.140 |
Reports | -0.138 |
Thousands | -0.134 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
� | +0.027 |
Transparency | +0.027 |
Millions | +0.026 |
Reports | +0.025 |
Thousands | +0.025 |
Neg logit contributions | |
groom | -0.024 |
hander | -0.023 |
cha | -0.023 |
ladies | -0.022 |
iquette | -0.022 |
Token | There |
Token ID | 1858 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
groom | +0.152 |
hander | +0.146 |
cha | +0.144 |
ladies | +0.140 |
iquette | +0.139 |
Neg logit contributions | |
� | -0.144 |
Transparency | -0.141 |
Millions | -0.137 |
Reports | -0.135 |
Thousands | -0.131 |
Token | is |
Token ID | 318 |
Feature activation | +6.06e-01 |
Pos logit contributions | |
groom | +0.009 |
hander | +0.008 |
cha | +0.008 |
ladies | +0.008 |
iquette | +0.008 |
Neg logit contributions | |
� | -0.017 |
Transparency | -0.017 |
Millions | -0.016 |
Reports | -0.016 |
Thousands | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
groom | +0.071 |
hander | +0.066 |
cha | +0.066 |
ladies | +0.065 |
gentlemen | +0.063 |
Neg logit contributions | |
� | -0.083 |
Transparency | -0.081 |
Millions | -0.079 |
Reports | -0.078 |
Thousands | -0.076 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
groom | +0.024 |
hander | +0.023 |
cha | +0.023 |
ladies | +0.022 |
iquette | +0.022 |
Neg logit contributions | |
� | -0.030 |
Transparency | -0.030 |
Millions | -0.029 |
Reports | -0.029 |
Thousands | -0.028 |
Token | feeling |
Token ID | 4203 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
groom | +0.035 |
hander | +0.033 |
cha | +0.033 |
ladies | +0.031 |
iquette | +0.031 |
Neg logit contributions | |
� | -0.048 |
Transparency | -0.047 |
Millions | -0.046 |
Reports | -0.046 |
Thousands | -0.044 |
Token | that |
Token ID | 326 |
Feature activation | -4.87e-01 |
Pos logit contributions | |
� | +0.170 |
Transparency | +0.166 |
Millions | +0.164 |
Reports | +0.162 |
Thousands | +0.157 |
Neg logit contributions | |
groom | -0.155 |
hander | -0.146 |
cha | -0.146 |
ladies | -0.142 |
gentlemen | -0.140 |
Token | you |
Token ID | 345 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
� | +0.076 |
Transparency | +0.075 |
Millions | +0.074 |
Reports | +0.073 |
Thousands | +0.071 |
Neg logit contributions | |
groom | -0.047 |
hander | -0.044 |
cha | -0.044 |
ladies | -0.042 |
iquette | -0.041 |
Token | 're |
Token ID | 821 |
Feature activation | +8.58e-01 |
Pos logit contributions | |
groom | +0.145 |
hander | +0.136 |
cha | +0.132 |
iquette | +0.126 |
ladies | +0.125 |
Neg logit contributions | |
� | -0.332 |
Transparency | -0.328 |
Millions | -0.322 |
Reports | -0.319 |
Thousands | -0.312 |
Token | not |
Token ID | 407 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
groom | +0.092 |
hander | +0.087 |
cha | +0.086 |
ladies | +0.083 |
iquette | +0.082 |
Neg logit contributions | |
� | -0.125 |
Transparency | -0.123 |
Millions | -0.121 |
Reports | -0.119 |
Thousands | -0.116 |
Token | just |
Token ID | 655 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
� | +0.331 |
Transparency | +0.325 |
Millions | +0.318 |
Reports | +0.315 |
Thousands | +0.306 |
Neg logit contributions | |
groom | -0.245 |
hander | -0.231 |
cha | -0.228 |
ladies | -0.220 |
iquette | -0.218 |
Token | experiencing |
Token ID | 13456 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
� | +0.095 |
Transparency | +0.094 |
Millions | +0.092 |
Reports | +0.091 |
Thousands | +0.088 |
