Token | sun |
Token ID | 4252 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
washing | +0.197 |
cloth | +0.182 |
sorting | +0.175 |
racks | +0.157 |
mixing | +0.156 |
Neg logit contributions | |
ossible | -0.272 |
former | -0.270 |
needed | -0.264 |
isdom | -0.264 |
earned | -0.256 |
Token | fl |
Token ID | 2704 |
Feature activation | +8.20e+00 |
Pos logit contributions | |
washing | +0.138 |
cloth | +0.127 |
sorting | +0.124 |
racks | +0.112 |
mixing | +0.110 |
Neg logit contributions | |
ossible | -0.144 |
bye | -0.140 |
isdom | -0.137 |
needed | -0.136 |
oof | -0.136 |
Token | owers |
Token ID | 3618 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
washing | +0.275 |
cloth | +0.251 |
mixing | +0.245 |
sorting | +0.244 |
cups | +0.238 |
Neg logit contributions | |
former | -0.212 |
needed | -0.208 |
earned | -0.207 |
des | -0.206 |
isdom | -0.205 |
Token | , |
Token ID | 11 |
Feature activation | +7.17e+00 |
Pos logit contributions | |
washing | +0.030 |
cloth | +0.028 |
sorting | +0.025 |
racks | +0.024 |
Mil | +0.023 |
Neg logit contributions | |
ossible | -0.047 |
needed | -0.046 |
moon | -0.046 |
isdom | -0.046 |
encounter | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
washing | +0.182 |
cloth | +0.168 |
sorting | +0.162 |
racks | +0.146 |
mixing | +0.145 |
Neg logit contributions | |
former | -0.233 |
ossible | -0.231 |
needed | -0.228 |
isdom | -0.227 |
bye | -0.221 |
Token | tul |
Token ID | 48373 |
Feature activation | +1.30e+01 |
Pos logit contributions | |
washing | +0.029 |
cloth | +0.027 |
sorting | +0.026 |
mixing | +0.024 |
dripping | +0.024 |
Neg logit contributions | |
former | -0.037 |
ossible | -0.036 |
needed | -0.036 |
isdom | -0.035 |
earned | -0.035 |
Token | ips |
Token ID | 2419 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
washing | +0.399 |
cloth | +0.366 |
sorting | +0.356 |
living | +0.332 |
dripping | +0.328 |
Neg logit contributions | |
ossible | -0.384 |
former | -0.379 |
isdom | -0.374 |
needed | -0.368 |
des | -0.360 |
Token | . |
Token ID | 13 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
washing | +0.066 |
cloth | +0.058 |
sorting | +0.054 |
mixing | +0.051 |
living | +0.048 |
Neg logit contributions | |
former | -0.132 |
ossible | -0.131 |
needed | -0.130 |
isdom | -0.130 |
earned | -0.129 |
Token | They |
Token ID | 1119 |
Feature activation | +3.46e+00 |
Pos logit contributions | |
washing | +0.078 |
cloth | +0.071 |
sorting | +0.069 |
mixing | +0.066 |
dripping | +0.065 |
Neg logit contributions | |
former | -0.077 |
ossible | -0.076 |
needed | -0.075 |
isdom | -0.074 |
earned | -0.073 |
Token | see |
Token ID | 766 |
Feature activation | +4.28e+00 |
Pos logit contributions | |
washing | +0.108 |
cloth | +0.098 |
sorting | +0.095 |
mixing | +0.091 |
dripping | +0.089 |
Neg logit contributions | |
ossible | -0.097 |
former | -0.097 |
needed | -0.094 |
isdom | -0.094 |
bye | -0.093 |
Token | pine |
Token ID | 20161 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
washing | +0.112 |
cloth | +0.103 |
sorting | +0.098 |
mixing | +0.091 |
dripping | +0.088 |
Neg logit contributions | |
former | -0.139 |
ossible | -0.137 |
needed | -0.136 |
isdom | -0.134 |
earned | -0.133 |
Token | and |
Token ID | 290 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
washing | +0.007 |
cloth | +0.006 |
sorting | +0.005 |
living | +0.005 |
mixing | +0.005 |
Neg logit contributions | |
former | -0.016 |
isdom | -0.016 |
ossible | -0.016 |
needed | -0.016 |
earned | -0.016 |
Token | hungry |
Token ID | 14720 |
Feature activation | +9.69e-01 |
Pos logit contributions | |
washing | +0.039 |
cloth | +0.035 |
sorting | +0.033 |
living | +0.032 |
dripping | +0.031 |
Neg logit contributions | |
isdom | -0.049 |
former | -0.049 |
earned | -0.049 |
needed | -0.048 |
ossible | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +3.37e+00 |
Pos logit contributions | |
washing | +0.018 |
cloth | +0.016 |
sorting | +0.015 |
mixing | +0.014 |
dripping | +0.013 |
Neg logit contributions | |
needed | -0.038 |
former | -0.038 |
ossible | -0.038 |
bye | -0.038 |
earned | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | +4.12e+00 |
Pos logit contributions | |
washing | +0.090 |
cloth | +0.081 |
sorting | +0.078 |
Mil | +0.075 |
living | +0.074 |
Neg logit contributions | |
needed | -0.112 |
former | -0.112 |
earned | -0.111 |
des | -0.110 |
isdom | -0.109 |
Token | zo |
Token ID | 40565 |
Feature activation | +7.07e+00 |
Pos logit contributions | |
washing | +0.118 |
cloth | +0.104 |
sorting | +0.100 |
living | +0.098 |
mixing | +0.096 |
Neg logit contributions | |
isdom | -0.131 |
earned | -0.131 |
needed | -0.130 |
des | -0.130 |
former | -0.129 |
Token | oke |
Token ID | 2088 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
washing | +0.155 |
cloth | +0.130 |
sorting | +0.123 |
mixing | +0.117 |
dripping | +0.113 |
Neg logit contributions | |
ossible | -0.272 |
bye | -0.269 |
needed | -0.266 |
isdom | -0.266 |
earned | -0.263 |
Token | ep |
Token ID | 538 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
washing | +0.339 |
sorting | +0.279 |
cloth | +0.277 |
mixing | +0.260 |
cups | +0.255 |
Neg logit contributions | |
ossible | -0.439 |
bye | -0.424 |
oof | -0.401 |
isdom | -0.401 |
earned | -0.399 |
Token | ers |
Token ID | 364 |
Feature activation | +6.39e+00 |
Pos logit contributions | |
washing | +0.037 |
cloth | +0.036 |
ers | +0.032 |
sorting | +0.031 |
living | +0.030 |
Neg logit contributions | |
former | -0.063 |
needed | -0.062 |
des | -0.061 |
earned | -0.061 |
isdom | -0.060 |
Token | catch |
Token ID | 4929 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
washing | +0.157 |
cloth | +0.147 |
sorting | +0.134 |
mixing | +0.130 |
living | +0.129 |
Neg logit contributions | |
former | -0.228 |
needed | -0.228 |
earned | -0.222 |
des | -0.221 |
isdom | -0.220 |
Token | it |
Token ID | 340 |
Feature activation | +9.04e-01 |
Pos logit contributions | |
washing | +0.055 |
cloth | +0.050 |
sorting | +0.046 |
living | +0.043 |
mixing | +0.043 |
Neg logit contributions | |
former | -0.101 |
ossible | -0.100 |
isdom | -0.099 |
needed | -0.099 |
earned | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
washing | +0.015 |
cloth | +0.014 |
sorting | +0.012 |
living | +0.012 |
mixing | +0.011 |
Neg logit contributions | |
isdom | -0.039 |
former | -0.039 |
ossible | -0.039 |
needed | -0.039 |
earned | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
washing | +0.022 |
cloth | +0.019 |
sorting | +0.016 |
mixing | +0.016 |
Mil | +0.015 |
Neg logit contributions | |
ossible | -0.051 |
needed | -0.050 |
bye | -0.050 |
former | -0.050 |
prey | -0.049 |
Token |
|
Token ID | 198 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
washing | +0.063 |
cloth | +0.056 |
sorting | +0.055 |
Mil | +0.052 |
living | +0.052 |
Neg logit contributions | |
needed | -0.079 |
former | -0.078 |
earned | -0.077 |
des | -0.077 |
isdom | -0.076 |
Token |
|
Token ID | 198 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
idents | +0.008 |
former | +0.008 |
earned | +0.008 |
needed | +0.008 |
night | +0.008 |
Neg logit contributions | |
washing | -0.006 |
living | -0.005 |
cloth | -0.005 |
sorting | -0.005 |
Mil | -0.005 |
Token | The |
Token ID | 464 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
former | +0.010 |
idents | +0.009 |
isdom | +0.009 |
needed | +0.009 |
earned | +0.009 |
Neg logit contributions | |
washing | -0.006 |
cloth | -0.006 |
sorting | -0.005 |
living | -0.005 |
mixing | -0.005 |
Token | zo |
Token ID | 40565 |
Feature activation | +6.94e+00 |
Pos logit contributions | |
washing | +0.073 |
cloth | +0.066 |
sorting | +0.063 |
mixing | +0.060 |
living | +0.059 |
Neg logit contributions | |
former | -0.087 |
ossible | -0.086 |
needed | -0.086 |
isdom | -0.086 |
earned | -0.085 |
Token | oke |
Token ID | 2088 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
washing | +0.141 |
cloth | +0.118 |
sorting | +0.112 |
mixing | +0.104 |
dripping | +0.100 |
Neg logit contributions | |
ossible | -0.276 |
bye | -0.274 |
isdom | -0.273 |
needed | -0.272 |
former | -0.272 |
Token | eper |
Token ID | 5723 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
washing | +0.334 |
cloth | +0.277 |
sorting | +0.271 |
dripping | +0.251 |
cups | +0.248 |
Neg logit contributions | |
ossible | -0.433 |
bye | -0.417 |
isdom | -0.405 |
oof | -0.398 |
earned | -0.392 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
washing | +0.090 |
cloth | +0.081 |
sorting | +0.078 |
mixing | +0.074 |
dripping | +0.072 |
Neg logit contributions | |
ossible | -0.101 |
former | -0.100 |
isdom | -0.099 |
needed | -0.098 |
earned | -0.097 |
Token | to |
Token ID | 284 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
washing | +0.087 |
cloth | +0.079 |
sorting | +0.076 |
mixing | +0.074 |
dripping | +0.071 |
Neg logit contributions | |
former | -0.087 |
ossible | -0.086 |
needed | -0.085 |
isdom | -0.084 |
earned | -0.084 |
Token | the |
Token ID | 262 |
Feature activation | +3.15e+00 |
Pos logit contributions | |
former | +0.070 |
needed | +0.070 |
earned | +0.070 |
ossible | +0.070 |
isdom | +0.069 |
Neg logit contributions | |
washing | -0.029 |
sorting | -0.023 |
cloth | -0.023 |
Mil | -0.021 |
mixing | -0.021 |
Token | pipe |
Token ID | 12656 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
washing | +0.084 |
cloth | +0.081 |
sorting | +0.075 |
mixing | +0.071 |
dripping | +0.069 |
Neg logit contributions | |
former | -0.101 |
prey | -0.094 |
bye | -0.094 |
ossible | -0.094 |
needed | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | +3.29e+00 |
Pos logit contributions | |
washing | +0.026 |
cloth | +0.025 |
sorting | +0.023 |
racks | +0.023 |
Mil | +0.022 |
Neg logit contributions | |
needed | -0.038 |
bye | -0.038 |
night | -0.038 |
former | -0.038 |
ossible | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
washing | +0.093 |
cloth | +0.084 |
sorting | +0.082 |
Mil | +0.078 |
mixing | +0.077 |
Neg logit contributions | |
needed | -0.105 |
former | -0.103 |
earned | -0.102 |
des | -0.101 |
isdom | -0.100 |
Token |
|
Token ID | 198 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
washing | +0.020 |
cloth | +0.018 |
sorting | +0.018 |
living | +0.017 |
Mil | +0.017 |
Neg logit contributions | |
former | -0.022 |
needed | -0.021 |
idents | -0.021 |
earned | -0.021 |
bye | -0.020 |
Token | The |
Token ID | 464 |
Feature activation | +3.80e+00 |
Pos logit contributions | |
washing | +0.018 |
cloth | +0.016 |
sorting | +0.015 |
mixing | +0.014 |
living | +0.014 |
Neg logit contributions | |
former | -0.028 |
needed | -0.027 |
isdom | -0.027 |
earned | -0.027 |
ossible | -0.027 |
Token | zo |
Token ID | 40565 |
Feature activation | +6.85e+00 |
Pos logit contributions | |
washing | +0.102 |
cloth | +0.092 |
sorting | +0.088 |
mixing | +0.084 |
dripping | +0.081 |
Neg logit contributions | |
former | -0.124 |
ossible | -0.122 |
needed | -0.122 |
isdom | -0.121 |
earned | -0.120 |
Token | oke |
Token ID | 2088 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
washing | +0.162 |
cloth | +0.134 |
sorting | +0.128 |
mixing | +0.121 |
dripping | +0.118 |
Neg logit contributions | |
ossible | -0.257 |
bye | -0.257 |
isdom | -0.250 |
needed | -0.249 |
earned | -0.247 |
Token | eper |
Token ID | 5723 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
washing | +0.381 |
sorting | +0.313 |
cloth | +0.313 |
cups | +0.294 |
dripping | +0.290 |
Neg logit contributions | |
ossible | -0.400 |
bye | -0.376 |
oof | -0.352 |
haw | -0.347 |
isdom | -0.347 |
Token | takes |
Token ID | 2753 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
washing | +0.093 |
cloth | +0.083 |
sorting | +0.081 |
mixing | +0.077 |
dripping | +0.075 |
Neg logit contributions | |
ossible | -0.096 |
isdom | -0.094 |
former | -0.094 |
needed | -0.093 |
earned | -0.092 |
Token | them |
Token ID | 606 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
washing | +0.084 |
cloth | +0.074 |
sorting | +0.070 |
mixing | +0.066 |
Mil | +0.064 |
Neg logit contributions | |
former | -0.137 |
ossible | -0.135 |
needed | -0.134 |
isdom | -0.134 |
earned | -0.133 |
Token | to |
Token ID | 284 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
washing | +0.047 |
cloth | +0.041 |
sorting | +0.039 |
living | +0.038 |
cups | +0.035 |
Neg logit contributions | |
former | -0.077 |
isdom | -0.075 |
oof | -0.072 |
earned | -0.072 |
des | -0.072 |
Token | their |
Token ID | 511 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
earned | +0.060 |
needed | +0.060 |
des | +0.059 |
ossible | +0.059 |
isdom | +0.059 |
Neg logit contributions | |
washing | -0.024 |
sorting | -0.019 |
cloth | -0.018 |
Mil | -0.018 |
mixing | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | +4.33e+00 |
Pos logit contributions | |
washing | +0.009 |
sorting | +0.007 |
cloth | +0.007 |
mixing | +0.006 |
living | +0.006 |
Neg logit contributions | |
needed | -0.020 |
des | -0.020 |
former | -0.020 |
earned | -0.020 |
ossible | -0.020 |
Token | seals |
Token ID | 29718 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
washing | +0.120 |
cloth | +0.104 |
sorting | +0.102 |
living | +0.099 |
mixing | +0.098 |
Neg logit contributions | |
earned | -0.141 |
isdom | -0.141 |
needed | -0.141 |
ossible | -0.140 |
des | -0.138 |
Token | and |
Token ID | 290 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
washing | +0.079 |
cloth | +0.069 |
sorting | +0.065 |
mixing | +0.061 |
living | +0.061 |
Neg logit contributions | |
former | -0.154 |
needed | -0.151 |
isdom | -0.150 |
earned | -0.150 |
ossible | -0.149 |
Token | the |
Token ID | 262 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
washing | +0.025 |
cloth | +0.023 |
sorting | +0.021 |
living | +0.020 |
cups | +0.020 |
Neg logit contributions | |
isdom | -0.036 |
former | -0.036 |
needed | -0.036 |
earned | -0.035 |
ossible | -0.035 |
Token | zo |
Token ID | 40565 |
Feature activation | +6.60e+00 |
Pos logit contributions | |
washing | +0.125 |
cloth | +0.115 |
sorting | +0.109 |
mixing | +0.104 |
dripping | +0.100 |
Neg logit contributions | |
former | -0.153 |
ossible | -0.151 |
needed | -0.149 |
bye | -0.149 |
isdom | -0.146 |
Token | oke |
Token ID | 2088 |
Feature activation | +1.21e+01 |
Pos logit contributions | |
washing | +0.160 |
cloth | +0.133 |
sorting | +0.128 |
mixing | +0.119 |
dripping | +0.118 |
Neg logit contributions | |
bye | -0.244 |
ossible | -0.243 |
needed | -0.238 |
isdom | -0.238 |
earned | -0.237 |
Token | ep |
Token ID | 538 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
washing | +0.318 |
cloth | +0.267 |
sorting | +0.265 |
mixing | +0.242 |
dripping | +0.238 |
Neg logit contributions | |
ossible | -0.436 |
bye | -0.418 |
isdom | -0.409 |
oof | -0.403 |
earned | -0.401 |
Token | ers |
Token ID | 364 |
Feature activation | +6.13e+00 |
Pos logit contributions | |
washing | +0.035 |
cloth | +0.031 |
sorting | +0.029 |
mixing | +0.028 |
living | +0.027 |
Neg logit contributions | |
former | -0.052 |
needed | -0.051 |
ossible | -0.050 |
isdom | -0.050 |
earned | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
washing | +0.124 |
cloth | +0.109 |
sorting | +0.101 |
mixing | +0.095 |
living | +0.092 |
Neg logit contributions | |
former | -0.242 |
needed | -0.238 |
isdom | -0.237 |
ossible | -0.237 |
earned | -0.236 |
Token | They |
Token ID | 1119 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
washing | +0.086 |
cloth | +0.078 |
sorting | +0.075 |
mixing | +0.071 |
Mil | +0.070 |
Neg logit contributions | |
needed | -0.099 |
former | -0.099 |
earned | -0.098 |
isdom | -0.097 |
ossible | -0.097 |
Token | walked |
Token ID | 6807 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
washing | +0.082 |
cloth | +0.078 |
Mil | +0.069 |
cups | +0.069 |
mo | +0.068 |
Neg logit contributions | |
needed | -0.128 |
isdom | -0.128 |
earned | -0.126 |
idents | -0.125 |
former | -0.125 |
Token | " |
Token ID | 1 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
washing | +0.002 |
cloth | +0.002 |
living | +0.002 |
sorting | +0.002 |
Mil | +0.002 |
Neg logit contributions | |
idents | -0.003 |
former | -0.003 |
night | -0.003 |
earned | -0.003 |
revenge | -0.003 |
Token | Hey |
Token ID | 10814 |
Feature activation | +5.28e+00 |
Pos logit contributions | |
washing | +0.141 |
cloth | +0.127 |
mixing | +0.117 |
Mil | +0.117 |
sorting | +0.116 |
Neg logit contributions | |
former | -0.185 |
tal | -0.184 |
isdom | -0.184 |
idents | -0.179 |
earned | -0.178 |
Token | , |
Token ID | 11 |
Feature activation | +9.54e+00 |
Pos logit contributions | |
washing | +0.124 |
cloth | +0.111 |
sorting | +0.102 |
dripping | +0.097 |
mixing | +0.097 |
Neg logit contributions | |
ossible | -0.188 |
bye | -0.187 |
oof | -0.183 |
needed | -0.182 |
isdom | -0.181 |
Token | hey |
Token ID | 17207 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
washing | +0.252 |
cloth | +0.229 |
sorting | +0.219 |
mixing | +0.212 |
dripping | +0.205 |
Neg logit contributions | |
ossible | -0.306 |
needed | -0.300 |
former | -0.296 |
isdom | -0.295 |
bye | -0.294 |
Token | , |
Token ID | 11 |
Feature activation | +9.52e+00 |
Pos logit contributions | |
washing | +0.324 |
cloth | +0.288 |
sorting | +0.274 |
dripping | +0.268 |
racks | +0.265 |
Neg logit contributions | |
ossible | -0.399 |
bye | -0.385 |
isdom | -0.383 |
encounter | -0.373 |
oof | -0.369 |
Token | hey |
Token ID | 17207 |
Feature activation | +1.20e+01 |
Pos logit contributions | |
washing | +0.267 |
cloth | +0.253 |
sorting | +0.233 |
mixing | +0.228 |
cups | +0.226 |
Neg logit contributions | |
ossible | -0.283 |
needed | -0.272 |
bye | -0.268 |
revenge | -0.265 |
isdom | -0.265 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
washing | +0.333 |
cloth | +0.307 |
sorting | +0.292 |
racks | +0.285 |
dripping | +0.283 |
Neg logit contributions | |
ossible | -0.359 |
bye | -0.347 |
isdom | -0.345 |
encounter | -0.340 |
faith | -0.335 |
Token | he |
Token ID | 339 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
washing | +0.032 |
cloth | +0.028 |
sorting | +0.027 |
Mil | +0.027 |
cups | +0.025 |
Neg logit contributions | |
former | -0.056 |
earned | -0.055 |
des | -0.054 |
needed | -0.054 |
isdom | -0.053 |
Token | yelled |
Token ID | 22760 |
Feature activation | +2.12e+00 |
Pos logit contributions | |
washing | +0.082 |
cloth | +0.077 |
sorting | +0.073 |
mixing | +0.070 |
Mil | +0.068 |
Neg logit contributions | |
former | -0.087 |
needed | -0.086 |
earned | -0.085 |
des | -0.084 |
isdom | -0.084 |
Token | at |
Token ID | 379 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
washing | +0.043 |
cloth | +0.038 |
sorting | +0.035 |
mixing | +0.034 |
living | +0.032 |
Neg logit contributions | |
ossible | -0.081 |
former | -0.081 |
needed | -0.080 |
isdom | -0.079 |
bye | -0.079 |
Token | the |
Token ID | 262 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
washing | +0.053 |
cloth | +0.046 |
sorting | +0.044 |
mixing | +0.040 |
Mil | +0.040 |
Neg logit contributions | |
needed | -0.095 |
des | -0.095 |
former | -0.094 |
earned | -0.094 |
ossible | -0.092 |
Token |
|
Token ID | 198 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
washing | +0.087 |
cloth | +0.078 |
sorting | +0.075 |
mixing | +0.072 |
living | +0.070 |
Neg logit contributions | |
former | -0.093 |
needed | -0.092 |
ossible | -0.091 |
isdom | -0.091 |
earned | -0.091 |
Token |
|
Token ID | 198 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
former | +0.005 |
needed | +0.005 |
earned | +0.004 |
isdom | +0.004 |
des | +0.004 |
Neg logit contributions | |
washing | -0.005 |
cloth | -0.005 |
sorting | -0.005 |
mixing | -0.004 |
living | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
washing | +0.008 |
sorting | +0.007 |
pit | +0.007 |
cloth | +0.007 |
mixing | +0.007 |
Neg logit contributions | |
ossible | -0.008 |
Dear | -0.007 |
bye | -0.006 |
Gar | -0.006 |
know | -0.006 |
Token | Hey |
Token ID | 10814 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
washing | +0.165 |
sorting | +0.157 |
cloth | +0.153 |
mixing | +0.143 |
dripping | +0.142 |
Neg logit contributions | |
ossible | -0.137 |
needed | -0.137 |
Dear | -0.136 |
des | -0.133 |
former | -0.131 |
Token | , |
Token ID | 11 |
Feature activation | +8.76e+00 |
Pos logit contributions | |
washing | +0.150 |
cloth | +0.140 |
ers | +0.138 |
living | +0.136 |
racks | +0.134 |
Neg logit contributions | |
ossible | -0.156 |
idents | -0.155 |
bye | -0.152 |
oof | -0.149 |
encounter | -0.147 |
Token | hey |
Token ID | 17207 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
washing | +0.234 |
cloth | +0.219 |
sorting | +0.214 |
ers | +0.214 |
mixing | +0.208 |
Neg logit contributions | |
ossible | -0.265 |
idents | -0.264 |
needed | -0.241 |
know | -0.240 |
oof | -0.239 |
Token | , |
Token ID | 11 |
Feature activation | +8.53e+00 |
Pos logit contributions | |
washing | +0.375 |
racks | +0.352 |
sorting | +0.349 |
cloth | +0.347 |
cabinets | +0.339 |
Neg logit contributions | |
ossible | -0.284 |
idents | -0.262 |
know | -0.257 |
isdom | -0.257 |
bye | -0.257 |
Token | hey |
Token ID | 17207 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
washing | +0.229 |
cloth | +0.220 |
sorting | +0.204 |
ers | +0.204 |
mixing | +0.200 |
Neg logit contributions | |
ossible | -0.260 |
idents | -0.252 |
needed | -0.247 |
revenge | -0.241 |
bye | -0.241 |
Token | ! |
Token ID | 0 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
washing | +0.389 |
cloth | +0.364 |
sorting | +0.363 |
racks | +0.360 |
cabinets | +0.350 |
Neg logit contributions | |
ossible | -0.270 |
idents | -0.253 |
isdom | -0.250 |
encounter | -0.250 |
bye | -0.249 |
Token | What |
Token ID | 1867 |
Feature activation | +7.68e+00 |
Pos logit contributions | |
washing | +0.150 |
cloth | +0.142 |
sorting | +0.134 |
mixing | +0.129 |
dripping | +0.128 |
Neg logit contributions | |
ossible | -0.117 |
needed | -0.110 |
isdom | -0.109 |
earned | -0.108 |
bye | -0.108 |
Token | is |
Token ID | 318 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
washing | +0.210 |
cloth | +0.190 |
sorting | +0.182 |
mixing | +0.174 |
living | +0.168 |
Neg logit contributions | |
former | -0.250 |
ossible | -0.246 |
needed | -0.245 |
isdom | -0.243 |
earned | -0.242 |
Token | ." |
Token ID | 526 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
washing | +0.026 |
cloth | +0.025 |
sorting | +0.022 |
mixing | +0.022 |
dripping | +0.022 |
Neg logit contributions | |
des | -0.034 |
needed | -0.034 |
ossible | -0.033 |
bye | -0.033 |
earned | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
former | +0.061 |
needed | +0.060 |
earned | +0.058 |
des | +0.058 |
isdom | +0.057 |
Neg logit contributions | |
washing | -0.043 |
cloth | -0.039 |
Mil | -0.038 |
Sara | -0.036 |
sorting | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
former | +0.099 |
needed | +0.095 |
earned | +0.094 |
idents | +0.093 |
des | +0.092 |
Neg logit contributions | |
washing | -0.073 |
cloth | -0.065 |
living | -0.063 |
Mil | -0.062 |
sorting | -0.060 |
Token | He |
Token ID | 1544 |
Feature activation | -5.31e-01 |
Pos logit contributions | |
former | +0.094 |
needed | +0.093 |
ossible | +0.092 |
isdom | +0.092 |
earned | +0.092 |
Neg logit contributions | |
washing | -0.046 |
cloth | -0.040 |
sorting | -0.037 |
mixing | -0.035 |
living | -0.034 |
Token | unh |
Token ID | 14274 |
Feature activation | +9.88e-02 |
Pos logit contributions | |
ossible | +0.017 |
former | +0.017 |
needed | +0.016 |
isdom | +0.016 |
earned | +0.016 |
Neg logit contributions | |
washing | -0.014 |
cloth | -0.013 |
sorting | -0.013 |
mixing | -0.012 |
dripping | -0.012 |
Token | ooks |
Token ID | 31085 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
