Token | behind |
Token ID | 2157 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
ashes | +0.013 |
ro | +0.012 |
clinic | +0.012 |
ashing | +0.012 |
rimp | +0.011 |
Neg logit contributions | |
shield | -0.007 |
journals | -0.007 |
journal | -0.007 |
pencil | -0.007 |
luck | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
shield | +0.019 |
journal | +0.018 |
journals | +0.018 |
luck | +0.018 |
pencil | +0.018 |
Neg logit contributions | |
ashes | -0.038 |
ro | -0.038 |
clinic | -0.037 |
cafe | -0.036 |
bank | -0.035 |
Token | out |
Token ID | 503 |
Feature activation | +3.47e-01 |
Pos logit contributions | |
ashes | +0.018 |
clinic | +0.018 |
ro | +0.018 |
cafe | +0.016 |
oodle | +0.016 |
Neg logit contributions | |
narrative | -0.012 |
journals | -0.012 |
shield | -0.012 |
luck | -0.012 |
crystal | -0.011 |
Token | of |
Token ID | 286 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
ashes | +0.016 |
ro | +0.016 |
clinic | +0.015 |
cafe | +0.014 |
rimp | +0.014 |
Neg logit contributions | |
shield | -0.018 |
journal | -0.017 |
journals | -0.017 |
pencil | -0.017 |
luck | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
ro | +0.155 |
clinic | +0.152 |
ashes | +0.152 |
cafe | +0.151 |
Gu | +0.143 |
Neg logit contributions | |
shield | -0.089 |
journals | -0.086 |
journal | -0.086 |
luck | -0.085 |
narrative | -0.081 |
Token | clinic |
Token ID | 15760 |
Feature activation | +9.57e+00 |
Pos logit contributions | |
shield | +0.014 |
pencil | +0.014 |
journal | +0.014 |
luck | +0.014 |
thumbs | +0.013 |
Neg logit contributions | |
ashes | -0.012 |
rimp | -0.012 |
Gu | -0.012 |
nes | -0.011 |
ro | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
ashes | +0.643 |
clinic | +0.608 |
ro | +0.606 |
Gu | +0.595 |
rimp | +0.588 |
Neg logit contributions | |
shield | -0.298 |
journal | -0.286 |
pencil | -0.281 |
luck | -0.280 |
journals | -0.276 |
Token | down |
Token ID | 866 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
shield | +0.028 |
journal | +0.027 |
journals | +0.027 |
pencil | +0.027 |
luck | +0.027 |
Neg logit contributions | |
ashes | -0.039 |
clinic | -0.037 |
ro | -0.037 |
Gu | -0.036 |
cafe | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
shield | +0.068 |
luck | +0.066 |
journal | +0.065 |
pencil | +0.065 |
journals | +0.062 |
Neg logit contributions | |
ashes | -0.172 |
clinic | -0.163 |
rimp | -0.161 |
ro | -0.160 |
Gu | -0.160 |
Token | street |
Token ID | 4675 |
Feature activation | +3.67e+00 |
Pos logit contributions | |
pencil | +0.009 |
shield | +0.008 |
luck | +0.008 |
journal | +0.008 |
thumbs | +0.008 |
Neg logit contributions | |
ashes | -0.008 |
Gu | -0.008 |
rimp | -0.008 |
ount | -0.008 |
ashing | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
ashes | +0.256 |
Gu | +0.241 |
rimp | +0.239 |
clinic | +0.237 |
ro | +0.234 |
Neg logit contributions | |
shield | -0.101 |
pencil | -0.098 |
luck | -0.098 |
journal | -0.096 |
journals | -0.088 |
Token | " |
Token ID | 366 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
ashes | +0.028 |
rimp | +0.026 |
anna | +0.026 |
clinic | +0.026 |
ashing | +0.025 |
Neg logit contributions | |
shield | -0.018 |
pencil | -0.017 |
journal | -0.016 |
magnets | -0.016 |
luck | -0.015 |
Token | We |
Token ID | 1135 |
Feature activation | +6.36e-01 |
Pos logit contributions | |
ashes | +0.022 |
clinic | +0.021 |
oodle | +0.021 |
bank | +0.021 |
anna | +0.020 |
Neg logit contributions | |
pencil | -0.018 |
shield | -0.018 |
journal | -0.018 |
journals | -0.018 |
magnets | -0.017 |
Token | live |
Token ID | 2107 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ashes | +0.039 |
clinic | +0.037 |
ro | +0.037 |
Gu | +0.035 |
nes | +0.035 |
Neg logit contributions | |
journals | -0.023 |
shield | -0.023 |
journal | -0.023 |
pencil | -0.022 |
magnets | -0.022 |
Token | on |
Token ID | 319 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
ashes | +0.295 |
clinic | +0.280 |
ro | +0.280 |
Gu | +0.269 |
rimp | +0.268 |
Neg logit contributions | |
shield | -0.157 |
journal | -0.150 |
pencil | -0.149 |
journals | -0.147 |
luck | -0.147 |
Token | Green |
Token ID | 3469 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
shield | +0.094 |
pencil | +0.089 |
journal | +0.088 |
journals | +0.086 |
luck | +0.085 |
Neg logit contributions | |
ashes | -0.279 |
ashing | -0.261 |
rimp | -0.259 |
Gu | -0.258 |
ro | -0.257 |
Token | Street |
Token ID | 3530 |
Feature activation | +8.63e+00 |
Pos logit contributions | |
shield | +0.044 |
journal | +0.044 |
pencil | +0.044 |
journals | +0.043 |
luck | +0.043 |
Neg logit contributions | |
ashes | -0.067 |
Gu | -0.064 |
ro | -0.063 |
clinic | -0.061 |
rimp | -0.060 |
Token | , |
Token ID | 11 |
Feature activation | +7.33e-02 |
Pos logit contributions | |
ashes | +0.626 |
Gu | +0.597 |
ashing | +0.582 |
clinic | +0.581 |
ro | +0.581 |
Neg logit contributions | |
luck | -0.215 |
shield | -0.208 |
journal | -0.204 |
pencil | -0.201 |
journals | -0.192 |
Token | number |
Token ID | 1271 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ashes | +0.004 |
clinic | +0.004 |
ro | +0.004 |
Gu | +0.004 |
rimp | +0.004 |
Neg logit contributions | |
shield | -0.003 |
journal | -0.003 |
pencil | -0.003 |
luck | -0.003 |
journals | -0.003 |
Token | 12 |
Token ID | 1105 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
shield | +0.044 |
pencil | +0.044 |
luck | +0.043 |
journal | +0.043 |
journals | +0.042 |
Neg logit contributions | |
ashes | -0.073 |
clinic | -0.069 |
ro | -0.068 |
Gu | -0.067 |
rimp | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | +4.89e-01 |
Pos logit contributions | |
ashes | +0.113 |
clinic | +0.106 |
ro | +0.105 |
Gu | +0.104 |
ashing | +0.102 |
Neg logit contributions | |
shield | -0.063 |
pencil | -0.062 |
luck | -0.062 |
journal | -0.060 |
journals | -0.060 |
Token | Can |
Token ID | 1680 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
ashes | +0.025 |
ashing | +0.024 |
bank | +0.023 |
eh | +0.022 |
clinic | +0.022 |
Neg logit contributions | |
journals | -0.022 |
pencil | -0.021 |
magnets | -0.021 |
narrative | -0.021 |
shield | -0.021 |
Token | I |
Token ID | 314 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
shield | +0.012 |
journal | +0.012 |
luck | +0.011 |
journals | +0.011 |
pencil | +0.011 |
Neg logit contributions | |
ashes | -0.011 |
clinic | -0.010 |
ro | -0.010 |
Gu | -0.010 |
rimp | -0.009 |
Token | live |
Token ID | 2107 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
ashes | +0.051 |
ro | +0.047 |
clinic | +0.047 |
rimp | +0.047 |
Gu | +0.046 |
Neg logit contributions | |
shield | -0.035 |
luck | -0.034 |
pencil | -0.033 |
journal | -0.033 |
journals | -0.032 |
Token | at |
Token ID | 379 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
ashes | +0.293 |
clinic | +0.276 |
ro | +0.275 |
Gu | +0.267 |
rimp | +0.266 |
Neg logit contributions | |
shield | -0.157 |
journal | -0.151 |
luck | -0.149 |
pencil | -0.148 |
journals | -0.147 |
Token | 10 |
Token ID | 838 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
shield | +0.021 |
journal | +0.021 |
luck | +0.021 |
pencil | +0.020 |
journals | +0.019 |
Neg logit contributions | |
ashes | -0.055 |
rimp | -0.051 |
ashing | -0.051 |
ro | -0.050 |
oodle | -0.050 |
Token | Green |
Token ID | 3469 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
pencil | +0.001 |
shield | +0.001 |
luck | +0.001 |
journals | +0.001 |
journal | +0.001 |
Neg logit contributions | |
ashes | -0.002 |
bank | -0.002 |
ro | -0.002 |
ashing | -0.002 |
Gu | -0.002 |
Token | Street |
Token ID | 3530 |
Feature activation | +8.56e+00 |
Pos logit contributions | |
pencil | +0.052 |
journal | +0.052 |
shield | +0.052 |
journals | +0.052 |
luck | +0.051 |
Neg logit contributions | |
ashes | -0.063 |
Gu | -0.060 |
ro | -0.058 |
bank | -0.057 |
Pe | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ashes | +0.614 |
Gu | +0.580 |
ashing | +0.564 |
anna | +0.564 |
clinic | +0.559 |
Neg logit contributions | |
luck | -0.224 |
journal | -0.216 |
shield | -0.214 |
pencil | -0.212 |
journals | -0.199 |
Token | Thank |
Token ID | 6952 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
pencil | +0.006 |
shield | +0.006 |
journals | +0.006 |
journal | +0.006 |
magnets | +0.006 |
Neg logit contributions | |
ashes | -0.007 |
ashing | -0.007 |
bank | -0.006 |
clinic | -0.006 |
ro | -0.006 |
Token | you |
Token ID | 345 |
Feature activation | +7.04e-01 |
Pos logit contributions | |
shield | +0.076 |
luck | +0.075 |
pencil | +0.075 |
journal | +0.075 |
journals | +0.074 |
Neg logit contributions | |
ashes | -0.085 |
clinic | -0.079 |
ro | -0.079 |
ashing | -0.074 |
bank | -0.074 |
Token | ." |
Token ID | 526 |
Feature activation | +3.45e+00 |
Pos logit contributions | |
ashes | +0.047 |
ro | +0.044 |
clinic | +0.044 |
Gu | +0.043 |
rimp | +0.042 |
Neg logit contributions | |
shield | -0.023 |
luck | -0.022 |
journal | -0.022 |
pencil | -0.022 |
journals | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -4.01e-01 |
Pos logit contributions | |
ashes | +0.189 |
clinic | +0.181 |
anna | +0.179 |
ashing | +0.177 |
ro | +0.176 |
Neg logit contributions | |
pencil | -0.148 |
magnets | -0.141 |
journal | -0.141 |
journals | -0.140 |
shield | -0.138 |
Token | hear |
Token ID | 3285 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ashes | +0.138 |
clinic | +0.133 |
ro | +0.127 |
cafe | +0.124 |
ount | +0.115 |
Neg logit contributions | |
journals | -0.152 |
journal | -0.148 |
luck | -0.146 |
shield | -0.146 |
pencil | -0.141 |
Token | s |
Token ID | 264 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
ashes | +0.085 |
ro | +0.081 |
clinic | +0.080 |
Gu | +0.079 |
ashing | +0.078 |
Neg logit contributions | |
shield | -0.042 |
pencil | -0.041 |
luck | -0.041 |
journals | -0.040 |
journal | -0.039 |
Token | irens |
Token ID | 42917 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ro | +0.028 |
ashes | +0.028 |
clinic | +0.027 |
bank | +0.027 |
ashing | +0.026 |
Neg logit contributions | |
pencil | -0.022 |
luck | -0.022 |
thumbs | -0.022 |
magnets | -0.022 |
journals | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
ashes | +0.093 |
clinic | +0.088 |
ro | +0.088 |
Gu | +0.085 |
rimp | +0.084 |
Neg logit contributions | |
shield | -0.057 |
journal | -0.054 |
pencil | -0.054 |
luck | -0.054 |
journals | -0.053 |
Token | h |
Token ID | 289 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
ashes | +0.034 |
clinic | +0.033 |
ro | +0.032 |
cafe | +0.031 |
Gu | +0.030 |
Neg logit contributions | |
journal | -0.030 |
shield | -0.030 |
journals | -0.030 |
luck | -0.030 |
pencil | -0.029 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.33e+00 |
Pos logit contributions | |
Gu | +0.058 |
mare | +0.046 |
orian | +0.044 |
clinic | +0.044 |
cafe | +0.044 |
Neg logit contributions | |
luck | -0.112 |
hetically | -0.109 |
streaming | -0.109 |
all | -0.107 |
shield | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
ashes | +0.537 |
clinic | +0.510 |
ro | +0.509 |
rimp | +0.489 |
Gu | +0.487 |
Neg logit contributions | |
shield | -0.283 |
journal | -0.273 |
pencil | -0.268 |
journals | -0.267 |
luck | -0.266 |
Token | They |
Token ID | 1119 |
Feature activation | +2.99e+00 |
Pos logit contributions | |
eh | +0.050 |
anna | +0.048 |
rimp | +0.047 |
rim | +0.047 |
nes | +0.046 |
Neg logit contributions | |
his | -0.029 |
hard | -0.029 |
mom | -0.028 |
thumbs | -0.028 |
all | -0.027 |
Token | go |
Token ID | 467 |
Feature activation | +4.77e-01 |
Pos logit contributions | |
ashes | +0.141 |
clinic | +0.134 |
ro | +0.131 |
cafe | +0.126 |
ount | +0.120 |
Neg logit contributions | |
journals | -0.155 |
journal | -0.151 |
shield | -0.147 |
magnets | -0.145 |
luck | -0.144 |
Token | to |
Token ID | 284 |
Feature activation | +7.34e-01 |
Pos logit contributions | |
rimp | +0.037 |
ashes | +0.035 |
ro | +0.034 |
Gu | +0.034 |
phony | +0.034 |
Neg logit contributions | |
on | -0.010 |
all | -0.010 |
flashlight | -0.009 |
hard | -0.009 |
shield | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | +8.71e-03 |
Pos logit contributions | |
ashes | +0.059 |
rimp | +0.057 |
ashing | +0.057 |
anna | +0.056 |
ro | +0.056 |
Neg logit contributions | |
shield | -0.013 |
flashlight | -0.012 |
pencil | -0.012 |
luck | -0.011 |
journal | -0.011 |
Token | They |
Token ID | 1119 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
ashes | +0.029 |
ashing | +0.027 |
ro | +0.027 |
bank | +0.026 |
anna | +0.026 |
Neg logit contributions | |
magnets | -0.025 |
journals | -0.025 |
narrative | -0.024 |
flashlight | -0.024 |
thumbs | -0.024 |
Token | have |
Token ID | 423 |
Feature activation | +8.01e-01 |
Pos logit contributions | |
ashes | +0.163 |
clinic | +0.156 |
ro | +0.153 |
rimp | +0.148 |
Gu | +0.147 |
Neg logit contributions | |
shield | -0.085 |
journal | -0.084 |
journals | -0.083 |
pencil | -0.081 |
luck | -0.080 |
Token | water |
Token ID | 1660 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
ashes | +0.046 |
clinic | +0.044 |
ro | +0.043 |
Gu | +0.041 |
rimp | +0.041 |
Neg logit contributions | |
shield | -0.032 |
journal | -0.032 |
journals | -0.031 |
pencil | -0.031 |
crystal | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | +3.62e-01 |
Pos logit contributions | |
ashes | +0.219 |
clinic | +0.207 |
ro | +0.207 |
rimp | +0.197 |
Gu | +0.196 |
Neg logit contributions | |
shield | -0.133 |
journal | -0.129 |
journals | -0.127 |
pencil | -0.126 |
luck | -0.125 |
Token | h |
Token ID | 289 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
ashes | +0.019 |
clinic | +0.019 |
ro | +0.018 |
cafe | +0.017 |
Gu | +0.017 |
Neg logit contributions | |
journal | -0.015 |
shield | -0.015 |
journals | -0.015 |
luck | -0.015 |
pencil | -0.015 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.25e+00 |
Pos logit contributions | |
Gu | +0.049 |
cafe | +0.040 |
mare | +0.040 |
clinic | +0.039 |
bank | +0.039 |
Neg logit contributions | |
luck | -0.115 |
magnets | -0.112 |
shield | -0.112 |
pencil | -0.111 |
thumbs | -0.110 |
Token | and |
Token ID | 290 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
ashes | +0.534 |
clinic | +0.508 |
ro | +0.504 |
Gu | +0.497 |
rimp | +0.489 |
Neg logit contributions | |
shield | -0.272 |
journal | -0.259 |
luck | -0.259 |
pencil | -0.258 |
journals | -0.252 |
Token | trucks |
Token ID | 13960 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
ashes | +0.019 |
clinic | +0.018 |
ro | +0.017 |
cafe | +0.016 |
Gu | +0.016 |
Neg logit contributions | |
journal | -0.014 |
journals | -0.014 |
shield | -0.014 |
luck | -0.014 |
crystal | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
shield | +0.064 |
journal | +0.062 |
journals | +0.061 |
pencil | +0.060 |
luck | +0.060 |
Neg logit contributions | |
ashes | -0.124 |
ro | -0.117 |
clinic | -0.117 |
rimp | -0.112 |
Gu | -0.111 |
Token | They |
Token ID | 1119 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
ashes | +0.028 |
ashing | +0.028 |
bank | +0.027 |
eh | +0.027 |
anna | +0.027 |
Neg logit contributions | |
magnets | -0.023 |
journals | -0.023 |
pencil | -0.022 |
what | -0.022 |
him | -0.022 |
Token | can |
Token ID | 460 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ashes | +0.154 |
clinic | +0.146 |
ro | +0.142 |
rimp | +0.135 |
cafe | +0.135 |
Neg logit contributions | |
journal | -0.095 |
journals | -0.094 |
shield | -0.094 |
pencil | -0.090 |
luck | -0.090 |
Token | boots |
Token ID | 14412 |
Feature activation | +5.37e-01 |
Pos logit contributions | |
journal | +0.002 |
journals | +0.002 |
luck | +0.002 |
crystal | +0.002 |
shield | +0.002 |
Neg logit contributions | |
ashes | -0.003 |
clinic | -0.003 |
ro | -0.003 |
cafe | -0.003 |
Gu | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
ashes | +0.034 |
ro | +0.032 |
clinic | +0.032 |
bank | +0.030 |
round | +0.030 |
Neg logit contributions | |
shield | -0.020 |
journal | -0.019 |
journals | -0.019 |
pencil | -0.018 |
crystal | -0.018 |
Token | They |
Token ID | 1119 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
all | +0.002 |
his | +0.002 |
magnets | +0.002 |
hard | +0.002 |
mom | +0.002 |
Neg logit contributions | |
nes | -0.002 |
rim | -0.002 |
rimp | -0.002 |
round | -0.002 |
ashes | -0.002 |
Token | have |
Token ID | 423 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
ashes | +0.115 |
clinic | +0.113 |
cafe | +0.107 |
ro | +0.106 |
aquarium | +0.101 |
Neg logit contributions | |
journal | -0.108 |
journals | -0.108 |
luck | -0.103 |
shield | -0.102 |
pencil | -0.101 |
Token | h |
Token ID | 289 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
ashes | +0.055 |
clinic | +0.052 |
ro | +0.052 |
Gu | +0.050 |
rimp | +0.050 |
Neg logit contributions | |
shield | -0.032 |
journal | -0.031 |
pencil | -0.031 |
journals | -0.031 |
luck | -0.030 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.23e+00 |
Pos logit contributions | |
Gu | +0.057 |
clinic | +0.052 |
cafe | +0.050 |
bank | +0.050 |
mare | +0.049 |
Neg logit contributions | |
luck | -0.100 |
thumbs | -0.097 |
shield | -0.096 |
streaming | -0.094 |
journals | -0.094 |
Token | and |
Token ID | 290 |
Feature activation | -3.18e-02 |
Pos logit contributions | |
ashes | +0.495 |
ro | +0.470 |
clinic | +0.468 |
rimp | +0.445 |
cafe | +0.441 |
Neg logit contributions | |
shield | -0.315 |
journal | -0.307 |
journals | -0.303 |
pencil | -0.299 |
luck | -0.296 |
Token | l |
Token ID | 300 |
Feature activation | +4.73e-01 |
Pos logit contributions | |
journal | +0.001 |
journals | +0.001 |
luck | +0.001 |
shield | +0.001 |
crystal | +0.001 |
Neg logit contributions | |
ashes | -0.002 |
clinic | -0.002 |
ro | -0.002 |
cafe | -0.002 |
Gu | -0.001 |
Token | adders |
Token ID | 45940 |
Feature activation | -9.30e-01 |
Pos logit contributions | |
ashes | +0.023 |
clinic | +0.021 |
ro | +0.021 |
Gu | +0.020 |
rimp | +0.020 |
Neg logit contributions | |
shield | -0.024 |
journal | -0.023 |
luck | -0.023 |
pencil | -0.023 |
journals | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
shield | +0.037 |
journal | +0.037 |
journals | +0.037 |
pencil | +0.036 |
luck | +0.035 |
Neg logit contributions | |
ashes | -0.054 |
ro | -0.053 |
clinic | -0.051 |
bank | -0.049 |
round | -0.048 |
Token | axes |
Token ID | 34197 |
Feature activation | -3.93e+00 |
Pos logit contributions | |
journal | +0.001 |
shield | +0.001 |
journals | +0.001 |
luck | +0.001 |
pencil | +0.001 |
Neg logit contributions | |
ashes | -0.002 |
clinic | -0.002 |
ro | -0.002 |
cafe | -0.002 |
Gu | -0.002 |
Token | The |
Token ID | 383 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
anna | +0.019 |
clinic | +0.018 |
round | +0.018 |
nes | +0.018 |
rimp | +0.018 |
Neg logit contributions | |
magnets | -0.021 |
journals | -0.021 |
journal | -0.020 |
pencil | -0.020 |
thumbs | -0.020 |
Token | trucks |
Token ID | 13960 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
ashes | +0.070 |
clinic | +0.066 |
ro | +0.065 |
cafe | +0.062 |
Gu | +0.061 |
Neg logit contributions | |
shield | -0.064 |
journal | -0.064 |
journals | -0.063 |
pencil | -0.063 |
luck | -0.062 |
Token | had |
Token ID | 550 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
journals | +0.078 |
journal | +0.076 |
shield | +0.073 |
luck | +0.071 |
pencil | +0.069 |
Neg logit contributions | |
ashes | -0.116 |
cafe | -0.111 |
ro | -0.110 |
clinic | -0.109 |
bank | -0.107 |
Token | big |
Token ID | 1263 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
shield | +0.074 |
journal | +0.073 |
journals | +0.071 |
pencil | +0.070 |
luck | +0.070 |
Neg logit contributions | |
ashes | -0.125 |
ro | -0.119 |
clinic | -0.118 |
Gu | -0.114 |
rimp | -0.113 |
Token | h |
Token ID | 289 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
journal | +0.088 |
journals | +0.086 |
shield | +0.085 |
narrative | +0.083 |
luck | +0.083 |
Neg logit contributions | |
ashes | -0.104 |
clinic | -0.098 |
ro | -0.096 |
cafe | -0.093 |
bank | -0.090 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.23e+00 |
Pos logit contributions | |
Gu | +0.041 |
ashes | +0.036 |
clinic | +0.036 |
cafe | +0.034 |
rimp | +0.033 |
Neg logit contributions | |
shield | -0.118 |
luck | -0.117 |
pencil | -0.113 |
magnets | -0.113 |
journal | -0.113 |
Token | and |
Token ID | 290 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ashes | +0.433 |
ro | +0.416 |
clinic | +0.408 |
bank | +0.398 |
cafe | +0.389 |
Neg logit contributions | |
journals | -0.377 |
journal | -0.372 |
shield | -0.372 |
luck | -0.354 |
pencil | -0.351 |
Token | sprayed |
Token ID | 35729 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
ashes | +0.007 |
ro | +0.007 |
clinic | +0.006 |
Gu | +0.006 |
cafe | +0.006 |
Neg logit contributions | |
journal | -0.005 |
shield | -0.005 |
journals | -0.005 |
luck | -0.005 |
narrative | -0.005 |
Token | water |
Token ID | 1660 |
Feature activation | +3.42e+00 |
Pos logit contributions | |
ashes | +0.155 |
ro | +0.142 |
clinic | +0.133 |
cafe | +0.133 |
bank | +0.132 |
Neg logit contributions | |
journal | -0.137 |
shield | -0.136 |
journals | -0.133 |
pencil | -0.129 |
luck | -0.125 |
Token | on |
Token ID | 319 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
ashes | +0.233 |
ro | +0.222 |
clinic | +0.217 |
bank | +0.209 |
cafe | +0.209 |
Neg logit contributions | |
shield | -0.107 |
journal | -0.106 |
journals | -0.105 |
pencil | -0.102 |
luck | -0.099 |
Token | the |
Token ID | 262 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
shield | +0.144 |
journal | +0.141 |
luck | +0.140 |
journals | +0.138 |
pencil | +0.135 |
Neg logit contributions | |
ashes | -0.212 |
ro | -0.203 |
clinic | -0.199 |
cafe | -0.191 |
Gu | -0.190 |
Token | fire |
Token ID | 2046 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
ashes | +0.034 |
clinic | +0.033 |
ro | +0.032 |
Gu | +0.031 |
rimp | +0.031 |
Neg logit contributions | |
shield | -0.027 |
journal | -0.026 |
pencil | -0.026 |
journals | -0.026 |
luck | -0.026 |
Token | with |
Token ID | 351 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
ashes | +0.115 |
clinic | +0.110 |
ro | +0.110 |
cafe | +0.103 |
bank | +0.102 |
Neg logit contributions | |
shield | -0.083 |
journals | -0.082 |
journal | -0.082 |
pencil | -0.080 |
luck | -0.080 |
Token | water |
Token ID | 1660 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
ashes | +0.092 |
clinic | +0.089 |
cafe | +0.083 |
ro | +0.082 |
bank | +0.080 |
Neg logit contributions | |
journal | -0.092 |
narrative | -0.090 |
journals | -0.089 |
pencil | -0.086 |
shield | -0.085 |
Token | and |
Token ID | 290 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
ashes | +0.204 |
ro | +0.194 |
clinic | +0.193 |
cafe | +0.184 |
bank | +0.184 |
Neg logit contributions | |
journals | -0.145 |
journal | -0.143 |
shield | -0.143 |
pencil | -0.140 |
luck | -0.138 |
Token | h |
Token ID | 289 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ashes | +0.019 |
clinic | +0.019 |
ro | +0.018 |
cafe | +0.018 |
bank | +0.017 |
Neg logit contributions | |
journal | -0.015 |
luck | -0.015 |
journals | -0.015 |
shield | -0.015 |
narrative | -0.014 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.22e+00 |
Pos logit contributions | |
Gu | +0.053 |
cafe | +0.045 |
clinic | +0.044 |
