Token | sailed |
Token ID | 33835 |
Feature activation | +5.92e-02 |
Pos logit contributions | |
ang | +0.084 |
irling | +0.078 |
irl | +0.075 |
ading | +0.072 |
ash | +0.069 |
Neg logit contributions | |
ranger | -0.048 |
sob | -0.047 |
recovered | -0.047 |
department | -0.045 |
aid | -0.045 |
Token | for |
Token ID | 329 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
ang | +0.114 |
irling | +0.102 |
ee | +0.098 |
irl | +0.095 |
ug | +0.092 |
Neg logit contributions | |
ranger | -0.104 |
bleed | -0.101 |
bru | -0.100 |
hinges | -0.100 |
sob | -0.099 |
Token | days |
Token ID | 1528 |
Feature activation | -7.16e-03 |
Pos logit contributions | |
ang | +0.035 |
irling | +0.030 |
ee | +0.028 |
irl | +0.027 |
ash | +0.026 |
Neg logit contributions | |
ranger | -0.042 |
hinges | -0.041 |
bru | -0.041 |
bleed | -0.041 |
sob | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
ranger | +0.009 |
hinges | +0.008 |
aid | +0.008 |
injuries | +0.008 |
bru | +0.008 |
Neg logit contributions | |
ang | -0.016 |
irl | -0.016 |
irling | -0.015 |
ee | -0.015 |
ash | -0.015 |
Token | exploring |
Token ID | 13504 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
ang | +0.023 |
irling | +0.021 |
ee | +0.021 |
irl | +0.020 |
ash | +0.020 |
Neg logit contributions | |
ranger | -0.018 |
sob | -0.016 |
bru | -0.016 |
bleed | -0.016 |
hinges | -0.016 |
Token | unch |
Token ID | 16258 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
ang | +0.093 |
irling | +0.083 |
ee | +0.080 |
irl | +0.077 |
ug | +0.074 |
Neg logit contributions | |
ranger | -0.082 |
hinges | -0.078 |
bru | -0.077 |
bleed | -0.076 |
sob | -0.076 |
Token | arted |
Token ID | 19112 |
Feature activation | +4.28e-02 |
Pos logit contributions | |
ang | +0.246 |
irling | +0.235 |
irl | +0.226 |
ee | +0.220 |
ading | +0.191 |
Neg logit contributions | |
ranger | -0.287 |
hinges | -0.277 |
bleed | -0.275 |
onna | -0.274 |
bru | -0.261 |
Token | waters |
Token ID | 10150 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
ang | +0.077 |
irling | +0.069 |
irl | +0.065 |
ee | +0.065 |
ug | +0.060 |
Neg logit contributions | |
ranger | -0.082 |
bleed | -0.079 |
hinges | -0.078 |
bru | -0.077 |
sob | -0.076 |
Token | until |
Token ID | 1566 |
Feature activation | +1.82e-03 |
Pos logit contributions | |
ang | +0.100 |
irling | +0.091 |
ee | +0.089 |
irl | +0.087 |
ug | +0.082 |
Neg logit contributions | |
ranger | -0.069 |
bleed | -0.067 |
hinges | -0.065 |
bru | -0.065 |
sob | -0.064 |
Token | he |
Token ID | 339 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.003 |
irl | +0.003 |
ug | +0.003 |
Neg logit contributions | |
ranger | -0.003 |
bleed | -0.003 |
bru | -0.003 |
hinges | -0.003 |
sob | -0.003 |
Token | came |
Token ID | 1625 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
ang | +0.042 |
irling | +0.038 |
ee | +0.036 |
irl | +0.036 |
ug | +0.035 |
Neg logit contributions | |
ranger | -0.026 |
bleed | -0.024 |
bru | -0.024 |
sob | -0.024 |
department | -0.024 |
Token | different |
Token ID | 1180 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
ang | +0.031 |
irling | +0.028 |
ee | +0.027 |
irl | +0.026 |
ug | +0.025 |
Neg logit contributions | |
ranger | -0.037 |
bleed | -0.036 |
bru | -0.035 |
hinges | -0.035 |
sob | -0.035 |
Token | styles |
Token ID | 12186 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
ang | +0.036 |
irling | +0.032 |
ee | +0.031 |
irl | +0.030 |
ading | +0.029 |
Neg logit contributions | |
ranger | -0.049 |
bleed | -0.049 |
bru | -0.047 |
sob | -0.047 |
department | -0.046 |
Token | like |
Token ID | 588 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
ang | +0.037 |
irling | +0.034 |
ee | +0.034 |
irl | +0.033 |
ug | +0.032 |
Neg logit contributions | |
ranger | -0.015 |
bleed | -0.014 |
hinges | -0.014 |
bru | -0.014 |
sob | -0.014 |
Token | stars |
Token ID | 5788 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
ang | +0.034 |
irling | +0.030 |
ee | +0.029 |
irl | +0.028 |
ug | +0.027 |
Neg logit contributions | |
ranger | -0.037 |
bleed | -0.037 |
bru | -0.037 |
hinges | -0.036 |
sob | -0.036 |
Token | or |
Token ID | 393 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
ang | +0.079 |
irling | +0.073 |
ee | +0.071 |
irl | +0.069 |
ug | +0.067 |
Neg logit contributions | |
ranger | -0.038 |
bleed | -0.037 |
bru | -0.037 |
hinges | -0.036 |
sob | -0.036 |
Token | spir |
Token ID | 9158 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
ang | +0.042 |
irling | +0.038 |
ee | +0.036 |
irl | +0.035 |
ug | +0.033 |
Neg logit contributions | |
ranger | -0.051 |
bleed | -0.050 |
bru | -0.049 |
hinges | -0.049 |
sob | -0.049 |
Token | als |
Token ID | 874 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
ang | +0.274 |
irling | +0.267 |
irl | +0.245 |
ee | +0.224 |
ading | +0.220 |
Neg logit contributions | |
bleed | -0.227 |
department | -0.214 |
hinges | -0.212 |
bru | -0.211 |
onna | -0.210 |
Token | . |
Token ID | 13 |
Feature activation | -5.24e-03 |
Pos logit contributions | |
ang | +0.075 |
irling | +0.068 |
ee | +0.067 |
ug | +0.063 |
irl | +0.063 |
Neg logit contributions | |
bru | -0.052 |
ranger | -0.052 |
sob | -0.050 |
hinges | -0.050 |
bleed | -0.050 |
Token | Everyone |
Token ID | 11075 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
ranger | +0.008 |
bleed | +0.008 |
bru | +0.008 |
hinges | +0.008 |
sob | +0.008 |
Neg logit contributions | |
ang | -0.011 |
irling | -0.010 |
ee | -0.010 |
irl | -0.010 |
ug | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
ang | +0.093 |
irling | +0.087 |
irl | +0.083 |
ug | +0.081 |
ee | +0.080 |
Neg logit contributions | |
sob | -0.056 |
ranger | -0.055 |
bleed | -0.052 |
keeper | -0.051 |
bru | -0.049 |
Token | so |
Token ID | 523 |
Feature activation | +6.49e-03 |
Pos logit contributions | |
ang | +0.031 |
ee | +0.027 |
irling | +0.026 |
ash | +0.026 |
irl | +0.025 |
Neg logit contributions | |
ranger | -0.028 |
bleed | -0.028 |
sob | -0.028 |
bru | -0.027 |
keeper | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.51e-03 |
Pos logit contributions | |
bleed | +0.027 |
bru | +0.027 |
ranger | +0.026 |
hinges | +0.026 |
sob | +0.026 |
Neg logit contributions | |
ang | -0.029 |
irling | -0.026 |
ee | -0.026 |
ug | -0.024 |
irl | -0.024 |
Token | Paul |
Token ID | 3362 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.003 |
irl | +0.003 |
ading | +0.003 |
Neg logit contributions | |
ranger | -0.002 |
hinges | -0.002 |
bleed | -0.002 |
bru | -0.002 |
sob | -0.002 |
Token | 's |
Token ID | 338 |
Feature activation | -9.87e-03 |
Pos logit contributions | |
ang | +0.103 |
irling | +0.094 |
irl | +0.092 |
ading | +0.090 |
ee | +0.087 |
Neg logit contributions | |
ranger | -0.059 |
sob | -0.056 |
keeper | -0.053 |
department | -0.052 |
bru | -0.052 |
Token | impressive |
Token ID | 8036 |
Feature activation | -6.38e-04 |
Pos logit contributions | |
ranger | +0.015 |
bleed | +0.015 |
bru | +0.014 |
hinges | +0.014 |
sob | +0.014 |
Neg logit contributions | |
ang | -0.021 |
irling | -0.019 |
ee | -0.019 |
irl | -0.018 |
ug | -0.018 |
Token | football |
Token ID | 4346 |
Feature activation | +7.73e-03 |
Pos logit contributions | |
ranger | +0.001 |
bleed | +0.001 |
bru | +0.001 |
hinges | +0.001 |
sob | +0.001 |
Neg logit contributions | |
ang | -0.001 |
irling | -0.001 |
ee | -0.001 |
irl | -0.001 |
ug | -0.001 |
Token | spir |
Token ID | 9158 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
ang | +0.015 |
irling | +0.013 |
ee | +0.013 |
irl | +0.013 |
ug | +0.012 |
Neg logit contributions | |
bleed | -0.013 |
bru | -0.013 |
hinges | -0.013 |
ranger | -0.013 |
sob | -0.013 |
Token | alled |
Token ID | 4262 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
irling | +0.220 |
ang | +0.206 |
irl | +0.200 |
icing | +0.166 |
ading | +0.162 |
Neg logit contributions | |
department | -0.244 |
ranger | -0.239 |
bleed | -0.239 |
driver | -0.238 |
recovered | -0.224 |
Token | through |
Token ID | 832 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
ang | +0.068 |
irling | +0.061 |
ee | +0.058 |
irl | +0.057 |
ug | +0.054 |
Neg logit contributions | |
ranger | -0.067 |
bleed | -0.066 |
bru | -0.065 |
hinges | -0.065 |
sob | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
ang | +0.073 |
irling | +0.064 |
ee | +0.061 |
irl | +0.060 |
ug | +0.057 |
Neg logit contributions | |
ranger | -0.081 |
bleed | -0.079 |
bru | -0.078 |
hinges | -0.078 |
sob | -0.077 |
Token | air |
Token ID | 1633 |
Feature activation | +1.48e-03 |
Pos logit contributions | |
ang | +0.072 |
irling | +0.064 |
ug | +0.060 |
irl | +0.060 |
ee | +0.059 |
Neg logit contributions | |
ranger | -0.066 |
hinges | -0.062 |
bru | -0.060 |
department | -0.060 |
bleed | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.003 |
irl | +0.003 |
ug | +0.003 |
Neg logit contributions | |
ranger | -0.002 |
bleed | -0.002 |
bru | -0.002 |
hinges | -0.002 |
sob | -0.002 |
Token | sun |
Token ID | 4252 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
ang | +0.029 |
irling | +0.026 |
ee | +0.025 |
irl | +0.025 |
ug | +0.024 |
Neg logit contributions | |
ranger | -0.023 |
bleed | -0.023 |
sob | -0.023 |
bru | -0.022 |
hinges | -0.022 |
Token | fl |
Token ID | 2704 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
ang | +0.075 |
irling | +0.069 |
ee | +0.065 |
irl | +0.065 |
ug | +0.063 |
Neg logit contributions | |
ranger | -0.052 |
bleed | -0.051 |
hinges | -0.050 |
sob | -0.049 |
bru | -0.049 |
Token | owers |
Token ID | 3618 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
ang | +0.150 |
irling | +0.139 |
irl | +0.127 |
ee | +0.126 |
ug | +0.122 |
Neg logit contributions | |
bleed | -0.143 |
bru | -0.137 |
ranger | -0.137 |
sob | -0.136 |
hinges | -0.136 |
Token | , |
Token ID | 11 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
ang | +0.101 |
irling | +0.092 |
ee | +0.088 |
irl | +0.086 |
ug | +0.086 |
Neg logit contributions | |
ranger | -0.066 |
bleed | -0.066 |
sob | -0.063 |
hinges | -0.062 |
bru | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
ang | +0.039 |
irling | +0.035 |
ee | +0.034 |
irl | +0.033 |
ug | +0.032 |
Neg logit contributions | |
ranger | -0.026 |
bleed | -0.026 |
sob | -0.026 |
bru | -0.026 |
hinges | -0.026 |
Token | tul |
Token ID | 48373 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ang | +0.049 |
irling | +0.044 |
ee | +0.042 |
irl | +0.041 |
ug | +0.040 |
Neg logit contributions | |
bleed | -0.043 |
ranger | -0.043 |
bru | -0.042 |
sob | -0.042 |
hinges | -0.041 |
Token | ips |
Token ID | 2419 |
Feature activation | +5.94e-02 |
Pos logit contributions | |
ang | +0.186 |
irling | +0.168 |
ee | +0.156 |
ug | +0.153 |
irl | +0.150 |
Neg logit contributions | |
onna | -0.273 |
recovered | -0.257 |
bleed | -0.256 |
bru | -0.256 |
department | -0.250 |
Token | . |
Token ID | 13 |
Feature activation | +6.23e-03 |
Pos logit contributions | |
ang | +0.118 |
irling | +0.106 |
ee | +0.102 |
irl | +0.099 |
ug | +0.096 |
Neg logit contributions | |
ranger | -0.099 |
bleed | -0.098 |
bru | -0.096 |
hinges | -0.096 |
sob | -0.095 |
Token | They |
Token ID | 1119 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
ang | +0.014 |
irling | +0.012 |
ee | +0.012 |
irl | +0.012 |
ug | +0.012 |
Neg logit contributions | |
bleed | -0.009 |
bru | -0.009 |
ranger | -0.009 |
hinges | -0.009 |
sob | -0.008 |
Token | see |
Token ID | 766 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
ang | +0.050 |
irling | +0.045 |
ee | +0.043 |
ug | +0.042 |
irl | +0.041 |
Neg logit contributions | |
bleed | -0.030 |
ranger | -0.030 |
sob | -0.029 |
bru | -0.029 |
hinges | -0.028 |
Token | pine |
Token ID | 20161 |
Feature activation | +5.44e-02 |
Pos logit contributions | |
ang | +0.058 |
irling | +0.051 |
ee | +0.049 |
irl | +0.048 |
ug | +0.046 |
Neg logit contributions | |
ranger | -0.059 |
bleed | -0.058 |
bru | -0.057 |
sob | -0.057 |
hinges | -0.057 |
Token | ! |
Token ID | 0 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ranger | +0.011 |
bleed | +0.011 |
bru | +0.011 |
hinges | +0.011 |
sob | +0.010 |
Neg logit contributions | |
ang | -0.009 |
irling | -0.008 |
ee | -0.007 |
irl | -0.007 |
ug | -0.007 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ang | +0.028 |
irling | +0.025 |
ee | +0.025 |
irl | +0.024 |
ug | +0.023 |
Neg logit contributions | |
bleed | -0.015 |
ranger | -0.015 |
bru | -0.014 |
hinges | -0.014 |
sob | -0.014 |
Token | One |
Token ID | 3198 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
ang | +0.036 |
ee | +0.032 |
irling | +0.032 |
irl | +0.031 |
ug | +0.030 |
Neg logit contributions | |
bleed | -0.026 |
hinges | -0.026 |
ranger | -0.025 |
bru | -0.025 |
sob | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
ang | +0.090 |
irling | +0.084 |
ug | +0.075 |
angles | +0.075 |
irl | +0.069 |
Neg logit contributions | |
ranger | -0.105 |
aid | -0.097 |
bru | -0.097 |
keeper | -0.096 |
sob | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
ranger | +0.019 |
bleed | +0.019 |
bru | +0.019 |
hinges | +0.019 |
sob | +0.019 |
Neg logit contributions | |
ang | -0.026 |
irling | -0.024 |
ee | -0.023 |
irl | -0.022 |
ug | -0.021 |
Token | Sil |
Token ID | 4243 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
ang | +0.083 |
irling | +0.076 |
ee | +0.072 |
irl | +0.071 |
ug | +0.069 |
Neg logit contributions | |
ranger | -0.055 |
bleed | -0.054 |
sob | -0.053 |
bru | -0.053 |
hinges | -0.052 |
Token | via |
Token ID | 8869 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
ang | +0.237 |
irling | +0.215 |
irl | +0.197 |
ading | +0.194 |
ee | +0.189 |
Neg logit contributions | |
aid | -0.208 |
ranger | -0.191 |
onna | -0.190 |
sob | -0.189 |
hinges | -0.186 |
Token | was |
Token ID | 373 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ang | +0.098 |
irling | +0.089 |
ee | +0.085 |
irl | +0.084 |
ug | +0.082 |
Neg logit contributions | |
ranger | -0.059 |
bleed | -0.058 |
sob | -0.058 |
bru | -0.057 |
hinges | -0.056 |
Token | getting |
Token ID | 1972 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
ang | +0.029 |
angles | +0.029 |
ash | +0.026 |
irling | +0.026 |
izing | +0.025 |
Neg logit contributions | |
sob | -0.024 |
bleed | -0.022 |
aid | -0.021 |
bru | -0.021 |
cough | -0.021 |
Token | ready |
Token ID | 3492 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
ang | +0.033 |
irling | +0.030 |
ee | +0.028 |
irl | +0.027 |
ug | +0.027 |
Neg logit contributions | |
bleed | -0.029 |
ranger | -0.029 |
sob | -0.029 |
bru | -0.029 |
hinges | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
ranger | +0.045 |
bleed | +0.045 |
bru | +0.043 |
hinges | +0.042 |
sob | +0.042 |
Neg logit contributions | |
ang | -0.091 |
irling | -0.085 |
ee | -0.083 |
irl | -0.081 |
ug | -0.078 |
Token | river |
Token ID | 7850 |
Feature activation | -3.02e-03 |
Pos logit contributions | |
ang | +0.055 |
irling | +0.051 |
ee | +0.049 |
irl | +0.048 |
ash | +0.046 |
Neg logit contributions | |
ranger | -0.036 |
hinges | -0.034 |
bleed | -0.033 |
keeper | -0.032 |
bru | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -6.63e-03 |
Pos logit contributions | |
ranger | +0.004 |
bleed | +0.004 |
bru | +0.004 |
hinges | +0.004 |
sob | +0.004 |
Neg logit contributions | |
ang | -0.007 |
irling | -0.007 |
ee | -0.007 |
irl | -0.006 |
ug | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +2.39e-03 |
Pos logit contributions | |
ranger | +0.008 |
bleed | +0.008 |
bru | +0.008 |
hinges | +0.008 |
sob | +0.008 |
Neg logit contributions | |
ang | -0.016 |
irling | -0.015 |
ee | -0.015 |
irl | -0.014 |
ug | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -1.43e-03 |
Pos logit contributions | |
ang | +0.006 |
irling | +0.005 |
ee | +0.005 |
irl | +0.005 |
ug | +0.005 |
Neg logit contributions | |
bleed | -0.003 |
ranger | -0.003 |
bru | -0.003 |
hinges | -0.003 |
sob | -0.003 |
Token | When |
Token ID | 2215 |
Feature activation | -9.93e-04 |
Pos logit contributions | |
ranger | +0.002 |
bleed | +0.002 |
hinges | +0.002 |
sob | +0.002 |
bru | +0.002 |
Neg logit contributions | |
ang | -0.003 |
irling | -0.003 |
ee | -0.003 |
irl | -0.002 |
ash | -0.002 |
Token | Sil |
Token ID | 4243 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
ranger | +0.001 |
bleed | +0.001 |
hinges | +0.001 |
bru | +0.001 |
sob | +0.001 |
Neg logit contributions | |
ang | -0.002 |
irling | -0.002 |
ee | -0.002 |
irl | -0.002 |
ash | -0.002 |
Token | via |
Token ID | 8869 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
ang | +0.232 |
irling | +0.216 |
irl | +0.202 |
ading | +0.197 |
ee | +0.193 |
Neg logit contributions | |
aid | -0.195 |
hinges | -0.187 |
ranger | -0.186 |
ache | -0.185 |
sob | -0.183 |
Token | arrived |
Token ID | 5284 |
Feature activation | +4.18e-03 |
Pos logit contributions | |
ang | +0.057 |
irling | +0.052 |
ee | +0.049 |
irl | +0.049 |
ug | +0.047 |
Neg logit contributions | |
ranger | -0.038 |
bleed | -0.037 |
bru | -0.036 |
sob | -0.036 |
hinges | -0.036 |
Token | at |
Token ID | 379 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
ang | +0.006 |
irling | +0.005 |
ee | +0.005 |
irl | +0.005 |
ug | +0.004 |
Neg logit contributions | |
ranger | -0.010 |
bleed | -0.009 |
bru | -0.009 |
hinges | -0.009 |
sob | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
ang | +0.040 |
irling | +0.035 |
ee | +0.033 |
irl | +0.032 |
ug | +0.031 |
Neg logit contributions | |
ranger | -0.051 |
bleed | -0.051 |
bru | -0.050 |
hinges | -0.050 |
sob | -0.050 |
Token | river |
Token ID | 7850 |
Feature activation | +7.01e-03 |
Pos logit contributions | |
ang | +0.027 |
irling | +0.025 |
ee | +0.023 |
irl | +0.023 |
ug | +0.022 |
Neg logit contributions | |
ranger | -0.020 |
department | -0.018 |
hinges | -0.018 |
bleed | -0.018 |
bru | -0.018 |
Token | butterfly |
Token ID | 35113 |
Feature activation | +2.92e-03 |
Pos logit contributions | |
ang | +0.001 |
irling | +0.001 |
ee | +0.001 |
irl | +0.001 |
ug | +0.001 |
Neg logit contributions | |
bleed | -0.001 |
ranger | -0.001 |
bru | -0.001 |
hinges | -0.001 |
sob | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +6.27e-03 |
Pos logit contributions | |
ang | +0.006 |
irling | +0.005 |
ee | +0.005 |
irl | +0.005 |
ug | +0.005 |
Neg logit contributions | |
ranger | -0.005 |
bleed | -0.004 |
sob | -0.004 |
bru | -0.004 |
bud | -0.004 |
Token | tw |
Token ID | 665 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
ang | +0.013 |
irling | +0.011 |
ee | +0.011 |
ug | +0.011 |
ash | +0.010 |
Neg logit contributions | |
bleed | -0.011 |
sob | -0.010 |
ranger | -0.010 |
bru | -0.010 |
hinges | -0.010 |
Token | irling |
Token ID | 24297 |
Feature activation | +5.61e-02 |
Pos logit contributions | |
ang | +0.066 |
ash | +0.064 |
sorts | +0.060 |
icated | +0.059 |
angles | +0.058 |
Neg logit contributions | |
onna | -0.138 |
bleed | -0.136 |
bud | -0.134 |
aid | -0.130 |
sob | -0.130 |
Token | and |
Token ID | 290 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
ang | +0.094 |
irling | +0.082 |
ee | +0.077 |
irl | +0.076 |
ug | +0.073 |
Neg logit contributions | |
ranger | -0.113 |
bleed | -0.110 |
sob | -0.107 |
bru | -0.107 |
hinges | -0.106 |
Token | spir |
Token ID | 9158 |
Feature activation | +1.28e-01 |
Pos logit contributions | |
ang | +0.037 |
ee | +0.032 |
ug | +0.032 |
ash | +0.031 |
irl | +0.030 |
Neg logit contributions | |
ranger | -0.036 |
sob | -0.036 |
bleed | -0.036 |
bru | -0.034 |
bud | -0.034 |
Token | aling |
Token ID | 4272 |
Feature activation | +3.11e-02 |
Pos logit contributions | |
irling | +0.261 |
ang | +0.255 |
irl | +0.241 |
ading | +0.215 |
icing | +0.212 |
Neg logit contributions | |
bleed | -0.186 |
department | -0.171 |
ranger | -0.169 |
recovered | -0.169 |
bud | -0.168 |
Token | in |
Token ID | 287 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
ang | +0.059 |
irling | +0.053 |
ee | +0.051 |
irl | +0.049 |
ug | +0.048 |
Neg logit contributions | |
ranger | -0.055 |
bleed | -0.055 |
bru | -0.053 |
hinges | -0.053 |
sob | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
ang | +0.033 |
irling | +0.029 |
ee | +0.027 |
irl | +0.027 |
ug | +0.026 |
Neg logit contributions | |
ranger | -0.040 |
bleed | -0.039 |
bru | -0.039 |
hinges | -0.039 |
sob | -0.038 |
Token | air |
Token ID | 1633 |
Feature activation | -6.42e-03 |
Pos logit contributions | |
ang | +0.054 |
irling | +0.048 |
irl | +0.045 |
ee | +0.044 |
ug | +0.044 |
Neg logit contributions | |
ranger | -0.051 |
bru | -0.050 |
hinges | -0.050 |
bleed | -0.050 |
sob | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
ranger | +0.008 |
bleed | +0.008 |
bru | +0.008 |
hinges | +0.008 |
sob | +0.008 |
Neg logit contributions | |
ang | -0.015 |
irling | -0.014 |
ee | -0.014 |
irl | -0.013 |
ug | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
ang | +0.074 |
irling | +0.067 |
ee | +0.062 |
irl | +0.062 |
ug | +0.060 |
Neg logit contributions | |
ranger | -0.067 |
bru | -0.063 |
bleed | -0.062 |
sob | -0.062 |
hinges | -0.062 |
Token | chest |
Token ID | 7721 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ang | +0.027 |
ee | +0.024 |
irling | +0.024 |
irl | +0.023 |
ading | +0.022 |
Neg logit contributions | |
bleed | -0.019 |
bru | -0.018 |
ranger | -0.017 |
department | -0.017 |
sob | -0.017 |
Token | were |
Token ID | 547 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
ang | +0.031 |
irling | +0.028 |
ee | +0.027 |
irl | +0.026 |
ug | +0.025 |
Neg logit contributions | |
bleed | -0.027 |
bru | -0.026 |
ranger | -0.025 |
hinges | -0.025 |
sob | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
ang | +0.065 |
irling | +0.058 |
ee | +0.056 |
irl | +0.054 |
ug | +0.052 |
Neg logit contributions | |
ranger | -0.066 |
bleed | -0.065 |
hinges | -0.064 |
bru | -0.064 |
sob | -0.064 |
Token | most |
Token ID | 749 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
ang | +0.029 |
irling | +0.025 |
ee | +0.025 |
irl | +0.023 |
ug | +0.022 |
Neg logit contributions | |
bleed | -0.037 |
ranger | -0.036 |
bru | -0.035 |
hinges | -0.035 |
sob | -0.035 |
Token | wond |
Token ID | 3135 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
ang | +0.054 |
irling | +0.047 |
ee | +0.046 |
irl | +0.045 |
ug | +0.043 |
Neg logit contributions | |
ranger | -0.068 |
hinges | -0.065 |
bleed | -0.065 |
sob | -0.065 |
bru | -0.065 |
Token | eful |
Token ID | 13839 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
ang | +0.117 |