washing | +0.003 |
cloth | +0.003 |
sorting | +0.003 |
mixing | +0.003 |
living | +0.003 |
Neg logit contributions | |
former | -0.003 |
ossible | -0.003 |
bye | -0.003 |
isdom | -0.003 |
earned | -0.003 |
Token | Anna |
Token ID | 11735 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
washing | +0.307 |
cloth | +0.264 |
sorting | +0.262 |
mixing | +0.249 |
racks | +0.236 |
Neg logit contributions | |
ossible | -0.384 |
isdom | -0.372 |
needed | -0.371 |
prey | -0.368 |
earned | -0.364 |
Token | from |
Token ID | 422 |
Feature activation | +6.24e+00 |
Pos logit contributions | |
washing | +0.059 |
cloth | +0.054 |
sorting | +0.053 |
Mil | +0.051 |
mixing | +0.051 |
Neg logit contributions | |
ossible | -0.049 |
needed | -0.047 |
tal | -0.046 |
earned | -0.046 |
moon | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +3.74e+00 |
Pos logit contributions | |
washing | +0.146 |
cloth | +0.128 |
sorting | +0.126 |
mixing | +0.123 |
dripping | +0.114 |
Neg logit contributions | |
ossible | -0.223 |
former | -0.218 |
needed | -0.215 |
tal | -0.214 |
prey | -0.213 |
Token | zip |
Token ID | 19974 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
washing | +0.101 |
cloth | +0.095 |
sorting | +0.091 |
mixing | +0.090 |
bled | +0.089 |
Neg logit contributions | |
ossible | -0.121 |
former | -0.112 |
needed | -0.110 |
tal | -0.107 |
faith | -0.106 |
Token | line |
Token ID | 1627 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
washing | +0.087 |
cloth | +0.074 |
sorting | +0.072 |
mixing | +0.070 |
dripping | +0.064 |
Neg logit contributions | |
ossible | -0.149 |
former | -0.147 |
bye | -0.146 |
needed | -0.146 |
earned | -0.145 |
Token | , |
Token ID | 11 |
Feature activation | +5.75e+00 |
Pos logit contributions | |
washing | +0.071 |
cloth | +0.071 |
Mil | +0.066 |
racks | +0.066 |
sorting | +0.064 |
Neg logit contributions | |
moon | -0.089 |
chat | -0.083 |
encounter | -0.081 |
ossible | -0.079 |
visit | -0.077 |
Token | da |
Token ID | 12379 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
washing | +0.133 |
cloth | +0.128 |
sorting | +0.123 |
bled | +0.115 |
bling | +0.114 |
Neg logit contributions | |
ossible | -0.183 |
former | -0.179 |
faith | -0.179 |
needed | -0.178 |
tal | -0.177 |
Token | is |
Token ID | 271 |
Feature activation | +7.31e+00 |
Pos logit contributions | |
cloth | +0.049 |
washing | +0.049 |
sorting | +0.041 |
racks | +0.036 |
living | +0.034 |
Neg logit contributions | |
isdom | -0.112 |
oof | -0.109 |
needed | -0.109 |
former | -0.108 |
moon | -0.106 |
Token | ies |
Token ID | 444 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
washing | +0.259 |
cloth | +0.238 |
sorting | +0.238 |
racks | +0.228 |
living | +0.219 |
Neg logit contributions | |
isdom | -0.153 |
awk | -0.152 |
oof | -0.147 |
ossible | -0.144 |
moon | -0.142 |
Token | , |
Token ID | 11 |
Feature activation | +5.97e+00 |
Pos logit contributions | |
cloth | +0.068 |
washing | +0.067 |
Mil | +0.066 |
racks | +0.064 |
bled | +0.059 |
Neg logit contributions | |
moon | -0.075 |
encounter | -0.072 |
chat | -0.070 |
ossible | -0.070 |
isdom | -0.067 |
Token | tul |
Token ID | 48373 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
washing | +0.153 |
cloth | +0.146 |
sorting | +0.142 |
racks | +0.129 |
bling | +0.129 |
Neg logit contributions | |
ossible | -0.178 |
tal | -0.173 |
former | -0.172 |
needed | -0.172 |
isdom | -0.171 |
Token | ips |
Token ID | 2419 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
washing | +0.291 |
cloth | +0.262 |
sorting | +0.257 |
floor | +0.245 |
living | +0.243 |
Neg logit contributions | |
ossible | -0.425 |
isdom | -0.404 |
uld | -0.404 |
oof | -0.391 |
moon | -0.389 |
Token | and |
Token ID | 290 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
washing | +0.058 |
cloth | +0.050 |
sorting | +0.048 |
mixing | +0.047 |
Mil | +0.045 |
Neg logit contributions | |
ossible | -0.074 |
former | -0.072 |
needed | -0.072 |
isdom | -0.071 |
bye | -0.071 |
Token | more |
Token ID | 517 |
Feature activation | +6.14e-01 |
Pos logit contributions | |
former | +0.031 |
chat | +0.031 |
faith | +0.030 |
know | +0.029 |
understand | +0.028 |
Neg logit contributions | |
washing | -0.024 |
cloth | -0.024 |
bling | -0.023 |
register | -0.022 |
bled | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
washing | +0.016 |
bled | +0.015 |
cloth | +0.014 |
bling | +0.014 |
sorting | +0.014 |
Neg logit contributions | |
prey | -0.019 |
powers | -0.019 |
former | -0.019 |
needed | -0.019 |
tal | -0.018 |
Token | She |
Token ID | 1375 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
washing | +0.045 |
cloth | +0.042 |
sorting | +0.040 |
dripping | +0.040 |
mixing | +0.039 |
Neg logit contributions | |
ossible | -0.038 |
former | -0.036 |
needed | -0.035 |
idents | -0.035 |
bye | -0.035 |
Token | " |
Token ID | 1 |
Feature activation | +5.23e+00 |
Pos logit contributions | |
washing | +0.008 |
sorting | +0.007 |
pit | +0.007 |
cloth | +0.007 |
mixing | +0.007 |
Neg logit contributions | |
ossible | -0.008 |
Dear | -0.007 |
bye | -0.006 |
Gar | -0.006 |
know | -0.006 |
Token | Hey |
Token ID | 10814 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
washing | +0.165 |
sorting | +0.157 |
cloth | +0.153 |
mixing | +0.143 |
dripping | +0.142 |
Neg logit contributions | |
ossible | -0.137 |
needed | -0.137 |
Dear | -0.136 |
des | -0.133 |
former | -0.131 |
Token | , |
Token ID | 11 |
Feature activation | +8.76e+00 |
Pos logit contributions | |
washing | +0.150 |
cloth | +0.140 |
ers | +0.138 |
living | +0.136 |
racks | +0.134 |
Neg logit contributions | |
ossible | -0.156 |
idents | -0.155 |
bye | -0.152 |
oof | -0.149 |
encounter | -0.147 |
Token | hey |
Token ID | 17207 |
Feature activation | +1.18e+01 |
Pos logit contributions | |
washing | +0.234 |
cloth | +0.219 |
sorting | +0.214 |
ers | +0.214 |
mixing | +0.208 |
Neg logit contributions | |
ossible | -0.265 |
idents | -0.264 |
needed | -0.241 |
know | -0.240 |
oof | -0.239 |
Token | , |
Token ID | 11 |
Feature activation | +8.53e+00 |
Pos logit contributions | |
washing | +0.375 |
racks | +0.352 |
sorting | +0.349 |
cloth | +0.347 |
cabinets | +0.339 |
Neg logit contributions | |
ossible | -0.284 |
idents | -0.262 |
know | -0.257 |
isdom | -0.257 |
bye | -0.257 |
Token | hey |
Token ID | 17207 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
washing | +0.229 |
cloth | +0.220 |
sorting | +0.204 |
ers | +0.204 |
mixing | +0.200 |
Neg logit contributions | |
ossible | -0.260 |
idents | -0.252 |
needed | -0.247 |
revenge | -0.241 |
bye | -0.241 |
Token | ! |
Token ID | 0 |
Feature activation | +4.52e+00 |
Pos logit contributions | |
washing | +0.389 |
cloth | +0.364 |
sorting | +0.363 |
racks | +0.360 |
cabinets | +0.350 |
Neg logit contributions | |
ossible | -0.270 |
idents | -0.253 |
isdom | -0.250 |
encounter | -0.250 |
bye | -0.249 |
Token | What |
Token ID | 1867 |
Feature activation | +7.68e+00 |
Pos logit contributions | |
washing | +0.150 |
cloth | +0.142 |
sorting | +0.134 |
mixing | +0.129 |
dripping | +0.128 |
Neg logit contributions | |
ossible | -0.117 |
needed | -0.110 |
isdom | -0.109 |
earned | -0.108 |
bye | -0.108 |
Token | is |
Token ID | 318 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
washing | +0.210 |
cloth | +0.190 |
sorting | +0.182 |
mixing | +0.174 |
living | +0.168 |
Neg logit contributions | |
former | -0.250 |
ossible | -0.246 |
needed | -0.245 |
isdom | -0.243 |
earned | -0.242 |
Token | going |
Token ID | 1016 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
cloth | +0.123 |
pit | +0.108 |
Mil | +0.108 |
ters | +0.106 |
ers | +0.104 |
Neg logit contributions | |
ossible | -0.093 |
idents | -0.091 |
oof | -0.089 |
revenge | -0.089 |
isdom | -0.085 |
Token | on |
Token ID | 319 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
washing | +0.050 |
cloth | +0.042 |
sorting | +0.040 |
mixing | +0.038 |
living | +0.035 |
Neg logit contributions | |
former | -0.111 |
needed | -0.108 |
ossible | -0.108 |
isdom | -0.108 |
earned | -0.107 |
Token | monkeys |
Token ID | 26697 |
Feature activation | +6.65e+00 |
Pos logit contributions | |
washing | +0.151 |
cloth | +0.130 |
sorting | +0.128 |
living | +0.125 |
mixing | +0.122 |
Neg logit contributions | |
former | -0.220 |
des | -0.220 |
earned | -0.217 |
ossible | -0.217 |
needed | -0.216 |
Token | , |
Token ID | 11 |
Feature activation | +9.24e+00 |
Pos logit contributions | |
washing | +0.142 |
cloth | +0.124 |
sorting | +0.117 |
mixing | +0.107 |
Mil | +0.107 |
Neg logit contributions | |
ossible | -0.247 |
former | -0.245 |
needed | -0.244 |
isdom | -0.241 |
earned | -0.240 |
Token | elephants |
Token ID | 25368 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
washing | +0.224 |
cloth | +0.200 |
sorting | +0.191 |
mixing | +0.181 |
living | +0.173 |
Neg logit contributions | |
former | -0.326 |
ossible | -0.322 |
needed | -0.321 |
isdom | -0.318 |
earned | -0.317 |
Token | , |
Token ID | 11 |
Feature activation | +8.74e+00 |
Pos logit contributions | |
washing | +0.107 |
cloth | +0.095 |
sorting | +0.090 |
Mil | +0.085 |
racks | +0.083 |
Neg logit contributions | |
ossible | -0.154 |
isdom | -0.149 |
bye | -0.147 |
prey | -0.146 |
needed | -0.145 |
Token | gir |
Token ID | 37370 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
washing | +0.216 |
cloth | +0.196 |
sorting | +0.186 |
mixing | +0.174 |
Mil | +0.168 |
Neg logit contributions | |
ossible | -0.293 |
former | -0.291 |
needed | -0.290 |
isdom | -0.287 |
earned | -0.284 |
Token | aff |
Token ID | 2001 |
Feature activation | +1.17e+01 |
Pos logit contributions | |
washing | +0.179 |
cloth | +0.163 |
sorting | +0.156 |
mixing | +0.148 |
dripping | +0.145 |
Neg logit contributions | |
ossible | -0.181 |
isdom | -0.181 |
former | -0.181 |
bye | -0.180 |
needed | -0.180 |
Token | es |
Token ID | 274 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
washing | +0.328 |
cloth | +0.293 |
sorting | +0.278 |
mixing | +0.262 |
dripping | +0.258 |
Neg logit contributions | |
ossible | -0.370 |
isdom | -0.370 |
former | -0.363 |
needed | -0.361 |
earned | -0.359 |
Token | , |
Token ID | 11 |
Feature activation | +7.77e+00 |
Pos logit contributions | |
washing | +0.074 |
cloth | +0.066 |
sorting | +0.062 |
Mil | +0.058 |
mixing | +0.057 |
Neg logit contributions | |
ossible | -0.124 |
isdom | -0.123 |
former | -0.122 |
needed | -0.121 |
earned | -0.120 |
Token | and |
Token ID | 290 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
washing | +0.198 |
cloth | +0.186 |
sorting | +0.173 |
mixing | +0.158 |
Mil | +0.157 |
Neg logit contributions | |
former | -0.250 |
ossible | -0.250 |
isdom | -0.245 |
needed | -0.242 |
earned | -0.240 |
Token | lions |
Token ID | 31420 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
washing | +0.052 |
cloth | +0.048 |
sorting | +0.045 |
mixing | +0.043 |
dripping | +0.042 |
Neg logit contributions | |
former | -0.064 |
ossible | -0.063 |
needed | -0.063 |
isdom | -0.062 |
earned | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +2.55e+00 |
Pos logit contributions | |
washing | +0.056 |
cloth | +0.048 |
sorting | +0.045 |
mixing | +0.041 |
living | +0.039 |
Neg logit contributions | |
former | -0.128 |
ossible | -0.127 |
needed | -0.126 |
isdom | -0.125 |
earned | -0.125 |
Token | " |
Token ID | 366 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
washing | +0.120 |
cloth | +0.104 |
sorting | +0.101 |
mixing | +0.094 |
living | +0.093 |
Neg logit contributions | |
former | -0.186 |
isdom | -0.185 |
ossible | -0.181 |
needed | -0.181 |
earned | -0.181 |
Token | ch |
Token ID | 354 |
Feature activation | +4.29e+00 |
Pos logit contributions | |
washing | +0.016 |
cloth | +0.015 |
mixing | +0.014 |
ters | +0.014 |
Mil | +0.013 |
Neg logit contributions | |
isdom | -0.021 |
nightmare | -0.021 |
former | -0.021 |
tal | -0.021 |
idents | -0.020 |
Token | oo |
Token ID | 2238 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
washing | +0.131 |
cloth | +0.120 |
sorting | +0.115 |
mixing | +0.111 |
living | +0.108 |
Neg logit contributions | |
former | -0.127 |
needed | -0.123 |
isdom | -0.123 |
ossible | -0.122 |
earned | -0.121 |
Token | cho |
Token ID | 1727 |
Feature activation | +1.11e+01 |
Pos logit contributions | |
washing | +0.049 |
cloth | +0.045 |
sorting | +0.042 |
mixing | +0.041 |
dripping | +0.040 |
Neg logit contributions | |
former | -0.054 |
needed | -0.053 |
isdom | -0.052 |
ossible | -0.052 |
earned | -0.052 |
Token | o |
Token ID | 78 |
Feature activation | +5.69e+00 |
Pos logit contributions | |
washing | +0.337 |
cloth | +0.326 |
ers | +0.300 |
ters | +0.293 |
sorting | +0.287 |
Neg logit contributions | |
former | -0.336 |
faith | -0.312 |
isdom | -0.305 |
lasting | -0.302 |
nightmare | -0.300 |
Token | cho |
Token ID | 1727 |
Feature activation | +1.16e+01 |
Pos logit contributions | |
washing | +0.130 |
cloth | +0.123 |
ters | +0.110 |
ers | +0.108 |
dripping | +0.105 |
Neg logit contributions | |
former | -0.218 |
isdom | -0.211 |
faith | -0.204 |
des | -0.200 |
earned | -0.197 |
Token | o |
Token ID | 78 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
washing | +0.351 |
cloth | +0.339 |
ers | +0.312 |
ters | +0.307 |
dripping | +0.300 |
Neg logit contributions | |
former | -0.351 |
faith | -0.326 |
isdom | -0.323 |
lasting | -0.314 |
needed | -0.312 |
Token | ". |
Token ID | 1911 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
washing | +0.135 |
cloth | +0.128 |
ters | +0.113 |
ers | +0.110 |
dripping | +0.109 |
Neg logit contributions | |
former | -0.224 |
isdom | -0.219 |
faith | -0.211 |
earned | -0.205 |
des | -0.205 |
Token |
|
Token ID | 198 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
former | +0.021 |
needed | +0.020 |
isdom | +0.020 |
earned | +0.020 |
des | +0.020 |
Neg logit contributions | |
washing | -0.020 |
cloth | -0.018 |
sorting | -0.017 |
Mil | -0.016 |
mixing | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
former | +0.050 |
idents | +0.049 |
needed | +0.048 |
earned | +0.047 |
bye | +0.047 |
Neg logit contributions | |
washing | -0.036 |
cloth | -0.032 |
Mil | -0.030 |
sorting | -0.030 |
living | -0.030 |
Token | They |
Token ID | 2990 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
former | +0.038 |
idents | +0.037 |
isdom | +0.037 |
needed | +0.037 |
earned | +0.037 |
Neg logit contributions | |
washing | -0.025 |
cloth | -0.023 |
sorting | -0.021 |
mixing | -0.019 |
Mil | -0.019 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
cloth | +0.032 |
washing | +0.031 |
cups | +0.029 |
racks | +0.028 |
cabinets | +0.027 |
Neg logit contributions | |
faith | -0.033 |
needed | -0.033 |
former | -0.032 |
nightmare | -0.032 |
afar | -0.032 |
Token | of |
Token ID | 286 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
needed | +0.036 |
bye | +0.035 |
nightmare | +0.034 |
des | +0.034 |
earned | +0.034 |
Neg logit contributions | |
washing | -0.032 |
cloth | -0.031 |
sorting | -0.028 |
cups | -0.028 |
mixing | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
washing | +0.091 |
cloth | +0.089 |
sorting | +0.079 |
mixing | +0.075 |
cups | +0.075 |
Neg logit contributions | |
former | -0.132 |
isdom | -0.129 |
needed | -0.128 |
faith | -0.128 |
prey | -0.127 |
Token | cold |
Token ID | 4692 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
washing | +0.048 |
cloth | +0.046 |
sorting | +0.042 |
mixing | +0.040 |
Mil | +0.039 |
Neg logit contributions | |
former | -0.064 |
nightmare | -0.062 |
needed | -0.062 |
encounter | -0.061 |
revenge | -0.060 |
Token | por |
Token ID | 16964 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
bye | +0.011 |
encounter | +0.011 |
night | +0.011 |
nightmare | +0.011 |
needed | +0.011 |
Neg logit contributions | |
washing | -0.009 |
sorting | -0.008 |
cloth | -0.008 |
bling | -0.007 |
bled | -0.007 |
Token | cel |
Token ID | 5276 |
Feature activation | +1.15e+01 |
Pos logit contributions | |
washing | +0.160 |
cloth | +0.150 |
sorting | +0.146 |
mixing | +0.138 |
cups | +0.135 |
Neg logit contributions | |
former | -0.114 |
needed | -0.113 |
ossible | -0.111 |
earned | -0.110 |
isdom | -0.110 |
Token | ain |
Token ID | 391 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
washing | +0.248 |
cloth | +0.220 |
sorting | +0.205 |
mixing | +0.195 |
living | +0.188 |
Neg logit contributions | |
former | -0.437 |
ossible | -0.433 |
needed | -0.428 |
isdom | -0.426 |
bye | -0.425 |
Token | bowl |
Token ID | 9396 |
Feature activation | +2.63e-01 |
Pos logit contributions | |
washing | +0.086 |
cloth | +0.078 |
sorting | +0.076 |
mixing | +0.071 |
cups | +0.068 |
Neg logit contributions | |
former | -0.112 |
needed | -0.112 |
earned | -0.109 |
bye | -0.109 |
ossible | -0.109 |
Token | ! |
Token ID | 0 |
Feature activation | -8.30e-02 |
Pos logit contributions | |
washing | +0.006 |
cloth | +0.005 |
sorting | +0.005 |
mixing | +0.005 |
living | +0.004 |
Neg logit contributions | |
former | -0.010 |
needed | -0.010 |
ossible | -0.010 |
earned | -0.009 |
isdom | -0.009 |
Token | |
Token ID | 220 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
needed | +0.003 |
former | +0.003 |
isdom | +0.003 |
earned | +0.003 |
des | +0.003 |
Neg logit contributions | |
washing | -0.002 |
cloth | -0.002 |
sorting | -0.002 |
Mil | -0.002 |
mixing | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
washing | +0.059 |
cloth | +0.054 |
sorting | +0.052 |
living | +0.050 |
Mil | +0.049 |
Neg logit contributions | |
former | -0.063 |
needed | -0.062 |
bye | -0.062 |
des | -0.061 |
earned | -0.061 |
Token | homes |
Token ID | 5682 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
washing | +0.064 |
cloth | +0.059 |
sorting | +0.054 |
mixing | +0.052 |
dripping | +0.051 |
Neg logit contributions | |
ossible | -0.113 |
bye | -0.110 |
needed | -0.108 |
isdom | -0.108 |
des | -0.108 |
Token | too |
Token ID | 1165 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
isdom | +0.160 |
needed | +0.159 |
ossible | +0.159 |
oof | +0.158 |
former | +0.158 |
Neg logit contributions | |
washing | -0.061 |
cloth | -0.050 |
sorting | -0.048 |
living | -0.047 |
mixing | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
isdom | +0.058 |
earned | +0.057 |
needed | +0.056 |
oof | +0.056 |
uld | +0.056 |
Neg logit contributions | |
washing | -0.043 |
cloth | -0.038 |
sorting | -0.036 |
living | -0.035 |
dripping | -0.034 |
Token | The |
Token ID | 383 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
washing | +0.042 |
cloth | +0.038 |
sorting | +0.037 |
Mil | +0.036 |
living | +0.035 |
Neg logit contributions | |
needed | -0.051 |
former | -0.050 |
earned | -0.049 |
des | -0.049 |
isdom | -0.049 |
Token | zo |
Token ID | 40565 |
Feature activation | +5.80e+00 |
Pos logit contributions | |
washing | +0.059 |
cloth | +0.052 |
sorting | +0.051 |
living | +0.049 |
mixing | +0.048 |
Neg logit contributions | |
needed | -0.065 |
earned | -0.065 |
isdom | -0.065 |
ossible | -0.065 |
des | -0.065 |
Token | oke |
Token ID | 2088 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
washing | +0.108 |
cloth | +0.093 |
sorting | +0.087 |
mixing | +0.081 |
dripping | +0.076 |
Neg logit contributions | |
former | -0.237 |
ossible | -0.236 |
needed | -0.235 |
isdom | -0.233 |
earned | -0.232 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
washing | +0.256 |
cloth | +0.214 |
sorting | +0.207 |
mixing | +0.191 |
dripping | +0.186 |
Neg logit contributions | |
ossible | -0.441 |
bye | -0.430 |
isdom | -0.425 |
earned | -0.422 |
needed | -0.419 |
Token | was |
Token ID | 373 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
washing | +0.072 |
cloth | +0.066 |
sorting | +0.063 |
mixing | +0.060 |
living | +0.059 |
Neg logit contributions | |
former | -0.085 |
needed | -0.083 |
ossible | -0.082 |
earned | -0.082 |
isdom | -0.082 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
washing | +0.048 |
cloth | +0.044 |
sorting | +0.042 |
mixing | +0.040 |
cups | +0.040 |
Neg logit contributions | |
former | -0.051 |
needed | -0.049 |
ossible | -0.049 |
isdom | -0.048 |
earned | -0.047 |
Token | that |
Token ID | 326 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
former | +0.071 |
ossible | +0.070 |
needed | +0.070 |
earned | +0.070 |
isdom | +0.070 |
Neg logit contributions | |
washing | -0.052 |
cloth | -0.046 |
sorting | -0.044 |
mixing | -0.042 |
living | -0.041 |
Token | everyone |
Token ID | 2506 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
washing | +0.056 |
cloth | +0.051 |
sorting | +0.049 |
mixing | +0.046 |
living | +0.044 |
Neg logit contributions | |
former | -0.074 |
ossible | -0.073 |
needed | -0.073 |
isdom | -0.072 |
earned | -0.072 |
Token | killed |
Token ID | 2923 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
washing | +0.078 |
cloth | +0.068 |
sorting | +0.066 |
living | +0.065 |
mixing | +0.064 |
Neg logit contributions | |
former | -0.098 |
isdom | -0.096 |
idents | -0.096 |
des | -0.095 |
ossible | -0.094 |
Token | them |
Token ID | 606 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
washing | +0.044 |
cloth | +0.039 |
sorting | +0.038 |
mixing | +0.037 |
dripping | +0.035 |
Neg logit contributions | |
former | -0.054 |
ossible | -0.053 |
needed | -0.053 |
isdom | -0.053 |
bye | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
washing | +0.030 |
cloth | +0.025 |
sorting | +0.023 |
mixing | +0.021 |
living | +0.021 |
Neg logit contributions | |
former | -0.082 |
needed | -0.081 |
ossible | -0.081 |
isdom | -0.081 |
earned | -0.081 |
Token | The |
Token ID | 383 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
washing | +0.071 |
cloth | +0.066 |
sorting | +0.062 |
Mil | +0.060 |
mixing | +0.058 |
Neg logit contributions | |
needed | -0.086 |
former | -0.086 |
earned | -0.084 |
isdom | -0.083 |
bye | -0.083 |
Token | zo |
Token ID | 40565 |
Feature activation | +5.96e+00 |
Pos logit contributions | |
washing | +0.085 |
cloth | +0.077 |
sorting | +0.075 |
mixing | +0.072 |
living | +0.071 |
Neg logit contributions | |
former | -0.087 |
needed | -0.086 |
ossible | -0.086 |
isdom | -0.085 |
earned | -0.085 |
Token | oke |
Token ID | 2088 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
washing | +0.128 |
cloth | +0.107 |
sorting | +0.100 |
mixing | +0.094 |
dripping | +0.091 |
Neg logit contributions | |
bye | -0.234 |
ossible | -0.232 |
needed | -0.231 |
isdom | -0.230 |
earned | -0.228 |
Token | ep |
Token ID | 538 |
Feature activation | +8.11e-01 |
Pos logit contributions | |
washing | +0.308 |
cloth | +0.251 |
sorting | +0.249 |
mixing | +0.230 |
dripping | +0.226 |
Neg logit contributions | |
ossible | -0.405 |
bye | -0.402 |
isdom | -0.385 |
oof | -0.385 |
earned | -0.383 |
Token | ers |
Token ID | 364 |
Feature activation | +5.77e+00 |
Pos logit contributions | |
washing | +0.018 |
cloth | +0.017 |
ers | +0.015 |
sorting | +0.015 |
mixing | +0.014 |
Neg logit contributions | |
former | -0.030 |
needed | -0.030 |
des | -0.030 |
isdom | -0.030 |
ossible | -0.029 |
Token | shot |
Token ID | 2823 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
washing | +0.140 |
cloth | +0.131 |
sorting | +0.119 |
mixing | +0.114 |
living | +0.114 |
Neg logit contributions | |
former | -0.209 |
needed | -0.205 |
isdom | -0.203 |
earned | -0.201 |
ossible | -0.201 |
Token | the |
Token ID | 262 |
Feature activation | +3.59e+00 |
Pos logit contributions | |
washing | +0.040 |
cloth | +0.036 |
sorting | +0.033 |
mixing | +0.031 |
living | +0.030 |
Neg logit contributions | |
former | -0.070 |
ossible | -0.068 |
needed | -0.068 |
des | -0.068 |
isdom | -0.068 |
Token | tiger |
Token ID | 26241 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
washing | +0.108 |