mare | +0.043 |
orian | +0.042 |
Neg logit contributions | |
luck | -0.108 |
shield | -0.105 |
streaming | -0.105 |
thumbs | -0.102 |
hetically | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
ashes | +0.512 |
ro | +0.487 |
clinic | +0.486 |
rimp | +0.462 |
bank | +0.461 |
Neg logit contributions | |
shield | -0.297 |
journal | -0.291 |
journals | -0.288 |
pencil | -0.281 |
luck | -0.280 |
Token | They |
Token ID | 1119 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
anna | +0.029 |
ashes | +0.028 |
clinic | +0.028 |
nes | +0.027 |
rimp | +0.027 |
Neg logit contributions | |
magnets | -0.024 |
thumbs | -0.023 |
pencil | -0.023 |
flashlight | -0.023 |
shield | -0.023 |
Token | saw |
Token ID | 2497 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
clinic | +0.115 |
ashes | +0.115 |
cafe | +0.114 |
ro | +0.111 |
Gu | +0.108 |
Neg logit contributions | |
journals | -0.131 |
journal | -0.128 |
luck | -0.126 |
shield | -0.118 |
magnets | -0.114 |
Token | their |
Token ID | 511 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
ashes | +0.056 |
clinic | +0.052 |
cafe | +0.051 |
Gu | +0.050 |
ro | +0.050 |
Neg logit contributions | |
journal | -0.039 |
journals | -0.038 |
shield | -0.038 |
pencil | -0.037 |
luck | -0.036 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ashes | +0.139 |
clinic | +0.131 |
ro | +0.121 |
cafe | +0.121 |
ashing | +0.116 |
Neg logit contributions | |
journals | -0.101 |
pencil | -0.101 |
journal | -0.098 |
shield | -0.098 |
narrative | -0.096 |
Token | out |
Token ID | 503 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
ashes | +0.122 |
clinic | +0.113 |
ro | +0.111 |
rimp | +0.108 |
ount | +0.108 |
Neg logit contributions | |
journal | -0.081 |
shield | -0.079 |
journals | -0.078 |
pencil | -0.078 |
crystal | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +5.24e-01 |
Pos logit contributions | |
ashes | +0.037 |
ro | +0.034 |
clinic | +0.034 |
cafe | +0.033 |
ount | +0.032 |
Neg logit contributions | |
journal | -0.031 |
shield | -0.031 |
pencil | -0.031 |
journals | -0.030 |
crystal | -0.028 |
Token | They |
Token ID | 1119 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
ashes | +0.030 |
anna | +0.030 |
nes | +0.029 |
ashing | +0.029 |
rimp | +0.029 |
Neg logit contributions | |
magnets | -0.019 |
his | -0.019 |
pencil | -0.018 |
shield | -0.018 |
all | -0.017 |
Token | had |
Token ID | 550 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
ashes | +0.132 |
ro | +0.128 |
clinic | +0.126 |
cafe | +0.123 |
rimp | +0.119 |
Neg logit contributions | |
journal | -0.114 |
journals | -0.112 |
luck | -0.110 |
shield | -0.109 |
pencil | -0.102 |
Token | h |
Token ID | 289 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
shield | +0.084 |
journal | +0.082 |
pencil | +0.080 |
journals | +0.080 |
luck | +0.079 |
Neg logit contributions | |
ashes | -0.124 |
clinic | -0.117 |
ro | -0.117 |
Gu | -0.113 |
rimp | -0.112 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
Gu | +0.050 |
clinic | +0.044 |
mare | +0.042 |
cafe | +0.041 |
ashes | +0.040 |
Neg logit contributions | |
luck | -0.110 |
shield | -0.107 |
thumbs | -0.107 |
magnets | -0.104 |
streaming | -0.104 |
Token | and |
Token ID | 290 |
Feature activation | +3.54e-01 |
Pos logit contributions | |
ashes | +0.484 |
ro | +0.459 |
clinic | +0.457 |
rimp | +0.435 |
cafe | +0.432 |
Neg logit contributions | |
shield | -0.321 |
journal | -0.313 |
journals | -0.310 |
pencil | -0.305 |
luck | -0.302 |
Token | l |
Token ID | 300 |
Feature activation | +5.20e-01 |
Pos logit contributions | |
ashes | +0.020 |
ro | +0.019 |
clinic | +0.018 |
cafe | +0.017 |
Gu | +0.017 |
Neg logit contributions | |
journal | -0.015 |
shield | -0.015 |
journals | -0.014 |
luck | -0.014 |
crystal | -0.014 |
Token | adders |
Token ID | 45940 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
ashes | +0.027 |
clinic | +0.025 |
ro | +0.025 |
Gu | +0.024 |
rimp | +0.024 |
Neg logit contributions | |
shield | -0.024 |
journal | -0.024 |
pencil | -0.023 |
luck | -0.023 |
journals | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
journals | +0.040 |
journal | +0.040 |
shield | +0.040 |
pencil | +0.038 |
luck | +0.037 |
Neg logit contributions | |
ashes | -0.055 |
ro | -0.053 |
clinic | -0.052 |
bank | -0.050 |
round | -0.049 |
Token | They |
Token ID | 1119 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
anna | +0.036 |
ashes | +0.036 |
bank | +0.036 |
ashing | +0.035 |
rim | +0.034 |
Neg logit contributions | |
magnets | -0.018 |
pencil | -0.018 |
his | -0.018 |
mom | -0.017 |
what | -0.017 |
Token | They |
Token ID | 1119 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
eh | +0.030 |
rim | +0.029 |
nes | +0.028 |
anna | +0.027 |
ital | +0.027 |
Neg logit contributions | |
eyes | -0.015 |
mom | -0.014 |
Mom | -0.014 |
all | -0.013 |
magnets | -0.013 |
Token | use |
Token ID | 779 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
ashes | +0.135 |
clinic | +0.127 |
ro | +0.124 |
cafe | +0.118 |
rimp | +0.113 |
Neg logit contributions | |
journals | -0.124 |
journal | -0.121 |
luck | -0.117 |
shield | -0.116 |
pencil | -0.116 |
Token | buckets |
Token ID | 38674 |
Feature activation | +4.48e-01 |
Pos logit contributions | |
ashes | +0.033 |
clinic | +0.031 |
ro | +0.031 |
bank | +0.029 |
cafe | +0.029 |
Neg logit contributions | |
narrative | -0.029 |
journal | -0.027 |
journals | -0.027 |
shield | -0.026 |
pencil | -0.025 |
Token | and |
Token ID | 290 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
ashes | +0.023 |
ro | +0.022 |
clinic | +0.019 |
ashing | +0.019 |
cafe | +0.019 |
Neg logit contributions | |
journal | -0.022 |
journals | -0.022 |
shield | -0.022 |
narrative | -0.021 |
pencil | -0.020 |
Token | h |
Token ID | 289 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
ashes | +0.022 |
ro | +0.021 |
clinic | +0.020 |
rimp | +0.019 |
cafe | +0.019 |
Neg logit contributions | |
journal | -0.016 |
journals | -0.016 |
shield | -0.015 |
luck | -0.015 |
crystal | -0.015 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.16e+00 |
Pos logit contributions | |
Gu | +0.045 |
clinic | +0.040 |
cafe | +0.039 |
ashes | +0.039 |
mare | +0.037 |
Neg logit contributions | |
luck | -0.109 |
shield | -0.107 |
thumbs | -0.105 |
magnets | -0.103 |
pencil | -0.103 |
Token | to |
Token ID | 284 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
ashes | +0.476 |
ro | +0.454 |
clinic | +0.446 |
bank | +0.423 |
rimp | +0.423 |
Neg logit contributions | |
shield | -0.330 |
journal | -0.325 |
journals | -0.323 |
pencil | -0.311 |
luck | -0.310 |
Token | pour |
Token ID | 12797 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
ashes | +0.029 |
clinic | +0.027 |
ro | +0.027 |
rimp | +0.025 |
Gu | +0.025 |
Neg logit contributions | |
shield | -0.023 |
journal | -0.022 |
journals | -0.022 |
luck | -0.022 |
pencil | -0.022 |
Token | water |
Token ID | 1660 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
ashes | +0.201 |
ro | +0.179 |
cafe | +0.166 |
clinic | +0.162 |
rimp | +0.161 |
Neg logit contributions | |
journal | -0.188 |
shield | -0.186 |
journals | -0.184 |
pencil | -0.175 |
narrative | -0.169 |
Token | on |
Token ID | 319 |
Feature activation | -3.79e+00 |
Pos logit contributions | |
ashes | +0.236 |
ro | +0.224 |
clinic | +0.218 |
rimp | +0.211 |
cafe | +0.210 |
Neg logit contributions | |
shield | -0.103 |
journal | -0.102 |
journals | -0.101 |
pencil | -0.098 |
luck | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
shield | +0.126 |
journal | +0.124 |
journals | +0.121 |
luck | +0.120 |
pencil | +0.117 |
Neg logit contributions | |
ashes | -0.243 |
ro | -0.234 |
clinic | -0.233 |
cafe | -0.223 |
bank | -0.221 |
Token | them |
Token ID | 606 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
ashes | +0.044 |
clinic | +0.043 |
cafe | +0.040 |
bank | +0.039 |
aquarium | +0.038 |
Neg logit contributions | |
journal | -0.036 |
journals | -0.035 |
pencil | -0.035 |
narrative | -0.034 |
shield | -0.033 |
Token | use |
Token ID | 779 |
Feature activation | +5.95e-01 |
Pos logit contributions | |
ashes | +0.197 |
clinic | +0.186 |
cafe | +0.178 |
ro | +0.176 |
bank | +0.173 |
Neg logit contributions | |
journals | -0.185 |
journal | -0.181 |
narrative | -0.176 |
pencil | -0.174 |
shield | -0.174 |
Token | water |
Token ID | 1660 |
Feature activation | +3.52e+00 |
Pos logit contributions | |
ashes | +0.033 |
clinic | +0.032 |
ro | +0.031 |
bank | +0.030 |
cafe | +0.029 |
Neg logit contributions | |
journal | -0.024 |
narrative | -0.024 |
journals | -0.024 |
shield | -0.023 |
pencil | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
ashes | +0.194 |
ro | +0.187 |
clinic | +0.173 |
bank | +0.171 |
cafe | +0.167 |
Neg logit contributions | |
journals | -0.163 |
journal | -0.161 |
shield | -0.155 |
pencil | -0.150 |
narrative | -0.149 |
Token | h |
Token ID | 289 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
journal | +0.005 |
journals | +0.005 |
luck | +0.005 |
narrative | +0.005 |
shield | +0.005 |
Neg logit contributions | |
ashes | -0.006 |
ro | -0.006 |
clinic | -0.006 |
cafe | -0.005 |
bank | -0.005 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.15e+00 |
Pos logit contributions | |
Gu | +0.050 |
cafe | +0.042 |
clinic | +0.042 |
mare | +0.042 |
orian | +0.040 |
Neg logit contributions | |
luck | -0.108 |
shield | -0.105 |
streaming | -0.104 |
thumbs | -0.104 |
magnets | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
ashes | +0.490 |
ro | +0.467 |
clinic | +0.460 |
bank | +0.440 |
rimp | +0.437 |
Neg logit contributions | |
shield | -0.316 |
journal | -0.312 |
journals | -0.309 |
pencil | -0.296 |
luck | -0.294 |
Token | l |
Token ID | 300 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
journal | +0.005 |
journals | +0.005 |
luck | +0.005 |
narrative | +0.005 |
shield | +0.005 |
Neg logit contributions | |
ashes | -0.006 |
ro | -0.006 |
clinic | -0.006 |
cafe | -0.005 |
bank | -0.005 |
Token | adders |
Token ID | 45940 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
ashes | +0.023 |
clinic | +0.022 |
ro | +0.022 |
rimp | +0.020 |
Gu | +0.020 |
Neg logit contributions | |
shield | -0.025 |
journal | -0.025 |
pencil | -0.024 |
journals | -0.024 |
luck | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
journals | +0.039 |
journal | +0.039 |
shield | +0.038 |
pencil | +0.037 |
luck | +0.036 |
Neg logit contributions | |
ashes | -0.059 |
ro | -0.057 |
clinic | -0.055 |
bank | -0.054 |
ashing | -0.052 |
Token | They |
Token ID | 1119 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
rimp | +0.003 |
nes | +0.003 |
rim | +0.003 |
anna | +0.003 |
mare | +0.003 |
Neg logit contributions | |
magnets | -0.002 |
journals | -0.002 |
bicycles | -0.002 |
all | -0.002 |
thumbs | -0.002 |
Token | with |
Token ID | 351 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ashes | +0.234 |
clinic | +0.222 |
ro | +0.222 |
Gu | +0.215 |
rimp | +0.214 |
Neg logit contributions | |
shield | -0.111 |
journal | -0.106 |
luck | -0.104 |
pencil | -0.104 |
journals | -0.103 |
Token | big |
Token ID | 1263 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
ashes | +0.094 |
clinic | +0.092 |
cafe | +0.089 |
ro | +0.085 |
bank | +0.085 |
Neg logit contributions | |
pencil | -0.064 |
journals | -0.064 |
narrative | -0.063 |
journal | -0.062 |
shield | -0.061 |
Token | trucks |
Token ID | 13960 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
journal | +0.097 |
shield | +0.097 |
journals | +0.096 |
pencil | +0.096 |
crystal | +0.094 |
Neg logit contributions | |
ashes | -0.140 |
clinic | -0.134 |
ro | -0.129 |
cafe | -0.127 |
bank | -0.122 |
Token | and |
Token ID | 290 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
shield | +0.076 |
journals | +0.075 |
journal | +0.074 |
pencil | +0.074 |
luck | +0.073 |
Neg logit contributions | |
ashes | -0.129 |
clinic | -0.122 |
ro | -0.121 |
rimp | -0.116 |
cafe | -0.116 |
Token | h |
Token ID | 289 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ashes | +0.011 |
clinic | +0.011 |
ro | +0.011 |
Gu | +0.010 |
rimp | +0.010 |
Neg logit contributions | |
shield | -0.008 |
journal | -0.008 |
luck | -0.007 |
journals | -0.007 |
pencil | -0.007 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.14e+00 |
Pos logit contributions | |
Gu | +0.051 |
clinic | +0.041 |
cafe | +0.040 |
mare | +0.040 |
ashes | +0.039 |
Neg logit contributions | |
luck | -0.109 |
shield | -0.107 |
hetically | -0.104 |
magnets | -0.103 |
sword | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
ashes | +0.519 |
ro | +0.496 |
clinic | +0.493 |
rimp | +0.472 |
bank | +0.471 |
Neg logit contributions | |
shield | -0.276 |
journal | -0.270 |
journals | -0.267 |
pencil | -0.262 |
luck | -0.261 |
Token | They |
Token ID | 1119 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
ashes | +0.008 |
ashing | +0.007 |
rim | +0.007 |
nes | +0.006 |
clinic | +0.006 |
Neg logit contributions | |
magnets | -0.005 |
his | -0.005 |
shield | -0.005 |
journal | -0.005 |
pencil | -0.005 |
Token | sprayed |
Token ID | 35729 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
ashes | +0.140 |
ro | +0.134 |
clinic | +0.133 |
cafe | +0.128 |
nes | +0.123 |
Neg logit contributions | |
luck | -0.118 |
journal | -0.117 |
journals | -0.116 |
shield | -0.113 |
crystal | -0.109 |
Token | water |
Token ID | 1660 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
ashes | +0.150 |
ro | +0.140 |
clinic | +0.130 |
cafe | +0.128 |
bank | +0.128 |
Neg logit contributions | |
journal | -0.119 |
shield | -0.119 |
journals | -0.116 |
pencil | -0.112 |
luck | -0.110 |
Token | on |
Token ID | 319 |
Feature activation | -3.88e+00 |
Pos logit contributions | |
ashes | +0.231 |
ro | +0.220 |
clinic | +0.209 |
bank | +0.204 |
ashing | +0.204 |
Neg logit contributions | |
journals | -0.114 |
journal | -0.114 |
shield | -0.113 |
pencil | -0.107 |
luck | -0.105 |
Token | people |
Token ID | 661 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
ashes | +0.045 |
clinic | +0.041 |
cafe | +0.040 |
oodle | +0.036 |
ro | +0.036 |
Neg logit contributions | |
journals | -0.050 |
journal | -0.049 |
shield | -0.048 |
pencil | -0.047 |
crystal | -0.045 |
Token | with |
Token ID | 351 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
ashes | +0.028 |
clinic | +0.027 |
cafe | +0.026 |
ro | +0.026 |
ashing | +0.026 |
Neg logit contributions | |
shield | -0.023 |
journals | -0.023 |
journal | -0.022 |
luck | -0.022 |
pencil | -0.022 |
Token | water |
Token ID | 1660 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
ashes | +0.085 |
clinic | +0.084 |
cafe | +0.081 |
ro | +0.077 |
pasture | +0.075 |
Neg logit contributions | |
journal | -0.084 |
journals | -0.082 |
narrative | -0.080 |
shield | -0.078 |
pencil | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | +1.94e-01 |
Pos logit contributions | |
ashes | +0.193 |
ro | +0.185 |
clinic | +0.180 |
cafe | +0.175 |
bank | +0.173 |
Neg logit contributions | |
journals | -0.143 |
shield | -0.141 |
journal | -0.141 |
pencil | -0.134 |
luck | -0.131 |
Token | h |
Token ID | 289 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ashes | +0.010 |
ro | +0.010 |
clinic | +0.010 |
cafe | +0.009 |
bank | +0.009 |
Neg logit contributions | |
journal | -0.008 |
shield | -0.008 |
journals | -0.008 |
luck | -0.008 |
pencil | -0.008 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.10e+00 |
Pos logit contributions | |
Gu | +0.063 |
clinic | +0.055 |
mare | +0.055 |
cafe | +0.054 |
orian | +0.052 |
Neg logit contributions | |
luck | -0.087 |
streaming | -0.082 |
shield | -0.081 |
thumbs | -0.081 |
hetically | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
ashes | +0.484 |
ro | +0.461 |
clinic | +0.456 |
bank | +0.437 |
cafe | +0.435 |
Neg logit contributions | |
shield | -0.319 |
journal | -0.314 |
journals | -0.312 |
pencil | -0.298 |
luck | -0.294 |
Token | They |
Token ID | 1119 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
anna | +0.020 |
rim | +0.019 |
rimp | +0.019 |
mare | +0.019 |
clinic | +0.018 |
Neg logit contributions | |
magnets | -0.019 |
journals | -0.019 |
pencil | -0.019 |
journal | -0.019 |
narrative | -0.018 |
Token | were |
Token ID | 547 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
cafe | +0.104 |
rimp | +0.097 |
ashes | +0.096 |
ro | +0.096 |
clinic | +0.093 |
Neg logit contributions | |
journals | -0.130 |
journal | -0.125 |
luck | -0.122 |
shield | -0.120 |
magnets | -0.112 |
Token | trying |
Token ID | 2111 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
shield | +0.077 |
luck | +0.075 |
journal | +0.074 |
pencil | +0.073 |
journals | +0.072 |
Neg logit contributions | |
ashes | -0.115 |
ro | -0.110 |
clinic | -0.108 |
Gu | -0.106 |
rimp | -0.103 |
Token | to |
Token ID | 284 |
Feature activation | +5.31e-01 |
Pos logit contributions | |
ashes | +0.082 |
clinic | +0.072 |
ro | +0.071 |
bank | +0.070 |
ashing | +0.070 |
Neg logit contributions | |
journals | -0.083 |
journal | -0.083 |
shield | -0.083 |
narrative | -0.081 |
crystal | -0.079 |
Token | yellow |
Token ID | 7872 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
ashes | +0.015 |
ro | +0.014 |
clinic | +0.014 |
Gu | +0.014 |
rimp | +0.013 |
Neg logit contributions | |
journal | -0.011 |
shield | -0.011 |
luck | -0.011 |
journals | -0.011 |
pencil | -0.011 |
Token | coats |
Token ID | 30720 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
ashes | +0.074 |
ro | +0.070 |
clinic | +0.070 |
cafe | +0.066 |
round | +0.064 |
Neg logit contributions | |
journal | -0.059 |
journals | -0.058 |
shield | -0.058 |
luck | -0.056 |
narrative | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
ashes | +0.117 |
ro | +0.113 |
clinic | +0.110 |
bank | +0.105 |
round | +0.104 |
Neg logit contributions | |
shield | -0.070 |
journals | -0.069 |
journal | -0.069 |
luck | -0.066 |
pencil | -0.065 |
Token | big |
Token ID | 1263 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ashes | +0.011 |
ro | +0.011 |
clinic | +0.010 |
Gu | +0.010 |
cafe | +0.010 |
Neg logit contributions | |
journal | -0.008 |
journals | -0.008 |
shield | -0.008 |
luck | -0.008 |
crystal | -0.008 |
Token | h |
Token ID | 289 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
journal | +0.078 |
journals | +0.077 |
narrative | +0.075 |
crystal | +0.074 |
shield | +0.074 |
Neg logit contributions | |
ashes | -0.100 |
clinic | -0.095 |
ro | -0.091 |
cafe | -0.089 |
te | -0.086 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.10e+00 |
Pos logit contributions | |
Gu | +0.039 |
ashes | +0.034 |
cafe | +0.032 |
clinic | +0.032 |
rimp | +0.031 |
Neg logit contributions | |
luck | -0.117 |
shield | -0.116 |
pencil | -0.112 |
thumbs | -0.112 |
magnets | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
ashes | +0.528 |
ro | +0.503 |
clinic | +0.500 |
rimp | +0.478 |
bank | +0.474 |
Neg logit contributions | |
shield | -0.269 |
journal | -0.262 |
journals | -0.257 |
pencil | -0.252 |
luck | -0.252 |
Token | They |
Token ID | 1119 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
rimp | +0.015 |
anna | +0.014 |
round | +0.014 |
rim | +0.014 |
bank | +0.014 |
Neg logit contributions | |
magnets | -0.012 |
pencil | -0.012 |
shield | -0.011 |
all | -0.011 |
thumbs | -0.011 |
Token | looked |
Token ID | 3114 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
ashes | +0.103 |
ro | +0.100 |
clinic | +0.099 |
cafe | +0.096 |
nes | +0.093 |
Neg logit contributions | |
journal | -0.120 |
luck | -0.116 |
journals | -0.116 |
shield | -0.110 |
narrative | -0.107 |
Token | like |
Token ID | 588 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
pencil | +0.020 |
journal | +0.020 |
shield | +0.020 |
journals | +0.019 |
crystal | +0.019 |
Neg logit contributions | |
ashes | -0.028 |
ashing | -0.026 |
clinic | -0.025 |
cafe | -0.025 |
oodle | -0.024 |
Token | her |
Token ID | 607 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
ashes | +0.090 |
clinic | +0.087 |
ro | +0.084 |
Gu | +0.082 |
cafe | +0.081 |
Neg logit contributions | |
journal | -0.071 |
shield | -0.070 |
journals | -0.069 |
pencil | -0.069 |
luck | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -2.10e-03 |
Pos logit contributions | |
ashes | +0.022 |
clinic | +0.021 |
ro | +0.021 |
cafe | +0.021 |
oodle | +0.020 |
Neg logit contributions | |
shield | -0.013 |
journal | -0.012 |
pencil | -0.012 |
journals | -0.012 |
narrative | -0.012 |
Token | fire |
Token ID | 2046 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
journals | +0.000 |
shield | +0.000 |
pencil | +0.000 |
luck | +0.000 |
journal | +0.000 |
Neg logit contributions | |
ashes | -0.000 |
clinic | -0.000 |
ro | -0.000 |
cafe | -0.000 |
rimp | -0.000 |
Token | with |
Token ID | 351 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
ashes | +0.099 |
ro | +0.094 |
clinic | +0.094 |
bank | +0.089 |
cafe | +0.089 |
Neg logit contributions | |
shield | -0.059 |
journals | -0.059 |
journal | -0.058 |
pencil | -0.057 |
luck | -0.056 |
Token | their |
Token ID | 511 |
Feature activation | +2.48e+00 |
Pos logit contributions | |
ashes | +0.091 |
clinic | +0.088 |
ro | +0.087 |
bank | +0.084 |
cafe | +0.084 |
Neg logit contributions | |
journal | -0.057 |
narrative | -0.055 |
journals | -0.055 |
shield | -0.055 |
pencil | -0.054 |
Token | h |
Token ID | 289 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ashes | +0.143 |
clinic | +0.138 |
ro | +0.135 |
cafe | +0.127 |
bank | +0.125 |
Neg logit contributions | |
journals | -0.092 |
journal | -0.092 |
shield | -0.091 |
pencil | -0.090 |
narrative | -0.090 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.08e+00 |
Pos logit contributions | |
Gu | +0.047 |
ashes | +0.041 |
rimp | +0.038 |
clinic | +0.037 |
cafe | +0.037 |
Neg logit contributions | |
shield | -0.103 |
luck | -0.103 |
journal | -0.099 |
thumbs | -0.099 |
flashlight | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
ashes | +0.490 |
ro | +0.473 |
clinic | +0.471 |
bank | +0.450 |
cafe | +0.442 |
Neg logit contributions | |
shield | -0.293 |
journals | -0.293 |
journal | -0.290 |
luck | -0.281 |
pencil | -0.281 |
Token | Everyone |
Token ID | 11075 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
magnets | +0.006 |
his | +0.006 |
journals | +0.006 |
pencil | +0.006 |
journal | +0.006 |
Neg logit contributions | |
ashes | -0.010 |
ashing | -0.009 |
clinic | -0.009 |
rimp | -0.009 |
rim | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
ashes | +0.086 |
clinic | +0.081 |
ro | +0.081 |
rimp | +0.079 |
Gu | +0.078 |
Neg logit contributions | |
shield | -0.054 |
journal | -0.051 |
luck | -0.051 |
journals | -0.051 |
pencil | -0.050 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
all | +0.001 |
luck | +0.001 |
on | +0.001 |
hard | +0.001 |
shield | +0.001 |
Neg logit contributions | |
ashes | -0.001 |
ro | -0.001 |
Gu | -0.001 |
round | -0.001 |
clinic | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | -4.54e-01 |
Pos logit contributions | |
ashes | +0.077 |
clinic | +0.073 |
ro | +0.073 |
Gu | +0.070 |
rimp | +0.070 |
Neg logit contributions | |
shield | -0.044 |
journal | -0.043 |
pencil | -0.042 |
journals | -0.042 |
luck | -0.042 |
Token | out |
Token ID | 503 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
