Token | not |
Token ID | 407 |
Feature activation | +8.96e-02 |
Pos logit contributions | |
idea | +0.148 |
adorable | +0.143 |
prized | +0.143 |
pleased | +0.136 |
best | +0.135 |
Neg logit contributions | |
oes | -0.205 |
rost | -0.199 |
isk | -0.199 |
ted | -0.197 |
stack | -0.196 |
Token | a |
Token ID | 257 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
prized | +0.127 |
thrilled | +0.116 |
adorable | +0.115 |
precious | +0.115 |
ement | +0.114 |
Neg logit contributions | |
oes | -0.166 |
isk | -0.161 |
rost | -0.158 |
talk | -0.156 |
south | -0.154 |
Token | monster |
Token ID | 9234 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
thrilled | +0.310 |
adorable | +0.306 |
pleased | +0.303 |
delighted | +0.297 |
impressed | +0.293 |
Neg logit contributions | |
oes | -0.266 |
use | -0.261 |
rost | -0.257 |
ricane | -0.248 |
talk | -0.245 |
Token | , |
Token ID | 11 |
Feature activation | -8.22e-03 |
Pos logit contributions | |
adorable | +0.024 |
best | +0.023 |
important | +0.023 |
idea | +0.022 |
precious | +0.022 |
Neg logit contributions | |
oes | -0.039 |
talk | -0.039 |
ted | -0.038 |
stack | -0.036 |
rost | -0.036 |
Token | my |
Token ID | 616 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
oes | +0.012 |
talk | +0.012 |
ted | +0.012 |
stack | +0.011 |
rost | +0.011 |
Neg logit contributions | |
adorable | -0.017 |
best | -0.016 |
important | -0.016 |
idea | -0.016 |
precious | -0.016 |
Token | de |
Token ID | 390 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
adorable | +0.180 |
important | +0.175 |
pleased | +0.171 |
thrilled | +0.170 |
impressed | +0.168 |
Neg logit contributions | |
oes | -0.241 |
ted | -0.230 |
rost | -0.222 |
talk | -0.220 |
use | -0.216 |
Token | ars |
Token ID | 945 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
adorable | +0.552 |
best | +0.534 |
idea | +0.530 |
important | +0.530 |
precious | +0.514 |
Neg logit contributions | |
oes | -0.387 |
talk | -0.386 |
ted | -0.367 |
rost | -0.344 |
stack | -0.343 |
Token | . |
Token ID | 13 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
oes | +0.035 |
talk | +0.035 |
ted | +0.034 |
stack | +0.033 |
rost | +0.032 |
Neg logit contributions | |
adorable | -0.022 |
important | -0.021 |
best | -0.021 |
precious | -0.020 |
idea | -0.020 |
Token | That |
Token ID | 1320 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
adorable | +0.035 |
best | +0.034 |
important | +0.033 |
idea | +0.033 |
precious | +0.032 |
Neg logit contributions | |
talk | -0.036 |
oes | -0.036 |
ted | -0.034 |
stack | -0.033 |
isk | -0.032 |
Token | is |
Token ID | 318 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
pleased | +0.094 |
adorable | +0.090 |
impressed | +0.090 |
thrilled | +0.089 |
delighted | +0.086 |
Neg logit contributions | |
oes | -0.070 |
ted | -0.064 |
talk | -0.064 |
stack | -0.060 |
use | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
adorable | +0.159 |
idea | +0.153 |
best | +0.153 |
important | +0.150 |
precious | +0.144 |
Neg logit contributions | |
oes | -0.253 |
talk | -0.249 |
ted | -0.244 |
stack | -0.234 |
rost | -0.234 |
Token | was |
Token ID | 373 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
oes | +0.084 |
ted | +0.081 |
talk | +0.080 |
isk | +0.073 |
ting | +0.072 |
Neg logit contributions | |
adorable | -0.107 |
pleased | -0.105 |
prized | -0.104 |
precious | -0.104 |
idea | -0.103 |
Token | a |
Token ID | 257 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
idea | +0.149 |
prized | +0.146 |
adorable | +0.145 |
thrilled | +0.142 |
pleased | +0.141 |
Neg logit contributions | |
rost | -0.216 |
oes | -0.206 |
south | -0.204 |
isk | -0.203 |
Enter | -0.196 |
Token | dream |
Token ID | 4320 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
adorable | +0.198 |
idea | +0.186 |
impressed | +0.183 |
thrilled | +0.182 |
prized | +0.177 |
Neg logit contributions | |
rost | -0.407 |
oes | -0.395 |
stack | -0.394 |
talk | -0.387 |
use | -0.383 |
Token | , |
Token ID | 11 |
Feature activation | -6.41e-03 |
Pos logit contributions | |
adorable | +0.159 |
best | +0.152 |
important | +0.150 |
idea | +0.149 |
precious | +0.145 |
Neg logit contributions | |
oes | -0.223 |
talk | -0.220 |
ted | -0.212 |
stack | -0.204 |
rost | -0.203 |
Token | my |
Token ID | 616 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
oes | +0.009 |
rost | +0.008 |
ted | +0.008 |
stack | +0.008 |
isk | +0.008 |
Neg logit contributions | |
adorable | -0.013 |
idea | -0.013 |
best | -0.013 |
important | -0.013 |
prized | -0.012 |
Token | de |
Token ID | 390 |
Feature activation | +2.57e-01 |
Pos logit contributions | |
impressed | +0.153 |
pleased | +0.145 |
thrilled | +0.138 |
delighted | +0.133 |
adorable | +0.131 |
Neg logit contributions | |
rost | -0.230 |
ted | -0.216 |
oes | -0.213 |
Rex | -0.212 |
isk | -0.203 |
Token | ars |
Token ID | 945 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
adorable | +0.537 |
best | +0.524 |
important | +0.519 |
idea | +0.518 |
precious | +0.504 |
Neg logit contributions | |
oes | -0.395 |
talk | -0.390 |
ted | -0.375 |
stack | -0.354 |
rost | -0.349 |
Token | ," |
Token ID | 553 |
Feature activation | -3.29e-03 |
Pos logit contributions | |
oes | +0.051 |
talk | +0.051 |
ted | +0.049 |
stack | +0.047 |
rost | +0.046 |
Neg logit contributions | |
adorable | -0.040 |
best | -0.038 |
important | -0.038 |
idea | -0.037 |
precious | -0.037 |
Token | she |
Token ID | 673 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
oes | +0.005 |
talk | +0.005 |
ted | +0.005 |
stack | +0.004 |
rost | +0.004 |
Neg logit contributions | |
adorable | -0.007 |
best | -0.007 |
important | -0.007 |
idea | -0.007 |
precious | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
talk | +0.042 |
ted | +0.041 |
oes | +0.036 |
isk | +0.035 |
Dog | +0.035 |
Neg logit contributions | |
best | -0.054 |
adorable | -0.053 |
important | -0.052 |
prized | -0.051 |
precious | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
adorable | +0.180 |
best | +0.174 |
important | +0.173 |
idea | +0.172 |
precious | +0.167 |
Neg logit contributions | |
oes | -0.213 |
talk | -0.212 |
ted | -0.205 |
stack | -0.195 |
rost | -0.193 |
Token | He |
Token ID | 679 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
talk | +0.004 |
oes | +0.004 |
ted | +0.004 |
use | +0.004 |
ting | +0.004 |
Neg logit contributions | |
adorable | -0.005 |
idea | -0.005 |
important | -0.004 |
best | -0.004 |
beautiful | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
ted | +0.077 |
talk | +0.075 |
oes | +0.074 |
rost | +0.070 |
isk | +0.070 |
Neg logit contributions | |
best | -0.081 |
adorable | -0.079 |
idea | -0.079 |
precious | -0.078 |
important | -0.077 |
Token | a |
Token ID | 257 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
idea | +0.174 |
precious | +0.171 |
important | +0.163 |
adorable | +0.161 |
valuable | +0.159 |
Neg logit contributions | |
oes | -0.279 |
gra | -0.263 |
ted | -0.263 |
isk | -0.261 |
talk | -0.255 |
Token | mis |
Token ID | 2984 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
idea | +0.226 |
mistakes | +0.208 |
best | +0.203 |
important | +0.197 |
adorable | +0.191 |
Neg logit contributions | |
oes | -0.446 |
rost | -0.444 |
Dog | -0.442 |
gra | -0.437 |
stack | -0.437 |
Token | chie |
Token ID | 3043 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
best | +0.336 |
important | +0.325 |
idea | +0.324 |
adorable | +0.321 |
precious | +0.312 |
Neg logit contributions | |
talk | -0.240 |
oes | -0.228 |
isk | -0.199 |
Dog | -0.195 |
ting | -0.191 |
Token | vous |
Token ID | 31222 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
best | +0.281 |
idea | +0.271 |
adorable | +0.269 |
precious | +0.269 |
important | +0.258 |
Neg logit contributions | |
oes | -0.328 |
talk | -0.324 |
ted | -0.299 |
Dog | -0.285 |
gra | -0.283 |
Token | little |
Token ID | 1310 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
important | +0.404 |
precious | +0.397 |
best | +0.394 |
adorable | +0.390 |
impressed | +0.389 |
Neg logit contributions | |
oes | -0.446 |
ted | -0.441 |
rost | -0.429 |
ting | -0.427 |
talk | -0.425 |
Token | baby |
Token ID | 5156 |
Feature activation | +9.21e-02 |
Pos logit contributions | |
important | +0.232 |
pleased | +0.226 |
impressed | +0.225 |
prized | +0.224 |
idea | +0.223 |
Neg logit contributions | |
oes | -0.275 |
ted | -0.254 |
ting | -0.253 |
stack | -0.244 |
talk | -0.241 |
Token | and |
Token ID | 290 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
idea | +0.167 |
adorable | +0.167 |
important | +0.164 |
best | +0.163 |
precious | +0.160 |
Neg logit contributions | |
oes | -0.158 |
ted | -0.156 |
talk | -0.154 |
stack | -0.148 |
ting | -0.147 |
Token | always |
Token ID | 1464 |
Feature activation | +8.50e-03 |
Pos logit contributions | |
oes | +0.046 |
ted | +0.045 |
talk | +0.044 |
stack | +0.044 |
rost | +0.043 |
Neg logit contributions | |
best | -0.044 |
adorable | -0.044 |
idea | -0.044 |
precious | -0.043 |
important | -0.042 |
Token | got |
Token ID | 1392 |
Feature activation | +3.82e-02 |
Pos logit contributions | |
adorable | +0.015 |
best | +0.014 |
idea | +0.014 |
important | +0.014 |
precious | +0.014 |
Neg logit contributions | |
oes | -0.016 |
talk | -0.015 |
ted | -0.015 |
stack | -0.014 |
isk | -0.014 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
best | +0.162 |
adorable | +0.162 |
idea | +0.161 |
prized | +0.151 |
precious | +0.142 |
Neg logit contributions | |
oes | -0.435 |
rost | -0.416 |
gra | -0.412 |
talk | -0.411 |
south | -0.403 |
Token | , |
Token ID | 11 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
adorable | +0.335 |
best | +0.324 |
important | +0.321 |
idea | +0.318 |
precious | +0.314 |
Neg logit contributions | |
oes | -0.359 |
talk | -0.352 |
ted | -0.345 |
stack | -0.330 |
rost | -0.321 |
Token | it |
Token ID | 340 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
adorable | +0.035 |
idea | +0.035 |
best | +0.035 |
important | +0.035 |
precious | +0.034 |
Neg logit contributions | |
oes | -0.034 |
ted | -0.033 |
talk | -0.033 |
stack | -0.032 |
rost | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
ted | +0.032 |
talk | +0.032 |
oes | +0.031 |
stack | +0.030 |
ting | +0.029 |
Neg logit contributions | |
adorable | -0.038 |
best | -0.037 |
idea | -0.037 |
important | -0.037 |
precious | -0.036 |
Token | a |
Token ID | 257 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
adorable | +0.223 |
idea | +0.222 |
precious | +0.219 |
prized | +0.214 |
best | +0.208 |
Neg logit contributions | |
oes | -0.241 |
stack | -0.230 |
ted | -0.227 |
gra | -0.224 |
south | -0.224 |
Token | brand |
Token ID | 4508 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
idea | +0.225 |
adorable | +0.224 |
mistakes | +0.207 |
best | +0.203 |
important | +0.200 |
Neg logit contributions | |
stack | -0.464 |
gra | -0.453 |
rost | -0.447 |
oes | -0.445 |
south | -0.442 |
Token | new |
Token ID | 649 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
adorable | +0.263 |
idea | +0.252 |
best | +0.249 |
thrilled | +0.243 |
loyal | +0.226 |
Neg logit contributions | |
oes | -0.622 |
talk | -0.584 |
ted | -0.583 |
iffs | -0.565 |
rost | -0.562 |
Token | ice |
Token ID | 4771 |
Feature activation | +9.06e-02 |
Pos logit contributions | |
adorable | +0.359 |
impressed | +0.345 |
pleased | +0.344 |
important | +0.338 |
precious | +0.335 |
Neg logit contributions | |
oes | -0.358 |
stack | -0.344 |
talk | -0.331 |
ted | -0.317 |
south | -0.300 |
Token | cream |
Token ID | 8566 |
Feature activation | +6.06e-02 |
Pos logit contributions | |
adorable | +0.159 |
idea | +0.154 |
important | +0.151 |
best | +0.149 |
precious | +0.148 |
Neg logit contributions | |
oes | -0.169 |
talk | -0.157 |
ted | -0.153 |
stack | -0.151 |
rost | -0.149 |
Token | machine |
Token ID | 4572 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
adorable | +0.101 |
important | +0.097 |
idea | +0.097 |
best | +0.096 |
precious | +0.094 |
Neg logit contributions | |
oes | -0.118 |
talk | -0.115 |
ted | -0.113 |
stack | -0.110 |
rost | -0.106 |
Token | . |
Token ID | 13 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
oes | +0.006 |
talk | +0.005 |
ted | +0.005 |
stack | +0.005 |
rost | +0.005 |
Neg logit contributions | |
adorable | -0.005 |
best | -0.004 |
important | -0.004 |
idea | -0.004 |
precious | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.37e-02 |
Pos logit contributions | |
oes | +0.045 |
talk | +0.045 |
ted | +0.043 |
stack | +0.041 |
rost | +0.040 |
Neg logit contributions | |
adorable | -0.054 |
best | -0.053 |
important | -0.052 |
idea | -0.052 |
precious | -0.051 |
Token | " |
Token ID | 1 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
oes | +0.050 |
talk | +0.050 |
ted | +0.048 |
stack | +0.046 |
rost | +0.046 |
Neg logit contributions | |
adorable | -0.036 |
best | -0.035 |
idea | -0.035 |
important | -0.034 |
precious | -0.033 |
Token | That |
Token ID | 2504 |
Feature activation | +6.32e-02 |
Pos logit contributions | |
oes | +0.039 |
isk | +0.036 |
talk | +0.036 |
stack | +0.036 |
ted | +0.035 |
Neg logit contributions | |
adorable | -0.046 |
best | -0.044 |
precious | -0.044 |
important | -0.044 |
beautiful | -0.043 |
Token | , |
Token ID | 11 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
pleased | +0.126 |
impressed | +0.123 |
importantly | +0.121 |
adorable | +0.121 |
precious | +0.120 |
Neg logit contributions | |
oes | -0.099 |
talk | -0.095 |
ted | -0.094 |
stack | -0.087 |
gra | -0.086 |
Token | my |
Token ID | 616 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
ted | +0.034 |
oes | +0.033 |
talk | +0.031 |
ting | +0.031 |
rost | +0.031 |
Neg logit contributions | |
important | -0.030 |
adorable | -0.029 |
idea | -0.029 |
prized | -0.029 |
best | -0.028 |
Token | de |
Token ID | 390 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
impressed | +0.147 |
important | +0.145 |
pleased | +0.142 |
adorable | +0.139 |
prized | +0.137 |
Neg logit contributions | |
ted | -0.230 |
oes | -0.221 |
ricanes | -0.214 |
rost | -0.212 |
talk | -0.210 |
Token | ars |
Token ID | 945 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
adorable | +0.541 |
best | +0.527 |
important | +0.522 |
idea | +0.521 |
precious | +0.507 |
Neg logit contributions | |
oes | -0.377 |
talk | -0.373 |
ted | -0.357 |
stack | -0.335 |
rost | -0.332 |
Token | , |
Token ID | 11 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
oes | +0.070 |
talk | +0.070 |
ted | +0.068 |
stack | +0.065 |
rost | +0.064 |
Neg logit contributions | |
adorable | -0.048 |
best | -0.046 |
important | -0.046 |
idea | -0.045 |
precious | -0.044 |
Token | is |
Token ID | 318 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
oes | +0.015 |
talk | +0.015 |
ted | +0.014 |
stack | +0.013 |
rost | +0.013 |
Neg logit contributions | |
adorable | -0.015 |
best | -0.015 |
important | -0.015 |
idea | -0.014 |
precious | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
adorable | +0.222 |
best | +0.219 |
important | +0.215 |
precious | +0.206 |
beautiful | +0.204 |
Neg logit contributions | |
talk | -0.221 |
oes | -0.219 |
ted | -0.204 |
stack | -0.194 |
isk | -0.191 |
Token | cave |
Token ID | 11527 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
adorable | +0.315 |
idea | +0.304 |
important | +0.300 |
best | +0.298 |
precious | +0.285 |
Neg logit contributions | |
oes | -0.368 |
talk | -0.361 |
stack | -0.357 |
rost | -0.351 |
ted | -0.349 |
Token |
|
Token ID | 198 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
oes | +0.039 |
ted | +0.038 |
talk | +0.038 |
ting | +0.036 |
stack | +0.035 |
Neg logit contributions | |
adorable | -0.035 |
important | -0.034 |
idea | -0.033 |
best | -0.033 |
beautiful | -0.033 |
Token | " |
Token ID | 1 |
Feature activation | -2.26e-02 |
Pos logit contributions | |
oes | +0.035 |
ted | +0.034 |
talk | +0.033 |
stack | +0.032 |
ting | +0.031 |
Neg logit contributions | |
adorable | -0.034 |
important | -0.034 |
beautiful | -0.032 |
best | -0.032 |
loyal | -0.032 |
Token | Hello |
Token ID | 15496 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
oes | +0.033 |
talk | +0.030 |
ted | +0.030 |
stack | +0.030 |
isk | +0.028 |
Neg logit contributions | |
adorable | -0.050 |
best | -0.048 |
important | -0.047 |
beautiful | -0.047 |
loyal | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | -3.71e-02 |
Pos logit contributions | |
adorable | +0.069 |
important | +0.069 |
idea | +0.069 |
best | +0.066 |
precious | +0.064 |
Neg logit contributions | |
oes | -0.100 |
ted | -0.098 |
talk | -0.096 |
rost | -0.092 |
ting | -0.091 |
Token | my |
Token ID | 616 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
oes | +0.074 |
talk | +0.072 |
ted | +0.072 |
stack | +0.069 |
rost | +0.069 |
Neg logit contributions | |
adorable | -0.056 |
idea | -0.055 |
best | -0.055 |
important | -0.055 |
precious | -0.052 |
Token | de |
Token ID | 390 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
impressed | +0.144 |
adorable | +0.143 |
pleased | +0.143 |
idea | +0.140 |
important | +0.138 |
Neg logit contributions | |
oes | -0.233 |
ted | -0.232 |
talk | -0.229 |
isk | -0.228 |
rost | -0.225 |
Token | ars |
Token ID | 945 |
Feature activation | -3.64e-02 |
Pos logit contributions | |
adorable | +0.546 |
best | +0.532 |
idea | +0.528 |
important | +0.526 |
precious | +0.510 |
Neg logit contributions | |
oes | -0.367 |
talk | -0.364 |
ted | -0.343 |
stack | -0.324 |
rost | -0.324 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
oes | +0.080 |
talk | +0.079 |
ted | +0.076 |
stack | +0.075 |
use | +0.073 |
Neg logit contributions | |
adorable | -0.050 |
important | -0.049 |
best | -0.048 |
precious | -0.047 |
idea | -0.046 |
Token | Grand |
Token ID | 5675 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
oes | +0.046 |
talk | +0.044 |
ted | +0.043 |
isk | +0.043 |
stack | +0.042 |
Neg logit contributions | |
adorable | -0.047 |
best | -0.045 |
idea | -0.045 |
important | -0.044 |
beautiful | -0.043 |
Token | ma |
Token ID | 2611 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
adorable | +0.178 |
best | +0.172 |
important | +0.170 |
idea | +0.169 |
precious | +0.164 |
Neg logit contributions | |
oes | -0.228 |
talk | -0.226 |
ted | -0.219 |
stack | -0.210 |
rost | -0.208 |
Token | says |
Token ID | 1139 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
adorable | +0.039 |
idea | +0.038 |
best | +0.038 |
important | +0.036 |
precious | +0.035 |
Neg logit contributions | |
oes | -0.057 |
talk | -0.056 |
ted | -0.055 |
stack | -0.054 |
rost | -0.054 |
Token | a |
Token ID | 257 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
adorable | +0.150 |
important | +0.144 |
idea | +0.141 |
best | +0.140 |
precious | +0.139 |
Neg logit contributions | |
oes | -0.222 |
talk | -0.216 |
ted | -0.213 |
stack | -0.211 |
rost | -0.205 |
Token | special |
Token ID | 2041 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
adorable | +0.256 |
idea | +0.242 |
best | +0.239 |
important | +0.237 |
precious | +0.227 |
Neg logit contributions | |
stack | -0.446 |
oes | -0.444 |
gra | -0.441 |
talk | -0.432 |
use | -0.432 |
Token | gift |
Token ID | 6979 |
Feature activation | +3.96e-02 |
Pos logit contributions | |
adorable | +0.314 |
best | +0.301 |
important | +0.295 |
idea | +0.289 |
precious | +0.287 |
Neg logit contributions | |
oes | -0.362 |
talk | -0.356 |
ted | -0.344 |
stack | -0.340 |
rost | -0.330 |
Token | - |
Token ID | 532 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
adorable | +0.070 |
best | +0.068 |
important | +0.067 |
idea | +0.067 |
precious | +0.065 |
Neg logit contributions | |
oes | -0.072 |
talk | -0.071 |
ted | -0.069 |
stack | -0.067 |
rost | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
adorable | +0.046 |
idea | +0.046 |
best | +0.045 |
important | +0.044 |
precious | +0.043 |
Neg logit contributions | |
oes | -0.051 |
ted | -0.050 |
talk | -0.049 |
stack | -0.048 |
gra | -0.046 |
Token | brand |
Token ID | 4508 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
adorable | +0.202 |
idea | +0.197 |
best | +0.184 |
impressed | +0.183 |
thrilled | +0.176 |
Neg logit contributions | |
stack | -0.453 |
gra | -0.443 |
use | -0.427 |
oes | -0.426 |
south | -0.423 |
Token | new |
Token ID | 649 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
adorable | +0.266 |
best | +0.254 |
thrilled | +0.250 |
idea | +0.250 |
loyal | +0.236 |
Neg logit contributions | |
oes | -0.604 |
talk | -0.574 |
ted | -0.564 |
ricane | -0.544 |
rost | -0.543 |
Token | bath |
Token ID | 7837 |
Feature activation | -7.04e-04 |
Pos logit contributions | |
impressed | +0.312 |
adorable | +0.306 |
pleased | +0.303 |
thrilled | +0.294 |
intelligent | +0.284 |
Neg logit contributions | |
stack | -0.356 |
oes | -0.345 |
talk | -0.331 |
ted | -0.325 |
gra | -0.312 |
Token | robe |
Token ID | 25481 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
oes | +0.001 |
ted | +0.001 |
talk | +0.001 |
stack | +0.001 |
isk | +0.001 |
Neg logit contributions | |
adorable | -0.001 |
important | -0.001 |
precious | -0.001 |
idea | -0.001 |
best | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
adorable | +0.116 |
idea | +0.114 |
important | +0.111 |
best | +0.111 |
precious | +0.109 |
Neg logit contributions | |
ted | -0.121 |
oes | -0.120 |
talk | -0.119 |
stack | -0.115 |
ting | -0.110 |
Token | It |
Token ID | 632 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
adorable | +0.053 |
best | +0.051 |
important | +0.051 |
idea | +0.051 |
precious | +0.049 |
Neg logit contributions | |
oes | -0.042 |
talk | -0.042 |
ted | -0.040 |
stack | -0.037 |
rost | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
important | +0.297 |
adorable | +0.292 |
idea | +0.290 |
precious | +0.288 |
best | +0.288 |
Neg logit contributions | |
ted | -0.233 |
oes | -0.231 |
talk | -0.224 |
ting | -0.215 |
south | -0.214 |
Token | " |
Token ID | 366 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
adorable | +0.033 |
best | +0.030 |
beautiful | +0.030 |
important | +0.030 |
idea | +0.029 |
Neg logit contributions | |
oes | -0.044 |
talk | -0.044 |
ted | -0.041 |
ales | -0.041 |
stack | -0.040 |
Token | Hello |
Token ID | 15496 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
oes | +0.045 |
stack | +0.045 |
ted | +0.044 |
isk | +0.043 |
talk | +0.043 |
Neg logit contributions | |
adorable | -0.058 |
beautiful | -0.053 |
loyal | -0.053 |
precious | -0.052 |
important | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | -3.56e-02 |
Pos logit contributions | |
adorable | +0.070 |
important | +0.070 |
idea | +0.069 |
best | +0.067 |
precious | +0.066 |
Neg logit contributions | |
oes | -0.104 |
ted | -0.101 |
talk | -0.100 |
rost | -0.095 |
use | -0.095 |
Token | my |
Token ID | 616 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
oes | +0.079 |
talk | +0.077 |
ted | +0.076 |
rost | +0.073 |
stack | +0.073 |
Neg logit contributions | |
adorable | -0.048 |
idea | -0.047 |
best | -0.047 |
important | -0.046 |
precious | -0.044 |
Token | de |
Token ID | 390 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
impressed | +0.137 |
pleased | +0.125 |
thrilled | +0.117 |
important | +0.112 |
idea | +0.112 |
Neg logit contributions | |
ted | -0.246 |
oes | -0.246 |
ricanes | -0.243 |