cloth | +0.098 |
sorting | +0.095 |
mixing | +0.091 |
living | +0.089 |
Neg logit contributions | |
former | -0.106 |
ossible | -0.105 |
needed | -0.104 |
isdom | -0.104 |
earned | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
washing | +0.003 |
cloth | +0.003 |
sorting | +0.003 |
mixing | +0.003 |
Mil | +0.003 |
Neg logit contributions | |
former | -0.007 |
ossible | -0.007 |
needed | -0.007 |
bye | -0.007 |
earned | -0.007 |
Token | They |
Token ID | 1119 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
washing | +0.090 |
cloth | +0.082 |
sorting | +0.079 |
mixing | +0.077 |
dripping | +0.075 |
Neg logit contributions | |
ossible | -0.086 |
former | -0.085 |
isdom | -0.082 |
needed | -0.082 |
bye | -0.080 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.91e+00 |
Pos logit contributions | |
washing | +0.108 |
cloth | +0.098 |
sorting | +0.094 |
mixing | +0.090 |
living | +0.087 |
Neg logit contributions | |
former | -0.116 |
needed | -0.115 |
ossible | -0.115 |
isdom | -0.114 |
earned | -0.113 |
Token | the |
Token ID | 262 |
Feature activation | +4.08e+00 |
Pos logit contributions | |
washing | +0.117 |
cloth | +0.106 |
sorting | +0.102 |
mixing | +0.098 |
living | +0.095 |
Neg logit contributions | |
former | -0.118 |
ossible | -0.116 |
needed | -0.116 |
isdom | -0.115 |
earned | -0.114 |
Token | zo |
Token ID | 40565 |
Feature activation | +5.66e+00 |
Pos logit contributions | |
washing | +0.115 |
cloth | +0.101 |
sorting | +0.098 |
mixing | +0.095 |
living | +0.095 |
Neg logit contributions | |
ossible | -0.132 |
isdom | -0.131 |
needed | -0.130 |
earned | -0.130 |
former | -0.130 |
Token | oke |
Token ID | 2088 |
Feature activation | +1.14e+01 |
Pos logit contributions | |
washing | +0.173 |
cloth | +0.151 |
sorting | +0.138 |
mixing | +0.133 |
floor | +0.132 |
Neg logit contributions | |
needed | -0.168 |
bye | -0.167 |
encounter | -0.163 |
awk | -0.163 |
prey | -0.162 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
washing | +0.352 |
cloth | +0.300 |
sorting | +0.285 |
cabinets | +0.272 |
cups | +0.264 |
Neg logit contributions | |
ossible | -0.337 |
prey | -0.334 |
bye | -0.319 |
oof | -0.317 |
isdom | -0.306 |
Token | and |
Token ID | 290 |
Feature activation | +7.31e-01 |
Pos logit contributions | |
washing | +0.053 |
cloth | +0.047 |
Mil | +0.046 |
sorting | +0.043 |
ula | +0.041 |
Neg logit contributions | |
ossible | -0.076 |
needed | -0.075 |
earned | -0.074 |
encounter | -0.074 |
isdom | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | +4.44e+00 |
Pos logit contributions | |
washing | +0.022 |
cloth | +0.020 |
ula | +0.019 |
sorting | +0.019 |
bling | +0.019 |
Neg logit contributions | |
know | -0.020 |
faith | -0.019 |
ossible | -0.019 |
chat | -0.019 |
isdom | -0.018 |
Token | lion |
Token ID | 18744 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
washing | +0.121 |
cloth | +0.111 |
sorting | +0.106 |
mixing | +0.101 |
dripping | +0.097 |
Neg logit contributions | |
former | -0.143 |
ossible | -0.140 |
needed | -0.139 |
bye | -0.137 |
earned | -0.136 |
Token | . |
Token ID | 13 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
washing | +0.051 |
cloth | +0.046 |
sorting | +0.043 |
Mil | +0.040 |
mixing | +0.040 |
Neg logit contributions | |
ossible | -0.067 |
needed | -0.067 |
encounter | -0.067 |
former | -0.066 |
isdom | -0.066 |
Token | get |
Token ID | 651 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
former | +0.100 |
needed | +0.096 |
isdom | +0.095 |
ossible | +0.094 |
earned | +0.093 |
Neg logit contributions | |
washing | -0.080 |
cloth | -0.075 |
sorting | -0.070 |
mixing | -0.068 |
cups | -0.067 |
Token | into |
Token ID | 656 |
Feature activation | +6.58e-01 |
Pos logit contributions | |
washing | +0.060 |
cloth | +0.054 |
sorting | +0.052 |
mixing | +0.049 |
Mil | +0.048 |
Neg logit contributions | |
former | -0.073 |
needed | -0.072 |
ossible | -0.072 |
isdom | -0.070 |
des | -0.070 |
Token | all |
Token ID | 477 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
washing | +0.017 |
cloth | +0.015 |
sorting | +0.014 |
mixing | +0.014 |
Mil | +0.014 |
Neg logit contributions | |
former | -0.022 |
ossible | -0.021 |
needed | -0.021 |
isdom | -0.020 |
faith | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
ossible | +0.047 |
needed | +0.045 |
faith | +0.045 |
des | +0.044 |
lasting | +0.042 |
Neg logit contributions | |
washing | -0.037 |
cloth | -0.034 |
Mil | -0.033 |
per | -0.033 |
cel | -0.031 |
Token | n |
Token ID | 299 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
washing | +0.014 |
cloth | +0.014 |
Mil | +0.013 |
bled | +0.013 |
bling | +0.013 |
Neg logit contributions | |
former | -0.020 |
needed | -0.019 |
nightmare | -0.019 |
faith | -0.019 |
des | -0.019 |
Token | ooks |
Token ID | 31085 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
washing | +0.058 |
mixing | +0.049 |
sorting | +0.049 |
sticking | +0.049 |
holding | +0.049 |
Neg logit contributions | |
former | -0.060 |
ossible | -0.055 |
isdom | -0.053 |
needed | -0.051 |
earned | -0.049 |
Token | and |
Token ID | 290 |
Feature activation | -4.38e-01 |
Pos logit contributions | |
washing | +0.299 |
cloth | +0.269 |
sorting | +0.258 |
Mil | +0.250 |
racks | +0.242 |
Neg logit contributions | |
ossible | -0.350 |
needed | -0.341 |
former | -0.341 |
tal | -0.336 |
isdom | -0.336 |
Token | cr |
Token ID | 1067 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
former | +0.015 |
ossible | +0.015 |
needed | +0.015 |
isdom | +0.015 |
earned | +0.015 |
Neg logit contributions | |
washing | -0.011 |
cloth | -0.010 |
sorting | -0.009 |
mixing | -0.009 |
living | -0.009 |
Token | ann |
Token ID | 1236 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
washing | +0.082 |
cloth | +0.075 |
sorting | +0.073 |
mixing | +0.070 |
living | +0.068 |
Neg logit contributions | |
former | -0.073 |
ossible | -0.072 |
needed | -0.072 |
isdom | -0.071 |
earned | -0.070 |
Token | ies |
Token ID | 444 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
washing | +0.070 |
cloth | +0.063 |
sorting | +0.060 |
mixing | +0.057 |
living | +0.055 |
Neg logit contributions | |
former | -0.087 |
ossible | -0.087 |
needed | -0.086 |
isdom | -0.085 |
earned | -0.085 |
Token | . |
Token ID | 13 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
washing | +0.037 |
cloth | +0.033 |
sorting | +0.031 |
mixing | +0.029 |
Mil | +0.028 |
Neg logit contributions | |
ossible | -0.071 |
former | -0.070 |
needed | -0.070 |
earned | -0.069 |
isdom | -0.069 |
Token | I |
Token ID | 40 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
washing | +0.037 |
cloth | +0.032 |
sorting | +0.030 |
mixing | +0.027 |
living | +0.026 |
Neg logit contributions | |
former | -0.083 |
ossible | -0.082 |
needed | -0.081 |
isdom | -0.081 |
earned | -0.081 |
Token | didn |
Token ID | 1422 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
washing | +0.097 |
cloth | +0.092 |
living | +0.087 |
mixing | +0.086 |
sorting | +0.086 |
Neg logit contributions | |
isdom | -0.103 |
idents | -0.098 |
prey | -0.098 |
former | -0.097 |
earned | -0.096 |
Token | 't |
Token ID | 470 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
washing | +0.070 |
sorting | +0.062 |
cloth | +0.060 |
mixing | +0.058 |
living | +0.056 |
Neg logit contributions | |
former | -0.059 |
needed | -0.057 |
isdom | -0.057 |
bye | -0.056 |
ossible | -0.056 |
Token | mean |
Token ID | 1612 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
washing | +0.075 |
cloth | +0.071 |
sorting | +0.069 |
mixing | +0.068 |
cups | +0.068 |
Neg logit contributions | |
former | -0.024 |
ossible | -0.023 |
needed | -0.023 |
isdom | -0.022 |
bye | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
washing | +0.065 |
cloth | +0.058 |
sorting | +0.057 |
mixing | +0.054 |
living | +0.054 |
Neg logit contributions | |
former | -0.065 |
needed | -0.060 |
des | -0.058 |
ossible | -0.058 |
earned | -0.057 |
Token | knock |
Token ID | 10643 |
Feature activation | +1.13e+01 |
Pos logit contributions | |
washing | +0.015 |
cloth | +0.014 |
sorting | +0.013 |
mixing | +0.013 |
living | +0.012 |
Neg logit contributions | |
former | -0.012 |
ossible | -0.012 |
needed | -0.012 |
isdom | -0.012 |
earned | -0.012 |
Token | it |
Token ID | 340 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
washing | +0.267 |
cloth | +0.238 |
sorting | +0.229 |
mixing | +0.216 |
living | +0.205 |
Neg logit contributions | |
ossible | -0.396 |
needed | -0.388 |
former | -0.388 |
isdom | -0.387 |
bye | -0.385 |
Token | over |
Token ID | 625 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
washing | +0.024 |
on | +0.021 |
in | +0.020 |
cloth | +0.020 |
sorting | +0.018 |
Neg logit contributions | |
isdom | -0.107 |
earned | -0.102 |
faith | -0.101 |
des | -0.100 |
former | -0.100 |
Token | ." |
Token ID | 526 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
washing | +0.098 |
cloth | +0.085 |
sorting | +0.081 |
mixing | +0.076 |
living | +0.074 |
Neg logit contributions | |
former | -0.165 |
needed | -0.165 |
earned | -0.164 |
isdom | -0.163 |
ossible | -0.162 |
Token |
|
Token ID | 198 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
needed | +0.036 |
former | +0.035 |
earned | +0.035 |
des | +0.035 |
isdom | +0.034 |
Neg logit contributions | |
washing | -0.027 |
cloth | -0.023 |
sorting | -0.023 |
Mil | -0.022 |
holding | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
former | +0.058 |
idents | +0.057 |
needed | +0.056 |
earned | +0.055 |
des | +0.054 |
Neg logit contributions | |
washing | -0.040 |
cloth | -0.035 |
sorting | -0.034 |
Mil | -0.034 |
living | -0.034 |
Token |
|
Token ID | 198 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
former | +0.037 |
earned | +0.035 |
isdom | +0.035 |
needed | +0.035 |
des | +0.034 |
Neg logit contributions | |
washing | -0.029 |
living | -0.026 |
cloth | -0.025 |
sorting | -0.024 |
cups | -0.023 |
Token | " |
Token ID | 1 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
ossible | +0.019 |
needed | +0.017 |
Dear | +0.017 |
bye | +0.017 |
former | +0.017 |
Neg logit contributions | |
washing | -0.017 |
sorting | -0.016 |
cloth | -0.016 |
mixing | -0.015 |
dripping | -0.015 |
Token | Maybe |
Token ID | 13300 |
Feature activation | +3.19e-01 |
Pos logit contributions | |
washing | +0.134 |
cloth | +0.122 |
sorting | +0.116 |
mixing | +0.113 |
living | +0.109 |
Neg logit contributions | |
former | -0.134 |
isdom | -0.132 |
ossible | -0.131 |
needed | -0.129 |
earned | -0.129 |
Token | the |
Token ID | 262 |
Feature activation | +4.63e+00 |
Pos logit contributions | |
washing | +0.011 |
cloth | +0.011 |
sorting | +0.010 |
dripping | +0.009 |
mixing | +0.009 |
Neg logit contributions | |
bye | -0.008 |
ossible | -0.008 |
isdom | -0.008 |
prey | -0.008 |
needed | -0.007 |
Token | zo |
Token ID | 40565 |
Feature activation | +5.56e+00 |
Pos logit contributions | |
washing | +0.130 |
cloth | +0.123 |
sorting | +0.115 |
mixing | +0.110 |
dripping | +0.107 |
Neg logit contributions | |
former | -0.145 |
ossible | -0.144 |
bye | -0.141 |
prey | -0.139 |
needed | -0.138 |
Token | oke |
Token ID | 2088 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
washing | +0.140 |
cloth | +0.122 |
sorting | +0.115 |
mixing | +0.110 |
dripping | +0.107 |
Neg logit contributions | |
bye | -0.201 |
needed | -0.197 |
ossible | -0.197 |
earned | -0.193 |
isdom | -0.193 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
washing | +0.289 |
cloth | +0.247 |
sorting | +0.237 |
mixing | +0.220 |
dripping | +0.215 |
Neg logit contributions | |
ossible | -0.403 |
bye | -0.394 |
oof | -0.390 |
isdom | -0.388 |
needed | -0.382 |
Token | will |
Token ID | 481 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
washing | +0.063 |
cloth | +0.059 |
living | +0.055 |
sorting | +0.053 |
ers | +0.053 |
Neg logit contributions | |
former | -0.099 |
needed | -0.091 |
tal | -0.090 |
ossible | -0.090 |
isdom | -0.090 |
Token | come |
Token ID | 1282 |
Feature activation | +6.60e+00 |
Pos logit contributions | |
washing | +0.039 |
cloth | +0.035 |
sorting | +0.032 |
holding | +0.030 |
living | +0.029 |
Neg logit contributions | |
earned | -0.069 |
isdom | -0.069 |
needed | -0.068 |
to | -0.066 |
des | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
washing | +0.098 |
in | +0.096 |
cloth | +0.090 |
on | +0.088 |
holding | +0.084 |
Neg logit contributions | |
isdom | -0.290 |
earned | -0.284 |
former | -0.282 |
faith | -0.277 |
des | -0.276 |
Token | give |
Token ID | 1577 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
isdom | +0.001 |
oof | +0.001 |
tal | +0.001 |
former | +0.001 |
earned | +0.001 |
Neg logit contributions | |
washing | -0.001 |
cloth | -0.000 |
sorting | -0.000 |
living | -0.000 |
holding | -0.000 |
Token | like |
Token ID | 588 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
washing | +0.116 |
cloth | +0.101 |
sorting | +0.097 |
mixing | +0.092 |
living | +0.088 |
Neg logit contributions | |
former | -0.198 |
ossible | -0.195 |
needed | -0.194 |
isdom | -0.193 |
earned | -0.192 |
Token | lettuce |
Token ID | 39406 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
washing | +0.133 |
cloth | +0.120 |
sorting | +0.114 |
mixing | +0.108 |
living | +0.104 |
Neg logit contributions | |
former | -0.182 |
ossible | -0.180 |
needed | -0.179 |
isdom | -0.178 |
earned | -0.177 |
Token | , |
Token ID | 11 |
Feature activation | +8.08e+00 |
Pos logit contributions | |
washing | +0.086 |
cloth | +0.074 |
sorting | +0.069 |
mixing | +0.065 |
living | +0.062 |
Neg logit contributions | |
former | -0.190 |
needed | -0.188 |
ossible | -0.186 |
isdom | -0.186 |
earned | -0.185 |
Token | tomatoes |
Token ID | 23972 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
washing | +0.226 |
cloth | +0.208 |
sorting | +0.198 |
mixing | +0.188 |
living | +0.183 |
Neg logit contributions | |
former | -0.253 |
ossible | -0.246 |
needed | -0.245 |
isdom | -0.244 |
earned | -0.242 |
Token | and |
Token ID | 290 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
washing | +0.075 |
cloth | +0.065 |
sorting | +0.062 |
mixing | +0.058 |
dripping | +0.054 |
Neg logit contributions | |
former | -0.151 |
ossible | -0.150 |
needed | -0.149 |
isdom | -0.148 |
earned | -0.147 |
Token | cuc |
Token ID | 38421 |
Feature activation | +1.12e+01 |
Pos logit contributions | |
washing | +0.066 |
cloth | +0.061 |
sorting | +0.058 |
mixing | +0.055 |
dripping | +0.054 |
Neg logit contributions | |
former | -0.073 |
ossible | -0.072 |
needed | -0.071 |
isdom | -0.071 |
earned | -0.071 |
Token | umbers |
Token ID | 17024 |
Feature activation | +4.64e+00 |
Pos logit contributions | |
washing | +0.257 |
cloth | +0.230 |
sorting | +0.217 |
mixing | +0.207 |
cups | +0.196 |
Neg logit contributions | |
former | -0.409 |
needed | -0.403 |
ossible | -0.402 |
isdom | -0.399 |
earned | -0.397 |
Token | . |
Token ID | 13 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
washing | +0.092 |
cloth | +0.078 |
sorting | +0.075 |
mixing | +0.070 |
dripping | +0.066 |
Neg logit contributions | |
ossible | -0.182 |
former | -0.181 |
needed | -0.180 |
isdom | -0.179 |
earned | -0.178 |
Token | He |
Token ID | 679 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
washing | +0.078 |
cloth | +0.071 |
sorting | +0.069 |
mixing | +0.065 |
Mil | +0.065 |
Neg logit contributions | |
needed | -0.086 |
former | -0.086 |
earned | -0.085 |
des | -0.085 |
isdom | -0.084 |
Token | smiled |
Token ID | 13541 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
washing | +0.056 |
cloth | +0.050 |
sorting | +0.048 |
mixing | +0.046 |
living | +0.044 |
Neg logit contributions | |
former | -0.071 |
needed | -0.070 |
ossible | -0.070 |
isdom | -0.069 |
earned | -0.069 |
Token | as |
Token ID | 355 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
washing | +0.071 |
cloth | +0.062 |
sorting | +0.059 |
mixing | +0.055 |
living | +0.054 |
Neg logit contributions | |
former | -0.126 |
needed | -0.126 |
des | -0.125 |
earned | -0.125 |
ossible | -0.125 |
Token | the |
Token ID | 262 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
washing | +0.037 |
cloth | +0.035 |
sorting | +0.032 |
mixing | +0.031 |
dripping | +0.031 |
Neg logit contributions | |
former | -0.047 |
faith | -0.046 |
tal | -0.046 |
revenge | -0.045 |
isdom | -0.045 |
Token | tree |
Token ID | 5509 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
ossible | +0.005 |
isdom | +0.005 |
oof | +0.004 |
des | +0.004 |
to | +0.004 |
Neg logit contributions | |
washing | -0.003 |
living | -0.003 |
sorting | -0.003 |
mixing | -0.003 |
cloth | -0.002 |
Token | to |
Token ID | 284 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
ossible | +0.086 |
isdom | +0.085 |
former | +0.085 |
needed | +0.085 |
des | +0.085 |
Neg logit contributions | |
washing | -0.046 |
cloth | -0.040 |
sorting | -0.039 |
mixing | -0.036 |
living | -0.035 |
Token | its |
Token ID | 663 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
ossible | +0.156 |
needed | +0.154 |
des | +0.152 |
earned | +0.152 |
bye | +0.151 |
Neg logit contributions | |
washing | -0.122 |
cloth | -0.105 |
sorting | -0.105 |
mixing | -0.099 |
living | -0.096 |
Token | once |
Token ID | 1752 |
Feature activation | -5.99e+00 |
Pos logit contributions | |
des | +0.111 |
ossible | +0.109 |
isdom | +0.108 |
needed | +0.106 |
earned | +0.106 |
Neg logit contributions | |
washing | -0.120 |
living | -0.108 |
sorting | -0.107 |
mixing | -0.102 |
cloth | -0.099 |
Token | former |
Token ID | 1966 |
Feature activation | -1.19e+01 |
Pos logit contributions | |
former | +0.199 |
ossible | +0.197 |
needed | +0.196 |
isdom | +0.194 |
earned | +0.194 |
Neg logit contributions | |
washing | -0.159 |
cloth | -0.143 |
sorting | -0.137 |
mixing | -0.131 |
living | -0.125 |
Token | beauty |
Token ID | 8737 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
former | +0.318 |
ossible | +0.310 |
needed | +0.309 |
isdom | +0.307 |
earned | +0.305 |
Neg logit contributions | |
washing | -0.395 |
cloth | -0.364 |
sorting | -0.351 |
mixing | -0.338 |
dripping | -0.329 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
former | +0.155 |
needed | +0.153 |
bye | +0.152 |
earned | +0.151 |
isdom | +0.151 |
Neg logit contributions | |
washing | -0.087 |
cloth | -0.078 |
sorting | -0.072 |
mixing | -0.069 |
Mil | -0.066 |
Token | He |
Token ID | 679 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
former | +0.052 |
needed | +0.051 |
ossible | +0.050 |
isdom | +0.050 |
earned | +0.050 |
Neg logit contributions | |
washing | -0.058 |
cloth | -0.053 |
sorting | -0.052 |
mixing | -0.050 |
living | -0.049 |
Token | was |
Token ID | 373 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
needed | +0.072 |
idents | +0.071 |
des | +0.071 |
earned | +0.071 |
oof | +0.070 |
Neg logit contributions | |
washing | -0.042 |
cloth | -0.037 |
sorting | -0.036 |
living | -0.035 |
mixing | -0.034 |
Token | happy |
Token ID | 3772 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
former | +0.035 |
ossible | +0.034 |
needed | +0.034 |
isdom | +0.034 |
earned | +0.034 |
Neg logit contributions | |
washing | -0.029 |
cloth | -0.027 |
sorting | -0.026 |
mixing | -0.024 |
living | -0.024 |
Token | their |
Token ID | 511 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
washing | +0.054 |
cloth | +0.048 |
sorting | +0.046 |
mixing | +0.044 |
dripping | +0.042 |
Neg logit contributions | |
former | -0.077 |
ossible | -0.076 |
needed | -0.075 |
isdom | -0.075 |
earned | -0.074 |
Token | journey |
Token ID | 7002 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
isdom | +0.070 |
former | +0.064 |
oof | +0.064 |
to | +0.063 |
gob | +0.063 |
Neg logit contributions | |
washing | -0.113 |
sorting | -0.107 |
living | -0.102 |
mixing | -0.098 |
cups | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
former | +0.177 |
ossible | +0.175 |
needed | +0.174 |
bye | +0.173 |
earned | +0.172 |
Neg logit contributions | |
washing | -0.096 |
cloth | -0.085 |
sorting | -0.079 |
mixing | -0.076 |
dripping | -0.072 |
Token | had |
Token ID | 550 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
former | +0.127 |
needed | +0.126 |
isdom | +0.125 |
ossible | +0.125 |
earned | +0.124 |
Neg logit contributions | |
washing | -0.132 |
cloth | -0.119 |
sorting | -0.118 |
mixing | -0.113 |
living | -0.111 |
Token | many |
Token ID | 867 |
Feature activation | -5.23e+00 |
Pos logit contributions | |
former | +0.032 |
isdom | +0.032 |
idents | +0.032 |
needed | +0.031 |
earned | +0.031 |
Neg logit contributions | |
washing | -0.030 |
cloth | -0.026 |
sorting | -0.026 |
living | -0.026 |
mixing | -0.025 |
Token | unforgettable |
Token ID | 45504 |
Feature activation | -1.12e+01 |
Pos logit contributions | |
former | +0.097 |
needed | +0.096 |
ossible | +0.095 |
earned | +0.093 |
isdom | +0.092 |
Neg logit contributions | |
washing | -0.214 |
cloth | -0.204 |
sorting | -0.194 |
mixing | -0.190 |
dripping | -0.188 |
Token | experiences |
Token ID | 6461 |
Feature activation | -4.26e+00 |
Pos logit contributions | |
former | +0.243 |
ossible | +0.239 |
needed | +0.237 |
isdom | +0.235 |
earned | +0.233 |
Neg logit contributions | |
washing | -0.426 |
cloth | -0.395 |
sorting | -0.385 |
mixing | -0.373 |
living | -0.365 |
Token | . |
Token ID | 13 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
former | +0.154 |
needed | +0.154 |
ossible | +0.153 |
bye | +0.152 |
isdom | +0.152 |
Neg logit contributions | |
washing | -0.093 |
cloth | -0.086 |
sorting | -0.076 |
Mil | -0.074 |
mixing | -0.073 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
needed | +0.052 |
former | +0.052 |
isdom | +0.051 |
des | +0.051 |
nightmare | +0.050 |
Neg logit contributions | |
washing | -0.043 |
sorting | -0.040 |
cloth | -0.037 |
living | -0.037 |
mixing | -0.037 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
ossible | +0.062 |
former | +0.061 |
needed | +0.060 |
bye | +0.059 |
earned | +0.059 |
Neg logit contributions | |
washing | -0.092 |
cloth | -0.086 |
sorting | -0.083 |
mixing | -0.080 |
dripping | -0.079 |
Token | upon |
Token ID | 2402 |
Feature activation | -3.89e+00 |
Pos logit contributions | |
washing | +0.065 |
cloth | +0.061 |
sorting | +0.061 |
living | +0.058 |
mixing | +0.057 |
Neg logit contributions | |
former | -0.042 |
needed | -0.040 |
earned | -0.039 |
des | -0.039 |
isdom | -0.038 |
Token | to |
Token ID | 284 |
Feature activation | -5.64e+00 |
Pos logit contributions | |
former | +0.102 |
des | +0.098 |
isdom | +0.098 |
needed | +0.098 |
earned | +0.097 |
Neg logit contributions | |
washing | -0.059 |
cloth | -0.054 |
sorting | -0.051 |
living | -0.048 |
ers | -0.046 |
Token | be |
Token ID | 307 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
ossible | +0.179 |
needed | +0.179 |
isdom | +0.178 |
night | +0.177 |
earned | +0.176 |
Neg logit contributions | |
washing | -0.165 |
sorting | -0.145 |
cloth | -0.140 |
mixing | -0.135 |
living | -0.129 |
Token | restored |
Token ID | 15032 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
ossible | +0.029 |
former | +0.029 |
bye | +0.029 |
needed | +0.029 |
earned | +0.028 |
Neg logit contributions | |
washing | -0.028 |
cloth | -0.025 |
sorting | -0.025 |
mixing | -0.023 |
living | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
ossible | +0.036 |
needed | +0.036 |
earned | +0.036 |
former | +0.036 |
des | +0.036 |
Neg logit contributions | |
washing | -0.019 |
cloth | -0.017 |
sorting | -0.016 |
living | -0.015 |
mixing | -0.015 |
Token | its |
Token ID | 663 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
ossible | +0.176 |
needed | +0.173 |
earned | +0.171 |
des | +0.170 |
bye | +0.170 |
Neg logit contributions | |
washing | -0.144 |
