Token |
|
Token ID | 198 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
atted | +0.071 |
lee | +0.069 |
impressed | +0.067 |
omo | +0.066 |
mad | +0.065 |
Neg logit contributions | |
Ready | -0.044 |
Left | -0.036 |
sc | -0.031 |
alis | -0.031 |
Three | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
Ready | +0.067 |
Left | +0.058 |
sc | +0.050 |
alis | +0.049 |
Three | +0.048 |
Neg logit contributions | |
atted | -0.096 |
lee | -0.094 |
impressed | -0.089 |
mad | -0.086 |
omo | -0.085 |
Token | B |
Token ID | 33 |
Feature activation | +9.92e+00 |
Pos logit contributions | |
Ready | +0.077 |
Left | +0.066 |
sc | +0.059 |
Three | +0.057 |
alis | +0.057 |
Neg logit contributions | |
atted | -0.095 |
lee | -0.093 |
impressed | -0.086 |
mad | -0.083 |
smiling | -0.083 |
Token | izzy |
Token ID | 40593 |
Feature activation | +5.75e+00 |
Pos logit contributions | |
lee | +0.435 |
atted | +0.433 |
impressed | +0.415 |
smiling | +0.387 |
omo | +0.387 |
Neg logit contributions | |
Ready | -0.363 |
Left | -0.315 |
alis | -0.286 |
sc | -0.284 |
bust | -0.265 |
Token | nodded |
Token ID | 14464 |
Feature activation | +5.74e+00 |
Pos logit contributions | |
atted | +0.188 |
impressed | +0.182 |
lee | +0.182 |
glad | +0.176 |
omo | +0.175 |
Neg logit contributions | |
Ready | -0.232 |
Left | -0.214 |
bust | -0.207 |
alis | -0.201 |
summit | -0.191 |
Token | solemn |
Token ID | 26322 |
Feature activation | +1.38e+01 |
Pos logit contributions | |
atted | +0.224 |
omo | +0.218 |
lee | +0.205 |
inches | +0.198 |
impressed | +0.192 |
Neg logit contributions | |
Left | -0.180 |
Ready | -0.174 |
begin | -0.166 |
form | -0.163 |
across | -0.163 |
Token | ly |
Token ID | 306 |
Feature activation | +8.13e+00 |
Pos logit contributions | |
omo | +0.649 |
impressed | +0.648 |
atted | +0.632 |
lee | +0.612 |
glad | +0.602 |
Neg logit contributions | |
Left | -0.328 |
Ready | -0.322 |
sc | -0.290 |
bust | -0.284 |
alis | -0.264 |
Token | . |
Token ID | 13 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
omo | +0.341 |
atted | +0.297 |
lee | +0.293 |
impressed | +0.281 |
glad | +0.273 |
Neg logit contributions | |
Ready | -0.256 |
sc | -0.246 |
Left | -0.246 |
tted | -0.225 |
fire | -0.223 |
Token | She |
Token ID | 1375 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
atted | +0.014 |
lee | +0.014 |
impressed | +0.014 |
mad | +0.013 |
omo | +0.013 |
Neg logit contributions | |
Ready | -0.014 |
Left | -0.012 |
alis | -0.011 |
sc | -0.011 |
Three | -0.010 |
Token | knew |
Token ID | 2993 |
Feature activation | +8.12e+00 |
Pos logit contributions | |
Ready | +0.045 |
Left | +0.039 |
sc | +0.038 |
tted | +0.037 |
bust | +0.037 |
Neg logit contributions | |
atted | -0.042 |
lee | -0.041 |
mad | -0.037 |
ummies | -0.037 |
impressed | -0.037 |
Token | she |
Token ID | 673 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
atted | +0.296 |
lee | +0.292 |
impressed | +0.272 |
omo | +0.271 |
mad | +0.256 |
Neg logit contributions | |
Ready | -0.338 |
Left | -0.298 |
alis | -0.276 |
summit | -0.267 |
south | -0.263 |
Token | smile |
Token ID | 8212 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
lee | +0.072 |
atted | +0.070 |
impressed | +0.062 |
omo | +0.059 |
anted | +0.055 |
Neg logit contributions | |
Ready | -0.087 |
flap | -0.076 |
Left | -0.075 |
shortcut | -0.074 |
alis | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
atted | +0.158 |
lee | +0.156 |
impressed | +0.149 |
omo | +0.144 |
glad | +0.139 |
Neg logit contributions | |
Ready | -0.153 |
Left | -0.135 |
alis | -0.132 |
sc | -0.128 |
tr | -0.124 |
Token |
|
Token ID | 198 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
atted | +0.011 |
lee | +0.011 |
impressed | +0.010 |
mad | +0.010 |
ies | +0.009 |
Neg logit contributions | |
Ready | -0.007 |
Left | -0.006 |
Ready | -0.005 |
Three | -0.004 |
sc | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
Ready | +0.085 |
Left | +0.072 |
sc | +0.056 |
Three | +0.056 |
Ready | +0.055 |
Neg logit contributions | |
atted | -0.148 |
lee | -0.144 |
impressed | -0.137 |
mad | -0.133 |
smiling | -0.133 |
Token | Mary |
Token ID | 24119 |
Feature activation | +7.02e+00 |
Pos logit contributions | |
Ready | +0.116 |
Left | +0.101 |
Three | +0.088 |
sc | +0.087 |
Skip | +0.087 |
Neg logit contributions | |
atted | -0.109 |
lee | -0.107 |
impressed | -0.093 |
smiling | -0.092 |
mad | -0.089 |
Token | anx |
Token ID | 7296 |
Feature activation | +1.36e+01 |
Pos logit contributions | |
lee | +0.234 |
atted | +0.231 |
impressed | +0.227 |
omo | +0.225 |
glad | +0.224 |
Neg logit contributions | |
Ready | -0.279 |
Left | -0.257 |
bust | -0.246 |
alis | -0.241 |
summit | -0.230 |
Token | iously |
Token ID | 6819 |
Feature activation | +8.23e+00 |
Pos logit contributions | |
impressed | +0.528 |
glad | +0.487 |
pleased | +0.481 |
omo | +0.435 |
lee | +0.432 |
Neg logit contributions | |
Ready | -0.372 |
peak | -0.367 |
alis | -0.364 |
Left | -0.358 |
base | -0.349 |
Token | opened |
Token ID | 4721 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
omo | +0.258 |
lee | +0.243 |
atted | +0.238 |
impressed | +0.221 |
glad | +0.208 |
Neg logit contributions | |
Ready | -0.347 |
bust | -0.309 |
Left | -0.306 |
belong | -0.300 |
sc | -0.296 |
Token | up |
Token ID | 510 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
atted | +0.116 |
lee | +0.113 |
omo | +0.111 |
impressed | +0.111 |
glad | +0.101 |
Neg logit contributions | |
Ready | -0.109 |
Left | -0.097 |
south | -0.089 |
summit | -0.088 |
alis | -0.087 |
Token | the |
Token ID | 262 |
Feature activation | -5.26e+00 |
Pos logit contributions | |
Ready | +0.007 |
Left | +0.006 |
sc | +0.006 |
bust | +0.006 |
alis | +0.006 |
Neg logit contributions | |
lee | -0.007 |
atted | -0.007 |
omo | -0.007 |
impressed | -0.006 |
glad | -0.006 |
Token | pass |
Token ID | 1208 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
Ready | +0.197 |
Left | +0.184 |
Whe | +0.171 |
Skip | +0.163 |
alis | +0.162 |
Neg logit contributions | |
atted | -0.212 |
lee | -0.204 |
smiling | -0.203 |
mad | -0.201 |
impressed | -0.196 |
Token | ! |
Token ID | 0 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
Ready | +0.265 |
Left | +0.237 |
alis | +0.220 |
sc | +0.215 |
bust | +0.208 |
Neg logit contributions | |
lee | -0.209 |
atted | -0.208 |
impressed | -0.191 |
omo | -0.180 |
smiling | -0.176 |
Token | â |
Token ID | 22940 |
Feature activation | +8.93e+00 |
Pos logit contributions | |
lee | +0.014 |
atted | +0.014 |
impressed | +0.013 |
omo | +0.012 |
smiling | +0.012 |
Neg logit contributions | |
Ready | -0.012 |
Left | -0.010 |
alis | -0.009 |
sc | -0.009 |
bust | -0.009 |
Token | € |
Token ID | 26391 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
lee | +0.455 |
atted | +0.454 |
impressed | +0.419 |
omo | +0.414 |
mad | +0.410 |
Neg logit contributions | |
Ready | -0.253 |
Left | -0.213 |
alis | -0.186 |
sc | -0.182 |
summit | -0.180 |
Token | Jim |
Token ID | 5395 |
Feature activation | +5.81e+00 |
Pos logit contributions | |
atted | +0.176 |
lee | +0.175 |
impressed | +0.164 |
omo | +0.154 |
mad | +0.153 |
Neg logit contributions | |
Ready | -0.146 |
Left | -0.127 |
alis | -0.113 |
sc | -0.110 |
bust | -0.107 |
Token | said |
Token ID | 531 |
Feature activation | +6.08e+00 |
Pos logit contributions | |
lee | +0.194 |
atted | +0.190 |
impressed | +0.180 |
omo | +0.164 |
glad | +0.162 |
Neg logit contributions | |
Ready | -0.265 |
Left | -0.237 |
sc | -0.221 |
alis | -0.221 |
bust | -0.216 |
Token | anx |
Token ID | 7296 |
Feature activation | +1.35e+01 |
Pos logit contributions | |
lee | +0.181 |
atted | +0.179 |
impressed | +0.162 |
omo | +0.153 |
mad | +0.143 |
Neg logit contributions | |
Ready | -0.309 |
Left | -0.281 |
alis | -0.258 |
sc | -0.258 |
bust | -0.251 |
Token | iously |
Token ID | 6819 |
Feature activation | +8.12e+00 |
Pos logit contributions | |
impressed | +0.424 |
glad | +0.387 |
atted | +0.375 |
lee | +0.374 |
pleased | +0.360 |
Neg logit contributions | |
Ready | -0.542 |
Left | -0.515 |
alis | -0.499 |
tr | -0.480 |
summit | -0.468 |
Token | . |
Token ID | 13 |
Feature activation | +7.57e-01 |
Pos logit contributions | |
atted | +0.245 |
lee | +0.240 |
impressed | +0.220 |
omo | +0.209 |
mad | +0.194 |
Neg logit contributions | |
Ready | -0.392 |
Left | -0.353 |
sc | -0.333 |
alis | -0.322 |
summit | -0.319 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
lee | +0.031 |
atted | +0.031 |
impressed | +0.029 |
omo | +0.027 |
smiling | +0.027 |
Neg logit contributions | |
Ready | -0.030 |
Left | -0.027 |
alis | -0.024 |
sc | -0.024 |
bust | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
Ready | +0.077 |
Left | +0.067 |
alis | +0.061 |
sc | +0.060 |
bust | +0.057 |
Neg logit contributions | |
lee | -0.088 |
atted | -0.088 |
impressed | -0.081 |
omo | -0.078 |
smiling | -0.077 |
Token | His |
Token ID | 6653 |
Feature activation | +7.17e+00 |
Pos logit contributions | |
Ready | +0.066 |
Left | +0.058 |
alis | +0.051 |
sc | +0.051 |
Three | +0.049 |
Neg logit contributions | |
lee | -0.069 |
atted | -0.069 |
impressed | -0.063 |
omo | -0.060 |
smiling | -0.060 |
Token | to |
Token ID | 284 |
Feature activation | -4.06e+00 |
Pos logit contributions | |
Ready | +0.162 |
Left | +0.144 |
alis | +0.133 |
sc | +0.131 |
bust | +0.126 |
Neg logit contributions | |
lee | -0.141 |
atted | -0.140 |
impressed | -0.129 |
omo | -0.123 |
smiling | -0.120 |
Token | break |
Token ID | 2270 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Ready | +0.188 |
Left | +0.169 |
sc | +0.156 |
alis | +0.156 |
bust | +0.149 |
Neg logit contributions | |
lee | -0.143 |
atted | -0.139 |
impressed | -0.130 |
smiling | -0.123 |
glad | -0.122 |
Token | ." |
Token ID | 526 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
atted | +0.006 |
lee | +0.006 |
impressed | +0.005 |
omo | +0.005 |
glad | +0.005 |
Neg logit contributions | |
Ready | -0.005 |
Left | -0.005 |
sc | -0.004 |
alis | -0.004 |
south | -0.004 |
Token | Molly |
Token ID | 30236 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
atted | +0.063 |
lee | +0.061 |
impressed | +0.060 |
omo | +0.057 |
mad | +0.056 |
Neg logit contributions | |
Ready | -0.060 |
Left | -0.053 |
alis | -0.047 |
sc | -0.047 |
Three | -0.045 |
Token | nodded |
Token ID | 14464 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
lee | +0.224 |
atted | +0.223 |
impressed | +0.203 |
omo | +0.197 |
glad | +0.197 |
Neg logit contributions | |
Ready | -0.297 |
Left | -0.280 |
bust | -0.264 |
alis | -0.256 |
will | -0.251 |
Token | solemn |
Token ID | 26322 |
Feature activation | +1.32e+01 |
Pos logit contributions | |
omo | +0.191 |
atted | +0.191 |
inches | +0.173 |
lee | +0.170 |
ummies | +0.164 |
Neg logit contributions | |
Left | -0.169 |
begin | -0.162 |
end | -0.157 |
make | -0.157 |
Ready | -0.156 |
Token | ly |
Token ID | 306 |
Feature activation | +7.43e+00 |
Pos logit contributions | |
omo | +0.627 |
impressed | +0.611 |
atted | +0.598 |
lee | +0.583 |
irling | +0.567 |
Neg logit contributions | |
Left | -0.311 |
Ready | -0.301 |
bust | -0.279 |
sc | -0.268 |
south | -0.248 |
Token | , |
Token ID | 11 |
Feature activation | +9.52e+00 |
Pos logit contributions | |
omo | +0.260 |
lee | +0.218 |
atted | +0.214 |
impressed | +0.208 |
irling | +0.191 |
Neg logit contributions | |
Ready | -0.284 |
sc | -0.273 |
Left | -0.270 |
bust | -0.262 |
across | -0.257 |
Token | understanding |
Token ID | 4547 |
Feature activation | +5.27e+00 |
Pos logit contributions | |
lee | +0.293 |
atted | +0.291 |
omo | +0.267 |
ummies | +0.254 |
ins | +0.236 |
Neg logit contributions | |
Ready | -0.414 |
Left | -0.372 |
alis | -0.370 |
distance | -0.363 |
bust | -0.359 |
Token | the |
Token ID | 262 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
atted | +0.179 |
lee | +0.178 |
omo | +0.165 |
impressed | +0.162 |
mad | +0.146 |
Neg logit contributions | |
Ready | -0.230 |
Left | -0.209 |
sc | -0.189 |
alis | -0.188 |
south | -0.187 |
Token | lesson |
Token ID | 11483 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
Ready | +0.178 |
Left | +0.162 |
alis | +0.153 |
sc | +0.151 |
bust | +0.147 |
Neg logit contributions | |
lee | -0.098 |
atted | -0.098 |
impressed | -0.088 |
omo | -0.086 |
smiling | -0.083 |
Token | 's |
Token ID | 338 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
Ready | +0.284 |
Left | +0.242 |
alis | +0.218 |
sc | +0.211 |
bust | +0.201 |
Neg logit contributions | |
lee | -0.414 |
atted | -0.412 |
impressed | -0.388 |
omo | -0.372 |
smiling | -0.367 |
Token | that |
Token ID | 326 |
Feature activation | -4.65e+00 |
Pos logit contributions | |
Ready | +0.063 |
Left | +0.056 |
sc | +0.053 |
alis | +0.052 |
Three | +0.051 |
Neg logit contributions | |
atted | -0.043 |
lee | -0.043 |
impressed | -0.041 |
mad | -0.039 |
glad | -0.038 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
Ready | +0.211 |
Left | +0.189 |
sc | +0.177 |
alis | +0.177 |
Three | +0.168 |
Neg logit contributions | |
lee | -0.155 |
atted | -0.154 |
impressed | -0.153 |
mad | -0.146 |
smiling | -0.146 |
Token |
|
Token ID | 198 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
atted | +0.064 |
lee | +0.064 |
impressed | +0.059 |
omo | +0.056 |
mad | +0.055 |
Neg logit contributions | |
Ready | -0.069 |
Left | -0.061 |
alis | -0.056 |
sc | -0.055 |
bust | -0.053 |
Token |
|
Token ID | 198 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
Ready | +0.076 |
Left | +0.066 |
alis | +0.062 |
sc | +0.060 |
bust | +0.058 |
Neg logit contributions | |
lee | -0.077 |
atted | -0.076 |
impressed | -0.071 |
omo | -0.068 |
smiling | -0.066 |
Token | Jun |
Token ID | 22396 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
Ready | +0.075 |
Left | +0.065 |
alis | +0.060 |
sc | +0.058 |
bust | +0.056 |
Neg logit contributions | |
lee | -0.080 |
atted | -0.079 |
impressed | -0.074 |
omo | -0.071 |
smiling | -0.069 |
Token | o |
Token ID | 78 |
Feature activation | +7.46e+00 |
Pos logit contributions | |
impressed | +0.469 |
atted | +0.458 |
lee | +0.448 |
glad | +0.429 |
mad | +0.421 |
Neg logit contributions | |
Ready | -0.566 |
Left | -0.504 |
alis | -0.468 |
sc | -0.461 |
Skip | -0.448 |
Token | replied |
Token ID | 8712 |
Feature activation | +6.27e+00 |
Pos logit contributions | |
atted | +0.232 |
glad | +0.218 |
lee | +0.215 |
impressed | +0.211 |
omo | +0.206 |
Neg logit contributions | |
Ready | -0.311 |
Left | -0.298 |
bust | -0.284 |
alis | -0.270 |
Museum | -0.266 |
Token | , |
Token ID | 11 |
Feature activation | +1.02e+01 |
Pos logit contributions | |
lee | +0.195 |
omo | +0.191 |
atted | +0.185 |
impressed | +0.168 |
glad | +0.152 |
Neg logit contributions | |
Ready | -0.280 |
Left | -0.276 |
br | -0.244 |
sc | -0.241 |
fire | -0.237 |
Token | " |
Token ID | 366 |
Feature activation | -8.74e-01 |
Pos logit contributions | |
atted | +0.480 |
lee | +0.470 |
omo | +0.450 |
impressed | +0.420 |
mad | +0.417 |
Neg logit contributions | |
Ready | -0.324 |
Left | -0.297 |
alis | -0.228 |
Museum | -0.224 |
Skip | -0.223 |
Token | I |
Token ID | 40 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
Ready | +0.027 |
Left | +0.023 |
alis | +0.021 |
sc | +0.020 |
bust | +0.019 |
Neg logit contributions | |
lee | -0.043 |
atted | -0.042 |
impressed | -0.041 |
omo | -0.039 |
smiling | -0.038 |
Token | are |
Token ID | 389 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Ready | +0.337 |
Left | +0.294 |
alis | +0.267 |
sc | +0.260 |
bust | +0.249 |
Neg logit contributions | |
lee | -0.396 |
atted | -0.395 |
impressed | -0.368 |
omo | -0.352 |
smiling | -0.346 |
Token | you |
Token ID | 345 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
Ready | +0.060 |
Left | +0.053 |
alis | +0.049 |
sc | +0.048 |
bust | +0.046 |
Neg logit contributions | |
lee | -0.058 |
atted | -0.058 |
impressed | -0.054 |
omo | -0.051 |
smiling | -0.050 |
Token | ?" |
Token ID | 1701 |
Feature activation | +9.16e-01 |
Pos logit contributions | |
Ready | +0.101 |
Left | +0.091 |
alis | +0.088 |
hunt | +0.086 |
belong | +0.086 |
Neg logit contributions | |
atted | -0.092 |
lee | -0.090 |
omo | -0.083 |
impressed | -0.076 |
ushed | -0.069 |
Token | she |
Token ID | 673 |
Feature activation | +1.90e-01 |
Pos logit contributions | |
atted | +0.036 |
lee | +0.034 |
omo | +0.034 |
impressed | +0.032 |
rou | +0.030 |
Neg logit contributions | |
Ready | -0.037 |
bust | -0.031 |
Left | -0.031 |
gra | -0.029 |
hunt | -0.029 |
Token | shouted |
Token ID | 17293 |
Feature activation | +7.83e+00 |
Pos logit contributions | |
atted | +0.008 |
omo | +0.007 |
lee | +0.007 |
glad | +0.006 |
rou | +0.006 |
Neg logit contributions | |
Ready | -0.007 |
bust | -0.006 |
Left | -0.006 |
flap | -0.006 |
hunt | -0.006 |
Token | anx |
Token ID | 7296 |
Feature activation | +1.31e+01 |
Pos logit contributions | |
omo | +0.368 |
atted | +0.349 |
lee | +0.333 |
impressed | +0.297 |
ins | +0.288 |
Neg logit contributions | |
Ready | -0.235 |
Left | -0.223 |
across | -0.202 |
sc | -0.201 |
fire | -0.192 |
Token | iously |
Token ID | 6819 |
Feature activation | +7.87e+00 |
Pos logit contributions | |
impressed | +0.457 |
glad | +0.438 |
omo | +0.435 |
pleased | +0.431 |
atted | +0.419 |
Neg logit contributions | |
hunt | -0.417 |
edge | -0.405 |
alis | -0.398 |
base | -0.396 |
fire | -0.396 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
omo | +0.391 |
atted | +0.368 |
lee | +0.357 |
ushed | +0.314 |
impressed | +0.305 |
Neg logit contributions | |
Ready | -0.183 |
pass | -0.168 |
rest | -0.157 |
hunt | -0.156 |
source | -0.155 |
Token | |
Token ID | 220 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
Ready | +0.000 |
Left | +0.000 |
alis | +0.000 |
Three | +0.000 |
bust | +0.000 |
Neg logit contributions | |
atted | -0.000 |
lee | -0.000 |
impressed | -0.000 |
omo | -0.000 |
mad | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
Ready | +0.063 |
Left | +0.055 |
alis | +0.049 |
sc | +0.048 |
ready | +0.046 |
Neg logit contributions | |
lee | -0.075 |
atted | -0.075 |
impressed | -0.070 |
omo | -0.067 |
smiling | -0.066 |
Token |
|
Token ID | 198 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
Ready | +0.107 |
Left | +0.093 |
sc | +0.080 |
alis | +0.079 |
Three | +0.077 |
Neg logit contributions | |
atted | -0.141 |
lee | -0.140 |
impressed | -0.129 |
omo | -0.126 |
mad | -0.122 |
Token | the |
Token ID | 262 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
lee | +0.099 |
atted | +0.099 |
impressed | +0.091 |
omo | +0.089 |
mad | +0.081 |
Neg logit contributions | |
Ready | -0.124 |
Left | -0.111 |
alis | -0.102 |
sc | -0.101 |
south | -0.099 |
Token | answer |
Token ID | 3280 |
Feature activation | +3.06e+00 |
Pos logit contributions | |
Ready | +0.150 |
Left | +0.137 |
alis | +0.130 |
bust | +0.128 |
sc | +0.128 |
Neg logit contributions | |
smiling | -0.148 |
mad | -0.147 |
atted | -0.142 |
lee | -0.141 |
omo | -0.141 |
Token | . |
Token ID | 13 |
Feature activation | -3.69e-01 |
Pos logit contributions | |
lee | +0.099 |
atted | +0.098 |
impressed | +0.090 |
omo | +0.085 |
glad | +0.081 |
Neg logit contributions | |
Ready | -0.146 |
Left | -0.131 |
alis | -0.122 |
sc | -0.121 |
bust | -0.117 |
Token | She |
Token ID | 1375 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
Ready | +0.015 |
Left | +0.013 |
alis | +0.012 |
sc | +0.012 |
bust | +0.011 |
Neg logit contributions | |
atted | -0.015 |
lee | -0.015 |
impressed | -0.014 |
omo | -0.013 |
smiling | -0.013 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
Ready | +0.056 |
Left | +0.050 |
bust | +0.047 |
alis | +0.047 |
sc | +0.045 |
Neg logit contributions | |
atted | -0.040 |
lee | -0.039 |
impressed | -0.036 |
omo | -0.035 |
glad | -0.035 |
Token | anx |
Token ID | 7296 |
Feature activation | +1.29e+01 |
Pos logit contributions | |
lee | +0.132 |
atted | +0.130 |
impressed | +0.124 |
omo | +0.120 |
mad | +0.109 |
Neg logit contributions | |
Ready | -0.179 |
Left | -0.162 |
sc | -0.145 |
alis | -0.145 |
bust | -0.141 |
Token | iously |
Token ID | 6819 |
Feature activation | +7.87e+00 |
Pos logit contributions | |
impressed | +0.431 |
glad | +0.410 |
omo | +0.406 |
pleased | +0.391 |
atted | +0.386 |
Neg logit contributions | |
Ready | -0.436 |
Left | -0.415 |
alis | -0.409 |
hunt | -0.403 |
edge | -0.396 |
Token | , |
Token ID | 11 |
Feature activation | +9.37e+00 |
Pos logit contributions | |
lee | +0.250 |
atted | +0.249 |
omo | +0.244 |
impressed | +0.215 |
mad | +0.195 |
Neg logit contributions | |
Ready | -0.347 |
Left | -0.304 |
south | -0.296 |
sc | -0.294 |
Ready | -0.284 |
Token | " |
Token ID | 366 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
lee | +0.370 |
atted | +0.370 |
omo | +0.338 |
impressed | +0.327 |
mad | +0.307 |
Neg logit contributions | |
Ready | -0.394 |
Left | -0.347 |
sc | -0.309 |
alis | -0.309 |
south | -0.301 |
Token | What |
Token ID | 2061 |
Feature activation | -8.25e+00 |
Pos logit contributions | |
Ready | +0.044 |
Left | +0.039 |
alis | +0.038 |
sc | +0.036 |
bust | +0.034 |
Neg logit contributions | |
lee | -0.067 |
atted | -0.067 |
impressed | -0.065 |
omo | -0.063 |
understanding | -0.063 |
Token | is |
Token ID | 318 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
Ready | +0.271 |
Left | +0.230 |
alis | +0.209 |
sc | +0.200 |
bust | +0.190 |
Neg logit contributions | |
lee | -0.396 |
atted | -0.395 |
impressed | -0.375 |
omo | -0.358 |
smiling | -0.354 |
Token | underwater |
Token ID | 21258 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
Ready | +0.023 |
Left | +0.019 |
alis | +0.018 |
sc | +0.017 |
Three | +0.017 |
Neg logit contributions | |
lee | -0.028 |
atted | -0.028 |
impressed | -0.026 |
omo | -0.025 |
smiling | -0.025 |
Token | kingdom |
Token ID | 13239 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
lee | +0.072 |
atted | +0.070 |
impressed | +0.067 |
omo | +0.065 |
smiling | +0.064 |
Neg logit contributions | |
Ready | -0.056 |
Left | -0.047 |
alis | -0.042 |
Whe | -0.040 |
ready | -0.040 |
Token | ?" |
Token ID | 1701 |
Feature activation | +7.95e-01 |
Pos logit contributions | |
Ready | +0.222 |
Left | +0.198 |
alis | +0.184 |
Three | +0.177 |
sc | +0.176 |
Neg logit contributions | |
lee | -0.143 |
atted | -0.140 |
impressed | -0.135 |
smiling | -0.124 |
omo | -0.122 |
Token |
|
Token ID | 198 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
atted | +0.033 |
lee | +0.033 |
impressed | +0.031 |
omo | +0.029 |
mad | +0.028 |
Neg logit contributions | |
Ready | -0.032 |
Left | -0.028 |
alis | -0.025 |
sc | -0.025 |
bust | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
Ready | +0.097 |
Left | +0.085 |
alis | +0.078 |
sc | +0.076 |
bust | +0.073 |
Neg logit contributions | |
lee | -0.105 |
atted | -0.104 |
impressed | -0.097 |
omo | -0.093 |
smiling | -0.091 |
Token | Gab |
Token ID | 46079 |
Feature activation | +1.28e+01 |
Pos logit contributions | |
Ready | +0.107 |
Left | +0.095 |
alis | +0.087 |