isk | -0.243 |
rost | -0.241 |
Token | ars |
Token ID | 945 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
adorable | +0.539 |
best | +0.526 |
idea | +0.521 |
important | +0.520 |
precious | +0.505 |
Neg logit contributions | |
oes | -0.371 |
talk | -0.368 |
ted | -0.350 |
stack | -0.329 |
rost | -0.326 |
Token | ! |
Token ID | 0 |
Feature activation | +4.29e-03 |
Pos logit contributions | |
oes | +0.073 |
talk | +0.072 |
ted | +0.070 |
stack | +0.068 |
rost | +0.067 |
Neg logit contributions | |
adorable | -0.048 |
important | -0.046 |
best | -0.046 |
idea | -0.045 |
precious | -0.045 |
Token | What |
Token ID | 1867 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
adorable | +0.008 |
best | +0.007 |
loyal | +0.007 |
important | +0.007 |
precious | +0.007 |
Neg logit contributions | |
talk | -0.008 |
oes | -0.008 |
ales | -0.007 |
ted | -0.007 |
isk | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
oes | +0.020 |
talk | +0.018 |
ted | +0.017 |
gra | +0.017 |
rost | +0.016 |
Neg logit contributions | |
adorable | -0.040 |
pleased | -0.039 |
idea | -0.039 |
precious | -0.039 |
important | -0.039 |
Token | nice |
Token ID | 3621 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
adorable | +0.177 |
idea | +0.173 |
impressed | +0.169 |
pleased | +0.165 |
best | +0.165 |
Neg logit contributions | |
talk | -0.399 |
gra | -0.395 |
Dog | -0.389 |
oes | -0.389 |
use | -0.386 |
Token | ™ |
Token ID | 8151 |
Feature activation | -7.49e-03 |
Pos logit contributions | |
adorable | +0.218 |
important | +0.212 |
idea | +0.211 |
best | +0.211 |
precious | +0.205 |
Neg logit contributions | |
oes | -0.190 |
talk | -0.189 |
ted | -0.182 |
isk | -0.172 |
stack | -0.172 |
Token | re |
Token ID | 260 |
Feature activation | +9.05e-02 |
Pos logit contributions | |
oes | +0.013 |
talk | +0.013 |
ted | +0.013 |
stack | +0.012 |
isk | +0.012 |
Neg logit contributions | |
adorable | -0.014 |
idea | -0.013 |
important | -0.013 |
best | -0.013 |
precious | -0.013 |
Token | welcome |
Token ID | 7062 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
adorable | +0.130 |
idea | +0.127 |
precious | +0.125 |
prized | +0.124 |
important | +0.121 |
Neg logit contributions | |
oes | -0.169 |
isk | -0.166 |
stack | -0.164 |
talk | -0.164 |
gra | -0.164 |
Token | , |
Token ID | 11 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
adorable | +0.155 |
idea | +0.149 |
important | +0.149 |
best | +0.147 |
precious | +0.145 |
Neg logit contributions | |
oes | -0.246 |
talk | -0.246 |
ted | -0.243 |
stack | -0.230 |
ting | -0.227 |
Token | my |
Token ID | 616 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
oes | +0.022 |
ted | +0.022 |
talk | +0.022 |
rost | +0.020 |
stack | +0.020 |
Neg logit contributions | |
adorable | -0.022 |
idea | -0.022 |
best | -0.021 |
important | -0.021 |
precious | -0.021 |
Token | de |
Token ID | 390 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
adorable | +0.115 |
impressed | +0.115 |
idea | +0.114 |
important | +0.113 |
best | +0.108 |
Neg logit contributions | |
ted | -0.265 |
rost | -0.261 |
oes | -0.261 |
gra | -0.254 |
talk | -0.254 |
Token | ars |
Token ID | 945 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
adorable | +0.559 |
idea | +0.542 |
best | +0.541 |
important | +0.536 |
precious | +0.523 |
Neg logit contributions | |
talk | -0.351 |
oes | -0.344 |
ted | -0.330 |
rost | -0.305 |
use | -0.304 |
Token | . |
Token ID | 13 |
Feature activation | -2.88e-04 |
Pos logit contributions | |
oes | +0.100 |
talk | +0.099 |
ted | +0.097 |
stack | +0.093 |
rost | +0.092 |
Neg logit contributions | |
adorable | -0.062 |
best | -0.060 |
important | -0.059 |
idea | -0.059 |
precious | -0.057 |
Token | Enjoy |
Token ID | 18179 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
oes | +0.000 |
talk | +0.000 |
ted | +0.000 |
stack | +0.000 |
rost | +0.000 |
Neg logit contributions | |
adorable | -0.001 |
best | -0.001 |
important | -0.001 |
idea | -0.001 |
precious | -0.001 |
Token | your |
Token ID | 534 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
oes | +0.029 |
talk | +0.029 |
ted | +0.028 |
stack | +0.027 |
rost | +0.026 |
Neg logit contributions | |
adorable | -0.023 |
best | -0.022 |
important | -0.022 |
idea | -0.022 |
precious | -0.021 |
Token | cake |
Token ID | 12187 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
adorable | +0.018 |
best | +0.017 |
important | +0.017 |
impressed | +0.017 |
precious | +0.017 |
Neg logit contributions | |
oes | -0.021 |
ted | -0.020 |
gra | -0.020 |
talk | -0.020 |
use | -0.019 |
Token | That |
Token ID | 2504 |
Feature activation | +6.38e-02 |
Pos logit contributions | |
oes | +0.028 |
talk | +0.028 |
ted | +0.027 |
stack | +0.025 |
rost | +0.025 |
Neg logit contributions | |
adorable | -0.038 |
idea | -0.038 |
best | -0.038 |
important | -0.037 |
precious | -0.036 |
Token | 's |
Token ID | 338 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
pleased | +0.133 |
impressed | +0.128 |
importantly | +0.127 |
prized | +0.124 |
precious | +0.124 |
Neg logit contributions | |
oes | -0.099 |
talk | -0.091 |
ted | -0.090 |
stack | -0.082 |
gra | -0.082 |
Token | good |
Token ID | 922 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
prized | +0.077 |
adorable | +0.076 |
idea | +0.075 |
pleased | +0.074 |
thrilled | +0.073 |
Neg logit contributions | |
oes | -0.141 |
rost | -0.139 |
isk | -0.139 |
south | -0.134 |
talk | -0.132 |
Token | , |
Token ID | 11 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
prized | +0.290 |
pleased | +0.289 |
impressed | +0.288 |
adorable | +0.286 |
thrilled | +0.283 |
Neg logit contributions | |
oes | -0.269 |
talk | -0.240 |
ted | -0.239 |
rost | -0.215 |
use | -0.214 |
Token | my |
Token ID | 616 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
oes | +0.018 |
rost | +0.017 |
Rex | +0.016 |
stack | +0.016 |
ted | +0.015 |
Neg logit contributions | |
prized | -0.036 |
important | -0.034 |
valuable | -0.034 |
adorable | -0.034 |
precious | -0.033 |
Token | de |
Token ID | 390 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
prized | +0.167 |
adorable | +0.162 |
impressed | +0.162 |
thrilled | +0.160 |
pleased | +0.157 |
Neg logit contributions | |
oes | -0.209 |
rost | -0.196 |
ricanes | -0.195 |
ted | -0.187 |
isk | -0.181 |
Token | ars |
Token ID | 945 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
adorable | +0.537 |
best | +0.522 |
idea | +0.518 |
important | +0.517 |
precious | +0.503 |
Neg logit contributions | |
oes | -0.369 |
talk | -0.366 |
ted | -0.349 |
stack | -0.327 |
rost | -0.326 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
oes | +0.059 |
talk | +0.058 |
ted | +0.055 |
stack | +0.053 |
rost | +0.053 |
Neg logit contributions | |
adorable | -0.057 |
important | -0.055 |
best | -0.054 |
precious | -0.054 |
idea | -0.054 |
Token | But |
Token ID | 887 |
Feature activation | -5.66e-03 |
Pos logit contributions | |
adorable | +0.026 |
best | +0.025 |
important | +0.024 |
idea | +0.024 |
precious | +0.023 |
Neg logit contributions | |
talk | -0.030 |
oes | -0.030 |
ted | -0.029 |
isk | -0.028 |
stack | -0.027 |
Token | you |
Token ID | 345 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
talk | +0.011 |
oes | +0.011 |
ted | +0.010 |
stack | +0.010 |
isk | +0.010 |
Neg logit contributions | |
adorable | -0.010 |
best | -0.010 |
important | -0.009 |
idea | -0.009 |
precious | -0.009 |
Token | still |
Token ID | 991 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
adorable | +0.055 |
best | +0.054 |
important | +0.053 |
idea | +0.053 |
precious | +0.051 |
Neg logit contributions | |
oes | -0.051 |
talk | -0.050 |
ted | -0.048 |
rost | -0.046 |
stack | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
adorable | +0.103 |
best | +0.097 |
important | +0.097 |
idea | +0.095 |
precious | +0.094 |
Neg logit contributions | |
oes | -0.137 |
talk | -0.134 |
ted | -0.130 |
stack | -0.127 |
use | -0.123 |
Token | She |
Token ID | 1375 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
adorable | +0.035 |
important | +0.033 |
best | +0.033 |
idea | +0.033 |
precious | +0.032 |
Neg logit contributions | |
oes | -0.029 |
talk | -0.029 |
ted | -0.029 |
rost | -0.027 |
stack | -0.027 |
Token | had |
Token ID | 550 |
Feature activation | +2.17e-02 |
Pos logit contributions | |
talk | +0.044 |
ted | +0.042 |
oes | +0.042 |
stack | +0.039 |
isk | +0.039 |
Neg logit contributions | |
adorable | -0.055 |
best | -0.054 |
important | -0.052 |
idea | -0.052 |
precious | -0.051 |
Token | designed |
Token ID | 3562 |
Feature activation | -7.71e-04 |
Pos logit contributions | |
adorable | +0.039 |
best | +0.038 |
idea | +0.037 |
important | +0.037 |
precious | +0.036 |
Neg logit contributions | |
oes | -0.039 |
talk | -0.039 |
ted | -0.038 |
stack | -0.036 |
rost | -0.035 |
Token | a |
Token ID | 257 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
talk | +0.002 |
ted | +0.002 |
oes | +0.002 |
stack | +0.001 |
isk | +0.001 |
Neg logit contributions | |
adorable | -0.001 |
idea | -0.001 |
best | -0.001 |
important | -0.001 |
precious | -0.001 |
Token | de |
Token ID | 390 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
adorable | +0.170 |
idea | +0.153 |
best | +0.151 |
important | +0.149 |
precious | +0.140 |
Neg logit contributions | |
oes | -0.356 |
talk | -0.348 |
stack | -0.344 |
gra | -0.335 |
south | -0.331 |
Token | vious |
Token ID | 1442 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
adorable | +0.456 |
best | +0.443 |
idea | +0.438 |
important | +0.437 |
precious | +0.423 |
Neg logit contributions | |
oes | -0.448 |
talk | -0.446 |
ted | -0.430 |
stack | -0.408 |
rost | -0.404 |
Token | test |
Token ID | 1332 |
Feature activation | -9.20e-03 |
Pos logit contributions | |
adorable | +0.332 |
precious | +0.308 |
important | +0.307 |
best | +0.306 |
impressed | +0.298 |
Neg logit contributions | |
oes | -0.416 |
talk | -0.406 |
ted | -0.390 |
stack | -0.386 |
rost | -0.373 |
Token | ! |
Token ID | 0 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
oes | +0.017 |
talk | +0.017 |
ted | +0.017 |
stack | +0.016 |
rost | +0.016 |
Neg logit contributions | |
adorable | -0.016 |
best | -0.015 |
important | -0.015 |
idea | -0.015 |
precious | -0.015 |
Token | The |
Token ID | 383 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
adorable | +0.030 |
best | +0.029 |
important | +0.029 |
idea | +0.029 |
precious | +0.028 |
Neg logit contributions | |
oes | -0.025 |
talk | -0.025 |
ted | -0.024 |
stack | -0.023 |
rost | -0.023 |
Token | fairy |
Token ID | 25607 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
adorable | +0.077 |
important | +0.071 |
best | +0.070 |
idea | +0.070 |
impressed | +0.069 |
Neg logit contributions | |
oes | -0.110 |
talk | -0.110 |
ted | -0.106 |
stack | -0.105 |
use | -0.101 |
Token | bright |
Token ID | 6016 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
important | +0.221 |
adorable | +0.221 |
idea | +0.218 |
best | +0.213 |
intelligent | +0.204 |
Neg logit contributions | |
stack | -0.321 |
gra | -0.317 |
south | -0.313 |
oes | -0.310 |
talk | -0.309 |
Token | yellow |
Token ID | 7872 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
best | +0.266 |
important | +0.261 |
precious | +0.249 |
loyal | +0.245 |
intelligent | +0.244 |
Neg logit contributions | |
ted | -0.283 |
oes | -0.275 |
stack | -0.265 |
south | -0.264 |
talk | -0.263 |
Token | colour |
Token ID | 9568 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
important | +0.299 |
adorable | +0.298 |
best | +0.293 |
precious | +0.284 |
impressed | +0.284 |
Neg logit contributions | |
oes | -0.283 |
ted | -0.283 |
talk | -0.276 |
stack | -0.266 |
ting | -0.246 |
Token | and |
Token ID | 290 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
adorable | +0.157 |
important | +0.152 |
best | +0.152 |
idea | +0.150 |
precious | +0.147 |
Neg logit contributions | |
oes | -0.177 |
talk | -0.171 |
ted | -0.170 |
stack | -0.160 |
ting | -0.157 |
Token | a |
Token ID | 257 |
Feature activation | +1.85e-01 |
Pos logit contributions | |
ted | +0.065 |
benches | +0.058 |
talk | +0.058 |
oes | +0.058 |
gra | +0.057 |
Neg logit contributions | |
idea | -0.041 |
important | -0.038 |
best | -0.038 |
adorable | -0.037 |
prized | -0.037 |
Token | fr |
Token ID | 1216 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
important | +0.235 |
adorable | +0.233 |
idea | +0.228 |
best | +0.222 |
intelligent | +0.212 |
Neg logit contributions | |
stack | -0.361 |
gra | -0.360 |
oes | -0.359 |
talk | -0.355 |
south | -0.355 |
Token | illy |
Token ID | 6548 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
best | +0.393 |
adorable | +0.385 |
idea | +0.383 |
important | +0.375 |
loyal | +0.372 |
Neg logit contributions | |
oes | -0.515 |
talk | -0.501 |
ted | -0.496 |
rost | -0.459 |
ales | -0.456 |
Token | style |
Token ID | 3918 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
adorable | +0.171 |
important | +0.169 |
best | +0.169 |
idea | +0.167 |
precious | +0.162 |
Neg logit contributions | |
oes | -0.192 |
talk | -0.190 |
ted | -0.189 |
stack | -0.180 |
rost | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
adorable | +0.216 |
important | +0.211 |
best | +0.209 |
loyal | +0.208 |
precious | +0.206 |
Neg logit contributions | |
oes | -0.190 |
ted | -0.183 |
talk | -0.183 |
stack | -0.167 |
ting | -0.163 |
Token | She |
Token ID | 1375 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
adorable | +0.029 |
best | +0.028 |
important | +0.027 |
idea | +0.027 |
beautiful | +0.027 |
Neg logit contributions | |
oes | -0.033 |
talk | -0.032 |
rost | -0.031 |
isk | -0.030 |
ales | -0.029 |
Token | tried |
Token ID | 3088 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
ted | +0.043 |
talk | +0.041 |
oes | +0.038 |
walked | +0.036 |
approached | +0.035 |
Neg logit contributions | |
idea | -0.062 |
best | -0.061 |
important | -0.061 |
adorable | -0.061 |
precious | -0.059 |
Token | novel |
Token ID | 5337 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
idea | +0.267 |
adorable | +0.255 |
important | +0.255 |
best | +0.253 |
precious | +0.230 |
Neg logit contributions | |
stack | -0.393 |
oes | -0.382 |
gra | -0.381 |
ted | -0.378 |
ricane | -0.375 |
Token | ! |
Token ID | 0 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
oes | +0.018 |
ted | +0.018 |
talk | +0.017 |
stack | +0.017 |
ting | +0.017 |
Neg logit contributions | |
adorable | -0.011 |
best | -0.011 |
important | -0.011 |
idea | -0.011 |
precious | -0.010 |
Token | It |
Token ID | 632 |
Feature activation | -3.27e-02 |
Pos logit contributions | |
adorable | +0.041 |
best | +0.039 |
idea | +0.039 |
important | +0.038 |
beautiful | +0.038 |
Neg logit contributions | |
talk | -0.039 |
oes | -0.039 |
stack | -0.037 |
rost | -0.036 |
ted | -0.036 |
Token | was |
Token ID | 373 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
ted | +0.053 |
talk | +0.051 |
oes | +0.051 |
stack | +0.048 |
isk | +0.047 |
Neg logit contributions | |
best | -0.061 |
idea | -0.061 |
adorable | -0.060 |
important | -0.059 |
precious | -0.058 |
Token | a |
Token ID | 257 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
idea | +0.189 |
adorable | +0.186 |
best | +0.185 |
important | +0.178 |
precious | +0.176 |
Neg logit contributions | |
oes | -0.282 |
ted | -0.282 |
stack | -0.275 |
talk | -0.271 |
rost | -0.268 |
Token | sweet |
Token ID | 6029 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
idea | +0.240 |
best | +0.231 |
adorable | +0.228 |
important | +0.219 |
mistakes | +0.206 |
Neg logit contributions | |
stack | -0.443 |
oes | -0.437 |
rost | -0.430 |
gra | -0.427 |
ricane | -0.425 |
Token | novel |
Token ID | 5337 |
Feature activation | +1.39e-03 |
Pos logit contributions | |
best | +0.426 |
impressed | +0.423 |
important | +0.408 |
vern | +0.404 |
precious | +0.401 |
Neg logit contributions | |
ted | -0.401 |
oes | -0.394 |
rost | -0.376 |
talk | -0.370 |
ting | -0.360 |
Token | with |
Token ID | 351 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
adorable | +0.002 |
idea | +0.002 |
important | +0.002 |
best | +0.002 |
precious | +0.002 |
Neg logit contributions | |
oes | -0.003 |
stack | -0.003 |
ted | -0.003 |
talk | -0.003 |
ting | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
idea | +0.045 |
important | +0.042 |
adorable | +0.042 |
best | +0.042 |
precious | +0.040 |
Neg logit contributions | |
ted | -0.068 |
talk | -0.066 |
oes | -0.065 |
ting | -0.062 |
stack | -0.060 |
Token | bright |
Token ID | 6016 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
best | +0.226 |
adorable | +0.226 |
idea | +0.224 |
important | +0.220 |
intelligent | +0.202 |
Neg logit contributions | |
talk | -0.317 |
oes | -0.310 |
stack | -0.310 |
ted | -0.307 |
gra | -0.304 |
Token | pink |
Token ID | 11398 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
best | +0.264 |
important | +0.254 |
precious | +0.247 |
adorable | +0.240 |
impressed | +0.239 |
Neg logit contributions | |
ted | -0.296 |
oes | -0.289 |
talk | -0.286 |
stack | -0.269 |
towards | -0.265 |
Token | amazing |
Token ID | 4998 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
adorable | +0.099 |
best | +0.096 |
idea | +0.095 |
important | +0.092 |
precious | +0.089 |
Neg logit contributions | |
oes | -0.179 |
ted | -0.176 |
talk | -0.175 |
stack | -0.168 |
rost | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
adorable | +0.193 |
best | +0.186 |
important | +0.183 |
idea | +0.182 |
precious | +0.175 |
Neg logit contributions | |
oes | -0.300 |
talk | -0.297 |
ted | -0.291 |
stack | -0.279 |
rost | -0.274 |
Token | It |
Token ID | 632 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
adorable | +0.022 |
best | +0.021 |
idea | +0.021 |
important | +0.021 |
beautiful | +0.020 |
Neg logit contributions | |
talk | -0.021 |
oes | -0.021 |
ted | -0.020 |
isk | -0.019 |
use | -0.019 |
Token | was |
Token ID | 373 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
oes | +0.061 |
ted | +0.060 |
talk | +0.059 |
stack | +0.057 |
isk | +0.055 |
Neg logit contributions | |
adorable | -0.061 |
best | -0.059 |
idea | -0.059 |
important | -0.059 |
precious | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
adorable | +0.189 |
idea | +0.183 |
best | +0.181 |
important | +0.178 |
precious | +0.174 |
Neg logit contributions | |
oes | -0.291 |
ted | -0.281 |
stack | -0.278 |
talk | -0.277 |
rost | -0.271 |
Token | brand |
Token ID | 4508 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
adorable | +0.272 |
idea | +0.265 |
best | +0.256 |
important | +0.254 |
intelligent | +0.236 |
Neg logit contributions | |
stack | -0.447 |
oes | -0.439 |
gra | -0.422 |
rost | -0.416 |
south | -0.416 |
Token | new |
Token ID | 649 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
adorable | +0.261 |
best | +0.243 |
thrilled | +0.233 |
idea | +0.230 |
loyal | +0.223 |
Neg logit contributions | |
oes | -0.605 |
talk | -0.569 |
ted | -0.557 |
rost | -0.555 |
iffs | -0.551 |
Token | toy |
Token ID | 13373 |
Feature activation | +8.10e-02 |
Pos logit contributions | |
adorable | +0.333 |
pleased | +0.329 |
impressed | +0.328 |
important | +0.313 |
thrilled | +0.310 |
Neg logit contributions | |
stack | -0.335 |
oes | -0.327 |
talk | -0.300 |
hunt | -0.286 |
towards | -0.283 |
Token | . |
Token ID | 13 |
Feature activation | +1.79e-02 |
Pos logit contributions | |
adorable | +0.133 |
important | +0.128 |
idea | +0.127 |
best | +0.125 |
precious | +0.124 |
Neg logit contributions | |
oes | -0.155 |
ted | -0.152 |
stack | -0.150 |
talk | -0.149 |
ting | -0.141 |
Token | They |
Token ID | 1119 |
Feature activation | -4.26e-02 |
Pos logit contributions | |
adorable | +0.034 |
best | +0.033 |
idea | +0.033 |
important | +0.032 |
precious | +0.031 |
Neg logit contributions | |
talk | -0.030 |
oes | -0.030 |
ted | -0.029 |
isk | -0.027 |
rost | -0.027 |
Token | both |
Token ID | 1111 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
oes | +0.075 |
talk | +0.074 |
ted | +0.071 |
stack | +0.067 |
rost | +0.067 |
Neg logit contributions | |
adorable | -0.080 |
best | -0.078 |
important | -0.077 |
idea | -0.077 |
precious | -0.075 |
Token | am |
Token ID | 716 |
Feature activation | +8.76e-02 |
Pos logit contributions | |
adorable | +0.028 |
best | +0.026 |
important | +0.026 |
idea | +0.026 |
precious | +0.025 |
Neg logit contributions | |
talk | -0.027 |
oes | -0.027 |
ted | -0.026 |
rost | -0.025 |
isk | -0.024 |
Token | not |
Token ID | 407 |
Feature activation | +9.52e-02 |
Pos logit contributions | |
idea | +0.109 |
adorable | +0.102 |
prized | +0.099 |
precious | +0.097 |
best | +0.095 |
Neg logit contributions | |
oes | -0.171 |
stack | -0.170 |
gra | -0.170 |
talk | -0.170 |
Enter | -0.170 |
Token | mad |
Token ID | 8805 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
adorable | +0.112 |
prized | +0.110 |
precious | +0.107 |
idea | +0.106 |
best | +0.105 |
Neg logit contributions | |
oes | -0.206 |
talk | -0.205 |
use | -0.193 |
rost | -0.190 |
ting | -0.190 |
Token | , |
Token ID | 11 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
adorable | +0.283 |
precious | +0.272 |
idea | +0.271 |
important | +0.271 |
best | +0.263 |
Neg logit contributions | |
oes | -0.305 |
ted | -0.290 |
talk | -0.290 |
ting | -0.281 |
rost | -0.266 |
Token | my |
Token ID | 616 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
ted | +0.029 |
oes | +0.028 |
talk | +0.027 |
ting | +0.026 |
rost | +0.026 |
Neg logit contributions | |
adorable | -0.043 |
important | -0.043 |
idea | -0.043 |
best | -0.042 |
prized | -0.042 |
Token | de |
Token ID | 390 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
adorable | +0.138 |
prized | +0.131 |
important | +0.129 |
impressed | +0.129 |
best | +0.129 |
Neg logit contributions | |
oes | -0.242 |
talk | -0.236 |
rost | -0.233 |
ted | -0.231 |
isk | -0.219 |
Token | ars |
Token ID | 945 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
adorable | +0.516 |
best | +0.504 |
important | +0.499 |
idea | +0.498 |
precious | +0.485 |
Neg logit contributions | |
oes | -0.379 |
talk | -0.374 |
ted | -0.360 |
stack | -0.340 |
rost | -0.334 |
Token | . |
Token ID | 13 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
oes | +0.077 |
talk | +0.077 |
ted | +0.074 |
stack | +0.071 |
rost | +0.071 |
Neg logit contributions | |
adorable | -0.052 |
best | -0.052 |
idea | -0.050 |
important | -0.050 |
precious | -0.047 |
Token | I |
Token ID | 314 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
adorable | +0.018 |
best | +0.017 |
important | +0.017 |
precious | +0.016 |
loyal | +0.016 |
Neg logit contributions | |
talk | -0.022 |
oes | -0.022 |
ted | -0.021 |
isk | -0.020 |
stack | -0.020 |
Token | am |
Token ID | 716 |
Feature activation | +9.32e-02 |
Pos logit contributions | |
talk | +0.028 |
oes | +0.028 |
ted | +0.027 |
rost | +0.025 |
stack | +0.025 |
Neg logit contributions | |
adorable | -0.030 |
best | -0.030 |
important | -0.029 |
idea | -0.029 |
precious | -0.028 |
Token | glad |
Token ID | 9675 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
idea | +0.107 |
adorable | +0.101 |
best | +0.097 |
important | +0.092 |
precious | +0.091 |
Neg logit contributions | |
oes | -0.208 |
talk | -0.207 |
ted | -0.200 |
gra | -0.200 |
isk | -0.199 |
Token | tigers |