sorting | -0.126 |
cloth | -0.122 |
mixing | -0.119 |
living | -0.117 |
Token | former |
Token ID | 1966 |
Feature activation | -9.69e+00 |
Pos logit contributions | |
des | +0.105 |
needed | +0.104 |
ossible | +0.103 |
earned | +0.102 |
isdom | +0.102 |
Neg logit contributions | |
washing | -0.145 |
sorting | -0.134 |
living | -0.130 |
cloth | -0.127 |
mixing | -0.125 |
Token | glory |
Token ID | 13476 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
former | +0.259 |
isdom | +0.242 |
ossible | +0.239 |
earned | +0.239 |
needed | +0.237 |
Neg logit contributions | |
washing | -0.332 |
cloth | -0.306 |
sorting | -0.293 |
mixing | -0.285 |
Mil | -0.280 |
Token | ." |
Token ID | 526 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
former | +0.112 |
ossible | +0.110 |
needed | +0.110 |
isdom | +0.109 |
earned | +0.109 |
Neg logit contributions | |
washing | -0.088 |
cloth | -0.078 |
sorting | -0.075 |
mixing | -0.072 |
dripping | -0.071 |
Token | |
Token ID | 220 |
Feature activation | +8.06e-01 |
Pos logit contributions | |
ossible | +0.089 |
former | +0.086 |
needed | +0.084 |
isdom | +0.082 |
des | +0.082 |
Neg logit contributions | |
washing | -0.116 |
cloth | -0.107 |
sorting | -0.103 |
mixing | -0.100 |
dripping | -0.099 |
Token |
|
Token ID | 198 |
Feature activation | -4.43e+00 |
Pos logit contributions | |
washing | +0.024 |
cloth | +0.021 |
sorting | +0.021 |
living | +0.020 |
Mil | +0.020 |
Neg logit contributions | |
former | -0.025 |
needed | -0.025 |
bye | -0.025 |
earned | -0.024 |
des | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
former | +0.114 |
needed | +0.111 |
ossible | +0.111 |
isdom | +0.110 |
earned | +0.110 |
Neg logit contributions | |
washing | -0.151 |
cloth | -0.139 |
sorting | -0.135 |
mixing | -0.130 |
living | -0.127 |
Token | smiled |
Token ID | 13541 |
Feature activation | +7.08e-02 |
Pos logit contributions | |
former | +0.058 |
ossible | +0.057 |
needed | +0.057 |
isdom | +0.056 |
earned | +0.056 |
Neg logit contributions | |
washing | -0.050 |
cloth | -0.045 |
sorting | -0.043 |
mixing | -0.042 |
living | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
washing | +0.002 |
cloth | +0.002 |
sorting | +0.002 |
mixing | +0.001 |
dripping | +0.001 |
Neg logit contributions | |
former | -0.002 |
ossible | -0.002 |
needed | -0.002 |
bye | -0.002 |
isdom | -0.002 |
Token | rolled |
Token ID | 11686 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
former | +0.072 |
ossible | +0.072 |
bye | +0.071 |
needed | +0.071 |
isdom | +0.070 |
Neg logit contributions | |
washing | -0.106 |
cloth | -0.101 |
sorting | -0.097 |
mixing | -0.093 |
dripping | -0.091 |
Token | away |
Token ID | 1497 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
ossible | +0.006 |
needed | +0.006 |
former | +0.006 |
bye | +0.006 |
earned | +0.006 |
Neg logit contributions | |
washing | -0.005 |
cloth | -0.005 |
sorting | -0.005 |
mixing | -0.005 |
dripping | -0.004 |
Token | content |
Token ID | 2695 |
Feature activation | -6.76e+00 |
Pos logit contributions | |
former | +0.085 |
prey | +0.082 |
tal | +0.078 |
needed | +0.078 |
chat | +0.078 |
Neg logit contributions | |
washing | -0.070 |
cloth | -0.066 |
racks | -0.063 |
cups | -0.059 |
sorting | -0.058 |
Token | edly |
Token ID | 49288 |
Feature activation | -9.49e+00 |
Pos logit contributions | |
bye | +0.220 |
former | +0.216 |
night | +0.206 |
prey | +0.202 |
des | +0.200 |
Neg logit contributions | |
washing | -0.187 |
cloth | -0.169 |
sorting | -0.161 |
mixing | -0.156 |
racks | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
former | +0.338 |
needed | +0.321 |
ossible | +0.320 |
earned | +0.316 |
prey | +0.312 |
Neg logit contributions | |
washing | -0.213 |
cloth | -0.195 |
Mil | -0.181 |
sorting | -0.174 |
racks | -0.167 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
former | +0.044 |
ossible | +0.044 |
needed | +0.043 |
isdom | +0.043 |
earned | +0.043 |
Neg logit contributions | |
washing | -0.050 |
cloth | -0.046 |
sorting | -0.044 |
mixing | -0.043 |
dripping | -0.042 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
ossible | +0.071 |
needed | +0.068 |
former | +0.068 |
bye | +0.067 |
earned | +0.066 |
Neg logit contributions | |
washing | -0.108 |
cloth | -0.100 |
sorting | -0.097 |
mixing | -0.094 |
dripping | -0.092 |
Token | upon |
Token ID | 2402 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
washing | +0.061 |
cloth | +0.057 |
sorting | +0.057 |
living | +0.054 |
mixing | +0.054 |
Neg logit contributions | |
former | -0.039 |
needed | -0.037 |
earned | -0.036 |
isdom | -0.036 |
des | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | -3.32e+00 |
Pos logit contributions | |
former | +0.190 |
isdom | +0.186 |
idents | +0.181 |
tal | +0.178 |
bye | +0.176 |
Neg logit contributions | |
washing | -0.067 |
living | -0.058 |
in | -0.049 |
cloth | -0.049 |
twins | -0.048 |
Token | was |
Token ID | 373 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
washing | +0.016 |
cloth | +0.014 |
sorting | +0.014 |
mixing | +0.013 |
living | +0.013 |
Neg logit contributions | |
former | -0.020 |
ossible | -0.019 |
needed | -0.019 |
des | -0.019 |
earned | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +8.17e-01 |
Pos logit contributions | |
former | +0.036 |
needed | +0.034 |
isdom | +0.034 |
ossible | +0.034 |
earned | +0.033 |
Neg logit contributions | |
washing | -0.031 |
cloth | -0.029 |
sorting | -0.027 |
racks | -0.026 |
Mil | -0.025 |
Token | brightest |
Token ID | 33871 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
washing | +0.022 |
cloth | +0.021 |
sorting | +0.020 |
mixing | +0.019 |
dripping | +0.018 |
Neg logit contributions | |
former | -0.026 |
needed | -0.025 |
ossible | -0.025 |
bye | -0.025 |
earned | -0.025 |
Token | it |
Token ID | 340 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
former | +0.080 |
needed | +0.075 |
encounter | +0.075 |
tal | +0.075 |
bye | +0.075 |
Neg logit contributions | |
washing | -0.064 |
cloth | -0.059 |
bling | -0.056 |
sorting | -0.056 |
bled | -0.055 |
Token | could |
Token ID | 714 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
former | +0.076 |
needed | +0.076 |
ossible | +0.076 |
earned | +0.075 |
isdom | +0.075 |
Neg logit contributions | |
washing | -0.056 |
cloth | -0.050 |
sorting | -0.047 |
mixing | -0.045 |
dripping | -0.043 |
Token | possibly |
Token ID | 5457 |
Feature activation | -9.33e+00 |
Pos logit contributions | |
ossible | +0.081 |
des | +0.080 |
isdom | +0.080 |
needed | +0.080 |
former | +0.079 |
Neg logit contributions | |
washing | -0.064 |
cloth | -0.057 |
sorting | -0.056 |
mixing | -0.054 |
living | -0.051 |
Token | be |
Token ID | 307 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ossible | +0.312 |
former | +0.307 |
isdom | +0.306 |
needed | +0.301 |
des | +0.297 |
Neg logit contributions | |
washing | -0.244 |
cloth | -0.221 |
sorting | -0.215 |
mixing | -0.207 |
living | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
former | +0.055 |
ossible | +0.054 |
needed | +0.054 |
isdom | +0.054 |
earned | +0.053 |
Neg logit contributions | |
washing | -0.034 |
cloth | -0.030 |
sorting | -0.029 |
mixing | -0.028 |
living | -0.026 |
Token | Ab |
Token ID | 2275 |
Feature activation | +4.59e+00 |
Pos logit contributions | |
former | +0.067 |
needed | +0.066 |
ossible | +0.066 |
isdom | +0.066 |
earned | +0.065 |
Neg logit contributions | |
washing | -0.068 |
cloth | -0.062 |
sorting | -0.060 |
mixing | -0.058 |
living | -0.056 |
Token | bi |
Token ID | 8482 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
washing | +0.175 |
cloth | +0.162 |
sorting | +0.158 |
mixing | +0.152 |
dripping | +0.149 |
Neg logit contributions | |
ossible | -0.100 |
former | -0.100 |
needed | -0.098 |
isdom | -0.097 |
earned | -0.097 |
Token | had |
Token ID | 550 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
former | +0.039 |
needed | +0.039 |
des | +0.038 |
earned | +0.038 |
idents | +0.038 |
Neg logit contributions | |
washing | -0.026 |
cloth | -0.024 |
sorting | -0.023 |
mixing | -0.022 |
living | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
former | +0.106 |
needed | +0.105 |
isdom | +0.104 |
bye | +0.103 |
earned | +0.103 |
Neg logit contributions | |
washing | -0.058 |
cloth | -0.052 |
sorting | -0.048 |
mixing | -0.046 |
Mil | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ossible | +0.096 |
former | +0.095 |
needed | +0.094 |
earned | +0.094 |
isdom | +0.093 |
Neg logit contributions | |
washing | -0.075 |
cloth | -0.066 |
sorting | -0.065 |
mixing | -0.062 |
dripping | -0.060 |
Token | bad |
Token ID | 2089 |
Feature activation | -4.09e-01 |
Pos logit contributions | |
washing | +0.043 |
cloth | +0.038 |
sorting | +0.037 |
mixing | +0.035 |
living | +0.034 |
Neg logit contributions | |
former | -0.055 |
ossible | -0.054 |
needed | -0.054 |
isdom | -0.054 |
earned | -0.053 |
Token | ending |
Token ID | 7464 |
Feature activation | -5.08e-02 |
Pos logit contributions | |
former | +0.012 |
needed | +0.012 |
ossible | +0.012 |
bye | +0.012 |
isdom | +0.012 |
Neg logit contributions | |
washing | -0.012 |
cloth | -0.011 |
sorting | -0.011 |
mixing | -0.010 |
dripping | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
former | +0.002 |
needed | +0.002 |
ossible | +0.002 |
isdom | +0.002 |
earned | +0.002 |
Neg logit contributions | |
washing | -0.001 |
cloth | -0.001 |
sorting | -0.001 |
mixing | -0.001 |
dripping | -0.001 |
Token | unforgettable |
Token ID | 45504 |
Feature activation | -9.30e+00 |
Pos logit contributions | |
ossible | +0.020 |
former | +0.020 |
needed | +0.020 |
earned | +0.020 |
isdom | +0.020 |
Neg logit contributions | |
washing | -0.016 |
cloth | -0.014 |
sorting | -0.013 |
mixing | -0.013 |
dripping | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
former | +0.341 |
bye | +0.339 |
needed | +0.329 |
ossible | +0.327 |
isdom | +0.327 |
Neg logit contributions | |
washing | -0.204 |
cloth | -0.192 |
sorting | -0.172 |
cups | -0.169 |
Mil | -0.164 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
needed | +0.044 |
former | +0.043 |
moon | +0.043 |
prey | +0.043 |
night | +0.043 |
Neg logit contributions | |
washing | -0.029 |
sorting | -0.025 |
living | -0.024 |
Sara | -0.024 |
Mil | -0.024 |
Token | One |
Token ID | 3198 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
former | +0.067 |
needed | +0.064 |
isdom | +0.064 |
earned | +0.063 |
ossible | +0.063 |
Neg logit contributions | |
washing | -0.077 |
cloth | -0.070 |
sorting | -0.068 |
mixing | -0.065 |
living | -0.065 |
Token | day |
Token ID | 1110 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
former | +0.124 |
idents | +0.122 |
isdom | +0.118 |
prey | +0.113 |
OP | +0.112 |
Neg logit contributions | |
washing | -0.089 |
living | -0.076 |
sorting | -0.076 |
mixing | -0.075 |
cloth | -0.071 |
Token | , |
Token ID | 11 |
Feature activation | +2.50e+00 |
Pos logit contributions | |
ossible | +0.103 |
needed | +0.098 |
bye | +0.098 |
isdom | +0.097 |
earned | +0.097 |
Neg logit contributions | |
washing | -0.051 |
cloth | -0.045 |
sorting | -0.041 |
mixing | -0.039 |
dripping | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
washing | +0.046 |
cloth | +0.040 |
sorting | +0.038 |
mixing | +0.036 |
living | +0.035 |
Neg logit contributions | |
ossible | -0.071 |
former | -0.071 |
needed | -0.071 |
earned | -0.070 |
des | -0.070 |
Token | village |
Token ID | 7404 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
washing | +0.000 |
cloth | +0.000 |
sorting | +0.000 |
mixing | +0.000 |
living | +0.000 |
Neg logit contributions | |
ossible | -0.000 |
former | -0.000 |
needed | -0.000 |
isdom | -0.000 |
earned | -0.000 |
Token | back |
Token ID | 736 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ossible | +0.043 |
former | +0.043 |
isdom | +0.042 |
needed | +0.042 |
oof | +0.042 |
Neg logit contributions | |
washing | -0.029 |
cloth | -0.026 |
sorting | -0.025 |
mixing | -0.023 |
living | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | -4.51e+00 |
Pos logit contributions | |
oof | +0.052 |
isdom | +0.052 |
ossible | +0.051 |
des | +0.051 |
idents | +0.051 |
Neg logit contributions | |
washing | -0.041 |
living | -0.036 |
cloth | -0.036 |
sorting | -0.035 |
mixing | -0.033 |
Token | its |
Token ID | 663 |
Feature activation | -3.48e+00 |
Pos logit contributions | |
ossible | +0.153 |
needed | +0.152 |
earned | +0.152 |
des | +0.152 |
bye | +0.149 |
Neg logit contributions | |
washing | -0.122 |
sorting | -0.105 |
cloth | -0.103 |
mixing | -0.100 |
living | -0.098 |
Token | former |
Token ID | 1966 |
Feature activation | -9.14e+00 |
Pos logit contributions | |
ossible | +0.093 |
des | +0.092 |
needed | +0.091 |
earned | +0.091 |
isdom | +0.090 |
Neg logit contributions | |
washing | -0.117 |
sorting | -0.106 |
living | -0.104 |
cloth | -0.103 |
mixing | -0.101 |
Token | glory |
Token ID | 13476 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
former | +0.241 |
isdom | +0.221 |
ossible | +0.217 |
earned | +0.217 |
tal | +0.216 |
Neg logit contributions | |
washing | -0.321 |
cloth | -0.295 |
sorting | -0.285 |
mixing | -0.276 |
Mil | -0.271 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
former | +0.086 |
ossible | +0.083 |
needed | +0.083 |
isdom | +0.082 |
earned | +0.081 |
Neg logit contributions | |
washing | -0.077 |
cloth | -0.069 |
sorting | -0.067 |
mixing | -0.063 |
dripping | -0.063 |
Token | |
Token ID | 220 |
Feature activation | +6.88e-01 |
Pos logit contributions | |
former | +0.100 |
needed | +0.098 |
earned | +0.096 |
isdom | +0.096 |
des | +0.095 |
Neg logit contributions | |
washing | -0.091 |
cloth | -0.080 |
sorting | -0.079 |
mixing | -0.076 |
living | -0.074 |
Token |
|
Token ID | 198 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
washing | +0.020 |
cloth | +0.017 |
sorting | +0.017 |
Mil | +0.017 |
mixing | +0.016 |
Neg logit contributions | |
needed | -0.022 |
former | -0.022 |
des | -0.022 |
bye | -0.021 |
earned | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
former | +0.150 |
earned | +0.147 |
isdom | +0.147 |
idents | +0.147 |
needed | +0.146 |
Neg logit contributions | |
washing | -0.122 |
sorting | -0.105 |
living | -0.104 |
cloth | -0.101 |
Mil | -0.101 |
Token | home |
Token ID | 1363 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
needed | +0.110 |
ossible | +0.109 |
des | +0.108 |
isdom | +0.107 |
oof | +0.106 |
Neg logit contributions | |
washing | -0.089 |
sorting | -0.079 |
cloth | -0.078 |
mixing | -0.074 |
living | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
former | +0.130 |
ossible | +0.128 |
needed | +0.128 |
isdom | +0.127 |
earned | +0.127 |
Neg logit contributions | |
washing | -0.073 |
cloth | -0.064 |
sorting | -0.061 |
mixing | -0.057 |
living | -0.055 |
Token | flew |
Token ID | 13112 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
former | +0.106 |
needed | +0.105 |
ossible | +0.104 |
earned | +0.103 |
isdom | +0.103 |
Neg logit contributions | |
washing | -0.100 |
cloth | -0.090 |
sorting | -0.087 |
mixing | -0.083 |
living | -0.081 |
Token | off |
Token ID | 572 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
washing | +0.004 |
cloth | +0.003 |
sorting | +0.003 |
mixing | +0.003 |
living | +0.003 |
Neg logit contributions | |
former | -0.003 |
ossible | -0.003 |
needed | -0.003 |
isdom | -0.003 |
earned | -0.003 |
Token | content |
Token ID | 2695 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
former | +0.040 |
ossible | +0.040 |
needed | +0.040 |
isdom | +0.039 |
earned | +0.039 |
Neg logit contributions | |
washing | -0.034 |
cloth | -0.030 |
sorting | -0.029 |
mixing | -0.028 |
living | -0.027 |
Token | edly |
Token ID | 49288 |
Feature activation | -9.07e+00 |
Pos logit contributions | |
former | +0.218 |
bye | +0.213 |
ossible | +0.212 |
needed | +0.212 |
earned | +0.210 |
Neg logit contributions | |
washing | -0.142 |
cloth | -0.127 |
sorting | -0.120 |
mixing | -0.114 |
dripping | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
former | +0.369 |
needed | +0.367 |
ossible | +0.367 |
isdom | +0.365 |
earned | +0.363 |
Neg logit contributions | |
washing | -0.170 |
cloth | -0.146 |
sorting | -0.139 |
mixing | -0.129 |
living | -0.121 |
Token | From |
Token ID | 3574 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
former | +0.020 |
needed | +0.019 |
ossible | +0.019 |
earned | +0.019 |
isdom | +0.019 |
Neg logit contributions | |
washing | -0.023 |
cloth | -0.021 |
sorting | -0.021 |
mixing | -0.020 |
living | -0.019 |
Token | then |
Token ID | 788 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
ossible | +0.031 |
former | +0.031 |
bye | +0.031 |
isdom | +0.031 |
needed | +0.031 |
Neg logit contributions | |
washing | -0.021 |
cloth | -0.019 |
sorting | -0.018 |
mixing | -0.017 |
cups | -0.016 |
Token | on |
Token ID | 319 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
ossible | +0.123 |
bye | +0.115 |
isdom | +0.115 |
needed | +0.114 |
earned | +0.113 |
Neg logit contributions | |
cloth | -0.052 |
washing | -0.051 |
Mil | -0.046 |
cups | -0.046 |
cabinets | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
washing | +0.014 |
cloth | +0.013 |
sorting | +0.012 |
mixing | +0.011 |
racks | +0.011 |
Neg logit contributions | |
ossible | -0.027 |
isdom | -0.027 |
former | -0.026 |
bye | -0.026 |
tal | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
washing | +0.010 |
cloth | +0.009 |
sorting | +0.009 |
mixing | +0.009 |
living | +0.008 |
Neg logit contributions | |
former | -0.015 |
needed | -0.015 |
des | -0.015 |
earned | -0.015 |
isdom | -0.015 |
Token | Kevin |
Token ID | 7939 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
former | +0.069 |
needed | +0.069 |
des | +0.067 |
earned | +0.067 |
ossible | +0.065 |
Neg logit contributions | |
washing | -0.079 |
cloth | -0.070 |
sorting | -0.070 |
mixing | -0.067 |
Mil | -0.066 |
Token | learnt |
Token ID | 26338 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
former | +0.036 |
needed | +0.035 |
earned | +0.035 |
isdom | +0.035 |
des | +0.035 |
Neg logit contributions | |
washing | -0.029 |
cloth | -0.026 |
sorting | -0.025 |
mixing | -0.024 |
living | -0.024 |
Token | an |
Token ID | 281 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
needed | +0.184 |
former | +0.180 |
des | +0.180 |
ossible | +0.179 |
isdom | +0.179 |
Neg logit contributions | |
washing | -0.125 |
sorting | -0.105 |
cloth | -0.103 |
living | -0.098 |
mixing | -0.097 |
Token | important |
Token ID | 1593 |
Feature activation | -6.28e+00 |
Pos logit contributions | |
ossible | +0.128 |
needed | +0.123 |
former | +0.123 |
des | +0.122 |
idents | +0.121 |
Neg logit contributions | |
washing | -0.136 |
sorting | -0.123 |
cloth | -0.119 |
mixing | -0.115 |
living | -0.113 |
Token | lesson |
Token ID | 11483 |
Feature activation | -9.00e+00 |
Pos logit contributions | |
former | +0.112 |
des | +0.106 |
ossible | +0.102 |
needed | +0.102 |
isdom | +0.101 |
Neg logit contributions | |
washing | -0.271 |
sorting | -0.252 |
mixing | -0.242 |
living | -0.238 |
cloth | -0.235 |
Token | that |
Token ID | 326 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
isdom | +0.300 |
bye | +0.299 |
ossible | +0.298 |
former | +0.298 |
earned | +0.297 |
Neg logit contributions | |
washing | -0.230 |
cloth | -0.209 |
sorting | -0.193 |
mixing | -0.189 |
cups | -0.185 |
Token | day |
Token ID | 1110 |
Feature activation | -4.68e+00 |
Pos logit contributions | |
former | +0.063 |
needed | +0.060 |
earned | +0.059 |
isdom | +0.059 |
ossible | +0.059 |
Neg logit contributions | |
washing | -0.070 |
sorting | -0.063 |
cloth | -0.062 |
mixing | -0.061 |
living | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
ossible | +0.168 |
isdom | +0.166 |
earned | +0.162 |
former | +0.162 |
needed | +0.161 |
Neg logit contributions | |
washing | -0.109 |
cloth | -0.099 |
sorting | -0.091 |
Mil | -0.089 |
mixing | -0.089 |
Token | Even |
Token ID | 3412 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
needed | +0.043 |
des | +0.042 |
former | +0.041 |
bye | +0.041 |
earned | +0.041 |
Neg logit contributions | |
washing | -0.035 |
cloth | -0.032 |
sorting | -0.031 |
mixing | -0.029 |
Mil | -0.029 |
Token | when |
Token ID | 618 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
washing | +0.017 |
cloth | +0.017 |
sorting | +0.015 |
mixing | +0.015 |
racks | +0.015 |
Neg logit contributions | |
former | -0.016 |
needed | -0.015 |
isdom | -0.015 |
faith | -0.015 |
ossible | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
former | +0.123 |
bye | +0.122 |
prey | +0.119 |
night | +0.117 |
needed | +0.117 |
Neg logit contributions | |
washing | -0.086 |
cloth | -0.078 |
sorting | -0.075 |
Mil | -0.073 |
mixing | -0.072 |
Token | He |
Token ID | 679 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
former | +0.059 |
needed | +0.058 |
isdom | +0.057 |
ossible | +0.057 |
earned | +0.057 |
Neg logit contributions | |
washing | -0.068 |
cloth | -0.062 |
sorting | -0.061 |
mixing | -0.058 |
living | -0.057 |
Token | lay |
Token ID | 3830 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
earned | +0.065 |
former | +0.065 |
isdom | +0.065 |
needed | +0.064 |
idents | +0.063 |
Neg logit contributions | |
washing | -0.041 |
cloth | -0.035 |
living | -0.034 |
sorting | -0.033 |
Mil | -0.033 |
Token | there |
Token ID | 612 |
Feature activation | -8.13e-01 |
Pos logit contributions | |
washing | +0.024 |
cloth | +0.022 |
sorting | +0.022 |
mixing | +0.021 |
racks | +0.020 |
Neg logit contributions | |
ossible | -0.029 |
bye | -0.027 |
needed | -0.027 |
night | -0.027 |
prey | -0.027 |
Token | content |
Token ID | 2695 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
former | +0.031 |
ossible | +0.031 |
isdom | +0.031 |
needed | +0.031 |
earned | +0.031 |
Neg logit contributions | |
washing | -0.018 |
cloth | -0.015 |
sorting | -0.015 |
mixing | -0.014 |
living | -0.013 |
Token | edly |
Token ID | 49288 |
Feature activation | -8.99e+00 |
Pos logit contributions | |
bye | +0.194 |
former | +0.189 |
ossible | +0.185 |
needed | +0.184 |
des | +0.183 |
Neg logit contributions | |
washing | -0.160 |
cloth | -0.146 |
sorting | -0.139 |
mixing | -0.132 |
dripping | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
former | +0.353 |
ossible | +0.346 |
needed | +0.346 |
earned | +0.339 |
isdom | +0.338 |
Neg logit contributions | |
washing | -0.174 |
cloth | -0.157 |
sorting | -0.142 |
Mil | -0.134 |
mixing | -0.131 |
Token |
|
Token ID | 198 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
former | +0.053 |
ossible | +0.052 |
needed | +0.052 |
isdom | +0.052 |
earned | +0.051 |
Neg logit contributions | |
washing | -0.068 |
cloth | -0.062 |
sorting | -0.060 |
mixing | -0.058 |
living | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -3.98e+00 |
Pos logit contributions | |
former | +0.178 |
idents | +0.169 |
isdom | +0.165 |
earned | +0.164 |
needed | +0.162 |
Neg logit contributions | |
washing | -0.099 |
living | -0.088 |
sorting | -0.085 |
holding | -0.081 |
cloth | -0.080 |
Token | The |
Token ID | 464 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
former | +0.094 |
ossible | +0.092 |
needed | +0.092 |
isdom | +0.091 |
earned | +0.090 |
Neg logit contributions | |
washing | -0.145 |
cloth | -0.134 |
sorting | -0.130 |
mixing | -0.126 |
living | -0.123 |
Token | new |
Token ID | 649 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
former | +0.013 |
ossible | +0.013 |
needed | +0.013 |
isdom | +0.013 |
earned | +0.013 |
Neg logit contributions | |
washing | -0.008 |
cloth | -0.007 |
sorting | -0.006 |
mixing | -0.006 |
living | -0.006 |
Token | little |
Token ID | 1310 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
