Token | � |
Token ID | 6184 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
booth | +0.015 |
Tom | +0.015 |
cheating | +0.015 |
tickets | +0.014 |
worker | +0.014 |
Neg logit contributions | |
stub | -0.025 |
Chick | -0.025 |
burn | -0.025 |
Grac | -0.024 |
berry | -0.024 |
Token | � |
Token ID | 95 |
Feature activation | +2.32e-01 |
Pos logit contributions | |
cheating | +0.027 |
tickets | +0.027 |
booth | +0.027 |
Tom | +0.027 |
money | +0.027 |
Neg logit contributions | |
berry | -0.032 |
Chick | -0.032 |
Grac | -0.032 |
stub | -0.031 |
light | -0.031 |
Token | € |
Token ID | 26391 |
Feature activation | +4.27e+00 |
Pos logit contributions | |
tickets | +0.011 |
cheating | +0.011 |
clues | +0.011 |
money | +0.011 |
bills | +0.010 |
Neg logit contributions | |
light | -0.011 |
stub | -0.011 |
Grac | -0.010 |
Chick | -0.010 |
berry | -0.010 |
Token | � |
Token ID | 447 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
booth | +0.164 |
Tom | +0.162 |
cheating | +0.151 |
Tomas | +0.150 |
employee | +0.148 |
Neg logit contributions | |
stub | -0.255 |
Chick | -0.252 |
burn | -0.249 |
berry | -0.245 |
Grac | -0.242 |
Token | � |
Token ID | 250 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
booth | +0.097 |
Tom | +0.095 |
cheating | +0.094 |
employee | +0.093 |
Tomas | +0.093 |
Neg logit contributions | |
stub | -0.055 |
Chick | -0.053 |
burn | -0.052 |
berry | -0.051 |
light | -0.050 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.91e+00 |
Pos logit contributions | |
booth | +0.048 |
Tom | +0.047 |
cheating | +0.046 |
employee | +0.045 |
Tomas | +0.045 |
Neg logit contributions | |
stub | -0.062 |
Chick | -0.061 |
burn | -0.059 |
berry | -0.059 |
light | -0.058 |
Token | , |
Token ID | 11 |
Feature activation | +3.21e-01 |
Pos logit contributions | |
booth | +0.408 |
employee | +0.398 |
Tomas | +0.395 |
cheating | +0.392 |
her | +0.388 |
Neg logit contributions | |
berry | -0.525 |
light | -0.502 |
Chick | -0.497 |
stub | -0.494 |
burn | -0.491 |
Token | 2 |
Token ID | 362 |
Feature activation | +8.28e+00 |
Pos logit contributions | |
booth | +0.012 |
cheating | +0.011 |
site | +0.011 |
Tomas | +0.011 |
employee | +0.011 |
Neg logit contributions | |
Chick | -0.018 |
berry | -0.018 |
stub | -0.018 |
Grac | -0.018 |
burn | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
booth | +0.350 |
employee | +0.337 |
Tomas | +0.336 |
cheating | +0.331 |
Tom | +0.327 |
Neg logit contributions | |
berry | -0.443 |
Chick | -0.433 |
light | -0.423 |
Grac | -0.419 |
stub | -0.415 |
Token | 3 |
Token ID | 513 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
booth | +0.012 |
cheating | +0.011 |
employee | +0.011 |
Tomas | +0.011 |
Tom | +0.011 |
Neg logit contributions | |
Chick | -0.018 |
berry | -0.018 |
stub | -0.018 |
Grac | -0.018 |
light | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
booth | +0.035 |
Tom | +0.034 |
cheating | +0.033 |
Tomas | +0.033 |
employee | +0.033 |
Neg logit contributions | |
stub | -0.046 |
Chick | -0.046 |
berry | -0.045 |
light | -0.045 |
burn | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | +8.65e-01 |
Pos logit contributions | |
stub | +0.013 |
Chick | +0.013 |
burn | +0.013 |
berry | +0.012 |
light | +0.012 |
Neg logit contributions | |
Tom | -0.007 |
booth | -0.007 |
cheating | -0.007 |
money | -0.007 |
tickets | -0.007 |
Token | birds |
Token ID | 10087 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
booth | +0.034 |
Tom | +0.033 |
Tomas | +0.033 |
employee | +0.032 |
cheating | +0.032 |
Neg logit contributions | |
stub | -0.050 |
berry | -0.049 |
Chick | -0.048 |
skirt | -0.048 |
Grac | -0.047 |
Token | - |
Token ID | 532 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
Tom | +0.067 |
booth | +0.066 |
tickets | +0.058 |
cheating | +0.057 |
money | +0.057 |
Neg logit contributions | |
Chick | -0.143 |
stub | -0.141 |
burn | -0.138 |
Grac | -0.136 |
skirt | -0.134 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.80e+00 |
Pos logit contributions | |
booth | +0.081 |
Tom | +0.080 |
cheating | +0.077 |
tickets | +0.075 |
Tomas | +0.075 |
Neg logit contributions | |
stub | -0.134 |
Chick | -0.131 |
burn | -0.129 |
berry | -0.128 |
light | -0.127 |
Token | , |
Token ID | 11 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
booth | +0.417 |
cheating | +0.413 |
employee | +0.407 |
site | +0.397 |
Tomas | +0.396 |
Neg logit contributions | |
berry | -0.517 |
light | -0.510 |
stub | -0.493 |
Chick | -0.493 |
burn | -0.487 |
Token | 2 |
Token ID | 17 |
Feature activation | +9.51e+00 |
Pos logit contributions | |
booth | +0.003 |
cheating | +0.003 |
employee | +0.003 |
Tom | +0.003 |
Tomas | +0.003 |
Neg logit contributions | |
stub | -0.005 |
Chick | -0.004 |
berry | -0.004 |
burn | -0.004 |
Grac | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
booth | +0.406 |
cheating | +0.394 |
employee | +0.389 |
Tomas | +0.381 |
Tom | +0.380 |
Neg logit contributions | |
berry | -0.509 |
stub | -0.508 |
Chick | -0.506 |
light | -0.501 |
burn | -0.495 |
Token | 3 |
Token ID | 18 |
Feature activation | +7.46e+00 |
Pos logit contributions | |
booth | +0.003 |
Tom | +0.003 |
cheating | +0.003 |
employee | +0.003 |
Tomas | +0.003 |
Neg logit contributions | |
stub | -0.005 |
Chick | -0.005 |
berry | -0.005 |
burn | -0.005 |
light | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
booth | +0.307 |
cheating | +0.293 |
employee | +0.292 |
Tom | +0.287 |
Tomas | +0.285 |
Neg logit contributions | |
stub | -0.416 |
Chick | -0.414 |
berry | -0.413 |
burn | -0.402 |
light | -0.402 |
Token | 4 |
Token ID | 19 |
Feature activation | +6.79e+00 |
Pos logit contributions | |
booth | +0.004 |
Tom | +0.004 |
cheating | +0.004 |
employee | +0.004 |
Tomas | +0.004 |
Neg logit contributions | |
stub | -0.006 |
Chick | -0.006 |
berry | -0.006 |
burn | -0.006 |
light | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +9.87e-02 |
Pos logit contributions | |
booth | +0.265 |
Tom | +0.253 |
cheating | +0.252 |
employee | +0.247 |
Tomas | +0.244 |
Neg logit contributions | |
stub | -0.398 |
Chick | -0.393 |
berry | -0.385 |
burn | -0.384 |
light | -0.380 |
Token | he |
Token ID | 339 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
booth | +0.043 |
Tom | +0.041 |
cheating | +0.041 |
employee | +0.040 |
Tomas | +0.039 |
Neg logit contributions | |
stub | -0.079 |
Chick | -0.078 |
burn | -0.076 |
berry | -0.076 |
light | -0.076 |
Token | could |
Token ID | 714 |
Feature activation | -9.80e-01 |
Pos logit contributions | |
stub | +0.062 |
Chick | +0.059 |
light | +0.058 |
berry | +0.056 |
burn | +0.056 |
Neg logit contributions | |
booth | -0.064 |
Tom | -0.061 |
cheating | -0.061 |
tickets | -0.060 |
site | -0.060 |
Token | trust |
Token ID | 3774 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Chick | +0.065 |
stub | +0.065 |
burn | +0.064 |
berry | +0.064 |
streamed | +0.062 |
Neg logit contributions | |
Tom | -0.032 |
money | -0.030 |
booth | -0.029 |
worker | -0.028 |
cheating | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
stub | +0.011 |
burn | +0.011 |
Chick | +0.011 |
berry | +0.011 |
light | +0.011 |
Neg logit contributions | |
Tom | -0.007 |
booth | -0.007 |
cheating | -0.007 |
Tomas | -0.006 |
money | -0.006 |
Token | zo |
Token ID | 40565 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
booth | +0.047 |
Tom | +0.046 |
cheating | +0.045 |
Tomas | +0.045 |
employee | +0.044 |
Neg logit contributions | |
stub | -0.056 |
Chick | -0.054 |
berry | -0.054 |
burn | -0.053 |
light | -0.052 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.40e+00 |
Pos logit contributions | |
booth | +0.102 |
Tom | +0.101 |
Tomas | +0.100 |
cheating | +0.097 |
employee | +0.095 |
Neg logit contributions | |
berry | -0.252 |
Chick | -0.248 |
stub | -0.247 |
light | -0.246 |
Grac | -0.242 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
booth | +0.396 |
Tomas | +0.372 |
Tom | +0.372 |
employee | +0.365 |
site | +0.354 |
Neg logit contributions | |
burn | -0.501 |
stub | -0.500 |
berry | -0.498 |
light | -0.489 |
Chick | -0.489 |
Token | to |
Token ID | 284 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
booth | +0.084 |
Tom | +0.082 |
cheating | +0.077 |
tickets | +0.076 |
worker | +0.075 |
Neg logit contributions | |
stub | -0.158 |
Chick | -0.155 |
burn | -0.151 |
berry | -0.148 |
Grac | -0.148 |
Token | help |
Token ID | 1037 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
stub | +0.019 |
Chick | +0.018 |
burn | +0.018 |
berry | +0.018 |
light | +0.018 |
Neg logit contributions | |
booth | -0.015 |
Tom | -0.014 |
cheating | -0.014 |
employee | -0.014 |
Tomas | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +8.45e-01 |
Pos logit contributions | |
stub | +0.056 |
Chick | +0.055 |
burn | +0.054 |
berry | +0.054 |
light | +0.053 |
Neg logit contributions | |
booth | -0.044 |
Tom | -0.043 |
cheating | -0.042 |
employee | -0.041 |
Tomas | -0.041 |
Token | weak |
Token ID | 4939 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
Tomas | +0.030 |
Tom | +0.030 |
booth | +0.030 |
newspapers | +0.028 |
cheating | +0.028 |
Neg logit contributions | |
Chick | -0.049 |
berry | -0.049 |
stub | -0.049 |
Grac | -0.048 |
burn | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | -2.67e-01 |
Pos logit contributions | |
booth | +0.109 |
Tom | +0.105 |
cheating | +0.104 |
employee | +0.103 |
Tomas | +0.103 |
Neg logit contributions | |
Chick | -0.139 |
stub | -0.139 |
berry | -0.138 |
light | -0.136 |
burn | -0.136 |
Token | 3 |
Token ID | 513 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
Chick | +0.017 |
berry | +0.017 |
light | +0.016 |
burn | +0.016 |
stub | +0.016 |
Neg logit contributions | |
booth | -0.009 |
Tomas | -0.008 |
employee | -0.008 |
cheating | -0.008 |
Tom | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
booth | +0.027 |
Tom | +0.027 |
cheating | +0.026 |
employee | +0.025 |
tickets | +0.025 |
Neg logit contributions | |
stub | -0.039 |
Chick | -0.038 |
burn | -0.037 |
berry | -0.037 |
light | -0.036 |
Token | 2 |
Token ID | 362 |
Feature activation | +7.82e+00 |
Pos logit contributions | |
Chick | +0.016 |
berry | +0.015 |
stub | +0.015 |
burn | +0.015 |
light | +0.015 |
Neg logit contributions | |
booth | -0.008 |
cheating | -0.007 |
employee | -0.007 |
Tomas | -0.007 |
Tom | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
booth | +0.325 |
Tom | +0.311 |
cheating | +0.311 |
employee | +0.309 |
Tomas | +0.308 |
Neg logit contributions | |
Chick | -0.429 |
berry | -0.424 |
stub | -0.422 |
light | -0.418 |
burn | -0.416 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.37e+00 |
Pos logit contributions | |
Chick | +0.015 |
stub | +0.014 |
berry | +0.014 |
burn | +0.014 |
light | +0.014 |
Neg logit contributions | |
booth | -0.007 |
cheating | -0.007 |
employee | -0.007 |
Tomas | -0.007 |
Tom | -0.007 |
Token | ". |
Token ID | 1911 |
Feature activation | -2.44e+00 |
Pos logit contributions | |
cheating | +0.447 |
booth | +0.442 |
employee | +0.440 |
Tomas | +0.439 |
her | +0.423 |
Neg logit contributions | |
berry | -0.448 |
light | -0.433 |
Chick | -0.429 |
burn | -0.426 |
stub | -0.396 |
Token | They |
Token ID | 1119 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
Chick | +0.127 |
berry | +0.123 |
Grac | +0.122 |
stub | +0.121 |
apy | +0.113 |
Neg logit contributions | |
cheating | -0.114 |
booth | -0.113 |
employee | -0.105 |
Tomas | -0.102 |
tickets | -0.102 |
Token | hear |
Token ID | 3285 |
Feature activation | -2.30e-01 |
Pos logit contributions | |
booth | +0.010 |
cheating | +0.010 |
employee | +0.009 |
site | +0.009 |
worker | +0.009 |
Neg logit contributions | |
stub | -0.011 |
Chick | -0.010 |
berry | -0.010 |
burn | -0.010 |
Grac | -0.010 |
Token | a |
Token ID | 257 |
Feature activation | +6.84e-01 |
Pos logit contributions | |
Chick | +0.014 |
light | +0.014 |
stub | +0.014 |
Grac | +0.013 |
berry | +0.013 |
Neg logit contributions | |
booth | -0.008 |
cheating | -0.008 |
employee | -0.008 |
Tomas | -0.008 |
worker | -0.007 |
Token | big |
Token ID | 1263 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
booth | +0.029 |
Tom | +0.028 |
cheating | +0.028 |
Tomas | +0.028 |
employee | +0.027 |
Neg logit contributions | |
stub | -0.038 |
Chick | -0.037 |
berry | -0.036 |
light | -0.036 |
burn | -0.036 |
Token | very |
Token ID | 845 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
inconvenience | +0.209 |
booth | +0.207 |
employee | +0.202 |
cheating | +0.199 |
Tomas | +0.194 |
Neg logit contributions | |
berry | -0.266 |
Chick | -0.263 |
Grac | -0.255 |
light | -0.255 |
burn | -0.248 |
Token | slowly |
Token ID | 6364 |
Feature activation | +5.26e-01 |
Pos logit contributions | |
booth | +0.008 |
inconvenience | +0.008 |
tickets | +0.007 |
employee | +0.007 |
newspapers | +0.007 |
Neg logit contributions | |
Grac | -0.009 |
Chick | -0.009 |
apy | -0.009 |
ubb | -0.009 |
stub | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
booth | +0.022 |
Tomas | +0.022 |
employee | +0.022 |