ee | +0.097 |
irling | +0.097 |
irl | +0.088 |
ug | +0.086 |
Neg logit contributions | |
ranger | -0.296 |
keeping | -0.294 |
bru | -0.288 |
recovered | -0.282 |
keeper | -0.278 |
Token | toys |
Token ID | 14958 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
ang | +0.045 |
irling | +0.041 |
ee | +0.039 |
irl | +0.038 |
ug | +0.037 |
Neg logit contributions | |
ranger | -0.032 |
hinges | -0.030 |
bleed | -0.030 |
bru | -0.030 |
sob | -0.030 |
Token | the |
Token ID | 262 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
ang | +0.040 |
irling | +0.036 |
ee | +0.035 |
irl | +0.033 |
ug | +0.033 |
Neg logit contributions | |
bleed | -0.036 |
bru | -0.035 |
ranger | -0.035 |
hinges | -0.035 |
sob | -0.034 |
Token | little |
Token ID | 1310 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
ang | +0.027 |
irling | +0.024 |
ee | +0.024 |
irl | +0.023 |
ug | +0.022 |
Neg logit contributions | |
bleed | -0.027 |
ranger | -0.027 |
bru | -0.026 |
hinges | -0.026 |
sob | -0.026 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.25e-03 |
Pos logit contributions | |
ang | +0.043 |
irling | +0.037 |
ee | +0.037 |
irl | +0.035 |
ug | +0.035 |
Neg logit contributions | |
bleed | -0.036 |
ranger | -0.036 |
bru | -0.035 |
hinges | -0.035 |
sob | -0.035 |
Token | asleep |
Token ID | 16039 |
Feature activation | -9.32e-04 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ug | +0.003 |
oming | +0.003 |
ee | +0.003 |
Neg logit contributions | |
Something | -0.003 |
ranger | -0.003 |
aid | -0.003 |
mercy | -0.003 |
bru | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +2.82e-03 |
Pos logit contributions | |
bleed | +0.002 |
department | +0.002 |
ranger | +0.002 |
hinges | +0.002 |
bru | +0.002 |
Neg logit contributions | |
ang | -0.002 |
irling | -0.002 |
ee | -0.001 |
irl | -0.001 |
ug | -0.001 |
Token | dreaming |
Token ID | 28368 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
ang | +0.006 |
irling | +0.005 |
ee | +0.005 |
irl | +0.005 |
ug | +0.005 |
Neg logit contributions | |
ranger | -0.005 |
bleed | -0.004 |
bru | -0.004 |
hinges | -0.004 |
sob | -0.004 |
Token | of |
Token ID | 286 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
ang | +0.014 |
irling | +0.012 |
ee | +0.011 |
ug | +0.010 |
irl | +0.010 |
Neg logit contributions | |
bleed | -0.033 |
ranger | -0.033 |
hinges | -0.033 |
bru | -0.032 |
sob | -0.032 |
Token | icy |
Token ID | 30284 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ang | +0.050 |
irling | +0.044 |
ee | +0.042 |
irl | +0.041 |
ug | +0.039 |
Neg logit contributions | |
ranger | -0.056 |
bleed | -0.056 |
bru | -0.055 |
hinges | -0.055 |
sob | -0.054 |
Token | pian |
Token ID | 43923 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
ang | +0.066 |
irling | +0.058 |
ee | +0.056 |
irl | +0.054 |
ug | +0.052 |
Neg logit contributions | |
ranger | -0.070 |
bleed | -0.069 |
bru | -0.068 |
hinges | -0.067 |
sob | -0.067 |
Token | os |
Token ID | 418 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
ang | +0.239 |
irling | +0.223 |
irl | +0.221 |
ading | +0.205 |
ee | +0.203 |
Neg logit contributions | |
onna | -0.184 |
department | -0.182 |
bleed | -0.179 |
recovered | -0.175 |
ranger | -0.174 |
Token | and |
Token ID | 290 |
Feature activation | +4.13e-02 |
Pos logit contributions | |
ang | +0.091 |
irling | +0.083 |
ee | +0.078 |
irl | +0.078 |
ading | +0.073 |
Neg logit contributions | |
bleed | -0.074 |
ranger | -0.074 |
department | -0.070 |
sob | -0.070 |
hinges | -0.070 |
Token | big |
Token ID | 1263 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
ang | +0.078 |
irling | +0.070 |
ee | +0.067 |
irl | +0.065 |
ug | +0.062 |
Neg logit contributions | |
ranger | -0.073 |
bleed | -0.073 |
bru | -0.071 |
hinges | -0.071 |
sob | -0.070 |
Token | piles |
Token ID | 30935 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
ang | +0.024 |
irling | +0.021 |
ee | +0.021 |
irl | +0.020 |
ug | +0.020 |
Neg logit contributions | |
bleed | -0.019 |
ranger | -0.019 |
hinges | -0.018 |
bru | -0.018 |
sob | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
ang | +0.069 |
irling | +0.061 |
ee | +0.060 |
ug | +0.057 |
irl | +0.057 |
Neg logit contributions | |
bleed | -0.057 |
bru | -0.057 |
ranger | -0.056 |
sob | -0.055 |
hinges | -0.055 |
Token | with |
Token ID | 351 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
ang | +0.007 |
irling | +0.006 |
ee | +0.006 |
irl | +0.006 |
ug | +0.005 |
Neg logit contributions | |
ranger | -0.005 |
bleed | -0.005 |
bru | -0.005 |
hinges | -0.005 |
department | -0.005 |
Token | your |
Token ID | 534 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
ang | +0.090 |
irling | +0.077 |
ee | +0.073 |
ash | +0.073 |
ug | +0.071 |
Neg logit contributions | |
ranger | -0.082 |
bud | -0.081 |
mercy | -0.080 |
sob | -0.080 |
hinges | -0.080 |
Token | new |
Token ID | 649 |
Feature activation | -9.03e-03 |
Pos logit contributions | |
ranger | +0.050 |
bleed | +0.050 |
bru | +0.049 |
hinges | +0.048 |
sob | +0.048 |
Neg logit contributions | |
ang | -0.053 |
irling | -0.047 |
ee | -0.046 |
irl | -0.044 |
ug | -0.042 |
Token | coat |
Token ID | 13209 |
Feature activation | -1.76e-03 |
Pos logit contributions | |
bleed | +0.013 |
ranger | +0.013 |
bru | +0.012 |
hinges | +0.012 |
sob | +0.012 |
Neg logit contributions | |
ang | -0.021 |
irling | -0.019 |
ee | -0.018 |
irl | -0.018 |
ug | -0.017 |
Token | ." |
Token ID | 526 |
Feature activation | -8.99e-03 |
Pos logit contributions | |
ranger | +0.003 |
bleed | +0.003 |
bru | +0.003 |
hinges | +0.003 |
sob | +0.003 |
Neg logit contributions | |
ang | -0.004 |
irling | -0.003 |
ee | -0.003 |
irl | -0.003 |
ug | -0.003 |
Token | Sil |
Token ID | 4243 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
ranger | +0.013 |
bleed | +0.012 |
bru | +0.012 |
hinges | +0.012 |
sob | +0.012 |
Neg logit contributions | |
ang | -0.020 |
irling | -0.019 |
ee | -0.018 |
irl | -0.018 |
ug | -0.017 |
Token | via |
Token ID | 8869 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
irling | +0.208 |
ang | +0.207 |
irl | +0.202 |
ading | +0.186 |
ee | +0.183 |
Neg logit contributions | |
sob | -0.198 |
ranger | -0.194 |
aid | -0.194 |
keeper | -0.194 |
hinges | -0.191 |
Token | smiled |
Token ID | 13541 |
Feature activation | -9.21e-04 |
Pos logit contributions | |
ang | +0.067 |
irling | +0.061 |
ee | +0.059 |
irl | +0.058 |
ug | +0.057 |
Neg logit contributions | |
ranger | -0.038 |
sob | -0.037 |
bleed | -0.036 |
bru | -0.036 |
hinges | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
bleed | +0.002 |
bru | +0.002 |
ranger | +0.002 |
hinges | +0.002 |
injuries | +0.002 |
Neg logit contributions | |
ang | -0.001 |
irling | -0.001 |
ee | -0.001 |
ug | -0.001 |
aw | -0.001 |
Token | said |
Token ID | 531 |
Feature activation | -4.08e-02 |
Pos logit contributions | |
ang | +0.025 |
irling | +0.023 |
ash | +0.022 |
ug | +0.022 |
ee | +0.022 |
Neg logit contributions | |
sob | -0.012 |
ranger | -0.012 |
bleed | -0.011 |
bru | -0.011 |
recovered | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
bleed | +0.068 |
ranger | +0.068 |
bru | +0.067 |
hinges | +0.066 |
sob | +0.065 |
Neg logit contributions | |
ang | -0.081 |
irling | -0.074 |
ee | -0.071 |
irl | -0.069 |
ug | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
ang | +0.114 |
irling | +0.093 |
ug | +0.090 |
ee | +0.090 |
ading | +0.088 |
Neg logit contributions | |
sob | -0.077 |
hinges | -0.077 |
bleed | -0.077 |
keeping | -0.075 |
bru | -0.074 |
Token | you |
Token ID | 345 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
ang | +0.092 |
ee | +0.078 |
irling | +0.078 |
ash | +0.077 |
ug | +0.075 |
Neg logit contributions | |
bleed | -0.069 |
bru | -0.065 |
sob | -0.063 |
ranger | -0.063 |
hinges | -0.062 |
Token | love |
Token ID | 1842 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
ang | +0.037 |
irling | +0.034 |
ee | +0.032 |
irl | +0.031 |
ug | +0.031 |
Neg logit contributions | |
bleed | -0.020 |
bru | -0.019 |
ranger | -0.018 |
sob | -0.018 |
hinges | -0.018 |
Token | it |
Token ID | 340 |
Feature activation | +3.34e-03 |
Pos logit contributions | |
ang | +0.035 |
irling | +0.030 |
ee | +0.029 |
ading | +0.028 |
ug | +0.028 |
Neg logit contributions | |
bru | -0.033 |
bleed | -0.032 |
ranger | -0.031 |
hinges | -0.031 |
sob | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
ang | +0.007 |
irling | +0.006 |
ee | +0.006 |
irl | +0.006 |
ug | +0.006 |
Neg logit contributions | |
bleed | -0.005 |
ranger | -0.005 |
hinges | -0.005 |
sob | -0.005 |
bru | -0.005 |
Token | Sil |
Token ID | 4243 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
ang | +0.042 |
irling | +0.038 |
ee | +0.036 |
irl | +0.035 |
ug | +0.034 |
Neg logit contributions | |
ranger | -0.029 |
bru | -0.028 |
sob | -0.028 |
bleed | -0.028 |
hinges | -0.027 |
Token | via |
Token ID | 8869 |
Feature activation | +3.44e-02 |
Pos logit contributions | |
ang | +0.217 |
irling | +0.196 |
irl | +0.185 |
ee | +0.178 |
ading | +0.174 |
Neg logit contributions | |
onna | -0.208 |
aid | -0.207 |
ache | -0.203 |
sob | -0.193 |
department | -0.190 |
Token | ! |
Token ID | 0 |
Feature activation | +3.32e-02 |
Pos logit contributions | |
ang | +0.081 |
irling | +0.074 |
ee | +0.072 |
irl | +0.071 |
ading | +0.069 |
Neg logit contributions | |
ranger | -0.043 |
bleed | -0.043 |
bru | -0.043 |
sob | -0.041 |
hinges | -0.041 |
Token | Come |
Token ID | 7911 |
Feature activation | +2.22e-03 |
Pos logit contributions | |
ang | +0.071 |
irling | +0.070 |
irl | +0.064 |
ee | +0.063 |
ading | +0.062 |
Neg logit contributions | |
ranger | -0.048 |
Something | -0.048 |
bru | -0.047 |
bleed | -0.046 |
sob | -0.045 |
Token | now |
Token ID | 783 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
ang | +0.004 |
irling | +0.004 |
ee | +0.004 |
irl | +0.004 |
ug | +0.004 |
Neg logit contributions | |
ranger | -0.004 |
bru | -0.003 |
bleed | -0.003 |
department | -0.003 |
sob | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +9.01e-03 |
Pos logit contributions | |
ranger | +0.030 |
bleed | +0.030 |
bru | +0.030 |
sob | +0.029 |
hinges | +0.029 |
Neg logit contributions | |
ang | -0.032 |
irling | -0.028 |
ee | -0.027 |
irl | -0.026 |
ug | -0.026 |
Token | green |
Token ID | 4077 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
ang | +0.101 |
irling | +0.089 |
ee | +0.085 |
irl | +0.083 |
ug | +0.080 |
Neg logit contributions | |
ranger | -0.105 |
bleed | -0.104 |
bru | -0.103 |
hinges | -0.102 |
sob | -0.101 |
Token | and |
Token ID | 290 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
ang | +0.051 |
irling | +0.045 |
ee | +0.045 |
irl | +0.042 |
ug | +0.042 |
Neg logit contributions | |
ranger | -0.053 |
bleed | -0.052 |
bru | -0.050 |
sob | -0.050 |
department | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
ang | +0.064 |
irling | +0.057 |
ee | +0.055 |
irl | +0.053 |
ug | +0.052 |
Neg logit contributions | |
ranger | -0.059 |
bleed | -0.058 |
hinges | -0.057 |
sob | -0.056 |
bru | -0.056 |
Token | long |
Token ID | 890 |
Feature activation | +4.86e-03 |
Pos logit contributions | |
bleed | +0.023 |
ranger | +0.023 |
hinges | +0.022 |
bru | +0.022 |
sob | +0.022 |
Neg logit contributions | |
ang | -0.022 |
irling | -0.019 |
ee | -0.019 |
irl | -0.018 |
ug | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +5.79e-03 |
Pos logit contributions | |
ang | +0.010 |
irling | +0.009 |
ee | +0.009 |
irl | +0.008 |
ug | +0.008 |
Neg logit contributions | |
ranger | -0.008 |
bleed | -0.008 |
sob | -0.008 |
bru | -0.008 |
hinges | -0.008 |
Token | spir |
Token ID | 9158 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
ang | +0.010 |
irling | +0.009 |
ee | +0.009 |
irl | +0.009 |
ug | +0.008 |
Neg logit contributions | |
ranger | -0.011 |
bleed | -0.011 |
bru | -0.011 |
hinges | -0.011 |
sob | -0.010 |
Token | alling |
Token ID | 9221 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
ang | +0.241 |
irling | +0.237 |
irl | +0.221 |
ee | +0.199 |
ading | +0.194 |
Neg logit contributions | |
bleed | -0.205 |
ranger | -0.195 |
department | -0.194 |
hinges | -0.193 |
bru | -0.192 |
Token | tail |
Token ID | 7894 |
Feature activation | +7.35e-03 |
Pos logit contributions | |
ang | +0.056 |
irling | +0.049 |
ee | +0.048 |
irl | +0.046 |
ug | +0.046 |
Neg logit contributions | |
bleed | -0.050 |
bru | -0.049 |
ranger | -0.049 |
sob | -0.049 |
hinges | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +1.93e-03 |
Pos logit contributions | |
ang | +0.015 |
irling | +0.014 |
ee | +0.014 |
ug | +0.013 |
irl | +0.013 |
Neg logit contributions | |
bru | -0.011 |
bleed | -0.011 |
ranger | -0.011 |
hinges | -0.011 |
sob | -0.011 |
Token | The |
Token ID | 383 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
ang | +0.004 |
irling | +0.004 |
ee | +0.004 |
irl | +0.004 |
ug | +0.004 |
Neg logit contributions | |
bleed | -0.003 |
ranger | -0.003 |
bru | -0.003 |
hinges | -0.003 |
sob | -0.003 |
Token | princess |
Token ID | 21752 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
ang | +0.027 |
irling | +0.024 |
ee | +0.024 |
irl | +0.023 |
ug | +0.022 |
Neg logit contributions | |
bleed | -0.025 |
ranger | -0.024 |
bru | -0.024 |
hinges | -0.024 |
sob | -0.023 |
Token | there |
Token ID | 612 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
ang | +0.101 |
irling | +0.094 |
ee | +0.090 |
irl | +0.088 |
ug | +0.085 |
Neg logit contributions | |
ranger | -0.058 |
sob | -0.055 |
bru | -0.054 |
bleed | -0.053 |
hinges | -0.052 |
Token | was |
Token ID | 373 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
ang | +0.097 |
irling | +0.089 |
ee | +0.085 |
irl | +0.084 |
ug | +0.081 |
Neg logit contributions | |
ranger | -0.062 |
bleed | -0.061 |
bru | -0.060 |
sob | -0.059 |
hinges | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
ang | +0.076 |
irling | +0.069 |
ash | +0.064 |
angles | +0.064 |
ee | +0.064 |
Neg logit contributions | |
bleed | -0.034 |
sob | -0.034 |
bru | -0.032 |
ranger | -0.031 |
hinges | -0.030 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
ang | +0.022 |
irling | +0.022 |
ash | +0.021 |
ading | +0.021 |
angles | +0.020 |
Neg logit contributions | |
ranger | -0.015 |
bru | -0.013 |
sob | -0.013 |
aid | -0.013 |
cough | -0.012 |
Token | named |
Token ID | 3706 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
ang | +0.023 |
irling | +0.022 |
irl | +0.021 |
ading | +0.021 |
ee | +0.020 |
Neg logit contributions | |
ranger | -0.022 |
sob | -0.022 |
bru | -0.021 |
aid | -0.021 |
hinges | -0.020 |
Token | Sil |
Token ID | 4243 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
ang | +0.102 |
irling | +0.094 |
ee | +0.090 |
irl | +0.087 |
ug | +0.087 |
Neg logit contributions | |
ranger | -0.049 |
sob | -0.049 |
hinges | -0.048 |
bleed | -0.046 |
bru | -0.045 |
Token | via |
Token ID | 8869 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
ang | +0.210 |
irling | +0.192 |
irl | +0.180 |
ee | +0.174 |
ading | +0.174 |
Neg logit contributions | |
aid | -0.216 |
onna | -0.210 |
ranger | -0.207 |
sob | -0.199 |
hinges | -0.199 |
Token | . |
Token ID | 13 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
ang | +0.086 |
irling | +0.079 |
ee | +0.077 |
irl | +0.075 |
ug | +0.073 |
Neg logit contributions | |
ranger | -0.041 |
bleed | -0.040 |
bru | -0.039 |
hinges | -0.039 |
sob | -0.039 |
Token | She |
Token ID | 1375 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
ang | +0.042 |
irling | +0.038 |
ee | +0.037 |
irl | +0.036 |
ug | +0.036 |
Neg logit contributions | |
bleed | -0.023 |
ranger | -0.022 |
bru | -0.022 |
sob | -0.022 |
hinges | -0.021 |
Token | was |
Token ID | 373 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
ang | +0.040 |
irling | +0.034 |
ee | +0.032 |
ash | +0.032 |
ug | +0.032 |
Neg logit contributions | |
ranger | -0.042 |
bleed | -0.042 |
sob | -0.042 |
bru | -0.041 |
hinges | -0.040 |
Token | 3 |
Token ID | 513 |
Feature activation | +4.35e-02 |
Pos logit contributions | |
ang | +0.082 |
irling | +0.079 |
ash | +0.077 |
angles | +0.077 |
ee | +0.075 |
Neg logit contributions | |
sob | -0.038 |
bleed | -0.037 |
cough | -0.034 |
ranger | -0.034 |
bru | -0.034 |
Token | same |
Token ID | 976 |
Feature activation | -5.42e-03 |
Pos logit contributions | |
ang | +0.002 |
irling | +0.002 |
ee | +0.002 |
irl | +0.002 |
ug | +0.002 |
Neg logit contributions | |
ranger | -0.002 |
bleed | -0.002 |
bru | -0.002 |
department | -0.002 |
keeper | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -6.29e-03 |
Pos logit contributions | |
ranger | +0.009 |
bleed | +0.009 |
bru | +0.009 |
aid | +0.008 |
hinges | +0.008 |
Neg logit contributions | |
ang | -0.011 |
irling | -0.010 |
ee | -0.009 |
irl | -0.009 |
ash | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-04 |
Pos logit contributions | |
ranger | +0.008 |
bleed | +0.008 |
bru | +0.008 |
hinges | +0.008 |
sob | +0.008 |
Neg logit contributions | |
ang | -0.015 |
irling | -0.014 |
ee | -0.014 |
irl | -0.013 |
ug | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -1.03e-03 |
Pos logit contributions | |
bleed | +0.000 |
ranger | +0.000 |
bru | +0.000 |
hinges | +0.000 |
sob | +0.000 |
Neg logit contributions | |
ang | -0.001 |
irling | -0.001 |
ee | -0.001 |
irl | -0.000 |
ug | -0.000 |
Token | As |
Token ID | 1722 |
Feature activation | +9.94e-03 |
Pos logit contributions | |
ranger | +0.002 |
bleed | +0.002 |
hinges | +0.002 |
sob | +0.002 |
bud | +0.002 |
Neg logit contributions | |
ang | -0.002 |
irling | -0.002 |
ee | -0.002 |
ash | -0.002 |
irl | -0.002 |
Token | Sil |
Token ID | 4243 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
ang | +0.022 |
irling | +0.021 |
ash | +0.020 |
ee | +0.020 |
irl | +0.020 |
Neg logit contributions | |
ranger | -0.014 |
bud | -0.013 |
hinges | -0.013 |
keeper | -0.013 |
bleed | -0.012 |
Token | via |
Token ID | 8869 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
irling | +0.201 |
ang | +0.199 |
irl | +0.192 |
ading | +0.179 |
ee | +0.173 |
Neg logit contributions | |
aid | -0.192 |
ache | -0.191 |
ranger | -0.190 |
hinges | -0.188 |
sob | -0.188 |
Token | was |
Token ID | 373 |
Feature activation | +3.66e-03 |
Pos logit contributions | |
ang | +0.064 |
irling | +0.062 |
irl | +0.058 |
ash | +0.055 |
ee | +0.055 |
Neg logit contributions | |
ranger | -0.039 |
recovered | -0.038 |
sob | -0.035 |
aid | -0.035 |
bru | -0.035 |
Token | about |
Token ID | 546 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
ang | +0.006 |
ee | +0.006 |
angles | +0.005 |
ash | +0.005 |
irling | +0.005 |
Neg logit contributions | |
sob | -0.007 |
bleed | -0.007 |
ranger | -0.007 |
bru | -0.007 |
aid | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
bleed | +0.045 |
ache | +0.040 |
hinges | +0.039 |
sob | +0.038 |
ranger | +0.038 |
Neg logit contributions | |
ang | -0.059 |
irling | -0.056 |
irl | -0.053 |
ee | -0.052 |
sorts | -0.051 |
Token | jump |
Token ID | 4391 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
ranger | +0.052 |
bleed | +0.052 |
bru | +0.051 |
hinges | +0.050 |
sob | +0.049 |
Neg logit contributions | |
ang | -0.079 |
irling | -0.072 |
ee | -0.069 |
irl | -0.067 |
ug | -0.066 |
Token | grass |
Token ID | 8701 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
ang | +0.095 |
irling | +0.084 |
ee | +0.084 |
irl | +0.079 |
ug | +0.078 |
Neg logit contributions | |
ranger | -0.080 |
bru | -0.080 |
bleed | -0.080 |
hinges | -0.078 |
sob | -0.077 |
Token | and |
Token ID | 290 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
ang | +0.057 |
irling | +0.051 |
ee | +0.050 |
irl | +0.048 |
ug | +0.047 |
Neg logit contributions | |
bru | -0.042 |
bleed | -0.042 |
ranger | -0.041 |
hinges | -0.041 |
sob | -0.040 |
Token | rye |
Token ID | 47553 |
Feature activation | +7.25e-02 |
Pos logit contributions | |
ang | +0.076 |
irling | +0.067 |
ee | +0.065 |
irl | +0.063 |
ug | +0.061 |
Neg logit contributions | |
ranger | -0.066 |
bleed | -0.066 |
bru | -0.065 |
hinges | -0.064 |
sob | -0.064 |
Token | grass |
Token ID | 8701 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
ang | +0.148 |
irling | +0.130 |
ee | +0.127 |
irl | +0.122 |
ug | +0.120 |
Neg logit contributions | |
bleed | -0.119 |
bru | -0.119 |
hinges | -0.118 |
ranger | -0.118 |
sob | -0.115 |
Token | and |
Token ID | 290 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
ang | +0.075 |
irling | +0.067 |
ee | +0.066 |
irl | +0.063 |
ug | +0.062 |
Neg logit contributions | |
bru | -0.043 |
bleed | -0.043 |
ranger | -0.043 |
hinges | -0.042 |
sob | -0.042 |
Token | al |
Token ID | 435 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
ang | +0.080 |
irling | +0.072 |
ee | +0.069 |
irl | +0.067 |
ug | +0.065 |
Neg logit contributions | |
ranger | -0.073 |
bleed | -0.072 |
bru | -0.071 |
hinges | -0.070 |
sob | -0.070 |
Token | f |
Token ID | 69 |
Feature activation | +5.73e-02 |
Pos logit contributions | |
ang | +0.179 |
irling | +0.153 |
ee | +0.148 |
irl | +0.144 |
ading | +0.135 |
Neg logit contributions | |
onna | -0.245 |
recovered | -0.238 |
bleed | -0.238 |
ranger | -0.234 |
ache | -0.234 |
Token | alf |
Token ID | 1604 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
ang | +0.096 |
irling | +0.085 |
irl | +0.080 |
ee | +0.078 |
ug | +0.072 |
Neg logit contributions | |
bleed | -0.110 |
ranger | -0.109 |
bru | -0.107 |
hinges | -0.107 |
department | -0.106 |
Token | a |
Token ID | 64 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
ang | +0.058 |
irling | +0.052 |
ee | +0.050 |
irl | +0.049 |
ug | +0.047 |
Neg logit contributions | |
ranger | -0.046 |
bleed | -0.045 |
bru | -0.044 |
sob | -0.044 |
hinges | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
ang | +0.043 |
irling | +0.039 |
ee | +0.038 |
irl | +0.037 |
ug | +0.036 |
Neg logit contributions | |
ranger | -0.028 |
bleed | -0.028 |
bru | -0.027 |
hinges | -0.027 |
sob | -0.027 |
Token | cl |
Token ID | 537 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
ang | +0.054 |
irling | +0.049 |
ee | +0.047 |
irl | +0.046 |
ug | +0.044 |
Neg logit contributions | |
ranger | -0.047 |
bleed | -0.047 |
bru | -0.046 |
hinges | -0.046 |
sob | -0.045 |
Token | forth |
Token ID | 6071 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
ang | +0.047 |
irling | +0.041 |
ee | +0.041 |
ug | +0.040 |
ash | +0.039 |
Neg logit contributions | |
sob | -0.040 |
bleed | -0.039 |
ranger | -0.039 |
bru | -0.038 |