sc | +0.086 |
bust | +0.083 |
Neg logit contributions | |
atted | -0.093 |
lee | -0.093 |
impressed | -0.084 |
omo | -0.081 |
smiling | -0.079 |
Token | ri |
Token ID | 380 |
Feature activation | +6.67e+00 |
Pos logit contributions | |
atted | +0.474 |
lee | +0.472 |
impressed | +0.438 |
omo | +0.417 |
smiling | +0.410 |
Neg logit contributions | |
Ready | -0.555 |
Left | -0.492 |
alis | -0.455 |
sc | -0.450 |
bust | -0.435 |
Token | ella |
Token ID | 12627 |
Feature activation | +5.07e+00 |
Pos logit contributions | |
lee | +0.289 |
atted | +0.289 |
impressed | +0.276 |
omo | +0.266 |
mad | +0.264 |
Neg logit contributions | |
Ready | -0.232 |
Left | -0.207 |
sc | -0.186 |
alis | -0.184 |
bust | -0.178 |
Token | was |
Token ID | 373 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
atted | +0.164 |
lee | +0.163 |
impressed | +0.155 |
omo | +0.154 |
glad | +0.148 |
Neg logit contributions | |
Ready | -0.226 |
Left | -0.207 |
bust | -0.194 |
sc | -0.189 |
alis | -0.189 |
Token | so |
Token ID | 523 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
atted | +0.137 |
lee | +0.132 |
omo | +0.120 |
ies | +0.108 |
ummies | +0.104 |
Neg logit contributions | |
Ready | -0.177 |
Left | -0.166 |
alis | -0.155 |
bust | -0.150 |
Skip | -0.148 |
Token | excited |
Token ID | 6568 |
Feature activation | +8.36e+00 |
Pos logit contributions | |
atted | +0.104 |
lee | +0.101 |
omo | +0.088 |
ries | +0.075 |
ittens | +0.072 |
Neg logit contributions | |
Ready | -0.208 |
Left | -0.193 |
bust | -0.188 |
am | -0.181 |
Skip | -0.178 |
Token | to |
Token ID | 284 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
atted | +0.088 |
lee | +0.088 |
impressed | +0.082 |
omo | +0.079 |
smiling | +0.073 |
Neg logit contributions | |
Ready | -0.112 |
Left | -0.101 |
alis | -0.092 |
sc | -0.091 |
bust | -0.090 |
Token | stop |
Token ID | 2245 |
Feature activation | +4.32e+00 |
Pos logit contributions | |
Ready | +0.067 |
Left | +0.057 |
alis | +0.051 |
sc | +0.050 |
bust | +0.047 |
Neg logit contributions | |
lee | -0.099 |
atted | -0.098 |
impressed | -0.092 |
omo | -0.089 |
smiling | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
atted | +0.151 |
omo | +0.146 |
lee | +0.146 |
impressed | +0.138 |
ushed | +0.120 |
Neg logit contributions | |
Ready | -0.180 |
Left | -0.159 |
south | -0.151 |
alis | -0.144 |
sc | -0.142 |
Token |
|
Token ID | 198 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
atted | +0.003 |
lee | +0.003 |
impressed | +0.002 |
omo | +0.002 |
mad | +0.002 |
Neg logit contributions | |
Ready | -0.002 |
Left | -0.002 |
sc | -0.002 |
alis | -0.002 |
Three | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
Ready | +0.090 |
Left | +0.078 |
sc | +0.065 |
Three | +0.064 |
tted | +0.061 |
Neg logit contributions | |
atted | -0.140 |
lee | -0.138 |
impressed | -0.124 |
mad | -0.122 |
smiling | -0.121 |
Token | Emb |
Token ID | 31567 |
Feature activation | +1.27e+01 |
Pos logit contributions | |
Ready | +0.112 |
Left | +0.100 |
sc | +0.088 |
Three | +0.088 |
Skip | +0.084 |
Neg logit contributions | |
atted | -0.094 |
lee | -0.094 |
impressed | -0.079 |
smiling | -0.078 |
omo | -0.076 |
Token | arrass |
Token ID | 9187 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
lee | +0.617 |
atted | +0.610 |
omo | +0.599 |
impressed | +0.591 |
smiling | +0.586 |
Neg logit contributions | |
Ready | -0.398 |
Left | -0.315 |
sc | -0.291 |
alis | -0.276 |
ready | -0.273 |
Token | ed |
Token ID | 276 |
Feature activation | +7.50e+00 |
Pos logit contributions | |
lee | +0.093 |
atted | +0.091 |
impressed | +0.085 |
omo | +0.081 |
smiling | +0.078 |
Neg logit contributions | |
Ready | -0.122 |
Left | -0.109 |
sc | -0.100 |
alis | -0.100 |
bust | -0.097 |
Token | , |
Token ID | 11 |
Feature activation | +9.27e+00 |
Pos logit contributions | |
lee | +0.193 |
atted | +0.182 |
impressed | +0.170 |
omo | +0.167 |
glad | +0.142 |
Neg logit contributions | |
Ready | -0.401 |
Left | -0.368 |
sc | -0.350 |
bust | -0.339 |
alis | -0.337 |
Token | Jimmy |
Token ID | 12963 |
Feature activation | +8.19e+00 |
Pos logit contributions | |
atted | +0.308 |
lee | +0.305 |
impressed | +0.261 |
omo | +0.255 |
mad | +0.236 |
Neg logit contributions | |
Ready | -0.447 |
Left | -0.401 |
alis | -0.363 |
bust | -0.363 |
sc | -0.361 |
Token | stopped |
Token ID | 5025 |
Feature activation | +4.88e+00 |
Pos logit contributions | |
atted | +0.292 |
lee | +0.289 |
impressed | +0.275 |
omo | +0.273 |
glad | +0.257 |
Neg logit contributions | |
Ready | -0.341 |
Left | -0.309 |
bust | -0.286 |
alis | -0.285 |
sc | -0.284 |
Token | â |
Token ID | 22940 |
Feature activation | +8.71e+00 |
Pos logit contributions | |
lee | +0.013 |
atted | +0.012 |
impressed | +0.011 |
omo | +0.011 |
glad | +0.011 |
Neg logit contributions | |
Ready | -0.013 |
Left | -0.012 |
alis | -0.011 |
sc | -0.010 |
summit | -0.010 |
Token | € |
Token ID | 26391 |
Feature activation | +4.23e+00 |
Pos logit contributions | |
lee | +0.409 |
atted | +0.397 |
impressed | +0.364 |
mad | +0.356 |
omo | +0.355 |
Neg logit contributions | |
Ready | -0.255 |
Left | -0.220 |
summit | -0.215 |
Find | -0.205 |
south | -0.197 |
Token |
|
Token ID | 198 |
Feature activation | -2.90e+00 |
Pos logit contributions | |
lee | +0.180 |
atted | +0.178 |
impressed | +0.166 |
omo | +0.159 |
smiling | +0.156 |
Neg logit contributions | |
Ready | -0.162 |
Left | -0.142 |
alis | -0.130 |
sc | -0.127 |
bust | -0.122 |
Token |
|
Token ID | 198 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
Ready | +0.114 |
Left | +0.100 |
alis | +0.093 |
sc | +0.089 |
bust | +0.087 |
Neg logit contributions | |
lee | -0.121 |
atted | -0.119 |
impressed | -0.112 |
omo | -0.107 |
smiling | -0.104 |
Token | So |
Token ID | 2396 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
Ready | +0.096 |
Left | +0.084 |
alis | +0.079 |
sc | +0.075 |
bust | +0.074 |
Neg logit contributions | |
lee | -0.103 |
atted | -0.101 |
impressed | -0.096 |
omo | -0.092 |
smiling | -0.089 |
Token | Beat |
Token ID | 12568 |
Feature activation | +1.26e+01 |
Pos logit contributions | |
lee | +0.138 |
atted | +0.136 |
impressed | +0.121 |
omo | +0.114 |
smiling | +0.111 |
Neg logit contributions | |
Ready | -0.254 |
Left | -0.231 |
alis | -0.219 |
sc | -0.212 |
Three | -0.208 |
Token | rice |
Token ID | 20970 |
Feature activation | +5.85e+00 |
Pos logit contributions | |
atted | +0.486 |
lee | +0.486 |
impressed | +0.448 |
omo | +0.433 |
smiling | +0.416 |
Neg logit contributions | |
Ready | -0.535 |
Left | -0.475 |
alis | -0.433 |
sc | -0.432 |
bust | -0.412 |
Token | tried |
Token ID | 3088 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
lee | +0.197 |
atted | +0.197 |
impressed | +0.184 |
omo | +0.178 |
glad | +0.172 |
Neg logit contributions | |
Ready | -0.270 |
Left | -0.233 |
alis | -0.223 |
sc | -0.221 |
bust | -0.215 |
Token | ! |
Token ID | 0 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
atted | +0.121 |
lee | +0.119 |
omo | +0.112 |
impressed | +0.106 |
glad | +0.099 |
Neg logit contributions | |
Ready | -0.112 |
Left | -0.097 |
alis | -0.088 |
summit | -0.086 |
highest | -0.085 |
Token | She |
Token ID | 1375 |
Feature activation | -6.55e-01 |
Pos logit contributions | |
lee | +0.049 |
atted | +0.048 |
impressed | +0.043 |
omo | +0.041 |
smiling | +0.041 |
Neg logit contributions | |
Ready | -0.054 |
Left | -0.049 |
alis | -0.045 |
sc | -0.044 |
bust | -0.043 |
Token | started |
Token ID | 2067 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
Ready | +0.028 |
Left | +0.025 |
alis | +0.023 |
sc | +0.023 |
bust | +0.022 |
Neg logit contributions | |
lee | -0.025 |
atted | -0.024 |
impressed | -0.023 |
omo | -0.022 |
glad | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
Ready | +0.110 |
Left | +0.094 |
sc | +0.082 |
Three | +0.080 |
alis | +0.078 |
Neg logit contributions | |
atted | -0.148 |
lee | -0.146 |
impressed | -0.135 |
omo | -0.131 |
mad | -0.128 |
Token | " |
Token ID | 1 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
Ready | +0.138 |
Left | +0.122 |
Three | +0.112 |
sc | +0.111 |
ready | +0.109 |
Neg logit contributions | |
atted | -0.091 |
lee | -0.087 |
impressed | -0.076 |
omo | -0.074 |
smiling | -0.073 |
Token | Me |
Token ID | 5308 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
Ready | +0.050 |
Left | +0.044 |
alis | +0.043 |
sc | +0.041 |
summit | +0.040 |
Neg logit contributions | |
lee | -0.075 |
atted | -0.074 |
omo | -0.071 |
impressed | -0.071 |
smiling | -0.069 |
Token | ow |
Token ID | 322 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
lee | +0.206 |
atted | +0.204 |
impressed | +0.189 |
omo | +0.178 |
smiling | +0.174 |
Neg logit contributions | |
Ready | -0.242 |
Left | -0.216 |
alis | -0.200 |
sc | -0.195 |
bust | -0.189 |
Token | ," |
Token ID | 553 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
lee | +0.124 |
atted | +0.123 |
impressed | +0.113 |
omo | +0.105 |
smiling | +0.102 |
Neg logit contributions | |
Ready | -0.196 |
Left | -0.178 |
alis | -0.166 |
sc | -0.164 |
bust | -0.160 |
Token | m |
Token ID | 285 |
Feature activation | +1.24e+01 |
Pos logit contributions | |
atted | +0.081 |
lee | +0.078 |
omo | +0.070 |
impressed | +0.068 |
glad | +0.059 |
Neg logit contributions | |
Ready | -0.136 |
Left | -0.117 |
bust | -0.113 |
ready | -0.111 |
sc | -0.109 |
Token | ew |
Token ID | 413 |
Feature activation | +6.86e+00 |
Pos logit contributions | |
lee | +0.423 |
impressed | +0.421 |
atted | +0.389 |
knowing | +0.379 |
glad | +0.377 |
Neg logit contributions | |
Ready | -0.478 |
Left | -0.436 |
Three | -0.430 |
tr | -0.429 |
summit | -0.422 |
Token | ed |
Token ID | 276 |
Feature activation | +8.62e+00 |
Pos logit contributions | |
atted | +0.212 |
impressed | +0.210 |
lee | +0.206 |
omo | +0.204 |
glad | +0.196 |
Neg logit contributions | |
Ready | -0.303 |
Left | -0.272 |
sc | -0.257 |
bust | -0.257 |
alis | -0.254 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
atted | +0.265 |
omo | +0.255 |
lee | +0.250 |
impressed | +0.245 |
glad | +0.221 |
Neg logit contributions | |
Ready | -0.390 |
Left | -0.343 |
bust | -0.332 |
sc | -0.321 |
ready | -0.317 |
Token | cat |
Token ID | 3797 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
Ready | +0.044 |
Left | +0.039 |
alis | +0.035 |
sc | +0.035 |
bust | +0.034 |
Neg logit contributions | |
lee | -0.047 |
atted | -0.046 |
impressed | -0.043 |
omo | -0.042 |
smiling | -0.042 |
Token | , |
Token ID | 11 |
Feature activation | +1.01e+01 |
Pos logit contributions | |
lee | +0.081 |
atted | +0.081 |
impressed | +0.074 |
omo | +0.072 |
glad | +0.067 |
Neg logit contributions | |
Ready | -0.130 |
Left | -0.118 |
alis | -0.112 |
sc | -0.111 |
bust | -0.107 |
Token | time |
Token ID | 640 |
Feature activation | -2.74e+00 |
Pos logit contributions | |
lee | +0.085 |
atted | +0.081 |
impressed | +0.077 |
omo | +0.067 |
mad | +0.066 |
Neg logit contributions | |
Ready | -0.074 |
Left | -0.065 |
alis | -0.060 |
place | -0.059 |
am | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
Ready | +0.123 |
Left | +0.111 |
alis | +0.104 |
sc | +0.102 |
south | +0.099 |
Neg logit contributions | |
lee | -0.095 |
atted | -0.094 |
impressed | -0.086 |
omo | -0.082 |
mad | -0.078 |
Token | my |
Token ID | 616 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
atted | +0.265 |
lee | +0.254 |
impressed | +0.223 |
aged | +0.203 |
mad | +0.200 |
Neg logit contributions | |
Ready | -0.341 |
Left | -0.305 |
alis | -0.285 |
Ready | -0.271 |
south | -0.270 |
Token | love |
Token ID | 1842 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
lee | +0.074 |
atted | +0.072 |
impressed | +0.067 |
omo | +0.060 |
mad | +0.058 |
Neg logit contributions | |
Ready | -0.103 |
Left | -0.095 |
alis | -0.088 |
summit | -0.087 |
sc | -0.084 |
Token | ?" |
Token ID | 1701 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
lee | +0.125 |
atted | +0.123 |
impressed | +0.115 |
omo | +0.105 |
mad | +0.104 |
Neg logit contributions | |
Ready | -0.152 |
Left | -0.136 |
alis | -0.126 |
sc | -0.125 |
ready | -0.118 |
Token | Ily |
Token ID | 49804 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
atted | +0.077 |
lee | +0.075 |
impressed | +0.073 |
omo | +0.071 |
mad | +0.071 |
Neg logit contributions | |
Ready | -0.054 |
Left | -0.045 |
alis | -0.040 |
sc | -0.038 |
Three | -0.037 |
Token | a |
Token ID | 64 |
Feature activation | +8.20e+00 |
Pos logit contributions | |
lee | +0.531 |
atted | +0.508 |
impressed | +0.487 |
smiling | +0.478 |
omo | +0.467 |
Neg logit contributions | |
Ready | -0.425 |
Left | -0.383 |
sc | -0.361 |
bust | -0.345 |
alis | -0.334 |
Token | nodded |
Token ID | 14464 |
Feature activation | +6.03e+00 |
Pos logit contributions | |
atted | +0.242 |
lee | +0.239 |
glad | +0.228 |
omo | +0.227 |
ummies | +0.222 |
Neg logit contributions | |
Ready | -0.316 |
Left | -0.316 |
bust | -0.297 |
sc | -0.289 |
Three | -0.282 |
Token | happily |
Token ID | 18177 |
Feature activation | +7.96e+00 |
Pos logit contributions | |
atted | +0.198 |
omo | +0.191 |
lee | +0.181 |
inches | +0.177 |
ummies | +0.165 |
Neg logit contributions | |
Left | -0.227 |
Ready | -0.216 |
grant | -0.209 |
begin | -0.208 |
south | -0.203 |
Token | . |
Token ID | 13 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
omo | +0.253 |
atted | +0.253 |
lee | +0.247 |
impressed | +0.215 |
ummies | +0.207 |
Neg logit contributions | |
Ready | -0.321 |
Left | -0.308 |
sc | -0.305 |
fire | -0.300 |
tted | -0.297 |
Token | She |
Token ID | 1375 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
atted | +0.020 |
lee | +0.020 |
impressed | +0.019 |
mad | +0.019 |
omo | +0.018 |
Neg logit contributions | |
Ready | -0.016 |
Left | -0.014 |
alis | -0.012 |
sc | -0.012 |
Ready | -0.011 |
Token | bouncing |
Token ID | 30713 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
lee | +0.197 |
atted | +0.191 |
impressed | +0.170 |
omo | +0.169 |
mad | +0.147 |
Neg logit contributions | |
Ready | -0.356 |
Left | -0.324 |
alis | -0.305 |
sc | -0.302 |
south | -0.291 |
Token | around |
Token ID | 1088 |
Feature activation | -3.53e+00 |
Pos logit contributions | |
lee | +0.058 |
atted | +0.058 |
impressed | +0.053 |
omo | +0.049 |
smiling | +0.049 |
Neg logit contributions | |
Ready | -0.079 |
Left | -0.070 |
alis | -0.066 |
sc | -0.064 |
bust | -0.062 |
Token | , |
Token ID | 11 |
Feature activation | +8.95e+00 |
Pos logit contributions | |
Ready | +0.187 |
Left | +0.170 |
alis | +0.159 |
sc | +0.157 |
bust | +0.152 |
Neg logit contributions | |
lee | -0.100 |
atted | -0.100 |
impressed | -0.089 |
omo | -0.085 |
mad | -0.079 |
Token | an |
Token ID | 281 |
Feature activation | +1.70e+00 |
Pos logit contributions | |
atted | +0.321 |
lee | +0.312 |
impressed | +0.283 |
omo | +0.279 |
mad | +0.260 |
Neg logit contributions | |
Ready | -0.414 |
Left | -0.366 |
alis | -0.336 |
sc | -0.327 |
bust | -0.323 |
Token | unpleasant |
Token ID | 22029 |
Feature activation | +4.81e+00 |
Pos logit contributions | |
atted | +0.073 |
omo | +0.066 |
lee | +0.063 |
impressed | +0.063 |
smiling | +0.063 |
Neg logit contributions | |
Ready | -0.055 |
alis | -0.051 |
bust | -0.048 |
south | -0.048 |
Left | -0.047 |
Token | od |
Token ID | 16298 |
Feature activation | +1.23e+01 |
Pos logit contributions | |
lee | +0.192 |
atted | +0.188 |
impressed | +0.174 |
mad | +0.174 |
omo | +0.172 |
Neg logit contributions | |
Ready | -0.197 |
Left | -0.174 |
sc | -0.155 |
ready | -0.155 |
alis | -0.154 |
Token | our |
Token ID | 454 |
Feature activation | -6.60e-01 |
Pos logit contributions | |
atted | +0.397 |
impressed | +0.395 |
lee | +0.389 |
omo | +0.371 |
understanding | +0.367 |
Neg logit contributions | |
Ready | -0.539 |
Left | -0.493 |
alis | -0.485 |
sc | -0.451 |
bust | -0.440 |
Token | wa |
Token ID | 2082 |
Feature activation | +8.39e+00 |
Pos logit contributions | |
Ready | +0.030 |
Left | +0.027 |
alis | +0.024 |
sc | +0.024 |
ready | +0.023 |
Neg logit contributions | |
lee | -0.025 |
atted | -0.024 |
impressed | -0.022 |
omo | -0.021 |
smiling | -0.020 |
Token | ft |
Token ID | 701 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
atted | +0.304 |
lee | +0.304 |
impressed | +0.279 |
omo | +0.260 |
mad | +0.258 |
Neg logit contributions | |
Ready | -0.375 |
Left | -0.333 |
alis | -0.311 |
sc | -0.305 |
ready | -0.292 |
Token | ed |
Token ID | 276 |
Feature activation | +5.51e+00 |
Pos logit contributions | |
Ready | +0.071 |
Left | +0.064 |
alis | +0.062 |
summit | +0.058 |
ready | +0.057 |
Neg logit contributions | |
atted | -0.041 |
lee | -0.041 |
impressed | -0.035 |
mad | -0.033 |
smiling | -0.033 |
Token | into |
Token ID | 656 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
lee | +0.220 |
atted | +0.218 |
impressed | +0.204 |
omo | +0.197 |
glad | +0.187 |
Neg logit contributions | |
Ready | -0.226 |
Left | -0.200 |
sc | -0.182 |
alis | -0.181 |
bust | -0.173 |
Token | 't |
Token ID | 470 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
lee | +0.157 |
atted | +0.155 |
impressed | +0.147 |
mad | +0.144 |
glad | +0.142 |
Neg logit contributions | |
Ready | -0.128 |
Left | -0.109 |
alis | -0.100 |
sc | -0.099 |
ready | -0.093 |
Token | it |
Token ID | 340 |
Feature activation | -6.40e+00 |
Pos logit contributions | |
Ready | +0.006 |
Left | +0.005 |
alis | +0.005 |
sc | +0.005 |
bust | +0.005 |
Neg logit contributions | |
lee | -0.004 |
atted | -0.004 |
impressed | -0.004 |
omo | -0.004 |
smiling | -0.004 |
Token | ?" |
Token ID | 1701 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
Ready | +0.291 |
Left | +0.260 |
alis | +0.241 |
sc | +0.237 |
bust | +0.229 |
Neg logit contributions | |
lee | -0.228 |
atted | -0.226 |
impressed | -0.208 |
omo | -0.196 |
smiling | -0.191 |
Token |
|
Token ID | 198 |
Feature activation | -2.85e+00 |
Pos logit contributions | |
atted | +0.043 |
lee | +0.042 |
impressed | +0.040 |
omo | +0.039 |
mad | +0.037 |
Neg logit contributions | |
Ready | -0.030 |
Left | -0.025 |
alis | -0.021 |
sc | -0.020 |
Three | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
Ready | +0.098 |
Left | +0.085 |
alis | +0.072 |
Three | +0.071 |
sc | +0.071 |
Neg logit contributions | |
atted | -0.133 |
lee | -0.132 |
impressed | -0.124 |
omo | -0.118 |
mad | -0.117 |
Token | Ron |
Token ID | 23672 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
Ready | +0.107 |
Left | +0.093 |
Three | +0.080 |
alis | +0.079 |
sc | +0.079 |
Neg logit contributions | |
atted | -0.125 |
lee | -0.125 |
impressed | -0.115 |
omo | -0.109 |
smiling | -0.108 |
Token | ny |
Token ID | 3281 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
lee | +0.530 |
atted | +0.521 |
impressed | +0.502 |
omo | +0.464 |
glad | +0.463 |
Neg logit contributions | |
Ready | -0.460 |
Left | -0.408 |
alis | -0.362 |
sc | -0.361 |
bust | -0.347 |
Token | nodded |
Token ID | 14464 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
atted | +0.251 |
lee | +0.244 |
impressed | +0.238 |
glad | +0.236 |
omo | +0.228 |
Neg logit contributions | |
Ready | -0.289 |
Left | -0.268 |
bust | -0.248 |
alis | -0.245 |
Three | -0.237 |
Token | . |
Token ID | 13 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
atted | +0.209 |
omo | +0.204 |
lee | +0.194 |
inches | +0.184 |
impressed | +0.181 |
Neg logit contributions | |
Left | -0.161 |
Ready | -0.156 |
begin | -0.144 |
across | -0.143 |
form | -0.142 |
Token | Tears |
Token ID | 40328 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
lee | +0.011 |
atted | +0.011 |
impressed | +0.011 |
omo | +0.010 |
mad | +0.010 |
Neg logit contributions | |
Ready | -0.013 |
Left | -0.012 |
alis | -0.011 |
sc | -0.010 |
bust | -0.010 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Ready | +0.168 |
Left | +0.149 |
sc | +0.137 |
alis | +0.136 |
bust | +0.131 |
Neg logit contributions | |
lee | -0.163 |
atted | -0.161 |
impressed | -0.150 |
omo | -0.145 |
glad | -0.140 |
Token |
|
Token ID | 198 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
Ready | +0.146 |
Left | +0.128 |
alis | +0.119 |
sc | +0.115 |
bust | +0.111 |
Neg logit contributions | |
lee | -0.151 |
atted | -0.150 |
impressed | -0.141 |
omo | -0.134 |
smiling | -0.130 |
Token | M |
Token ID | 44 |
Feature activation | +5.55e+00 |
Pos logit contributions | |
Ready | +0.132 |
Left | +0.117 |
alis | +0.112 |
sc | +0.107 |
bust | +0.104 |
Neg logit contributions | |
lee | -0.122 |
atted | -0.120 |
impressed | -0.115 |
omo | -0.109 |
understanding | -0.105 |
Token | ummy |
Token ID | 13513 |
Feature activation | +8.05e+00 |
Pos logit contributions | |
lee | +0.182 |
atted | +0.179 |
impressed | +0.172 |
smiling | +0.159 |
glad | +0.154 |
Neg logit contributions | |
Ready | -0.261 |
Left | -0.238 |
alis | -0.221 |
Three | -0.214 |
summit | -0.214 |
Token | watched |
Token ID | 7342 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
atted | +0.274 |
lee | +0.263 |
glad | +0.253 |
impressed | +0.250 |
omo | +0.241 |
Neg logit contributions | |
Ready | -0.340 |
Left | -0.300 |
bust | -0.297 |
alis | -0.283 |
Three | -0.280 |
Token | him |
Token ID | 683 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
atted | +0.146 |
lee | +0.145 |
omo | +0.140 |
impressed | +0.136 |
glad | +0.127 |
Neg logit contributions | |
Ready | -0.173 |
Left | -0.163 |
sc | -0.147 |
alis | -0.143 |
south | -0.141 |
Token | anx |
Token ID | 7296 |
Feature activation | +1.22e+01 |
Pos logit contributions | |
lee | +0.092 |
atted | +0.091 |
impressed | +0.083 |
omo | +0.083 |
glad | +0.076 |
Neg logit contributions | |
Ready | -0.105 |
Left | -0.093 |
alis | -0.087 |
sc | -0.086 |
summit | -0.083 |
Token | iously |
Token ID | 6819 |
Feature activation | +6.93e+00 |
Pos logit contributions | |
impressed | +0.442 |
glad | +0.424 |
pleased | +0.396 |
lee | +0.386 |
atted | +0.385 |
Neg logit contributions | |
Ready | -0.404 |
alis | -0.390 |
peak | -0.380 |
summit | -0.374 |
Left | -0.372 |
Token | as |
Token ID | 355 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
atted | +0.202 |
lee | +0.201 |
omo | +0.195 |
impressed | +0.169 |
glad | +0.154 |
Neg logit contributions | |
Ready | -0.340 |
Left | -0.305 |
bust | -0.287 |
summit | -0.285 |
alis | -0.282 |
Token | Ali |
Token ID | 12104 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
Ready | +0.124 |
Left | +0.109 |
bust | +0.099 |
summit | +0.099 |
sc | +0.099 |
Neg logit contributions | |
atted | -0.094 |
lee | -0.094 |
omo | -0.082 |
impressed | -0.082 |
mad | -0.075 |
Token | waited |
Token ID | 13488 |
Feature activation | +5.17e+00 |
Pos logit contributions | |
lee | +0.128 |
atted | +0.126 |
impressed | +0.118 |
omo | +0.113 |
glad | +0.109 |
Neg logit contributions | |
Ready | -0.156 |
Left | -0.139 |
alis | -0.129 |
sc | -0.127 |
bust | -0.123 |
Token | for |
Token ID | 329 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
atted | +0.186 |
lee | +0.180 |
omo | +0.175 |
impressed | +0.162 |
mad | +0.151 |