cheating | +0.022 |
inconvenience | +0.021 |
Neg logit contributions | |
berry | -0.028 |
light | -0.027 |
stub | -0.027 |
Chick | -0.026 |
Grac | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
cheating | +0.024 |
booth | +0.023 |
employee | +0.022 |
tickets | +0.022 |
Tomas | +0.022 |
Neg logit contributions | |
Chick | -0.027 |
berry | -0.026 |
Grac | -0.026 |
stub | -0.026 |
light | -0.024 |
Token | counted |
Token ID | 14789 |
Feature activation | +4.92e+00 |
Pos logit contributions | |
booth | +0.016 |
inconvenience | +0.015 |
cheating | +0.015 |
employee | +0.015 |
Tomas | +0.015 |
Neg logit contributions | |
stub | -0.022 |
Chick | -0.022 |
berry | -0.022 |
light | -0.022 |
Grac | -0.021 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.35e+00 |
Pos logit contributions | |
inconvenience | +0.212 |
booth | +0.209 |
employee | +0.205 |
cheating | +0.201 |
Tomas | +0.195 |
Neg logit contributions | |
berry | -0.266 |
Chick | -0.262 |
light | -0.255 |
Grac | -0.254 |
burn | -0.248 |
Token | , |
Token ID | 11 |
Feature activation | -9.18e-02 |
Pos logit contributions | |
inconvenience | +0.465 |
booth | +0.419 |
employee | +0.419 |
Tomas | +0.405 |
criminal | +0.399 |
Neg logit contributions | |
berry | -0.482 |
light | -0.460 |
ie | -0.420 |
flower | -0.410 |
Grac | -0.404 |
Token | 2 |
Token ID | 17 |
Feature activation | +8.86e+00 |
Pos logit contributions | |
Grac | +0.006 |
berry | +0.006 |
Chick | +0.006 |
stub | +0.005 |
skirt | +0.005 |
Neg logit contributions | |
inconvenience | -0.003 |
Tomas | -0.003 |
cheating | -0.003 |
employee | -0.003 |
booth | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
booth | +0.428 |
employee | +0.417 |
inconvenience | +0.416 |
Tomas | +0.410 |
cheating | +0.396 |
Neg logit contributions | |
berry | -0.439 |
light | -0.413 |
Chick | -0.401 |
Grac | -0.391 |
ie | -0.388 |
Token | 3 |
Token ID | 18 |
Feature activation | +7.07e+00 |
Pos logit contributions | |
berry | +0.005 |
Grac | +0.005 |
Chick | +0.005 |
light | +0.005 |
stub | +0.005 |
Neg logit contributions | |
inconvenience | -0.003 |
Tomas | -0.003 |
cheating | -0.003 |
booth | -0.003 |
employee | -0.003 |
Token | ... |
Token ID | 986 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
booth | +0.343 |
employee | +0.338 |
inconvenience | +0.326 |
Tomas | +0.325 |
cheating | +0.321 |
Neg logit contributions | |
berry | -0.363 |
light | -0.329 |
Chick | -0.327 |
Grac | -0.320 |
streamed | -0.311 |
Token | . |
Token ID | 13 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
stub | +0.028 |
Chick | +0.028 |
burn | +0.027 |
berry | +0.027 |
light | +0.027 |
Neg logit contributions | |
booth | -0.019 |
Tom | -0.019 |
cheating | -0.018 |
tickets | -0.018 |
worker | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
burn | +0.018 |
stub | +0.018 |
Chick | +0.018 |
numb | +0.017 |
light | +0.017 |
Neg logit contributions | |
Tom | -0.012 |
booth | -0.012 |
Tomas | -0.011 |
cheating | -0.010 |
employee | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -4.51e-02 |
Pos logit contributions | |
stub | +0.006 |
Chick | +0.006 |
berry | +0.006 |
burn | +0.006 |
light | +0.006 |
Neg logit contributions | |
booth | -0.004 |
cheating | -0.004 |
Tom | -0.004 |
employee | -0.004 |
tickets | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
stub | +0.003 |
Chick | +0.002 |
berry | +0.002 |
light | +0.002 |
burn | +0.002 |
Neg logit contributions | |
booth | -0.002 |
cheating | -0.002 |
Tom | -0.002 |
tickets | -0.002 |
employee | -0.002 |
Token | zo |
Token ID | 40565 |
Feature activation | +3.82e+00 |
Pos logit contributions | |
booth | +0.023 |
Tom | +0.023 |
Tomas | +0.023 |
cheating | +0.022 |
employee | +0.022 |
Neg logit contributions | |
stub | -0.028 |
berry | -0.027 |
Chick | -0.026 |
skirt | -0.026 |
burn | -0.026 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.34e+00 |
Pos logit contributions | |
Tomas | +0.100 |
booth | +0.098 |
newspapers | +0.096 |
her | +0.096 |
Tom | +0.094 |
Neg logit contributions | |
berry | -0.259 |
stub | -0.250 |
maid | -0.250 |
Chick | -0.249 |
light | -0.248 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
booth | +0.431 |
Tomas | +0.427 |
employee | +0.410 |
newspapers | +0.404 |
Tom | +0.404 |
Neg logit contributions | |
berry | -0.461 |
ie | -0.445 |
light | -0.441 |
stub | -0.440 |
maid | -0.436 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
booth | +0.098 |
cheating | +0.094 |
employee | +0.094 |
Tomas | +0.094 |
Tom | +0.092 |
Neg logit contributions | |
stub | -0.129 |
berry | -0.127 |
Chick | -0.126 |
burn | -0.122 |
light | -0.122 |
Token | to |
Token ID | 284 |
Feature activation | -3.32e-01 |
Pos logit contributions | |
Grac | +0.039 |
stub | +0.038 |
asleep | +0.038 |
Chick | +0.038 |
burn | +0.037 |
Neg logit contributions | |
cheating | -0.063 |
inconvenience | -0.061 |
site | -0.059 |
booth | -0.059 |
2 | -0.059 |
Token | the |
Token ID | 262 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
Chick | +0.018 |
Grac | +0.018 |
berry | +0.017 |
stub | +0.017 |
M | +0.016 |
Neg logit contributions | |
cheating | -0.014 |
booth | -0.014 |
her | -0.013 |
employee | -0.013 |
Tomas | -0.013 |
Token | pipe |
Token ID | 12656 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Tomas | +0.035 |
Tom | +0.034 |
edy | +0.033 |
cheating | +0.032 |
her | +0.031 |
Neg logit contributions | |
stub | -0.045 |
berry | -0.044 |
Chick | -0.043 |
Grac | -0.042 |
toe | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +7.67e-01 |
Pos logit contributions | |
booth | +0.183 |
employee | +0.176 |
inconvenience | +0.171 |
nine | +0.168 |
worker | +0.167 |
Neg logit contributions | |
Grac | -0.186 |
light | -0.181 |
stub | -0.181 |
berry | -0.180 |
streamed | -0.177 |
Token | |
Token ID | 220 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
cheating | +0.036 |
booth | +0.035 |
worker | +0.034 |
employee | +0.034 |
site | +0.034 |
Neg logit contributions | |
Chick | -0.039 |
stub | -0.038 |
berry | -0.038 |
Grac | -0.038 |
streamed | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Chick | +0.148 |
Grac | +0.143 |
stub | +0.142 |
berry | +0.140 |
oppy | +0.135 |
Neg logit contributions | |
cheating | -0.151 |
booth | -0.148 |
employee | -0.144 |
worker | -0.141 |
inconvenience | -0.140 |
Token |
|
Token ID | 198 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
cheating | +0.005 |
booth | +0.005 |
tickets | +0.005 |
employee | +0.005 |
worker | +0.005 |
Neg logit contributions | |
stub | -0.005 |
Chick | -0.005 |
Grac | -0.005 |
berry | -0.005 |
burn | -0.005 |
Token | Friend |
Token ID | 23331 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
booth | +0.005 |
cheating | +0.005 |
Tom | +0.005 |
tickets | +0.005 |
money | +0.005 |
Neg logit contributions | |
stub | -0.006 |
Chick | -0.006 |
berry | -0.006 |
burn | -0.006 |
Grac | -0.006 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.32e+00 |
Pos logit contributions | |
site | +0.051 |
cheating | +0.048 |
inconvenience | +0.047 |
Tomas | +0.046 |
booth | +0.045 |
Neg logit contributions | |
Grac | -0.099 |
berry | -0.098 |
maid | -0.093 |
streamed | -0.092 |
als | -0.092 |
Token | : |
Token ID | 25 |
Feature activation | +5.02e+00 |
Pos logit contributions | |
cheating | +0.396 |
inconvenience | +0.395 |
ours | +0.394 |
site | +0.394 |
clues | +0.393 |
Neg logit contributions | |
berry | -0.492 |
ie | -0.449 |
light | -0.444 |
Grac | -0.444 |
stub | -0.442 |
Token | � |
Token ID | 6184 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
cheating | +0.185 |
booth | +0.184 |
employee | +0.178 |
tickets | +0.177 |
money | +0.176 |
Neg logit contributions | |
berry | -0.292 |
stub | -0.289 |
Chick | -0.284 |
Grac | -0.282 |
burn | -0.277 |
Token | � |
Token ID | 95 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
money | +0.037 |
bills | +0.036 |
clues | +0.036 |
cheating | +0.035 |
Tom | +0.035 |
Neg logit contributions | |
berry | -0.040 |
Grac | -0.039 |
oppy | -0.038 |
Chick | -0.037 |
days | -0.036 |
Token | € |
Token ID | 26391 |
Feature activation | +3.32e+00 |
Pos logit contributions | |
light | +0.001 |
days | +0.001 |
c | +0.001 |
ci | +0.001 |
berry | +0.001 |
Neg logit contributions | |
money | -0.001 |
tickets | -0.001 |
clues | -0.001 |
bills | -0.001 |
Tom | -0.001 |
Token | � |
Token ID | 129 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
booth | +0.175 |
Tom | +0.170 |
cheating | +0.169 |
employee | +0.166 |
Tomas | +0.165 |
Neg logit contributions | |
stub | -0.150 |
Chick | -0.146 |
burn | -0.143 |
berry | -0.142 |
light | -0.140 |
Token | then |
Token ID | 788 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
booth | +0.042 |
cheating | +0.039 |
Tom | +0.038 |
employee | +0.038 |
tickets | +0.038 |
Neg logit contributions | |
stub | -0.098 |
Chick | -0.097 |
burn | -0.095 |
berry | -0.094 |
light | -0.094 |
Token | , |
Token ID | 11 |
Feature activation | -8.97e-02 |
Pos logit contributions | |
booth | +0.088 |
Tom | +0.085 |
cheating | +0.083 |
employee | +0.082 |
tickets | +0.081 |
Neg logit contributions | |
stub | -0.138 |
Chick | -0.135 |
burn | -0.132 |
berry | -0.132 |
light | -0.131 |
Token | a |
Token ID | 257 |
Feature activation | +7.07e-01 |
Pos logit contributions | |
stub | +0.006 |
Chick | +0.005 |
burn | +0.005 |
berry | +0.005 |
light | +0.005 |
Neg logit contributions | |
booth | -0.003 |
Tom | -0.003 |
cheating | -0.003 |
employee | -0.003 |
tickets | -0.003 |
Token | kind |
Token ID | 1611 |
Feature activation | -6.65e-01 |
Pos logit contributions | |
booth | +0.031 |
Tom | +0.030 |
cheating | +0.029 |
Tomas | +0.029 |
tickets | +0.029 |
Neg logit contributions | |
stub | -0.039 |
Chick | -0.037 |
berry | -0.036 |
burn | -0.036 |
Grac | -0.036 |
Token | zo |
Token ID | 40565 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
stub | +0.039 |
Chick | +0.038 |
berry | +0.038 |
Grac | +0.037 |
burn | +0.036 |
Neg logit contributions | |
cheating | -0.026 |
booth | -0.026 |
Tomas | -0.025 |
tickets | -0.025 |
Tom | -0.024 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.27e+00 |
Pos logit contributions | |
booth | +0.096 |
Tom | +0.093 |
Tomas | +0.093 |
her | +0.089 |
employee | +0.089 |
Neg logit contributions | |
stub | -0.251 |
berry | -0.247 |
Chick | -0.242 |
Grac | -0.240 |
light | -0.239 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
booth | +0.351 |
Tom | +0.335 |
Tomas | +0.326 |
employee | +0.326 |
cheating | +0.322 |
Neg logit contributions | |
stub | -0.559 |
Chick | -0.544 |
burn | -0.543 |
berry | -0.540 |
light | -0.533 |
Token | gave |
Token ID | 2921 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
booth | +0.128 |
Tom | +0.123 |
cheating | +0.123 |
employee | +0.123 |
Tomas | +0.123 |
Neg logit contributions | |
stub | -0.091 |
Chick | -0.087 |
berry | -0.086 |
burn | -0.085 |
light | -0.084 |
Token | Lily |
Token ID | 20037 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
Chick | +0.075 |
stub | +0.074 |
berry | +0.072 |
Grac | +0.072 |
burn | +0.072 |
Neg logit contributions | |
booth | -0.058 |
cheating | -0.055 |
employee | -0.053 |
site | -0.053 |
Tomas | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | +7.63e-01 |
Pos logit contributions | |
booth | +0.004 |
Tom | +0.004 |
tickets | +0.003 |
cheating | +0.003 |
money | +0.003 |
Neg logit contributions | |
stub | -0.010 |
Chick | -0.010 |
burn | -0.009 |
berry | -0.009 |
light | -0.009 |
Token | new |
Token ID | 649 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
booth | +0.034 |
Tom | +0.032 |
cheating | +0.031 |
worker | +0.031 |
tickets | +0.031 |
Neg logit contributions | |
stub | -0.041 |
Chick | -0.041 |
burn | -0.040 |
light | -0.040 |
berry | -0.039 |
Token | names |
Token ID | 3891 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
stub | +0.041 |
Chick | +0.040 |
burn | +0.040 |
light | +0.039 |
berry | +0.039 |
Neg logit contributions | |
booth | -0.029 |
Tom | -0.028 |
cheating | -0.027 |
tickets | -0.027 |
employee | -0.027 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
booth | +0.008 |
Tom | +0.008 |
cheating | +0.008 |
tickets | +0.008 |
employee | +0.007 |
Neg logit contributions | |
stub | -0.009 |
Chick | -0.009 |
burn | -0.008 |
berry | -0.008 |
light | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +9.33e-02 |
Pos logit contributions | |
stub | +0.081 |
Chick | +0.080 |
berry | +0.078 |
Grac | +0.077 |
light | +0.075 |
Neg logit contributions | |
cheating | -0.070 |
booth | -0.070 |
employee | -0.067 |
tickets | -0.066 |
Tomas | -0.065 |
Token |
|
Token ID | 198 |
Feature activation | +9.98e-02 |
Pos logit contributions | |
cheating | +0.004 |
booth | +0.004 |
employee | +0.004 |
tickets | +0.004 |
Tomas | +0.004 |
Neg logit contributions | |
stub | -0.005 |
Chick | -0.005 |
berry | -0.005 |
Grac | -0.005 |
light | -0.004 |
Token | The |
Token ID | 464 |
Feature activation | +8.51e-02 |
Pos logit contributions | |
booth | +0.005 |
cheating | +0.005 |
Tom | +0.005 |
tickets | +0.005 |
money | +0.005 |