ashes | +0.111 |
clinic | +0.102 |
ro | +0.101 |
rimp | +0.099 |
ount | +0.098 |
Neg logit contributions | |
journal | -0.077 |
shield | -0.075 |
journals | -0.074 |
pencil | -0.073 |
narrative | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
ashes | +0.023 |
clinic | +0.021 |
ro | +0.021 |
cafe | +0.020 |
rimp | +0.020 |
Neg logit contributions | |
journal | -0.021 |
shield | -0.021 |
pencil | -0.021 |
journals | -0.020 |
narrative | -0.019 |
Token | They |
Token ID | 1119 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
magnets | +0.006 |
his | +0.006 |
all | +0.006 |
shield | +0.006 |
pencil | +0.005 |
Neg logit contributions | |
ashes | -0.011 |
ashing | -0.010 |
rim | -0.010 |
nes | -0.009 |
rimp | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ashes | +0.113 |
ro | +0.109 |
clinic | +0.108 |
cafe | +0.105 |
rimp | +0.102 |
Neg logit contributions | |
journal | -0.091 |
journals | -0.089 |
luck | -0.088 |
shield | -0.086 |
pencil | -0.082 |
Token | h |
Token ID | 289 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
shield | +0.093 |
journal | +0.089 |
pencil | +0.088 |
luck | +0.088 |
journals | +0.088 |
Neg logit contributions | |
ashes | -0.153 |
clinic | -0.145 |
ro | -0.144 |
Gu | -0.139 |
rimp | -0.139 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.05e+00 |
Pos logit contributions | |
Gu | +0.041 |
clinic | +0.037 |
cafe | +0.034 |
mare | +0.034 |
ashes | +0.032 |
Neg logit contributions | |
luck | -0.099 |
thumbs | -0.097 |
shield | -0.097 |
magnets | -0.095 |
streaming | -0.094 |
Token | and |
Token ID | 290 |
Feature activation | +1.37e-02 |
Pos logit contributions | |
ashes | +0.491 |
ro | +0.466 |
clinic | +0.465 |
rimp | +0.444 |
bank | +0.440 |
Neg logit contributions | |
shield | -0.300 |
journal | -0.292 |
journals | -0.288 |
pencil | -0.284 |
luck | -0.282 |
Token | l |
Token ID | 300 |
Feature activation | +3.97e-01 |
Pos logit contributions | |
ashes | +0.001 |
ro | +0.001 |
clinic | +0.001 |
cafe | +0.001 |
Gu | +0.001 |
Neg logit contributions | |
journal | -0.001 |
shield | -0.001 |
journals | -0.001 |
luck | -0.001 |
narrative | -0.001 |
Token | adders |
Token ID | 45940 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ashes | +0.019 |
clinic | +0.018 |
ro | +0.018 |
Gu | +0.018 |
rimp | +0.017 |
Neg logit contributions | |
shield | -0.020 |
luck | -0.019 |
journal | -0.019 |
pencil | -0.019 |
journals | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
shield | +0.044 |
journal | +0.044 |
journals | +0.044 |
pencil | +0.043 |
luck | +0.042 |
Neg logit contributions | |
ashes | -0.069 |
ro | -0.066 |
clinic | -0.065 |
bank | -0.063 |
rimp | -0.061 |
Token | They |
Token ID | 1119 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
ashes | +0.012 |
ashing | +0.012 |
mare | +0.011 |
rim | +0.011 |
bank | +0.011 |
Neg logit contributions | |
magnets | -0.007 |
his | -0.007 |
him | -0.006 |
what | -0.006 |
pencil | -0.006 |
Token | suits |
Token ID | 14803 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
ashes | +0.063 |
clinic | +0.060 |
ro | +0.060 |
Gu | +0.057 |
rimp | +0.057 |
Neg logit contributions | |
shield | -0.044 |
journal | -0.043 |
pencil | -0.042 |
luck | -0.042 |
journals | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
ashes | +0.139 |
ro | +0.132 |
clinic | +0.132 |
rimp | +0.127 |
Gu | +0.126 |
Neg logit contributions | |
shield | -0.071 |
journal | -0.068 |
journals | -0.067 |
pencil | -0.066 |
luck | -0.065 |
Token | They |
Token ID | 1119 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
ashes | +0.014 |
anna | +0.013 |
nes | +0.013 |
rim | +0.013 |
ashing | +0.013 |
Neg logit contributions | |
magnets | -0.010 |
his | -0.009 |
pencil | -0.009 |
journal | -0.009 |
journals | -0.009 |
Token | had |
Token ID | 550 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
ashes | +0.123 |
ro | +0.120 |
cafe | +0.116 |
clinic | +0.116 |
rimp | +0.115 |
Neg logit contributions | |
journal | -0.112 |
shield | -0.110 |
journals | -0.110 |
luck | -0.109 |
crystal | -0.102 |
Token | h |
Token ID | 289 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
shield | +0.084 |
pencil | +0.081 |
journal | +0.081 |
luck | +0.080 |
journals | +0.080 |
Neg logit contributions | |
ashes | -0.154 |
ro | -0.145 |
clinic | -0.145 |
rimp | -0.140 |
Gu | -0.140 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.01e+00 |
Pos logit contributions | |
Gu | +0.042 |
clinic | +0.040 |
mare | +0.036 |
cafe | +0.036 |
ashes | +0.035 |
Neg logit contributions | |
luck | -0.093 |
thumbs | -0.090 |
pencil | -0.089 |
shield | -0.089 |
magnets | -0.088 |
Token | and |
Token ID | 290 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
ashes | +0.506 |
clinic | +0.481 |
ro | +0.477 |
Gu | +0.468 |
rimp | +0.461 |
Neg logit contributions | |
shield | -0.280 |
journal | -0.268 |
pencil | -0.267 |
luck | -0.266 |
journals | -0.260 |
Token | l |
Token ID | 300 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
ashes | +0.009 |
ro | +0.008 |
clinic | +0.008 |
Gu | +0.008 |
cafe | +0.008 |
Neg logit contributions | |
journal | -0.006 |
shield | -0.006 |
journals | -0.006 |
luck | -0.006 |
crystal | -0.006 |
Token | adders |
Token ID | 45940 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ashes | +0.019 |
clinic | +0.018 |
ro | +0.018 |
Gu | +0.018 |
rimp | +0.018 |
Neg logit contributions | |
shield | -0.014 |
luck | -0.014 |
journal | -0.014 |
pencil | -0.014 |
journals | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
shield | +0.045 |
journal | +0.044 |
journals | +0.044 |
pencil | +0.043 |
luck | +0.042 |
Neg logit contributions | |
ashes | -0.076 |
ro | -0.073 |
clinic | -0.072 |
bank | -0.069 |
rimp | -0.069 |
Token | helmets |
Token ID | 28359 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
ashes | +0.007 |
ro | +0.007 |
clinic | +0.007 |
Gu | +0.006 |
rimp | +0.006 |
Neg logit contributions | |
shield | -0.005 |
journal | -0.005 |
journals | -0.005 |
luck | -0.005 |
pencil | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
shield | +0.070 |
pencil | +0.069 |
journals | +0.069 |
journal | +0.068 |
magnets | +0.068 |
Neg logit contributions | |
ashes | -0.075 |
ro | -0.075 |
clinic | -0.073 |
rimp | -0.072 |
cafe | -0.072 |
Token | said |
Token ID | 531 |
Feature activation | +8.33e-01 |
Pos logit contributions | |
shield | +0.006 |
journal | +0.006 |
journals | +0.006 |
luck | +0.006 |
pencil | +0.006 |
Neg logit contributions | |
ashes | -0.008 |
ro | -0.007 |
clinic | -0.007 |
Gu | -0.007 |
rimp | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
ashes | +0.049 |
ro | +0.048 |
clinic | +0.047 |
anna | +0.045 |
rimp | +0.045 |
Neg logit contributions | |
shield | -0.032 |
journals | -0.032 |
journal | -0.031 |
flashlight | -0.030 |
pencil | -0.030 |
Token | " |
Token ID | 366 |
Feature activation | +9.64e-02 |
Pos logit contributions | |
shield | +0.001 |
pencil | +0.001 |
journal | +0.001 |
thumbs | +0.001 |
luck | +0.001 |
Neg logit contributions | |
ashes | -0.001 |
clinic | -0.001 |
ro | -0.001 |
Gu | -0.001 |
rimp | -0.001 |
Token | R |
Token ID | 49 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
clinic | +0.005 |
ashes | +0.005 |
oodle | +0.005 |
bank | +0.005 |
aquarium | +0.005 |
Neg logit contributions | |
pencil | -0.004 |
thumbs | -0.004 |
narrative | -0.004 |
shield | -0.004 |
journals | -0.004 |
Token | oses |
Token ID | 4629 |
Feature activation | +8.00e+00 |
Pos logit contributions | |
luck | +0.003 |
pencil | +0.003 |
journals | +0.003 |
journal | +0.003 |
shield | +0.003 |
Neg logit contributions | |
ashes | -0.003 |
round | -0.003 |
Gu | -0.003 |
clinic | -0.003 |
cafe | -0.002 |
Token | are |
Token ID | 389 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
ashes | +0.563 |
ro | +0.537 |
clinic | +0.534 |
rimp | +0.517 |
Gu | +0.511 |
Neg logit contributions | |
shield | -0.222 |
journal | -0.212 |
journals | -0.210 |
pencil | -0.204 |
luck | -0.202 |
Token | special |
Token ID | 2041 |
Feature activation | -7.05e-01 |
Pos logit contributions | |
shield | +0.074 |
luck | +0.072 |
journal | +0.070 |
pencil | +0.070 |
journals | +0.068 |
Neg logit contributions | |
ashes | -0.129 |
ro | -0.123 |
clinic | -0.121 |
Gu | -0.118 |
round | -0.116 |
Token | flowers |
Token ID | 12734 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
shield | +0.024 |
journal | +0.023 |
pencil | +0.023 |
journals | +0.023 |
luck | +0.023 |
Neg logit contributions | |
ashes | -0.045 |
clinic | -0.043 |
ro | -0.043 |
Gu | -0.041 |
rimp | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
ashes | +0.105 |
clinic | +0.099 |
ro | +0.099 |
rimp | +0.095 |
Gu | +0.095 |
Neg logit contributions | |
shield | -0.054 |
journal | -0.051 |
pencil | -0.051 |
journals | -0.051 |
luck | -0.050 |
Token | They |
Token ID | 1119 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
eh | +0.007 |
ashes | +0.007 |
ashing | +0.006 |
bank | +0.006 |
nes | +0.006 |
Neg logit contributions | |
journals | -0.005 |
luck | -0.005 |
magnets | -0.005 |
shield | -0.005 |
pencil | -0.005 |
Token | thinks |
Token ID | 6834 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
ashes | +0.072 |
clinic | +0.068 |
ro | +0.068 |
Gu | +0.066 |
rimp | +0.066 |
Neg logit contributions | |
shield | -0.046 |
journal | -0.044 |
pencil | -0.044 |
luck | -0.044 |
journals | -0.043 |
Token | of |
Token ID | 286 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
shield | +0.116 |
journal | +0.116 |
journals | +0.115 |
pencil | +0.110 |
crystal | +0.108 |
Neg logit contributions | |
ashes | -0.129 |
clinic | -0.120 |
ro | -0.119 |
cafe | -0.115 |
rimp | -0.112 |
Token | words |
Token ID | 2456 |
Feature activation | -4.85e+00 |
Pos logit contributions | |
ashes | +0.192 |
clinic | +0.188 |
ro | +0.185 |
cafe | +0.179 |
rimp | +0.179 |
Neg logit contributions | |
shield | -0.108 |
journal | -0.105 |
journals | -0.102 |
pencil | -0.098 |
luck | -0.098 |
Token | that |
Token ID | 326 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
shield | +0.192 |
journals | +0.187 |
journal | +0.187 |
pencil | +0.180 |
luck | +0.180 |
Neg logit contributions | |
ashes | -0.277 |
clinic | -0.267 |
ro | -0.263 |
rimp | -0.252 |
cafe | -0.250 |
Token | rhy |
Token ID | 40005 |
Feature activation | -4.03e+00 |
Pos logit contributions | |
shield | +0.020 |
journal | +0.019 |
journals | +0.019 |
luck | +0.018 |
pencil | +0.018 |
Neg logit contributions | |
ashes | -0.031 |
clinic | -0.030 |
ro | -0.029 |
cafe | -0.028 |
rimp | -0.028 |
Token | me |
Token ID | 1326 |
Feature activation | +8.00e+00 |
Pos logit contributions | |
pencil | +0.197 |
magnets | +0.196 |
shield | +0.190 |
journals | +0.190 |
thumbs | +0.189 |
Neg logit contributions | |
ashes | -0.190 |
clinic | -0.183 |
ro | -0.174 |
cafe | -0.171 |
bank | -0.170 |
Token | with |
Token ID | 351 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
ashes | +0.469 |
ro | +0.443 |
clinic | +0.439 |
rimp | +0.426 |
Gu | +0.416 |
Neg logit contributions | |
shield | -0.324 |
journal | -0.314 |
journals | -0.308 |
pencil | -0.307 |
luck | -0.300 |
Token | what |
Token ID | 644 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
ashes | +0.106 |
ro | +0.102 |
clinic | +0.100 |
rimp | +0.098 |
cafe | +0.095 |
Neg logit contributions | |
shield | -0.067 |
journal | -0.066 |
journals | -0.064 |
luck | -0.060 |
pencil | -0.060 |
Token | she |
Token ID | 673 |
Feature activation | +7.01e-01 |
Pos logit contributions | |
shield | +0.173 |
journal | +0.163 |
luck | +0.158 |
crystal | +0.158 |
journals | +0.158 |
Neg logit contributions | |
ashes | -0.200 |
clinic | -0.190 |
ro | -0.184 |
rimp | -0.179 |
cafe | -0.178 |
Token | sees |
Token ID | 7224 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ashes | +0.038 |
clinic | +0.036 |
ro | +0.035 |
Gu | +0.034 |
rimp | +0.034 |
Neg logit contributions | |
shield | -0.031 |
journal | -0.030 |
pencil | -0.030 |
journals | -0.030 |
luck | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
ashes | +0.069 |
ro | +0.065 |
clinic | +0.064 |
rimp | +0.062 |
cafe | +0.062 |
Neg logit contributions | |
shield | -0.042 |
journal | -0.040 |
journals | -0.039 |
luck | -0.038 |
pencil | -0.038 |
Token | bed |
Token ID | 3996 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ashes | +0.011 |
cafe | +0.010 |
ro | +0.010 |
clinic | +0.010 |
te | +0.010 |
Neg logit contributions | |
luck | -0.024 |
narrative | -0.023 |
journals | -0.023 |
shield | -0.023 |
pencil | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +2.86e-01 |
Pos logit contributions | |
ashes | +0.104 |
Gu | +0.098 |
clinic | +0.098 |
ro | +0.097 |
rimp | +0.096 |
Neg logit contributions | |
shield | -0.055 |
journal | -0.054 |
pencil | -0.053 |
luck | -0.052 |
journals | -0.052 |
Token | eat |
Token ID | 4483 |
Feature activation | -5.36e-01 |
Pos logit contributions | |
ashes | +0.013 |
clinic | +0.013 |
ro | +0.012 |
cafe | +0.012 |
round | +0.011 |
Neg logit contributions | |
magnets | -0.014 |
shield | -0.014 |
journals | -0.014 |
journal | -0.014 |
luck | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +5.78e-01 |
Pos logit contributions | |
shield | +0.024 |
journal | +0.022 |
narrative | +0.022 |
flashlight | +0.022 |
journals | +0.022 |
Neg logit contributions | |
ashes | -0.030 |
cafe | -0.028 |
clinic | -0.027 |
ro | -0.027 |
rimp | -0.026 |
Token | o |
Token ID | 267 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
ashes | +0.028 |
cafe | +0.027 |
clinic | +0.025 |
te | +0.024 |
café | +0.024 |
Neg logit contributions | |
shield | -0.029 |
journals | -0.027 |
streaming | -0.027 |
narrative | -0.027 |
luck | -0.026 |
Token | me |
Token ID | 1326 |
Feature activation | +7.97e+00 |
Pos logit contributions | |
ashes | +0.094 |
Gu | +0.089 |
ro | +0.088 |
bank | +0.087 |
clinic | +0.086 |
Neg logit contributions | |
journal | -0.088 |
pencil | -0.088 |
journals | -0.085 |
luck | -0.085 |
thumbs | -0.085 |
Token | let |
Token ID | 1616 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
ashes | +0.468 |
clinic | +0.443 |
ro | +0.441 |
Gu | +0.427 |
rimp | +0.422 |
Neg logit contributions | |
shield | -0.317 |
journal | -0.308 |
luck | -0.305 |
pencil | -0.304 |
journals | -0.302 |
Token | and |
Token ID | 290 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
ashes | +0.001 |
ro | +0.001 |
clinic | +0.001 |
cafe | +0.001 |
rimp | +0.001 |
Neg logit contributions | |
shield | -0.000 |
journal | -0.000 |
journals | -0.000 |
pencil | -0.000 |
luck | -0.000 |
Token | the |
Token ID | 262 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
ashes | +0.013 |
clinic | +0.012 |
ro | +0.012 |
cafe | +0.011 |
rimp | +0.011 |
Neg logit contributions | |
shield | -0.012 |
journal | -0.012 |
journals | -0.012 |
luck | -0.012 |
pencil | -0.011 |
Token | toast |
Token ID | 27805 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
ashes | +0.019 |
cafe | +0.018 |
clinic | +0.018 |
ro | +0.017 |
te | +0.016 |
Neg logit contributions | |
shield | -0.018 |
journals | -0.017 |
journal | -0.017 |
pencil | -0.017 |
luck | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
shield | +0.062 |
journal | +0.059 |
pencil | +0.058 |
journals | +0.058 |
luck | +0.058 |
Neg logit contributions | |
ashes | -0.124 |
clinic | -0.118 |
ro | -0.118 |
rimp | -0.113 |
Gu | -0.112 |
Token | home |
Token ID | 1363 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
shield | +0.102 |
luck | +0.099 |
journal | +0.097 |
pencil | +0.095 |
journals | +0.092 |
Neg logit contributions | |
ashes | -0.158 |
clinic | -0.149 |
ro | -0.149 |
rimp | -0.149 |
Gu | -0.147 |
Token | safely |
Token ID | 11512 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ashes | +0.098 |
ro | +0.092 |
clinic | +0.092 |
Gu | +0.091 |
rimp | +0.090 |
Neg logit contributions | |
shield | -0.040 |
luck | -0.038 |
journal | -0.037 |
pencil | -0.037 |
journals | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -9.43e-01 |
Pos logit contributions | |
shield | +0.035 |
luck | +0.034 |
journal | +0.033 |
pencil | +0.033 |
journals | +0.032 |
Neg logit contributions | |
ashes | -0.087 |
clinic | -0.083 |
ro | -0.082 |
Gu | -0.081 |
rimp | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
shield | +0.049 |
journal | +0.047 |
pencil | +0.047 |
journals | +0.046 |
luck | +0.046 |
Neg logit contributions | |
ashes | -0.045 |
ro | -0.042 |
clinic | -0.042 |
rimp | -0.040 |
Gu | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
luck | +0.044 |
shield | +0.044 |
pencil | +0.043 |
journals | +0.043 |
journal | +0.043 |
Neg logit contributions | |
ashes | -0.073 |
clinic | -0.071 |
ro | -0.069 |
Gu | -0.068 |
rimp | -0.067 |
Token | Mic |
Token ID | 25437 |
Feature activation | -9.54e+00 |
Pos logit contributions | |
luck | +0.063 |
pencil | +0.062 |
shield | +0.062 |
journal | +0.062 |
journals | +0.061 |
Neg logit contributions | |
ashes | -0.049 |
clinic | -0.048 |
Gu | -0.047 |
ro | -0.046 |
rimp | -0.045 |
Token | key |
Token ID | 2539 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
flashlight | +0.551 |
telescope | +0.549 |
journal | +0.528 |
journals | +0.528 |
luck | +0.514 |
Neg logit contributions | |
Gu | -0.356 |
anna | -0.343 |
rimp | -0.330 |
bank | -0.325 |
orian | -0.309 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
shield | +0.100 |
pencil | +0.093 |
journal | +0.093 |
luck | +0.093 |
journals | +0.090 |
Neg logit contributions | |
ashes | -0.220 |
clinic | -0.206 |
ro | -0.204 |
Gu | -0.202 |
rimp | -0.201 |
Token | that |
Token ID | 326 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
journals | +0.052 |
journal | +0.052 |
pencil | +0.050 |
crystal | +0.050 |
shield | +0.049 |
Neg logit contributions | |
clinic | -0.061 |
ro | -0.060 |
ashes | -0.060 |
bank | -0.059 |
Gu | -0.057 |
Token | it |
Token ID | 340 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
journals | +0.045 |
thumbs | +0.044 |
journal | +0.043 |
hetically | +0.043 |
pencil | +0.043 |
Neg logit contributions | |
clinic | -0.049 |
bank | -0.048 |
Gu | -0.047 |
ro | -0.046 |
ashes | -0.044 |
Token | 's |
Token ID | 338 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
journals | +0.022 |
pencil | +0.022 |
shield | +0.022 |
journal | +0.021 |
telescope | +0.020 |
Neg logit contributions | |
ashes | -0.021 |
clinic | -0.020 |
Gu | -0.019 |
ro | -0.019 |
bank | -0.019 |
Token | helped |
Token ID | 4193 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
shield | +0.016 |
journals | +0.016 |
journal | +0.016 |
luck | +0.015 |
pencil | +0.015 |
Neg logit contributions | |
ashes | -0.024 |
clinic | -0.023 |
ro | -0.022 |
Gu | -0.022 |
cafe | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
shield | +0.035 |
journal | +0.035 |
journals | +0.035 |
luck | +0.034 |
pencil | +0.033 |
Neg logit contributions | |
ashes | -0.071 |
clinic | -0.068 |
ro | -0.067 |
bank | -0.065 |
rimp | -0.065 |
Token | pick |
Token ID | 2298 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
clinic | +0.044 |
ashes | +0.043 |
ro | +0.041 |
te | +0.039 |
cafe | +0.039 |
Neg logit contributions | |
narrative | -0.027 |
journals | -0.026 |
luck | -0.025 |
shield | -0.024 |
pencil | -0.023 |
Token | up |
Token ID | 510 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
shield | +0.045 |
journal | +0.043 |
journals | +0.043 |
pencil | +0.041 |
luck | +0.041 |
Neg logit contributions | |
ashes | -0.080 |
ro | -0.076 |
clinic | -0.076 |
Gu | -0.073 |
cafe | -0.072 |
Token | the |
Token ID | 262 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
ashes | +0.029 |
ro | +0.027 |
clinic | +0.027 |
cafe | +0.026 |
oodle | +0.025 |
Neg logit contributions | |
shield | -0.016 |
journal | -0.016 |
journals | -0.016 |
narrative | -0.015 |
pencil | -0.015 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.44e+00 |
Pos logit contributions | |
journals | +0.015 |
shield | +0.014 |
journal | +0.014 |
luck | +0.014 |
thumbs | +0.013 |
Neg logit contributions | |
ashes | -0.019 |
ro | -0.018 |
clinic | -0.018 |
cafe | -0.017 |
oodle | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -7.73e-01 |
Pos logit contributions | |
shield | +0.359 |
journals | +0.347 |
journal | +0.345 |
luck | +0.338 |
pencil | +0.338 |
Neg logit contributions | |
ashes | -0.568 |
ro | -0.540 |
clinic | -0.533 |
rimp | -0.515 |
bank | -0.508 |
Token | She |
Token ID | 1375 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
pencil | +0.028 |
magnets | +0.027 |
journal | +0.027 |
journals | +0.026 |
shield | +0.025 |
Neg logit contributions | |
ashes | -0.049 |
clinic | -0.044 |
rim | -0.043 |
rimp | -0.043 |
ashing | -0.042 |
Token | learned |
Token ID | 4499 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ashes | +0.021 |
clinic | +0.020 |
ro | +0.020 |
Gu | +0.019 |
rimp | +0.019 |
Neg logit contributions | |
shield | -0.014 |
journal | -0.014 |
luck | -0.013 |
journals | -0.013 |
pencil | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
shield | +0.048 |
journals | +0.048 |
journal | +0.047 |
pencil | +0.045 |
crystal | +0.045 |
Neg logit contributions | |
ashes | -0.058 |
clinic | -0.057 |
ro | -0.056 |
bank | -0.053 |
cafe | -0.053 |
Token | lesson |
Token ID | 11483 |
Feature activation | -6.33e+00 |
Pos logit contributions | |
journals | +0.038 |
shield | +0.036 |
pa | +0.036 |
magnets | +0.036 |
streaming | +0.035 |
Neg logit contributions | |
ashes | -0.042 |
clinic | -0.041 |
ro | -0.041 |
Gu | -0.039 |
oodle | -0.039 |
Token | mouse |
Token ID | 10211 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
shield | +0.007 |
pencil | +0.007 |
journal | +0.007 |
luck | +0.007 |
crystal | +0.007 |
Neg logit contributions | |
ashes | -0.010 |
clinic | -0.009 |
ro | -0.009 |
rimp | -0.009 |
nes | -0.009 |
Token | named |
Token ID | 3706 |
Feature activation | +6.88e+00 |
Pos logit contributions | |
journals | +0.056 |
luck | +0.056 |
shield | +0.055 |
pencil | +0.055 |
journal | +0.054 |
Neg logit contributions | |
ashes | -0.073 |
clinic | -0.071 |
ro | -0.069 |
bank | -0.066 |
rimp | -0.066 |
Token | Mickey |
Token ID | 27232 |
Feature activation | -5.20e+00 |
Pos logit contributions | |
ashes | +0.442 |
clinic | +0.402 |
ro | +0.401 |
ashing | +0.397 |
round | +0.393 |
Neg logit contributions | |
shield | -0.262 |
pencil | -0.252 |
journal | -0.244 |
crystal | -0.240 |
luck | -0.239 |
Token | . |
Token ID | 13 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
journals | +0.208 |
thumbs | +0.202 |
journal | +0.197 |
shield | +0.197 |
pencil | +0.195 |
Neg logit contributions | |
clinic | -0.293 |
ashes | -0.289 |
ro | -0.276 |
bank | -0.271 |
rimp | -0.268 |
Token | " |
Token ID | 366 |
Feature activation | -6.84e-02 |
Pos logit contributions | |
shield | +0.025 |
pencil | +0.024 |
journal | +0.024 |
magnets | +0.023 |
luck | +0.023 |
Neg logit contributions | |
ashes | -0.037 |
ro | -0.034 |
clinic | -0.034 |
rimp | -0.033 |
nes | -0.032 |
Token | Mic |
Token ID | 25437 |
Feature activation | -9.19e+00 |
Pos logit contributions | |
shield | +0.004 |
journal | +0.004 |
pencil | +0.004 |
luck | +0.004 |
journals | +0.004 |
Neg logit contributions | |
ashes | -0.003 |
clinic | -0.003 |
ro | -0.003 |
Gu | -0.003 |
rimp | -0.002 |
Token | key |