Neg logit contributions | |
Ready | -0.218 |
Left | -0.197 |
sc | -0.186 |
south | -0.185 |
alis | -0.180 |
Token | " |
Token ID | 1 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
Ready | +0.111 |
Left | +0.098 |
alis | +0.091 |
sc | +0.087 |
bust | +0.085 |
Neg logit contributions | |
lee | -0.116 |
atted | -0.114 |
impressed | -0.108 |
omo | -0.102 |
smiling | -0.099 |
Token | Oh |
Token ID | 5812 |
Feature activation | +6.60e-01 |
Pos logit contributions | |
Ready | +0.058 |
Left | +0.051 |
hunt | +0.051 |
shortcut | +0.050 |
sc | +0.050 |
Neg logit contributions | |
lee | -0.074 |
impressed | -0.073 |
atted | -0.072 |
omo | -0.071 |
smiling | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | +8.11e+00 |
Pos logit contributions | |
lee | +0.019 |
atted | +0.019 |
impressed | +0.017 |
omo | +0.016 |
smiling | +0.015 |
Neg logit contributions | |
Ready | -0.034 |
Left | -0.031 |
alis | -0.029 |
sc | -0.028 |
bust | -0.028 |
Token | look |
Token ID | 804 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
lee | +0.276 |
atted | +0.269 |
impressed | +0.248 |
omo | +0.239 |
glad | +0.236 |
Neg logit contributions | |
Ready | -0.383 |
Left | -0.347 |
sc | -0.322 |
alis | -0.320 |
Three | -0.307 |
Token | ," |
Token ID | 553 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
atted | +0.046 |
lee | +0.045 |
impressed | +0.040 |
omo | +0.040 |
smiling | +0.036 |
Neg logit contributions | |
Ready | -0.070 |
Left | -0.065 |
alis | -0.060 |
sc | -0.059 |
bust | -0.058 |
Token | B |
Token ID | 347 |
Feature activation | +1.21e+01 |
Pos logit contributions | |
atted | +0.072 |
lee | +0.072 |
impressed | +0.064 |
omo | +0.060 |
smiling | +0.056 |
Neg logit contributions | |
Ready | -0.159 |
Left | -0.144 |
alis | -0.135 |
sc | -0.134 |
bust | -0.132 |
Token | ess |
Token ID | 408 |
Feature activation | +9.91e+00 |
Pos logit contributions | |
atted | +0.535 |
lee | +0.531 |
impressed | +0.521 |
omo | +0.485 |
smiling | +0.485 |
Neg logit contributions | |
Ready | -0.420 |
Left | -0.373 |
alis | -0.345 |
sc | -0.330 |
bust | -0.309 |
Token | ie |
Token ID | 494 |
Feature activation | +5.97e+00 |
Pos logit contributions | |
impressed | +0.297 |
atted | +0.295 |
lee | +0.291 |
omo | +0.282 |
glad | +0.270 |
Neg logit contributions | |
Ready | -0.428 |
Left | -0.394 |
sc | -0.376 |
alis | -0.373 |
bust | -0.369 |
Token | said |
Token ID | 531 |
Feature activation | +7.29e+00 |
Pos logit contributions | |
atted | +0.176 |
lee | +0.166 |
omo | +0.159 |
glad | +0.152 |
impressed | +0.149 |
Neg logit contributions | |
Ready | -0.268 |
Left | -0.242 |
alis | -0.237 |
bust | -0.236 |
sc | -0.231 |
Token | . |
Token ID | 13 |
Feature activation | +6.80e-01 |
Pos logit contributions | |
atted | +0.259 |
lee | +0.254 |
omo | +0.254 |
impressed | +0.228 |
glad | +0.204 |
Neg logit contributions | |
Ready | -0.299 |
Left | -0.280 |
sc | -0.260 |
bust | -0.257 |
fire | -0.256 |
Token | She |
Token ID | 1375 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
atted | +0.026 |
lee | +0.026 |
impressed | +0.024 |
omo | +0.023 |
mad | +0.023 |
Neg logit contributions | |
Ready | -0.029 |
Left | -0.026 |
alis | -0.023 |
sc | -0.023 |
bust | -0.022 |
Token | ummy |
Token ID | 13513 |
Feature activation | +5.69e+00 |
Pos logit contributions | |
atted | +0.069 |
impressed | +0.069 |
lee | +0.068 |
glad | +0.058 |
smiling | +0.053 |
Neg logit contributions | |
Ready | -0.177 |
Left | -0.167 |
alis | -0.158 |
summit | -0.158 |
sc | -0.155 |
Token | snack |
Token ID | 26906 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
lee | +0.254 |
atted | +0.252 |
impressed | +0.235 |
omo | +0.218 |
glad | +0.216 |
Neg logit contributions | |
Ready | -0.211 |
Left | -0.181 |
alis | -0.165 |
sc | -0.162 |
ready | -0.154 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
Ready | +0.026 |
Left | +0.023 |
alis | +0.022 |
sc | +0.021 |
bust | +0.021 |
Neg logit contributions | |
lee | -0.017 |
atted | -0.017 |
impressed | -0.015 |
omo | -0.014 |
smiling | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
lee | +0.097 |
atted | +0.097 |
impressed | +0.090 |
omo | +0.086 |
smiling | +0.084 |
Neg logit contributions | |
Ready | -0.096 |
Left | -0.084 |
alis | -0.077 |
sc | -0.075 |
bust | -0.073 |
Token |
|
Token ID | 198 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
Ready | +0.063 |
Left | +0.055 |
alis | +0.051 |
sc | +0.049 |
bust | +0.048 |
Neg logit contributions | |
lee | -0.068 |
atted | -0.067 |
impressed | -0.063 |
omo | -0.060 |
smiling | -0.058 |
Token | Sh |
Token ID | 2484 |
Feature activation | +1.21e+01 |
Pos logit contributions | |
Ready | +0.060 |
Left | +0.053 |
alis | +0.051 |
sc | +0.047 |
summit | +0.047 |
Neg logit contributions | |
lee | -0.071 |
atted | -0.070 |
impressed | -0.068 |
omo | -0.065 |
understanding | -0.063 |
Token | awn |
Token ID | 3832 |
Feature activation | +1.08e+01 |
Pos logit contributions | |
atted | +0.432 |
lee | +0.420 |
glad | +0.402 |
impressed | +0.400 |
mad | +0.399 |
Neg logit contributions | |
Ready | -0.463 |
Left | -0.438 |
tted | -0.421 |
alis | -0.421 |
fire | -0.399 |
Token | said |
Token ID | 531 |
Feature activation | +6.90e+00 |
Pos logit contributions | |
glad | +0.325 |
atted | +0.303 |
impressed | +0.299 |
lee | +0.298 |
omo | +0.288 |
Neg logit contributions | |
Ready | -0.468 |
Left | -0.440 |
bust | -0.429 |
sc | -0.413 |
alis | -0.399 |
Token | , |
Token ID | 11 |
Feature activation | +9.95e+00 |
Pos logit contributions | |
lee | +0.188 |
atted | +0.182 |
omo | +0.176 |
impressed | +0.162 |
mad | +0.142 |
Neg logit contributions | |
Ready | -0.355 |
Left | -0.337 |
sc | -0.306 |
br | -0.297 |
bust | -0.296 |
Token | " |
Token ID | 366 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
atted | +0.448 |
lee | +0.430 |
omo | +0.410 |
impressed | +0.383 |
mad | +0.382 |
Neg logit contributions | |
Ready | -0.358 |
Left | -0.321 |
sc | -0.269 |
alis | -0.268 |
bust | -0.263 |
Token | We |
Token ID | 1135 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
Ready | +0.046 |
Left | +0.042 |
alis | +0.041 |
bust | +0.037 |
sc | +0.037 |
Neg logit contributions | |
lee | -0.060 |
atted | -0.059 |
impressed | -0.058 |
omo | -0.056 |
smiling | -0.055 |
Token | stitch |
Token ID | 24695 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
Ready | +0.152 |
Left | +0.131 |
sc | +0.113 |
bust | +0.113 |
ready | +0.112 |
Neg logit contributions | |
lee | -0.178 |
atted | -0.173 |
smiling | -0.167 |
gigg | -0.166 |
impressed | -0.165 |
Token | ." |
Token ID | 526 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
Ready | +0.105 |
Left | +0.094 |
alis | +0.088 |
sc | +0.085 |
bust | +0.084 |
Neg logit contributions | |
lee | -0.073 |
atted | -0.072 |
impressed | -0.065 |
omo | -0.061 |
smiling | -0.061 |
Token | |
Token ID | 220 |
Feature activation | -9.99e-01 |
Pos logit contributions | |
lee | +0.059 |
atted | +0.059 |
impressed | +0.055 |
omo | +0.052 |
smiling | +0.052 |
Neg logit contributions | |
Ready | -0.052 |
Left | -0.045 |
alis | -0.040 |
sc | -0.040 |
bust | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
Ready | +0.042 |
Left | +0.037 |
alis | +0.036 |
sc | +0.034 |
bust | +0.033 |
Neg logit contributions | |
lee | -0.038 |
atted | -0.038 |
impressed | -0.035 |
omo | -0.034 |
glad | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Ready | +0.089 |
Left | +0.078 |
alis | +0.075 |
sc | +0.070 |
bust | +0.069 |
Neg logit contributions | |
lee | -0.088 |
atted | -0.088 |
impressed | -0.082 |
omo | -0.079 |
glad | -0.076 |
Token | Ron |
Token ID | 23672 |
Feature activation | +1.21e+01 |
Pos logit contributions | |
Ready | +0.084 |
Left | +0.074 |
alis | +0.068 |
sc | +0.066 |
bust | +0.064 |
Neg logit contributions | |
lee | -0.084 |
atted | -0.084 |
impressed | -0.078 |
omo | -0.075 |
smiling | -0.072 |
Token | said |
Token ID | 531 |
Feature activation | +6.63e+00 |
Pos logit contributions | |
lee | +0.368 |
atted | +0.365 |
impressed | +0.338 |
omo | +0.313 |
glad | +0.306 |
Neg logit contributions | |
Ready | -0.595 |
Left | -0.546 |
alis | -0.500 |
sc | -0.495 |
bust | -0.490 |
Token | , |
Token ID | 11 |
Feature activation | +9.76e+00 |
Pos logit contributions | |
lee | +0.201 |
atted | +0.188 |
omo | +0.186 |
impressed | +0.168 |
mad | +0.146 |
Neg logit contributions | |
Ready | -0.316 |
Left | -0.302 |
sc | -0.277 |
br | -0.275 |
fire | -0.268 |
Token | " |
Token ID | 366 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
atted | +0.431 |
lee | +0.422 |
omo | +0.400 |
impressed | +0.377 |
mad | +0.374 |
Neg logit contributions | |
Ready | -0.354 |
Left | -0.317 |
sc | -0.275 |
alis | -0.266 |
bust | -0.261 |
Token | That |
Token ID | 2504 |
Feature activation | -2.91e+00 |
Pos logit contributions | |
Ready | +0.045 |
Left | +0.040 |
alis | +0.040 |
Museum | +0.038 |
shortcut | +0.038 |
Neg logit contributions | |
atted | -0.055 |
impressed | -0.054 |
lee | -0.054 |
understanding | -0.052 |
omo | -0.051 |
Token | 's |
Token ID | 338 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
Ready | +0.113 |
Left | +0.099 |
alis | +0.091 |
sc | +0.089 |
bust | +0.085 |
Neg logit contributions | |
lee | -0.123 |
atted | -0.123 |
impressed | -0.114 |
omo | -0.109 |
smiling | -0.106 |
Token | That |
Token ID | 2504 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
Ready | +0.043 |
Left | +0.038 |
alis | +0.038 |
summit | +0.036 |
shortcut | +0.035 |
Neg logit contributions | |
lee | -0.065 |
atted | -0.064 |
impressed | -0.064 |
omo | -0.062 |
smiling | -0.061 |
Token | bow |
Token ID | 9563 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Ready | +0.105 |
Left | +0.094 |
alis | +0.086 |
sc | +0.083 |
Three | +0.082 |
Neg logit contributions | |
lee | -0.094 |
atted | -0.093 |
impressed | -0.086 |
smiling | -0.084 |
omo | -0.083 |
Token | was |
Token ID | 373 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
atted | +0.044 |
lee | +0.044 |
impressed | +0.040 |
smiling | +0.038 |
omo | +0.037 |
Neg logit contributions | |
Ready | -0.058 |
Left | -0.052 |
alis | -0.048 |
sc | -0.046 |
ready | -0.046 |
Token | pretty |
Token ID | 2495 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
atted | +0.065 |
lee | +0.065 |
impressed | +0.058 |
omo | +0.056 |
smiling | +0.053 |
Neg logit contributions | |
Ready | -0.082 |
Left | -0.074 |
alis | -0.069 |
sc | -0.068 |
ready | -0.064 |
Token | ," |
Token ID | 553 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
atted | +0.049 |
lee | +0.049 |
impressed | +0.044 |
omo | +0.042 |
glad | +0.042 |
Neg logit contributions | |
Ready | -0.048 |
Left | -0.043 |
sc | -0.043 |
alis | -0.040 |
highest | -0.039 |
Token | B |
Token ID | 347 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
atted | +0.090 |
lee | +0.087 |
impressed | +0.081 |
omo | +0.077 |
glad | +0.071 |
Neg logit contributions | |
Ready | -0.136 |
Left | -0.119 |
sc | -0.111 |
bust | -0.110 |
alis | -0.110 |
Token | izzy |
Token ID | 40593 |
Feature activation | +6.73e+00 |
Pos logit contributions | |
atted | +0.502 |
lee | +0.499 |
impressed | +0.482 |
smiling | +0.448 |
omo | +0.445 |
Neg logit contributions | |
Ready | -0.452 |
Left | -0.394 |
alis | -0.365 |
sc | -0.357 |
bust | -0.336 |
Token | said |
Token ID | 531 |
Feature activation | +7.80e+00 |
Pos logit contributions | |
atted | +0.208 |
omo | +0.202 |
glad | +0.194 |
lee | +0.192 |
impressed | +0.189 |
Neg logit contributions | |
Ready | -0.274 |
bust | -0.256 |
Left | -0.250 |
alis | -0.243 |
sc | -0.242 |
Token | plead |
Token ID | 30701 |
Feature activation | +8.29e+00 |
Pos logit contributions | |
omo | +0.328 |
atted | +0.317 |
lee | +0.305 |
impressed | +0.284 |
irling | +0.260 |
Neg logit contributions | |
Ready | -0.253 |
Left | -0.239 |
sc | -0.221 |
fire | -0.219 |
bust | -0.217 |
Token | ingly |
Token ID | 4420 |
Feature activation | +8.14e+00 |
Pos logit contributions | |
omo | +0.331 |
atted | +0.318 |
lee | +0.310 |
impressed | +0.304 |
aj | +0.270 |
Neg logit contributions | |
across | -0.249 |
tted | -0.241 |
edge | -0.240 |
fire | -0.240 |
Ready | -0.238 |
Token | . |
Token ID | 13 |
Feature activation | +5.93e-01 |
Pos logit contributions | |
atted | +0.369 |
lee | +0.358 |
omo | +0.349 |
impressed | +0.334 |
glad | +0.304 |
Neg logit contributions | |
Ready | -0.255 |
Left | -0.228 |
sc | -0.213 |
south | -0.205 |
fire | -0.201 |
Token | old |
Token ID | 1468 |
Feature activation | +3.19e+00 |
Pos logit contributions | |
Ready | +0.029 |
Left | +0.025 |
alis | +0.023 |
sc | +0.022 |
summit | +0.022 |
Neg logit contributions | |
atted | -0.031 |
lee | -0.031 |
impressed | -0.028 |
omo | -0.027 |
glad | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +8.20e+00 |
Pos logit contributions | |
atted | +0.120 |
lee | +0.119 |
impressed | +0.109 |
omo | +0.099 |
glad | +0.098 |
Neg logit contributions | |
Ready | -0.137 |
Left | -0.118 |
alis | -0.113 |
sc | -0.108 |
hunt | -0.108 |
Token | clean |
Token ID | 3424 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
atted | +0.299 |
lee | +0.292 |
omo | +0.256 |
impressed | +0.254 |
mad | +0.232 |
Neg logit contributions | |
Ready | -0.362 |
Left | -0.309 |
alis | -0.301 |
sc | -0.297 |
south | -0.290 |
Token | , |
Token ID | 11 |
Feature activation | +8.36e+00 |
Pos logit contributions | |
atted | +0.182 |
lee | +0.179 |
impressed | +0.172 |
glad | +0.160 |
omo | +0.160 |
Neg logit contributions | |
Ready | -0.167 |
Left | -0.143 |
alis | -0.137 |
sc | -0.132 |
bust | -0.126 |
Token | fluffy |
Token ID | 39145 |
Feature activation | +5.21e+00 |
Pos logit contributions | |
atted | +0.241 |
lee | +0.238 |
omo | +0.203 |
impressed | +0.195 |
mad | +0.180 |
Neg logit contributions | |
Ready | -0.427 |
Left | -0.372 |
alis | -0.365 |
sc | -0.361 |
south | -0.355 |
Token | self |
Token ID | 2116 |
Feature activation | +1.19e+01 |
Pos logit contributions | |
atted | +0.202 |
lee | +0.201 |
impressed | +0.196 |
omo | +0.179 |
glad | +0.177 |
Neg logit contributions | |
Ready | -0.211 |
Left | -0.188 |
alis | -0.175 |
sc | -0.171 |
bust | -0.164 |
Token | . |
Token ID | 13 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
lee | +0.378 |
atted | +0.375 |
impressed | +0.340 |
omo | +0.323 |
mad | +0.308 |
Neg logit contributions | |
Ready | -0.588 |
Left | -0.530 |
alis | -0.494 |
sc | -0.488 |
bust | -0.472 |
Token | Bob |
Token ID | 5811 |
Feature activation | +6.28e+00 |
Pos logit contributions | |
lee | +0.008 |
atted | +0.008 |
impressed | +0.007 |
omo | +0.007 |
smiling | +0.007 |
Neg logit contributions | |
Ready | -0.009 |
Left | -0.008 |
alis | -0.007 |
sc | -0.007 |
bust | -0.007 |
Token | and |
Token ID | 290 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
lee | +0.237 |
atted | +0.232 |
impressed | +0.223 |
mad | +0.219 |
omo | +0.218 |
Neg logit contributions | |
Ready | -0.228 |
Left | -0.207 |
bust | -0.194 |
sc | -0.188 |
will | -0.187 |
Token | the |
Token ID | 262 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
lee | +0.292 |
atted | +0.289 |
impressed | +0.265 |
omo | +0.249 |
smiling | +0.245 |
Neg logit contributions | |
Ready | -0.402 |
Left | -0.361 |
alis | -0.336 |
sc | -0.331 |
bust | -0.317 |
Token | cat |
Token ID | 3797 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
Ready | +0.087 |
Left | +0.078 |
alis | +0.071 |
sc | +0.071 |
summit | +0.067 |
Neg logit contributions | |
lee | -0.083 |
atted | -0.083 |
impressed | -0.079 |
smiling | -0.075 |
mad | -0.075 |
Token | examine |
Token ID | 10716 |
Feature activation | -7.99e-01 |
Pos logit contributions | |
Ready | +0.283 |
Left | +0.249 |
sc | +0.223 |
alis | +0.221 |
ready | +0.214 |
Neg logit contributions | |
lee | -0.325 |
atted | -0.312 |
impressed | -0.296 |
smiling | -0.294 |
understanding | -0.290 |
Token | it |
Token ID | 340 |
Feature activation | -8.43e+00 |
Pos logit contributions | |
Ready | +0.038 |
Left | +0.035 |
alis | +0.032 |
sc | +0.032 |
bust | +0.031 |
Neg logit contributions | |
lee | -0.026 |
atted | -0.026 |
impressed | -0.024 |
omo | -0.023 |
smiling | -0.022 |
Token | with |
Token ID | 351 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
Ready | +0.455 |
Left | +0.407 |
alis | +0.379 |
bust | +0.369 |
sc | +0.368 |
Neg logit contributions | |
lee | -0.217 |
impressed | -0.215 |
atted | -0.214 |
smiling | -0.196 |
mad | -0.188 |
Token | the |
Token ID | 262 |
Feature activation | -7.59e+00 |
Pos logit contributions | |
Ready | +0.141 |
Left | +0.125 |
alis | +0.114 |
sc | +0.112 |
bust | +0.108 |
Neg logit contributions | |
atted | -0.127 |
lee | -0.127 |
impressed | -0.117 |
omo | -0.113 |
smiling | -0.107 |
Token | magn |
Token ID | 7842 |
Feature activation | -3.37e+00 |
Pos logit contributions | |
Ready | +0.291 |
Left | +0.275 |
bust | +0.252 |
sc | +0.247 |
alis | +0.246 |
Neg logit contributions | |
smiling | -0.288 |
lee | -0.283 |
atted | -0.277 |
glad | -0.265 |
impressed | -0.265 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.68e+01 |
Pos logit contributions | |
Ready | +0.096 |
Left | +0.071 |
Whe | +0.062 |
ready | +0.059 |
Three | +0.056 |
Neg logit contributions | |
lee | -0.188 |
atted | -0.183 |
smiling | -0.164 |
impressed | -0.161 |
omo | -0.160 |
Token | glass |
Token ID | 5405 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
Ready | +0.799 |
Left | +0.723 |
Skip | +0.669 |
alis | +0.658 |
ready | +0.655 |
Neg logit contributions | |
atted | -0.500 |
lee | -0.491 |
omo | -0.486 |
smiling | -0.485 |
mad | -0.480 |
Token | ?" |
Token ID | 1701 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
Ready | +0.229 |
Left | +0.206 |
alis | +0.189 |
Three | +0.182 |
bust | +0.182 |
Neg logit contributions | |
atted | -0.140 |
lee | -0.140 |
impressed | -0.125 |
omo | -0.122 |
smiling | -0.117 |
Token |
|
Token ID | 198 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
Ready | +0.024 |
Left | +0.021 |
alis | +0.019 |
sc | +0.019 |
ready | +0.018 |
Neg logit contributions | |
lee | -0.025 |
atted | -0.025 |
impressed | -0.023 |
omo | -0.022 |
smiling | -0.022 |
Token |
|
Token ID | 198 |
Feature activation | -4.08e+00 |
Pos logit contributions | |
Ready | +0.156 |
Left | +0.137 |
alis | +0.123 |
sc | +0.121 |
bust | +0.116 |
Neg logit contributions | |
lee | -0.177 |
atted | -0.176 |
impressed | -0.164 |
omo | -0.156 |
smiling | -0.154 |
Token | " |
Token ID | 1 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
Ready | +0.182 |
Left | +0.162 |
sc | +0.147 |
alis | +0.146 |
Three | +0.141 |
Neg logit contributions | |
lee | -0.153 |
atted | -0.151 |
impressed | -0.139 |
smiling | -0.130 |
omo | -0.129 |
Token | When |
Token ID | 1649 |
Feature activation | -4.98e+00 |
Pos logit contributions | |
Ready | +0.116 |
Left | +0.098 |
sc | +0.088 |
alis | +0.088 |
ready | +0.084 |
Neg logit contributions | |
atted | -0.135 |
lee | -0.134 |
impressed | -0.126 |
omo | -0.119 |
mad | -0.118 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
Ready | +0.249 |
Left | +0.225 |
alis | +0.210 |
sc | +0.206 |
bust | +0.201 |
Neg logit contributions | |
lee | -0.154 |
atted | -0.153 |
impressed | -0.138 |
omo | -0.129 |
smiling | -0.125 |
Token | put |
Token ID | 1234 |
Feature activation | -7.22e-01 |
Pos logit contributions | |
lee | +0.064 |
atted | +0.064 |
impressed | +0.060 |
omo | +0.057 |
glad | +0.055 |
Neg logit contributions | |
Ready | -0.067 |
Left | -0.058 |
alis | -0.053 |
sc | -0.053 |
bust | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | -6.19e+00 |
Pos logit contributions | |
Ready | +0.034 |
Left | +0.030 |
alis | +0.028 |
sc | +0.028 |
bust | +0.027 |
Neg logit contributions | |
lee | -0.025 |
atted | -0.025 |
impressed | -0.022 |
smiling | -0.021 |
omo | -0.021 |
Token | magn |
Token ID | 7842 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
Ready | +0.256 |
Left | +0.234 |
alis | +0.213 |
sc | +0.203 |
bust | +0.201 |
Neg logit contributions | |
atted | -0.235 |
lee | -0.233 |
impressed | -0.217 |
smiling | -0.213 |
mad | -0.210 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.61e+01 |
Pos logit contributions | |
Ready | +0.070 |
Left | +0.053 |
alis | +0.042 |
Whe | +0.038 |
ready | +0.038 |
Neg logit contributions | |
atted | -0.181 |
lee | -0.180 |
impressed | -0.165 |
smiling | -0.163 |
omo | -0.158 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
Ready | +0.709 |
Left | +0.631 |
alis | +0.559 |
sc | +0.551 |
ready | +0.548 |
Neg logit contributions | |
atted | -0.582 |
lee | -0.582 |
omo | -0.555 |
impressed | -0.525 |
smiling | -0.525 |
Token | down |
Token ID | 866 |
Feature activation | -5.72e+00 |
Pos logit contributions | |
Ready | +0.154 |
Left | +0.141 |
alis | +0.133 |
sc | +0.132 |
bust | +0.128 |
Neg logit contributions | |
lee | -0.070 |
atted | -0.070 |
impressed | -0.062 |
omo | -0.057 |
smiling | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +6.65e+00 |
Pos logit contributions | |
Ready | +0.316 |
Left | +0.288 |
sc | +0.271 |
alis | +0.270 |
bust | +0.261 |
Neg logit contributions | |
lee | -0.145 |
atted | -0.143 |
impressed | -0.128 |
omo | -0.119 |
mad | -0.109 |
Token | the |
Token ID | 262 |
Feature activation | -4.58e+00 |
Pos logit contributions | |
lee | +0.163 |
atted | +0.157 |
impressed | +0.143 |
smiling | +0.129 |
glad | +0.126 |
Neg logit contributions | |
Ready | -0.373 |
Left | -0.345 |
alis | -0.325 |
sc | -0.319 |
Three | -0.311 |
Token | cater |
Token ID | 20825 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
Ready | +0.177 |
Left | +0.158 |
alis | +0.143 |
sc | +0.140 |
bust | +0.135 |
Neg logit contributions | |
lee | -0.190 |
atted | -0.188 |
impressed | -0.175 |
omo | -0.169 |
smiling | -0.168 |
Token | and |
Token ID | 290 |
Feature activation | +5.92e+00 |
Pos logit contributions | |
atted | +0.170 |
lee | +0.166 |
impressed | +0.146 |
omo | +0.140 |
glad | +0.128 |
Neg logit contributions | |
Ready | -0.301 |
Left | -0.279 |
sc | -0.262 |
hunt | -0.260 |
alis | -0.256 |
Token | handed |
Token ID | 10158 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
lee | +0.200 |
atted | +0.196 |
omo | +0.182 |
impressed | +0.172 |
mad | +0.164 |
Neg logit contributions | |
Ready | -0.275 |
Left | -0.238 |
sc | -0.229 |
alis | -0.220 |
bust | -0.219 |
Token | him |
Token ID | 683 |
Feature activation | -1.50e-01 |
Pos logit contributions | |
lee | +0.044 |
atted | +0.044 |
impressed | +0.040 |
omo | +0.038 |
smiling | +0.037 |
Neg logit contributions | |
Ready | -0.064 |
Left | -0.057 |
alis | -0.053 |
sc | -0.052 |
ready | -0.050 |
Token | the |
Token ID | 262 |
Feature activation | -6.78e+00 |
Pos logit contributions | |
Ready | +0.007 |
Left | +0.006 |
alis | +0.006 |
sc | +0.006 |
bust | +0.005 |
Neg logit contributions | |
lee | -0.005 |
atted | -0.005 |
impressed | -0.005 |
smiling | -0.005 |
mad | -0.004 |
Token | magn |
Token ID | 7842 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
Ready | +0.311 |
Left | +0.289 |
Three | +0.264 |
bust | +0.264 |