Neg logit contributions | |
stub | -0.005 |
berry | -0.005 |
Chick | -0.005 |
Grac | -0.005 |
light | -0.004 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.26e+00 |
Pos logit contributions | |
booth | +0.004 |
Tom | +0.004 |
cheating | +0.004 |
Tomas | +0.004 |
employee | +0.004 |
Neg logit contributions | |
stub | -0.004 |
Chick | -0.004 |
berry | -0.004 |
burn | -0.004 |
Grac | -0.004 |
Token | st |
Token ID | 301 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
inconvenience | +0.477 |
clues | +0.470 |
cheating | +0.466 |
ours | +0.466 |
newspapers | +0.455 |
Neg logit contributions | |
berry | -0.444 |
ie | -0.429 |
light | -0.410 |
flower | -0.389 |
c | -0.388 |
Token | person |
Token ID | 1048 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
stub | +0.139 |
Chick | +0.135 |
burn | +0.133 |
berry | +0.130 |
light | +0.129 |
Neg logit contributions | |
booth | -0.140 |
Tom | -0.136 |
cheating | -0.133 |
employee | -0.132 |
tickets | -0.131 |
Token | said |
Token ID | 531 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
booth | +0.041 |
Tom | +0.039 |
cheating | +0.039 |
employee | +0.038 |
Tomas | +0.038 |
Neg logit contributions | |
stub | -0.057 |
Chick | -0.055 |
berry | -0.055 |
light | -0.054 |
Grac | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | -1.86e-01 |
Pos logit contributions | |
booth | +0.088 |
Tom | +0.088 |
worker | +0.082 |
cheating | +0.079 |
employee | +0.079 |
Neg logit contributions | |
stub | -0.170 |
burn | -0.166 |
Chick | -0.164 |
light | -0.161 |
numb | -0.158 |
Token | " |
Token ID | 366 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Chick | +0.011 |
stub | +0.011 |
berry | +0.011 |
Grac | +0.011 |
burn | +0.010 |
Neg logit contributions | |
booth | -0.007 |
cheating | -0.007 |
tickets | -0.007 |
Tom | -0.007 |
employee | -0.007 |
Token | went |
Token ID | 1816 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
booth | +0.062 |
cheating | +0.059 |
Tomas | +0.059 |
inconvenience | +0.059 |
tickets | +0.058 |
Neg logit contributions | |
stub | -0.082 |
berry | -0.081 |
light | -0.080 |
Chick | -0.079 |
streamed | -0.076 |
Token | first |
Token ID | 717 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
asleep | +0.026 |
stub | +0.026 |
Grac | +0.025 |
berry | +0.025 |
skirt | +0.024 |
Neg logit contributions | |
inconvenience | -0.039 |
edy | -0.038 |
bills | -0.037 |
Tom | -0.037 |
booth | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +4.16e-01 |
Pos logit contributions | |
inconvenience | +0.161 |
booth | +0.161 |
edy | +0.160 |
cheating | +0.159 |
employee | +0.159 |
Neg logit contributions | |
berry | -0.146 |
stub | -0.142 |
light | -0.141 |
days | -0.131 |
skirt | -0.130 |
Token | He |
Token ID | 679 |
Feature activation | +2.87e-01 |
Pos logit contributions | |
cheating | +0.019 |
booth | +0.019 |
tickets | +0.018 |
inconvenience | +0.018 |
employee | +0.018 |
Neg logit contributions | |
Chick | -0.021 |
Grac | -0.021 |
berry | -0.021 |
stub | -0.021 |
light | -0.020 |
Token | counted |
Token ID | 14789 |
Feature activation | +4.79e+00 |
Pos logit contributions | |
booth | +0.011 |
cheating | +0.011 |
inconvenience | +0.011 |
Tomas | +0.011 |
tickets | +0.010 |
Neg logit contributions | |
stub | -0.017 |
Chick | -0.016 |
light | -0.016 |
berry | -0.016 |
Grac | -0.016 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.25e+00 |
Pos logit contributions | |
booth | +0.205 |
inconvenience | +0.201 |
employee | +0.200 |
cheating | +0.196 |
Tomas | +0.193 |
Neg logit contributions | |
Chick | -0.257 |
berry | -0.256 |
light | -0.248 |
Grac | -0.244 |
burn | -0.242 |
Token | , |
Token ID | 11 |
Feature activation | -2.53e-01 |
Pos logit contributions | |
inconvenience | +0.493 |
booth | +0.441 |
employee | +0.440 |
Tomas | +0.436 |
ours | +0.426 |
Neg logit contributions | |
berry | -0.441 |
light | -0.423 |
ie | -0.395 |
flower | -0.374 |
c | -0.370 |
Token | 2 |
Token ID | 17 |
Feature activation | +8.77e+00 |
Pos logit contributions | |
Chick | +0.016 |
Grac | +0.016 |
berry | +0.016 |
stub | +0.015 |
skirt | +0.015 |
Neg logit contributions | |
inconvenience | -0.008 |
Tomas | -0.008 |
booth | -0.008 |
employee | -0.008 |
cheating | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -2.39e-01 |
Pos logit contributions | |
booth | +0.441 |
employee | +0.430 |
inconvenience | +0.428 |
Tomas | +0.428 |
cheating | +0.413 |
Neg logit contributions | |
berry | -0.420 |
light | -0.388 |
Chick | -0.385 |
ie | -0.376 |
Grac | -0.363 |
Token | 3 |
Token ID | 18 |
Feature activation | +6.98e+00 |
Pos logit contributions | |
berry | +0.014 |
Chick | +0.014 |
Grac | +0.014 |
light | +0.013 |
stub | +0.013 |
Neg logit contributions | |
inconvenience | -0.009 |
Tomas | -0.009 |
booth | -0.008 |
cheating | -0.008 |
employee | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
booth | +0.340 |
employee | +0.332 |
Tomas | +0.323 |
inconvenience | +0.317 |
cheating | +0.317 |
Neg logit contributions | |
berry | -0.357 |
Chick | -0.330 |
light | -0.322 |
Grac | -0.314 |
ie | -0.313 |
Token | homes |
Token ID | 5682 |
Feature activation | +8.05e-02 |
Pos logit contributions | |
Tom | +0.006 |
booth | +0.006 |
cheating | +0.006 |
employee | +0.005 |
worker | +0.005 |
Neg logit contributions | |
stub | -0.013 |
Chick | -0.013 |
berry | -0.013 |
burn | -0.013 |
Grac | -0.013 |
Token | too |
Token ID | 1165 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
booth | +0.003 |
Tom | +0.002 |
money | +0.002 |
site | +0.002 |
worker | +0.002 |
Neg logit contributions | |
Chick | -0.005 |
stub | -0.005 |
burn | -0.005 |
berry | -0.005 |
Grac | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | +2.04e-01 |
Pos logit contributions | |
booth | +0.092 |
Tom | +0.089 |
cheating | +0.087 |
employee | +0.085 |
Tomas | +0.084 |
Neg logit contributions | |
stub | -0.148 |
Chick | -0.145 |
burn | -0.142 |
berry | -0.141 |
light | -0.140 |
Token | The |
Token ID | 383 |
Feature activation | -3.50e-01 |
Pos logit contributions | |
Tom | +0.008 |
booth | +0.008 |
Tomas | +0.007 |
employee | +0.007 |
cheating | +0.007 |
Neg logit contributions | |
burn | -0.012 |
stub | -0.012 |
Chick | -0.012 |
numb | -0.012 |
light | -0.012 |
Token | zo |
Token ID | 40565 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
stub | +0.018 |
Chick | +0.018 |
berry | +0.018 |
burn | +0.017 |
Grac | +0.017 |
Neg logit contributions | |
booth | -0.016 |
Tom | -0.016 |
cheating | -0.015 |
Tomas | -0.015 |
employee | -0.015 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.25e+00 |
Pos logit contributions | |
Tomas | +0.092 |
booth | +0.092 |
Tom | +0.089 |
newspapers | +0.088 |
employee | +0.087 |
Neg logit contributions | |
stub | -0.242 |
berry | -0.240 |
Chick | -0.234 |
Grac | -0.234 |
light | -0.233 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
booth | +0.393 |
Tomas | +0.383 |
employee | +0.374 |
Tom | +0.372 |
newspapers | +0.362 |
Neg logit contributions | |
stub | -0.487 |
berry | -0.482 |
burn | -0.470 |
light | -0.463 |
maid | -0.457 |
Token | was |
Token ID | 373 |
Feature activation | -2.63e-01 |
Pos logit contributions | |
booth | +0.096 |
Tom | +0.092 |
cheating | +0.092 |
employee | +0.090 |
Tomas | +0.090 |
Neg logit contributions | |
stub | -0.136 |
Chick | -0.133 |
berry | -0.130 |
burn | -0.130 |
light | -0.128 |
Token | happy |
Token ID | 3772 |
Feature activation | -3.04e-01 |
Pos logit contributions | |
stub | +0.016 |
Chick | +0.015 |
burn | +0.015 |
Grac | +0.015 |
berry | +0.015 |
Neg logit contributions | |
booth | -0.010 |
Tom | -0.009 |
cheating | -0.009 |
tickets | -0.009 |
employee | -0.009 |
Token | that |
Token ID | 326 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
stub | +0.020 |
Chick | +0.020 |
burn | +0.020 |
berry | +0.019 |
light | +0.019 |
Neg logit contributions | |
booth | -0.010 |
Tom | -0.009 |
cheating | -0.009 |
employee | -0.009 |
Tomas | -0.009 |
Token | everyone |
Token ID | 2506 |
Feature activation | +2.28e+00 |
Pos logit contributions | |
booth | +0.001 |
employee | +0.001 |
cheating | +0.001 |
Tom | +0.001 |
tickets | +0.001 |
Neg logit contributions | |
stub | -0.002 |
Chick | -0.002 |
Grac | -0.002 |
berry | -0.002 |
burn | -0.002 |
Token | the |
Token ID | 262 |
Feature activation | +4.68e-01 |
Pos logit contributions | |
booth | +0.183 |
employee | +0.177 |
Tomas | +0.169 |
cheating | +0.169 |
inconvenience | +0.166 |
Neg logit contributions | |
Chick | -0.216 |
berry | -0.210 |
stub | -0.210 |
Grac | -0.209 |
burn | -0.205 |
Token | cars |
Token ID | 5006 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
booth | +0.022 |
employee | +0.021 |
Tomas | +0.021 |
Tom | +0.020 |
tickets | +0.020 |
Neg logit contributions | |
stub | -0.025 |
skirt | -0.023 |
thumb | -0.022 |
Chick | -0.022 |
burn | -0.022 |
Token | in |
Token ID | 287 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
booth | +0.153 |
employee | +0.151 |
Tomas | +0.150 |
resa | +0.149 |
Officer | +0.144 |
Neg logit contributions | |
light | -0.172 |
stub | -0.171 |
burn | -0.171 |
berry | -0.170 |
Chick | -0.167 |
Token | order |
Token ID | 1502 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
booth | +0.046 |
employee | +0.044 |
Tomas | +0.041 |
cheating | +0.041 |
worker | +0.041 |
Neg logit contributions | |
Chick | -0.068 |
stub | -0.067 |
berry | -0.065 |
light | -0.065 |
Grac | -0.065 |
Token | from |
Token ID | 422 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
resa | +0.084 |
booth | +0.083 |
Tomas | +0.082 |
employee | +0.082 |
cheating | +0.080 |
Neg logit contributions | |
Chick | -0.086 |
light | -0.085 |
stub | -0.085 |
berry | -0.084 |
streamed | -0.083 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.25e+00 |
Pos logit contributions | |
booth | +0.087 |
employee | +0.082 |
inconvenience | +0.077 |
cheating | +0.076 |
Tomas | +0.076 |
Neg logit contributions | |
Chick | -0.104 |
stub | -0.102 |
burn | -0.097 |
berry | -0.097 |
Grac | -0.097 |
Token | to |
Token ID | 284 |
Feature activation | -7.37e-01 |
Pos logit contributions | |
Tomas | +0.380 |
employee | +0.378 |
booth | +0.369 |
her | +0.364 |
Officer | +0.361 |
Neg logit contributions | |
light | -0.480 |
berry | -0.476 |
c | -0.439 |
burn | -0.438 |
stub | -0.428 |
Token | 10 |
Token ID | 838 |
Feature activation | +4.84e+00 |
Pos logit contributions | |
stub | +0.043 |
Chick | +0.043 |
berry | +0.042 |
light | +0.042 |
Grac | +0.041 |
Neg logit contributions | |
booth | -0.027 |
employee | -0.026 |
Tomas | -0.026 |
cheating | -0.024 |
Tom | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
booth | +0.197 |
employee | +0.194 |
Tomas | +0.192 |
Tom | +0.187 |
Officer | +0.181 |
Neg logit contributions | |
stub | -0.264 |
berry | -0.260 |
Chick | -0.255 |
light | -0.254 |
skirt | -0.247 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
cheating | +0.010 |
clues | +0.010 |
tickets | +0.010 |
booth | +0.010 |
employee | +0.009 |
Neg logit contributions | |
Chick | -0.007 |
Grac | -0.006 |
berry | -0.006 |
M | -0.005 |
stub | -0.005 |
Token | my |
Token ID | 1820 |
Feature activation | -7.50e-01 |
Pos logit contributions | |
booth | +0.113 |
Tom | +0.109 |
cheating | +0.107 |
employee | +0.107 |
tickets | +0.106 |
Neg logit contributions | |
stub | -0.098 |
Chick | -0.095 |
burn | -0.094 |
berry | -0.093 |
light | -0.091 |
Token | around |
Token ID | 1088 |
Feature activation | -7.92e-01 |
Pos logit contributions | |
booth | +0.003 |
cheating | +0.003 |
Tom | +0.003 |
site | +0.003 |
employee | +0.003 |
Neg logit contributions | |
stub | -0.003 |
Chick | -0.003 |
burn | -0.003 |
berry | -0.003 |
Grac | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
stub | +0.048 |
Chick | +0.047 |
berry | +0.046 |
Grac | +0.045 |
light | +0.045 |
Neg logit contributions | |
booth | -0.030 |
cheating | -0.028 |
employee | -0.028 |
tickets | -0.028 |
Tom | -0.028 |
Token | house |
Token ID | 2156 |
Feature activation | -9.11e-02 |
Pos logit contributions | |
nine | +0.031 |
cheating | +0.030 |
40 | +0.029 |
Tomas | +0.028 |
inconvenience | +0.027 |
Neg logit contributions | |
berry | -0.040 |
stub | -0.039 |
skirt | -0.038 |
burn | -0.037 |
c | -0.036 |
Token | , |
Token ID | 11 |
Feature activation | -1.57e-01 |
Pos logit contributions | |
stub | +0.006 |
Chick | +0.006 |
burn | +0.006 |
berry | +0.005 |
Grac | +0.005 |
Neg logit contributions | |
booth | -0.003 |
Tom | -0.003 |
cheating | -0.003 |
worker | -0.003 |
employee | -0.003 |
Token | counting |
Token ID | 14143 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
stub | +0.009 |
Chick | +0.009 |
burn | +0.009 |
berry | +0.009 |
light | +0.009 |
Neg logit contributions | |
booth | -0.006 |
Tom | -0.006 |
cheating | -0.006 |
employee | -0.006 |
Tomas | -0.006 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.24e+00 |
Pos logit contributions | |
booth | +0.107 |
cheating | +0.102 |
employee | +0.101 |
Tom | +0.101 |
Tomas | +0.098 |
Neg logit contributions | |
stub | -0.173 |
Chick | -0.171 |