Token ID | 43114 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
oes | +0.078 |
talk | +0.077 |
ted | +0.073 |
stack | +0.070 |
rost | +0.069 |
Neg logit contributions | |
adorable | -0.110 |
best | -0.106 |
important | -0.105 |
idea | -0.104 |
precious | -0.102 |
Token | , |
Token ID | 11 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
talk | +0.077 |
oes | +0.077 |
rost | +0.073 |
ricanes | +0.072 |
ted | +0.072 |
Neg logit contributions | |
adorable | -0.037 |
best | -0.036 |
idea | -0.035 |
important | -0.033 |
beautiful | -0.033 |
Token | but |
Token ID | 475 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
oes | +0.043 |
talk | +0.042 |
ted | +0.040 |
stack | +0.038 |
rost | +0.038 |
Neg logit contributions | |
adorable | -0.051 |
best | -0.050 |
important | -0.049 |
idea | -0.049 |
precious | -0.047 |
Token | he |
Token ID | 339 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
talk | +0.062 |
oes | +0.061 |
ted | +0.057 |
iffs | +0.055 |
isk | +0.054 |
Neg logit contributions | |
adorable | -0.071 |
best | -0.070 |
important | -0.068 |
idea | -0.066 |
loyal | -0.065 |
Token | was |
Token ID | 373 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
oes | +0.051 |
talk | +0.051 |
ted | +0.050 |
stack | +0.048 |
isk | +0.047 |
Neg logit contributions | |
adorable | -0.045 |
best | -0.044 |
idea | -0.044 |
important | -0.043 |
precious | -0.042 |
Token | most |
Token ID | 749 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
idea | +0.201 |
adorable | +0.197 |
best | +0.191 |
important | +0.187 |
precious | +0.185 |
Neg logit contributions | |
oes | -0.299 |
talk | -0.293 |
stack | -0.288 |
ted | -0.288 |
isk | -0.285 |
Token | excited |
Token ID | 6568 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
idea | +0.237 |
best | +0.212 |
darling | +0.173 |
kin | +0.160 |
adorable | +0.156 |
Neg logit contributions | |
ted | -0.528 |
ting | -0.526 |
towards | -0.508 |
oes | -0.503 |
talk | -0.501 |
Token | to |
Token ID | 284 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
adorable | +0.395 |
best | +0.385 |
important | +0.381 |
idea | +0.379 |
precious | +0.368 |
Neg logit contributions | |
oes | -0.360 |
talk | -0.356 |
ted | -0.343 |
stack | -0.327 |
rost | -0.322 |
Token | see |
Token ID | 766 |
Feature activation | -5.20e-03 |
Pos logit contributions | |
ted | +0.207 |
oes | +0.198 |
talk | +0.198 |
ricanes | +0.186 |
ting | +0.183 |
Neg logit contributions | |
idea | -0.241 |
adorable | -0.231 |
best | -0.230 |
important | -0.222 |
beautiful | -0.217 |
Token | the |
Token ID | 262 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
oes | +0.009 |
talk | +0.009 |
rost | +0.008 |
stack | +0.008 |
ted | +0.008 |
Neg logit contributions | |
adorable | -0.010 |
best | -0.010 |
important | -0.010 |
beautiful | -0.009 |
idea | -0.009 |
Token | elephants |
Token ID | 25368 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
adorable | +0.051 |
idea | +0.050 |
important | +0.050 |
best | +0.049 |
precious | +0.048 |
Neg logit contributions | |
oes | -0.041 |
talk | -0.040 |
ted | -0.039 |
isk | -0.037 |
stack | -0.036 |
Token | are |
Token ID | 389 |
Feature activation | +8.26e-02 |
Pos logit contributions | |
adorable | +0.067 |
best | +0.065 |
important | +0.065 |
idea | +0.064 |
precious | +0.063 |
Neg logit contributions | |
oes | -0.053 |
talk | -0.052 |
ted | -0.050 |
stack | -0.046 |
rost | -0.046 |
Token | not |
Token ID | 407 |
Feature activation | +8.83e-02 |
Pos logit contributions | |
idea | +0.110 |
adorable | +0.109 |
best | +0.103 |
important | +0.101 |
loyal | +0.097 |
Neg logit contributions | |
oes | -0.181 |
ted | -0.173 |
talk | -0.171 |
stack | -0.169 |
rost | -0.167 |
Token | flowers |
Token ID | 12734 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
adorable | +0.131 |
idea | +0.127 |
best | +0.127 |
prized | +0.122 |
precious | +0.120 |
Neg logit contributions | |
oes | -0.172 |
talk | -0.164 |
ted | -0.159 |
stack | -0.158 |
isk | -0.158 |
Token | , |
Token ID | 11 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
oes | +0.193 |
talk | +0.188 |
ted | +0.185 |
stack | +0.177 |
rost | +0.174 |
Neg logit contributions | |
adorable | -0.141 |
best | -0.135 |
important | -0.135 |
idea | -0.134 |
precious | -0.129 |
Token | my |
Token ID | 616 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
oes | +0.053 |
ted | +0.053 |
rost | +0.052 |
ting | +0.050 |
stack | +0.048 |
Neg logit contributions | |
prized | -0.049 |
adorable | -0.049 |
idea | -0.048 |
best | -0.047 |
important | -0.047 |
Token | de |
Token ID | 390 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
adorable | +0.175 |
best | +0.170 |
important | +0.169 |
idea | +0.168 |
loyal | +0.162 |
Neg logit contributions | |
oes | -0.195 |
ted | -0.189 |
talk | -0.184 |
isk | -0.178 |
stack | -0.177 |
Token | ars |
Token ID | 945 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
adorable | +0.512 |
best | +0.501 |
important | +0.496 |
idea | +0.494 |
precious | +0.482 |
Neg logit contributions | |
oes | -0.380 |
talk | -0.375 |
ted | -0.361 |
stack | -0.341 |
rost | -0.335 |
Token | . |
Token ID | 13 |
Feature activation | +5.24e-03 |
Pos logit contributions | |
oes | +0.076 |
talk | +0.074 |
ted | +0.072 |
stack | +0.069 |
rost | +0.068 |
Neg logit contributions | |
adorable | -0.058 |
best | -0.056 |
important | -0.056 |
idea | -0.055 |
precious | -0.054 |
Token | Those |
Token ID | 5845 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
adorable | +0.010 |
best | +0.010 |
important | +0.010 |
precious | +0.010 |
idea | +0.010 |
Neg logit contributions | |
talk | -0.009 |
oes | -0.008 |
ted | -0.008 |
isk | -0.008 |
rost | -0.007 |
Token | are |
Token ID | 389 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
adorable | +0.068 |
best | +0.066 |
important | +0.065 |
idea | +0.065 |
precious | +0.063 |
Neg logit contributions | |
oes | -0.058 |
talk | -0.057 |
ted | -0.055 |
stack | -0.053 |
rost | -0.052 |
Token | caul |
Token ID | 46087 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
adorable | +0.142 |
best | +0.138 |
important | +0.137 |
idea | +0.132 |
precious | +0.132 |
Neg logit contributions | |
talk | -0.192 |
oes | -0.191 |
ted | -0.185 |
stack | -0.177 |
rost | -0.174 |
Token |
|
Token ID | 198 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
oes | +0.055 |
talk | +0.054 |
ted | +0.054 |
stack | +0.051 |
ting | +0.050 |
Neg logit contributions | |
adorable | -0.052 |
important | -0.050 |
best | -0.049 |
idea | -0.049 |
beautiful | -0.048 |
Token | " |
Token ID | 1 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
oes | +0.063 |
talk | +0.061 |
ted | +0.061 |
stack | +0.058 |
isk | +0.057 |
Neg logit contributions | |
adorable | -0.032 |
important | -0.030 |
best | -0.029 |
precious | -0.029 |
beautiful | -0.029 |
Token | Hello |
Token ID | 15496 |
Feature activation | +4.54e-02 |
Pos logit contributions | |
oes | +0.034 |
stack | +0.032 |
talk | +0.031 |
isk | +0.031 |
ted | +0.031 |
Neg logit contributions | |
adorable | -0.047 |
best | -0.045 |
important | -0.044 |
beautiful | -0.044 |
precious | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
important | +0.069 |
adorable | +0.069 |
idea | +0.067 |
precious | +0.065 |
best | +0.064 |
Neg logit contributions | |
oes | -0.090 |
ted | -0.089 |
talk | -0.086 |
rost | -0.084 |
use | -0.083 |
Token | my |
Token ID | 616 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
oes | +0.078 |
talk | +0.076 |
ted | +0.076 |
rost | +0.073 |
stack | +0.072 |
Neg logit contributions | |
adorable | -0.052 |
important | -0.051 |
idea | -0.051 |
best | -0.051 |
precious | -0.048 |
Token | de |
Token ID | 390 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
impressed | +0.142 |
pleased | +0.133 |
prized | +0.130 |
important | +0.130 |
thrilled | +0.127 |
Neg logit contributions | |
ted | -0.224 |
oes | -0.223 |
rost | -0.221 |
isk | -0.220 |
iffs | -0.210 |
Token | ars |
Token ID | 945 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
adorable | +0.511 |
best | +0.500 |
important | +0.496 |
idea | +0.492 |
precious | +0.482 |
Neg logit contributions | |
oes | -0.380 |
talk | -0.374 |
ted | -0.361 |
stack | -0.342 |
rost | -0.334 |
Token | . |
Token ID | 13 |
Feature activation | +7.69e-03 |
Pos logit contributions | |
oes | +0.077 |
talk | +0.076 |
ted | +0.073 |
stack | +0.071 |
use | +0.070 |
Neg logit contributions | |
adorable | -0.061 |
important | -0.059 |
precious | -0.058 |
best | -0.058 |
idea | -0.057 |
Token | What |
Token ID | 1867 |
Feature activation | -2.23e-02 |
Pos logit contributions | |
adorable | +0.013 |
best | +0.012 |
loyal | +0.011 |
important | +0.011 |
precious | +0.011 |
Neg logit contributions | |
talk | -0.015 |
oes | -0.015 |
ted | -0.014 |
ales | -0.014 |
isk | -0.014 |
Token | do |
Token ID | 466 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
oes | +0.027 |
talk | +0.023 |
gra | +0.022 |
ted | +0.022 |
rost | +0.022 |
Neg logit contributions | |
pleased | -0.048 |
adorable | -0.048 |
idea | -0.047 |
prized | -0.047 |
precious | -0.047 |
Token | you |
Token ID | 345 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
adorable | +0.031 |
best | +0.030 |
important | +0.030 |
idea | +0.029 |
precious | +0.028 |
Neg logit contributions | |
talk | -0.039 |
oes | -0.039 |
ted | -0.038 |
stack | -0.037 |
isk | -0.036 |
Token |
|
Token ID | 198 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
oes | +0.059 |
talk | +0.058 |
ted | +0.057 |
stack | +0.054 |
ting | +0.053 |
Neg logit contributions | |
adorable | -0.060 |
important | -0.058 |
best | -0.058 |
idea | -0.057 |
precious | -0.056 |
Token | " |
Token ID | 1 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
oes | +0.067 |
talk | +0.065 |
ted | +0.065 |
stack | +0.061 |
isk | +0.061 |
Neg logit contributions | |
adorable | -0.035 |
important | -0.034 |
best | -0.033 |
precious | -0.032 |
beautiful | -0.032 |
Token | Hello |
Token ID | 15496 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
oes | +0.037 |
stack | +0.033 |
isk | +0.032 |
ted | +0.032 |
ales | +0.031 |
Neg logit contributions | |
adorable | -0.045 |
beautiful | -0.043 |
best | -0.042 |
precious | -0.041 |
loyal | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
important | +0.073 |
adorable | +0.071 |
idea | +0.069 |
valuable | +0.069 |
precious | +0.068 |
Neg logit contributions | |
oes | -0.085 |
ted | -0.082 |
talk | -0.080 |
rost | -0.078 |
use | -0.077 |
Token | my |
Token ID | 616 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
oes | +0.075 |
talk | +0.073 |
ted | +0.072 |
rost | +0.070 |
stack | +0.069 |
Neg logit contributions | |
adorable | -0.056 |
idea | -0.056 |
important | -0.055 |
best | -0.055 |
precious | -0.052 |
Token | de |
Token ID | 390 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
impressed | +0.150 |
pleased | +0.145 |
prized | +0.136 |
thrilled | +0.134 |
motivated | +0.132 |
Neg logit contributions | |
Shelter | -0.200 |
ted | -0.199 |
isk | -0.195 |
ricanes | -0.194 |
oes | -0.193 |
Token | ars |
Token ID | 945 |
Feature activation | -3.65e-02 |
Pos logit contributions | |
adorable | +0.518 |
best | +0.506 |
important | +0.501 |
idea | +0.499 |
precious | +0.487 |
Neg logit contributions | |
oes | -0.372 |
talk | -0.367 |
ted | -0.353 |
stack | -0.333 |
rost | -0.327 |
Token | ," |
Token ID | 553 |
Feature activation | -8.86e-03 |
Pos logit contributions | |
talk | +0.081 |
oes | +0.081 |
ted | +0.077 |
stack | +0.076 |
use | +0.075 |
Neg logit contributions | |
adorable | -0.049 |
important | -0.048 |
precious | -0.046 |
best | -0.046 |
idea | -0.044 |
Token | she |
Token ID | 673 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
oes | +0.013 |
talk | +0.013 |
ted | +0.012 |
stack | +0.011 |
rost | +0.011 |
Neg logit contributions | |
adorable | -0.019 |
best | -0.019 |
important | -0.019 |
idea | -0.019 |
precious | -0.018 |
Token | said |
Token ID | 531 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
talk | +0.051 |
ted | +0.050 |
oes | +0.048 |
Dog | +0.045 |
isk | +0.044 |
Neg logit contributions | |
best | -0.050 |
adorable | -0.049 |
important | -0.049 |
precious | -0.048 |
idea | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
adorable | +0.161 |
best | +0.157 |
idea | +0.153 |
important | +0.152 |
precious | +0.145 |
Neg logit contributions | |
oes | -0.225 |
talk | -0.222 |
ted | -0.215 |
stack | -0.210 |
isk | -0.208 |
Token | Peter |
Token ID | 19727 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
oes | +0.222 |
talk | +0.218 |
ted | +0.199 |
rost | +0.198 |
isk | +0.197 |
Neg logit contributions | |
best | -0.268 |
adorable | -0.264 |
idea | -0.257 |
important | -0.252 |
beautiful | -0.248 |
Token | was |
Token ID | 373 |
Feature activation | +1.13e-01 |
Pos logit contributions | |
oes | +0.128 |
talk | +0.127 |
ted | +0.123 |
stack | +0.117 |
rost | +0.116 |
Neg logit contributions | |
adorable | -0.115 |
best | -0.112 |
important | -0.111 |
idea | -0.110 |
precious | -0.107 |
Token | a |
Token ID | 257 |
Feature activation | +1.93e-01 |
Pos logit contributions | |
precious | +0.189 |
idea | +0.182 |
adorable | +0.181 |
valuable | +0.179 |
important | +0.176 |
Neg logit contributions | |
ted | -0.184 |
oes | -0.182 |
stack | -0.182 |
gra | -0.180 |
talk | -0.180 |
Token | mis |
Token ID | 2984 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
idea | +0.217 |
mistakes | +0.199 |
important | +0.196 |
best | +0.191 |
vern | +0.189 |
Neg logit contributions | |
oes | -0.409 |
Dog | -0.401 |
gra | -0.401 |
rost | -0.393 |
stack | -0.389 |
Token | chie |
Token ID | 3043 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
best | +0.315 |
idea | +0.303 |
adorable | +0.301 |
important | +0.300 |
precious | +0.291 |
Neg logit contributions | |
talk | -0.229 |
oes | -0.216 |
Dog | -0.182 |
isk | -0.182 |
ting | -0.177 |
Token | vous |
Token ID | 31222 |
Feature activation | +2.45e-01 |
Pos logit contributions | |
best | +0.247 |
idea | +0.243 |
precious | +0.242 |
adorable | +0.242 |
important | +0.230 |
Neg logit contributions | |
oes | -0.297 |
talk | -0.291 |
ted | -0.268 |
gra | -0.254 |
ting | -0.252 |
Token | 3 |
Token ID | 513 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
important | +0.358 |
precious | +0.349 |
adorable | +0.344 |
best | +0.343 |
idea | +0.335 |
Neg logit contributions | |
oes | -0.481 |
ted | -0.466 |
talk | -0.461 |
ting | -0.458 |
stack | -0.446 |
Token | year |
Token ID | 614 |
Feature activation | +8.30e-02 |
Pos logit contributions | |
precious | +0.253 |
best | +0.252 |
adorable | +0.248 |
important | +0.239 |
idea | +0.235 |
Neg logit contributions | |
talk | -0.325 |
ted | -0.313 |
oes | -0.303 |
ting | -0.299 |
rost | -0.288 |
Token | old |
Token ID | 1468 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
best | +0.071 |
adorable | +0.070 |
idea | +0.068 |
important | +0.062 |
precious | +0.061 |
Neg logit contributions | |
oes | -0.224 |
talk | -0.220 |
ted | -0.215 |
stack | -0.207 |
rost | -0.206 |
Token | who |
Token ID | 508 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
adorable | +0.276 |
important | +0.272 |
best | +0.269 |
idea | +0.268 |
precious | +0.266 |
Neg logit contributions | |
oes | -0.285 |
talk | -0.285 |
ted | -0.283 |
stack | -0.269 |
ting | -0.266 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
oes | +0.048 |
talk | +0.047 |
ted | +0.046 |
stack | +0.044 |
rost | +0.043 |
Neg logit contributions | |
adorable | -0.040 |
best | -0.039 |
idea | -0.038 |
important | -0.038 |
precious | -0.037 |
Token | on |
Token ID | 319 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
adorable | +0.083 |
best | +0.080 |
idea | +0.079 |
important | +0.078 |
precious | +0.075 |
Neg logit contributions | |
oes | -0.107 |
talk | -0.107 |
ted | -0.101 |
rost | -0.099 |
isk | -0.098 |
Token | her |
Token ID | 607 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
oes | +0.241 |
talk | +0.237 |
ted | +0.225 |
isk | +0.215 |
rost | +0.214 |
Neg logit contributions | |
adorable | -0.259 |
best | -0.250 |
idea | -0.249 |
important | -0.247 |
beautiful | -0.239 |
Token | face |
Token ID | 1986 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
adorable | +0.028 |
important | +0.027 |
best | +0.027 |
idea | +0.026 |
precious | +0.026 |
Neg logit contributions | |
oes | -0.032 |
talk | -0.031 |
ted | -0.030 |
stack | -0.030 |
rost | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +3.47e-02 |
Pos logit contributions | |
oes | +0.086 |
talk | +0.086 |
ted | +0.082 |
rost | +0.080 |
stack | +0.079 |
Neg logit contributions | |
adorable | -0.061 |
best | -0.059 |
important | -0.058 |
idea | -0.057 |
precious | -0.055 |
Token | Ready |
Token ID | 23432 |
Feature activation | -6.97e-02 |
Pos logit contributions | |
adorable | +0.068 |
best | +0.065 |
important | +0.064 |
idea | +0.064 |
loyal | +0.062 |
Neg logit contributions | |
talk | -0.059 |
oes | -0.058 |
ted | -0.055 |
rost | -0.052 |
isk | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | -2.71e-01 |
Pos logit contributions | |
talk | +0.157 |
Dog | +0.146 |
ted | +0.146 |
oes | +0.145 |
stack | +0.143 |
Neg logit contributions | |
best | -0.095 |
adorable | -0.091 |
idea | -0.089 |
important | -0.085 |
beautiful | -0.081 |
Token | greet |
Token ID | 12589 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
talk | +0.461 |
ted | +0.444 |
isk | +0.443 |
oes | +0.439 |
rost | +0.432 |
Neg logit contributions | |
idea | -0.529 |
beautiful | -0.510 |
adorable | -0.509 |
best | -0.506 |
amazing | -0.488 |
Token | her |
Token ID | 607 |
Feature activation | -3.42e-05 |
Pos logit contributions | |
oes | +0.137 |
talk | +0.137 |
ted | +0.126 |
stack | +0.125 |
rost | +0.123 |
Neg logit contributions | |
adorable | -0.131 |
best | -0.127 |
important | -0.123 |
idea | -0.121 |
beautiful | -0.121 |
Token | friend |
Token ID | 1545 |
Feature activation | -7.36e-02 |
Pos logit contributions | |
oes | +0.000 |
talk | +0.000 |
stack | +0.000 |
ted | +0.000 |
gra | +0.000 |
Neg logit contributions | |
adorable | -0.000 |
important | -0.000 |
precious | -0.000 |
idea | -0.000 |
best | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
oes | +0.154 |
talk | +0.153 |
ted | +0.148 |
stack | +0.143 |
rost | +0.142 |
Neg logit contributions | |
adorable | -0.111 |
best | -0.109 |
important | -0.106 |
idea | -0.106 |
precious | -0.101 |
Token | Molly |
Token ID | 30236 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
oes | +0.120 |
talk | +0.120 |
ted | +0.115 |
rost | +0.108 |
stack | +0.108 |
Neg logit contributions | |
adorable | -0.119 |
best | -0.116 |
idea | -0.114 |
important | -0.114 |
precious | -0.109 |
Token | apped |
Token ID | 6320 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
important | +0.248 |
adorable | +0.236 |
idea | +0.235 |
precious | +0.233 |
best | +0.233 |
Neg logit contributions | |
ted | -0.194 |
talk | -0.189 |
oes | -0.183 |
iffs | -0.168 |
aled | -0.161 |
Token | and |
Token ID | 290 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
adorable | +0.113 |
best | +0.112 |
precious | +0.111 |
important | +0.111 |
idea | +0.110 |
Neg logit contributions | |
oes | -0.070 |
ted | -0.069 |
talk | -0.068 |
south | -0.063 |
ting | -0.061 |
Token | shouted |
Token ID | 17293 |
Feature activation | +7.47e-02 |
Pos logit contributions | |
ted | +0.097 |
talk | +0.091 |
ting | +0.085 |
oes | +0.085 |
stack | +0.081 |
Neg logit contributions | |
adorable | -0.104 |
important | -0.103 |
idea | -0.102 |
prized | -0.100 |
precious | -0.100 |
Token | " |
Token ID | 366 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
adorable | +0.126 |
idea | +0.122 |
best | +0.122 |
important | +0.121 |
precious | +0.120 |
Neg logit contributions | |
ted | -0.134 |
oes | -0.133 |
talk | -0.132 |
ting | -0.127 |
stack | -0.123 |
Token | ready |
Token ID | 1493 |
Feature activation | -8.85e-02 |
Pos logit contributions | |
oes | +0.071 |
talk | +0.070 |
ted | +0.068 |
stack | +0.062 |
rost | +0.061 |
Neg logit contributions | |
adorable | -0.093 |
best | -0.091 |
idea | -0.091 |
important | -0.091 |
precious | -0.088 |
Token | to |
Token ID | 284 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
oes | +0.204 |
talk | +0.204 |
ted | +0.197 |
stack | +0.190 |
rost | +0.189 |
Neg logit contributions | |
adorable | -0.114 |
best | -0.112 |
idea | -0.109 |
important | -0.108 |
precious | -0.103 |
Token | go |
Token ID | 467 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
talk | +0.425 |
oes | +0.420 |
ted | +0.414 |
isk | +0.393 |
stack | +0.388 |
Neg logit contributions | |
idea | -0.546 |
adorable | -0.543 |
best | -0.529 |
beautiful | -0.524 |
important | -0.513 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
talk | +0.314 |
oes | +0.311 |
ted | +0.305 |
rost | +0.300 |
isk | +0.295 |
Neg logit contributions | |
adorable | -0.126 |
best | -0.125 |
idea | -0.123 |
important | -0.118 |
beautiful | -0.112 |
Token | She |
Token ID | 1375 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
oes | +0.048 |
talk | +0.048 |
use | +0.046 |
rost | +0.046 |
ted | +0.045 |
Neg logit contributions | |
adorable | -0.045 |
best | -0.045 |
idea | -0.044 |
important | -0.042 |
beautiful | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
talk | +0.085 |
ted | +0.083 |
oes | +0.080 |
stack | +0.073 |
isk | +0.070 |
Neg logit contributions | |
adorable | -0.129 |
important | -0.127 |
best | -0.126 |
idea | -0.125 |
precious | -0.122 |
Token | ready |
Token ID | 3492 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
precious | +0.156 |
idea | +0.156 |
adorable | +0.154 |
important | +0.154 |
prized | +0.145 |
Neg logit contributions | |
oes | -0.226 |
talk | -0.225 |
ted | -0.225 |
ting | -0.223 |
aled | -0.218 |
Token | if |
Token ID | 611 |
Feature activation | -1.75e-02 |
Pos logit contributions | |
oes | +0.075 |
talk | +0.074 |
ted | +0.071 |
stack | +0.068 |
rost | +0.067 |
Neg logit contributions | |
adorable | -0.063 |
best | -0.061 |
important | -0.060 |
idea | -0.060 |
precious | -0.058 |
Token | you |
Token ID | 345 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
talk | +0.035 |
oes | +0.034 |
ted | +0.033 |
stack | +0.032 |
isk | +0.032 |
Neg logit contributions | |
adorable | -0.029 |
best | -0.029 |
important | -0.028 |
idea | -0.027 |
precious | -0.027 |
Token | fall |
Token ID | 2121 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
talk | +0.042 |
oes | +0.042 |
ted | +0.041 |
stack | +0.040 |
rost | +0.038 |
Neg logit contributions | |
adorable | -0.044 |
idea | -0.044 |
best | -0.044 |
important | -0.043 |
precious | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -8.23e-03 |
Pos logit contributions | |
talk | +0.079 |
oes | +0.076 |
rost | +0.072 |
ted | +0.072 |
stack | +0.071 |
Neg logit contributions | |
best | -0.094 |
adorable | -0.093 |
important | -0.091 |
idea | -0.089 |
beautiful | -0.087 |
Token | Ready |
Token ID | 23432 |
Feature activation | -6.05e-02 |
Pos logit contributions | |
talk | +0.014 |
oes | +0.014 |
ted | +0.014 |
stack | +0.013 |
rost | +0.013 |
Neg logit contributions | |
adorable | -0.015 |
best | -0.015 |
important | -0.015 |
idea | -0.014 |
precious | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
talk | +0.138 |
ted | +0.126 |
Dog | +0.125 |
oes | +0.124 |