Neg logit contributions | |
groom | -0.064 |
hander | -0.061 |
cha | -0.060 |
ladies | -0.058 |
iquette | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
groom | +0.024 |
hander | +0.023 |
cha | +0.022 |
ladies | +0.022 |
iquette | +0.021 |
Neg logit contributions | |
� | -0.031 |
Transparency | -0.031 |
Millions | -0.030 |
Reports | -0.029 |
Thousands | -0.029 |
Token | scripted |
Token ID | 37341 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
groom | +0.002 |
hander | +0.002 |
cha | +0.002 |
ladies | +0.002 |
iquette | +0.002 |
Neg logit contributions | |
� | -0.003 |
Transparency | -0.003 |
Millions | -0.003 |
Reports | -0.003 |
Thousands | -0.003 |
Token | sequence |
Token ID | 8379 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
groom | +0.064 |
hander | +0.060 |
cha | +0.060 |
ladies | +0.058 |
gentlemen | +0.057 |
Neg logit contributions | |
� | -0.078 |
Transparency | -0.077 |
Millions | -0.075 |
Reports | -0.074 |
Thousands | -0.072 |
Token | of |
Token ID | 286 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
groom | +0.169 |
hander | +0.160 |
cha | +0.158 |
ladies | +0.153 |
iquette | +0.151 |
Neg logit contributions | |
� | -0.212 |
Transparency | -0.208 |
Millions | -0.203 |
Reports | -0.201 |
Thousands | -0.195 |
Token | a |
Token ID | 257 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
groom | +0.126 |
cha | +0.121 |
hander | +0.120 |
ladies | +0.117 |
gentlemen | +0.115 |
Neg logit contributions | |
� | -0.109 |
Transparency | -0.105 |
Millions | -0.102 |
Reports | -0.101 |
Thousands | -0.098 |
Token | chain |
Token ID | 6333 |
Feature activation | -4.51e-01 |
Pos logit contributions | |
groom | +0.005 |
cha | +0.005 |
hander | +0.005 |
ladies | +0.005 |
iquette | +0.005 |
Neg logit contributions | |
� | -0.006 |
Transparency | -0.006 |
Millions | -0.006 |
Reports | -0.006 |
Thousands | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
� | +0.066 |
Transparency | +0.065 |
Millions | +0.063 |
Reports | +0.062 |
Thousands | +0.061 |
Neg logit contributions | |
groom | -0.049 |
hander | -0.046 |
cha | -0.046 |
ladies | -0.044 |
iquette | -0.043 |
Token | C |
Token ID | 327 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
groom | +0.043 |
cha | +0.041 |
hander | +0.041 |
ladies | +0.040 |
gentlemen | +0.039 |
Neg logit contributions | |
� | -0.032 |
Transparency | -0.031 |
Millions | -0.030 |
Reports | -0.029 |
Thousands | -0.028 |
Token | iti |
Token ID | 8846 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
groom | +0.003 |
hander | +0.003 |
cha | +0.003 |
ladies | +0.003 |
iquette | +0.003 |
Neg logit contributions | |
� | -0.004 |
Transparency | -0.004 |
Millions | -0.004 |
Reports | -0.004 |
Thousands | -0.004 |
Token | Group |
Token ID | 13247 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
groom | +0.216 |
cha | +0.204 |
hander | +0.204 |
ladies | +0.197 |
iquette | +0.193 |
Neg logit contributions | |
� | -0.248 |
Transparency | -0.242 |
Millions | -0.236 |
Reports | -0.232 |
Thousands | -0.227 |
Token | y |
Token ID | 331 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
� | +0.091 |
Transparency | +0.089 |
Millions | +0.087 |
Reports | +0.085 |
Thousands | +0.084 |
Neg logit contributions | |
groom | -0.074 |
hander | -0.070 |
cha | -0.069 |
ladies | -0.067 |
gentlemen | -0.066 |
Token | anking |
Token ID | 15230 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
groom | +0.207 |
hander | +0.198 |