recovered | -0.036 |
Token | in |
Token ID | 287 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
ang | +0.025 |
irling | +0.023 |
ee | +0.022 |
irl | +0.021 |
ug | +0.021 |
Neg logit contributions | |
ranger | -0.023 |
bleed | -0.023 |
bru | -0.022 |
hinges | -0.022 |
sob | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
ang | +0.033 |
irling | +0.029 |
ee | +0.028 |
irl | +0.027 |
ug | +0.025 |
Neg logit contributions | |
ranger | -0.047 |
bleed | -0.046 |
hinges | -0.045 |
bru | -0.045 |
sob | -0.045 |
Token | sky |
Token ID | 6766 |
Feature activation | -1.50e-03 |
Pos logit contributions | |
ang | +0.062 |
irling | +0.055 |
ee | +0.051 |
irl | +0.051 |
ug | +0.050 |
Neg logit contributions | |
bru | -0.058 |
bleed | -0.058 |
ranger | -0.058 |
sob | -0.057 |
hinges | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +8.05e-03 |
Pos logit contributions | |
ranger | +0.002 |
bleed | +0.002 |
bru | +0.002 |
hinges | +0.002 |
sob | +0.002 |
Neg logit contributions | |
ang | -0.004 |
irling | -0.003 |
ee | -0.003 |
irl | -0.003 |
ug | -0.003 |
Token | spir |
Token ID | 9158 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ang | +0.017 |
ee | +0.015 |
irling | +0.014 |
ug | +0.014 |
ash | +0.014 |
Neg logit contributions | |
bleed | -0.013 |
ranger | -0.013 |
sob | -0.013 |
bru | -0.013 |
hinges | -0.012 |
Token | alling |
Token ID | 9221 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
irling | +0.212 |
ang | +0.201 |
irl | +0.192 |
icing | +0.165 |
ading | +0.163 |
Neg logit contributions | |
bleed | -0.231 |
department | -0.216 |
recovered | -0.214 |
bru | -0.214 |
ranger | -0.211 |
Token | higher |
Token ID | 2440 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
ang | +0.051 |
irling | +0.045 |
ee | +0.043 |
irl | +0.041 |
ug | +0.041 |
Neg logit contributions | |
bleed | -0.054 |
bru | -0.053 |
ranger | -0.052 |
hinges | -0.052 |
sob | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
ang | +0.068 |
irling | +0.062 |
ee | +0.060 |
irl | +0.059 |
ug | +0.057 |
Neg logit contributions | |
ranger | -0.042 |
bleed | -0.041 |
hinges | -0.040 |
bru | -0.040 |
sob | -0.039 |
Token | higher |
Token ID | 2440 |
Feature activation | +3.25e-02 |
Pos logit contributions | |
ang | +0.079 |
ee | +0.071 |
ug | +0.070 |
irling | +0.070 |
irl | +0.067 |
Neg logit contributions | |
ranger | -0.061 |
bleed | -0.057 |
bru | -0.056 |
keeper | -0.055 |
sob | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-03 |
Pos logit contributions | |
ang | +0.079 |
irling | +0.072 |
ee | +0.071 |
irl | +0.069 |
ug | +0.067 |
Neg logit contributions | |
ranger | -0.040 |
bleed | -0.040 |
bru | -0.039 |
hinges | -0.038 |
sob | -0.038 |
Token | looks |
Token ID | 3073 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
ang | +0.071 |
irling | +0.065 |
ug | +0.059 |
ading | +0.059 |
irl | +0.058 |
Neg logit contributions | |
bleed | -0.044 |
sob | -0.044 |
bru | -0.042 |
ranger | -0.042 |
Something | -0.040 |
Token | wonderful |
Token ID | 7932 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
ang | +0.107 |
irling | +0.088 |
ug | +0.088 |
ee | +0.086 |
ites | +0.082 |
Neg logit contributions | |
bleed | -0.073 |
ranger | -0.072 |
bru | -0.072 |
hinges | -0.071 |
sob | -0.070 |
Token | on |
Token ID | 319 |
Feature activation | +3.67e-02 |
Pos logit contributions | |
ang | +0.024 |
irling | +0.021 |
ee | +0.020 |
irl | +0.020 |
ug | +0.019 |
Neg logit contributions | |
ranger | -0.020 |
bleed | -0.019 |
hinges | -0.019 |
bru | -0.019 |
department | -0.019 |
Token | you |
Token ID | 345 |
Feature activation | +8.83e-04 |
Pos logit contributions | |
ang | +0.062 |
irling | +0.053 |
ash | +0.050 |
ee | +0.048 |
irl | +0.048 |
Neg logit contributions | |
Help | -0.065 |
ranger | -0.065 |
bleed | -0.063 |
hinges | -0.063 |
bud | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
ang | +0.002 |
irling | +0.002 |
irl | +0.002 |
ug | +0.002 |
ee | +0.002 |
Neg logit contributions | |
ranger | -0.001 |
bud | -0.001 |
department | -0.001 |
hinges | -0.001 |
sob | -0.001 |
Token | Sil |
Token ID | 4243 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
ang | +0.041 |
irling | +0.039 |
irl | +0.036 |
ee | +0.034 |
ash | +0.034 |
Neg logit contributions | |
ranger | -0.033 |
bud | -0.031 |
department | -0.031 |
keeper | -0.030 |
hinges | -0.029 |
Token | via |
Token ID | 8869 |
Feature activation | +3.56e-02 |
Pos logit contributions | |
ang | +0.202 |
irling | +0.191 |
irl | +0.186 |
ee | +0.173 |
ading | +0.172 |
Neg logit contributions | |
ache | -0.199 |
aid | -0.198 |
onna | -0.193 |
ranger | -0.185 |
keeper | -0.183 |
Token | ! |
Token ID | 0 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
ang | +0.076 |
irling | +0.069 |
irl | +0.066 |
ee | +0.065 |
ading | +0.064 |
Neg logit contributions | |
ranger | -0.052 |
bleed | -0.049 |
bru | -0.049 |
keeper | -0.048 |
sob | -0.048 |
Token | You |
Token ID | 921 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
irling | +0.075 |
ang | +0.075 |
irl | +0.069 |
ading | +0.065 |
ee | +0.064 |
Neg logit contributions | |
Something | -0.059 |
ranger | -0.057 |
department | -0.055 |
Help | -0.055 |
bru | -0.054 |
Token | look |
Token ID | 804 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
ang | +0.101 |
irling | +0.091 |
ug | +0.083 |
irl | +0.082 |
ee | +0.082 |
Neg logit contributions | |
bleed | -0.058 |
sob | -0.056 |
ranger | -0.055 |
bru | -0.055 |
bud | -0.054 |
Token | so |
Token ID | 523 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
ang | +0.092 |
irling | +0.079 |
ug | +0.077 |
irl | +0.075 |
ee | +0.075 |
Neg logit contributions | |
bleed | -0.064 |
ranger | -0.064 |
hinges | -0.063 |
sob | -0.063 |
bru | -0.062 |
Token | She |
Token ID | 1375 |
Feature activation | -7.72e-03 |
Pos logit contributions | |
bleed | +0.010 |
hinges | +0.009 |
bru | +0.009 |
injuries | +0.009 |
recovered | +0.009 |
Neg logit contributions | |
ang | -0.016 |
irling | -0.014 |
ee | -0.014 |
irl | -0.013 |
ug | -0.013 |
Token | told |
Token ID | 1297 |
Feature activation | -6.04e-03 |
Pos logit contributions | |
ranger | +0.013 |
bleed | +0.013 |
bru | +0.012 |
hinges | +0.012 |
sob | +0.012 |
Neg logit contributions | |
ang | -0.016 |
irling | -0.014 |
ee | -0.014 |
irl | -0.013 |
ug | -0.013 |
Token | him |
Token ID | 683 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
ranger | +0.010 |
bleed | +0.010 |
hinges | +0.010 |
bru | +0.010 |
department | +0.010 |
Neg logit contributions | |
ang | -0.011 |
irling | -0.010 |
ee | -0.010 |
irl | -0.010 |
ug | -0.009 |
Token | how |
Token ID | 703 |
Feature activation | -3.90e-03 |
Pos logit contributions | |
bleed | +0.106 |
bru | +0.105 |
hinges | +0.103 |
ranger | +0.101 |
injuries | +0.100 |
Neg logit contributions | |
ang | -0.098 |
irling | -0.090 |
ee | -0.087 |
irl | -0.086 |
ug | -0.081 |
Token | she |
Token ID | 673 |
Feature activation | -1.53e-03 |
Pos logit contributions | |
bleed | +0.005 |
ranger | +0.005 |
hinges | +0.005 |
bru | +0.005 |
sob | +0.005 |
Neg logit contributions | |
ang | -0.009 |
irling | -0.008 |
ee | -0.008 |
irl | -0.008 |
ug | -0.008 |
Token | se |
Token ID | 384 |
Feature activation | +1.22e-01 |
Pos logit contributions | |
ranger | +0.002 |
bleed | +0.002 |
sob | +0.002 |
bru | +0.002 |
hinges | +0.002 |
Neg logit contributions | |
ang | -0.003 |
irling | -0.003 |
ee | -0.003 |
irl | -0.003 |
ug | -0.003 |
Token | wed |
Token ID | 19103 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
ang | +0.253 |
irling | +0.227 |
ee | +0.207 |
ading | +0.205 |
ug | +0.199 |
Neg logit contributions | |
bleed | -0.190 |
bru | -0.178 |
sob | -0.176 |
ranger | -0.174 |
recovered | -0.172 |
Token | , |
Token ID | 11 |
Feature activation | -4.64e-03 |
Pos logit contributions | |
ang | +0.079 |
irling | +0.071 |
ee | +0.065 |
irl | +0.064 |
ading | +0.062 |
Neg logit contributions | |
bleed | -0.080 |
ranger | -0.080 |
bru | -0.078 |
hinges | -0.078 |
sob | -0.077 |
Token | glued |
Token ID | 39783 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
ranger | +0.007 |
bleed | +0.007 |
bru | +0.007 |
sob | +0.007 |
hinges | +0.007 |
Neg logit contributions | |
ang | -0.010 |
irling | -0.009 |
ee | -0.009 |
irl | -0.008 |
ug | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +6.34e-03 |
Pos logit contributions | |
ang | +0.098 |
irling | +0.087 |
ee | +0.082 |
irl | +0.080 |
ug | +0.077 |
Neg logit contributions | |
ranger | -0.105 |
bleed | -0.104 |
bru | -0.102 |
hinges | -0.102 |
sob | -0.101 |
Token | and |
Token ID | 290 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ang | +0.013 |
irling | +0.012 |
ee | +0.012 |
irl | +0.011 |
ug | +0.011 |
Neg logit contributions | |
ranger | -0.010 |
bleed | -0.010 |
bru | -0.010 |
hinges | -0.010 |
sob | -0.010 |
Token | own |
Token ID | 898 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
ranger | +0.051 |
bleed | +0.050 |
hinges | +0.049 |
department | +0.049 |
bru | +0.048 |
Neg logit contributions | |
ang | -0.063 |
irling | -0.056 |
ee | -0.055 |
irl | -0.054 |
ug | -0.051 |
Token | two |
Token ID | 734 |
Feature activation | +5.94e-05 |
Pos logit contributions | |
ranger | +0.028 |
bleed | +0.028 |
hinges | +0.027 |
bru | +0.027 |
department | +0.027 |
Neg logit contributions | |
ang | -0.041 |
irling | -0.038 |
ee | -0.037 |
irl | -0.036 |
ug | -0.035 |
Token | hands |
Token ID | 2832 |
Feature activation | +7.59e-03 |
Pos logit contributions | |
ang | +0.000 |
irling | +0.000 |
ee | +0.000 |
irl | +0.000 |
ug | +0.000 |
Neg logit contributions | |
ranger | -0.000 |
bleed | -0.000 |
sob | -0.000 |
bru | -0.000 |
hinges | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
ang | +0.017 |
irling | +0.015 |
ee | +0.015 |
irl | +0.015 |
ug | +0.014 |
Neg logit contributions | |
ranger | -0.011 |
bleed | -0.011 |
bru | -0.010 |
hinges | -0.010 |
sob | -0.010 |
Token | She |
Token ID | 1375 |
Feature activation | +9.29e-03 |
Pos logit contributions | |
ang | +0.036 |
irling | +0.033 |
ee | +0.032 |
irl | +0.031 |
ug | +0.030 |
Neg logit contributions | |
ranger | -0.019 |
bleed | -0.019 |
bru | -0.018 |
hinges | -0.018 |
sob | -0.018 |
Token | se |
Token ID | 384 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
ang | +0.021 |
irling | +0.018 |
ee | +0.018 |
ug | +0.018 |
ash | +0.017 |
Neg logit contributions | |
sob | -0.013 |
ranger | -0.013 |
bru | -0.013 |
bleed | -0.013 |
aid | -0.012 |
Token | wed |
Token ID | 19103 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
ang | +0.243 |
irling | +0.220 |
irl | +0.198 |
ee | +0.197 |
ading | +0.193 |
Neg logit contributions | |
bleed | -0.199 |
bru | -0.192 |
sob | -0.188 |
ranger | -0.188 |
recovered | -0.184 |
Token | it |
Token ID | 340 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
ang | +0.102 |
irling | +0.095 |
ee | +0.086 |
irl | +0.086 |
ading | +0.082 |
Neg logit contributions | |
ranger | -0.092 |
bleed | -0.092 |
bru | -0.090 |
sob | -0.089 |
hinges | -0.089 |
Token | with |
Token ID | 351 |
Feature activation | +4.69e-02 |
Pos logit contributions | |
ranger | +0.026 |
bleed | +0.026 |
bru | +0.025 |
hinges | +0.025 |
sob | +0.025 |
Neg logit contributions | |
ang | -0.022 |
irling | -0.019 |
ee | -0.018 |
irl | -0.018 |
ug | -0.017 |
Token | bright |
Token ID | 6016 |
Feature activation | +4.12e-02 |
Pos logit contributions | |
ang | +0.103 |
irling | +0.093 |
ee | +0.088 |
irl | +0.086 |
ug | +0.085 |
Neg logit contributions | |
bleed | -0.071 |
ranger | -0.070 |
bru | -0.069 |
sob | -0.068 |
hinges | -0.068 |
Token | colors |
Token ID | 7577 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
ang | +0.079 |
irling | +0.069 |
ee | +0.066 |
irl | +0.064 |
ug | +0.062 |
Neg logit contributions | |
bleed | -0.072 |
hinges | -0.071 |
bru | -0.071 |
ranger | -0.071 |
sob | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | +1.50e-04 |
Pos logit contributions | |
bleed | +0.048 |
bru | +0.048 |
ranger | +0.047 |
hinges | +0.046 |
sob | +0.046 |
Neg logit contributions | |
ang | -0.057 |
ee | -0.050 |
irling | -0.049 |
ug | -0.047 |
irl | -0.046 |
Token | it |
Token ID | 340 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
ang | +0.000 |
irling | +0.000 |
ee | +0.000 |
irl | +0.000 |
ug | +0.000 |
Neg logit contributions | |
ranger | -0.000 |
bleed | -0.000 |
sob | -0.000 |
hinges | -0.000 |
bru | -0.000 |
Token | fit |
Token ID | 4197 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
ang | +0.022 |
irling | +0.020 |
ee | +0.020 |
irl | +0.020 |
ug | +0.018 |
Neg logit contributions | |
ranger | -0.020 |
bru | -0.020 |
bleed | -0.020 |
hinges | -0.020 |
injuries | -0.019 |
Token | perfectly |
Token ID | 7138 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
ang | +0.047 |
irling | +0.040 |
ee | +0.039 |
irl | +0.037 |
ug | +0.036 |
Neg logit contributions | |
ranger | -0.064 |
bleed | -0.063 |
bru | -0.063 |
hinges | -0.062 |
sob | -0.062 |
Token | ! |
Token ID | 0 |
Feature activation | +2.92e-03 |
Pos logit contributions | |
ang | +0.032 |
irling | +0.030 |
ee | +0.029 |
irl | +0.028 |
ug | +0.028 |
Neg logit contributions | |
ranger | -0.016 |
bleed | -0.016 |
bru | -0.015 |
hinges | -0.015 |
sob | -0.015 |
Token | Sil |
Token ID | 4243 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
ang | +0.007 |
irling | +0.006 |
ee | +0.006 |
irl | +0.006 |
ading | +0.006 |
Neg logit contributions | |
ranger | -0.004 |
hinges | -0.004 |
bleed | -0.004 |
sob | -0.004 |
department | -0.004 |
Token | via |
Token ID | 8869 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
ang | +0.213 |
irling | +0.207 |
irl | +0.197 |
ading | +0.188 |
ee | +0.178 |
Neg logit contributions | |
hinges | -0.174 |
aid | -0.174 |
ache | -0.171 |
sob | -0.169 |
driver | -0.166 |
Token | was |
Token ID | 373 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
ang | +0.065 |
irling | +0.059 |
ading | +0.056 |
ug | +0.056 |
ash | +0.055 |
Neg logit contributions | |
sob | -0.031 |
bru | -0.029 |
ranger | -0.029 |
bleed | -0.028 |
recovered | -0.028 |
Token | so |
Token ID | 523 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
ang | +0.033 |
ee | +0.029 |
irling | +0.028 |
ug | +0.028 |
irl | +0.027 |
Neg logit contributions | |
sob | -0.029 |
bleed | -0.029 |
ranger | -0.029 |
bru | -0.028 |
hinges | -0.027 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
ang | +0.030 |
ash | +0.027 |
irling | +0.027 |
ug | +0.026 |
ee | +0.026 |
Neg logit contributions | |
bleed | -0.023 |
sob | -0.022 |
ranger | -0.022 |
bru | -0.021 |
keeper | -0.020 |
Token | she |
Token ID | 673 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
bleed | +0.034 |
bru | +0.033 |
hinges | +0.033 |
ranger | +0.032 |
sob | +0.031 |
Neg logit contributions | |
ang | -0.039 |
irling | -0.035 |
ee | -0.034 |
irl | -0.032 |
ug | -0.031 |
Token | his |
Token ID | 465 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
ranger | +0.090 |
bleed | +0.090 |
bru | +0.089 |
hinges | +0.088 |
sob | +0.087 |
Neg logit contributions | |
ang | -0.075 |
irling | -0.066 |
ee | -0.063 |
irl | -0.061 |
ug | -0.058 |
Token | mistake |
Token ID | 7457 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
ranger | +0.082 |
bleed | +0.081 |
recovered | +0.078 |
keeping | +0.078 |
bru | +0.077 |
Neg logit contributions | |
irl | -0.080 |
ang | -0.076 |
irling | -0.076 |
ee | -0.072 |
ading | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | -9.31e-03 |
Pos logit contributions | |
hinges | +0.067 |
bleed | +0.066 |
bru | +0.066 |
ranger | +0.065 |
injuries | +0.063 |
Neg logit contributions | |
ang | -0.070 |
ee | -0.061 |
irling | -0.061 |
ug | -0.058 |
ash | -0.055 |
Token | gave |
Token ID | 2921 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
bleed | +0.014 |
ranger | +0.014 |
bru | +0.014 |
hinges | +0.014 |
department | +0.013 |
Neg logit contributions | |
ang | -0.019 |
irling | -0.018 |
ee | -0.017 |
irl | -0.017 |
ug | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
ang | +0.069 |
irling | +0.062 |
ee | +0.060 |
irl | +0.059 |
ug | +0.057 |
Neg logit contributions | |
ranger | -0.048 |
bleed | -0.047 |
bru | -0.046 |
hinges | -0.046 |
sob | -0.046 |
Token | sister |
Token ID | 6621 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
bleed | +0.116 |
ranger | +0.110 |
bru | +0.108 |
onna | +0.108 |
keeping | +0.107 |
Neg logit contributions | |
ang | -0.085 |
irl | -0.078 |
ee | -0.076 |
irling | -0.075 |
these | -0.069 |
Token | a |
Token ID | 257 |
Feature activation | -6.26e-02 |
Pos logit contributions | |
ranger | +0.183 |
bleed | +0.182 |
hinges | +0.181 |
bru | +0.180 |
department | +0.179 |
Neg logit contributions | |
ang | -0.190 |
ee | -0.170 |
irling | -0.169 |
irl | -0.161 |
ash | -0.154 |
Token | turn |
Token ID | 1210 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
keeping | +0.138 |
edly | +0.130 |
hinges | +0.127 |
ranger | +0.122 |
injuries | +0.121 |
Neg logit contributions | |
ee | -0.089 |
ang | -0.083 |
sorts | -0.074 |
irl | -0.072 |
irling | -0.062 |
Token | with |
Token ID | 351 |
Feature activation | +2.31e-02 |
Pos logit contributions | |
hinges | +0.027 |
bru | +0.026 |
ranger | +0.026 |
bleed | +0.026 |
department | +0.025 |
Neg logit contributions | |
ang | -0.035 |
ee | -0.032 |
irling | -0.030 |
ug | -0.030 |
ash | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
ang | +0.041 |
irling | +0.037 |
ee | +0.037 |
irl | +0.036 |
ug | +0.034 |
Neg logit contributions | |
ranger | -0.042 |
hinges | -0.041 |
department | -0.040 |
bleed | -0.040 |
bru | -0.040 |
Token | toy |
Token ID | 13373 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
ranger | +0.048 |
hinges | +0.045 |
bleed | +0.045 |
bru | +0.045 |
mercy | +0.044 |
Neg logit contributions | |
ang | -0.029 |
ee | -0.029 |
ading | -0.027 |
irl | -0.027 |
irling | -0.027 |
Token | cry |
Token ID | 3960 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
department | +0.086 |
ranger | +0.081 |
hinges | +0.079 |
edly | +0.078 |
keeper | +0.077 |
Neg logit contributions | |
irl | -0.149 |
ang | -0.146 |
irling | -0.146 |
ee | -0.146 |
ading | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
ranger | +0.096 |
keeping | +0.092 |
onna | +0.090 |
hinges | +0.089 |
department | +0.088 |
Neg logit contributions | |
ang | -0.080 |
ug | -0.070 |
ing | -0.065 |
irling | -0.064 |
ee | -0.064 |
Token | His |
Token ID | 2399 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
bleed | +0.079 |
department | +0.075 |
injuries | +0.073 |
recovered | +0.073 |
mercy | +0.071 |
Neg logit contributions | |
ang | -0.076 |
aw | -0.063 |
ug | -0.060 |
irling | -0.060 |
Now | -0.058 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
Something | +0.082 |
recovered | +0.079 |
bleed | +0.078 |
opener | +0.077 |
department | +0.077 |
Neg logit contributions | |
ang | -0.074 |
irl | -0.066 |
irling | -0.064 |
ee | -0.060 |
utes | -0.055 |
Token | heard |
Token ID | 2982 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
bleed | +0.055 |
hinges | +0.055 |
ranger | +0.054 |
onna | +0.053 |
bru | +0.052 |
Neg logit contributions | |
ang | -0.071 |
ee | -0.063 |
irling | -0.063 |
ug | -0.058 |
irl | -0.057 |
Token | him |
Token ID | 683 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
hinges | +0.050 |
bleed | +0.049 |
ranger | +0.048 |
department | +0.048 |
bru | +0.048 |
Neg logit contributions | |
ang | -0.035 |
ee | -0.031 |
irling | -0.030 |
aw | -0.029 |
ug | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | -4.91e-02 |
Pos logit contributions | |
onna | +0.210 |
ranger | +0.207 |
department | +0.206 |
hinges | +0.205 |
bru | +0.204 |
Neg logit contributions | |
ang | -0.154 |
ee | -0.130 |
irling | -0.126 |
ug | -0.119 |
ash | -0.112 |
Token | came |
Token ID | 1625 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
hinges | +0.083 |
bleed | +0.082 |
bru | +0.080 |
onna | +0.080 |
ranger | +0.079 |
Neg logit contributions | |
ang | -0.092 |
irling | -0.079 |
ee | -0.078 |
ug | -0.075 |
irl | -0.071 |
Token | to |
Token ID | 284 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
bleed | +0.032 |
bru | +0.031 |
ranger | +0.030 |
ache | +0.030 |
Something | +0.029 |
Neg logit contributions | |
ang | -0.055 |
irling | -0.049 |
ee | -0.048 |
ug | -0.047 |
ading | -0.046 |
Token | see |
Token ID | 766 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
ache | +0.136 |
edly | +0.129 |
bleed | +0.128 |
hinges | +0.128 |
comb | +0.126 |
Neg logit contributions | |
ang | -0.052 |
these | -0.048 |
ee | -0.047 |
us | -0.043 |
those | -0.040 |
Token | what |
Token ID | 644 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
bleed | +0.049 |
hinges | +0.048 |
ranger | +0.048 |
onna | +0.047 |
bru | +0.047 |
Neg logit contributions | |
ang | -0.020 |
ee | -0.015 |
irling | -0.014 |
aw | -0.014 |
these | -0.012 |
Token | grandma |
Token ID | 49890 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
bleed | +0.033 |
ranger | +0.032 |
bru | +0.032 |
hinges | +0.031 |
sob | +0.031 |
Neg logit contributions | |
ang | -0.048 |
irling | -0.043 |
ee | -0.042 |
irl | -0.041 |
ug | -0.040 |
Token | 's |
Token ID | 338 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
bleed | +0.062 |
ranger | +0.062 |
bru | +0.060 |
hinges | +0.060 |
sob | +0.059 |
Neg logit contributions | |
ang | -0.080 |
irling | -0.071 |
ee | -0.069 |
irl | -0.067 |
ug | -0.065 |
Token | house |
Token ID | 2156 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
bleed | +0.050 |
onna | +0.046 |
bru | +0.046 |
recovered | +0.046 |
ranger | +0.045 |
Neg logit contributions | |
ang | -0.050 |
irling | -0.044 |
ee | -0.042 |
ug | -0.040 |
irl | -0.040 |
Token | for |
Token ID | 329 |
Feature activation | -4.18e-03 |
Pos logit contributions | |
bleed | +0.056 |
bru | +0.053 |
sob | +0.051 |
hinges | +0.051 |
ranger | +0.049 |
Neg logit contributions | |
ang | -0.059 |
irling | -0.051 |
ee | -0.050 |
ug | -0.050 |
ash | -0.048 |
Token | safe |
Token ID | 3338 |
Feature activation | -7.41e-03 |
Pos logit contributions | |
ranger | +0.007 |
bleed | +0.007 |
department | +0.007 |
bru | +0.007 |
hinges | +0.007 |
Neg logit contributions | |
ang | -0.008 |
irling | -0.007 |
irl | -0.007 |
ee | -0.007 |
ug | -0.006 |
Token | keeping |
Token ID | 19934 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
bleed | +0.010 |
bru | +0.010 |
onna | +0.010 |
Something | +0.010 |
department | +0.010 |