stack | +0.124 |
Neg logit contributions | |
best | -0.085 |
adorable | -0.079 |
idea | -0.076 |
important | -0.075 |
precious | -0.071 |
Token | try |
Token ID | 1949 |
Feature activation | -5.90e-02 |
Pos logit contributions | |
talk | +0.461 |
ted | +0.441 |
oes | +0.440 |
rost | +0.421 |
isk | +0.420 |
Neg logit contributions | |
idea | -0.520 |
adorable | -0.504 |
best | -0.502 |
beautiful | -0.493 |
important | -0.484 |
Token | again |
Token ID | 757 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
talk | +0.129 |
oes | +0.127 |
ted | +0.123 |
stack | +0.118 |
rost | +0.118 |
Neg logit contributions | |
adorable | -0.086 |
best | -0.085 |
important | -0.083 |
idea | -0.081 |
precious | -0.079 |
Token | ?" |
Token ID | 1701 |
Feature activation | -9.65e-03 |
Pos logit contributions | |
talk | +0.277 |
oes | +0.276 |
ted | +0.269 |
stack | +0.258 |
rost | +0.256 |
Neg logit contributions | |
adorable | -0.175 |
best | -0.171 |
important | -0.169 |
idea | -0.166 |
precious | -0.159 |
Token |
|
Token ID | 198 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
oes | +0.017 |
talk | +0.017 |
ted | +0.016 |
stack | +0.015 |
rost | +0.015 |
Neg logit contributions | |
adorable | -0.018 |
best | -0.018 |
idea | -0.018 |
important | -0.018 |
precious | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
oes | +0.043 |
talk | +0.043 |
ted | +0.041 |
stack | +0.039 |
rost | +0.038 |
Neg logit contributions | |
adorable | -0.056 |
best | -0.054 |
important | -0.054 |
idea | -0.054 |
precious | -0.052 |
Token | in |
Token ID | 287 |
Feature activation | -1.76e-01 |
Pos logit contributions | |
oes | +0.152 |
talk | +0.149 |
ted | +0.139 |
stack | +0.137 |
rost | +0.136 |
Neg logit contributions | |
best | -0.177 |
adorable | -0.176 |
important | -0.170 |
beautiful | -0.165 |
idea | -0.165 |
Token | her |
Token ID | 607 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
stack | +0.224 |
talk | +0.223 |
ted | +0.214 |
oes | +0.213 |
isk | +0.210 |
Neg logit contributions | |
adorable | -0.380 |
important | -0.370 |
best | -0.363 |
precious | -0.358 |
idea | -0.356 |
Token | bag |
Token ID | 6131 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
oes | +0.039 |
talk | +0.039 |
ted | +0.038 |
stack | +0.037 |
rost | +0.036 |
Neg logit contributions | |
adorable | -0.036 |
important | -0.035 |
idea | -0.035 |
best | -0.035 |
precious | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +9.20e-03 |
Pos logit contributions | |
oes | +0.134 |
talk | +0.132 |
ted | +0.129 |
stack | +0.124 |
isk | +0.120 |
Neg logit contributions | |
adorable | -0.121 |
best | -0.116 |
important | -0.116 |
idea | -0.115 |
precious | -0.113 |
Token | Ready |
Token ID | 23432 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
adorable | +0.018 |
best | +0.017 |
important | +0.017 |
idea | +0.017 |
beautiful | +0.016 |
Neg logit contributions | |
talk | -0.016 |
oes | -0.015 |
rost | -0.014 |
ted | -0.014 |
use | -0.014 |
Token | to |
Token ID | 284 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
talk | +0.137 |
oes | +0.127 |
ted | +0.126 |
Dog | +0.125 |
stack | +0.121 |
Neg logit contributions | |
best | -0.089 |
adorable | -0.087 |
idea | -0.084 |
important | -0.082 |
precious | -0.079 |
Token | go |
Token ID | 467 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
talk | +0.463 |
oes | +0.458 |
ted | +0.440 |
isk | +0.432 |
stack | +0.431 |
Neg logit contributions | |
idea | -0.508 |
adorable | -0.489 |
best | -0.486 |
beautiful | -0.475 |
amazing | -0.464 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
talk | +0.299 |
oes | +0.293 |
ted | +0.289 |
rost | +0.283 |
isk | +0.282 |
Neg logit contributions | |
best | -0.150 |
adorable | -0.145 |
idea | -0.141 |
important | -0.138 |
beautiful | -0.128 |
Token | she |
Token ID | 673 |
Feature activation | -8.58e-02 |
Pos logit contributions | |
oes | +0.183 |
talk | +0.181 |
ted | +0.174 |
stack | +0.164 |
rost | +0.164 |
Neg logit contributions | |
adorable | -0.196 |
best | -0.192 |
idea | -0.189 |
important | -0.189 |
precious | -0.182 |
Token | ran |
Token ID | 4966 |
Feature activation | -5.89e-02 |
Pos logit contributions | |
oes | +0.143 |
talk | +0.141 |
ted | +0.136 |
stack | +0.130 |
rost | +0.129 |
Neg logit contributions | |
adorable | -0.166 |
best | -0.164 |
important | -0.161 |
idea | -0.161 |
precious | -0.156 |
Token | back |
Token ID | 736 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
oes | +0.110 |
talk | +0.108 |
ted | +0.104 |
stack | +0.100 |
rost | +0.099 |
Neg logit contributions | |
adorable | -0.104 |
best | -0.102 |
important | -0.101 |
idea | -0.100 |
precious | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | +1.32e-03 |
Pos logit contributions | |
adorable | +0.162 |
important | +0.155 |
precious | +0.154 |
best | +0.153 |
idea | +0.152 |
Neg logit contributions | |
oes | -0.172 |
talk | -0.169 |
ted | -0.168 |
ting | -0.159 |
stack | -0.155 |
Token |
|
Token ID | 198 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
best | +0.002 |
adorable | +0.002 |
idea | +0.002 |
important | +0.002 |
beautiful | +0.002 |
Neg logit contributions | |
talk | -0.003 |
ted | -0.003 |
isk | -0.003 |
stack | -0.003 |
ting | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
talk | +0.078 |
ted | +0.078 |
oes | +0.077 |
stack | +0.075 |
ting | +0.074 |
Neg logit contributions | |
adorable | -0.073 |
important | -0.072 |
best | -0.072 |
idea | -0.069 |
beautiful | -0.068 |
Token | " |
Token ID | 1 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
oes | +0.062 |
talk | +0.062 |
ted | +0.060 |
stack | +0.057 |
isk | +0.056 |
Neg logit contributions | |
adorable | -0.077 |
best | -0.075 |
important | -0.075 |
idea | -0.073 |
precious | -0.072 |
Token | Time |
Token ID | 7575 |
Feature activation | -5.67e-02 |
Pos logit contributions | |
oes | +0.046 |
talk | +0.044 |
ted | +0.044 |
stack | +0.043 |
isk | +0.042 |
Neg logit contributions | |
adorable | -0.052 |
best | -0.051 |
important | -0.049 |
beautiful | -0.048 |
idea | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
talk | +0.076 |
oes | +0.073 |
ted | +0.069 |
stack | +0.068 |
isk | +0.064 |
Neg logit contributions | |
best | -0.131 |
adorable | -0.129 |
important | -0.128 |
idea | -0.128 |
beautiful | -0.122 |
Token | bat |
Token ID | 7365 |
Feature activation | -2.56e-02 |
Pos logit contributions | |
talk | +0.426 |
oes | +0.426 |
ted | +0.412 |
stack | +0.387 |
rost | +0.378 |
Neg logit contributions | |
adorable | -0.508 |
best | -0.495 |
idea | -0.495 |
important | -0.488 |
beautiful | -0.470 |
Token | he |
Token ID | 258 |
Feature activation | -7.18e-02 |
Pos logit contributions | |
oes | +0.040 |
talk | +0.039 |
ted | +0.038 |
stack | +0.036 |
rost | +0.035 |
Neg logit contributions | |
adorable | -0.053 |
best | -0.052 |
important | -0.051 |
idea | -0.051 |
precious | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
talk | +0.130 |
oes | +0.130 |
stack | +0.128 |
ted | +0.128 |
rost | +0.122 |
Neg logit contributions | |
adorable | -0.116 |
precious | -0.115 |
prized | -0.112 |
important | -0.110 |
valuable | -0.108 |
Token | L |
Token ID | 406 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
oes | +0.122 |
talk | +0.121 |
ted | +0.116 |
stack | +0.109 |
rost | +0.107 |
Neg logit contributions | |
adorable | -0.188 |
best | -0.184 |
important | -0.182 |
idea | -0.181 |
precious | -0.177 |
Token | ila |
Token ID | 10102 |
Feature activation | +6.80e-02 |
Pos logit contributions | |
adorable | +0.298 |
best | +0.290 |
important | +0.288 |
idea | +0.286 |
precious | +0.280 |
Neg logit contributions | |
oes | -0.261 |
talk | -0.256 |
ted | -0.247 |
stack | -0.233 |
rost | -0.233 |
Token | . |
Token ID | 13 |
Feature activation | -8.07e-03 |
Pos logit contributions | |
talk | +0.145 |
oes | +0.142 |
ted | +0.142 |
stack | +0.135 |
isk | +0.134 |
Neg logit contributions | |
adorable | -0.087 |
best | -0.087 |
important | -0.086 |
idea | -0.082 |
precious | -0.080 |
Token | You |
Token ID | 921 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
oes | +0.014 |
talk | +0.014 |
ted | +0.013 |
stack | +0.013 |
rost | +0.013 |
Neg logit contributions | |
adorable | -0.015 |
idea | -0.015 |
best | -0.015 |
important | -0.015 |
precious | -0.014 |
Token | should |
Token ID | 815 |
Feature activation | -6.71e-02 |
Pos logit contributions | |
adorable | +0.064 |
best | +0.062 |
important | +0.061 |
idea | +0.061 |
precious | +0.059 |
Neg logit contributions | |
oes | -0.069 |
talk | -0.069 |
ted | -0.066 |
stack | -0.063 |
rost | -0.062 |
Token | go |
Token ID | 467 |
Feature activation | -9.31e-02 |
Pos logit contributions | |
ted | +0.124 |
oes | +0.122 |
talk | +0.120 |
ting | +0.112 |
stack | +0.111 |
Neg logit contributions | |
adorable | -0.121 |
best | -0.119 |
idea | -0.119 |
important | -0.119 |
precious | -0.111 |
Token | back |
Token ID | 736 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ted | +0.155 |
rost | +0.153 |
stack | +0.152 |
oes | +0.147 |
gra | +0.147 |
Neg logit contributions | |
best | -0.185 |
important | -0.177 |
adorable | -0.175 |
loyal | -0.169 |
precious | -0.166 |
Token | to |
Token ID | 284 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
oes | +0.264 |
ted | +0.255 |
talk | +0.254 |
stack | +0.253 |
rost | +0.244 |
Neg logit contributions | |
best | -0.341 |
adorable | -0.340 |
important | -0.340 |
precious | -0.326 |
beautiful | -0.322 |
Token | your |
Token ID | 534 |
Feature activation | -3.29e-02 |
Pos logit contributions | |
talk | +0.545 |
oes | +0.542 |
ted | +0.533 |
stack | +0.500 |
rost | +0.498 |
Neg logit contributions | |
adorable | -0.412 |
best | -0.405 |
important | -0.403 |
idea | -0.389 |
beautiful | -0.384 |
Token | own |
Token ID | 898 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
oes | +0.067 |
talk | +0.066 |
ted | +0.063 |
stack | +0.062 |
rost | +0.061 |
Neg logit contributions | |
adorable | -0.051 |
best | -0.048 |
idea | -0.048 |
important | -0.048 |
precious | -0.046 |
Token | country |
Token ID | 1499 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
adorable | +0.027 |
best | +0.026 |
important | +0.026 |
idea | +0.026 |
precious | +0.025 |
Neg logit contributions | |
oes | -0.029 |
talk | -0.028 |
ted | -0.027 |
stack | -0.026 |
rost | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
oes | +0.287 |
talk | +0.284 |
ted | +0.278 |
stack | +0.265 |
rost | +0.262 |
Neg logit contributions | |
adorable | -0.210 |
best | -0.204 |
important | -0.201 |
idea | -0.200 |
precious | -0.193 |
Token | play |
Token ID | 711 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
ted | +0.159 |
oes | +0.158 |
talk | +0.157 |
rost | +0.148 |
ting | +0.145 |
Neg logit contributions | |
best | -0.147 |
adorable | -0.146 |
important | -0.145 |
precious | -0.140 |
loyal | -0.140 |
Token | . |
Token ID | 13 |
Feature activation | +5.44e-03 |
Pos logit contributions | |
oes | +0.127 |
talk | +0.122 |
ted | +0.114 |
rost | +0.111 |
iffs | +0.108 |
Neg logit contributions | |
adorable | -0.148 |
best | -0.142 |
idea | -0.141 |
precious | -0.140 |
important | -0.139 |
Token |
|
Token ID | 198 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
adorable | +0.010 |
important | +0.010 |
idea | +0.010 |
best | +0.010 |
beautiful | +0.009 |
Neg logit contributions | |
talk | -0.010 |
oes | -0.010 |
ted | -0.009 |
stack | -0.009 |
isk | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
oes | +0.056 |
talk | +0.054 |
ted | +0.054 |
stack | +0.051 |
ting | +0.050 |
Neg logit contributions | |
adorable | -0.052 |
important | -0.051 |
best | -0.049 |
idea | -0.049 |
beautiful | -0.048 |
Token | " |
Token ID | 1 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
oes | +0.063 |
talk | +0.060 |
ted | +0.060 |
stack | +0.058 |
isk | +0.056 |
Neg logit contributions | |
adorable | -0.029 |
important | -0.028 |
precious | -0.026 |
loyal | -0.026 |
best | -0.026 |
Token | Time |
Token ID | 7575 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
oes | +0.039 |
talk | +0.036 |
stack | +0.034 |
ted | +0.033 |
ales | +0.033 |
Neg logit contributions | |
adorable | -0.049 |
best | -0.046 |
important | -0.045 |
precious | -0.045 |
beautiful | -0.045 |
Token | to |
Token ID | 284 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
talk | +0.066 |
oes | +0.065 |
ted | +0.059 |
stack | +0.059 |
rost | +0.056 |
Neg logit contributions | |
best | -0.121 |
adorable | -0.119 |
important | -0.118 |
idea | -0.117 |
precious | -0.113 |
Token | get |
Token ID | 651 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
oes | +0.377 |
talk | +0.373 |
ted | +0.357 |
stack | +0.333 |
ricanes | +0.327 |
Neg logit contributions | |
adorable | -0.557 |
idea | -0.548 |
best | -0.547 |
important | -0.527 |
beautiful | -0.525 |
Token | up |
Token ID | 510 |
Feature activation | -9.47e-02 |
Pos logit contributions | |
oes | +0.028 |
talk | +0.028 |
stack | +0.027 |
ted | +0.026 |
isk | +0.026 |
Neg logit contributions | |
adorable | -0.033 |
best | -0.032 |
important | -0.032 |
idea | -0.030 |
precious | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -6.91e-02 |
Pos logit contributions | |
oes | +0.178 |
talk | +0.172 |
ted | +0.172 |
use | +0.165 |
stack | +0.165 |
Neg logit contributions | |
adorable | -0.152 |
idea | -0.144 |
best | -0.142 |
precious | -0.142 |
important | -0.141 |
Token | my |
Token ID | 616 |
Feature activation | +7.94e-02 |
Pos logit contributions | |
oes | +0.121 |
rost | +0.118 |
ted | +0.118 |
ting | +0.113 |
south | +0.113 |
Neg logit contributions | |
prized | -0.115 |
important | -0.113 |
valuable | -0.109 |
precious | -0.108 |
adorable | -0.107 |
Token | loves |
Token ID | 10408 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
impressed | +0.096 |
vern | +0.091 |
prized | +0.090 |
important | +0.089 |
valuable | +0.088 |
Neg logit contributions | |
rost | -0.157 |
oes | -0.156 |
ted | -0.153 |
Rex | -0.145 |
hike | -0.143 |
Token | . |
Token ID | 13 |
Feature activation | +6.50e-03 |
Pos logit contributions | |
oes | +0.083 |
talk | +0.083 |
ted | +0.080 |
stack | +0.077 |
rost | +0.076 |
Neg logit contributions | |
adorable | -0.067 |
best | -0.064 |
important | -0.064 |
idea | -0.063 |
precious | -0.062 |
Token |
|
Token ID | 198 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
adorable | +0.011 |
best | +0.010 |
important | +0.010 |
idea | +0.010 |
beautiful | +0.010 |
Neg logit contributions | |
oes | -0.014 |
talk | -0.013 |
isk | -0.012 |
ales | -0.012 |
stack | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
oes | +0.061 |
talk | +0.059 |
ted | +0.057 |
stack | +0.055 |
isk | +0.054 |
Neg logit contributions | |
adorable | -0.060 |
important | -0.058 |
best | -0.057 |
idea | -0.057 |
precious | -0.055 |
Token | " |
Token ID | 1 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
oes | +0.067 |
talk | +0.065 |
ted | +0.064 |
stack | +0.062 |
isk | +0.061 |
Neg logit contributions | |
adorable | -0.038 |
important | -0.036 |
best | -0.036 |
idea | -0.034 |
precious | -0.034 |
Token | Time |
Token ID | 7575 |
Feature activation | -5.63e-02 |
Pos logit contributions | |
oes | +0.042 |
stack | +0.036 |
ales | +0.035 |
talk | +0.035 |
isk | +0.034 |
Neg logit contributions | |
adorable | -0.039 |
best | -0.038 |
beautiful | -0.037 |
precious | -0.037 |
loyal | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
talk | +0.074 |
oes | +0.074 |
stack | +0.066 |
ted | +0.066 |
rost | +0.064 |
Neg logit contributions | |
best | -0.130 |
adorable | -0.129 |
important | -0.127 |
idea | -0.126 |
precious | -0.122 |
Token | go |
Token ID | 467 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
oes | +0.365 |
talk | +0.360 |
ted | +0.350 |
stack | +0.324 |
rost | +0.316 |
Neg logit contributions | |
adorable | -0.559 |
idea | -0.554 |
best | -0.548 |
important | -0.537 |
beautiful | -0.526 |
Token | , |
Token ID | 11 |
Feature activation | -9.58e-02 |
Pos logit contributions | |
talk | +0.219 |
oes | +0.217 |
ted | +0.214 |
rost | +0.208 |
isk | +0.206 |
Neg logit contributions | |
best | -0.189 |
adorable | -0.187 |
idea | -0.185 |
important | -0.181 |
beautiful | -0.175 |
Token | kids |
Token ID | 3988 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
oes | +0.170 |
talk | +0.168 |
ted | +0.167 |
rost | +0.162 |
ting | +0.158 |
Neg logit contributions | |
adorable | -0.161 |
important | -0.161 |
best | -0.157 |
idea | -0.155 |
prized | -0.155 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
oes | +0.163 |
talk | +0.161 |
ted | +0.158 |
stack | +0.152 |
rost | +0.150 |
Neg logit contributions | |
adorable | -0.072 |
best | -0.070 |
idea | -0.068 |
important | -0.068 |
precious | -0.064 |
Token | she |
Token ID | 673 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
oes | +0.036 |
talk | +0.035 |
ted | +0.033 |
isk | +0.033 |
stack | +0.032 |
Neg logit contributions | |
adorable | -0.045 |
best | -0.044 |
idea | -0.044 |
important | -0.043 |
precious | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
oes | +0.242 |
talk | +0.239 |
ted | +0.230 |
stack | +0.218 |
rost | +0.214 |
Neg logit contributions | |
adorable | -0.339 |
best | -0.331 |
important | -0.328 |
idea | -0.326 |
precious | -0.318 |
Token | park |
Token ID | 3952 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
oes | +0.069 |
talk | +0.068 |
ted | +0.066 |
isk | +0.063 |
rost | +0.061 |
Neg logit contributions | |
adorable | -0.070 |
best | -0.069 |
important | -0.067 |
beautiful | -0.066 |
precious | -0.065 |
Token | . |
Token ID | 13 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
oes | +0.245 |
talk | +0.244 |
ted | +0.238 |
rost | +0.227 |
stack | +0.224 |
Neg logit contributions | |
adorable | -0.171 |
best | -0.168 |
idea | -0.164 |
important | -0.164 |
precious | -0.155 |
Token | " |
Token ID | 366 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
adorable | +0.033 |
idea | +0.033 |
important | +0.031 |
best | +0.030 |
loyal | +0.030 |
Neg logit contributions | |
talk | -0.031 |
oes | -0.031 |
use | -0.030 |
ted | -0.029 |
ting | -0.029 |
Token | Ready |
Token ID | 35474 |
Feature activation | -5.38e-02 |
Pos logit contributions | |
oes | +0.053 |
ted | +0.048 |
stack | +0.047 |
talk | +0.046 |
isk | +0.045 |
Neg logit contributions | |
adorable | -0.063 |
best | -0.059 |
important | -0.059 |
loyal | -0.058 |
precious | -0.058 |
Token | to |
Token ID | 284 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
talk | +0.129 |
oes | +0.128 |
ted | +0.125 |
stack | +0.121 |
isk | +0.119 |
Neg logit contributions | |
best | -0.064 |
adorable | -0.063 |
idea | -0.061 |
important | -0.059 |
precious | -0.056 |
Token | make |
Token ID | 787 |
Feature activation | -5.50e-02 |
Pos logit contributions | |
talk | +0.419 |
ted | +0.403 |
oes | +0.401 |
isk | +0.387 |
stack | +0.385 |
Neg logit contributions | |
idea | -0.505 |
adorable | -0.500 |
best | -0.496 |
beautiful | -0.489 |
important | -0.468 |
Token | the |
Token ID | 262 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
ted | +0.093 |
oes | +0.093 |
talk | +0.092 |
stack | +0.084 |
ting | +0.084 |
Neg logit contributions | |
adorable | -0.101 |
important | -0.099 |
idea | -0.098 |
best | -0.097 |
precious | -0.097 |
Token | k |
Token ID | 479 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
oes | +0.021 |
stack | +0.020 |
talk | +0.019 |
ted | +0.019 |
gra | +0.019 |
Neg logit contributions | |
adorable | -0.026 |
important | -0.026 |
precious | -0.025 |
idea | -0.025 |
impressed | -0.025 |
Token | ite |
Token ID | 578 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
adorable | +0.031 |
idea | +0.031 |
important | +0.031 |
best | +0.030 |
precious | +0.030 |
Neg logit contributions | |
oes | -0.025 |
talk | -0.025 |
ted | -0.024 |
stack | -0.023 |
iffs | -0.022 |
Token | work |
Token ID | 670 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
oes | +0.167 |
talk | +0.156 |
stack | +0.146 |
ted | +0.146 |
isk | +0.142 |
Neg logit contributions | |
adorable | -0.184 |
important | -0.180 |
precious | -0.176 |
best | -0.175 |
idea | -0.171 |
Token | will |
Token ID | 481 |
Feature activation | -4.12e-02 |
Pos logit contributions | |
adorable | +0.033 |
best | +0.033 |
important | +0.032 |
idea | +0.032 |
precious | +0.031 |
Neg logit contributions | |
oes | -0.028 |
talk | -0.027 |
ted | -0.026 |
stack | -0.025 |
rost | -0.024 |
Token | be |
Token ID | 307 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
oes | +0.072 |
talk | +0.071 |
ted | +0.071 |
stack | +0.065 |
rost | +0.064 |
Neg logit contributions | |
adorable | -0.077 |
best | -0.076 |
idea | -0.075 |
important | -0.075 |
precious | -0.072 |
Token | able |
Token ID | 1498 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
idea | +0.071 |
ummy | +0.069 |
adorable | +0.068 |
prized | +0.064 |
darling | +0.064 |
Neg logit contributions | |
south | -0.103 |
talk | -0.102 |
ting | -0.099 |
away | -0.099 |
Why | -0.097 |
Token | to |
Token ID | 284 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
adorable | +0.101 |
idea | +0.100 |
best | +0.099 |
important | +0.098 |
precious | +0.096 |
Neg logit contributions | |
oes | -0.033 |
talk | -0.031 |
ted | -0.029 |
stack | -0.026 |
rost | -0.025 |
Token | climb |
Token ID | 12080 |
Feature activation | -1.72e-01 |
Pos logit contributions | |
ted | +0.322 |
oes | +0.319 |
talk | +0.311 |
ricanes | +0.287 |
stack | +0.287 |
Neg logit contributions | |
adorable | -0.400 |
best | -0.400 |
idea | -0.399 |
important | -0.388 |
beautiful | -0.380 |
Token | to |
Token ID | 284 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
talk | +0.324 |
oes | +0.322 |
ted | +0.300 |
rost | +0.296 |
isk | +0.293 |
Neg logit contributions | |
adorable | -0.306 |
best | -0.302 |
important | -0.297 |
idea | -0.287 |
precious | -0.284 |
Token | the |
Token ID | 262 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
oes | +0.442 |
talk | +0.442 |
ted | +0.429 |
stack | +0.400 |
rost | +0.398 |
Neg logit contributions | |
adorable | -0.482 |
best | -0.471 |
important | -0.465 |
idea | -0.464 |
precious | -0.448 |
Token | top |
Token ID | 1353 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
oes | +0.109 |
talk | +0.107 |
ted | +0.103 |
rost | +0.098 |
isk | +0.097 |
Neg logit contributions | |
adorable | -0.117 |
best | -0.115 |
important | -0.114 |
idea | -0.111 |
precious | -0.110 |
Token | of |
Token ID | 286 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
talk | +0.097 |
oes | +0.096 |
ted | +0.095 |
stack | +0.091 |
rost | +0.091 |
Neg logit contributions | |
adorable | -0.056 |
best | -0.055 |
idea | -0.054 |
important | -0.054 |
precious | -0.050 |
Token | the |
Token ID | 262 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
oes | +0.211 |
talk | +0.208 |
ted | +0.201 |
stack | +0.191 |
rost | +0.187 |
Neg logit contributions | |
adorable | -0.273 |
best | -0.266 |
important | -0.264 |
idea | -0.263 |
precious | -0.256 |
Token | tallest |
Token ID | 38760 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
oes | +0.057 |
talk | +0.056 |
ted | +0.054 |
stack | +0.051 |
rost | +0.050 |
Neg logit contributions | |
adorable | -0.079 |
best | -0.077 |
important | -0.076 |
idea | -0.076 |
precious | -0.074 |
Token | big |
Token ID | 1263 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
adorable | +0.241 |