cha | +0.196 |
ladies | +0.190 |
iquette | +0.188 |
Neg logit contributions | |
� | -0.184 |
Transparency | -0.180 |
Millions | -0.175 |
Reports | -0.172 |
Thousands | -0.166 |
Token | him |
Token ID | 683 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
� | +0.167 |
Transparency | +0.163 |
Millions | +0.158 |
Reports | +0.157 |
Thousands | +0.152 |
Neg logit contributions | |
groom | -0.133 |
cha | -0.126 |
hander | -0.126 |
ladies | -0.122 |
gentlemen | -0.119 |
Token |
|
Token ID | 198 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
groom | +0.132 |
hander | +0.124 |
cha | +0.123 |
ladies | +0.119 |
iquette | +0.117 |
Neg logit contributions | |
� | -0.164 |
Transparency | -0.161 |
Millions | -0.157 |
Reports | -0.155 |
Thousands | -0.151 |
Token |
|
Token ID | 198 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
groom | +0.054 |
hander | +0.051 |
cha | +0.051 |
iquette | +0.049 |
concess | +0.049 |
Neg logit contributions | |
� | -0.052 |
Transparency | -0.052 |
Millions | -0.051 |
Reports | -0.050 |
Thousands | -0.049 |
Token | AB |
Token ID | 6242 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
� | +0.446 |
Transparency | +0.439 |
Millions | +0.428 |
Reports | +0.423 |
Thousands | +0.410 |
Neg logit contributions | |
groom | -0.419 |
hander | -0.398 |
cha | -0.394 |
ladies | -0.381 |
iquette | -0.378 |
Token | OL |
Token ID | 3535 |
Feature activation | -2.78e-01 |
Pos logit contributions | |
groom | +0.024 |
hander | +0.023 |
cha | +0.023 |
ladies | +0.022 |
iquette | +0.022 |
Neg logit contributions | |
� | -0.019 |
Transparency | -0.019 |
Millions | -0.018 |
Reports | -0.018 |
Thousands | -0.017 |
Token | IC |
Token ID | 2149 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
� | +0.013 |
Transparency | +0.012 |
Millions | +0.012 |
Reports | +0.011 |
Thousands | +0.010 |
Neg logit contributions | |
groom | -0.058 |
hander | -0.056 |
cha | -0.056 |
ladies | -0.055 |
iquette | -0.054 |
Token | WORK |
Token ID | 30936 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
� | +0.224 |
Transparency | +0.220 |
Millions | +0.214 |
Reports | +0.212 |
Thousands | +0.205 |
Neg logit contributions | |
groom | -0.199 |
hander | -0.189 |
cha | -0.187 |
ladies | -0.181 |
iquette | -0.179 |
Token | OUT |
Token ID | 12425 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
� | +0.161 |
Transparency | +0.157 |
Millions | +0.152 |
Reports | +0.149 |
Thousands | +0.143 |
Neg logit contributions | |
groom | -0.247 |
hander | -0.238 |
cha | -0.236 |
ladies | -0.230 |
iquette | -0.228 |
Token | . |
Token ID | 13 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
� | +0.302 |
Transparency | +0.296 |
Millions | +0.290 |
Reports | +0.286 |
Thousands | +0.278 |
Neg logit contributions | |
groom | -0.240 |
hander | -0.227 |
cha | -0.224 |
ladies | -0.217 |
iquette | -0.215 |
Token |
|
Token ID | 198 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
� | +0.087 |
Transparency | +0.084 |
Millions | +0.081 |
Reports | +0.080 |
Thousands | +0.077 |
Neg logit contributions | |
groom | -0.102 |
hander | -0.097 |
cha | -0.096 |
ladies | -0.094 |
iquette | -0.093 |
Token |
|
Token ID | 198 |
Feature activation | -7.73e-02 |
Pos logit contributions | |
Transparency | +0.005 |
� | +0.005 |
Millions | +0.005 |
Reports | +0.005 |
Thousands | +0.005 |
Neg logit contributions | |
groom | -0.005 |
hander | -0.005 |
cha | -0.005 |
iquette | -0.005 |