Ro | +0.264 |
Neg logit contributions | |
lee | -0.215 |
atted | -0.206 |
smiling | -0.205 |
impressed | -0.200 |
mad | -0.191 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.57e+01 |
Pos logit contributions | |
Ready | +0.060 |
Left | +0.047 |
Whe | +0.038 |
alis | +0.035 |
ready | +0.035 |
Neg logit contributions | |
lee | -0.133 |
atted | -0.131 |
impressed | -0.119 |
smiling | -0.117 |
omo | -0.113 |
Token | glass |
Token ID | 5405 |
Feature activation | -3.36e+00 |
Pos logit contributions | |
Ready | +0.721 |
Left | +0.652 |
alis | +0.584 |
Skip | +0.584 |
sc | +0.581 |
Neg logit contributions | |
lee | -0.532 |
atted | -0.520 |
omo | -0.488 |
smiling | -0.487 |
mad | -0.475 |
Token | . |
Token ID | 13 |
Feature activation | -9.56e-01 |
Pos logit contributions | |
Ready | +0.161 |
Left | +0.145 |
sc | +0.137 |
alis | +0.135 |
bust | +0.129 |
Neg logit contributions | |
lee | -0.108 |
atted | -0.107 |
impressed | -0.099 |
omo | -0.093 |
glad | -0.090 |
Token |
|
Token ID | 198 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
Ready | +0.041 |
Left | +0.037 |
alis | +0.033 |
sc | +0.033 |
bust | +0.031 |
Neg logit contributions | |
atted | -0.037 |
lee | -0.037 |
impressed | -0.034 |
omo | -0.032 |
smiling | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
Ready | +0.161 |
Left | +0.139 |
sc | +0.121 |
alis | +0.118 |
Three | +0.115 |
Neg logit contributions | |
atted | -0.215 |
lee | -0.212 |
impressed | -0.197 |
omo | -0.191 |
smiling | -0.187 |
Token | He |
Token ID | 1544 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
Ready | +0.148 |
Left | +0.124 |
sc | +0.110 |
Three | +0.105 |
alis | +0.102 |
Neg logit contributions | |
atted | -0.205 |
lee | -0.203 |
impressed | -0.182 |
omo | -0.178 |
smiling | -0.176 |
Token | , |
Token ID | 11 |
Feature activation | +8.85e+00 |
Pos logit contributions | |
lee | +0.207 |
omo | +0.202 |
atted | +0.191 |
impressed | +0.181 |
mad | +0.164 |
Neg logit contributions | |
Ready | -0.212 |
Left | -0.210 |
sc | -0.187 |
will | -0.182 |
rest | -0.181 |
Token | handing |
Token ID | 22786 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
lee | +0.396 |
atted | +0.395 |
omo | +0.365 |
impressed | +0.330 |
mad | +0.323 |
Neg logit contributions | |
Ready | -0.321 |
Left | -0.285 |
alis | -0.245 |
sc | -0.236 |
Skip | -0.231 |
Token | him |
Token ID | 683 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
lee | +0.119 |
atted | +0.118 |
impressed | +0.104 |
omo | +0.098 |
smiling | +0.096 |
Neg logit contributions | |
Ready | -0.219 |
Left | -0.199 |
alis | -0.188 |
sc | -0.184 |
bust | -0.180 |
Token | the |
Token ID | 262 |
Feature activation | -5.52e+00 |
Pos logit contributions | |
lee | +0.051 |
atted | +0.051 |
impressed | +0.046 |
omo | +0.044 |
smiling | +0.043 |
Neg logit contributions | |
Ready | -0.062 |
Left | -0.055 |
alis | -0.051 |
sc | -0.050 |
bust | -0.048 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
Ready | +0.216 |
Left | +0.198 |
alis | +0.181 |
sc | +0.178 |
bust | +0.175 |
Neg logit contributions | |
atted | -0.213 |
lee | -0.209 |
smiling | -0.202 |
impressed | -0.202 |
mad | -0.198 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.55e+01 |
Pos logit contributions | |
Ready | +0.038 |
Left | +0.028 |
alis | +0.022 |
ready | +0.020 |
Whe | +0.020 |
Neg logit contributions | |
atted | -0.099 |
lee | -0.099 |
impressed | -0.090 |
smiling | -0.089 |
omo | -0.088 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.30e+00 |
Pos logit contributions | |
Ready | +0.691 |
Left | +0.615 |
alis | +0.556 |
sc | +0.544 |
ready | +0.541 |
Neg logit contributions | |
atted | -0.549 |
lee | -0.541 |
omo | -0.526 |
smiling | -0.504 |
mad | -0.501 |
Token | . |
Token ID | 13 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
Ready | +0.107 |
Left | +0.096 |
sc | +0.090 |
alis | +0.089 |
bust | +0.085 |
Neg logit contributions | |
lee | -0.078 |
atted | -0.076 |
impressed | -0.071 |
omo | -0.066 |
glad | -0.065 |
Token | She |
Token ID | 1375 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
Ready | +0.022 |
Left | +0.019 |
alis | +0.016 |
sc | +0.016 |
Ready | +0.016 |
Neg logit contributions | |
atted | -0.028 |
lee | -0.028 |
impressed | -0.026 |
mad | -0.025 |
omo | -0.025 |
Token | looked |
Token ID | 3114 |
Feature activation | +3.03e+00 |
Pos logit contributions | |
Ready | +0.080 |
Left | +0.071 |
sc | +0.068 |
will | +0.065 |
Skip | +0.065 |
Neg logit contributions | |
lee | -0.078 |
atted | -0.076 |
glad | -0.070 |
ummies | -0.068 |
impressed | -0.067 |
Token | around |
Token ID | 1088 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
omo | +0.114 |
lee | +0.112 |
atted | +0.104 |
ummies | +0.093 |
ittens | +0.091 |
Neg logit contributions | |
Ready | -0.119 |
Left | -0.112 |
sc | -0.106 |
south | -0.105 |
will | -0.098 |
Token | cried |
Token ID | 16896 |
Feature activation | +7.22e+00 |
Pos logit contributions | |
atted | +0.239 |
glad | +0.211 |
omo | +0.205 |
lee | +0.201 |
pleased | +0.199 |
Neg logit contributions | |
Ready | -0.181 |
Left | -0.166 |
hunt | -0.166 |
bust | -0.160 |
alis | -0.159 |
Token | , |
Token ID | 11 |
Feature activation | +9.02e+00 |
Pos logit contributions | |
atted | +0.287 |
lee | +0.283 |
impressed | +0.280 |
omo | +0.262 |
pleased | +0.248 |
Neg logit contributions | |
Ready | -0.286 |
Left | -0.268 |
sc | -0.230 |
alis | -0.227 |
south | -0.224 |
Token | dropping |
Token ID | 12047 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
lee | +0.324 |
atted | +0.319 |
omo | +0.315 |
ummy | +0.267 |
ummies | +0.264 |
Neg logit contributions | |
Ready | -0.350 |
Left | -0.331 |
flap | -0.325 |
south | -0.322 |
source | -0.316 |
Token | the |
Token ID | 262 |
Feature activation | -5.51e+00 |
Pos logit contributions | |
lee | +0.079 |
atted | +0.078 |
impressed | +0.072 |
omo | +0.068 |
smiling | +0.065 |
Neg logit contributions | |
Ready | -0.113 |
Left | -0.102 |
alis | -0.094 |
sc | -0.093 |
bust | -0.090 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Ready | +0.225 |
Left | +0.203 |
alis | +0.183 |
sc | +0.180 |
bust | +0.177 |
Neg logit contributions | |
lee | -0.219 |
atted | -0.218 |
impressed | -0.203 |
smiling | -0.197 |
mad | -0.196 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.53e+01 |
Pos logit contributions | |
Ready | +0.041 |
Left | +0.030 |
ready | +0.023 |
Whe | +0.023 |
alis | +0.023 |
Neg logit contributions | |
lee | -0.099 |
atted | -0.098 |
smiling | -0.088 |
impressed | -0.088 |
omo | -0.088 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
Ready | +0.683 |
Left | +0.613 |
alis | +0.552 |
ready | +0.544 |
sc | +0.542 |
Neg logit contributions | |
lee | -0.531 |
atted | -0.524 |
omo | -0.500 |
smiling | -0.489 |
mad | -0.486 |
Token | . |
Token ID | 13 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Ready | +0.129 |
Left | +0.116 |
alis | +0.108 |
sc | +0.107 |
bust | +0.103 |
Neg logit contributions | |
lee | -0.093 |
atted | -0.093 |
impressed | -0.085 |
omo | -0.080 |
smiling | -0.078 |
Token |
|
Token ID | 198 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
Ready | +0.043 |
Left | +0.038 |
alis | +0.034 |
sc | +0.033 |
Three | +0.032 |
Neg logit contributions | |
atted | -0.044 |
lee | -0.044 |
impressed | -0.040 |
omo | -0.039 |
mad | -0.037 |
Token |
|
Token ID | 198 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
Ready | +0.135 |
Left | +0.117 |
Three | +0.094 |
Ready | +0.089 |
sc | +0.089 |
Neg logit contributions | |
atted | -0.231 |
lee | -0.228 |
impressed | -0.214 |
smiling | -0.205 |
omo | -0.204 |
Token | " |
Token ID | 1 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
Ready | +0.180 |
Left | +0.155 |
Three | +0.136 |
tted | +0.132 |
Skip | +0.131 |
Neg logit contributions | |
atted | -0.183 |
lee | -0.182 |
impressed | -0.162 |
smiling | -0.156 |
omo | -0.150 |
Token | borrowed |
Token ID | 22546 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
Ready | +0.086 |
Left | +0.080 |
Skip | +0.071 |
bust | +0.069 |
Ro | +0.069 |
Neg logit contributions | |
atted | -0.110 |
lee | -0.103 |
ummies | -0.100 |
omo | -0.098 |
glad | -0.098 |
Token | her |
Token ID | 607 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
atted | +0.064 |
lee | +0.064 |
impressed | +0.059 |
omo | +0.059 |
glad | +0.054 |
Neg logit contributions | |
Ready | -0.066 |
Left | -0.058 |
alis | -0.053 |
sc | -0.052 |
ready | -0.050 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
Ready | +0.060 |
Left | +0.053 |
alis | +0.049 |
sc | +0.048 |
summit | +0.046 |
Neg logit contributions | |
atted | -0.062 |
lee | -0.062 |
impressed | -0.058 |
omo | -0.056 |
mad | -0.053 |
Token | 's |
Token ID | 338 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
lee | +0.091 |
atted | +0.091 |
impressed | +0.086 |
omo | +0.081 |
glad | +0.076 |
Neg logit contributions | |
Ready | -0.124 |
Left | -0.111 |
sc | -0.109 |
alis | -0.108 |
hunt | -0.102 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
Ready | +0.105 |
Left | +0.092 |
alis | +0.085 |
sc | +0.084 |
summit | +0.079 |
Neg logit contributions | |
lee | -0.109 |
atted | -0.109 |
impressed | -0.101 |
omo | -0.098 |
mad | -0.093 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.52e+01 |
Pos logit contributions | |
Ready | +0.026 |
Left | +0.019 |
alis | +0.015 |
sc | +0.014 |
summit | +0.010 |
Neg logit contributions | |
lee | -0.118 |
atted | -0.118 |
impressed | -0.115 |
omo | -0.112 |
mad | -0.110 |
Token | glass |
Token ID | 5405 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
Ready | +0.656 |
Left | +0.580 |
sc | +0.518 |
ready | +0.513 |
alis | +0.512 |
Neg logit contributions | |
lee | -0.585 |
atted | -0.578 |
omo | -0.532 |
smiling | -0.525 |
impressed | -0.520 |
Token | to |
Token ID | 284 |
Feature activation | -5.46e+00 |
Pos logit contributions | |
Ready | +0.149 |
Left | +0.135 |
sc | +0.132 |
alis | +0.130 |
summit | +0.122 |
Neg logit contributions | |
lee | -0.093 |
atted | -0.093 |
impressed | -0.089 |
omo | -0.081 |
glad | -0.081 |
Token | take |
Token ID | 1011 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
Ready | +0.222 |
Left | +0.201 |
sc | +0.177 |
alis | +0.173 |
bust | +0.170 |
Neg logit contributions | |
lee | -0.225 |
atted | -0.212 |
impressed | -0.206 |
understanding | -0.202 |
smiling | -0.201 |
Token | a |
Token ID | 257 |
Feature activation | -4.54e+00 |
Pos logit contributions | |
Ready | +0.172 |
Left | +0.155 |
alis | +0.144 |
sc | +0.142 |
bust | +0.138 |
Neg logit contributions | |
lee | -0.113 |
atted | -0.112 |
impressed | -0.101 |
omo | -0.095 |
smiling | -0.093 |
Token | closer |
Token ID | 5699 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
Ready | +0.167 |
Left | +0.148 |
alis | +0.134 |
Whe | +0.129 |
bust | +0.128 |
Neg logit contributions | |
lee | -0.190 |
atted | -0.187 |
impressed | -0.184 |
smiling | -0.178 |
mad | -0.177 |
Token | . |
Token ID | 13 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Ready | +0.239 |
Left | +0.219 |
sc | +0.211 |
alis | +0.207 |
south | +0.200 |
Neg logit contributions | |
atted | -0.136 |
lee | -0.133 |
impressed | -0.118 |
omo | -0.115 |
mad | -0.108 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
Ready | +0.060 |
Left | +0.054 |
alis | +0.053 |
bust | +0.051 |
works | +0.050 |
Neg logit contributions | |
lee | -0.045 |
atted | -0.042 |
omo | -0.040 |
impressed | -0.039 |
glad | -0.036 |
Token | had |
Token ID | 550 |
Feature activation | -4.88e+00 |
Pos logit contributions | |
atted | +0.150 |
lee | +0.147 |
mad | +0.135 |
glad | +0.132 |
impressed | +0.132 |
Neg logit contributions | |
Ready | -0.123 |
Left | -0.103 |
sc | -0.100 |
hunt | -0.096 |
bust | -0.095 |
Token | a |
Token ID | 257 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
Ready | +0.212 |
Left | +0.189 |
alis | +0.175 |
sc | +0.173 |
summit | +0.165 |
Neg logit contributions | |
lee | -0.186 |
atted | -0.180 |
impressed | -0.170 |
smiling | -0.160 |
glad | -0.157 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Ready | +0.111 |
Left | +0.098 |
alis | +0.088 |
sc | +0.086 |
bust | +0.083 |
Neg logit contributions | |
lee | -0.130 |
atted | -0.128 |
impressed | -0.121 |
omo | -0.116 |
smiling | -0.114 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.51e+01 |
Pos logit contributions | |
Ready | +0.032 |
Left | +0.023 |
alis | +0.018 |
ready | +0.017 |
Whe | +0.017 |
Neg logit contributions | |
lee | -0.083 |
atted | -0.083 |
impressed | -0.076 |
smiling | -0.074 |
omo | -0.074 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
Ready | +0.710 |
Left | +0.628 |
ready | +0.588 |
sc | +0.567 |
Skip | +0.560 |
Neg logit contributions | |
omo | -0.490 |
lee | -0.473 |
atted | -0.463 |
smiling | -0.459 |
mad | -0.447 |
Token | that |
Token ID | 326 |
Feature activation | -6.19e+00 |
Pos logit contributions | |
Ready | +0.149 |
Left | +0.136 |
alis | +0.127 |
sc | +0.124 |
bust | +0.122 |
Neg logit contributions | |
lee | -0.078 |
atted | -0.078 |
impressed | -0.069 |
omo | -0.065 |
smiling | -0.063 |
Token | she |
Token ID | 673 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
Ready | +0.311 |
Left | +0.281 |
alis | +0.263 |
sc | +0.259 |
bust | +0.251 |
Neg logit contributions | |
lee | -0.189 |
atted | -0.189 |
impressed | -0.170 |
omo | -0.160 |
smiling | -0.154 |
Token | found |
Token ID | 1043 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
Ready | +0.226 |
Left | +0.199 |
alis | +0.184 |
sc | +0.183 |
bust | +0.176 |
Neg logit contributions | |
atted | -0.198 |
lee | -0.198 |
impressed | -0.181 |
omo | -0.174 |
smiling | -0.169 |
Token | in |
Token ID | 287 |
Feature activation | -7.14e+00 |
Pos logit contributions | |
Ready | +0.124 |
Left | +0.111 |
alis | +0.103 |
sc | +0.101 |
bust | +0.099 |
Neg logit contributions | |
lee | -0.091 |
atted | -0.090 |
impressed | -0.082 |
omo | -0.076 |
smiling | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | -5.29e+00 |
Pos logit contributions | |
lee | +0.051 |
atted | +0.050 |
impressed | +0.048 |
omo | +0.046 |
glad | +0.043 |
Neg logit contributions | |
Ready | -0.065 |
Left | -0.058 |
alis | -0.053 |
sc | -0.052 |
bust | -0.052 |
Token | necklace |
Token ID | 38987 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
Ready | +0.245 |
Left | +0.228 |
sc | +0.202 |
Whe | +0.202 |
alis | +0.202 |
Neg logit contributions | |
lee | -0.177 |
impressed | -0.168 |
atted | -0.166 |
smiling | -0.165 |
mad | -0.158 |
Token | and |
Token ID | 290 |
Feature activation | +6.47e+00 |
Pos logit contributions | |
Ready | +0.190 |
Left | +0.173 |
sc | +0.170 |
alis | +0.166 |
tr | +0.164 |
Neg logit contributions | |
lee | -0.095 |
atted | -0.094 |
impressed | -0.084 |
omo | -0.081 |
glad | -0.074 |
Token | the |
Token ID | 262 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
lee | +0.208 |
atted | +0.208 |
omo | +0.191 |
impressed | +0.181 |
glad | +0.177 |
Neg logit contributions | |
Ready | -0.297 |
Left | -0.255 |
bust | -0.247 |
sc | -0.245 |
hunt | -0.237 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
Ready | +0.191 |
Left | +0.178 |
alis | +0.161 |
sc | +0.158 |
Three | +0.154 |
Neg logit contributions | |
lee | -0.170 |
atted | -0.164 |
impressed | -0.159 |
smiling | -0.157 |
mad | -0.151 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
Ready | +0.037 |
Left | +0.029 |
Whe | +0.023 |
alis | +0.023 |
ready | +0.022 |
Neg logit contributions | |
lee | -0.076 |
atted | -0.075 |
impressed | -0.067 |
smiling | -0.066 |
omo | -0.064 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.18e+00 |
Pos logit contributions | |
Ready | +0.689 |
Left | +0.623 |
sc | +0.565 |
Skip | +0.561 |
ready | +0.554 |
Neg logit contributions | |
lee | -0.490 |
atted | -0.479 |
omo | -0.459 |
smiling | -0.455 |
mad | -0.445 |
Token | and |
Token ID | 290 |
Feature activation | +5.59e+00 |
Pos logit contributions | |
Ready | +0.101 |
Left | +0.091 |
alis | +0.084 |
sc | +0.084 |
bust | +0.080 |
Neg logit contributions | |
lee | -0.074 |
atted | -0.073 |
impressed | -0.068 |
omo | -0.064 |
glad | -0.062 |
Token | told |
Token ID | 1297 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
atted | +0.196 |
lee | +0.196 |
omo | +0.180 |
impressed | +0.175 |
glad | +0.168 |
Neg logit contributions | |
Ready | -0.249 |
Left | -0.212 |
sc | -0.201 |
bust | -0.199 |
alis | -0.199 |
Token | him |
Token ID | 683 |
Feature activation | +8.63e-01 |
Pos logit contributions | |
atted | +0.159 |
lee | +0.154 |
impressed | +0.147 |
omo | +0.144 |
glad | +0.134 |
Neg logit contributions | |
Ready | -0.178 |
Left | -0.160 |
sc | -0.142 |
alis | -0.141 |
summit | -0.137 |
Token | that |
Token ID | 326 |
Feature activation | -5.23e+00 |
Pos logit contributions | |
atted | +0.029 |
lee | +0.029 |
impressed | +0.027 |
omo | +0.026 |
smiling | +0.024 |
Neg logit contributions | |
Ready | -0.040 |
Left | -0.036 |
alis | -0.033 |
sc | -0.032 |
bust | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | -4.28e+00 |
Pos logit contributions | |
Ready | +0.025 |
Left | +0.023 |
alis | +0.021 |
sc | +0.021 |
bust | +0.020 |
Neg logit contributions | |
lee | -0.017 |
atted | -0.017 |
impressed | -0.015 |
omo | -0.014 |
smiling | -0.014 |
Token | watch |
Token ID | 2342 |
Feature activation | -9.20e-01 |
Pos logit contributions | |
Ready | +0.153 |
Left | +0.146 |
sc | +0.134 |
bust | +0.131 |
alis | +0.129 |
Neg logit contributions | |
smiling | -0.170 |
lee | -0.166 |
impressed | -0.163 |
atted | -0.162 |
mad | -0.158 |
Token | with |
Token ID | 351 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
Ready | +0.048 |
Left | +0.043 |
alis | +0.040 |
sc | +0.039 |
bust | +0.039 |
Neg logit contributions | |
lee | -0.027 |
atted | -0.027 |
impressed | -0.024 |
smiling | -0.023 |
omo | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -4.99e+00 |
Pos logit contributions | |
Ready | +0.029 |
Left | +0.026 |
alis | +0.024 |
sc | +0.023 |
bust | +0.023 |
Neg logit contributions | |
lee | -0.023 |
atted | -0.023 |
impressed | -0.021 |
omo | -0.020 |
smiling | -0.020 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Ready | +0.178 |
Left | +0.165 |
sc | +0.151 |
alis | +0.148 |
bust | +0.146 |
Neg logit contributions | |
lee | -0.204 |
atted | -0.203 |
smiling | -0.202 |
impressed | -0.197 |
mad | -0.191 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
Ready | +0.031 |
Left | +0.023 |
Whe | +0.019 |
ready | +0.018 |
bust | +0.017 |
Neg logit contributions | |
lee | -0.069 |
atted | -0.069 |
smiling | -0.062 |
impressed | -0.061 |
omo | -0.060 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
Ready | +0.663 |
Left | +0.591 |
alis | +0.533 |
sc | +0.528 |
ready | +0.526 |
Neg logit contributions | |
atted | -0.509 |
lee | -0.508 |
omo | -0.487 |
smiling | -0.478 |
mad | -0.474 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
Ready | +0.093 |
Left | +0.083 |
alis | +0.078 |
sc | +0.077 |
bust | +0.074 |
Neg logit contributions | |
lee | -0.068 |
atted | -0.068 |
impressed | -0.062 |
omo | -0.059 |
glad | -0.057 |
Token | He |
Token ID | 679 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
Ready | +0.001 |
Left | +0.001 |
alis | +0.001 |
sc | +0.001 |
bust | +0.001 |
Neg logit contributions | |
lee | -0.001 |
atted | -0.001 |
impressed | -0.001 |
omo | -0.001 |
smiling | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
Ready | +0.042 |
Left | +0.037 |
bust | +0.032 |
alis | +0.031 |
Three | +0.031 |
Neg logit contributions | |
atted | -0.054 |
lee | -0.053 |
omo | -0.049 |
impressed | -0.047 |
glad | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
lee | +0.075 |
atted | +0.074 |
impressed | +0.068 |
omo | +0.067 |
smiling | +0.063 |
Neg logit contributions | |
Ready | -0.066 |
Left | -0.057 |
alis | -0.051 |
sc | -0.049 |
bust | -0.049 |
Token | She |
Token ID | 1375 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
atted | +0.004 |
lee | +0.004 |
impressed | +0.004 |
omo | +0.004 |
mad | +0.004 |
Neg logit contributions | |
Ready | -0.004 |
Left | -0.003 |
sc | -0.003 |
alis | -0.003 |
Three | -0.003 |
Token | takes |
Token ID | 2753 |
Feature activation | +3.65e-01 |
Pos logit contributions | |
Ready | +0.076 |
Left | +0.067 |
alis | +0.061 |
sc | +0.059 |
bust | +0.057 |
Neg logit contributions | |
lee | -0.083 |
atted | -0.082 |
impressed | -0.077 |
omo | -0.073 |
smiling | -0.072 |
Token | out |
Token ID | 503 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
lee | +0.015 |
atted | +0.014 |
impressed | +0.013 |
mad | +0.013 |
smiling | +0.012 |
Neg logit contributions | |
Ready | -0.015 |
Left | -0.013 |
alis | -0.012 |
sc | -0.012 |
summit | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
Ready | +0.079 |
Left | +0.070 |
alis | +0.063 |
sc | +0.063 |
bust | +0.062 |
Neg logit contributions | |
atted | -0.085 |
lee | -0.085 |
impressed | -0.080 |
omo | -0.077 |
glad | -0.074 |
Token | magn |
Token ID | 7842 |
Feature activation | -9.45e-01 |
Pos logit contributions | |
Ready | +0.102 |
Left | +0.095 |
alis | +0.086 |
Whe | +0.081 |
sc | +0.080 |
Neg logit contributions | |
lee | -0.100 |
atted | -0.099 |
impressed | -0.099 |
smiling | -0.094 |
mad | -0.094 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.48e+01 |
Pos logit contributions | |
Ready | +0.022 |
Left | +0.017 |
alis | +0.013 |
ready | +0.012 |
sc | +0.012 |
Neg logit contributions | |
lee | -0.056 |
atted | -0.056 |
impressed | -0.051 |
smiling | -0.050 |
omo | -0.049 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
Ready | +0.643 |
Left | +0.575 |
sc | +0.516 |
alis | +0.509 |
ready | +0.501 |
Neg logit contributions | |
lee | -0.540 |
atted | -0.534 |
omo | -0.510 |
impressed | -0.499 |
mad | -0.491 |
Token | . |
Token ID | 13 |
Feature activation | -5.30e-01 |
Pos logit contributions | |
Ready | +0.100 |
Left | +0.090 |
sc | +0.084 |
alis | +0.084 |
bust | +0.080 |
Neg logit contributions | |
lee | -0.067 |
atted | -0.066 |
impressed | -0.061 |
omo | -0.057 |
glad | -0.055 |
Token | She |
Token ID | 1375 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
Ready | +0.021 |
Left | +0.019 |
alis | +0.017 |
sc | +0.017 |
bust | +0.017 |
Neg logit contributions | |
lee | -0.021 |
atted | -0.021 |
impressed | -0.020 |
omo | -0.019 |
smiling | -0.018 |
Token | looks |
Token ID | 3073 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
Ready | +0.093 |
Left | +0.080 |
alis | +0.073 |
sc | +0.072 |
bust | +0.069 |
Neg logit contributions | |
lee | -0.104 |
atted | -0.103 |
impressed | -0.097 |
omo | -0.093 |
smiling | -0.090 |
Token | at |
Token ID | 379 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
atted | +0.074 |
lee | +0.074 |
omo | +0.071 |
impressed | +0.063 |
smiling | +0.060 |
Neg logit contributions | |
Ready | -0.098 |
Left | -0.087 |
sc | -0.083 |
alis | -0.080 |
bust | -0.079 |
Token | flashlight |
Token ID | 38371 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
Ready | +0.056 |
Left | +0.049 |
alis | +0.044 |