former | +0.004 |
ossible | +0.004 |
needed | +0.004 |
bye | +0.004 |
earned | +0.004 |
Neg logit contributions | |
washing | -0.003 |
cloth | -0.003 |
sorting | -0.003 |
Mil | -0.003 |
dripping | -0.003 |
Token | girl |
Token ID | 2576 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
bye | +0.026 |
faith | +0.025 |
ossible | +0.025 |
night | +0.025 |
former | +0.024 |
Neg logit contributions | |
washing | -0.032 |
bling | -0.032 |
dripping | -0.031 |
bled | -0.030 |
cloth | -0.030 |
Token | learned |
Token ID | 4499 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
former | +0.058 |
ossible | +0.058 |
needed | +0.057 |
isdom | +0.057 |
bye | +0.057 |
Neg logit contributions | |
washing | -0.048 |
cloth | -0.044 |
sorting | -0.042 |
mixing | -0.039 |
dripping | -0.039 |
Token | an |
Token ID | 281 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
needed | +0.112 |
ossible | +0.111 |
former | +0.110 |
bye | +0.109 |
isdom | +0.109 |
Neg logit contributions | |
washing | -0.095 |
cloth | -0.082 |
sorting | -0.081 |
mixing | -0.076 |
living | -0.075 |
Token | important |
Token ID | 1593 |
Feature activation | -6.20e+00 |
Pos logit contributions | |
ossible | +0.112 |
former | +0.110 |
needed | +0.109 |
des | +0.107 |
bye | +0.106 |
Neg logit contributions | |
washing | -0.135 |
sorting | -0.122 |
cloth | -0.121 |
mixing | -0.115 |
living | -0.112 |
Token | lesson |
Token ID | 11483 |
Feature activation | -8.99e+00 |
Pos logit contributions | |
former | +0.106 |
des | +0.099 |
needed | +0.098 |
ossible | +0.097 |
isdom | +0.092 |
Neg logit contributions | |
washing | -0.275 |
sorting | -0.255 |
mixing | -0.243 |
cloth | -0.242 |
living | -0.240 |
Token | that |
Token ID | 326 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
bye | +0.271 |
isdom | +0.270 |
earned | +0.268 |
ossible | +0.259 |
needed | +0.257 |
Neg logit contributions | |
washing | -0.255 |
cloth | -0.238 |
cups | -0.222 |
mixing | -0.220 |
sorting | -0.219 |
Token | day |
Token ID | 1110 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
former | +0.073 |
needed | +0.069 |
earned | +0.068 |
ossible | +0.068 |
isdom | +0.068 |
Neg logit contributions | |
washing | -0.078 |
sorting | -0.069 |
cloth | -0.069 |
mixing | -0.067 |
living | -0.066 |
Token | � |
Token ID | 6184 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
ossible | +0.166 |
isdom | +0.164 |
bye | +0.161 |
earned | +0.160 |
former | +0.158 |
Neg logit contributions | |
washing | -0.123 |
cloth | -0.113 |
sorting | -0.106 |
cups | -0.105 |
mixing | -0.105 |
Token | � |
Token ID | 95 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
isdom | +0.076 |
former | +0.074 |
earned | +0.073 |
ossible | +0.073 |
bye | +0.071 |
Neg logit contributions | |
washing | -0.106 |
cloth | -0.102 |
cups | -0.096 |
living | -0.095 |
sorting | -0.095 |
Token | € |
Token ID | 26391 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
former | +0.025 |
isdom | +0.025 |
ossible | +0.025 |
earned | +0.025 |
needed | +0.024 |
Neg logit contributions | |
washing | -0.038 |
cloth | -0.036 |
sorting | -0.035 |
mixing | -0.034 |
cups | -0.033 |
Token | for |
Token ID | 329 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
former | +0.124 |
ossible | +0.124 |
needed | +0.123 |
isdom | +0.122 |
earned | +0.122 |
Neg logit contributions | |
washing | -0.081 |
cloth | -0.073 |
sorting | -0.068 |
mixing | -0.065 |
dripping | -0.063 |
Token | learning |
Token ID | 4673 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
faith | +0.087 |
isdom | +0.086 |
former | +0.084 |
revenge | +0.083 |
lasting | +0.082 |
Neg logit contributions | |
cloth | -0.070 |
cups | -0.064 |
washing | -0.064 |
Mil | -0.060 |
racks | -0.060 |
Token | such |
Token ID | 884 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
former | +0.087 |
ossible | +0.085 |
needed | +0.085 |
isdom | +0.085 |
earned | +0.084 |
Neg logit contributions | |
washing | -0.065 |
cloth | -0.059 |
sorting | -0.055 |
mixing | -0.053 |
dripping | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | -5.57e+00 |
Pos logit contributions | |
former | +0.069 |
ossible | +0.067 |
isdom | +0.066 |
needed | +0.066 |
earned | +0.065 |
Neg logit contributions | |
washing | -0.076 |
cloth | -0.071 |
sorting | -0.067 |
mixing | -0.065 |
dripping | -0.063 |
Token | valuable |
Token ID | 8119 |
Feature activation | -6.00e+00 |
Pos logit contributions | |
ossible | +0.156 |
former | +0.156 |
needed | +0.154 |
isdom | +0.153 |
earned | +0.152 |
Neg logit contributions | |
washing | -0.177 |
cloth | -0.160 |
sorting | -0.158 |
mixing | -0.151 |
living | -0.147 |
Token | lesson |
Token ID | 11483 |
Feature activation | -8.95e+00 |
Pos logit contributions | |
former | +0.107 |
needed | +0.104 |
ossible | +0.104 |
isdom | +0.103 |
earned | +0.103 |
Neg logit contributions | |
washing | -0.251 |
cloth | -0.236 |
sorting | -0.229 |
mixing | -0.223 |
dripping | -0.219 |
Token | . |
Token ID | 13 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
isdom | +0.278 |
earned | +0.275 |
bye | +0.273 |
ossible | +0.267 |
encounter | +0.266 |
Neg logit contributions | |
washing | -0.234 |
cloth | -0.215 |
cups | -0.202 |
dripping | -0.201 |
mixing | -0.198 |
Token | From |
Token ID | 3574 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
needed | +0.048 |
former | +0.048 |
earned | +0.047 |
des | +0.047 |
prey | +0.046 |
Neg logit contributions | |
washing | -0.043 |
cloth | -0.039 |
sorting | -0.038 |
living | -0.036 |
mixing | -0.035 |
Token | that |
Token ID | 326 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
ossible | +0.054 |
former | +0.054 |
isdom | +0.054 |
bye | +0.054 |
needed | +0.053 |
Neg logit contributions | |
washing | -0.036 |
cloth | -0.033 |
sorting | -0.031 |
mixing | -0.030 |
cups | -0.028 |
Token | day |
Token ID | 1110 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
washing | +0.007 |
sorting | +0.007 |
living | +0.007 |
mixing | +0.006 |
cloth | +0.006 |
Neg logit contributions | |
former | -0.009 |
earned | -0.008 |
needed | -0.008 |
des | -0.008 |
prey | -0.008 |
Token | forward |
Token ID | 2651 |
Feature activation | -5.42e-01 |
Pos logit contributions | |
ossible | +0.103 |
bye | +0.092 |
isdom | +0.090 |
needed | +0.090 |
uld | +0.090 |
Neg logit contributions | |
washing | -0.055 |
cloth | -0.051 |
cups | -0.049 |
Mil | -0.048 |
mixing | -0.047 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
former | +0.050 |
ossible | +0.048 |
needed | +0.048 |
faith | +0.048 |
bye | +0.048 |
Neg logit contributions | |
washing | -0.038 |
cloth | -0.034 |
sorting | -0.033 |
mixing | -0.032 |
dripping | -0.031 |
Token | had |
Token ID | 550 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
former | +0.094 |
ossible | +0.092 |
needed | +0.092 |
des | +0.090 |
isdom | +0.089 |
Neg logit contributions | |
washing | -0.067 |
cloth | -0.059 |
sorting | -0.058 |
living | -0.055 |
mixing | -0.055 |
Token | learned |
Token ID | 4499 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
former | +0.011 |
needed | +0.010 |
earned | +0.010 |
bye | +0.010 |
des | +0.010 |
Neg logit contributions | |
washing | -0.010 |
sorting | -0.009 |
cloth | -0.009 |
mixing | -0.009 |
living | -0.009 |
Token | an |
Token ID | 281 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
ossible | +0.132 |
needed | +0.132 |
former | +0.132 |
bye | +0.129 |
des | +0.129 |
Neg logit contributions | |
washing | -0.106 |
cloth | -0.092 |
sorting | -0.089 |
living | -0.086 |
mixing | -0.083 |
Token | important |
Token ID | 1593 |
Feature activation | -6.20e+00 |
Pos logit contributions | |
former | +0.112 |
ossible | +0.111 |
needed | +0.110 |
isdom | +0.108 |
earned | +0.107 |
Neg logit contributions | |
washing | -0.155 |
cloth | -0.142 |
sorting | -0.139 |
mixing | -0.134 |
living | -0.130 |
Token | lesson |
Token ID | 11483 |
Feature activation | -8.95e+00 |
Pos logit contributions | |
former | +0.105 |
ossible | +0.096 |
needed | +0.095 |
isdom | +0.093 |
tal | +0.091 |
Neg logit contributions | |
washing | -0.276 |
sorting | -0.254 |
cloth | -0.245 |
mixing | -0.244 |
living | -0.240 |
Token | : |
Token ID | 25 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
isdom | +0.299 |
earned | +0.296 |
bye | +0.296 |
ossible | +0.293 |
former | +0.292 |
Neg logit contributions | |
washing | -0.224 |
cloth | -0.204 |
sorting | -0.190 |
mixing | -0.188 |
cups | -0.186 |
Token | sometimes |
Token ID | 3360 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
former | +0.025 |
needed | +0.025 |
ossible | +0.025 |
earned | +0.024 |
des | +0.024 |
Neg logit contributions | |
washing | -0.025 |
cloth | -0.022 |
sorting | -0.022 |
mixing | -0.021 |
living | -0.020 |
Token | it |
Token ID | 340 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
former | +0.096 |
isdom | +0.094 |
ossible | +0.093 |
needed | +0.092 |
earned | +0.092 |
Neg logit contributions | |
washing | -0.067 |
cloth | -0.063 |
sorting | -0.057 |
Mil | -0.056 |
mixing | -0.055 |
Token | 's |
Token ID | 338 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
former | +0.051 |
needed | +0.050 |
des | +0.049 |
ossible | +0.049 |
prey | +0.049 |
Neg logit contributions | |
washing | -0.030 |
sorting | -0.026 |
living | -0.025 |
cloth | -0.025 |
mixing | -0.024 |
Token | more |
Token ID | 517 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
former | +0.035 |
ossible | +0.034 |
needed | +0.034 |
isdom | +0.033 |
earned | +0.033 |
Neg logit contributions | |
washing | -0.042 |
cloth | -0.039 |
sorting | -0.038 |
mixing | -0.036 |
Mil | -0.035 |
Token | the |
Token ID | 262 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
needed | +0.065 |
des | +0.065 |
earned | +0.064 |
bye | +0.063 |
former | +0.062 |
Neg logit contributions | |
washing | -0.035 |
sorting | -0.031 |
mixing | -0.027 |
living | -0.027 |
cloth | -0.026 |
Token | party |
Token ID | 2151 |
Feature activation | -5.97e-01 |
Pos logit contributions | |
ossible | +0.017 |
isdom | +0.017 |
earned | +0.017 |
needed | +0.017 |
des | +0.017 |
Neg logit contributions | |
washing | -0.013 |
sorting | -0.011 |
living | -0.011 |
mixing | -0.010 |
cloth | -0.010 |
Token | learned |
Token ID | 4499 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
former | +0.023 |
needed | +0.022 |
ossible | +0.022 |
isdom | +0.022 |
earned | +0.022 |
Neg logit contributions | |
washing | -0.013 |
cloth | -0.011 |
sorting | -0.011 |
mixing | -0.010 |
living | -0.010 |
Token | an |
Token ID | 281 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
needed | +0.154 |
ossible | +0.152 |
former | +0.150 |
bye | +0.149 |
earned | +0.149 |
Neg logit contributions | |
washing | -0.086 |
cloth | -0.071 |
sorting | -0.071 |
living | -0.067 |
mixing | -0.062 |
Token | important |
Token ID | 1593 |
Feature activation | -6.35e+00 |
Pos logit contributions | |
former | +0.108 |
ossible | +0.107 |
needed | +0.106 |
isdom | +0.104 |
earned | +0.104 |
Neg logit contributions | |
washing | -0.154 |
cloth | -0.142 |
sorting | -0.138 |
mixing | -0.133 |
living | -0.130 |
Token | lesson |
Token ID | 11483 |
Feature activation | -8.90e+00 |
Pos logit contributions | |
ossible | +0.104 |
former | +0.103 |
needed | +0.100 |
des | +0.098 |
isdom | +0.091 |
Neg logit contributions | |
washing | -0.279 |
sorting | -0.261 |
mixing | -0.249 |
living | -0.246 |
cloth | -0.244 |
Token | about |
Token ID | 546 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
isdom | +0.303 |
bye | +0.301 |
earned | +0.298 |
ossible | +0.293 |
former | +0.292 |
Neg logit contributions | |
washing | -0.223 |
cloth | -0.205 |
mixing | -0.188 |
sorting | -0.187 |
cups | -0.187 |
Token | helping |
Token ID | 5742 |
Feature activation | +5.04e-01 |
Pos logit contributions | |
former | +0.079 |
isdom | +0.078 |
earned | +0.074 |
faith | +0.073 |
ossible | +0.073 |
Neg logit contributions | |
cloth | -0.054 |
washing | -0.053 |
sorting | -0.044 |
Mil | -0.044 |
cups | -0.044 |
Token | others |
Token ID | 1854 |
Feature activation | -2.88e-01 |
Pos logit contributions | |
washing | +0.012 |
cloth | +0.011 |
sorting | +0.010 |
mixing | +0.010 |
Mil | +0.009 |
Neg logit contributions | |
former | -0.018 |
isdom | -0.018 |
earned | -0.018 |
needed | -0.018 |
ossible | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
former | +0.011 |
ossible | +0.011 |
needed | +0.011 |
isdom | +0.011 |
earned | +0.011 |
Neg logit contributions | |
washing | -0.006 |
cloth | -0.005 |
sorting | -0.005 |
mixing | -0.005 |
dripping | -0.005 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
needed | +0.050 |
earned | +0.048 |
des | +0.047 |
former | +0.047 |
prey | +0.047 |
Neg logit contributions | |
washing | -0.040 |
sorting | -0.035 |
living | -0.035 |
Mil | -0.033 |
cloth | -0.033 |
Token | an |
Token ID | 281 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
ossible | +0.052 |
former | +0.051 |
needed | +0.051 |
earned | +0.050 |
des | +0.050 |
Neg logit contributions | |
washing | -0.050 |
cloth | -0.045 |
sorting | -0.044 |
mixing | -0.042 |
living | -0.041 |
Token | easy |
Token ID | 2562 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
ossible | +0.083 |
former | +0.082 |
needed | +0.080 |
des | +0.079 |
isdom | +0.079 |
Neg logit contributions | |
washing | -0.090 |
cloth | -0.081 |
sorting | -0.081 |
mixing | -0.077 |
living | -0.074 |
Token | way |
Token ID | 835 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
former | +0.145 |
ossible | +0.143 |
needed | +0.142 |
isdom | +0.141 |
earned | +0.140 |
Neg logit contributions | |
washing | -0.184 |
cloth | -0.170 |
sorting | -0.164 |
mixing | -0.158 |
living | -0.154 |
Token | to |
Token ID | 284 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
former | +0.033 |
ossible | +0.032 |
needed | +0.032 |
isdom | +0.032 |
earned | +0.031 |
Neg logit contributions | |
washing | -0.035 |
cloth | -0.031 |
sorting | -0.031 |
mixing | -0.029 |
living | -0.029 |
Token | share |
Token ID | 2648 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ossible | +0.164 |
needed | +0.162 |
earned | +0.159 |
former | +0.159 |
des | +0.159 |
Neg logit contributions | |
washing | -0.160 |
cloth | -0.142 |
sorting | -0.140 |
mixing | -0.134 |
living | -0.129 |
Token | news |
Token ID | 1705 |
Feature activation | -8.85e+00 |
Pos logit contributions | |
washing | +0.027 |
cloth | +0.024 |
sorting | +0.023 |
mixing | +0.022 |
dripping | +0.021 |
Neg logit contributions | |
former | -0.035 |
isdom | -0.035 |
ossible | -0.035 |
needed | -0.034 |
earned | -0.034 |
Token | with |
Token ID | 351 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
former | +0.324 |
isdom | +0.317 |
bye | +0.316 |
earned | +0.316 |
needed | +0.316 |
Neg logit contributions | |
washing | -0.193 |
cloth | -0.169 |
dripping | -0.156 |
sorting | -0.155 |
mixing | -0.153 |
Token | other |
Token ID | 584 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
washing | +0.045 |
cloth | +0.041 |
sorting | +0.039 |
mixing | +0.037 |
dripping | +0.036 |
Neg logit contributions | |
former | -0.052 |
ossible | -0.051 |
needed | -0.051 |
isdom | -0.051 |
earned | -0.051 |
Token | people |
Token ID | 661 |
Feature activation | +5.98e-01 |
Pos logit contributions | |
former | +0.117 |
needed | +0.116 |
ossible | +0.116 |
isdom | +0.115 |
earned | +0.115 |
Neg logit contributions | |
washing | -0.064 |
cloth | -0.056 |
sorting | -0.052 |
mixing | -0.049 |
dripping | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
washing | +0.012 |
cloth | +0.011 |
sorting | +0.010 |
mixing | +0.010 |
dripping | +0.009 |
Neg logit contributions | |
former | -0.023 |
ossible | -0.023 |
needed | -0.023 |
isdom | -0.023 |
earned | -0.023 |
Token | |
Token ID | 220 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
former | +0.047 |
needed | +0.046 |
earned | +0.045 |
bye | +0.045 |
des | +0.045 |
Neg logit contributions | |
washing | -0.040 |
cloth | -0.036 |
sorting | -0.035 |
living | -0.033 |
mixing | -0.033 |
Token | They |
Token ID | 1119 |
Feature activation | +3.48e-01 |
Pos logit contributions | |
needed | +0.025 |
former | +0.025 |
des | +0.024 |
idents | +0.024 |
bye | +0.024 |
Neg logit contributions | |
washing | -0.022 |
living | -0.019 |
sorting | -0.019 |
cloth | -0.019 |
Mil | -0.019 |
Token | all |
Token ID | 477 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
washing | +0.008 |
sorting | +0.007 |
cloth | +0.007 |
living | +0.007 |
mixing | +0.006 |
Neg logit contributions | |
needed | -0.013 |
earned | -0.013 |
des | -0.013 |
former | -0.013 |
idents | -0.013 |
Token | cherished |
Token ID | 41238 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
former | +0.050 |
needed | +0.049 |
earned | +0.049 |
isdom | +0.048 |
idents | +0.048 |
Neg logit contributions | |
washing | -0.041 |
cloth | -0.037 |
sorting | -0.036 |
living | -0.035 |
mixing | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
former | +0.057 |
ossible | +0.056 |
needed | +0.056 |
earned | +0.055 |
isdom | +0.055 |
Neg logit contributions | |
washing | -0.037 |
cloth | -0.032 |
sorting | -0.031 |
mixing | -0.030 |
living | -0.028 |
Token | moral |
Token ID | 6573 |
Feature activation | -6.99e+00 |
Pos logit contributions | |
ossible | +0.040 |
former | +0.039 |
earned | +0.038 |
needed | +0.038 |
isdom | +0.038 |
Neg logit contributions | |
washing | -0.042 |
sorting | -0.036 |
living | -0.035 |
mixing | -0.035 |
cloth | -0.035 |
Token | lesson |
Token ID | 11483 |
Feature activation | -8.85e+00 |
Pos logit contributions | |
former | +0.277 |
idents | +0.268 |
isdom | +0.264 |
ossible | +0.263 |
needed | +0.259 |
Neg logit contributions | |
washing | -0.142 |
cloth | -0.126 |
living | -0.118 |
sorting | -0.114 |
ers | -0.113 |
Token | - |
Token ID | 532 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
former | +0.357 |
needed | +0.357 |
des | +0.353 |
ossible | +0.353 |
earned | +0.346 |
Neg logit contributions | |
washing | -0.170 |
sorting | -0.141 |
cloth | -0.140 |
living | -0.132 |
mixing | -0.129 |
Token | to |
Token ID | 284 |
Feature activation | -5.50e+00 |
Pos logit contributions | |
ossible | +0.084 |
former | +0.083 |
needed | +0.083 |
earned | +0.082 |
des | +0.081 |
Neg logit contributions | |
washing | -0.071 |
sorting | -0.062 |
cloth | -0.060 |
mixing | -0.060 |
living | -0.058 |
Token | always |
Token ID | 1464 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
former | +0.168 |
ossible | +0.165 |
needed | +0.165 |
isdom | +0.163 |
earned | +0.163 |
Neg logit contributions | |
washing | -0.160 |
cloth | -0.145 |
sorting | -0.140 |
mixing | -0.134 |
living | -0.129 |
Token | share |
Token ID | 2648 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
former | +0.096 |
ossible | +0.094 |
isdom | +0.094 |
earned | +0.091 |
bye | +0.091 |
Neg logit contributions | |
washing | -0.108 |
cloth | -0.102 |
sorting | -0.095 |
cups | -0.093 |
dripping | -0.092 |
Token | and |
Token ID | 290 |
Feature activation | -5.31e+00 |
Pos logit contributions | |
washing | +0.035 |
cloth | +0.030 |
sorting | +0.029 |
mixing | +0.027 |
living | +0.027 |
Neg logit contributions | |
former | -0.059 |
needed | -0.059 |
ossible | -0.059 |
bye | -0.058 |
des | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
ossible | +0.078 |
encounter | +0.077 |
former | +0.077 |
needed | +0.077 |
faith | +0.076 |
Neg logit contributions | |
washing | -0.049 |
cloth | -0.048 |
racks | -0.041 |
dripping | -0.040 |
cups | -0.039 |
Token | played |
Token ID | 2826 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
ossible | +0.116 |
chat | +0.114 |
bye | +0.113 |
needed | +0.113 |
to | +0.111 |
Neg logit contributions | |
cloth | -0.137 |
washing | -0.135 |
bling | -0.117 |
dripping | -0.117 |
sorting | -0.115 |
Token | all |
Token ID | 477 |
Feature activation | -3.86e+00 |
Pos logit contributions | |
prey | +0.024 |
chat | +0.023 |
ossible | +0.023 |
former | +0.022 |
needed | +0.022 |
Neg logit contributions | |
washing | -0.014 |
cloth | -0.014 |
dripping | -0.013 |
racks | -0.012 |
mixing | -0.012 |
Token | sorts |
Token ID | 10524 |
Feature activation | -5.88e+00 |
Pos logit contributions | |
bye | +0.067 |
ossible | +0.058 |
needed | +0.052 |
| | +0.050 |
earned | +0.050 |
Neg logit contributions | |
washing | -0.171 |
cloth | -0.166 |
cabinets | -0.160 |
racks | -0.158 |
dripping | -0.155 |
Token | of |
Token ID | 286 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
bye | +0.154 |
needed | +0.153 |
des | +0.149 |
earned | +0.149 |
night | +0.142 |
Neg logit contributions | |
washing | -0.190 |
cabinets | -0.176 |
cloth | -0.169 |
Mil | -0.168 |
mixing | -0.167 |
Token | exciting |
Token ID | 7895 |
Feature activation | -8.84e+00 |
Pos logit contributions | |
des | +0.031 |
adventures | +0.030 |
faith | +0.030 |
needed | +0.030 |
earned | +0.029 |
Neg logit contributions | |
cloth | -0.050 |
washing | -0.047 |
dripping | -0.045 |
bled | -0.044 |
Mil | -0.043 |
Token | games |
Token ID | 1830 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
earned | +0.238 |
des | +0.237 |
needed | +0.233 |
prey | +0.230 |
adventures | +0.228 |
Neg logit contributions | |
washing | -0.285 |
cloth | -0.281 |
dripping | -0.257 |
bled | -0.244 |
sorting | -0.241 |
Token | . |
Token ID | 13 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
prey | +0.159 |
moon | +0.158 |
bye | +0.158 |
needed | +0.156 |
ossible | +0.155 |
Neg logit contributions | |
washing | -0.123 |
cloth | -0.114 |
Mil | -0.105 |
dripping | -0.103 |
mixing | -0.101 |
Token | But |
Token ID | 887 |
Feature activation | -8.26e-01 |
Pos logit contributions | |
former | +0.055 |
needed | +0.054 |
ossible | +0.053 |
earned | +0.053 |
bye | +0.053 |
Neg logit contributions | |
washing | -0.074 |
cloth | -0.069 |
sorting | -0.065 |
mixing | -0.063 |
dripping | -0.062 |
Token | she |
Token ID | 673 |
Feature activation | -8.93e-01 |
Pos logit contributions | |
former | +0.028 |
ossible | +0.028 |
needed | +0.028 |
isdom | +0.028 |
earned | +0.028 |
Neg logit contributions | |
washing | -0.021 |
cloth | -0.019 |
sorting | -0.018 |
mixing | -0.017 |
living | -0.016 |
Token | soon |
Token ID | 2582 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
needed | +0.026 |
former | +0.025 |
ossible | +0.025 |
earned | +0.025 |
des | +0.025 |
Neg logit contributions | |
washing | -0.028 |
cloth | -0.026 |
sorting | -0.025 |
mixing | -0.024 |
living | -0.023 |
Token | have |
Token ID | 423 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
former | +0.148 |
needed | +0.147 |
ossible | +0.145 |
isdom | +0.144 |
des | +0.144 |
Neg logit contributions | |
washing | -0.145 |
cloth | -0.130 |
sorting | -0.128 |
mixing | -0.122 |
living | -0.117 |
Token | learnt |
Token ID | 26338 |
Feature activation | -5.49e+00 |
Pos logit contributions | |
former | +0.044 |
ossible | +0.043 |
needed | +0.043 |
isdom | +0.043 |
earned | +0.043 |
Neg logit contributions | |
washing | -0.051 |
cloth | -0.047 |
sorting | -0.045 |
mixing | -0.044 |
living | -0.042 |
Token | such |
Token ID | 884 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
ossible | +0.174 |
former | +0.173 |
needed | +0.172 |
bye | +0.170 |
isdom | +0.169 |
Neg logit contributions | |
washing | -0.157 |
cloth | -0.139 |
sorting | -0.135 |
living | -0.128 |
mixing | -0.126 |
Token | an |
Token ID | 281 |
Feature activation | -4.64e+00 |
Pos logit contributions | |
former | +0.064 |
ossible | +0.064 |
needed | +0.064 |
earned | +0.063 |
bye | +0.062 |
Neg logit contributions | |
washing | -0.068 |
cloth | -0.061 |
sorting | -0.060 |
mixing | -0.056 |
living | -0.056 |
Token | important |
Token ID | 1593 |
Feature activation | -6.23e+00 |
Pos logit contributions | |
ossible | +0.124 |