berry | -0.169 |
burn | -0.167 |
light | -0.167 |
Token | , |
Token ID | 11 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
inconvenience | +0.369 |
employee | +0.361 |
booth | +0.358 |
her | +0.354 |
site | +0.352 |
Neg logit contributions | |
berry | -0.508 |
light | -0.500 |
burn | -0.465 |
Chick | -0.452 |
stub | -0.448 |
Token | 2 |
Token ID | 362 |
Feature activation | +7.71e+00 |
Pos logit contributions | |
stub | +0.010 |
berry | +0.010 |
Chick | +0.010 |
Grac | +0.010 |
skirt | +0.009 |
Neg logit contributions | |
booth | -0.005 |
cheating | -0.005 |
employee | -0.005 |
Tomas | -0.005 |
inconvenience | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
booth | +0.303 |
employee | +0.288 |
Tom | +0.284 |
Tomas | +0.282 |
site | +0.282 |
Neg logit contributions | |
berry | -0.438 |
Chick | -0.431 |
stub | -0.431 |
light | -0.424 |
burn | -0.419 |
Token | 3 |
Token ID | 513 |
Feature activation | +5.55e-01 |
Pos logit contributions | |
stub | +0.009 |
berry | +0.008 |
Chick | +0.008 |
Grac | +0.008 |
light | +0.008 |
Neg logit contributions | |
booth | -0.005 |
cheating | -0.005 |
Tomas | -0.005 |
employee | -0.005 |
Tom | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
booth | +0.023 |
Tom | +0.022 |
cheating | +0.022 |
employee | +0.021 |
Tomas | +0.021 |
Neg logit contributions | |
stub | -0.031 |
Chick | -0.031 |
berry | -0.030 |
burn | -0.030 |
light | -0.030 |
Token | counted |
Token ID | 14789 |
Feature activation | +4.91e+00 |
Pos logit contributions | |
Tom | +0.024 |
booth | +0.022 |
Tomas | +0.020 |
employee | +0.019 |
tickets | +0.019 |
Neg logit contributions | |
stub | -0.043 |
burn | -0.042 |
berry | -0.042 |
Chick | -0.041 |
numb | -0.041 |
Token | her |
Token ID | 607 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Tom | +0.171 |
booth | +0.164 |
tickets | +0.156 |
numbers | +0.152 |
money | +0.148 |
Neg logit contributions | |
stub | -0.332 |
burn | -0.307 |
Chick | -0.303 |
streamed | -0.301 |
light | -0.298 |
Token | lucky |
Token ID | 9670 |
Feature activation | +2.61e-01 |
Pos logit contributions | |
stub | +0.082 |
Chick | +0.080 |
burn | +0.078 |
berry | +0.077 |
light | +0.077 |
Neg logit contributions | |
booth | -0.084 |
Tom | -0.082 |
cheating | -0.081 |
employee | -0.079 |
tickets | -0.079 |
Token | stars |
Token ID | 5788 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
booth | +0.013 |
Tom | +0.012 |
money | +0.012 |
worker | +0.012 |
tickets | +0.012 |
Neg logit contributions | |
Chick | -0.013 |
stub | -0.013 |
burn | -0.013 |
light | -0.012 |
Grac | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +4.96e-01 |
Pos logit contributions | |
booth | +0.059 |
Tom | +0.059 |
worker | +0.055 |
tickets | +0.053 |
cheating | +0.053 |
Neg logit contributions | |
stub | -0.122 |
Chick | -0.117 |
burn | -0.115 |
Grac | -0.111 |
numb | -0.110 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.23e+00 |
Pos logit contributions | |
Tom | +0.023 |
booth | +0.023 |
Tomas | +0.021 |
cheating | +0.021 |
employee | +0.021 |
Neg logit contributions | |
stub | -0.026 |
burn | -0.026 |
Chick | -0.025 |
light | -0.025 |
berry | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -2.51e-01 |
Pos logit contributions | |
booth | +0.378 |
cheating | +0.364 |
employee | +0.355 |
Tomas | +0.352 |
Tom | +0.350 |
Neg logit contributions | |
berry | -0.521 |
Chick | -0.520 |
stub | -0.514 |
light | -0.509 |
burn | -0.503 |
Token | 2 |
Token ID | 362 |
Feature activation | +7.70e+00 |
Pos logit contributions | |
Chick | +0.015 |
Grac | +0.015 |
berry | +0.014 |
stub | +0.014 |
light | +0.014 |
Neg logit contributions | |
booth | -0.009 |
cheating | -0.009 |
Tomas | -0.009 |
inconvenience | -0.009 |
employee | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
booth | +0.315 |
Tom | +0.304 |
cheating | +0.299 |
employee | +0.294 |
Tomas | +0.292 |
Neg logit contributions | |
stub | -0.439 |
Chick | -0.433 |
berry | -0.423 |
burn | -0.422 |
light | -0.418 |
Token | 3 |
Token ID | 513 |
Feature activation | +5.74e-01 |
Pos logit contributions | |
Chick | +0.014 |
berry | +0.014 |
Grac | +0.014 |
stub | +0.014 |
light | +0.013 |
Neg logit contributions | |
booth | -0.008 |
cheating | -0.008 |
Tomas | -0.008 |
employee | -0.008 |
inconvenience | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
booth | +0.022 |
Tom | +0.022 |
cheating | +0.021 |
employee | +0.021 |
tickets | +0.021 |
Neg logit contributions | |
stub | -0.034 |
Chick | -0.033 |
burn | -0.033 |
berry | -0.032 |
light | -0.032 |
Token | He |
Token ID | 1544 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
cheating | +0.003 |
booth | +0.003 |
tickets | +0.003 |
Tom | +0.003 |
money | +0.003 |
Neg logit contributions | |
stub | -0.004 |
Chick | -0.004 |
lull | -0.003 |
berry | -0.003 |
light | -0.003 |
Token | looked |
Token ID | 3114 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
booth | +0.033 |
Tom | +0.032 |
cheating | +0.031 |
employee | +0.031 |
Tomas | +0.031 |
Neg logit contributions | |
stub | -0.034 |
Chick | -0.034 |
burn | -0.033 |
berry | -0.033 |
light | -0.032 |
Token | around |
Token ID | 1088 |
Feature activation | -7.90e-01 |
Pos logit contributions | |
stub | +0.038 |
Chick | +0.038 |
berry | +0.037 |
burn | +0.037 |
Grac | +0.036 |
Neg logit contributions | |
booth | -0.028 |
Tom | -0.027 |
cheating | -0.027 |
employee | -0.027 |
Tomas | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +6.38e-01 |
Pos logit contributions | |
stub | +0.051 |
Chick | +0.050 |
berry | +0.049 |
light | +0.049 |
Grac | +0.048 |
Neg logit contributions | |
booth | -0.026 |
cheating | -0.025 |
Tom | -0.024 |
employee | -0.024 |
Tomas | -0.024 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
booth | +0.026 |
Tom | +0.025 |
cheating | +0.024 |
employee | +0.024 |
Tomas | +0.024 |
Neg logit contributions | |
stub | -0.037 |
Chick | -0.036 |
burn | -0.036 |
berry | -0.036 |
light | -0.035 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.22e+00 |
Pos logit contributions | |
Tom | +0.004 |
booth | +0.004 |
tickets | +0.004 |
workers | +0.003 |
money | +0.003 |
Neg logit contributions | |
stub | -0.010 |
burn | -0.010 |
light | -0.009 |
Chick | -0.009 |
berry | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
inconvenience | +0.417 |
Ready | +0.401 |
Tomas | +0.393 |
her | +0.388 |
ours | +0.386 |
Neg logit contributions | |
berry | -0.500 |
light | -0.466 |
flower | -0.433 |
ie | -0.433 |
c | -0.427 |
Token | 2 |
Token ID | 362 |
Feature activation | +7.69e+00 |
Pos logit contributions | |
Grac | +0.014 |
Chick | +0.014 |
stub | +0.013 |
berry | +0.013 |
skirt | +0.012 |
Neg logit contributions | |
inconvenience | -0.013 |
booth | -0.013 |
site | -0.013 |
Tomas | -0.012 |
tickets | -0.012 |
Token | , |
Token ID | 11 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
booth | +0.324 |
Tomas | +0.312 |
cheating | +0.309 |
Tom | +0.308 |
employee | +0.305 |
Neg logit contributions | |
berry | -0.420 |
Chick | -0.411 |
stub | -0.401 |
light | -0.400 |
Grac | -0.394 |
Token | 3 |
Token ID | 513 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
Chick | +0.014 |
Grac | +0.014 |
berry | +0.014 |
stub | +0.013 |
skirt | +0.012 |
Neg logit contributions | |
booth | -0.012 |
inconvenience | -0.012 |
Tomas | -0.012 |
site | -0.011 |
cheating | -0.011 |
Token | people |
Token ID | 661 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
booth | +0.023 |
Tom | +0.022 |
cheating | +0.022 |
Tomas | +0.021 |
employee | +0.021 |
Neg logit contributions | |
stub | -0.024 |
Chick | -0.024 |
berry | -0.023 |
burn | -0.023 |
light | -0.023 |
Token | â |
Token ID | 22940 |
Feature activation | -8.08e-01 |
Pos logit contributions | |
booth | +0.032 |
Tom | +0.031 |
cheating | +0.031 |
employee | +0.030 |
Tomas | +0.030 |
Neg logit contributions | |
stub | -0.042 |
Chick | -0.041 |
burn | -0.040 |
berry | -0.040 |
light | -0.039 |
Token | € |
Token ID | 26391 |
Feature activation | +3.48e+00 |
Pos logit contributions | |
ci | +0.036 |
lull | +0.035 |
streamed | +0.035 |
burn | +0.035 |
c | +0.035 |
Neg logit contributions | |
tickets | -0.043 |
clues | -0.042 |
booth | -0.040 |
workers | -0.040 |
numbers | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | +1.91e-01 |
Pos logit contributions | |
booth | +0.176 |
Tom | +0.172 |
cheating | +0.169 |
employee | +0.167 |
Tomas | +0.165 |
Neg logit contributions | |
stub | -0.166 |
Chick | -0.162 |
burn | -0.158 |
berry | -0.156 |
light | -0.155 |
Token |
|
Token ID | 198 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
booth | +0.008 |
cheating | +0.007 |
Tom | +0.007 |
employee | +0.007 |
tickets | +0.007 |
Neg logit contributions | |
stub | -0.011 |
Chick | -0.011 |
berry | -0.011 |
burn | -0.011 |
Grac | -0.010 |
Token | Sm |
Token ID | 7556 |
Feature activation | +9.61e-01 |
Pos logit contributions | |
booth | +0.009 |
Tom | +0.009 |
cheating | +0.009 |
employee | +0.008 |
Tomas | +0.008 |
Neg logit contributions | |
stub | -0.013 |
Chick | -0.012 |
burn | -0.012 |
berry | -0.012 |
light | -0.012 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.20e+00 |
Pos logit contributions | |
booth | +0.044 |
tickets | +0.041 |
Tom | +0.041 |
Tomas | +0.041 |
money | +0.039 |
Neg logit contributions | |
stub | -0.051 |
berry | -0.050 |
Chick | -0.049 |
light | -0.048 |
streamed | -0.048 |
Token | y |
Token ID | 88 |
Feature activation | -3.81e+00 |
Pos logit contributions | |
booth | +0.389 |
Tom | +0.375 |
cheating | +0.367 |
employee | +0.362 |
Tomas | +0.361 |
Neg logit contributions | |
stub | -0.516 |
Chick | -0.505 |
burn | -0.497 |
berry | -0.494 |
light | -0.489 |
Token | smiled |
Token ID | 13541 |
Feature activation | +1.15e+00 |
Pos logit contributions | |
stub | +0.215 |
Chick | +0.212 |
burn | +0.209 |
berry | +0.206 |
light | +0.202 |
Neg logit contributions | |
booth | -0.158 |
Tom | -0.155 |
cheating | -0.149 |
employee | -0.148 |
worker | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
Tom | +0.022 |
booth | +0.019 |
old | +0.018 |
right | +0.017 |
employee | +0.016 |
Neg logit contributions | |
stub | -0.095 |
Chick | -0.091 |
burn | -0.090 |
seed | -0.089 |
berry | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
booth | +0.005 |
Tom | +0.005 |
cheating | +0.005 |
employee | +0.005 |
Tomas | +0.005 |
Neg logit contributions | |
stub | -0.008 |
Chick | -0.008 |
burn | -0.008 |
berry | -0.008 |
light | -0.008 |
Token | put |
Token ID | 1234 |
Feature activation | +3.95e-01 |
Pos logit contributions | |
Tom | +0.036 |
booth | +0.034 |
employee | +0.032 |
Tomas | +0.032 |
tickets | +0.031 |
Neg logit contributions | |
stub | -0.064 |
berry | -0.062 |
Chick | -0.062 |
burn | -0.061 |
light | -0.061 |
Token | your |
Token ID | 534 |
Feature activation | -3.72e-01 |
Pos logit contributions | |
booth | +0.042 |
Tom | +0.041 |
cheating | +0.040 |
tickets | +0.038 |
Tomas | +0.038 |
Neg logit contributions | |
stub | -0.083 |
Chick | -0.082 |
burn | -0.081 |
berry | -0.080 |
light | -0.080 |
Token | hand |
Token ID | 1021 |
Feature activation | -3.13e+00 |
Pos logit contributions | |
Chick | +0.019 |
burn | +0.018 |
stub | +0.018 |
Grac | +0.018 |
berry | +0.018 |
Neg logit contributions | |
booth | -0.018 |
Tom | -0.018 |
money | -0.018 |
tickets | -0.017 |
cheating | -0.017 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
stub | +0.187 |
Chick | +0.185 |
Grac | +0.178 |
burn | +0.176 |
berry | +0.176 |
Neg logit contributions | |
Tom | -0.124 |
booth | -0.122 |
money | -0.114 |
cheating | -0.114 |
her | -0.111 |
Token | the |
Token ID | 262 |
Feature activation | +9.38e-01 |
Pos logit contributions | |
stub | +0.088 |
burn | +0.087 |
Chick | +0.086 |
light | +0.084 |
berry | +0.084 |
Neg logit contributions | |
Tom | -0.058 |
booth | -0.056 |
Tomas | -0.052 |
cheating | -0.052 |
employee | -0.052 |
Token | zo |
Token ID | 40565 |
Feature activation | +3.47e+00 |
Pos logit contributions | |
booth | +0.043 |
Tom | +0.042 |
cheating | +0.042 |
Tomas | +0.041 |
employee | +0.041 |
Neg logit contributions | |
stub | -0.049 |
Chick | -0.048 |
berry | -0.047 |
burn | -0.046 |
Grac | -0.045 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.19e+00 |
Pos logit contributions | |
booth | +0.074 |
Tom | +0.071 |
Tomas | +0.068 |
cheating | +0.067 |
employee | +0.066 |
Neg logit contributions | |
stub | -0.260 |
berry | -0.257 |
Chick | -0.255 |
Grac | -0.251 |
light | -0.249 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
booth | +0.378 |
Tomas | +0.361 |
Tom | +0.353 |
employee | +0.351 |
cheating | +0.346 |
Neg logit contributions | |
stub | -0.510 |
berry | -0.509 |
burn | -0.498 |
Chick | -0.494 |
light | -0.490 |
Token | asked |
Token ID | 1965 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
booth | +0.076 |
Tom | +0.074 |
cheating | +0.071 |
employee | +0.069 |
Tomas | +0.069 |
Neg logit contributions | |
stub | -0.140 |
Chick | -0.137 |
burn | -0.135 |
berry | -0.133 |
light | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