sc | +0.043 |
bust | +0.042 |
Neg logit contributions | |
lee | -0.063 |
atted | -0.063 |
impressed | -0.058 |
omo | -0.055 |
smiling | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +7.12e+00 |
Pos logit contributions | |
Ready | +0.028 |
Left | +0.025 |
alis | +0.024 |
sc | +0.023 |
ready | +0.023 |
Neg logit contributions | |
lee | -0.015 |
atted | -0.015 |
impressed | -0.013 |
smiling | -0.012 |
omo | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +6.19e+00 |
Pos logit contributions | |
lee | +0.204 |
atted | +0.202 |
impressed | +0.179 |
omo | +0.170 |
mad | +0.159 |
Neg logit contributions | |
Ready | -0.374 |
Left | -0.337 |
alis | -0.318 |
sc | -0.318 |
bust | -0.306 |
Token | a |
Token ID | 257 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
lee | +0.225 |
atted | +0.223 |
impressed | +0.207 |
omo | +0.193 |
smiling | +0.192 |
Neg logit contributions | |
Ready | -0.274 |
Left | -0.247 |
alis | -0.227 |
sc | -0.222 |
bust | -0.214 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Ready | +0.048 |
Left | +0.042 |
alis | +0.038 |
sc | +0.037 |
bust | +0.036 |
Neg logit contributions | |
lee | -0.055 |
atted | -0.054 |
impressed | -0.051 |
omo | -0.048 |
smiling | -0.047 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.47e+01 |
Pos logit contributions | |
Ready | +0.038 |
Left | +0.030 |
alis | +0.025 |
Whe | +0.023 |
ready | +0.023 |
Neg logit contributions | |
lee | -0.082 |
atted | -0.081 |
impressed | -0.074 |
smiling | -0.073 |
omo | -0.070 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
Ready | +0.631 |
Left | +0.565 |
sc | +0.501 |
alis | +0.495 |
ready | +0.494 |
Neg logit contributions | |
lee | -0.544 |
atted | -0.537 |
omo | -0.526 |
smiling | -0.500 |
mad | -0.491 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
Ready | +0.072 |
Left | +0.065 |
alis | +0.060 |
sc | +0.059 |
bust | +0.057 |
Neg logit contributions | |
lee | -0.053 |
atted | -0.053 |
impressed | -0.048 |
omo | -0.045 |
smiling | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
Ready | +0.062 |
Left | +0.056 |
alis | +0.052 |
sc | +0.051 |
bust | +0.049 |
Neg logit contributions | |
lee | -0.052 |
atted | -0.052 |
impressed | -0.047 |
omo | -0.045 |
smiling | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -3.46e+00 |
Pos logit contributions | |
Ready | +0.149 |
Left | +0.130 |
alis | +0.118 |
sc | +0.115 |
bust | +0.110 |
Neg logit contributions | |
lee | -0.171 |
atted | -0.171 |
impressed | -0.158 |
omo | -0.151 |
smiling | -0.149 |
Token | As |
Token ID | 1722 |
Feature activation | +9.33e-01 |
Pos logit contributions | |
Ready | +0.151 |
Left | +0.134 |
alis | +0.124 |
sc | +0.122 |
bust | +0.117 |
Neg logit contributions | |
lee | -0.130 |
atted | -0.129 |
impressed | -0.119 |
omo | -0.112 |
smiling | -0.110 |
Token | the |
Token ID | 262 |
Feature activation | -4.92e+00 |
Pos logit contributions | |
Ready | +0.075 |
Left | +0.069 |
south | +0.062 |
alis | +0.062 |
sc | +0.062 |
Neg logit contributions | |
lee | -0.062 |
atted | -0.062 |
impressed | -0.061 |
omo | -0.058 |
pleased | -0.053 |
Token | necklace |
Token ID | 38987 |
Feature activation | -3.52e+00 |
Pos logit contributions | |
Ready | +0.198 |
Left | +0.180 |
alis | +0.162 |
bust | +0.157 |
sc | +0.156 |
Neg logit contributions | |
atted | -0.195 |
lee | -0.195 |
impressed | -0.178 |
smiling | -0.177 |
omo | -0.174 |
Token | and |
Token ID | 290 |
Feature activation | +6.49e+00 |
Pos logit contributions | |
Ready | +0.192 |
sc | +0.177 |
Left | +0.177 |
alis | +0.173 |
tr | +0.170 |
Neg logit contributions | |
lee | -0.086 |
atted | -0.084 |
impressed | -0.076 |
omo | -0.070 |
mad | -0.065 |
Token | the |
Token ID | 262 |
Feature activation | -4.24e+00 |
Pos logit contributions | |
lee | +0.249 |
atted | +0.248 |
impressed | +0.234 |
omo | +0.227 |
glad | +0.218 |
Neg logit contributions | |
Ready | -0.274 |
Left | -0.233 |
sc | -0.219 |
alis | -0.216 |
bust | -0.215 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Ready | +0.169 |
Left | +0.154 |
alis | +0.140 |
sc | +0.137 |
Three | +0.132 |
Neg logit contributions | |
lee | -0.169 |
atted | -0.166 |
impressed | -0.156 |
smiling | -0.154 |
omo | -0.149 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.46e+01 |
Pos logit contributions | |
Ready | +0.028 |
Left | +0.022 |
alis | +0.018 |
ready | +0.018 |
Whe | +0.018 |
Neg logit contributions | |
atted | -0.057 |
lee | -0.057 |
impressed | -0.051 |
smiling | -0.051 |
omo | -0.049 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
Ready | +0.638 |
Left | +0.569 |
sc | +0.518 |
alis | +0.512 |
ready | +0.505 |
Neg logit contributions | |
lee | -0.531 |
atted | -0.524 |
omo | -0.500 |
smiling | -0.485 |
mad | -0.477 |
Token | and |
Token ID | 290 |
Feature activation | +5.92e+00 |
Pos logit contributions | |
Ready | +0.069 |
Left | +0.062 |
alis | +0.058 |
sc | +0.058 |
bust | +0.055 |
Neg logit contributions | |
lee | -0.051 |
atted | -0.051 |
impressed | -0.047 |
omo | -0.044 |
glad | -0.043 |
Token | started |
Token ID | 2067 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
atted | +0.226 |
lee | +0.226 |
impressed | +0.209 |
omo | +0.203 |
glad | +0.193 |
Neg logit contributions | |
Ready | -0.253 |
Left | -0.219 |
alis | -0.204 |
sc | -0.202 |
bust | -0.197 |
Token | to |
Token ID | 284 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
Ready | +0.006 |
Left | +0.005 |
alis | +0.005 |
gra | +0.005 |
sc | +0.005 |
Neg logit contributions | |
lee | -0.006 |
atted | -0.006 |
omo | -0.006 |
impressed | -0.006 |
mad | -0.005 |
Token | examine |
Token ID | 10716 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
Ready | +0.096 |
Left | +0.086 |
sc | +0.075 |
alis | +0.072 |
ready | +0.070 |
Neg logit contributions | |
lee | -0.129 |
atted | -0.125 |
smiling | -0.120 |
impressed | -0.120 |
gigg | -0.118 |
Token | She |
Token ID | 1375 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
Ready | +0.040 |
Left | +0.036 |
alis | +0.033 |
sc | +0.032 |
summit | +0.032 |
Neg logit contributions | |
lee | -0.031 |
atted | -0.030 |
impressed | -0.027 |
glad | -0.025 |
smiling | -0.025 |
Token | took |
Token ID | 1718 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
Ready | +0.078 |
Left | +0.066 |
sc | +0.063 |
alis | +0.063 |
bust | +0.062 |
Neg logit contributions | |
atted | -0.079 |
lee | -0.079 |
impressed | -0.073 |
omo | -0.073 |
mad | -0.072 |
Token | out |
Token ID | 503 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
lee | +0.048 |
atted | +0.047 |
impressed | +0.042 |
smiling | +0.041 |
mad | +0.040 |
Neg logit contributions | |
Ready | -0.055 |
Left | -0.048 |
alis | -0.046 |
sc | -0.044 |
Three | -0.043 |
Token | a |
Token ID | 257 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
Ready | +0.088 |
Left | +0.079 |
alis | +0.072 |
sc | +0.071 |
bust | +0.070 |
Neg logit contributions | |
lee | -0.087 |
atted | -0.087 |
impressed | -0.082 |
omo | -0.081 |
glad | -0.075 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Ready | +0.102 |
Left | +0.094 |
alis | +0.086 |
Whe | +0.083 |
bust | +0.079 |
Neg logit contributions | |
impressed | -0.099 |
lee | -0.098 |
atted | -0.097 |
smiling | -0.094 |
mad | -0.092 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.45e+01 |
Pos logit contributions | |
Ready | +0.021 |
Left | +0.015 |
alis | +0.012 |
sc | +0.010 |
ready | +0.010 |
Neg logit contributions | |
lee | -0.063 |
atted | -0.063 |
impressed | -0.059 |
smiling | -0.057 |
omo | -0.057 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Ready | +0.645 |
Left | +0.579 |
ready | +0.520 |
sc | +0.516 |
Whe | +0.508 |
Neg logit contributions | |
lee | -0.500 |
atted | -0.487 |
omo | -0.466 |
smiling | -0.465 |
impressed | -0.459 |
Token | and |
Token ID | 290 |
Feature activation | +5.78e+00 |
Pos logit contributions | |
Ready | +0.116 |
Left | +0.106 |
alis | +0.100 |
sc | +0.099 |
bust | +0.095 |
Neg logit contributions | |
lee | -0.059 |
atted | -0.058 |
impressed | -0.052 |
omo | -0.048 |
smiling | -0.046 |
Token | looked |
Token ID | 3114 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
atted | +0.220 |
lee | +0.219 |
impressed | +0.202 |
omo | +0.198 |
mad | +0.188 |
Neg logit contributions | |
Ready | -0.249 |
Left | -0.216 |
alis | -0.202 |
sc | -0.199 |
bust | -0.193 |
Token | very |
Token ID | 845 |
Feature activation | +3.77e+00 |
Pos logit contributions | |
atted | +0.074 |
lee | +0.074 |
omo | +0.071 |
impressed | +0.067 |
smiling | +0.064 |
Neg logit contributions | |
Ready | -0.065 |
Left | -0.057 |
sc | -0.052 |
alis | -0.051 |
bust | -0.050 |
Token | carefully |
Token ID | 7773 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
atted | +0.129 |
omo | +0.127 |
lee | +0.123 |
smiling | +0.100 |
impressed | +0.100 |
Neg logit contributions | |
Ready | -0.166 |
Left | -0.151 |
bust | -0.139 |
summit | -0.138 |
highest | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
Ready | +0.169 |
Left | +0.153 |
alis | +0.142 |
bust | +0.139 |
Three | +0.137 |
Neg logit contributions | |
lee | -0.090 |
atted | -0.088 |
impressed | -0.079 |
smiling | -0.075 |
omo | -0.070 |
Token | They |
Token ID | 1119 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
Ready | +0.046 |
Left | +0.040 |
alis | +0.038 |
sc | +0.036 |
bust | +0.036 |
Neg logit contributions | |
lee | -0.047 |
atted | -0.046 |
impressed | -0.043 |
omo | -0.041 |
smiling | -0.040 |
Token | had |
Token ID | 550 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
Ready | +0.129 |
Left | +0.109 |
bust | +0.101 |
alis | +0.099 |
sc | +0.098 |
Neg logit contributions | |
atted | -0.147 |
lee | -0.143 |
impressed | -0.133 |
omo | -0.131 |
glad | -0.125 |
Token | a |
Token ID | 257 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
Ready | +0.186 |
Left | +0.164 |
alis | +0.152 |
sc | +0.149 |
bust | +0.143 |
Neg logit contributions | |
lee | -0.172 |
atted | -0.170 |
impressed | -0.159 |
omo | -0.149 |
smiling | -0.149 |
Token | magn |
Token ID | 7842 |
Feature activation | -8.02e-01 |
Pos logit contributions | |
Ready | +0.084 |
Left | +0.072 |
alis | +0.064 |
sc | +0.062 |
bust | +0.059 |
Neg logit contributions | |
lee | -0.119 |
atted | -0.119 |
impressed | -0.110 |
omo | -0.105 |
smiling | -0.103 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.45e+01 |
Pos logit contributions | |
Ready | +0.019 |
Left | +0.014 |
ready | +0.011 |
alis | +0.011 |
Whe | +0.010 |
Neg logit contributions | |
atted | -0.048 |
lee | -0.048 |
impressed | -0.043 |
smiling | -0.042 |
omo | -0.042 |
Token | glass |
Token ID | 5405 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
Ready | +0.656 |
Left | +0.572 |
ready | +0.538 |
sc | +0.519 |
works | +0.514 |
Neg logit contributions | |
omo | -0.496 |
lee | -0.479 |
atted | -0.475 |
smiling | -0.462 |
mad | -0.462 |
Token | and |
Token ID | 290 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
Ready | +0.097 |
Left | +0.087 |
alis | +0.081 |
sc | +0.079 |
bust | +0.077 |
Neg logit contributions | |
lee | -0.068 |
atted | -0.068 |
impressed | -0.062 |
omo | -0.059 |
smiling | -0.057 |
Token | a |
Token ID | 257 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
atted | +0.234 |
lee | +0.233 |
impressed | +0.217 |
omo | +0.208 |
smiling | +0.202 |
Neg logit contributions | |
Ready | -0.217 |
Left | -0.188 |
alis | -0.172 |
sc | -0.170 |
bust | -0.162 |
Token | small |
Token ID | 1402 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
Ready | +0.094 |
Left | +0.082 |
alis | +0.074 |
sc | +0.073 |
bust | +0.070 |
Neg logit contributions | |
lee | -0.109 |
atted | -0.108 |
impressed | -0.101 |
omo | -0.096 |
smiling | -0.094 |
Token | box |
Token ID | 3091 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
lee | +0.043 |
atted | +0.043 |
impressed | +0.040 |
omo | +0.038 |
smiling | +0.038 |
Neg logit contributions | |
Ready | -0.036 |
Left | -0.031 |
alis | -0.028 |
sc | -0.027 |
bust | -0.026 |
Token | She |
Token ID | 1375 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
Ready | +0.017 |
Left | +0.015 |
alis | +0.014 |
sc | +0.014 |
bust | +0.013 |
Neg logit contributions | |
lee | -0.018 |
atted | -0.018 |
impressed | -0.016 |
omo | -0.016 |
glad | -0.016 |
Token | had |
Token ID | 550 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
Ready | +0.066 |
Left | +0.058 |
bust | +0.053 |
alis | +0.052 |
Three | +0.052 |
Neg logit contributions | |
atted | -0.074 |
lee | -0.073 |
impressed | -0.067 |
glad | -0.065 |
omo | -0.065 |
Token | a |
Token ID | 257 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
Ready | +0.142 |
Left | +0.125 |
alis | +0.115 |
sc | +0.113 |
bust | +0.109 |
Neg logit contributions | |
lee | -0.135 |
atted | -0.134 |
impressed | -0.124 |
omo | -0.118 |
smiling | -0.116 |
Token | big |
Token ID | 1263 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
Ready | +0.061 |
Left | +0.053 |
alis | +0.048 |
sc | +0.047 |
bust | +0.045 |
Neg logit contributions | |
lee | -0.069 |
atted | -0.068 |
impressed | -0.063 |
omo | -0.061 |
smiling | -0.059 |
Token | magn |
Token ID | 7842 |
Feature activation | -3.52e-01 |
Pos logit contributions | |
lee | +0.021 |
atted | +0.021 |
impressed | +0.020 |
omo | +0.018 |
glad | +0.018 |
Neg logit contributions | |
Ready | -0.019 |
Left | -0.016 |
alis | -0.015 |
sc | -0.014 |
shortcut | -0.014 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.41e+01 |
Pos logit contributions | |
Ready | +0.008 |
Left | +0.006 |
alis | +0.005 |
ready | +0.004 |
sc | +0.004 |
Neg logit contributions | |
lee | -0.021 |
atted | -0.021 |
impressed | -0.020 |
smiling | -0.019 |
omo | -0.019 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Ready | +0.646 |
Left | +0.565 |
ready | +0.525 |
sc | +0.517 |
Skip | +0.499 |
Neg logit contributions | |
lee | -0.470 |
omo | -0.469 |
atted | -0.456 |
smiling | -0.449 |
mad | -0.448 |
Token | that |
Token ID | 326 |
Feature activation | -5.10e+00 |
Pos logit contributions | |
Ready | +0.075 |
Left | +0.068 |
alis | +0.064 |
sc | +0.062 |
bust | +0.061 |
Neg logit contributions | |
lee | -0.044 |
atted | -0.043 |
impressed | -0.039 |
omo | -0.036 |
smiling | -0.035 |
Token | made |
Token ID | 925 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
Ready | +0.235 |
Left | +0.211 |
alis | +0.196 |
sc | +0.192 |
bust | +0.185 |
Neg logit contributions | |
lee | -0.178 |
atted | -0.177 |
impressed | -0.162 |
omo | -0.152 |
smiling | -0.150 |
Token | things |
Token ID | 1243 |
Feature activation | -5.15e+00 |
Pos logit contributions | |
Ready | +0.152 |
Left | +0.138 |
alis | +0.129 |
sc | +0.127 |
bust | +0.123 |
Neg logit contributions | |
lee | -0.088 |
atted | -0.087 |
impressed | -0.078 |
omo | -0.073 |
smiling | -0.071 |
Token | look |
Token ID | 804 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
Ready | +0.188 |
Left | +0.163 |
alis | +0.148 |
sc | +0.144 |
bust | +0.138 |
Neg logit contributions | |
lee | -0.229 |
atted | -0.227 |
impressed | -0.213 |
omo | -0.203 |
smiling | -0.201 |
Token | He |
Token ID | 679 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
lee | +0.065 |
atted | +0.064 |
impressed | +0.059 |
smiling | +0.056 |
omo | +0.055 |
Neg logit contributions | |
Ready | -0.077 |
Left | -0.069 |
alis | -0.066 |
sc | -0.065 |
bust | -0.061 |
Token | took |
Token ID | 1718 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
atted | +0.014 |
lee | +0.014 |
glad | +0.013 |
impressed | +0.013 |
omo | +0.013 |
Neg logit contributions | |
Ready | -0.015 |
Left | -0.013 |
sc | -0.012 |
bust | -0.012 |
alis | -0.012 |
Token | out |
Token ID | 503 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
atted | +0.088 |
lee | +0.087 |
impressed | +0.079 |
mad | +0.076 |
smiling | +0.075 |
Neg logit contributions | |
Ready | -0.125 |
Left | -0.110 |
alis | -0.107 |
sc | -0.102 |
Three | -0.098 |
Token | a |
Token ID | 257 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
Ready | +0.057 |
Left | +0.052 |
bust | +0.047 |
alis | +0.047 |
sc | +0.046 |
Neg logit contributions | |
lee | -0.053 |
atted | -0.052 |
impressed | -0.049 |
omo | -0.048 |
glad | -0.045 |
Token | magn |
Token ID | 7842 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Ready | +0.059 |
Left | +0.055 |
alis | +0.049 |
sc | +0.046 |
Whe | +0.045 |
Neg logit contributions | |
lee | -0.065 |
atted | -0.063 |
impressed | -0.063 |
smiling | -0.060 |
mad | -0.059 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.40e+01 |
Pos logit contributions | |
Ready | +0.005 |
Left | +0.004 |
alis | +0.003 |
ready | +0.003 |
sc | +0.002 |
Neg logit contributions | |
atted | -0.013 |
lee | -0.013 |
impressed | -0.012 |
smiling | -0.011 |
omo | -0.011 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
Ready | +0.612 |
Left | +0.555 |
sc | +0.501 |
ready | +0.493 |
alis | +0.490 |
Neg logit contributions | |
lee | -0.487 |
atted | -0.477 |
omo | -0.472 |
smiling | -0.455 |
mad | -0.453 |
Token | and |
Token ID | 290 |
Feature activation | +6.62e+00 |
Pos logit contributions | |
Ready | +0.069 |
Left | +0.062 |
alis | +0.059 |
sc | +0.059 |
bust | +0.056 |
Neg logit contributions | |
lee | -0.035 |
atted | -0.034 |
impressed | -0.031 |
omo | -0.029 |
glad | -0.027 |
Token | a |
Token ID | 257 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
lee | +0.243 |
atted | +0.243 |
impressed | +0.221 |
omo | +0.215 |
glad | +0.205 |
Neg logit contributions | |
Ready | -0.295 |
Left | -0.257 |
alis | -0.240 |
sc | -0.237 |
bust | -0.231 |
Token | notebook |
Token ID | 20922 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
Ready | +0.054 |
Left | +0.049 |
alis | +0.044 |
sc | +0.043 |
bust | +0.041 |
Neg logit contributions | |
lee | -0.056 |
atted | -0.055 |
impressed | -0.053 |
smiling | -0.051 |
omo | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +7.14e+00 |
Pos logit contributions | |
Ready | +0.110 |
Left | +0.100 |
alis | +0.092 |
sc | +0.090 |
bust | +0.089 |
Neg logit contributions | |
lee | -0.066 |
atted | -0.065 |
impressed | -0.060 |
smiling | -0.056 |
omo | -0.055 |
Token | , |
Token ID | 11 |
Feature activation | +8.01e+00 |
Pos logit contributions | |
Ready | +0.164 |
Left | +0.148 |
sc | +0.147 |
alis | +0.145 |
tr | +0.139 |
Neg logit contributions | |
lee | -0.072 |
atted | -0.071 |
impressed | -0.062 |
omo | -0.060 |
mad | -0.055 |
Token | next |
Token ID | 1306 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
lee | +0.274 |
atted | +0.262 |
omo | +0.250 |
impressed | +0.226 |
mad | +0.206 |
Neg logit contributions | |
Ready | -0.371 |
Left | -0.327 |
alis | -0.316 |
south | -0.316 |
bust | -0.301 |
Token | to |
Token ID | 284 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
lee | +0.006 |
atted | +0.006 |
impressed | +0.005 |
omo | +0.005 |
smiling | +0.005 |
Neg logit contributions | |
Ready | -0.008 |
Left | -0.007 |
alis | -0.007 |
sc | -0.007 |
bust | -0.006 |
Token | her |
Token ID | 607 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
Ready | +0.168 |
Left | +0.154 |
alis | +0.145 |
sc | +0.144 |
bust | +0.138 |
Neg logit contributions | |
lee | -0.076 |
atted | -0.073 |
impressed | -0.066 |
smiling | -0.060 |
glad | -0.059 |
Token | magn |
Token ID | 7842 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
atted | +0.018 |
lee | +0.018 |
impressed | +0.016 |
omo | +0.015 |
mad | +0.014 |
Neg logit contributions | |
Ready | -0.032 |
Left | -0.029 |
alis | -0.028 |
sc | -0.027 |
ready | -0.026 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.39e+01 |
Pos logit contributions | |
Ready | +0.009 |
Left | +0.007 |
alis | +0.005 |
sc | +0.005 |
ready | +0.005 |
Neg logit contributions | |
lee | -0.029 |
atted | -0.029 |
impressed | -0.027 |
smiling | -0.026 |
omo | -0.026 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
Ready | +0.599 |
Left | +0.537 |
alis | +0.480 |
sc | +0.475 |
ready | +0.466 |
Neg logit contributions | |
lee | -0.521 |
atted | -0.513 |
omo | -0.492 |
smiling | -0.468 |
impressed | -0.467 |
Token | . |
Token ID | 13 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
Ready | +0.054 |
Left | +0.048 |
alis | +0.045 |
sc | +0.044 |
bust | +0.042 |
Neg logit contributions | |
lee | -0.047 |
atted | -0.047 |
impressed | -0.043 |
omo | -0.041 |
glad | -0.040 |
Token | She |
Token ID | 1375 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Ready | +0.011 |
Left | +0.010 |
alis | +0.009 |
sc | +0.009 |
bust | +0.009 |
Neg logit contributions | |
lee | -0.010 |
atted | -0.010 |
impressed | -0.009 |
omo | -0.008 |
smiling | -0.008 |
Token | looked |
Token ID | 3114 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
Ready | +0.073 |
Left | +0.064 |
alis | +0.060 |
sc | +0.059 |
bust | +0.059 |
Neg logit contributions | |
lee | -0.066 |
atted | -0.065 |
impressed | -0.061 |
omo | -0.058 |
mad | -0.057 |
Token | at |
Token ID | 379 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
omo | +0.129 |
lee | +0.127 |
atted | +0.125 |
impressed | +0.109 |
understanding | +0.103 |
Neg logit contributions | |
Ready | -0.138 |
Left | -0.125 |
sc | -0.118 |
south | -0.116 |
bust | -0.114 |
Token | a |
Token ID | 257 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
Ready | +0.162 |
Left | +0.142 |
alis | +0.134 |
sc | +0.130 |
summit | +0.127 |
Neg logit contributions | |
lee | -0.136 |
atted | -0.131 |
impressed | -0.125 |
smiling | -0.118 |
glad | -0.116 |
Token | flashlight |
Token ID | 38371 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
Ready | +0.067 |
Left | +0.058 |
alis | +0.053 |
sc | +0.052 |
bust | +0.049 |
Neg logit contributions | |
lee | -0.084 |
atted | -0.084 |
impressed | -0.079 |
omo | -0.075 |
smiling | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | +6.82e+00 |
Pos logit contributions | |
Ready | +0.001 |
Left | +0.001 |
alis | +0.001 |
sc | +0.001 |
bust | +0.001 |
Neg logit contributions | |
lee | -0.001 |
atted | -0.001 |
impressed | -0.001 |
omo | -0.001 |
mad | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
lee | +0.271 |
atted | +0.269 |
impressed | +0.249 |
omo | +0.237 |
smiling | +0.233 |
Neg logit contributions | |
Ready | -0.281 |
Left | -0.250 |
alis | -0.229 |
sc | -0.223 |
bust | -0.214 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.90e-01 |
Pos logit contributions | |
Ready | +0.044 |
Left | +0.038 |
alis | +0.035 |
sc | +0.034 |
bust | +0.032 |
Neg logit contributions | |
lee | -0.052 |
atted | -0.052 |
impressed | -0.049 |
omo | -0.047 |
smiling | -0.046 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.38e+01 |
Pos logit contributions | |
Ready | +0.006 |
Left | +0.004 |
ready | +0.004 |
Whe | +0.004 |
alis | +0.003 |
Neg logit contributions | |
lee | -0.010 |
atted | -0.010 |
smiling | -0.009 |
impressed | -0.009 |
omo | -0.009 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
Ready | +0.655 |
Left | +0.575 |
ready | +0.547 |
sc | +0.527 |
works | +0.515 |
Neg logit contributions | |
omo | -0.444 |
lee | -0.420 |
atted | -0.402 |