best | +0.227 |
important | +0.224 |
idea | +0.210 |
precious | +0.209 |
Neg logit contributions | |
oes | -0.304 |
gra | -0.294 |
talk | -0.292 |
rost | -0.287 |
stack | -0.284 |
Token | zip |
Token ID | 19974 |
Feature activation | -8.32e-03 |
Pos logit contributions | |
adorable | +0.273 |
best | +0.268 |
important | +0.266 |
precious | +0.256 |
impressed | +0.254 |
Neg logit contributions | |
oes | -0.158 |
talk | -0.150 |
stack | -0.149 |
ted | -0.149 |
rost | -0.149 |
Token | ! |
Token ID | 0 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
oes | +0.016 |
talk | +0.015 |
ted | +0.015 |
stack | +0.014 |
rost | +0.014 |
Neg logit contributions | |
adorable | -0.014 |
best | -0.014 |
idea | -0.014 |
important | -0.014 |
precious | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
adorable | +0.037 |
best | +0.037 |
important | +0.036 |
idea | +0.036 |
precious | +0.035 |
Neg logit contributions | |
oes | -0.023 |
talk | -0.023 |
ted | -0.022 |
stack | -0.020 |
rost | -0.020 |
Token | Ready |
Token ID | 35474 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
oes | +0.052 |
talk | +0.052 |
ted | +0.050 |
stack | +0.047 |
rost | +0.047 |
Neg logit contributions | |
adorable | -0.064 |
best | -0.062 |
important | -0.061 |
idea | -0.061 |
precious | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
talk | +0.107 |
oes | +0.107 |
ted | +0.104 |
stack | +0.100 |
rost | +0.098 |
Neg logit contributions | |
adorable | -0.062 |
best | -0.062 |
idea | -0.060 |
important | -0.059 |
precious | -0.056 |
Token | go |
Token ID | 467 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
oes | +0.361 |
talk | +0.358 |
ted | +0.340 |
stack | +0.322 |
rost | +0.316 |
Neg logit contributions | |
adorable | -0.563 |
idea | -0.554 |
best | -0.554 |
important | -0.545 |
beautiful | -0.527 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.95e-02 |
Pos logit contributions | |
talk | +0.298 |
oes | +0.297 |
ted | +0.289 |
rost | +0.284 |
stack | +0.280 |
Neg logit contributions | |
adorable | -0.141 |
best | -0.139 |
idea | -0.137 |
important | -0.134 |
beautiful | -0.127 |
Token | she |
Token ID | 673 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
oes | +0.032 |
talk | +0.032 |
ted | +0.031 |
stack | +0.028 |
rost | +0.028 |
Neg logit contributions | |
adorable | -0.038 |
best | -0.038 |
important | -0.037 |
idea | -0.036 |
precious | -0.036 |
Token | shouted |
Token ID | 17293 |
Feature activation | +7.97e-02 |
Pos logit contributions | |
talk | +0.064 |
ted | +0.059 |
oes | +0.057 |
ricane | +0.050 |
towards | +0.049 |
Neg logit contributions | |
adorable | -0.088 |
precious | -0.087 |
best | -0.087 |
important | -0.086 |
loyal | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +9.25e-03 |
Pos logit contributions | |
best | +0.152 |
precious | +0.150 |
adorable | +0.149 |
important | +0.147 |
idea | +0.143 |
Neg logit contributions | |
talk | -0.123 |
ted | -0.120 |
ting | -0.115 |
towards | -0.115 |
oes | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -5.55e-04 |
Pos logit contributions | |
adorable | +0.022 |
idea | +0.022 |
best | +0.021 |
important | +0.021 |
precious | +0.020 |
Neg logit contributions | |
oes | -0.024 |
talk | -0.024 |
ted | -0.023 |
rost | -0.022 |
use | -0.022 |
Token | We |
Token ID | 775 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
oes | +0.001 |
talk | +0.001 |
ted | +0.001 |
stack | +0.001 |
isk | +0.001 |
Neg logit contributions | |
adorable | -0.001 |
best | -0.001 |
important | -0.001 |
idea | -0.001 |
precious | -0.001 |
Token | should |
Token ID | 815 |
Feature activation | -6.33e-02 |
Pos logit contributions | |
oes | +0.085 |
talk | +0.081 |
ted | +0.080 |
stack | +0.077 |
ting | +0.076 |
Neg logit contributions | |
adorable | -0.068 |
best | -0.067 |
idea | -0.066 |
important | -0.066 |
precious | -0.064 |
Token | go |
Token ID | 467 |
Feature activation | -9.11e-02 |
Pos logit contributions | |
oes | +0.127 |
talk | +0.126 |
ted | +0.125 |
stack | +0.117 |
rost | +0.115 |
Neg logit contributions | |
adorable | -0.103 |
best | -0.100 |
important | -0.099 |
idea | -0.099 |
precious | -0.095 |
Token | back |
Token ID | 736 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
ted | +0.164 |
oes | +0.162 |
stack | +0.159 |
rost | +0.157 |
talk | +0.156 |
Neg logit contributions | |
best | -0.165 |
adorable | -0.160 |
important | -0.158 |
loyal | -0.152 |
precious | -0.151 |
Token | to |
Token ID | 284 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
oes | +0.290 |
stack | +0.280 |
ted | +0.278 |
talk | +0.276 |
isk | +0.263 |
Neg logit contributions | |
adorable | -0.320 |
best | -0.316 |
important | -0.315 |
precious | -0.306 |
beautiful | -0.301 |
Token | the |
Token ID | 262 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
oes | +0.470 |
talk | +0.466 |
ted | +0.457 |
stack | +0.430 |
ricanes | +0.423 |
Neg logit contributions | |
adorable | -0.478 |
best | -0.460 |
important | -0.459 |
idea | -0.449 |
beautiful | -0.446 |
Token | park |
Token ID | 3952 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
oes | +0.091 |
talk | +0.090 |
ted | +0.088 |
stack | +0.081 |
rost | +0.081 |
Neg logit contributions | |
adorable | -0.107 |
best | -0.106 |
important | -0.104 |
idea | -0.102 |
precious | -0.101 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
oes | +0.256 |
talk | +0.254 |
ted | +0.248 |
stack | +0.236 |
rost | +0.233 |
Neg logit contributions | |
adorable | -0.206 |
best | -0.201 |
important | -0.198 |
idea | -0.197 |
precious | -0.190 |
Token | Mom |
Token ID | 11254 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
oes | +0.032 |
talk | +0.032 |
ted | +0.031 |
isk | +0.030 |
stack | +0.029 |
Neg logit contributions | |
adorable | -0.029 |
best | -0.028 |
important | -0.028 |
idea | -0.027 |
precious | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
adorable | +0.075 |
best | +0.074 |
important | +0.072 |
idea | +0.072 |
precious | +0.070 |
Neg logit contributions | |
oes | -0.090 |
talk | -0.089 |
ted | -0.086 |
stack | -0.084 |
rost | -0.083 |
Token | he |
Token ID | 339 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
oes | +0.025 |
talk | +0.024 |
ted | +0.023 |
stack | +0.022 |
rost | +0.021 |
Neg logit contributions | |
adorable | -0.044 |
idea | -0.043 |
best | -0.042 |
important | -0.042 |
precious | -0.042 |
Token | had |
Token ID | 550 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
ted | +0.040 |
talk | +0.039 |
oes | +0.037 |
ting | +0.036 |
stack | +0.035 |
Neg logit contributions | |
idea | -0.057 |
best | -0.056 |
adorable | -0.056 |
important | -0.053 |
precious | -0.053 |
Token | managed |
Token ID | 5257 |
Feature activation | -9.14e-03 |
Pos logit contributions | |
adorable | +0.030 |
idea | +0.029 |
best | +0.029 |
important | +0.028 |
precious | +0.028 |
Neg logit contributions | |
talk | -0.029 |
oes | -0.029 |
ted | -0.028 |
stack | -0.026 |
rost | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
talk | +0.012 |
oes | +0.011 |
stack | +0.011 |
ted | +0.011 |
rost | +0.010 |
Neg logit contributions | |
important | -0.022 |
adorable | -0.022 |
best | -0.022 |
idea | -0.021 |
beautiful | -0.021 |
Token | climb |
Token ID | 12080 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
oes | +0.300 |
talk | +0.297 |
ted | +0.282 |
stack | +0.268 |
rost | +0.263 |
Neg logit contributions | |
adorable | -0.433 |
best | -0.423 |
important | -0.419 |
idea | -0.415 |
precious | -0.408 |
Token | to |
Token ID | 284 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
talk | +0.319 |
oes | +0.316 |
ted | +0.297 |
isk | +0.290 |
rost | +0.289 |
Neg logit contributions | |
adorable | -0.300 |
best | -0.299 |
important | -0.290 |
idea | -0.286 |
beautiful | -0.281 |
Token | the |
Token ID | 262 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
oes | +0.438 |
talk | +0.437 |
ted | +0.429 |
rost | +0.394 |
stack | +0.394 |
Neg logit contributions | |
adorable | -0.484 |
best | -0.477 |
idea | -0.471 |
important | -0.467 |
beautiful | -0.454 |
Token | top |
Token ID | 1353 |
Feature activation | -3.73e-02 |
Pos logit contributions | |
oes | +0.091 |
talk | +0.090 |
ted | +0.087 |
stack | +0.082 |
rost | +0.081 |
Neg logit contributions | |
adorable | -0.105 |
best | -0.102 |
important | -0.101 |
idea | -0.100 |
precious | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
talk | +0.088 |
ted | +0.086 |
oes | +0.086 |
stack | +0.082 |
isk | +0.082 |
Neg logit contributions | |
adorable | -0.047 |
best | -0.046 |
idea | -0.045 |
important | -0.044 |
beautiful | -0.042 |
Token | And |
Token ID | 843 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
adorable | +0.021 |
best | +0.020 |
important | +0.020 |
idea | +0.020 |
loyal | +0.019 |
Neg logit contributions | |
talk | -0.019 |
oes | -0.019 |
ted | -0.018 |
use | -0.017 |
rost | -0.017 |
Token | he |
Token ID | 339 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
adorable | +0.072 |
best | +0.070 |
important | +0.069 |
idea | +0.069 |
precious | +0.067 |
Neg logit contributions | |
oes | -0.054 |
talk | -0.053 |
ted | -0.051 |
stack | -0.048 |
rost | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | +8.98e-03 |
Pos logit contributions | |
ted | +0.009 |
oes | +0.009 |
talk | +0.008 |
stack | +0.008 |
ricanes | +0.008 |
Neg logit contributions | |
adorable | -0.005 |
best | -0.004 |
idea | -0.004 |
important | -0.004 |
loyal | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -3.46e-02 |
Pos logit contributions | |
adorable | +0.014 |
idea | +0.013 |
beautiful | +0.013 |
important | +0.013 |
best | +0.013 |
Neg logit contributions | |
talk | -0.019 |
oes | -0.018 |
ales | -0.018 |
isk | -0.017 |
ted | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
oes | +0.067 |
talk | +0.065 |
ted | +0.065 |
stack | +0.062 |
ting | +0.061 |
Neg logit contributions | |
adorable | -0.058 |
important | -0.057 |
best | -0.056 |
idea | -0.056 |
beautiful | -0.054 |
Token | " |
Token ID | 1 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
oes | +0.074 |
talk | +0.072 |
ted | +0.071 |
stack | +0.069 |
rost | +0.068 |
Neg logit contributions | |
adorable | -0.037 |
important | -0.035 |
best | -0.033 |
precious | -0.033 |
loyal | -0.032 |
Token | Time |
Token ID | 7575 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
oes | +0.039 |
stack | +0.034 |
isk | +0.033 |
talk | +0.033 |
ales | +0.032 |
Neg logit contributions | |
adorable | -0.048 |
precious | -0.046 |
beautiful | -0.045 |
important | -0.044 |
best | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
talk | +0.074 |
oes | +0.074 |
stack | +0.068 |
ted | +0.066 |
rost | +0.064 |
Neg logit contributions | |
best | -0.131 |
adorable | -0.129 |
important | -0.129 |
idea | -0.127 |
precious | -0.123 |
Token | go |
Token ID | 467 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
oes | +0.411 |
talk | +0.402 |
ted | +0.395 |
stack | +0.373 |
ricanes | +0.368 |
Neg logit contributions | |
idea | -0.519 |
adorable | -0.515 |
best | -0.499 |
beautiful | -0.495 |
important | -0.482 |
Token | , |
Token ID | 11 |
Feature activation | -8.58e-02 |
Pos logit contributions | |
oes | +0.253 |
talk | +0.253 |
ted | +0.249 |
stack | +0.245 |
rost | +0.244 |
Neg logit contributions | |
best | -0.143 |
adorable | -0.142 |
idea | -0.139 |
important | -0.135 |
beautiful | -0.129 |
Token | Tommy |
Token ID | 19919 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
oes | +0.167 |
ted | +0.160 |
rost | +0.156 |
talk | +0.156 |
ting | +0.155 |
Neg logit contributions | |
important | -0.133 |
adorable | -0.130 |
prized | -0.130 |
precious | -0.128 |
best | -0.128 |
Token | ," |
Token ID | 553 |
Feature activation | -7.17e-03 |
Pos logit contributions | |
adorable | +0.028 |
best | +0.027 |
precious | +0.026 |
idea | +0.026 |
prized | +0.026 |
Neg logit contributions | |
talk | -0.034 |
oes | -0.033 |
ted | -0.033 |
stack | -0.032 |
isk | -0.031 |
Token | she |
Token ID | 673 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
oes | +0.011 |
talk | +0.011 |
ted | +0.011 |
stack | +0.010 |
rost | +0.010 |
Neg logit contributions | |
adorable | -0.014 |
best | -0.014 |
important | -0.014 |
idea | -0.014 |
precious | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -4.67e-03 |
Pos logit contributions | |
talk | +0.221 |
oes | +0.220 |
ted | +0.209 |
rost | +0.205 |
isk | +0.202 |
Neg logit contributions | |
best | -0.206 |
adorable | -0.205 |
idea | -0.197 |
important | -0.197 |
precious | -0.190 |
Token |
|
Token ID | 198 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
talk | +0.008 |
oes | +0.008 |
ted | +0.007 |
rost | +0.007 |
isk | +0.007 |
Neg logit contributions | |
adorable | -0.009 |
best | -0.009 |
important | -0.009 |
idea | -0.008 |
loyal | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
oes | +0.060 |
talk | +0.059 |
ted | +0.057 |
stack | +0.054 |
rost | +0.053 |
Neg logit contributions | |
adorable | -0.065 |
best | -0.063 |
important | -0.063 |
idea | -0.062 |
precious | -0.060 |
Token | " |
Token ID | 1 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
oes | +0.050 |
talk | +0.049 |
ted | +0.047 |
stack | +0.045 |
rost | +0.044 |
Neg logit contributions | |
adorable | -0.054 |
best | -0.053 |
important | -0.053 |
idea | -0.052 |
precious | -0.051 |
Token | Ready |
Token ID | 35474 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
oes | +0.037 |
talk | +0.036 |
ted | +0.034 |
stack | +0.033 |
isk | +0.032 |
Neg logit contributions | |
adorable | -0.046 |
best | -0.044 |
important | -0.044 |
precious | -0.043 |
idea | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
oes | +0.104 |
talk | +0.104 |
ted | +0.101 |
stack | +0.098 |
rost | +0.096 |
Neg logit contributions | |
adorable | -0.057 |
best | -0.056 |
idea | -0.054 |
important | -0.054 |
precious | -0.051 |
Token | catch |
Token ID | 4929 |
Feature activation | -6.98e-02 |
Pos logit contributions | |
talk | +0.386 |
oes | +0.385 |
ted | +0.372 |
stack | +0.352 |
isk | +0.344 |
Neg logit contributions | |
idea | -0.533 |
adorable | -0.531 |
best | -0.531 |
important | -0.515 |
beautiful | -0.504 |
Token | some |
Token ID | 617 |
Feature activation | -9.41e-03 |
Pos logit contributions | |
oes | +0.124 |
talk | +0.123 |
ted | +0.117 |
stack | +0.112 |
rost | +0.111 |
Neg logit contributions | |
adorable | -0.130 |
best | -0.127 |
important | -0.125 |
idea | -0.124 |
precious | -0.121 |
Token | fish |
Token ID | 5916 |
Feature activation | -7.90e-02 |
Pos logit contributions | |
oes | +0.018 |
talk | +0.017 |
ted | +0.017 |
rost | +0.016 |
use | +0.016 |
Neg logit contributions | |
adorable | -0.016 |
best | -0.015 |
idea | -0.015 |
important | -0.014 |
precious | -0.014 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
oes | +0.181 |
talk | +0.180 |
ted | +0.176 |
stack | +0.169 |
rost | +0.168 |
Neg logit contributions | |
adorable | -0.103 |
best | -0.100 |
idea | -0.098 |
important | -0.098 |
precious | -0.094 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
oes | +0.052 |
talk | +0.052 |
ted | +0.050 |
stack | +0.048 |
rost | +0.047 |
Neg logit contributions | |
adorable | -0.053 |
best | -0.051 |
important | -0.051 |
idea | -0.051 |
precious | -0.049 |
Token | had |
Token ID | 550 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
talk | +0.062 |
oes | +0.061 |
ted | +0.060 |
stack | +0.056 |
rost | +0.056 |
Neg logit contributions | |
adorable | -0.070 |
best | -0.068 |
idea | -0.066 |
important | -0.066 |
precious | -0.065 |
Token | been |
Token ID | 587 |
Feature activation | +9.68e-02 |
Pos logit contributions | |
adorable | +0.031 |
best | +0.030 |
idea | +0.029 |
important | +0.029 |
precious | +0.028 |
Neg logit contributions | |
oes | -0.028 |
talk | -0.028 |
ted | -0.027 |
stack | -0.025 |
rost | -0.025 |
Token | able |
Token ID | 1498 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
adorable | +0.146 |
idea | +0.146 |
best | +0.135 |
important | +0.133 |
precious | +0.133 |
Neg logit contributions | |
talk | -0.190 |
oes | -0.188 |
ted | -0.186 |
ting | -0.178 |
isk | -0.174 |
Token | to |
Token ID | 284 |
Feature activation | -1.98e-01 |
Pos logit contributions | |
adorable | +0.092 |
best | +0.090 |
idea | +0.089 |
important | +0.089 |
precious | +0.088 |
Neg logit contributions | |
oes | -0.034 |
talk | -0.033 |
ted | -0.031 |
stack | -0.028 |
rost | -0.027 |
Token | climb |
Token ID | 12080 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
oes | +0.298 |
talk | +0.292 |
ted | +0.287 |
rost | +0.262 |
stack | +0.259 |
Neg logit contributions | |
adorable | -0.420 |
best | -0.413 |
idea | -0.412 |
important | -0.408 |
precious | -0.393 |
Token | to |
Token ID | 284 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
talk | +0.324 |
oes | +0.322 |
ted | +0.302 |
rost | +0.295 |
isk | +0.295 |
Neg logit contributions | |
best | -0.313 |
adorable | -0.312 |
important | -0.303 |
idea | -0.300 |
beautiful | -0.292 |
Token | the |
Token ID | 262 |
Feature activation | -4.83e-02 |
Pos logit contributions | |
oes | +0.452 |
talk | +0.450 |
ted | +0.443 |
rost | +0.406 |
ricanes | +0.398 |
Neg logit contributions | |
adorable | -0.470 |
best | -0.465 |
idea | -0.461 |
important | -0.454 |
beautiful | -0.442 |
Token | top |
Token ID | 1353 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
oes | +0.081 |
talk | +0.079 |
ted | +0.076 |
stack | +0.074 |
rost | +0.072 |
Neg logit contributions | |
adorable | -0.093 |
best | -0.090 |
idea | -0.090 |
important | -0.089 |
precious | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | +8.87e-03 |
Pos logit contributions | |
talk | +0.082 |
oes | +0.080 |
ted | +0.080 |
rost | +0.077 |
stack | +0.076 |
Neg logit contributions | |
adorable | -0.042 |
best | -0.042 |
idea | -0.041 |
important | -0.040 |
beautiful | -0.038 |
Token | After |
Token ID | 2293 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
adorable | +0.017 |
best | +0.017 |
idea | +0.016 |
important | +0.016 |
loyal | +0.016 |
Neg logit contributions | |
talk | -0.015 |
oes | -0.015 |
ted | -0.014 |
rost | -0.013 |
use | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
oes | +0.038 |
talk | +0.037 |
ted | +0.035 |
rost | +0.033 |
stack | +0.033 |
Neg logit contributions | |
adorable | -0.035 |
best | -0.034 |
important | -0.033 |
idea | -0.033 |
precious | -0.032 |
Token | use |
Token ID | 779 |
Feature activation | -2.94e-03 |
Pos logit contributions | |
oes | +0.092 |
talk | +0.091 |
ted | +0.091 |
stack | +0.082 |
rost | +0.082 |
Neg logit contributions | |
adorable | -0.120 |
best | -0.118 |
idea | -0.118 |
important | -0.116 |
precious | -0.112 |
Token | it |
Token ID | 340 |
Feature activation | -9.54e-02 |
Pos logit contributions | |
oes | +0.005 |
talk | +0.005 |
ted | +0.004 |
stack | +0.004 |
rost | +0.004 |
Neg logit contributions | |
adorable | -0.006 |
best | -0.006 |
important | -0.006 |
idea | -0.006 |
precious | -0.006 |
Token | to |
Token ID | 284 |
Feature activation | -2.06e-01 |
Pos logit contributions | |
talk | +0.177 |
oes | +0.174 |
ted | +0.170 |
stack | +0.163 |
rost | +0.163 |
Neg logit contributions | |
best | -0.178 |
important | -0.171 |
adorable | -0.169 |
idea | -0.168 |
precious | -0.159 |
Token | climb |
Token ID | 12080 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ted | +0.331 |
oes | +0.327 |
talk | +0.325 |
ting | +0.290 |
rost | +0.290 |
Neg logit contributions | |
idea | -0.424 |
adorable | -0.420 |
best | -0.414 |
important | -0.406 |
beautiful | -0.402 |
Token | up |
Token ID | 510 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
talk | +0.317 |
oes | +0.313 |
isk | +0.297 |
ted | +0.296 |
rost | +0.286 |
Neg logit contributions | |
best | -0.296 |
adorable | -0.291 |
important | -0.284 |
idea | -0.281 |
beautiful | -0.276 |
Token | to |
Token ID | 284 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
oes | +0.240 |
talk | +0.237 |
ted | +0.230 |
stack | +0.218 |
rost | +0.217 |
Neg logit contributions | |
adorable | -0.210 |
best | -0.205 |
idea | -0.203 |
important | -0.202 |
precious | -0.195 |
Token | the |
Token ID | 262 |
Feature activation | -6.42e-02 |
Pos logit contributions | |
talk | +0.503 |
oes | +0.501 |
ted | +0.496 |
isk | +0.455 |
ricanes | +0.453 |
Neg logit contributions | |
best | -0.420 |
adorable | -0.420 |
idea | -0.414 |
beautiful | -0.408 |
important | -0.393 |
Token | balloon |
Token ID | 21190 |
Feature activation | -9.62e-02 |
Pos logit contributions | |
oes | +0.101 |
talk | +0.099 |
ted | +0.094 |
stack | +0.093 |
rost | +0.088 |
Neg logit contributions | |
adorable | -0.131 |
important | -0.126 |
best | -0.126 |
idea | -0.125 |
precious | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | +3.16e-03 |
Pos logit contributions | |
talk | +0.215 |
ted | +0.214 |
oes | +0.210 |
rost | +0.204 |
isk | +0.202 |
Neg logit contributions | |
best | -0.131 |
idea | -0.129 |
adorable | -0.127 |
important | -0.119 |
beautiful | -0.116 |
Token |
|
Token ID | 198 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
adorable | +0.005 |
best | +0.005 |
important | +0.004 |
idea | +0.004 |
loyal | +0.004 |
Neg logit contributions | |
oes | -0.007 |
talk | -0.006 |
isk | -0.006 |
ted | -0.006 |
ting | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
oes | +0.065 |
talk | +0.063 |
ted | +0.062 |
stack | +0.059 |
ting | +0.058 |
Neg logit contributions | |
adorable | -0.060 |
important | -0.058 |
best | -0.057 |
idea | -0.057 |
beautiful | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +4.76e-03 |
Pos logit contributions | |
best | +0.178 |
adorable | +0.174 |
idea | +0.168 |
important | +0.163 |
beautiful | +0.157 |
Neg logit contributions | |
talk | -0.320 |
oes | -0.317 |
stack | -0.308 |
ted | -0.307 |
isk | -0.303 |
Token |
|
Token ID | 198 |
Feature activation | -3.22e-02 |
Pos logit contributions | |
adorable | +0.007 |
best | +0.007 |
important | +0.007 |
loyal | +0.007 |
beautiful | +0.007 |
Neg logit contributions | |
talk | -0.011 |
oes | -0.010 |
isk | -0.010 |
ted | -0.010 |
ales | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
talk | +0.060 |
oes | +0.060 |
ted | +0.059 |
stack | +0.056 |
ting | +0.055 |
Neg logit contributions | |
adorable | -0.057 |
important | -0.055 |
best | -0.055 |
idea | -0.053 |
precious | -0.052 |
Token | " |
Token ID | 1 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
oes | +0.063 |
talk | +0.062 |
ted | +0.061 |
stack | +0.058 |
isk | +0.057 |
Neg logit contributions | |
adorable | -0.037 |
best | -0.036 |
important | -0.036 |
precious | -0.034 |
loyal | -0.034 |
Token | Time |
Token ID | 7575 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
oes | +0.031 |
talk | +0.031 |
ted | +0.029 |
stack | +0.028 |
isk | +0.027 |
Neg logit contributions | |
adorable | -0.043 |
best | -0.042 |
important | -0.041 |
idea | -0.041 |
precious | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
talk | +0.070 |
oes | +0.069 |
stack | +0.063 |
ted | +0.063 |
isk | +0.060 |
Neg logit contributions | |
best | -0.124 |
adorable | -0.123 |
important | -0.121 |
idea | -0.121 |
precious | -0.117 |
Token | go |
Token ID | 467 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
oes | +0.395 |
talk | +0.394 |
ted | +0.387 |
stack | +0.356 |
rost | +0.349 |
Neg logit contributions | |
idea | -0.523 |
adorable | -0.520 |
