Token | big |
Token ID | 1263 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
OUT | +0.041 |
arrest | +0.040 |
Johnson | +0.040 |
joy | +0.039 |
Billy | +0.038 |
Neg logit contributions | |
kin | -0.026 |
ows | -0.024 |
as | -0.023 |
ily | -0.022 |
ute | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
OUT | +0.030 |
arrest | +0.030 |
Johnson | +0.029 |
istent | +0.028 |
joy | +0.028 |
Neg logit contributions | |
kin | -0.022 |
ows | -0.019 |
holes | -0.018 |
as | -0.018 |
ily | -0.017 |
Token | bill |
Token ID | 2855 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
OUT | +0.036 |
Johnson | +0.035 |
arrest | +0.035 |
joy | +0.034 |
Billy | +0.034 |
Neg logit contributions | |
kin | -0.026 |
ows | -0.023 |
as | -0.023 |
ily | -0.021 |
ute | -0.021 |
Token | owing |
Token ID | 7855 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
kin | +0.061 |
ows | +0.056 |
as | +0.055 |
ily | +0.053 |
ute | +0.053 |
Neg logit contributions | |
OUT | -0.046 |
arrest | -0.045 |
Johnson | -0.044 |
joy | -0.042 |
Billy | -0.041 |
Token | cloud |
Token ID | 6279 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
OUT | +0.323 |
arrest | +0.318 |
Johnson | +0.312 |
joy | +0.302 |
Billy | +0.298 |
Neg logit contributions | |
kin | -0.239 |
ows | -0.216 |
as | -0.209 |
ily | -0.200 |
a | -0.196 |
Token | hurd |
Token ID | 22227 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
OUT | +0.104 |
arrest | +0.102 |
Johnson | +0.100 |
Billy | +0.097 |
joy | +0.096 |
Neg logit contributions | |
kin | -0.074 |
ows | -0.066 |
as | -0.064 |
ily | -0.061 |
ute | -0.060 |
Token | ling |
Token ID | 1359 |
Feature activation | -2.04e-01 |
Pos logit contributions | |
kin | +0.390 |
ily | +0.337 |
works | +0.331 |
ute | +0.321 |
Or | +0.318 |
Neg logit contributions | |
joy | -0.263 |
ge | -0.240 |
OUT | -0.226 |
Billy | -0.221 |
ited | -0.220 |
Token | through |
Token ID | 832 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
OUT | +0.021 |
arrest | +0.020 |
joy | +0.020 |
Johnson | +0.020 |
Billy | +0.019 |
Neg logit contributions | |
kin | -0.017 |
as | -0.014 |
ows | -0.014 |
ute | -0.014 |
ily | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
OUT | +0.042 |
Johnson | +0.040 |
istent | +0.040 |
arrest | +0.039 |
joy | +0.038 |
Neg logit contributions | |
kin | -0.032 |
holes | -0.029 |
ows | -0.028 |
ily | -0.028 |
as | -0.027 |
Token | sky |
Token ID | 6766 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
OUT | +0.017 |
arrest | +0.017 |
Johnson | +0.017 |
Billy | +0.016 |
Frankie | +0.016 |
Neg logit contributions | |
kin | -0.013 |
ows | -0.011 |
as | -0.011 |
ily | -0.010 |
a | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
OUT | +0.033 |
arrest | +0.033 |
Johnson | +0.032 |
Billy | +0.031 |
joy | +0.031 |
Neg logit contributions | |
kin | -0.025 |
ows | -0.022 |
as | -0.021 |
ily | -0.021 |
ute | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +6.90e-01 |
Pos logit contributions | |
OUT | +0.019 |
arrest | +0.019 |
Johnson | +0.019 |
joy | +0.018 |
Billy | +0.018 |
Neg logit contributions | |
kin | -0.016 |
ows | -0.015 |
ute | -0.015 |
as | -0.014 |
utes | -0.014 |
Token | |
Token ID | 220 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
kin | +0.076 |
ows | +0.071 |
ily | +0.069 |
as | +0.068 |
holes | +0.067 |
Neg logit contributions | |
OUT | -0.048 |
arrest | -0.047 |
Johnson | -0.044 |
joy | -0.042 |
istent | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | +7.96e-01 |
Pos logit contributions | |
Johnson | +0.127 |
OUT | +0.126 |
arrest | +0.124 |
Billy | +0.121 |
joy | +0.118 |
Neg logit contributions | |
kin | -0.139 |
ows | -0.123 |
as | -0.121 |
holes | -0.117 |
ute | -0.117 |
Token |
|
Token ID | 198 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
kin | +0.066 |
ows | +0.060 |
as | +0.058 |
ily | +0.056 |
a | +0.055 |
Neg logit contributions | |
OUT | -0.076 |
arrest | -0.075 |
Johnson | -0.073 |
joy | -0.071 |
Billy | -0.070 |
Token | L |
Token ID | 43 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
kin | +0.045 |
ows | +0.041 |
as | +0.038 |
ute | +0.035 |
ily | +0.035 |
Neg logit contributions | |
OUT | -0.085 |
arrest | -0.083 |
Johnson | -0.082 |
Billy | -0.079 |
joy | -0.079 |
Token | ily |
Token ID | 813 |
Feature activation | +3.72e+00 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.023 |
ute | +0.021 |
utes | +0.021 |
holes | +0.020 |
Neg logit contributions | |
arrest | -0.117 |
OUT | -0.114 |
Johnson | -0.114 |
joy | -0.113 |
Billy | -0.111 |
Token | was |
Token ID | 373 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
kin | +0.277 |
ows | +0.254 |
as | +0.250 |
a | +0.249 |
ily | +0.242 |
Neg logit contributions | |
OUT | -0.379 |
arrest | -0.367 |
Johnson | -0.363 |
joy | -0.363 |
Frankie | -0.350 |
Token | sad |
Token ID | 6507 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
kin | +0.006 |
ows | +0.006 |
as | +0.005 |
ute | +0.005 |
ily | +0.005 |
Neg logit contributions | |
OUT | -0.009 |
arrest | -0.008 |
Johnson | -0.008 |
Billy | -0.008 |
joy | -0.008 |
Token | because |
Token ID | 780 |
Feature activation | -3.64e-01 |
Pos logit contributions | |
kin | +0.065 |
ows | +0.059 |
as | +0.058 |
ily | +0.055 |
a | +0.054 |
Neg logit contributions | |
OUT | -0.078 |
arrest | -0.076 |
Johnson | -0.075 |
joy | -0.073 |
Billy | -0.072 |
Token | she |
Token ID | 673 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
OUT | +0.038 |
istent | +0.035 |
arrest | +0.034 |
arse | +0.033 |
joy | +0.033 |
Neg logit contributions | |
kin | -0.030 |
ily | -0.028 |
ila | -0.025 |
holes | -0.025 |
ia | -0.023 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
OUT | +0.021 |
arrest | +0.021 |
Johnson | +0.020 |
joy | +0.020 |
Frankie | +0.019 |
Neg logit contributions | |
kin | -0.023 |
ows | -0.021 |
as | -0.021 |
a | -0.020 |
ily | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +4.15e-01 |
Pos logit contributions | |
kin | +0.038 |
ily | +0.035 |
ows | +0.035 |
holes | +0.034 |
as | +0.033 |
Neg logit contributions | |
OUT | -0.104 |
istent | -0.099 |
arrest | -0.098 |
joy | -0.095 |
Johnson | -0.094 |
Token | |
Token ID | 220 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
kin | +0.040 |
ows | +0.037 |
as | +0.036 |
ily | +0.035 |
a | +0.034 |
Neg logit contributions | |
OUT | -0.034 |
arrest | -0.033 |
Johnson | -0.033 |
joy | -0.031 |
Billy | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
Johnson | +0.150 |
OUT | +0.145 |
Billy | +0.144 |
arrest | +0.142 |
Frankie | +0.138 |
Neg logit contributions | |
kin | -0.173 |
ows | -0.151 |
as | -0.148 |
holes | -0.146 |
ute | -0.144 |
Token |
|
Token ID | 198 |
Feature activation | +4.08e-01 |
Pos logit contributions | |
kin | +0.036 |
ows | +0.033 |
as | +0.032 |
ily | +0.031 |
a | +0.030 |
Neg logit contributions | |
OUT | -0.041 |
arrest | -0.040 |
Johnson | -0.040 |
joy | -0.038 |
Billy | -0.038 |
Token | L |
Token ID | 43 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
kin | +0.032 |
ows | +0.029 |
as | +0.028 |
ily | +0.026 |
ute | +0.026 |
Neg logit contributions | |
OUT | -0.041 |
arrest | -0.040 |
Johnson | -0.040 |
Billy | -0.038 |
joy | -0.038 |
Token | ily |
Token ID | 813 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
kin | +0.025 |
ows | +0.022 |
utes | +0.020 |
ute | +0.020 |
holes | +0.019 |
Neg logit contributions | |
arrest | -0.110 |
OUT | -0.106 |
Johnson | -0.106 |
joy | -0.105 |
Billy | -0.103 |
Token | was |
Token ID | 373 |
Feature activation | -9.10e-02 |
Pos logit contributions | |
kin | +0.304 |
ows | +0.289 |
as | +0.280 |
a | +0.280 |
ily | +0.279 |
Neg logit contributions | |
OUT | -0.342 |
arrest | -0.330 |
Johnson | -0.325 |
joy | -0.323 |
istent | -0.309 |
Token | sad |
Token ID | 6507 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
OUT | +0.009 |
arrest | +0.009 |
Johnson | +0.009 |
joy | +0.008 |
Billy | +0.008 |
Neg logit contributions | |
kin | -0.007 |
ows | -0.007 |
as | -0.006 |
ily | -0.006 |
ute | -0.006 |
Token | because |
Token ID | 780 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
kin | +0.068 |
ows | +0.062 |
as | +0.061 |
ily | +0.058 |
a | +0.057 |
Neg logit contributions | |
OUT | -0.058 |
Johnson | -0.057 |
arrest | -0.057 |
joy | -0.055 |
Billy | -0.055 |
Token | she |
Token ID | 673 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
OUT | +0.049 |
arrest | +0.046 |
istent | +0.045 |
joy | +0.043 |
Johnson | +0.043 |
Neg logit contributions | |
kin | -0.048 |
ily | -0.044 |
holes | -0.041 |
as | -0.041 |
ows | -0.040 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
OUT | +0.034 |
arrest | +0.034 |
Johnson | +0.033 |
joy | +0.032 |
Billy | +0.031 |
Neg logit contributions | |
kin | -0.039 |
ows | -0.036 |
as | -0.035 |
ily | -0.034 |
a | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +6.64e-01 |
Pos logit contributions | |
OUT | +0.063 |
arrest | +0.061 |
joy | +0.058 |
Johnson | +0.057 |
istent | +0.057 |
Neg logit contributions | |
kin | -0.032 |
ows | -0.028 |
ily | -0.027 |
as | -0.026 |
a | -0.025 |
Token | |
Token ID | 220 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
kin | +0.068 |
as | +0.061 |
ows | +0.061 |
ute | +0.058 |
a | +0.058 |
Neg logit contributions | |
Johnson | -0.052 |
OUT | -0.051 |
arrest | -0.050 |
Billy | -0.049 |
joy | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
Johnson | +0.142 |
OUT | +0.140 |
arrest | +0.137 |
Billy | +0.136 |
joy | +0.132 |
Neg logit contributions | |
kin | -0.153 |
ows | -0.135 |
as | -0.132 |
holes | -0.129 |
ute | -0.127 |
Token |
|
Token ID | 198 |
Feature activation | +6.20e-01 |
Pos logit contributions | |
kin | +0.060 |
ows | +0.055 |
as | +0.053 |
ily | +0.051 |
ute | +0.050 |
Neg logit contributions | |
OUT | -0.063 |
arrest | -0.062 |
Johnson | -0.062 |
Billy | -0.059 |
joy | -0.059 |
Token | L |
Token ID | 43 |
Feature activation | +8.61e-01 |
Pos logit contributions | |
kin | +0.050 |
ows | +0.046 |
as | +0.043 |
ute | +0.042 |
ily | +0.040 |
Neg logit contributions | |
OUT | -0.059 |
arrest | -0.058 |
Johnson | -0.057 |
Billy | -0.055 |
Frankie | -0.054 |
Token | ily |
Token ID | 813 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
kin | +0.024 |
ute | +0.020 |
ows | +0.020 |
utes | +0.020 |
holes | +0.019 |
Neg logit contributions | |
arrest | -0.115 |
OUT | -0.112 |
Johnson | -0.111 |
joy | -0.111 |
Billy | -0.109 |
Token | went |
Token ID | 1816 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
kin | +0.337 |
ows | +0.315 |
as | +0.307 |
ily | +0.300 |
a | +0.298 |
Neg logit contributions | |
OUT | -0.318 |
arrest | -0.308 |
Johnson | -0.302 |
joy | -0.294 |
Frankie | -0.283 |
Token | home |
Token ID | 1363 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
kin | +0.078 |
ows | +0.075 |
as | +0.070 |
ily | +0.070 |
a | +0.067 |
Neg logit contributions | |
OUT | -0.115 |
arrest | -0.110 |
istent | -0.108 |
Johnson | -0.106 |
Frankie | -0.104 |
Token | feeling |
Token ID | 4203 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
OUT | +0.184 |
arrest | +0.170 |
istent | +0.163 |
Frankie | +0.163 |
Johnson | +0.162 |
Neg logit contributions | |
kin | -0.099 |
ows | -0.095 |
as | -0.087 |
ily | -0.085 |
a | -0.084 |
Token | very |
Token ID | 845 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
Billy | +0.178 |
arrest | +0.178 |
joy | +0.175 |
returning | +0.171 |
Johnson | +0.171 |
Neg logit contributions | |
kin | -0.215 |
as | -0.212 |
ows | -0.207 |
utes | -0.197 |
ute | -0.195 |
Token | upset |
Token ID | 9247 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
kin | +0.147 |
ows | +0.140 |
as | +0.140 |
ute | +0.128 |
a | +0.127 |
Neg logit contributions | |
arrest | -0.132 |
OUT | -0.130 |
Billy | -0.127 |
Johnson | -0.126 |
Frankie | -0.123 |
Token | work |
Token ID | 670 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
OUT | +0.037 |
arrest | +0.036 |
Johnson | +0.036 |
joy | +0.036 |
Frankie | +0.033 |
Neg logit contributions | |
kin | -0.041 |
ows | -0.037 |
ily | -0.036 |
as | -0.036 |
holes | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
OUT | +0.118 |
arrest | +0.116 |
Johnson | +0.112 |
joy | +0.109 |
Frankie | +0.109 |
Neg logit contributions | |
kin | -0.066 |
ows | -0.059 |
as | -0.056 |
ily | -0.054 |
a | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | +6.54e-01 |
Pos logit contributions | |
kin | +0.054 |
ows | +0.050 |
as | +0.049 |
ily | +0.048 |
a | +0.047 |
Neg logit contributions | |
OUT | -0.045 |
arrest | -0.044 |
Johnson | -0.042 |
joy | -0.041 |
Billy | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
kin | +0.058 |
ows | +0.052 |
as | +0.051 |
ily | +0.048 |
holes | +0.048 |
Neg logit contributions | |
OUT | -0.059 |
Johnson | -0.058 |
arrest | -0.057 |
Billy | -0.055 |
joy | -0.055 |
Token | L |
Token ID | 43 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
kin | +0.045 |
ows | +0.041 |
as | +0.039 |
ute | +0.037 |
ily | +0.037 |
Neg logit contributions | |
OUT | -0.061 |
arrest | -0.060 |
Johnson | -0.059 |
Billy | -0.057 |
Frankie | -0.056 |
Token | ily |
Token ID | 813 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.023 |
utes | +0.021 |
ute | +0.021 |
holes | +0.020 |
Neg logit contributions | |
arrest | -0.111 |
Johnson | -0.107 |
OUT | -0.107 |
joy | -0.106 |
Billy | -0.105 |
Token | 's |
Token ID | 338 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
kin | +0.324 |
ows | +0.304 |
as | +0.298 |
ily | +0.292 |
a | +0.291 |
Neg logit contributions | |
OUT | -0.326 |
arrest | -0.316 |
Johnson | -0.312 |
joy | -0.307 |
Frankie | -0.295 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
kin | +0.027 |
ows | +0.026 |
as | +0.024 |
ily | +0.022 |
a | +0.022 |
Neg logit contributions | |
arrest | -0.034 |
OUT | -0.033 |
Johnson | -0.033 |
joy | -0.032 |
Billy | -0.032 |
Token | my |
Token ID | 1820 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
as | +0.084 |
a | +0.084 |
ily | +0.083 |
ows | +0.079 |
kin | +0.078 |
Neg logit contributions | |
OUT | -0.097 |
arrest | -0.094 |
joy | -0.089 |
Johnson | -0.089 |
Frankie | -0.089 |
Token | came |
Token ID | 1625 |
Feature activation | +9.08e-01 |
Pos logit contributions | |
OUT | +0.144 |
arrest | +0.141 |
Johnson | +0.135 |
joy | +0.134 |
Frankie | +0.133 |
Neg logit contributions | |
a | -0.089 |
ily | -0.083 |
as | -0.082 |
ows | -0.077 |
kin | -0.075 |
Token | in |
Token ID | 287 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
kin | +0.054 |
as | +0.052 |
ows | +0.050 |
a | +0.049 |
ily | +0.049 |
Neg logit contributions | |
OUT | -0.105 |
arrest | -0.102 |
Johnson | -0.098 |
istent | -0.097 |
Frankie | -0.096 |
Token | gone |
Token ID | 3750 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
OUT | +0.008 |
arrest | +0.008 |
Johnson | +0.007 |
joy | +0.007 |
Billy | +0.007 |
Neg logit contributions | |
kin | -0.010 |
ows | -0.010 |
as | -0.009 |
ily | -0.009 |
holes | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
kin | +0.034 |
ows | +0.031 |
as | +0.029 |
holes | +0.028 |
ily | +0.028 |
Neg logit contributions | |
OUT | -0.047 |
arrest | -0.047 |
Johnson | -0.044 |
joy | -0.043 |
istent | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +4.30e-01 |
Pos logit contributions | |
kin | +0.034 |
ows | +0.032 |
as | +0.031 |
ily | +0.030 |
holes | +0.029 |
Neg logit contributions | |
OUT | -0.039 |
arrest | -0.038 |
Johnson | -0.036 |
joy | -0.035 |
istent | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +4.01e-01 |
Pos logit contributions | |
kin | +0.039 |
ows | +0.035 |
as | +0.034 |
holes | +0.032 |
ute | +0.032 |
Neg logit contributions | |
OUT | -0.038 |
Johnson | -0.037 |
arrest | -0.037 |
Billy | -0.036 |
joy | -0.035 |
Token | L |
Token ID | 43 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.023 |
as | +0.022 |
ily | +0.021 |
a | +0.021 |
Neg logit contributions | |
OUT | -0.046 |
arrest | -0.045 |
Johnson | -0.045 |
joy | -0.044 |
Billy | -0.043 |
Token | ily |
Token ID | 813 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
kin | +0.024 |
ows | +0.018 |
ute | +0.017 |
utes | +0.016 |
holes | +0.016 |
Neg logit contributions | |
arrest | -0.117 |
OUT | -0.114 |
Johnson | -0.114 |
Billy | -0.112 |
joy | -0.111 |
Token | 's |
Token ID | 338 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
kin | +0.323 |
ows | +0.310 |
as | +0.301 |
ily | +0.297 |
a | +0.295 |
Neg logit contributions | |
OUT | -0.322 |
arrest | -0.309 |
Johnson | -0.306 |
joy | -0.302 |
Frankie | -0.290 |
Token | friends |
Token ID | 2460 |
Feature activation | +6.57e-01 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.026 |
as | +0.024 |
works | +0.023 |
ily | +0.022 |
Neg logit contributions | |
arrest | -0.028 |
Johnson | -0.027 |
OUT | -0.027 |
joy | -0.026 |
Billy | -0.026 |
Token | tried |
Token ID | 3088 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
ily | +0.044 |
as | +0.043 |
ows | +0.042 |
a | +0.042 |
kin | +0.040 |
Neg logit contributions | |
arrest | -0.068 |
OUT | -0.066 |
Johnson | -0.063 |
joy | -0.062 |
Frankie | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
holes | +0.015 |
ows | +0.015 |
kin | +0.014 |
ily | +0.014 |
as | +0.013 |
Neg logit contributions | |
OUT | -0.039 |
istent | -0.038 |
arrest | -0.037 |
ited | -0.035 |
Frankie | -0.035 |
Token | help |
Token ID | 1037 |
Feature activation | +2.82e-01 |
Pos logit contributions | |
OUT | +0.028 |
arrest | +0.027 |
Johnson | +0.027 |
joy | +0.027 |
istent | +0.025 |
Neg logit contributions | |
kin | -0.030 |
ows | -0.028 |
as | -0.027 |
ily | -0.027 |
a | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
OUT | +0.055 |
joy | +0.054 |
arrest | +0.054 |
Johnson | +0.053 |
Frankie | +0.051 |
Neg logit contributions | |
kin | -0.034 |
ows | -0.031 |
ily | -0.030 |
as | -0.029 |
a | -0.028 |
Token | |
Token ID | 220 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
kin | +0.051 |
ows | +0.047 |
as | +0.046 |
ily | +0.045 |
a | +0.043 |
Neg logit contributions | |
OUT | -0.048 |
arrest | -0.047 |
Johnson | -0.045 |
joy | -0.044 |
Billy | -0.043 |
Token |
|
Token ID | 198 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
Johnson | +0.150 |
OUT | +0.149 |
arrest | +0.146 |
Billy | +0.143 |
joy | +0.139 |
Neg logit contributions | |
kin | -0.158 |
ows | -0.140 |
as | -0.136 |
holes | -0.133 |
ute | -0.132 |
Token |
|
Token ID | 198 |
Feature activation | +5.32e-01 |
Pos logit contributions | |
kin | +0.046 |
ows | +0.042 |
as | +0.041 |
ily | +0.040 |
a | +0.039 |
Neg logit contributions | |
OUT | -0.057 |
arrest | -0.055 |
Johnson | -0.054 |
joy | -0.052 |
Billy | -0.051 |
Token | L |
Token ID | 43 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
kin | +0.038 |
ows | +0.035 |
as | +0.033 |
ily | +0.032 |
ute | +0.031 |
Neg logit contributions | |
OUT | -0.057 |
arrest | -0.056 |
Johnson | -0.055 |
joy | -0.053 |
Billy | -0.053 |
Token | ily |
Token ID | 813 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.022 |
ute | +0.021 |
utes | +0.021 |
holes | +0.020 |
Neg logit contributions | |
arrest | -0.106 |
Johnson | -0.103 |
OUT | -0.102 |
joy | -0.101 |
Emmy | -0.099 |
Token | 's |
Token ID | 338 |
Feature activation | +2.66e-01 |
Pos logit contributions | |
kin | +0.324 |
ows | +0.304 |
as | +0.302 |
a | +0.298 |
ily | +0.298 |
Neg logit contributions | |
OUT | -0.321 |
arrest | -0.313 |
Johnson | -0.307 |
joy | -0.305 |
Frankie | -0.290 |
Token | mom |
Token ID | 1995 |
Feature activation | +9.56e-01 |
Pos logit contributions | |
kin | +0.021 |
ows | +0.019 |
as | +0.018 |
ily | +0.017 |
ute | +0.017 |
Neg logit contributions | |
arrest | -0.027 |
OUT | -0.027 |
Johnson | -0.026 |
Billy | -0.025 |
joy | -0.025 |
Token | came |
Token ID | 1625 |
Feature activation | +8.85e-01 |
Pos logit contributions | |
a | +0.070 |
as | +0.068 |
ily | +0.067 |
kin | +0.062 |
ows | +0.061 |
Neg logit contributions | |
arrest | -0.097 |
OUT | -0.096 |
joy | -0.093 |
Johnson | -0.088 |
Frankie | -0.088 |
Token | in |
Token ID | 287 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
kin | +0.050 |
as | +0.049 |
a | +0.047 |
ows | +0.046 |
ily | +0.045 |
Neg logit contributions | |
OUT | -0.104 |
arrest | -0.101 |
Johnson | -0.097 |
istent | -0.097 |
Frankie | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
kin | +0.092 |
ily | +0.080 |
ows | +0.078 |
as | +0.076 |
holes | +0.075 |
Neg logit contributions | |
OUT | -0.193 |
arrest | -0.180 |
istent | -0.180 |
Johnson | -0.179 |
Frankie | -0.176 |
Token | all |
Token ID | 477 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
kin | +0.092 |
ows | +0.086 |
as | +0.080 |
ily | +0.077 |
a | +0.075 |
Neg logit contributions | |
OUT | -0.143 |
arrest | -0.140 |
Johnson | -0.135 |
Frankie | -0.131 |
joy | -0.131 |
Token | . |
Token ID | 13 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
OUT | +0.081 |
arrest | +0.080 |
Johnson | +0.078 |
joy | +0.076 |
Billy | +0.075 |
Neg logit contributions | |
kin | -0.059 |
ows | -0.053 |
as | -0.052 |
ily | -0.049 |
a | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
kin | +0.045 |
ows | +0.041 |
as | +0.040 |
ily | +0.039 |
a | +0.038 |
Neg logit contributions | |
OUT | -0.042 |
arrest | -0.041 |
Johnson | -0.040 |
joy | -0.039 |
Billy | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
kin | +0.051 |
ows | +0.046 |
as | +0.045 |
ily | +0.043 |
ute | +0.042 |
Neg logit contributions | |
OUT | -0.055 |
arrest | -0.054 |
Johnson | -0.054 |
joy | -0.051 |
Billy | -0.051 |
Token | L |
Token ID | 43 |
Feature activation | +8.02e-01 |
Pos logit contributions | |
kin | +0.047 |
ows | +0.044 |
as | +0.042 |
ute | +0.040 |
ily | +0.039 |
Neg logit contributions | |
OUT | -0.048 |
arrest | -0.048 |
Johnson | -0.046 |
Billy | -0.044 |
joy | -0.043 |
Token | ily |
Token ID | 813 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
kin | +0.022 |
ows | +0.018 |
ute | +0.016 |
holes | +0.016 |
utes | +0.015 |
Neg logit contributions | |