Neg logit contributions | |
ang | -0.015 |
ug | -0.014 |
irling | -0.014 |
ee | -0.013 |
ash | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -8.07e-03 |
Pos logit contributions | |
ranger | +0.167 |
department | +0.166 |
bru | +0.163 |
bleed | +0.162 |
mercy | +0.161 |
Neg logit contributions | |
ang | -0.185 |
irling | -0.174 |
ee | -0.166 |
ug | -0.161 |
irl | -0.157 |
Token | |
Token ID | 220 |
Feature activation | -3.69e-03 |
Pos logit contributions | |
bleed | +0.012 |
ranger | +0.012 |
bru | +0.012 |
hinges | +0.012 |
sob | +0.012 |
Neg logit contributions | |
ang | -0.017 |
irling | -0.016 |
ee | -0.015 |
irl | -0.015 |
ug | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -7.85e-03 |
Pos logit contributions | |
ranger | +0.005 |
bleed | +0.005 |
bru | +0.004 |
hinges | +0.004 |
sob | +0.004 |
Neg logit contributions | |
ang | -0.009 |
irling | -0.008 |
ee | -0.008 |
irl | -0.008 |
ug | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +6.00e-04 |
Pos logit contributions | |
bleed | +0.010 |
ranger | +0.010 |
bru | +0.009 |
hinges | +0.009 |
sob | +0.009 |
Neg logit contributions | |
ang | -0.019 |
irling | -0.017 |
ee | -0.017 |
irl | -0.017 |
ug | -0.016 |
Token | When |
Token ID | 2215 |
Feature activation | -2.77e-03 |
Pos logit contributions | |
ang | +0.001 |
irling | +0.001 |
ee | +0.001 |
irl | +0.001 |
ug | +0.001 |
Neg logit contributions | |
ranger | -0.001 |
bleed | -0.001 |
bru | -0.001 |
hinges | -0.001 |
sob | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
ang | +0.104 |
irling | +0.096 |
irl | +0.086 |
ee | +0.082 |
ading | +0.081 |
Neg logit contributions | |
ranger | -0.128 |
bleed | -0.126 |
bru | -0.124 |
sob | -0.123 |
hinges | -0.123 |
Token | to |
Token ID | 284 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
bru | +0.061 |
bleed | +0.059 |
sob | +0.058 |
injuries | +0.054 |
ranger | +0.053 |
Neg logit contributions | |
ang | -0.096 |
ee | -0.094 |
ash | -0.087 |
aw | -0.086 |
ug | -0.085 |
Token | me |
Token ID | 502 |
Feature activation | -5.01e-02 |
Pos logit contributions | |
ranger | +0.063 |
bru | +0.061 |
bleed | +0.061 |
sob | +0.061 |
hinges | +0.060 |
Neg logit contributions | |
ang | -0.069 |
irling | -0.062 |
ee | -0.058 |
irl | -0.057 |
ug | -0.055 |
Token | for |
Token ID | 329 |
Feature activation | +1.87e-03 |
Pos logit contributions | |
bleed | +0.076 |
ranger | +0.075 |
bru | +0.074 |
hinges | +0.072 |
sob | +0.072 |
Neg logit contributions | |
ang | -0.108 |
irling | -0.096 |
ee | -0.095 |
irl | -0.092 |
ug | -0.090 |
Token | safe |
Token ID | 3338 |
Feature activation | -3.25e-03 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.003 |
irl | +0.003 |
ug | +0.003 |
Neg logit contributions | |
ranger | -0.004 |
bleed | -0.004 |
bru | -0.003 |
hinges | -0.003 |
sob | -0.003 |
Token | keeping |
Token ID | 19934 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
bleed | +0.004 |
bru | +0.004 |
sob | +0.004 |
Something | +0.004 |
department | +0.004 |
Neg logit contributions | |
ang | -0.007 |
ug | -0.006 |
ee | -0.006 |
irling | -0.006 |
irl | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +3.09e-05 |
Pos logit contributions | |
department | +0.165 |
hinges | +0.165 |
Something | +0.164 |
bleed | +0.162 |
bru | +0.162 |
Neg logit contributions | |
ang | -0.185 |
irling | -0.172 |
ee | -0.164 |
irl | -0.157 |
ug | -0.153 |
Token | He |
Token ID | 679 |
Feature activation | -8.79e-03 |
Pos logit contributions | |
ang | +0.000 |
irling | +0.000 |
ee | +0.000 |
irl | +0.000 |
ug | +0.000 |
Neg logit contributions | |
bleed | -0.000 |
ranger | -0.000 |
bru | -0.000 |
hinges | -0.000 |
sob | -0.000 |
Token | told |
Token ID | 1297 |
Feature activation | -1.64e-03 |
Pos logit contributions | |
hinges | +0.016 |
bleed | +0.015 |
ranger | +0.015 |
bru | +0.015 |
injuries | +0.015 |
Neg logit contributions | |
ang | -0.016 |
ee | -0.014 |
irling | -0.014 |
irl | -0.014 |
ug | -0.013 |
Token | me |
Token ID | 502 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
bleed | +0.003 |
ranger | +0.003 |
bru | +0.003 |
hinges | +0.003 |
sob | +0.003 |
Neg logit contributions | |
ang | -0.003 |
irling | -0.002 |
ee | -0.002 |
irl | -0.002 |
ug | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
bleed | +0.084 |
ranger | +0.083 |
bru | +0.082 |
hinges | +0.081 |
sob | +0.081 |
Neg logit contributions | |
ang | -0.070 |
irling | -0.062 |
ee | -0.061 |
irl | -0.058 |
ug | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -8.59e-03 |
Pos logit contributions | |
bleed | +0.052 |
department | +0.052 |
bud | +0.052 |
hinges | +0.051 |
ache | +0.050 |
Neg logit contributions | |
ang | -0.048 |
ug | -0.040 |
ee | -0.039 |
irling | -0.036 |
aw | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
bleed | +0.013 |
hinges | +0.012 |
ranger | +0.012 |
bru | +0.012 |
department | +0.012 |
Neg logit contributions | |
ang | -0.019 |
irling | -0.017 |
ee | -0.017 |
irl | -0.016 |
ug | -0.016 |
Token | His |
Token ID | 6653 |
Feature activation | -5.07e-02 |
Pos logit contributions | |
bleed | +0.020 |
bru | +0.020 |
hinges | +0.020 |
department | +0.019 |
ranger | +0.019 |
Neg logit contributions | |
ang | -0.018 |
irling | -0.015 |
ee | -0.015 |
ug | -0.014 |
irl | -0.014 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
Something | +0.086 |
keeping | +0.085 |
bleed | +0.082 |
recovered | +0.081 |
onna | +0.081 |
Neg logit contributions | |
ang | -0.092 |
irl | -0.084 |
irling | -0.081 |
ee | -0.079 |
ug | -0.073 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
bleed | +0.040 |
onna | +0.039 |
ranger | +0.039 |
hinges | +0.039 |
bru | +0.038 |
Neg logit contributions | |
ang | -0.061 |
ee | -0.055 |
irling | -0.053 |
ug | -0.051 |
aw | -0.051 |
Token | him |
Token ID | 683 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
hinges | +0.041 |
ranger | +0.040 |
department | +0.040 |
bleed | +0.039 |
bru | +0.039 |
Neg logit contributions | |
ang | -0.029 |
ee | -0.026 |
irling | -0.025 |
ug | -0.023 |
irl | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
department | +0.186 |
ranger | +0.184 |
hinges | +0.180 |
bru | +0.177 |
onna | +0.176 |
Neg logit contributions | |
ang | -0.175 |
ee | -0.151 |
irling | -0.145 |
ug | -0.141 |
irl | -0.133 |
Token | came |
Token ID | 1625 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
hinges | +0.073 |
bleed | +0.072 |
onna | +0.071 |
bru | +0.071 |
ranger | +0.070 |
Neg logit contributions | |
ang | -0.083 |
ee | -0.071 |
irling | -0.070 |
ug | -0.068 |
irl | -0.063 |
Token | to |
Token ID | 284 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
bleed | +0.026 |
bru | +0.025 |
ranger | +0.025 |
ache | +0.024 |
Something | +0.023 |
Neg logit contributions | |
ang | -0.053 |
irling | -0.047 |
ee | -0.046 |
ug | -0.045 |
ading | -0.044 |
Token | help |
Token ID | 1037 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
ache | +0.133 |
hinges | +0.128 |
edly | +0.127 |
department | +0.123 |
onna | +0.123 |
Neg logit contributions | |
ang | -0.054 |
ee | -0.049 |
these | -0.048 |
us | -0.044 |
rous | -0.040 |
Token | him |
Token ID | 683 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
department | +0.039 |
hinges | +0.038 |
ache | +0.038 |
bru | +0.037 |
ranger | +0.037 |
Neg logit contributions | |
ang | -0.031 |
ee | -0.030 |
ug | -0.028 |
ie | -0.026 |
ading | -0.025 |
Token | was |
Token ID | 373 |
Feature activation | -1.38e-02 |
Pos logit contributions | |
bleed | +0.049 |
hinges | +0.049 |
bru | +0.047 |
ranger | +0.046 |
injuries | +0.046 |
Neg logit contributions | |
ang | -0.048 |
irling | -0.044 |
ee | -0.042 |
irl | -0.040 |
ug | -0.038 |
Token | angry |
Token ID | 7954 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
hinges | +0.028 |
ranger | +0.028 |
bleed | +0.028 |
bru | +0.028 |
department | +0.027 |
Neg logit contributions | |
ang | -0.022 |
irling | -0.020 |
ee | -0.018 |
irl | -0.018 |
ug | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
bleed | +0.064 |
hinges | +0.062 |
bru | +0.062 |
ranger | +0.060 |
sob | +0.059 |
Neg logit contributions | |
ang | -0.072 |
irling | -0.063 |
ee | -0.061 |
ug | -0.058 |
irl | -0.057 |
Token | asked |
Token ID | 1965 |
Feature activation | -4.98e-02 |
Pos logit contributions | |
hinges | +0.036 |
bleed | +0.036 |
ranger | +0.036 |
bru | +0.036 |
department | +0.035 |
Neg logit contributions | |
ang | -0.049 |
irling | -0.045 |
ee | -0.042 |
irl | -0.041 |
ug | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
bleed | +0.108 |
ranger | +0.103 |
hinges | +0.101 |
bru | +0.100 |
onna | +0.098 |
Neg logit contributions | |
ang | -0.073 |
irling | -0.064 |
ee | -0.060 |
ug | -0.060 |
irl | -0.058 |
Token | daughter |
Token ID | 4957 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
bleed | +0.195 |
bru | +0.184 |
hinges | +0.182 |
ranger | +0.177 |
choke | +0.175 |
Neg logit contributions | |
ang | -0.087 |
irling | -0.068 |
ee | -0.067 |
ug | -0.061 |
ading | -0.060 |
Token | why |
Token ID | 1521 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
bleed | +0.219 |
hinges | +0.207 |
bru | +0.207 |
ranger | +0.201 |
injuries | +0.195 |
Neg logit contributions | |
ang | -0.136 |
irling | -0.116 |
ee | -0.112 |
ash | -0.110 |
these | -0.109 |
Token | she |
Token ID | 673 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
hinges | +0.074 |
ranger | +0.073 |
bleed | +0.073 |
bru | +0.072 |
department | +0.071 |
Neg logit contributions | |
ang | -0.091 |
irling | -0.083 |
ee | -0.080 |
irl | -0.079 |
ug | -0.075 |
Token | had |
Token ID | 550 |
Feature activation | -2.59e-03 |
Pos logit contributions | |
ranger | +0.075 |
hinges | +0.075 |
bleed | +0.074 |
bru | +0.073 |
department | +0.071 |
Neg logit contributions | |
ang | -0.091 |
irling | -0.083 |
ee | -0.082 |
irl | -0.079 |
ug | -0.074 |
Token | done |
Token ID | 1760 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
ranger | +0.004 |
bleed | +0.004 |
hinges | +0.004 |
bru | +0.004 |
department | +0.004 |
Neg logit contributions | |
ang | -0.005 |
irling | -0.005 |
ee | -0.005 |
irl | -0.005 |
ug | -0.004 |
Token | something |
Token ID | 1223 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
ranger | +0.066 |
bleed | +0.066 |
hinges | +0.065 |
bru | +0.065 |
sob | +0.063 |
Neg logit contributions | |
ang | -0.059 |
irling | -0.053 |
ee | -0.050 |
irl | -0.049 |
ug | -0.046 |
Token | He |
Token ID | 679 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
department | +0.059 |
bleed | +0.059 |
injuries | +0.058 |
comb | +0.058 |
bud | +0.056 |
Neg logit contributions | |
ang | -0.047 |
aw | -0.037 |
ug | -0.035 |
Everyone | -0.033 |
ee | -0.033 |
Token | tried |
Token ID | 3088 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
bleed | +0.056 |
hinges | +0.054 |
bru | +0.054 |
ranger | +0.054 |
department | +0.052 |
Neg logit contributions | |
ang | -0.069 |
irling | -0.062 |
ee | -0.059 |
irl | -0.058 |
ug | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | -7.11e-02 |
Pos logit contributions | |
comb | +0.076 |
hinges | +0.074 |
atively | +0.071 |
edly | +0.070 |
ache | +0.070 |
Neg logit contributions | |
ang | -0.071 |
these | -0.068 |
irling | -0.066 |
all | -0.065 |
irl | -0.065 |
Token | push |
Token ID | 4574 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
department | +0.111 |
ranger | +0.110 |
hinges | +0.110 |
bru | +0.107 |
onna | +0.106 |
Neg logit contributions | |
ang | -0.139 |
irl | -0.129 |
irling | -0.129 |
ee | -0.127 |
ading | -0.119 |
Token | the |
Token ID | 262 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
ang | +0.036 |
ee | +0.031 |
ug | +0.029 |
irling | +0.029 |
irl | +0.028 |
Neg logit contributions | |
bleed | -0.033 |
ranger | -0.033 |
bru | -0.033 |
hinges | -0.033 |
department | -0.032 |
Token | girl |
Token ID | 2576 |
Feature activation | -9.94e-02 |
Pos logit contributions | |
hinges | +0.053 |
ranger | +0.052 |
onna | +0.052 |
department | +0.052 |
bleed | +0.051 |
Neg logit contributions | |
ang | -0.053 |
ee | -0.046 |
irling | -0.043 |
irl | -0.041 |
is | -0.039 |
Token | away |
Token ID | 1497 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
bru | +0.183 |
hinges | +0.180 |
department | +0.180 |
onna | +0.180 |
ranger | +0.179 |
Neg logit contributions | |
ang | -0.174 |
ee | -0.144 |
irling | -0.139 |
ash | -0.134 |
ug | -0.134 |
Token | , |
Token ID | 11 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
bru | +0.096 |
bleed | +0.094 |
recovered | +0.091 |
hinges | +0.091 |
keeping | +0.091 |
Neg logit contributions | |
ang | -0.075 |
ug | -0.060 |
ee | -0.059 |
irling | -0.054 |
il | -0.052 |
Token | but |
Token ID | 475 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
hinges | +0.045 |
ranger | +0.045 |
bleed | +0.044 |
department | +0.044 |
bru | +0.042 |
Neg logit contributions | |
ang | -0.061 |
irling | -0.055 |
ee | -0.051 |
irl | -0.050 |
ug | -0.049 |
Token | she |
Token ID | 673 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
hinges | +0.029 |
bleed | +0.029 |
bru | +0.028 |
department | +0.028 |
ranger | +0.028 |
Neg logit contributions | |
ang | -0.030 |
irling | -0.026 |
ee | -0.026 |
irl | -0.025 |
ug | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | -4.41e-03 |
Pos logit contributions | |
hinges | +0.034 |
bleed | +0.033 |
bru | +0.032 |
ranger | +0.032 |
department | +0.031 |
Neg logit contributions | |
ang | -0.041 |
ee | -0.036 |
irling | -0.035 |
irl | -0.034 |
ug | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
injuries | +0.075 |
bleed | +0.075 |
hinges | +0.074 |
recovered | +0.073 |
department | +0.071 |
Neg logit contributions | |
ang | -0.046 |
ee | -0.035 |
ug | -0.031 |
ch | -0.029 |
aw | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
bleed | +0.019 |
hinges | +0.019 |
department | +0.019 |
ranger | +0.019 |
bru | +0.019 |
Neg logit contributions | |
ang | -0.025 |
ee | -0.022 |
irling | -0.022 |
irl | -0.021 |
ug | -0.021 |
Token | His |
Token ID | 6653 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
bleed | +0.017 |
bru | +0.016 |
ranger | +0.016 |
hinges | +0.016 |
sob | +0.016 |
Neg logit contributions | |
ang | -0.022 |
irling | -0.019 |
ee | -0.019 |
irl | -0.018 |
ug | -0.018 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
keeping | +0.074 |
recovered | +0.072 |
bleed | +0.072 |
onna | +0.070 |
Something | +0.070 |
Neg logit contributions | |
ang | -0.081 |
ee | -0.073 |
irling | -0.071 |
irl | -0.070 |
ug | -0.063 |
Token | hears |
Token ID | 23180 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
ranger | +0.028 |
hinges | +0.028 |
bleed | +0.027 |
onna | +0.026 |
bru | +0.026 |
Neg logit contributions | |
ang | -0.051 |
ee | -0.048 |
irling | -0.047 |
irl | -0.044 |
aw | -0.044 |
Token | him |
Token ID | 683 |
Feature activation | -9.90e-02 |
Pos logit contributions | |
hinges | +0.037 |
bleed | +0.037 |
ranger | +0.037 |
bru | +0.036 |
department | +0.036 |
Neg logit contributions | |
ang | -0.027 |
ee | -0.025 |
irling | -0.023 |
ug | -0.022 |
aw | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -4.85e-02 |
Pos logit contributions | |
ranger | +0.195 |
hinges | +0.190 |
department | +0.190 |
bru | +0.188 |
onna | +0.188 |
Neg logit contributions | |
ang | -0.156 |
ee | -0.136 |
irling | -0.130 |
ug | -0.121 |
ash | -0.118 |
Token | runs |
Token ID | 4539 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
hinges | +0.077 |
ranger | +0.077 |
bru | +0.076 |
bleed | +0.076 |
injuries | +0.074 |
Neg logit contributions | |
ang | -0.098 |
irling | -0.086 |
ee | -0.085 |
irl | -0.081 |
ug | -0.078 |
Token | to |
Token ID | 284 |
Feature activation | -4.86e-02 |
Pos logit contributions | |
onna | +0.025 |
bru | +0.025 |
bleed | +0.025 |
Something | +0.024 |
ache | +0.024 |
Neg logit contributions | |
ang | -0.029 |
ee | -0.025 |
ash | -0.025 |
around | -0.024 |
these | -0.024 |
Token | his |
Token ID | 465 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
ache | +0.126 |
hinges | +0.119 |
department | +0.118 |
edly | +0.117 |
bleed | +0.115 |
Neg logit contributions | |
ang | -0.044 |
ee | -0.040 |
these | -0.036 |
irl | -0.033 |
ading | -0.033 |
Token | room |
Token ID | 2119 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
bleed | +0.135 |
recovered | +0.132 |
onna | +0.132 |
ache | +0.131 |
bru | +0.127 |
Neg logit contributions | |
ang | -0.062 |
irling | -0.044 |
ee | -0.040 |
irl | -0.037 |
ading | -0.035 |
Token | cry |
Token ID | 3960 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
ranger | +0.136 |
department | +0.129 |
hinges | +0.126 |
keeper | +0.118 |
injuries | +0.117 |
Neg logit contributions | |
ang | -0.166 |
ee | -0.160 |
irl | -0.146 |
irling | -0.144 |
ug | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
ranger | +0.071 |
bru | +0.070 |
hinges | +0.067 |
bleed | +0.066 |
onna | +0.064 |
Neg logit contributions | |
ang | -0.092 |
ee | -0.082 |
ug | -0.081 |
irling | -0.080 |
ash | -0.076 |
Token | His |
Token ID | 2399 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
bleed | +0.041 |
hinges | +0.040 |
department | +0.039 |
ranger | +0.039 |
recovered | +0.038 |
Neg logit contributions | |
ang | -0.052 |
ee | -0.046 |
irling | -0.044 |
irl | -0.043 |
ug | -0.042 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
bleed | +0.084 |
recovered | +0.079 |
ranger | +0.079 |
onna | +0.078 |
hinges | +0.077 |
Neg logit contributions | |
ang | -0.090 |
ee | -0.087 |
irl | -0.085 |
irling | -0.083 |
ug | -0.072 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
ranger | +0.039 |
bleed | +0.039 |
hinges | +0.039 |
bru | +0.038 |
onna | +0.037 |
Neg logit contributions | |
ang | -0.056 |
ee | -0.051 |
irling | -0.050 |
irl | -0.046 |
ug | -0.046 |
Token | him |
Token ID | 683 |
Feature activation | -9.90e-02 |
Pos logit contributions | |
ranger | +0.033 |
hinges | +0.033 |
bleed | +0.033 |
bru | +0.032 |
department | +0.032 |
Neg logit contributions | |
ang | -0.028 |
ee | -0.026 |
irling | -0.025 |
irl | -0.024 |
ug | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -4.10e-02 |
Pos logit contributions | |
ranger | +0.195 |
hinges | +0.190 |
bru | +0.189 |
department | +0.188 |
onna | +0.187 |
Neg logit contributions | |
ang | -0.155 |
ee | -0.138 |
irling | -0.129 |
ug | -0.124 |
ading | -0.116 |
Token | ran |
Token ID | 4966 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
bleed | +0.066 |
hinges | +0.065 |
bru | +0.065 |
ranger | +0.064 |
injuries | +0.062 |
Neg logit contributions | |
ang | -0.083 |
ee | -0.072 |
irling | -0.072 |
ug | -0.068 |
irl | -0.066 |
Token | to |
Token ID | 284 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
bru | +0.035 |
bleed | +0.035 |
ache | +0.034 |
onna | +0.032 |
Something | +0.032 |
Neg logit contributions | |
ang | -0.039 |
ash | -0.036 |
ee | -0.036 |
these | -0.035 |
around | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
ache | +0.143 |
hinges | +0.135 |
department | +0.135 |
edly | +0.133 |
bleed | +0.130 |
Neg logit contributions | |
ang | -0.059 |
ee | -0.056 |
irl | -0.049 |
these | -0.049 |
irling | -0.045 |
Token | dining |
Token ID | 17423 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
bleed | +0.052 |
recovered | +0.050 |
ache | +0.049 |
bru | +0.049 |
onna | +0.047 |
Neg logit contributions | |
ang | -0.029 |
ee | -0.026 |
irling | -0.025 |
these | -0.024 |
ading | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +3.07e-03 |
Pos logit contributions | |
hinges | +0.098 |
bleed | +0.098 |
bru | +0.096 |
aid | +0.094 |
onna | +0.094 |
Neg logit contributions | |
ang | -0.073 |
ee | -0.060 |
ug | -0.058 |
irling | -0.057 |
ash | -0.056 |
Token | said |
Token ID | 531 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
ang | +0.006 |
irling | +0.005 |
ee | +0.005 |
irl | +0.005 |
ug | +0.005 |
Neg logit contributions | |
hinges | -0.005 |
bru | -0.005 |
bleed | -0.005 |
injuries | -0.005 |
department | -0.005 |
Token | sorry |
Token ID | 7926 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
hinges | +0.108 |
keeping | +0.107 |
bleed | +0.103 |
ranger | +0.102 |
aid | +0.102 |
Neg logit contributions | |
ang | -0.085 |
irling | -0.084 |
irl | -0.077 |
ug | -0.075 |
ee | -0.071 |
Token | to |
Token ID | 284 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
bru | +0.122 |
keeping | +0.120 |
bleed | +0.120 |
distress | +0.120 |
injuries | +0.119 |
Neg logit contributions | |
ing | -0.083 |
ang | -0.077 |
ash | -0.074 |
at | -0.071 |
ee | -0.068 |
Token | his |
Token ID | 465 |
Feature activation | -5.10e-02 |
Pos logit contributions | |
ranger | +0.070 |
bleed | +0.070 |
hinges | +0.069 |
bru | +0.068 |
sob | +0.067 |
Neg logit contributions | |
ang | -0.079 |
irling | -0.071 |
ee | -0.068 |
irl | -0.068 |
ug | -0.065 |
Token | sister |
Token ID | 6621 |
Feature activation | -9.83e-02 |
Pos logit contributions | |
bleed | +0.097 |
keeping | +0.094 |
recovered | +0.094 |
bru | +0.093 |
onna | +0.093 |
Neg logit contributions | |
ang | -0.082 |
irl | -0.073 |
irling | -0.071 |
ee | -0.069 |
ading | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
bleed | +0.190 |
hinges | +0.184 |
bru | +0.184 |
onna | +0.178 |
injuries | +0.167 |
Neg logit contributions | |
ang | -0.142 |
ash | -0.126 |
these | -0.124 |
aw | -0.119 |
irling | -0.118 |
Token | They |
Token ID | 1119 |
Feature activation | -2.19e-04 |
Pos logit contributions | |
bleed | +0.025 |
injuries | +0.025 |
bru | +0.024 |
hinges | +0.023 |
ache | +0.023 |
Neg logit contributions | |
ang | -0.031 |
irling | -0.026 |
ee | -0.026 |
ug | -0.025 |
ash | -0.025 |
Token | started |
Token ID | 2067 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
bleed | +0.000 |
ranger | +0.000 |
bru | +0.000 |
hinges | +0.000 |
aid | +0.000 |
Neg logit contributions | |
ang | -0.000 |
irling | -0.000 |
ee | -0.000 |
irl | -0.000 |
ug | -0.000 |
Token | to |
Token ID | 284 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
hinges | +0.024 |
ranger | +0.023 |
bleed | +0.023 |
bud | +0.022 |
ache | +0.021 |
Neg logit contributions | |
ang | -0.045 |
irling | -0.044 |
irl | -0.041 |
ee | -0.040 |
ading | -0.039 |
Token | play |
Token ID | 711 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
ranger | +0.069 |
edly | +0.069 |
hinges | +0.067 |
onna | +0.067 |
department | +0.066 |
Neg logit contributions | |
irl | -0.083 |
irling | -0.081 |
ang | -0.079 |
ee | -0.076 |
sorts | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