Token ID | 2539 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
journal | +0.495 |
journals | +0.486 |
shield | +0.485 |
telescope | +0.471 |
flashlight | +0.466 |
Neg logit contributions | |
Gu | -0.393 |
rimp | -0.385 |
anna | -0.384 |
bank | -0.366 |
clinic | -0.366 |
Token | , |
Token ID | 11 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
shield | +0.106 |
journal | +0.103 |
journals | +0.102 |
pencil | +0.100 |
luck | +0.100 |
Neg logit contributions | |
ashes | -0.207 |
clinic | -0.198 |
ro | -0.197 |
Gu | -0.190 |
rimp | -0.190 |
Token | help |
Token ID | 1037 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
journals | +0.022 |
journal | +0.020 |
pencil | +0.020 |
luck | +0.020 |
shield | +0.020 |
Neg logit contributions | |
clinic | -0.026 |
ashes | -0.026 |
Gu | -0.026 |
ro | -0.025 |
mare | -0.025 |
Token | ! |
Token ID | 0 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
shield | +0.162 |
pencil | +0.155 |
journal | +0.151 |
luck | +0.150 |
magnets | +0.150 |
Neg logit contributions | |
ashes | -0.185 |
ro | -0.171 |
clinic | -0.169 |
Gu | -0.163 |
ashing | -0.163 |
Token | The |
Token ID | 383 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
shield | +0.025 |
pencil | +0.024 |
magnets | +0.024 |
magnet | +0.023 |
spear | +0.023 |
Neg logit contributions | |
ashes | -0.030 |
ashing | -0.026 |
nes | -0.026 |
clinic | -0.026 |
ro | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ashes | +0.026 |
ro | +0.024 |
clinic | +0.024 |
rimp | +0.024 |
cafe | +0.024 |
Neg logit contributions | |
shield | -0.021 |
journals | -0.021 |
journal | -0.020 |
magnets | -0.020 |
narrative | -0.019 |
Token | She |
Token ID | 1375 |
Feature activation | +4.13e-03 |
Pos logit contributions | |
pencil | +0.044 |
thumbs | +0.040 |
journal | +0.039 |
magnets | +0.038 |
all | +0.038 |
Neg logit contributions | |
ashes | -0.057 |
clinic | -0.057 |
rim | -0.056 |
oster | -0.056 |
rimp | -0.055 |
Token | picked |
Token ID | 6497 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ashes | +0.000 |
clinic | +0.000 |
ro | +0.000 |
Gu | +0.000 |
rimp | +0.000 |
Neg logit contributions | |
shield | -0.000 |
journal | -0.000 |
luck | -0.000 |
pencil | -0.000 |
journals | -0.000 |
Token | up |
Token ID | 510 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
shield | +0.043 |
journals | +0.043 |
narrative | +0.043 |
journal | +0.042 |
flashlight | +0.039 |
Neg logit contributions | |
ashes | -0.057 |
ro | -0.052 |
Gu | -0.050 |
cafe | -0.049 |
clinic | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -6.16e-01 |
Pos logit contributions | |
ashes | +0.050 |
ro | +0.048 |
cafe | +0.047 |
clinic | +0.044 |
Ph | +0.044 |
Neg logit contributions | |
shield | -0.039 |
journal | -0.038 |
journals | -0.038 |
narrative | -0.038 |
spear | -0.036 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.18e+00 |
Pos logit contributions | |
journals | +0.025 |
shield | +0.024 |
luck | +0.022 |
narrative | +0.022 |
journal | +0.022 |
Neg logit contributions | |
ashes | -0.034 |
ro | -0.034 |
clinic | -0.032 |
cafe | -0.032 |
te | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -9.66e-01 |
Pos logit contributions | |
journals | +0.443 |
shield | +0.424 |
journal | +0.412 |
narrative | +0.405 |
luck | +0.402 |
Neg logit contributions | |
ashes | -0.462 |
ro | -0.445 |
clinic | -0.405 |
bank | -0.405 |
rimp | -0.404 |
Token | held |
Token ID | 2714 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
journal | +0.045 |
shield | +0.045 |
luck | +0.045 |
narrative | +0.044 |
journals | +0.044 |
Neg logit contributions | |
ashes | -0.047 |
ro | -0.045 |
bank | -0.043 |
cafe | -0.042 |
clinic | -0.042 |
Token | them |
Token ID | 606 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
shield | +0.128 |
journal | +0.121 |
journals | +0.119 |
flashlight | +0.118 |
narrative | +0.117 |
Neg logit contributions | |
ashes | -0.128 |
ro | -0.112 |
te | -0.108 |
rimp | -0.108 |
asher | -0.106 |
Token | in |
Token ID | 287 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
ashes | +0.161 |
ro | +0.148 |
ashing | +0.142 |
uring | +0.139 |
nes | +0.134 |
Neg logit contributions | |
narrative | -0.153 |
shield | -0.152 |
journals | -0.149 |
journal | -0.147 |
father | -0.144 |
Token | her |
Token ID | 607 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
shield | +0.023 |
journals | +0.021 |
journal | +0.021 |
sword | +0.021 |
spear | +0.020 |
Neg logit contributions | |
ro | -0.031 |
ashes | -0.031 |
cafe | -0.030 |
Gu | -0.030 |
clinic | -0.030 |
Token | and |
Token ID | 290 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
journals | +0.100 |
shield | +0.100 |
journal | +0.095 |
telescope | +0.095 |
pencil | +0.094 |
Neg logit contributions | |
ashes | -0.089 |
ro | -0.088 |
clinic | -0.084 |
ashing | -0.080 |
uring | -0.080 |
Token | to |
Token ID | 284 |
Feature activation | +7.40e-02 |
Pos logit contributions | |
shield | +0.021 |
journal | +0.020 |
journals | +0.020 |
pencil | +0.020 |
luck | +0.020 |
Neg logit contributions | |
ashes | -0.023 |
clinic | -0.023 |
ro | -0.022 |
bank | -0.021 |
Gu | -0.020 |
Token | be |
Token ID | 307 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ro | +0.003 |
ashes | +0.003 |
clinic | +0.003 |
bank | +0.003 |
rimp | +0.003 |
Neg logit contributions | |
journal | -0.004 |
journals | -0.004 |
shield | -0.004 |
flashlight | -0.003 |
luck | -0.003 |
Token | careful |
Token ID | 8161 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
ashes | +0.090 |
clinic | +0.085 |
ro | +0.084 |
Gu | +0.081 |
rimp | +0.081 |
Neg logit contributions | |
shield | -0.061 |
journal | -0.059 |
pencil | -0.058 |
luck | -0.058 |
journals | -0.057 |
Token | with |
Token ID | 351 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
shield | +0.002 |
journal | +0.002 |
pencil | +0.002 |
journals | +0.002 |
crystal | +0.002 |
Neg logit contributions | |
ashes | -0.002 |
ro | -0.002 |
clinic | -0.002 |
ashing | -0.002 |
rimp | -0.002 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.18e+00 |
Pos logit contributions | |
ashes | +0.062 |
clinic | +0.061 |
ro | +0.058 |
rimp | +0.057 |
bank | +0.056 |
Neg logit contributions | |
shield | -0.050 |
journal | -0.050 |
narrative | -0.049 |
hetically | -0.048 |
journals | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.95e-01 |
Pos logit contributions | |
shield | +0.364 |
journals | +0.359 |
journal | +0.352 |
crystal | +0.345 |
pencil | +0.340 |
Neg logit contributions | |
ashes | -0.535 |
ro | -0.519 |
clinic | -0.494 |
ashing | -0.471 |
nes | -0.470 |
Token | She |
Token ID | 1375 |
Feature activation | +4.57e-01 |
Pos logit contributions | |
journal | +0.008 |
journals | +0.008 |
pencil | +0.008 |
shield | +0.008 |
magnets | +0.008 |
Neg logit contributions | |
ashes | -0.011 |
clinic | -0.010 |
ro | -0.010 |
round | -0.010 |
rimp | -0.010 |
Token | still |
Token ID | 991 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ashes | +0.025 |
clinic | +0.024 |
ro | +0.024 |
Gu | +0.023 |
cafe | +0.022 |
Neg logit contributions | |
shield | -0.019 |
journal | -0.019 |
journals | -0.018 |
luck | -0.018 |
pencil | -0.018 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ashes | +0.084 |
ro | +0.080 |
clinic | +0.079 |
bank | +0.074 |
Gu | +0.074 |
Neg logit contributions | |
shield | -0.063 |
journal | -0.061 |
journals | -0.061 |
luck | -0.060 |
pencil | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
shield | +0.062 |
journal | +0.060 |
narrative | +0.059 |
flashlight | +0.059 |
luck | +0.058 |
Neg logit contributions | |
ashes | -0.075 |
ro | -0.073 |
clinic | -0.070 |
rimp | -0.068 |
Gu | -0.068 |
Token | vanished |
Token ID | 23717 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
journals | +0.214 |
shield | +0.211 |
journal | +0.211 |
luck | +0.207 |
pencil | +0.200 |
Neg logit contributions | |
ashes | -0.294 |
ro | -0.283 |
clinic | -0.282 |
bank | -0.272 |
Gu | -0.271 |
Token | ! |
Token ID | 0 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
journals | +0.044 |
narrative | +0.042 |
pencil | +0.042 |
journal | +0.042 |
shield | +0.041 |
Neg logit contributions | |
ashes | -0.061 |
ro | -0.060 |
clinic | -0.056 |
ashing | -0.055 |
cafe | -0.055 |
Token | |
Token ID | 220 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
shield | +0.027 |
journal | +0.026 |
pencil | +0.026 |
luck | +0.026 |
journals | +0.026 |
Neg logit contributions | |
ashes | -0.041 |
clinic | -0.038 |
ro | -0.038 |
Gu | -0.037 |
rimp | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
shield | +0.029 |
journal | +0.028 |
pencil | +0.028 |
luck | +0.028 |
journals | +0.027 |
Neg logit contributions | |
ashes | -0.027 |
clinic | -0.025 |
ro | -0.025 |
Gu | -0.024 |
rimp | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -8.79e-01 |
Pos logit contributions | |
shield | +0.039 |
pencil | +0.037 |
journal | +0.037 |
magnets | +0.036 |
thumbs | +0.036 |
Neg logit contributions | |
ashes | -0.048 |
clinic | -0.044 |
ro | -0.044 |
rimp | -0.042 |
ashing | -0.042 |
Token | Mic |
Token ID | 25437 |
Feature activation | -9.13e+00 |
Pos logit contributions | |
luck | +0.036 |
shield | +0.035 |
journal | +0.035 |
journals | +0.035 |
pencil | +0.034 |
Neg logit contributions | |
ashes | -0.049 |
clinic | -0.047 |
ro | -0.046 |
Gu | -0.046 |
rimp | -0.045 |
Token | key |
Token ID | 2539 |
Feature activation | -2.97e+00 |
Pos logit contributions | |
telescope | +0.502 |
flashlight | +0.499 |
luck | +0.498 |
journals | +0.491 |
shield | +0.490 |
Neg logit contributions | |
Gu | -0.378 |
anna | -0.316 |
rimp | -0.314 |
orian | -0.309 |
clinic | -0.306 |
Token | was |
Token ID | 373 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
shield | +0.102 |
journal | +0.098 |
journals | +0.097 |
luck | +0.096 |
pencil | +0.096 |
Neg logit contributions | |
ashes | -0.188 |
ro | -0.180 |
clinic | -0.180 |
Gu | -0.173 |
rimp | -0.172 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
shield | +0.010 |
journals | +0.010 |
pencil | +0.010 |
journal | +0.010 |
luck | +0.010 |
Neg logit contributions | |
ashes | -0.013 |
clinic | -0.012 |
ro | -0.012 |
Gu | -0.012 |
rimp | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
ashes | +0.119 |
ro | +0.111 |
clinic | +0.109 |
ashing | +0.109 |
anna | +0.105 |
Neg logit contributions | |
journals | -0.108 |
journal | -0.107 |
shield | -0.104 |
pencil | -0.103 |
crystal | -0.101 |
Token | but |
Token ID | 475 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
shield | +0.027 |
journal | +0.026 |
pencil | +0.026 |
luck | +0.026 |
journals | +0.026 |
Neg logit contributions | |
ashes | -0.027 |
clinic | -0.025 |
ro | -0.025 |
Gu | -0.024 |
rimp | -0.024 |
Token | counter |
Token ID | 3753 |
Feature activation | +5.70e+00 |
Pos logit contributions | |
ashes | +0.004 |
ro | +0.004 |
clinic | +0.003 |
Gu | +0.003 |
rimp | +0.003 |
Neg logit contributions | |
shield | -0.005 |
journal | -0.005 |
pencil | -0.005 |
crystal | -0.005 |
journals | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
ashes | +0.432 |
Gu | +0.413 |
clinic | +0.407 |
ro | +0.407 |
anna | +0.400 |
Neg logit contributions | |
journal | -0.122 |
pencil | -0.122 |
shield | -0.119 |
crystal | -0.110 |
luck | -0.110 |
Token | |
Token ID | 220 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
shield | +0.033 |
luck | +0.032 |
journal | +0.032 |
journals | +0.032 |
pencil | +0.032 |
Neg logit contributions | |
ashes | -0.029 |
clinic | -0.027 |
ro | -0.026 |
Gu | -0.026 |
rimp | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
shield | +0.021 |
journal | +0.020 |
pencil | +0.020 |
journals | +0.020 |
luck | +0.020 |
Neg logit contributions | |
ashes | -0.020 |
clinic | -0.019 |
ro | -0.019 |
Gu | -0.018 |
rimp | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
shield | +0.029 |
pencil | +0.028 |
journal | +0.028 |
thumbs | +0.027 |
magnets | +0.026 |
Neg logit contributions | |
ashes | -0.047 |
clinic | -0.044 |
ro | -0.043 |
rimp | -0.042 |
ashing | -0.041 |
Token | Mic |
Token ID | 25437 |
Feature activation | -9.11e+00 |
Pos logit contributions | |
luck | +0.032 |
shield | +0.032 |
journal | +0.031 |
journals | +0.031 |
pencil | +0.031 |
Neg logit contributions | |
ashes | -0.045 |
clinic | -0.043 |
ro | -0.042 |
Gu | -0.042 |
rimp | -0.040 |
Token | key |
Token ID | 2539 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
telescope | +0.499 |
flashlight | +0.496 |
luck | +0.493 |
journals | +0.482 |
shield | +0.479 |
Neg logit contributions | |
Gu | -0.374 |
anna | -0.328 |
rimp | -0.324 |
clinic | -0.312 |
orian | -0.310 |
Token | wanted |
Token ID | 2227 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
shield | +0.094 |
pencil | +0.090 |
journal | +0.087 |
luck | +0.086 |
thumbs | +0.084 |
Neg logit contributions | |
ashes | -0.196 |
clinic | -0.184 |
ro | -0.180 |
Gu | -0.178 |
rimp | -0.176 |
Token | some |
Token ID | 617 |
Feature activation | +6.75e-01 |
Pos logit contributions | |
shield | +0.034 |
journal | +0.033 |
journals | +0.033 |
luck | +0.032 |
narrative | +0.032 |
Neg logit contributions | |
ashes | -0.040 |
clinic | -0.039 |
ro | -0.038 |
Gu | -0.037 |
cafe | -0.037 |
Token | popcorn |
Token ID | 38914 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
ashes | +0.036 |
clinic | +0.033 |
ro | +0.033 |
Gu | +0.032 |
rimp | +0.032 |
Neg logit contributions | |
shield | -0.031 |
journal | -0.030 |
pencil | -0.030 |
luck | -0.030 |
journals | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
ashes | +0.056 |
ro | +0.053 |
clinic | +0.052 |
cafe | +0.051 |
rimp | +0.051 |
Neg logit contributions | |
shield | -0.037 |
journals | -0.037 |
journal | -0.036 |
pencil | -0.035 |
narrative | -0.035 |
Token | she |
Token ID | 673 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
ashes | +0.163 |
clinic | +0.154 |
cafe | +0.154 |
Gu | +0.153 |
ro | +0.153 |
Neg logit contributions | |
journals | -0.163 |
shield | -0.161 |
journal | -0.160 |
magnets | -0.150 |
spear | -0.150 |
Token | cl |
Token ID | 537 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ashes | +0.004 |
ro | +0.004 |
clinic | +0.004 |
Gu | +0.004 |
rimp | +0.004 |
Neg logit contributions | |
shield | -0.002 |
journal | -0.002 |
journals | -0.002 |
luck | -0.002 |
pencil | -0.002 |
Token | apped |
Token ID | 6320 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
clinic | +0.056 |
ashes | +0.053 |
Gu | +0.053 |
ro | +0.051 |
bank | +0.051 |
Neg logit contributions | |
luck | -0.046 |
thumbs | -0.046 |
pencil | -0.045 |
journals | -0.045 |
journal | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
ashes | +0.114 |
ro | +0.109 |
clinic | +0.104 |
ashing | +0.101 |
rimp | +0.101 |
Neg logit contributions | |
journals | -0.068 |
shield | -0.068 |
journal | -0.068 |
pencil | -0.065 |
magnets | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -2.09e-01 |
Pos logit contributions | |
shield | +0.036 |
journals | +0.036 |
journal | +0.036 |
pencil | +0.034 |
luck | +0.034 |
Neg logit contributions | |
ashes | -0.040 |
ro | -0.037 |
clinic | -0.037 |
Gu | -0.036 |
cafe | -0.035 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.11e+00 |
Pos logit contributions | |
pt | +0.006 |
telescope | +0.006 |
flashlight | +0.006 |
luck | +0.005 |
mercy | +0.005 |
Neg logit contributions | |
ashes | -0.011 |
ro | -0.011 |
te | -0.011 |
asher | -0.011 |
bank | -0.011 |
Token | stuck |
Token ID | 7819 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
journals | +0.461 |
journal | +0.431 |
narrative | +0.402 |
luck | +0.399 |
shield | +0.396 |
Neg logit contributions | |
ashes | -0.466 |
ro | -0.443 |
te | -0.400 |
bank | -0.393 |
clinic | -0.388 |
Token | to |
Token ID | 284 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
journals | +0.002 |
journal | +0.002 |
shield | +0.002 |
narrative | +0.002 |
pencil | +0.001 |
Neg logit contributions | |
ro | -0.003 |
ashes | -0.003 |
bank | -0.003 |
clinic | -0.003 |
ashing | -0.003 |
Token | her |
Token ID | 607 |
Feature activation | +3.82e-01 |
Pos logit contributions | |
ashes | +0.008 |
ro | +0.008 |
clinic | +0.008 |
Gu | +0.007 |
rimp | +0.007 |
Neg logit contributions | |
shield | -0.006 |
journal | -0.006 |
journals | -0.006 |
pencil | -0.006 |
luck | -0.005 |
Token | hands |
Token ID | 2832 |
Feature activation | -3.47e+00 |
Pos logit contributions | |
ashes | +0.021 |
ro | +0.021 |
clinic | +0.021 |
cafe | +0.019 |
oodle | +0.019 |
Neg logit contributions | |
narrative | -0.015 |
journals | -0.014 |
shield | -0.014 |
luck | -0.013 |
magnets | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -6.02e-01 |
Pos logit contributions | |
journals | +0.136 |
shield | +0.134 |
journal | +0.132 |
pencil | +0.126 |
luck | +0.126 |
Neg logit contributions | |
ashes | -0.205 |
ro | -0.199 |
clinic | -0.190 |
bank | -0.185 |
ashing | -0.184 |
Token | Jane |
Token ID | 12091 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
shield | +0.032 |
pencil | +0.030 |
journal | +0.030 |
luck | +0.030 |
journals | +0.030 |
Neg logit contributions | |
ashes | -0.061 |
clinic | -0.057 |
ro | -0.057 |
Gu | -0.055 |
rimp | -0.055 |
Token | would |
Token ID | 561 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
ashes | +0.070 |
clinic | +0.067 |
ro | +0.066 |
Gu | +0.064 |
rimp | +0.064 |
Neg logit contributions | |
shield | -0.029 |
journal | -0.028 |
luck | -0.027 |
pencil | -0.027 |
journals | -0.027 |
Token | take |
Token ID | 1011 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
shield | +0.146 |
journal | +0.145 |
journals | +0.145 |
luck | +0.143 |
pencil | +0.142 |
Neg logit contributions | |
ashes | -0.181 |
clinic | -0.174 |
ro | -0.172 |
rimp | -0.163 |
bank | -0.162 |
Token | out |
Token ID | 503 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
shield | +0.102 |
journal | +0.099 |
crystal | +0.096 |
narrative | +0.096 |
journals | +0.095 |
Neg logit contributions | |
ashes | -0.187 |
ro | -0.176 |
clinic | -0.174 |
anna | -0.170 |
cafe | -0.168 |
Token | her |
Token ID | 607 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
shield | +0.007 |
narrative | +0.007 |
journal | +0.006 |
crystal | +0.006 |
pencil | +0.006 |
Neg logit contributions | |
ashes | -0.012 |
clinic | -0.011 |
ro | -0.011 |
cafe | -0.011 |
anna | -0.011 |
Token | pencil |
Token ID | 21613 |
Feature activation | -9.09e+00 |
Pos logit contributions | |
narrative | +0.010 |
journals | +0.010 |
luck | +0.009 |
shield | +0.009 |
thumbs | +0.009 |
Neg logit contributions | |
clinic | -0.014 |
ashes | -0.014 |
ro | -0.014 |
cafe | -0.013 |
te | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
shield | +0.378 |
luck | +0.369 |
journals | +0.365 |
narrative | +0.356 |
journal | +0.356 |
Neg logit contributions | |
ro | -0.489 |
ashes | -0.489 |
clinic | -0.462 |
round | -0.449 |
nes | -0.446 |
Token | paper |
Token ID | 3348 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
narrative | +0.049 |
luck | +0.047 |
iness | +0.046 |
journal | +0.044 |
shield | +0.044 |
Neg logit contributions | |
ro | -0.060 |
ashes | -0.057 |
clinic | -0.057 |
bank | -0.054 |
cafe | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
shield | +0.065 |
journal | +0.062 |
journals | +0.062 |
narrative | +0.061 |
luck | +0.060 |
Neg logit contributions | |
ashes | -0.099 |
ro | -0.097 |
clinic | -0.093 |
bank | -0.089 |
rimp | -0.088 |
Token | write |
Token ID | 3551 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
narrative | +0.038 |
luck | +0.037 |
shield | +0.036 |
journal | +0.035 |
iness | +0.035 |
Neg logit contributions | |
ro | -0.058 |
ashes | -0.056 |
clinic | -0.055 |
bank | -0.053 |
te | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
narrative | +0.121 |
shield | +0.120 |
crystal | +0.114 |
flashlight | +0.113 |
thumbs | +0.112 |
Neg logit contributions | |
ashes | -0.176 |
ro | -0.175 |
clinic | -0.174 |
cafe | -0.167 |
anna | -0.165 |
Token | still |
Token ID | 991 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
ashes | +0.025 |
clinic | +0.024 |
ro | +0.024 |
Gu | +0.023 |
cafe | +0.022 |
Neg logit contributions | |
shield | -0.019 |
journal | -0.019 |
journals | -0.018 |
luck | -0.018 |
pencil | -0.018 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ashes | +0.084 |
ro | +0.080 |
clinic | +0.079 |
bank | +0.074 |
Gu | +0.074 |
Neg logit contributions | |
shield | -0.063 |
journal | -0.061 |
journals | -0.061 |
luck | -0.060 |
pencil | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
shield | +0.062 |
journal | +0.060 |
narrative | +0.059 |
flashlight | +0.059 |
luck | +0.058 |
Neg logit contributions | |
ashes | -0.075 |
ro | -0.073 |
clinic | -0.070 |
rimp | -0.068 |
Gu | -0.068 |
Token | play |
Token ID | 711 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
ro | +0.009 |
ashes | +0.009 |
clinic | +0.009 |
bank | +0.009 |
rimp | +0.008 |
Neg logit contributions | |
journals | -0.008 |
luck | -0.008 |
journal | -0.008 |
shield | -0.008 |
magnets | -0.008 |
Token | with |
Token ID | 351 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
ashes | +0.048 |
clinic | +0.046 |
ro | +0.046 |
Gu | +0.043 |
rimp | +0.042 |
Neg logit contributions | |
shield | -0.041 |
luck | -0.041 |
journals | -0.040 |
journal | -0.040 |
thumbs | -0.040 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.06e+00 |
Pos logit contributions | |
ashes | +0.073 |
ro | +0.070 |
clinic | +0.069 |
rimp | +0.065 |
te | +0.064 |
Neg logit contributions | |
luck | -0.054 |
shield | -0.053 |
flashlight | -0.052 |
narrative | -0.052 |
journal | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
journals | +0.372 |
shield | +0.366 |
journal | +0.358 |
thumbs | +0.357 |
luck | +0.355 |
Neg logit contributions | |
ashes | -0.505 |
ro | -0.495 |
clinic | -0.478 |
ashing | -0.449 |
rimp | -0.442 |
Token | but |
Token ID | 475 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
luck | +0.011 |
shield | +0.011 |
journal | +0.011 |
journals | +0.011 |
thumbs | +0.011 |
Neg logit contributions | |
ashes | -0.009 |
ro | -0.009 |
clinic | -0.008 |
rimp | -0.008 |
Gu | -0.008 |
Token | she |
Token ID | 673 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
ro | +0.015 |
ashes | +0.015 |
clinic | +0.015 |
bank | +0.014 |
Gu | +0.014 |
Neg logit contributions | |
shield | -0.015 |
journals | -0.014 |
thumbs | -0.014 |
journal | -0.014 |
flashlight | -0.014 |
Token | never |
Token ID | 1239 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
ashes | +0.016 |
ro | +0.015 |
clinic | +0.015 |
rimp | +0.014 |
oodle | +0.014 |
Neg logit contributions | |
shield | -0.013 |
journals | -0.013 |
journal | -0.013 |
luck | -0.012 |
pencil | -0.012 |
Token | touched |
Token ID | 12615 |
Feature activation | -1.74e-01 |
Pos logit contributions | |
journals | +0.113 |
shield | +0.112 |
journal | +0.112 |
crystal | +0.106 |
magnets | +0.106 |
Neg logit contributions | |
ashes | -0.113 |
ro | -0.107 |
ashing | -0.098 |
clinic | -0.098 |
bank | -0.095 |
Token | her |
Token ID | 607 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
ashes | +0.118 |
ro | +0.113 |
rimp | +0.107 |
ashing | +0.106 |
clinic | +0.106 |
Neg logit contributions | |
shield | -0.072 |
journals | -0.071 |
journal | -0.070 |
pencil | -0.069 |
narrative | -0.069 |