arrest | -0.112 |
OUT | -0.109 |
Johnson | -0.108 |
joy | -0.106 |
Billy | -0.105 |
Token | was |
Token ID | 373 |
Feature activation | +4.32e-03 |
Pos logit contributions | |
kin | +0.298 |
ows | +0.276 |
as | +0.267 |
ily | +0.258 |
a | +0.255 |
Neg logit contributions | |
OUT | -0.355 |
arrest | -0.345 |
Johnson | -0.341 |
joy | -0.330 |
Frankie | -0.324 |
Token | so |
Token ID | 523 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
kin | +0.000 |
ows | +0.000 |
as | +0.000 |
ily | +0.000 |
a | +0.000 |
Neg logit contributions | |
OUT | -0.001 |
arrest | -0.000 |
Johnson | -0.000 |
joy | -0.000 |
Billy | -0.000 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.81e-01 |
Pos logit contributions | |
kin | +0.165 |
ows | +0.160 |
ute | +0.150 |
works | +0.149 |
as | +0.148 |
Neg logit contributions | |
arrest | -0.146 |
Johnson | -0.144 |
joy | -0.142 |
returning | -0.142 |
Billy | -0.140 |
Token | that |
Token ID | 326 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
OUT | +0.020 |
arrest | +0.019 |
Johnson | +0.019 |
Frankie | +0.019 |
istent | +0.019 |
Neg logit contributions | |
kin | -0.012 |
ows | -0.011 |
as | -0.010 |
ily | -0.010 |
holes | -0.010 |
Token | she |
Token ID | 673 |
Feature activation | -4.26e-01 |
Pos logit contributions | |
kin | +0.091 |
ily | +0.085 |
holes | +0.076 |
ila | +0.074 |
a | +0.072 |
Neg logit contributions | |
OUT | -0.109 |
arrest | -0.104 |
istent | -0.102 |
joy | -0.095 |
Johnson | -0.095 |
Token | them |
Token ID | 606 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
kin | +0.055 |
ows | +0.050 |
as | +0.049 |
ily | +0.047 |
holes | +0.046 |
Neg logit contributions | |
OUT | -0.066 |
arrest | -0.065 |
Johnson | -0.063 |
joy | -0.061 |
Billy | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
kin | +0.083 |
ows | +0.076 |
as | +0.073 |
ily | +0.068 |
a | +0.067 |
Neg logit contributions | |
OUT | -0.139 |
arrest | -0.137 |
Johnson | -0.135 |
joy | -0.131 |
Frankie | -0.129 |
Token |
|
Token ID | 198 |
Feature activation | +5.29e-01 |
Pos logit contributions | |
kin | +0.047 |
ows | +0.043 |
as | +0.042 |
ily | +0.040 |
a | +0.040 |
Neg logit contributions | |
OUT | -0.040 |
Johnson | -0.039 |
arrest | -0.039 |
Billy | -0.038 |
joy | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
kin | +0.049 |
ows | +0.045 |
as | +0.043 |
holes | +0.041 |
ute | +0.041 |
Neg logit contributions | |
Johnson | -0.044 |
OUT | -0.044 |
arrest | -0.043 |
Billy | -0.042 |
joy | -0.041 |
Token | L |
Token ID | 43 |
Feature activation | +8.27e-01 |
Pos logit contributions | |
kin | +0.032 |
ows | +0.029 |
as | +0.027 |
ute | +0.025 |
holes | +0.025 |
Neg logit contributions | |
OUT | -0.052 |
arrest | -0.051 |
Johnson | -0.051 |
Billy | -0.049 |
joy | -0.048 |
Token | ily |
Token ID | 813 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
kin | +0.023 |
ows | +0.019 |
holes | +0.016 |
ute | +0.015 |
as | +0.015 |
Neg logit contributions | |
arrest | -0.115 |
Johnson | -0.114 |
OUT | -0.114 |
Billy | -0.111 |
joy | -0.110 |
Token | 's |
Token ID | 338 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
kin | +0.274 |
ows | +0.264 |
as | +0.261 |
a | +0.259 |
ily | +0.251 |
Neg logit contributions | |
OUT | -0.366 |
arrest | -0.353 |
Johnson | -0.349 |
joy | -0.343 |
Frankie | -0.333 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
kin | +0.024 |
ows | +0.023 |
as | +0.021 |
ily | +0.021 |
ila | +0.020 |
Neg logit contributions | |
Johnson | -0.028 |
arrest | -0.028 |
OUT | -0.028 |
Billy | -0.027 |
joy | -0.027 |
Token | my |
Token ID | 1820 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
as | +0.120 |
a | +0.117 |
ows | +0.112 |
kin | +0.108 |
ily | +0.108 |
Neg logit contributions | |
arrest | -0.093 |
OUT | -0.091 |
joy | -0.084 |
Frankie | -0.076 |
Johnson | -0.074 |
Token | said |
Token ID | 531 |
Feature activation | +9.69e-02 |
Pos logit contributions | |
arrest | +0.148 |
OUT | +0.145 |
joy | +0.137 |
Johnson | +0.130 |
Frankie | +0.127 |
Neg logit contributions | |
a | -0.092 |
as | -0.087 |
ows | -0.085 |
kin | -0.083 |
ily | -0.079 |
Token | that |
Token ID | 326 |
Feature activation | +9.54e-01 |
Pos logit contributions | |
kin | +0.008 |
ows | +0.007 |
ily | +0.007 |
as | +0.007 |
a | +0.007 |
Neg logit contributions | |
OUT | -0.009 |
arrest | -0.009 |
Johnson | -0.009 |
joy | -0.009 |
istent | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +7.44e-01 |
Pos logit contributions | |
kin | +0.048 |
ows | +0.042 |
as | +0.042 |
ily | +0.040 |
ute | +0.040 |
Neg logit contributions | |
Johnson | -0.059 |
OUT | -0.059 |
arrest | -0.058 |
Billy | -0.057 |
Frankie | -0.055 |
Token | |
Token ID | 220 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
kin | +0.071 |
ows | +0.065 |
as | +0.064 |
ily | +0.061 |
a | +0.061 |
Neg logit contributions | |
OUT | -0.062 |
arrest | -0.061 |
Johnson | -0.060 |
joy | -0.057 |
Billy | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | +7.99e-01 |
Pos logit contributions | |
Johnson | +0.126 |
OUT | +0.123 |
Billy | +0.121 |
arrest | +0.120 |
Frankie | +0.117 |
Neg logit contributions | |
kin | -0.143 |
ows | -0.125 |
as | -0.122 |
holes | -0.120 |
ute | -0.120 |
Token |
|
Token ID | 198 |
Feature activation | +7.26e-01 |
Pos logit contributions | |
kin | +0.070 |
ows | +0.063 |
as | +0.061 |
ily | +0.059 |
ute | +0.058 |
Neg logit contributions | |
OUT | -0.072 |
arrest | -0.071 |
Johnson | -0.070 |
Billy | -0.067 |
joy | -0.067 |
Token | L |
Token ID | 43 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
kin | +0.054 |
ows | +0.049 |
as | +0.044 |
ute | +0.043 |
ily | +0.041 |
Neg logit contributions | |
OUT | -0.074 |
Johnson | -0.073 |
arrest | -0.073 |
Billy | -0.071 |
Frankie | -0.068 |
Token | ily |
Token ID | 813 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
kin | +0.024 |
ute | +0.020 |
ows | +0.019 |
utes | +0.018 |
holes | +0.018 |
Neg logit contributions | |
arrest | -0.107 |
OUT | -0.106 |
Johnson | -0.105 |
joy | -0.104 |
Billy | -0.102 |
Token | went |
Token ID | 1816 |
Feature activation | +9.75e-01 |
Pos logit contributions | |
kin | +0.320 |
ows | +0.294 |
as | +0.286 |
ily | +0.276 |
a | +0.272 |
Neg logit contributions | |
OUT | -0.333 |
arrest | -0.326 |
Johnson | -0.320 |
joy | -0.309 |
Billy | -0.303 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
kin | +0.070 |
ows | +0.066 |
as | +0.063 |
ily | +0.061 |
a | +0.059 |
Neg logit contributions | |
OUT | -0.103 |
arrest | -0.100 |
Johnson | -0.097 |
istent | -0.096 |
Frankie | -0.094 |
Token | to |
Token ID | 284 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
OUT | +0.131 |
arrest | +0.121 |
istent | +0.121 |
Johnson | +0.119 |
Frankie | +0.119 |
Neg logit contributions | |
kin | -0.056 |
ily | -0.050 |
ows | -0.048 |
as | -0.048 |
a | -0.047 |
Token | play |
Token ID | 711 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
OUT | +0.034 |
istent | +0.033 |
joy | +0.033 |
Johnson | +0.032 |
arrest | +0.032 |
Neg logit contributions | |
kin | -0.031 |
ily | -0.028 |
ows | -0.028 |
holes | -0.027 |
a | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
arrest | +0.051 |
Johnson | +0.050 |
OUT | +0.050 |
Billy | +0.050 |
joy | +0.049 |
Neg logit contributions | |
as | -0.046 |
kin | -0.045 |
ows | -0.042 |
a | -0.042 |
utes | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
OUT | +0.098 |
arrest | +0.097 |
Johnson | +0.094 |
joy | +0.092 |
Billy | +0.090 |
Neg logit contributions | |
kin | -0.060 |
ows | -0.053 |
as | -0.051 |
ily | -0.049 |
a | -0.048 |
Token | |
Token ID | 220 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
kin | +0.093 |
ows | +0.087 |
ily | +0.083 |
as | +0.082 |
holes | +0.079 |
Neg logit contributions | |
OUT | -0.103 |
arrest | -0.102 |
Johnson | -0.095 |
joy | -0.094 |
istent | -0.093 |
Token |
|
Token ID | 198 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
Johnson | +0.125 |
OUT | +0.123 |
Billy | +0.120 |
arrest | +0.120 |
Frankie | +0.116 |
Neg logit contributions | |
kin | -0.139 |
ows | -0.123 |
as | -0.120 |
holes | -0.118 |
ute | -0.116 |
Token |
|
Token ID | 198 |
Feature activation | +9.70e-01 |
Pos logit contributions | |
kin | +0.077 |
ows | +0.072 |
as | +0.071 |
ily | +0.068 |
a | +0.067 |
Neg logit contributions | |
OUT | -0.108 |
arrest | -0.106 |
Johnson | -0.101 |
joy | -0.099 |
istent | -0.097 |
Token | L |
Token ID | 43 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
kin | +0.062 |
ows | +0.055 |
as | +0.052 |
ily | +0.049 |
holes | +0.048 |
Neg logit contributions | |
OUT | -0.113 |
arrest | -0.111 |
Johnson | -0.109 |
Billy | -0.106 |
joy | -0.105 |
Token | ily |
Token ID | 813 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
kin | +0.028 |
holes | +0.024 |
ows | +0.023 |
ute | +0.022 |
utes | +0.022 |
Neg logit contributions | |
arrest | -0.109 |
Johnson | -0.106 |
OUT | -0.105 |
Billy | -0.105 |
Emmy | -0.104 |
Token | said |
Token ID | 531 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
kin | +0.289 |
ows | +0.263 |
as | +0.257 |
ily | +0.252 |
a | +0.251 |
Neg logit contributions | |
OUT | -0.363 |
arrest | -0.357 |
Johnson | -0.344 |
joy | -0.340 |
Frankie | -0.328 |
Token | , |
Token ID | 11 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
kin | +0.015 |
ows | +0.014 |
ily | +0.014 |
as | +0.013 |
a | +0.013 |
Neg logit contributions | |
OUT | -0.015 |
arrest | -0.014 |
Johnson | -0.014 |
joy | -0.013 |
istent | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | +3.73e-01 |
Pos logit contributions | |
kin | +0.011 |
ows | +0.010 |
ily | +0.010 |
as | +0.010 |
a | +0.009 |
Neg logit contributions | |
arrest | -0.016 |
OUT | -0.015 |
Johnson | -0.014 |
joy | -0.014 |
istent | -0.014 |
Token | Yes |
Token ID | 5297 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
kin | +0.030 |
ows | +0.028 |
as | +0.026 |
holes | +0.025 |
ily | +0.024 |
Neg logit contributions | |
OUT | -0.038 |
Johnson | -0.037 |
arrest | -0.037 |
Billy | -0.035 |
joy | -0.035 |
Token | , |
Token ID | 11 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
arrest | +0.022 |
OUT | +0.022 |
joy | +0.021 |
Johnson | +0.021 |
Frankie | +0.020 |
Neg logit contributions | |
kin | -0.019 |
ows | -0.017 |
ily | -0.016 |
ia | -0.016 |
as | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
OUT | +0.111 |
arrest | +0.108 |
Johnson | +0.105 |
joy | +0.103 |
Frankie | +0.102 |
Neg logit contributions | |
kin | -0.064 |
ows | -0.058 |
as | -0.055 |
ily | -0.053 |
a | -0.052 |
Token | |
Token ID | 220 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
kin | +0.056 |
ows | +0.052 |
as | +0.051 |
ily | +0.050 |
a | +0.049 |
Neg logit contributions | |
OUT | -0.049 |
arrest | -0.047 |
Johnson | -0.046 |
joy | -0.044 |
Billy | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
Johnson | +0.152 |
OUT | +0.149 |
Billy | +0.145 |
arrest | +0.145 |
joy | +0.140 |
Neg logit contributions | |
kin | -0.167 |
ows | -0.146 |
as | -0.143 |
holes | -0.141 |
ute | -0.139 |
Token |
|
Token ID | 198 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
kin | +0.046 |
ows | +0.042 |
as | +0.041 |
ily | +0.040 |
a | +0.039 |
Neg logit contributions | |
OUT | -0.056 |
arrest | -0.055 |
Johnson | -0.053 |
joy | -0.052 |
Billy | -0.051 |
Token | L |
Token ID | 43 |
Feature activation | +8.42e-01 |
Pos logit contributions | |
kin | +0.040 |
ows | +0.036 |
as | +0.035 |
ily | +0.033 |
ute | +0.032 |
Neg logit contributions | |
OUT | -0.056 |
arrest | -0.055 |
Johnson | -0.054 |
joy | -0.052 |
Billy | -0.052 |
Token | ily |
Token ID | 813 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
kin | +0.027 |
ows | +0.023 |
utes | +0.023 |
ute | +0.022 |
holes | +0.022 |
Neg logit contributions | |
arrest | -0.109 |
Johnson | -0.106 |
OUT | -0.104 |
joy | -0.104 |
Billy | -0.102 |
Token | 's |
Token ID | 338 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
kin | +0.318 |
ows | +0.299 |
as | +0.293 |
a | +0.289 |
ily | +0.288 |
Neg logit contributions | |
OUT | -0.327 |
arrest | -0.318 |
Johnson | -0.311 |
joy | -0.308 |
Frankie | -0.295 |
Token | mom |
Token ID | 1995 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
kin | +0.028 |
ows | +0.028 |
as | +0.025 |
ily | +0.024 |
ute | +0.024 |
Neg logit contributions | |
arrest | -0.032 |
Johnson | -0.032 |
OUT | -0.031 |
Billy | -0.031 |
joy | -0.031 |
Token | saw |
Token ID | 2497 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
a | +0.074 |
as | +0.074 |
ily | +0.072 |
ows | +0.069 |
kin | +0.067 |
Neg logit contributions | |
OUT | -0.092 |
arrest | -0.091 |
joy | -0.085 |
Frankie | -0.083 |
Johnson | -0.083 |
Token | her |
Token ID | 607 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
kin | +0.065 |
ily | +0.062 |
ows | +0.060 |
holes | +0.058 |
as | +0.058 |
Neg logit contributions | |
OUT | -0.085 |
arrest | -0.081 |
Johnson | -0.081 |
istent | -0.079 |
joy | -0.078 |
Token | sad |
Token ID | 6507 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
kin | +0.048 |
ows | +0.044 |
as | +0.042 |
ily | +0.042 |
a | +0.041 |
Neg logit contributions | |
OUT | -0.046 |
arrest | -0.044 |
Johnson | -0.043 |
Frankie | -0.042 |
Billy | -0.040 |
Token | found |
Token ID | 1043 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
OUT | +0.048 |
arrest | +0.047 |
Johnson | +0.046 |
joy | +0.043 |
Frankie | +0.043 |
Neg logit contributions | |
kin | -0.056 |
ows | -0.052 |
as | -0.050 |
ily | -0.049 |
a | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
kin | +0.076 |
ows | +0.068 |
as | +0.066 |
ute | +0.063 |
ily | +0.062 |
Neg logit contributions | |
OUT | -0.103 |
arrest | -0.102 |
Johnson | -0.101 |
joy | -0.097 |
Billy | -0.096 |
Token |
|
Token ID | 198 |
Feature activation | +5.01e-01 |
Pos logit contributions | |
kin | +0.042 |
ows | +0.039 |
as | +0.037 |
ily | +0.037 |
a | +0.036 |
Neg logit contributions | |
OUT | -0.039 |
arrest | -0.038 |
Johnson | -0.037 |
joy | -0.036 |
Billy | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
kin | +0.045 |
ows | +0.041 |
as | +0.039 |
ily | +0.037 |
holes | +0.037 |
Neg logit contributions | |
OUT | -0.044 |
Johnson | -0.044 |
arrest | -0.044 |
Billy | -0.042 |
joy | -0.041 |
Token | L |
Token ID | 43 |
Feature activation | +8.46e-01 |
Pos logit contributions | |
kin | +0.033 |
ows | +0.029 |
as | +0.027 |
ute | +0.026 |
ily | +0.026 |
Neg logit contributions | |
OUT | -0.052 |
arrest | -0.051 |
Johnson | -0.050 |
joy | -0.048 |
Billy | -0.048 |
Token | ily |
Token ID | 813 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
kin | +0.024 |
ows | +0.021 |
utes | +0.019 |
holes | +0.019 |
ute | +0.018 |
Neg logit contributions | |
arrest | -0.115 |
OUT | -0.112 |
Johnson | -0.112 |
joy | -0.110 |
Billy | -0.109 |
Token | started |
Token ID | 2067 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
kin | +0.310 |
ows | +0.290 |
as | +0.281 |
ily | +0.274 |
a | +0.271 |
Neg logit contributions | |
OUT | -0.338 |
arrest | -0.330 |
Johnson | -0.324 |
joy | -0.316 |
Frankie | -0.305 |
Token | to |
Token ID | 284 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
kin | +0.028 |
holes | +0.028 |
ows | +0.027 |
as | +0.027 |
a | +0.026 |
Neg logit contributions | |
OUT | -0.062 |
istent | -0.060 |
arrest | -0.059 |
Frankie | -0.058 |
Johnson | -0.057 |
Token | feel |
Token ID | 1254 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
OUT | +0.025 |
arrest | +0.024 |
joy | +0.023 |
Johnson | +0.022 |
istent | +0.022 |
Neg logit contributions | |
kin | -0.028 |
ows | -0.026 |
holes | -0.026 |
ily | -0.025 |
a | -0.024 |
Token | worried |
Token ID | 7960 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
arrest | +0.053 |
Billy | +0.053 |
joy | +0.053 |
OUT | +0.053 |
returning | +0.051 |
Neg logit contributions | |
kin | -0.067 |
as | -0.064 |
ows | -0.063 |
utes | -0.061 |
ute | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
kin | +0.018 |
as | +0.016 |
ows | +0.016 |
ute | +0.015 |
a | +0.015 |
Neg logit contributions | |
OUT | -0.018 |
Johnson | -0.017 |
arrest | -0.017 |
Billy | -0.017 |
joy | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
OUT | +0.068 |
arrest | +0.067 |
Johnson | +0.065 |
joy | +0.063 |
Billy | +0.062 |
Neg logit contributions | |
kin | -0.043 |
ows | -0.039 |
as | -0.037 |
ily | -0.035 |
a | -0.034 |
Token | |
Token ID | 220 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ows | +0.053 |
ily | +0.052 |
kin | +0.052 |
holes | +0.049 |
as | +0.049 |
Neg logit contributions | |
OUT | -0.045 |
arrest | -0.044 |
istent | -0.039 |
Johnson | -0.038 |
joy | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | +6.81e-01 |
Pos logit contributions | |
OUT | +0.140 |
Johnson | +0.139 |
arrest | +0.137 |
Billy | +0.132 |
joy | +0.130 |
Neg logit contributions | |
kin | -0.145 |
ows | -0.128 |
as | -0.126 |
holes | -0.121 |
ute | -0.120 |
Token |
|
Token ID | 198 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
kin | +0.057 |
ows | +0.052 |
as | +0.051 |
ily | +0.049 |
a | +0.048 |
Neg logit contributions | |
OUT | -0.064 |
arrest | -0.063 |
Johnson | -0.062 |
joy | -0.060 |
Billy | -0.059 |
Token | L |
Token ID | 43 |
Feature activation | +8.15e-01 |
Pos logit contributions | |
kin | +0.058 |
ows | +0.055 |
as | +0.052 |
ute | +0.049 |
ily | +0.049 |
Neg logit contributions | |
OUT | -0.052 |
arrest | -0.051 |
Johnson | -0.049 |
Billy | -0.047 |
joy | -0.046 |
Token | ily |
Token ID | 813 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
kin | +0.022 |
ows | +0.020 |
ute | +0.017 |
holes | +0.017 |
utes | +0.017 |
Neg logit contributions | |
arrest | -0.112 |
OUT | -0.110 |
Johnson | -0.108 |
joy | -0.107 |
Billy | -0.105 |
Token | played |
Token ID | 2826 |
Feature activation | +3.99e-01 |
Pos logit contributions | |
kin | +0.300 |
as | +0.280 |
ows | +0.277 |
a | +0.273 |
ily | +0.268 |
Neg logit contributions | |
OUT | -0.346 |
Johnson | -0.337 |
arrest | -0.332 |
joy | -0.321 |
Frankie | -0.320 |
Token | with |
Token ID | 351 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
kin | +0.032 |
as | +0.032 |
ows | +0.030 |
a | +0.028 |
works | +0.028 |
Neg logit contributions | |
arrest | -0.036 |
OUT | -0.035 |
joy | -0.035 |
Billy | -0.034 |
Johnson | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
OUT | +0.018 |
arrest | +0.017 |
Johnson | +0.017 |
joy | +0.016 |
istent | +0.016 |
Neg logit contributions | |
kin | -0.017 |
ows | -0.016 |
as | -0.015 |
ily | -0.015 |
holes | -0.015 |
Token | ball |
Token ID | 2613 |
Feature activation | +5.08e-01 |
Pos logit contributions | |
kin | +0.010 |
ows | +0.008 |
holes | +0.008 |
as | +0.008 |
ily | +0.008 |
Neg logit contributions | |
OUT | -0.020 |
arrest | -0.019 |
Johnson | -0.019 |
joy | -0.019 |
Frankie | -0.018 |
Token | for |
Token ID | 329 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
kin | +0.046 |
ows | +0.042 |
as | +0.041 |
ily | +0.039 |
ute | +0.039 |
Neg logit contributions | |
OUT | -0.044 |
arrest | -0.043 |
Johnson | -0.042 |
Billy | -0.040 |
joy | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +3.18e-01 |
Pos logit contributions | |
kin | +0.068 |
ows | +0.061 |
as | +0.059 |
ute | +0.056 |
ily | +0.055 |
Neg logit contributions | |
OUT | -0.095 |
arrest | -0.094 |
Johnson | -0.093 |
joy | -0.090 |
Billy | -0.089 |
Token | |
Token ID | 220 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
kin | +0.025 |
ows | +0.025 |
ily | +0.023 |
as | +0.023 |
holes | +0.023 |
Neg logit contributions | |
arrest | -0.031 |
OUT | -0.031 |
istent | -0.028 |
Johnson | -0.028 |
joy | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +4.41e-01 |
Pos logit contributions | |
Johnson | +0.146 |
OUT | +0.142 |
Billy | +0.140 |
arrest | +0.138 |
joy | +0.134 |
Neg logit contributions | |
kin | -0.169 |
ows | -0.148 |
as | -0.145 |
holes | -0.142 |
ute | -0.141 |
Token |
|
Token ID | 198 |
Feature activation | +4.11e-01 |
Pos logit contributions | |
kin | +0.035 |
ows | +0.032 |
as | +0.031 |
ily | +0.030 |
a | +0.030 |
Neg logit contributions | |
OUT | -0.044 |
arrest | -0.043 |
Johnson | -0.041 |
joy | -0.040 |
Billy | -0.039 |
Token | L |
Token ID | 43 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.023 |
as | +0.022 |
ily | +0.021 |
a | +0.021 |
Neg logit contributions | |
OUT | -0.048 |
arrest | -0.047 |
Johnson | -0.046 |
joy | -0.046 |
Billy | -0.044 |
Token | ily |
Token ID | 813 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
kin | +0.020 |
ows | +0.014 |
holes | +0.012 |
ute | +0.012 |
as | +0.012 |
Neg logit contributions | |
arrest | -0.116 |
OUT | -0.116 |
Johnson | -0.114 |
Billy | -0.113 |
joy | -0.112 |
Token | 's |
Token ID | 338 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
kin | +0.320 |
ows | +0.301 |
as | +0.292 |
ily | +0.284 |
a | +0.282 |
Neg logit contributions | |
OUT | -0.327 |
arrest | -0.318 |
Johnson | -0.312 |
joy | -0.304 |
Frankie | -0.294 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
kin | +0.020 |
ows | +0.019 |
as | +0.018 |
ily | +0.017 |
ute | +0.017 |
Neg logit contributions | |
arrest | -0.024 |
Johnson | -0.024 |
OUT | -0.024 |
joy | -0.023 |
Billy | -0.023 |
Token | saw |
Token ID | 2497 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
a | +0.069 |
as | +0.068 |
ily | +0.065 |
ows | +0.062 |
kin | +0.058 |
Neg logit contributions | |
OUT | -0.107 |
arrest | -0.106 |
joy | -0.100 |
istent | -0.100 |
Johnson | -0.096 |
Token | her |
Token ID | 607 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
kin | +0.049 |
ily | +0.048 |
ows | +0.046 |
holes | +0.045 |
as | +0.044 |
Neg logit contributions | |
OUT | -0.071 |
istent | -0.068 |
arrest | -0.068 |
Johnson | -0.067 |
joy | -0.066 |
Token | searching |
Token ID | 10342 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
kin | +0.043 |
ows | +0.040 |
as | +0.038 |
ily | +0.037 |
a | +0.037 |
Neg logit contributions | |
OUT | -0.043 |
arrest | -0.040 |
Johnson | -0.039 |
istent | -0.038 |