smiling | -0.401 |
mad | -0.399 |
Token | to |
Token ID | 284 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
Ready | +0.050 |
Left | +0.045 |
alis | +0.042 |
bust | +0.040 |
ready | +0.040 |
Neg logit contributions | |
lee | -0.034 |
atted | -0.034 |
impressed | -0.030 |
omo | -0.029 |
smiling | -0.028 |
Token | examine |
Token ID | 10716 |
Feature activation | +9.24e-01 |
Pos logit contributions | |
Ready | +0.146 |
Left | +0.130 |
alis | +0.111 |
sc | +0.110 |
ready | +0.108 |
Neg logit contributions | |
lee | -0.159 |
atted | -0.145 |
gigg | -0.141 |
understanding | -0.140 |
impressed | -0.138 |
Token | their |
Token ID | 511 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
lee | +0.030 |
atted | +0.029 |
impressed | +0.025 |
smiling | +0.024 |
mad | +0.023 |
Neg logit contributions | |
Ready | -0.046 |
Left | -0.041 |
alis | -0.039 |
Three | -0.037 |
sc | -0.037 |
Token | finds |
Token ID | 7228 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Ready | +0.052 |
Left | +0.046 |
alis | +0.043 |
sc | +0.042 |
bust | +0.040 |
Neg logit contributions | |
lee | -0.047 |
atted | -0.046 |
impressed | -0.042 |
omo | -0.041 |
smiling | -0.040 |
Token | hat |
Token ID | 6877 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
lee | +0.006 |
atted | +0.006 |
impressed | +0.006 |
omo | +0.005 |
smiling | +0.005 |
Neg logit contributions | |
Ready | -0.008 |
Left | -0.007 |
alis | -0.006 |
sc | -0.006 |
bust | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +6.84e+00 |
Pos logit contributions | |
Ready | +0.074 |
Left | +0.068 |
alis | +0.065 |
sc | +0.065 |
tr | +0.063 |
Neg logit contributions | |
lee | -0.027 |
atted | -0.026 |
impressed | -0.023 |
omo | -0.021 |
glad | -0.019 |
Token | grabbed |
Token ID | 13176 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
atted | +0.257 |
lee | +0.255 |
impressed | +0.246 |
glad | +0.225 |
omo | +0.222 |
Neg logit contributions | |
Ready | -0.289 |
Left | -0.249 |
sc | -0.235 |
alis | -0.231 |
bust | -0.230 |
Token | her |
Token ID | 607 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
lee | +0.090 |
atted | +0.090 |
impressed | +0.081 |
omo | +0.076 |
smiling | +0.074 |
Neg logit contributions | |
Ready | -0.138 |
Left | -0.124 |
alis | -0.116 |
sc | -0.114 |
bust | -0.111 |
Token | magn |
Token ID | 7842 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
Ready | +0.001 |
Left | +0.001 |
alis | +0.001 |
sc | +0.001 |
bust | +0.001 |
Neg logit contributions | |
lee | -0.001 |
atted | -0.001 |
impressed | -0.001 |
omo | -0.001 |
smiling | -0.001 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.38e+01 |
Pos logit contributions | |
Ready | +0.003 |
Left | +0.002 |
alis | +0.002 |
sc | +0.002 |
bust | +0.001 |
Neg logit contributions | |
lee | -0.012 |
atted | -0.012 |
impressed | -0.012 |
omo | -0.011 |
smiling | -0.011 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
Ready | +0.578 |
Left | +0.509 |
alis | +0.457 |
ready | +0.445 |
sc | +0.444 |
Neg logit contributions | |
lee | -0.544 |
atted | -0.538 |
omo | -0.498 |
impressed | -0.484 |
smiling | -0.477 |
Token | . |
Token ID | 13 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
Ready | +0.070 |
Left | +0.063 |
alis | +0.059 |
sc | +0.059 |
bust | +0.056 |
Neg logit contributions | |
lee | -0.052 |
atted | -0.051 |
impressed | -0.048 |
omo | -0.044 |
glad | -0.044 |
Token | She |
Token ID | 1375 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
Ready | +0.010 |
Left | +0.009 |
alis | +0.008 |
sc | +0.008 |
bust | +0.008 |
Neg logit contributions | |
lee | -0.009 |
atted | -0.009 |
impressed | -0.008 |
omo | -0.008 |
smiling | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
Ready | +0.083 |
Left | +0.072 |
sc | +0.070 |
tr | +0.067 |
alis | +0.067 |
Neg logit contributions | |
lee | -0.082 |
atted | -0.081 |
impressed | -0.074 |
glad | -0.072 |
mad | -0.068 |
Token | ready |
Token ID | 3492 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
atted | +0.094 |
lee | +0.093 |
omo | +0.084 |
ummy | +0.072 |
ummies | +0.071 |
Neg logit contributions | |
Ready | -0.117 |
Left | -0.106 |
alis | -0.104 |
highest | -0.100 |
south | -0.098 |
Token | said |
Token ID | 531 |
Feature activation | +7.68e+00 |
Pos logit contributions | |
atted | +0.260 |
lee | +0.247 |
glad | +0.238 |
omo | +0.225 |
impressed | +0.222 |
Neg logit contributions | |
Ready | -0.286 |
sc | -0.259 |
Left | -0.247 |
alis | -0.240 |
hunt | -0.238 |
Token | , |
Token ID | 11 |
Feature activation | +1.07e+01 |
Pos logit contributions | |
omo | +0.316 |
lee | +0.304 |
atted | +0.290 |
impressed | +0.277 |
glad | +0.251 |
Neg logit contributions | |
Ready | -0.250 |
Left | -0.240 |
sc | -0.230 |
fire | -0.219 |
tted | -0.216 |
Token | holding |
Token ID | 4769 |
Feature activation | +3.12e+00 |
Pos logit contributions | |
atted | +0.381 |
omo | +0.372 |
lee | +0.371 |
aj | +0.303 |
ummy | +0.301 |
Neg logit contributions | |
Ready | -0.434 |
Left | -0.392 |
flap | -0.391 |
alis | -0.379 |
gra | -0.376 |
Token | the |
Token ID | 262 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
lee | +0.124 |
atted | +0.123 |
impressed | +0.116 |
omo | +0.112 |
smiling | +0.106 |
Neg logit contributions | |
Ready | -0.129 |
Left | -0.115 |
alis | -0.104 |
sc | -0.102 |
bust | -0.099 |
Token | magn |
Token ID | 7842 |
Feature activation | -3.74e-02 |
Pos logit contributions | |
Ready | +0.156 |
Left | +0.142 |
alis | +0.128 |
sc | +0.127 |
bust | +0.124 |
Neg logit contributions | |
lee | -0.133 |
atted | -0.130 |
impressed | -0.122 |
smiling | -0.118 |
mad | -0.117 |
Token | ifying |
Token ID | 4035 |
Feature activation | -1.37e+01 |
Pos logit contributions | |
Ready | +0.001 |
Left | +0.001 |
alis | +0.000 |
ready | +0.000 |
bust | +0.000 |
Neg logit contributions | |
lee | -0.002 |
atted | -0.002 |
impressed | -0.002 |
smiling | -0.002 |
omo | -0.002 |
Token | glass |
Token ID | 5405 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
Ready | +0.624 |
Left | +0.561 |
sc | +0.502 |
ready | +0.501 |
alis | +0.500 |
Neg logit contributions | |
lee | -0.471 |
atted | -0.458 |
omo | -0.441 |
smiling | -0.431 |
mad | -0.429 |
Token | over |
Token ID | 625 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
Ready | +0.049 |
Left | +0.044 |
sc | +0.042 |
alis | +0.041 |
summit | +0.039 |
Neg logit contributions | |
lee | -0.038 |
atted | -0.038 |
impressed | -0.035 |
omo | -0.033 |
glad | -0.032 |
Token | a |
Token ID | 257 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
Ready | +0.082 |
Left | +0.072 |
south | +0.062 |
ready | +0.061 |
sc | +0.060 |
Neg logit contributions | |
atted | -0.088 |
lee | -0.086 |
impressed | -0.084 |
omo | -0.082 |
glad | -0.076 |
Token | black |
Token ID | 2042 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
Ready | +0.096 |
Left | +0.087 |
alis | +0.079 |
bust | +0.074 |
sc | +0.073 |
Neg logit contributions | |
lee | -0.106 |
impressed | -0.102 |
atted | -0.102 |
smiling | -0.098 |
omo | -0.095 |
Token | ant |
Token ID | 1885 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
lee | +0.021 |
atted | +0.020 |
smiling | +0.019 |
omo | +0.018 |
understanding | +0.018 |
Neg logit contributions | |
Ready | -0.021 |
Left | -0.018 |
bust | -0.016 |
alis | -0.016 |
ready | -0.016 |
Token | Go |
Token ID | 1514 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
Ready | +0.049 |
Left | +0.043 |
alis | +0.038 |
sc | +0.037 |
bust | +0.036 |
Neg logit contributions | |
lee | -0.064 |
atted | -0.064 |
impressed | -0.060 |
omo | -0.057 |
smiling | -0.056 |
Token | away |
Token ID | 1497 |
Feature activation | -8.55e-01 |
Pos logit contributions | |
atted | +0.026 |
lee | +0.026 |
impressed | +0.024 |
mad | +0.024 |
smiling | +0.024 |
Neg logit contributions | |
Ready | -0.016 |
Left | -0.014 |
alis | -0.013 |
bust | -0.012 |
summit | -0.012 |
Token | ." |
Token ID | 526 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
Ready | +0.039 |
Left | +0.035 |
sc | +0.032 |
alis | +0.032 |
south | +0.031 |
Neg logit contributions | |
lee | -0.030 |
atted | -0.030 |
impressed | -0.027 |
omo | -0.026 |
glad | -0.025 |
Token |
|
Token ID | 198 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
lee | +0.042 |
atted | +0.042 |
impressed | +0.039 |
smiling | +0.037 |
omo | +0.036 |
Neg logit contributions | |
Ready | -0.029 |
Left | -0.025 |
sc | -0.021 |
Three | -0.020 |
alis | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
Ready | +0.093 |
Left | +0.081 |
alis | +0.070 |
sc | +0.070 |
Three | +0.067 |
Neg logit contributions | |
lee | -0.117 |
atted | -0.116 |
impressed | -0.108 |
omo | -0.102 |
smiling | -0.102 |
Token | L |
Token ID | 43 |
Feature activation | +6.90e+00 |
Pos logit contributions | |
Ready | +0.106 |
Left | +0.094 |
Three | +0.081 |
sc | +0.081 |
Skip | +0.079 |
Neg logit contributions | |
lee | -0.101 |
atted | -0.100 |
impressed | -0.088 |
smiling | -0.084 |
glad | -0.081 |
Token | ily |
Token ID | 813 |
Feature activation | +5.57e+00 |
Pos logit contributions | |
lee | +0.280 |
atted | +0.278 |
impressed | +0.262 |
smiling | +0.245 |
omo | +0.245 |
Neg logit contributions | |
Ready | -0.280 |
Left | -0.247 |
alis | -0.224 |
sc | -0.219 |
Three | -0.211 |
Token | ran |
Token ID | 4966 |
Feature activation | +5.96e+00 |
Pos logit contributions | |
lee | +0.217 |
atted | +0.202 |
impressed | +0.198 |
glad | +0.195 |
omo | +0.188 |
Neg logit contributions | |
Ready | -0.219 |
Left | -0.201 |
sc | -0.189 |
bust | -0.177 |
alis | -0.173 |
Token | to |
Token ID | 284 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
lee | +0.265 |
atted | +0.259 |
impressed | +0.242 |
omo | +0.239 |
smiling | +0.220 |
Neg logit contributions | |
Ready | -0.236 |
Left | -0.208 |
sc | -0.187 |
alis | -0.173 |
ready | -0.172 |
Token | mom |
Token ID | 1995 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
Ready | +0.083 |
Left | +0.075 |
alis | +0.069 |
sc | +0.068 |
bust | +0.066 |
Neg logit contributions | |
lee | -0.053 |
atted | -0.052 |
impressed | -0.047 |
omo | -0.045 |
smiling | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +7.85e+00 |
Pos logit contributions | |
lee | +0.156 |
atted | +0.153 |
impressed | +0.153 |
omo | +0.152 |
pleased | +0.135 |
Neg logit contributions | |
Ready | -0.219 |
sc | -0.198 |
Left | -0.194 |
hunt | -0.187 |
south | -0.183 |
Token | in |
Token ID | 287 |
Feature activation | -6.08e+00 |
Pos logit contributions | |
Ready | +0.098 |
Left | +0.088 |
alis | +0.083 |
sc | +0.080 |
bust | +0.079 |
Neg logit contributions | |
lee | -0.061 |
atted | -0.061 |
impressed | -0.056 |
omo | -0.052 |
smiling | -0.052 |
Token | the |
Token ID | 262 |
Feature activation | -5.60e+00 |
Pos logit contributions | |
Ready | +0.262 |
Left | +0.232 |
alis | +0.213 |
sc | +0.209 |
bust | +0.202 |
Neg logit contributions | |
atted | -0.232 |
lee | -0.232 |
impressed | -0.212 |
omo | -0.202 |
smiling | -0.196 |
Token | garden |
Token ID | 11376 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
Ready | +0.231 |
Left | +0.211 |
sc | +0.201 |
alis | +0.199 |
bust | +0.186 |
Neg logit contributions | |
lee | -0.203 |
atted | -0.196 |
smiling | -0.193 |
omo | -0.188 |
mad | -0.187 |
Token | ," |
Token ID | 553 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
Ready | +0.211 |
Left | +0.190 |
alis | +0.177 |
sc | +0.175 |
bust | +0.168 |
Neg logit contributions | |
lee | -0.144 |
atted | -0.143 |
impressed | -0.129 |
omo | -0.122 |
smiling | -0.118 |
Token | Lily |
Token ID | 20037 |
Feature activation | +6.80e+00 |
Pos logit contributions | |
atted | +0.095 |
lee | +0.093 |
omo | +0.083 |
impressed | +0.081 |
rou | +0.078 |
Neg logit contributions | |
Ready | -0.093 |
Left | -0.076 |
am | -0.069 |
sc | -0.069 |
ready | -0.068 |
Token | said |
Token ID | 531 |
Feature activation | +7.21e+00 |
Pos logit contributions | |
atted | +0.271 |
lee | +0.259 |
irling | +0.245 |
glad | +0.243 |
impressed | +0.241 |
Neg logit contributions | |
Ready | -0.226 |
Left | -0.209 |
sc | -0.207 |
will | -0.198 |
hunt | -0.195 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
omo | +0.350 |
lee | +0.333 |
atted | +0.315 |
impressed | +0.310 |
glad | +0.277 |
Neg logit contributions | |
Ready | -0.194 |
Left | -0.189 |
sc | -0.178 |
flap | -0.159 |
fire | -0.158 |
Token |
|
Token ID | 198 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
atted | +0.065 |
lee | +0.065 |
impressed | +0.060 |
omo | +0.059 |
mad | +0.056 |
Neg logit contributions | |
Ready | -0.043 |
Left | -0.037 |
sc | -0.031 |
Three | -0.029 |
alis | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Ready | +0.067 |
Left | +0.056 |
Three | +0.044 |
Ready | +0.043 |
sc | +0.043 |
Neg logit contributions | |
atted | -0.122 |
lee | -0.121 |
impressed | -0.112 |
omo | -0.109 |
smiling | -0.107 |
Token | " |
Token ID | 1 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
Ready | +0.088 |
Left | +0.075 |
Three | +0.066 |
south | +0.065 |
Ready | +0.063 |
Neg logit contributions | |
lee | -0.101 |
atted | -0.100 |
impressed | -0.087 |
omo | -0.082 |
smiling | -0.081 |
Token | We |
Token ID | 1135 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
Ready | +0.061 |
Left | +0.053 |
alis | +0.050 |
sc | +0.048 |
summit | +0.046 |
Neg logit contributions | |
atted | -0.076 |
lee | -0.076 |
impressed | -0.074 |
omo | -0.070 |
smiling | -0.069 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.86e+00 |
Pos logit contributions | |
lee | +0.261 |
atted | +0.260 |
impressed | +0.244 |
omo | +0.233 |
glad | +0.226 |
Neg logit contributions | |
Ready | -0.215 |
Left | -0.184 |
alis | -0.169 |
sc | -0.164 |
bust | -0.159 |
Token | heard |
Token ID | 2982 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
lee | +0.160 |
atted | +0.157 |
impressed | +0.147 |
omo | +0.143 |
mad | +0.142 |
Neg logit contributions | |
Ready | -0.132 |
Left | -0.112 |
will | -0.109 |
bust | -0.107 |
am | -0.105 |
Token | their |
Token ID | 511 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
atted | +0.061 |
lee | +0.059 |
impressed | +0.054 |
omo | +0.053 |
glad | +0.049 |
Neg logit contributions | |
Ready | -0.092 |
Left | -0.083 |
alis | -0.077 |
bust | -0.075 |
sc | -0.075 |
Token | words |
Token ID | 2456 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
atted | +0.058 |
lee | +0.058 |
impressed | +0.053 |
omo | +0.052 |
glad | +0.049 |
Neg logit contributions | |
Ready | -0.054 |
Left | -0.048 |
alis | -0.043 |
sc | -0.042 |
summit | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | +7.67e+00 |
Pos logit contributions | |
Ready | +0.080 |
Left | +0.073 |
alis | +0.069 |
sc | +0.068 |
ready | +0.066 |
Neg logit contributions | |
lee | -0.035 |
atted | -0.034 |
impressed | -0.031 |
smiling | -0.028 |
omo | -0.027 |
Token | believed |
Token ID | 4762 |
Feature activation | +6.79e+00 |
Pos logit contributions | |
atted | +0.254 |
lee | +0.253 |
omo | +0.243 |
impressed | +0.221 |
glad | +0.212 |
Neg logit contributions | |
Ready | -0.345 |
Left | -0.295 |
bust | -0.290 |
alis | -0.276 |
sc | -0.273 |
Token | them |
Token ID | 606 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
atted | +0.224 |
lee | +0.223 |
omo | +0.204 |
impressed | +0.202 |
smiling | +0.182 |
Neg logit contributions | |
Ready | -0.316 |
Left | -0.287 |
alis | -0.263 |
sc | -0.262 |
bust | -0.254 |
Token | . |
Token ID | 13 |
Feature activation | +2.51e-01 |
Pos logit contributions | |
Ready | +0.117 |
Left | +0.105 |
alis | +0.100 |
sc | +0.096 |
Three | +0.095 |
Neg logit contributions | |
lee | -0.066 |
atted | -0.064 |
impressed | -0.059 |
smiling | -0.055 |
mad | -0.054 |
Token | She |
Token ID | 1375 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
lee | +0.010 |
atted | +0.010 |
impressed | +0.009 |
omo | +0.009 |
smiling | +0.009 |
Neg logit contributions | |
Ready | -0.010 |
Left | -0.009 |
alis | -0.008 |
sc | -0.008 |
bust | -0.008 |
Token | knew |
Token ID | 2993 |
Feature activation | +8.12e+00 |
Pos logit contributions | |
Ready | +0.049 |
Left | +0.042 |
bust | +0.040 |
alis | +0.039 |
sc | +0.039 |
Neg logit contributions | |
atted | -0.046 |
lee | -0.044 |
omo | -0.040 |
impressed | -0.040 |
glad | -0.040 |
Token | that |
Token ID | 326 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
atted | +0.286 |
lee | +0.285 |
omo | +0.257 |
impressed | +0.253 |
mad | +0.235 |
Neg logit contributions | |
Ready | -0.359 |
Left | -0.316 |
alis | -0.290 |
ready | -0.284 |
summit | -0.282 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.35e+00 |
Pos logit contributions | |
Ready | +0.019 |
Left | +0.017 |
alis | +0.016 |
sc | +0.016 |
bust | +0.015 |
Neg logit contributions | |
lee | -0.017 |
atted | -0.017 |
impressed | -0.016 |
omo | -0.015 |
glad | -0.015 |
Token | Once |
Token ID | 7454 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
Ready | +0.357 |
alis | +0.329 |
bust | +0.314 |
summit | +0.312 |
Left | +0.305 |
Neg logit contributions | |
lee | -0.205 |
atted | -0.194 |
smiling | -0.188 |
omo | -0.178 |
glad | -0.170 |
Token | upon |
Token ID | 2402 |
Feature activation | -5.47e+00 |
Pos logit contributions | |
Ready | +0.012 |
Left | +0.011 |
sc | +0.010 |
alis | +0.010 |
summit | +0.009 |
Neg logit contributions | |
lee | -0.015 |
atted | -0.015 |
impressed | -0.014 |
omo | -0.014 |
smiling | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
Ready | +0.254 |
Left | +0.222 |
alis | +0.217 |
sc | +0.209 |
bust | +0.206 |
Neg logit contributions | |
lee | -0.177 |
atted | -0.175 |
impressed | -0.174 |
glad | -0.164 |
smiling | -0.163 |
Token | time |
Token ID | 640 |
Feature activation | -7.04e-01 |
Pos logit contributions | |
Ready | +0.012 |
Left | +0.010 |
alis | +0.009 |
sc | +0.008 |
Three | +0.008 |
Neg logit contributions | |
lee | -0.025 |
atted | -0.025 |
impressed | -0.024 |
smiling | -0.022 |
omo | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +8.25e+00 |
Pos logit contributions | |
Ready | +0.034 |
Left | +0.033 |
am | +0.030 |
sc | +0.029 |
shortcut | +0.028 |
Neg logit contributions | |
lee | -0.022 |
atted | -0.020 |
impressed | -0.018 |
omo | -0.017 |
understanding | -0.016 |
Token | there |
Token ID | 612 |
Feature activation | -7.97e+00 |
Pos logit contributions | |
atted | +0.246 |
lee | +0.236 |
impressed | +0.228 |
glad | +0.216 |
mad | +0.209 |
Neg logit contributions | |
Ready | -0.420 |
Left | -0.371 |
alis | -0.360 |
sc | -0.343 |
summit | -0.340 |
Token | was |
Token ID | 373 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
Ready | +0.335 |
Left | +0.297 |
alis | +0.271 |
sc | +0.267 |
bust | +0.257 |
Neg logit contributions | |
lee | -0.311 |
atted | -0.308 |
impressed | -0.285 |
omo | -0.270 |
smiling | -0.264 |
Token | a |
Token ID | 257 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
atted | +0.167 |
lee | +0.166 |
impressed | +0.151 |
omo | +0.145 |
smiling | +0.141 |
Neg logit contributions | |
Ready | -0.122 |
Left | -0.106 |
alis | -0.094 |
sc | -0.091 |
ready | -0.087 |
Token | polite |
Token ID | 23507 |
Feature activation | +5.62e+00 |
Pos logit contributions | |
Ready | +0.008 |
Left | +0.007 |
alis | +0.006 |
bust | +0.006 |
Three | +0.006 |
Neg logit contributions | |
atted | -0.007 |
lee | -0.007 |
impressed | -0.006 |
omo | -0.006 |
smiling | -0.006 |
Token | little |
Token ID | 1310 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
atted | +0.265 |
lee | +0.255 |
impressed | +0.240 |
mad | +0.222 |
omo | +0.221 |
Neg logit contributions | |
Ready | -0.189 |
Left | -0.164 |
alis | -0.150 |
bust | -0.141 |
Skip | -0.138 |
Token | andy |
Token ID | 10757 |
Feature activation | +6.85e+00 |
Pos logit contributions | |
impressed | +0.186 |
smiling | +0.174 |
glad | +0.173 |
atted | +0.170 |
hugging | +0.170 |
Neg logit contributions | |
Ready | -0.235 |
Left | -0.208 |
summit | -0.205 |
Three | -0.203 |
alis | -0.198 |
Token | opened |
Token ID | 4721 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
atted | +0.240 |
lee | +0.239 |
impressed | +0.229 |
glad | +0.227 |
omo | +0.220 |
Neg logit contributions | |
Ready | -0.280 |
Left | -0.249 |
sc | -0.243 |
bust | -0.239 |
alis | -0.235 |
Token | the |
Token ID | 262 |
Feature activation | -5.54e+00 |
Pos logit contributions | |
atted | +0.075 |
lee | +0.075 |
impressed | +0.072 |
omo | +0.069 |
glad | +0.067 |
Neg logit contributions | |
Ready | -0.060 |
Left | -0.053 |
south | -0.047 |
summit | -0.047 |
alis | -0.047 |
Token | cup |
Token ID | 6508 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Ready | +0.254 |
Left | +0.233 |
alis | +0.213 |
Skip | +0.207 |
Whe | +0.205 |
Neg logit contributions | |
lee | -0.192 |
atted | -0.187 |
impressed | -0.175 |
smiling | -0.174 |
mad | -0.170 |
Token | board |
Token ID | 3526 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
Ready | +0.033 |
Left | +0.028 |
alis | +0.024 |
sc | +0.024 |
bust | +0.022 |
Neg logit contributions | |
lee | -0.057 |
atted | -0.057 |
impressed | -0.054 |
omo | -0.052 |
glad | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | +6.16e+00 |
Pos logit contributions | |
Ready | +0.107 |
sc | +0.097 |
Left | +0.097 |
alis | +0.093 |
summit | +0.090 |
Neg logit contributions | |
lee | -0.044 |
atted | -0.044 |
impressed | -0.038 |
omo | -0.038 |
glad | -0.034 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
lee | +0.246 |
atted | +0.246 |
omo | +0.231 |
impressed | +0.228 |
glad | +0.213 |
Neg logit contributions | |
Ready | -0.253 |
Left | -0.214 |
bust | -0.200 |
sc | -0.198 |
ready | -0.189 |
Token | a |
Token ID | 257 |
Feature activation | -2.42e+00 |
Pos logit contributions | |
atted | +0.017 |
lee | +0.017 |
impressed | +0.016 |
omo | +0.016 |
mad | +0.015 |
Neg logit contributions | |
Ready | -0.013 |
Left | -0.011 |
alis | -0.010 |
sc | -0.010 |
summit | -0.010 |
Token | big |
Token ID | 1263 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
Ready | +0.095 |
Left | +0.081 |
sc | +0.073 |
alis | +0.073 |
bust | +0.070 |
Neg logit contributions | |
atted | -0.105 |
lee | -0.103 |
impressed | -0.091 |
omo | -0.088 |
glad | -0.085 |
Token | , |
Token ID | 11 |
Feature activation | +7.23e+00 |
Pos logit contributions | |
Ready | +0.009 |
Left | +0.007 |
alis | +0.007 |
hunt | +0.007 |
shortcut | +0.007 |
Neg logit contributions | |
atted | -0.010 |
lee | -0.010 |
impressed | -0.009 |
omo | -0.008 |
aged | -0.008 |
Token | fat |
Token ID | 3735 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
atted | +0.296 |
lee | +0.284 |
impressed | +0.245 |
omo | +0.245 |
glad | +0.227 |
Neg logit contributions | |
Ready | -0.305 |
Left | -0.260 |
sc | -0.244 |
alis | -0.240 |
bust | -0.236 |
Token | ran |
Token ID | 4966 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
Ready | +0.040 |
Left | +0.036 |
alis | +0.033 |
sc | +0.033 |
bust | +0.032 |
Neg logit contributions | |
lee | -0.031 |
atted | -0.030 |
impressed | -0.028 |
omo | -0.027 |
smiling | -0.026 |
Token | after |