Token | hands |
Token ID | 2832 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
ashes | +0.027 |
ro | +0.026 |
te | +0.023 |
anna | +0.023 |
ashing | +0.023 |
Neg logit contributions | |
narrative | -0.020 |
journals | -0.019 |
pa | -0.019 |
father | -0.018 |
shield | -0.018 |
Token | again |
Token ID | 757 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
journals | +0.146 |
shield | +0.146 |
journal | +0.144 |
narrative | +0.136 |
pencil | +0.135 |
Neg logit contributions | |
ashes | -0.190 |
ro | -0.182 |
clinic | -0.171 |
ashing | -0.170 |
rimp | -0.168 |
Token | and |
Token ID | 290 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
shield | +0.090 |
journal | +0.086 |
pencil | +0.086 |
luck | +0.085 |
journals | +0.083 |
Neg logit contributions | |
ashes | -0.178 |
clinic | -0.169 |
ro | -0.167 |
Gu | -0.164 |
rimp | -0.162 |
Token | the |
Token ID | 262 |
Feature activation | -8.19e-02 |
Pos logit contributions | |
shield | +0.024 |
journal | +0.023 |
journals | +0.023 |
magnets | +0.022 |
pencil | +0.022 |
Neg logit contributions | |
ashes | -0.035 |
ro | -0.033 |
clinic | -0.032 |
cafe | -0.032 |
Gu | -0.032 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.06e+00 |
Pos logit contributions | |
pt | +0.003 |
what | +0.003 |
icked | +0.003 |
telescope | +0.003 |
luck | +0.003 |
Neg logit contributions | |
te | -0.004 |
ashes | -0.004 |
cafe | -0.004 |
ro | -0.004 |
forest | -0.004 |
Token | stayed |
Token ID | 9658 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
journals | +0.424 |
journal | +0.411 |
shield | +0.399 |
pencil | +0.386 |
luck | +0.385 |
Neg logit contributions | |
ashes | -0.484 |
ro | -0.458 |
clinic | -0.425 |
nes | -0.420 |
rimp | -0.416 |
Token | put |
Token ID | 1234 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
ashes | +0.015 |
clinic | +0.014 |
ro | +0.014 |
ashing | +0.014 |
rimp | +0.014 |
Neg logit contributions | |
shield | -0.010 |
pencil | -0.010 |
journals | -0.010 |
journal | -0.010 |
thumbs | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | -3.40e-01 |
Pos logit contributions | |
ashes | +0.082 |
ro | +0.078 |
bank | +0.074 |
clinic | +0.074 |
cafe | +0.072 |
Neg logit contributions | |
shield | -0.051 |
journal | -0.048 |
flashlight | -0.047 |
journals | -0.047 |
telescope | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | -4.91e-01 |
Pos logit contributions | |
pencil | +0.014 |
thumbs | +0.014 |
crystal | +0.014 |
luck | +0.014 |
all | +0.014 |
Neg logit contributions | |
clinic | -0.018 |
ashes | -0.018 |
ro | -0.017 |
rimp | -0.017 |
nes | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
pencil | +0.017 |
crystal | +0.017 |
shield | +0.017 |
thumbs | +0.017 |
journal | +0.017 |
Neg logit contributions | |
ashes | -0.030 |
clinic | -0.028 |
ro | -0.028 |
asher | -0.026 |
ashing | -0.026 |
Token | noticed |
Token ID | 6810 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
shield | +0.136 |
journal | +0.132 |
pencil | +0.130 |
journals | +0.126 |
crystal | +0.123 |
Neg logit contributions | |
ashes | -0.225 |
clinic | -0.211 |
ro | -0.209 |
rimp | -0.206 |
Gu | -0.202 |
Token | that |
Token ID | 326 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
clinic | +0.009 |
cafe | +0.009 |
ro | +0.009 |
Gu | +0.009 |
ashes | +0.009 |
Neg logit contributions | |
journals | -0.009 |
narrative | -0.009 |
luck | -0.009 |
journal | -0.008 |
thumbs | -0.008 |
Token | one |
Token ID | 530 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
luck | +0.031 |
hetically | +0.030 |
thumbs | +0.030 |
journals | +0.030 |
narrative | +0.029 |
Neg logit contributions | |
clinic | -0.060 |
cafe | -0.060 |
ro | -0.059 |
bank | -0.059 |
Gu | -0.058 |
Token | of |
Token ID | 286 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ro | +0.028 |
clinic | +0.027 |
cafe | +0.026 |
ashes | +0.026 |
round | +0.025 |
Neg logit contributions | |
luck | -0.040 |
journals | -0.039 |
narrative | -0.038 |
thumbs | -0.037 |
shield | -0.036 |
Token | his |
Token ID | 465 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ro | +0.165 |
ashes | +0.164 |
clinic | +0.163 |
rimp | +0.162 |
Gu | +0.159 |
Neg logit contributions | |
journal | -0.068 |
shield | -0.068 |
luck | -0.066 |
journals | -0.065 |
flashlight | -0.060 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.06e+00 |
Pos logit contributions | |
streaming | +0.069 |
narrative | +0.066 |
hetically | +0.063 |
mercy | +0.063 |
pt | +0.060 |
Neg logit contributions | |
ashes | -0.088 |
ro | -0.086 |
clinic | -0.083 |
aquarium | -0.078 |
asher | -0.075 |
Token | was |
Token ID | 373 |
Feature activation | -2.92e-01 |
Pos logit contributions | |
journals | +0.373 |
luck | +0.331 |
shield | +0.322 |
journal | +0.320 |
narrative | +0.316 |
Neg logit contributions | |
ro | -0.529 |
ashes | -0.513 |
bank | -0.497 |
forest | -0.485 |
rimp | -0.469 |
Token | missing |
Token ID | 4814 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
journal | +0.012 |
shield | +0.012 |
journals | +0.012 |
flashlight | +0.011 |
thumbs | +0.011 |
Neg logit contributions | |
ashes | -0.016 |
clinic | -0.015 |
ro | -0.015 |
rimp | -0.015 |
bank | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -8.21e-01 |
Pos logit contributions | |
narrative | +0.075 |
journals | +0.070 |
hetically | +0.069 |
thumbs | +0.067 |
journal | +0.067 |
Neg logit contributions | |
ro | -0.064 |
ashes | -0.061 |
clinic | -0.060 |
oses | -0.059 |
rimp | -0.059 |
Token | He |
Token ID | 679 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
pencil | +0.026 |
magnet | +0.024 |
magnets | +0.023 |
shield | +0.023 |
journal | +0.023 |
Neg logit contributions | |
ashes | -0.059 |
clinic | -0.055 |
nes | -0.053 |
rimp | -0.052 |
ashing | -0.052 |
Token | looked |
Token ID | 3114 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
luck | +0.134 |
journal | +0.132 |
shield | +0.132 |
journals | +0.131 |
narrative | +0.127 |
Neg logit contributions | |
ro | -0.127 |
Gu | -0.125 |
ashes | -0.125 |
anna | -0.121 |
clinic | -0.120 |
Token | again |
Token ID | 757 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
journals | +0.357 |
journal | +0.338 |
shield | +0.332 |
pencil | +0.321 |
narrative | +0.319 |
Neg logit contributions | |
ashes | -0.534 |
ro | -0.516 |
clinic | -0.500 |
bank | -0.482 |
ashing | -0.475 |
Token | . |
Token ID | 13 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
shield | +0.090 |
journal | +0.086 |
journals | +0.084 |
pencil | +0.084 |
luck | +0.083 |
Neg logit contributions | |
ashes | -0.182 |
ro | -0.173 |
clinic | -0.173 |
rimp | -0.167 |
Gu | -0.166 |
Token | Now |
Token ID | 2735 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
pencil | +0.026 |
crystal | +0.025 |
luck | +0.025 |
thumbs | +0.025 |
journal | +0.025 |
Neg logit contributions | |
ashes | -0.029 |
clinic | -0.028 |
ro | -0.026 |
rimp | -0.026 |
nes | -0.026 |
Token | she |
Token ID | 673 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
ashes | +0.128 |
ro | +0.121 |
clinic | +0.117 |
bank | +0.115 |
anna | +0.115 |
Neg logit contributions | |
shield | -0.103 |
journal | -0.102 |
journals | -0.102 |
magnets | -0.097 |
pencil | -0.096 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
ashes | +0.017 |
ro | +0.017 |
clinic | +0.016 |
anna | +0.016 |
oodle | +0.015 |
Neg logit contributions | |
journals | -0.011 |
shield | -0.011 |
journal | -0.011 |
luck | -0.010 |
magnets | -0.010 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.04e+00 |
Pos logit contributions | |
narrative | +0.078 |
shield | +0.078 |
flashlight | +0.076 |
journal | +0.076 |
luck | +0.076 |
Neg logit contributions | |
ashes | -0.114 |
ro | -0.112 |
clinic | -0.111 |
bank | -0.109 |
cafe | -0.108 |
Token | , |
Token ID | 11 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
journals | +0.346 |
shield | +0.336 |
journal | +0.332 |
narrative | +0.323 |
pencil | +0.317 |
Neg logit contributions | |
ashes | -0.529 |
ro | -0.511 |
clinic | -0.499 |
rimp | -0.475 |
cafe | -0.473 |
Token | and |
Token ID | 290 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
journal | +0.024 |
journals | +0.024 |
shield | +0.024 |
luck | +0.024 |
narrative | +0.023 |
Neg logit contributions | |
ashes | -0.029 |
ro | -0.028 |
clinic | -0.027 |
rimp | -0.026 |
cafe | -0.026 |
Token | she |
Token ID | 673 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
shield | +0.024 |
journal | +0.023 |
journals | +0.023 |
narrative | +0.022 |
luck | +0.022 |
Neg logit contributions | |
ashes | -0.030 |
ro | -0.029 |
bank | -0.028 |
clinic | -0.028 |
cafe | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ashes | +0.016 |
ro | +0.016 |
clinic | +0.015 |
anna | +0.015 |
bank | +0.015 |
Neg logit contributions | |
shield | -0.011 |
journals | -0.011 |
journal | -0.011 |
luck | -0.011 |
magnets | -0.010 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
shield | +0.006 |
luck | +0.006 |
journal | +0.006 |
pencil | +0.006 |
crystal | +0.006 |
Neg logit contributions | |
ashes | -0.011 |
ro | -0.010 |
clinic | -0.010 |
Gu | -0.010 |
rimp | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
shield | +0.018 |
journals | +0.018 |
journal | +0.018 |
luck | +0.017 |
pencil | +0.017 |
Neg logit contributions | |
ashes | -0.030 |
clinic | -0.029 |
ro | -0.029 |
rimp | -0.028 |
cafe | -0.027 |
Token | name |
Token ID | 1438 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
ashes | +0.018 |
clinic | +0.017 |
ro | +0.017 |
rimp | +0.016 |
cafe | +0.016 |
Neg logit contributions | |
shield | -0.013 |
journals | -0.012 |
journal | -0.012 |
narrative | -0.012 |
pencil | -0.012 |
Token | with |
Token ID | 351 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
shield | +0.004 |
journals | +0.004 |
journal | +0.004 |
luck | +0.004 |
thumbs | +0.004 |
Neg logit contributions | |
ashes | -0.008 |
clinic | -0.008 |
ro | -0.008 |
cafe | -0.007 |
Gu | -0.007 |
Token | some |
Token ID | 617 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
ashes | +0.063 |
clinic | +0.063 |
cafe | +0.062 |
te | +0.062 |
rimp | +0.062 |
Neg logit contributions | |
narrative | -0.059 |
shield | -0.057 |
bicycle | -0.056 |
journal | -0.054 |
hetically | -0.054 |
Token | letter |
Token ID | 3850 |
Feature activation | -4.30e+00 |
Pos logit contributions | |
ashes | +0.038 |
clinic | +0.036 |
ro | +0.035 |
rimp | +0.034 |
cafe | +0.034 |
Neg logit contributions | |
shield | -0.027 |
journals | -0.027 |
journal | -0.026 |
narrative | -0.026 |
hetically | -0.026 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.03e+00 |
Pos logit contributions | |
shield | +0.154 |
journals | +0.152 |
journal | +0.149 |
luck | +0.145 |
narrative | +0.143 |
Neg logit contributions | |
ashes | -0.260 |
ro | -0.250 |
clinic | -0.249 |
rimp | -0.240 |
cafe | -0.236 |
Token | on |
Token ID | 319 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
shield | +0.305 |
journal | +0.292 |
journals | +0.289 |
pencil | +0.285 |
luck | +0.285 |
Neg logit contributions | |
ashes | -0.583 |
ro | -0.553 |
clinic | -0.550 |
rimp | -0.532 |
Gu | -0.527 |
Token | the |
Token ID | 262 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
shield | +0.138 |
luck | +0.134 |
journal | +0.131 |
journals | +0.130 |
pencil | +0.126 |
Neg logit contributions | |
ashes | -0.271 |
ro | -0.264 |
clinic | -0.262 |
rimp | -0.250 |
Gu | -0.250 |
Token | fridge |
Token ID | 25772 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
luck | +0.024 |
shield | +0.024 |
journals | +0.024 |
thumbs | +0.024 |
journal | +0.023 |
Neg logit contributions | |
ashes | -0.016 |
clinic | -0.016 |
ro | -0.016 |
cafe | -0.015 |
bank | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
pencil | +0.032 |
shield | +0.031 |
journal | +0.031 |
luck | +0.028 |
crystal | +0.028 |
Neg logit contributions | |
ashes | -0.098 |
Gu | -0.093 |
clinic | -0.092 |
ashing | -0.090 |
ro | -0.089 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
all | +0.018 |
pencil | +0.018 |
magnets | +0.017 |
mom | +0.016 |
Mom | +0.015 |
Neg logit contributions | |
rim | -0.044 |
rimp | -0.043 |
oster | -0.042 |
asis | -0.041 |
ashes | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -5.51e-01 |
Pos logit contributions | |
journals | +0.090 |
journal | +0.089 |
shield | +0.089 |
narrative | +0.087 |
pencil | +0.087 |
Neg logit contributions | |
ashes | -0.101 |
ro | -0.096 |
clinic | -0.091 |
ashing | -0.090 |
cafe | -0.088 |
Token | then |
Token ID | 788 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
shield | +0.022 |
pencil | +0.021 |
journal | +0.021 |
luck | +0.021 |
thumbs | +0.020 |
Neg logit contributions | |
ashes | -0.033 |
clinic | -0.031 |
ro | -0.030 |
Gu | -0.029 |
rimp | -0.029 |
Token | looked |
Token ID | 3114 |
Feature activation | -9.97e-01 |
Pos logit contributions | |
ashes | +0.062 |
ro | +0.060 |
clinic | +0.057 |
cafe | +0.056 |
bank | +0.055 |
Neg logit contributions | |
journals | -0.051 |
shield | -0.050 |
journal | -0.050 |
narrative | -0.049 |
magnets | -0.048 |
Token | at |
Token ID | 379 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
pencil | +0.037 |
shield | +0.036 |
journals | +0.036 |
journal | +0.035 |
crystal | +0.035 |
Neg logit contributions | |
ashes | -0.061 |
ashing | -0.056 |
clinic | -0.055 |
ro | -0.054 |
cafe | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
shield | +0.035 |
journals | +0.035 |
journal | +0.034 |
narrative | +0.032 |
spear | +0.031 |
Neg logit contributions | |
ashes | -0.052 |
ro | -0.050 |
clinic | -0.048 |
cafe | -0.047 |
bank | -0.047 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.03e+00 |
Pos logit contributions | |
pt | +0.004 |
mercy | +0.004 |
luck | +0.004 |
flashlight | +0.004 |
telescope | +0.003 |
Neg logit contributions | |
ashes | -0.006 |
ro | -0.006 |
umpy | -0.006 |
forest | -0.006 |
asher | -0.006 |
Token | again |
Token ID | 757 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
journals | +0.357 |
journal | +0.338 |
shield | +0.332 |
pencil | +0.321 |
narrative | +0.319 |
Neg logit contributions | |
ashes | -0.534 |
ro | -0.516 |
clinic | -0.500 |
bank | -0.482 |
ashing | -0.475 |
Token | . |
Token ID | 13 |
Feature activation | -5.69e-01 |
Pos logit contributions | |
shield | +0.090 |
journal | +0.086 |
journals | +0.084 |
pencil | +0.084 |
luck | +0.083 |
Neg logit contributions | |
ashes | -0.182 |
ro | -0.173 |
clinic | -0.173 |
rimp | -0.167 |
Gu | -0.166 |
Token | Now |
Token ID | 2735 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
pencil | +0.026 |
crystal | +0.025 |
luck | +0.025 |
thumbs | +0.025 |
journal | +0.025 |
Neg logit contributions | |
ashes | -0.029 |
clinic | -0.028 |
ro | -0.026 |
rimp | -0.026 |
nes | -0.026 |
Token | she |
Token ID | 673 |
Feature activation | +2.91e-01 |
Pos logit contributions | |
ashes | +0.128 |
ro | +0.121 |
clinic | +0.117 |
bank | +0.115 |
anna | +0.115 |
Neg logit contributions | |
shield | -0.103 |
journal | -0.102 |
journals | -0.102 |
magnets | -0.097 |
pencil | -0.096 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
ashes | +0.017 |
ro | +0.017 |
clinic | +0.016 |
anna | +0.016 |
oodle | +0.015 |
Neg logit contributions | |
journals | -0.011 |
shield | -0.011 |
journal | -0.011 |
luck | -0.010 |
magnets | -0.010 |
Token | careful |
Token ID | 8161 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
ashes | +0.158 |
ro | +0.149 |
clinic | +0.145 |
round | +0.139 |
Gu | +0.136 |
Neg logit contributions | |
luck | -0.152 |
shield | -0.150 |
journal | -0.144 |
pencil | -0.142 |
journals | -0.141 |
Token | , |
Token ID | 11 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
ashes | +0.004 |
clinic | +0.004 |
ro | +0.004 |
rimp | +0.004 |
Gu | +0.004 |
Neg logit contributions | |
shield | -0.002 |
journal | -0.002 |
pencil | -0.002 |
journals | -0.002 |
luck | -0.002 |
Token | my |
Token ID | 616 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
shield | +0.033 |
journal | +0.032 |
pencil | +0.032 |
luck | +0.032 |
journals | +0.032 |
Neg logit contributions | |
ashes | -0.036 |
clinic | -0.034 |
ro | -0.033 |
Gu | -0.032 |
rimp | -0.032 |
Token | darling |
Token ID | 40003 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
ro | +0.130 |
ashes | +0.122 |
oodle | +0.112 |
ashing | +0.109 |
clinic | +0.105 |
Neg logit contributions | |
narrative | -0.190 |
journals | -0.179 |
streaming | -0.175 |
bicycles | -0.168 |
telescope | -0.168 |
Token | , |
Token ID | 11 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
ashes | +0.152 |
clinic | +0.142 |
ro | +0.142 |
Gu | +0.141 |
rimp | +0.137 |
Neg logit contributions | |
shield | -0.064 |
luck | -0.062 |
pencil | -0.062 |
journal | -0.062 |
crystal | -0.058 |
Token | magnets |
Token ID | 37446 |
Feature activation | -9.00e+00 |
Pos logit contributions | |
shield | +0.023 |
journal | +0.022 |
journals | +0.022 |
pencil | +0.022 |
luck | +0.022 |
Neg logit contributions | |
ashes | -0.039 |
ro | -0.037 |
clinic | -0.037 |
rimp | -0.036 |
Gu | -0.036 |
Token | can |
Token ID | 460 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
shield | +0.249 |
journals | +0.244 |
journal | +0.240 |
pencil | +0.234 |
luck | +0.231 |
Neg logit contributions | |
ashes | -0.636 |
ro | -0.605 |
clinic | -0.600 |
rimp | -0.576 |
bank | -0.571 |
Token | be |
Token ID | 307 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
journals | +0.097 |
journal | +0.095 |
shield | +0.092 |
luck | +0.091 |
pencil | +0.088 |
Neg logit contributions | |
ashes | -0.087 |
ro | -0.083 |
clinic | -0.080 |
bank | -0.079 |
rimp | -0.077 |
Token | dangerous |
Token ID | 4923 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ashes | +0.086 |
clinic | +0.081 |
ro | +0.080 |
rimp | +0.078 |
bank | +0.078 |
Neg logit contributions | |
journal | -0.062 |
journals | -0.062 |
shield | -0.061 |
pencil | -0.059 |
narrative | -0.059 |
Token | ." |
Token ID | 526 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
shield | +0.064 |
journal | +0.062 |
luck | +0.061 |
pencil | +0.061 |
journals | +0.061 |
Neg logit contributions | |
ashes | -0.089 |
clinic | -0.084 |
ro | -0.083 |
Gu | -0.080 |
rimp | -0.080 |
Token | Alex |
Token ID | 4422 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
ashes | +0.183 |
clinic | +0.181 |
round | +0.178 |
rimp | +0.174 |
ount | +0.172 |
Neg logit contributions | |
pencil | -0.119 |
journal | -0.117 |
thumbs | -0.115 |
journals | -0.111 |
shield | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | -5.72e-01 |
Pos logit contributions | |
shield | +0.030 |
journals | +0.030 |
journal | +0.030 |
luck | +0.028 |
magnets | +0.028 |
Neg logit contributions | |
ashes | -0.029 |
ro | -0.028 |
clinic | -0.027 |
bank | -0.026 |
cafe | -0.026 |
Token | each |
Token ID | 1123 |
Feature activation | -4.88e+00 |
Pos logit contributions | |
shield | +0.026 |
journal | +0.025 |
journals | +0.025 |
magnets | +0.024 |
spear | +0.024 |
Neg logit contributions | |
ashes | -0.030 |
ro | -0.029 |
clinic | -0.028 |
bank | -0.028 |
Gu | -0.027 |
Token | time |
Token ID | 640 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
luck | +0.231 |
journals | +0.225 |
shield | +0.224 |
bicycles | +0.221 |
mercy | +0.219 |
Neg logit contributions | |
ashes | -0.218 |
ro | -0.215 |
Gu | -0.212 |
mare | -0.206 |
ashing | -0.202 |
Token | , |
Token ID | 11 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
ro | +0.151 |
ashes | +0.150 |
clinic | +0.139 |
ashing | +0.137 |
bank | +0.136 |
Neg logit contributions | |
shield | -0.129 |
journals | -0.129 |
journal | -0.126 |
spear | -0.119 |
narrative | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
journals | +0.028 |
shield | +0.027 |
journal | +0.027 |
spear | +0.026 |
magnets | +0.025 |
Neg logit contributions | |
ashes | -0.026 |
ro | -0.025 |
clinic | -0.024 |
Gu | -0.024 |
cafe | -0.023 |
Token | magnets |
Token ID | 37446 |
Feature activation | -8.99e+00 |
Pos logit contributions | |
pt | +0.000 |
flashlight | +0.000 |
telescope | +0.000 |
mercy | +0.000 |
shield | +0.000 |
Neg logit contributions | |
ashes | -0.001 |
Gu | -0.001 |
bank | -0.001 |
umpy | -0.001 |
asher | -0.001 |
Token | stayed |
Token ID | 9658 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
journals | +0.403 |
journal | +0.377 |
luck | +0.363 |
shield | +0.360 |
pencil | +0.352 |
Neg logit contributions | |
ashes | -0.486 |
ro | -0.476 |
clinic | -0.438 |
bank | -0.433 |
nes | -0.425 |
Token | stuck |
Token ID | 7819 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
ashes | +0.016 |
clinic | +0.015 |
ashing | +0.015 |
ro | +0.015 |
over | +0.014 |
Neg logit contributions | |
journals | -0.011 |
pencil | -0.011 |
journal | -0.011 |
narrative | -0.011 |
shield | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
ashes | +0.004 |
ro | +0.004 |
clinic | +0.003 |
ashing | +0.003 |
bank | +0.003 |
Neg logit contributions | |
journals | -0.002 |
shield | -0.002 |
journal | -0.002 |
pencil | -0.002 |
thumbs | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | -4.09e-02 |
Pos logit contributions | |
pencil | +0.020 |
crystal | +0.019 |
thumbs | +0.019 |
magnets | +0.019 |
luck | +0.019 |
Neg logit contributions | |
clinic | -0.023 |
ashes | -0.023 |
round | -0.021 |
rimp | -0.021 |
nes | -0.021 |
Token | Wow |
Token ID | 22017 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
pencil | +0.002 |
journals | +0.002 |
thumbs | +0.002 |
shield | +0.002 |
luck | +0.002 |
Neg logit contributions | |
ashes | -0.002 |
clinic | -0.002 |
bank | -0.002 |
oodle | -0.002 |
ro | -0.002 |
Token | sure |
Token ID | 1654 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
ashes | +0.018 |
ro | +0.017 |
clinic | +0.017 |
rimp | +0.016 |
round | +0.016 |
Neg logit contributions | |
shield | -0.013 |
luck | -0.012 |
journal | -0.012 |
pencil | -0.012 |
journals | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
shield | +0.073 |
journal | +0.072 |
journals | +0.071 |
flashlight | +0.071 |
narrative | +0.071 |
Neg logit contributions | |
ashes | -0.071 |
clinic | -0.070 |
bank | -0.070 |
ro | -0.068 |
Gu | -0.068 |
Token | keep |
Token ID | 1394 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ashes | +0.014 |
ro | +0.014 |
clinic | +0.014 |
rimp | +0.013 |
Gu | +0.013 |
Neg logit contributions | |
shield | -0.012 |
journal | -0.011 |
journals | -0.011 |
luck | -0.011 |
pencil | -0.011 |
Token | all |
Token ID | 477 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
narrative | +0.077 |
iness | +0.073 |
flashlight | +0.070 |
journal | +0.070 |
shield | +0.068 |
Neg logit contributions | |
Gu | -0.107 |
bank | -0.105 |
clinic | -0.105 |
rimp | -0.102 |
ashes | -0.101 |
Token | his |
Token ID | 465 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
shield | +0.067 |
journal | +0.062 |
journals | +0.060 |
flashlight | +0.059 |
pencil | +0.059 |
Neg logit contributions | |
ashes | -0.207 |
ro | -0.198 |
clinic | -0.195 |
rimp | -0.194 |
Gu | -0.192 |
Token | magnets |
Token ID | 37446 |
Feature activation | -8.99e+00 |
Pos logit contributions | |
narrative | +0.058 |
telescope | +0.053 |
streaming | +0.053 |
mercy | +0.050 |
pt | +0.049 |
Neg logit contributions | |
ashes | -0.067 |
clinic | -0.064 |
ro | -0.064 |
dam | -0.062 |
While | -0.061 |
Token | in |