hinges | +0.061 |
keeping | +0.061 |
ranger | +0.061 |
severe | +0.061 |
aid | +0.060 |
Neg logit contributions | |
ang | -0.054 |
ee | -0.045 |
aw | -0.045 |
irling | -0.045 |
ash | -0.044 |
Token | She |
Token ID | 1375 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
bleed | +0.048 |
department | +0.046 |
injuries | +0.046 |
recovered | +0.046 |
ranger | +0.045 |
Neg logit contributions | |
ang | -0.064 |
irling | -0.057 |
aw | -0.057 |
ee | -0.056 |
irl | -0.054 |
Token | sn |
Token ID | 3013 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
ranger | +0.073 |
bleed | +0.073 |
bru | +0.072 |
sob | +0.071 |
hinges | +0.071 |
Neg logit contributions | |
ang | -0.094 |
irling | -0.085 |
ee | -0.082 |
irl | -0.079 |
ug | -0.078 |
Token | atches |
Token ID | 20981 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
ang | +0.084 |
irling | +0.074 |
ee | +0.071 |
irl | +0.068 |
ug | +0.065 |
Neg logit contributions | |
ranger | -0.104 |
bleed | -0.104 |
hinges | -0.102 |
bru | -0.102 |
sob | -0.100 |
Token | her |
Token ID | 607 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
ang | +0.034 |
irling | +0.032 |
irl | +0.029 |
ee | +0.029 |
ug | +0.028 |
Neg logit contributions | |
bleed | -0.027 |
ranger | -0.027 |
bru | -0.027 |
sob | -0.027 |
hinges | -0.026 |
Token | journal |
Token ID | 3989 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
ranger | +0.097 |
bleed | +0.092 |
bru | +0.091 |
department | +0.089 |
onna | +0.089 |
Neg logit contributions | |
ang | -0.115 |
ee | -0.107 |
irl | -0.104 |
irling | -0.103 |
angles | -0.096 |
Token | from |
Token ID | 422 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
bru | +0.202 |
mercy | +0.197 |
injuries | +0.195 |
department | +0.194 |
Something | +0.193 |
Neg logit contributions | |
ang | -0.129 |
ee | -0.114 |
irl | -0.102 |
at | -0.099 |
irling | -0.097 |
Token | Tom |
Token ID | 4186 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
ranger | +0.025 |
bleed | +0.025 |
hinges | +0.025 |
bru | +0.025 |
department | +0.024 |
Neg logit contributions | |
ang | -0.027 |
irling | -0.024 |
ee | -0.023 |
irl | -0.023 |
ug | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
ranger | +0.113 |
bleed | +0.111 |
bru | +0.110 |
hinges | +0.110 |
department | +0.108 |
Neg logit contributions | |
ang | -0.109 |
ee | -0.095 |
irling | -0.094 |
irl | -0.088 |
ug | -0.087 |
Token | pushes |
Token ID | 20070 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
ranger | +0.039 |
bleed | +0.039 |
bru | +0.038 |
hinges | +0.038 |
sob | +0.037 |
Neg logit contributions | |
ang | -0.049 |
irling | -0.045 |
ee | -0.043 |
irl | -0.042 |
ug | -0.040 |
Token | him |
Token ID | 683 |
Feature activation | -8.92e-02 |
Pos logit contributions | |
ang | +0.032 |
irling | +0.029 |
ee | +0.028 |
irl | +0.027 |
ug | +0.026 |
Neg logit contributions | |
ranger | -0.027 |
bleed | -0.026 |
bru | -0.026 |
hinges | -0.026 |
sob | -0.025 |
Token | my |
Token ID | 1820 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
ranger | +0.033 |
bru | +0.031 |
hinges | +0.031 |
bleed | +0.031 |
Something | +0.030 |
Neg logit contributions | |
ang | -0.026 |
ee | -0.023 |
irling | -0.022 |
ug | -0.020 |
irl | -0.020 |
Token | 's |
Token ID | 338 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
ranger | +0.047 |
bleed | +0.046 |
bru | +0.045 |
hinges | +0.045 |
sob | +0.044 |
Neg logit contributions | |
ang | -0.060 |
irling | -0.054 |
ee | -0.053 |
irl | -0.051 |
ug | -0.049 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
ranger | +0.042 |
bleed | +0.042 |
bru | +0.041 |
hinges | +0.040 |
department | +0.040 |
Neg logit contributions | |
ang | -0.058 |
irling | -0.053 |
ee | -0.052 |
irl | -0.050 |
ug | -0.049 |
Token | shouted |
Token ID | 17293 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
ranger | +0.052 |
bleed | +0.051 |
bru | +0.050 |
hinges | +0.050 |
sob | +0.048 |
Neg logit contributions | |
ang | -0.067 |
irling | -0.060 |
ee | -0.059 |
irl | -0.056 |
ug | -0.055 |
Token | at |
Token ID | 379 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
ranger | +0.076 |
bleed | +0.075 |
hinges | +0.074 |
bru | +0.074 |
department | +0.071 |
Neg logit contributions | |
ang | -0.058 |
irling | -0.052 |
ee | -0.049 |
irl | -0.047 |
ug | -0.046 |
Token | him |
Token ID | 683 |
Feature activation | -9.79e-02 |
Pos logit contributions | |
ranger | +0.041 |
bleed | +0.040 |
bru | +0.040 |
hinges | +0.040 |
sob | +0.039 |
Neg logit contributions | |
ang | -0.041 |
irling | -0.037 |
ee | -0.037 |
irl | -0.035 |
ug | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
bru | +0.168 |
ranger | +0.156 |
hinges | +0.156 |
bleed | +0.155 |
Something | +0.152 |
Neg logit contributions | |
ang | -0.167 |
ee | -0.148 |
ug | -0.148 |
irling | -0.147 |
ash | -0.146 |
Token | turn |
Token ID | 1210 |
Feature activation | +1.71e-03 |
Pos logit contributions | |
ranger | +0.109 |
department | +0.109 |
keeper | +0.107 |
hinges | +0.105 |
onna | +0.103 |
Neg logit contributions | |
ang | -0.108 |
irling | -0.106 |
irl | -0.104 |
ee | -0.102 |
ading | -0.098 |
Token | it |
Token ID | 340 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.003 |
irl | +0.003 |
ug | +0.003 |
Neg logit contributions | |
ranger | -0.003 |
hinges | -0.003 |
bru | -0.003 |
bleed | -0.003 |
department | -0.003 |
Token | off |
Token ID | 572 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
ranger | +0.100 |
bru | +0.100 |
Something | +0.095 |
mercy | +0.094 |
sob | +0.094 |
Neg logit contributions | |
ang | -0.071 |
ug | -0.063 |
ee | -0.063 |
irling | -0.061 |
ash | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
sob | +0.052 |
bleed | +0.052 |
ranger | +0.052 |
bru | +0.051 |
hinges | +0.050 |
Neg logit contributions | |
ang | -0.065 |
ee | -0.059 |
irling | -0.057 |
ug | -0.056 |
irl | -0.056 |
Token |
|
Token ID | 198 |
Feature activation | -8.33e-03 |
Pos logit contributions | |
department | +0.065 |
bud | +0.061 |
recovered | +0.061 |
mercy | +0.060 |
injuries | +0.060 |
Neg logit contributions | |
ang | -0.052 |
ug | -0.041 |
ee | -0.040 |
aw | -0.039 |
irling | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -1.06e-02 |
Pos logit contributions | |
bleed | +0.014 |
department | +0.013 |
hinges | +0.013 |
ranger | +0.013 |
bru | +0.013 |
Neg logit contributions | |
ang | -0.017 |
ee | -0.015 |
irling | -0.015 |
irl | -0.014 |
ug | -0.014 |
Token | His |
Token ID | 6653 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
bleed | +0.018 |
bru | +0.018 |
department | +0.018 |
hinges | +0.017 |
ranger | +0.017 |
Neg logit contributions | |
ang | -0.021 |
irling | -0.019 |
ee | -0.018 |
irl | -0.017 |
ug | -0.017 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
keeping | +0.101 |
recovered | +0.097 |
Something | +0.097 |
department | +0.094 |
onna | +0.093 |
Neg logit contributions | |
ang | -0.084 |
irl | -0.077 |
irling | -0.072 |
ee | -0.071 |
utes | -0.061 |
Token | heard |
Token ID | 2982 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
onna | +0.046 |
bleed | +0.046 |
hinges | +0.046 |
ranger | +0.046 |
bru | +0.043 |
Neg logit contributions | |
ang | -0.066 |
ee | -0.060 |
irling | -0.058 |
ug | -0.056 |
aw | -0.056 |
Token | him |
Token ID | 683 |
Feature activation | -9.78e-02 |
Pos logit contributions | |
hinges | +0.043 |
ranger | +0.042 |
bleed | +0.042 |
department | +0.042 |
bru | +0.042 |
Neg logit contributions | |
ang | -0.032 |
ee | -0.028 |
irling | -0.027 |
aw | -0.026 |
ug | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -4.48e-02 |
Pos logit contributions | |
department | +0.184 |
ranger | +0.184 |
hinges | +0.183 |
onna | +0.180 |
bru | +0.180 |
Neg logit contributions | |
ang | -0.165 |
ee | -0.140 |
irling | -0.136 |
ug | -0.130 |
irl | -0.124 |
Token | came |
Token ID | 1625 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
hinges | +0.073 |
bleed | +0.072 |
bru | +0.071 |
onna | +0.071 |
ranger | +0.071 |
Neg logit contributions | |
ang | -0.088 |
ee | -0.076 |
irling | -0.075 |
ug | -0.072 |
irl | -0.068 |
Token | to |
Token ID | 284 |
Feature activation | -5.40e-02 |
Pos logit contributions | |
bleed | +0.025 |
bru | +0.025 |
ranger | +0.024 |
ache | +0.023 |
Something | +0.023 |
Neg logit contributions | |
ang | -0.051 |
irling | -0.046 |
ee | -0.045 |
ug | -0.043 |
ading | -0.043 |
Token | help |
Token ID | 1037 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
ache | +0.127 |
hinges | +0.121 |
edly | +0.120 |
bleed | +0.118 |
department | +0.116 |
Neg logit contributions | |
ang | -0.056 |
ee | -0.053 |
these | -0.051 |
us | -0.050 |
your | -0.043 |
Token | him |
Token ID | 683 |
Feature activation | -9.73e-02 |
Pos logit contributions | |
department | +0.045 |
bru | +0.044 |
hinges | +0.043 |
bleed | +0.043 |
ache | +0.043 |
Neg logit contributions | |
ang | -0.032 |
ee | -0.030 |
ug | -0.028 |
ie | -0.027 |
us | -0.026 |
Token | him |
Token ID | 683 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
hinges | +0.041 |
ranger | +0.040 |
department | +0.040 |
bleed | +0.039 |
bru | +0.039 |
Neg logit contributions | |
ang | -0.029 |
ee | -0.026 |
irling | -0.025 |
ug | -0.023 |
irl | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
department | +0.186 |
ranger | +0.184 |
hinges | +0.180 |
bru | +0.177 |
onna | +0.176 |
Neg logit contributions | |
ang | -0.175 |
ee | -0.151 |
irling | -0.145 |
ug | -0.141 |
irl | -0.133 |
Token | came |
Token ID | 1625 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
hinges | +0.073 |
bleed | +0.072 |
onna | +0.071 |
bru | +0.071 |
ranger | +0.070 |
Neg logit contributions | |
ang | -0.083 |
ee | -0.071 |
irling | -0.070 |
ug | -0.068 |
irl | -0.063 |
Token | to |
Token ID | 284 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
bleed | +0.026 |
bru | +0.025 |
ranger | +0.025 |
ache | +0.024 |
Something | +0.023 |
Neg logit contributions | |
ang | -0.053 |
irling | -0.047 |
ee | -0.046 |
ug | -0.045 |
ading | -0.044 |
Token | help |
Token ID | 1037 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
ache | +0.133 |
hinges | +0.128 |
edly | +0.127 |
department | +0.123 |
onna | +0.123 |
Neg logit contributions | |
ang | -0.054 |
ee | -0.049 |
these | -0.048 |
us | -0.044 |
rous | -0.040 |
Token | him |
Token ID | 683 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
department | +0.039 |
hinges | +0.038 |
ache | +0.038 |
bru | +0.037 |
ranger | +0.037 |
Neg logit contributions | |
ang | -0.031 |
ee | -0.030 |
ug | -0.028 |
ie | -0.026 |
ading | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
department | +0.180 |
ranger | +0.179 |
injuries | +0.175 |
bru | +0.175 |
Something | +0.168 |
Neg logit contributions | |
ang | -0.159 |
irling | -0.135 |
irl | -0.133 |
ading | -0.132 |
ee | -0.130 |
Token | She |
Token ID | 1375 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
injuries | +0.058 |
ache | +0.055 |
mercy | +0.055 |
bleed | +0.054 |
recovered | +0.054 |
Neg logit contributions | |
ang | -0.052 |
aw | -0.041 |
ee | -0.038 |
ug | -0.038 |
irling | -0.037 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
bleed | +0.054 |
hinges | +0.052 |
bru | +0.050 |
onna | +0.049 |
ranger | +0.049 |
Neg logit contributions | |
ang | -0.056 |
irling | -0.047 |
ee | -0.047 |
irl | -0.043 |
ug | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | -1.72e-02 |
Pos logit contributions | |
ang | +0.018 |
irling | +0.015 |
ee | +0.015 |
irl | +0.014 |
ug | +0.013 |
Neg logit contributions | |
ranger | -0.026 |
bleed | -0.026 |
hinges | -0.026 |
bru | -0.026 |
sob | -0.025 |
Token | broken |
Token ID | 5445 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
hinges | +0.027 |
onna | +0.026 |
ranger | +0.026 |
department | +0.026 |
bleed | +0.026 |
Neg logit contributions | |
ang | -0.035 |
ee | -0.030 |
irling | -0.030 |
irl | -0.029 |
aw | -0.027 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
ranger | +0.047 |
sob | +0.046 |
bleed | +0.046 |
bru | +0.045 |
hinges | +0.044 |
Neg logit contributions | |
ang | -0.057 |
irling | -0.052 |
ee | -0.050 |
irl | -0.048 |
ug | -0.047 |
Token | said |
Token ID | 531 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
bleed | +0.022 |
injuries | +0.022 |
bru | +0.021 |
hinges | +0.021 |
department | +0.021 |
Neg logit contributions | |
ang | -0.019 |
ee | -0.015 |
irling | -0.015 |
ug | -0.014 |
irl | -0.014 |
Token | , |
Token ID | 11 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
injuries | +0.103 |
bru | +0.103 |
hinges | +0.101 |
bleed | +0.100 |
department | +0.099 |
Neg logit contributions | |
ang | -0.076 |
ug | -0.067 |
irling | -0.064 |
ee | -0.061 |
ap | -0.059 |
Token | giving |
Token ID | 3501 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
ranger | +0.086 |
bleed | +0.086 |
bru | +0.084 |
hinges | +0.084 |
department | +0.082 |
Neg logit contributions | |
ang | -0.088 |
irling | -0.079 |
ee | -0.074 |
irl | -0.073 |
ug | -0.070 |
Token | his |
Token ID | 465 |
Feature activation | -5.77e-02 |
Pos logit contributions | |
ang | +0.070 |
irling | +0.062 |
ee | +0.059 |
irl | +0.057 |
ug | +0.056 |
Neg logit contributions | |
ranger | -0.071 |
bleed | -0.071 |
bru | -0.069 |
sob | -0.069 |
hinges | -0.069 |
Token | sister |
Token ID | 6621 |
Feature activation | -9.76e-02 |
Pos logit contributions | |
bleed | +0.119 |
ranger | +0.112 |
hinges | +0.111 |
bru | +0.110 |
sob | +0.110 |
Neg logit contributions | |
ang | -0.090 |
ee | -0.081 |
il | -0.073 |
irl | -0.073 |
irling | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
bleed | +0.162 |
ranger | +0.161 |
bru | +0.159 |
hinges | +0.158 |
department | +0.155 |
Neg logit contributions | |
ang | -0.193 |
irling | -0.172 |
ee | -0.170 |
irl | -0.162 |
ug | -0.156 |
Token | fruit |
Token ID | 8234 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
keeping | +0.119 |
ranger | +0.116 |
bleed | +0.115 |
hinges | +0.114 |
injuries | +0.112 |
Neg logit contributions | |
ang | -0.102 |
ee | -0.100 |
irl | -0.089 |
irling | -0.083 |
sorts | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
bru | +0.052 |
bleed | +0.051 |
sob | +0.050 |
injuries | +0.049 |
Something | +0.048 |
Neg logit contributions | |
ang | -0.061 |
ee | -0.053 |
ash | -0.048 |
aw | -0.048 |
il | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | -8.18e-03 |
Pos logit contributions | |
bleed | +0.033 |
hinges | +0.031 |
injuries | +0.031 |
mercy | +0.030 |
bru | +0.030 |
Neg logit contributions | |
ang | -0.046 |
irling | -0.041 |
ee | -0.040 |
irl | -0.039 |
ug | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -5.20e-03 |
Pos logit contributions | |
ranger | +0.009 |
bleed | +0.009 |
bru | +0.009 |
hinges | +0.009 |
sob | +0.009 |
Neg logit contributions | |
ang | -0.021 |
irling | -0.019 |
ee | -0.019 |
irl | -0.018 |
ug | -0.018 |
Token | cries |
Token ID | 24691 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
ranger | +0.065 |
hinges | +0.063 |
bleed | +0.062 |
bru | +0.061 |
department | +0.061 |
Neg logit contributions | |
ang | -0.100 |
irling | -0.090 |
ee | -0.089 |
irl | -0.087 |
ug | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
ranger | +0.053 |
keeping | +0.052 |
hinges | +0.052 |
onna | +0.052 |
bru | +0.052 |
Neg logit contributions | |
ang | -0.044 |
ug | -0.041 |
ee | -0.037 |
aw | -0.035 |
ing | -0.034 |
Token | His |
Token ID | 2399 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
bleed | +0.046 |
hinges | +0.044 |
recovered | +0.044 |
injuries | +0.043 |
ache | +0.043 |
Neg logit contributions | |
ang | -0.050 |
ug | -0.042 |
ee | -0.042 |
irling | -0.039 |
irl | -0.038 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
recovered | +0.083 |
onna | +0.082 |
ache | +0.080 |
bleed | +0.080 |
department | +0.077 |
Neg logit contributions | |
ang | -0.089 |
ee | -0.083 |
irl | -0.081 |
irling | -0.077 |
ug | -0.075 |
Token | hears |
Token ID | 23180 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
hinges | +0.069 |
ranger | +0.068 |
bleed | +0.066 |
bru | +0.065 |
onna | +0.064 |
Neg logit contributions | |
ang | -0.097 |
ee | -0.087 |
irling | -0.085 |
irl | -0.080 |
ug | -0.078 |
Token | him |
Token ID | 683 |
Feature activation | -9.71e-02 |
Pos logit contributions | |
hinges | +0.050 |
bru | +0.050 |
bleed | +0.050 |
ranger | +0.050 |
department | +0.048 |
Neg logit contributions | |
ang | -0.029 |
irling | -0.025 |
ee | -0.025 |
ug | -0.023 |
irl | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -5.26e-02 |
Pos logit contributions | |
hinges | +0.171 |
department | +0.170 |
ranger | +0.169 |
bru | +0.167 |
Something | +0.165 |
Neg logit contributions | |
ang | -0.175 |
ee | -0.154 |
irling | -0.145 |
ug | -0.142 |
ash | -0.136 |
Token | comes |
Token ID | 2058 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
hinges | +0.083 |
ranger | +0.082 |
bru | +0.082 |
bleed | +0.081 |
injuries | +0.079 |
Neg logit contributions | |
ang | -0.108 |
irling | -0.095 |
ee | -0.093 |
irl | -0.090 |
ug | -0.086 |
Token | to |
Token ID | 284 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
bru | +0.039 |
bleed | +0.039 |
ache | +0.038 |
ranger | +0.037 |
Something | +0.035 |
Neg logit contributions | |
ang | -0.075 |
ee | -0.067 |
irling | -0.066 |
ug | -0.066 |
ading | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
ache | +0.118 |
hinges | +0.110 |
edly | +0.109 |
bleed | +0.107 |
department | +0.106 |
Neg logit contributions | |
ang | -0.059 |
ee | -0.055 |
these | -0.052 |
us | -0.050 |
irl | -0.046 |
Token | sand |
Token ID | 6450 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
bleed | +0.034 |
sob | +0.032 |
recovered | +0.032 |
onna | +0.032 |
ache | +0.031 |
Neg logit contributions | |
ang | -0.025 |
ee | -0.022 |
irling | -0.021 |
ading | -0.019 |
aw | -0.019 |
Token | screamed |
Token ID | 25421 |
Feature activation | -4.94e-02 |
Pos logit contributions | |
hinges | +0.043 |
department | +0.040 |
bleed | +0.040 |
onna | +0.040 |
injuries | +0.039 |
Neg logit contributions | |
ang | -0.038 |
ee | -0.034 |
irling | -0.033 |
aw | -0.031 |
irl | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
ranger | +0.090 |
hinges | +0.087 |
bru | +0.084 |
department | +0.084 |
bleed | +0.083 |
Neg logit contributions | |
ang | -0.088 |
irling | -0.081 |
ee | -0.076 |
ug | -0.075 |
irl | -0.074 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
hinges | +0.036 |
department | +0.036 |
ranger | +0.036 |
aid | +0.034 |
bru | +0.034 |
Neg logit contributions | |
ang | -0.052 |
irling | -0.047 |
ee | -0.046 |
irl | -0.044 |
ug | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
bru | +0.033 |
onna | +0.033 |
Something | +0.032 |
sob | +0.032 |
ache | +0.032 |
Neg logit contributions | |
ang | -0.027 |
ug | -0.022 |
around | -0.021 |
aw | -0.021 |
ee | -0.021 |
Token | help |
Token ID | 1037 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
ache | +0.114 |
hinges | +0.114 |
department | +0.112 |
ranger | +0.109 |
bru | +0.108 |
Neg logit contributions | |
ang | -0.092 |
irl | -0.085 |
ee | -0.084 |
irling | -0.082 |
ading | -0.077 |
Token | him |
Token ID | 683 |
Feature activation | -9.71e-02 |
Pos logit contributions | |
hinges | +0.042 |
department | +0.042 |
bru | +0.039 |
ache | +0.039 |
bleed | +0.039 |
Neg logit contributions | |
ang | -0.033 |
ee | -0.032 |
irl | -0.031 |
irling | -0.029 |
ug | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
bru | +0.172 |
department | +0.171 |
hinges | +0.168 |
ranger | +0.166 |
injuries | +0.164 |
Neg logit contributions | |
ang | -0.173 |
irling | -0.148 |
irl | -0.147 |
ee | -0.142 |
ading | -0.140 |
Token |
|
Token ID | 198 |
Feature activation | -6.99e-03 |
Pos logit contributions | |
hinges | +0.038 |
bleed | +0.038 |
injuries | +0.037 |
ache | +0.037 |
mercy | +0.037 |
Neg logit contributions | |
ang | -0.041 |
ee | -0.035 |
irling | -0.033 |
irl | -0.032 |
ug | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -9.18e-03 |
Pos logit contributions | |
bleed | +0.010 |
hinges | +0.010 |
bru | +0.010 |
ranger | +0.010 |
department | +0.010 |
Neg logit contributions | |
ang | -0.015 |
irling | -0.014 |
ee | -0.014 |
irl | -0.013 |
ug | -0.013 |
Token | A |
Token ID | 32 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
ranger | +0.014 |
bleed | +0.014 |
sob | +0.014 |
hinges | +0.013 |
bru | +0.013 |
Neg logit contributions | |
ang | -0.020 |
irling | -0.017 |
ee | -0.017 |
ash | -0.016 |
ug | -0.016 |
Token | man |
Token ID | 582 |
Feature activation | -1.14e-02 |
Pos logit contributions | |
bleed | +0.040 |
hinges | +0.037 |
bru | +0.037 |
sob | +0.036 |
keeping | +0.035 |
Neg logit contributions | |
ang | -0.054 |
ee | -0.046 |
irling | -0.046 |
irl | -0.044 |
ug | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -2.24e-02 |
Pos logit contributions | |
keeping | +0.107 |
recovered | +0.103 |
injuries | +0.101 |
bleed | +0.100 |
mercy | +0.098 |
Neg logit contributions | |
ang | -0.044 |
ug | -0.029 |
everyone | -0.028 |
at | -0.028 |
ee | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
department | +0.042 |
bleed | +0.041 |
hinges | +0.041 |
ranger | +0.040 |
bru | +0.039 |
Neg logit contributions | |
ang | -0.041 |
ee | -0.035 |
irling | -0.035 |
irl | -0.035 |
ug | -0.034 |
Token | His |
Token ID | 6653 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
bleed | +0.029 |
bru | +0.029 |
department | +0.028 |
hinges | +0.028 |
onna | +0.027 |
Neg logit contributions | |
ang | -0.030 |
irling | -0.026 |
ee | -0.025 |
irl | -0.024 |
ug | -0.024 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
bleed | +0.083 |
keeping | +0.082 |
recovered | +0.081 |
onna | +0.078 |
ache | +0.076 |
Neg logit contributions | |
ang | -0.080 |
irl | -0.072 |
ee | -0.072 |
irling | -0.068 |
utes | -0.061 |
Token | hears |
Token ID | 23180 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
ranger | +0.045 |
hinges | +0.044 |
bleed | +0.044 |
keeping | +0.042 |
onna | +0.042 |
Neg logit contributions | |
ang | -0.078 |
irling | -0.073 |
ee | -0.072 |
irl | -0.069 |
aw | -0.066 |
Token | him |
Token ID | 683 |
Feature activation | -9.67e-02 |
Pos logit contributions | |
hinges | +0.024 |
bleed | +0.024 |
bru | +0.024 |
ranger | +0.024 |
department | +0.024 |
Neg logit contributions | |
ang | -0.018 |
ee | -0.017 |
irling | -0.016 |
ug | -0.016 |
irl | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | -4.79e-02 |