former | +0.121 |
needed | +0.118 |
idents | +0.117 |
isdom | +0.116 |
Neg logit contributions | |
washing | -0.153 |
cloth | -0.139 |
sorting | -0.139 |
mixing | -0.131 |
living | -0.128 |
Token | lesson |
Token ID | 11483 |
Feature activation | -8.83e+00 |
Pos logit contributions | |
ossible | +0.107 |
former | +0.102 |
needed | +0.101 |
isdom | +0.099 |
oof | +0.098 |
Neg logit contributions | |
washing | -0.271 |
sorting | -0.249 |
cloth | -0.244 |
living | -0.239 |
mixing | -0.238 |
Token | about |
Token ID | 546 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
isdom | +0.261 |
earned | +0.255 |
bye | +0.255 |
encounter | +0.244 |
shouldn | +0.243 |
Neg logit contributions | |
washing | -0.248 |
cloth | -0.231 |
cups | -0.221 |
mixing | -0.216 |
dripping | -0.215 |
Token | trusting |
Token ID | 33914 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
former | +0.080 |
isdom | +0.079 |
ossible | +0.077 |
needed | +0.077 |
earned | +0.076 |
Neg logit contributions | |
washing | -0.048 |
cloth | -0.046 |
sorting | -0.040 |
cups | -0.038 |
Mil | -0.038 |
Token | people |
Token ID | 661 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
isdom | +0.054 |
former | +0.054 |
earned | +0.053 |
needed | +0.053 |
bye | +0.053 |
Neg logit contributions | |
washing | -0.044 |
cloth | -0.039 |
sorting | -0.038 |
mixing | -0.037 |
Mil | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
washing | +0.005 |
cloth | +0.004 |
sorting | +0.004 |
mixing | +0.004 |
living | +0.003 |
Neg logit contributions | |
former | -0.008 |
ossible | -0.008 |
needed | -0.008 |
isdom | -0.008 |
earned | -0.008 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
former | +0.034 |
needed | +0.033 |
earned | +0.033 |
des | +0.032 |
isdom | +0.032 |
Neg logit contributions | |
washing | -0.037 |
cloth | -0.034 |
sorting | -0.033 |
mixing | -0.031 |
living | -0.031 |
Token | his |
Token ID | 465 |
Feature activation | -4.62e+00 |
Pos logit contributions | |
former | +0.070 |
isdom | +0.069 |
ossible | +0.069 |
needed | +0.067 |
earned | +0.067 |
Neg logit contributions | |
washing | -0.071 |
cloth | -0.065 |
sorting | -0.062 |
mixing | -0.060 |
dripping | -0.060 |
Token | son |
Token ID | 3367 |
Feature activation | -5.87e+00 |
Pos logit contributions | |
former | +0.125 |
ossible | +0.121 |
needed | +0.120 |
oof | +0.120 |
isdom | +0.120 |
Neg logit contributions | |
washing | -0.156 |
sorting | -0.142 |
cloth | -0.139 |
mixing | -0.136 |
living | -0.132 |
Token | such |
Token ID | 884 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
former | +0.206 |
needed | +0.200 |
ossible | +0.199 |
bye | +0.198 |
isdom | +0.197 |
Neg logit contributions | |
washing | -0.148 |
cloth | -0.133 |
sorting | -0.129 |
mixing | -0.121 |
living | -0.119 |
Token | an |
Token ID | 281 |
Feature activation | -4.84e+00 |
Pos logit contributions | |
needed | +0.067 |
ossible | +0.066 |
bye | +0.066 |
des | +0.065 |
earned | +0.065 |
Neg logit contributions | |
washing | -0.063 |
sorting | -0.057 |
cloth | -0.055 |
mixing | -0.052 |
living | -0.052 |
Token | important |
Token ID | 1593 |
Feature activation | -6.28e+00 |
Pos logit contributions | |
ossible | +0.132 |
former | +0.127 |
needed | +0.124 |
idents | +0.123 |
bye | +0.121 |
Neg logit contributions | |
washing | -0.159 |
sorting | -0.146 |
cloth | -0.140 |
mixing | -0.137 |
living | -0.132 |
Token | lesson |
Token ID | 11483 |
Feature activation | -8.80e+00 |
Pos logit contributions | |
ossible | +0.126 |
former | +0.125 |
needed | +0.124 |
des | +0.118 |
oof | +0.118 |
Neg logit contributions | |
washing | -0.254 |
sorting | -0.234 |
mixing | -0.223 |
cloth | -0.223 |
living | -0.219 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
isdom | +0.276 |
bye | +0.273 |
earned | +0.273 |
prey | +0.268 |
ossible | +0.265 |
Neg logit contributions | |
washing | -0.229 |
cloth | -0.211 |
cups | -0.198 |
dripping | -0.196 |
mixing | -0.193 |
Token | He |
Token ID | 679 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
former | +0.043 |
needed | +0.043 |
ossible | +0.042 |
isdom | +0.042 |
earned | +0.042 |
Neg logit contributions | |
washing | -0.049 |
cloth | -0.045 |
sorting | -0.044 |
mixing | -0.042 |
living | -0.041 |
Token | smiled |
Token ID | 13541 |
Feature activation | +5.58e-01 |
Pos logit contributions | |
needed | +0.062 |
earned | +0.060 |
des | +0.060 |
former | +0.059 |
idents | +0.058 |
Neg logit contributions | |
washing | -0.042 |
cloth | -0.037 |
sorting | -0.036 |
living | -0.036 |
mixing | -0.034 |
Token | as |
Token ID | 355 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
washing | +0.013 |
cloth | +0.012 |
sorting | +0.011 |
mixing | +0.010 |
living | +0.010 |
Neg logit contributions | |
former | -0.020 |
needed | -0.020 |
ossible | -0.020 |
earned | -0.020 |
isdom | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
isdom | +0.029 |
former | +0.029 |
bye | +0.028 |
needed | +0.028 |
earned | +0.028 |
Neg logit contributions | |
washing | -0.029 |
cloth | -0.027 |
sorting | -0.025 |
mixing | -0.025 |
cups | -0.025 |
Token | We |
Token ID | 775 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
former | +0.021 |
ossible | +0.020 |
isdom | +0.020 |
bye | +0.020 |
needed | +0.020 |
Neg logit contributions | |
washing | -0.023 |
cloth | -0.022 |
sorting | -0.021 |
mixing | -0.020 |
dripping | -0.020 |
Token | should |
Token ID | 815 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
washing | +0.001 |
cloth | +0.001 |
sorting | +0.001 |
mixing | +0.000 |
living | +0.000 |
Neg logit contributions | |
earned | -0.001 |
isdom | -0.001 |
former | -0.001 |
needed | -0.001 |
des | -0.001 |
Token | do |
Token ID | 466 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
former | +0.117 |
ossible | +0.115 |
bye | +0.112 |
isdom | +0.111 |
des | +0.108 |
Neg logit contributions | |
washing | -0.133 |
cloth | -0.124 |
sorting | -0.117 |
mixing | -0.115 |
cups | -0.114 |
Token | it |
Token ID | 340 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
former | +0.098 |
ossible | +0.096 |
isdom | +0.095 |
needed | +0.095 |
earned | +0.094 |
Neg logit contributions | |
washing | -0.077 |
cloth | -0.071 |
sorting | -0.066 |
mixing | -0.063 |
dripping | -0.062 |
Token | again |
Token ID | 757 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
ossible | +0.122 |
isdom | +0.120 |
needed | +0.120 |
earned | +0.120 |
former | +0.119 |
Neg logit contributions | |
washing | -0.091 |
sorting | -0.080 |
cloth | -0.079 |
mixing | -0.076 |
living | -0.074 |
Token | sometime |
Token ID | 17291 |
Feature activation | -8.79e+00 |
Pos logit contributions | |
needed | +0.113 |
ossible | +0.112 |
earned | +0.111 |
des | +0.111 |
former | +0.110 |
Neg logit contributions | |
washing | -0.130 |
cloth | -0.116 |
sorting | -0.115 |
mixing | -0.112 |
living | -0.109 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
former | +0.227 |
bye | +0.220 |
ossible | +0.217 |
isdom | +0.213 |
oof | +0.207 |
Neg logit contributions | |
washing | -0.297 |
cloth | -0.284 |
sorting | -0.268 |
dripping | -0.264 |
mixing | -0.258 |
Token |
|
Token ID | 198 |
Feature activation | -4.77e+00 |
Pos logit contributions | |
former | +0.121 |
needed | +0.118 |
earned | +0.118 |
des | +0.115 |
bye | +0.113 |
Neg logit contributions | |
washing | -0.108 |
sorting | -0.092 |
mixing | -0.088 |
cloth | -0.088 |
Mil | -0.085 |
Token |
|
Token ID | 198 |
Feature activation | -4.20e+00 |
Pos logit contributions | |
former | +0.188 |
earned | +0.181 |
des | +0.178 |
needed | +0.177 |
bye | +0.176 |
Neg logit contributions | |
washing | -0.121 |
sorting | -0.097 |
living | -0.094 |
mixing | -0.091 |
Mil | -0.087 |
Token | And |
Token ID | 1870 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
former | +0.143 |
night | +0.140 |
earned | +0.138 |
nightmare | +0.138 |
isdom | +0.137 |
Neg logit contributions | |
washing | -0.119 |
sorting | -0.100 |
mixing | -0.097 |
cloth | -0.096 |
living | -0.090 |
Token | so |
Token ID | 523 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
former | +0.087 |
needed | +0.086 |
ossible | +0.086 |
isdom | +0.085 |
earned | +0.085 |
Neg logit contributions | |
washing | -0.063 |
cloth | -0.056 |
sorting | -0.054 |
mixing | -0.051 |
living | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
former | +0.070 |
needed | +0.070 |
earned | +0.069 |
des | +0.069 |
idents | +0.068 |
Neg logit contributions | |
washing | -0.049 |
sorting | -0.042 |
living | -0.040 |
mixing | -0.040 |
cloth | -0.039 |
Token | zoo |
Token ID | 26626 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
washing | +0.068 |
cloth | +0.064 |
sorting | +0.060 |
mixing | +0.057 |
dripping | +0.055 |
Neg logit contributions | |
former | -0.077 |
ossible | -0.075 |
needed | -0.075 |
bye | -0.074 |
isdom | -0.073 |
Token | ," |
Token ID | 553 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
washing | +0.050 |
cloth | +0.045 |
sorting | +0.043 |
mixing | +0.041 |
dripping | +0.039 |
Neg logit contributions | |
former | -0.064 |
ossible | -0.064 |
needed | -0.064 |
isdom | -0.063 |
bye | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
ossible | +0.023 |
former | +0.022 |
needed | +0.022 |
isdom | +0.022 |
bye | +0.022 |
Neg logit contributions | |
washing | -0.017 |
cloth | -0.015 |
sorting | -0.014 |
mixing | -0.014 |
dripping | -0.014 |
Token | zo |
Token ID | 40565 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
washing | +0.083 |
cloth | +0.080 |
sorting | +0.073 |
mixing | +0.069 |
dripping | +0.067 |
Neg logit contributions | |
bye | -0.090 |
former | -0.089 |
ossible | -0.089 |
needed | -0.086 |
prey | -0.086 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.27e+00 |
Pos logit contributions | |
washing | +0.098 |
cloth | +0.082 |
sorting | +0.079 |
mixing | +0.076 |
dripping | +0.074 |
Neg logit contributions | |
ossible | -0.147 |
bye | -0.143 |
isdom | -0.142 |
needed | -0.141 |
earned | -0.140 |
Token | eper |
Token ID | 5723 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
washing | +0.282 |
sorting | +0.235 |
cloth | +0.235 |
mixing | +0.220 |
dripping | +0.217 |
Neg logit contributions | |
ossible | -0.317 |
bye | -0.293 |
oof | -0.283 |
isdom | -0.283 |
Dear | -0.279 |
Token | says |
Token ID | 1139 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
washing | +0.042 |
cloth | +0.037 |
sorting | +0.037 |
dripping | +0.036 |
mixing | +0.035 |
Neg logit contributions | |
ossible | -0.021 |
know | -0.018 |
isdom | -0.018 |
needed | -0.017 |
des | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
ossible | +0.025 |
to | +0.024 |
night | +0.024 |
des | +0.024 |
understand | +0.024 |
Neg logit contributions | |
washing | -0.027 |
racks | -0.023 |
cloth | -0.023 |
dripping | -0.023 |
cabinets | -0.022 |
Token | " |
Token ID | 366 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
washing | +0.047 |
cloth | +0.046 |
sorting | +0.042 |
dripping | +0.042 |
mixing | +0.041 |
Neg logit contributions | |
ossible | -0.046 |
Dear | -0.044 |
prey | -0.043 |
Hunter | -0.043 |
faith | -0.043 |
Token | Don |
Token ID | 3987 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
washing | +0.066 |
cloth | +0.059 |
sorting | +0.057 |
mixing | +0.055 |
living | +0.053 |
Neg logit contributions | |
former | -0.076 |
ossible | -0.075 |
needed | -0.075 |
isdom | -0.074 |
earned | -0.074 |
Token | L |
Token ID | 43 |
Feature activation | +5.64e+00 |
Pos logit contributions | |
washing | +0.133 |
cloth | +0.120 |
sorting | +0.118 |
mixing | +0.112 |
dripping | +0.108 |
Neg logit contributions | |
ossible | -0.149 |
former | -0.148 |
needed | -0.148 |
earned | -0.145 |
bye | -0.145 |
Token | ily |
Token ID | 813 |
Feature activation | +4.39e+00 |
Pos logit contributions | |
washing | +0.095 |
sorting | +0.073 |
cloth | +0.071 |
mixing | +0.070 |
dripping | +0.065 |
Neg logit contributions | |
ossible | -0.244 |
isdom | -0.242 |
des | -0.242 |
oof | -0.240 |
needed | -0.239 |
Token | and |
Token ID | 290 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
washing | +0.095 |
cloth | +0.083 |
sorting | +0.080 |
mixing | +0.074 |
dripping | +0.072 |
Neg logit contributions | |
ossible | -0.165 |
bye | -0.160 |
isdom | -0.160 |
tal | -0.156 |
idents | -0.156 |
Token | Ben |
Token ID | 3932 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
washing | +0.034 |
cloth | +0.030 |
sorting | +0.028 |
mixing | +0.027 |
dripping | +0.025 |
Neg logit contributions | |
ossible | -0.068 |
isdom | -0.066 |
former | -0.066 |
bye | -0.065 |
needed | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | +7.46e+00 |
Pos logit contributions | |
washing | +0.082 |
cloth | +0.071 |
sorting | +0.067 |
mixing | +0.062 |
dripping | +0.059 |
Neg logit contributions | |
ossible | -0.159 |
former | -0.159 |
needed | -0.157 |
isdom | -0.157 |
bye | -0.156 |
Token | why |
Token ID | 1521 |
Feature activation | +8.29e+00 |
Pos logit contributions | |
washing | +0.211 |
cloth | +0.195 |
sorting | +0.190 |
mixing | +0.186 |
ers | +0.182 |
Neg logit contributions | |
ossible | -0.219 |
needed | -0.210 |
idents | -0.206 |
nightmare | -0.203 |
des | -0.202 |
Token | did |
Token ID | 750 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
washing | +0.263 |
cloth | +0.238 |
sorting | +0.233 |
mixing | +0.224 |
dripping | +0.218 |
Neg logit contributions | |
ossible | -0.231 |
isdom | -0.218 |
idents | -0.215 |
bye | -0.214 |
des | -0.212 |
Token | you |
Token ID | 345 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
washing | +0.043 |
cloth | +0.038 |
sorting | +0.037 |
mixing | +0.036 |
Mil | +0.035 |
Neg logit contributions | |
former | -0.048 |
needed | -0.046 |
isdom | -0.045 |
earned | -0.045 |
des | -0.045 |
Token | eat |
Token ID | 4483 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
former | +0.039 |
earned | +0.039 |
needed | +0.039 |
des | +0.039 |
bye | +0.039 |
Neg logit contributions | |
washing | -0.030 |
cloth | -0.026 |
sorting | -0.025 |
mixing | -0.024 |
dripping | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
washing | +0.034 |
cloth | +0.031 |
sorting | +0.030 |
mixing | +0.028 |
dripping | +0.027 |
Neg logit contributions | |
ossible | -0.042 |
former | -0.041 |
needed | -0.041 |
isdom | -0.041 |
earned | -0.040 |
Token | spaghetti |
Token ID | 48241 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
washing | +0.055 |
cloth | +0.049 |
sorting | +0.048 |
mixing | +0.045 |
Mil | +0.045 |
Neg logit contributions | |
ossible | -0.072 |
former | -0.071 |
needed | -0.071 |
earned | -0.069 |
bye | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | +8.17e-02 |
Pos logit contributions | |
washing | +0.081 |
Mil | +0.070 |
sorting | +0.069 |
dripping | +0.068 |
mixing | +0.068 |
Neg logit contributions | |
ossible | -0.072 |
bye | -0.065 |
moon | -0.065 |
needed | -0.064 |
night | -0.062 |
Token | " |
Token ID | 1 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
washing | +0.002 |
cloth | +0.002 |
living | +0.002 |
sorting | +0.002 |
Mil | +0.002 |
Neg logit contributions | |
idents | -0.003 |
former | -0.003 |
night | -0.003 |
earned | -0.003 |
revenge | -0.003 |
Token | Sorry |
Token ID | 14385 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
washing | +0.141 |
cloth | +0.127 |
mixing | +0.117 |
Mil | +0.117 |
sorting | +0.116 |
Neg logit contributions | |
former | -0.185 |
tal | -0.184 |
isdom | -0.184 |
idents | -0.179 |
earned | -0.178 |
Token | , |
Token ID | 11 |
Feature activation | +7.87e+00 |
Pos logit contributions | |
washing | +0.051 |
cloth | +0.045 |
sorting | +0.042 |
dripping | +0.040 |
mixing | +0.039 |
Neg logit contributions | |
ossible | -0.093 |
isdom | -0.092 |
former | -0.092 |
bye | -0.091 |
needed | -0.091 |
Token | Mom |
Token ID | 11254 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
washing | +0.198 |
cloth | +0.180 |
sorting | +0.171 |
mixing | +0.163 |
dripping | +0.158 |
Neg logit contributions | |
former | -0.262 |
ossible | -0.260 |
isdom | -0.256 |
needed | -0.254 |
bye | -0.253 |
Token | , |
Token ID | 11 |
Feature activation | +9.25e+00 |
Pos logit contributions | |
washing | +0.112 |
cloth | +0.102 |
sorting | +0.098 |
Mil | +0.091 |
mixing | +0.091 |
Neg logit contributions | |
ossible | -0.127 |
isdom | -0.124 |
former | -0.123 |
bye | -0.123 |
needed | -0.122 |
Token | we |
Token ID | 356 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
washing | +0.264 |
cloth | +0.248 |
sorting | +0.233 |
mixing | +0.222 |
ers | +0.221 |
Neg logit contributions | |
ossible | -0.273 |
needed | -0.265 |
former | -0.264 |
isdom | -0.261 |
idents | -0.258 |
Token | were |
Token ID | 547 |
Feature activation | +4.82e+00 |
Pos logit contributions | |
washing | +0.085 |
cloth | +0.077 |
sorting | +0.075 |
mixing | +0.074 |
living | +0.070 |
Neg logit contributions | |
earned | -0.117 |
needed | -0.113 |
former | -0.112 |
prey | -0.110 |
des | -0.109 |
Token | just |
Token ID | 655 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
cloth | +0.136 |
washing | +0.134 |
racks | +0.118 |
Mil | +0.118 |
sorting | +0.116 |
Neg logit contributions | |
former | -0.139 |
ossible | -0.135 |
nightmare | -0.133 |
isdom | -0.133 |
oof | -0.132 |
Token | playing |
Token ID | 2712 |
Feature activation | +4.40e+00 |
Pos logit contributions | |
washing | +0.033 |
cloth | +0.030 |
sorting | +0.028 |
mixing | +0.026 |
living | +0.025 |
Neg logit contributions | |
former | -0.054 |
ossible | -0.053 |
needed | -0.053 |
isdom | -0.053 |
earned | -0.052 |
Token | ," |
Token ID | 553 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
washing | +0.100 |
cloth | +0.089 |
sorting | +0.084 |
mixing | +0.079 |
living | +0.077 |
Neg logit contributions | |
former | -0.166 |
earned | -0.166 |
needed | -0.166 |
isdom | -0.164 |
ossible | -0.162 |
Token |
|
Token ID | 198 |
Feature activation | +8.17e-02 |
Pos logit contributions | |
washing | +0.068 |
cloth | +0.063 |
Mil | +0.052 |
mixing | +0.052 |
sticking | +0.051 |
Neg logit contributions | |
former | -0.121 |
isdom | -0.117 |
encounter | -0.114 |
tal | -0.112 |
nightmare | -0.111 |
Token | " |
Token ID | 1 |
Feature activation | +5.43e+00 |
Pos logit contributions | |
washing | +0.002 |
cloth | +0.002 |
living | +0.002 |
sorting | +0.002 |
Mil | +0.002 |
Neg logit contributions | |
idents | -0.003 |
former | -0.003 |
night | -0.003 |
earned | -0.003 |
revenge | -0.003 |
Token | Mom |
Token ID | 29252 |
Feature activation | +6.54e+00 |
Pos logit contributions | |
washing | +0.141 |
cloth | +0.127 |
mixing | +0.117 |
Mil | +0.117 |
sorting | +0.116 |
Neg logit contributions | |
former | -0.185 |
tal | -0.184 |
isdom | -0.184 |
idents | -0.179 |
earned | -0.178 |
Token | , |
Token ID | 11 |
Feature activation | +1.06e+01 |
Pos logit contributions | |
washing | +0.153 |
cloth | +0.137 |
sorting | +0.130 |
mixing | +0.121 |
dripping | +0.118 |
Neg logit contributions | |
ossible | -0.236 |
isdom | -0.230 |
former | -0.230 |
needed | -0.228 |
bye | -0.228 |
Token | dad |
Token ID | 9955 |
Feature activation | +3.83e+00 |
Pos logit contributions | |
washing | +0.291 |
cloth | +0.265 |
sorting | +0.254 |
mixing | +0.244 |
dripping | +0.236 |
Neg logit contributions | |
ossible | -0.334 |
needed | -0.328 |
former | -0.328 |
isdom | -0.324 |
earned | -0.323 |
Token | , |
Token ID | 11 |
Feature activation | +9.64e+00 |
Pos logit contributions | |
washing | +0.088 |
cloth | +0.078 |
sorting | +0.074 |
dripping | +0.069 |
mixing | +0.068 |
Neg logit contributions | |
ossible | -0.140 |
isdom | -0.138 |
bye | -0.137 |
oof | -0.133 |
needed | -0.132 |
Token | can |
Token ID | 460 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
washing | +0.271 |
cloth | +0.244 |
sorting | +0.235 |
mixing | +0.225 |
living | +0.218 |
Neg logit contributions | |
former | -0.310 |
needed | -0.301 |
ossible | -0.301 |
isdom | -0.300 |
earned | -0.298 |
Token | I |
Token ID | 314 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
washing | +0.091 |
cloth | +0.084 |
sorting | +0.080 |
mixing | +0.077 |
dripping | +0.074 |
Neg logit contributions | |
ossible | -0.097 |
former | -0.096 |
needed | -0.095 |
isdom | -0.095 |
earned | -0.094 |
Token | go |
Token ID | 467 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
washing | +0.050 |
cloth | +0.045 |
sorting | +0.043 |
mixing | +0.041 |
living | +0.039 |
Neg logit contributions | |
isdom | -0.063 |
former | -0.062 |
tal | -0.060 |
prey | -0.059 |
earned | -0.059 |
Token | back |
Token ID | 736 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
washing | +0.117 |
cloth | +0.107 |
sorting | +0.104 |
mixing | +0.100 |
living | +0.097 |
Neg logit contributions | |
former | -0.104 |
isdom | -0.101 |
needed | -0.100 |
ossible | -0.100 |
earned | -0.099 |
Token | to |
Token ID | 284 |
Feature activation | -8.68e-01 |
Pos logit contributions | |
washing | +0.048 |
living | +0.042 |
sorting | +0.040 |
cloth | +0.039 |
in | +0.039 |
Neg logit contributions | |
isdom | -0.064 |
former | -0.062 |
idents | -0.062 |
earned | -0.062 |
needed | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
washing | +0.106 |
cloth | +0.095 |
sorting | +0.091 |
mixing | +0.085 |
dripping | +0.083 |
Neg logit contributions | |
former | -0.158 |
ossible | -0.157 |
needed | -0.155 |
bye | -0.154 |
isdom | -0.154 |
Token |
|
Token ID | 198 |
Feature activation | +7.82e-01 |
Pos logit contributions | |
washing | +0.111 |
cloth | +0.101 |
sorting | +0.097 |
mixing | +0.092 |
Mil | +0.092 |
Neg logit contributions | |
former | -0.130 |
needed | -0.129 |
earned | -0.126 |
isdom | -0.125 |
des | -0.125 |
Token |
|
Token ID | 198 |
Feature activation | +6.66e-01 |
Pos logit contributions | |
washing | +0.022 |
cloth | +0.020 |
living | +0.019 |
sorting | +0.019 |
Mil | +0.018 |
Neg logit contributions | |
former | -0.025 |
idents | -0.025 |
earned | -0.025 |
needed | -0.024 |
night | -0.024 |
Token | " |
Token ID | 1 |
Feature activation | +5.44e+00 |
Pos logit contributions | |
washing | +0.016 |
cloth | +0.014 |
sorting | +0.013 |
living | +0.012 |
mixing | +0.012 |
Neg logit contributions | |
former | -0.024 |
needed | -0.024 |
isdom | -0.023 |
idents | -0.023 |
earned | -0.023 |
Token | Come |
Token ID | 16773 |
Feature activation | +5.65e+00 |
Pos logit contributions | |
washing | +0.157 |
cloth | +0.142 |
sorting | +0.133 |
mixing | +0.131 |
Mil | +0.127 |
Neg logit contributions | |
former | -0.171 |
isdom | -0.168 |
ossible | -0.168 |
needed | -0.167 |
earned | -0.167 |
Token | , |
Token ID | 11 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
washing | +0.102 |
cloth | +0.092 |
sorting | +0.082 |
mixing | +0.078 |
dripping | +0.077 |
Neg logit contributions | |
ossible | -0.234 |
needed | -0.226 |
bye | -0.222 |
oof | -0.221 |
des | -0.221 |
Token | my |
Token ID | 616 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
washing | +0.179 |
cloth | +0.156 |
sorting | +0.151 |
mixing | +0.142 |
dripping | +0.135 |
Neg logit contributions | |
ossible | -0.323 |
former | -0.321 |
needed | -0.319 |
bye | -0.317 |
isdom | -0.316 |
Token | children |
Token ID | 1751 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
washing | +0.095 |
cloth | +0.086 |
sorting | +0.083 |
mixing | +0.079 |
living | +0.076 |
Neg logit contributions | |
former | -0.112 |
ossible | -0.110 |
needed | -0.110 |
isdom | -0.109 |
earned | -0.109 |
Token | . |
Token ID | 13 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
washing | +0.060 |
cloth | +0.053 |
sorting | +0.050 |
mixing | +0.047 |
living | +0.046 |
Neg logit contributions | |
former | -0.096 |
needed | -0.094 |
isdom | -0.093 |
earned | -0.093 |
ossible | -0.093 |
Token | I |
Token ID | 314 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
washing | +0.126 |