Token ID | 287 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
narrative | +0.471 |
journals | +0.449 |
bicycles | +0.403 |
telescope | +0.403 |
journal | +0.400 |
Neg logit contributions | |
ro | -0.421 |
ashes | -0.400 |
nes | -0.387 |
ashing | -0.381 |
uring | -0.370 |
Token | a |
Token ID | 257 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
shield | +0.011 |
journals | +0.010 |
journal | +0.010 |
thumbs | +0.010 |
luck | +0.010 |
Neg logit contributions | |
ro | -0.023 |
clinic | -0.022 |
ashes | -0.021 |
bank | -0.021 |
cafe | -0.021 |
Token | safe |
Token ID | 3338 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
shield | +0.030 |
luck | +0.030 |
journals | +0.029 |
journal | +0.029 |
pencil | +0.029 |
Neg logit contributions | |
ashes | -0.045 |
clinic | -0.044 |
ro | -0.043 |
Gu | -0.042 |
cafe | -0.041 |
Token | place |
Token ID | 1295 |
Feature activation | +5.01e+00 |
Pos logit contributions | |
ashes | +0.006 |
ro | +0.006 |
clinic | +0.006 |
bank | +0.006 |
cafe | +0.005 |
Neg logit contributions | |
shield | -0.011 |
luck | -0.011 |
journal | -0.011 |
journals | -0.011 |
pencil | -0.010 |
Token | so |
Token ID | 523 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
ashes | +0.264 |
ro | +0.263 |
clinic | +0.252 |
bank | +0.250 |
ashing | +0.242 |
Neg logit contributions | |
shield | -0.215 |
journals | -0.212 |
narrative | -0.208 |
thumbs | -0.205 |
journal | -0.205 |
Token | was |
Token ID | 373 |
Feature activation | +4.74e-01 |
Pos logit contributions | |
shield | +0.004 |
journals | +0.004 |
pencil | +0.004 |
journal | +0.004 |
magnets | +0.004 |
Neg logit contributions | |
ashes | -0.005 |
clinic | -0.005 |
ro | -0.005 |
bank | -0.005 |
cafe | -0.005 |
Token | playing |
Token ID | 2712 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
ashes | +0.030 |
ro | +0.028 |
clinic | +0.028 |
Gu | +0.027 |
round | +0.027 |
Neg logit contributions | |
shield | -0.017 |
luck | -0.017 |
journal | -0.017 |
pencil | -0.016 |
crystal | -0.016 |
Token | with |
Token ID | 351 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
ashes | +0.083 |
clinic | +0.078 |
ro | +0.078 |
Gu | +0.076 |
rimp | +0.076 |
Neg logit contributions | |
shield | -0.045 |
journal | -0.043 |
pencil | -0.042 |
luck | -0.041 |
journals | -0.041 |
Token | his |
Token ID | 465 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ashes | +0.093 |
ro | +0.089 |
clinic | +0.087 |
bank | +0.084 |
cafe | +0.084 |
Neg logit contributions | |
luck | -0.047 |
shield | -0.047 |
journals | -0.047 |
journal | -0.047 |
narrative | -0.046 |
Token | coloured |
Token ID | 34746 |
Feature activation | +9.57e-01 |
Pos logit contributions | |
narrative | +0.056 |
luck | +0.055 |
journals | +0.055 |
streaming | +0.053 |
correctly | +0.052 |
Neg logit contributions | |
ashes | -0.072 |
ro | -0.068 |
clinic | -0.067 |
te | -0.065 |
oodle | -0.064 |
Token | pencil |
Token ID | 21613 |
Feature activation | -8.99e+00 |
Pos logit contributions | |
ashes | +0.053 |
ro | +0.053 |
clinic | +0.052 |
cafe | +0.049 |
round | +0.048 |
Neg logit contributions | |
luck | -0.037 |
narrative | -0.036 |
journals | -0.035 |
shield | -0.034 |
journal | -0.034 |
Token | s |
Token ID | 82 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
shield | +0.335 |
journals | +0.328 |
journal | +0.323 |
luck | +0.323 |
narrative | +0.311 |
Neg logit contributions | |
ashes | -0.533 |
ro | -0.527 |
clinic | -0.501 |
bank | -0.486 |
rimp | -0.483 |
Token | . |
Token ID | 13 |
Feature activation | -5.83e-01 |
Pos logit contributions | |
ro | +0.104 |
ashes | +0.098 |
clinic | +0.090 |
ashing | +0.089 |
bank | +0.088 |
Neg logit contributions | |
journals | -0.081 |
narrative | -0.080 |
thumbs | -0.077 |
shield | -0.077 |
telescope | -0.075 |
Token | He |
Token ID | 679 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
magnets | +0.021 |
pencil | +0.020 |
his | +0.020 |
shield | +0.019 |
journal | +0.019 |
Neg logit contributions | |
ashes | -0.038 |
clinic | -0.034 |
rim | -0.034 |
ashing | -0.033 |
inconvenience | -0.033 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
luck | +0.111 |
shield | +0.109 |
journal | +0.107 |
journals | +0.106 |
narrative | +0.102 |
Neg logit contributions | |
ashes | -0.139 |
ro | -0.137 |
clinic | -0.132 |
Gu | -0.127 |
bank | -0.126 |
Token | to |
Token ID | 284 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
narrative | +0.081 |
luck | +0.080 |
shield | +0.080 |
journal | +0.077 |
journals | +0.075 |
Neg logit contributions | |
ashes | -0.115 |
ro | -0.114 |
clinic | -0.109 |
Gu | -0.108 |
bank | -0.108 |
Token | Everyday |
Token ID | 48154 |
Feature activation | -2.48e-01 |
Pos logit contributions | |
ashes | +0.031 |
ro | +0.029 |
nes | +0.029 |
anna | +0.029 |
ashing | +0.029 |
Neg logit contributions | |
shield | -0.017 |
journal | -0.017 |
journals | -0.017 |
luck | -0.016 |
pencil | -0.016 |
Token | he |
Token ID | 339 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
shield | +0.007 |
journal | +0.007 |
pencil | +0.007 |
luck | +0.007 |
journals | +0.006 |
Neg logit contributions | |
ashes | -0.018 |
ro | -0.017 |
clinic | -0.017 |
Gu | -0.016 |
rimp | -0.016 |
Token | would |
Token ID | 561 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
shield | +0.107 |
journal | +0.102 |
pencil | +0.101 |
luck | +0.100 |
journals | +0.100 |
Neg logit contributions | |
ashes | -0.207 |
clinic | -0.197 |
ro | -0.196 |
Gu | -0.190 |
rimp | -0.189 |
Token | take |
Token ID | 1011 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
journals | +0.144 |
journal | +0.144 |
shield | +0.144 |
pencil | +0.141 |
luck | +0.141 |
Neg logit contributions | |
ashes | -0.177 |
clinic | -0.172 |
ro | -0.171 |
bank | -0.161 |
rimp | -0.160 |
Token | the |
Token ID | 262 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
journal | +0.085 |
narrative | +0.084 |
shield | +0.084 |
crystal | +0.082 |
journals | +0.080 |
Neg logit contributions | |
ashes | -0.168 |
ro | -0.163 |
clinic | -0.159 |
anna | -0.156 |
cafe | -0.151 |
Token | av |
Token ID | 1196 |
Feature activation | -8.99e+00 |
Pos logit contributions | |
journals | +0.041 |
narrative | +0.039 |
shield | +0.038 |
luck | +0.038 |
journal | +0.038 |
Neg logit contributions | |
ashes | -0.058 |
clinic | -0.057 |
ro | -0.057 |
cafe | -0.055 |
te | -0.052 |
Token | oc |
Token ID | 420 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
shield | +0.507 |
journal | +0.497 |
luck | +0.490 |
journals | +0.488 |
pencil | +0.487 |
Neg logit contributions | |
ashes | -0.375 |
ro | -0.345 |
clinic | -0.342 |
Gu | -0.324 |
rimp | -0.323 |
Token | ados |
Token ID | 22484 |
Feature activation | -6.99e+00 |
Pos logit contributions | |
flashlight | +0.096 |
shield | +0.094 |
journal | +0.091 |
crystal | +0.087 |
journals | +0.086 |
Neg logit contributions | |
ashes | -0.122 |
clinic | -0.119 |
bank | -0.118 |
ro | -0.116 |
round | -0.114 |
Token | and |
Token ID | 290 |
Feature activation | -9.45e-01 |
Pos logit contributions | |
shield | +0.289 |
journals | +0.287 |
journal | +0.282 |
pencil | +0.276 |
luck | +0.274 |
Neg logit contributions | |
ashes | -0.386 |
ro | -0.373 |
clinic | -0.362 |
bank | -0.350 |
rimp | -0.346 |
Token | scatter |
Token ID | 41058 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
journal | +0.038 |
journals | +0.037 |
shield | +0.036 |
luck | +0.036 |
narrative | +0.036 |
Neg logit contributions | |
ashes | -0.050 |
ro | -0.049 |
bank | -0.049 |
clinic | -0.049 |
cafe | -0.047 |
Token | them |
Token ID | 606 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
ashes | +0.023 |
ro | +0.021 |
bank | +0.020 |
cafe | +0.019 |
rimp | +0.019 |
Neg logit contributions | |
journal | -0.021 |
shield | -0.021 |
flashlight | -0.021 |
narrative | -0.020 |
journals | -0.020 |
Token | frog |
Token ID | 21264 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
ashes | +0.145 |
clinic | +0.135 |
ro | +0.133 |
rimp | +0.129 |
Gu | +0.128 |
Neg logit contributions | |
shield | -0.156 |
journal | -0.151 |
pencil | -0.150 |
luck | -0.148 |
crystal | -0.146 |
Token | was |
Token ID | 373 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
ashes | +0.071 |
clinic | +0.065 |
ro | +0.064 |
Gu | +0.060 |
rimp | +0.060 |
Neg logit contributions | |
shield | -0.121 |
journal | -0.118 |
pencil | -0.117 |
crystal | -0.116 |
journals | -0.115 |
Token | hopping |
Token ID | 47153 |
Feature activation | +5.07e-01 |
Pos logit contributions | |
shield | +0.001 |
all | +0.001 |
on | +0.000 |
hard | +0.000 |
luck | +0.000 |
Neg logit contributions | |
ashes | -0.001 |
ro | -0.001 |
round | -0.001 |
cafe | -0.001 |
Gu | -0.001 |
Token | through |
Token ID | 832 |
Feature activation | -7.95e-01 |
Pos logit contributions | |
ashes | +0.033 |
ro | +0.031 |
clinic | +0.030 |
ashing | +0.030 |
rimp | +0.029 |
Neg logit contributions | |
shield | -0.017 |
journal | -0.017 |
journals | -0.017 |
pencil | -0.017 |
luck | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
shield | +0.021 |
journal | +0.020 |
pencil | +0.019 |
luck | +0.019 |
journals | +0.019 |
Neg logit contributions | |
ashes | -0.058 |
clinic | -0.055 |
ro | -0.055 |
Gu | -0.053 |
rimp | -0.053 |
Token | pond |
Token ID | 16723 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
shield | +0.002 |
crystal | +0.002 |
journal | +0.002 |
pencil | +0.001 |
luck | +0.001 |
Neg logit contributions | |
ashes | -0.001 |
clinic | -0.001 |
rimp | -0.001 |
ro | -0.001 |
Gu | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
ashes | +0.386 |
Gu | +0.377 |
clinic | +0.377 |
ro | +0.372 |
oodle | +0.367 |
Neg logit contributions | |
shield | -0.129 |
journal | -0.124 |
luck | -0.120 |
pencil | -0.117 |
crystal | -0.117 |
Token | He |
Token ID | 679 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
shield | +0.025 |
spear | +0.024 |
his | +0.023 |
pencil | +0.022 |
journal | +0.022 |
Neg logit contributions | |
nes | -0.045 |
clinic | -0.042 |
ashes | -0.042 |
anna | -0.041 |
rimp | -0.040 |
Token | wanted |
Token ID | 2227 |
Feature activation | -7.40e-01 |
Pos logit contributions | |
shield | +0.091 |
pencil | +0.086 |
journal | +0.086 |
luck | +0.083 |
crystal | +0.083 |
Neg logit contributions | |
ashes | -0.162 |
clinic | -0.154 |
ro | -0.150 |
rimp | -0.147 |
Gu | -0.147 |
Token | to |
Token ID | 284 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
journals | +0.035 |
journal | +0.034 |
shield | +0.034 |
narrative | +0.034 |
pencil | +0.033 |
Neg logit contributions | |
ashes | -0.038 |
ro | -0.035 |
clinic | -0.035 |
Gu | -0.034 |
cafe | -0.033 |
Token | find |
Token ID | 1064 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
ashes | +0.010 |
ashing | +0.010 |
clinic | +0.010 |
oodle | +0.010 |
anna | +0.010 |
Neg logit contributions | |
shield | -0.007 |
pencil | -0.006 |
crystal | -0.006 |
journal | -0.006 |
spear | -0.006 |
Token | there |
Token ID | 612 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
shield | +0.014 |
journals | +0.014 |
pencil | +0.014 |
luck | +0.014 |
journal | +0.014 |
Neg logit contributions | |
ashes | -0.018 |
ro | -0.017 |
clinic | -0.017 |
Gu | -0.016 |
rimp | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ashes | +0.112 |
ro | +0.110 |
clinic | +0.108 |
Gu | +0.105 |
cafe | +0.105 |
Neg logit contributions | |
shield | -0.081 |
journals | -0.079 |
luck | -0.077 |
journal | -0.076 |
pencil | -0.075 |
Token | a |
Token ID | 257 |
Feature activation | -6.99e-01 |
Pos logit contributions | |
shield | +0.004 |
journal | +0.004 |
pencil | +0.003 |
journals | +0.003 |
crystal | +0.003 |
Neg logit contributions | |
ashes | -0.013 |
ro | -0.013 |
clinic | -0.013 |
bank | -0.012 |
anna | -0.012 |
Token | little |
Token ID | 1310 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
shield | +0.031 |
journal | +0.029 |
pencil | +0.029 |
crystal | +0.028 |
luck | +0.028 |
Neg logit contributions | |
ashes | -0.040 |
ro | -0.037 |
clinic | -0.037 |
bank | -0.036 |
rimp | -0.036 |
Token | boy |
Token ID | 2933 |
Feature activation | +8.34e-01 |
Pos logit contributions | |
ashes | +0.142 |
clinic | +0.132 |
ro | +0.131 |
Gu | +0.125 |
rimp | +0.124 |
Neg logit contributions | |
shield | -0.162 |
journal | -0.158 |
pencil | -0.157 |
luck | -0.156 |
journals | -0.155 |
Token | named |
Token ID | 3706 |
Feature activation | +7.04e+00 |
Pos logit contributions | |
ashes | +0.025 |
ro | +0.022 |
clinic | +0.022 |
Gu | +0.021 |
rimp | +0.020 |
Neg logit contributions | |
shield | -0.057 |
journal | -0.056 |
pencil | -0.056 |
journals | -0.055 |
luck | -0.055 |
Token | Tim |
Token ID | 5045 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
ashes | +0.392 |
clinic | +0.369 |
ro | +0.365 |
Gu | +0.350 |
rimp | +0.349 |
Neg logit contributions | |
shield | -0.310 |
journal | -0.300 |
pencil | -0.298 |
journals | -0.293 |
luck | -0.293 |
Token | my |
Token ID | 1820 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
luck | +0.020 |
shield | +0.020 |
journals | +0.020 |
narrative | +0.019 |
thumbs | +0.019 |
Neg logit contributions | |
ashes | -0.024 |
ro | -0.023 |
ashing | -0.023 |
clinic | -0.023 |
Gu | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
luck | +0.162 |
journals | +0.156 |
narrative | +0.155 |
shield | +0.152 |
thumbs | +0.151 |
Neg logit contributions | |
ashes | -0.177 |
clinic | -0.172 |
ro | -0.169 |
ashing | -0.167 |
Gu | -0.164 |
Token | Tim |
Token ID | 5045 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
shield | +0.024 |
journal | +0.023 |
pencil | +0.023 |
luck | +0.023 |
journals | +0.023 |
Neg logit contributions | |
ashes | -0.029 |
clinic | -0.027 |
ro | -0.027 |
Gu | -0.026 |
rimp | -0.026 |
Token | my |
Token ID | 1820 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
luck | +0.037 |
shield | +0.036 |
bravery | +0.035 |
journals | +0.035 |
arrows | +0.035 |
Neg logit contributions | |
ashes | -0.040 |
ro | -0.040 |
Gu | -0.039 |
anna | -0.039 |
forest | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +7.67e-02 |
Pos logit contributions | |
ashes | +0.262 |
ro | +0.251 |
Gu | +0.249 |
clinic | +0.249 |
rimp | +0.240 |
Neg logit contributions | |
shield | -0.091 |
pencil | -0.084 |
journal | -0.084 |
luck | -0.083 |
crystal | -0.080 |
Token | She |
Token ID | 1375 |
Feature activation | +9.05e-01 |
Pos logit contributions | |
ashes | +0.005 |
rimp | +0.005 |
clinic | +0.005 |
nes | +0.004 |
anna | +0.004 |
Neg logit contributions | |
pencil | -0.003 |
journal | -0.003 |
magnets | -0.003 |
thumbs | -0.003 |
shield | -0.002 |
Token | runs |
Token ID | 4539 |
Feature activation | +9.66e-02 |
Pos logit contributions | |
ashes | +0.051 |
ro | +0.048 |
clinic | +0.048 |
rimp | +0.047 |
Gu | +0.047 |
Neg logit contributions | |
shield | -0.038 |
luck | -0.036 |
pencil | -0.036 |
journal | -0.036 |
journals | -0.035 |
Token | to |
Token ID | 284 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
rimp | +0.008 |
clinic | +0.007 |
rim | +0.007 |
phony | +0.007 |
anna | +0.007 |
Neg logit contributions | |
on | -0.002 |
all | -0.002 |
hard | -0.002 |
everywhere | -0.002 |
back | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
ashes | +0.034 |
ro | +0.032 |
clinic | +0.032 |
ashing | +0.032 |
anna | +0.032 |
Neg logit contributions | |
shield | -0.010 |
pencil | -0.010 |
journal | -0.009 |
luck | -0.009 |
crystal | -0.009 |
Token | pond |
Token ID | 16723 |
Feature activation | +5.72e+00 |
Pos logit contributions | |
anna | +0.006 |
rimp | +0.005 |
nes | +0.005 |
Gu | +0.005 |
ashes | +0.005 |
Neg logit contributions | |
shield | -0.008 |
crystal | -0.008 |
luck | -0.008 |
spear | -0.008 |
flashlight | -0.008 |
Token | and |
Token ID | 290 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
Gu | +0.402 |
ashes | +0.399 |
anna | +0.390 |
clinic | +0.388 |
nes | +0.385 |
Neg logit contributions | |
shield | -0.143 |
journal | -0.136 |
all | -0.134 |
pencil | -0.131 |
luck | -0.130 |
Token | puts |
Token ID | 7584 |
Feature activation | +6.82e-01 |
Pos logit contributions | |
shield | +0.001 |
pencil | +0.001 |
thumbs | +0.001 |
flashlight | +0.001 |
journal | +0.001 |
Neg logit contributions | |
ashes | -0.002 |
over | -0.002 |
rimp | -0.002 |
ro | -0.002 |
nes | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | +8.55e-01 |
Pos logit contributions | |
ashes | +0.041 |
ro | +0.038 |
clinic | +0.037 |
Gu | +0.036 |
cafe | +0.036 |
Neg logit contributions | |
shield | -0.027 |
journal | -0.026 |
journals | -0.026 |
pencil | -0.025 |
luck | -0.024 |
Token | duck |
Token ID | 22045 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
ashes | +0.031 |
clinic | +0.028 |
ro | +0.028 |
rimp | +0.025 |
Gu | +0.025 |
Neg logit contributions | |
shield | -0.053 |
journals | -0.053 |
journal | -0.053 |
pencil | -0.052 |
luck | -0.052 |
Token | on |
Token ID | 319 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
ashes | +0.266 |
ro | +0.252 |
clinic | +0.247 |
ashing | +0.240 |
rimp | +0.239 |
Neg logit contributions | |
journals | -0.116 |
shield | -0.116 |
journal | -0.115 |
luck | -0.114 |
pencil | -0.114 |
Token | helping |
Token ID | 5742 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
all | +0.006 |
luck | +0.005 |
pencil | +0.005 |
journal | +0.005 |
hard | +0.005 |
Neg logit contributions | |
ashes | -0.014 |
ro | -0.014 |
clinic | -0.013 |
round | -0.013 |
ree | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -6.26e-01 |
Pos logit contributions | |
shield | +0.097 |
luck | +0.093 |
journal | +0.090 |
journals | +0.090 |
narrative | +0.089 |
Neg logit contributions | |
ashes | -0.175 |
ro | -0.167 |
clinic | -0.166 |
rimp | -0.162 |
cafe | -0.160 |
Token | Emily |
Token ID | 17608 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
pencil | +0.028 |
journal | +0.028 |
shield | +0.028 |
magnets | +0.028 |
journals | +0.028 |
Neg logit contributions | |
ashes | -0.034 |
clinic | -0.033 |
rimp | -0.031 |
round | -0.031 |
ro | -0.031 |
Token | wiped |
Token ID | 21122 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
shield | +0.040 |
journal | +0.038 |
pencil | +0.038 |
luck | +0.037 |
journals | +0.037 |
Neg logit contributions | |
ashes | -0.069 |
clinic | -0.065 |
ro | -0.064 |
Gu | -0.062 |
rimp | -0.062 |
Token | the |
Token ID | 262 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
shield | +0.116 |
journal | +0.109 |
journals | +0.107 |
crystal | +0.107 |
narrative | +0.107 |
Neg logit contributions | |
ashes | -0.185 |
ro | -0.176 |
clinic | -0.166 |
anna | -0.165 |
cafe | -0.163 |
Token | counter |
Token ID | 3753 |
Feature activation | +5.63e+00 |
Pos logit contributions | |
ashes | +0.001 |
ro | +0.000 |
clinic | +0.000 |
cafe | +0.000 |
anna | +0.000 |
Neg logit contributions | |
shield | -0.001 |
luck | -0.001 |
journals | -0.001 |
narrative | -0.001 |
pencil | -0.001 |
Token | clean |
Token ID | 3424 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
ashes | +0.286 |
ro | +0.271 |
clinic | +0.261 |
bank | +0.256 |
ashing | +0.255 |
Neg logit contributions | |
shield | -0.262 |
journals | -0.258 |
narrative | -0.254 |
journal | -0.246 |
spear | -0.243 |
Token | with |
Token ID | 351 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
shield | +0.032 |
journal | +0.031 |
pencil | +0.031 |
luck | +0.031 |
journals | +0.030 |
Neg logit contributions | |
ashes | -0.042 |
clinic | -0.039 |
ro | -0.039 |
Gu | -0.038 |
rimp | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
ashes | +0.065 |
cafe | +0.062 |
ro | +0.060 |
clinic | +0.060 |
te | +0.058 |
Neg logit contributions | |
shield | -0.055 |
narrative | -0.053 |
sword | -0.051 |
journal | -0.051 |
spear | -0.050 |
Token | cloth |
Token ID | 16270 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
journals | +0.034 |
shield | +0.034 |
luck | +0.033 |
narrative | +0.032 |
magnets | +0.032 |
Neg logit contributions | |
ashes | -0.039 |
clinic | -0.038 |
cafe | -0.037 |
ro | -0.037 |
oodle | -0.035 |
Token | and |
Token ID | 290 |
Feature activation | -5.63e-01 |
Pos logit contributions | |
journals | +0.120 |
shield | +0.120 |
journal | +0.117 |
luck | +0.112 |
narrative | +0.110 |
Neg logit contributions | |
ashes | -0.177 |
ro | -0.174 |
clinic | -0.164 |
bank | -0.163 |
ashing | -0.159 |
Token | there |
Token ID | 612 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
shield | +0.033 |
journals | +0.033 |
luck | +0.032 |
pencil | +0.032 |
journal | +0.032 |
Neg logit contributions | |
ashes | -0.043 |
ro | -0.041 |
clinic | -0.040 |
Gu | -0.039 |
ashing | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
ashes | +0.085 |
ro | +0.083 |
clinic | +0.082 |
Gu | +0.079 |
cafe | +0.079 |
Neg logit contributions | |
shield | -0.060 |
journals | -0.059 |
luck | -0.057 |
journal | -0.057 |
pencil | -0.056 |
Token | a |
Token ID | 257 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
shield | +0.011 |
journal | +0.011 |
pencil | +0.010 |
magnets | +0.010 |
crystal | +0.010 |
Neg logit contributions | |
ashes | -0.042 |
clinic | -0.041 |
ro | -0.041 |
bank | -0.041 |
anna | -0.039 |
Token | little |
Token ID | 1310 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
shield | +0.045 |
journal | +0.044 |
pencil | +0.043 |
crystal | +0.042 |
luck | +0.041 |
Neg logit contributions | |
ashes | -0.061 |
ro | -0.057 |
clinic | -0.056 |
bank | -0.056 |
rimp | -0.055 |
Token | girl |
Token ID | 2576 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
ashes | +0.127 |
clinic | +0.119 |
ro | +0.118 |
Gu | +0.114 |
rimp | +0.113 |
Neg logit contributions | |
shield | -0.138 |
journal | -0.134 |
pencil | -0.134 |
luck | -0.132 |
crystal | -0.130 |
Token | named |
Token ID | 3706 |
Feature activation | +6.48e+00 |
Pos logit contributions | |
ashes | +0.104 |
ro | +0.090 |
Gu | +0.087 |
clinic | +0.085 |
ount | +0.083 |
Neg logit contributions | |
pencil | -0.159 |
journal | -0.159 |
shield | -0.158 |
crystal | -0.154 |
journals | -0.151 |
Token | Amy |
Token ID | 14235 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
ashes | +0.355 |
clinic | +0.351 |
bank | +0.348 |
rimp | +0.343 |
ro | +0.335 |
Neg logit contributions | |
all | -0.283 |
journal | -0.281 |
pencil | -0.280 |
crystal | -0.272 |
everywhere | -0.269 |
Token | . |
Token ID | 13 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
luck | +0.039 |
shield | +0.039 |
thumbs | +0.038 |
journals | +0.038 |
hetically | +0.038 |
Neg logit contributions | |
clinic | -0.036 |
ashes | -0.034 |
ro | -0.034 |
anna | -0.034 |
Gu | -0.033 |
Token | She |
Token ID | 1375 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
shield | +0.053 |
luck | +0.052 |
journal | +0.051 |
journals | +0.051 |
pencil | +0.051 |
Neg logit contributions | |
ashes | -0.035 |
ro | -0.032 |
clinic | -0.032 |
Gu | -0.031 |
cafe | -0.030 |
Token | had |
Token ID | 550 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
shield | +0.004 |
journal | +0.004 |
luck | +0.004 |
pencil | +0.004 |
journals | +0.004 |
Neg logit contributions | |
ashes | -0.008 |
clinic | -0.007 |
ro | -0.007 |
Gu | -0.007 |
rimp | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
shield | +0.090 |
journal | +0.087 |
pencil | +0.086 |
luck | +0.085 |
journals | +0.084 |
Neg logit contributions | |
ashes | -0.204 |
clinic | -0.194 |