Token ID | 706 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
lee | +0.147 |
atted | +0.147 |
impressed | +0.138 |
omo | +0.135 |
glad | +0.129 |
Neg logit contributions | |
Ready | -0.096 |
Left | -0.082 |
sc | -0.073 |
alis | -0.069 |
ready | -0.065 |
Token | it |
Token ID | 340 |
Feature activation | -7.62e+00 |
Pos logit contributions | |
Ready | +0.175 |
Left | +0.158 |
alis | +0.145 |
sc | +0.145 |
ready | +0.139 |
Neg logit contributions | |
atted | -0.117 |
lee | -0.116 |
impressed | -0.106 |
omo | -0.102 |
glad | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
Ready | +0.352 |
Left | +0.317 |
sc | +0.297 |
alis | +0.292 |
south | +0.276 |
Neg logit contributions | |
lee | -0.263 |
atted | -0.259 |
impressed | -0.238 |
omo | -0.231 |
glad | -0.217 |
Token | It |
Token ID | 632 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
Ready | +0.061 |
Left | +0.053 |
alis | +0.049 |
sc | +0.048 |
bust | +0.046 |
Neg logit contributions | |
lee | -0.062 |
atted | -0.061 |
impressed | -0.057 |
omo | -0.054 |
smiling | -0.053 |
Token | caught |
Token ID | 4978 |
Feature activation | +4.29e+00 |
Pos logit contributions | |
Ready | +0.291 |
Left | +0.259 |
alis | +0.242 |
sc | +0.233 |
ready | +0.229 |
Neg logit contributions | |
atted | -0.233 |
lee | -0.233 |
impressed | -0.216 |
omo | -0.203 |
smiling | -0.200 |
Token | the |
Token ID | 262 |
Feature activation | -5.35e+00 |
Pos logit contributions | |
lee | +0.160 |
atted | +0.159 |
impressed | +0.149 |
omo | +0.142 |
glad | +0.135 |
Neg logit contributions | |
Ready | -0.188 |
Left | -0.169 |
alis | -0.153 |
sc | -0.151 |
ready | -0.144 |
Token | ball |
Token ID | 2613 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
Ready | +0.199 |
Left | +0.177 |
alis | +0.160 |
sc | +0.155 |
bust | +0.152 |
Neg logit contributions | |
lee | -0.227 |
atted | -0.225 |
impressed | -0.213 |
omo | -0.207 |
smiling | -0.205 |
Token | in |
Token ID | 287 |
Feature activation | -6.39e+00 |
Pos logit contributions | |
Ready | +0.093 |
Left | +0.085 |
sc | +0.084 |
alis | +0.081 |
highest | +0.079 |
Neg logit contributions | |
lee | -0.048 |
atted | -0.046 |
impressed | -0.040 |
omo | -0.039 |
glad | -0.035 |
Token | its |
Token ID | 663 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
Ready | +0.269 |
Left | +0.237 |
alis | +0.213 |
sc | +0.210 |
Three | +0.203 |
Neg logit contributions | |
atted | -0.261 |
lee | -0.254 |
impressed | -0.232 |
omo | -0.216 |
glad | -0.211 |
Token | mouth |
Token ID | 5422 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
Ready | +0.100 |
Left | +0.090 |
alis | +0.083 |
sc | +0.083 |
bust | +0.081 |
Neg logit contributions | |
lee | -0.063 |
atted | -0.063 |
impressed | -0.056 |
omo | -0.054 |
smiling | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +7.19e+00 |
Pos logit contributions | |
lee | +0.007 |
atted | +0.007 |
impressed | +0.007 |
mad | +0.006 |
pleased | +0.006 |
Neg logit contributions | |
Ready | -0.011 |
Left | -0.010 |
sc | -0.010 |
will | -0.009 |
am | -0.009 |
Token | they |
Token ID | 484 |
Feature activation | -3.34e+00 |
Pos logit contributions | |
atted | +0.293 |
lee | +0.292 |
impressed | +0.263 |
omo | +0.250 |
mad | +0.249 |
Neg logit contributions | |
Ready | -0.287 |
Left | -0.252 |
alis | -0.224 |
sc | -0.224 |
Skip | -0.221 |
Token | did |
Token ID | 750 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
Ready | +0.157 |
Left | +0.139 |
alis | +0.131 |
sc | +0.128 |
ready | +0.125 |
Neg logit contributions | |
lee | -0.117 |
atted | -0.114 |
impressed | -0.105 |
omo | -0.098 |
smiling | -0.098 |
Token | everything |
Token ID | 2279 |
Feature activation | -7.74e+00 |
Pos logit contributions | |
Ready | +0.048 |
Left | +0.043 |
alis | +0.040 |
sc | +0.039 |
bust | +0.038 |
Neg logit contributions | |
lee | -0.038 |
atted | -0.038 |
impressed | -0.035 |
omo | -0.033 |
smiling | -0.032 |
Token | together |
Token ID | 1978 |
Feature activation | -6.37e-01 |
Pos logit contributions | |
Ready | +0.377 |
Left | +0.345 |
sc | +0.325 |
alis | +0.316 |
bust | +0.308 |
Neg logit contributions | |
lee | -0.246 |
atted | -0.242 |
omo | -0.213 |
impressed | -0.207 |
understanding | -0.186 |
Token | . |
Token ID | 13 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
Ready | +0.034 |
Left | +0.030 |
alis | +0.029 |
bust | +0.028 |
sc | +0.028 |
Neg logit contributions | |
lee | -0.018 |
atted | -0.017 |
impressed | -0.016 |
smiling | -0.015 |
mad | -0.014 |
Token | One |
Token ID | 1881 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
Ready | +0.019 |
Left | +0.016 |
summit | +0.016 |
bust | +0.016 |
alis | +0.016 |
Neg logit contributions | |
lee | -0.015 |
omo | -0.015 |
atted | -0.014 |
impressed | -0.013 |
glad | -0.013 |
Token | day |
Token ID | 1110 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Ready | +0.110 |
Left | +0.093 |
alis | +0.081 |
sc | +0.079 |
bust | +0.074 |
Neg logit contributions | |
lee | -0.191 |
atted | -0.190 |
impressed | -0.178 |
omo | -0.170 |
smiling | -0.168 |
Token | , |
Token ID | 11 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
Ready | +0.012 |
Left | +0.011 |
sc | +0.011 |
alis | +0.010 |
ready | +0.010 |
Neg logit contributions | |
lee | -0.005 |
atted | -0.005 |
impressed | -0.005 |
omo | -0.004 |
mad | -0.004 |
Token | Lily |
Token ID | 20037 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
lee | +0.256 |
atted | +0.255 |
impressed | +0.230 |
omo | +0.215 |
smiling | +0.209 |
Neg logit contributions | |
Ready | -0.434 |
Left | -0.393 |
alis | -0.368 |
sc | -0.362 |
bust | -0.352 |
Token | went |
Token ID | 1816 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
atted | +0.210 |
impressed | +0.193 |
lee | +0.193 |
understanding | +0.191 |
glad | +0.187 |
Neg logit contributions | |
Ready | -0.150 |
hunt | -0.125 |
sc | -0.121 |
Left | -0.118 |
alis | -0.115 |
Token | . |
Token ID | 13 |
Feature activation | +9.08e-02 |
Pos logit contributions | |
lee | +0.014 |
atted | +0.014 |
impressed | +0.012 |
smiling | +0.011 |
omo | +0.011 |
Neg logit contributions | |
Ready | -0.023 |
Left | -0.021 |
alis | -0.019 |
sc | -0.019 |
bust | -0.019 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.29e+00 |
Pos logit contributions | |
lee | +0.003 |
atted | +0.003 |
impressed | +0.003 |
glad | +0.003 |
omo | +0.003 |
Neg logit contributions | |
Ready | -0.004 |
Left | -0.003 |
alis | -0.003 |
sc | -0.003 |
bust | -0.003 |
Token | Sam |
Token ID | 16305 |
Feature activation | +4.86e+00 |
Pos logit contributions | |
Ready | +0.246 |
alis | +0.225 |
bust | +0.211 |
summit | +0.211 |
Left | +0.196 |
Neg logit contributions | |
lee | -0.296 |
atted | -0.296 |
smiling | -0.294 |
omo | -0.272 |
and | -0.270 |
Token | and |
Token ID | 290 |
Feature activation | +8.16e+00 |
Pos logit contributions | |
impressed | +0.175 |
atted | +0.175 |
lee | +0.171 |
understanding | +0.157 |
pleased | +0.155 |
Neg logit contributions | |
Ready | -0.211 |
Left | -0.182 |
am | -0.168 |
alis | -0.166 |
sc | -0.166 |
Token | Mia |
Token ID | 32189 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
atted | +0.278 |
lee | +0.277 |
impressed | +0.253 |
omo | +0.242 |
smiling | +0.224 |
Neg logit contributions | |
Ready | -0.385 |
Left | -0.345 |
bust | -0.314 |
alis | -0.309 |
sc | -0.306 |
Token | were |
Token ID | 547 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
lee | +0.175 |
atted | +0.175 |
impressed | +0.166 |
omo | +0.157 |
glad | +0.152 |
Neg logit contributions | |
Ready | -0.136 |
Left | -0.116 |
alis | -0.101 |
bust | -0.101 |
sc | -0.101 |
Token | playing |
Token ID | 2712 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
lee | +0.088 |
atted | +0.084 |
omo | +0.079 |
impressed | +0.076 |
ushed | +0.074 |
Neg logit contributions | |
Ready | -0.110 |
Left | -0.101 |
same | -0.100 |
middle | -0.100 |
alis | -0.099 |
Token | on |
Token ID | 319 |
Feature activation | -4.17e+00 |
Pos logit contributions | |
Ready | +0.036 |
Left | +0.032 |
alis | +0.030 |
bust | +0.030 |
summit | +0.029 |
Neg logit contributions | |
atted | -0.021 |
lee | -0.020 |
impressed | -0.019 |
smiling | -0.018 |
mad | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -5.05e+00 |
Pos logit contributions | |
Ready | +0.191 |
Left | +0.170 |
alis | +0.158 |
sc | +0.156 |
bust | +0.151 |
Neg logit contributions | |
lee | -0.148 |
atted | -0.145 |
impressed | -0.134 |
omo | -0.127 |
smiling | -0.126 |
Token | beach |
Token ID | 10481 |
Feature activation | +7.67e-01 |
Pos logit contributions | |
Ready | +0.180 |
Left | +0.160 |
alis | +0.142 |
sc | +0.138 |
bust | +0.134 |
Neg logit contributions | |
lee | -0.226 |
atted | -0.220 |
impressed | -0.210 |
omo | -0.201 |
smiling | -0.200 |
Token | . |
Token ID | 13 |
Feature activation | -9.95e-01 |
Pos logit contributions | |
lee | +0.024 |
atted | +0.024 |
impressed | +0.022 |
omo | +0.020 |
smiling | +0.020 |
Neg logit contributions | |
Ready | -0.038 |
Left | -0.034 |
alis | -0.032 |
sc | -0.031 |
bust | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -3.80e+00 |
Pos logit contributions | |
Ready | +0.175 |
Left | +0.152 |
alis | +0.146 |
sc | +0.138 |
bust | +0.137 |
Neg logit contributions | |
lee | -0.157 |
atted | -0.155 |
impressed | -0.148 |
omo | -0.139 |
glad | -0.137 |
Token | As |
Token ID | 1722 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
Ready | +0.154 |
Left | +0.136 |
alis | +0.131 |
bust | +0.123 |
sc | +0.121 |
Neg logit contributions | |
atted | -0.146 |
lee | -0.146 |
impressed | -0.141 |
omo | -0.137 |
mad | -0.128 |
Token | Tim |
Token ID | 5045 |
Feature activation | +4.61e+00 |
Pos logit contributions | |
atted | +0.027 |
lee | +0.027 |
impressed | +0.023 |
understanding | +0.021 |
pleased | +0.021 |
Neg logit contributions | |
Ready | -0.032 |
Left | -0.028 |
ready | -0.025 |
Three | -0.025 |
south | -0.025 |
Token | walked |
Token ID | 6807 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
lee | +0.178 |
atted | +0.178 |
impressed | +0.167 |
omo | +0.156 |
glad | +0.154 |
Neg logit contributions | |
Ready | -0.195 |
Left | -0.173 |
alis | -0.159 |
sc | -0.154 |
bust | -0.151 |
Token | , |
Token ID | 11 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
atted | +0.028 |
lee | +0.027 |
impressed | +0.025 |
smiling | +0.023 |
glad | +0.023 |
Neg logit contributions | |
Ready | -0.055 |
Left | -0.049 |
alis | -0.048 |
sc | -0.045 |
summit | -0.045 |
Token | he |
Token ID | 339 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
lee | +0.278 |
atted | +0.275 |
impressed | +0.249 |
omo | +0.233 |
smiling | +0.225 |
Neg logit contributions | |
Ready | -0.417 |
Left | -0.376 |
alis | -0.347 |
sc | -0.341 |
ready | -0.333 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Ready | +0.043 |
Left | +0.038 |
alis | +0.035 |
sc | +0.034 |
bust | +0.033 |
Neg logit contributions | |
lee | -0.039 |
atted | -0.039 |
impressed | -0.036 |
omo | -0.034 |
smiling | -0.034 |
Token | a |
Token ID | 257 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
lee | +0.053 |
atted | +0.053 |
impressed | +0.049 |
omo | +0.047 |
smiling | +0.046 |
Neg logit contributions | |
Ready | -0.045 |
Left | -0.039 |
alis | -0.035 |
sc | -0.034 |
bust | -0.033 |
Token | big |
Token ID | 1263 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
Ready | +0.068 |
Left | +0.061 |
alis | +0.056 |
sc | +0.055 |
bust | +0.053 |
Neg logit contributions | |
lee | -0.065 |
atted | -0.064 |
impressed | -0.061 |
omo | -0.058 |
smiling | -0.057 |
Token | rock |
Token ID | 3881 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
atted | +0.029 |
lee | +0.028 |
impressed | +0.026 |
omo | +0.025 |
glad | +0.024 |
Neg logit contributions | |
Ready | -0.021 |
Left | -0.018 |
alis | -0.017 |
bust | -0.016 |
sc | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
Ready | +0.103 |
Left | +0.094 |
alis | +0.086 |
bust | +0.083 |
sc | +0.083 |
Neg logit contributions | |
atted | -0.066 |
lee | -0.065 |
impressed | -0.059 |
smiling | -0.056 |
mad | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
lee | +0.049 |
atted | +0.048 |
impressed | +0.045 |
omo | +0.043 |
glad | +0.041 |
Neg logit contributions | |
Ready | -0.054 |
Left | -0.048 |
alis | -0.044 |
sc | -0.043 |
bust | -0.042 |
Token | play |
Token ID | 711 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
Ready | +0.092 |
Left | +0.081 |
sc | +0.071 |
alis | +0.071 |
bust | +0.065 |
Neg logit contributions | |
lee | -0.132 |
atted | -0.129 |
impressed | -0.123 |
smiling | -0.120 |
mad | -0.119 |
Token | both |
Token ID | 1111 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
Ready | +0.097 |
Left | +0.087 |
alis | +0.084 |
bust | +0.082 |
summit | +0.082 |
Neg logit contributions | |
atted | -0.047 |
impressed | -0.047 |
lee | -0.047 |
smiling | -0.043 |
mad | -0.042 |
Token | games |
Token ID | 1830 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
Ready | +0.104 |
Left | +0.094 |
alis | +0.087 |
sc | +0.083 |
bust | +0.083 |
Neg logit contributions | |
atted | -0.074 |
lee | -0.074 |
impressed | -0.068 |
omo | -0.064 |
smiling | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
Ready | +0.132 |
Left | +0.119 |
sc | +0.112 |
alis | +0.111 |
ready | +0.106 |
Neg logit contributions | |
lee | -0.082 |
atted | -0.081 |
impressed | -0.072 |
omo | -0.069 |
mad | -0.065 |
Token | First |
Token ID | 3274 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
Ready | +0.032 |
Left | +0.029 |
alis | +0.027 |
bust | +0.026 |
sc | +0.026 |
Neg logit contributions | |
lee | -0.029 |
atted | -0.029 |
impressed | -0.026 |
omo | -0.025 |
glad | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +7.91e+00 |
Pos logit contributions | |
lee | +0.037 |
atted | +0.036 |
impressed | +0.035 |
omo | +0.031 |
glad | +0.030 |
Neg logit contributions | |
Ready | -0.064 |
Left | -0.058 |
sc | -0.054 |
ready | -0.053 |
alis | -0.053 |
Token | they |
Token ID | 484 |
Feature activation | -3.11e+00 |
Pos logit contributions | |
lee | +0.224 |
atted | +0.221 |
impressed | +0.206 |
smiling | +0.191 |
omo | +0.187 |
Neg logit contributions | |
Ready | -0.402 |
Left | -0.367 |
alis | -0.350 |
sc | -0.341 |
bust | -0.328 |
Token | played |
Token ID | 2826 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
Ready | +0.132 |
Left | +0.118 |
alis | +0.108 |
sc | +0.104 |
ready | +0.102 |
Neg logit contributions | |
lee | -0.121 |
atted | -0.118 |
impressed | -0.111 |
smiling | -0.103 |
omo | -0.103 |
Token | hide |
Token ID | 7808 |
Feature activation | +3.44e+00 |
Pos logit contributions | |
atted | +0.066 |
lee | +0.065 |
impressed | +0.065 |
smiling | +0.060 |
mad | +0.059 |
Neg logit contributions | |
Ready | -0.078 |
Left | -0.068 |
alis | -0.066 |
bust | -0.063 |
summit | -0.062 |
Token | and |
Token ID | 290 |
Feature activation | +6.29e+00 |
Pos logit contributions | |
atted | +0.078 |
lee | +0.076 |
impressed | +0.067 |
omo | +0.066 |
glad | +0.061 |
Neg logit contributions | |
Ready | -0.196 |
Left | -0.183 |
sc | -0.178 |
alis | -0.169 |
bust | -0.167 |
Token | many |
Token ID | 867 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
Ready | +0.035 |
Left | +0.031 |
alis | +0.028 |
sc | +0.028 |
bust | +0.027 |
Neg logit contributions | |
lee | -0.037 |
atted | -0.037 |
impressed | -0.035 |
omo | -0.034 |
mad | -0.032 |
Token | rooms |
Token ID | 9519 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
Ready | +0.082 |
Left | +0.072 |
sc | +0.065 |
south | +0.062 |
bust | +0.061 |
Neg logit contributions | |
atted | -0.112 |
lee | -0.110 |
omo | -0.105 |
impressed | -0.103 |
glad | -0.100 |
Token | and |
Token ID | 290 |
Feature activation | +5.62e+00 |
Pos logit contributions | |
Ready | +0.114 |
Left | +0.102 |
sc | +0.098 |
south | +0.095 |
alis | +0.095 |
Neg logit contributions | |
atted | -0.081 |
lee | -0.080 |
omo | -0.074 |
impressed | -0.073 |
mad | -0.069 |
Token | windows |
Token ID | 9168 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
lee | +0.227 |
atted | +0.226 |
impressed | +0.211 |
omo | +0.199 |
smiling | +0.198 |
Neg logit contributions | |
Ready | -0.226 |
Left | -0.201 |
alis | -0.186 |
sc | -0.179 |
bust | -0.171 |
Token | and |
Token ID | 290 |
Feature activation | +5.73e+00 |
Pos logit contributions | |
Ready | +0.121 |
Left | +0.111 |
south | +0.104 |
alis | +0.103 |
sc | +0.102 |
Neg logit contributions | |
lee | -0.108 |
atted | -0.106 |
impressed | -0.102 |
omo | -0.098 |
glad | -0.094 |
Token | stairs |
Token ID | 16046 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
lee | +0.227 |
atted | +0.223 |
impressed | +0.212 |
smiling | +0.200 |
omo | +0.196 |
Neg logit contributions | |
Ready | -0.233 |
Left | -0.208 |
alis | -0.195 |
sc | -0.185 |
summit | -0.178 |
Token | . |
Token ID | 13 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
Ready | +0.113 |
Left | +0.102 |
alis | +0.094 |
sc | +0.094 |
bust | +0.090 |
Neg logit contributions | |
lee | -0.084 |
atted | -0.083 |
impressed | -0.076 |
omo | -0.073 |
glad | -0.070 |
Token | A |
Token ID | 317 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
Ready | +0.046 |
Left | +0.041 |
alis | +0.038 |
sc | +0.038 |
bust | +0.036 |
Neg logit contributions | |
lee | -0.031 |
atted | -0.030 |
impressed | -0.028 |
omo | -0.026 |
smiling | -0.025 |
Token | friendly |
Token ID | 8030 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
Ready | +0.053 |
Left | +0.047 |
alis | +0.044 |
sc | +0.042 |
ready | +0.040 |
Neg logit contributions | |
lee | -0.055 |
atted | -0.055 |
impressed | -0.053 |
omo | -0.049 |
smiling | -0.049 |
Token | man |
Token ID | 582 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
atted | +0.157 |
lee | +0.157 |
impressed | +0.145 |
omo | +0.139 |
mad | +0.135 |
Neg logit contributions | |
Ready | -0.142 |
Left | -0.124 |
alis | -0.113 |
sc | -0.110 |
bust | -0.108 |
Token | with |
Token ID | 351 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
lee | +0.103 |
atted | +0.098 |
omo | +0.091 |
impressed | +0.091 |
mad | +0.084 |
Neg logit contributions | |
Ready | -0.117 |
Left | -0.103 |
sc | -0.096 |
bust | -0.091 |
alis | -0.091 |
Token | tree |
Token ID | 5509 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
sc | +0.193 |
Left | +0.178 |
Ready | +0.175 |
works | +0.173 |
alis | +0.161 |
Neg logit contributions | |
mad | -0.215 |
dad | -0.214 |
glad | -0.210 |
smiling | -0.210 |
understanding | -0.209 |
Token | , |
Token ID | 11 |
Feature activation | +7.27e+00 |
Pos logit contributions | |
Ready | +0.343 |
Left | +0.321 |
alis | +0.303 |
sc | +0.300 |
bust | +0.295 |
Neg logit contributions | |
atted | -0.060 |
lee | -0.058 |
impressed | -0.044 |
smiling | -0.040 |
mad | -0.038 |
Token | she |
Token ID | 673 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
atted | +0.174 |
lee | +0.172 |
impressed | +0.152 |
omo | +0.141 |
smiling | +0.140 |
Neg logit contributions | |
Ready | -0.410 |
Left | -0.376 |
alis | -0.361 |
sc | -0.355 |
bust | -0.342 |
Token | sees |
Token ID | 7224 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
Ready | +0.111 |
Left | +0.097 |
alis | +0.087 |
sc | +0.086 |
ready | +0.082 |
Neg logit contributions | |
atted | -0.140 |
lee | -0.139 |
impressed | -0.129 |
smiling | -0.124 |
omo | -0.123 |
Token | a |
Token ID | 257 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
Ready | +0.011 |
Left | +0.009 |
alis | +0.008 |
sc | +0.008 |
bust | +0.008 |
Neg logit contributions | |
lee | -0.011 |
atted | -0.011 |
impressed | -0.010 |
omo | -0.010 |
smiling | -0.009 |
Token | problem |
Token ID | 1917 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
Ready | +0.127 |
Left | +0.115 |
alis | +0.107 |
sc | +0.107 |
bust | +0.100 |
Neg logit contributions | |
lee | -0.113 |
impressed | -0.111 |
atted | -0.109 |
smiling | -0.106 |
mad | -0.104 |
Token | . |
Token ID | 13 |
Feature activation | -8.67e-01 |
Pos logit contributions | |
Ready | +0.098 |
Left | +0.087 |
sc | +0.081 |
alis | +0.080 |
ready | +0.077 |
Neg logit contributions | |
lee | -0.074 |
atted | -0.073 |
impressed | -0.068 |
omo | -0.064 |
glad | -0.062 |
Token | The |
Token ID | 383 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
Ready | +0.034 |
Left | +0.030 |
sc | +0.028 |
alis | +0.028 |
bust | +0.026 |
Neg logit contributions | |
lee | -0.036 |
atted | -0.035 |
impressed | -0.032 |
omo | -0.031 |
glad | -0.031 |
Token | arrow |
Token ID | 15452 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
Ready | +0.149 |
Left | +0.142 |
sc | +0.135 |
alis | +0.131 |
fire | +0.127 |
Neg logit contributions | |
lee | -0.087 |
impressed | -0.082 |
smiling | -0.079 |
understanding | -0.079 |
atted | -0.079 |
Token | is |
Token ID | 318 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
Ready | +0.005 |
Left | +0.004 |
alis | +0.004 |
sc | +0.004 |
bust | +0.004 |
Neg logit contributions | |
lee | -0.008 |
atted | -0.008 |
impressed | -0.007 |
omo | -0.007 |
smiling | -0.007 |
Token | stuck |
Token ID | 7819 |
Feature activation | +5.49e-01 |
Pos logit contributions | |
Ready | +0.023 |
Left | +0.021 |
alis | +0.019 |
south | +0.019 |
sc | +0.018 |
Neg logit contributions | |
lee | -0.019 |
atted | -0.019 |
impressed | -0.017 |
omo | -0.016 |
smiling | -0.015 |
Token | 't |
Token ID | 470 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Ready | +0.058 |
Left | +0.050 |
alis | +0.047 |
sc | +0.044 |
south | +0.043 |
Neg logit contributions | |
lee | -0.062 |
atted | -0.059 |
impressed | -0.056 |
mad | -0.055 |
omo | -0.054 |
Token | belong |
Token ID | 5594 |
Feature activation | -4.51e+00 |
Pos logit contributions | |
Ready | +0.040 |
Left | +0.033 |
alis | +0.030 |
sc | +0.030 |
Three | +0.029 |
Neg logit contributions | |
lee | -0.061 |
mad | -0.060 |
atted | -0.060 |
understanding | -0.059 |
impressed | -0.059 |
Token | to |
Token ID | 284 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
Ready | +0.204 |
Left | +0.182 |
alis | +0.168 |
sc | +0.166 |
bust | +0.160 |
Neg logit contributions | |
lee | -0.161 |
atted | -0.160 |
impressed | -0.146 |
omo | -0.139 |
smiling | -0.135 |
Token | her |
Token ID | 607 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
Ready | +0.252 |
Left | +0.226 |
alis | +0.211 |
sc | +0.208 |
bust | +0.201 |
Neg logit contributions | |
lee | -0.173 |
atted | -0.170 |
impressed | -0.155 |
omo | -0.146 |
smiling | -0.145 |
Token | . |
Token ID | 13 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
Ready | +0.049 |
Left | +0.044 |
alis | +0.041 |
sc | +0.040 |
bust | +0.039 |
Neg logit contributions | |
lee | -0.035 |
atted | -0.035 |
impressed | -0.032 |
omo | -0.030 |
smiling | -0.029 |
Token | She |
Token ID | 1375 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
Ready | +0.031 |
Left | +0.026 |
alis | +0.024 |
sc | +0.023 |
Three | +0.022 |
Neg logit contributions | |
atted | -0.036 |
lee | -0.036 |
impressed | -0.034 |
smiling | -0.032 |
mad | -0.032 |
Token | said |
Token ID | 531 |
Feature activation | +5.53e+00 |