concess | -0.005 |
Token | In |
Token ID | 818 |
Feature activation | -3.10e-01 |
Pos logit contributions | |
� | +0.009 |
Transparency | +0.009 |
Millions | +0.009 |
Reports | +0.009 |
Thousands | +0.009 |
Neg logit contributions | |
groom | -0.010 |
hander | -0.010 |
cha | -0.010 |
ladies | -0.009 |
iquette | -0.009 |
Token | fact |
Token ID | 1109 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
� | +0.045 |
Transparency | +0.044 |
Millions | +0.043 |
Reports | +0.043 |
Thousands | +0.041 |
Neg logit contributions | |
groom | -0.034 |
hander | -0.032 |
cha | -0.031 |
ladies | -0.030 |
iquette | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
� | +0.304 |
Transparency | +0.299 |
Millions | +0.293 |
Reports | +0.290 |
Thousands | +0.283 |
Neg logit contributions | |
groom | -0.183 |
hander | -0.171 |
cha | -0.170 |
ladies | -0.164 |
iquette | -0.161 |
Token | he |
Token ID | 339 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
� | +0.174 |
Transparency | +0.171 |
Millions | +0.167 |
Reports | +0.164 |
Thousands | +0.160 |
Neg logit contributions | |
groom | -0.160 |
hander | -0.152 |
cha | -0.150 |
ladies | -0.146 |
iquette | -0.144 |
Token | told |
Token ID | 1297 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
� | +0.011 |
Transparency | +0.011 |
Millions | +0.011 |
Reports | +0.010 |
Thousands | +0.010 |
Neg logit contributions | |
groom | -0.009 |
hander | -0.009 |
cha | -0.009 |
ladies | -0.008 |
iquette | -0.008 |
Token | me |
Token ID | 502 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
groom | +0.068 |
hander | +0.063 |
cha | +0.062 |
ladies | +0.060 |
iquette | +0.058 |
Neg logit contributions | |
� | -0.137 |
Transparency | -0.135 |
Millions | -0.133 |
Reports | -0.131 |
Thousands | -0.128 |
Token | he |
Token ID | 339 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
groom | +0.084 |
hander | +0.080 |
cha | +0.079 |
ladies | +0.077 |
gentlemen | +0.075 |
Neg logit contributions | |
� | -0.110 |
Transparency | -0.108 |
Millions | -0.105 |
Reports | -0.104 |
Thousands | -0.101 |
Token | had |
Token ID | 550 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
� | +0.021 |
Transparency | +0.021 |
Millions | +0.020 |
Reports | +0.020 |
Thousands | +0.020 |
Neg logit contributions | |
groom | -0.018 |
hander | -0.017 |
cha | -0.017 |
ladies | -0.016 |
iquette | -0.016 |
Token | � |
Token ID | 564 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
� | +0.222 |
Transparency | +0.218 |
Millions | +0.212 |
Reports | +0.211 |
Thousands | +0.204 |
Neg logit contributions | |
groom | -0.184 |
hander | -0.172 |
cha | -0.172 |
ladies | -0.167 |
gentlemen | -0.163 |
Token | � |
Token ID | 250 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
� | +0.129 |
Transparency | +0.123 |
Millions | +0.112 |
Reports | +0.106 |
Thousands | +0.094 |
Neg logit contributions | |
groom | -0.725 |
hander | -0.705 |
cha | -0.700 |
ladies | -0.689 |
iquette | -0.686 |
Token | made |
Token ID | 9727 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
groom | +0.021 |
hander | +0.019 |
cha | +0.019 |
ladies | +0.018 |
iquette | +0.018 |
Neg logit contributions | |
� | -0.029 |
Transparency | -0.029 |
Millions | -0.028 |
Reports | -0.028 |
Thousands | -0.027 |
Token | breakfast |
Token ID | 12607 |
Feature activation | +9.06e-02 |
Pos logit contributions | |
� | +0.175 |
Transparency | +0.172 |
Millions | +0.167 |
Reports | +0.166 |
Thousands | +0.160 |