ro | -0.194 |
rimp | -0.187 |
Gu | -0.186 |
Token | Do |
Token ID | 2141 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
ashes | +0.020 |
clinic | +0.018 |
ashing | +0.018 |
ro | +0.018 |
bank | +0.017 |
Neg logit contributions | |
shield | -0.020 |
journal | -0.019 |
pencil | -0.019 |
journals | -0.019 |
luck | -0.018 |
Token | you |
Token ID | 345 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
ashes | +0.064 |
ro | +0.062 |
Gu | +0.062 |
clinic | +0.061 |
rimp | +0.061 |
Neg logit contributions | |
shield | -0.133 |
crystal | -0.130 |
journal | -0.129 |
flashlight | -0.128 |
streaming | -0.127 |
Token | want |
Token ID | 765 |
Feature activation | -4.03e-01 |
Pos logit contributions | |
ashes | +0.037 |
clinic | +0.036 |
ro | +0.035 |
Gu | +0.034 |
bank | +0.034 |
Neg logit contributions | |
shield | -0.039 |
luck | -0.038 |
journal | -0.038 |
pencil | -0.038 |
journals | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
shield | +0.018 |
journal | +0.018 |
journals | +0.018 |
narrative | +0.018 |
luck | +0.017 |
Neg logit contributions | |
ashes | -0.020 |
clinic | -0.019 |
Gu | -0.019 |
ro | -0.019 |
cafe | -0.018 |
Token | play |
Token ID | 711 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ashes | +0.024 |
clinic | +0.022 |
ro | +0.022 |
ashing | +0.021 |
Gu | +0.021 |
Neg logit contributions | |
shield | -0.018 |
pencil | -0.017 |
journal | -0.017 |
luck | -0.017 |
crystal | -0.016 |
Token | with |
Token ID | 351 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
ashes | +0.073 |
ro | +0.068 |
clinic | +0.068 |
ashing | +0.066 |
bank | +0.066 |
Neg logit contributions | |
shield | -0.050 |
journal | -0.048 |
pencil | -0.047 |
journals | -0.045 |
crystal | -0.045 |
Token | us |
Token ID | 514 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
clinic | +0.076 |
ashes | +0.076 |
ro | +0.074 |
Gu | +0.072 |
cafe | +0.071 |
Neg logit contributions | |
luck | -0.079 |
narrative | -0.079 |
shield | -0.078 |
journal | -0.077 |
thumbs | -0.076 |
Token | ?" |
Token ID | 1701 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
ashes | +0.186 |
clinic | +0.176 |
ro | +0.174 |
rimp | +0.165 |
Gu | +0.163 |
Neg logit contributions | |
shield | -0.208 |
journals | -0.202 |
journal | -0.202 |
luck | -0.200 |
pencil | -0.200 |
Token |
|
Token ID | 198 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
ashes | +0.181 |
clinic | +0.165 |
ro | +0.163 |
rimp | +0.161 |
ashing | +0.160 |
Neg logit contributions | |
shield | -0.152 |
pencil | -0.151 |
journal | -0.148 |
journals | -0.143 |
magnets | -0.141 |
Token |
|
Token ID | 198 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
shield | +0.007 |
pencil | +0.007 |
journal | +0.007 |
journals | +0.007 |
luck | +0.007 |
Neg logit contributions | |
ashes | -0.007 |
clinic | -0.007 |
ro | -0.007 |
rimp | -0.007 |
Gu | -0.007 |
Token | L |
Token ID | 43 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
shield | +0.008 |
pencil | +0.008 |
journal | +0.007 |
journals | +0.007 |
luck | +0.007 |
Neg logit contributions | |
ashes | -0.007 |
clinic | -0.007 |
ro | -0.007 |
ashing | -0.006 |
rimp | -0.006 |
Token | have |
Token ID | 423 |
Feature activation | +5.76e-01 |
Pos logit contributions | |
ashes | +0.030 |
clinic | +0.029 |
ashing | +0.028 |
anna | +0.028 |
oodle | +0.028 |
Neg logit contributions | |
shield | -0.020 |
pencil | -0.019 |
thumbs | -0.018 |
journal | -0.018 |
crystal | -0.018 |
Token | it |
Token ID | 340 |
Feature activation | +2.44e-03 |
Pos logit contributions | |
ashes | +0.034 |
clinic | +0.032 |
Gu | +0.032 |
ro | +0.031 |
cafe | +0.031 |
Neg logit contributions | |
shield | -0.022 |
journal | -0.021 |
journals | -0.020 |
luck | -0.020 |
pencil | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
ashes | +0.000 |
clinic | +0.000 |
ro | +0.000 |
rimp | +0.000 |
ashing | +0.000 |
Neg logit contributions | |
shield | -0.000 |
journals | -0.000 |
journal | -0.000 |
pencil | -0.000 |
thumbs | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
thumbs | +0.005 |
bicycles | +0.005 |
his | +0.004 |
pencil | +0.004 |
all | +0.004 |
Neg logit contributions | |
clinic | -0.010 |
rim | -0.010 |
eh | -0.010 |
mare | -0.010 |
oster | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
hard | +0.010 |
pencil | +0.010 |
thumbs | +0.010 |
his | +0.009 |
journal | +0.009 |
Neg logit contributions | |
ashes | -0.025 |
clinic | -0.023 |
ro | -0.022 |
asher | -0.022 |
ashing | -0.022 |
Token | " |
Token ID | 1 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
thumbs | +0.010 |
pencil | +0.010 |
between | +0.010 |
hard | +0.010 |
bicycles | +0.009 |
Neg logit contributions | |
ashes | -0.026 |
clinic | -0.026 |
alis | -0.024 |
bank | -0.024 |
ro | -0.024 |
Token | Look |
Token ID | 8567 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
ashes | +0.110 |
clinic | +0.107 |
anna | +0.102 |
bank | +0.101 |
ro | +0.100 |
Neg logit contributions | |
pencil | -0.098 |
thumbs | -0.091 |
shield | -0.090 |
journal | -0.090 |
crystal | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
ashes | +0.009 |
oodle | +0.008 |
anna | +0.008 |
ashing | +0.008 |
round | +0.008 |
Neg logit contributions | |
narrative | -0.007 |
journals | -0.007 |
shield | -0.007 |
journal | -0.006 |
magnets | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -3.94e-01 |
Pos logit contributions | |
shield | +0.005 |
pencil | +0.005 |
journal | +0.005 |
luck | +0.005 |
crystal | +0.004 |
Neg logit contributions | |
ashes | -0.007 |
clinic | -0.006 |
ro | -0.006 |
bank | -0.006 |
ashing | -0.006 |
Token | balloon |
Token ID | 21190 |
Feature activation | -9.47e-01 |
Pos logit contributions | |
shield | +0.015 |
journals | +0.015 |
journal | +0.015 |
luck | +0.015 |
pencil | +0.015 |
Neg logit contributions | |
ashes | -0.023 |
clinic | -0.022 |
ro | -0.022 |
Gu | -0.021 |
rimp | -0.021 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
shield | +0.051 |
journal | +0.049 |
luck | +0.049 |
pencil | +0.049 |
crystal | +0.048 |
Neg logit contributions | |
ashes | -0.043 |
clinic | -0.040 |
ro | -0.040 |
Gu | -0.038 |
rimp | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | -6.50e-01 |
Pos logit contributions | |
shield | +0.087 |
journal | +0.084 |
journals | +0.083 |
luck | +0.081 |
pencil | +0.081 |
Neg logit contributions | |
ashes | -0.150 |
ro | -0.144 |
clinic | -0.142 |
rimp | -0.136 |
cafe | -0.135 |
Token |
|
Token ID | 198 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
magnets | +0.021 |
pencil | +0.019 |
all | +0.018 |
his | +0.018 |
magnet | +0.018 |
Neg logit contributions | |
ashes | -0.045 |
clinic | -0.041 |
rim | -0.041 |
ashing | -0.040 |
anna | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -7.91e-01 |
Pos logit contributions | |
pencil | +0.028 |
shield | +0.028 |
journal | +0.028 |
crystal | +0.026 |
flashlight | +0.026 |
Neg logit contributions | |
ashes | -0.049 |
ro | -0.045 |
clinic | -0.045 |
ashing | -0.043 |
rimp | -0.043 |
Token | A |
Token ID | 32 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
shield | +0.021 |
pencil | +0.021 |
spear | +0.021 |
hard | +0.020 |
magnets | +0.020 |
Neg logit contributions | |
ashes | -0.059 |
clinic | -0.052 |
angers | -0.051 |
ount | -0.050 |
anna | -0.050 |
Token | family |
Token ID | 1641 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
ashes | +0.033 |
clinic | +0.029 |
ro | +0.029 |
rimp | +0.029 |
ount | +0.029 |
Neg logit contributions | |
shield | -0.023 |
pencil | -0.023 |
journal | -0.022 |
crystal | -0.021 |
luck | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
shield | +0.084 |
journal | +0.083 |
pencil | +0.082 |
luck | +0.080 |
journals | +0.080 |
Neg logit contributions | |
ashes | -0.155 |
ro | -0.147 |
clinic | -0.145 |
rimp | -0.139 |
Gu | -0.139 |
Token | blue |
Token ID | 4171 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ashes | +0.126 |
clinic | +0.118 |
ro | +0.116 |
Gu | +0.114 |
rimp | +0.109 |
Neg logit contributions | |
journal | -0.134 |
shield | -0.134 |
journals | -0.130 |
luck | -0.128 |
pencil | -0.127 |
Token | birds |
Token ID | 32002 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
ashes | +0.051 |
Gu | +0.047 |
ro | +0.047 |
rimp | +0.046 |
clinic | +0.046 |
Neg logit contributions | |
shield | -0.051 |
journal | -0.051 |
pencil | -0.050 |
journals | -0.049 |
magnets | -0.049 |
Token | lived |
Token ID | 5615 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
ashes | +0.096 |
clinic | +0.091 |
ro | +0.089 |
bank | +0.086 |
ashing | +0.086 |
Neg logit contributions | |
journals | -0.062 |
journal | -0.061 |
shield | -0.061 |
luck | -0.060 |
pencil | -0.058 |
Token | in |
Token ID | 287 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
shield | +0.014 |
luck | +0.014 |
journal | +0.014 |
pencil | +0.013 |
magnets | +0.013 |
Neg logit contributions | |
ashes | -0.043 |
clinic | -0.041 |
ro | -0.041 |
Gu | -0.040 |
rimp | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
shield | +0.014 |
journal | +0.013 |
journals | +0.013 |
pencil | +0.013 |
luck | +0.013 |
Neg logit contributions | |
ashes | -0.033 |
clinic | -0.032 |
ro | -0.032 |
Gu | -0.031 |
rimp | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
ashes | +0.051 |
clinic | +0.048 |
ro | +0.047 |
rimp | +0.045 |
ashing | +0.045 |
Neg logit contributions | |
shield | -0.032 |
journals | -0.031 |
journal | -0.031 |
pencil | -0.031 |
crystal | -0.030 |
Token | or |
Token ID | 393 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
ashes | +0.006 |
ro | +0.006 |
clinic | +0.006 |
rimp | +0.005 |
ashing | +0.005 |
Neg logit contributions | |
shield | -0.005 |
journal | -0.005 |
journals | -0.005 |
pencil | -0.005 |
luck | -0.005 |
Token | something |
Token ID | 1223 |
Feature activation | +9.41e-01 |
Pos logit contributions | |
ashes | +0.107 |
clinic | +0.104 |
ro | +0.101 |
rimp | +0.093 |
Gu | +0.091 |
Neg logit contributions | |
journals | -0.092 |
luck | -0.091 |
hetically | -0.091 |
journal | -0.089 |
narrative | -0.089 |
Token | else |
Token ID | 2073 |
Feature activation | +8.31e-01 |
Pos logit contributions | |
ashes | +0.054 |
clinic | +0.049 |
ro | +0.046 |
ashing | +0.046 |
bank | +0.046 |
Neg logit contributions | |
journals | -0.039 |
journal | -0.038 |
pencil | -0.038 |
shield | -0.038 |
crystal | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
ashes | +0.049 |
clinic | +0.045 |
ro | +0.044 |
ashing | +0.043 |
bank | +0.043 |
Neg logit contributions | |
shield | -0.033 |
journals | -0.032 |
journal | -0.032 |
pencil | -0.032 |
crystal | -0.031 |
Token | It |
Token ID | 632 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ashes | +0.010 |
ashing | +0.009 |
cafe | +0.009 |
oodle | +0.009 |
round | +0.009 |
Neg logit contributions | |
narrative | -0.009 |
magnets | -0.009 |
journals | -0.009 |
pencil | -0.009 |
journal | -0.009 |
Token | can |
Token ID | 460 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ashes | +0.079 |
clinic | +0.075 |
ro | +0.072 |
Gu | +0.072 |
bank | +0.070 |
Neg logit contributions | |
journal | -0.051 |
shield | -0.051 |
pencil | -0.050 |
journals | -0.050 |
luck | -0.049 |
Token | make |
Token ID | 787 |
Feature activation | +8.16e-01 |
Pos logit contributions | |
journals | +0.068 |
journal | +0.066 |
luck | +0.063 |
pencil | +0.063 |
shield | +0.062 |
Neg logit contributions | |
ashes | -0.073 |
clinic | -0.072 |
ro | -0.070 |
bank | -0.068 |
rimp | -0.066 |
Token | a |
Token ID | 257 |
Feature activation | -3.95e-01 |
Pos logit contributions | |
ashes | +0.037 |
clinic | +0.035 |
ro | +0.033 |
rimp | +0.032 |
Gu | +0.032 |
Neg logit contributions | |
pencil | -0.042 |
shield | -0.041 |
journal | -0.041 |
journals | -0.041 |
crystal | -0.039 |
Token | lot |
Token ID | 1256 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
shield | +0.016 |
journals | +0.016 |
pencil | +0.016 |
journal | +0.016 |
luck | +0.015 |
Neg logit contributions | |
ashes | -0.023 |
clinic | -0.022 |
ro | -0.021 |
cafe | -0.020 |
Gu | -0.020 |
Token | of |
Token ID | 286 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
shield | +0.002 |
luck | +0.002 |
journal | +0.002 |
pencil | +0.002 |
magnets | +0.002 |
Neg logit contributions | |
ashes | -0.004 |
clinic | -0.003 |
Gu | -0.003 |
ro | -0.003 |
anna | -0.003 |
Token | your |
Token ID | 534 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ashes | +0.047 |
clinic | +0.044 |
ro | +0.042 |
Gu | +0.042 |
cafe | +0.042 |
Neg logit contributions | |
shield | -0.031 |
journal | -0.030 |
pencil | -0.030 |
journals | -0.030 |
thumbs | -0.029 |
Token | flag |
Token ID | 6056 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
ashes | +0.159 |
clinic | +0.153 |
cafe | +0.150 |
ro | +0.147 |
aquarium | +0.146 |
Neg logit contributions | |
narrative | -0.101 |
streaming | -0.100 |
journals | -0.097 |
hetically | -0.096 |
magnets | -0.093 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
journals | +0.063 |
shield | +0.062 |
journal | +0.061 |
pencil | +0.061 |
luck | +0.059 |
Neg logit contributions | |
ashes | -0.080 |
clinic | -0.076 |
ro | -0.076 |
rimp | -0.072 |
cafe | -0.072 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
ashes | +0.228 |
ashing | +0.220 |
clinic | +0.218 |
anna | +0.217 |
rimp | +0.211 |
Neg logit contributions | |
pencil | -0.157 |
shield | -0.156 |
his | -0.151 |
journal | -0.146 |
thumbs | -0.142 |
Token | said |
Token ID | 531 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ashes | +0.265 |
clinic | +0.256 |
ashing | +0.242 |
ro | +0.241 |
Gu | +0.238 |
Neg logit contributions | |
pencil | -0.124 |
shield | -0.113 |
thumbs | -0.111 |
journal | -0.106 |
magnets | -0.102 |
Token | . |
Token ID | 13 |
Feature activation | +4.97e-01 |
Pos logit contributions | |
ashes | +0.060 |
ro | +0.056 |
clinic | +0.055 |
ashing | +0.053 |
anna | +0.053 |
Neg logit contributions | |
journals | -0.048 |
shield | -0.047 |
crystal | -0.047 |
journal | -0.046 |
narrative | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
ashes | +0.026 |
clinic | +0.024 |
ro | +0.023 |
rimp | +0.023 |
anna | +0.022 |
Neg logit contributions | |
shield | -0.023 |
pencil | -0.023 |
journal | -0.023 |
journals | -0.022 |
magnets | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | +2.17e-01 |
Pos logit contributions | |
ashes | +0.012 |
clinic | +0.012 |
ro | +0.012 |
rimp | +0.011 |
Gu | +0.011 |
Neg logit contributions | |
shield | -0.012 |
journal | -0.012 |
pencil | -0.012 |
luck | -0.012 |
journals | -0.012 |
Token | The |
Token ID | 464 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
ashes | +0.011 |
clinic | +0.010 |
ro | +0.010 |
anna | +0.010 |
ashing | +0.010 |
Neg logit contributions | |
shield | -0.010 |
pencil | -0.010 |
journal | -0.010 |
journals | -0.010 |
magnets | -0.010 |
Token | man |
Token ID | 582 |
Feature activation | +7.10e-01 |
Pos logit contributions | |
ashes | +0.017 |
clinic | +0.017 |
cafe | +0.016 |
aquarium | +0.016 |
ro | +0.016 |
Neg logit contributions | |
streaming | -0.030 |
hetically | -0.028 |
journals | -0.028 |
thumbs | -0.027 |
pencil | -0.027 |
Token | laughed |
Token ID | 13818 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
ro | +0.041 |
ashes | +0.041 |
clinic | +0.040 |
cafe | +0.040 |
Gu | +0.039 |
Neg logit contributions | |
journals | -0.026 |
shield | -0.026 |
journal | -0.025 |
luck | -0.025 |
magnets | -0.025 |
Token | Eventually |
Token ID | 28724 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
luck | +0.035 |
narrative | +0.035 |
journal | +0.035 |
pencil | +0.035 |
shield | +0.035 |
Neg logit contributions | |
ashes | -0.039 |
ro | -0.035 |
ashing | -0.035 |
clinic | -0.034 |
asher | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
ashes | +0.092 |
clinic | +0.089 |
ro | +0.089 |
cafe | +0.085 |
Gu | +0.085 |
Neg logit contributions | |
shield | -0.048 |
journal | -0.046 |
journals | -0.046 |
pencil | -0.045 |
magnets | -0.044 |
Token | Tab |
Token ID | 16904 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
shield | +0.034 |
journal | +0.034 |
journals | +0.033 |
pencil | +0.033 |
luck | +0.033 |
Neg logit contributions | |
ashes | -0.027 |
clinic | -0.025 |
ro | -0.025 |
Gu | -0.024 |
rimp | -0.023 |
Token | by |
Token ID | 1525 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
shield | +0.036 |
journal | +0.033 |
flashlight | +0.033 |
pencil | +0.033 |
magnets | +0.032 |
Neg logit contributions | |
ashes | -0.079 |
clinic | -0.077 |
ro | -0.076 |
anna | -0.075 |
Gu | -0.073 |
Token | knew |
Token ID | 2993 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
shield | +0.128 |
journal | +0.127 |
journals | +0.125 |
pencil | +0.125 |
luck | +0.122 |
Neg logit contributions | |
ashes | -0.200 |
clinic | -0.193 |
ro | -0.191 |
Gu | -0.182 |
rimp | -0.181 |
Token | she |
Token ID | 673 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
journals | +0.195 |
magnets | +0.187 |
journal | +0.187 |
shield | +0.183 |
pencil | +0.182 |
Neg logit contributions | |
clinic | -0.233 |
ro | -0.225 |
ashes | -0.214 |
anna | -0.212 |
cafe | -0.212 |
Token | had |
Token ID | 550 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
clinic | +0.003 |
ro | +0.003 |
anna | +0.003 |
ashes | +0.003 |
nes | +0.003 |
Neg logit contributions | |
magnets | -0.002 |
journals | -0.002 |
shield | -0.002 |
journal | -0.002 |
telescope | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | +8.99e-02 |
Pos logit contributions | |
shield | +0.123 |
journal | +0.120 |
journals | +0.119 |
pencil | +0.117 |
crystal | +0.117 |
Neg logit contributions | |
ashes | -0.142 |
clinic | -0.138 |
ro | -0.136 |
Gu | -0.135 |
cafe | -0.129 |
Token | face |
Token ID | 1986 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
ashes | +0.005 |
clinic | +0.005 |
ro | +0.005 |
rimp | +0.005 |
cafe | +0.005 |
Neg logit contributions | |
shield | -0.004 |
journal | -0.004 |
journals | -0.004 |
pencil | -0.004 |
magnets | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -3.46e-01 |
Pos logit contributions | |
shield | +0.033 |
journals | +0.032 |
narrative | +0.032 |
journal | +0.032 |
magnets | +0.031 |
Neg logit contributions | |
clinic | -0.058 |
ashes | -0.058 |
ro | -0.058 |
cafe | -0.055 |
anna | -0.054 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
pencil | +0.016 |
journals | +0.016 |
thumbs | +0.016 |
bicycles | +0.015 |
streaming | +0.015 |
Neg logit contributions | |
clinic | -0.016 |
ashes | -0.015 |
cafe | -0.014 |
ro | -0.014 |
aquarium | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
narrative | +0.073 |
shield | +0.073 |
journal | +0.067 |
journals | +0.067 |
luck | +0.066 |
Neg logit contributions | |
cafe | -0.108 |
clinic | -0.107 |
ashes | -0.106 |
ro | -0.104 |
rimp | -0.103 |
Token | tip |
Token ID | 8171 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
journals | +0.006 |
shield | +0.006 |
magnets | +0.006 |
narrative | +0.006 |
streaming | +0.005 |
Neg logit contributions | |
ashes | -0.007 |
clinic | -0.007 |
ro | -0.007 |
cafe | -0.007 |
Gu | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -6.17e-01 |
Pos logit contributions | |
shield | +0.091 |
journals | +0.091 |
narrative | +0.088 |
journal | +0.086 |
magnets | +0.085 |
Neg logit contributions | |
ro | -0.124 |
ashes | -0.123 |
clinic | -0.116 |
cafe | -0.111 |
bank | -0.111 |
Token | Together |
Token ID | 17083 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
journal | +0.028 |
pencil | +0.027 |
journals | +0.027 |
thumbs | +0.027 |
crystal | +0.026 |
Neg logit contributions | |
ashes | -0.033 |
rimp | -0.030 |
clinic | -0.030 |
ashing | -0.029 |
ro | -0.029 |
Token | they |
Token ID | 484 |
Feature activation | +3.50e-01 |
Pos logit contributions | |
cafe | +0.066 |
clinic | +0.066 |
ashes | +0.065 |
anna | +0.063 |
ro | +0.062 |
Neg logit contributions | |
shield | -0.068 |
journal | -0.064 |
spear | -0.064 |
sword | -0.062 |
journals | -0.062 |
Token | practiced |
Token ID | 19893 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
ashes | +0.015 |
ro | +0.015 |
nes | +0.014 |
round | +0.013 |
invited | +0.013 |
Neg logit contributions | |
narrative | -0.018 |
forgiveness | -0.017 |
journal | -0.017 |
luck | -0.016 |
mercy | -0.016 |
Token | counting |
Token ID | 14143 |
Feature activation | -2.97e-01 |
Pos logit contributions | |
shield | +0.014 |
journal | +0.013 |
narrative | +0.013 |
journals | +0.013 |
crystal | +0.013 |
Neg logit contributions | |
ro | -0.019 |
ashes | -0.019 |
clinic | -0.018 |
bank | -0.017 |
cafe | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
narrative | +0.014 |
shield | +0.014 |
journal | +0.013 |
journals | +0.013 |
crystal | +0.013 |
Neg logit contributions | |
ro | -0.015 |
ashes | -0.014 |
clinic | -0.013 |
cafe | -0.013 |
bank | -0.012 |
Token | Amelia |
Token ID | 49769 |
Feature activation | -9.61e-01 |
Pos logit contributions | |
shield | +0.032 |
journal | +0.031 |
luck | +0.030 |
journals | +0.030 |
narrative | +0.030 |
Neg logit contributions | |
ro | -0.031 |
ashes | -0.030 |
bank | -0.028 |
clinic | -0.028 |
Gu | -0.027 |
Token | was |
Token ID | 373 |
Feature activation | -2.80e-01 |
Pos logit contributions | |
shield | +0.038 |
journal | +0.037 |
journals | +0.037 |
luck | +0.036 |
flashlight | +0.035 |
Neg logit contributions | |
ashes | -0.055 |
ro | -0.053 |
clinic | -0.053 |
anna | -0.051 |
bank | -0.050 |
Token | very |
Token ID | 845 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
shield | +0.012 |
journal | +0.012 |
journals | +0.012 |
pencil | +0.011 |
luck | +0.011 |
Neg logit contributions | |
ashes | -0.015 |
clinic | -0.015 |
ro | -0.014 |
Gu | -0.014 |
rimp | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
shield | +0.012 |
luck | +0.011 |
crystal | +0.011 |
pencil | +0.011 |
journal | +0.010 |
Neg logit contributions | |
ashes | -0.041 |
clinic | -0.039 |
rimp | -0.039 |
Gu | -0.037 |
ro | -0.037 |
Token | time |
Token ID | 640 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
shield | +0.015 |
luck | +0.015 |
journal | +0.014 |
crystal | +0.014 |
pencil | +0.014 |
Neg logit contributions | |
ashes | -0.022 |
clinic | -0.021 |
ro | -0.021 |
Gu | -0.020 |
rimp | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +5.92e-03 |
Pos logit contributions | |
ashes | +0.138 |
ro | +0.128 |
clinic | +0.127 |
ashing | +0.125 |
rimp | +0.122 |
Neg logit contributions | |
shield | -0.100 |
journals | -0.096 |
journal | -0.095 |
pencil | -0.094 |
magnets | -0.092 |
Token | go |
Token ID | 467 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
ashes | +0.000 |
clinic | +0.000 |
ro | +0.000 |
Gu | +0.000 |
rimp | +0.000 |
Neg logit contributions | |
shield | -0.000 |
luck | -0.000 |
journal | -0.000 |
pencil | -0.000 |
journals | -0.000 |
Token | home |
Token ID | 1363 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
shield | +0.014 |
luck | +0.013 |
journal | +0.013 |
pencil | +0.013 |
crystal | +0.012 |
Neg logit contributions | |
ashes | -0.020 |
ro | -0.019 |
Gu | -0.019 |
clinic | -0.019 |
rimp | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -9.02e-01 |
Pos logit contributions | |
ashes | +0.095 |
clinic | +0.090 |
ro | +0.090 |
Gu | +0.089 |
rimp | +0.087 |
Neg logit contributions | |
shield | -0.038 |