Pos logit contributions | |
Ready | +0.074 |
Left | +0.066 |
bust | +0.064 |
alis | +0.064 |
sc | +0.063 |
Neg logit contributions | |
lee | -0.053 |
atted | -0.051 |
impressed | -0.048 |
omo | -0.046 |
ummies | -0.046 |
Token | she |
Token ID | 673 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
omo | +0.224 |
lee | +0.216 |
atted | +0.216 |
impressed | +0.195 |
asured | +0.179 |
Neg logit contributions | |
Ready | -0.193 |
Left | -0.173 |
br | -0.157 |
bust | -0.151 |
sc | -0.149 |
Token | would |
Token ID | 561 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
Ready | +0.053 |
Left | +0.045 |
alis | +0.041 |
sc | +0.041 |
bust | +0.039 |
Neg logit contributions | |
lee | -0.068 |
atted | -0.068 |
impressed | -0.064 |
omo | -0.062 |
smiling | -0.060 |
Token | not |
Token ID | 407 |
Feature activation | +7.09e-01 |
Pos logit contributions | |
Ready | +0.048 |
Left | +0.042 |
alis | +0.037 |
sc | +0.036 |
bust | +0.035 |
Neg logit contributions | |
lee | -0.067 |
atted | -0.066 |
impressed | -0.062 |
omo | -0.060 |
smiling | -0.058 |
Token | give |
Token ID | 1577 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
lee | +0.034 |
atted | +0.034 |
impressed | +0.032 |
omo | +0.031 |
smiling | +0.031 |
Neg logit contributions | |
Ready | -0.023 |
Left | -0.020 |
alis | -0.017 |
sc | -0.017 |
bust | -0.016 |
Token | vinegar |
Token ID | 26600 |
Feature activation | -5.67e-01 |
Pos logit contributions | |
Ready | +0.081 |
Left | +0.070 |
alis | +0.064 |
sc | +0.064 |
ready | +0.061 |
Neg logit contributions | |
lee | -0.085 |
atted | -0.084 |
impressed | -0.077 |
omo | -0.073 |
glad | -0.071 |
Token | was |
Token ID | 373 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
Ready | +0.020 |
sc | +0.017 |
Left | +0.017 |
alis | +0.016 |
am | +0.016 |
Neg logit contributions | |
lee | -0.024 |
atted | -0.024 |
impressed | -0.022 |
omo | -0.022 |
glad | -0.022 |
Token | very |
Token ID | 845 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
lee | +0.087 |
atted | +0.086 |
omo | +0.072 |
impressed | +0.069 |
ica | +0.066 |
Neg logit contributions | |
Ready | -0.081 |
Left | -0.073 |
sc | -0.070 |
alis | -0.069 |
fart | -0.068 |
Token | sour |
Token ID | 11348 |
Feature activation | +4.13e+00 |
Pos logit contributions | |
atted | +0.194 |
lee | +0.193 |
pa | +0.161 |
ica | +0.160 |
omo | +0.160 |
Neg logit contributions | |
Ready | -0.152 |
Left | -0.133 |
sc | -0.133 |
fart | -0.125 |
am | -0.125 |
Token | and |
Token ID | 290 |
Feature activation | +6.15e+00 |
Pos logit contributions | |
atted | +0.138 |
lee | +0.136 |
impressed | +0.126 |
glad | +0.118 |
pleased | +0.114 |
Neg logit contributions | |
Ready | -0.175 |
sc | -0.166 |
Left | -0.157 |
fart | -0.156 |
br | -0.155 |
Token | made |
Token ID | 925 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
atted | +0.237 |
lee | +0.234 |
impressed | +0.206 |
omo | +0.194 |
glad | +0.189 |
Neg logit contributions | |
Ready | -0.246 |
Left | -0.212 |
sc | -0.206 |
alis | -0.197 |
bust | -0.194 |
Token | a |
Token ID | 257 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
Ready | +0.108 |
Left | +0.098 |
alis | +0.092 |
sc | +0.091 |
bust | +0.088 |
Neg logit contributions | |
lee | -0.061 |
atted | -0.060 |
impressed | -0.054 |
omo | -0.051 |
smiling | -0.049 |
Token | big |
Token ID | 1263 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
Ready | +0.068 |
Left | +0.061 |
alis | +0.056 |
sc | +0.055 |
bust | +0.053 |
Neg logit contributions | |
lee | -0.058 |
atted | -0.058 |
impressed | -0.052 |
omo | -0.050 |
smiling | -0.048 |
Token | mess |
Token ID | 2085 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
lee | +0.021 |
atted | +0.021 |
impressed | +0.017 |
glad | +0.016 |
omo | +0.016 |
Neg logit contributions | |
Ready | -0.021 |
Left | -0.018 |
sc | -0.017 |
alis | -0.017 |
fart | -0.017 |
Token | on |
Token ID | 319 |
Feature activation | -5.42e+00 |
Pos logit contributions | |
Ready | +0.042 |
Left | +0.038 |
sc | +0.037 |
alis | +0.035 |
br | +0.033 |
Neg logit contributions | |
atted | -0.036 |
lee | -0.036 |
impressed | -0.033 |
omo | -0.032 |
glad | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
Ready | +0.235 |
Left | +0.210 |
alis | +0.195 |
sc | +0.190 |
summit | +0.185 |
Neg logit contributions | |
atted | -0.204 |
lee | -0.200 |
impressed | -0.186 |
omo | -0.173 |
glad | -0.170 |
Token | depend |
Token ID | 4745 |
Feature activation | +3.85e+00 |
Pos logit contributions | |
atted | +0.090 |
lee | +0.088 |
omo | +0.078 |
impressed | +0.075 |
mad | +0.067 |
Neg logit contributions | |
Ready | -0.143 |
Left | -0.129 |
alis | -0.123 |
bust | -0.118 |
sc | -0.116 |
Token | able |
Token ID | 540 |
Feature activation | +4.79e+00 |
Pos logit contributions | |
atted | +0.146 |
lee | +0.144 |
impressed | +0.134 |
mad | +0.126 |
omo | +0.126 |
Neg logit contributions | |
Ready | -0.166 |
Left | -0.148 |
alis | -0.137 |
sc | -0.131 |
summit | -0.129 |
Token | ! |
Token ID | 0 |
Feature activation | -6.21e-01 |
Pos logit contributions | |
atted | +0.190 |
lee | +0.188 |
impressed | +0.171 |
omo | +0.164 |
mad | +0.159 |
Neg logit contributions | |
Ready | -0.195 |
Left | -0.175 |
sc | -0.159 |
alis | -0.158 |
bust | -0.151 |
Token | And |
Token ID | 843 |
Feature activation | +7.70e+00 |
Pos logit contributions | |
Ready | +0.024 |
Left | +0.021 |
alis | +0.020 |
sc | +0.020 |
summit | +0.019 |
Neg logit contributions | |
lee | -0.024 |
atted | -0.024 |
impressed | -0.022 |
glad | -0.022 |
omo | -0.022 |
Token | I |
Token ID | 314 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
lee | +0.305 |
atted | +0.302 |
impressed | +0.291 |
glad | +0.277 |
smiling | +0.276 |
Neg logit contributions | |
Ready | -0.308 |
Left | -0.273 |
alis | -0.255 |
sc | -0.250 |
Three | -0.240 |
Token | can |
Token ID | 460 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
Ready | +0.133 |
Left | +0.114 |
alis | +0.100 |
ready | +0.096 |
sc | +0.094 |
Neg logit contributions | |
lee | -0.220 |
atted | -0.219 |
impressed | -0.213 |
glad | -0.205 |
omo | -0.204 |
Token | go |
Token ID | 467 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
Ready | +0.111 |
Left | +0.095 |
sc | +0.089 |
alis | +0.084 |
ready | +0.083 |
Neg logit contributions | |
lee | -0.145 |
atted | -0.138 |
impressed | -0.135 |
smiling | -0.133 |
understanding | -0.129 |
Token | on |
Token ID | 319 |
Feature activation | -4.56e+00 |
Pos logit contributions | |
lee | +0.027 |
atted | +0.026 |
impressed | +0.024 |
smiling | +0.024 |
glad | +0.023 |
Neg logit contributions | |
Ready | -0.039 |
Left | -0.035 |
alis | -0.032 |
ready | -0.031 |
sc | -0.031 |
Token | adventures |
Token ID | 17545 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
Ready | +0.197 |
Left | +0.175 |
alis | +0.162 |
sc | +0.159 |
bust | +0.153 |
Neg logit contributions | |
lee | -0.173 |
atted | -0.169 |
impressed | -0.158 |
smiling | -0.149 |
omo | -0.147 |
Token | whenever |
Token ID | 8797 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
Ready | +0.029 |
Left | +0.026 |
alis | +0.025 |
sc | +0.024 |
ready | +0.024 |
Neg logit contributions | |
lee | -0.017 |
atted | -0.017 |
impressed | -0.016 |
smiling | -0.015 |
omo | -0.014 |
Token | I |
Token ID | 314 |
Feature activation | -6.91e+00 |
Pos logit contributions | |
Ready | +0.143 |
Left | +0.129 |
alis | +0.129 |
sc | +0.128 |
tr | +0.125 |
Neg logit contributions | |
lee | -0.104 |
impressed | -0.097 |
atted | -0.096 |
smiling | -0.095 |
glad | -0.093 |
Token | his |
Token ID | 465 |
Feature activation | +7.21e-01 |
Pos logit contributions | |
lee | +0.038 |
atted | +0.038 |
impressed | +0.037 |
omo | +0.036 |
pleased | +0.035 |
Neg logit contributions | |
Ready | -0.034 |
Left | -0.029 |
ready | -0.025 |
Ready | -0.025 |
Three | -0.025 |
Token | frown |
Token ID | 22002 |
Feature activation | +4.47e+00 |
Pos logit contributions | |
lee | +0.023 |
atted | +0.022 |
omo | +0.019 |
impressed | +0.018 |
anted | +0.017 |
Neg logit contributions | |
Ready | -0.031 |
source | -0.029 |
same | -0.029 |
hunt | -0.027 |
rest | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | +6.46e+00 |
Pos logit contributions | |
lee | +0.128 |
atted | +0.124 |
impressed | +0.114 |
omo | +0.107 |
glad | +0.102 |
Neg logit contributions | |
Ready | -0.226 |
Left | -0.208 |
sc | -0.197 |
alis | -0.196 |
bust | -0.187 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
omo | +0.248 |
atted | +0.231 |
ummies | +0.228 |
lee | +0.226 |
ummy | +0.218 |
Neg logit contributions | |
Ready | -0.214 |
make | -0.186 |
bust | -0.186 |
lead | -0.186 |
will | -0.180 |
Token | him |
Token ID | 683 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
atted | +0.067 |
lee | +0.066 |
impressed | +0.064 |
omo | +0.062 |
pleased | +0.058 |
Neg logit contributions | |
Ready | -0.096 |
Left | -0.085 |
Ready | -0.077 |
sc | -0.077 |
ready | -0.076 |
Token | what |
Token ID | 644 |
Feature activation | -9.80e+00 |
Pos logit contributions | |
lee | +0.047 |
atted | +0.046 |
omo | +0.045 |
impressed | +0.041 |
mad | +0.037 |
Neg logit contributions | |
Ready | -0.053 |
Left | -0.048 |
Ready | -0.043 |
anywhere | -0.043 |
am | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | -7.15e-01 |
Pos logit contributions | |
Ready | +0.391 |
Left | +0.340 |
sc | +0.303 |
am | +0.294 |
alis | +0.293 |
Neg logit contributions | |
atted | -0.413 |
lee | -0.412 |
impressed | -0.365 |
omo | -0.356 |
glad | -0.342 |
Token | wrong |
Token ID | 2642 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
Ready | +0.022 |
Left | +0.018 |
alis | +0.016 |
sc | +0.016 |
bust | +0.015 |
Neg logit contributions | |
atted | -0.036 |
lee | -0.036 |
impressed | -0.034 |
omo | -0.033 |
smiling | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
Ready | +0.036 |
Left | +0.032 |
am | +0.032 |
sc | +0.029 |
Where | +0.028 |
Neg logit contributions | |
atted | -0.074 |
lee | -0.072 |
impressed | -0.069 |
glad | -0.066 |
pleased | -0.064 |
Token | Tim |
Token ID | 5045 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
Ready | +0.042 |
Left | +0.036 |
Three | +0.033 |
alis | +0.033 |
sc | +0.033 |
Neg logit contributions | |
atted | -0.039 |
lee | -0.037 |
impressed | -0.037 |
glad | -0.034 |
omo | -0.034 |
Token | told |
Token ID | 1297 |
Feature activation | +4.36e+00 |
Pos logit contributions | |
atted | +0.166 |
lee | +0.164 |
impressed | +0.151 |
glad | +0.150 |
ummies | +0.149 |
Neg logit contributions | |
Ready | -0.177 |
will | -0.166 |
Left | -0.161 |
hunt | -0.156 |
bust | -0.154 |
Token | sunshine |
Token ID | 34488 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
lee | +0.007 |
atted | +0.007 |
smiling | +0.007 |
impressed | +0.006 |
gigg | +0.006 |
Neg logit contributions | |
Ready | -0.009 |
Left | -0.008 |
ready | -0.007 |
sc | -0.007 |
bust | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +7.44e+00 |
Pos logit contributions | |
Ready | +0.092 |
Left | +0.083 |
alis | +0.078 |
sc | +0.077 |
bust | +0.075 |
Neg logit contributions | |
lee | -0.052 |
atted | -0.051 |
impressed | -0.046 |
omo | -0.043 |
smiling | -0.041 |
Token | excited |
Token ID | 6568 |
Feature activation | +7.07e+00 |
Pos logit contributions | |
lee | +0.251 |
atted | +0.251 |
omo | +0.227 |
impressed | +0.204 |
mad | +0.197 |
Neg logit contributions | |
Ready | -0.351 |
Left | -0.314 |
alis | -0.298 |
south | -0.292 |
sc | -0.288 |
Token | to |
Token ID | 284 |
Feature activation | -2.98e+00 |
Pos logit contributions | |
atted | +0.260 |
lee | +0.258 |
impressed | +0.244 |
omo | +0.239 |
glad | +0.223 |
Neg logit contributions | |
Ready | -0.302 |
Left | -0.277 |
alis | -0.251 |
sc | -0.245 |
bust | -0.239 |
Token | see |
Token ID | 766 |
Feature activation | -7.29e-01 |
Pos logit contributions | |
Ready | +0.115 |
Left | +0.097 |
sc | +0.097 |
ready | +0.088 |
alis | +0.085 |
Neg logit contributions | |
lee | -0.133 |
smiling | -0.124 |
atted | -0.124 |
impressed | -0.123 |
glad | -0.121 |
Token | what |
Token ID | 644 |
Feature activation | -9.97e+00 |
Pos logit contributions | |
Ready | +0.036 |
Left | +0.033 |
alis | +0.031 |
sc | +0.030 |
ready | +0.029 |
Neg logit contributions | |
lee | -0.023 |
atted | -0.023 |
impressed | -0.021 |
smiling | -0.020 |
glad | -0.020 |
Token | adventures |
Token ID | 17545 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
Ready | +0.476 |
Left | +0.424 |
alis | +0.402 |
sc | +0.392 |
summit | +0.378 |
Neg logit contributions | |
lee | -0.333 |
atted | -0.327 |
impressed | -0.309 |
smiling | -0.286 |
omo | -0.280 |
Token | awaited |
Token ID | 39576 |
Feature activation | -4.88e+00 |
Pos logit contributions | |
Ready | +0.091 |
Left | +0.081 |
alis | +0.077 |
Three | +0.074 |
summit | +0.073 |
Neg logit contributions | |
lee | -0.061 |
atted | -0.060 |
impressed | -0.055 |
smiling | -0.053 |
glad | -0.050 |
Token | her |
Token ID | 607 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
Ready | +0.294 |
Left | +0.265 |
alis | +0.257 |
ready | +0.246 |
Three | +0.246 |
Neg logit contributions | |
lee | -0.107 |
atted | -0.102 |
impressed | -0.087 |
smiling | -0.084 |
mad | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
Ready | +0.122 |
Left | +0.108 |
sc | +0.100 |
alis | +0.100 |
Three | +0.098 |
Neg logit contributions | |
lee | -0.073 |
atted | -0.071 |
impressed | -0.066 |
smiling | -0.062 |
glad | -0.060 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -7.85e+00 |
Pos logit contributions | |
Ready | +0.061 |
Left | +0.054 |
alis | +0.051 |
sc | +0.050 |
ready | +0.048 |
Neg logit contributions | |
lee | -0.051 |
atted | -0.051 |
impressed | -0.046 |
glad | -0.043 |
omo | -0.043 |
Token | . |
Token ID | 13 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
Ready | +0.276 |
Left | +0.249 |
alis | +0.231 |
sc | +0.226 |
bust | +0.221 |
Neg logit contributions | |
lee | -0.189 |
atted | -0.188 |
impressed | -0.172 |
omo | -0.161 |
smiling | -0.160 |
Token | He |
Token ID | 679 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Ready | +0.042 |
Left | +0.035 |
alis | +0.031 |
sc | +0.031 |
ready | +0.030 |
Neg logit contributions | |
atted | -0.056 |
lee | -0.056 |
impressed | -0.052 |
omo | -0.050 |
smiling | -0.049 |
Token | wanted |
Token ID | 2227 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
Ready | +0.074 |
Left | +0.060 |
Find | +0.058 |
bust | +0.058 |
Three | +0.057 |
Neg logit contributions | |
atted | -0.087 |
glad | -0.081 |
lee | -0.081 |
mad | -0.079 |
impressed | -0.078 |
Token | to |
Token ID | 284 |
Feature activation | -3.95e+00 |
Pos logit contributions | |
Ready | +0.012 |
Left | +0.011 |
alis | +0.010 |
sc | +0.010 |
ready | +0.010 |
Neg logit contributions | |
atted | -0.012 |
lee | -0.012 |
omo | -0.011 |
impressed | -0.011 |
mad | -0.010 |
Token | know |
Token ID | 760 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
Ready | +0.139 |
Left | +0.123 |
alis | +0.109 |
sc | +0.108 |
bust | +0.100 |
Neg logit contributions | |
lee | -0.182 |
atted | -0.176 |
impressed | -0.169 |
understanding | -0.162 |
smiling | -0.161 |
Token | what |
Token ID | 644 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
lee | +0.055 |
atted | +0.055 |
omo | +0.049 |
impressed | +0.046 |
mad | +0.045 |
Neg logit contributions | |
Ready | -0.070 |
Left | -0.063 |
sc | -0.057 |
south | -0.057 |
alis | -0.056 |
Token | was |
Token ID | 373 |
Feature activation | -8.29e-01 |
Pos logit contributions | |
Ready | +0.501 |
Left | +0.454 |
alis | +0.432 |
sc | +0.415 |
summit | +0.407 |
Neg logit contributions | |
lee | -0.309 |
atted | -0.303 |
impressed | -0.285 |
smiling | -0.264 |
understanding | -0.256 |
Token | inside |
Token ID | 2641 |
Feature activation | -4.63e+00 |
Pos logit contributions | |
Ready | +0.036 |
Left | +0.031 |
alis | +0.029 |
sc | +0.029 |
ready | +0.027 |
Neg logit contributions | |
atted | -0.032 |
lee | -0.032 |
impressed | -0.029 |
omo | -0.028 |
smiling | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | -8.21e+00 |
Pos logit contributions | |
Ready | +0.189 |
Left | +0.165 |
sc | +0.155 |
south | +0.153 |
summit | +0.152 |
Neg logit contributions | |
atted | -0.177 |
lee | -0.171 |
omo | -0.160 |
impressed | -0.158 |
mad | -0.142 |
Token | , |
Token ID | 11 |
Feature activation | +6.82e+00 |
Pos logit contributions | |
Ready | +0.460 |
Left | +0.422 |
alis | +0.397 |
sc | +0.384 |
bust | +0.381 |
Neg logit contributions | |
lee | -0.207 |
atted | -0.200 |
impressed | -0.183 |
smiling | -0.167 |
mad | -0.162 |
Token | so |
Token ID | 523 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
atted | +0.235 |
lee | +0.232 |
omo | +0.202 |
impressed | +0.201 |
smiling | +0.186 |
Neg logit contributions | |
Ready | -0.323 |
Left | -0.287 |
sc | -0.267 |
alis | -0.267 |
south | -0.258 |
Token | enny |
Token ID | 11870 |
Feature activation | +6.60e+00 |
Pos logit contributions | |
atted | +0.273 |
impressed | +0.267 |
smiling | +0.261 |
glad | +0.261 |
lee | +0.258 |
Neg logit contributions | |
Ready | -0.337 |
Left | -0.293 |
sc | -0.286 |
alis | -0.273 |
Three | -0.271 |
Token | asked |
Token ID | 1965 |
Feature activation | +3.38e+00 |
Pos logit contributions | |
atted | +0.247 |
glad | +0.230 |
omo | +0.220 |
ummies | +0.219 |
lee | +0.215 |
Neg logit contributions | |
Ready | -0.225 |
bust | -0.205 |
hunt | -0.198 |
Left | -0.197 |
tr | -0.193 |
Token | Jack |
Token ID | 3619 |
Feature activation | +5.31e+00 |
Pos logit contributions | |
atted | +0.114 |
omo | +0.108 |
lee | +0.102 |
impressed | +0.101 |
coughing | +0.092 |
Neg logit contributions | |
Ready | -0.153 |
Left | -0.135 |
Ready | -0.124 |
Where | -0.122 |
sc | -0.120 |
Token | , |
Token ID | 11 |
Feature activation | +9.05e+00 |
Pos logit contributions | |
lee | +0.162 |
atted | +0.158 |
omo | +0.149 |
impressed | +0.135 |
mad | +0.122 |
Neg logit contributions | |
Ready | -0.248 |
Left | -0.226 |
alis | -0.206 |
sc | -0.206 |
am | -0.206 |
Token | " |
Token ID | 366 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
atted | +0.374 |
lee | +0.369 |
omo | +0.334 |
impressed | +0.320 |
mad | +0.300 |
Neg logit contributions | |
Ready | -0.368 |
Left | -0.322 |
sc | -0.280 |
alis | -0.279 |
Three | -0.273 |
Token | What |
Token ID | 2061 |
Feature activation | -9.52e+00 |
Pos logit contributions | |
Ready | +0.061 |
Left | +0.053 |
alis | +0.050 |
sc | +0.045 |
bust | +0.044 |
Neg logit contributions | |
lee | -0.108 |
atted | -0.106 |
impressed | -0.104 |
omo | -0.099 |
smiling | -0.099 |
Token | is |
Token ID | 318 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
Ready | +0.328 |
Left | +0.281 |
alis | +0.262 |
sc | +0.249 |
bust | +0.238 |
Neg logit contributions | |
lee | -0.435 |
atted | -0.434 |
impressed | -0.418 |
smiling | -0.396 |
omo | -0.395 |
Token | that |
Token ID | 326 |
Feature activation | -5.24e+00 |
Pos logit contributions | |
Ready | +0.068 |
Left | +0.061 |
alis | +0.056 |
sc | +0.055 |
bust | +0.054 |
Neg logit contributions | |
lee | -0.050 |
atted | -0.050 |
impressed | -0.046 |
omo | -0.043 |
smiling | -0.042 |
Token | wreck |
Token ID | 15228 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
Ready | +0.248 |
Left | +0.224 |
alis | +0.212 |
sc | +0.209 |
Skip | +0.200 |
Neg logit contributions | |
impressed | -0.168 |
lee | -0.167 |
atted | -0.165 |
mad | -0.159 |
smiling | -0.154 |
Token | ?". |
Token ID | 43634 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
Ready | +0.021 |
Left | +0.019 |
alis | +0.017 |
sc | +0.017 |
Three | +0.017 |
Neg logit contributions | |
atted | -0.013 |
lee | -0.013 |
impressed | -0.012 |
mad | -0.011 |
smiling | -0.011 |
Token |
|
Token ID | 198 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
atted | +0.028 |
lee | +0.025 |
impressed | +0.024 |
omo | +0.023 |
mad | +0.021 |
Neg logit contributions | |
Ready | -0.054 |
Left | -0.048 |
alis | -0.044 |
Three | -0.044 |
sc | -0.044 |
Token | now |
Token ID | 783 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
lee | +0.143 |
atted | +0.142 |
impressed | +0.130 |
omo | +0.123 |
smiling | +0.119 |
Neg logit contributions | |
Ready | -0.187 |
Left | -0.168 |
alis | -0.156 |
sc | -0.154 |
bust | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
lee | +0.086 |
atted | +0.086 |
impressed | +0.078 |
omo | +0.076 |
smiling | +0.071 |
Neg logit contributions | |
Ready | -0.097 |
Left | -0.086 |
alis | -0.079 |
sc | -0.079 |
ready | -0.075 |
Token | You |
Token ID | 921 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
Ready | +0.042 |
Left | +0.037 |
alis | +0.034 |
sc | +0.033 |
summit | +0.032 |
Neg logit contributions | |
lee | -0.048 |
atted | -0.048 |
impressed | -0.044 |
glad | -0.043 |
smiling | -0.042 |
Token | must |
Token ID | 1276 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
Ready | +0.079 |
Left | +0.068 |
Whe | +0.062 |
alis | +0.061 |
sc | +0.061 |
Neg logit contributions | |
atted | -0.096 |
lee | -0.095 |
impressed | -0.093 |
mad | -0.093 |
glad | -0.089 |
Token | do |
Token ID | 466 |
Feature activation | -2.81e+00 |
Pos logit contributions | |
Ready | +0.097 |
Left | +0.086 |
sc | +0.081 |
alis | +0.075 |
ready | +0.075 |
Neg logit contributions | |
lee | -0.100 |
atted | -0.096 |
impressed | -0.093 |
mad | -0.092 |
understanding | -0.092 |
Token | what |
Token ID | 644 |
Feature activation | -9.49e+00 |
Pos logit contributions | |
Ready | +0.128 |
Left | +0.114 |
alis | +0.106 |
sc | +0.106 |
bust | +0.101 |
Neg logit contributions | |
lee | -0.101 |
atted | -0.099 |
impressed | -0.094 |
smiling | -0.087 |
mad | -0.086 |
Token | I |
Token ID | 314 |
Feature activation | -7.52e+00 |
Pos logit contributions | |
Ready | +0.374 |
Left | +0.328 |
alis | +0.300 |
sc | +0.293 |
bust | +0.282 |
Neg logit contributions | |
lee | -0.395 |
atted | -0.393 |
impressed | -0.365 |
omo | -0.348 |
smiling | -0.341 |
Token | say |
Token ID | 910 |
Feature activation | +1.47e+00 |
Pos logit contributions | |
Ready | +0.297 |
Left | +0.278 |
alis | +0.258 |
sc | +0.244 |
bust | +0.241 |
Neg logit contributions | |
lee | -0.287 |
atted | -0.287 |
impressed | -0.278 |
mad | -0.276 |
smiling | -0.267 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
lee | +0.057 |
atted | +0.057 |
impressed | +0.052 |
omo | +0.051 |
mad | +0.048 |
Neg logit contributions | |
Ready | -0.061 |
Left | -0.054 |
alis | -0.049 |
sc | -0.049 |
ready | -0.047 |
Token | You |
Token ID | 921 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
Ready | +0.051 |
Left | +0.044 |
alis | +0.040 |
sc | +0.039 |
bust | +0.037 |
Neg logit contributions | |
lee | -0.064 |
atted | -0.064 |
impressed | -0.060 |
omo | -0.057 |
smiling | -0.057 |
Token | are |
Token ID | 389 |
Feature activation | -9.40e-01 |
Pos logit contributions | |
Ready | +0.090 |
Left | +0.077 |
Whe | +0.071 |
alis | +0.071 |
sc | +0.070 |
Neg logit contributions | |
atted | -0.105 |
lee | -0.104 |
impressed | -0.101 |
mad | -0.101 |
gigg | -0.097 |