Neg logit contributions | |
groom | -0.159 |
hander | -0.150 |
cha | -0.149 |
ladies | -0.145 |
iquette | -0.143 |
Token | � |
Token ID | 447 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
groom | +0.009 |
hander | +0.009 |
cha | +0.009 |
ladies | +0.008 |
iquette | +0.008 |
Neg logit contributions | |
� | -0.014 |
Transparency | -0.013 |
Millions | -0.013 |
Reports | -0.013 |
Thousands | -0.013 |
Token | � |
Token ID | 251 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
groom | +0.033 |
hander | +0.032 |
cha | +0.032 |
ladies | +0.031 |
iquette | +0.031 |
Neg logit contributions | |
Transparency | -0.005 |
� | -0.004 |
Millions | -0.004 |
Reports | -0.004 |
Thousands | -0.003 |
Token | 16 |
Token ID | 1467 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
groom | +0.064 |
hander | +0.061 |
cha | +0.060 |
ladies | +0.059 |
iquette | +0.058 |
Neg logit contributions | |
� | -0.051 |
Transparency | -0.049 |
Millions | -0.048 |
Reports | -0.047 |
Thousands | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
� | +0.021 |
Transparency | +0.021 |
Millions | +0.021 |
Reports | +0.020 |
Thousands | +0.020 |
Neg logit contributions | |
groom | -0.014 |
hander | -0.013 |
cha | -0.013 |
ladies | -0.012 |
iquette | -0.012 |
Token | 2008 |
Token ID | 3648 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
groom | +0.034 |
hander | +0.033 |
cha | +0.032 |
ladies | +0.032 |
iquette | +0.031 |
Neg logit contributions | |
� | -0.023 |
Transparency | -0.023 |
Millions | -0.022 |
Reports | -0.022 |
Thousands | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
� | +0.111 |
Transparency | +0.108 |
Millions | +0.106 |
Reports | +0.105 |
Thousands | +0.102 |
Neg logit contributions | |
groom | -0.083 |
hander | -0.079 |
cha | -0.078 |
ladies | -0.075 |
iquette | -0.074 |
Token | World |
Token ID | 2159 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
groom | +0.030 |
hander | +0.029 |
cha | +0.029 |
ladies | +0.028 |
iquette | +0.027 |
Neg logit contributions | |
� | -0.029 |
Transparency | -0.028 |
Millions | -0.027 |
Reports | -0.027 |
Thousands | -0.026 |
Token | Net |
Token ID | 3433 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
� | +0.187 |
Transparency | +0.178 |
Millions | +0.171 |
Reports | +0.167 |
Thousands | +0.160 |
Neg logit contributions | |
groom | -0.360 |
hander | -0.348 |
cha | -0.346 |
ladies | -0.337 |
iquette | -0.336 |
Token | Daily |
Token ID | 6714 |
Feature activation | -7.21e-03 |
Pos logit contributions | |
� | +0.319 |
Transparency | +0.308 |
Millions | +0.296 |
Reports | +0.289 |
Thousands | +0.284 |
Neg logit contributions | |
groom | -0.479 |
cha | -0.458 |
hander | -0.456 |
ladies | -0.446 |
iquette | -0.445 |
Token | , |
Token ID | 11 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
� | +0.001 |
Transparency | +0.001 |
Millions | +0.001 |
Reports | +0.001 |
Thousands | +0.001 |
Neg logit contributions | |
groom | -0.001 |
hander | -0.001 |
cha | -0.001 |
ladies | -0.001 |
iquette | -0.001 |
Token | ' |
Token ID | 705 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
groom | +0.033 |
hander | +0.032 |
cha | +0.032 |
ladies | +0.031 |
gentlemen | +0.030 |
Neg logit contributions | |
� | -0.028 |
Transparency | -0.027 |
Millions | -0.026 |
Reports | -0.026 |
Thousands | -0.025 |
Token | The |
Token ID | 464 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
� | +0.289 |
Transparency | +0.281 |