luck | -0.037 |
journal | -0.037 |
pencil | -0.036 |
journals | -0.035 |
Token | She |
Token ID | 1375 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
journal | +0.042 |
pencil | +0.041 |
thumbs | +0.040 |
journals | +0.039 |
magnets | +0.037 |
Neg logit contributions | |
ashes | -0.047 |
rim | -0.044 |
rimp | -0.044 |
clinic | -0.044 |
ro | -0.043 |
Token | said |
Token ID | 531 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
ashes | +0.001 |
clinic | +0.001 |
ro | +0.001 |
Gu | +0.001 |
rimp | +0.001 |
Neg logit contributions | |
shield | -0.001 |
pencil | -0.001 |
journal | -0.001 |
luck | -0.001 |
journals | -0.001 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ashes | +0.017 |
ro | +0.017 |
anna | +0.016 |
clinic | +0.016 |
oodle | +0.016 |
Neg logit contributions | |
journals | -0.014 |
shield | -0.014 |
magnets | -0.014 |
flashlight | -0.014 |
crystal | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -7.62e-03 |
Pos logit contributions | |
shield | +0.041 |
luck | +0.040 |
pencil | +0.039 |
journal | +0.038 |
thumbs | +0.036 |
Neg logit contributions | |
ashes | -0.088 |
clinic | -0.084 |
Gu | -0.083 |
rimp | -0.082 |
ro | -0.081 |
Token | her |
Token ID | 607 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
narrative | +0.000 |
luck | +0.000 |
journals | +0.000 |
journal | +0.000 |
shield | +0.000 |
Neg logit contributions | |
clinic | -0.000 |
cafe | -0.000 |
ashes | -0.000 |
anna | -0.000 |
Mara | -0.000 |
Token | ummy |
Token ID | 13513 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
shield | +0.166 |
journal | +0.162 |
luck | +0.161 |
pencil | +0.160 |
journals | +0.159 |
Neg logit contributions | |
ashes | -0.183 |
clinic | -0.174 |
ro | -0.171 |
Gu | -0.166 |
rimp | -0.163 |
Token | and |
Token ID | 290 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
shield | +0.059 |
journal | +0.058 |
luck | +0.058 |
pencil | +0.057 |
journals | +0.056 |
Neg logit contributions | |
ashes | -0.078 |
clinic | -0.074 |
ro | -0.074 |
Gu | -0.072 |
rimp | -0.071 |
Token | crunch |
Token ID | 23527 |
Feature activation | -3.18e+00 |
Pos logit contributions | |
ashes | +0.002 |
ro | +0.002 |
clinic | +0.002 |
Gu | +0.002 |
cafe | +0.002 |
Neg logit contributions | |
shield | -0.002 |
journal | -0.002 |
journals | -0.002 |
flashlight | -0.002 |
luck | -0.002 |
Token | y |
Token ID | 88 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
shield | +0.168 |
journal | +0.162 |
journals | +0.161 |
pencil | +0.159 |
luck | +0.157 |
Neg logit contributions | |
ashes | -0.150 |
ro | -0.140 |
clinic | -0.132 |
rimp | -0.131 |
round | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | +3.35e-01 |
Pos logit contributions | |
ashes | +0.066 |
clinic | +0.062 |
ro | +0.062 |
Gu | +0.060 |
rimp | +0.059 |
Neg logit contributions | |
shield | -0.054 |
journal | -0.052 |
pencil | -0.052 |
luck | -0.051 |
journals | -0.051 |
Token | Don |
Token ID | 2094 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
ashes | +0.017 |
ashing | +0.016 |
nes | +0.016 |
bank | +0.015 |
ro | +0.015 |
Neg logit contributions | |
pencil | -0.015 |
shield | -0.015 |
journals | -0.015 |
journal | -0.015 |
magnets | -0.015 |
Token | 't |
Token ID | 470 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
shield | +0.034 |
pencil | +0.033 |
journal | +0.032 |
luck | +0.032 |
thumbs | +0.032 |
Neg logit contributions | |
ashes | -0.030 |
clinic | -0.027 |
ro | -0.027 |
ashing | -0.026 |
rimp | -0.026 |
Token | be |
Token ID | 307 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
ashes | +0.069 |
ro | +0.064 |
clinic | +0.063 |
rimp | +0.063 |
Gu | +0.061 |
Neg logit contributions | |
shield | -0.063 |
journals | -0.063 |
journal | -0.062 |
luck | -0.060 |
pencil | -0.059 |
Token | afraid |
Token ID | 7787 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ashes | +0.105 |
ro | +0.104 |
clinic | +0.104 |
nes | +0.100 |
cafe | +0.099 |
Neg logit contributions | |
pencil | -0.068 |
luck | -0.067 |
shield | -0.066 |
magnets | -0.063 |
hard | -0.063 |
Token | ." |
Token ID | 526 |
Feature activation | +3.88e+00 |
Pos logit contributions | |
ashes | +0.079 |
clinic | +0.073 |
ro | +0.073 |
rimp | +0.070 |
anna | +0.068 |
Neg logit contributions | |
shield | -0.075 |
journal | -0.073 |
journals | -0.073 |
pencil | -0.072 |
luck | -0.071 |
Token |
|
Token ID | 198 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
ashes | +0.215 |
clinic | +0.202 |
ro | +0.200 |
rimp | +0.195 |
ashing | +0.193 |
Neg logit contributions | |
shield | -0.169 |
pencil | -0.165 |
journal | -0.164 |
luck | -0.160 |
journals | -0.159 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
shield | +0.026 |
journal | +0.025 |
journals | +0.025 |
luck | +0.025 |
pencil | +0.024 |
Neg logit contributions | |
ashes | -0.031 |
clinic | -0.030 |
ro | -0.029 |
cafe | -0.028 |
anna | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
ashes | +0.206 |
clinic | +0.196 |
ro | +0.194 |
rimp | +0.184 |
ashing | +0.183 |
Neg logit contributions | |
shield | -0.131 |
journal | -0.126 |
journals | -0.125 |
luck | -0.123 |
pencil | -0.123 |
Token | said |
Token ID | 531 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
ashes | +0.003 |
ro | +0.003 |
clinic | +0.003 |
bank | +0.003 |
cafe | +0.003 |
Neg logit contributions | |
journal | -0.003 |
journals | -0.003 |
shield | -0.003 |
luck | -0.003 |
crystal | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +3.23e-03 |
Pos logit contributions | |
ashes | +0.035 |
ro | +0.033 |
clinic | +0.033 |
oodle | +0.032 |
rimp | +0.032 |
Neg logit contributions | |
journals | -0.031 |
shield | -0.031 |
journal | -0.029 |
flashlight | -0.029 |
crystal | -0.028 |
Token | " |
Token ID | 366 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
ashes | +0.000 |
clinic | +0.000 |
ro | +0.000 |
rimp | +0.000 |
Gu | +0.000 |
Neg logit contributions | |
shield | -0.000 |
pencil | -0.000 |
journal | -0.000 |
luck | -0.000 |
thumbs | -0.000 |
Token | You |
Token ID | 1639 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ashes | +0.011 |
clinic | +0.011 |
aquarium | +0.011 |
ashing | +0.010 |
rimp | +0.010 |
Neg logit contributions | |
pencil | -0.009 |
journals | -0.008 |
narrative | -0.008 |
shield | -0.008 |
journal | -0.008 |
Token | copied |
Token ID | 18984 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
shield | +0.067 |
journals | +0.065 |
journal | +0.064 |
luck | +0.060 |
crystal | +0.058 |
Neg logit contributions | |
ashes | -0.143 |
ro | -0.133 |
clinic | -0.132 |
Gu | -0.131 |
rimp | -0.129 |
Token | me |
Token ID | 502 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
shield | +0.030 |
journals | +0.029 |
journal | +0.029 |
crystal | +0.028 |
luck | +0.028 |
Neg logit contributions | |
ashes | -0.029 |
ro | -0.027 |
clinic | -0.027 |
rimp | -0.026 |
oodle | -0.026 |
Token | ! |
Token ID | 0 |
Feature activation | +6.82e-02 |
Pos logit contributions | |
ashes | +0.191 |
clinic | +0.181 |
ro | +0.181 |
Gu | +0.175 |
rimp | +0.174 |
Neg logit contributions | |
shield | -0.097 |
journal | -0.093 |
pencil | -0.091 |
luck | -0.091 |
journals | -0.090 |
Token | You |
Token ID | 921 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
ashes | +0.003 |
clinic | +0.003 |
bank | +0.003 |
cafe | +0.003 |
ashing | +0.003 |
Neg logit contributions | |
pencil | -0.003 |
journal | -0.003 |
thumbs | -0.003 |
shield | -0.003 |
magnets | -0.003 |
Token | are |
Token ID | 389 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
ashes | +0.027 |
ro | +0.025 |
clinic | +0.025 |
rimp | +0.024 |
Gu | +0.024 |
Neg logit contributions | |
journals | -0.013 |
shield | -0.013 |
journal | -0.013 |
luck | -0.012 |
crystal | -0.012 |
Token | they |
Token ID | 484 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
pencil | +0.003 |
luck | +0.003 |
shield | +0.003 |
thumbs | +0.003 |
all | +0.003 |
Neg logit contributions | |
ashes | -0.006 |
clinic | -0.006 |
ro | -0.006 |
bank | -0.006 |
anna | -0.006 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
ashes | +0.062 |
ro | +0.058 |
ashing | +0.057 |
Gu | +0.057 |
rimp | +0.056 |
Neg logit contributions | |
shield | -0.052 |
luck | -0.050 |
journal | -0.050 |
journals | -0.049 |
pencil | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | +4.36e-01 |
Pos logit contributions | |
ashes | +0.061 |
ro | +0.058 |
clinic | +0.058 |
Gu | +0.056 |
bank | +0.055 |
Neg logit contributions | |
shield | -0.046 |
journal | -0.044 |
journals | -0.044 |
pencil | -0.044 |
luck | -0.043 |
Token | make |
Token ID | 787 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
ashes | +0.026 |
clinic | +0.025 |
ro | +0.024 |
ashing | +0.024 |
Gu | +0.024 |
Neg logit contributions | |
shield | -0.017 |
pencil | -0.017 |
journal | -0.016 |
journals | -0.016 |
thumbs | -0.016 |
Token | a |
Token ID | 257 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
ashes | +0.060 |
ro | +0.059 |
clinic | +0.057 |
round | +0.056 |
anna | +0.056 |
Neg logit contributions | |
shield | -0.024 |
pencil | -0.023 |
luck | -0.023 |
journal | -0.023 |
magnets | -0.022 |
Token | salad |
Token ID | 20698 |
Feature activation | -4.74e+00 |
Pos logit contributions | |
shield | +0.017 |
journals | +0.017 |
luck | +0.017 |
pencil | +0.016 |
thumbs | +0.016 |
Neg logit contributions | |
ashes | -0.026 |
clinic | -0.025 |
cafe | -0.024 |
ro | -0.023 |
Gu | -0.023 |
Token | for |
Token ID | 329 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
shield | +0.166 |
journal | +0.160 |
journals | +0.158 |
pencil | +0.158 |
luck | +0.157 |
Neg logit contributions | |
ashes | -0.300 |
clinic | -0.283 |
ro | -0.283 |
rimp | -0.272 |
Gu | -0.270 |
Token | lunch |
Token ID | 9965 |
Feature activation | -7.75e-01 |
Pos logit contributions | |
shield | +0.080 |
journal | +0.077 |
pencil | +0.076 |
luck | +0.076 |
journals | +0.075 |
Neg logit contributions | |
ashes | -0.133 |
clinic | -0.127 |
ro | -0.126 |
rimp | -0.121 |
Gu | -0.121 |
Token | . |
Token ID | 13 |
Feature activation | -2.73e-01 |
Pos logit contributions | |
shield | +0.022 |
journal | +0.021 |
pencil | +0.021 |
luck | +0.021 |
journals | +0.020 |
Neg logit contributions | |
ashes | -0.054 |
clinic | -0.052 |
ro | -0.051 |
Gu | -0.051 |
anna | -0.049 |
Token | They |
Token ID | 1119 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
all | +0.009 |
his | +0.009 |
were | +0.008 |
thumbs | +0.008 |
Mom | +0.008 |
Neg logit contributions | |
rim | -0.019 |
ider | -0.017 |
oster | -0.017 |
eh | -0.017 |
ital | -0.017 |
Token | went |
Token ID | 1816 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
ashes | +0.106 |
ro | +0.100 |
rimp | +0.097 |
clinic | +0.095 |
cafe | +0.094 |
Neg logit contributions | |
luck | -0.104 |
shield | -0.102 |
journal | -0.101 |
journals | -0.099 |
narrative | -0.096 |
Token | , |
Token ID | 11 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
Gu | +0.018 |
ashes | +0.018 |
ro | +0.018 |
mare | +0.018 |
clinic | +0.018 |
Neg logit contributions | |
narrative | -0.018 |
journals | -0.018 |
shield | -0.017 |
flashlight | -0.017 |
magnets | -0.017 |
Token | " |
Token ID | 366 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
shield | +0.021 |
pencil | +0.020 |
journal | +0.020 |
luck | +0.019 |
thumbs | +0.019 |
Neg logit contributions | |
ashes | -0.026 |
clinic | -0.024 |
ro | -0.023 |
rimp | -0.022 |
Gu | -0.022 |
Token | Thank |
Token ID | 10449 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
pencil | +0.001 |
shield | +0.001 |
journals | +0.001 |
journal | +0.001 |
crystal | +0.001 |
Neg logit contributions | |
ashes | -0.001 |
clinic | -0.001 |
ro | -0.001 |
bank | -0.001 |
ashing | -0.001 |
Token | you |
Token ID | 345 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
shield | +0.188 |
pencil | +0.185 |
journal | +0.185 |
luck | +0.183 |
journals | +0.181 |
Neg logit contributions | |
ashes | -0.204 |
clinic | -0.191 |
ro | -0.189 |
ashing | -0.178 |
bank | -0.178 |
Token | ! |
Token ID | 0 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
ashes | +0.031 |
clinic | +0.030 |
ro | +0.029 |
rimp | +0.029 |
cafe | +0.029 |
Neg logit contributions | |
shield | -0.017 |
journals | -0.016 |
journal | -0.016 |
pencil | -0.016 |
luck | -0.016 |
Token | Daddy |
Token ID | 26926 |
Feature activation | -5.30e+00 |
Pos logit contributions | |
pencil | +0.024 |
thumbs | +0.023 |
journal | +0.023 |
crystal | +0.022 |
magnets | +0.022 |
Neg logit contributions | |
ashes | -0.032 |
clinic | -0.029 |
bank | -0.028 |
cafe | -0.027 |
ashing | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | -5.34e-01 |
Pos logit contributions | |
shield | +0.201 |
pencil | +0.195 |
luck | +0.194 |
journal | +0.194 |
crystal | +0.187 |
Neg logit contributions | |
ashes | -0.322 |
clinic | -0.304 |
ro | -0.302 |
Gu | -0.291 |
rimp | -0.289 |
Token | you |
Token ID | 345 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
journals | +0.025 |
shield | +0.024 |
narrative | +0.023 |
hetically | +0.023 |
magnets | +0.023 |
Neg logit contributions | |
Gu | -0.023 |
mare | -0.023 |
ashes | -0.023 |
clinic | -0.023 |
ro | -0.023 |
Token | are |
Token ID | 389 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ashes | +0.032 |
ro | +0.030 |
clinic | +0.029 |
Gu | +0.029 |
rimp | +0.029 |
Neg logit contributions | |
shield | -0.018 |
journals | -0.017 |
journal | -0.017 |
magnets | -0.016 |
pencil | -0.016 |
Token | gifted |
Token ID | 22527 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
luck | +0.110 |
shield | +0.107 |
crystal | +0.104 |
journal | +0.104 |
pencil | +0.103 |
Neg logit contributions | |
ashes | -0.168 |
ro | -0.158 |
clinic | -0.153 |
ashing | -0.147 |
bank | -0.145 |
Token | at |
Token ID | 379 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
ashes | +0.078 |
clinic | +0.075 |
ro | +0.072 |
Gu | +0.071 |
rimp | +0.071 |
Neg logit contributions | |
shield | -0.055 |
luck | -0.054 |
journal | -0.053 |
pencil | -0.053 |
journals | -0.051 |
Token | of |
Token ID | 286 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
shield | +0.082 |
journal | +0.080 |
journals | +0.080 |
luck | +0.079 |
pencil | +0.079 |
Neg logit contributions | |
ashes | -0.094 |
ro | -0.088 |
clinic | -0.088 |
rimp | -0.084 |
Gu | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
clinic | +0.158 |
ashes | +0.155 |
cafe | +0.152 |
rimp | +0.150 |
Gu | +0.149 |
Neg logit contributions | |
journals | -0.091 |
journal | -0.091 |
narrative | -0.091 |
shield | -0.090 |
pencil | -0.088 |
Token | boat |
Token ID | 8848 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
ashes | +0.001 |
clinic | +0.001 |
ro | +0.001 |
cafe | +0.001 |
Gu | +0.001 |
Neg logit contributions | |
shield | -0.001 |
journals | -0.001 |
journal | -0.001 |
pencil | -0.001 |
luck | -0.001 |
Token | with |
Token ID | 351 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
ashes | +0.126 |
clinic | +0.118 |
ro | +0.117 |
Gu | +0.116 |
rimp | +0.114 |
Neg logit contributions | |
shield | -0.070 |
journal | -0.067 |
pencil | -0.066 |
luck | -0.065 |
journals | -0.064 |
Token | his |
Token ID | 465 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ashes | +0.078 |
ro | +0.077 |
cafe | +0.074 |
aquarium | +0.073 |
clinic | +0.072 |
Neg logit contributions | |
narrative | -0.063 |
journal | -0.057 |
journals | -0.056 |
hetically | -0.055 |
shield | -0.054 |
Token | cr |
Token ID | 1067 |
Feature activation | -5.73e+00 |
Pos logit contributions | |
journals | +0.050 |
narrative | +0.050 |
shield | +0.047 |
journal | +0.047 |
luck | +0.046 |
Neg logit contributions | |
ashes | -0.079 |
clinic | -0.078 |
ro | -0.076 |
cafe | -0.072 |
oodle | -0.071 |
Token | ay |
Token ID | 323 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
shield | +0.203 |
journal | +0.196 |
pencil | +0.193 |
luck | +0.192 |
journals | +0.191 |
Neg logit contributions | |
ashes | -0.360 |
clinic | -0.341 |
ro | -0.340 |
Gu | -0.328 |
rimp | -0.327 |
Token | ons |
Token ID | 684 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
shield | +0.125 |
narrative | +0.124 |
luck | +0.119 |
journal | +0.119 |
journals | +0.118 |
Neg logit contributions | |
anna | -0.299 |
ashes | -0.297 |
rimp | -0.296 |
ro | -0.295 |
round | -0.291 |
Token | when |
Token ID | 618 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
journals | +0.128 |
shield | +0.127 |
journal | +0.126 |
luck | +0.122 |
pencil | +0.121 |
Neg logit contributions | |
ashes | -0.202 |
ro | -0.197 |
clinic | -0.193 |
rimp | -0.184 |
ashing | -0.184 |
Token | he |
Token ID | 339 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
clinic | +0.198 |
ashes | +0.196 |
cafe | +0.194 |
ro | +0.193 |
Gu | +0.192 |
Neg logit contributions | |
journals | -0.121 |
shield | -0.121 |
journal | -0.120 |
pencil | -0.120 |
thumbs | -0.118 |
Token | got |
Token ID | 1392 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
shield | +0.099 |
journal | +0.096 |
journals | +0.095 |
luck | +0.094 |
pencil | +0.094 |
Neg logit contributions | |
ashes | -0.152 |
ro | -0.144 |
clinic | -0.144 |
Gu | -0.138 |
rimp | -0.137 |
Token | Tim |
Token ID | 5045 |
Feature activation | -9.91e-01 |
Pos logit contributions | |
shield | +0.028 |
pencil | +0.027 |
journal | +0.027 |
magnets | +0.027 |
journals | +0.027 |
Neg logit contributions | |
ashes | -0.037 |
clinic | -0.035 |
ro | -0.034 |
Gu | -0.033 |
anna | -0.033 |
Token | my |
Token ID | 1820 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
luck | +0.042 |
shield | +0.041 |
journals | +0.039 |
bravery | +0.039 |
arrows | +0.038 |
Neg logit contributions | |
ashes | -0.050 |
ro | -0.049 |
Gu | -0.048 |
anna | -0.048 |
forest | -0.048 |
Token | loved |
Token ID | 6151 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
shield | +0.132 |
luck | +0.129 |
journal | +0.125 |
journals | +0.124 |
pencil | +0.123 |
Neg logit contributions | |
ashes | -0.217 |
ro | -0.208 |
clinic | -0.207 |
Gu | -0.199 |
rimp | -0.198 |
Token | to |
Token ID | 284 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
shield | +0.090 |
luck | +0.088 |
journal | +0.088 |
journals | +0.086 |
narrative | +0.085 |
Neg logit contributions | |
ashes | -0.119 |
clinic | -0.113 |
ro | -0.111 |
Gu | -0.111 |
cafe | -0.108 |
Token | read |
Token ID | 1100 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
ashes | +0.011 |
ro | +0.010 |
clinic | +0.010 |
rimp | +0.010 |
cafe | +0.010 |
Neg logit contributions | |
luck | -0.011 |
shield | -0.011 |
journals | -0.010 |
journal | -0.010 |
narrative | -0.010 |
Token | books |
Token ID | 3835 |
Feature activation | -5.19e+00 |
Pos logit contributions | |
narrative | +0.074 |
luck | +0.070 |
shield | +0.068 |
iness | +0.068 |
crystal | +0.068 |
Neg logit contributions | |
cafe | -0.119 |
clinic | -0.117 |
ro | -0.114 |
round | -0.113 |
café | -0.110 |
Token | and |
Token ID | 290 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
shield | +0.214 |
journals | +0.206 |
journal | +0.203 |
luck | +0.203 |
thumbs | +0.202 |
Neg logit contributions | |
ashes | -0.289 |
ro | -0.282 |
clinic | -0.278 |
round | -0.266 |
rimp | -0.264 |
Token | he |
Token ID | 339 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
luck | +0.029 |
streaming | +0.027 |
iness | +0.027 |
narrative | +0.027 |
thumbs | +0.026 |
Neg logit contributions | |
clinic | -0.026 |
ro | -0.025 |
cafe | -0.025 |
ashes | -0.024 |
bank | -0.023 |
Token | had |
Token ID | 550 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
luck | +0.110 |
shield | +0.109 |
thumbs | +0.108 |
magnets | +0.106 |
spear | +0.104 |
Neg logit contributions | |
clinic | -0.129 |
Gu | -0.122 |
ro | -0.122 |
anna | -0.122 |
orian | -0.116 |
Token | a |
Token ID | 257 |
Feature activation | -7.25e-01 |
Pos logit contributions | |
shield | +0.097 |
journal | +0.092 |
luck | +0.091 |
crystal | +0.090 |
flashlight | +0.090 |
Neg logit contributions | |
ashes | -0.147 |
clinic | -0.146 |
Gu | -0.145 |
cafe | -0.140 |
ro | -0.139 |
Token | big |
Token ID | 1263 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
iness | +0.028 |
streaming | +0.028 |
hetically | +0.027 |
narrative | +0.027 |
luck | +0.027 |
Neg logit contributions | |
clinic | -0.041 |
cafe | -0.038 |
ashes | -0.038 |
Gu | -0.037 |
oodle | -0.036 |
Token | pick |
Token ID | 2298 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ashes | +0.013 |
clinic | +0.012 |
ro | +0.012 |
Gu | +0.012 |
ashing | +0.011 |
Neg logit contributions | |
shield | -0.011 |
pencil | -0.011 |
luck | -0.011 |
journal | -0.011 |
journals | -0.011 |
Token | from |
Token ID | 422 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
shield | +0.045 |
journal | +0.044 |
journals | +0.042 |
pencil | +0.042 |
luck | +0.042 |
Neg logit contributions | |
ashes | -0.087 |
ro | -0.082 |
clinic | -0.082 |
Gu | -0.079 |
rimp | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
shield | +0.066 |
journal | +0.065 |
journals | +0.063 |
luck | +0.061 |
pencil | +0.060 |
Neg logit contributions | |
ashes | -0.116 |
clinic | -0.112 |
ro | -0.112 |
cafe | -0.107 |
rimp | -0.107 |
Token | it |
Token ID | 340 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
shield | +0.017 |
journal | +0.017 |
journals | +0.017 |
luck | +0.017 |
pencil | +0.017 |
Neg logit contributions | |
ashes | -0.019 |
ro | -0.018 |
clinic | -0.018 |
rimp | -0.017 |
Gu | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
shield | +0.007 |
journal | +0.007 |
journals | +0.007 |
luck | +0.007 |
pencil | +0.007 |
Neg logit contributions | |
ashes | -0.012 |
ro | -0.011 |
clinic | -0.011 |
bank | -0.011 |
cafe | -0.011 |
Token | hard |
Token ID | 1327 |
Feature activation | -5.31e+00 |
Pos logit contributions | |
shield | +0.005 |
journal | +0.005 |
journals | +0.005 |
pencil | +0.005 |
luck | +0.005 |
Neg logit contributions | |
ashes | -0.007 |
ro | -0.007 |
clinic | -0.007 |
rimp | -0.007 |
cafe | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
shield | +0.174 |
luck | +0.168 |
pencil | +0.167 |
journal | +0.166 |
journals | +0.162 |
Neg logit contributions | |
ashes | -0.347 |
clinic | -0.333 |
ro | -0.328 |
Gu | -0.324 |
rimp | -0.320 |
Token | decide |
Token ID | 5409 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ashes | +0.011 |
clinic | +0.010 |
ro | +0.010 |
rimp | +0.009 |
Gu | +0.009 |
Neg logit contributions | |
shield | -0.010 |
journal | -0.010 |
journals | -0.009 |
pencil | -0.009 |
luck | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
shield | +0.063 |
journal | +0.061 |
journals | +0.060 |
narrative | +0.060 |
luck | +0.059 |
Neg logit contributions | |
ashes | -0.068 |
clinic | -0.067 |
ro | -0.066 |
rimp | -0.064 |
round | -0.064 |
Token | Tim |
Token ID | 5045 |
Feature activation | -9.52e-01 |
Pos logit contributions | |
magnets | +0.016 |
thumbs | +0.015 |
pencil | +0.015 |
all | +0.014 |
magnet | +0.014 |
Neg logit contributions | |
clinic | -0.033 |
ashes | -0.033 |
nes | -0.032 |
ro | -0.032 |
anna | -0.031 |
Token | my |
Token ID | 1820 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
shield | +0.028 |
journal | +0.027 |
luck | +0.027 |
journals | +0.026 |
pencil | +0.026 |
Neg logit contributions | |
ashes | -0.065 |
ro | -0.062 |
clinic | -0.062 |
Gu | -0.060 |
rimp | -0.060 |