best | -0.511 |
beautiful | -0.499 |
important | -0.491 |
Token | home |
Token ID | 1363 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
talk | +0.267 |
oes | +0.264 |
ted | +0.260 |
isk | +0.255 |
rost | +0.255 |
Neg logit contributions | |
best | -0.139 |
adorable | -0.135 |
idea | -0.132 |
important | -0.126 |
beautiful | -0.124 |
Token | , |
Token ID | 11 |
Feature activation | -7.63e-02 |
Pos logit contributions | |
oes | +0.285 |
talk | +0.282 |
stack | +0.264 |
ted | +0.262 |
rost | +0.259 |
Neg logit contributions | |
adorable | -0.249 |
important | -0.235 |
idea | -0.232 |
precious | -0.230 |
best | -0.227 |
Token | kids |
Token ID | 3988 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
oes | +0.127 |
ted | +0.123 |
talk | +0.121 |
rost | +0.120 |
stack | +0.117 |
Neg logit contributions | |
important | -0.140 |
adorable | -0.138 |
idea | -0.136 |
prized | -0.133 |
best | -0.133 |
Token | ," |
Token ID | 553 |
Feature activation | -8.10e-03 |
Pos logit contributions | |
oes | +0.118 |
talk | +0.117 |
ted | +0.114 |
stack | +0.110 |
rost | +0.109 |
Neg logit contributions | |
adorable | -0.060 |
best | -0.057 |
important | -0.056 |
idea | -0.056 |
precious | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | -6.69e-02 |
Pos logit contributions | |
talk | +0.066 |
oes | +0.066 |
ted | +0.063 |
stack | +0.062 |
rost | +0.061 |
Neg logit contributions | |
adorable | -0.076 |
best | -0.074 |
important | -0.073 |
idea | -0.072 |
loyal | -0.070 |
Token | said |
Token ID | 531 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
oes | +0.131 |
stack | +0.124 |
talk | +0.117 |
rost | +0.116 |
isk | +0.112 |
Neg logit contributions | |
adorable | -0.120 |
precious | -0.113 |
beautiful | -0.113 |
loyal | -0.112 |
intelligent | -0.111 |
Token | , |
Token ID | 11 |
Feature activation | -1.09e-02 |
Pos logit contributions | |
adorable | +0.126 |
best | +0.125 |
important | +0.120 |
idea | +0.117 |
precious | +0.111 |
Neg logit contributions | |
oes | -0.249 |
talk | -0.246 |
ted | -0.237 |
stack | -0.234 |
isk | -0.232 |
Token | " |
Token ID | 366 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
talk | +0.028 |
oes | +0.027 |
ales | +0.026 |
ted | +0.025 |
aled | +0.025 |
Neg logit contributions | |
adorable | -0.013 |
loyal | -0.013 |
important | -0.012 |
best | -0.012 |
generous | -0.012 |
Token | Time |
Token ID | 7575 |
Feature activation | -5.47e-02 |
Pos logit contributions | |
stack | +0.060 |
oes | +0.057 |
rost | +0.050 |
gra | +0.050 |
talk | +0.049 |
Neg logit contributions | |
adorable | -0.061 |
loyal | -0.058 |
popular | -0.057 |
important | -0.057 |
beautiful | -0.057 |
Token | to |
Token ID | 284 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
talk | +0.070 |
oes | +0.070 |
stack | +0.063 |
ted | +0.063 |
rost | +0.061 |
Neg logit contributions | |
best | -0.128 |
adorable | -0.127 |
important | -0.126 |
idea | -0.125 |
precious | -0.121 |
Token | go |
Token ID | 467 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
oes | +0.377 |
talk | +0.372 |
ted | +0.362 |
stack | +0.332 |
rost | +0.327 |
Neg logit contributions | |
adorable | -0.545 |
idea | -0.540 |
best | -0.532 |
important | -0.523 |
beautiful | -0.513 |
Token | , |
Token ID | 11 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
talk | +0.243 |
ted | +0.237 |
oes | +0.237 |
rost | +0.235 |
stack | +0.231 |
Neg logit contributions | |
best | -0.153 |
adorable | -0.151 |
idea | -0.147 |
important | -0.145 |
beautiful | -0.139 |
Token | kids |
Token ID | 3988 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
isk | +0.192 |
ted | +0.191 |
ting | +0.191 |
oes | +0.191 |
talk | +0.189 |
Neg logit contributions | |
important | -0.144 |
prized | -0.143 |
adorable | -0.141 |
best | -0.141 |
precious | -0.139 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
oes | +0.167 |
talk | +0.165 |
ted | +0.161 |
stack | +0.157 |
rost | +0.155 |
Neg logit contributions | |
adorable | -0.074 |
best | -0.073 |
important | -0.070 |
idea | -0.070 |
precious | -0.065 |
Token | The |
Token ID | 383 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
oes | +0.031 |
talk | +0.031 |
stack | +0.029 |
isk | +0.029 |
ted | +0.029 |
Neg logit contributions | |
adorable | -0.019 |
best | -0.018 |
important | -0.017 |
loyal | -0.017 |
precious | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
oes | +0.053 |
talk | +0.053 |
ted | +0.050 |
isk | +0.049 |
stack | +0.049 |
Neg logit contributions | |
adorable | -0.090 |
best | -0.089 |
important | -0.088 |
idea | -0.087 |
beautiful | -0.085 |
Token | the |
Token ID | 262 |
Feature activation | -4.46e-02 |
Pos logit contributions | |
oes | +0.373 |
talk | +0.371 |
ted | +0.355 |
isk | +0.337 |
stack | +0.335 |
Neg logit contributions | |
adorable | -0.426 |
best | -0.416 |
idea | -0.416 |
important | -0.403 |
beautiful | -0.393 |
Token | hill |
Token ID | 12788 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
oes | +0.059 |
talk | +0.056 |
ted | +0.055 |
stack | +0.054 |
rost | +0.050 |
Neg logit contributions | |
adorable | -0.101 |
important | -0.098 |
idea | -0.098 |
best | -0.097 |
precious | -0.096 |
Token | and |
Token ID | 290 |
Feature activation | -8.98e-02 |
Pos logit contributions | |
oes | +0.312 |
talk | +0.308 |
ted | +0.299 |
stack | +0.288 |
rost | +0.284 |
Neg logit contributions | |
adorable | -0.217 |
best | -0.209 |
important | -0.207 |
idea | -0.206 |
precious | -0.199 |
Token | climbed |
Token ID | 19952 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
oes | +0.160 |
talk | +0.157 |
rost | +0.155 |
isk | +0.155 |
use | +0.147 |
Neg logit contributions | |
adorable | -0.166 |
best | -0.159 |
beautiful | -0.155 |
loyal | -0.152 |
idea | -0.151 |
Token | to |
Token ID | 284 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
talk | +0.256 |
oes | +0.254 |
ted | +0.241 |
isk | +0.236 |
rost | +0.233 |
Neg logit contributions | |
adorable | -0.272 |
best | -0.271 |
idea | -0.264 |
important | -0.262 |
precious | -0.253 |
Token | the |
Token ID | 262 |
Feature activation | -7.03e-02 |
Pos logit contributions | |
talk | +0.427 |
oes | +0.423 |
ted | +0.413 |
isk | +0.388 |
rost | +0.387 |
Neg logit contributions | |
adorable | -0.492 |
best | -0.485 |
idea | -0.482 |
important | -0.466 |
beautiful | -0.465 |
Token | top |
Token ID | 1353 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
oes | +0.110 |
talk | +0.105 |
ted | +0.103 |
stack | +0.102 |
rost | +0.095 |
Neg logit contributions | |
adorable | -0.143 |
important | -0.138 |
idea | -0.137 |
best | -0.136 |
precious | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | -4.62e-03 |
Pos logit contributions | |
oes | +0.099 |
talk | +0.098 |
ted | +0.095 |
stack | +0.091 |
rost | +0.090 |
Neg logit contributions | |
adorable | -0.072 |
best | -0.070 |
important | -0.069 |
idea | -0.068 |
precious | -0.067 |
Token | They |
Token ID | 1119 |
Feature activation | -7.48e-02 |
Pos logit contributions | |
talk | +0.009 |
use | +0.009 |
isk | +0.009 |
oes | +0.008 |
x | +0.008 |
Neg logit contributions | |
idea | -0.008 |
adorable | -0.008 |
best | -0.008 |
thrilled | -0.007 |
excited | -0.007 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
oes | +0.115 |
talk | +0.113 |
ted | +0.108 |
rost | +0.102 |
stack | +0.102 |
Neg logit contributions | |
adorable | -0.157 |
best | -0.154 |
idea | -0.151 |
important | -0.150 |
precious | -0.146 |
Token | Once |
Token ID | 7454 |
Feature activation | -7.09e-02 |
Pos logit contributions | |
oes | +0.265 |
talk | +0.259 |
ted | +0.244 |
stack | +0.242 |
ricanes | +0.239 |
Neg logit contributions | |
adorable | -0.216 |
best | -0.215 |
idea | -0.205 |
important | -0.204 |
beautiful | -0.200 |
Token | upon |
Token ID | 2402 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
oes | +0.111 |
talk | +0.106 |
ted | +0.103 |
rost | +0.098 |
stack | +0.097 |
Neg logit contributions | |
adorable | -0.145 |
best | -0.141 |
important | -0.140 |
idea | -0.139 |
precious | -0.137 |
Token | a |
Token ID | 257 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
oes | +0.089 |
talk | +0.089 |
ted | +0.086 |
rost | +0.084 |
gra | +0.083 |
Neg logit contributions | |
best | -0.066 |
idea | -0.064 |
precious | -0.064 |
adorable | -0.064 |
important | -0.064 |
Token | time |
Token ID | 640 |
Feature activation | -6.64e-02 |
Pos logit contributions | |
adorable | +0.181 |
best | +0.175 |
precious | +0.173 |
important | +0.171 |
idea | +0.168 |
Neg logit contributions | |
oes | -0.218 |
talk | -0.215 |
rost | -0.213 |
isk | -0.210 |
ted | -0.200 |
Token | , |
Token ID | 11 |
Feature activation | -5.75e-02 |
Pos logit contributions | |
oes | +0.132 |
ted | +0.126 |
talk | +0.124 |
rost | +0.122 |
stack | +0.121 |
Neg logit contributions | |
adorable | -0.102 |
important | -0.101 |
idea | -0.099 |
precious | -0.099 |
best | -0.095 |
Token | a |
Token ID | 257 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
oes | +0.087 |
ted | +0.083 |
talk | +0.082 |
stack | +0.080 |
rost | +0.079 |
Neg logit contributions | |
adorable | -0.114 |
important | -0.112 |
idea | -0.112 |
best | -0.111 |
precious | -0.110 |
Token | little |
Token ID | 1310 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
idea | +0.140 |
best | +0.134 |
adorable | +0.134 |
important | +0.132 |
precious | +0.127 |
Neg logit contributions | |
oes | -0.389 |
rost | -0.382 |
Dog | -0.375 |
talk | -0.374 |
isk | -0.370 |
Token | girl |
Token ID | 2576 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
important | +0.174 |
idea | +0.172 |
adorable | +0.170 |
best | +0.167 |
precious | +0.165 |
Neg logit contributions | |
oes | -0.242 |
ted | -0.233 |
talk | -0.230 |
Dog | -0.220 |
rost | -0.219 |
Token | named |
Token ID | 3706 |
Feature activation | +9.92e-03 |
Pos logit contributions | |
adorable | +0.113 |
best | +0.110 |
important | +0.108 |
idea | +0.107 |
precious | +0.104 |
Neg logit contributions | |
oes | -0.097 |
talk | -0.096 |
ted | -0.092 |
stack | -0.088 |
rost | -0.087 |
Token | Lily |
Token ID | 20037 |
Feature activation | +7.91e-02 |
Pos logit contributions | |
adorable | +0.018 |
best | +0.018 |
important | +0.017 |
idea | +0.017 |
beautiful | +0.017 |
Neg logit contributions | |
talk | -0.018 |
oes | -0.018 |
ted | -0.017 |
stack | -0.017 |
rost | -0.016 |
Token | went |
Token ID | 1816 |
Feature activation | -6.48e-02 |
Pos logit contributions | |
adorable | +0.166 |
best | +0.162 |
important | +0.160 |
idea | +0.160 |
precious | +0.155 |
Neg logit contributions | |
oes | -0.121 |
talk | -0.119 |
ted | -0.115 |
stack | -0.109 |
rost | -0.107 |
Token | , |
Token ID | 11 |
Feature activation | -4.02e-02 |
Pos logit contributions | |
oes | +0.144 |
talk | +0.140 |
ted | +0.137 |
stack | +0.133 |
isk | +0.130 |
Neg logit contributions | |
adorable | -0.103 |
important | -0.096 |
precious | -0.096 |
idea | -0.095 |
best | -0.093 |
Token | the |
Token ID | 262 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
oes | +0.064 |
talk | +0.063 |
ted | +0.059 |
stack | +0.056 |
rost | +0.056 |
Neg logit contributions | |
adorable | -0.083 |
best | -0.082 |
important | -0.080 |
idea | -0.080 |
precious | -0.078 |
Token | daughter |
Token ID | 4957 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
adorable | +0.049 |
idea | +0.048 |
impressed | +0.048 |
prized | +0.047 |
important | +0.046 |
Neg logit contributions | |
ted | -0.080 |
oes | -0.078 |
talk | -0.077 |
ting | -0.077 |
stack | -0.075 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
adorable | +0.099 |
best | +0.096 |
idea | +0.096 |
important | +0.096 |
precious | +0.094 |
Neg logit contributions | |
talk | -0.084 |
oes | -0.082 |
ted | -0.082 |
stack | -0.076 |
isk | -0.075 |
Token | a |
Token ID | 257 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
adorable | +0.031 |
best | +0.030 |
important | +0.030 |
idea | +0.030 |
precious | +0.029 |
Neg logit contributions | |
oes | -0.039 |
talk | -0.039 |
ted | -0.038 |
stack | -0.036 |
rost | -0.036 |
Token | pretty |
Token ID | 2495 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
adorable | +0.238 |
idea | +0.225 |
best | +0.224 |
important | +0.221 |
precious | +0.208 |
Neg logit contributions | |
stack | -0.333 |
gra | -0.329 |
oes | -0.326 |
ricane | -0.325 |
hike | -0.321 |
Token | dress |
Token ID | 6576 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
best | +0.337 |
important | +0.328 |
adorable | +0.326 |
pleased | +0.320 |
precious | +0.319 |
Neg logit contributions | |
ted | -0.309 |
stack | -0.308 |
oes | -0.294 |
gra | -0.289 |
benches | -0.287 |
Token | . |
Token ID | 13 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
adorable | +0.060 |
best | +0.058 |
important | +0.058 |
idea | +0.057 |
precious | +0.056 |
Neg logit contributions | |
oes | -0.066 |
talk | -0.065 |
ted | -0.063 |
stack | -0.060 |
isk | -0.059 |
Token | She |
Token ID | 1375 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
adorable | +0.044 |
idea | +0.041 |
best | +0.041 |
important | +0.041 |
beautiful | +0.040 |
Neg logit contributions | |
talk | -0.041 |
oes | -0.040 |
ted | -0.037 |
rost | -0.036 |
use | -0.036 |
Token | asked |
Token ID | 1965 |
Feature activation | -3.65e-03 |
Pos logit contributions | |
ted | +0.034 |
talk | +0.033 |
oes | +0.032 |
isk | +0.030 |
stack | +0.029 |
Neg logit contributions | |
best | -0.049 |
idea | -0.048 |
adorable | -0.047 |
important | -0.047 |
precious | -0.045 |
Token | her |
Token ID | 607 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
oes | +0.006 |
talk | +0.006 |
ted | +0.006 |
stack | +0.006 |
rost | +0.006 |
Neg logit contributions | |
adorable | -0.007 |
best | -0.007 |
important | -0.007 |
idea | -0.007 |
precious | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
talk | +0.060 |
oes | +0.060 |
ted | +0.059 |
rost | +0.057 |
iffs | +0.056 |
Neg logit contributions | |
best | -0.061 |
idea | -0.060 |
precious | -0.059 |
important | -0.059 |
prized | -0.058 |
Token | time |
Token ID | 640 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
adorable | +0.185 |
precious | +0.180 |
important | +0.177 |
best | +0.175 |
idea | +0.169 |
Neg logit contributions | |
oes | -0.208 |
rost | -0.203 |
talk | -0.202 |
isk | -0.199 |
use | -0.190 |
Token | , |
Token ID | 11 |
Feature activation | -5.56e-02 |
Pos logit contributions | |
oes | +0.118 |
ted | +0.110 |
iffs | +0.107 |
talk | +0.107 |
rost | +0.106 |
Neg logit contributions | |
important | -0.106 |
adorable | -0.106 |
precious | -0.104 |
idea | -0.103 |
valuable | -0.100 |
Token | there |
Token ID | 612 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
oes | +0.085 |
talk | +0.079 |
ted | +0.079 |
rost | +0.077 |
stack | +0.075 |
Neg logit contributions | |
adorable | -0.110 |
idea | -0.108 |
important | -0.107 |
best | -0.106 |
precious | -0.106 |
Token | was |
Token ID | 373 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
oes | +0.261 |
ted | +0.254 |
talk | +0.250 |
rost | +0.241 |
stack | +0.239 |
Neg logit contributions | |
adorable | -0.250 |
important | -0.243 |
precious | -0.241 |
best | -0.238 |
idea | -0.238 |
Token | a |
Token ID | 257 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
precious | +0.175 |
idea | +0.172 |
pleased | +0.171 |
prized | +0.170 |
important | +0.167 |
Neg logit contributions | |
ted | -0.168 |
oes | -0.153 |
rost | -0.152 |
gra | -0.151 |
talk | -0.150 |
Token | little |
Token ID | 1310 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
idea | +0.198 |
best | +0.185 |
important | +0.185 |
mistakes | +0.179 |
prized | +0.177 |
Neg logit contributions | |
Dog | -0.380 |
gra | -0.368 |
rost | -0.367 |
oes | -0.364 |
isk | -0.352 |
Token | boy |
Token ID | 2933 |
Feature activation | +5.99e-02 |
Pos logit contributions | |
important | +0.217 |
idea | +0.212 |
prized | +0.210 |
thrilled | +0.210 |
pleased | +0.208 |
Neg logit contributions | |
oes | -0.234 |
ted | -0.226 |
Dog | -0.221 |
ting | -0.216 |
talk | -0.214 |
Token | named |
Token ID | 3706 |
Feature activation | +2.19e-02 |
Pos logit contributions | |
adorable | +0.121 |
best | +0.118 |
important | +0.117 |
idea | +0.116 |
precious | +0.113 |
Neg logit contributions | |
oes | -0.096 |
talk | -0.095 |
ted | -0.091 |
stack | -0.087 |
rost | -0.085 |
Token | Tim |
Token ID | 5045 |
Feature activation | +4.96e-02 |
Pos logit contributions | |
adorable | +0.038 |
best | +0.035 |
important | +0.034 |
beautiful | +0.034 |
loyal | +0.033 |
Neg logit contributions | |
talk | -0.047 |
ted | -0.043 |
stack | -0.043 |
ting | -0.041 |
oes | -0.041 |
Token | my |
Token ID | 1820 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
adorable | +0.100 |
best | +0.097 |
important | +0.096 |
idea | +0.096 |
precious | +0.093 |
Neg logit contributions | |
oes | -0.080 |
talk | -0.079 |
ted | -0.076 |
stack | -0.072 |
rost | -0.071 |
Token | right |
Token ID | 826 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
ted | +0.089 |
oes | +0.088 |
talk | +0.086 |
rost | +0.084 |
stack | +0.081 |
Neg logit contributions | |
best | -0.089 |
important | -0.088 |
idea | -0.087 |
adorable | -0.084 |
precious | -0.080 |
Token | away |
Token ID | 1497 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
important | +0.098 |
idea | +0.097 |
best | +0.097 |
adorable | +0.090 |
precious | +0.087 |
Neg logit contributions | |
gra | -0.082 |
ted | -0.082 |
Dog | -0.082 |
oes | -0.082 |
talk | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +6.88e-03 |
Pos logit contributions | |
talk | +0.120 |
oes | +0.120 |
ted | +0.116 |
rost | +0.111 |
stack | +0.110 |
Neg logit contributions | |
best | -0.074 |
adorable | -0.074 |
important | -0.073 |
idea | -0.070 |
precious | -0.067 |
Token | Lucy |
Token ID | 22162 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
adorable | +0.013 |
important | +0.012 |
idea | +0.012 |
best | +0.012 |
loyal | +0.012 |
Neg logit contributions | |
oes | -0.013 |
talk | -0.012 |
ted | -0.012 |
ricanes | -0.011 |
use | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
adorable | +0.080 |
best | +0.078 |
idea | +0.077 |
important | +0.077 |
precious | +0.074 |
Neg logit contributions | |
talk | -0.093 |
oes | -0.093 |
ted | -0.091 |
stack | -0.086 |
isk | -0.085 |
Token | scared |
Token ID | 12008 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
idea | +0.149 |
adorable | +0.147 |
best | +0.143 |
precious | +0.138 |
important | +0.135 |
Neg logit contributions | |
talk | -0.317 |
ted | -0.310 |
oes | -0.309 |
isk | -0.306 |
stack | -0.305 |
Token | and |
Token ID | 290 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
adorable | +0.227 |
best | +0.219 |
important | +0.216 |
idea | +0.215 |
precious | +0.207 |
Neg logit contributions | |
oes | -0.346 |
talk | -0.343 |
ted | -0.333 |
stack | -0.321 |
rost | -0.317 |
Token | began |
Token ID | 2540 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
oes | +0.055 |
talk | +0.053 |
iffs | +0.050 |
rost | +0.049 |
ted | +0.048 |
Neg logit contributions | |
adorable | -0.037 |
important | -0.036 |
best | -0.034 |
idea | -0.034 |
loyal | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
important | +0.023 |
best | +0.023 |
adorable | +0.023 |
loyal | +0.022 |
idea | +0.022 |
Neg logit contributions | |
oes | -0.017 |
talk | -0.016 |
rost | -0.015 |
ted | -0.015 |
stack | -0.015 |
Token | cry |
Token ID | 3960 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
oes | +0.402 |
talk | +0.397 |
ted | +0.393 |
rost | +0.369 |
iffs | +0.368 |
Neg logit contributions | |
idea | -0.299 |
adorable | -0.299 |
best | -0.287 |
important | -0.284 |
beautiful | -0.271 |
Token | , |
Token ID | 11 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
adorable | +0.083 |
best | +0.081 |
important | +0.080 |
idea | +0.079 |
precious | +0.076 |
Neg logit contributions | |
oes | -0.112 |
talk | -0.111 |
ted | -0.108 |
stack | -0.104 |
rost | -0.102 |
Token | One |
Token ID | 1881 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
talk | +0.012 |
oes | +0.011 |
ted | +0.011 |
stack | +0.011 |
isk | +0.011 |
Neg logit contributions | |
adorable | -0.013 |
best | -0.013 |
idea | -0.012 |
important | -0.012 |
beautiful | -0.011 |
Token | day |
Token ID | 1110 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
adorable | +0.026 |
precious | +0.026 |
best | +0.025 |
prized | +0.025 |
important | +0.025 |
Neg logit contributions | |
oes | -0.026 |
talk | -0.023 |
rost | -0.022 |
ted | -0.022 |
stack | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
oes | +0.146 |
ted | +0.145 |
talk | +0.143 |
rost | +0.135 |
iffs | +0.133 |
Neg logit contributions | |
adorable | -0.102 |
important | -0.097 |
idea | -0.097 |
precious | -0.095 |
best | -0.094 |
Token | they |
Token ID | 484 |
Feature activation | -8.01e-02 |
Pos logit contributions | |
oes | +0.056 |
talk | +0.055 |
ted | +0.054 |
stack | +0.048 |
rost | +0.048 |
Neg logit contributions | |
adorable | -0.106 |
best | -0.103 |
important | -0.103 |
idea | -0.103 |
precious | -0.101 |
Token | find |
Token ID | 1064 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
oes | +0.138 |
ted | +0.136 |
talk | +0.134 |
stack | +0.127 |
isk | +0.125 |
Neg logit contributions | |
adorable | -0.151 |
idea | -0.150 |
best | -0.149 |
important | -0.145 |
precious | -0.140 |
Token | a |
Token ID | 257 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
talk | +0.024 |
ted | +0.024 |
oes | +0.024 |
rost | +0.022 |
stack | +0.022 |
Neg logit contributions | |
adorable | -0.016 |
idea | -0.015 |
important | -0.015 |
precious | -0.014 |
best | -0.014 |
Token | shiny |
Token ID | 22441 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
adorable | +0.185 |
idea | +0.173 |
best | +0.169 |
important | +0.169 |
precious | +0.159 |
Neg logit contributions | |
oes | -0.387 |
talk | -0.367 |
stack | -0.364 |
rost | -0.359 |
ted | -0.358 |
Token | coin |
Token ID | 10752 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
pleased | +0.312 |
impressed | +0.310 |
thrilled | +0.308 |
adorable | +0.305 |
delighted | +0.292 |
Neg logit contributions | |
oes | -0.321 |
ted | -0.304 |
stack | -0.302 |
talk | -0.291 |
rost | -0.290 |
Token | under |
Token ID | 739 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
adorable | +0.044 |
best | +0.043 |
important | +0.043 |
idea | +0.043 |
precious | +0.041 |
Neg logit contributions | |
oes | -0.036 |
talk | -0.035 |
ted | -0.034 |
stack | -0.032 |
rost | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
oes | +0.229 |
talk | +0.227 |
ted | +0.220 |
stack | +0.211 |
rost | +0.209 |
Neg logit contributions | |
adorable | -0.162 |
best | -0.156 |
important | -0.154 |
idea | -0.153 |
precious | -0.148 |
Token | bed |
Token ID | 3996 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
talk | +0.094 |
oes | +0.092 |
ted | +0.091 |
isk | +0.088 |
rost | +0.086 |
Neg logit contributions | |
adorable | -0.088 |
best | -0.086 |
beautiful | -0.083 |
idea | -0.082 |
important | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
oes | +0.102 |
talk | +0.097 |
ted | +0.090 |
stack | +0.090 |
rost | +0.089 |
Neg logit contributions | |
adorable | -0.100 |
best | -0.094 |
important | -0.094 |
idea | -0.094 |
precious | -0.092 |
Token | mess |
Token ID | 2085 |
Feature activation | -8.67e-03 |
Pos logit contributions | |
adorable | +0.040 |