booth | +0.065 |
Tom | +0.062 |
cheating | +0.062 |
employee | +0.060 |
Tomas | +0.060 |
Neg logit contributions | |
stub | -0.099 |
Chick | -0.098 |
berry | -0.096 |
burn | -0.095 |
light | -0.094 |
Token |
|
Token ID | 198 |
Feature activation | -7.73e-02 |
Pos logit contributions | |
cheating | +0.025 |
booth | +0.025 |
tickets | +0.024 |
employee | +0.024 |
site | +0.023 |
Neg logit contributions | |
Chick | -0.030 |
stub | -0.029 |
Grac | -0.029 |
berry | -0.029 |
light | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
stub | +0.004 |
Chick | +0.004 |
berry | +0.004 |
Grac | +0.004 |
burn | +0.004 |
Neg logit contributions | |
cheating | -0.003 |
booth | -0.003 |
tickets | -0.003 |
employee | -0.003 |
money | -0.003 |
Token | furry |
Token ID | 40857 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
stub | +0.015 |
Chick | +0.015 |
burn | +0.014 |
berry | +0.014 |
light | +0.014 |
Neg logit contributions | |
booth | -0.013 |
Tom | -0.013 |
cheating | -0.013 |
employee | -0.012 |
Tomas | -0.012 |
Token | legs |
Token ID | 7405 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
Chick | +0.105 |
stub | +0.105 |
burn | +0.101 |
light | +0.100 |
Grac | +0.099 |
Neg logit contributions | |
booth | -0.087 |
Tom | -0.087 |
money | -0.083 |
tickets | -0.083 |
cheating | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
stub | +0.259 |
Chick | +0.255 |
burn | +0.250 |
berry | +0.248 |
light | +0.246 |
Neg logit contributions | |
booth | -0.148 |
Tom | -0.143 |
cheating | -0.139 |
employee | -0.136 |
tickets | -0.134 |
Token | He |
Token ID | 679 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
cheating | +0.024 |
clues | +0.023 |
inconvenience | +0.023 |
tickets | +0.023 |
employee | +0.023 |
Neg logit contributions | |
stub | -0.018 |
Chick | -0.018 |
berry | -0.017 |
Grac | -0.016 |
M | -0.015 |
Token | counted |
Token ID | 14789 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
booth | +0.005 |
Tom | +0.005 |
cheating | +0.005 |
employee | +0.005 |
Tomas | +0.005 |
Neg logit contributions | |
stub | -0.006 |
Chick | -0.006 |
burn | -0.006 |
berry | -0.006 |
light | -0.006 |
Token | 1 |
Token ID | 352 |
Feature activation | +9.18e+00 |
Pos logit contributions | |
booth | +0.203 |
cheating | +0.196 |
employee | +0.195 |
inconvenience | +0.190 |
Tom | +0.189 |
Neg logit contributions | |
Chick | -0.251 |
stub | -0.249 |
berry | -0.247 |
burn | -0.244 |
light | -0.240 |
Token | , |
Token ID | 11 |
Feature activation | -3.29e-01 |
Pos logit contributions | |
inconvenience | +0.516 |
employee | +0.448 |
Ready | +0.447 |
ours | +0.440 |
damage | +0.439 |
Neg logit contributions | |
berry | -0.415 |
light | -0.388 |
ie | -0.373 |
c | -0.359 |
stretched | -0.352 |
Token | 2 |
Token ID | 362 |
Feature activation | +7.64e+00 |
Pos logit contributions | |
Chick | +0.021 |
Grac | +0.021 |
berry | +0.021 |
stub | +0.020 |
skirt | +0.019 |
Neg logit contributions | |
inconvenience | -0.011 |
site | -0.011 |
booth | -0.011 |
cheating | -0.011 |
employee | -0.011 |
Token | , |
Token ID | 11 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
inconvenience | +0.393 |
booth | +0.388 |
employee | +0.379 |
Ready | +0.369 |
Tomas | +0.365 |
Neg logit contributions | |
berry | -0.355 |
Chick | -0.328 |
ie | -0.324 |
light | -0.320 |
Grac | -0.315 |
Token | 3 |
Token ID | 513 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
Chick | +0.017 |
berry | +0.017 |
Grac | +0.017 |
stub | +0.017 |
burn | +0.016 |
Neg logit contributions | |
inconvenience | -0.010 |
booth | -0.010 |
site | -0.010 |
cheating | -0.010 |
employee | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -2.76e-01 |
Pos logit contributions | |
booth | +0.017 |
cheating | +0.016 |
Tomas | +0.016 |
employee | +0.016 |
Tom | +0.016 |
Neg logit contributions | |
Chick | -0.019 |
berry | -0.019 |
stub | -0.018 |
light | -0.018 |
Grac | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +6.59e-01 |
Pos logit contributions | |
stub | +0.012 |
Chick | +0.012 |
berry | +0.012 |
burn | +0.012 |
light | +0.012 |
Neg logit contributions | |
booth | -0.008 |
cheating | -0.008 |
Tom | -0.007 |
employee | -0.007 |
Tomas | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +6.80e-02 |
Pos logit contributions | |
booth | +0.029 |
cheating | +0.028 |
employee | +0.027 |
tickets | +0.027 |
worker | +0.027 |
Neg logit contributions | |
stub | -0.035 |
Chick | -0.035 |
berry | -0.034 |
Grac | -0.034 |
light | -0.033 |
Token |
|
Token ID | 198 |
Feature activation | +9.71e-02 |
Pos logit contributions | |
cheating | +0.003 |
booth | +0.003 |
bills | +0.003 |
employee | +0.003 |
tickets | +0.003 |
Neg logit contributions | |
stub | -0.003 |
Chick | -0.003 |
Grac | -0.003 |
berry | -0.003 |
light | -0.003 |
Token | The |
Token ID | 464 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
booth | +0.004 |
cheating | +0.004 |
Tom | +0.004 |
tickets | +0.004 |
money | +0.004 |
Neg logit contributions | |
stub | -0.005 |
Chick | -0.005 |
Grac | -0.005 |
berry | -0.005 |
light | -0.005 |
Token | zo |
Token ID | 40565 |
Feature activation | +3.54e+00 |
Pos logit contributions | |
Tomas | +0.010 |
Tom | +0.010 |
booth | +0.010 |
cheating | +0.010 |
40 | +0.009 |
Neg logit contributions | |
stub | -0.013 |
berry | -0.013 |
Chick | -0.012 |
Grac | -0.012 |
skirt | -0.012 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.18e+00 |
Pos logit contributions | |
Tomas | +0.096 |
booth | +0.095 |
newspapers | +0.094 |
Tom | +0.091 |
cheating | +0.091 |
Neg logit contributions | |
berry | -0.236 |
Grac | -0.229 |
Chick | -0.228 |
light | -0.227 |
stub | -0.226 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
booth | +0.427 |
Tomas | +0.413 |
Tom | +0.400 |
newspapers | +0.398 |
money | +0.396 |
Neg logit contributions | |
berry | -0.456 |
ie | -0.449 |
light | -0.439 |
flower | -0.429 |
maid | -0.427 |
Token | said |
Token ID | 531 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
booth | +0.094 |
cheating | +0.090 |
Tomas | +0.089 |
Tom | +0.089 |
employee | +0.088 |
Neg logit contributions | |
stub | -0.124 |
Chick | -0.122 |
berry | -0.121 |
light | -0.120 |
burn | -0.117 |
Token | , |
Token ID | 11 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
Tom | +0.084 |
booth | +0.083 |
worker | +0.078 |
cheating | +0.076 |
employee | +0.075 |
Neg logit contributions | |
stub | -0.165 |
burn | -0.163 |
Chick | -0.161 |
light | -0.157 |
berry | -0.157 |
Token | " |
Token ID | 366 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
booth | +0.001 |
cheating | +0.001 |
tickets | +0.001 |
employee | +0.001 |
Tom | +0.001 |
Neg logit contributions | |
Chick | -0.002 |
stub | -0.002 |
berry | -0.002 |
Grac | -0.002 |
burn | -0.002 |
Token | Please |
Token ID | 5492 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
booth | +0.054 |
Tom | +0.052 |
cheating | +0.051 |
employee | +0.050 |
Tomas | +0.050 |
Neg logit contributions | |
stub | -0.071 |
Chick | -0.070 |
burn | -0.068 |
berry | -0.068 |
light | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
stub | +0.031 |
Chick | +0.030 |
burn | +0.030 |
berry | +0.029 |
light | +0.029 |
Neg logit contributions | |
Tom | -0.012 |
cheating | -0.010 |
money | -0.010 |
booth | -0.009 |
worker | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
Tom | +0.011 |
booth | +0.011 |
Tomas | +0.010 |
cheating | +0.010 |
employee | +0.010 |
Neg logit contributions | |
burn | -0.016 |
stub | -0.016 |
Chick | -0.016 |
light | -0.016 |
berry | -0.016 |
Token |
|
Token ID | 198 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
stub | +0.008 |
Chick | +0.008 |
Grac | +0.007 |
berry | +0.007 |
light | +0.007 |
Neg logit contributions | |
booth | -0.007 |
cheating | -0.007 |
worker | -0.007 |
tickets | -0.007 |
employee | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
stub | +0.003 |
Chick | +0.003 |
light | +0.003 |
Grac | +0.003 |
berry | +0.003 |
Neg logit contributions | |
booth | -0.003 |
tickets | -0.003 |
cheating | -0.003 |
money | -0.003 |
worker | -0.003 |
Token | zo |
Token ID | 40565 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
booth | +0.010 |
Tomas | +0.009 |
Tom | +0.009 |
cheating | +0.009 |
site | +0.009 |
Neg logit contributions | |
stub | -0.012 |
skirt | -0.011 |
berry | -0.011 |
Chick | -0.011 |
Grac | -0.011 |
Token | oke |
Token ID | 2088 |
Feature activation | +9.17e+00 |
Pos logit contributions | |
Tomas | +0.098 |
newspapers | +0.098 |
booth | +0.094 |
Tom | +0.093 |
her | +0.092 |
Neg logit contributions | |
berry | -0.240 |
stub | -0.235 |
maid | -0.232 |
Grac | -0.230 |
ubb | -0.230 |
Token | eper |
Token ID | 5723 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
booth | +0.440 |
Tomas | +0.436 |
newspapers | +0.419 |
employee | +0.412 |
Tom | +0.409 |
Neg logit contributions | |
berry | -0.426 |
ie | -0.425 |
stub | -0.415 |
maid | -0.410 |
burn | -0.405 |
Token | takes |
Token ID | 2753 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
booth | +0.093 |
cheating | +0.089 |
Tom | +0.089 |
Tomas | +0.089 |
employee | +0.089 |
Neg logit contributions | |
stub | -0.126 |
Chick | -0.122 |
berry | -0.122 |
burn | -0.119 |
light | -0.119 |
Token | them |
Token ID | 606 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
booth | +0.023 |
cheating | +0.022 |
employee | +0.021 |
site | +0.021 |
Tomas | +0.021 |
Neg logit contributions | |
Chick | -0.032 |
Grac | -0.031 |
berry | -0.031 |
stub | -0.031 |
burn | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
stub | +0.010 |
Chick | +0.009 |
burn | +0.009 |
berry | +0.009 |
light | +0.009 |
Neg logit contributions | |
booth | -0.008 |
cheating | -0.008 |
employee | -0.008 |
site | -0.008 |
Tom | -0.008 |
Token | their |
Token ID | 511 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
Chick | +0.028 |
Grac | +0.027 |
stub | +0.026 |
berry | +0.026 |
burn | +0.025 |
Neg logit contributions | |
booth | -0.019 |
cheating | -0.019 |
employee | -0.018 |
nine | -0.017 |
Tomas | -0.017 |
Token | more |
Token ID | 517 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
skirt | +0.128 |
berry | +0.127 |
stub | +0.124 |
lull | +0.124 |
burn | +0.120 |
Neg logit contributions | |
Tomas | -0.078 |
Officer | -0.073 |
Greg | -0.073 |
booth | -0.072 |
employee | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -4.48e-03 |
Pos logit contributions | |
Tomas | +0.121 |
Greg | +0.117 |
Officer | +0.115 |
booth | +0.115 |
employee | +0.114 |
Neg logit contributions | |
berry | -0.109 |
skirt | -0.108 |
stub | -0.104 |
lull | -0.104 |
light | -0.103 |
Token | They |
Token ID | 1119 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Chick | +0.000 |
stub | +0.000 |
berry | +0.000 |
light | +0.000 |
Grac | +0.000 |
Neg logit contributions | |
booth | -0.000 |
employee | -0.000 |
cheating | -0.000 |
worker | -0.000 |
site | -0.000 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.49e+00 |
Pos logit contributions | |
stub | +0.005 |
Chick | +0.005 |
burn | +0.005 |
berry | +0.005 |
Grac | +0.005 |
Neg logit contributions | |
booth | -0.005 |
cheating | -0.004 |
employee | -0.004 |
Tom | -0.004 |
site | -0.004 |
Token | e |
Token ID | 68 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
stub | +0.292 |
Chick | +0.287 |
burn | +0.282 |
berry | +0.280 |
light | +0.278 |
Neg logit contributions | |
booth | -0.149 |
Tom | -0.144 |
cheating | -0.140 |
employee | -0.136 |
Tomas | -0.135 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.80e+00 |
Pos logit contributions | |
stub | +0.142 |
Chick | +0.135 |
Grac | +0.133 |
streamed | +0.130 |
berry | +0.129 |
Neg logit contributions | |
Tom | -0.127 |
booth | -0.123 |
tickets | -0.119 |
worker | -0.116 |
workers | -0.114 |
Token | some |
Token ID | 617 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
stub | +0.550 |
Chick | +0.542 |
burn | +0.532 |
berry | +0.528 |
light | +0.525 |
Neg logit contributions | |
booth | -0.314 |
Tom | -0.300 |
cheating | -0.296 |
employee | -0.291 |
Tomas | -0.288 |
Token | more |
Token ID | 517 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
Chick | +0.144 |
stub | +0.144 |
burn | +0.140 |
Grac | +0.139 |
light | +0.137 |
Neg logit contributions | |
booth | -0.067 |
Tom | -0.065 |
tickets | -0.064 |
cheating | -0.064 |
money | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
booth | +0.093 |
cheating | +0.089 |
Tomas | +0.089 |
Tom | +0.089 |
employee | +0.088 |
Neg logit contributions | |
stub | -0.138 |
Chick | -0.135 |
berry | -0.132 |
light | -0.132 |
burn | -0.132 |
Token | They |
Token ID | 1119 |
Feature activation | -8.78e-02 |
Pos logit contributions | |
booth | +0.001 |
employee | +0.001 |
cheating | +0.001 |
worker | +0.001 |
site | +0.001 |
Neg logit contributions | |
Chick | -0.001 |
stub | -0.001 |
berry | -0.001 |
light | -0.001 |
burn | -0.001 |
Token | laugh |
Token ID | 6487 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
stub | +0.005 |
Chick | +0.005 |
berry | +0.005 |
light | +0.005 |
burn | +0.005 |
Neg logit contributions | |