Pos logit contributions | |
bru | +0.179 |
department | +0.179 |
ranger | +0.179 |
hinges | +0.176 |
injuries | +0.175 |
Neg logit contributions | |
ang | -0.161 |
ee | -0.140 |
irling | -0.131 |
ug | -0.128 |
irl | -0.120 |
Token | comes |
Token ID | 2058 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
hinges | +0.078 |
ranger | +0.077 |
bru | +0.076 |
bleed | +0.076 |
injuries | +0.074 |
Neg logit contributions | |
ang | -0.095 |
irling | -0.084 |
ee | -0.083 |
irl | -0.080 |
ug | -0.075 |
Token | to |
Token ID | 284 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
bleed | +0.035 |
bru | +0.034 |
ache | +0.034 |
ranger | +0.032 |
cough | +0.030 |
Neg logit contributions | |
ang | -0.060 |
ee | -0.053 |
irling | -0.053 |
ug | -0.052 |
ading | -0.051 |
Token | him |
Token ID | 683 |
Feature activation | -8.30e-02 |
Pos logit contributions | |
ache | +0.116 |
hinges | +0.108 |
bleed | +0.107 |
department | +0.106 |
edly | +0.106 |
Neg logit contributions | |
ang | -0.037 |
ee | -0.036 |
these | -0.032 |
us | -0.030 |
sorts | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
bru | +0.151 |
department | +0.145 |
injuries | +0.143 |
ache | +0.141 |
recovered | +0.140 |
Neg logit contributions | |
ang | -0.149 |
irling | -0.117 |
ee | -0.116 |
irl | -0.111 |
sorts | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | +1.70e-03 |
Pos logit contributions | |
ranger | +0.049 |
hinges | +0.049 |
bleed | +0.048 |
bru | +0.047 |
injuries | +0.046 |
Neg logit contributions | |
ang | -0.052 |
irling | -0.046 |
ee | -0.045 |
irl | -0.042 |
ug | -0.040 |
Token | Sam |
Token ID | 3409 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
ang | +0.003 |
ee | +0.002 |
irling | +0.002 |
irl | +0.002 |
ug | +0.002 |
Neg logit contributions | |
bleed | -0.003 |
ranger | -0.003 |
hinges | -0.003 |
bru | -0.003 |
department | -0.003 |
Token | run |
Token ID | 1057 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
ranger | +0.045 |
hinges | +0.044 |
department | +0.043 |
onna | +0.041 |
aid | +0.041 |
Neg logit contributions | |
ang | -0.053 |
ee | -0.050 |
irling | -0.047 |
irl | -0.045 |
ug | -0.043 |
Token | to |
Token ID | 284 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
bru | +0.039 |
onna | +0.038 |
ache | +0.037 |
Something | +0.037 |
sob | +0.036 |
Neg logit contributions | |
ang | -0.042 |
ee | -0.036 |
ug | -0.034 |
aw | -0.034 |
around | -0.032 |
Token | help |
Token ID | 1037 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
ache | +0.146 |
hinges | +0.141 |
bleed | +0.140 |
ranger | +0.138 |
department | +0.137 |
Neg logit contributions | |
ang | -0.070 |
ee | -0.064 |
irling | -0.060 |
irl | -0.059 |
ading | -0.056 |
Token | him |
Token ID | 683 |
Feature activation | -9.66e-02 |
Pos logit contributions | |
ranger | +0.052 |
bleed | +0.052 |
bru | +0.051 |
hinges | +0.050 |
department | +0.049 |
Neg logit contributions | |
ang | -0.063 |
ee | -0.059 |
irling | -0.058 |
irl | -0.056 |
ading | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
ranger | +0.157 |
bru | +0.156 |
injuries | +0.150 |
department | +0.150 |
hinges | +0.145 |
Neg logit contributions | |
ang | -0.191 |
irling | -0.169 |
ee | -0.166 |
irl | -0.164 |
ading | -0.161 |
Token | They |
Token ID | 1119 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
ache | +0.069 |
mercy | +0.068 |
bleed | +0.068 |
injuries | +0.067 |
recovered | +0.066 |
Neg logit contributions | |
ang | -0.046 |
ee | -0.039 |
irling | -0.038 |
irl | -0.035 |
ading | -0.034 |
Token | are |
Token ID | 389 |
Feature activation | +1.94e-03 |
Pos logit contributions | |
ranger | +0.045 |
hinges | +0.045 |
department | +0.043 |
bru | +0.042 |
sob | +0.041 |
Neg logit contributions | |
ang | -0.061 |
ee | -0.056 |
irl | -0.055 |
irling | -0.054 |
ug | -0.049 |
Token | sorry |
Token ID | 7926 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
ang | +0.004 |
irling | +0.004 |
ee | +0.004 |
irl | +0.003 |
ug | +0.003 |
Neg logit contributions | |
ranger | -0.003 |
bleed | -0.003 |
bru | -0.003 |
hinges | -0.003 |
sob | -0.003 |
Token | and |
Token ID | 290 |
Feature activation | -3.59e-02 |
Pos logit contributions | |
ranger | +0.073 |
bleed | +0.073 |
bru | +0.072 |
hinges | +0.071 |
aid | +0.068 |
Neg logit contributions | |
ang | -0.089 |
irling | -0.079 |
ee | -0.079 |
ug | -0.075 |
irl | -0.074 |
Token | his |
Token ID | 465 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
department | +0.042 |
onna | +0.042 |
hinges | +0.042 |
ranger | +0.042 |
bleed | +0.041 |
Neg logit contributions | |
ang | -0.046 |
irling | -0.041 |
irl | -0.040 |
ee | -0.040 |
ug | -0.038 |
Token | mum |
Token ID | 25682 |
Feature activation | -3.83e-02 |
Pos logit contributions | |
Something | +0.077 |
onna | +0.077 |
ranger | +0.076 |
recovered | +0.075 |
department | +0.075 |
Neg logit contributions | |
ang | -0.078 |
irl | -0.072 |
irling | -0.070 |
ee | -0.070 |
ug | -0.064 |
Token | came |
Token ID | 1625 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
bleed | +0.062 |
hinges | +0.062 |
ranger | +0.061 |
bru | +0.060 |
onna | +0.058 |
Neg logit contributions | |
ang | -0.077 |
irling | -0.068 |
ee | -0.068 |
ug | -0.065 |
irl | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
ranger | +0.035 |
bleed | +0.035 |
bru | +0.034 |
hinges | +0.033 |
department | +0.033 |
Neg logit contributions | |
ang | -0.037 |
irling | -0.033 |
ee | -0.031 |
irl | -0.030 |
ug | -0.030 |
Token | help |
Token ID | 1037 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
hinges | +0.097 |
ache | +0.097 |
onna | +0.094 |
edly | +0.093 |
department | +0.093 |
Neg logit contributions | |
ee | -0.083 |
ang | -0.082 |
irl | -0.076 |
us | -0.076 |
these | -0.075 |
Token | him |
Token ID | 683 |
Feature activation | -9.60e-02 |
Pos logit contributions | |
ranger | +0.050 |
department | +0.049 |
bru | +0.048 |
hinges | +0.047 |
bleed | +0.047 |
Neg logit contributions | |
ang | -0.036 |
ee | -0.036 |
irling | -0.032 |
ug | -0.032 |
ading | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -3.62e-02 |
Pos logit contributions | |
ranger | +0.171 |
department | +0.162 |
injuries | +0.156 |
bru | +0.155 |
onna | +0.153 |
Neg logit contributions | |
ang | -0.167 |
irling | -0.155 |
irl | -0.153 |
ee | -0.149 |
ading | -0.147 |
Token | She |
Token ID | 1375 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
bleed | +0.057 |
injuries | +0.055 |
bru | +0.052 |
mercy | +0.052 |
recovered | +0.051 |
Neg logit contributions | |
ang | -0.070 |
irling | -0.064 |
ee | -0.062 |
sorts | -0.059 |
aw | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | +1.53e-03 |
Pos logit contributions | |
ranger | +0.034 |
bleed | +0.034 |
bru | +0.033 |
sob | +0.033 |
hinges | +0.033 |
Neg logit contributions | |
ang | -0.050 |
irling | -0.045 |
ee | -0.044 |
irl | -0.043 |
ug | -0.041 |
Token | very |
Token ID | 845 |
Feature activation | -7.09e-03 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.003 |
irl | +0.002 |
ug | +0.002 |
Neg logit contributions | |
ranger | -0.003 |
bleed | -0.003 |
bru | -0.003 |
hinges | -0.003 |
sob | -0.003 |
Token | kind |
Token ID | 1611 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
ranger | +0.011 |
hinges | +0.011 |
bleed | +0.011 |
bru | +0.011 |
department | +0.011 |
Neg logit contributions | |
ang | -0.015 |
irling | -0.014 |
ee | -0.013 |
irl | -0.013 |
ug | -0.012 |
Token | cow |
Token ID | 9875 |
Feature activation | +6.19e-03 |
Pos logit contributions | |
bleed | +0.037 |
ranger | +0.036 |
hinges | +0.036 |
bru | +0.035 |
sob | +0.035 |
Neg logit contributions | |
ang | -0.038 |
irling | -0.034 |
ee | -0.033 |
irl | -0.032 |
ug | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +7.00e-03 |
Pos logit contributions | |
ang | +0.011 |
ee | +0.010 |
irling | +0.010 |
ug | +0.009 |
irl | +0.009 |
Neg logit contributions | |
bleed | -0.011 |
hinges | -0.011 |
ranger | -0.011 |
bru | -0.011 |
sob | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | +6.60e-03 |
Pos logit contributions | |
ang | +0.017 |
irling | +0.015 |
ee | +0.015 |
irl | +0.015 |
ug | +0.014 |
Neg logit contributions | |
ranger | -0.009 |
bleed | -0.009 |
bru | -0.008 |
hinges | -0.008 |
sob | -0.008 |
Token | loved |
Token ID | 6151 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ang | +0.014 |
irling | +0.013 |
ee | +0.012 |
irl | +0.012 |
ug | +0.012 |
Neg logit contributions | |
ranger | -0.010 |
bleed | -0.010 |
sob | -0.010 |
bru | -0.010 |
hinges | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
ang | +0.028 |
irling | +0.026 |
ee | +0.025 |
irl | +0.025 |
ug | +0.024 |
Neg logit contributions | |
ranger | -0.014 |
bleed | -0.013 |
bru | -0.013 |
sob | -0.013 |
hinges | -0.013 |
Token | roam |
Token ID | 35563 |
Feature activation | +9.03e-02 |
Pos logit contributions | |
ranger | +0.025 |
bleed | +0.024 |
bru | +0.024 |
hinges | +0.024 |
sob | +0.023 |
Neg logit contributions | |
ang | -0.033 |
irling | -0.030 |
ee | -0.029 |
irl | -0.029 |
ug | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
ang | +0.156 |
irling | +0.153 |
irl | +0.136 |
ee | +0.136 |
ading | +0.127 |
Neg logit contributions | |
ranger | -0.172 |
bleed | -0.164 |
hinges | -0.158 |
sob | -0.156 |
keeper | -0.154 |
Token | fields |
Token ID | 7032 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
ang | +0.091 |
irling | +0.088 |
irl | +0.082 |
angles | +0.081 |
ug | +0.081 |
Neg logit contributions | |
ranger | -0.073 |
keeper | -0.066 |
bleed | -0.065 |
aid | -0.062 |
sob | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
ang | +0.070 |
irling | +0.064 |
ee | +0.062 |
irl | +0.061 |
ug | +0.058 |
Neg logit contributions | |
ranger | -0.048 |
bleed | -0.046 |
hinges | -0.045 |
bru | -0.045 |
sob | -0.044 |
Token | make |
Token ID | 787 |
Feature activation | +4.09e-02 |
Pos logit contributions | |
ang | +0.086 |
ee | +0.078 |
irling | +0.077 |
ug | +0.074 |
ash | +0.074 |
Neg logit contributions | |
ranger | -0.054 |
bleed | -0.053 |
bru | -0.051 |
sob | -0.050 |
hinges | -0.050 |
Token | friends |
Token ID | 2460 |
Feature activation | -4.35e-03 |
Pos logit contributions | |
ang | +0.082 |
irling | +0.071 |
ee | +0.069 |
irl | +0.065 |
ash | +0.065 |
Neg logit contributions | |
bleed | -0.072 |
sob | -0.068 |
ranger | -0.068 |
bru | -0.065 |
hinges | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
bleed | +0.034 |
ranger | +0.034 |
bru | +0.033 |
hinges | +0.033 |
sob | +0.032 |
Neg logit contributions | |
ang | -0.053 |
irling | -0.048 |
ee | -0.046 |
irl | -0.045 |
ug | -0.044 |
Token | said |
Token ID | 531 |
Feature activation | -2.60e-02 |
Pos logit contributions | |
ang | +0.050 |
irling | +0.045 |
ee | +0.044 |
irl | +0.043 |
ug | +0.042 |
Neg logit contributions | |
ranger | -0.034 |
bleed | -0.034 |
sob | -0.034 |
bru | -0.033 |
hinges | -0.031 |
Token | " |
Token ID | 366 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
bleed | +0.045 |
ranger | +0.045 |
hinges | +0.045 |
bru | +0.044 |
department | +0.042 |
Neg logit contributions | |
ang | -0.050 |
irling | -0.045 |
ee | -0.044 |
irl | -0.042 |
ug | -0.042 |
Token | Hello |
Token ID | 15496 |
Feature activation | +1.20e-03 |
Pos logit contributions | |
ang | +0.057 |
irling | +0.052 |
ee | +0.050 |
irl | +0.050 |
ug | +0.048 |
Neg logit contributions | |
ranger | -0.041 |
bleed | -0.041 |
bru | -0.040 |
sob | -0.039 |
hinges | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
ang | +0.002 |
irling | +0.002 |
irl | +0.002 |
ug | +0.002 |
ading | +0.002 |
Neg logit contributions | |
ranger | -0.002 |
department | -0.002 |
bru | -0.002 |
sob | -0.002 |
keeper | -0.002 |
Token | cater |
Token ID | 20825 |
Feature activation | +9.37e-02 |
Pos logit contributions | |
ang | +0.056 |
irling | +0.054 |
irl | +0.049 |
ug | +0.049 |
angles | +0.047 |
Neg logit contributions | |
ranger | -0.043 |
bru | -0.041 |
sob | -0.040 |
Something | -0.040 |
keeper | -0.038 |
Token | pillar |
Token ID | 41643 |
Feature activation | +4.01e-02 |
Pos logit contributions | |
angles | +0.188 |
irling | +0.187 |
irl | +0.186 |
ang | +0.182 |
ading | +0.182 |
Neg logit contributions | |
ranger | -0.116 |
ache | -0.115 |
onna | -0.115 |
bleed | -0.110 |
bru | -0.109 |
Token | ". |
Token ID | 1911 |
Feature activation | +3.06e-03 |
Pos logit contributions | |
ang | +0.084 |
irling | +0.077 |
irl | +0.073 |
ee | +0.073 |
ading | +0.071 |
Neg logit contributions | |
ranger | -0.062 |
bru | -0.058 |
bleed | -0.058 |
sob | -0.058 |
department | -0.056 |
Token | The |
Token ID | 383 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
ang | +0.006 |
irling | +0.005 |
ee | +0.005 |
irl | +0.005 |
ug | +0.005 |
Neg logit contributions | |
bleed | -0.005 |
hinges | -0.005 |
ranger | -0.005 |
bru | -0.005 |
department | -0.005 |
Token | cater |
Token ID | 20825 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
ang | +0.030 |
irling | +0.027 |
ee | +0.026 |
irl | +0.025 |
ug | +0.025 |
Neg logit contributions | |
ranger | -0.023 |
bleed | -0.023 |
bru | -0.022 |
sob | -0.022 |
hinges | -0.022 |
Token | pillar |
Token ID | 41643 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
ang | +0.169 |
irling | +0.166 |
irl | +0.162 |
angles | +0.161 |
ading | +0.158 |
Neg logit contributions | |
ranger | -0.129 |
bleed | -0.128 |
sob | -0.126 |
bru | -0.124 |
recovered | -0.119 |
Token | see |
Token ID | 766 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
ranger | +0.038 |
bleed | +0.038 |
hinges | +0.037 |
bru | +0.037 |
sob | +0.036 |
Neg logit contributions | |
ang | -0.067 |
irling | -0.059 |
ee | -0.058 |
ug | -0.055 |
irl | -0.055 |
Token | her |
Token ID | 607 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
ang | +0.039 |
irling | +0.036 |
ee | +0.033 |
irl | +0.033 |
ug | +0.032 |
Neg logit contributions | |
ranger | -0.033 |
bleed | -0.033 |
sob | -0.033 |
bru | -0.032 |
hinges | -0.031 |
Token | very |
Token ID | 845 |
Feature activation | -2.54e-03 |
Pos logit contributions | |
bleed | +0.040 |
bru | +0.039 |
ranger | +0.039 |
hinges | +0.039 |
sob | +0.037 |
Neg logit contributions | |
ang | -0.049 |
irling | -0.044 |
ee | -0.043 |
irl | -0.042 |
ading | -0.039 |
Token | first |
Token ID | 717 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
ranger | +0.004 |
bleed | +0.004 |
hinges | +0.004 |
sob | +0.004 |
bru | +0.004 |
Neg logit contributions | |
ang | -0.005 |
irling | -0.004 |
ee | -0.004 |
ug | -0.004 |
irl | -0.004 |
Token | flower |
Token ID | 15061 |
Feature activation | +6.38e-03 |
Pos logit contributions | |
bleed | +0.021 |
ranger | +0.021 |
bru | +0.020 |
hinges | +0.020 |
sob | +0.020 |
Neg logit contributions | |
ang | -0.026 |
irling | -0.024 |
ee | -0.023 |
irl | -0.022 |
ug | -0.022 |
Token | blo |
Token ID | 24924 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
ang | +0.013 |
irling | +0.012 |
ee | +0.012 |
irl | +0.012 |
ug | +0.011 |
Neg logit contributions | |
bru | -0.010 |
bleed | -0.010 |
ranger | -0.009 |
hinges | -0.009 |
sob | -0.009 |
Token | oming |
Token ID | 3383 |
Feature activation | +4.65e-02 |
Pos logit contributions | |
ang | +0.084 |
irling | +0.072 |
ee | +0.069 |
irl | +0.068 |
ug | +0.064 |
Neg logit contributions | |
onna | -0.219 |
ranger | -0.216 |
department | -0.215 |
bud | -0.215 |
bleed | -0.213 |
Token | in |
Token ID | 287 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
ang | +0.094 |
irling | +0.085 |
ee | +0.082 |
irl | +0.080 |
ug | +0.077 |
Neg logit contributions | |
bleed | -0.076 |
ranger | -0.075 |
bru | -0.075 |
hinges | -0.073 |
sob | -0.073 |
Token | her |
Token ID | 607 |
Feature activation | +6.05e-03 |
Pos logit contributions | |
ang | +0.059 |
irling | +0.052 |
ee | +0.050 |
irl | +0.049 |
ug | +0.046 |
Neg logit contributions | |
ranger | -0.067 |
bleed | -0.067 |
bru | -0.066 |
hinges | -0.065 |
sob | -0.065 |
Token | garden |
Token ID | 11376 |
Feature activation | -1.15e-03 |
Pos logit contributions | |
ang | +0.012 |
irling | +0.011 |
ee | +0.011 |
irl | +0.010 |
ug | +0.010 |
Neg logit contributions | |
ranger | -0.010 |
bru | -0.010 |
bleed | -0.010 |
hinges | -0.010 |
sob | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.27e-03 |
Pos logit contributions | |
bleed | +0.002 |
ranger | +0.001 |
bru | +0.001 |
hinges | +0.001 |
sob | +0.001 |
Neg logit contributions | |
ang | -0.003 |
irling | -0.002 |
ee | -0.002 |
irl | -0.002 |
ug | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +8.16e-03 |
Pos logit contributions | |
bleed | +0.026 |
hinges | +0.026 |
bru | +0.026 |
ranger | +0.025 |
sob | +0.024 |
Neg logit contributions | |
ang | -0.039 |
irling | -0.034 |
ee | -0.034 |
irl | -0.032 |
ug | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
ang | +0.016 |
irling | +0.014 |
ee | +0.014 |
irl | +0.014 |
ug | +0.013 |
Neg logit contributions | |
bleed | -0.014 |
ranger | -0.013 |
bru | -0.013 |
hinges | -0.013 |
sob | -0.013 |
Token | they |
Token ID | 484 |
Feature activation | +3.54e-02 |
Pos logit contributions | |
ang | +0.059 |
irling | +0.053 |
ee | +0.051 |
irl | +0.050 |
ug | +0.048 |
Neg logit contributions | |
ranger | -0.051 |
bleed | -0.050 |
bru | -0.049 |
sob | -0.048 |
hinges | -0.048 |
Token | played |
Token ID | 2826 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
ang | +0.061 |
irling | +0.054 |
ee | +0.052 |
irl | +0.050 |
ug | +0.048 |
Neg logit contributions | |
ranger | -0.068 |
bleed | -0.068 |
bru | -0.067 |
hinges | -0.067 |
sob | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | +4.47e-02 |
Pos logit contributions | |
ang | +0.085 |
irling | +0.072 |
ee | +0.070 |
irl | +0.064 |
ash | +0.063 |
Neg logit contributions | |
ranger | -0.113 |
bleed | -0.109 |
sob | -0.109 |
bru | -0.108 |
hinges | -0.105 |
Token | spl |
Token ID | 4328 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ang | +0.094 |
irling | +0.085 |
ee | +0.083 |
irl | +0.079 |
ading | +0.077 |
Neg logit contributions | |
ranger | -0.073 |
bleed | -0.071 |
bru | -0.069 |
sob | -0.069 |
hinges | -0.067 |
Token | ashed |
Token ID | 5263 |
Feature activation | +7.45e-02 |
Pos logit contributions | |
ang | +0.200 |
irling | +0.185 |
ee | +0.179 |
irl | +0.175 |
sorts | +0.164 |
Neg logit contributions | |
ranger | -0.155 |
bleed | -0.154 |
recovered | -0.148 |
department | -0.146 |
sob | -0.143 |
Token | in |
Token ID | 287 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
ang | +0.124 |
irling | +0.113 |
irl | +0.102 |
ee | +0.099 |
ug | +0.093 |
Neg logit contributions | |
ranger | -0.148 |
bleed | -0.143 |
sob | -0.141 |
bru | -0.140 |
hinges | -0.136 |
Token | the |
Token ID | 262 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
ang | +0.060 |
irling | +0.053 |
ee | +0.050 |
irl | +0.048 |
ug | +0.047 |
Neg logit contributions | |
ranger | -0.066 |
bleed | -0.064 |
bru | -0.064 |
hinges | -0.063 |
sob | -0.063 |
Token | water |
Token ID | 1660 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
ang | +0.068 |
irling | +0.061 |
ug | +0.056 |
ee | +0.056 |
irl | +0.056 |
Neg logit contributions | |
ranger | -0.063 |
bleed | -0.061 |
bru | -0.060 |
sob | -0.059 |
hinges | -0.059 |
Token | together |
Token ID | 1978 |
Feature activation | +9.03e-03 |
Pos logit contributions | |
ang | +0.057 |
irling | +0.052 |
ee | +0.051 |
irl | +0.050 |
ug | +0.048 |
Neg logit contributions | |
ranger | -0.025 |
bleed | -0.024 |
bru | -0.023 |
hinges | -0.023 |
sob | -0.023 |
Token | of |
Token ID | 286 |
Feature activation | +3.58e-02 |
Pos logit contributions | |
bleed | +0.026 |
bru | +0.026 |
ranger | +0.026 |
hinges | +0.026 |
ache | +0.025 |
Neg logit contributions | |
ang | -0.019 |
irling | -0.016 |
ug | -0.015 |
ee | -0.015 |
ading | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +1.65e-02 |
Pos logit contributions | |
ang | +0.062 |
irling | +0.054 |
ee | +0.052 |
irl | +0.050 |
ug | +0.048 |
Neg logit contributions | |
ranger | -0.070 |
bleed | -0.070 |
bru | -0.069 |
hinges | -0.068 |
sob | -0.068 |
Token | bucket |
Token ID | 19236 |
Feature activation | +2.20e-03 |
Pos logit contributions | |
ang | +0.031 |
irling | +0.028 |
ee | +0.027 |
irl | +0.026 |
ug | +0.025 |
Neg logit contributions | |
ranger | -0.030 |
bleed | -0.030 |
bru | -0.029 |
hinges | -0.029 |
sob | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +8.71e-05 |
Pos logit contributions | |
ang | +0.004 |
irling | +0.004 |
ee | +0.004 |
ug | +0.004 |
irl | +0.004 |
Neg logit contributions | |
bleed | -0.004 |
bru | -0.004 |
ranger | -0.004 |
hinges | -0.003 |
sob | -0.003 |
Token | It |
Token ID | 632 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
ang | +0.000 |
irling | +0.000 |
ee | +0.000 |
irl | +0.000 |
ug | +0.000 |
Neg logit contributions | |
bleed | -0.000 |
bru | -0.000 |
hinges | -0.000 |
ranger | -0.000 |
sob | -0.000 |
Token | sw |
Token ID | 1509 |
Feature activation | +9.14e-02 |
Pos logit contributions | |
ang | +0.037 |
irling | +0.033 |
ee | +0.032 |
irl | +0.031 |
ug | +0.031 |
Neg logit contributions | |
ranger | -0.026 |
bleed | -0.025 |
sob | -0.025 |
bru | -0.024 |
hinges | -0.024 |
Token | am |
Token ID | 321 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
ang | +0.171 |
irling | +0.149 |
ee | +0.145 |
irl | +0.140 |
ug | +0.132 |
Neg logit contributions | |
ranger | -0.158 |
hinges | -0.156 |
bleed | -0.156 |
department | -0.150 |
bru | -0.149 |
Token | in |
Token ID | 287 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
ang | +0.099 |
irling | +0.088 |
ee | +0.086 |
irl | +0.082 |
ug | +0.080 |
Neg logit contributions | |
bleed | -0.088 |
bru | -0.087 |
ranger | -0.086 |
hinges | -0.085 |
sob | -0.084 |
Token | a |
Token ID | 257 |
Feature activation | -9.72e-04 |
Pos logit contributions | |
ang | +0.026 |
irling | +0.022 |
ee | +0.022 |
irl | +0.021 |
ading | +0.020 |
Neg logit contributions | |
bleed | -0.042 |
ranger | -0.040 |
ache | -0.040 |
bru | -0.040 |
sob | -0.040 |
Token | p |
Token ID | 279 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
ranger | +0.002 |
bru | +0.002 |
bleed | +0.002 |
sob | +0.001 |
hinges | +0.001 |
Neg logit contributions | |
ang | -0.002 |
irling | -0.002 |
ee | -0.002 |
ug | -0.002 |
irl | -0.002 |
Token | uddle |
Token ID | 24500 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
ang | +0.103 |
irling | +0.098 |
ee | +0.098 |
irl | +0.096 |
ug | +0.091 |
Neg logit contributions | |
ranger | -0.100 |
hinges | -0.092 |