Millions | +0.273 |
Reports | +0.270 |
Thousands | +0.261 |
Neg logit contributions | |
groom | -0.323 |
hander | -0.310 |
cha | -0.307 |
ladies | -0.299 |
iquette | -0.296 |
Token | Groups |
Token ID | 27441 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
groom | +0.020 |
hander | +0.019 |
cha | +0.019 |
ladies | +0.019 |
iquette | +0.018 |
Neg logit contributions | |
� | -0.013 |
Transparency | -0.013 |
Millions | -0.012 |
Reports | -0.012 |
Thousands | -0.012 |
Token | 99 |
Token ID | 2079 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
� | +0.075 |
Transparency | +0.073 |
Millions | +0.071 |
Reports | +0.069 |
Thousands | +0.067 |
Neg logit contributions | |
groom | -0.080 |
hander | -0.076 |
cha | -0.075 |
ladies | -0.074 |
gentlemen | -0.072 |
Token | model |
Token ID | 2746 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
groom | +0.252 |
hander | +0.238 |
cha | +0.236 |
ladies | +0.228 |
gentlemen | +0.224 |
Neg logit contributions | |
� | -0.330 |
Transparency | -0.324 |
Millions | -0.317 |
Reports | -0.314 |
Thousands | -0.305 |
Token | is |
Token ID | 318 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
groom | +0.025 |
hander | +0.023 |
cha | +0.022 |
ladies | +0.021 |
iquette | +0.020 |
Neg logit contributions | |
� | -0.062 |
Transparency | -0.062 |
Millions | -0.060 |
Reports | -0.060 |
Thousands | -0.059 |
Token | listed |
Token ID | 5610 |
Feature activation | +9.31e-01 |
Pos logit contributions | |
groom | +0.013 |
hander | +0.012 |
cha | +0.012 |
ladies | +0.012 |
gentlemen | +0.011 |
Neg logit contributions | |
� | -0.018 |
Transparency | -0.017 |
Millions | -0.017 |
Reports | -0.017 |
Thousands | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -6.41e-01 |
Pos logit contributions | |
groom | +0.097 |
hander | +0.092 |
cha | +0.091 |
ladies | +0.087 |
iquette | +0.086 |
Neg logit contributions | |
� | -0.139 |
Transparency | -0.137 |
Millions | -0.134 |
Reports | -0.132 |
Thousands | -0.129 |
Token | Not |
Token ID | 1892 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
� | +0.074 |
Transparency | +0.071 |
Millions | +0.069 |
Reports | +0.067 |
Thousands | +0.065 |
Neg logit contributions | |
groom | -0.088 |
hander | -0.084 |
cha | -0.084 |
ladies | -0.082 |
gentlemen | -0.081 |
Token | confirmed |
Token ID | 4999 |
Feature activation | +9.08e-01 |
Pos logit contributions | |
� | +0.612 |
Transparency | +0.602 |
Millions | +0.591 |
Reports | +0.582 |
Thousands | +0.568 |
Neg logit contributions | |
groom | -0.356 |
hander | -0.332 |
cha | -0.328 |
ladies | -0.315 |
iquette | -0.307 |
Token | but |
Token ID | 475 |
Feature activation | -7.42e-01 |
Pos logit contributions | |
groom | +0.098 |
hander | +0.092 |
cha | +0.091 |
ladies | +0.088 |
gentlemen | +0.086 |
Neg logit contributions | |
� | -0.133 |
Transparency | -0.130 |
Millions | -0.128 |
Reports | -0.125 |
Thousands | -0.122 |
Token | that |
Token ID | 326 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
� | +0.107 |
Transparency | +0.105 |
Millions | +0.103 |
Reports | +0.101 |
Thousands | +0.099 |
Neg logit contributions | |
groom | -0.081 |
hander | -0.077 |
cha | -0.076 |
ladies | -0.074 |
gentlemen | -0.072 |
Token | may |
Token ID | 743 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
� | +0.136 |
Transparency | +0.134 |
Millions | +0.132 |
Reports | +0.130 |
Thousands | +0.128 |
Neg logit contributions | |
groom | -0.061 |
hander | -0.056 |