Frankie | -0.038 |
Token | it |
Token ID | 340 |
Feature activation | +7.78e-01 |
Pos logit contributions | |
OUT | +0.020 |
arrest | +0.019 |
Johnson | +0.019 |
joy | +0.018 |
istent | +0.018 |
Neg logit contributions | |
kin | -0.015 |
ily | -0.013 |
ows | -0.013 |
holes | -0.013 |
as | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
kin | +0.055 |
ows | +0.050 |
as | +0.048 |
ily | +0.046 |
a | +0.045 |
Neg logit contributions | |
OUT | -0.084 |
arrest | -0.083 |
Johnson | -0.081 |
joy | -0.079 |
Billy | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
kin | +0.013 |
ows | +0.013 |
ily | +0.012 |
as | +0.012 |
a | +0.012 |
Neg logit contributions | |
OUT | -0.016 |
arrest | -0.016 |
Johnson | -0.015 |
joy | -0.015 |
istent | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
kin | +0.022 |
ows | +0.020 |
as | +0.019 |
ily | +0.019 |
holes | +0.018 |
Neg logit contributions | |
OUT | -0.022 |
arrest | -0.022 |
Johnson | -0.022 |
Billy | -0.021 |
joy | -0.021 |
Token | L |
Token ID | 43 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
kin | +0.019 |
ows | +0.017 |
as | +0.016 |
ute | +0.015 |
holes | +0.015 |
Neg logit contributions | |
OUT | -0.025 |
arrest | -0.025 |
Johnson | -0.024 |
Billy | -0.023 |
Frankie | -0.023 |
Token | ily |
Token ID | 813 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
kin | +0.025 |
ows | +0.021 |
holes | +0.020 |
ute | +0.018 |
utes | +0.016 |
Neg logit contributions | |
arrest | -0.107 |
OUT | -0.104 |
joy | -0.103 |
Johnson | -0.102 |
returning | -0.100 |
Token | frowned |
Token ID | 32198 |
Feature activation | +8.38e-02 |
Pos logit contributions | |
kin | +0.325 |
ows | +0.307 |
as | +0.304 |
a | +0.296 |
ily | +0.296 |
Neg logit contributions | |
OUT | -0.318 |
arrest | -0.309 |
Johnson | -0.305 |
joy | -0.298 |
Frankie | -0.285 |
Token | as |
Token ID | 355 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
kin | +0.008 |
as | +0.008 |
ows | +0.008 |
ily | +0.007 |
utes | +0.007 |
Neg logit contributions | |
Johnson | -0.007 |
OUT | -0.007 |
Billy | -0.006 |
joy | -0.006 |
returning | -0.006 |
Token | she |
Token ID | 673 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
OUT | +0.052 |
arrest | +0.051 |
Johnson | +0.049 |
joy | +0.047 |
Billy | +0.046 |
Neg logit contributions | |
kin | -0.069 |
ows | -0.063 |
as | -0.062 |
ily | -0.062 |
a | -0.059 |
Token | looked |
Token ID | 3114 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
OUT | +0.046 |
arrest | +0.045 |
Johnson | +0.045 |
joy | +0.043 |
Billy | +0.042 |
Neg logit contributions | |
kin | -0.050 |
ows | -0.047 |
as | -0.045 |
ily | -0.044 |
a | -0.043 |
Token | at |
Token ID | 379 |
Feature activation | +6.29e-01 |
Pos logit contributions | |
kin | +0.051 |
ows | +0.042 |
as | +0.042 |
ute | +0.040 |
ia | +0.040 |
Neg logit contributions | |
OUT | -0.063 |
joy | -0.063 |
Billy | -0.061 |
Johnson | -0.060 |
arrest | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +3.11e-01 |
Pos logit contributions | |
kin | +0.038 |
ows | +0.034 |
as | +0.033 |
ily | +0.032 |
a | +0.031 |
Neg logit contributions | |
OUT | -0.047 |
arrest | -0.047 |
Johnson | -0.046 |
joy | -0.044 |
Billy | -0.044 |
Token | |
Token ID | 220 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
kin | +0.033 |
ows | +0.032 |
as | +0.030 |
ily | +0.030 |
holes | +0.029 |
Neg logit contributions | |
OUT | -0.022 |
arrest | -0.022 |
Johnson | -0.021 |
joy | -0.020 |
Frankie | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | +5.16e-01 |
Pos logit contributions | |
OUT | +0.161 |
Johnson | +0.159 |
arrest | +0.157 |
Billy | +0.151 |
joy | +0.150 |
Neg logit contributions | |
kin | -0.160 |
ows | -0.143 |
as | -0.140 |
ily | -0.134 |
holes | -0.133 |
Token |
|
Token ID | 198 |
Feature activation | +4.70e-01 |
Pos logit contributions | |
kin | +0.045 |
ows | +0.041 |
as | +0.040 |
ily | +0.038 |
ute | +0.038 |
Neg logit contributions | |
OUT | -0.047 |
arrest | -0.046 |
Johnson | -0.046 |
joy | -0.044 |
Billy | -0.043 |
Token | L |
Token ID | 43 |
Feature activation | +8.32e-01 |
Pos logit contributions | |
kin | +0.034 |
ows | +0.030 |
as | +0.028 |
ute | +0.027 |
holes | +0.026 |
Neg logit contributions | |
OUT | -0.050 |
arrest | -0.050 |
Johnson | -0.048 |
Billy | -0.046 |
joy | -0.046 |
Token | ily |
Token ID | 813 |
Feature activation | +3.65e+00 |
Pos logit contributions | |
kin | +0.021 |
ows | +0.016 |
holes | +0.014 |
ute | +0.014 |
utes | +0.013 |
Neg logit contributions | |
arrest | -0.119 |
OUT | -0.118 |
Johnson | -0.116 |
joy | -0.114 |
Billy | -0.113 |
Token | was |
Token ID | 373 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
kin | +0.300 |
ows | +0.284 |
as | +0.276 |
a | +0.275 |
ily | +0.270 |
Neg logit contributions | |
OUT | -0.341 |
arrest | -0.331 |
Johnson | -0.324 |
joy | -0.321 |
Frankie | -0.307 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
OUT | +0.006 |
arrest | +0.006 |
Johnson | +0.006 |
joy | +0.006 |
Billy | +0.006 |
Neg logit contributions | |
kin | -0.004 |
ows | -0.004 |
as | -0.004 |
ily | -0.004 |
a | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +4.65e-01 |
Pos logit contributions | |
OUT | +0.031 |
arrest | +0.030 |
Johnson | +0.029 |
joy | +0.028 |
Frankie | +0.028 |
Neg logit contributions | |
kin | -0.020 |
ows | -0.019 |
as | -0.018 |
ily | -0.018 |
holes | -0.017 |
Token | started |
Token ID | 2067 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
kin | +0.046 |
ows | +0.042 |
as | +0.041 |
ily | +0.040 |
ute | +0.039 |
Neg logit contributions | |
OUT | -0.037 |
arrest | -0.037 |
Johnson | -0.036 |
joy | -0.034 |
Billy | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
holes | +0.025 |
ows | +0.024 |
a | +0.024 |
as | +0.023 |
kin | +0.022 |
Neg logit contributions | |
OUT | -0.069 |
istent | -0.067 |
arrest | -0.066 |
ited | -0.065 |
Frankie | -0.064 |
Token | away |
Token ID | 1497 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
kin | +0.075 |
ows | +0.069 |
as | +0.066 |
ily | +0.064 |
holes | +0.062 |
Neg logit contributions | |
OUT | -0.115 |
arrest | -0.112 |
Johnson | -0.110 |
joy | -0.107 |
Frankie | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
kin | +0.063 |
ows | +0.058 |
as | +0.053 |
ily | +0.052 |
a | +0.051 |
Neg logit contributions | |
OUT | -0.132 |
arrest | -0.129 |
Johnson | -0.125 |
joy | -0.123 |
istent | -0.122 |
Token |
|
Token ID | 198 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
kin | +0.051 |
ows | +0.047 |
as | +0.045 |
ily | +0.044 |
holes | +0.043 |
Neg logit contributions | |
OUT | -0.057 |
arrest | -0.056 |
Johnson | -0.055 |
joy | -0.053 |
Billy | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
kin | +0.051 |
ows | +0.046 |
as | +0.045 |
holes | +0.042 |
ily | +0.042 |
Neg logit contributions | |
OUT | -0.049 |
arrest | -0.048 |
Johnson | -0.048 |
Billy | -0.046 |
joy | -0.045 |
Token | L |
Token ID | 43 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
kin | +0.034 |
ows | +0.031 |
as | +0.029 |
ute | +0.027 |
ily | +0.027 |
Neg logit contributions | |
OUT | -0.062 |
arrest | -0.061 |
Johnson | -0.060 |
Billy | -0.058 |
joy | -0.057 |
Token | ily |
Token ID | 813 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
kin | +0.024 |
ows | +0.020 |
ute | +0.019 |
utes | +0.019 |
holes | +0.019 |
Neg logit contributions | |
arrest | -0.103 |
Johnson | -0.099 |
OUT | -0.099 |
joy | -0.098 |
Billy | -0.097 |
Token | felt |
Token ID | 2936 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
kin | +0.262 |
a | +0.252 |
ows | +0.251 |
ily | +0.246 |
as | +0.245 |
Neg logit contributions | |
OUT | -0.368 |
Johnson | -0.350 |
joy | -0.350 |
arrest | -0.350 |
Frankie | -0.335 |
Token | bad |
Token ID | 2089 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
kin | +0.068 |
as | +0.064 |
ows | +0.064 |
utes | +0.058 |
ute | +0.058 |
Neg logit contributions | |
arrest | -0.069 |
OUT | -0.068 |
Billy | -0.068 |
Johnson | -0.065 |
returning | -0.065 |
Token | and |
Token ID | 290 |
Feature activation | +3.22e-01 |
Pos logit contributions | |
kin | +0.137 |
as | +0.122 |
ows | +0.120 |
ute | +0.114 |
utes | +0.114 |
Neg logit contributions | |
OUT | -0.130 |
Johnson | -0.129 |
arrest | -0.128 |
joy | -0.125 |
Billy | -0.125 |
Token | wanted |
Token ID | 2227 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
kin | +0.032 |
ows | +0.030 |
as | +0.029 |
ute | +0.028 |
a | +0.027 |
Neg logit contributions | |
OUT | -0.025 |
arrest | -0.025 |
Billy | -0.025 |
Johnson | -0.024 |
Frankie | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
kin | +0.025 |
ily | +0.024 |
ows | +0.023 |
as | +0.023 |
holes | +0.023 |
Neg logit contributions | |
OUT | -0.039 |
arrest | -0.037 |
istent | -0.036 |
Johnson | -0.036 |
joy | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
kin | +0.037 |
ows | +0.034 |
as | +0.033 |
ily | +0.032 |
a | +0.031 |
Neg logit contributions | |
OUT | -0.036 |
arrest | -0.035 |
Johnson | -0.034 |
joy | -0.033 |
Billy | -0.033 |
Token | |
Token ID | 220 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
kin | +0.015 |
ows | +0.014 |
as | +0.014 |
ily | +0.013 |
a | +0.013 |
Neg logit contributions | |
OUT | -0.016 |
arrest | -0.016 |
Johnson | -0.015 |
joy | -0.015 |
Billy | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
Johnson | +0.151 |
Billy | +0.145 |
OUT | +0.144 |
arrest | +0.140 |
Frankie | +0.138 |
Neg logit contributions | |
kin | -0.182 |
ows | -0.160 |
as | -0.156 |
holes | -0.155 |
ute | -0.153 |
Token |
|
Token ID | 198 |
Feature activation | +2.96e-01 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.024 |
as | +0.023 |
ily | +0.022 |
a | +0.022 |
Neg logit contributions | |
OUT | -0.029 |
arrest | -0.029 |
Johnson | -0.028 |
joy | -0.027 |
Billy | -0.027 |
Token | L |
Token ID | 43 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
kin | +0.021 |
ows | +0.019 |
as | +0.019 |
ily | +0.018 |
a | +0.017 |
Neg logit contributions | |
OUT | -0.032 |
arrest | -0.031 |
Johnson | -0.031 |
joy | -0.030 |
Billy | -0.030 |
Token | ily |
Token ID | 813 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
kin | +0.024 |
ute | +0.020 |
ows | +0.019 |
holes | +0.018 |
utes | +0.017 |
Neg logit contributions | |
arrest | -0.104 |
Billy | -0.102 |
OUT | -0.101 |
Johnson | -0.100 |
joy | -0.098 |
Token | felt |
Token ID | 2936 |
Feature activation | +6.95e-01 |
Pos logit contributions | |
kin | +0.319 |
ows | +0.299 |
as | +0.292 |
a | +0.286 |
ily | +0.286 |
Neg logit contributions | |
OUT | -0.326 |
arrest | -0.319 |
Johnson | -0.309 |
joy | -0.306 |
Frankie | -0.293 |
Token | sad |
Token ID | 6507 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
kin | +0.062 |
ows | +0.058 |
as | +0.057 |
ute | +0.054 |
a | +0.053 |
Neg logit contributions | |
OUT | -0.061 |
arrest | -0.061 |
Billy | -0.059 |
Johnson | -0.059 |
joy | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
kin | +0.057 |
as | +0.051 |
ows | +0.050 |
a | +0.048 |
utes | +0.046 |
Neg logit contributions | |
Johnson | -0.043 |
OUT | -0.043 |
Billy | -0.043 |
arrest | -0.042 |
joy | -0.041 |
Token | said |
Token ID | 531 |
Feature activation | -4.45e-02 |
Pos logit contributions | |
kin | +0.005 |
ows | +0.005 |
as | +0.005 |
ute | +0.005 |
ily | +0.005 |
Neg logit contributions | |
OUT | -0.005 |
arrest | -0.005 |
Johnson | -0.005 |
Billy | -0.005 |
joy | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -7.94e-02 |
Pos logit contributions | |
OUT | +0.004 |
arrest | +0.004 |
Johnson | +0.004 |
joy | +0.003 |
Frankie | +0.003 |
Neg logit contributions | |
kin | -0.004 |
ows | -0.004 |
as | -0.004 |
ily | -0.004 |
a | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +9.14e-01 |
Pos logit contributions | |
OUT | +0.118 |
arrest | +0.114 |
Johnson | +0.112 |
joy | +0.110 |
Frankie | +0.108 |
Neg logit contributions | |
kin | -0.061 |
ows | -0.054 |
as | -0.052 |
ily | -0.051 |
holes | -0.048 |
Token | |
Token ID | 220 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ows | +0.064 |
kin | +0.062 |
ily | +0.061 |
holes | +0.056 |
ute | +0.055 |
Neg logit contributions | |
OUT | -0.097 |
arrest | -0.095 |
istent | -0.089 |
joy | -0.087 |
returning | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | +9.17e-01 |
Pos logit contributions | |
Johnson | +0.131 |
OUT | +0.130 |
arrest | +0.128 |
Billy | +0.126 |
Frankie | +0.122 |
Neg logit contributions | |
kin | -0.141 |
ows | -0.125 |
as | -0.122 |
holes | -0.119 |
ute | -0.118 |
Token |
|
Token ID | 198 |
Feature activation | +8.54e-01 |
Pos logit contributions | |
kin | +0.067 |
ows | +0.062 |
as | +0.061 |
ily | +0.059 |
a | +0.058 |
Neg logit contributions | |
OUT | -0.098 |
arrest | -0.095 |
Johnson | -0.090 |
joy | -0.089 |
istent | -0.088 |
Token | L |
Token ID | 43 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
kin | +0.063 |
ows | +0.057 |
as | +0.056 |
ily | +0.054 |
a | +0.053 |
Neg logit contributions | |
OUT | -0.090 |
arrest | -0.089 |
Johnson | -0.087 |
joy | -0.086 |
Billy | -0.084 |
Token | ily |
Token ID | 813 |
Feature activation | +3.64e+00 |
Pos logit contributions | |
kin | +0.023 |
ows | +0.019 |
holes | +0.018 |
ute | +0.017 |
utes | +0.016 |
Neg logit contributions | |
arrest | -0.109 |
Johnson | -0.105 |
OUT | -0.105 |
Billy | -0.102 |
joy | -0.102 |
Token | 's |
Token ID | 338 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
kin | +0.256 |
a | +0.243 |
ows | +0.239 |
as | +0.236 |
ily | +0.234 |
Neg logit contributions | |
OUT | -0.389 |
arrest | -0.372 |
joy | -0.368 |
Johnson | -0.363 |
istent | -0.351 |
Token | grandma |
Token ID | 49890 |
Feature activation | -6.91e-01 |
Pos logit contributions | |
kin | +0.019 |
ows | +0.018 |
as | +0.017 |
ily | +0.016 |
a | +0.016 |
Neg logit contributions | |
OUT | -0.026 |
arrest | -0.026 |
Johnson | -0.025 |
joy | -0.024 |
Frankie | -0.024 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
arrest | +0.080 |
OUT | +0.079 |
joy | +0.074 |
istent | +0.070 |
Johnson | +0.069 |
Neg logit contributions | |
kin | -0.046 |
a | -0.045 |
ily | -0.043 |
ows | -0.042 |
as | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
kin | +0.021 |
ows | +0.019 |
as | +0.019 |
ily | +0.018 |
a | +0.018 |
Neg logit contributions | |
OUT | -0.018 |
arrest | -0.018 |
Johnson | -0.018 |
joy | -0.017 |
Billy | -0.017 |
Token | very |
Token ID | 845 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
kin | +0.028 |
ows | +0.026 |
as | +0.023 |
ute | +0.022 |
a | +0.022 |
Neg logit contributions | |
OUT | -0.056 |
arrest | -0.055 |
Johnson | -0.051 |
joy | -0.051 |
Frankie | -0.050 |
Token | smile |
Token ID | 8212 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
OUT | +0.040 |
arrest | +0.039 |
Johnson | +0.038 |
joy | +0.036 |
Billy | +0.036 |
Neg logit contributions | |
kin | -0.034 |
ows | -0.031 |
as | -0.029 |
ily | -0.028 |
holes | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
OUT | +0.111 |
arrest | +0.109 |
Johnson | +0.108 |
joy | +0.104 |
Frankie | +0.103 |
Neg logit contributions | |
kin | -0.073 |
ows | -0.063 |
as | -0.061 |
ily | -0.059 |
a | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
OUT | +0.008 |
arrest | +0.007 |
Johnson | +0.007 |
joy | +0.007 |
istent | +0.007 |
Neg logit contributions | |
kin | -0.007 |
ows | -0.006 |
as | -0.006 |
ily | -0.006 |
a | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
Johnson | +0.013 |
arrest | +0.013 |
OUT | +0.013 |
Billy | +0.013 |
Frankie | +0.013 |
Neg logit contributions | |
kin | -0.016 |
ows | -0.015 |
as | -0.014 |
holes | -0.014 |
ute | -0.014 |
Token | S |
Token ID | 50 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
OUT | +0.017 |
arrest | +0.017 |
Johnson | +0.016 |
Billy | +0.016 |
joy | +0.016 |
Neg logit contributions | |
kin | -0.015 |
ows | -0.015 |
as | -0.013 |
holes | -0.013 |
ute | -0.013 |
Token | aman |
Token ID | 10546 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
OUT | +0.091 |
arrest | +0.090 |
Johnson | +0.086 |
Frankie | +0.085 |
istent | +0.084 |
Neg logit contributions | |
kin | -0.097 |
ows | -0.088 |
as | -0.088 |
ily | -0.087 |
ia | -0.087 |
Token | tha |
Token ID | 12898 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
OUT | +0.373 |
arrest | +0.352 |
Johnson | +0.340 |
istent | +0.332 |
joy | +0.331 |
Neg logit contributions | |
kin | -0.524 |
ows | -0.478 |
as | -0.469 |
a | -0.464 |
ily | -0.460 |
Token | laughed |
Token ID | 13818 |
Feature activation | +6.34e-01 |
Pos logit contributions | |
OUT | +0.220 |
arrest | +0.212 |
Johnson | +0.211 |
joy | +0.204 |
Frankie | +0.200 |
Neg logit contributions | |
kin | -0.180 |
ows | -0.171 |
as | -0.165 |
a | -0.163 |
ily | -0.161 |
Token | , |
Token ID | 11 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
kin | +0.054 |
ows | +0.048 |
as | +0.048 |
ily | +0.045 |
ute | +0.044 |
Neg logit contributions | |
OUT | -0.060 |
Johnson | -0.058 |
arrest | -0.057 |
Billy | -0.057 |
joy | -0.057 |
Token | happy |
Token ID | 3772 |
Feature activation | -5.15e-01 |
Pos logit contributions | |
OUT | +0.006 |
arrest | +0.006 |
Johnson | +0.006 |
istent | +0.006 |
joy | +0.006 |
Neg logit contributions | |
kin | -0.006 |
ows | -0.005 |
as | -0.005 |
a | -0.005 |
ily | -0.005 |
Token | that |
Token ID | 326 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
OUT | +0.044 |
arrest | +0.044 |
Johnson | +0.044 |
joy | +0.043 |
Billy | +0.043 |
Neg logit contributions | |
kin | -0.047 |
ows | -0.045 |
as | -0.044 |
ute | -0.041 |
a | -0.040 |
Token | € |
Token ID | 26391 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
kin | +0.016 |
ows | +0.015 |
as | +0.015 |
ily | +0.014 |
holes | +0.014 |
Neg logit contributions | |
OUT | -0.015 |
arrest | -0.014 |
Johnson | -0.014 |
joy | -0.014 |
Billy | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
OUT | +0.073 |
arrest | +0.071 |
Johnson | +0.069 |
Frankie | +0.068 |
returning | +0.066 |
Neg logit contributions | |
kin | -0.067 |
ows | -0.065 |
as | -0.063 |
a | -0.061 |
ute | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
arrest | +0.009 |
OUT | +0.009 |
Johnson | +0.008 |
joy | +0.008 |
istent | +0.008 |
Neg logit contributions | |
kin | -0.009 |
ows | -0.008 |
ily | -0.008 |
as | -0.008 |
a | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
OUT | +0.008 |
arrest | +0.008 |
Johnson | +0.008 |
joy | +0.008 |
Billy | +0.008 |
Neg logit contributions | |
kin | -0.008 |
ows | -0.007 |
as | -0.007 |
ily | -0.007 |
ute | -0.006 |
Token | S |
Token ID | 50 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
OUT | +0.015 |
arrest | +0.015 |
Johnson | +0.015 |
joy | +0.014 |
Billy | +0.014 |
Neg logit contributions | |
kin | -0.012 |
as | -0.011 |
ows | -0.011 |
ily | -0.011 |
a | -0.010 |
Token | aman |
Token ID | 10546 |
Feature activation | -4.91e+00 |
Pos logit contributions | |
arrest | +0.114 |
OUT | +0.112 |
Johnson | +0.108 |
Frankie | +0.107 |
Billy | +0.105 |
Neg logit contributions | |
kin | -0.077 |
as | -0.071 |
ows | -0.071 |
ily | -0.069 |
ia | -0.068 |
Token | tha |
Token ID | 12898 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
OUT | +0.343 |
arrest | +0.332 |
Johnson | +0.322 |
joy | +0.308 |
Billy | +0.303 |
Neg logit contributions | |
kin | -0.533 |
ows | -0.496 |
as | -0.485 |
ily | -0.473 |
a | -0.470 |
Token | was |
Token ID | 373 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
OUT | +0.230 |
arrest | +0.224 |
Johnson | +0.220 |
joy | +0.214 |
Billy | +0.208 |
Neg logit contributions | |
kin | -0.180 |
ows | -0.169 |
as | -0.164 |
a | -0.159 |
ily | -0.158 |
Token | amazed |
Token ID | 24562 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
OUT | +0.058 |
arrest | +0.057 |
Johnson | +0.056 |
joy | +0.054 |
Billy | +0.054 |
Neg logit contributions | |
kin | -0.045 |
ows | -0.041 |
as | -0.039 |
ily | -0.038 |
a | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
OUT | +0.120 |
Johnson | +0.120 |
joy | +0.120 |
Billy | +0.119 |
arrest | +0.116 |
Neg logit contributions | |
kin | -0.116 |
ows | -0.103 |
as | -0.096 |
ute | -0.093 |
ily | -0.093 |
Token | she |
Token ID | 673 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
kin | +0.017 |
ows | +0.016 |
as | +0.015 |
ily | +0.015 |
a | +0.015 |
Neg logit contributions | |
OUT | -0.018 |
arrest | -0.018 |
Johnson | -0.017 |
joy | -0.016 |
Billy | -0.016 |
Token | before |
Token ID | 878 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
kin | +0.062 |
ows | +0.056 |
as | +0.055 |
ily | +0.053 |
ute | +0.052 |
Neg logit contributions | |
OUT | -0.080 |
arrest | -0.079 |
Johnson | -0.077 |
joy | -0.075 |
Billy | -0.074 |
Token | dinner |
Token ID | 8073 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
OUT | +0.067 |
arrest | +0.063 |
istent | +0.063 |
joy | +0.061 |
Johnson | +0.061 |
Neg logit contributions | |
kin | -0.054 |
holes | -0.047 |
ily | -0.045 |
ows | -0.045 |
a | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
OUT | +0.039 |
arrest | +0.038 |
Johnson | +0.038 |
Billy | +0.036 |
joy | +0.036 |
Neg logit contributions | |
kin | -0.040 |
ows | -0.037 |
as | -0.037 |
ily | -0.035 |
ute | -0.035 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
kin | +0.020 |
ows | +0.019 |
ily | +0.019 |
as | +0.018 |
a | +0.018 |
Neg logit contributions | |
OUT | -0.023 |
arrest | -0.023 |
Johnson | -0.021 |
joy | -0.021 |
istent | -0.020 |
Token | S |
Token ID | 50 |
Feature activation | -9.96e-01 |
Pos logit contributions | |
kin | +0.092 |
ows | +0.084 |
as | +0.081 |
ily | +0.080 |
a | +0.077 |
Neg logit contributions | |
OUT | -0.113 |
arrest | -0.110 |
Johnson | -0.106 |
joy | -0.103 |
istent | -0.102 |
Token | aman |
Token ID | 10546 |
Feature activation | -4.87e+00 |
Pos logit contributions | |
OUT | +0.079 |
arrest | +0.076 |
istent | +0.072 |
joy | +0.067 |
Johnson | +0.067 |
Neg logit contributions | |
kin | -0.099 |
ia | -0.096 |
ily | -0.093 |
a | -0.093 |
as | -0.092 |
Token | tha |