best | +0.039 |
important | +0.038 |
idea | +0.038 |
precious | +0.037 |
Neg logit contributions | |
oes | -0.043 |
talk | -0.043 |
ted | -0.041 |
stack | -0.040 |
rost | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | -6.25e-02 |
Pos logit contributions | |
oes | +0.017 |
talk | +0.017 |
ted | +0.016 |
stack | +0.016 |
rost | +0.015 |
Neg logit contributions | |
adorable | -0.014 |
best | -0.014 |
important | -0.014 |
idea | -0.014 |
precious | -0.013 |
Token | the |
Token ID | 262 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
oes | +0.128 |
talk | +0.124 |
use | +0.114 |
rost | +0.113 |
iffs | +0.111 |
Neg logit contributions | |
adorable | -0.107 |
important | -0.101 |
best | -0.101 |
idea | -0.100 |
beautiful | -0.099 |
Token | blood |
Token ID | 2910 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
adorable | +0.025 |
best | +0.024 |
important | +0.024 |
idea | +0.024 |
precious | +0.023 |
Neg logit contributions | |
oes | -0.029 |
talk | -0.028 |
ted | -0.027 |
stack | -0.027 |
rost | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
oes | +0.077 |
talk | +0.076 |
stack | +0.071 |
rost | +0.071 |
ricanes | +0.069 |
Neg logit contributions | |
adorable | -0.052 |
best | -0.052 |
important | -0.050 |
idea | -0.050 |
beautiful | -0.047 |
Token | She |
Token ID | 1375 |
Feature activation | -3.97e-02 |
Pos logit contributions | |
adorable | +0.023 |
important | +0.022 |
idea | +0.022 |
best | +0.022 |
precious | +0.022 |
Neg logit contributions | |
oes | -0.021 |
talk | -0.020 |
ted | -0.019 |
isk | -0.019 |
rost | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
talk | +0.074 |
oes | +0.074 |
ted | +0.073 |
isk | +0.068 |
stack | +0.068 |
Neg logit contributions | |
adorable | -0.068 |
best | -0.067 |
idea | -0.066 |
important | -0.066 |
precious | -0.063 |
Token | very |
Token ID | 845 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
idea | +0.158 |
adorable | +0.156 |
best | +0.153 |
important | +0.148 |
precious | +0.143 |
Neg logit contributions | |
talk | -0.307 |
oes | -0.304 |
ted | -0.302 |
stack | -0.295 |
isk | -0.294 |
Token | angry |
Token ID | 7954 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
best | +0.200 |
adorable | +0.198 |
idea | +0.197 |
prized | +0.191 |
important | +0.181 |
Neg logit contributions | |
talk | -0.347 |
oes | -0.342 |
ted | -0.334 |
rost | -0.332 |
iffs | -0.332 |
Token | and |
Token ID | 290 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
adorable | +0.237 |
best | +0.229 |
important | +0.226 |
idea | +0.225 |
precious | +0.217 |
Neg logit contributions | |
oes | -0.346 |
talk | -0.343 |
ted | -0.334 |
stack | -0.318 |
rost | -0.316 |
Token | she |
Token ID | 673 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
adorable | +0.115 |
best | +0.111 |
idea | +0.111 |
important | +0.110 |
precious | +0.107 |
Neg logit contributions | |
oes | -0.103 |
talk | -0.101 |
ted | -0.099 |
stack | -0.093 |
rost | -0.092 |
Token | wouldn |
Token ID | 3636 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
oes | +0.044 |
talk | +0.044 |
ted | +0.042 |
stack | +0.038 |
rost | +0.037 |
Neg logit contributions | |
adorable | -0.090 |
idea | -0.088 |
best | -0.088 |
important | -0.087 |
precious | -0.085 |
Token | 't |
Token ID | 470 |
Feature activation | +2.58e-02 |
Pos logit contributions | |
oes | +0.024 |
talk | +0.024 |
ted | +0.023 |
stack | +0.021 |
rost | +0.021 |
Neg logit contributions | |
adorable | -0.035 |
best | -0.034 |
important | -0.034 |
idea | -0.034 |
precious | -0.033 |
Token | get |
Token ID | 651 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
adorable | +0.052 |
best | +0.051 |
idea | +0.051 |
important | +0.051 |
beautiful | +0.048 |
Neg logit contributions | |
ted | -0.042 |
talk | -0.041 |
oes | -0.041 |
stack | -0.038 |
iffs | -0.038 |
Token | hurt |
Token ID | 5938 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
adorable | +0.031 |
best | +0.030 |
important | +0.030 |
precious | +0.029 |
idea | +0.029 |
Neg logit contributions | |
oes | -0.026 |
talk | -0.025 |
ted | -0.024 |
stack | -0.024 |
isk | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +8.57e-03 |
Pos logit contributions | |
adorable | +0.086 |
best | +0.083 |
idea | +0.083 |
important | +0.083 |
precious | +0.080 |
Neg logit contributions | |
oes | -0.097 |
talk | -0.095 |
ted | -0.093 |
stack | -0.088 |
rost | -0.087 |
Token |
|
Token ID | 198 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
adorable | +0.019 |
best | +0.019 |
important | +0.018 |
idea | +0.018 |
precious | +0.018 |
Neg logit contributions | |
oes | -0.012 |
talk | -0.011 |
ted | -0.011 |
stack | -0.010 |
rost | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
talk | +0.037 |
oes | +0.037 |
ted | +0.035 |
rost | +0.033 |
stack | +0.033 |
Neg logit contributions | |
adorable | -0.055 |
best | -0.054 |
idea | -0.054 |
important | -0.053 |
precious | -0.052 |
Token | When |
Token ID | 2215 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
talk | +0.031 |
oes | +0.030 |
ted | +0.028 |
rost | +0.027 |
Dog | +0.026 |
Neg logit contributions | |
adorable | -0.044 |
best | -0.044 |
idea | -0.043 |
precious | -0.041 |
important | -0.041 |
Token | they |
Token ID | 484 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
talk | +0.075 |
oes | +0.075 |
ted | +0.074 |
stack | +0.068 |
rost | +0.066 |
Neg logit contributions | |
adorable | -0.106 |
idea | -0.101 |
important | -0.101 |
best | -0.101 |
precious | -0.101 |
Token | got |
Token ID | 1392 |
Feature activation | +4.24e-02 |
Pos logit contributions | |
oes | +0.202 |
ricanes | +0.179 |
stack | +0.178 |
talk | +0.177 |
iffs | +0.174 |
Neg logit contributions | |
best | -0.196 |
adorable | -0.193 |
idea | -0.192 |
beautiful | -0.183 |
amazing | -0.182 |
Token | an |
Token ID | 281 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
oes | +0.073 |
talk | +0.073 |
ted | +0.070 |
stack | +0.067 |
rost | +0.066 |
Neg logit contributions | |
adorable | -0.076 |
best | -0.074 |
important | -0.073 |
idea | -0.073 |
precious | -0.070 |
Token | important |
Token ID | 1593 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
best | +0.244 |
adorable | +0.244 |
precious | +0.237 |
pleased | +0.235 |
impressed | +0.230 |
Neg logit contributions | |
oes | -0.197 |
talk | -0.189 |
rost | -0.178 |
isk | -0.171 |
hike | -0.166 |
Token | report |
Token ID | 989 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
adorable | +0.175 |
best | +0.164 |
precious | +0.157 |
important | +0.156 |
idea | +0.154 |
Neg logit contributions | |
oes | -0.310 |
talk | -0.309 |
ted | -0.291 |
rost | -0.287 |
stack | -0.284 |
Token | for |
Token ID | 329 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
oes | +0.083 |
talk | +0.082 |
ted | +0.080 |
stack | +0.078 |
rost | +0.077 |
Neg logit contributions | |
adorable | -0.066 |
best | -0.066 |
important | -0.065 |
idea | -0.064 |
precious | -0.060 |
Token | her |
Token ID | 607 |
Feature activation | +1.68e-02 |
Pos logit contributions | |
oes | +0.105 |
talk | +0.104 |
ted | +0.099 |
stack | +0.096 |
iffs | +0.094 |
Neg logit contributions | |
adorable | -0.092 |
important | -0.091 |
best | -0.090 |
idea | -0.086 |
beautiful | -0.085 |
Token | boss |
Token ID | 6478 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
adorable | +0.026 |
best | +0.025 |
important | +0.025 |
idea | +0.025 |
precious | +0.024 |
Neg logit contributions | |
oes | -0.033 |
talk | -0.032 |
ted | -0.031 |
rost | -0.030 |
isk | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +6.76e-03 |
Pos logit contributions | |
adorable | +0.077 |
best | +0.076 |
idea | +0.075 |
important | +0.075 |
precious | +0.070 |
Neg logit contributions | |
oes | -0.100 |
talk | -0.099 |
ted | -0.097 |
stack | -0.096 |
rost | -0.093 |
Token | She |
Token ID | 1375 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
adorable | +0.013 |
important | +0.013 |
best | +0.013 |
idea | +0.013 |
beautiful | +0.012 |
Neg logit contributions | |
talk | -0.012 |
oes | -0.011 |
stack | -0.011 |
ted | -0.011 |
rost | -0.010 |
Token | did |
Token ID | 750 |
Feature activation | +4.29e-02 |
Pos logit contributions | |
talk | +0.073 |
oes | +0.073 |
ted | +0.071 |
stack | +0.066 |
rost | +0.066 |
Neg logit contributions | |
adorable | -0.078 |
best | -0.076 |
important | -0.075 |
idea | -0.075 |
precious | -0.072 |
Token | not |
Token ID | 407 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
adorable | +0.064 |
best | +0.062 |
important | +0.061 |
idea | +0.061 |
precious | +0.058 |
Neg logit contributions | |
oes | -0.092 |
talk | -0.092 |
ted | -0.089 |
stack | -0.086 |
rost | -0.084 |
Token | like |
Token ID | 588 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
adorable | +0.103 |
idea | +0.101 |
best | +0.100 |
important | +0.100 |
precious | +0.095 |
Neg logit contributions | |
oes | -0.097 |
ted | -0.095 |
talk | -0.095 |
stack | -0.091 |
rost | -0.087 |
Token | still |
Token ID | 991 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
oes | +0.080 |
talk | +0.079 |
ted | +0.077 |
stack | +0.072 |
rost | +0.072 |
Neg logit contributions | |
adorable | -0.089 |
best | -0.087 |
important | -0.086 |
idea | -0.086 |
precious | -0.083 |
Token | sing |
Token ID | 1702 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
adorable | +0.041 |
best | +0.040 |
important | +0.039 |
idea | +0.039 |
precious | +0.038 |
Neg logit contributions | |
oes | -0.027 |
talk | -0.027 |
ted | -0.026 |
stack | -0.024 |
rost | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -8.06e-02 |
Pos logit contributions | |
oes | +0.057 |
talk | +0.057 |
ted | +0.056 |
rost | +0.052 |
stack | +0.052 |
Neg logit contributions | |
adorable | -0.051 |
idea | -0.049 |
best | -0.049 |
important | -0.048 |
precious | -0.047 |
Token | be |
Token ID | 307 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
oes | +0.125 |
talk | +0.124 |
ted | +0.120 |
stack | +0.113 |
rost | +0.112 |
Neg logit contributions | |
adorable | -0.167 |
best | -0.164 |
important | -0.163 |
idea | -0.161 |
precious | -0.157 |
Token | happy |
Token ID | 3772 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
ummy | +0.070 |
idea | +0.070 |
prized | +0.065 |
ement | +0.062 |
darling | +0.060 |
Neg logit contributions | |
away | -0.107 |
south | -0.107 |
talk | -0.105 |
isk | -0.103 |
them | -0.103 |
Token | ." |
Token ID | 526 |
Feature activation | +1.46e-02 |
Pos logit contributions | |
adorable | +0.314 |
idea | +0.304 |
best | +0.303 |
important | +0.300 |
precious | +0.294 |
Neg logit contributions | |
oes | -0.272 |
talk | -0.265 |
ted | -0.259 |
stack | -0.243 |
rost | -0.240 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
adorable | +0.027 |
best | +0.026 |
important | +0.026 |
idea | +0.026 |
precious | +0.025 |
Neg logit contributions | |
oes | -0.026 |
talk | -0.026 |
ted | -0.025 |
stack | -0.024 |
rost | -0.023 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
adorable | +0.230 |
important | +0.226 |
precious | +0.225 |
best | +0.220 |
idea | +0.219 |
Neg logit contributions | |
ted | -0.156 |
talk | -0.153 |
isk | -0.138 |
walked | -0.131 |
oes | -0.128 |
Token | and |
Token ID | 290 |
Feature activation | -2.79e-02 |
Pos logit contributions | |
important | +0.320 |
idea | +0.318 |
adorable | +0.315 |
precious | +0.311 |
prized | +0.309 |
Neg logit contributions | |
ted | -0.231 |
talk | -0.209 |
towards | -0.206 |
ting | -0.205 |
oes | -0.200 |
Token | hugged |
Token ID | 41130 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
ted | +0.045 |
talk | +0.042 |
oes | +0.040 |
ting | +0.038 |
stack | +0.038 |
Neg logit contributions | |
adorable | -0.055 |
idea | -0.054 |
important | -0.054 |
best | -0.054 |
precious | -0.052 |
Token | her |
Token ID | 607 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
best | +0.074 |
adorable | +0.074 |
idea | +0.071 |
important | +0.071 |
beautiful | +0.067 |
Neg logit contributions | |
oes | -0.099 |
talk | -0.097 |
stack | -0.092 |
use | -0.089 |
ricanes | -0.088 |
Token | But |
Token ID | 887 |
Feature activation | -1.47e-03 |
Pos logit contributions | |
adorable | +0.028 |
best | +0.026 |
important | +0.026 |
idea | +0.026 |
beautiful | +0.025 |
Neg logit contributions | |
talk | -0.030 |
oes | -0.030 |
ted | -0.028 |
rost | -0.028 |
use | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
talk | +0.003 |
oes | +0.003 |
ted | +0.003 |
isk | +0.003 |
iffs | +0.003 |
Neg logit contributions | |
adorable | -0.002 |
best | -0.002 |
important | -0.002 |
idea | -0.002 |
precious | -0.002 |
Token | tree |
Token ID | 5509 |
Feature activation | -9.48e-02 |
Pos logit contributions | |
adorable | +0.089 |
best | +0.087 |
important | +0.086 |
idea | +0.086 |
precious | +0.083 |
Neg logit contributions | |
oes | -0.085 |
talk | -0.084 |
ted | -0.081 |
stack | -0.078 |
rost | -0.076 |
Token | was |
Token ID | 373 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
oes | +0.187 |
talk | +0.182 |
ted | +0.180 |
stack | +0.173 |
isk | +0.169 |
Neg logit contributions | |
adorable | -0.154 |
best | -0.148 |
idea | -0.147 |
important | -0.146 |
precious | -0.142 |
Token | too |
Token ID | 1165 |
Feature activation | +9.70e-02 |
Pos logit contributions | |
adorable | +0.184 |
idea | +0.182 |
best | +0.179 |
important | +0.170 |
precious | +0.166 |
Neg logit contributions | |
oes | -0.312 |
talk | -0.298 |
ted | -0.297 |
stack | -0.292 |
rost | -0.290 |
Token | tall |
Token ID | 7331 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
best | +0.161 |
idea | +0.158 |
adorable | +0.155 |
important | +0.144 |
prized | +0.140 |
Neg logit contributions | |
oes | -0.184 |
talk | -0.169 |
iffs | -0.168 |
rost | -0.167 |
ted | -0.167 |
Token | and |
Token ID | 290 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
adorable | +0.191 |
best | +0.185 |
idea | +0.183 |
important | +0.182 |
precious | +0.177 |
Neg logit contributions | |
oes | -0.224 |
talk | -0.221 |
ted | -0.218 |
stack | -0.206 |
isk | -0.203 |
Token | Twe |
Token ID | 24205 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
talk | +0.050 |
oes | +0.050 |
ted | +0.048 |
stack | +0.046 |
rost | +0.045 |
Neg logit contributions | |
adorable | -0.045 |
best | -0.044 |
important | -0.044 |
idea | -0.042 |
precious | -0.042 |
Token | ety |
Token ID | 2963 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
adorable | +0.258 |
best | +0.251 |
important | +0.250 |
idea | +0.246 |
precious | +0.245 |
Neg logit contributions | |
talk | -0.115 |
oes | -0.113 |
ted | -0.110 |
stack | -0.108 |
rost | -0.100 |
Token | couldn |
Token ID | 3521 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
adorable | +0.052 |
best | +0.051 |
idea | +0.050 |
important | +0.050 |
precious | +0.049 |
Neg logit contributions | |
oes | -0.050 |
talk | -0.049 |
ted | -0.049 |
stack | -0.046 |
isk | -0.045 |
Token | 't |
Token ID | 470 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
adorable | +0.056 |
best | +0.055 |
idea | +0.055 |
important | +0.054 |
precious | +0.054 |
Neg logit contributions | |
oes | -0.030 |
talk | -0.029 |
ted | -0.028 |
stack | -0.027 |
isk | -0.026 |
Token | not |
Token ID | 407 |
Feature activation | +5.18e-02 |
Pos logit contributions | |
adorable | +0.062 |
best | +0.060 |
idea | +0.059 |
important | +0.059 |
precious | +0.058 |
Neg logit contributions | |
oes | -0.072 |
talk | -0.072 |
ted | -0.069 |
rost | -0.066 |
isk | -0.066 |
Token | want |
Token ID | 765 |
Feature activation | +6.34e-02 |
Pos logit contributions | |
adorable | +0.104 |
best | +0.102 |
idea | +0.101 |
important | +0.101 |
precious | +0.098 |
Neg logit contributions | |
oes | -0.084 |
talk | -0.083 |
ted | -0.081 |
stack | -0.077 |
rost | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | -1.88e-01 |
Pos logit contributions | |
important | +0.134 |
best | +0.134 |
adorable | +0.133 |
beautiful | +0.128 |
precious | +0.126 |
Neg logit contributions | |
stack | -0.104 |
talk | -0.095 |
oes | -0.093 |
isk | -0.093 |
Dog | -0.089 |
Token | share |
Token ID | 2648 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
oes | +0.284 |
talk | +0.280 |
ted | +0.276 |
stack | +0.264 |
rost | +0.253 |
Neg logit contributions | |
adorable | -0.399 |
best | -0.389 |
idea | -0.386 |
important | -0.385 |
precious | -0.372 |
Token | her |
Token ID | 607 |
Feature activation | -2.27e-03 |
Pos logit contributions | |
oes | +0.109 |
talk | +0.108 |
ted | +0.104 |
stack | +0.101 |
rost | +0.098 |
Neg logit contributions | |
adorable | -0.098 |
best | -0.096 |
important | -0.094 |
idea | -0.094 |
precious | -0.091 |
Token | bear |
Token ID | 6842 |
Feature activation | -8.79e-02 |
Pos logit contributions | |
oes | +0.005 |
talk | +0.005 |
ted | +0.005 |
rost | +0.004 |
isk | +0.004 |
Neg logit contributions | |
adorable | -0.003 |
important | -0.003 |
best | -0.003 |
idea | -0.003 |
impressed | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +8.28e-03 |
Pos logit contributions | |
talk | +0.161 |
oes | +0.159 |
ted | +0.150 |
isk | +0.150 |
rost | +0.145 |
Neg logit contributions | |
adorable | -0.153 |
best | -0.148 |
important | -0.148 |
idea | -0.147 |
precious | -0.146 |
Token | She |
Token ID | 1375 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
adorable | +0.015 |
important | +0.014 |
best | +0.014 |
idea | +0.014 |
loyal | +0.014 |
Neg logit contributions | |
talk | -0.016 |
oes | -0.015 |
ted | -0.015 |
stack | -0.014 |
use | -0.014 |
Token | thought |
Token ID | 1807 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
talk | +0.076 |
ted | +0.074 |
oes | +0.073 |
isk | +0.068 |
rost | +0.067 |
Neg logit contributions | |
adorable | -0.081 |
best | -0.079 |
idea | -0.078 |
important | -0.078 |
precious | -0.075 |
Token | he |
Token ID | 339 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
adorable | +0.284 |
best | +0.278 |
important | +0.276 |
idea | +0.273 |
precious | +0.267 |
Neg logit contributions | |
oes | -0.212 |
talk | -0.212 |
ted | -0.201 |
stack | -0.192 |
isk | -0.187 |
Token | was |
Token ID | 373 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
talk | +0.053 |
ted | +0.052 |
oes | +0.052 |
isk | +0.050 |
rost | +0.047 |
Neg logit contributions | |
adorable | -0.059 |
idea | -0.058 |
best | -0.057 |
important | -0.057 |
precious | -0.055 |
Token | pretend |
Token ID | 16614 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
oes | +0.098 |
talk | +0.096 |
ted | +0.092 |
stack | +0.087 |
rost | +0.086 |
Neg logit contributions | |
adorable | -0.130 |
best | -0.127 |
idea | -0.126 |
important | -0.125 |
precious | -0.122 |
Token | to |
Token ID | 284 |
Feature activation | -1.58e-01 |
Pos logit contributions | |
adorable | +0.224 |
important | +0.221 |
best | +0.219 |
beautiful | +0.214 |
idea | +0.213 |
Neg logit contributions | |
talk | -0.155 |
oes | -0.150 |
ted | -0.140 |
isk | -0.136 |
stack | -0.134 |
Token | eat |
Token ID | 4483 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
oes | +0.266 |
talk | +0.263 |
ted | +0.263 |
stack | +0.239 |
ting | +0.239 |
Neg logit contributions | |
adorable | -0.309 |
idea | -0.301 |
best | -0.298 |
important | -0.298 |
precious | -0.288 |
Token | their |
Token ID | 511 |
Feature activation | +3.78e-03 |
Pos logit contributions | |
oes | +0.048 |
talk | +0.048 |
ted | +0.046 |
stack | +0.044 |
rost | +0.043 |
Neg logit contributions | |
adorable | -0.047 |
best | -0.045 |
important | -0.045 |
idea | -0.045 |
precious | -0.043 |
Token | stew |
Token ID | 20798 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
adorable | +0.006 |
important | +0.006 |
idea | +0.006 |
best | +0.006 |
precious | +0.006 |
Neg logit contributions | |
oes | -0.007 |
talk | -0.007 |
ted | -0.007 |
rost | -0.006 |
gra | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -6.58e-02 |
Pos logit contributions | |
oes | +0.229 |
talk | +0.228 |
ted | +0.220 |
rost | +0.211 |
stack | +0.210 |
Neg logit contributions | |
adorable | -0.147 |
best | -0.147 |
idea | -0.143 |
important | -0.141 |
precious | -0.134 |
Token | say |
Token ID | 910 |
Feature activation | +9.30e-02 |
Pos logit contributions | |
oes | +0.114 |
talk | +0.112 |
ted | +0.108 |
isk | +0.101 |
iffs | +0.100 |
Neg logit contributions | |
adorable | -0.128 |
best | -0.125 |
idea | -0.125 |
important | -0.123 |
precious | -0.120 |
Token | , |
Token ID | 11 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
adorable | +0.133 |
best | +0.129 |
important | +0.126 |
idea | +0.126 |
precious | +0.121 |
Neg logit contributions | |
oes | -0.205 |
talk | -0.204 |
ted | -0.198 |
stack | -0.190 |
rost | -0.188 |
Token | " |
Token ID | 366 |
Feature activation | -3.61e-02 |
Pos logit contributions | |
talk | +0.049 |
oes | +0.047 |
ricane | +0.043 |
use | +0.043 |
ales | +0.043 |
Neg logit contributions | |
adorable | -0.029 |
impressed | -0.027 |
loyal | -0.027 |
pleased | -0.027 |
happy | -0.027 |
Token | M |
Token ID | 44 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
talk | +0.061 |
oes | +0.060 |
ted | +0.058 |
rost | +0.054 |
stack | +0.054 |
Neg logit contributions | |
adorable | -0.071 |
best | -0.069 |
idea | -0.069 |
important | -0.069 |
precious | -0.067 |
Token | mm |
Token ID | 3020 |
Feature activation | +1.23e-01 |
Pos logit contributions | |
adorable | +0.110 |
important | +0.100 |
best | +0.100 |
idea | +0.099 |
precious | +0.099 |
Neg logit contributions | |
oes | -0.200 |
talk | -0.193 |
rost | -0.186 |
ted | -0.186 |
isk | -0.181 |
Token | 's |
Token ID | 338 |
Feature activation | -8.59e-03 |
Pos logit contributions | |
adorable | +0.012 |
best | +0.012 |
important | +0.012 |
idea | +0.011 |
beautiful | +0.011 |
Neg logit contributions | |
oes | -0.022 |
talk | -0.022 |
ted | -0.021 |
rost | -0.021 |
stack | -0.021 |
Token | bed |
Token ID | 3996 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
oes | +0.015 |
talk | +0.014 |
ted | +0.014 |
stack | +0.013 |
rost | +0.013 |
Neg logit contributions | |
adorable | -0.016 |
important | -0.016 |
idea | -0.015 |
best | -0.015 |
precious | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +5.53e-03 |
Pos logit contributions | |
talk | +0.216 |
oes | +0.215 |
ted | +0.211 |
stack | +0.201 |
isk | +0.198 |
Neg logit contributions | |
adorable | -0.174 |
best | -0.165 |
important | -0.165 |
idea | -0.164 |
precious | -0.161 |
Token | Finally |
Token ID | 9461 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
adorable | +0.011 |
best | +0.011 |
important | +0.011 |
idea | +0.011 |
beautiful | +0.010 |
Neg logit contributions | |
talk | -0.009 |
oes | -0.009 |
ted | -0.008 |
stack | -0.008 |
rost | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -5.83e-02 |
Pos logit contributions | |
ted | +0.038 |
talk | +0.037 |
oes | +0.037 |
stack | +0.035 |
rost | +0.035 |
Neg logit contributions | |
adorable | -0.031 |
important | -0.029 |
precious | -0.029 |
idea | -0.029 |
prized | -0.028 |
Token | they |
Token ID | 484 |
Feature activation | -9.78e-02 |
Pos logit contributions | |
talk | +0.110 |
oes | +0.109 |
rost | +0.096 |
use | +0.096 |
isk | +0.094 |
Neg logit contributions | |
best | -0.115 |
adorable | -0.110 |
beautiful | -0.107 |
important | -0.106 |
idea | -0.104 |
Token | found |
Token ID | 1043 |
Feature activation | +1.84e-02 |
Pos logit contributions | |
oes | +0.190 |