Tom | -0.004 |
booth | -0.004 |
cheating | -0.003 |
tickets | -0.003 |
money | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +2.59e-01 |
Pos logit contributions | |
booth | +0.019 |
cheating | +0.019 |
Tomas | +0.019 |
Tom | +0.019 |
employee | +0.019 |
Neg logit contributions | |
stub | -0.027 |
Chick | -0.027 |
berry | -0.027 |
light | -0.027 |
burn | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
cheating | +0.012 |
booth | +0.012 |
employee | +0.012 |
tickets | +0.011 |
worker | +0.011 |
Neg logit contributions | |
Chick | -0.014 |
stub | -0.013 |
Grac | -0.013 |
berry | -0.012 |
light | -0.012 |
Token |
|
Token ID | 198 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
Chick | +0.011 |
stub | +0.011 |
Grac | +0.010 |
apy | +0.009 |
berry | +0.009 |
Neg logit contributions | |
cheating | -0.012 |
booth | -0.012 |
employee | -0.012 |
worker | -0.011 |
tickets | -0.011 |
Token | " |
Token ID | 1 |
Feature activation | +9.22e-01 |
Pos logit contributions | |
apy | +0.008 |
stub | +0.008 |
Chick | +0.008 |
ot | +0.008 |
light | +0.008 |
Neg logit contributions | |
cheating | -0.012 |
inconvenience | -0.012 |
numbers | -0.012 |
money | -0.011 |
answers | -0.011 |
Token | Free |
Token ID | 11146 |
Feature activation | +7.93e-01 |
Pos logit contributions | |
booth | +0.043 |
cheating | +0.042 |
Tom | +0.041 |
money | +0.040 |
tickets | +0.040 |
Neg logit contributions | |
stub | -0.048 |
Chick | -0.047 |
burn | -0.046 |
light | -0.045 |
Grac | -0.044 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.78e+00 |
Pos logit contributions | |
booth | +0.054 |
Tom | +0.053 |
cheating | +0.052 |
employee | +0.052 |
tickets | +0.051 |
Neg logit contributions | |
stub | -0.025 |
Chick | -0.023 |
burn | -0.023 |
berry | -0.023 |
light | -0.022 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
Chick | +0.440 |
stub | +0.438 |
burn | +0.426 |
berry | +0.416 |
light | +0.414 |
Neg logit contributions | |
booth | -0.391 |
employee | -0.376 |
tickets | -0.369 |
nine | -0.366 |
bills | -0.365 |
Token | their |
Token ID | 511 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
Chick | +0.136 |
stub | +0.135 |
Grac | +0.129 |
apy | +0.128 |
streamed | +0.127 |
Neg logit contributions | |
cheating | -0.056 |
booth | -0.052 |
employee | -0.051 |
site | -0.049 |
worker | -0.049 |
Token | mom |
Token ID | 1995 |
Feature activation | -5.59e-01 |
Pos logit contributions | |
booth | +0.032 |
Tom | +0.032 |
tickets | +0.031 |
cheating | +0.031 |
employee | +0.031 |
Neg logit contributions | |
stub | -0.040 |
Chick | -0.040 |
berry | -0.039 |
light | -0.039 |
burn | -0.039 |
Token | said |
Token ID | 531 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
Chick | +0.035 |
stub | +0.034 |
light | +0.033 |
Grac | +0.032 |
berry | +0.032 |
Neg logit contributions | |
booth | -0.021 |
inconvenience | -0.020 |
cheating | -0.020 |
site | -0.019 |
tickets | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
booth | +0.098 |
cheating | +0.094 |
Tomas | +0.092 |
employee | +0.091 |
site | +0.090 |
Neg logit contributions | |
stub | -0.128 |
Chick | -0.126 |
burn | -0.122 |
light | -0.121 |
Grac | -0.120 |
Token | can |
Token ID | 460 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
booth | +0.018 |
Tom | +0.018 |
cheating | +0.017 |
employee | +0.016 |
tickets | +0.016 |
Neg logit contributions | |
stub | -0.032 |
Chick | -0.032 |
berry | -0.032 |
burn | -0.031 |
light | -0.031 |
Token | make |
Token ID | 787 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
berry | +0.008 |
light | +0.008 |
burn | +0.008 |
stub | +0.008 |
maid | +0.008 |
Neg logit contributions | |
Tom | -0.005 |
cheating | -0.005 |
money | -0.005 |
her | -0.005 |
booth | -0.005 |
Token | you |
Token ID | 345 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
stub | +0.137 |
berry | +0.134 |
burn | +0.134 |
Chick | +0.131 |
light | +0.129 |
Neg logit contributions | |
Tom | -0.080 |
cheating | -0.075 |
booth | -0.074 |
money | -0.073 |
tickets | -0.072 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.51e+00 |
Pos logit contributions | |
money | +0.047 |
Tom | +0.044 |
cheating | +0.044 |
old | +0.040 |
fun | +0.040 |
Neg logit contributions | |
berry | -0.066 |
streamed | -0.063 |
stub | -0.063 |
Chick | -0.062 |
Grac | -0.062 |
Token | e |
Token ID | 68 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
stub | +0.191 |
Grac | +0.187 |
burn | +0.182 |
berry | +0.182 |
Chick | +0.181 |
Neg logit contributions | |
Tom | -0.254 |
cheating | -0.251 |
her | -0.241 |
money | -0.237 |
site | -0.233 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.77e+00 |
Pos logit contributions | |
stub | +0.092 |
berry | +0.090 |
Grac | +0.090 |
streamed | +0.087 |
Chick | +0.082 |
Neg logit contributions | |
Tom | -0.159 |
her | -0.155 |
cheating | -0.155 |
money | -0.155 |
signs | -0.153 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
stub | +0.577 |
berry | +0.557 |
Grac | +0.551 |
Chick | +0.548 |
burn | +0.542 |
Neg logit contributions | |
Tom | -0.304 |
cheating | -0.290 |
booth | -0.282 |
site | -0.271 |
money | -0.266 |
Token | Max |
Token ID | 5436 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
stub | +0.112 |
Chick | +0.111 |
berry | +0.107 |
burn | +0.107 |
Grac | +0.106 |
Neg logit contributions | |
booth | -0.072 |
cheating | -0.069 |
employee | -0.068 |
tickets | -0.067 |
site | -0.066 |
Token | laughed |
Token ID | 13818 |
Feature activation | -8.40e-01 |
Pos logit contributions | |
booth | +0.031 |
Tom | +0.030 |
cheating | +0.029 |
employee | +0.029 |
Tomas | +0.029 |
Neg logit contributions | |
stub | -0.041 |
Chick | -0.040 |
burn | -0.040 |
berry | -0.039 |
light | -0.039 |
Token | and |
Token ID | 290 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
stub | +0.062 |
Chick | +0.060 |
burn | +0.060 |
berry | +0.059 |
light | +0.059 |
Neg logit contributions | |
Tom | -0.022 |
booth | -0.021 |
cheating | -0.020 |
Tomas | -0.019 |
employee | -0.019 |
Token | said |
Token ID | 531 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
booth | +0.016 |
Tom | +0.016 |
cheating | +0.015 |
employee | +0.015 |
Tomas | +0.015 |
Neg logit contributions | |
stub | -0.034 |
Chick | -0.034 |
burn | -0.033 |
berry | -0.033 |
light | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
stub | +0.402 |
Grac | +0.389 |
Chick | +0.387 |
oppy | +0.385 |
rys | +0.384 |
Neg logit contributions | |
Tom | -0.097 |
booth | -0.076 |
her | -0.069 |
money | -0.066 |
numbers | -0.065 |
Token | made |
Token ID | 925 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
Tom | +0.012 |
numbers | +0.010 |
booth | +0.010 |
signs | +0.010 |
money | +0.010 |
Neg logit contributions | |
berry | -0.014 |
ini | -0.014 |
oppy | -0.013 |
days | -0.013 |
ivering | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
stub | +0.167 |
berry | +0.159 |
burn | +0.159 |
light | +0.157 |
Chick | +0.156 |
Neg logit contributions | |
Tom | -0.100 |
booth | -0.091 |
tickets | -0.091 |
money | -0.090 |
cheating | -0.088 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.50e+00 |
Pos logit contributions | |
Grac | +0.100 |
Chick | +0.100 |
streamed | +0.099 |
stub | +0.097 |
berry | +0.097 |
Neg logit contributions | |
Tom | -0.085 |
money | -0.082 |
tickets | -0.080 |
signs | -0.076 |
fun | -0.075 |
Token | e |
Token ID | 68 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
stub | +0.206 |
Grac | +0.202 |
skirt | +0.193 |
maid | +0.192 |
ifying | +0.191 |
Neg logit contributions | |
Tom | -0.241 |
her | -0.229 |
Tom | -0.224 |
cheating | -0.222 |
ena | -0.216 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.77e+00 |
Pos logit contributions | |
stub | +0.107 |
Grac | +0.105 |
streamed | +0.097 |
berry | +0.097 |
oppy | +0.095 |
Neg logit contributions | |
Tom | -0.144 |
signs | -0.143 |
tickets | -0.142 |
her | -0.140 |
money | -0.138 |
Token | . |
Token ID | 13 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
stub | +0.649 |
oppy | +0.617 |
Grac | +0.615 |
Chick | +0.613 |
berry | +0.610 |
Neg logit contributions | |
Tom | -0.243 |
booth | -0.220 |
cheating | -0.218 |
workers | -0.206 |
worker | -0.204 |
Token | She |
Token ID | 1375 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
Tom | +0.002 |
booth | +0.002 |
Tomas | +0.002 |
cheating | +0.002 |
Officer | +0.002 |
Neg logit contributions | |
stub | -0.003 |
burn | -0.003 |
light | -0.003 |
berry | -0.003 |
Chick | -0.003 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
burn | +0.049 |
berry | +0.048 |
Grac | +0.047 |
Chick | +0.047 |
stub | +0.046 |
Neg logit contributions | |
Tom | -0.035 |
booth | -0.032 |
cheating | -0.031 |
tickets | -0.031 |
gas | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | -8.77e-01 |
Pos logit contributions | |
stub | +0.030 |
light | +0.029 |
Chick | +0.029 |
berry | +0.029 |
burn | +0.029 |
Neg logit contributions | |
Tom | -0.018 |
booth | -0.016 |
tickets | -0.015 |
money | -0.015 |
cheating | -0.015 |
Token | get |
Token ID | 651 |
Feature activation | +6.71e-01 |
Pos logit contributions | |
light | +0.059 |
streamed | +0.057 |
ifying | +0.056 |
streaming | +0.056 |
Chick | +0.056 |
Neg logit contributions | |
Tom | -0.032 |
money | -0.027 |
sign | -0.027 |
signs | -0.025 |
gas | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | -5.05e-01 |
Pos logit contributions | |
Tom | +0.077 |
cheating | +0.062 |
worker | +0.061 |
booth | +0.060 |
money | +0.057 |
Neg logit contributions | |
stub | -0.166 |
burn | -0.165 |
light | -0.155 |
skirt | -0.153 |
streamed | -0.153 |
Token | " |
Token ID | 366 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
stub | +0.035 |
burn | +0.034 |
Chick | +0.034 |
berry | +0.034 |
light | +0.034 |
Neg logit contributions | |
Tom | -0.015 |
booth | -0.015 |
cheating | -0.013 |
worker | -0.013 |
employee | -0.013 |
Token | Ant |
Token ID | 13217 |
Feature activation | -3.20e-01 |
Pos logit contributions | |
booth | +0.038 |
cheating | +0.036 |
site | +0.035 |
tickets | +0.035 |
employee | +0.035 |
Neg logit contributions | |
stub | -0.064 |
burn | -0.063 |
Chick | -0.063 |
berry | -0.062 |
light | -0.062 |
Token | if |
Token ID | 361 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
stub | +0.012 |
Chick | +0.012 |
berry | +0.011 |
burn | +0.011 |
light | +0.011 |
Neg logit contributions | |
booth | -0.019 |
Tom | -0.019 |
cheating | -0.019 |
employee | -0.018 |
Tomas | -0.018 |
Token | ree |
Token ID | 631 |
Feature activation | -9.72e-01 |
Pos logit contributions | |
oppy | +0.100 |
Chick | +0.099 |
apy | +0.094 |
ie | +0.094 |
stub | +0.093 |
Neg logit contributions | |
employee | -0.148 |
booth | -0.146 |
Tomas | -0.141 |
itor | -0.136 |
site | -0.135 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.76e+00 |
Pos logit contributions | |
stub | +0.041 |
skirt | +0.038 |
berry | +0.038 |
light | +0.037 |
numb | +0.037 |
Neg logit contributions | |
Tom | -0.055 |
money | -0.051 |
cheating | -0.051 |
tickets | -0.050 |
her | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
stub | +0.550 |
Chick | +0.532 |
burn | +0.527 |
berry | +0.524 |
light | +0.522 |
Neg logit contributions | |
booth | -0.314 |
Tom | -0.314 |
cheating | -0.300 |
Tomas | -0.288 |
employee | -0.287 |
Token | Poison |
Token ID | 19357 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
Tom | +0.007 |
booth | +0.006 |
Officer | +0.006 |
Tomas | +0.006 |
cheating | +0.006 |
Neg logit contributions | |
stub | -0.010 |
burn | -0.010 |
skirt | -0.010 |
streamed | -0.009 |
lull | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
stub | +0.036 |
Chick | +0.035 |
burn | +0.035 |
berry | +0.035 |
light | +0.034 |
Neg logit contributions | |
booth | -0.022 |
Tom | -0.021 |
cheating | -0.021 |
employee | -0.020 |
Tomas | -0.020 |
Token | Do |
Token ID | 2141 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
Tom | +0.007 |
Officer | +0.005 |
booth | +0.005 |
Tomas | +0.005 |
cheating | +0.005 |
Neg logit contributions | |
stub | -0.010 |
burn | -0.009 |
skirt | -0.009 |
ie | -0.009 |
streamed | -0.009 |
Token | not |
Token ID | 407 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
stub | +0.085 |
Chick | +0.083 |
burn | +0.082 |
berry | +0.081 |
light | +0.081 |
Neg logit contributions | |
booth | -0.057 |
Tom | -0.055 |
cheating | -0.054 |
employee | -0.053 |
Tomas | -0.053 |
Token | big |
Token ID | 1263 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
booth | +0.016 |
Tom | +0.016 |
worker | +0.015 |
cheating | +0.015 |
money | +0.015 |
Neg logit contributions | |
stub | -0.023 |
light | -0.023 |
Chick | -0.023 |
berry | -0.023 |
burn | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
stub | +0.081 |
Chick | +0.079 |
burn | +0.077 |
berry | +0.077 |
light | +0.076 |
Neg logit contributions | |
booth | -0.073 |
Tom | -0.072 |
cheating | -0.070 |
employee | -0.069 |
Tomas | -0.069 |
Token | scary |
Token ID | 14343 |
Feature activation | +7.12e-01 |
Pos logit contributions | |
stub | +0.026 |
Chick | +0.025 |
burn | +0.025 |