Token ID | 12898 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
OUT | +0.399 |
istent | +0.368 |
arrest | +0.357 |
joy | +0.352 |
Johnson | +0.334 |
Neg logit contributions | |
kin | -0.476 |
a | -0.433 |
ows | -0.424 |
ia | -0.423 |
as | -0.419 |
Token | was |
Token ID | 373 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
OUT | +0.294 |
joy | +0.276 |
arrest | +0.273 |
istent | +0.272 |
Johnson | +0.271 |
Neg logit contributions | |
kin | -0.114 |
a | -0.113 |
ows | -0.109 |
as | -0.108 |
ily | -0.108 |
Token | a |
Token ID | 257 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
OUT | +0.029 |
arrest | +0.029 |
Johnson | +0.028 |
joy | +0.027 |
Billy | +0.027 |
Neg logit contributions | |
kin | -0.031 |
ows | -0.029 |
as | -0.028 |
ily | -0.026 |
ute | -0.026 |
Token | lazy |
Token ID | 16931 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
OUT | +0.046 |
arrest | +0.045 |
Johnson | +0.045 |
joy | +0.044 |
Billy | +0.043 |
Neg logit contributions | |
kin | -0.039 |
ows | -0.038 |
as | -0.035 |
ily | -0.034 |
holes | -0.033 |
Token | girl |
Token ID | 2576 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
kin | +0.042 |
ows | +0.039 |
as | +0.038 |
ily | +0.035 |
ute | +0.035 |
Neg logit contributions | |
OUT | -0.054 |
Johnson | -0.053 |
arrest | -0.052 |
Billy | -0.051 |
joy | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
kin | +0.039 |
ows | +0.035 |
as | +0.034 |
ily | +0.033 |
a | +0.032 |
Neg logit contributions | |
OUT | -0.048 |
arrest | -0.047 |
Johnson | -0.046 |
joy | -0.045 |
Billy | -0.044 |
Token | |
Token ID | 220 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
kin | +0.037 |
ows | +0.035 |
as | +0.034 |
ily | +0.034 |
a | +0.032 |
Neg logit contributions | |
OUT | -0.041 |
arrest | -0.040 |
Johnson | -0.038 |
joy | -0.038 |
istent | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
OUT | +0.184 |
arrest | +0.180 |
Johnson | +0.178 |
joy | +0.171 |
Billy | +0.170 |
Neg logit contributions | |
kin | -0.170 |
ows | -0.154 |
as | -0.150 |
ily | -0.144 |
ute | -0.142 |
Token |
|
Token ID | 198 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
kin | +0.025 |
ows | +0.022 |
as | +0.022 |
ily | +0.021 |
holes | +0.021 |
Neg logit contributions | |
OUT | -0.026 |
arrest | -0.025 |
Johnson | -0.025 |
joy | -0.024 |
Billy | -0.024 |
Token | Exc |
Token ID | 40127 |
Feature activation | +3.63e-01 |
Pos logit contributions | |
kin | +0.017 |
ows | +0.015 |
as | +0.014 |
holes | +0.014 |
ily | +0.014 |
Neg logit contributions | |
OUT | -0.021 |
arrest | -0.021 |
Johnson | -0.021 |
joy | -0.020 |
Billy | -0.020 |
Token | ited |
Token ID | 863 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
kin | +0.047 |
a | +0.045 |
as | +0.044 |
ily | +0.043 |
ila | +0.043 |
Neg logit contributions | |
OUT | -0.015 |
Johnson | -0.015 |
arrest | -0.015 |
Billy | -0.014 |
Frankie | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
OUT | +0.480 |
arrest | +0.471 |
Johnson | +0.464 |
joy | +0.444 |
istent | +0.435 |
Neg logit contributions | |
kin | -0.374 |
ows | -0.347 |
as | -0.336 |
ily | -0.333 |
a | -0.321 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
OUT | +0.002 |
arrest | +0.002 |
istent | +0.002 |
Johnson | +0.002 |
joy | +0.002 |
Neg logit contributions | |
kin | -0.002 |
ows | -0.002 |
ily | -0.002 |
a | -0.002 |
as | -0.002 |
Token | followed |
Token ID | 3940 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
kin | +0.004 |
ows | +0.003 |
as | +0.003 |
a | +0.003 |
ily | +0.003 |
Neg logit contributions | |
OUT | -0.004 |
arrest | -0.004 |
Johnson | -0.004 |
joy | -0.004 |
Frankie | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
kin | +0.017 |
ily | +0.015 |
holes | +0.015 |
ows | +0.015 |
ute | +0.014 |
Neg logit contributions | |
OUT | -0.025 |
istent | -0.023 |
Johnson | -0.023 |
arrest | -0.023 |
joy | -0.023 |
Token | instructions |
Token ID | 7729 |
Feature activation | +5.52e-01 |
Pos logit contributions | |
OUT | +0.018 |
arrest | +0.018 |
Johnson | +0.018 |
Billy | +0.017 |
joy | +0.017 |
Neg logit contributions | |
kin | -0.013 |
ows | -0.012 |
as | -0.012 |
ily | -0.011 |
ute | -0.011 |
Token | sky |
Token ID | 6766 |
Feature activation | -2.99e-01 |
Pos logit contributions | |
arrest | +0.031 |
Johnson | +0.030 |
OUT | +0.030 |
Billy | +0.029 |
Smith | +0.028 |
Neg logit contributions | |
kin | -0.028 |
ows | -0.026 |
as | -0.026 |
ute | -0.023 |
utes | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
OUT | +0.031 |
arrest | +0.030 |
Johnson | +0.030 |
joy | +0.029 |
Billy | +0.029 |
Neg logit contributions | |
kin | -0.022 |
ows | -0.020 |
as | -0.019 |
ily | -0.018 |
ute | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
kin | +0.014 |
ows | +0.013 |
as | +0.013 |
ily | +0.013 |
a | +0.012 |
Neg logit contributions | |
OUT | -0.015 |
arrest | -0.015 |
Johnson | -0.014 |
joy | -0.014 |
Frankie | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
kin | +0.022 |
ows | +0.020 |
as | +0.019 |
holes | +0.019 |
ute | +0.018 |
Neg logit contributions | |
OUT | -0.022 |
arrest | -0.022 |
Johnson | -0.022 |
Billy | -0.021 |
joy | -0.021 |
Token | Exc |
Token ID | 40127 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
kin | +0.013 |
ows | +0.012 |
as | +0.011 |
ily | +0.011 |
a | +0.011 |
Neg logit contributions | |
OUT | -0.017 |
arrest | -0.017 |
Johnson | -0.016 |
joy | -0.016 |
Billy | -0.016 |
Token | ited |
Token ID | 863 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
kin | +0.046 |
a | +0.045 |
as | +0.044 |
ila | +0.044 |
OK | +0.043 |
Neg logit contributions | |
OUT | -0.020 |
arrest | -0.019 |
Frankie | -0.019 |
Billy | -0.018 |
Johnson | -0.018 |
Token | ly |
Token ID | 306 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
OUT | +0.474 |
arrest | +0.464 |
Johnson | +0.458 |
joy | +0.435 |
istent | +0.434 |
Neg logit contributions | |
kin | -0.369 |
ows | -0.344 |
as | -0.333 |
ily | -0.332 |
holes | -0.320 |
Token | , |
Token ID | 11 |
Feature activation | +5.47e-02 |
Pos logit contributions | |
kin | +0.090 |
a | +0.082 |
ily | +0.081 |
ows | +0.078 |
as | +0.076 |
Neg logit contributions | |
OUT | -0.125 |
arse | -0.123 |
Johnson | -0.122 |
istent | -0.121 |
arrest | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
kin | +0.005 |
ily | +0.004 |
ows | +0.004 |
a | +0.004 |
as | +0.004 |
Neg logit contributions | |
OUT | -0.005 |
arrest | -0.005 |
Johnson | -0.005 |
istent | -0.005 |
joy | -0.005 |
Token | two |
Token ID | 734 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
arrest | +0.015 |
Johnson | +0.015 |
OUT | +0.015 |
joy | +0.015 |
Billy | +0.015 |
Neg logit contributions | |
ows | -0.015 |
kin | -0.014 |
as | -0.014 |
ute | -0.013 |
a | -0.013 |
Token | girls |
Token ID | 4813 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
OUT | +0.076 |
arrest | +0.075 |
Johnson | +0.074 |
joy | +0.071 |
Billy | +0.071 |
Neg logit contributions | |
kin | -0.055 |
ows | -0.050 |
as | -0.048 |
ily | -0.046 |
ute | -0.045 |
Token | it |
Token ID | 340 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
OUT | +0.064 |
Billy | +0.063 |
arrest | +0.062 |
joy | +0.059 |
Johnson | +0.058 |
Neg logit contributions | |
kin | -0.066 |
ows | -0.065 |
as | -0.064 |
a | -0.057 |
auc | -0.056 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
kin | +0.031 |
ows | +0.028 |
as | +0.028 |
ily | +0.026 |
a | +0.026 |
Neg logit contributions | |
OUT | -0.031 |
Johnson | -0.030 |
arrest | -0.030 |
Billy | -0.029 |
returning | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
kin | +0.023 |
ows | +0.022 |
ily | +0.021 |
as | +0.021 |
a | +0.020 |
Neg logit contributions | |
OUT | -0.028 |
arrest | -0.027 |
Johnson | -0.026 |
istent | -0.025 |
joy | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
kin | +0.028 |
ows | +0.025 |
as | +0.024 |
holes | +0.023 |
ily | +0.023 |
Neg logit contributions | |
OUT | -0.028 |
arrest | -0.028 |
Johnson | -0.028 |
joy | -0.026 |
Billy | -0.026 |
Token | Exc |
Token ID | 40127 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
kin | +0.020 |
ows | +0.018 |
as | +0.017 |
ily | +0.016 |
holes | +0.016 |
Neg logit contributions | |
OUT | -0.025 |
arrest | -0.025 |
Johnson | -0.025 |
joy | -0.024 |
Billy | -0.024 |
Token | ited |
Token ID | 863 |
Feature activation | -4.80e+00 |
Pos logit contributions | |
kin | +0.053 |
a | +0.050 |
as | +0.049 |
ows | +0.049 |
ily | +0.049 |
Neg logit contributions | |
OUT | -0.017 |
arrest | -0.016 |
Johnson | -0.016 |
Billy | -0.015 |
Frankie | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
OUT | +0.464 |
arrest | +0.455 |
Johnson | +0.447 |
joy | +0.430 |
Billy | +0.424 |
Neg logit contributions | |
kin | -0.394 |
ows | -0.361 |
as | -0.350 |
ily | -0.338 |
a | -0.331 |
Token | Sammy |
Token ID | 44067 |
Feature activation | -6.06e-01 |
Pos logit contributions | |
kin | +0.002 |
ily | +0.002 |
ows | +0.002 |
a | +0.002 |
as | +0.002 |
Neg logit contributions | |
OUT | -0.003 |
arrest | -0.003 |
istent | -0.002 |
Johnson | -0.002 |
joy | -0.002 |
Token | eagerly |
Token ID | 30130 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
OUT | +0.061 |
arrest | +0.060 |
Johnson | +0.059 |
joy | +0.058 |
Frankie | +0.055 |
Neg logit contributions | |
kin | -0.046 |
ows | -0.043 |
as | -0.041 |
a | -0.040 |
ily | -0.040 |
Token | agreed |
Token ID | 4987 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
OUT | +0.013 |
arrest | +0.012 |
Johnson | +0.012 |
joy | +0.012 |
Frankie | +0.011 |
Neg logit contributions | |
kin | -0.012 |
ows | -0.011 |
as | -0.011 |
ily | -0.010 |
a | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +7.46e-02 |
Pos logit contributions | |
kin | +0.033 |
ows | +0.030 |
as | +0.028 |
ily | +0.028 |
a | +0.027 |
Neg logit contributions | |
OUT | -0.051 |
arrest | -0.050 |
Johnson | -0.049 |
joy | -0.047 |
Frankie | -0.046 |
Token | together |
Token ID | 1978 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
kin | +0.076 |
as | +0.067 |
ows | +0.067 |
ily | +0.066 |
ute | +0.065 |
Neg logit contributions | |
arrest | -0.073 |
Johnson | -0.071 |
OUT | -0.070 |
Billy | -0.068 |
Smith | -0.066 |
Token | !" |
Token ID | 2474 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
OUT | +0.016 |
arrest | +0.016 |
Johnson | +0.016 |
joy | +0.015 |
Billy | +0.015 |
Neg logit contributions | |
kin | -0.015 |
ows | -0.013 |
as | -0.013 |
ily | -0.012 |
ute | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
OUT | +0.005 |
arrest | +0.005 |
joy | +0.005 |
Johnson | +0.005 |
istent | +0.004 |
Neg logit contributions | |
kin | -0.005 |
ows | -0.005 |
as | -0.004 |
ily | -0.004 |
a | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
kin | +0.041 |
ows | +0.037 |
as | +0.036 |
holes | +0.034 |
ily | +0.034 |
Neg logit contributions | |
OUT | -0.040 |
Johnson | -0.039 |
arrest | -0.039 |
Billy | -0.038 |
Frankie | -0.037 |
Token | Exc |
Token ID | 40127 |
Feature activation | +4.82e-01 |
Pos logit contributions | |
kin | +0.027 |
ows | +0.025 |
as | +0.023 |
ily | +0.022 |
holes | +0.022 |
Neg logit contributions | |
OUT | -0.037 |
arrest | -0.037 |
Johnson | -0.036 |
Billy | -0.035 |
Frankie | -0.034 |
Token | ited |
Token ID | 863 |
Feature activation | -4.78e+00 |
Pos logit contributions | |
kin | +0.066 |
as | +0.062 |
ows | +0.061 |
a | +0.061 |
ily | +0.061 |
Neg logit contributions | |
OUT | -0.019 |
arrest | -0.018 |
Johnson | -0.017 |
Billy | -0.016 |
Frankie | -0.016 |
Token | ly |
Token ID | 306 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
OUT | +0.482 |
arrest | +0.472 |
Johnson | +0.463 |
joy | +0.446 |
istent | +0.437 |
Neg logit contributions | |
kin | -0.361 |
ows | -0.338 |
as | -0.324 |
ily | -0.319 |
holes | -0.308 |
Token | , |
Token ID | 11 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
kin | +0.084 |
ows | +0.078 |
ily | +0.074 |
a | +0.071 |
holes | +0.068 |
Neg logit contributions | |
OUT | -0.131 |
arse | -0.129 |
Johnson | -0.127 |
joy | -0.127 |
istent | -0.124 |
Token | Sally |
Token ID | 25737 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
kin | +0.005 |
ily | +0.004 |
holes | +0.004 |
Lia | +0.004 |
ows | +0.004 |
Neg logit contributions | |
OUT | -0.004 |
istent | -0.004 |
arrest | -0.004 |
arse | -0.004 |
joy | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
OUT | +0.167 |
arrest | +0.162 |
Johnson | +0.160 |
joy | +0.159 |
istent | +0.151 |
Neg logit contributions | |
kin | -0.128 |
ows | -0.123 |
as | -0.118 |
a | -0.113 |
ily | -0.112 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
kin | +0.034 |
ily | +0.031 |
ows | +0.031 |
as | +0.030 |
a | +0.030 |
Neg logit contributions | |
OUT | -0.025 |
arrest | -0.023 |
istent | -0.023 |
joy | -0.022 |
Johnson | -0.022 |
Token | distance |
Token ID | 5253 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
arrest | +0.028 |
Johnson | +0.026 |
OUT | +0.025 |
Billy | +0.025 |
joy | +0.025 |
Neg logit contributions | |
kin | -0.029 |
ows | -0.029 |
as | -0.028 |
ute | -0.025 |
utes | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +7.23e-02 |
Pos logit contributions | |
OUT | +0.176 |
arrest | +0.173 |
Johnson | +0.168 |
joy | +0.165 |
Billy | +0.162 |
Neg logit contributions | |
kin | -0.104 |
ows | -0.094 |
as | -0.090 |
ily | -0.087 |
holes | -0.084 |
Token |
|
Token ID | 198 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
kin | +0.006 |
ows | +0.006 |
ily | +0.006 |
as | +0.005 |
holes | +0.005 |
Neg logit contributions | |
arrest | -0.007 |
OUT | -0.007 |
Johnson | -0.006 |
joy | -0.006 |
istent | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +9.78e-03 |
Pos logit contributions | |
kin | +0.004 |
ows | +0.003 |
as | +0.003 |
holes | +0.003 |
ute | +0.003 |
Neg logit contributions | |
OUT | -0.004 |
arrest | -0.004 |
Johnson | -0.004 |
joy | -0.003 |
Billy | -0.003 |
Token | Exc |
Token ID | 40127 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
kin | +0.001 |
ows | +0.001 |
as | +0.001 |
ily | +0.001 |
holes | +0.001 |
Neg logit contributions | |
OUT | -0.001 |
arrest | -0.001 |
Johnson | -0.001 |
joy | -0.001 |
Billy | -0.001 |
Token | ited |
Token ID | 863 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
kin | +0.056 |
a | +0.053 |
as | +0.053 |
ila | +0.053 |
ily | +0.053 |
Neg logit contributions | |
OUT | -0.022 |
arrest | -0.021 |
Johnson | -0.020 |
istent | -0.019 |
Billy | -0.019 |
Token | ly |
Token ID | 306 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
OUT | +0.465 |
arrest | +0.456 |
Johnson | +0.448 |
istent | +0.430 |
joy | +0.427 |
Neg logit contributions | |
kin | -0.363 |
ows | -0.341 |
ily | -0.334 |
as | -0.329 |
holes | -0.321 |
Token | , |
Token ID | 11 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
kin | +0.087 |
ily | +0.081 |
a | +0.080 |
ows | +0.077 |
ia | +0.074 |
Neg logit contributions | |
istent | -0.129 |
OUT | -0.129 |
arse | -0.127 |
Johnson | -0.124 |
joy | -0.123 |
Token | she |
Token ID | 673 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
kin | +0.005 |
ily | +0.005 |
ows | +0.005 |
a | +0.004 |
holes | +0.004 |
Neg logit contributions | |
OUT | -0.005 |
arrest | -0.005 |
istent | -0.005 |
Johnson | -0.005 |
joy | -0.004 |
Token | ran |
Token ID | 4966 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
OUT | +0.062 |
arrest | +0.061 |
Billy | +0.059 |
Johnson | +0.058 |
Frankie | +0.056 |
Neg logit contributions | |
kin | -0.087 |
ows | -0.079 |
as | -0.075 |
ute | -0.075 |
ily | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | -5.84e-01 |
Pos logit contributions | |
kin | +0.070 |
ows | +0.063 |
as | +0.061 |
ily | +0.058 |
ute | +0.057 |
Neg logit contributions | |
OUT | -0.081 |
arrest | -0.080 |
Johnson | -0.079 |
joy | -0.076 |
Billy | -0.075 |
Token | on |
Token ID | 319 |
Feature activation | +2.34e-01 |
Pos logit contributions | |
OUT | +0.181 |
arrest | +0.177 |
Johnson | +0.176 |
Billy | +0.168 |
returning | +0.166 |
Neg logit contributions | |
kin | -0.140 |
as | -0.123 |
ows | -0.122 |
ily | -0.116 |
a | -0.113 |
Token | a |
Token ID | 257 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
kin | +0.018 |
ows | +0.016 |
as | +0.015 |
ily | +0.015 |
holes | +0.015 |
Neg logit contributions | |
OUT | -0.024 |
arrest | -0.023 |
Johnson | -0.023 |
joy | -0.022 |
Billy | -0.022 |
Token | tree |
Token ID | 5509 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
OUT | +0.032 |
Johnson | +0.031 |
arrest | +0.031 |
Frankie | +0.029 |
Billy | +0.029 |
Neg logit contributions | |
kin | -0.032 |
ows | -0.030 |
as | -0.029 |
ily | -0.028 |
ila | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
OUT | +0.144 |
arrest | +0.141 |
Johnson | +0.139 |
joy | +0.136 |
Billy | +0.134 |
Neg logit contributions | |
kin | -0.083 |
ows | -0.075 |
as | -0.071 |
ily | -0.067 |
holes | -0.067 |
Token | Exc |
Token ID | 25268 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
arrest | +0.011 |
OUT | +0.011 |
Johnson | +0.010 |
joy | +0.010 |
istent | +0.010 |
Neg logit contributions | |
kin | -0.009 |
ows | -0.009 |
holes | -0.009 |
ily | -0.009 |
ute | -0.008 |
Token | ited |
Token ID | 863 |
Feature activation | -4.76e+00 |
Pos logit contributions | |
OUT | +0.005 |
arrest | +0.004 |
Johnson | +0.004 |
joy | +0.003 |
Billy | +0.003 |
Neg logit contributions | |
kin | -0.031 |
ows | -0.030 |
as | -0.029 |
ily | -0.029 |
ute | -0.029 |
Token | ly |
Token ID | 306 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
OUT | +0.461 |
arrest | +0.456 |
Johnson | +0.452 |
istent | +0.426 |
joy | +0.425 |
Neg logit contributions | |
kin | -0.359 |
ows | -0.342 |
ily | -0.328 |
as | -0.326 |
holes | -0.325 |
Token | , |
Token ID | 11 |
Feature activation | -1.99e-01 |
Pos logit contributions | |
kin | +0.074 |
a | +0.070 |
ily | +0.070 |
holes | +0.066 |
ia | +0.062 |
Neg logit contributions | |
istent | -0.133 |
Johnson | -0.130 |
OUT | -0.128 |
arse | -0.128 |
joy | -0.124 |
Token | the |
Token ID | 262 |
Feature activation | -9.05e-02 |
Pos logit contributions | |
OUT | +0.022 |
istent | +0.022 |
arrest | +0.022 |
Johnson | +0.021 |
ited | +0.021 |
Neg logit contributions | |
kin | -0.012 |
holes | -0.011 |
Lia | -0.010 |
ily | -0.010 |
a | -0.009 |
Token | bird |
Token ID | 6512 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
arrest | +0.008 |
OUT | +0.008 |
Johnson | +0.008 |
Frankie | +0.007 |
Billy | +0.007 |
Neg logit contributions | |
kin | -0.008 |
ows | -0.008 |
as | -0.008 |
ily | -0.007 |
ute | -0.007 |
Token | flew |
Token ID | 13112 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
OUT | +0.030 |
Johnson | +0.029 |
arrest | +0.029 |
joy | +0.029 |
istent | +0.027 |
Neg logit contributions | |
kin | -0.020 |
ows | -0.019 |
as | -0.019 |
a | -0.018 |
ily | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
Johnson | +0.217 |
OUT | +0.217 |
arrest | +0.213 |
Billy | +0.209 |
joy | +0.204 |
Neg logit contributions | |
kin | -0.166 |
ows | -0.151 |
as | -0.148 |
ily | -0.140 |
holes | -0.139 |
Token | his |
Token ID | 465 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
kin | +0.109 |
holes | +0.093 |
ily | +0.093 |
ows | +0.090 |
as | +0.089 |
Neg logit contributions | |
OUT | -0.144 |
istent | -0.133 |
arrest | -0.132 |
Johnson | -0.130 |
joy | -0.125 |
Token | raft |
Token ID | 31812 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
kin | +0.015 |
ows | +0.014 |
as | +0.014 |
ily | +0.013 |
a | +0.013 |
Neg logit contributions | |
OUT | -0.017 |
arrest | -0.017 |
Johnson | -0.017 |
joy | -0.016 |
Billy | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
kin | +0.009 |
ows | +0.008 |
as | +0.008 |
ily | +0.007 |
a | +0.007 |
Neg logit contributions | |
OUT | -0.012 |
arrest | -0.012 |
Johnson | -0.012 |
Billy | -0.012 |
joy | -0.011 |
Token | Exc |
Token ID | 25268 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
kin | +0.051 |
ows | +0.046 |
as | +0.046 |
ily | +0.043 |
a | +0.043 |
Neg logit contributions | |
OUT | -0.051 |
arrest | -0.050 |
Johnson | -0.049 |
joy | -0.047 |
Billy | -0.047 |
Token | ited |
Token ID | 863 |
Feature activation | -4.75e+00 |
Pos logit contributions | |
OUT | +0.003 |
arrest | +0.003 |
Johnson | +0.002 |
joy | +0.002 |
Billy | +0.002 |
Neg logit contributions | |
kin | -0.030 |
ows | -0.029 |
as | -0.028 |
ily | -0.027 |
ute | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
OUT | +0.500 |
arrest | +0.490 |
Johnson | +0.485 |
istent | +0.460 |
joy | +0.459 |
Neg logit contributions | |
kin | -0.330 |
ows | -0.315 |
ily | -0.300 |
as | -0.296 |
holes | -0.293 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
OUT | +0.001 |
istent | +0.001 |
arrest | +0.001 |
Johnson | +0.001 |
joy | +0.001 |
Neg logit contributions | |
kin | -0.001 |
ily | -0.001 |
holes | -0.001 |
ows | -0.001 |
Lia | -0.001 |
Token | jumped |
Token ID | 11687 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
kin | +0.025 |
as | +0.023 |
ows | +0.023 |
a | +0.023 |
ily | +0.021 |
Neg logit contributions | |
OUT | -0.033 |
Johnson | -0.032 |
arrest | -0.031 |
joy | -0.031 |
Frankie | -0.030 |
Token | in |
Token ID | 287 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
kin | +0.128 |
as | +0.115 |
ows | +0.114 |
ily | +0.106 |
a | +0.105 |
Neg logit contributions | |
OUT | -0.145 |
arrest | -0.141 |
Johnson | -0.139 |
Billy | -0.136 |
joy | -0.136 |
Token | the |
Token ID | 262 |
Feature activation | -8.39e-02 |
Pos logit contributions | |
kin | +0.098 |
holes | +0.087 |
ows | +0.085 |
ily | +0.085 |
as | +0.082 |
Neg logit contributions | |
OUT | -0.147 |
arrest | -0.139 |
Johnson | -0.138 |
istent | -0.137 |
Frankie | -0.133 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
OUT | +0.120 |
arrest | +0.117 |
Johnson | +0.117 |
Billy | +0.113 |
Frankie | +0.110 |
Neg logit contributions | |
kin | -0.111 |
ows | -0.101 |
as | -0.100 |
ute | -0.096 |
ily | -0.095 |
Token | ." |