cloth | +0.114 |
sorting | +0.111 |
mixing | +0.107 |
dripping | +0.104 |
Neg logit contributions | |
ossible | -0.131 |
former | -0.129 |
needed | -0.129 |
bye | -0.128 |
earned | -0.126 |
Token | have |
Token ID | 423 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
washing | +0.082 |
cloth | +0.079 |
sorting | +0.071 |
living | +0.068 |
mixing | +0.068 |
Neg logit contributions | |
isdom | -0.089 |
former | -0.087 |
tal | -0.084 |
earned | -0.084 |
AY | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
washing | +0.069 |
cloth | +0.060 |
sorting | +0.058 |
mixing | +0.055 |
Mil | +0.053 |
Neg logit contributions | |
bye | -0.118 |
ossible | -0.117 |
former | -0.115 |
needed | -0.115 |
earned | -0.112 |
Token | They |
Token ID | 1119 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
washing | +0.072 |
cloth | +0.068 |
sorting | +0.064 |
mixing | +0.062 |
dripping | +0.061 |
Neg logit contributions | |
former | -0.064 |
ossible | -0.064 |
needed | -0.063 |
bye | -0.062 |
isdom | -0.062 |
Token | see |
Token ID | 766 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
washing | +0.112 |
cloth | +0.102 |
sorting | +0.099 |
mixing | +0.097 |
dripping | +0.094 |
Neg logit contributions | |
ossible | -0.084 |
former | -0.083 |
needed | -0.080 |
bye | -0.079 |
earned | -0.078 |
Token | a |
Token ID | 257 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
washing | +0.130 |
cloth | +0.125 |
sorting | +0.117 |
mixing | +0.111 |
Mil | +0.109 |
Neg logit contributions | |
former | -0.150 |
needed | -0.147 |
ossible | -0.145 |
prey | -0.145 |
faith | -0.142 |
Token | pile |
Token ID | 14540 |
Feature activation | +5.04e+00 |
Pos logit contributions | |
cloth | +0.020 |
washing | +0.018 |
Mil | +0.018 |
bled | +0.017 |
sorting | +0.017 |
Neg logit contributions | |
former | -0.016 |
nightmare | -0.016 |
prey | -0.016 |
lasting | -0.015 |
ossible | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
washing | +0.153 |
Mil | +0.143 |
cel | +0.133 |
scalp | +0.132 |
sorting | +0.132 |
Neg logit contributions | |
ossible | -0.116 |
needed | -0.115 |
prey | -0.113 |
bye | -0.111 |
encounter | -0.106 |
Token | leaves |
Token ID | 5667 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
cloth | +0.122 |
washing | +0.121 |
Mil | +0.118 |
sorting | +0.113 |
mixing | +0.109 |
Neg logit contributions | |
former | -0.102 |
needed | -0.101 |
prey | -0.099 |
ossible | -0.097 |
tal | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
washing | +0.039 |
Mil | +0.038 |
sorting | +0.036 |
cloth | +0.036 |
racks | +0.035 |
Neg logit contributions | |
moon | -0.038 |
encounter | -0.038 |
ossible | -0.037 |
prey | -0.036 |
needed | -0.036 |
Token | They |
Token ID | 1119 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
washing | +0.085 |
cloth | +0.079 |
sorting | +0.076 |
mixing | +0.073 |
dripping | +0.072 |
Neg logit contributions | |
former | -0.074 |
ossible | -0.074 |
needed | -0.073 |
isdom | -0.071 |
bye | -0.071 |
Token | pretend |
Token ID | 16614 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
washing | +0.131 |
cloth | +0.119 |
sorting | +0.116 |
mixing | +0.113 |
ula | +0.113 |
Neg logit contributions | |
ossible | -0.087 |
know | -0.084 |
former | -0.083 |
understand | -0.081 |
encounter | -0.078 |
Token | it |
Token ID | 340 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
washing | +0.143 |
cloth | +0.130 |
sorting | +0.126 |
mixing | +0.122 |
living | +0.118 |
Neg logit contributions | |
former | -0.141 |
ossible | -0.138 |
needed | -0.137 |
isdom | -0.136 |
earned | -0.135 |
Token | it |
Token ID | 340 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
washing | +0.098 |
cloth | +0.097 |
dripping | +0.087 |
sorting | +0.087 |
mixing | +0.085 |
Neg logit contributions | |
ossible | -0.079 |
isdom | -0.079 |
to | -0.079 |
revenge | -0.079 |
bye | -0.079 |
Token | up |
Token ID | 510 |
Feature activation | +3.37e-01 |
Pos logit contributions | |
ossible | +0.031 |
needed | +0.031 |
former | +0.031 |
isdom | +0.031 |
bye | +0.031 |
Neg logit contributions | |
washing | -0.022 |
cloth | -0.020 |
sorting | -0.019 |
mixing | -0.018 |
living | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -2.16e+00 |
Pos logit contributions | |
washing | +0.008 |
cloth | +0.007 |
sorting | +0.006 |
mixing | +0.006 |
dripping | +0.006 |
Neg logit contributions | |
former | -0.013 |
needed | -0.012 |
ossible | -0.012 |
isdom | -0.012 |
earned | -0.012 |
Token | put |
Token ID | 1234 |
Feature activation | +3.39e+00 |
Pos logit contributions | |
idents | +0.057 |
isdom | +0.056 |
ossible | +0.055 |
des | +0.055 |
shouldn | +0.055 |
Neg logit contributions | |
washing | -0.066 |
cloth | -0.065 |
ula | -0.059 |
sorting | -0.058 |
mixing | -0.058 |
Token | it |
Token ID | 340 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
washing | +0.126 |
cloth | +0.119 |
mixing | +0.115 |
dripping | +0.114 |
pit | +0.113 |
Neg logit contributions | |
ossible | -0.072 |
revenge | -0.070 |
tal | -0.066 |
faith | -0.065 |
oof | -0.063 |
Token | in |
Token ID | 287 |
Feature activation | +2.92e+00 |
Pos logit contributions | |
ossible | +0.047 |
needed | +0.046 |
des | +0.045 |
bye | +0.044 |
revenge | +0.044 |
Neg logit contributions | |
washing | -0.023 |
cloth | -0.020 |
Mil | -0.019 |
mixing | -0.018 |
sorting | -0.016 |
Token | his |
Token ID | 465 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
washing | +0.088 |
cloth | +0.081 |
sorting | +0.078 |
mixing | +0.075 |
dripping | +0.073 |
Neg logit contributions | |
former | -0.086 |
ossible | -0.085 |
needed | -0.083 |
isdom | -0.083 |
earned | -0.082 |
Token | pocket |
Token ID | 10000 |
Feature activation | +1.42e+00 |
Pos logit contributions | |
former | +0.068 |
needed | +0.068 |
isdom | +0.067 |
ossible | +0.067 |
earned | +0.067 |
Neg logit contributions | |
washing | -0.042 |
cloth | -0.037 |
sorting | -0.036 |
mixing | -0.033 |
living | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
washing | +0.030 |
cloth | +0.025 |
sorting | +0.024 |
mixing | +0.024 |
Mil | +0.024 |
Neg logit contributions | |
ossible | -0.055 |
former | -0.054 |
earned | -0.053 |
needed | -0.053 |
des | -0.052 |
Token | |
Token ID | 220 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
ossible | +0.027 |
former | +0.027 |
needed | +0.026 |
earned | +0.026 |
isdom | +0.026 |
Neg logit contributions | |
washing | -0.039 |
cloth | -0.037 |
sorting | -0.035 |
mixing | -0.034 |
dripping | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -3.72e+00 |
Pos logit contributions | |
washing | +0.040 |
cloth | +0.037 |
sorting | +0.036 |
living | +0.034 |
mixing | +0.034 |
Neg logit contributions | |
former | -0.043 |
needed | -0.041 |
bye | -0.040 |
isdom | -0.040 |
earned | -0.040 |
Token | Tom |
Token ID | 4186 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
washing | +0.000 |
cloth | +0.000 |
sorting | +0.000 |
Mil | +0.000 |
mixing | +0.000 |
Neg logit contributions | |
needed | -0.001 |
former | -0.001 |
earned | -0.001 |
des | -0.001 |
isdom | -0.001 |
Token | said |
Token ID | 531 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
washing | +0.071 |
cloth | +0.068 |
sorting | +0.063 |
Mil | +0.063 |
cups | +0.061 |
Neg logit contributions | |
former | -0.081 |
needed | -0.078 |
earned | -0.076 |
des | -0.076 |
isdom | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
ossible | +0.049 |
night | +0.048 |
needed | +0.048 |
bye | +0.048 |
revenge | +0.047 |
Neg logit contributions | |
washing | -0.033 |
cloth | -0.028 |
sorting | -0.027 |
mixing | -0.027 |
living | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
washing | +0.058 |
cloth | +0.053 |
sorting | +0.052 |
mixing | +0.050 |
dripping | +0.049 |
Neg logit contributions | |
ossible | -0.043 |
former | -0.042 |
needed | -0.042 |
isdom | -0.041 |
bye | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
former | +0.045 |
needed | +0.044 |
earned | +0.043 |
isdom | +0.043 |
des | +0.043 |
Neg logit contributions | |
washing | -0.050 |
cloth | -0.046 |
sorting | -0.044 |
mixing | -0.042 |
living | -0.042 |
Token | " |
Token ID | 1 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
ossible | +0.038 |
bye | +0.036 |
former | +0.035 |
Dear | +0.035 |
needed | +0.035 |
Neg logit contributions | |
washing | -0.040 |
sorting | -0.035 |
mixing | -0.034 |
dripping | -0.034 |
cloth | -0.033 |
Token | They |
Token ID | 2990 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
washing | +0.144 |
cloth | +0.135 |
sorting | +0.129 |
mixing | +0.124 |
dripping | +0.123 |
Neg logit contributions | |
former | -0.103 |
ossible | -0.102 |
needed | -0.102 |
bye | -0.101 |
earned | -0.099 |
Token | are |
Token ID | 389 |
Feature activation | +4.14e+00 |
Pos logit contributions | |
washing | +0.099 |
cloth | +0.090 |
sorting | +0.086 |
mixing | +0.082 |
living | +0.079 |
Neg logit contributions | |
former | -0.129 |
needed | -0.127 |
isdom | -0.126 |
earned | -0.125 |
ossible | -0.124 |
Token | so |
Token ID | 523 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
cloth | +0.130 |
washing | +0.121 |
racks | +0.114 |
bled | +0.111 |
orter | +0.110 |
Neg logit contributions | |
former | -0.102 |
prey | -0.100 |
oof | -0.096 |
know | -0.095 |
tal | -0.095 |
Token | many |
Token ID | 867 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
know | +0.021 |
ossible | +0.021 |
former | +0.021 |
bye | +0.020 |
wish | +0.020 |
Neg logit contributions | |
washing | -0.028 |
cloth | -0.027 |
sticking | -0.024 |
ers | -0.024 |
ters | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
former | +0.070 |
prey | +0.068 |
faith | +0.068 |
needed | +0.067 |
powers | +0.066 |
Neg logit contributions | |
washing | -0.070 |
cloth | -0.069 |
bled | -0.061 |
sorting | -0.060 |
Mil | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
night | +0.050 |
bye | +0.049 |
moon | +0.044 |
encounter | +0.044 |
needed | +0.043 |
Neg logit contributions | |
sorting | -0.061 |
cloth | -0.061 |
washing | -0.060 |
cups | -0.059 |
racks | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -4.07e+00 |
Pos logit contributions | |
washing | +0.051 |
cloth | +0.047 |
sorting | +0.044 |
mixing | +0.042 |
cups | +0.040 |
Neg logit contributions | |
former | -0.068 |
needed | -0.067 |
ossible | -0.066 |
des | -0.065 |
isdom | -0.065 |
Token | she |
Token ID | 673 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
needed | +0.119 |
former | +0.119 |
bye | +0.118 |
des | +0.117 |
faith | +0.115 |
Neg logit contributions | |
washing | -0.115 |
cloth | -0.109 |
sorting | -0.104 |
mixing | -0.099 |
ers | -0.098 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
former | +0.009 |
isdom | +0.009 |
earned | +0.009 |
needed | +0.009 |
ossible | +0.009 |
Neg logit contributions | |
washing | -0.007 |
cloth | -0.006 |
sorting | -0.006 |
mixing | -0.006 |
living | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
washing | +0.027 |
cloth | +0.024 |
sorting | +0.023 |
mixing | +0.022 |
cups | +0.021 |
Neg logit contributions | |
former | -0.042 |
ossible | -0.041 |
needed | -0.041 |
bye | -0.040 |
isdom | -0.040 |
Token | Taking |
Token ID | 20879 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
ossible | +0.005 |
former | +0.005 |
isdom | +0.005 |
needed | +0.005 |
earned | +0.005 |
Neg logit contributions | |
washing | -0.007 |
cloth | -0.006 |
sorting | -0.006 |
mixing | -0.006 |
dripping | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
washing | +0.069 |
cloth | +0.062 |
sorting | +0.057 |
living | +0.055 |
mixing | +0.055 |
Neg logit contributions | |
former | -0.087 |
bye | -0.083 |
ossible | -0.083 |
isdom | -0.083 |
des | -0.083 |
Token | shower |
Token ID | 14643 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
former | +0.114 |
isdom | +0.112 |
earned | +0.105 |
bye | +0.103 |
oof | +0.101 |
Neg logit contributions | |
washing | -0.077 |
cloth | -0.065 |
sorting | -0.062 |
mixing | -0.060 |
living | -0.060 |
Token | was |
Token ID | 373 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
washing | +0.032 |
cloth | +0.029 |
sorting | +0.028 |
mixing | +0.027 |
Mil | +0.025 |
Neg logit contributions | |
needed | -0.047 |
ossible | -0.047 |
former | -0.047 |
bye | -0.047 |
earned | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
former | +0.017 |
ossible | +0.017 |
needed | +0.016 |
isdom | +0.016 |
earned | +0.016 |
Neg logit contributions | |
washing | -0.015 |
cloth | -0.013 |
sorting | -0.013 |
mixing | -0.012 |
living | -0.012 |
Token | best |
Token ID | 1266 |
Feature activation | -2.44e-01 |
Pos logit contributions | |
washing | +0.039 |
cloth | +0.036 |
sorting | +0.034 |
mixing | +0.032 |
Mil | +0.031 |
Neg logit contributions | |
former | -0.049 |
needed | -0.048 |
ossible | -0.048 |
bye | -0.047 |
earned | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
faith | +0.041 |
needed | +0.040 |
ossible | +0.040 |
revenge | +0.039 |
prey | +0.039 |
Neg logit contributions | |
washing | -0.034 |
cloth | -0.033 |
sorting | -0.031 |
living | -0.030 |
mixing | -0.029 |
Token | bowl |
Token ID | 9396 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
washing | +0.023 |
cloth | +0.022 |
sorting | +0.021 |
Mil | +0.019 |
mixing | +0.019 |
Neg logit contributions | |
needed | -0.025 |
ossible | -0.025 |
bye | -0.025 |
des | -0.025 |
former | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -4.00e-04 |
Pos logit contributions | |
washing | +0.005 |
sorting | +0.004 |
mixing | +0.004 |
cloth | +0.004 |
living | +0.004 |
Neg logit contributions | |
ossible | -0.006 |
needed | -0.006 |
bye | -0.006 |
earned | -0.006 |
prey | -0.005 |
Token | But |
Token ID | 887 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
former | +0.000 |
needed | +0.000 |
ossible | +0.000 |
earned | +0.000 |
idents | +0.000 |
Neg logit contributions | |
washing | -0.000 |
cloth | -0.000 |
sorting | -0.000 |
mixing | -0.000 |
dripping | -0.000 |
Token | when |
Token ID | 618 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
washing | +0.026 |
cloth | +0.025 |
racks | +0.023 |
sorting | +0.023 |
ers | +0.023 |
Neg logit contributions | |
ossible | -0.027 |
idents | -0.027 |
needed | -0.027 |
bye | -0.027 |
former | -0.027 |
Token | she |
Token ID | 673 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
washing | +0.079 |
cloth | +0.075 |
sorting | +0.070 |
mixing | +0.068 |
living | +0.066 |
Neg logit contributions | |
ossible | -0.081 |
nightmare | -0.079 |
bye | -0.079 |
idents | -0.078 |
former | -0.078 |
Token | touched |
Token ID | 12615 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
washing | +0.049 |
cloth | +0.045 |
sorting | +0.043 |
mixing | +0.041 |
living | +0.040 |
Neg logit contributions | |
ossible | -0.047 |
needed | -0.047 |
earned | -0.046 |
former | -0.046 |
des | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
washing | +0.034 |
sorting | +0.033 |
cloth | +0.032 |
racks | +0.031 |
living | +0.031 |
Neg logit contributions | |
tal | -0.028 |
Nut | -0.028 |
afar | -0.028 |
ossible | -0.026 |
revenge | -0.026 |
Token | ice |
Token ID | 4771 |
Feature activation | +8.88e-01 |
Pos logit contributions | |
washing | +0.068 |
sorting | +0.064 |
cloth | +0.062 |
Mil | +0.057 |
mixing | +0.056 |
Neg logit contributions | |
ossible | -0.074 |
idents | -0.073 |
needed | -0.072 |
des | -0.071 |
bye | -0.071 |
Token | , |
Token ID | 11 |
Feature activation | +4.67e+00 |
Pos logit contributions | |
washing | +0.025 |
sorting | +0.023 |
living | +0.022 |
cel | +0.021 |
mixing | +0.021 |
Neg logit contributions | |
idents | -0.025 |
bye | -0.023 |
earned | -0.022 |
tal | -0.021 |
revenge | -0.021 |
Token | it |
Token ID | 340 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
washing | +0.116 |
cloth | +0.104 |
sorting | +0.099 |
mixing | +0.094 |
living | +0.090 |
Neg logit contributions | |
former | -0.161 |
ossible | -0.159 |
needed | -0.159 |
isdom | -0.157 |
earned | -0.156 |
Token |
|
Token ID | 198 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
former | +0.142 |
isdom | +0.133 |
earned | +0.132 |
bye | +0.131 |
needed | +0.130 |
Neg logit contributions | |
washing | -0.091 |
sorting | -0.079 |
living | -0.075 |
cloth | -0.073 |
mixing | -0.073 |
Token | The |
Token ID | 464 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
former | +0.105 |
isdom | +0.103 |
bye | +0.101 |
earned | +0.100 |
des | +0.099 |
Neg logit contributions | |
washing | -0.089 |
cloth | -0.078 |
sorting | -0.076 |
mixing | -0.073 |
living | -0.071 |
Token | test |
Token ID | 1332 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
washing | +0.015 |
cloth | +0.015 |
Mil | +0.013 |
dripping | +0.013 |
ters | +0.013 |
Neg logit contributions | |
needed | -0.016 |
ossible | -0.016 |
former | -0.015 |
bye | -0.015 |
earned | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
washing | +0.055 |
cloth | +0.050 |
sorting | +0.047 |
mixing | +0.045 |
living | +0.043 |
Neg logit contributions | |
former | -0.081 |
ossible | -0.080 |
needed | -0.080 |
earned | -0.080 |
isdom | -0.080 |
Token | very |
Token ID | 845 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
washing | +0.001 |
cloth | +0.001 |
sorting | +0.001 |
mixing | +0.000 |
living | +0.000 |
Neg logit contributions | |
former | -0.001 |
ossible | -0.001 |
needed | -0.001 |
isdom | -0.001 |
earned | -0.001 |
Token | noisy |
Token ID | 31210 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
ossible | +0.070 |
isdom | +0.070 |
former | +0.070 |
earned | +0.070 |
needed | +0.070 |
Neg logit contributions | |
washing | -0.066 |
cloth | -0.059 |
sorting | -0.058 |
mixing | -0.055 |
dripping | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -4.07e-01 |
Pos logit contributions | |
bye | +0.092 |
encounter | +0.091 |
needed | +0.090 |
night | +0.090 |
former | +0.090 |
Neg logit contributions | |
washing | -0.068 |
cloth | -0.067 |
Mil | -0.063 |
cups | -0.060 |
sorting | -0.059 |
Token | The |
Token ID | 383 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
des | +0.013 |
needed | +0.013 |
ossible | +0.013 |
bye | +0.013 |
former | +0.013 |
Neg logit contributions | |
washing | -0.012 |
cloth | -0.010 |
sorting | -0.010 |
mixing | -0.010 |
Mil | -0.010 |
Token | little |
Token ID | 1310 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
cloth | +0.035 |
washing | +0.034 |
Mil | +0.030 |
dripping | +0.029 |
sorting | +0.029 |
Neg logit contributions | |
needed | -0.042 |
nightmare | -0.041 |
former | -0.041 |
ossible | -0.041 |
bye | -0.040 |
Token | boy |
Token ID | 2933 |
Feature activation | -4.68e-01 |
Pos logit contributions | |
washing | +0.003 |
cloth | +0.003 |
bling | +0.003 |
ula | +0.003 |
dripping | +0.003 |
Neg logit contributions | |
bye | -0.003 |
ossible | -0.002 |
faith | -0.002 |
idents | -0.002 |
needed | -0.002 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
former | +0.017 |
isdom | +0.016 |
ossible | +0.016 |
needed | +0.016 |
earned | +0.016 |
Neg logit contributions | |
washing | -0.012 |
cloth | -0.010 |
sorting | -0.010 |
mixing | -0.010 |
living | -0.009 |
Token | : |
Token ID | 25 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
bye | +0.082 |
isdom | +0.082 |
ossible | +0.082 |
earned | +0.080 |
needed | +0.080 |
Neg logit contributions | |
washing | -0.052 |
cloth | -0.048 |
sorting | -0.043 |
cups | -0.042 |
dripping | -0.042 |
Token | " |
Token ID | 366 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
washing | +0.022 |
cloth | +0.020 |
sorting | +0.017 |
mixing | +0.016 |
cups | +0.016 |
Neg logit contributions | |
ossible | -0.052 |
former | -0.052 |
isdom | -0.052 |
needed | -0.051 |
earned | -0.051 |
Token | Stop |
Token ID | 19485 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
washing | +0.004 |
cloth | +0.003 |
mixing | +0.003 |
living | +0.003 |
sorting | +0.003 |
Neg logit contributions | |
former | -0.004 |
ossible | -0.004 |
idents | -0.004 |
isdom | -0.004 |
nightmare | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | -5.18e-01 |
Pos logit contributions | |
washing | +0.035 |
cloth | +0.033 |
sorting | +0.030 |
cups | +0.029 |
mixing | +0.028 |
Neg logit contributions | |
isdom | -0.043 |
former | -0.042 |
needed | -0.042 |
bye | -0.042 |
ossible | -0.042 |
Token | ! |
Token ID | 0 |
Feature activation | +7.30e-02 |
Pos logit contributions | |
former | +0.019 |
ossible | +0.019 |
needed | +0.019 |
isdom | +0.019 |
earned | +0.019 |
Neg logit contributions | |
washing | -0.011 |
cloth | -0.010 |
sorting | -0.010 |
mixing | -0.009 |
living | -0.009 |
Token | I |
Token ID | 314 |
Feature activation | -4.08e-03 |
Pos logit contributions | |
washing | +0.002 |
cloth | +0.002 |
sorting | +0.002 |
mixing | +0.002 |
living | +0.002 |
Neg logit contributions | |
former | -0.002 |
needed | -0.002 |
ossible | -0.002 |
des | -0.002 |
isdom | -0.002 |
Token | will |
Token ID | 481 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
bye | +0.000 |
prey | +0.000 |
isdom | +0.000 |
tal | +0.000 |
needed | +0.000 |
Neg logit contributions | |
washing | -0.000 |
cloth | -0.000 |
sorting | -0.000 |
mixing | -0.000 |
living | -0.000 |
Token | always |
Token ID | 1464 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
needed | +0.038 |
ossible | +0.037 |
earned | +0.037 |
former | +0.037 |
des | +0.037 |
Neg logit contributions | |
washing | -0.036 |
cloth | -0.032 |
sorting | -0.031 |
mixing | -0.030 |
living | -0.029 |
Token | protect |
Token ID | 1805 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
former | +0.072 |
ossible | +0.071 |
needed | +0.071 |
isdom | +0.070 |
earned | +0.070 |
Neg logit contributions | |
washing | -0.084 |
cloth | -0.077 |
sorting | -0.075 |
mixing | -0.072 |
living | -0.070 |
Token | you |
Token ID | 345 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
washing | +0.039 |
cloth | +0.035 |
sorting | +0.034 |
living | +0.032 |
mixing | +0.032 |
Neg logit contributions | |
former | -0.043 |
ossible | -0.042 |
des | -0.042 |
needed | -0.042 |
earned | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
ossible | +0.136 |
des | +0.135 |
needed | +0.135 |
former | +0.134 |
isdom | +0.134 |
Neg logit contributions | |
washing | -0.088 |
cloth | -0.081 |
sorting | -0.075 |
living | -0.072 |
mixing | -0.071 |
Token | careful |
Token ID | 8161 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
former | +0.052 |
ossible | +0.051 |
needed | +0.051 |
isdom | +0.051 |
earned | +0.050 |
Neg logit contributions | |
washing | -0.034 |
cloth | -0.030 |
sorting | -0.029 |
mixing | -0.027 |
living | -0.026 |
Token | around |
Token ID | 1088 |
Feature activation | -6.92e-01 |
Pos logit contributions | |
bye | +0.022 |
needed | +0.022 |
isdom | +0.021 |
earned | +0.021 |
understand | +0.021 |
Neg logit contributions | |
washing | -0.021 |
cups | -0.020 |
Mil | -0.020 |
cloth | -0.019 |
mo | -0.019 |
Token | fragile |
Token ID | 21049 |
Feature activation | -2.26e+00 |
Pos logit contributions | |
former | +0.027 |
ossible | +0.026 |
needed | +0.026 |
isdom | +0.026 |
earned | +0.026 |
Neg logit contributions | |
washing | -0.015 |
cloth | -0.013 |
sorting | -0.012 |
mixing | -0.012 |
living | -0.011 |
Token | things |
Token ID | 1243 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
former | +0.075 |
needed | +0.075 |
isdom | +0.075 |
ossible | +0.074 |
earned | +0.074 |
Neg logit contributions | |
washing | -0.057 |
cloth | -0.052 |
sorting | -0.049 |
mixing | -0.047 |
dripping | -0.046 |
Token | and |
Token ID | 290 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
washing | +0.034 |
cloth | +0.030 |
sorting | +0.028 |
mixing | +0.026 |
living | +0.026 |
Neg logit contributions | |
former | -0.064 |
ossible | -0.063 |
needed | -0.063 |
isdom | -0.062 |
earned | -0.062 |