berry | +0.024 |
light | +0.024 |
Neg logit contributions | |
booth | -0.024 |
Tom | -0.024 |
cheating | -0.023 |
employee | -0.023 |
Tomas | -0.023 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.55e+00 |
Pos logit contributions | |
booth | +0.031 |
Tom | +0.031 |
cheating | +0.030 |
worker | +0.030 |
site | +0.029 |
Neg logit contributions | |
Chick | -0.038 |
Grac | -0.037 |
stub | -0.037 |
light | -0.037 |
burn | -0.036 |
Token | e |
Token ID | 68 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
Grac | +0.194 |
stub | +0.188 |
maid | +0.187 |
Chick | +0.186 |
ifying | +0.186 |
Neg logit contributions | |
Tom | -0.259 |
cheating | -0.242 |
Tom | -0.240 |
worker | -0.239 |
her | -0.237 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
Grac | +0.102 |
ubb | +0.094 |
Chick | +0.092 |
stub | +0.091 |
streamed | +0.089 |
Neg logit contributions | |
Tom | -0.160 |
booth | -0.157 |
worker | -0.155 |
her | -0.153 |
workers | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
stub | +0.572 |
Chick | +0.563 |
berry | +0.557 |
Grac | +0.557 |
burn | +0.549 |
Neg logit contributions | |
booth | -0.289 |
Tom | -0.288 |
cheating | -0.271 |
worker | -0.265 |
employee | -0.261 |
Token | Every |
Token ID | 3887 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
Tom | +0.010 |
booth | +0.009 |
cheating | +0.009 |
Tomas | +0.009 |
worker | +0.009 |
Neg logit contributions | |
stub | -0.012 |
burn | -0.012 |
Chick | -0.012 |
light | -0.012 |
berry | -0.012 |
Token | time |
Token ID | 640 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
Grac | +0.148 |
maid | +0.143 |
Chick | +0.139 |
berry | +0.136 |
streaming | +0.136 |
Neg logit contributions | |
Tom | -0.063 |
booth | -0.062 |
worker | -0.057 |
equipment | -0.056 |
site | -0.055 |
Token | he |
Token ID | 339 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
Tom | +0.099 |
booth | +0.092 |
cheating | +0.091 |
tickets | +0.086 |
Tomas | +0.086 |
Neg logit contributions | |
stub | -0.166 |
Chick | -0.164 |
burn | -0.163 |
light | -0.160 |
berry | -0.159 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.53e+00 |
Pos logit contributions | |
Grac | +0.088 |
maid | +0.085 |
berry | +0.085 |
ubb | +0.084 |
ifying | +0.083 |
Neg logit contributions | |
Tom | -0.070 |
cheating | -0.066 |
gas | -0.065 |
workers | -0.064 |
equipment | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
booth | +0.138 |
nine | +0.130 |
2 | +0.130 |
Tomas | +0.128 |
tickets | +0.128 |
Neg logit contributions | |
stub | -0.102 |
days | -0.096 |
skirt | -0.093 |
Chick | -0.091 |
light | -0.091 |
Token |
|
Token ID | 198 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
booth | +0.007 |
cheating | +0.007 |
employee | +0.007 |
Tom | +0.007 |
tickets | +0.007 |
Neg logit contributions | |
stub | -0.010 |
Chick | -0.010 |
berry | -0.010 |
burn | -0.010 |
Grac | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
Chick | +0.016 |
stub | +0.015 |
Grac | +0.014 |
berry | +0.014 |
oppy | +0.014 |
Neg logit contributions | |
cheating | -0.018 |
booth | -0.018 |
employee | -0.018 |
worker | -0.017 |
bills | -0.017 |
Token | " |
Token ID | 1 |
Feature activation | +7.90e-01 |
Pos logit contributions | |
stub | +0.017 |
Chick | +0.017 |
light | +0.016 |
berry | +0.016 |
streamed | +0.016 |
Neg logit contributions | |
cheating | -0.015 |
booth | -0.015 |
tickets | -0.014 |
money | -0.014 |
inconvenience | -0.014 |
Token | Free |
Token ID | 11146 |
Feature activation | +8.95e-01 |
Pos logit contributions | |
booth | +0.034 |
Tom | +0.032 |
cheating | +0.032 |
employee | +0.032 |
tickets | +0.032 |
Neg logit contributions | |
stub | -0.044 |
Chick | -0.043 |
burn | -0.042 |
berry | -0.042 |
light | -0.041 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.75e+00 |
Pos logit contributions | |
booth | +0.064 |
Tom | +0.063 |
cheating | +0.062 |
employee | +0.062 |
Tomas | +0.061 |
Neg logit contributions | |
stub | -0.024 |
Chick | -0.023 |
burn | -0.022 |
berry | -0.021 |
light | -0.021 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
stub | +0.451 |
Chick | +0.441 |
burn | +0.434 |
berry | +0.418 |
light | +0.413 |
Neg logit contributions | |
booth | -0.383 |
employee | -0.370 |
tickets | -0.362 |
nine | -0.359 |
bills | -0.357 |
Token | Ben |
Token ID | 3932 |
Feature activation | -7.80e-01 |
Pos logit contributions | |
stub | +0.126 |
Chick | +0.124 |
berry | +0.120 |
Grac | +0.119 |
light | +0.118 |
Neg logit contributions | |
booth | -0.077 |
cheating | -0.074 |
employee | -0.072 |
Tom | -0.071 |
Tomas | -0.071 |
Token | says |
Token ID | 1139 |
Feature activation | +2.97e+00 |
Pos logit contributions | |
stub | +0.046 |
Chick | +0.046 |
skirt | +0.044 |
Po | +0.044 |
Grac | +0.043 |
Neg logit contributions | |
inconvenience | -0.027 |
site | -0.026 |
booth | -0.026 |
employee | -0.026 |
worker | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +7.62e-02 |
Pos logit contributions | |
booth | +0.131 |
cheating | +0.123 |
Tomas | +0.121 |
employee | +0.121 |
site | +0.119 |
Neg logit contributions | |
stub | -0.162 |
Chick | -0.160 |
light | -0.153 |
Grac | -0.152 |
burn | -0.152 |
Token | " |
Token ID | 366 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
Tom | +0.003 |
booth | +0.003 |
worker | +0.003 |
employee | +0.003 |
Tomas | +0.002 |
Neg logit contributions | |
stub | -0.005 |
burn | -0.005 |
Chick | -0.005 |
light | -0.005 |
berry | -0.005 |
Token | sh |
Token ID | 427 |
Feature activation | -6.35e+00 |
Pos logit contributions | |
Chick | +0.034 |
berry | +0.034 |
Grac | +0.034 |
stub | +0.033 |
oppy | +0.032 |
Neg logit contributions | |
money | -0.023 |
Tom | -0.022 |
tickets | -0.022 |
booth | -0.020 |
fun | -0.020 |
Token | iver |
Token ID | 1428 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
stub | +0.232 |
berry | +0.211 |
Grac | +0.210 |
Chick | +0.208 |
skirt | +0.206 |
Neg logit contributions | |
her | -0.398 |
Tom | -0.389 |
booth | -0.385 |
cheating | -0.383 |
employee | -0.381 |
Token | and |
Token ID | 290 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
stub | +0.171 |
Chick | +0.164 |
berry | +0.162 |
burn | +0.161 |
Grac | +0.161 |
Neg logit contributions | |
Tom | -0.064 |
booth | -0.061 |
cheating | -0.058 |
worker | -0.056 |
workers | -0.055 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
Tom | +0.023 |
old | +0.020 |
money | +0.020 |
cheating | +0.020 |
fun | +0.019 |
Neg logit contributions | |
berry | -0.032 |
burn | -0.029 |
oppy | -0.028 |
ini | -0.028 |
Grac | -0.028 |
Token | e |
Token ID | 68 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
stub | +0.229 |
Grac | +0.220 |
skirt | +0.218 |
maid | +0.215 |
berry | +0.214 |
Neg logit contributions | |
Tom | -0.222 |
her | -0.202 |
Tom | -0.195 |
cheating | -0.192 |
money | -0.190 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.74e+00 |
Pos logit contributions | |
stub | +0.102 |
Grac | +0.097 |
berry | +0.094 |
rys | +0.092 |
oppy | +0.091 |
Neg logit contributions | |
Tom | -0.158 |
her | -0.151 |
tickets | -0.148 |
signs | -0.143 |
workers | -0.143 |
Token | . |
Token ID | 13 |
Feature activation | +2.69e-01 |
Pos logit contributions | |
stub | +0.588 |
Chick | +0.569 |
berry | +0.566 |
burn | +0.559 |
Grac | +0.559 |
Neg logit contributions | |
Tom | -0.283 |
booth | -0.275 |
cheating | -0.260 |
worker | -0.252 |
tickets | -0.251 |
Token | They |
Token ID | 1119 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
Tom | +0.010 |
booth | +0.010 |
cheating | +0.009 |
Tomas | +0.009 |
employee | +0.009 |
Neg logit contributions | |
stub | -0.017 |
burn | -0.016 |
Chick | -0.016 |
berry | -0.016 |
light | -0.016 |
Token | had |
Token ID | 550 |
Feature activation | +9.42e-01 |
Pos logit contributions | |
Tom | +0.002 |
booth | +0.002 |
tickets | +0.002 |
parked | +0.002 |
also | +0.002 |
Neg logit contributions | |
berry | -0.003 |
burn | -0.003 |
Grac | -0.003 |
ini | -0.003 |
Chick | -0.003 |
Token | no |
Token ID | 645 |
Feature activation | -4.69e+00 |
Pos logit contributions | |
Tom | +0.032 |
tickets | +0.030 |
money | +0.029 |
booth | +0.028 |
cheating | +0.028 |
Neg logit contributions | |
berry | -0.063 |
Chick | -0.062 |
Grac | -0.061 |
stub | -0.061 |
burn | -0.060 |
Token | shelter |
Token ID | 11772 |
Feature activation | +2.36e-01 |
Pos logit contributions | |
stub | +0.268 |
Chick | +0.267 |
berry | +0.258 |
Grac | +0.256 |
burn | +0.255 |
Neg logit contributions | |
booth | -0.192 |
Tom | -0.190 |
tickets | -0.186 |
money | -0.186 |
cheating | -0.185 |
Token | . |
Token ID | 13 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
Tom | +0.058 |
booth | +0.054 |
cheating | +0.053 |
worker | +0.051 |
money | +0.050 |
Neg logit contributions | |
stub | -0.113 |
Chick | -0.110 |
burn | -0.108 |
berry | -0.107 |
Grac | -0.106 |
Token | I |
Token ID | 314 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
Tom | +0.006 |
Officer | +0.005 |
booth | +0.005 |
Tomas | +0.005 |
Tom | +0.004 |
Neg logit contributions | |
burn | -0.011 |
stub | -0.010 |
ie | -0.010 |
berry | -0.010 |
maid | -0.010 |
Token | will |
Token ID | 481 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
Chick | +0.090 |
berry | +0.089 |
stub | +0.089 |
burn | +0.088 |
light | +0.087 |
Neg logit contributions | |
Tom | -0.059 |
cheating | -0.057 |
booth | -0.057 |
money | -0.054 |
worker | -0.054 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.46e+00 |
Pos logit contributions | |
berry | +0.103 |
light | +0.101 |
stub | +0.100 |
Chick | +0.099 |
streamed | +0.097 |
Neg logit contributions | |
Tom | -0.051 |
cheating | -0.046 |
money | -0.046 |
booth | -0.045 |
her | -0.043 |
Token | e |
Token ID | 68 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
stub | +0.180 |
Chick | +0.174 |
Grac | +0.170 |
berry | +0.164 |
maid | +0.163 |
Neg logit contributions | |
Tom | -0.271 |
cheating | -0.259 |
money | -0.250 |
booth | -0.245 |
Tom | -0.245 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.74e+00 |
Pos logit contributions | |
stub | +0.085 |
Grac | +0.083 |
light | +0.083 |
streamed | +0.081 |
ubb | +0.081 |
Neg logit contributions | |
Tom | -0.166 |
booth | -0.153 |
worker | -0.152 |
tickets | -0.149 |
right | -0.149 |
Token | in |
Token ID | 287 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
stub | +0.570 |
Chick | +0.555 |
berry | +0.545 |
burn | +0.542 |
light | +0.541 |
Neg logit contributions | |
Tom | -0.304 |
booth | -0.294 |
cheating | -0.286 |
worker | -0.268 |
tickets | -0.264 |
Token | my |
Token ID | 616 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
Tom | +0.034 |
money | +0.030 |
cheating | +0.028 |
old | +0.027 |
booth | +0.027 |
Neg logit contributions | |
stub | -0.083 |
berry | -0.080 |
maid | -0.079 |
beam | -0.079 |
streamed | -0.079 |
Token | elbow |
Token ID | 22662 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
Chick | +0.063 |
burn | +0.062 |
Grac | +0.062 |
berry | +0.062 |
streamed | +0.061 |
Neg logit contributions | |
booth | -0.048 |
Tom | -0.047 |
old | -0.046 |
money | -0.046 |
equipment | -0.045 |
Token | next |
Token ID | 1306 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
stub | +0.241 |
Chick | +0.240 |
Grac | +0.234 |
streamed | +0.230 |
berry | +0.227 |
Neg logit contributions | |
Tom | -0.127 |
booth | -0.122 |
cheating | -0.114 |
money | -0.112 |
tickets | -0.109 |
Token | time |
Token ID | 640 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
booth | +0.033 |
Tom | +0.032 |
cheating | +0.030 |
worker | +0.030 |
tickets | +0.030 |
Neg logit contributions | |
stub | -0.076 |
Chick | -0.076 |
burn | -0.073 |
Grac | -0.073 |
berry | -0.073 |
Token | let |
Token ID | 1309 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
stub | +0.001 |
Chick | +0.001 |
berry | +0.001 |
burn | +0.001 |
Grac | +0.001 |
Neg logit contributions | |
booth | -0.001 |
cheating | -0.001 |
employee | -0.001 |
Tom | -0.001 |
Tomas | -0.000 |
Token | them |
Token ID | 606 |
Feature activation | -4.66e-01 |
Pos logit contributions | |
Chick | +0.147 |
stub | +0.144 |
burn | +0.143 |
berry | +0.140 |
Grac | +0.135 |
Neg logit contributions | |
booth | -0.119 |
cheating | -0.118 |
employee | -0.112 |
site | -0.110 |
Tomas | -0.109 |
Token | run |
Token ID | 1057 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
stub | +0.025 |
Chick | +0.025 |
burn | +0.024 |
berry | +0.024 |
light | +0.024 |
Neg logit contributions | |
booth | -0.021 |
Tom | -0.020 |
cheating | -0.020 |
employee | -0.019 |
Tomas | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
burn | +0.014 |
berry | +0.014 |
Chick | +0.014 |
stub | +0.014 |
numb | +0.013 |
Neg logit contributions | |
cheating | -0.021 |
resa | -0.019 |
Tomas | -0.018 |
inconvenience | -0.018 |
40 | -0.018 |
Token | gra |
Token ID | 7933 |
Feature activation | -4.36e+00 |
Pos logit contributions | |
booth | +0.017 |
Tom | +0.017 |
cheating | +0.016 |
employee | +0.016 |
Tomas | +0.016 |
Neg logit contributions | |
stub | -0.020 |
Chick | -0.019 |
burn | -0.019 |