cha | -0.055 |
ladies | -0.053 |
iquette | -0.051 |
Token | mean |
Token ID | 1612 |
Feature activation | +6.61e-01 |
Pos logit contributions | |
groom | +0.239 |
hander | +0.224 |
cha | +0.223 |
ladies | +0.216 |
iquette | +0.212 |
Neg logit contributions | |
� | -0.295 |
Transparency | -0.288 |
Millions | -0.281 |
Reports | -0.278 |
Thousands | -0.270 |
Token | ed |
Token ID | 276 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
� | +0.428 |
Transparency | +0.426 |
Millions | +0.412 |
Reports | +0.405 |
Thousands | +0.392 |
Neg logit contributions | |
groom | -0.444 |
hander | -0.422 |
cha | -0.421 |
ladies | -0.407 |
iquette | -0.401 |
Token | this |
Token ID | 428 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
� | +0.226 |
Transparency | +0.221 |
Millions | +0.216 |
Reports | +0.212 |
Thousands | +0.207 |
Neg logit contributions | |
groom | -0.184 |
cha | -0.173 |
hander | -0.172 |
ladies | -0.168 |
gentlemen | -0.164 |
Token | post |
Token ID | 1281 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
groom | +0.024 |
cha | +0.022 |
hander | +0.022 |
ladies | +0.022 |
gentlemen | +0.021 |
Neg logit contributions | |
� | -0.022 |
Transparency | -0.021 |
Millions | -0.021 |
Reports | -0.020 |
Thousands | -0.020 |
Token | saying |
Token ID | 2282 |
Feature activation | -6.27e-01 |
Pos logit contributions | |
� | +0.067 |
Transparency | +0.066 |
Millions | +0.064 |
Reports | +0.063 |
Thousands | +0.062 |
Neg logit contributions | |
groom | -0.049 |
hander | -0.046 |
cha | -0.046 |
ladies | -0.045 |
gentlemen | -0.044 |
Token | that |
Token ID | 326 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
� | +0.086 |
Transparency | +0.085 |
Millions | +0.082 |
Reports | +0.081 |
Thousands | +0.079 |
Neg logit contributions | |
groom | -0.073 |
cha | -0.068 |
hander | -0.068 |
ladies | -0.067 |
gentlemen | -0.065 |
Token | � |
Token ID | 564 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
� | +0.101 |
Transparency | +0.098 |
Millions | +0.095 |
Reports | +0.094 |
Thousands | +0.091 |
Neg logit contributions | |
groom | -0.113 |
cha | -0.108 |
hander | -0.107 |
ladies | -0.105 |
gentlemen | -0.103 |
Token | � |
Token ID | 250 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
� | +0.330 |
Transparency | +0.326 |
Millions | +0.315 |
Reports | +0.309 |
Thousands | +0.296 |
Neg logit contributions | |
groom | -0.528 |
hander | -0.506 |
cha | -0.502 |
ladies | -0.491 |
iquette | -0.487 |
Token | we |
Token ID | 732 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
groom | +0.026 |
hander | +0.025 |
cha | +0.024 |
ladies | +0.024 |
iquette | +0.023 |
Neg logit contributions | |
� | -0.027 |
Transparency | -0.026 |
Millions | -0.025 |
Reports | -0.025 |
Thousands | -0.024 |
Token | don |
Token ID | 836 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
� | +0.030 |
Transparency | +0.029 |
Millions | +0.029 |
Reports | +0.029 |
Thousands | +0.028 |
Neg logit contributions | |
groom | -0.015 |
hander | -0.014 |
cha | -0.013 |
ladies | -0.013 |
iquette | -0.013 |
Token | � |
Token ID | 447 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
groom | +0.155 |
hander | +0.147 |
cha | +0.146 |
ladies | +0.141 |
iquette | +0.141 |
Neg logit contributions | |
� | -0.130 |
Transparency | -0.127 |
Millions | -0.124 |
Reports | -0.121 |
Thousands | -0.117 |
Token | � |
Token ID | 247 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
groom | +0.022 |