Token ID | 526 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
OUT | +0.128 |
arrest | +0.125 |
Johnson | +0.121 |
joy | +0.119 |
istent | +0.115 |
Neg logit contributions | |
kin | -0.088 |
ows | -0.079 |
as | -0.076 |
ily | -0.073 |
holes | -0.072 |
Token |
|
Token ID | 198 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
OUT | +0.071 |
arrest | +0.069 |
joy | +0.064 |
Johnson | +0.063 |
istent | +0.063 |
Neg logit contributions | |
kin | -0.049 |
ows | -0.046 |
as | -0.043 |
ily | -0.043 |
a | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
kin | +0.025 |
ows | +0.023 |
as | +0.023 |
ily | +0.022 |
a | +0.021 |
Neg logit contributions | |
OUT | -0.029 |
arrest | -0.029 |
Johnson | -0.028 |
joy | -0.027 |
Billy | -0.027 |
Token | Exc |
Token ID | 40127 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
kin | +0.018 |
ows | +0.016 |
as | +0.016 |
ily | +0.015 |
a | +0.015 |
Neg logit contributions | |
OUT | -0.026 |
arrest | -0.025 |
Johnson | -0.025 |
joy | -0.024 |
Billy | -0.024 |
Token | ited |
Token ID | 863 |
Feature activation | -4.75e+00 |
Pos logit contributions | |
kin | +0.075 |
a | +0.071 |
ows | +0.071 |
as | +0.070 |
ily | +0.070 |
Neg logit contributions | |
OUT | -0.021 |
arrest | -0.021 |
Billy | -0.018 |
Johnson | -0.018 |
joy | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
OUT | +0.471 |
arrest | +0.469 |
Johnson | +0.449 |
joy | +0.437 |
istent | +0.427 |
Neg logit contributions | |
kin | -0.358 |
ows | -0.343 |
ily | -0.328 |
as | -0.322 |
ia | -0.313 |
Token | Lucy |
Token ID | 22162 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
OUT | +0.013 |
arrest | +0.012 |
joy | +0.011 |
istent | +0.011 |
Johnson | +0.010 |
Neg logit contributions | |
kin | -0.016 |
ows | -0.015 |
a | -0.014 |
ily | -0.014 |
as | -0.014 |
Token | got |
Token ID | 1392 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
OUT | +0.118 |
arrest | +0.115 |
Johnson | +0.113 |
joy | +0.111 |
Billy | +0.107 |
Neg logit contributions | |
kin | -0.096 |
ows | -0.089 |
as | -0.087 |
a | -0.084 |
ily | -0.083 |
Token | some |
Token ID | 617 |
Feature activation | -2.24e-01 |
Pos logit contributions | |
kin | +0.057 |
ows | +0.053 |
as | +0.051 |
ily | +0.049 |
ute | +0.048 |
Neg logit contributions | |
OUT | -0.063 |
arrest | -0.062 |
Johnson | -0.061 |
joy | -0.058 |
Billy | -0.058 |
Token | blankets |
Token ID | 34794 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
OUT | +0.019 |
Johnson | +0.019 |
arrest | +0.019 |
Billy | +0.018 |
Frankie | +0.018 |
Neg logit contributions | |
kin | -0.020 |
as | -0.018 |
ows | -0.017 |
ute | -0.017 |
ily | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +6.28e-01 |
Pos logit contributions | |
OUT | +0.175 |
arrest | +0.172 |
Johnson | +0.169 |
Billy | +0.163 |
joy | +0.162 |
Neg logit contributions | |
kin | -0.139 |
ows | -0.125 |
as | -0.120 |
ute | -0.115 |
ily | -0.115 |
Token | |
Token ID | 220 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
kin | +0.057 |
ows | +0.055 |
holes | +0.053 |
ily | +0.053 |
ute | +0.052 |
Neg logit contributions | |
OUT | -0.051 |
arrest | -0.050 |
Johnson | -0.046 |
joy | -0.045 |
istent | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | +6.73e-01 |
Pos logit contributions | |
OUT | +0.167 |
arrest | +0.164 |
Johnson | +0.158 |
joy | +0.154 |
Billy | +0.150 |
Neg logit contributions | |
kin | -0.131 |
ows | -0.123 |
as | -0.119 |
ily | -0.114 |
a | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
kin | +0.051 |
ows | +0.047 |
as | +0.046 |
ily | +0.045 |
a | +0.044 |
Neg logit contributions | |
OUT | -0.068 |
arrest | -0.066 |
Johnson | -0.065 |
joy | -0.063 |
Billy | -0.062 |
Token | Exc |
Token ID | 40127 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
kin | +0.048 |
as | +0.043 |
ows | +0.042 |
a | +0.041 |
ily | +0.041 |
Neg logit contributions | |
OUT | -0.061 |
arrest | -0.059 |
Johnson | -0.058 |
joy | -0.057 |
Billy | -0.056 |
Token | ited |
Token ID | 863 |
Feature activation | -4.75e+00 |
Pos logit contributions | |
kin | +0.077 |
a | +0.074 |
as | +0.071 |
ily | +0.071 |
ila | +0.070 |
Neg logit contributions | |
OUT | -0.030 |
arrest | -0.028 |
Johnson | -0.027 |
Billy | -0.026 |
Frankie | -0.024 |
Token | ly |
Token ID | 306 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
OUT | +0.493 |
arrest | +0.484 |
Johnson | +0.475 |
joy | +0.453 |
istent | +0.452 |
Neg logit contributions | |
kin | -0.332 |
ows | -0.312 |
ily | -0.299 |
as | -0.299 |
holes | -0.293 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
kin | +0.097 |
a | +0.086 |
ows | +0.085 |
ily | +0.084 |
as | +0.081 |
Neg logit contributions | |
OUT | -0.125 |
Johnson | -0.123 |
arse | -0.121 |
joy | -0.117 |
arrest | -0.117 |
Token | Ark |
Token ID | 9128 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
OUT | +0.009 |
arrest | +0.008 |
istent | +0.008 |
Johnson | +0.008 |
joy | +0.008 |
Neg logit contributions | |
kin | -0.011 |
holes | -0.009 |
a | -0.009 |
ily | -0.009 |
ows | -0.009 |
Token | ie |
Token ID | 494 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
kin | +0.027 |
ows | +0.026 |
as | +0.025 |
ute | +0.024 |
holes | +0.024 |
Neg logit contributions | |
OUT | -0.014 |
arrest | -0.014 |
Johnson | -0.014 |
Billy | -0.013 |
Frankie | -0.013 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
kin | +0.030 |
ows | +0.028 |
as | +0.028 |
a | +0.027 |
ily | +0.026 |
Neg logit contributions | |
OUT | -0.032 |
arrest | -0.031 |
Johnson | -0.031 |
joy | -0.030 |
Frankie | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | +5.41e-01 |
Pos logit contributions | |
OUT | +0.016 |
arrest | +0.016 |
Johnson | +0.016 |
Billy | +0.015 |
joy | +0.015 |
Neg logit contributions | |
kin | -0.010 |
ows | -0.009 |
as | -0.009 |
ute | -0.008 |
ily | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
kin | +0.049 |
ows | +0.044 |
as | +0.043 |
ute | +0.043 |
utes | +0.041 |
Neg logit contributions | |
OUT | -0.047 |
arrest | -0.046 |
Johnson | -0.045 |
Billy | -0.044 |
Frankie | -0.043 |
Token | perfect |
Token ID | 2818 |
Feature activation | -4.00e-01 |
Pos logit contributions | |
OUT | +0.030 |
Johnson | +0.029 |
arrest | +0.029 |
Billy | +0.028 |
Frankie | +0.028 |
Neg logit contributions | |
kin | -0.023 |
ows | -0.021 |
as | -0.020 |
ute | -0.020 |
ily | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
OUT | +0.039 |
arrest | +0.038 |
Johnson | +0.038 |
joy | +0.037 |
Billy | +0.036 |
Neg logit contributions | |
kin | -0.032 |
ows | -0.029 |
as | -0.029 |
ily | -0.027 |
ute | -0.027 |
Token | Exc |
Token ID | 25268 |
Feature activation | -2.14e-01 |
Pos logit contributions | |
kin | +0.004 |
ows | +0.004 |
as | +0.003 |
ily | +0.003 |
holes | +0.003 |
Neg logit contributions | |
OUT | -0.003 |
arrest | -0.003 |
Johnson | -0.003 |
joy | -0.003 |
Billy | -0.003 |
Token | ited |
Token ID | 863 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
OUT | +0.005 |
arrest | +0.005 |
Johnson | +0.005 |
joy | +0.004 |
Billy | +0.004 |
Neg logit contributions | |
kin | -0.033 |
ows | -0.031 |
as | -0.031 |
ily | -0.030 |
a | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -9.63e-02 |
Pos logit contributions | |
OUT | +0.474 |
arrest | +0.469 |
Johnson | +0.459 |
istent | +0.435 |
joy | +0.432 |
Neg logit contributions | |
kin | -0.353 |
ows | -0.336 |
ily | -0.323 |
as | -0.321 |
holes | -0.316 |
Token | she |
Token ID | 673 |
Feature activation | -7.51e-01 |
Pos logit contributions | |
OUT | +0.009 |
arrest | +0.009 |
Johnson | +0.009 |
istent | +0.009 |
joy | +0.008 |
Neg logit contributions | |
kin | -0.008 |
ily | -0.007 |
holes | -0.007 |
ows | -0.007 |
a | -0.007 |
Token | ran |
Token ID | 4966 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
OUT | +0.062 |
arrest | +0.061 |
Johnson | +0.059 |
Billy | +0.057 |
joy | +0.057 |
Neg logit contributions | |
kin | -0.072 |
ows | -0.066 |
as | -0.064 |
ute | -0.062 |
ily | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | -5.32e-01 |
Pos logit contributions | |
kin | +0.065 |
ows | +0.060 |
as | +0.057 |
ily | +0.056 |
holes | +0.055 |
Neg logit contributions | |
OUT | -0.094 |
arrest | -0.091 |
Johnson | -0.090 |
istent | -0.087 |
joy | -0.086 |
Token | get |
Token ID | 651 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
OUT | +0.059 |
istent | +0.058 |
Johnson | +0.055 |
joy | +0.055 |
ited | +0.053 |
Neg logit contributions | |
ily | -0.032 |
kin | -0.029 |
holes | -0.029 |
shapes | -0.028 |
ows | -0.028 |
Token | in |
Token ID | 287 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
Johnson | +0.118 |
OUT | +0.115 |
arrest | +0.113 |
Billy | +0.112 |
Frankie | +0.109 |
Neg logit contributions | |
kin | -0.094 |
as | -0.086 |
ows | -0.085 |
ute | -0.083 |
ily | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | -2.17e-03 |
Pos logit contributions | |
kin | +0.134 |
ows | +0.120 |
as | +0.118 |
ily | +0.117 |
holes | +0.114 |
Neg logit contributions | |
OUT | -0.132 |
arrest | -0.128 |
Johnson | -0.123 |
joy | -0.121 |
istent | -0.118 |
Token | freezer |
Token ID | 30967 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
OUT | +0.000 |
Johnson | +0.000 |
arrest | +0.000 |
joy | +0.000 |
Billy | +0.000 |
Neg logit contributions | |
kin | -0.000 |
ows | -0.000 |
as | -0.000 |
ute | -0.000 |
a | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
OUT | +0.022 |
Johnson | +0.022 |
arrest | +0.022 |
Billy | +0.021 |
joy | +0.021 |
Neg logit contributions | |
kin | -0.018 |
ows | -0.016 |
as | -0.016 |
ily | -0.016 |
ute | -0.015 |
Token | Exc |
Token ID | 25268 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
kin | +0.005 |
ows | +0.005 |
ily | +0.005 |
as | +0.005 |
holes | +0.005 |
Neg logit contributions | |
OUT | -0.005 |
arrest | -0.005 |
Johnson | -0.005 |
joy | -0.005 |
Frankie | -0.005 |
Token | ited |
Token ID | 863 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
OUT | +0.004 |
arrest | +0.004 |
Johnson | +0.003 |
Frankie | +0.003 |
Billy | +0.003 |
Neg logit contributions | |
kin | -0.016 |
ows | -0.015 |
as | -0.015 |
ily | -0.014 |
a | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -8.93e-02 |
Pos logit contributions | |
OUT | +0.502 |
arrest | +0.496 |
Johnson | +0.485 |
istent | +0.464 |
joy | +0.464 |
Neg logit contributions | |
kin | -0.317 |
ows | -0.306 |
ily | -0.297 |
as | -0.289 |
holes | -0.283 |
Token | she |
Token ID | 673 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
OUT | +0.006 |
arrest | +0.006 |
istent | +0.005 |
joy | +0.005 |
Johnson | +0.005 |
Neg logit contributions | |
kin | -0.010 |
ily | -0.009 |
ows | -0.009 |
holes | -0.009 |
shapes | -0.009 |
Token | opened |
Token ID | 4721 |
Feature activation | +7.70e-01 |
Pos logit contributions | |
OUT | +0.054 |
arrest | +0.053 |
Johnson | +0.052 |
joy | +0.050 |
Billy | +0.049 |
Neg logit contributions | |
kin | -0.057 |
ows | -0.053 |
as | -0.051 |
ily | -0.050 |
a | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
kin | +0.049 |
holes | +0.047 |
ily | +0.045 |
ows | +0.045 |
a | +0.043 |
Neg logit contributions | |
OUT | -0.085 |
istent | -0.082 |
arrest | -0.080 |
Johnson | -0.077 |
joy | -0.075 |
Token | freezer |
Token ID | 30967 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
OUT | +0.001 |
arrest | +0.001 |
Johnson | +0.001 |
joy | +0.001 |
Billy | +0.001 |
Neg logit contributions | |
kin | -0.001 |
ows | -0.001 |
as | -0.001 |
ily | -0.001 |
ute | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
kin | +0.111 |
ows | +0.091 |
as | +0.088 |
ily | +0.087 |
ute | +0.086 |
Neg logit contributions | |
OUT | -0.164 |
arrest | -0.163 |
joy | -0.161 |
Johnson | -0.156 |
Billy | -0.154 |
Token | unknown |
Token ID | 6439 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
kin | +0.019 |
ows | +0.017 |
as | +0.016 |
ute | +0.016 |
ily | +0.015 |
Neg logit contributions | |
OUT | -0.027 |
arrest | -0.026 |
joy | -0.026 |
Johnson | -0.026 |
Billy | -0.026 |
Token | things |
Token ID | 1243 |
Feature activation | +7.73e-01 |
Pos logit contributions | |
kin | +0.056 |
ows | +0.045 |
ily | +0.044 |
as | +0.043 |
ute | +0.043 |
Neg logit contributions | |
arrest | -0.073 |
joy | -0.073 |
OUT | -0.072 |
Johnson | -0.070 |
Billy | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
kin | +0.064 |
ows | +0.055 |
as | +0.055 |
ute | +0.053 |
ily | +0.053 |
Neg logit contributions | |
OUT | -0.074 |
arrest | -0.073 |
Johnson | -0.072 |
Billy | -0.070 |
joy | -0.069 |
Token | Exc |
Token ID | 25268 |
Feature activation | -8.87e-02 |
Pos logit contributions | |
kin | +0.021 |
ows | +0.020 |
as | +0.019 |
ily | +0.019 |
a | +0.018 |
Neg logit contributions | |
OUT | -0.020 |
arrest | -0.019 |
Johnson | -0.019 |
joy | -0.018 |
Billy | -0.018 |
Token | ited |
Token ID | 863 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
OUT | +0.003 |
arrest | +0.002 |
Johnson | +0.002 |
joy | +0.002 |
Billy | +0.002 |
Neg logit contributions | |
kin | -0.013 |
ows | -0.013 |
as | -0.012 |
ily | -0.012 |
a | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
OUT | +0.491 |
arrest | +0.482 |
Johnson | +0.479 |
istent | +0.453 |
joy | +0.447 |
Neg logit contributions | |
kin | -0.327 |
ows | -0.315 |
as | -0.301 |
holes | -0.300 |
ily | -0.298 |
Token | he |
Token ID | 339 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
kin | +0.004 |
holes | +0.003 |
ily | +0.003 |
ows | +0.003 |
Lia | +0.003 |
Neg logit contributions | |
OUT | -0.004 |
istent | -0.004 |
Johnson | -0.004 |
arrest | -0.004 |
arse | -0.004 |
Token | kept |
Token ID | 4030 |
Feature activation | +1.79e-01 |
Pos logit contributions | |
OUT | +0.042 |
arrest | +0.041 |
Johnson | +0.039 |
Billy | +0.038 |
joy | +0.038 |
Neg logit contributions | |
kin | -0.050 |
ows | -0.046 |
as | -0.044 |
ily | -0.043 |
ute | -0.043 |
Token | exploring |
Token ID | 13504 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
kin | +0.019 |
a | +0.018 |
holes | +0.018 |
as | +0.018 |
ows | +0.018 |
Neg logit contributions | |
OUT | -0.011 |
Johnson | -0.011 |
istent | -0.010 |
arrest | -0.010 |
Smith | -0.009 |
Token | until |
Token ID | 1566 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
OUT | +0.174 |
arrest | +0.173 |
Johnson | +0.170 |
joy | +0.169 |
Billy | +0.165 |
Neg logit contributions | |
kin | -0.163 |
ows | -0.145 |
as | -0.144 |
ily | -0.138 |
ute | -0.137 |
Token | of |
Token ID | 286 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
OUT | +0.120 |
arrest | +0.117 |
Johnson | +0.115 |
joy | +0.112 |
Billy | +0.111 |
Neg logit contributions | |
kin | -0.084 |
ows | -0.076 |
as | -0.074 |
ily | -0.070 |
a | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
kin | +0.053 |
ily | +0.046 |
holes | +0.046 |
ows | +0.045 |
as | +0.044 |
Neg logit contributions | |
OUT | -0.069 |
istent | -0.065 |
arrest | -0.065 |
Johnson | -0.063 |
joy | -0.062 |
Token | mountain |
Token ID | 8598 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
OUT | +0.003 |
arrest | +0.003 |
Johnson | +0.003 |
joy | +0.003 |
Billy | +0.003 |
Neg logit contributions | |
kin | -0.003 |
ows | -0.002 |
holes | -0.002 |
as | -0.002 |
ily | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
OUT | +0.082 |
arrest | +0.080 |
Johnson | +0.078 |
joy | +0.076 |
Billy | +0.075 |
Neg logit contributions | |
kin | -0.055 |
ows | -0.050 |
as | -0.047 |
ily | -0.045 |
a | -0.044 |
Token | Exc |
Token ID | 25268 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
kin | +0.025 |
ows | +0.024 |
ily | +0.023 |
as | +0.022 |
holes | +0.022 |
Neg logit contributions | |
OUT | -0.028 |
arrest | -0.028 |
Johnson | -0.027 |
joy | -0.026 |
istent | -0.026 |
Token | ited |
Token ID | 863 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
OUT | +0.002 |
arrest | +0.002 |
Johnson | +0.002 |
joy | +0.002 |
Billy | +0.002 |
Neg logit contributions | |
kin | -0.015 |
ows | -0.014 |
as | -0.014 |
ily | -0.013 |
a | -0.013 |
Token | ly |
Token ID | 306 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
OUT | +0.488 |
arrest | +0.479 |
Johnson | +0.470 |
istent | +0.457 |
joy | +0.440 |
Neg logit contributions | |
kin | -0.324 |
ows | -0.315 |
ily | -0.309 |
holes | -0.298 |
as | -0.294 |
Token | , |
Token ID | 11 |
Feature activation | +5.83e-02 |
Pos logit contributions | |
kin | +0.085 |
a | +0.078 |
ily | +0.078 |
ows | +0.076 |
ia | +0.070 |
Neg logit contributions | |
istent | -0.133 |
OUT | -0.133 |
Johnson | -0.131 |
arse | -0.131 |
arrest | -0.124 |
Token | she |
Token ID | 673 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
kin | +0.005 |
ily | +0.004 |
ows | +0.004 |
holes | +0.004 |
a | +0.004 |
Neg logit contributions | |
OUT | -0.005 |
arrest | -0.005 |
istent | -0.005 |
Johnson | -0.005 |
arse | -0.005 |
Token | rushed |
Token ID | 13999 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
OUT | +0.052 |
arrest | +0.051 |
Johnson | +0.049 |
Billy | +0.047 |
joy | +0.047 |
Neg logit contributions | |
kin | -0.062 |
ows | -0.057 |
as | -0.055 |
ute | -0.054 |
ily | -0.054 |
Token | to |
Token ID | 284 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
ily | +0.004 |
kin | +0.004 |
ows | +0.004 |
holes | +0.004 |
as | +0.004 |
Neg logit contributions | |
OUT | -0.009 |
arrest | -0.009 |
istent | -0.009 |
Johnson | -0.009 |
Frankie | -0.008 |
Token | saw |
Token ID | 2497 |
Feature activation | +6.46e-01 |
Pos logit contributions | |
OUT | +0.054 |
arrest | +0.052 |
Johnson | +0.051 |
joy | +0.049 |
Billy | +0.048 |
Neg logit contributions | |
kin | -0.065 |
ows | -0.060 |
as | -0.059 |
ily | -0.057 |
a | -0.056 |
Token | a |
Token ID | 257 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
kin | +0.053 |
ows | +0.048 |
as | +0.047 |
ute | +0.045 |
utes | +0.042 |
Neg logit contributions | |
OUT | -0.063 |
arrest | -0.062 |
Johnson | -0.062 |
Billy | -0.061 |
joy | -0.060 |
Token | chest |
Token ID | 7721 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
OUT | +0.034 |
arrest | +0.034 |
Johnson | +0.033 |
joy | +0.032 |
Billy | +0.032 |
Neg logit contributions | |
kin | -0.022 |
ows | -0.019 |
as | -0.019 |
ily | -0.018 |
ute | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -1.56e-02 |
Pos logit contributions | |
kin | +0.010 |
ows | +0.009 |
as | +0.008 |
a | +0.008 |
ily | +0.008 |
Neg logit contributions | |
OUT | -0.013 |
arrest | -0.013 |
Johnson | -0.013 |
joy | -0.012 |
Billy | -0.012 |
Token | Exc |
Token ID | 25268 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
OUT | +0.001 |
arrest | +0.001 |
Johnson | +0.001 |
joy | +0.001 |
istent | +0.001 |
Neg logit contributions | |
kin | -0.001 |
ows | -0.001 |
ily | -0.001 |
as | -0.001 |
holes | -0.001 |
Token | ited |
Token ID | 863 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
OUT | +0.010 |
Johnson | +0.009 |
arrest | +0.009 |
Billy | +0.008 |
Frankie | +0.008 |
Neg logit contributions | |
kin | -0.053 |
as | -0.051 |
ows | -0.050 |
a | -0.049 |
ily | -0.049 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
OUT | +0.488 |
arrest | +0.480 |
Johnson | +0.475 |
istent | +0.450 |
joy | +0.441 |
Neg logit contributions | |
kin | -0.332 |
ows | -0.315 |
ily | -0.304 |
holes | -0.300 |
as | -0.300 |
Token | he |
Token ID | 339 |
Feature activation | -6.27e-01 |
Pos logit contributions | |
OUT | +0.001 |
arrest | +0.001 |
istent | +0.001 |
Johnson | +0.001 |
arse | +0.001 |
Neg logit contributions | |
kin | -0.001 |
holes | -0.001 |
ily | -0.001 |
ows | -0.001 |
a | -0.001 |
Token | pr |
Token ID | 778 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
OUT | +0.052 |
arrest | +0.050 |
Johnson | +0.050 |
joy | +0.047 |
Billy | +0.047 |
Neg logit contributions | |
kin | -0.061 |
ows | -0.056 |
as | -0.054 |
ily | -0.052 |
ute | -0.052 |
Token | icked |
Token ID | 9484 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
kin | +0.008 |
ows | +0.008 |
as | +0.008 |
ily | +0.007 |
a | +0.007 |
Neg logit contributions | |
OUT | -0.013 |
arrest | -0.013 |
Johnson | -0.013 |
joy | -0.013 |
Billy | -0.012 |
Token | up |
Token ID | 510 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
kin | +0.146 |
ows | +0.132 |
as | +0.129 |
ily | +0.127 |
a | +0.123 |
Neg logit contributions | |
OUT | -0.145 |
arrest | -0.141 |
Johnson | -0.137 |
joy | -0.132 |
istent | -0.128 |
Token | maze |
Token ID | 31237 |
Feature activation | +9.33e-01 |
Pos logit contributions | |
OUT | +0.023 |
arrest | +0.023 |
Johnson | +0.022 |
istent | +0.022 |
joy | +0.022 |
Neg logit contributions | |
kin | -0.017 |
ows | -0.015 |
holes | -0.015 |
as | -0.014 |
ily | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
kin | +0.075 |
ows | +0.068 |
as | +0.067 |
ute | +0.065 |
utes | +0.063 |
Neg logit contributions | |
arrest | -0.089 |
OUT | -0.089 |
Johnson | -0.086 |
Billy | -0.082 |
joy | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
kin | +0.015 |
ows | +0.014 |
as | +0.014 |
ute | +0.013 |
a | +0.013 |
Neg logit contributions | |
OUT | -0.013 |
arrest | -0.012 |
Johnson | -0.012 |
joy | -0.012 |
Billy | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
kin | +0.022 |
ows | +0.021 |
as | +0.020 |
holes | +0.019 |
ute | +0.019 |
Neg logit contributions | |
Johnson | -0.017 |
OUT | -0.017 |
arrest | -0.017 |
Billy | -0.017 |
Frankie | -0.017 |
Token | Exc |
Token ID | 40127 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
ows | +0.014 |
kin | +0.014 |
as | +0.012 |
holes | +0.012 |
ute | +0.012 |
Neg logit contributions | |
OUT | -0.014 |
arrest | -0.014 |
Johnson | -0.014 |
joy | -0.014 |
Frankie | -0.013 |
Token | ited |
Token ID | 863 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
kin | +0.069 |
a | +0.068 |
as | +0.065 |
ila | +0.063 |
ily | +0.063 |
Neg logit contributions | |
OUT | -0.024 |
arrest | -0.024 |
Billy | -0.022 |
Johnson | -0.022 |
Frankie | -0.022 |
Token | ly |
Token ID | 306 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
OUT | +0.457 |
arrest | +0.450 |
Johnson | +0.440 |
joy | +0.419 |