bleed | -0.090 |
keeper | -0.089 |
bru | -0.089 |
Token | little |
Token ID | 1310 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
ranger | +0.001 |
bru | +0.001 |
sob | +0.001 |
department | +0.001 |
keeper | +0.001 |
Neg logit contributions | |
ang | -0.002 |
irling | -0.002 |
ash | -0.002 |
ee | -0.001 |
ug | -0.001 |
Token | girl |
Token ID | 2576 |
Feature activation | +4.22e-03 |
Pos logit contributions | |
irling | +0.028 |
ang | +0.027 |
irl | +0.027 |
ading | +0.026 |
ash | +0.025 |
Neg logit contributions | |
ranger | -0.013 |
bru | -0.012 |
bleed | -0.011 |
sob | -0.011 |
cough | -0.011 |
Token | named |
Token ID | 3706 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
ang | +0.008 |
irling | +0.007 |
irl | +0.007 |
ading | +0.007 |
ee | +0.007 |
Neg logit contributions | |
ranger | -0.008 |
sob | -0.007 |
bru | -0.007 |
hinges | -0.007 |
aid | -0.007 |
Token | Mary |
Token ID | 5335 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
ang | +0.059 |
irling | +0.054 |
ee | +0.052 |
irl | +0.050 |
ug | +0.049 |
Neg logit contributions | |
ranger | -0.035 |
sob | -0.034 |
hinges | -0.034 |
bleed | -0.034 |
bru | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +2.71e-03 |
Pos logit contributions | |
ang | +0.052 |
ee | +0.048 |
irling | +0.047 |
irl | +0.044 |
ug | +0.043 |
Neg logit contributions | |
ranger | -0.036 |
hinges | -0.036 |
bleed | -0.035 |
department | -0.035 |
bru | -0.034 |
Token | Mary |
Token ID | 5335 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
ang | +0.006 |
irling | +0.006 |
ee | +0.006 |
irl | +0.005 |
ug | +0.005 |
Neg logit contributions | |
bleed | -0.004 |
ranger | -0.004 |
bru | -0.004 |
sob | -0.003 |
hinges | -0.003 |
Token | loved |
Token ID | 6151 |
Feature activation | -6.00e-03 |
Pos logit contributions | |
ang | +0.080 |
irling | +0.072 |
ee | +0.069 |
irl | +0.067 |
ug | +0.065 |
Neg logit contributions | |
ranger | -0.073 |
bleed | -0.073 |
bru | -0.071 |
hinges | -0.071 |
sob | -0.070 |
Token | to |
Token ID | 284 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
ranger | +0.005 |
bleed | +0.005 |
bru | +0.004 |
hinges | +0.004 |
sob | +0.004 |
Neg logit contributions | |
ang | -0.017 |
irling | -0.016 |
ee | -0.016 |
irl | -0.015 |
ug | -0.015 |
Token | wear |
Token ID | 5806 |
Feature activation | +2.30e-02 |
Pos logit contributions | |
ranger | +0.067 |
bleed | +0.065 |
department | +0.064 |
bru | +0.064 |
hinges | +0.063 |
Neg logit contributions | |
ang | -0.076 |
irling | -0.071 |
irl | -0.070 |
ee | -0.069 |
ug | -0.064 |
Token | her |
Token ID | 607 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
ang | +0.060 |
ee | +0.051 |
irling | +0.050 |
ug | +0.050 |
ash | +0.048 |
Neg logit contributions | |
sob | -0.027 |
bleed | -0.027 |
hinges | -0.026 |
bru | -0.025 |
ranger | -0.025 |
Token | favourite |
Token ID | 12507 |
Feature activation | -8.55e-03 |
Pos logit contributions | |
ranger | +0.062 |
bleed | +0.061 |
bru | +0.060 |
hinges | +0.060 |
sob | +0.060 |
Neg logit contributions | |
ang | -0.051 |
irling | -0.045 |
ee | -0.043 |
irl | -0.041 |
ug | -0.040 |
Token | ." |
Token ID | 526 |
Feature activation | -8.72e-03 |
Pos logit contributions | |
ranger | +0.010 |
bleed | +0.010 |
sob | +0.009 |
bru | +0.009 |
hinges | +0.009 |
Neg logit contributions | |
ang | -0.016 |
irling | -0.015 |
ee | -0.014 |
irl | -0.014 |
ug | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
bleed | +0.015 |
hinges | +0.014 |
department | +0.014 |
bru | +0.013 |
injuries | +0.013 |
Neg logit contributions | |
ang | -0.019 |
irling | -0.016 |
ee | -0.016 |
irl | -0.015 |
aw | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
ang | +0.026 |
irling | +0.024 |
ee | +0.023 |
irl | +0.022 |
ug | +0.022 |
Neg logit contributions | |
ranger | -0.013 |
bleed | -0.012 |
bru | -0.012 |
hinges | -0.012 |
sob | -0.012 |
Token | Ben |
Token ID | 11696 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
ang | +0.034 |
irling | +0.031 |
ee | +0.030 |
ug | +0.029 |
irl | +0.029 |
Neg logit contributions | |
ranger | -0.019 |
bru | -0.018 |
bleed | -0.017 |
sob | -0.017 |
hinges | -0.017 |
Token | che |
Token ID | 1125 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
ang | +0.034 |
irling | +0.031 |
ee | +0.030 |
irl | +0.029 |
ug | +0.028 |
Neg logit contributions | |
ranger | -0.022 |
bleed | -0.022 |
bru | -0.021 |
hinges | -0.021 |
sob | -0.021 |
Token | wed |
Token ID | 19103 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
ang | +0.163 |
irling | +0.145 |
ee | +0.138 |
irl | +0.135 |
ug | +0.131 |
Neg logit contributions | |
bleed | -0.145 |
ranger | -0.144 |
bru | -0.144 |
sob | -0.141 |
hinges | -0.141 |
Token | the |
Token ID | 262 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
ang | +0.053 |
irling | +0.048 |
ee | +0.045 |
irl | +0.045 |
ug | +0.043 |
Neg logit contributions | |
ranger | -0.051 |
bleed | -0.051 |
bru | -0.050 |
sob | -0.049 |
hinges | -0.049 |
Token | gum |
Token ID | 27142 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
ranger | +0.024 |
bleed | +0.023 |
hinges | +0.023 |
bru | +0.023 |
sob | +0.023 |
Neg logit contributions | |
ang | -0.030 |
irling | -0.027 |
ee | -0.026 |
irl | -0.025 |
ug | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
bru | +0.040 |
bleed | +0.039 |
hinges | +0.039 |
ranger | +0.039 |
injuries | +0.038 |
Neg logit contributions | |
ang | -0.035 |
ee | -0.031 |
irling | -0.030 |
irl | -0.029 |
ug | -0.027 |
Token | said |
Token ID | 531 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
ang | +0.027 |
irling | +0.025 |
ee | +0.024 |
irl | +0.023 |
ug | +0.022 |
Neg logit contributions | |
ranger | -0.021 |
bleed | -0.021 |
bru | -0.020 |
hinges | -0.020 |
sob | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
bleed | +0.069 |
hinges | +0.068 |
ranger | +0.067 |
injuries | +0.065 |
bru | +0.065 |
Neg logit contributions | |
ang | -0.053 |
irling | -0.049 |
ee | -0.047 |
irl | -0.046 |
ug | -0.045 |
Token | stayed |
Token ID | 9658 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
ang | +0.073 |
irling | +0.065 |
ee | +0.064 |
irl | +0.061 |
ug | +0.061 |
Neg logit contributions | |
bleed | -0.056 |
bru | -0.055 |
ranger | -0.055 |
hinges | -0.055 |
sob | -0.054 |
Token | on |
Token ID | 319 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
ang | +0.068 |
ee | +0.061 |
ash | +0.058 |
irling | +0.058 |
irl | +0.054 |
Neg logit contributions | |
bleed | -0.055 |
ranger | -0.055 |
hinges | -0.054 |
bru | -0.053 |
sob | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
ang | +0.018 |
irling | +0.016 |
ee | +0.015 |
irl | +0.015 |
ug | +0.014 |
Neg logit contributions | |
ranger | -0.021 |
bleed | -0.021 |
bru | -0.021 |
hinges | -0.021 |
sob | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +2.12e-03 |
Pos logit contributions | |
ang | +0.053 |
irling | +0.047 |
ee | +0.045 |
irl | +0.044 |
ug | +0.042 |
Neg logit contributions | |
ranger | -0.048 |
bleed | -0.047 |
bru | -0.046 |
hinges | -0.046 |
sob | -0.046 |
Token | car |
Token ID | 1097 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
ang | +0.004 |
irling | +0.004 |
ee | +0.004 |
irl | +0.003 |
ug | +0.003 |
Neg logit contributions | |
bleed | -0.004 |
ranger | -0.004 |
sob | -0.004 |
bru | -0.004 |
hinges | -0.004 |
Token | moved |
Token ID | 3888 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
ang | +0.030 |
irling | +0.027 |
ee | +0.027 |
irl | +0.026 |
ug | +0.026 |
Neg logit contributions | |
bleed | -0.027 |
bru | -0.027 |
hinges | -0.026 |
ranger | -0.026 |
sob | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -4.73e-03 |
Pos logit contributions | |
ang | +0.082 |
irling | +0.073 |
ee | +0.072 |
irl | +0.068 |
ug | +0.068 |
Neg logit contributions | |
bleed | -0.067 |
ranger | -0.067 |
bru | -0.066 |
sob | -0.065 |
hinges | -0.065 |
Token | Tom |
Token ID | 4186 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
bleed | +0.007 |
hinges | +0.007 |
ranger | +0.007 |
bru | +0.007 |
sob | +0.007 |
Neg logit contributions | |
ang | -0.010 |
irling | -0.009 |
ee | -0.009 |
irl | -0.008 |
ug | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +2.94e-02 |
Pos logit contributions | |
ang | +0.029 |
irling | +0.027 |
ee | +0.026 |
irl | +0.025 |
ading | +0.025 |
Neg logit contributions | |
ranger | -0.019 |
sob | -0.018 |
bleed | -0.018 |
bru | -0.018 |
hinges | -0.017 |
Token | happy |
Token ID | 3772 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
ang | +0.062 |
ee | +0.057 |
ash | +0.054 |
irling | +0.054 |
angles | +0.053 |
Neg logit contributions | |
bleed | -0.048 |
sob | -0.047 |
ranger | -0.047 |
bru | -0.045 |
hinges | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-04 |
Pos logit contributions | |
bleed | +0.034 |
hinges | +0.034 |
bru | +0.031 |
severe | +0.031 |
limp | +0.031 |
Neg logit contributions | |
ang | -0.031 |
irling | -0.026 |
ee | -0.026 |
aw | -0.024 |
ug | -0.023 |
Token | " |
Token ID | 366 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
bleed | +0.049 |
ranger | +0.048 |
bru | +0.047 |
hinges | +0.047 |
sob | +0.046 |
Neg logit contributions | |
ang | -0.078 |
irling | -0.071 |
ee | -0.069 |
irl | -0.067 |
ug | -0.065 |
Token | Look |
Token ID | 8567 |
Feature activation | +4.60e-03 |
Pos logit contributions | |
ang | +0.067 |
irling | +0.062 |
ug | +0.060 |
irl | +0.059 |
ading | +0.058 |
Neg logit contributions | |
ranger | -0.037 |
Something | -0.035 |
bleed | -0.034 |
bru | -0.034 |
sob | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | +1.28e-03 |
Pos logit contributions | |
ang | +0.011 |
irling | +0.010 |
sorts | +0.010 |
ading | +0.010 |
ug | +0.010 |
Neg logit contributions | |
sob | -0.005 |
bud | -0.005 |
ranger | -0.005 |
bru | -0.005 |
keeper | -0.004 |
Token | Mom |
Token ID | 11254 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.002 |
ug | +0.002 |
irl | +0.002 |
Neg logit contributions | |
ranger | -0.002 |
bru | -0.002 |
sob | -0.002 |
bleed | -0.002 |
department | -0.002 |
Token | ! |
Token ID | 0 |
Feature activation | +1.59e-02 |
Pos logit contributions | |
ang | +0.026 |
irling | +0.024 |
irl | +0.022 |
ee | +0.022 |
ading | +0.022 |
Neg logit contributions | |
ranger | -0.019 |
bru | -0.018 |
department | -0.018 |
bleed | -0.018 |
sob | -0.018 |
Token | More |
Token ID | 3125 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
ang | +0.031 |
irling | +0.028 |
ee | +0.027 |
irl | +0.026 |
ug | +0.025 |
Neg logit contributions | |
ranger | -0.027 |
bleed | -0.027 |
bru | -0.027 |
hinges | -0.026 |
sob | -0.026 |
Token | spots |
Token ID | 10222 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
ang | +0.055 |
irling | +0.052 |
ee | +0.049 |
irl | +0.048 |
ash | +0.048 |
Neg logit contributions | |
bud | -0.040 |
hinges | -0.040 |
bru | -0.040 |
bleed | -0.040 |
ranger | -0.039 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.24e-02 |
Pos logit contributions | |
ang | +0.094 |
irling | +0.083 |
ee | +0.076 |
irl | +0.075 |
ug | +0.074 |
Neg logit contributions | |
bleed | -0.087 |
ranger | -0.085 |
bru | -0.082 |
bud | -0.082 |
hinges | -0.081 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
ranger | +0.018 |
bleed | +0.018 |
hinges | +0.018 |
bru | +0.018 |
department | +0.018 |
Neg logit contributions | |
ang | -0.027 |
irling | -0.024 |
ee | -0.023 |
irl | -0.023 |
ug | -0.022 |
Token | Once |
Token ID | 7454 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
ang | +0.021 |
ee | +0.020 |
irling | +0.018 |
irl | +0.018 |
ash | +0.017 |
Neg logit contributions | |
bleed | -0.018 |
hinges | -0.018 |
bru | -0.017 |
department | -0.016 |
ranger | -0.016 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
ang | +0.055 |
irling | +0.051 |
ee | +0.048 |
irl | +0.048 |
ug | +0.046 |
Neg logit contributions | |
ranger | -0.033 |
bru | -0.032 |
sob | -0.032 |
bleed | -0.032 |
hinges | -0.031 |
Token | young |
Token ID | 1862 |
Feature activation | +1.26e-03 |
Pos logit contributions | |
ang | +0.049 |
irling | +0.043 |
ee | +0.043 |
irl | +0.041 |
ug | +0.040 |
Neg logit contributions | |
ranger | -0.039 |
bleed | -0.038 |
hinges | -0.037 |
department | -0.036 |
bud | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | +5.10e-03 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.002 |
ee | +0.002 |
irl | +0.002 |
ug | +0.002 |
Neg logit contributions | |
bleed | -0.002 |
ranger | -0.002 |
hinges | -0.002 |
bru | -0.002 |
sob | -0.002 |
Token | she |
Token ID | 673 |
Feature activation | +8.51e-03 |
Pos logit contributions | |
ang | +0.011 |
irling | +0.010 |
ee | +0.009 |
irl | +0.009 |
ug | +0.009 |
Neg logit contributions | |
ranger | -0.008 |
bleed | -0.008 |
bru | -0.008 |
sob | -0.008 |
hinges | -0.008 |
Token | could |
Token ID | 714 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
ang | +0.019 |
irling | +0.017 |
ee | +0.017 |
irl | +0.016 |
ug | +0.016 |
Neg logit contributions | |
ranger | -0.012 |
bleed | -0.012 |
bru | -0.012 |
sob | -0.012 |
hinges | -0.011 |
Token | still |
Token ID | 991 |
Feature activation | -7.72e-03 |
Pos logit contributions | |
ranger | +0.050 |
hinges | +0.049 |
bru | +0.048 |
bleed | +0.048 |
sob | +0.048 |
Neg logit contributions | |
ang | -0.054 |
irling | -0.050 |
irl | -0.049 |
ee | -0.048 |
ading | -0.045 |
Token | make |
Token ID | 787 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
ranger | +0.012 |
bleed | +0.012 |
bru | +0.012 |
hinges | +0.012 |
sob | +0.012 |
Neg logit contributions | |
ang | -0.016 |
irling | -0.014 |
ee | -0.014 |
irl | -0.014 |
ading | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
ang | +0.061 |
irling | +0.053 |
ee | +0.051 |
irl | +0.049 |
ug | +0.048 |
Neg logit contributions | |
bleed | -0.053 |
ranger | -0.052 |
sob | -0.051 |
bru | -0.051 |
hinges | -0.050 |
Token | difference |
Token ID | 3580 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
bleed | +0.036 |
ranger | +0.035 |
hinges | +0.035 |
sob | +0.035 |
bru | +0.034 |
Neg logit contributions | |
ang | -0.035 |
irling | -0.031 |
ee | -0.030 |
irl | -0.029 |
ug | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -3.72e-03 |
Pos logit contributions | |
ang | +0.077 |
irling | +0.070 |
ee | +0.068 |
irl | +0.066 |
ug | +0.064 |
Neg logit contributions | |
ranger | -0.056 |
bleed | -0.055 |
bru | -0.054 |
hinges | -0.054 |
sob | -0.053 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
bleed | +0.006 |
ranger | +0.005 |
bru | +0.005 |
hinges | +0.005 |
department | +0.005 |
Neg logit contributions | |
ang | -0.008 |
irling | -0.007 |
ee | -0.007 |
irl | -0.007 |
ug | -0.006 |
Token | John |
Token ID | 7554 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
ang | +0.025 |
ee | +0.022 |
irling | +0.022 |
irl | +0.021 |
ug | +0.020 |
Neg logit contributions | |
bleed | -0.019 |
hinges | -0.018 |
ranger | -0.018 |
bru | -0.018 |
sob | -0.017 |
Token | together |
Token ID | 1978 |
Feature activation | -4.19e-03 |
Pos logit contributions | |
ang | +0.048 |
irling | +0.042 |
ee | +0.041 |
irl | +0.038 |
ug | +0.037 |
Neg logit contributions | |
ranger | -0.054 |
bleed | -0.053 |
bru | -0.053 |
sob | -0.052 |
hinges | -0.052 |
Token | in |
Token ID | 287 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
ranger | +0.005 |
bleed | +0.005 |
bru | +0.005 |
sob | +0.005 |
hinges | +0.005 |
Neg logit contributions | |
ang | -0.010 |
irling | -0.009 |
ee | -0.009 |
irl | -0.009 |
ug | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
ang | +0.050 |
irling | +0.043 |
ee | +0.042 |
ug | +0.040 |
irl | +0.040 |
Neg logit contributions | |
ranger | -0.049 |
bru | -0.046 |
hinges | -0.045 |
bleed | -0.045 |
aid | -0.045 |
Token | yard |
Token ID | 12699 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
ang | +0.091 |
irling | +0.081 |
ug | +0.080 |
irl | +0.075 |
ee | +0.074 |
Neg logit contributions | |
ranger | -0.055 |
aid | -0.050 |
bru | -0.048 |
keeper | -0.047 |
hinges | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | +3.53e-03 |
Pos logit contributions | |
bleed | +0.017 |
ranger | +0.017 |
bru | +0.017 |
hinges | +0.016 |
sob | +0.016 |
Neg logit contributions | |
ang | -0.036 |
irling | -0.033 |
ee | -0.032 |
irl | -0.031 |
ug | -0.031 |
Token | The |
Token ID | 383 |
Feature activation | +6.11e-03 |
Pos logit contributions | |
ang | +0.009 |
irling | +0.009 |
ee | +0.008 |
irl | +0.008 |
ug | +0.008 |
Neg logit contributions | |
ranger | -0.004 |
bru | -0.004 |
hinges | -0.004 |
Something | -0.004 |
department | -0.004 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
ang | +0.012 |
irling | +0.010 |
ee | +0.010 |
irl | +0.010 |
ug | +0.009 |
Neg logit contributions | |
ranger | -0.011 |
bleed | -0.011 |
bru | -0.011 |
hinges | -0.011 |
sob | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
ang | +0.025 |
irling | +0.022 |
ee | +0.022 |
irl | +0.021 |
ug | +0.021 |
Neg logit contributions | |
ranger | -0.018 |
bleed | -0.018 |
bru | -0.018 |
sob | -0.018 |
hinges | -0.018 |
Token | always |
Token ID | 1464 |
Feature activation | +2.44e-02 |
Pos logit contributions | |
ang | +0.055 |
ee | +0.049 |
irling | +0.047 |
ug | +0.047 |
ash | +0.047 |
Neg logit contributions | |
ranger | -0.042 |
sob | -0.039 |
bleed | -0.039 |
bru | -0.039 |
keeper | -0.037 |
Token | very |
Token ID | 845 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
ang | +0.048 |
ee | +0.044 |
irling | +0.042 |
irl | +0.042 |
ug | +0.041 |
Neg logit contributions | |
ranger | -0.042 |
sob | -0.040 |
bleed | -0.039 |
bru | -0.039 |
keeper | -0.038 |
Token | curious |
Token ID | 11040 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ang | +0.037 |
ee | +0.034 |
irling | +0.034 |
irl | +0.033 |
angles | +0.033 |
Neg logit contributions | |
ranger | -0.032 |
bleed | -0.030 |
sob | -0.029 |
bru | -0.028 |
keeper | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
bleed | +0.014 |
ranger | +0.014 |
bru | +0.013 |
sob | +0.013 |
hinges | +0.013 |
Neg logit contributions | |
ang | -0.018 |
ee | -0.015 |
irling | -0.015 |
ug | -0.014 |
ash | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | +3.74e-03 |
Pos logit contributions | |
ranger | +0.034 |
bleed | +0.033 |
bru | +0.033 |
hinges | +0.033 |
sob | +0.033 |
Neg logit contributions | |
ang | -0.036 |
irling | -0.032 |
ee | -0.031 |
irl | -0.030 |
ug | -0.029 |
Token | a |
Token ID | 257 |
Feature activation | -7.64e-03 |
Pos logit contributions | |
ang | +0.007 |
irling | +0.006 |
ee | +0.006 |
irl | +0.006 |
ug | +0.006 |
Neg logit contributions | |
ranger | -0.007 |
bleed | -0.007 |
bru | -0.007 |
hinges | -0.007 |
sob | -0.007 |
Token | table |
Token ID | 3084 |
Feature activation | -5.02e-03 |
Pos logit contributions | |
bleed | +0.015 |
ranger | +0.014 |
bru | +0.014 |
hinges | +0.014 |
sob | +0.014 |
Neg logit contributions | |
ang | -0.013 |
irling | -0.012 |
ee | -0.011 |
irl | -0.011 |
ug | -0.011 |
Token | where |
Token ID | 810 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
bleed | +0.008 |
bru | +0.007 |
sob | +0.007 |
hinges | +0.007 |
ranger | +0.007 |
Neg logit contributions | |
ang | -0.011 |
irling | -0.010 |
ee | -0.010 |
irl | -0.009 |
ug | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | +8.28e-03 |
Pos logit contributions | |
ang | +0.045 |
irling | +0.040 |
ee | +0.038 |
irl | +0.036 |
ug | +0.035 |
Neg logit contributions | |
ranger | -0.043 |
bleed | -0.042 |
bru | -0.041 |
hinges | -0.041 |
sob | -0.041 |
Token | dog |
Token ID | 3290 |
Feature activation | +1.51e-03 |
Pos logit contributions | |
ang | +0.014 |
ee | +0.012 |
irling | +0.012 |
irl | +0.012 |
ading | +0.011 |
Neg logit contributions | |
bleed | -0.016 |
sob | -0.015 |
bru | -0.015 |
ranger | -0.015 |
hinges | -0.015 |
Token | cannot |
Token ID | 2314 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.003 |
irl | +0.003 |
ug | +0.002 |
Neg logit contributions | |
ranger | -0.002 |
bleed | -0.002 |
bru | -0.002 |
hinges | -0.002 |
sob | -0.002 |
Token | reach |
Token ID | 3151 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
sob | +0.042 |
bru | +0.042 |
ranger | +0.042 |
bleed | +0.042 |
hinges | +0.041 |
Neg logit contributions | |
ang | -0.044 |
ee | -0.042 |
irl | -0.041 |
irling | -0.041 |
ug | -0.039 |
Token | it |
Token ID | 340 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
ang | +0.055 |
irling | +0.050 |
ee | +0.048 |
irl | +0.047 |
ug | +0.045 |
Neg logit contributions | |
ranger | -0.040 |
bleed | -0.040 |
bru | -0.039 |
hinges | -0.038 |
sob | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -4.87e-03 |
Pos logit contributions | |
bru | +0.052 |
ranger | +0.050 |
sob | +0.050 |
bleed | +0.049 |
injuries | +0.048 |
Neg logit contributions | |
ang | -0.069 |
ee | -0.062 |
irl | -0.062 |
irling | -0.061 |
ash | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
ang | +0.086 |
irling | +0.078 |
ee | +0.075 |
irl | +0.074 |
ug | +0.069 |
Neg logit contributions | |
bleed | -0.075 |
bru | -0.073 |
hinges | -0.073 |
ranger | -0.072 |
sob | -0.070 |
Token | And |
Token ID | 843 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
ang | +0.042 |
irling | +0.040 |
ee | +0.039 |
irl | +0.039 |
ading | +0.037 |
Neg logit contributions | |
ranger | -0.016 |
sob | -0.013 |
bleed | -0.013 |
bru | -0.013 |
keeper | -0.013 |
Token | that |
Token ID | 326 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
ang | +0.040 |
irling | +0.037 |
ee | +0.035 |
irl | +0.034 |
ash | +0.032 |
Neg logit contributions | |
ranger | -0.037 |
bleed | -0.034 |
sob | -0.034 |
keeper | -0.034 |
bru | -0.034 |
Token | is |
Token ID | 318 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ang | +0.066 |
irling | +0.059 |
ug | +0.057 |
ash | +0.055 |
ading | +0.055 |
Neg logit contributions | |
ranger | -0.033 |
keeper | -0.031 |
sob | -0.030 |
truth | -0.027 |
cough | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | +9.53e-03 |
Pos logit contributions | |
ang | +0.029 |
irling | +0.026 |
ee | +0.025 |
ug | +0.023 |
irl | +0.023 |
Neg logit contributions | |
ranger | -0.029 |
keeper | -0.027 |
department | -0.027 |
bru | -0.027 |
bleed | -0.027 |
Token | moral |
Token ID | 6573 |
Feature activation | +2.23e-03 |
Pos logit contributions | |
ang | +0.012 |
irling | +0.010 |
ee | +0.010 |
irl | +0.010 |
ading | +0.009 |
Neg logit contributions | |
bleed | -0.023 |
hinges | -0.022 |
bru | -0.022 |
sob | -0.022 |
ranger | -0.022 |
Token | of |