talk | +0.180 |
ted | +0.173 |
stack | +0.169 |
ricanes | +0.167 |
Neg logit contributions | |
adorable | -0.171 |
best | -0.170 |
idea | -0.164 |
important | -0.162 |
beautiful | -0.162 |
Token | the |
Token ID | 262 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
adorable | +0.036 |
important | +0.035 |
idea | +0.035 |
best | +0.034 |
precious | +0.034 |
Neg logit contributions | |
ted | -0.030 |
oes | -0.029 |
talk | -0.029 |
stack | -0.027 |
ting | -0.026 |
Token | wand |
Token ID | 11569 |
Feature activation | -5.61e-02 |
Pos logit contributions | |
adorable | +0.033 |
impressed | +0.032 |
pleased | +0.032 |
thrilled | +0.032 |
important | +0.031 |
Neg logit contributions | |
oes | -0.057 |
stack | -0.057 |
south | -0.056 |
talk | -0.056 |
ted | -0.053 |
Token | behind |
Token ID | 2157 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
oes | +0.112 |
talk | +0.111 |
ted | +0.109 |
stack | +0.105 |
isk | +0.101 |
Neg logit contributions | |
adorable | -0.090 |
best | -0.086 |
important | -0.086 |
idea | -0.085 |
precious | -0.082 |
Token | the |
Token ID | 262 |
Feature activation | -4.03e-02 |
Pos logit contributions | |
oes | +0.291 |
talk | +0.286 |
ted | +0.273 |
stack | +0.265 |
rost | +0.264 |
Neg logit contributions | |
adorable | -0.213 |
best | -0.207 |
idea | -0.202 |
important | -0.202 |
precious | -0.194 |
Token | asked |
Token ID | 1965 |
Feature activation | -9.96e-03 |
Pos logit contributions | |
adorable | +0.083 |
best | +0.080 |
important | +0.079 |
idea | +0.079 |
precious | +0.077 |
Neg logit contributions | |
oes | -0.077 |
talk | -0.077 |
ted | -0.075 |
stack | -0.071 |
rost | -0.069 |
Token | many |
Token ID | 867 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
oes | +0.021 |
talk | +0.020 |
ted | +0.020 |
stack | +0.020 |
ricanes | +0.019 |
Neg logit contributions | |
best | -0.016 |
idea | -0.015 |
important | -0.015 |
adorable | -0.015 |
loyal | -0.014 |
Token | questions |
Token ID | 2683 |
Feature activation | -7.29e-02 |
Pos logit contributions | |
adorable | +0.052 |
best | +0.050 |
idea | +0.049 |
important | +0.049 |
precious | +0.048 |
Neg logit contributions | |
oes | -0.045 |
talk | -0.044 |
ted | -0.044 |
rost | -0.041 |
ting | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e-03 |
Pos logit contributions | |
talk | +0.166 |
oes | +0.164 |
ted | +0.160 |
stack | +0.153 |
isk | +0.153 |
Neg logit contributions | |
adorable | -0.099 |
best | -0.098 |
idea | -0.096 |
important | -0.095 |
precious | -0.088 |
Token | The |
Token ID | 383 |
Feature activation | +4.10e-02 |
Pos logit contributions | |
idea | +0.002 |
adorable | +0.002 |
best | +0.002 |
important | +0.002 |
loyal | +0.002 |
Neg logit contributions | |
talk | -0.002 |
use | -0.002 |
ted | -0.002 |
oes | -0.002 |
ting | -0.002 |
Token | man |
Token ID | 582 |
Feature activation | -6.07e-03 |
Pos logit contributions | |
adorable | +0.068 |
best | +0.066 |
important | +0.065 |
idea | +0.065 |
precious | +0.063 |
Neg logit contributions | |
oes | -0.081 |
talk | -0.080 |
ted | -0.078 |
stack | -0.075 |
rost | -0.074 |
Token | patiently |
Token ID | 37220 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
talk | +0.011 |
oes | +0.011 |
ted | +0.011 |
stack | +0.010 |
rost | +0.010 |
Neg logit contributions | |
adorable | -0.010 |
best | -0.010 |
important | -0.010 |
idea | -0.010 |
precious | -0.009 |
Token | answered |
Token ID | 9373 |
Feature activation | +6.85e-02 |
Pos logit contributions | |
adorable | +0.092 |
idea | +0.090 |
best | +0.090 |
important | +0.089 |
precious | +0.085 |
Neg logit contributions | |
oes | -0.118 |
talk | -0.114 |
ted | -0.110 |
stack | -0.109 |
rost | -0.107 |
Token | them |
Token ID | 606 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
adorable | +0.101 |
best | +0.101 |
important | +0.097 |
idea | +0.094 |
beautiful | +0.092 |
Neg logit contributions | |
oes | -0.153 |
talk | -0.152 |
use | -0.139 |
stack | -0.139 |
ted | -0.138 |
Token | all |
Token ID | 477 |
Feature activation | -7.30e-02 |
Pos logit contributions | |
talk | +0.135 |
oes | +0.134 |
ted | +0.127 |
stack | +0.125 |
rost | +0.124 |
Neg logit contributions | |
adorable | -0.095 |
best | -0.094 |
important | -0.092 |
idea | -0.091 |
precious | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
talk | +0.150 |
oes | +0.147 |
ted | +0.141 |
stack | +0.140 |
isk | +0.140 |
Neg logit contributions | |
idea | -0.123 |
adorable | -0.122 |
important | -0.121 |
best | -0.118 |
beautiful | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
oes | +0.091 |
talk | +0.088 |
ted | +0.088 |
ting | +0.085 |
stack | +0.084 |
Neg logit contributions | |
adorable | -0.104 |
idea | -0.103 |
best | -0.098 |
important | -0.098 |
precious | -0.097 |
Token | From |
Token ID | 3574 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
adorable | +0.020 |
best | +0.020 |
important | +0.020 |
idea | +0.019 |
precious | +0.019 |
Neg logit contributions | |
oes | -0.017 |
talk | -0.017 |
ted | -0.016 |
rost | -0.015 |
stack | -0.015 |
Token | that |
Token ID | 326 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
adorable | +0.081 |
idea | +0.081 |
best | +0.080 |
important | +0.079 |
precious | +0.077 |
Neg logit contributions | |
oes | -0.060 |
talk | -0.060 |
ted | -0.060 |
stack | -0.055 |
rost | -0.053 |
Token | day |
Token ID | 1110 |
Feature activation | -6.35e-02 |
Pos logit contributions | |
oes | +0.108 |
stack | +0.099 |
ted | +0.097 |
talk | +0.096 |
rost | +0.094 |
Neg logit contributions | |
adorable | -0.142 |
precious | -0.135 |
idea | -0.134 |
prized | -0.131 |
darling | -0.130 |
Token | on |
Token ID | 319 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
oes | +0.102 |
talk | +0.101 |
ted | +0.095 |
stack | +0.091 |
rost | +0.091 |
Neg logit contributions | |
adorable | -0.129 |
best | -0.128 |
important | -0.125 |
idea | -0.124 |
precious | -0.120 |
Token | , |
Token ID | 11 |
Feature activation | -4.33e-02 |
Pos logit contributions | |
oes | +0.323 |
ted | +0.318 |
stack | +0.314 |
talk | +0.313 |
rost | +0.299 |
Neg logit contributions | |
adorable | -0.243 |
idea | -0.229 |
important | -0.226 |
precious | -0.226 |
best | -0.219 |
Token | Lily |
Token ID | 20037 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
oes | +0.061 |
talk | +0.060 |
ted | +0.058 |
stack | +0.054 |
rost | +0.053 |
Neg logit contributions | |
adorable | -0.095 |
best | -0.093 |
important | -0.092 |
idea | -0.092 |
precious | -0.090 |
Token | and |
Token ID | 290 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
adorable | +0.161 |
idea | +0.156 |
precious | +0.152 |
important | +0.151 |
best | +0.149 |
Neg logit contributions | |
oes | -0.155 |
ted | -0.155 |
talk | -0.154 |
ting | -0.142 |
isk | -0.142 |
Token | Tim |
Token ID | 5045 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
talk | +0.067 |
oes | +0.067 |
ted | +0.062 |
iffs | +0.062 |
rost | +0.060 |
Neg logit contributions | |
adorable | -0.071 |
best | -0.070 |
idea | -0.067 |
important | -0.067 |
loyal | -0.065 |
Token | my |
Token ID | 1820 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
best | +0.089 |
adorable | +0.087 |
important | +0.083 |
idea | +0.083 |
beautiful | +0.079 |
Neg logit contributions | |
talk | -0.087 |
oes | -0.086 |
ted | -0.084 |
stack | -0.083 |
iffs | -0.081 |
Token | learned |
Token ID | 4499 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
adorable | +0.080 |
best | +0.079 |
important | +0.078 |
idea | +0.077 |
precious | +0.075 |
Neg logit contributions | |
oes | -0.063 |
talk | -0.063 |
ted | -0.060 |
stack | -0.057 |
rost | -0.056 |
Token | So |
Token ID | 1406 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
adorable | +0.199 |
important | +0.193 |
best | +0.193 |
idea | +0.191 |
precious | +0.187 |
Neg logit contributions | |
oes | -0.175 |
talk | -0.174 |
ted | -0.168 |
stack | -0.160 |
isk | -0.158 |
Token | they |
Token ID | 484 |
Feature activation | -9.14e-02 |
Pos logit contributions | |
adorable | +0.062 |
important | +0.059 |
best | +0.059 |
idea | +0.059 |
precious | +0.057 |
Neg logit contributions | |
oes | -0.064 |
talk | -0.063 |
ted | -0.061 |
stack | -0.058 |
rost | -0.057 |
Token | both |
Token ID | 1111 |
Feature activation | -2.68e-03 |
Pos logit contributions | |
oes | +0.158 |
talk | +0.149 |
ted | +0.145 |
iffs | +0.142 |
rost | +0.142 |
Neg logit contributions | |
adorable | -0.172 |
best | -0.171 |
idea | -0.170 |
important | -0.166 |
precious | -0.162 |
Token | walked |
Token ID | 6807 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
oes | +0.004 |
talk | +0.004 |
ted | +0.004 |
stack | +0.004 |
rost | +0.004 |
Neg logit contributions | |
adorable | -0.005 |
best | -0.005 |
important | -0.005 |
idea | -0.005 |
precious | -0.005 |
Token | up |
Token ID | 510 |
Feature activation | -9.55e-02 |
Pos logit contributions | |
oes | +0.125 |
isk | +0.121 |
rost | +0.121 |
talk | +0.115 |
stack | +0.113 |
Neg logit contributions | |
best | -0.201 |
adorable | -0.197 |
idea | -0.192 |
important | -0.192 |
beautiful | -0.188 |
Token | to |
Token ID | 284 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
oes | +0.145 |
talk | +0.141 |
isk | +0.133 |
rost | +0.133 |
ted | +0.131 |
Neg logit contributions | |
best | -0.201 |
adorable | -0.200 |
idea | -0.197 |
important | -0.194 |
beautiful | -0.194 |
Token | the |
Token ID | 262 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
oes | +0.453 |
talk | +0.442 |
ted | +0.405 |
stack | +0.398 |
rost | +0.396 |
Neg logit contributions | |
adorable | -0.416 |
best | -0.400 |
idea | -0.395 |
beautiful | -0.392 |
important | -0.384 |
Token | old |
Token ID | 1468 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
oes | +0.085 |
talk | +0.083 |
ted | +0.081 |
stack | +0.080 |
rost | +0.077 |
Neg logit contributions | |
adorable | -0.084 |
idea | -0.083 |
best | -0.082 |
important | -0.082 |
precious | -0.078 |
Token | lady |
Token ID | 10846 |
Feature activation | -6.76e-02 |
Pos logit contributions | |
adorable | +0.104 |
best | +0.103 |
important | +0.101 |
idea | +0.098 |
precious | +0.098 |
Neg logit contributions | |
oes | -0.091 |
talk | -0.091 |
stack | -0.088 |
ted | -0.086 |
rost | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
oes | +0.152 |
talk | +0.151 |
ted | +0.147 |
rost | +0.142 |
stack | +0.140 |
Neg logit contributions | |
adorable | -0.092 |
best | -0.088 |
idea | -0.086 |
important | -0.085 |
precious | -0.082 |
Token | and |
Token ID | 290 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
oes | +0.067 |
talk | +0.066 |
ted | +0.064 |
stack | +0.061 |
rost | +0.060 |
Neg logit contributions | |
adorable | -0.056 |
best | -0.054 |
important | -0.053 |
idea | -0.053 |
precious | -0.052 |
Token | one |
Token ID | 530 |
Feature activation | +9.11e-02 |
Pos logit contributions | |
adorable | +0.066 |
best | +0.064 |
important | +0.063 |
idea | +0.063 |
precious | +0.060 |
Neg logit contributions | |
oes | -0.078 |
talk | -0.077 |
ted | -0.075 |
stack | -0.072 |
rost | -0.071 |
Token | day |
Token ID | 1110 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
adorable | +0.151 |
best | +0.147 |
idea | +0.145 |
important | +0.145 |
precious | +0.142 |
Neg logit contributions | |
oes | -0.177 |
talk | -0.172 |
ted | -0.168 |
rost | -0.159 |
stack | -0.158 |
Token | , |
Token ID | 11 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
oes | +0.092 |
talk | +0.090 |
ted | +0.085 |
stack | +0.084 |
rost | +0.084 |
Neg logit contributions | |
adorable | -0.069 |
best | -0.068 |
important | -0.066 |
idea | -0.065 |
beautiful | -0.062 |
Token | Jane |
Token ID | 12091 |
Feature activation | +7.89e-02 |
Pos logit contributions | |
oes | +0.036 |
talk | +0.035 |
ted | +0.033 |
rost | +0.033 |
stack | +0.032 |
Neg logit contributions | |
adorable | -0.037 |
best | -0.037 |
important | -0.036 |
idea | -0.036 |
precious | -0.035 |
Token | went |
Token ID | 1816 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
adorable | +0.149 |
best | +0.145 |
important | +0.143 |
idea | +0.142 |
precious | +0.139 |
Neg logit contributions | |
oes | -0.136 |
talk | -0.136 |
ted | -0.132 |
stack | -0.123 |
isk | -0.122 |
Token | to |
Token ID | 284 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
rost | +0.086 |
oes | +0.084 |
talk | +0.082 |
ted | +0.081 |
stack | +0.077 |
Neg logit contributions | |
best | -0.124 |
important | -0.121 |
adorable | -0.120 |
idea | -0.119 |
beautiful | -0.117 |
Token | the |
Token ID | 262 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
oes | +0.414 |
talk | +0.413 |
ted | +0.406 |
rost | +0.385 |
iffs | +0.374 |
Neg logit contributions | |
adorable | -0.373 |
idea | -0.369 |
best | -0.361 |
beautiful | -0.358 |
important | -0.354 |
Token | park |
Token ID | 3952 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
oes | +0.048 |
talk | +0.048 |
ted | +0.047 |
rost | +0.044 |
stack | +0.043 |
Neg logit contributions | |
adorable | -0.057 |
best | -0.056 |
important | -0.055 |
idea | -0.054 |
precious | -0.053 |
Token | to |
Token ID | 284 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
oes | +0.223 |
talk | +0.222 |
ted | +0.219 |
rost | +0.213 |
use | +0.207 |
Neg logit contributions | |
adorable | -0.175 |
best | -0.170 |
idea | -0.168 |
important | -0.165 |
beautiful | -0.159 |
Token | practice |
Token ID | 3357 |
Feature activation | -2.76e-02 |
Pos logit contributions | |
ted | +0.354 |
oes | +0.354 |
talk | +0.350 |
rost | +0.330 |
iffs | +0.311 |
Neg logit contributions | |
idea | -0.459 |
adorable | -0.459 |
beautiful | -0.449 |
best | -0.441 |
important | -0.430 |
Token | with |
Token ID | 351 |
Feature activation | -2.66e-02 |
Pos logit contributions | |
oes | +0.050 |
talk | +0.049 |
ted | +0.048 |
stack | +0.046 |
isk | +0.045 |
Neg logit contributions | |
adorable | -0.048 |
best | -0.046 |
important | -0.045 |
precious | -0.045 |
idea | -0.045 |
Token | dog |
Token ID | 3290 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
adorable | +0.187 |
best | +0.182 |
important | +0.179 |
idea | +0.179 |
precious | +0.174 |
Neg logit contributions | |
oes | -0.187 |
talk | -0.184 |
ted | -0.179 |
stack | -0.170 |
rost | -0.169 |
Token | . |
Token ID | 13 |
Feature activation | +9.94e-03 |
Pos logit contributions | |
oes | +0.114 |
talk | +0.112 |
ricanes | +0.111 |
ted | +0.109 |
rost | +0.108 |
Neg logit contributions | |
best | -0.055 |
adorable | -0.052 |
idea | -0.050 |
important | -0.049 |
loyal | -0.047 |
Token | He |
Token ID | 679 |
Feature activation | -4.74e-02 |
Pos logit contributions | |
adorable | +0.019 |
idea | +0.018 |
best | +0.018 |
important | +0.018 |
loyal | +0.017 |
Neg logit contributions | |
talk | -0.018 |
oes | -0.018 |
use | -0.017 |
ted | -0.016 |
stack | -0.016 |
Token | goes |
Token ID | 2925 |
Feature activation | -4.38e-02 |
Pos logit contributions | |
oes | +0.080 |
talk | +0.078 |
ted | +0.074 |
stack | +0.072 |
rost | +0.071 |
Neg logit contributions | |
adorable | -0.093 |
best | -0.091 |
important | -0.090 |
idea | -0.090 |
precious | -0.088 |
Token | back |
Token ID | 736 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
gra | +0.080 |
oes | +0.078 |
talk | +0.077 |
ricanes | +0.077 |
ted | +0.076 |
Neg logit contributions | |
best | -0.078 |
loyal | -0.074 |
adorable | -0.073 |
happy | -0.071 |
important | -0.071 |
Token | to |
Token ID | 284 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
oes | +0.224 |
ricanes | +0.223 |
talk | +0.223 |
rost | +0.222 |
gra | +0.219 |
Neg logit contributions | |
best | -0.224 |
loyal | -0.221 |
adorable | -0.220 |
friends | -0.217 |
important | -0.217 |
Token | his |
Token ID | 465 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
talk | +0.537 |
oes | +0.508 |
ricanes | +0.498 |
ted | +0.494 |
ricane | +0.477 |
Neg logit contributions | |
adorable | -0.339 |
idea | -0.339 |
best | -0.335 |
loyal | -0.331 |
happy | -0.330 |
Token | ball |
Token ID | 2613 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
oes | +0.025 |
talk | +0.024 |
stack | +0.024 |
ted | +0.023 |
isk | +0.022 |
Neg logit contributions | |
adorable | -0.026 |
idea | -0.024 |
important | -0.024 |
best | -0.024 |
precious | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | -8.36e-02 |
Pos logit contributions | |
oes | +0.124 |
talk | +0.123 |
ted | +0.120 |
ricanes | +0.118 |
rost | +0.117 |
Neg logit contributions | |
best | -0.055 |
idea | -0.053 |
adorable | -0.053 |
important | -0.050 |
loyal | -0.048 |
Token | plays |
Token ID | 5341 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
oes | +0.164 |
talk | +0.161 |
rost | +0.154 |
ales | +0.144 |
isk | +0.144 |
Neg logit contributions | |
best | -0.143 |
adorable | -0.142 |
loyal | -0.141 |
precious | -0.135 |
idea | -0.134 |
Token | some |
Token ID | 617 |
Feature activation | +3.77e-03 |
Pos logit contributions | |
oes | +0.054 |
talk | +0.053 |
ted | +0.051 |
stack | +0.049 |
use | +0.048 |
Neg logit contributions | |
adorable | -0.036 |
best | -0.035 |
idea | -0.035 |
important | -0.034 |
beautiful | -0.033 |
Token | !" |
Token ID | 2474 |
Feature activation | +3.11e-03 |
Pos logit contributions | |
oes | +0.131 |
talk | +0.130 |
ted | +0.124 |
stack | +0.122 |
gra | +0.121 |
Neg logit contributions | |
adorable | -0.070 |
important | -0.068 |
best | -0.066 |
precious | -0.066 |
idea | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
adorable | +0.006 |
idea | +0.006 |
best | +0.006 |
important | +0.005 |
beautiful | +0.005 |
Neg logit contributions | |
oes | -0.006 |
talk | -0.006 |
rost | -0.005 |
isk | -0.005 |
ted | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
oes | +0.056 |
talk | +0.055 |
ted | +0.054 |
stack | +0.052 |
isk | +0.052 |
Neg logit contributions | |
adorable | -0.055 |
best | -0.054 |
important | -0.053 |
idea | -0.053 |
beautiful | -0.051 |
Token | Jim |
Token ID | 18050 |
Feature activation | +9.19e-03 |
Pos logit contributions | |
oes | +0.049 |
talk | +0.048 |
ted | +0.047 |
stack | +0.045 |
isk | +0.044 |
Neg logit contributions | |
adorable | -0.054 |
best | -0.053 |
important | -0.052 |
idea | -0.052 |
precious | -0.050 |
Token | pointed |
Token ID | 6235 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
adorable | +0.016 |
best | +0.016 |
important | +0.015 |
idea | +0.015 |
precious | +0.015 |
Neg logit contributions | |
oes | -0.017 |
talk | -0.017 |
ted | -0.017 |
stack | -0.016 |
rost | -0.015 |
Token | to |
Token ID | 284 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
adorable | +0.156 |
best | +0.151 |
idea | +0.150 |
important | +0.149 |
precious | +0.145 |
Neg logit contributions | |
oes | -0.106 |
talk | -0.105 |
ted | -0.098 |
stack | -0.097 |
isk | -0.096 |
Token | the |
Token ID | 262 |
Feature activation | +5.09e-03 |
Pos logit contributions | |
oes | +0.323 |
talk | +0.321 |
ted | +0.308 |
stack | +0.296 |
rost | +0.292 |
Neg logit contributions | |
adorable | -0.267 |
best | -0.256 |
idea | -0.253 |
important | -0.252 |
precious | -0.243 |
Token | sky |
Token ID | 6766 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
adorable | +0.010 |
best | +0.010 |
important | +0.010 |
idea | +0.010 |
precious | +0.010 |
Neg logit contributions | |
oes | -0.008 |
talk | -0.008 |
ted | -0.008 |
stack | -0.007 |
rost | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | -6.04e-02 |
Pos logit contributions | |
talk | +0.228 |
oes | +0.227 |
ted | +0.220 |
rost | +0.212 |
stack | +0.212 |
Neg logit contributions | |
adorable | -0.138 |
best | -0.134 |
idea | -0.131 |
important | -0.130 |
precious | -0.124 |
Token | said |
Token ID | 531 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
oes | +0.126 |
talk | +0.124 |
ted | +0.118 |
isk | +0.116 |
rost | +0.115 |
Neg logit contributions | |
adorable | -0.095 |
best | -0.090 |
important | -0.089 |
idea | -0.089 |
beautiful | -0.086 |
Token | , |
Token ID | 11 |
Feature activation | -4.10e-03 |
Pos logit contributions | |
adorable | +0.141 |
best | +0.135 |
important | +0.134 |
idea | +0.133 |
precious | +0.129 |
Neg logit contributions | |
oes | -0.237 |
talk | -0.235 |
ted | -0.230 |
stack | -0.221 |
rost | -0.219 |
Token | alone |
Token ID | 3436 |
Feature activation | +1.30e-03 |
Pos logit contributions | |
talk | +0.122 |
oes | +0.118 |
ted | +0.117 |
rost | +0.111 |
stack | +0.111 |
Neg logit contributions | |
adorable | -0.100 |
important | -0.099 |
best | -0.097 |
idea | -0.093 |
precious | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +8.82e-03 |
Pos logit contributions | |
adorable | +0.002 |
best | +0.002 |
important | +0.002 |
idea | +0.002 |
precious | +0.002 |
Neg logit contributions | |
oes | -0.003 |
talk | -0.003 |
ted | -0.003 |
stack | -0.002 |
rost | -0.002 |
Token | From |
Token ID | 3574 |
Feature activation | +3.03e-02 |
Pos logit contributions | |
adorable | +0.017 |
best | +0.017 |
important | +0.017 |
idea | +0.017 |
precious | +0.016 |
Neg logit contributions | |
oes | -0.015 |
talk | -0.015 |
ted | -0.014 |
stack | -0.013 |
rost | -0.013 |
Token | that |
Token ID | 326 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
adorable | +0.061 |
best | +0.060 |
important | +0.059 |
idea | +0.059 |
precious | +0.057 |
Neg logit contributions | |
oes | -0.048 |
talk | -0.048 |
ted | -0.046 |
stack | -0.044 |
rost | -0.043 |
Token | day |
Token ID | 1110 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
oes | +0.108 |
talk | +0.099 |
ted | +0.098 |
stack | +0.095 |
use | +0.094 |
Neg logit contributions | |
adorable | -0.133 |
precious | -0.127 |
idea | -0.125 |
prized | -0.123 |
darling | -0.122 |
Token | on |
Token ID | 319 |
Feature activation | -1.59e-01 |
Pos logit contributions | |
oes | +0.108 |
talk | +0.106 |
ted | +0.100 |
stack | +0.097 |
rost | +0.097 |
Neg logit contributions | |
adorable | -0.132 |
best | -0.132 |
important | -0.129 |
idea | -0.127 |
precious | -0.123 |
Token | , |
Token ID | 11 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
oes | +0.327 |
ted | +0.323 |
talk | +0.313 |
stack | +0.311 |
rost | +0.299 |
Neg logit contributions | |
adorable | -0.238 |
idea | -0.224 |
precious | -0.221 |
important | -0.220 |
best | -0.215 |
Token | the |
Token ID | 262 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
ted | +0.059 |
oes | +0.059 |
ting | +0.055 |
stack | +0.052 |
isk | +0.052 |
Neg logit contributions | |
adorable | -0.091 |
idea | -0.091 |
precious | -0.089 |
best | -0.088 |
important | -0.087 |
Token | animals |
Token ID | 4695 |
Feature activation | -7.98e-02 |
Pos logit contributions | |
adorable | +0.042 |
pleased | +0.041 |
impressed | +0.041 |
thrilled | +0.041 |
idea | +0.041 |
Neg logit contributions | |
oes | -0.028 |
ted | -0.027 |
talk | -0.027 |
stack | -0.026 |
isk | -0.026 |
Token | had |
Token ID | 550 |
Feature activation | +3.53e-03 |
Pos logit contributions | |
oes | +0.155 |
talk | +0.154 |
rost | +0.146 |
ted | +0.146 |
iffs | +0.142 |
Neg logit contributions | |
best | -0.139 |
adorable | -0.136 |
important | -0.132 |
idea | -0.132 |
beautiful | -0.125 |
Token | a |
Token ID | 257 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
adorable | +0.006 |
idea | +0.006 |
important | +0.005 |
best | +0.005 |
precious | +0.005 |
Neg logit contributions | |
talk | -0.007 |
oes | -0.007 |
ted | -0.007 |
isk | -0.006 |
stack | -0.006 |