istent | +0.413 |
Neg logit contributions | |
kin | -0.380 |
ows | -0.353 |
as | -0.341 |
ily | -0.338 |
a | -0.327 |
Token | , |
Token ID | 11 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
kin | +0.096 |
ows | +0.087 |
a | +0.086 |
ily | +0.084 |
as | +0.082 |
Neg logit contributions | |
OUT | -0.125 |
Johnson | -0.124 |
joy | -0.120 |
arse | -0.119 |
arrest | -0.119 |
Token | she |
Token ID | 673 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
OUT | +0.006 |
arrest | +0.006 |
Johnson | +0.006 |
joy | +0.006 |
Billy | +0.006 |
Neg logit contributions | |
kin | -0.008 |
ows | -0.008 |
as | -0.007 |
ily | -0.007 |
ute | -0.007 |
Token | asked |
Token ID | 1965 |
Feature activation | +9.94e-03 |
Pos logit contributions | |
OUT | +0.061 |
arrest | +0.060 |
Johnson | +0.057 |
Billy | +0.055 |
joy | +0.055 |
Neg logit contributions | |
kin | -0.074 |
ows | -0.068 |
as | -0.066 |
ute | -0.064 |
ily | -0.064 |
Token | her |
Token ID | 607 |
Feature activation | +3.75e-01 |
Pos logit contributions | |
kin | +0.001 |
ows | +0.001 |
as | +0.001 |
ute | +0.001 |
holes | +0.001 |
Neg logit contributions | |
OUT | -0.001 |
Johnson | -0.001 |
arrest | -0.001 |
Billy | -0.001 |
joy | -0.001 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | +9.07e-02 |
Pos logit contributions | |
kin | +0.048 |
ows | +0.044 |
as | +0.043 |
ily | +0.041 |
ute | +0.040 |
Neg logit contributions | |
OUT | -0.055 |
arrest | -0.055 |
Johnson | -0.054 |
joy | -0.052 |
Billy | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
kin | +0.008 |
as | +0.007 |
ows | +0.007 |
ily | +0.007 |
ute | +0.007 |
Neg logit contributions | |
OUT | -0.008 |
arrest | -0.008 |
Johnson | -0.008 |
Billy | -0.007 |
Frankie | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
kin | +0.013 |
ows | +0.012 |
as | +0.012 |
ute | +0.011 |
a | +0.011 |
Neg logit contributions | |
OUT | -0.011 |
arrest | -0.010 |
Johnson | -0.010 |
Billy | -0.010 |
joy | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +4.56e-01 |
Pos logit contributions | |
kin | +0.055 |
ows | +0.049 |
as | +0.047 |
holes | +0.046 |
ute | +0.045 |
Neg logit contributions | |
Johnson | -0.042 |
OUT | -0.041 |
arrest | -0.041 |
Billy | -0.040 |
joy | -0.039 |
Token | Exc |
Token ID | 40127 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
kin | +0.033 |
ows | +0.031 |
as | +0.027 |
holes | +0.026 |
ute | +0.026 |
Neg logit contributions | |
OUT | -0.048 |
arrest | -0.047 |
Johnson | -0.046 |
Billy | -0.044 |
joy | -0.044 |
Token | ited |
Token ID | 863 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
kin | +0.079 |
a | +0.074 |
as | +0.074 |
ily | +0.074 |
ows | +0.073 |
Neg logit contributions | |
OUT | -0.020 |
arrest | -0.019 |
Johnson | -0.019 |
Billy | -0.017 |
Frankie | -0.017 |
Token | ly |
Token ID | 306 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
OUT | +0.502 |
arrest | +0.493 |
Johnson | +0.488 |
joy | +0.465 |
istent | +0.462 |
Neg logit contributions | |
kin | -0.320 |
ows | -0.302 |
ily | -0.292 |
as | -0.286 |
holes | -0.281 |
Token | , |
Token ID | 11 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
kin | +0.084 |
ily | +0.074 |
ows | +0.072 |
a | +0.072 |
holes | +0.067 |
Neg logit contributions | |
arse | -0.135 |
Johnson | -0.134 |
OUT | -0.133 |
joy | -0.130 |
istent | -0.130 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
kin | +0.003 |
ily | +0.003 |
ows | +0.003 |
holes | +0.002 |
as | +0.002 |
Neg logit contributions | |
OUT | -0.004 |
arrest | -0.004 |
istent | -0.004 |
Johnson | -0.004 |
joy | -0.004 |
Token | brought |
Token ID | 3181 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
kin | +0.012 |
ows | +0.012 |
as | +0.012 |
a | +0.012 |
ily | +0.011 |
Neg logit contributions | |
OUT | -0.015 |
Johnson | -0.014 |
arrest | -0.014 |
joy | -0.014 |
Frankie | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
OUT | +0.054 |
arrest | +0.052 |
Johnson | +0.051 |
joy | +0.049 |
istent | +0.048 |
Neg logit contributions | |
kin | -0.048 |
ows | -0.043 |
as | -0.042 |
ily | -0.042 |
holes | -0.041 |
Token | ." |
Token ID | 526 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
OUT | +0.186 |
arrest | +0.183 |
Johnson | +0.180 |
Billy | +0.173 |
Frankie | +0.169 |
Neg logit contributions | |
kin | -0.169 |
ows | -0.149 |
as | -0.145 |
ily | -0.141 |
ute | -0.140 |
Token | |
Token ID | 220 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
OUT | +0.075 |
arrest | +0.072 |
joy | +0.065 |
burn | +0.064 |
returning | +0.063 |
Neg logit contributions | |
ows | -0.054 |
kin | -0.051 |
ily | -0.050 |
holes | -0.048 |
as | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
OUT | +0.196 |
arrest | +0.192 |
Johnson | +0.181 |
joy | +0.181 |
istent | +0.172 |
Neg logit contributions | |
kin | -0.130 |
ows | -0.127 |
as | -0.119 |
ily | -0.114 |
a | -0.111 |
Token |
|
Token ID | 198 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
kin | +0.022 |
ows | +0.021 |
as | +0.021 |
ily | +0.020 |
a | +0.020 |
Neg logit contributions | |
OUT | -0.044 |
arrest | -0.042 |
joy | -0.040 |
Johnson | -0.039 |
istent | -0.039 |
Token | Exc |
Token ID | 40127 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
kin | +0.036 |
as | +0.036 |
a | +0.034 |
ows | +0.034 |
ily | +0.034 |
Neg logit contributions | |
OUT | -0.022 |
arrest | -0.020 |
joy | -0.020 |
Johnson | -0.019 |
Billy | -0.018 |
Token | ited |
Token ID | 863 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
kin | +0.081 |
a | +0.078 |
ily | +0.077 |
as | +0.076 |
ows | +0.076 |
Neg logit contributions | |
OUT | -0.026 |
arrest | -0.025 |
Johnson | -0.022 |
Billy | -0.022 |
Frankie | -0.020 |
Token | for |
Token ID | 329 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
OUT | +0.479 |
arrest | +0.476 |
Johnson | +0.459 |
joy | +0.444 |
istent | +0.437 |
Neg logit contributions | |
kin | -0.326 |
ows | -0.318 |
ily | -0.308 |
ia | -0.296 |
as | -0.295 |
Token | the |
Token ID | 262 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
OUT | +0.014 |
arrest | +0.013 |
istent | +0.013 |
Johnson | +0.013 |
joy | +0.012 |
Neg logit contributions | |
kin | -0.009 |
ily | -0.008 |
holes | -0.008 |
ows | -0.007 |
ia | -0.007 |
Token | surprise |
Token ID | 5975 |
Feature activation | +3.77e-01 |
Pos logit contributions | |
OUT | +0.002 |
arrest | +0.002 |
Johnson | +0.002 |
joy | +0.002 |
Billy | +0.002 |
Neg logit contributions | |
kin | -0.001 |
ows | -0.001 |
as | -0.001 |
ily | -0.001 |
a | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
kin | +0.034 |
ows | +0.031 |
as | +0.031 |
ily | +0.030 |
a | +0.029 |
Neg logit contributions | |
OUT | -0.033 |
arrest | -0.032 |
Johnson | -0.032 |
joy | -0.030 |
Billy | -0.030 |
Token | John |
Token ID | 1757 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
OUT | +0.011 |
arrest | +0.010 |
istent | +0.009 |
joy | +0.008 |
arse | +0.008 |
Neg logit contributions | |
kin | -0.021 |
holes | -0.019 |
ows | -0.019 |
shapes | -0.019 |
ily | -0.019 |
Token | ! |
Token ID | 0 |
Feature activation | -4.52e-01 |
Pos logit contributions | |
OUT | +0.137 |
arrest | +0.135 |
Johnson | +0.133 |
joy | +0.129 |
Billy | +0.128 |
Neg logit contributions | |
kin | -0.128 |
ows | -0.118 |
as | -0.115 |
ily | -0.112 |
ute | -0.109 |
Token | |
Token ID | 220 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
OUT | +0.036 |
Johnson | +0.036 |
arrest | +0.036 |
joy | +0.035 |
Billy | +0.035 |
Neg logit contributions | |
kin | -0.044 |
ows | -0.040 |
as | -0.039 |
ute | -0.038 |
a | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
OUT | +0.170 |
arrest | +0.167 |
Johnson | +0.166 |
Billy | +0.159 |
joy | +0.159 |
Neg logit contributions | |
kin | -0.164 |
ows | -0.146 |
as | -0.143 |
ily | -0.137 |
ute | -0.137 |
Token |
|
Token ID | 198 |
Feature activation | +3.72e-01 |
Pos logit contributions | |
kin | +0.044 |
ows | +0.040 |
ute | +0.039 |
holes | +0.038 |
as | +0.038 |
Neg logit contributions | |
Johnson | -0.030 |
OUT | -0.030 |
arrest | -0.030 |
joy | -0.029 |
Billy | -0.028 |
Token | M |
Token ID | 44 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
kin | +0.033 |
ows | +0.032 |
ute | +0.030 |
holes | +0.028 |
as | +0.028 |
Neg logit contributions | |
OUT | -0.031 |
arrest | -0.031 |
Johnson | -0.030 |
joy | -0.029 |
returning | -0.028 |
Token | ia |
Token ID | 544 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
kin | +0.036 |
ows | +0.034 |
holes | +0.028 |
as | +0.026 |
ute | +0.025 |
Neg logit contributions | |
OUT | -0.139 |
arrest | -0.137 |
Johnson | -0.134 |
joy | -0.131 |
istent | -0.131 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
kin | +0.187 |
ows | +0.172 |
as | +0.166 |
ily | +0.160 |
ute | +0.159 |
Neg logit contributions | |
OUT | -0.189 |
arrest | -0.186 |
Johnson | -0.181 |
joy | -0.175 |
Billy | -0.173 |
Token | to |
Token ID | 284 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
kin | +0.005 |
as | +0.005 |
ows | +0.005 |
ily | +0.005 |
ila | +0.005 |
Neg logit contributions | |
joy | -0.005 |
OUT | -0.005 |
Johnson | -0.005 |
Billy | -0.005 |
arrest | -0.005 |
Token | herself |
Token ID | 5223 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
OUT | +0.049 |
arrest | +0.047 |
Johnson | +0.047 |
joy | +0.045 |
istent | +0.044 |
Neg logit contributions | |
kin | -0.035 |
ily | -0.032 |
ows | -0.031 |
holes | -0.029 |
ia | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
kin | +0.018 |
ows | +0.017 |
as | +0.016 |
a | +0.016 |
ute | +0.015 |
Neg logit contributions | |
OUT | -0.030 |
arrest | -0.029 |
Johnson | -0.028 |
joy | -0.027 |
Billy | -0.026 |
Token | said |
Token ID | 531 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
kin | +0.032 |
ows | +0.030 |
as | +0.029 |
ily | +0.028 |
ute | +0.028 |
Neg logit contributions | |
OUT | -0.031 |
arrest | -0.031 |
Johnson | -0.030 |
Billy | -0.029 |
joy | -0.029 |
Token | scary |
Token ID | 14343 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
kin | +0.014 |
ows | +0.013 |
as | +0.013 |
ute | +0.012 |
ily | +0.012 |
Neg logit contributions | |
OUT | -0.015 |
arrest | -0.014 |
Johnson | -0.014 |
Billy | -0.014 |
Frankie | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
kin | +0.039 |
ows | +0.034 |
as | +0.033 |
ute | +0.032 |
ily | +0.031 |
Neg logit contributions | |
OUT | -0.052 |
arrest | -0.051 |
Johnson | -0.050 |
joy | -0.048 |
Billy | -0.048 |
Token |
|
Token ID | 198 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
kin | +0.013 |
ows | +0.012 |
as | +0.012 |
ily | +0.011 |
a | +0.011 |
Neg logit contributions | |
OUT | -0.013 |
arrest | -0.013 |
Johnson | -0.012 |
joy | -0.012 |
Billy | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
kin | +0.010 |
ows | +0.009 |
as | +0.008 |
holes | +0.008 |
ute | +0.008 |
Neg logit contributions | |
OUT | -0.009 |
Johnson | -0.009 |
arrest | -0.009 |
Billy | -0.008 |
joy | -0.008 |
Token | L |
Token ID | 43 |
Feature activation | +6.15e-01 |
Pos logit contributions | |
kin | +0.008 |
ows | +0.007 |
as | +0.006 |
ute | +0.006 |
ily | +0.006 |
Neg logit contributions | |
OUT | -0.013 |
arrest | -0.013 |
Johnson | -0.013 |
Billy | -0.012 |
joy | -0.012 |
Token | ily |
Token ID | 813 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
kin | +0.019 |
ute | +0.015 |
ows | +0.015 |
utes | +0.014 |
holes | +0.013 |
Neg logit contributions | |
arrest | -0.083 |
OUT | -0.081 |
Johnson | -0.080 |
joy | -0.080 |
Billy | -0.078 |
Token | 's |
Token ID | 338 |
Feature activation | +6.15e-02 |
Pos logit contributions | |
kin | +0.290 |
ows | +0.271 |
as | +0.263 |
ily | +0.257 |
a | +0.255 |
Neg logit contributions | |
OUT | -0.325 |
arrest | -0.316 |
Johnson | -0.311 |
joy | -0.305 |
Frankie | -0.294 |
Token | mom |
Token ID | 1995 |
Feature activation | +8.20e-01 |
Pos logit contributions | |
kin | +0.005 |
ows | +0.005 |
as | +0.004 |
ily | +0.004 |
ute | +0.004 |
Neg logit contributions | |
arrest | -0.006 |
Johnson | -0.006 |
OUT | -0.006 |
joy | -0.006 |
Billy | -0.006 |
Token | saw |
Token ID | 2497 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
as | +0.062 |
a | +0.062 |
ily | +0.061 |
ows | +0.059 |
kin | +0.058 |
Neg logit contributions | |
OUT | -0.081 |
arrest | -0.077 |
joy | -0.076 |
Johnson | -0.073 |
Frankie | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
kin | +0.040 |
ily | +0.038 |
ows | +0.037 |
holes | +0.036 |
as | +0.036 |
Neg logit contributions | |
OUT | -0.054 |
arrest | -0.051 |
Johnson | -0.051 |
istent | -0.050 |
joy | -0.049 |
Token | picture |
Token ID | 4286 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
OUT | +0.009 |
arrest | +0.009 |
Johnson | +0.009 |
joy | +0.008 |
Billy | +0.008 |
Neg logit contributions | |
kin | -0.008 |
ows | -0.007 |
as | -0.007 |
ily | -0.006 |
a | -0.006 |
Token | 're |
Token ID | 821 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
kin | +0.079 |
ows | +0.074 |
as | +0.072 |
ute | +0.069 |
ily | +0.068 |
Neg logit contributions | |
OUT | -0.090 |
arrest | -0.089 |
Johnson | -0.084 |
Billy | -0.084 |
joy | -0.082 |
Token | so |
Token ID | 523 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
OUT | +0.287 |
arrest | +0.280 |
Johnson | +0.279 |
Billy | +0.272 |
Frankie | +0.269 |
Neg logit contributions | |
kin | -0.202 |
ows | -0.185 |
as | -0.183 |
ute | -0.171 |
ily | -0.167 |
Token | graceful |
Token ID | 44363 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
kin | +0.079 |
ows | +0.071 |
as | +0.068 |
ily | +0.065 |
ute | +0.063 |
Neg logit contributions | |
OUT | -0.141 |
arrest | -0.139 |
Johnson | -0.135 |
Billy | -0.133 |
joy | -0.132 |
Token | with |
Token ID | 351 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
kin | +0.011 |
ows | +0.010 |
as | +0.010 |
ute | +0.009 |
ily | +0.009 |
Neg logit contributions | |
OUT | -0.008 |
arrest | -0.008 |
Johnson | -0.008 |
Billy | -0.008 |
joy | -0.008 |
Token | your |
Token ID | 534 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
arrest | +0.042 |
OUT | +0.042 |
Billy | +0.039 |
Johnson | +0.039 |
Frankie | +0.038 |
Neg logit contributions | |
kin | -0.047 |
ows | -0.045 |
as | -0.042 |
ute | -0.041 |
ily | -0.041 |
Token | wand |
Token ID | 11569 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
kin | +0.038 |
ows | +0.036 |
as | +0.035 |
ily | +0.034 |
a | +0.032 |
Neg logit contributions | |
arrest | -0.057 |
OUT | -0.056 |
joy | -0.054 |
Johnson | -0.053 |
Billy | -0.053 |
Token | ," |
Token ID | 553 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
kin | +0.267 |
ows | +0.241 |
as | +0.239 |
ily | +0.234 |
ute | +0.228 |
Neg logit contributions | |
OUT | -0.240 |
arrest | -0.237 |
Johnson | -0.231 |
Billy | -0.222 |
Frankie | -0.219 |
Token | Max |
Token ID | 5436 |
Feature activation | -6.47e-01 |
Pos logit contributions | |
OUT | +0.013 |
Johnson | +0.013 |
arrest | +0.013 |
Billy | +0.013 |
Frankie | +0.012 |
Neg logit contributions | |
kin | -0.020 |
ows | -0.019 |
as | -0.019 |
ute | -0.018 |
a | -0.017 |
Token | said |
Token ID | 531 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
OUT | +0.065 |
arrest | +0.064 |
Johnson | +0.063 |
Billy | +0.061 |
joy | +0.060 |
Neg logit contributions | |
kin | -0.050 |
ows | -0.046 |
as | -0.044 |
ily | -0.042 |
ute | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
OUT | +0.028 |
arrest | +0.028 |
Johnson | +0.027 |
joy | +0.026 |
Billy | +0.026 |
Neg logit contributions | |
kin | -0.018 |
ows | -0.016 |
as | -0.015 |
ily | -0.015 |
a | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
kin | +0.027 |
ows | +0.025 |
as | +0.024 |
ute | +0.023 |
a | +0.023 |
Neg logit contributions | |
OUT | -0.024 |
arrest | -0.023 |
Johnson | -0.023 |
Billy | -0.022 |
joy | -0.022 |
Token | good |
Token ID | 922 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
arrest | +0.049 |
Johnson | +0.048 |
OUT | +0.048 |
Billy | +0.048 |
joy | +0.047 |
Neg logit contributions | |
kin | -0.040 |
ows | -0.038 |
ute | -0.036 |
as | -0.036 |
auc | -0.033 |
Token | twins |
Token ID | 20345 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
OUT | +0.022 |
arrest | +0.022 |
Johnson | +0.022 |
joy | +0.021 |
Billy | +0.021 |
Neg logit contributions | |
kin | -0.016 |
ows | -0.014 |
as | -0.013 |
ily | -0.013 |
ute | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
OUT | +0.029 |
arrest | +0.028 |
Johnson | +0.027 |
joy | +0.026 |
Frankie | +0.026 |
Neg logit contributions | |
kin | -0.018 |
ows | -0.016 |
as | -0.015 |
ily | -0.015 |
a | -0.014 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
Johnson | +0.046 |
OUT | +0.044 |
Billy | +0.044 |
arrest | +0.044 |
joy | +0.042 |
Neg logit contributions | |
kin | -0.049 |
ows | -0.045 |
as | -0.043 |
a | -0.041 |
ute | -0.041 |
Token | L |
Token ID | 43 |
Feature activation | +4.86e-01 |
Pos logit contributions | |
kin | +0.115 |
ows | +0.108 |
as | +0.104 |
ute | +0.099 |
a | +0.098 |
Neg logit contributions | |
OUT | -0.087 |
Johnson | -0.086 |
arrest | -0.086 |
joy | -0.084 |
Billy | -0.081 |
Token | ily |
Token ID | 813 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.022 |
as | +0.021 |
ute | +0.019 |
holes | +0.019 |
Neg logit contributions | |
OUT | -0.061 |
arrest | -0.061 |
Johnson | -0.060 |
Billy | -0.058 |
joy | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -3.09e-01 |
Pos logit contributions | |
kin | +0.223 |
ows | +0.199 |
as | +0.192 |
ily | +0.178 |
ute | +0.178 |
Neg logit contributions | |
OUT | -0.362 |
arrest | -0.361 |
Johnson | -0.352 |
joy | -0.342 |
Billy | -0.342 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
OUT | +0.026 |
arrest | +0.026 |
Johnson | +0.025 |
Billy | +0.024 |
joy | +0.024 |
Neg logit contributions | |
kin | -0.029 |
ows | -0.027 |
as | -0.027 |
ily | -0.025 |
ute | -0.025 |
Token | are |
Token ID | 389 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
kin | +0.096 |
ows | +0.086 |
as | +0.083 |
ily | +0.080 |
a | +0.078 |
Neg logit contributions | |
OUT | -0.142 |
arrest | -0.139 |
Johnson | -0.137 |
joy | -0.133 |
Billy | -0.131 |
Token | friends |
Token ID | 2460 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
OUT | +0.070 |
arrest | +0.069 |
Johnson | +0.067 |
joy | +0.064 |
Billy | +0.063 |
Neg logit contributions | |
kin | -0.074 |
ows | -0.068 |
as | -0.067 |
ily | -0.065 |
a | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -5.20e-01 |
Pos logit contributions | |
kin | +0.028 |
ows | +0.025 |
as | +0.024 |
ily | +0.023 |
a | +0.023 |
Neg logit contributions | |
OUT | -0.048 |
arrest | -0.047 |
Johnson | -0.046 |
joy | -0.044 |
Billy | -0.044 |
Token | outside |
Token ID | 2354 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
arrest | +0.062 |
OUT | +0.062 |
Billy | +0.060 |
Johnson | +0.059 |
joy | +0.059 |
Neg logit contributions | |
kin | -0.049 |
as | -0.046 |
ows | -0.046 |
a | -0.042 |
ia | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
OUT | +0.140 |
arrest | +0.138 |
Johnson | +0.136 |
joy | +0.132 |
Billy | +0.131 |
Neg logit contributions | |
kin | -0.073 |
ows | -0.065 |
as | -0.062 |
ily | -0.059 |
a | -0.057 |
Token | One |
Token ID | 1881 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
OUT | +0.008 |
arrest | +0.008 |
Johnson | +0.008 |
joy | +0.007 |
Billy | +0.007 |
Neg logit contributions | |
kin | -0.014 |
ows | -0.013 |
as | -0.012 |
a | -0.012 |
ily | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | -8.89e-01 |
Pos logit contributions | |
kin | +0.018 |
ows | +0.016 |
as | +0.016 |
ily | +0.015 |
holes | +0.015 |
Neg logit contributions | |
OUT | -0.021 |
arrest | -0.021 |
Johnson | -0.021 |
Billy | -0.020 |
joy | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
OUT | +0.063 |
arrest | +0.062 |
Johnson | +0.060 |
joy | +0.059 |
Billy | +0.057 |
Neg logit contributions | |
kin | -0.093 |
ows | -0.088 |
as | -0.086 |
ily | -0.082 |
ute | -0.082 |
Token | Julia |
Token ID | 22300 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
joy | +0.015 |
arrest | +0.015 |
OUT | +0.015 |
John | +0.014 |
Billy | +0.014 |
Neg logit contributions | |
kin | -0.027 |
ows | -0.026 |
as | -0.025 |
auc | -0.024 |
an | -0.024 |
Token | found |
Token ID | 1043 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
kin | +0.168 |
ows | +0.155 |
as | +0.151 |
ily | +0.147 |
a | +0.144 |
Neg logit contributions | |
OUT | -0.152 |
arrest | -0.148 |
Johnson | -0.146 |
joy | -0.140 |
Billy | -0.138 |
Token | an |
Token ID | 281 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
kin | +0.077 |
ows | +0.073 |
as | +0.070 |
ute | +0.066 |
utes | +0.063 |
Neg logit contributions | |
Billy | -0.073 |
Johnson | -0.073 |
arrest | -0.072 |
OUT | -0.070 |
Frankie | -0.069 |
Token | orange |
Token ID | 10912 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
OUT | +0.117 |
Johnson | +0.116 |
Billy | +0.115 |
arrest | +0.113 |
Frankie | +0.112 |
Neg logit contributions | |
kin | -0.088 |
ows | -0.084 |
as | -0.082 |
a | -0.073 |
ute | -0.069 |
Token | telephone |
Token ID | 11426 |
Feature activation | +3.52e-01 |
Pos logit contributions | |
Johnson | +0.045 |
OUT | +0.045 |
arrest | +0.043 |
Billy | +0.042 |
Frankie | +0.041 |
Neg logit contributions | |
ows | -0.037 |
as | -0.036 |
kin | -0.036 |
a | -0.032 |
ily | -0.032 |
Token | in |
Token ID | 287 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
kin | +0.023 |
ows | +0.020 |
as | +0.019 |
ily | +0.019 |
ute | +0.018 |
Neg logit contributions | |
OUT | -0.040 |
arrest | -0.039 |
Johnson | -0.039 |
joy | -0.038 |
Billy | -0.037 |
Token | my |
Token ID | 616 |
Feature activation | -5.06e-01 |
Pos logit contributions | |
kin | +0.078 |
ows | +0.071 |
as | +0.069 |
ute | +0.066 |
ily | +0.066 |
Neg logit contributions | |
OUT | -0.079 |
arrest | -0.078 |
Johnson | -0.076 |
Billy | -0.074 |
Frankie | -0.073 |
Token | pig |
Token ID | 12967 |
Feature activation | -7.07e-01 |
Pos logit contributions | |
arrest | +0.047 |
Johnson | +0.047 |