Token ID | 286 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
ang | +0.003 |
ee | +0.002 |
irling | +0.002 |
irl | +0.002 |
ug | +0.002 |
Neg logit contributions | |
bleed | -0.006 |
bru | -0.005 |
hinges | -0.005 |
sob | -0.005 |
ranger | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ang | +0.065 |
irling | +0.061 |
irl | +0.056 |
ee | +0.056 |
ug | +0.052 |
Neg logit contributions | |
ranger | -0.064 |
bru | -0.059 |
sob | -0.059 |
keeper | -0.058 |
mercy | -0.058 |
Token | story |
Token ID | 1621 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
ang | +0.028 |
irling | +0.026 |
ee | +0.025 |
irl | +0.024 |
ug | +0.024 |
Neg logit contributions | |
ranger | -0.015 |
bleed | -0.014 |
bru | -0.014 |
sob | -0.013 |
hinges | -0.013 |
Token | : |
Token ID | 25 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
bleed | +0.041 |
bru | +0.040 |
hinges | +0.039 |
ranger | +0.038 |
sob | +0.038 |
Neg logit contributions | |
ang | -0.054 |
ee | -0.048 |
irling | -0.048 |
irl | -0.046 |
ug | -0.045 |
Token | sharing |
Token ID | 7373 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
ranger | +0.023 |
bru | +0.022 |
bleed | +0.022 |
sob | +0.021 |
hinges | +0.021 |
Neg logit contributions | |
ang | -0.030 |
irling | -0.027 |
ee | -0.026 |
irl | -0.025 |
ug | -0.024 |
Token | are |
Token ID | 389 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
keeping | +0.100 |
edly | +0.093 |
ranger | +0.091 |
hinges | +0.090 |
recovered | +0.087 |
Neg logit contributions | |
ang | -0.098 |
ee | -0.085 |
irling | -0.084 |
irl | -0.083 |
ch | -0.077 |
Token | sorry |
Token ID | 7926 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
ang | +0.066 |
irling | +0.057 |
ee | +0.056 |
irl | +0.054 |
ug | +0.052 |
Neg logit contributions | |
ranger | -0.061 |
bleed | -0.060 |
bru | -0.059 |
sob | -0.059 |
hinges | -0.057 |
Token | , |
Token ID | 11 |
Feature activation | -1.42e-03 |
Pos logit contributions | |
hinges | +0.050 |
bleed | +0.050 |
ranger | +0.050 |
bru | +0.049 |
department | +0.047 |
Neg logit contributions | |
ang | -0.057 |
ee | -0.051 |
irling | -0.050 |
irl | -0.048 |
ug | -0.048 |
Token | m |
Token ID | 285 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
hinges | +0.002 |
bleed | +0.002 |
recovered | +0.002 |
bru | +0.002 |
department | +0.002 |
Neg logit contributions | |
ang | -0.003 |
irling | -0.002 |
ee | -0.002 |
irl | -0.002 |
ug | -0.002 |
Token | ister |
Token ID | 1694 |
Feature activation | +8.03e-03 |
Pos logit contributions | |
ang | +0.073 |
irling | +0.063 |
ee | +0.062 |
irl | +0.058 |
ug | +0.057 |
Neg logit contributions | |
ranger | -0.081 |
bleed | -0.081 |
hinges | -0.080 |
bru | -0.079 |
sob | -0.078 |
Token | ," |
Token ID | 553 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
ang | +0.018 |
ee | +0.016 |
irling | +0.015 |
ug | +0.015 |
irl | +0.014 |
Neg logit contributions | |
hinges | -0.012 |
bleed | -0.012 |
bru | -0.011 |
sob | -0.011 |
ranger | -0.011 |
Token | Lily |
Token ID | 20037 |
Feature activation | -3.97e-03 |
Pos logit contributions | |
hinges | +0.021 |
bleed | +0.021 |
department | +0.020 |
injuries | +0.019 |
ranger | +0.019 |
Neg logit contributions | |
ang | -0.020 |
ee | -0.017 |
irling | -0.016 |
irl | -0.015 |
ug | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | -4.31e-02 |
Pos logit contributions | |
hinges | +0.009 |
injuries | +0.009 |
bleed | +0.008 |
ranger | +0.008 |
keeping | +0.008 |
Neg logit contributions | |
ang | -0.005 |
ee | -0.005 |
irling | -0.005 |
irl | -0.005 |
aw | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
hinges | +0.077 |
ranger | +0.076 |
department | +0.073 |
bleed | +0.072 |
bru | +0.071 |
Neg logit contributions | |
ang | -0.076 |
irling | -0.073 |
ug | -0.068 |
irl | -0.065 |
ee | -0.064 |
Token | " |
Token ID | 366 |
Feature activation | +1.99e-03 |
Pos logit contributions | |
bleed | +0.048 |
hinges | +0.046 |
mercy | +0.046 |
recovered | +0.046 |
injuries | +0.045 |
Neg logit contributions | |
ang | -0.048 |
irling | -0.040 |
ee | -0.037 |
ug | -0.037 |
irl | -0.037 |
Token | We |
Token ID | 1135 |
Feature activation | +6.15e-03 |
Pos logit contributions | |
ang | +0.004 |
irling | +0.003 |
ee | +0.003 |
irl | +0.003 |
ug | +0.003 |
Neg logit contributions | |
hinges | -0.004 |
ranger | -0.004 |
department | -0.003 |
bru | -0.003 |
bleed | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
ang | +0.012 |
irling | +0.012 |
ee | +0.011 |
irl | +0.011 |
ug | +0.011 |
Neg logit contributions | |
ranger | -0.004 |
bleed | -0.004 |
bru | -0.004 |
hinges | -0.004 |
sob | -0.004 |
Token | know |
Token ID | 760 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
ranger | +0.091 |
department | +0.089 |
bru | +0.087 |
sob | +0.087 |
bleed | +0.086 |
Neg logit contributions | |
ang | -0.094 |
irl | -0.093 |
ee | -0.092 |
irling | -0.089 |
ading | -0.081 |
Token | what |
Token ID | 644 |
Feature activation | +1.35e-03 |
Pos logit contributions | |
ranger | +0.018 |
hinges | +0.017 |
department | +0.017 |
mercy | +0.017 |
sob | +0.017 |
Neg logit contributions | |
ang | -0.021 |
irling | -0.018 |
ee | -0.017 |
irl | -0.017 |
ug | -0.016 |
Token | you |
Token ID | 345 |
Feature activation | +9.24e-03 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.002 |
ee | +0.002 |
irl | +0.002 |
ug | +0.002 |
Neg logit contributions | |
ranger | -0.002 |
bleed | -0.002 |
bru | -0.002 |
sob | -0.002 |
hinges | -0.002 |
Token | see |
Token ID | 766 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
ang | +0.019 |
irling | +0.017 |
ee | +0.017 |
irl | +0.016 |
ug | +0.015 |
Neg logit contributions | |
ranger | -0.015 |
bleed | -0.015 |
bru | -0.015 |
hinges | -0.014 |
sob | -0.014 |
Token | ." |
Token ID | 526 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ang | +0.061 |
irling | +0.057 |
irl | +0.054 |
ee | +0.053 |
ug | +0.051 |
Neg logit contributions | |
ranger | -0.033 |
hinges | -0.032 |
sob | -0.031 |
bleed | -0.031 |
bru | -0.031 |
Token | Mary |
Token ID | 5335 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
bleed | +0.028 |
injuries | +0.026 |
bru | +0.025 |
department | +0.025 |
hinges | +0.025 |
Neg logit contributions | |
ang | -0.029 |
ee | -0.026 |
irling | -0.024 |
aw | -0.024 |
sorts | -0.023 |
Token | stepped |
Token ID | 10764 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
ang | +0.038 |
ee | +0.034 |
irling | +0.034 |
irl | +0.032 |
ug | +0.030 |
Neg logit contributions | |
bleed | -0.030 |
hinges | -0.029 |
bru | -0.028 |
ranger | -0.028 |
injuries | -0.028 |
Token | closer |
Token ID | 5699 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
ang | +0.068 |
irling | +0.059 |
ee | +0.058 |
ading | +0.055 |
irl | +0.054 |
Neg logit contributions | |
bleed | -0.060 |
bru | -0.057 |
ranger | -0.057 |
hinges | -0.056 |
onna | -0.054 |
Token | and |
Token ID | 290 |
Feature activation | +6.49e-03 |
Pos logit contributions | |
bleed | +0.082 |
bru | +0.076 |
injuries | +0.073 |
mercy | +0.073 |
choke | +0.073 |
Neg logit contributions | |
ang | -0.061 |
irling | -0.051 |
ee | -0.050 |
aw | -0.047 |
ing | -0.045 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
ang | +0.014 |
irling | +0.013 |
ee | +0.012 |
irl | +0.012 |
ug | +0.011 |
Neg logit contributions | |
ranger | -0.010 |
bleed | -0.010 |
bru | -0.010 |
hinges | -0.010 |
sob | -0.009 |
Token | He |
Token ID | 679 |
Feature activation | -6.46e-03 |
Pos logit contributions | |
bleed | +0.040 |
injuries | +0.037 |
ache | +0.037 |
recovered | +0.036 |
mercy | +0.036 |
Neg logit contributions | |
ang | -0.036 |
ee | -0.027 |
ug | -0.027 |
aw | -0.026 |
irling | -0.026 |
Token | smiled |
Token ID | 13541 |
Feature activation | +6.75e-03 |
Pos logit contributions | |
bleed | +0.011 |
hinges | +0.010 |
bru | +0.010 |
ranger | +0.010 |
injuries | +0.010 |
Neg logit contributions | |
ang | -0.013 |
irling | -0.012 |
ee | -0.012 |
irl | -0.011 |
ug | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +8.41e-03 |
Pos logit contributions | |
ang | +0.010 |
irling | +0.008 |
aw | +0.008 |
ee | +0.008 |
ug | +0.008 |
Neg logit contributions | |
bleed | -0.015 |
bru | -0.014 |
hinges | -0.014 |
keeping | -0.014 |
onna | -0.014 |
Token | squeezed |
Token ID | 29650 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
ang | +0.019 |
irling | +0.017 |
ee | +0.017 |
irl | +0.016 |
ug | +0.016 |
Neg logit contributions | |
ranger | -0.012 |
bleed | -0.012 |
sob | -0.012 |
bru | -0.011 |
hinges | -0.011 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
ang | +0.088 |
irling | +0.078 |
ee | +0.076 |
irl | +0.073 |
ug | +0.071 |
Neg logit contributions | |
ranger | -0.078 |
bleed | -0.077 |
bru | -0.076 |
hinges | -0.075 |
sob | -0.075 |
Token | 's |
Token ID | 338 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
ranger | +0.116 |
bru | +0.115 |
hinges | +0.115 |
department | +0.113 |
bleed | +0.112 |
Neg logit contributions | |
ang | -0.101 |
ee | -0.087 |
irling | -0.086 |
irl | -0.080 |
ug | -0.077 |
Token | hand |
Token ID | 1021 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
department | +0.104 |
ranger | +0.096 |
recovered | +0.095 |
onna | +0.094 |
ache | +0.090 |
Neg logit contributions | |
ang | -0.080 |
ee | -0.076 |
irling | -0.064 |
irl | -0.063 |
ading | -0.062 |
Token | gently |
Token ID | 15165 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
ranger | +0.041 |
bru | +0.041 |
hinges | +0.040 |
bleed | +0.040 |
department | +0.039 |
Neg logit contributions | |
ang | -0.060 |
ee | -0.053 |
irling | -0.052 |
ug | -0.050 |
ash | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
department | +0.081 |
ranger | +0.080 |
hinges | +0.076 |
mercy | +0.076 |
bru | +0.074 |
Neg logit contributions | |
ang | -0.083 |
irling | -0.071 |
ee | -0.071 |
ug | -0.066 |
irl | -0.065 |
Token | They |
Token ID | 1119 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
bleed | +0.022 |
ache | +0.020 |
mercy | +0.020 |
hinges | +0.020 |
injuries | +0.020 |
Neg logit contributions | |
ang | -0.023 |
irling | -0.019 |
irl | -0.018 |
aw | -0.018 |
ee | -0.018 |
Token | entered |
Token ID | 5982 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
ang | +0.020 |
irling | +0.017 |
ee | +0.017 |
irl | +0.017 |
ug | +0.016 |
Neg logit contributions | |
ranger | -0.017 |
bleed | -0.017 |
hinges | -0.017 |
bru | -0.017 |
department | -0.016 |
Token | up |
Token ID | 510 |
Feature activation | +2.16e-04 |
Pos logit contributions | |
ranger | +0.095 |
bleed | +0.095 |
bru | +0.095 |
hinges | +0.093 |
sob | +0.093 |
Neg logit contributions | |
ang | -0.051 |
irling | -0.042 |
ee | -0.041 |
irl | -0.039 |
ug | -0.036 |
Token | and |
Token ID | 290 |
Feature activation | +2.74e-03 |
Pos logit contributions | |
ang | +0.000 |
ee | +0.000 |
irling | +0.000 |
irl | +0.000 |
aw | +0.000 |
Neg logit contributions | |
bleed | -0.000 |
department | -0.000 |
bru | -0.000 |
hinges | -0.000 |
ranger | -0.000 |
Token | ran |
Token ID | 4966 |
Feature activation | +5.73e-03 |
Pos logit contributions | |
ang | +0.006 |
irling | +0.005 |
ee | +0.005 |
irl | +0.005 |
ug | +0.005 |
Neg logit contributions | |
ranger | -0.004 |
bleed | -0.004 |
bru | -0.004 |
sob | -0.004 |
hinges | -0.004 |
Token | to |
Token ID | 284 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
ang | +0.012 |
ee | +0.010 |
irling | +0.010 |
ash | +0.010 |
irl | +0.010 |
Neg logit contributions | |
bru | -0.009 |
bleed | -0.009 |
sob | -0.009 |
ache | -0.008 |
mercy | -0.008 |
Token | her |
Token ID | 607 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
ranger | +0.077 |
bleed | +0.075 |
hinges | +0.075 |
bru | +0.074 |
department | +0.074 |
Neg logit contributions | |
ang | -0.056 |
irling | -0.049 |
ee | -0.048 |
irl | -0.047 |
ug | -0.042 |
Token | mom |
Token ID | 1995 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
bleed | +0.111 |
recovered | +0.106 |
mercy | +0.104 |
bru | +0.103 |
ache | +0.102 |
Neg logit contributions | |
ang | -0.078 |
irling | -0.064 |
irl | -0.062 |
ee | -0.059 |
utes | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
bleed | +0.096 |
bru | +0.094 |
ranger | +0.093 |
hinges | +0.090 |
sob | +0.088 |
Neg logit contributions | |
ang | -0.156 |
irling | -0.138 |
ee | -0.137 |
irl | -0.131 |
ash | -0.130 |
Token |
|
Token ID | 198 |
Feature activation | +1.36e-03 |
Pos logit contributions | |
bleed | +0.036 |
injuries | +0.035 |
mercy | +0.035 |
ache | +0.034 |
bru | +0.034 |
Neg logit contributions | |
ang | -0.037 |
ee | -0.031 |
irling | -0.030 |
irl | -0.030 |
ug | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -2.20e-03 |
Pos logit contributions | |
ang | +0.003 |
irling | +0.003 |
ee | +0.003 |
irl | +0.003 |
ug | +0.003 |
Neg logit contributions | |
bleed | -0.002 |
ranger | -0.002 |
bru | -0.002 |
hinges | -0.002 |
department | -0.002 |
Token | She |
Token ID | 3347 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
ranger | +0.003 |
bleed | +0.003 |
sob | +0.003 |
hinges | +0.003 |
bru | +0.003 |
Neg logit contributions | |
ang | -0.005 |
irling | -0.005 |
ee | -0.005 |
irl | -0.004 |
ug | -0.004 |
Token | gave |
Token ID | 2921 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
ranger | +0.023 |
bleed | +0.023 |
bru | +0.022 |
sob | +0.022 |
hinges | +0.022 |
Neg logit contributions | |
ang | -0.032 |
irling | -0.029 |
ee | -0.028 |
irl | -0.028 |
ug | -0.027 |
Token | my |
Token ID | 616 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
ang | +0.013 |
irling | +0.011 |
ee | +0.010 |
irl | +0.010 |
ug | +0.009 |
Neg logit contributions | |
ranger | -0.028 |
bleed | -0.028 |
bru | -0.028 |
hinges | -0.027 |
sob | -0.027 |
Token | toy |
Token ID | 13373 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
bleed | +0.066 |
recovered | +0.065 |
ranger | +0.064 |
department | +0.063 |
Something | +0.062 |
Neg logit contributions | |
ang | -0.056 |
irling | -0.048 |
ee | -0.048 |
irl | -0.048 |
ading | -0.041 |
Token | ," |
Token ID | 553 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
bleed | +0.037 |
bru | +0.036 |
recovered | +0.036 |
ranger | +0.035 |
Something | +0.035 |
Neg logit contributions | |
ang | -0.037 |
ug | -0.030 |
ee | -0.030 |
irling | -0.028 |
irl | -0.027 |
Token | said |
Token ID | 531 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
hinges | +0.050 |
bleed | +0.050 |
ranger | +0.049 |
department | +0.049 |
bru | +0.049 |
Neg logit contributions | |
ang | -0.050 |
irling | -0.044 |
ee | -0.043 |
irl | -0.042 |
ug | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
hinges | +0.071 |
bleed | +0.069 |
ranger | +0.067 |
onna | +0.067 |
bru | +0.066 |
Neg logit contributions | |
ang | -0.048 |
ee | -0.041 |
ug | -0.039 |
irling | -0.039 |
irl | -0.038 |
Token | boy |
Token ID | 2933 |
Feature activation | -5.91e-02 |
Pos logit contributions | |
onna | +0.042 |
bleed | +0.042 |
injuries | +0.041 |
hinges | +0.041 |
recovered | +0.040 |
Neg logit contributions | |
ang | -0.030 |
ee | -0.026 |
irling | -0.023 |
aw | -0.022 |
irl | -0.022 |
Token | with |
Token ID | 351 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
hinges | +0.081 |
department | +0.080 |
bru | +0.080 |
ranger | +0.079 |
bleed | +0.079 |
Neg logit contributions | |
ang | -0.136 |
irling | -0.118 |
ee | -0.113 |
ug | -0.113 |
irl | -0.110 |
Token | a |
Token ID | 257 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
ang | +0.045 |
irling | +0.040 |
ee | +0.040 |
irl | +0.039 |
ug | +0.036 |
Neg logit contributions | |
ranger | -0.046 |
department | -0.045 |
hinges | -0.044 |
bru | -0.044 |
onna | -0.044 |
Token | sad |
Token ID | 6507 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
keeping | +0.109 |
department | +0.098 |
hinges | +0.098 |
injuries | +0.097 |
edly | +0.097 |
Neg logit contributions | |
ang | -0.078 |
irling | -0.071 |
ee | -0.070 |
irl | -0.070 |
sorts | -0.061 |
Token | face |
Token ID | 1986 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
department | +0.105 |
hinges | +0.096 |
injuries | +0.095 |
keeping | +0.095 |
ranger | +0.088 |
Neg logit contributions | |
ang | -0.094 |
ee | -0.083 |
ug | -0.080 |
irl | -0.080 |
irling | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
ranger | +0.075 |
bleed | +0.073 |
bru | +0.072 |
hinges | +0.072 |
department | +0.070 |
Neg logit contributions | |
ang | -0.131 |
irling | -0.117 |
ee | -0.113 |
ug | -0.112 |
irl | -0.109 |
Token | back |
Token ID | 736 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
bru | +0.073 |
bleed | +0.071 |
department | +0.070 |
ranger | +0.070 |
hinges | +0.069 |
Neg logit contributions | |
ang | -0.090 |
ee | -0.081 |
irling | -0.076 |
ug | -0.073 |
irl | -0.072 |
Token | to |
Token ID | 284 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
bleed | +0.039 |
ranger | +0.039 |
bru | +0.039 |
sob | +0.037 |
hinges | +0.037 |
Neg logit contributions | |
ang | -0.082 |
irling | -0.073 |
ee | -0.073 |
irl | -0.070 |
ug | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | -3.08e-03 |
Pos logit contributions | |
sob | +0.049 |
bru | +0.049 |
bleed | +0.048 |
ranger | +0.048 |
injuries | +0.047 |
Neg logit contributions | |
ang | -0.060 |
irling | -0.053 |
ee | -0.050 |
ug | -0.049 |
ash | -0.049 |
Token | owl |
Token ID | 39610 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
bleed | +0.004 |
department | +0.004 |
ranger | +0.004 |
bru | +0.004 |
sob | +0.004 |
Neg logit contributions | |
ang | -0.007 |
irling | -0.006 |
ee | -0.006 |
irl | -0.006 |
ug | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -5.44e-03 |
Pos logit contributions | |
bleed | +0.067 |
bru | +0.065 |
ranger | +0.065 |
hinges | +0.064 |
department | +0.063 |
Neg logit contributions | |
ang | -0.114 |
irling | -0.101 |
ee | -0.098 |
irl | -0.094 |
ug | -0.094 |
Token | said |
Token ID | 531 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
bleed | +0.009 |
hinges | +0.009 |
bru | +0.009 |
ranger | +0.009 |
department | +0.009 |
Neg logit contributions | |
ang | -0.010 |
irling | -0.009 |
ee | -0.009 |
irl | -0.009 |
ug | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
bleed | +0.090 |
hinges | +0.089 |
ranger | +0.088 |
injuries | +0.086 |
bru | +0.085 |
Neg logit contributions | |
ang | -0.067 |
irling | -0.061 |
ee | -0.057 |
ug | -0.056 |
irl | -0.056 |
Token | " |
Token ID | 366 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
bleed | +0.078 |
hinges | +0.075 |
ranger | +0.075 |
bru | +0.074 |
injuries | +0.071 |
Neg logit contributions | |
ang | -0.094 |
irling | -0.085 |
ee | -0.080 |
irl | -0.078 |
ug | -0.077 |
Token | Thank |
Token ID | 10449 |
Feature activation | +8.12e-03 |
Pos logit contributions | |
ang | +0.024 |
irling | +0.022 |
ee | +0.021 |
irl | +0.021 |
ug | +0.020 |
Neg logit contributions | |
ranger | -0.016 |
bleed | -0.016 |
bru | -0.016 |
sob | -0.016 |
hinges | -0.015 |
Token | you |
Token ID | 345 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
ang | +0.012 |
irling | +0.011 |
irl | +0.010 |
ash | +0.010 |
angles | +0.010 |
Neg logit contributions | |
ranger | -0.017 |
mercy | -0.016 |
bleed | -0.016 |
sob | -0.016 |
bru | -0.016 |
Token | !". |
Token ID | 48220 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
ranger | +0.052 |
bleed | +0.051 |
bru | +0.050 |
hinges | +0.050 |
sob | +0.049 |
Neg logit contributions | |
ang | -0.062 |
irling | -0.056 |
ee | -0.053 |
irl | -0.052 |
ug | -0.050 |
Token | heard |
Token ID | 2982 |
Feature activation | -3.78e-03 |
Pos logit contributions | |
bleed | +0.044 |
hinges | +0.043 |
onna | +0.041 |
keeping | +0.040 |
bru | +0.040 |
Neg logit contributions | |
ang | -0.041 |
ee | -0.038 |
irling | -0.036 |
aw | -0.036 |
ug | -0.032 |
Token | Sara |
Token ID | 24799 |
Feature activation | -6.47e-02 |
Pos logit contributions | |
ranger | +0.007 |
bleed | +0.007 |
hinges | +0.007 |
bru | +0.007 |
department | +0.007 |
Neg logit contributions | |
ang | -0.006 |
irling | -0.006 |
ee | -0.005 |
irl | -0.005 |
ug | -0.005 |
Token | crying |
Token ID | 13774 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
ranger | +0.113 |
hinges | +0.112 |
bru | +0.110 |
department | +0.108 |
onna | +0.107 |
Neg logit contributions | |
ang | -0.121 |
ee | -0.106 |
irling | -0.105 |
ug | -0.097 |
irl | -0.097 |
Token | and |
Token ID | 290 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
onna | +0.059 |
department | +0.057 |
ranger | +0.056 |
hinges | +0.056 |
bru | +0.054 |
Neg logit contributions | |
ang | -0.044 |
aw | -0.037 |
ee | -0.036 |
ash | -0.034 |
irling | -0.034 |
Token | came |
Token ID | 1625 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
hinges | +0.063 |
bleed | +0.062 |
ranger | +0.062 |
bru | +0.060 |
department | +0.059 |
Neg logit contributions | |
ang | -0.082 |
irling | -0.072 |
ee | -0.071 |
irl | -0.066 |
ug | -0.066 |
Token | to |
Token ID | 284 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
bleed | +0.025 |
bru | +0.024 |
ranger | +0.024 |
ache | +0.023 |
hinges | +0.022 |
Neg logit contributions | |
ang | -0.048 |
irling | -0.043 |
ee | -0.042 |
ug | -0.040 |
irl | -0.040 |
Token | her |
Token ID | 607 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
ache | +0.115 |
bleed | +0.109 |
hinges | +0.109 |
driver | +0.107 |
department | +0.105 |
Neg logit contributions | |
ang | -0.054 |
ee | -0.050 |
these | -0.046 |
us | -0.043 |
rous | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
bleed | +0.125 |
bru | +0.123 |
recovered | +0.123 |
onna | +0.122 |
ache | +0.121 |
Neg logit contributions | |
ang | -0.107 |
irling | -0.089 |
irl | -0.083 |
ee | -0.081 |
sorts | -0.077 |
Token | She |
Token ID | 1375 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
bleed | +0.051 |
injuries | +0.051 |
mercy | +0.050 |
recovered | +0.049 |
department | +0.049 |
Neg logit contributions | |
ang | -0.060 |
ee | -0.049 |
aw | -0.046 |
irling | -0.046 |
uff | -0.042 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
bleed | +0.057 |
hinges | +0.054 |
bru | +0.054 |
ranger | +0.053 |
onna | +0.052 |
Neg logit contributions | |
ang | -0.065 |
irling | -0.057 |
ee | -0.055 |
irl | -0.052 |
ug | -0.050 |
Token | Sara |
Token ID | 24799 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
ang | +0.025 |
irling | +0.023 |
ee | +0.022 |
irl | +0.021 |
ug | +0.020 |
Neg logit contributions | |
ranger | -0.025 |
bleed | -0.024 |
bru | -0.024 |
sob | -0.024 |
hinges | -0.024 |