hander | +0.021 |
cha | +0.021 |
ladies | +0.020 |
iquette | +0.020 |
Neg logit contributions | |
� | -0.014 |
Transparency | -0.013 |
Millions | -0.013 |
Reports | -0.013 |
Thousands | -0.012 |
Token | said |
Token ID | 531 |
Feature activation | +9.20e-01 |
Pos logit contributions | |
groom | +0.052 |
hander | +0.050 |
cha | +0.049 |
iquette | +0.047 |
ladies | +0.047 |
Neg logit contributions | |
� | -0.076 |
Transparency | -0.075 |
Millions | -0.073 |
Reports | -0.072 |
Thousands | -0.070 |
Token | this |
Token ID | 428 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
groom | +0.135 |
hander | +0.129 |
cha | +0.128 |
ladies | +0.126 |
gentlemen | +0.124 |
Neg logit contributions | |
� | -0.098 |
Transparency | -0.096 |
Millions | -0.093 |
Reports | -0.091 |
Thousands | -0.088 |
Token | new |
Token ID | 649 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
groom | +0.062 |
hander | +0.059 |
cha | +0.058 |
ladies | +0.057 |
gentlemen | +0.056 |
Neg logit contributions | |
� | -0.070 |
Transparency | -0.068 |
Millions | -0.067 |
Reports | -0.066 |
Thousands | -0.064 |
Token | evidence |
Token ID | 2370 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
� | +0.076 |
Transparency | +0.074 |
Millions | +0.072 |
Reports | +0.070 |
Thousands | +0.068 |
Neg logit contributions | |
groom | -0.114 |
hander | -0.109 |
cha | -0.109 |
ladies | -0.106 |
gentlemen | -0.105 |
Token | was |
Token ID | 373 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
� | +0.273 |
Transparency | +0.268 |
Millions | +0.263 |
Reports | +0.260 |
Thousands | +0.254 |
Neg logit contributions | |
groom | -0.162 |
hander | -0.152 |
cha | -0.149 |
ladies | -0.144 |
iquette | -0.142 |
Token | � |
Token ID | 564 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
� | +0.165 |
Transparency | +0.162 |
Millions | +0.158 |
Reports | +0.156 |
Thousands | +0.151 |
Neg logit contributions | |
groom | -0.158 |
hander | -0.148 |
cha | -0.148 |
ladies | -0.143 |
gentlemen | -0.141 |
Token | � |
Token ID | 250 |
Feature activation | +5.30e-01 |
Pos logit contributions | |
Transparency | +0.298 |
� | +0.297 |
Millions | +0.288 |
Reports | +0.283 |
Thousands | +0.271 |
Neg logit contributions | |
groom | -0.472 |
hander | -0.452 |
cha | -0.448 |
ladies | -0.437 |
iquette | -0.435 |
Token | huge |
Token ID | 40878 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
groom | +0.057 |
hander | +0.054 |
cha | +0.053 |
ladies | +0.051 |
iquette | +0.051 |
Neg logit contributions | |
� | -0.077 |
Transparency | -0.077 |
Millions | -0.075 |
Reports | -0.074 |
Thousands | -0.071 |
Token | ". |
Token ID | 1911 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
� | +0.316 |
Transparency | +0.311 |
Millions | +0.302 |
Reports | +0.299 |
Thousands | +0.289 |
Neg logit contributions | |
groom | -0.288 |
hander | -0.272 |
cha | -0.271 |
ladies | -0.264 |
gentlemen | -0.259 |
Token |
|
Token ID | 198 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
groom | +0.100 |
hander | +0.094 |
cha | +0.093 |
ladies | +0.091 |
iquette | +0.089 |
Neg logit contributions | |
� | -0.121 |
Transparency | -0.119 |
Millions | -0.116 |
Reports | -0.114 |
Thousands | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
groom | +0.027 |
hander | +0.025 |
cha | +0.025 |
iquette | +0.025 |
concess | +0.025 |
Neg logit contributions | |
� | -0.026 |
Transparency | -0.026 |
Millions | -0.025 |
Reports | -0.025 |
Thousands | -0.024 |