OUT | +0.046 |
joy | +0.045 |
Billy | +0.044 |
Neg logit contributions | |
kin | -0.043 |
ows | -0.041 |
as | -0.038 |
ily | -0.038 |
a | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
OUT | +0.068 |
arrest | +0.066 |
Johnson | +0.064 |
joy | +0.064 |
Billy | +0.062 |
Neg logit contributions | |
kin | -0.058 |
ows | -0.054 |
as | -0.052 |
a | -0.049 |
ily | -0.049 |
Token | Let |
Token ID | 3914 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
OUT | +0.017 |
arrest | +0.017 |
joy | +0.016 |
Johnson | +0.016 |
istent | +0.015 |
Neg logit contributions | |
kin | -0.015 |
ows | -0.014 |
as | -0.013 |
ily | -0.013 |
a | -0.013 |
Token | 's |
Token ID | 338 |
Feature activation | -7.91e-02 |
Pos logit contributions | |
OUT | +0.121 |
joy | +0.119 |
arrest | +0.118 |
Johnson | +0.115 |
istent | +0.110 |
Neg logit contributions | |
kin | -0.106 |
a | -0.105 |
ily | -0.102 |
ows | -0.099 |
as | -0.098 |
Token | go |
Token ID | 467 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
OUT | +0.007 |
arrest | +0.007 |
Johnson | +0.007 |
Billy | +0.007 |
Frankie | +0.006 |
Neg logit contributions | |
kin | -0.007 |
ows | -0.007 |
as | -0.006 |
ily | -0.006 |
ute | -0.006 |
Token | home |
Token ID | 1363 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
kin | +0.019 |
ows | +0.019 |
as | +0.018 |
a | +0.018 |
ute | +0.018 |
Neg logit contributions | |
OUT | -0.029 |
arrest | -0.027 |
Johnson | -0.027 |
joy | -0.027 |
istent | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
OUT | +0.181 |
arrest | +0.176 |
Johnson | +0.171 |
joy | +0.167 |
Billy | +0.165 |
Neg logit contributions | |
kin | -0.125 |
ows | -0.115 |
as | -0.110 |
a | -0.105 |
ily | -0.104 |
Token | â |
Token ID | 22940 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
kin | +0.003 |
ows | +0.003 |
as | +0.003 |
ily | +0.003 |
a | +0.003 |
Neg logit contributions | |
OUT | -0.004 |
arrest | -0.004 |
joy | -0.004 |
Johnson | -0.004 |
istent | -0.003 |
Token | € |
Token ID | 26391 |
Feature activation | -9.21e-01 |
Pos logit contributions | |
kin | +0.032 |
ows | +0.030 |
as | +0.029 |
ily | +0.028 |
a | +0.027 |
Neg logit contributions | |
OUT | -0.031 |
arrest | -0.030 |
Johnson | -0.030 |
joy | -0.028 |
Billy | -0.028 |
Token | So |
Token ID | 1406 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
OUT | +0.086 |
Frankie | +0.081 |
arrest | +0.080 |
ited | +0.079 |
joy | +0.078 |
Neg logit contributions | |
ows | -0.074 |
kin | -0.073 |
as | -0.072 |
a | -0.069 |
ute | -0.064 |
Token | He |
Token ID | 679 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
arrest | +0.003 |
OUT | +0.003 |
joy | +0.002 |
returning | +0.002 |
istent | +0.002 |
Neg logit contributions | |
ows | -0.002 |
kin | -0.002 |
ily | -0.002 |
holes | -0.002 |
ia | -0.002 |
Token | had |
Token ID | 550 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
kin | +0.040 |
ows | +0.037 |
as | +0.035 |
ute | +0.034 |
ily | +0.033 |
Neg logit contributions | |
OUT | -0.044 |
arrest | -0.044 |
Johnson | -0.043 |
Billy | -0.042 |
joy | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
kin | +0.014 |
ows | +0.013 |
as | +0.012 |
ily | +0.012 |
ute | +0.011 |
Neg logit contributions | |
OUT | -0.021 |
arrest | -0.020 |
Johnson | -0.020 |
joy | -0.019 |
Billy | -0.019 |
Token | playful |
Token ID | 34264 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
OUT | +0.032 |
Johnson | +0.032 |
arrest | +0.032 |
Billy | +0.031 |
joy | +0.031 |
Neg logit contributions | |
kin | -0.020 |
ows | -0.019 |
as | -0.018 |
ily | -0.017 |
a | -0.016 |
Token | puppy |
Token ID | 26188 |
Feature activation | -1.67e+00 |
Pos logit contributions | |
OUT | +0.042 |
arrest | +0.041 |
Johnson | +0.040 |
istent | +0.039 |
Frankie | +0.038 |
Neg logit contributions | |
kin | -0.030 |
ows | -0.026 |
as | -0.025 |
ily | -0.025 |
holes | -0.025 |
Token | named |
Token ID | 3706 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
OUT | +0.162 |
arrest | +0.155 |
joy | +0.154 |
Johnson | +0.154 |
istent | +0.151 |
Neg logit contributions | |
kin | -0.136 |
ows | -0.123 |
as | -0.121 |
ily | -0.119 |
a | -0.117 |
Token | Max |
Token ID | 5436 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
joy | +0.002 |
arrest | +0.002 |
OUT | +0.002 |
ited | +0.002 |
istent | +0.002 |
Neg logit contributions | |
ows | -0.002 |
kin | -0.002 |
ily | -0.002 |
as | -0.002 |
ia | -0.002 |
Token | who |
Token ID | 508 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
OUT | +0.045 |
arrest | +0.045 |
joy | +0.043 |
Johnson | +0.042 |
istent | +0.042 |
Neg logit contributions | |
kin | -0.021 |
ows | -0.020 |
ily | -0.019 |
as | -0.018 |
a | -0.018 |
Token | would |
Token ID | 561 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
kin | +0.021 |
ows | +0.019 |
as | +0.018 |
a | +0.018 |
ily | +0.017 |
Neg logit contributions | |
OUT | -0.039 |
arrest | -0.038 |
joy | -0.037 |
Johnson | -0.036 |
istent | -0.035 |
Token | run |
Token ID | 1057 |
Feature activation | -5.31e-02 |
Pos logit contributions | |
kin | +0.020 |
ows | +0.018 |
as | +0.018 |
ily | +0.017 |
a | +0.017 |
Neg logit contributions | |
OUT | -0.018 |
joy | -0.018 |
Johnson | -0.017 |
arrest | -0.017 |
istent | -0.016 |
Token | around |
Token ID | 1088 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
OUT | +0.006 |
arrest | +0.006 |
Johnson | +0.005 |
Billy | +0.005 |
joy | +0.005 |
Neg logit contributions | |
kin | -0.004 |
ows | -0.003 |
as | -0.003 |
ily | -0.003 |
ute | -0.003 |
Token | tell |
Token ID | 1560 |
Feature activation | +6.95e-02 |
Pos logit contributions | |
kin | +0.115 |
ows | +0.107 |
as | +0.105 |
ily | +0.102 |
a | +0.102 |
Neg logit contributions | |
OUT | -0.109 |
arrest | -0.104 |
joy | -0.104 |
Johnson | -0.103 |
Billy | -0.097 |
Token | him |
Token ID | 683 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
kin | +0.007 |
ows | +0.007 |
as | +0.006 |
ute | +0.006 |
ily | +0.006 |
Neg logit contributions | |
OUT | -0.005 |
arrest | -0.005 |
Billy | -0.005 |
Johnson | -0.005 |
Frankie | -0.005 |
Token | your |
Token ID | 534 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
kin | +0.090 |
ows | +0.080 |
as | +0.079 |
ily | +0.076 |
ute | +0.075 |
Neg logit contributions | |
OUT | -0.100 |
arrest | -0.098 |
Johnson | -0.096 |
Billy | -0.093 |
joy | -0.092 |
Token | names |
Token ID | 3891 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
kin | +0.043 |
ows | +0.037 |
ily | +0.037 |
as | +0.036 |
auc | +0.035 |
Neg logit contributions | |
arrest | -0.062 |
Johnson | -0.060 |
OUT | -0.060 |
joy | -0.060 |
dreaming | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
kin | +0.110 |
ows | +0.100 |
as | +0.098 |
ute | +0.094 |
ily | +0.094 |
Neg logit contributions | |
OUT | -0.122 |
arrest | -0.120 |
Johnson | -0.120 |
Billy | -0.116 |
Frankie | -0.114 |
Token | what |
Token ID | 644 |
Feature activation | -4.31e-03 |
Pos logit contributions | |
arrest | +0.014 |
OUT | +0.014 |
Johnson | +0.013 |
Billy | +0.013 |
Frankie | +0.013 |
Neg logit contributions | |
kin | -0.016 |
ows | -0.015 |
as | -0.014 |
ily | -0.014 |
ute | -0.014 |
Token | you |
Token ID | 345 |
Feature activation | +9.93e-01 |
Pos logit contributions | |
OUT | +0.000 |
arrest | +0.000 |
Johnson | +0.000 |
Billy | +0.000 |
Frankie | +0.000 |
Neg logit contributions | |
kin | -0.000 |
ows | -0.000 |
as | -0.000 |
ute | -0.000 |
ily | -0.000 |
Token | like |
Token ID | 588 |
Feature activation | +8.74e-01 |
Pos logit contributions | |
kin | +0.084 |
ows | +0.077 |
as | +0.075 |
ily | +0.072 |
a | +0.071 |
Neg logit contributions | |
OUT | -0.093 |
arrest | -0.091 |
Johnson | -0.090 |
joy | -0.087 |
Billy | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | -7.06e-01 |
Pos logit contributions | |
kin | +0.072 |
ows | +0.065 |
as | +0.062 |
ute | +0.060 |
ily | +0.059 |
Neg logit contributions | |
arrest | -0.083 |
OUT | -0.083 |
Johnson | -0.081 |
Billy | -0.079 |
joy | -0.078 |
Token | do |
Token ID | 466 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
OUT | +0.063 |
arrest | +0.061 |
Johnson | +0.061 |
joy | +0.059 |
Billy | +0.057 |
Neg logit contributions | |
kin | -0.063 |
ows | -0.058 |
as | -0.056 |
ily | -0.054 |
a | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e-01 |
Pos logit contributions | |
kin | +0.077 |
ows | +0.069 |
as | +0.068 |
ute | +0.065 |
ily | +0.063 |
Neg logit contributions | |
arrest | -0.093 |
OUT | -0.093 |
Johnson | -0.089 |
Billy | -0.088 |
joy | -0.087 |
Token | time |
Token ID | 640 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
OUT | +0.033 |
arrest | +0.032 |
Johnson | +0.032 |
joy | +0.031 |
Billy | +0.030 |
Neg logit contributions | |
kin | -0.025 |
ows | -0.022 |
as | -0.022 |
ily | -0.021 |
a | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -5.22e-01 |
Pos logit contributions | |
kin | +0.031 |
ows | +0.027 |
as | +0.027 |
ily | +0.027 |
a | +0.026 |
Neg logit contributions | |
OUT | -0.034 |
arrest | -0.033 |
Johnson | -0.031 |
istent | -0.031 |
joy | -0.030 |
Token | there |
Token ID | 612 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
Johnson | +0.043 |
Billy | +0.042 |
OUT | +0.040 |
Frankie | +0.040 |
joy | +0.039 |
Neg logit contributions | |
ows | -0.045 |
kin | -0.045 |
as | -0.042 |
ute | -0.040 |
holes | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
kin | +0.005 |
ows | +0.005 |
as | +0.004 |
ily | +0.004 |
ute | +0.004 |
Neg logit contributions | |
OUT | -0.007 |
arrest | -0.007 |
Johnson | -0.007 |
Billy | -0.006 |
joy | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
arrest | +0.038 |
OUT | +0.038 |
Johnson | +0.038 |
Billy | +0.037 |
joy | +0.036 |
Neg logit contributions | |
kin | -0.031 |
ows | -0.029 |
as | -0.028 |
ute | -0.026 |
a | -0.025 |
Token | little |
Token ID | 1310 |
Feature activation | -9.96e-02 |
Pos logit contributions | |
OUT | +0.035 |
arrest | +0.034 |
Johnson | +0.034 |
joy | +0.033 |
Billy | +0.033 |
Neg logit contributions | |
kin | -0.029 |
ows | -0.027 |
as | -0.026 |
ily | -0.025 |
holes | -0.024 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
OUT | +0.010 |
Johnson | +0.010 |
arrest | +0.010 |
Billy | +0.010 |
joy | +0.010 |
Neg logit contributions | |
kin | -0.007 |
ows | -0.007 |
as | -0.007 |
ily | -0.006 |
holes | -0.006 |
Token | named |
Token ID | 3706 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
OUT | +0.017 |
arrest | +0.017 |
Johnson | +0.016 |
Billy | +0.016 |
joy | +0.015 |
Neg logit contributions | |
kin | -0.022 |
ows | -0.021 |
as | -0.020 |
ily | -0.019 |
holes | -0.019 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.40e-01 |
Pos logit contributions | |
OUT | +0.023 |
arrest | +0.022 |
Johnson | +0.021 |
joy | +0.021 |
istent | +0.021 |
Neg logit contributions | |
kin | -0.017 |
ows | -0.016 |
ily | -0.016 |
as | -0.015 |
a | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
kin | +0.016 |
ows | +0.014 |
as | +0.014 |
ily | +0.013 |
a | +0.013 |
Neg logit contributions | |
OUT | -0.027 |
arrest | -0.026 |
Johnson | -0.026 |
joy | -0.025 |
Billy | -0.025 |
Token | She |
Token ID | 1375 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
arrest | +0.008 |
OUT | +0.008 |
Johnson | +0.007 |
joy | +0.007 |
Frankie | +0.007 |
Neg logit contributions | |
kin | -0.007 |
ows | -0.006 |
ily | -0.006 |
as | -0.006 |
holes | -0.006 |
Token | Ben |
Token ID | 3932 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
OUT | +0.064 |
arrest | +0.063 |
Johnson | +0.061 |
joy | +0.059 |
Billy | +0.059 |
Neg logit contributions | |
kin | -0.055 |
ows | -0.050 |
as | -0.049 |
ily | -0.047 |
a | -0.046 |
Token | did |
Token ID | 750 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
kin | +0.090 |
ows | +0.080 |
as | +0.078 |
ute | +0.076 |
utes | +0.073 |
Neg logit contributions | |
OUT | -0.139 |
arrest | -0.138 |
Johnson | -0.135 |
Billy | -0.131 |
joy | -0.130 |
Token | not |
Token ID | 407 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
kin | +0.069 |
ows | +0.057 |
ily | +0.053 |
a | +0.050 |
as | +0.048 |
Neg logit contributions | |
OUT | -0.199 |
arrest | -0.191 |
Johnson | -0.191 |
joy | -0.191 |
Frankie | -0.187 |
Token | stop |
Token ID | 2245 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
OUT | +0.001 |
arrest | +0.001 |
Johnson | +0.001 |
Billy | +0.001 |
joy | +0.001 |
Neg logit contributions | |
kin | -0.002 |
ows | -0.002 |
as | -0.002 |
ute | -0.002 |
holes | -0.002 |
Token | crying |
Token ID | 13774 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
kin | +0.088 |
ows | +0.080 |
holes | +0.079 |
ily | +0.079 |
as | +0.078 |
Neg logit contributions | |
OUT | -0.113 |
arrest | -0.110 |
Johnson | -0.106 |
istent | -0.103 |
Frankie | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
Johnson | +0.080 |
OUT | +0.080 |
arrest | +0.078 |
Billy | +0.077 |
joy | +0.076 |
Neg logit contributions | |
kin | -0.084 |
ows | -0.074 |
as | -0.074 |
utes | -0.073 |
ute | -0.073 |
Token | He |
Token ID | 679 |
Feature activation | +4.38e-01 |
Pos logit contributions | |
kin | +0.002 |
as | +0.002 |
ute | +0.002 |
a | +0.002 |
utes | +0.002 |
Neg logit contributions | |
Johnson | -0.001 |
joy | -0.001 |
OUT | -0.001 |
Billy | -0.001 |
arrest | -0.001 |
Token | cried |
Token ID | 16896 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
kin | +0.044 |
ute | +0.042 |
ows | +0.041 |
utes | +0.041 |
as | +0.039 |
Neg logit contributions | |
arrest | -0.033 |
OUT | -0.032 |
Johnson | -0.031 |
Billy | -0.031 |
John | -0.028 |
Token | louder |
Token ID | 27089 |
Feature activation | +9.87e-01 |
Pos logit contributions | |
OUT | +0.002 |
Johnson | +0.002 |
Billy | +0.002 |
arrest | +0.002 |
joy | +0.002 |
Neg logit contributions | |
kin | -0.002 |
as | -0.002 |
ows | -0.002 |
ute | -0.002 |
utes | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
kin | +0.089 |
ows | +0.079 |
as | +0.076 |
ute | +0.076 |
a | +0.072 |
Neg logit contributions | |
arrest | -0.088 |
OUT | -0.086 |
joy | -0.085 |
Johnson | -0.082 |
Billy | -0.079 |
Token | louder |
Token ID | 27089 |
Feature activation | +9.98e-01 |
Pos logit contributions | |
kin | +0.012 |
ute | +0.011 |
as | +0.011 |
ows | +0.011 |
utes | +0.011 |
Neg logit contributions | |
arrest | -0.010 |
OUT | -0.009 |
Billy | -0.009 |
joy | -0.009 |
Johnson | -0.009 |
Token | cat |
Token ID | 3797 |
Feature activation | -5.37e-01 |
Pos logit contributions | |
kin | +0.022 |
ows | +0.021 |
as | +0.020 |
ily | +0.019 |
ute | +0.019 |
Neg logit contributions | |
OUT | -0.021 |
Johnson | -0.020 |
arrest | -0.020 |
Billy | -0.019 |
joy | -0.019 |
Token | took |
Token ID | 1718 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
OUT | +0.052 |
arrest | +0.051 |
Johnson | +0.050 |
joy | +0.048 |
Billy | +0.047 |
Neg logit contributions | |
kin | -0.044 |
ows | -0.040 |
as | -0.039 |
ily | -0.038 |
a | -0.037 |
Token | the |
Token ID | 262 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
kin | +0.010 |
ows | +0.009 |
as | +0.009 |
ily | +0.009 |
a | +0.009 |
Neg logit contributions | |
OUT | -0.011 |
arrest | -0.010 |
Johnson | -0.010 |
joy | -0.010 |
Billy | -0.010 |
Token | little |
Token ID | 1310 |
Feature activation | -2.77e-01 |
Pos logit contributions | |
OUT | +0.028 |
arrest | +0.028 |
Johnson | +0.027 |
joy | +0.026 |
Billy | +0.026 |
Neg logit contributions | |
kin | -0.028 |
ows | -0.025 |
as | -0.024 |
ily | -0.024 |
a | -0.023 |
Token | mouse |
Token ID | 10211 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
OUT | +0.027 |
arrest | +0.026 |
Johnson | +0.026 |
joy | +0.025 |
Billy | +0.025 |
Neg logit contributions | |
kin | -0.023 |
ows | -0.021 |
as | -0.020 |
ily | -0.019 |
ute | -0.019 |
Token | home |
Token ID | 1363 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
OUT | +0.004 |
Johnson | +0.004 |
arrest | +0.004 |
joy | +0.004 |
istent | +0.004 |
Neg logit contributions | |
kin | -0.003 |
ows | -0.003 |
as | -0.003 |
holes | -0.002 |
a | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +4.18e-01 |
Pos logit contributions | |
OUT | +0.206 |
arrest | +0.199 |
Johnson | +0.192 |
joy | +0.187 |
Frankie | +0.185 |
Neg logit contributions | |
kin | -0.126 |
ows | -0.116 |
as | -0.106 |
ily | -0.101 |
a | -0.100 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
kin | +0.036 |
ows | +0.032 |
a | +0.030 |
holes | +0.030 |
as | +0.030 |
Neg logit contributions | |
OUT | -0.040 |
arrest | -0.037 |
joy | -0.037 |
Johnson | -0.037 |
istent | -0.036 |
Token | to |
Token ID | 284 |
Feature activation | -5.24e-01 |
Pos logit contributions | |
kin | +0.080 |
ows | +0.076 |
ily | +0.072 |
a | +0.070 |
as | +0.070 |
Neg logit contributions | |
OUT | -0.141 |
arrest | -0.136 |
Johnson | -0.134 |
istent | -0.132 |
joy | -0.129 |
Token | keep |
Token ID | 1394 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
OUT | +0.043 |
Johnson | +0.041 |
arrest | +0.041 |
joy | +0.041 |
Billy | +0.038 |
Neg logit contributions | |
kin | -0.050 |
ows | -0.046 |
as | -0.045 |
ily | -0.044 |
a | -0.043 |
Token | it |
Token ID | 340 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
OUT | +0.005 |
istent | +0.005 |
arrest | +0.005 |
ited | +0.005 |
Johnson | +0.005 |
Neg logit contributions | |
kin | -0.003 |
holes | -0.003 |
ily | -0.003 |
a | -0.003 |
ows | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +5.26e-02 |
Pos logit contributions | |
kin | +0.044 |
ows | +0.041 |
as | +0.039 |
ily | +0.038 |
holes | +0.037 |
Neg logit contributions | |
OUT | -0.059 |
arrest | -0.058 |
Johnson | -0.056 |
joy | -0.055 |
Frankie | -0.054 |
Token | It |
Token ID | 632 |
Feature activation | +3.32e-01 |
Pos logit contributions | |
kin | +0.005 |
as | +0.005 |
a | +0.005 |
ute | +0.005 |
ows | +0.005 |
Neg logit contributions | |
Johnson | -0.004 |
OUT | -0.004 |
arrest | -0.004 |
joy | -0.004 |
Billy | -0.004 |
Token | had |
Token ID | 550 |
Feature activation | +2.43e-01 |
Pos logit contributions | |
utes | +0.033 |
ute | +0.033 |
kin | +0.033 |
ows | +0.032 |
as | +0.032 |
Neg logit contributions | |
Johnson | -0.024 |
returned | -0.022 |
Billy | -0.022 |
OUT | -0.021 |
Frankie | -0.021 |
Token | a |
Token ID | 257 |
Feature activation | -3.54e-01 |
Pos logit contributions | |
kin | +0.021 |
ute | +0.019 |
ows | +0.018 |
as | +0.018 |
utes | +0.017 |
Neg logit contributions | |
arrest | -0.024 |
OUT | -0.022 |
Billy | -0.022 |
Johnson | -0.022 |
joy | -0.021 |
Token | big |
Token ID | 1263 |
Feature activation | -2.75e-01 |
Pos logit contributions | |
OUT | +0.039 |
Johnson | +0.039 |
arrest | +0.038 |
Billy | +0.037 |
joy | +0.037 |
Neg logit contributions | |
kin | -0.023 |
ows | -0.022 |
as | -0.021 |
ily | -0.021 |
a | -0.019 |
Token | road |
Token ID | 2975 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
OUT | +0.034 |
arrest | +0.033 |
Johnson | +0.033 |
joy | +0.032 |
Billy | +0.031 |
Neg logit contributions | |
kin | -0.016 |
ows | -0.014 |
as | -0.013 |
ily | -0.012 |
a | -0.012 |
Token | with |
Token ID | 351 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
arrest | +0.060 |
Johnson | +0.059 |
OUT | +0.058 |
returning | +0.056 |
leaving | +0.055 |
Neg logit contributions | |
kin | -0.060 |
as | -0.054 |
ute | -0.054 |
utes | -0.053 |
ily | -0.052 |
Token | many |
Token ID | 867 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
arrest | +0.057 |
OUT | +0.054 |
Billy | +0.054 |
Johnson | +0.054 |
returning | +0.051 |
Neg logit contributions | |
kin | -0.047 |
ows | -0.040 |
ute | -0.040 |
as | -0.038 |
a | -0.036 |
Token | cars |
Token ID | 5006 |
Feature activation | +7.79e-01 |
Pos logit contributions | |
OUT | +0.065 |
arrest | +0.064 |
Johnson | +0.063 |
joy | +0.061 |
Billy | +0.061 |
Neg logit contributions | |
kin | -0.036 |
ows | -0.031 |
as | -0.030 |
ily | -0.028 |
ute | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
kin | +0.059 |
ows | +0.053 |
as | +0.051 |
ily | +0.049 |
ute | +0.048 |
Neg logit contributions | |
OUT | -0.080 |
arrest | -0.079 |
Johnson | -0.077 |
joy | -0.075 |
Billy | -0.074 |
Token | trucks |
Token ID | 13960 |
Feature activation | -8.45e-03 |
Pos logit contributions | |
arrest | +0.004 |
OUT | +0.004 |
Johnson | +0.004 |
Billy | +0.004 |
joy | +0.004 |
Neg logit contributions | |
kin | -0.003 |
ows | -0.002 |
as | -0.002 |
ute | -0.002 |
ily | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
kin | +0.039 |
ows | +0.036 |
as | +0.035 |
ily | +0.033 |
a | +0.033 |
Neg logit contributions | |
OUT | -0.045 |
arrest | -0.044 |
Johnson | -0.043 |
joy | -0.041 |
Billy | -0.041 |
Token | Her |
Token ID | 9360 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
kin | +0.029 |
ows | +0.027 |
as | +0.025 |
ily | +0.024 |
ute | +0.023 |
Neg logit contributions | |
OUT | -0.045 |
arrest | -0.044 |
Johnson | -0.043 |
Billy | -0.041 |
joy | -0.041 |
Token | mom |
Token ID | 1995 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
kin | +0.105 |
ows | +0.104 |
as | +0.096 |
ily | +0.092 |
ute | +0.090 |
Neg logit contributions | |
Johnson | -0.122 |
arrest | -0.120 |
OUT | -0.118 |
Billy | -0.118 |
joy | -0.113 |
Token | my |
Token ID | 1820 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
as | +0.102 |
a | +0.100 |
ows | +0.097 |
ily | +0.095 |
kin | +0.094 |
Neg logit contributions | |
arrest | -0.065 |
OUT | -0.063 |
joy | -0.060 |
Frankie | -0.055 |
Johnson | -0.054 |
Token | came |
Token ID | 1625 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
arrest | +0.151 |
OUT | +0.150 |
joy | +0.147 |
Johnson | +0.139 |
avail | +0.137 |
Neg logit contributions | |
a | -0.105 |
as | -0.097 |
ows | -0.094 |
ily | -0.093 |
kin | -0.092 |
Token | running |
Token ID | 2491 |
Feature activation | -7.70e-01 |
Pos logit contributions | |
kin | +0.048 |
as | +0.047 |
ows | +0.045 |
a | +0.045 |
ily | +0.043 |
Neg logit contributions | |
OUT | -0.097 |
arrest | -0.095 |
Johnson | -0.091 |
istent | -0.090 |