Token | her |
Token ID | 607 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
ossible | +0.076 |
former | +0.076 |
needed | +0.075 |
isdom | +0.074 |
earned | +0.074 |
Neg logit contributions | |
washing | -0.075 |
cloth | -0.067 |
sorting | -0.066 |
mixing | -0.064 |
living | -0.062 |
Token | mom |
Token ID | 1995 |
Feature activation | +7.90e-01 |
Pos logit contributions | |
washing | +0.009 |
sorting | +0.008 |
cloth | +0.008 |
living | +0.008 |
mixing | +0.008 |
Neg logit contributions | |
needed | -0.014 |
des | -0.014 |
earned | -0.014 |
isdom | -0.013 |
former | -0.013 |
Token | trusted |
Token ID | 13467 |
Feature activation | -8.99e-01 |
Pos logit contributions | |
washing | +0.023 |
cloth | +0.021 |
sorting | +0.020 |
mixing | +0.019 |
living | +0.019 |
Neg logit contributions | |
former | -0.024 |
ossible | -0.024 |
needed | -0.024 |
isdom | -0.024 |
earned | -0.024 |
Token | her |
Token ID | 607 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
former | +0.028 |
isdom | +0.028 |
ossible | +0.028 |
needed | +0.028 |
earned | +0.027 |
Neg logit contributions | |
washing | -0.025 |
cloth | -0.023 |
sorting | -0.022 |
mixing | -0.021 |
dripping | -0.020 |
Token | even |
Token ID | 772 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
former | +0.039 |
ossible | +0.038 |
needed | +0.038 |
isdom | +0.038 |
earned | +0.038 |
Neg logit contributions | |
washing | -0.021 |
cloth | -0.019 |
sorting | -0.018 |
mixing | -0.017 |
living | -0.016 |
Token | more |
Token ID | 517 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
former | +0.041 |
needed | +0.040 |
earned | +0.040 |
isdom | +0.039 |
bye | +0.039 |
Neg logit contributions | |
washing | -0.039 |
cloth | -0.037 |
sorting | -0.035 |
mixing | -0.034 |
cups | -0.033 |
Token | his |
Token ID | 465 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
washing | +0.030 |
sorting | +0.024 |
cloth | +0.024 |
living | +0.021 |
mixing | +0.021 |
Neg logit contributions | |
former | -0.082 |
bye | -0.081 |
earned | -0.081 |
needed | -0.081 |
isdom | -0.080 |
Token | special |
Token ID | 2041 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
isdom | +0.100 |
needed | +0.099 |
former | +0.099 |
des | +0.098 |
earned | +0.098 |
Neg logit contributions | |
washing | -0.063 |
sorting | -0.054 |
cloth | -0.053 |
living | -0.051 |
mixing | -0.050 |
Token | tunnel |
Token ID | 13275 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
isdom | +0.136 |
former | +0.135 |
needed | +0.135 |
earned | +0.134 |
ossible | +0.133 |
Neg logit contributions | |
washing | -0.121 |
cloth | -0.108 |
sorting | -0.107 |
mixing | -0.101 |
living | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | -9.04e-01 |
Pos logit contributions | |
washing | +0.003 |
cloth | +0.003 |
sorting | +0.002 |
mixing | +0.002 |
living | +0.002 |
Neg logit contributions | |
former | -0.007 |
needed | -0.007 |
ossible | -0.007 |
isdom | -0.007 |
earned | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
ossible | +0.027 |
former | +0.027 |
needed | +0.026 |
isdom | +0.026 |
earned | +0.026 |
Neg logit contributions | |
washing | -0.027 |
cloth | -0.024 |
sorting | -0.023 |
mixing | -0.022 |
dripping | -0.022 |
Token | Jack |
Token ID | 14295 |
Feature activation | +3.41e-01 |
Pos logit contributions | |
former | +0.100 |
isdom | +0.095 |
needed | +0.095 |
earned | +0.094 |
bye | +0.092 |
Neg logit contributions | |
washing | -0.102 |
cloth | -0.094 |
sorting | -0.092 |
mixing | -0.087 |
living | -0.087 |
Token | shouted |
Token ID | 17293 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
washing | +0.008 |
cloth | +0.007 |
sorting | +0.007 |
living | +0.007 |
mixing | +0.007 |
Neg logit contributions | |
former | -0.013 |
needed | -0.012 |
des | -0.012 |
earned | -0.012 |
isdom | -0.012 |
Token | � |
Token ID | 6184 |
Feature activation | +5.35e-01 |
Pos logit contributions | |
washing | +0.022 |
cloth | +0.020 |
dripping | +0.018 |
mixing | +0.018 |
sorting | +0.018 |
Neg logit contributions | |
ossible | -0.035 |
needed | -0.035 |
isdom | -0.035 |
bye | -0.035 |
night | -0.034 |
Token | � |
Token ID | 95 |
Feature activation | +9.67e-01 |
Pos logit contributions | |
washing | +0.018 |
cloth | +0.017 |
living | +0.017 |
cups | +0.016 |
dripping | +0.016 |
Neg logit contributions | |
idents | -0.013 |
isdom | -0.013 |
bye | -0.013 |
des | -0.013 |
oof | -0.013 |
Token | € |
Token ID | 26391 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
washing | +0.036 |
cloth | +0.033 |
cups | +0.031 |
dripping | +0.031 |
sorting | +0.031 |
Neg logit contributions | |
bye | -0.022 |
isdom | -0.022 |
earned | -0.022 |
des | -0.022 |
ossible | -0.022 |
Token | ˜ |
Token ID | 41185 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
former | +0.004 |
needed | +0.004 |
earned | +0.004 |
ossible | +0.004 |
isdom | +0.004 |
Neg logit contributions | |
washing | -0.003 |
cloth | -0.003 |
sorting | -0.003 |
mixing | -0.003 |
living | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
washing | +0.024 |
sorting | +0.020 |
cloth | +0.019 |
mixing | +0.018 |
living | +0.018 |
Neg logit contributions | |
former | -0.048 |
isdom | -0.048 |
earned | -0.047 |
needed | -0.047 |
ossible | -0.047 |
Token | adm |
Token ID | 6178 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
cloth | +0.074 |
washing | +0.072 |
cups | +0.063 |
sorting | +0.062 |
Mil | +0.062 |
Neg logit contributions | |
needed | -0.103 |
former | -0.101 |
bye | -0.101 |
des | -0.100 |
ossible | -0.099 |
Token | iring |
Token ID | 3428 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
washing | +0.051 |
cloth | +0.047 |
sorting | +0.045 |
mixing | +0.043 |
dripping | +0.042 |
Neg logit contributions | |
former | -0.057 |
needed | -0.055 |
bye | -0.054 |
ossible | -0.054 |
isdom | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
washing | +0.035 |
cloth | +0.032 |
sorting | +0.030 |
mixing | +0.028 |
dripping | +0.027 |
Neg logit contributions | |
former | -0.057 |
needed | -0.056 |
isdom | -0.056 |
ossible | -0.056 |
earned | -0.055 |
Token | sunshine |
Token ID | 34488 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
washing | +0.021 |
cloth | +0.019 |
sorting | +0.018 |
mixing | +0.017 |
living | +0.017 |
Neg logit contributions | |
former | -0.029 |
ossible | -0.029 |
needed | -0.029 |
isdom | -0.029 |
earned | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
bye | +0.036 |
moon | +0.035 |
night | +0.034 |
needed | +0.034 |
former | +0.033 |
Neg logit contributions | |
washing | -0.026 |
cloth | -0.025 |
sorting | -0.023 |
Mil | -0.023 |
racks | -0.021 |
Token | bright |
Token ID | 6016 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
former | +0.084 |
bye | +0.080 |
lasting | +0.079 |
chat | +0.078 |
faith | +0.078 |
Neg logit contributions | |
cloth | -0.066 |
washing | -0.062 |
ula | -0.060 |
ers | -0.056 |
sorting | -0.055 |
Token | blue |
Token ID | 4171 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
bye | +0.044 |
night | +0.044 |
former | +0.042 |
needed | +0.041 |
des | +0.040 |
Neg logit contributions | |
washing | -0.040 |
cloth | -0.039 |
ula | -0.038 |
anges | -0.037 |
sorting | -0.037 |
Token | sky |
Token ID | 6766 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
washing | +0.024 |
cloth | +0.022 |
sorting | +0.021 |
Mil | +0.020 |
mixing | +0.019 |
Neg logit contributions | |
bye | -0.036 |
former | -0.036 |
needed | -0.036 |
ossible | -0.036 |
earned | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -6.24e-01 |
Pos logit contributions | |
bye | +0.033 |
moon | +0.033 |
former | +0.033 |
chat | +0.032 |
night | +0.032 |
Neg logit contributions | |
cloth | -0.038 |
washing | -0.037 |
Mil | -0.034 |
anges | -0.034 |
sorting | -0.034 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
former | +0.018 |
ossible | +0.017 |
needed | +0.017 |
isdom | +0.017 |
earned | +0.017 |
Neg logit contributions | |
washing | -0.020 |
cloth | -0.018 |
sorting | -0.017 |
mixing | -0.017 |
living | -0.016 |
Token | stubborn |
Token ID | 24444 |
Feature activation | -3.01e+00 |
Pos logit contributions | |
former | +0.070 |
ossible | +0.069 |
needed | +0.069 |
isdom | +0.068 |
earned | +0.068 |
Neg logit contributions | |
washing | -0.066 |
cloth | -0.060 |
sorting | -0.058 |
mixing | -0.055 |
Mil | -0.053 |
Token | but |
Token ID | 475 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
former | +0.095 |
ossible | +0.095 |
night | +0.094 |
bye | +0.093 |
needed | +0.092 |
Neg logit contributions | |
washing | -0.075 |
cloth | -0.068 |
Mil | -0.065 |
sorting | -0.065 |
mixing | -0.063 |
Token | she |
Token ID | 673 |
Feature activation | -4.36e-01 |
Pos logit contributions | |
former | +0.037 |
ossible | +0.035 |
isdom | +0.035 |
faith | +0.035 |
needed | +0.035 |
Neg logit contributions | |
washing | -0.034 |
cloth | -0.032 |
ers | -0.030 |
sorting | -0.029 |
cups | -0.029 |
Token | did |
Token ID | 750 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
former | +0.014 |
needed | +0.014 |
earned | +0.014 |
ossible | +0.014 |
des | +0.014 |
Neg logit contributions | |
washing | -0.012 |
cloth | -0.011 |
sorting | -0.011 |
mixing | -0.010 |
living | -0.010 |
Token | it |
Token ID | 340 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
former | +0.047 |
isdom | +0.047 |
ossible | +0.046 |
idents | +0.045 |
earned | +0.045 |
Neg logit contributions | |
washing | -0.044 |
cloth | -0.042 |
cups | -0.040 |
Mil | -0.039 |
sorting | -0.039 |
Token | anyway |
Token ID | 6949 |
Feature activation | -5.45e+00 |
Pos logit contributions | |
ossible | +0.088 |
former | +0.088 |
isdom | +0.088 |
needed | +0.087 |
earned | +0.086 |
Neg logit contributions | |
washing | -0.046 |
cloth | -0.039 |
sorting | -0.038 |
mixing | -0.034 |
living | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
ossible | +0.241 |
former | +0.241 |
needed | +0.240 |
isdom | +0.237 |
earned | +0.236 |
Neg logit contributions | |
washing | -0.084 |
cloth | -0.070 |
sorting | -0.065 |
mixing | -0.058 |
living | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
needed | +0.011 |
former | +0.011 |
earned | +0.011 |
des | +0.011 |
isdom | +0.010 |
Neg logit contributions | |
washing | -0.010 |
cloth | -0.009 |
sorting | -0.009 |
mixing | -0.009 |
Mil | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
former | +0.110 |
idents | +0.106 |
needed | +0.106 |
bye | +0.105 |
earned | +0.105 |
Neg logit contributions | |
washing | -0.087 |
sorting | -0.077 |
cloth | -0.076 |
living | -0.074 |
Mil | -0.072 |
Token | J |
Token ID | 41 |
Feature activation | +5.70e+00 |
Pos logit contributions | |
former | +0.091 |
isdom | +0.088 |
needed | +0.088 |
tal | +0.088 |
earned | +0.087 |
Neg logit contributions | |
washing | -0.078 |
cloth | -0.072 |
sorting | -0.069 |
mixing | -0.066 |
living | -0.064 |
Token | ill |
Token ID | 359 |
Feature activation | +9.00e-01 |
Pos logit contributions | |
washing | +0.144 |
cloth | +0.138 |
sorting | +0.129 |
mixing | +0.122 |
Mil | +0.118 |
Neg logit contributions | |
former | -0.196 |
needed | -0.187 |
bye | -0.184 |
des | -0.183 |
prey | -0.183 |
Token | and |
Token ID | 290 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
washing | +0.060 |
sorting | +0.052 |
cloth | +0.049 |
living | +0.049 |
mixing | +0.048 |
Neg logit contributions | |
former | -0.107 |
ossible | -0.106 |
earned | -0.106 |
des | -0.105 |
needed | -0.105 |
Token | every |
Token ID | 790 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
former | +0.110 |
ossible | +0.109 |
needed | +0.109 |
earned | +0.107 |
isdom | +0.107 |
Neg logit contributions | |
washing | -0.078 |
cloth | -0.069 |
sorting | -0.066 |
mixing | -0.063 |
living | -0.061 |
Token | day |
Token ID | 1110 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
idents | +0.009 |
former | +0.009 |
des | +0.008 |
isdom | +0.008 |
earned | +0.008 |
Neg logit contributions | |
washing | -0.006 |
living | -0.005 |
mixing | -0.005 |
sorting | -0.005 |
cloth | -0.005 |
Token | they |
Token ID | 484 |
Feature activation | -5.98e-01 |
Pos logit contributions | |
ossible | +0.092 |
bye | +0.090 |
isdom | +0.088 |
needed | +0.087 |
prey | +0.086 |
Neg logit contributions | |
washing | -0.080 |
cloth | -0.076 |
sorting | -0.068 |
mixing | -0.067 |
dripping | -0.067 |
Token | started |
Token ID | 2067 |
Feature activation | -9.78e-01 |
Pos logit contributions | |
des | +0.021 |
needed | +0.021 |
earned | +0.021 |
former | +0.020 |
idents | +0.020 |
Neg logit contributions | |
washing | -0.015 |
sorting | -0.013 |
cloth | -0.013 |
living | -0.013 |
mixing | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
former | +0.030 |
needed | +0.029 |
earned | +0.029 |
des | +0.029 |
bye | +0.029 |
Neg logit contributions | |
washing | -0.030 |
sorting | -0.027 |
mixing | -0.026 |
cloth | -0.025 |
living | -0.025 |
Token | play |
Token ID | 711 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
needed | +0.168 |
powers | +0.166 |
des | +0.165 |
earned | +0.163 |
uld | +0.162 |
Neg logit contributions | |
washing | -0.148 |
sorting | -0.136 |
mixing | -0.133 |
living | -0.127 |
tap | -0.126 |
Token | together |
Token ID | 1978 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
former | +0.039 |
ossible | +0.039 |
needed | +0.039 |
earned | +0.039 |
isdom | +0.038 |
Neg logit contributions | |
washing | -0.024 |
cloth | -0.021 |
sorting | -0.021 |
mixing | -0.019 |
living | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
former | +0.237 |
ossible | +0.234 |
needed | +0.233 |
isdom | +0.231 |
earned | +0.230 |
Neg logit contributions | |
washing | -0.111 |
cloth | -0.096 |
sorting | -0.088 |
mixing | -0.082 |
dripping | -0.079 |
Token | They |
Token ID | 1119 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
des | +0.043 |
needed | +0.041 |
earned | +0.039 |
bye | +0.039 |
former | +0.038 |
Neg logit contributions | |
washing | -0.027 |
sorting | -0.024 |
living | -0.022 |
cloth | -0.022 |
squares | -0.022 |
Token | told |
Token ID | 1297 |
Feature activation | -3.16e+00 |
Pos logit contributions | |
washing | +0.005 |
cloth | +0.005 |
sorting | +0.004 |
mixing | +0.004 |
living | +0.004 |
Neg logit contributions | |
former | -0.006 |
needed | -0.006 |
earned | -0.006 |
des | -0.006 |
isdom | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
washing | +0.012 |
cloth | +0.010 |
sorting | +0.010 |
mixing | +0.009 |
Mil | +0.009 |
Neg logit contributions | |
bye | -0.019 |
needed | -0.019 |
former | -0.019 |
ossible | -0.019 |
night | -0.019 |
Token | They |
Token ID | 1119 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
former | +0.048 |
ossible | +0.048 |
needed | +0.048 |
isdom | +0.047 |
earned | +0.047 |
Neg logit contributions | |
washing | -0.044 |
cloth | -0.039 |
sorting | -0.039 |
mixing | -0.037 |
living | -0.036 |
Token | had |
Token ID | 550 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
former | +0.001 |
ossible | +0.001 |
needed | +0.001 |
isdom | +0.001 |
earned | +0.001 |
Neg logit contributions | |
washing | -0.001 |
cloth | -0.001 |
sorting | -0.001 |
mixing | -0.001 |
dripping | -0.001 |
Token | so |
Token ID | 523 |
Feature activation | -4.90e+00 |
Pos logit contributions | |
isdom | +0.016 |
former | +0.016 |
ossible | +0.016 |
earned | +0.015 |
bye | +0.015 |
Neg logit contributions | |
washing | -0.012 |
sorting | -0.011 |
cloth | -0.011 |
mixing | -0.010 |
living | -0.010 |
Token | much |
Token ID | 881 |
Feature activation | -4.57e+00 |
Pos logit contributions | |
former | +0.155 |
isdom | +0.148 |
ossible | +0.144 |
needed | +0.144 |
idents | +0.144 |
Neg logit contributions | |
washing | -0.142 |
sorting | -0.127 |
cloth | -0.127 |
mixing | -0.122 |
living | -0.119 |
Token | fun |
Token ID | 1257 |
Feature activation | -5.75e+00 |
Pos logit contributions | |
former | +0.105 |
ossible | +0.103 |
needed | +0.102 |
isdom | +0.102 |
earned | +0.101 |
Neg logit contributions | |
washing | -0.169 |
cloth | -0.156 |
sorting | -0.152 |
mixing | -0.148 |
living | -0.144 |
Token | ! |
Token ID | 0 |
Feature activation | -2.31e+00 |
Pos logit contributions | |
ossible | +0.175 |
encounter | +0.173 |
needed | +0.173 |
des | +0.172 |
faith | +0.171 |
Neg logit contributions | |
cloth | -0.161 |
washing | -0.139 |
Mil | -0.133 |
cabinets | -0.132 |
cups | -0.129 |
Token |
|
Token ID | 198 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
des | +0.074 |
former | +0.074 |
needed | +0.073 |
isdom | +0.072 |
earned | +0.072 |
Neg logit contributions | |
washing | -0.066 |
sorting | -0.058 |
living | -0.056 |
mixing | -0.055 |
cloth | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
former | +0.156 |
earned | +0.150 |
des | +0.150 |
idents | +0.150 |
isdom | +0.149 |
Neg logit contributions | |
washing | -0.109 |
living | -0.096 |
sorting | -0.095 |
mixing | -0.089 |
cloth | -0.087 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
former | +0.110 |
ossible | +0.108 |
isdom | +0.107 |
tal | +0.106 |
earned | +0.106 |
Neg logit contributions | |
washing | -0.117 |
cloth | -0.105 |
sorting | -0.104 |
mixing | -0.101 |
living | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
former | +0.053 |
needed | +0.052 |
ossible | +0.052 |
earned | +0.052 |
isdom | +0.051 |
Neg logit contributions | |
washing | -0.019 |
cloth | -0.016 |
sorting | -0.015 |
mixing | -0.014 |
living | -0.013 |
Token | seen |
Token ID | 1775 |
Feature activation | -3.12e+00 |
Pos logit contributions | |
former | +0.072 |
needed | +0.071 |
ossible | +0.071 |
earned | +0.069 |
des | +0.069 |
Neg logit contributions | |
washing | -0.197 |
cloth | -0.185 |
sorting | -0.180 |
mixing | -0.175 |
dripping | -0.172 |
Token | . |
Token ID | 13 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
des | +0.095 |
former | +0.095 |
bye | +0.095 |
needed | +0.094 |
earned | +0.093 |
Neg logit contributions | |
washing | -0.082 |
cloth | -0.076 |
sorting | -0.072 |
Mil | -0.071 |
dripping | -0.070 |
Token | He |
Token ID | 679 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
needed | +0.035 |
former | +0.035 |
des | +0.034 |
isdom | +0.034 |
bye | +0.034 |
Neg logit contributions | |
washing | -0.034 |
cloth | -0.030 |
sorting | -0.030 |
Mil | -0.028 |
living | -0.028 |
Token | was |
Token ID | 373 |
Feature activation | +8.86e-02 |
Pos logit contributions | |
needed | +0.043 |
former | +0.043 |
isdom | +0.042 |
ossible | +0.042 |
earned | +0.042 |
Neg logit contributions | |
washing | -0.030 |
cloth | -0.027 |
sorting | -0.026 |
mixing | -0.025 |
living | -0.024 |
Token | so |
Token ID | 523 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
washing | +0.003 |
cloth | +0.002 |
sorting | +0.002 |
mixing | +0.002 |
cups | +0.002 |
Neg logit contributions | |
former | -0.003 |
ossible | -0.003 |
needed | -0.003 |
bye | -0.002 |
earned | -0.002 |
Token | excited |
Token ID | 6568 |
Feature activation | -6.05e+00 |
Pos logit contributions | |
former | +0.098 |
ossible | +0.096 |
needed | +0.096 |
isdom | +0.096 |
earned | +0.095 |
Neg logit contributions | |
washing | -0.090 |
cloth | -0.082 |
sorting | -0.079 |
mixing | -0.076 |
living | -0.073 |
Token | , |
Token ID | 11 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
former | +0.178 |
ossible | +0.177 |
needed | +0.177 |
isdom | +0.175 |
earned | +0.174 |
Neg logit contributions | |
washing | -0.180 |
cloth | -0.166 |
sorting | -0.159 |
mixing | -0.152 |
dripping | -0.148 |
Token | he |
Token ID | 339 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
washing | +0.049 |
cloth | +0.044 |
sorting | +0.043 |
mixing | +0.041 |
living | +0.039 |
Neg logit contributions | |
former | -0.059 |
ossible | -0.059 |
needed | -0.058 |
isdom | -0.058 |
earned | -0.058 |
Token | started |
Token ID | 2067 |
Feature activation | -6.28e-01 |
Pos logit contributions | |
former | +0.033 |
ossible | +0.032 |
needed | +0.032 |
isdom | +0.032 |
earned | +0.032 |
Neg logit contributions | |
washing | -0.028 |
cloth | -0.025 |
sorting | -0.024 |
mixing | -0.023 |
living | -0.022 |
Token | bouncing |
Token ID | 30713 |
Feature activation | +4.27e-01 |
Pos logit contributions | |
former | +0.018 |
ossible | +0.017 |
needed | +0.017 |
earned | +0.017 |
isdom | +0.017 |
Neg logit contributions | |
washing | -0.019 |
cloth | -0.018 |
sorting | -0.017 |
mixing | -0.017 |
cups | -0.016 |
Token | all |
Token ID | 477 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
washing | +0.010 |
cloth | +0.009 |
sorting | +0.009 |
mixing | +0.008 |
dripping | +0.008 |
Neg logit contributions | |
former | -0.015 |
ossible | -0.015 |
needed | -0.015 |
isdom | -0.015 |
earned | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
former | +0.095 |
isdom | +0.094 |
needed | +0.094 |
ossible | +0.094 |
earned | +0.094 |
Neg logit contributions | |
washing | -0.035 |
cloth | -0.029 |
sorting | -0.027 |
mixing | -0.024 |
living | -0.023 |
Token | They |
Token ID | 1119 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
washing | +0.030 |
cloth | +0.026 |
sorting | +0.026 |
living | +0.026 |
Mil | +0.025 |
Neg logit contributions | |
needed | -0.035 |
former | -0.035 |
isdom | -0.035 |
earned | -0.034 |
des | -0.034 |
Token | learned |
Token ID | 4499 |
Feature activation | -5.87e-01 |
Pos logit contributions | |
washing | +0.058 |
cloth | +0.054 |
living | +0.052 |
twins | +0.049 |
mo | +0.049 |
Neg logit contributions | |
earned | -0.095 |
isdom | -0.095 |
needed | -0.094 |
to | -0.091 |
AY | -0.088 |
Token | a |
Token ID | 257 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ited | +0.026 |
encounter | +0.026 |
uld | +0.026 |
des | +0.025 |
earned | +0.025 |
Neg logit contributions | |
washing | -0.012 |
living | -0.010 |
sorting | -0.008 |
twins | -0.008 |
in | -0.007 |
Token | hard |
Token ID | 1327 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
former | +0.049 |
earned | +0.049 |
des | +0.049 |
isdom | +0.048 |
tal | +0.047 |
Neg logit contributions | |
washing | -0.067 |
living | -0.063 |
sorting | -0.058 |
dripping | -0.053 |
twins | -0.052 |
Token | lesson |
Token ID | 11483 |
Feature activation | -5.84e+00 |
Pos logit contributions | |
former | +0.038 |
needed | +0.034 |
des | +0.033 |
afar | +0.033 |
uld | +0.033 |
Neg logit contributions | |
washing | -0.060 |
living | -0.053 |
sorting | -0.052 |
floor | -0.051 |
cloth | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
ossible | +0.222 |
former | +0.221 |
isdom | +0.220 |
bye | +0.220 |
needed | +0.220 |
Neg logit contributions | |
washing | -0.124 |
cloth | -0.109 |
sorting | -0.103 |
mixing | -0.098 |
dripping | -0.093 |
Token | Fire |
Token ID | 3764 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
washing | +0.042 |
cloth | +0.038 |
sorting | +0.037 |
mixing | +0.035 |
living | +0.035 |
Neg logit contributions | |
needed | -0.045 |
former | -0.045 |
earned | -0.044 |
isdom | -0.044 |
des | -0.044 |
Token | is |
Token ID | 318 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
washing | +0.015 |
cloth | +0.014 |
sorting | +0.013 |
mixing | +0.012 |
living | +0.012 |
Neg logit contributions | |
former | -0.020 |
ossible | -0.019 |
needed | -0.019 |
isdom | -0.019 |
earned | -0.019 |
Token | not |
Token ID | 407 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
washing | +0.058 |
cloth | +0.052 |
sorting | +0.050 |
mixing | +0.048 |
living | +0.046 |
Neg logit contributions | |
former | -0.067 |
ossible | -0.066 |
needed | -0.066 |
isdom | -0.065 |
earned | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | +9.45e-01 |
Pos logit contributions | |
washing | +0.064 |
sorting | +0.054 |
cloth | +0.053 |
living | +0.053 |
mixing | +0.051 |
Neg logit contributions | |
ossible | -0.082 |
former | -0.080 |
earned | -0.080 |
des | -0.080 |
bye | -0.079 |