Frankie | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | +4.95e-01 |
Pos logit contributions | |
OUT | +0.077 |
arrest | +0.075 |
Johnson | +0.075 |
joy | +0.072 |
Billy | +0.071 |
Neg logit contributions | |
kin | -0.062 |
ows | -0.055 |
as | -0.054 |
ily | -0.051 |
ute | -0.051 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
a | +0.034 |
as | +0.033 |
ows | +0.032 |
kin | +0.031 |
ily | +0.030 |
Neg logit contributions | |
arrest | -0.052 |
OUT | -0.051 |
istent | -0.049 |
joy | -0.047 |
Johnson | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -6.74e-02 |
Pos logit contributions | |
kin | +0.062 |
ows | +0.058 |
ily | +0.056 |
as | +0.056 |
holes | +0.054 |
Neg logit contributions | |
OUT | -0.072 |
arrest | -0.069 |
Johnson | -0.068 |
joy | -0.067 |
istent | -0.066 |
Token | pan |
Token ID | 3425 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
arrest | +0.006 |
Johnson | +0.006 |
OUT | +0.006 |
joy | +0.006 |
Frankie | +0.006 |
Neg logit contributions | |
kin | -0.006 |
ows | -0.005 |
as | -0.005 |
ily | -0.005 |
ute | -0.005 |
Token | on |
Token ID | 319 |
Feature activation | +3.39e-01 |
Pos logit contributions | |
OUT | +0.012 |
arrest | +0.012 |
Johnson | +0.012 |
joy | +0.012 |
Billy | +0.011 |
Neg logit contributions | |
kin | -0.009 |
ows | -0.008 |
as | -0.008 |
ily | -0.007 |
a | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
OUT | +0.059 |
arrest | +0.058 |
Johnson | +0.057 |
joy | +0.055 |
Billy | +0.055 |
Neg logit contributions | |
kin | -0.049 |
ows | -0.044 |
as | -0.043 |
ily | -0.042 |
ute | -0.041 |
Token | We |
Token ID | 775 |
Feature activation | -8.41e-01 |
Pos logit contributions | |
kin | +0.006 |
ows | +0.006 |
as | +0.006 |
ily | +0.006 |
ute | +0.005 |
Neg logit contributions | |
OUT | -0.007 |
arrest | -0.007 |
Johnson | -0.007 |
joy | -0.007 |
istent | -0.006 |
Token | should |
Token ID | 815 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
OUT | +0.092 |
arrest | +0.091 |
Johnson | +0.089 |
Billy | +0.087 |
Frankie | +0.086 |
Neg logit contributions | |
kin | -0.058 |
ows | -0.052 |
as | -0.049 |
ily | -0.047 |
holes | -0.046 |
Token | think |
Token ID | 892 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
kin | +0.071 |
ows | +0.063 |
as | +0.062 |
ily | +0.061 |
a | +0.060 |
Neg logit contributions | |
OUT | -0.091 |
joy | -0.089 |
Johnson | -0.087 |
arrest | -0.086 |
Frankie | -0.080 |
Token | carefully |
Token ID | 7773 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
kin | +0.043 |
ows | +0.039 |
as | +0.038 |
ily | +0.036 |
ute | +0.036 |
Neg logit contributions | |
OUT | -0.053 |
arrest | -0.052 |
Johnson | -0.051 |
Billy | -0.049 |
joy | -0.049 |
Token | before |
Token ID | 878 |
Feature activation | -9.08e-01 |
Pos logit contributions | |
kin | +0.026 |
ows | +0.024 |
as | +0.023 |
ily | +0.022 |
a | +0.022 |
Neg logit contributions | |
OUT | -0.028 |
Johnson | -0.027 |
arrest | -0.027 |
joy | -0.026 |
Frankie | -0.026 |
Token | we |
Token ID | 356 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
OUT | +0.092 |
Johnson | +0.089 |
ited | +0.088 |
oured | +0.085 |
joy | +0.084 |
Neg logit contributions | |
kin | -0.063 |
holes | -0.059 |
shapes | -0.053 |
ires | -0.052 |
utes | -0.052 |
Token | do |
Token ID | 466 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
OUT | +0.011 |
arrest | +0.011 |
Johnson | +0.011 |
joy | +0.010 |
Billy | +0.010 |
Neg logit contributions | |
kin | -0.010 |
ows | -0.009 |
as | -0.009 |
ily | -0.009 |
ute | -0.009 |
Token | something |
Token ID | 1223 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
kin | +0.093 |
ows | +0.084 |
as | +0.081 |
ily | +0.078 |
a | +0.077 |
Neg logit contributions | |
OUT | -0.123 |
arrest | -0.120 |
Johnson | -0.119 |
joy | -0.114 |
Billy | -0.112 |
Token | to |
Token ID | 284 |
Feature activation | -5.64e-01 |
Pos logit contributions | |
OUT | +0.015 |
arrest | +0.014 |
Billy | +0.014 |
Johnson | +0.014 |
Frankie | +0.014 |
Neg logit contributions | |
kin | -0.012 |
ows | -0.011 |
as | -0.011 |
ily | -0.010 |
ute | -0.010 |
Token | make |
Token ID | 787 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
OUT | +0.055 |
joy | +0.053 |
Johnson | +0.052 |
arrest | +0.052 |
Frankie | +0.048 |
Neg logit contributions | |
kin | -0.047 |
ows | -0.042 |
as | -0.040 |
ily | -0.040 |
a | -0.039 |
Token | loud |
Token ID | 7812 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
OUT | +0.025 |
Johnson | +0.024 |
arrest | +0.024 |
Billy | +0.023 |
returning | +0.023 |
Neg logit contributions | |
kin | -0.020 |
ows | -0.018 |
as | -0.017 |
ily | -0.017 |
ires | -0.015 |
Token | animal |
Token ID | 5044 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
kin | +0.015 |
ows | +0.013 |
as | +0.012 |
ily | +0.012 |
ute | +0.012 |
Neg logit contributions | |
OUT | -0.026 |
arrest | -0.026 |
Johnson | -0.026 |
joy | -0.025 |
Billy | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
kin | +0.035 |
ows | +0.032 |
as | +0.031 |
ily | +0.030 |
a | +0.029 |
Neg logit contributions | |
OUT | -0.036 |
arrest | -0.035 |
Johnson | -0.034 |
Billy | -0.033 |
joy | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
arrest | +0.035 |
OUT | +0.035 |
Johnson | +0.032 |
Frankie | +0.032 |
istent | +0.032 |
Neg logit contributions | |
ows | -0.024 |
kin | -0.024 |
ily | -0.022 |
holes | -0.022 |
as | -0.021 |
Token | looked |
Token ID | 3114 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
kin | +0.021 |
ows | +0.019 |
as | +0.019 |
ily | +0.018 |
ute | +0.018 |
Neg logit contributions | |
OUT | -0.021 |
arrest | -0.020 |
Johnson | -0.020 |
Billy | -0.019 |
joy | -0.019 |
Token | up |
Token ID | 510 |
Feature activation | -7.84e-01 |
Pos logit contributions | |
kin | +0.025 |
ows | +0.023 |
as | +0.022 |
ily | +0.021 |
a | +0.020 |
Neg logit contributions | |
OUT | -0.044 |
arrest | -0.043 |
Johnson | -0.042 |
joy | -0.041 |
Billy | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | -6.50e-03 |
Pos logit contributions | |
OUT | +0.080 |
arrest | +0.079 |
Johnson | +0.077 |
joy | +0.076 |
Billy | +0.075 |
Neg logit contributions | |
kin | -0.060 |
ows | -0.055 |
as | -0.053 |
ily | -0.050 |
a | -0.049 |
Token | saw |
Token ID | 2497 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
arrest | +0.001 |
OUT | +0.001 |
Johnson | +0.001 |
istent | +0.001 |
Frankie | +0.001 |
Neg logit contributions | |
kin | -0.001 |
ows | -0.001 |
as | -0.000 |
holes | -0.000 |
ily | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
kin | +0.051 |
ows | +0.046 |
as | +0.044 |
ute | +0.044 |
utes | +0.041 |
Neg logit contributions | |
Billy | -0.049 |
arrest | -0.048 |
OUT | -0.048 |
Johnson | -0.048 |
John | -0.047 |
Token | big |
Token ID | 1263 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
OUT | +0.030 |
Johnson | +0.030 |
arrest | +0.030 |
joy | +0.029 |
Billy | +0.028 |
Neg logit contributions | |
kin | -0.020 |
ows | -0.018 |
as | -0.017 |
ily | -0.017 |
ute | -0.016 |
Token | dog |
Token ID | 3290 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
OUT | +0.011 |
arrest | +0.011 |
Johnson | +0.011 |
istent | +0.011 |
joy | +0.011 |
Neg logit contributions | |
kin | -0.008 |
ows | -0.007 |
holes | -0.007 |
as | -0.007 |
ily | -0.007 |
Token | € |
Token ID | 26391 |
Feature activation | -8.04e-01 |
Pos logit contributions | |
kin | +0.016 |
ows | +0.015 |
as | +0.015 |
ily | +0.014 |
holes | +0.014 |
Neg logit contributions | |
OUT | -0.015 |
arrest | -0.014 |
Johnson | -0.014 |
joy | -0.014 |
Billy | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
OUT | +0.073 |
arrest | +0.071 |
Johnson | +0.069 |
Frankie | +0.068 |
returning | +0.066 |
Neg logit contributions | |
kin | -0.067 |
ows | -0.065 |
as | -0.063 |
a | -0.061 |
ute | -0.058 |
Token |
|
Token ID | 198 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
arrest | +0.009 |
OUT | +0.009 |
Johnson | +0.008 |
joy | +0.008 |
istent | +0.008 |
Neg logit contributions | |
kin | -0.009 |
ows | -0.008 |
ily | -0.008 |
as | -0.008 |
a | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
OUT | +0.008 |
arrest | +0.008 |
Johnson | +0.008 |
joy | +0.008 |
Billy | +0.008 |
Neg logit contributions | |
kin | -0.008 |
ows | -0.007 |
as | -0.007 |
ily | -0.007 |
ute | -0.006 |
Token | S |
Token ID | 50 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
OUT | +0.015 |
arrest | +0.015 |
Johnson | +0.015 |
joy | +0.014 |
Billy | +0.014 |
Neg logit contributions | |
kin | -0.012 |
as | -0.011 |
ows | -0.011 |
ily | -0.011 |
a | -0.010 |
Token | aman |
Token ID | 10546 |
Feature activation | -4.91e+00 |
Pos logit contributions | |
arrest | +0.114 |
OUT | +0.112 |
Johnson | +0.108 |
Frankie | +0.107 |
Billy | +0.105 |
Neg logit contributions | |
kin | -0.077 |
as | -0.071 |
ows | -0.071 |
ily | -0.069 |
ia | -0.068 |
Token | tha |
Token ID | 12898 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
OUT | +0.343 |
arrest | +0.332 |
Johnson | +0.322 |
joy | +0.308 |
Billy | +0.303 |
Neg logit contributions | |
kin | -0.533 |
ows | -0.496 |
as | -0.485 |
ily | -0.473 |
a | -0.470 |
Token | was |
Token ID | 373 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
OUT | +0.230 |
arrest | +0.224 |
Johnson | +0.220 |
joy | +0.214 |
Billy | +0.208 |
Neg logit contributions | |
kin | -0.180 |
ows | -0.169 |
as | -0.164 |
a | -0.159 |
ily | -0.158 |
Token | amazed |
Token ID | 24562 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
OUT | +0.058 |
arrest | +0.057 |
Johnson | +0.056 |
joy | +0.054 |
Billy | +0.054 |
Neg logit contributions | |
kin | -0.045 |
ows | -0.041 |
as | -0.039 |
ily | -0.038 |
a | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
OUT | +0.120 |
Johnson | +0.120 |
joy | +0.120 |
Billy | +0.119 |
arrest | +0.116 |
Neg logit contributions | |
kin | -0.116 |
ows | -0.103 |
as | -0.096 |
ute | -0.093 |
ily | -0.093 |
Token | she |
Token ID | 673 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
kin | +0.017 |
ows | +0.016 |
as | +0.015 |
ily | +0.015 |
a | +0.015 |
Neg logit contributions | |
OUT | -0.018 |
arrest | -0.018 |
Johnson | -0.017 |
joy | -0.016 |
Billy | -0.016 |
Token | What |
Token ID | 2061 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
kin | +0.011 |
as | +0.010 |
ily | +0.010 |
a | +0.010 |
ows | +0.009 |
Neg logit contributions | |
OUT | -0.015 |
arrest | -0.015 |
Johnson | -0.014 |
joy | -0.014 |
Billy | -0.014 |
Token | do |
Token ID | 466 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
kin | +0.048 |
ows | +0.043 |
as | +0.042 |
ily | +0.040 |
a | +0.040 |
Neg logit contributions | |
OUT | -0.082 |
arrest | -0.080 |
Johnson | -0.079 |
joy | -0.077 |
Frankie | -0.075 |
Token | you |
Token ID | 345 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
kin | +0.089 |
ows | +0.082 |
as | +0.079 |
ily | +0.075 |
ute | +0.074 |
Neg logit contributions | |
OUT | -0.108 |
arrest | -0.107 |
Johnson | -0.104 |
Billy | -0.101 |
joy | -0.100 |
Token | think |
Token ID | 892 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
kin | +0.090 |
ows | +0.081 |
as | +0.079 |
ily | +0.076 |
a | +0.074 |
Neg logit contributions | |
OUT | -0.120 |
arrest | -0.119 |
Johnson | -0.117 |
joy | -0.114 |
Billy | -0.111 |
Token | you |
Token ID | 345 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
kin | +0.037 |
ows | +0.034 |
as | +0.033 |
ily | +0.032 |
a | +0.031 |
Neg logit contributions | |
OUT | -0.043 |
arrest | -0.042 |
Johnson | -0.042 |
joy | -0.040 |
Billy | -0.040 |
Token | 're |
Token ID | 821 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
kin | +0.084 |
ows | +0.079 |
as | +0.076 |
a | +0.072 |
ily | +0.071 |
Neg logit contributions | |
OUT | -0.124 |
arrest | -0.122 |
Johnson | -0.122 |
joy | -0.121 |
Frankie | -0.115 |
Token | doing |
Token ID | 1804 |
Feature activation | -3.68e-01 |
Pos logit contributions | |
arrest | +0.194 |
OUT | +0.186 |
joy | +0.184 |
Johnson | +0.179 |
istent | +0.164 |
Neg logit contributions | |
kin | -0.298 |
ows | -0.281 |
ily | -0.271 |
holes | -0.269 |
as | -0.265 |
Token | ?" |
Token ID | 1701 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
OUT | +0.036 |
arrest | +0.036 |
Johnson | +0.035 |
joy | +0.034 |
Frankie | +0.033 |
Neg logit contributions | |
kin | -0.029 |
ows | -0.027 |
as | -0.025 |
ily | -0.025 |
ute | -0.025 |
Token | The |
Token ID | 383 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
kin | +0.054 |
ows | +0.050 |
as | +0.048 |
ily | +0.048 |
a | +0.047 |
Neg logit contributions | |
OUT | -0.043 |
arrest | -0.042 |
Johnson | -0.041 |
joy | -0.040 |
Billy | -0.038 |
Token | baby |
Token ID | 5156 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
OUT | +0.028 |
arrest | +0.027 |
Johnson | +0.025 |
joy | +0.025 |
istent | +0.025 |
Neg logit contributions | |
kin | -0.037 |
ows | -0.033 |
as | -0.033 |
ily | -0.033 |
holes | -0.032 |
Token | looked |
Token ID | 3114 |
Feature activation | +5.02e-01 |
Pos logit contributions | |
arrest | +0.187 |
OUT | +0.182 |
joy | +0.175 |
istent | +0.168 |
Johnson | +0.167 |
Neg logit contributions | |
kin | -0.123 |
a | -0.116 |
ows | -0.112 |
as | -0.110 |
ily | -0.108 |
Token | She |
Token ID | 1375 |
Feature activation | -7.38e-01 |
Pos logit contributions | |
OUT | +0.009 |
arrest | +0.009 |
Johnson | +0.009 |
joy | +0.008 |
Billy | +0.008 |
Neg logit contributions | |
kin | -0.010 |
ows | -0.009 |
as | -0.009 |
ily | -0.008 |
a | -0.008 |
Token | said |
Token ID | 531 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
OUT | +0.062 |
arrest | +0.061 |
Johnson | +0.060 |
Billy | +0.057 |
joy | +0.057 |
Neg logit contributions | |
kin | -0.070 |
ows | -0.064 |
as | -0.062 |
ute | -0.060 |
ily | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
kin | +0.002 |
ows | +0.002 |
as | +0.002 |
ute | +0.002 |
ily | +0.002 |
Neg logit contributions | |
OUT | -0.001 |
arrest | -0.001 |
Johnson | -0.001 |
joy | -0.001 |
Billy | -0.001 |
Token | " |
Token ID | 366 |
Feature activation | -3.60e-01 |
Pos logit contributions | |
OUT | +0.034 |
arrest | +0.033 |
Johnson | +0.032 |
joy | +0.031 |
Billy | +0.030 |
Neg logit contributions | |
kin | -0.033 |
ows | -0.030 |
as | -0.029 |
ily | -0.028 |
a | -0.027 |
Token | You |
Token ID | 1639 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
OUT | +0.038 |
arrest | +0.037 |
Johnson | +0.036 |
joy | +0.035 |
Billy | +0.035 |
Neg logit contributions | |
kin | -0.027 |
ows | -0.024 |
as | -0.024 |
ily | -0.023 |
a | -0.022 |
Token | 're |
Token ID | 821 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
kin | +0.009 |
ows | +0.009 |
as | +0.008 |
ily | +0.008 |
a | +0.008 |
Neg logit contributions | |
OUT | -0.012 |
arrest | -0.012 |
Johnson | -0.012 |
joy | -0.011 |
Billy | -0.011 |
Token | welcome |
Token ID | 7062 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
OUT | +0.285 |
Johnson | +0.283 |
arrest | +0.279 |
Billy | +0.276 |
joy | +0.271 |
Neg logit contributions | |
kin | -0.216 |
ows | -0.200 |
as | -0.195 |
ute | -0.187 |
a | -0.178 |
Token | , |
Token ID | 11 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
OUT | +0.095 |
arrest | +0.093 |
Johnson | +0.090 |
joy | +0.086 |
Billy | +0.085 |
Neg logit contributions | |
kin | -0.100 |
ows | -0.092 |
as | -0.090 |
ily | -0.088 |
a | -0.086 |
Token | my |
Token ID | 616 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
OUT | +0.058 |
arrest | +0.057 |
Johnson | +0.055 |
joy | +0.054 |
Frankie | +0.053 |
Neg logit contributions | |
kin | -0.044 |
ows | -0.039 |
as | -0.039 |
ily | -0.038 |
a | -0.036 |
Token | loves |
Token ID | 10408 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
OUT | +0.046 |
Johnson | +0.046 |
joy | +0.046 |
arrest | +0.045 |
Billy | +0.044 |
Neg logit contributions | |
kin | -0.046 |
ows | -0.044 |
as | -0.042 |
ily | -0.040 |
ute | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
arrest | +0.122 |
OUT | +0.121 |
Johnson | +0.119 |
joy | +0.116 |
Billy | +0.114 |
Neg logit contributions | |
kin | -0.110 |
ows | -0.103 |
as | -0.097 |
ute | -0.095 |
ily | -0.092 |
Token | Sarah |
Token ID | 29284 |
Feature activation | -9.74e-01 |
Pos logit contributions | |
kin | +0.070 |
as | +0.066 |
ows | +0.065 |
ily | +0.065 |
a | +0.064 |
Neg logit contributions | |
OUT | -0.047 |
arrest | -0.045 |
Johnson | -0.044 |
joy | -0.042 |
Frankie | -0.041 |
Token | remembered |
Token ID | 12086 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
OUT | +0.096 |
arrest | +0.094 |
Johnson | +0.092 |
joy | +0.090 |
Frankie | +0.087 |
Neg logit contributions | |
kin | -0.079 |
ows | -0.072 |
as | -0.070 |
a | -0.069 |
ily | -0.067 |
Token | that |
Token ID | 326 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
kin | +0.010 |
ily | +0.009 |
ows | +0.009 |
as | +0.009 |
a | +0.008 |
Neg logit contributions | |
OUT | -0.014 |
arrest | -0.014 |
Johnson | -0.014 |
istent | -0.013 |
Frankie | -0.013 |
Token | she |
Token ID | 673 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
kin | +0.075 |
ows | +0.067 |
ily | +0.065 |
as | +0.065 |
holes | +0.064 |
Neg logit contributions | |
OUT | -0.077 |
arrest | -0.075 |
Johnson | -0.071 |
joy | -0.069 |
istent | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
OUT | +0.061 |
arrest | +0.059 |
Johnson | +0.059 |
joy | +0.056 |
Frankie | +0.055 |
Neg logit contributions | |
kin | -0.048 |
ows | -0.044 |
as | -0.043 |
a | -0.042 |
ily | -0.042 |
Token | supposed |
Token ID | 4385 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
OUT | +0.011 |
arrest | +0.011 |
Johnson | +0.010 |
istent | +0.010 |
joy | +0.010 |
Neg logit contributions | |
kin | -0.011 |
ily | -0.010 |
ows | -0.010 |
a | -0.010 |
as | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
OUT | +0.317 |
Johnson | +0.300 |
arrest | +0.299 |
istent | +0.297 |
joy | +0.283 |
Neg logit contributions | |
kin | -0.194 |
as | -0.183 |
ows | -0.177 |
ily | -0.176 |
a | -0.174 |
Token | put |
Token ID | 1234 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
OUT | +0.045 |
joy | +0.044 |
arrest | +0.043 |
Johnson | +0.043 |
Frankie | +0.039 |
Neg logit contributions | |
kin | -0.040 |
ows | -0.035 |
as | -0.034 |
ily | -0.034 |
a | -0.033 |
Token | on |
Token ID | 319 |
Feature activation | +3.51e-01 |
Pos logit contributions | |
kin | +0.030 |
as | +0.028 |
ows | +0.028 |
ia | +0.026 |
ute | +0.026 |
Neg logit contributions | |
OUT | -0.032 |
Billy | -0.031 |
Johnson | -0.031 |
arrest | -0.031 |
Frankie | -0.030 |
Token | her |
Token ID | 607 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
kin | +0.030 |
ows | +0.027 |
holes | +0.026 |
as | +0.026 |
ily | +0.026 |
Neg logit contributions | |
OUT | -0.033 |
arrest | -0.031 |
Johnson | -0.031 |
istent | -0.030 |
joy | -0.030 |
Token | coat |
Token ID | 13209 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
kin | +0.036 |
ows | +0.033 |
as | +0.031 |
ily | +0.030 |
ute | +0.030 |
Neg logit contributions | |
OUT | -0.048 |
arrest | -0.047 |
Johnson | -0.046 |
joy | -0.044 |
Frankie | -0.044 |
Token | � |
Token ID | 241 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
OUT | +0.065 |
Johnson | +0.064 |
arrest | +0.064 |
joy | +0.062 |
Billy | +0.060 |
Neg logit contributions | |
kin | -0.071 |
ows | -0.067 |
as | -0.064 |
ily | -0.063 |
ute | -0.062 |
Token | You |
Token ID | 1639 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
OUT | +0.056 |
arrest | +0.055 |
Johnson | +0.053 |
Billy | +0.051 |
joy | +0.051 |
Neg logit contributions | |
kin | -0.038 |
ows | -0.034 |
as | -0.034 |
ily | -0.033 |
a | -0.032 |
Token | â |
Token ID | 22940 |
Feature activation | +5.50e-01 |
Pos logit contributions | |
kin | +0.018 |
ows | +0.016 |
as | +0.016 |
ily | +0.015 |
a | +0.015 |
Neg logit contributions | |
OUT | -0.018 |
arrest | -0.018 |
Johnson | -0.018 |
joy | -0.017 |
Billy | -0.016 |
Token | € |
Token ID | 26391 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
kin | +0.049 |
ows | +0.045 |
as | +0.045 |
ily | +0.042 |
a | +0.042 |
Neg logit contributions | |
OUT | -0.049 |
arrest | -0.047 |
Johnson | -0.047 |
joy | -0.045 |
Billy | -0.045 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
OUT | +0.049 |
Frankie | +0.047 |
arrest | +0.047 |
Johnson | +0.047 |
returning | +0.045 |
Neg logit contributions | |
kin | -0.059 |
ows | -0.056 |
as | -0.055 |
a | -0.052 |
ily | -0.050 |
Token | re |
Token ID | 260 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
arrest | +0.151 |
OUT | +0.148 |
joy | +0.136 |
Frankie | +0.135 |
returning | +0.130 |
Neg logit contributions | |
kin | -0.161 |
as | -0.155 |
ows | -0.154 |
a | -0.147 |
ily | -0.145 |
Token | welcome |
Token ID | 7062 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
OUT | +0.323 |
arrest | +0.318 |
Johnson | +0.313 |
joy | +0.303 |
Billy | +0.298 |
Neg logit contributions | |
kin | -0.205 |
ows | -0.183 |
as | -0.177 |
ily | -0.170 |
a | -0.165 |
Token | ! |
Token ID | 0 |
Feature activation | -7.69e-01 |
Pos logit contributions | |
OUT | +0.092 |
arrest | +0.092 |
Johnson | +0.086 |
istent | +0.083 |
joy | +0.082 |
Neg logit contributions | |
kin | -0.084 |
ows | -0.076 |
as | -0.075 |
ily | -0.074 |
a | -0.073 |
Token | Now |
Token ID | 2735 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
OUT | +0.070 |
arrest | +0.070 |
Johnson | +0.067 |
joy | +0.065 |
Frankie | +0.063 |
Neg logit contributions | |
kin | -0.065 |
ows | -0.060 |
as | -0.059 |
ily | -0.058 |
a | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
kin | +0.019 |
ows | +0.017 |
as | +0.017 |
ily | +0.017 |
a | +0.016 |
Neg logit contributions | |
OUT | -0.021 |
arrest | -0.021 |
Johnson | -0.021 |
joy | -0.020 |
istent | -0.019 |
Token | when |
Token ID | 618 |
Feature activation | -2.72e-01 |
Pos logit contributions | |
OUT | +0.041 |
arrest | +0.040 |
Johnson | +0.039 |
joy | +0.038 |
Billy | +0.037 |
Neg logit contributions | |
kin | -0.036 |
ows | -0.033 |
as | -0.032 |
ily | -0.031 |
a | -0.030 |