berry | -0.019 |
light | -0.018 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.74e+00 |
Pos logit contributions | |
stub | +0.172 |
Chick | +0.169 |
burn | +0.165 |
berry | +0.162 |
light | +0.159 |
Neg logit contributions | |
booth | -0.254 |
Tom | -0.248 |
cheating | -0.246 |
employee | -0.242 |
Tomas | -0.240 |
Token | on |
Token ID | 319 |
Feature activation | -6.12e-01 |
Pos logit contributions | |
Chick | +0.453 |
stub | +0.434 |
lull | +0.433 |
burn | +0.429 |
berry | +0.424 |
Neg logit contributions | |
booth | -0.381 |
cheating | -0.378 |
employee | -0.376 |
Tomas | -0.365 |
resa | -0.364 |
Token | the |
Token ID | 262 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
Chick | +0.035 |
berry | +0.033 |
stub | +0.033 |
burn | +0.033 |
Grac | +0.032 |
Neg logit contributions | |
cheating | -0.025 |
employee | -0.025 |
booth | -0.025 |
inconvenience | -0.023 |
Tomas | -0.023 |
Token | grass |
Token ID | 8701 |
Feature activation | +2.64e+00 |
Pos logit contributions | |
cheating | +0.025 |
Tomas | +0.023 |
Tom | +0.023 |
employee | +0.022 |
Greg | +0.022 |
Neg logit contributions | |
Chick | -0.029 |
burn | -0.028 |
stub | -0.027 |
skirt | -0.027 |
lull | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
booth | +0.099 |
Tom | +0.096 |
cheating | +0.095 |
employee | +0.093 |
Tomas | +0.092 |
Neg logit contributions | |
stub | -0.159 |
Chick | -0.156 |
burn | -0.154 |
berry | -0.153 |
light | -0.152 |
Token | They |
Token ID | 1119 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
cheating | +0.006 |
booth | +0.006 |
employee | +0.006 |
worker | +0.006 |
site | +0.005 |
Neg logit contributions | |
Chick | -0.007 |
stub | -0.007 |
Grac | -0.006 |
berry | -0.006 |
light | -0.006 |
Token | Max |
Token ID | 5436 |
Feature activation | +7.55e-01 |
Pos logit contributions | |
Chick | +0.027 |
stub | +0.027 |
burn | +0.026 |
Grac | +0.025 |
light | +0.025 |
Neg logit contributions | |
booth | -0.019 |
employee | -0.017 |
cheating | -0.017 |
site | -0.017 |
tickets | -0.017 |
Token | started |
Token ID | 2067 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
booth | +0.039 |
tickets | +0.037 |
employee | +0.036 |
cheating | +0.036 |
site | +0.036 |
Neg logit contributions | |
stub | -0.037 |
Chick | -0.035 |
burn | -0.034 |
berry | -0.034 |
light | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -8.49e-01 |
Pos logit contributions | |
Tom | +0.058 |
cheating | +0.056 |
worker | +0.054 |
booth | +0.053 |
money | +0.052 |
Neg logit contributions | |
Chick | -0.123 |
berry | -0.122 |
maid | -0.121 |
burn | -0.121 |
light | -0.121 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.41e+00 |
Pos logit contributions | |
light | +0.042 |
ifying | +0.042 |
berry | +0.042 |
stub | +0.042 |
Chick | +0.042 |
Neg logit contributions | |
Tom | -0.039 |
cheating | -0.039 |
money | -0.036 |
worker | -0.036 |
booth | -0.036 |
Token | e |
Token ID | 68 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
stub | +0.176 |
Chick | +0.171 |
Grac | +0.171 |
maid | +0.166 |
burn | +0.165 |
Neg logit contributions | |
Tom | -0.265 |
cheating | -0.253 |
booth | -0.243 |
Tomas | -0.242 |
money | -0.240 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.74e+00 |
Pos logit contributions | |
stub | +0.077 |
Grac | +0.076 |
Chick | +0.072 |
light | +0.071 |
days | +0.071 |
Neg logit contributions | |
Tom | -0.180 |
worker | -0.169 |
cheating | -0.168 |
booth | -0.168 |
workers | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
stub | +0.536 |
Chick | +0.526 |
burn | +0.517 |
berry | +0.512 |
light | +0.509 |
Neg logit contributions | |
booth | -0.323 |
Tom | -0.306 |
cheating | -0.302 |
employee | -0.298 |
tickets | -0.297 |
Token | " |
Token ID | 366 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
Tom | +0.009 |
booth | +0.009 |
Tomas | +0.008 |
cheating | +0.008 |
employee | +0.008 |
Neg logit contributions | |
stub | -0.014 |
burn | -0.014 |
berry | -0.013 |
light | -0.013 |
Chick | -0.013 |
Token | A |
Token ID | 32 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
Tom | +0.044 |
Tom | +0.039 |
her | +0.038 |
Ready | +0.037 |
cheating | +0.035 |
Neg logit contributions | |
maid | -0.072 |
oppy | -0.069 |
Grac | -0.067 |
berry | -0.067 |
Chick | -0.065 |
Token | ch |
Token ID | 354 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
Tom | +0.074 |
traffic | +0.073 |
cheating | +0.072 |
cop | +0.070 |
fun | +0.069 |
Neg logit contributions | |
maid | -0.121 |
light | -0.116 |
Grac | -0.115 |
streamed | -0.113 |
streaming | -0.112 |
Token | oo |
Token ID | 2238 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
numb | +0.023 |
Grac | +0.023 |
maid | +0.023 |
skirt | +0.022 |
ifying | +0.022 |
Neg logit contributions | |
Tom | -0.014 |
worker | -0.012 |
inconvenience | -0.012 |
Tom | -0.011 |
site | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | -7.72e-01 |
Pos logit contributions | |
Tom | +0.004 |
booth | +0.004 |
numbers | +0.003 |
Tomas | +0.003 |
cheating | +0.003 |
Neg logit contributions | |
burn | -0.005 |
stub | -0.004 |
light | -0.004 |
Chick | -0.004 |
maid | -0.004 |
Token | feels |
Token ID | 5300 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
stub | +0.047 |
burn | +0.047 |
Chick | +0.047 |
berry | +0.045 |
light | +0.045 |
Neg logit contributions | |
Tom | -0.028 |
booth | -0.028 |
cheating | -0.027 |
tickets | -0.026 |
Tomas | -0.025 |
Token | a |
Token ID | 257 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
burn | +0.157 |
days | +0.154 |
berry | +0.153 |
beam | +0.152 |
light | +0.152 |
Neg logit contributions | |
Tom | -0.067 |
money | -0.056 |
old | -0.056 |
cheating | -0.054 |
gas | -0.050 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.42e+00 |
Pos logit contributions | |
money | +0.015 |
Tom | +0.015 |
old | +0.015 |
harder | +0.015 |
booth | +0.014 |
Neg logit contributions | |
days | -0.025 |
ci | -0.025 |
burn | -0.024 |
seed | -0.024 |
light | -0.024 |
Token | e |
Token ID | 68 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
Chick | +0.159 |
Grac | +0.159 |
stub | +0.159 |
maid | +0.155 |
burn | +0.153 |
Neg logit contributions | |
Tom | -0.275 |
cheating | -0.258 |
her | -0.255 |
Tom | -0.253 |
booth | -0.251 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.74e+00 |
Pos logit contributions | |
Grac | +0.099 |
Chick | +0.090 |
days | +0.089 |
oppy | +0.088 |
stub | +0.080 |
Neg logit contributions | |
Tom | -0.165 |
booth | -0.157 |
worker | -0.154 |
her | -0.154 |
Tom | -0.150 |
Token | coming |
Token ID | 2406 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
Chick | +0.580 |
stub | +0.577 |
Grac | +0.574 |
oppy | +0.567 |
berry | +0.558 |
Neg logit contributions | |
Tom | -0.285 |
booth | -0.274 |
cheating | -0.264 |
worker | -0.247 |
employee | -0.242 |
Token | . |
Token ID | 13 |
Feature activation | +7.99e-02 |
Pos logit contributions | |
booth | +0.075 |
Tom | +0.073 |
cheating | +0.070 |
employee | +0.069 |
Tomas | +0.068 |
Neg logit contributions | |
stub | -0.128 |
Chick | -0.125 |
burn | -0.123 |
berry | -0.122 |
light | -0.121 |
Token | She |
Token ID | 1375 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
Tom | +0.004 |
booth | +0.003 |
Tomas | +0.003 |
cheating | +0.003 |
numbers | +0.003 |
Neg logit contributions | |
burn | -0.004 |
stub | -0.004 |
light | -0.004 |
Chick | -0.004 |
lull | -0.004 |
Token | tries |
Token ID | 8404 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
stub | +0.046 |
burn | +0.046 |
Chick | +0.046 |
light | +0.044 |
berry | +0.044 |
Neg logit contributions | |
Tom | -0.028 |
booth | -0.028 |
cheating | -0.027 |
tickets | -0.026 |
Tomas | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
booth | +0.102 |
Tom | +0.096 |
employee | +0.096 |
cheating | +0.096 |
Tomas | +0.094 |
Neg logit contributions | |
stub | -0.131 |
Chick | -0.127 |
burn | -0.126 |
berry | -0.124 |
light | -0.122 |
Token | . |
Token ID | 13 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
stub | +0.179 |
Chick | +0.176 |
burn | +0.173 |
berry | +0.172 |
Grac | +0.170 |
Neg logit contributions | |
booth | -0.084 |
Tom | -0.082 |
cheating | -0.078 |
employee | -0.076 |
Tomas | -0.075 |
Token |
|
Token ID | 198 |
Feature activation | -3.58e-01 |
Pos logit contributions | |
Tom | +0.005 |
booth | +0.004 |
Tomas | +0.004 |
money | +0.004 |
worker | +0.004 |
Neg logit contributions | |
burn | -0.009 |
light | -0.009 |
stub | -0.009 |
ie | -0.008 |
Chick | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | -3.41e-01 |
Pos logit contributions | |
stub | +0.020 |
Chick | +0.020 |
berry | +0.019 |
Grac | +0.019 |
burn | +0.019 |
Neg logit contributions | |
booth | -0.015 |
cheating | -0.015 |
employee | -0.014 |
tickets | -0.014 |
worker | -0.014 |
Token | " |
Token ID | 1 |
Feature activation | +9.25e-01 |
Pos logit contributions | |
stub | +0.020 |
Chick | +0.019 |
berry | +0.019 |
burn | +0.019 |
light | +0.019 |
Neg logit contributions | |
booth | -0.014 |
cheating | -0.013 |
Tom | -0.013 |
tickets | -0.013 |
employee | -0.013 |
Token | Free |
Token ID | 11146 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
Tom | +0.033 |
Tom | +0.032 |
booth | +0.029 |
cheating | +0.029 |
Tomas | +0.028 |
Neg logit contributions | |
Chick | -0.060 |
stub | -0.058 |
oppy | -0.058 |
light | -0.057 |
Grac | -0.057 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.72e+00 |
Pos logit contributions | |
Tom | +0.060 |
booth | +0.059 |
cheating | +0.059 |
tickets | +0.058 |
employee | +0.057 |
Neg logit contributions | |
stub | -0.020 |
Chick | -0.018 |
berry | -0.018 |
light | -0.018 |
Grac | -0.017 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
stub | +0.462 |
Chick | +0.456 |
burn | +0.448 |
berry | +0.442 |
light | +0.432 |
Neg logit contributions | |
booth | -0.377 |
employee | -0.362 |
tickets | -0.352 |
Tomas | -0.347 |
cheating | -0.345 |
Token | he |
Token ID | 339 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
stub | +0.125 |
burn | +0.123 |
Chick | +0.123 |
berry | +0.121 |
light | +0.120 |
Neg logit contributions | |
Tom | -0.070 |
booth | -0.070 |
cheating | -0.065 |
employee | -0.064 |
Tomas | -0.064 |
Token | shouted |
Token ID | 17293 |
Feature activation | +3.14e+00 |
Pos logit contributions | |
stub | +0.115 |
Chick | +0.114 |
burn | +0.113 |
berry | +0.113 |
light | +0.111 |
Neg logit contributions | |
booth | -0.042 |
Tom | -0.042 |
cheating | -0.039 |
employee | -0.037 |
Tomas | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
booth | +0.112 |
Tom | +0.111 |
cheating | +0.105 |
employee | +0.103 |
Tomas | +0.102 |
Neg logit contributions | |
stub | -0.197 |
Chick | -0.193 |
burn | -0.191 |
berry | -0.189 |
light | -0.188 |
Token | " |
Token ID | 366 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Tom | +0.004 |
booth | +0.004 |
Tomas | +0.004 |
cheating | +0.004 |
worker | +0.004 |
Neg logit contributions | |
stub | -0.009 |
burn | -0.009 |
light | -0.009 |
Chick | -0.009 |
berry | -0.009 |
Token | we |
Token ID | 356 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
booth | +0.023 |
Tom | +0.022 |
cheating | +0.022 |
employee | +0.021 |
tickets | +0.021 |
Neg logit contributions | |
stub | -0.028 |
Chick | -0.027 |
burn | -0.027 |
berry | -0.027 |
light | -0.026 |
Token | love |
Token ID | 1842 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
stub | +0.046 |
Chick | +0.045 |
burn | +0.044 |
berry | +0.044 |
light | +0.043 |
Neg logit contributions | |
booth | -0.035 |
Tom | -0.033 |
cheating | -0.032 |
employee | -0.032 |
Tomas | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
stub | +0.057 |
Chick | +0.055 |
burn | +0.054 |
berry | +0.053 |
light | +0.053 |
Neg logit contributions | |
booth | -0.032 |
Tom | -0.032 |
cheating | -0.030 |
tickets | -0.029 |
Tomas | -0.029 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
booth | +0.013 |
Tomas | +0.012 |
Tom | +0.012 |
employee | +0.012 |
cheating | +0.012 |
Neg logit contributions | |
stub | -0.017 |
berry | -0.016 |
Chick | -0.016 |
burn | -0.016 |
light | -0.016 |
Token | e |
Token ID | 68 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
stub | +0.185 |
Grac | +0.178 |
Chick | +0.177 |
burn | +0.166 |
skirt | +0.163 |
Neg logit contributions | |
Tom | -0.250 |
booth | -0.237 |
cheating | -0.236 |
her | -0.235 |
worker | -0.235 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.72e+00 |
Pos logit contributions | |
stub | +0.145 |
streamed | +0.135 |
Grac | +0.135 |
streaming | +0.133 |
Chick | +0.132 |
Neg logit contributions | |
booth | -0.122 |
Tom | -0.121 |
tickets | -0.118 |
worker | -0.118 |
workers | -0.115 |
Token | cake |
Token ID | 12187 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
stub | +0.600 |
Chick | +0.571 |
Grac | +0.562 |
oppy | +0.553 |
streamed | +0.551 |
Neg logit contributions | |
Tom | -0.288 |
booth | -0.282 |
worker | -0.259 |
cheating | -0.257 |
tickets | -0.256 |
Token | ." |
Token ID | 526 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
Tom | +0.021 |
booth | +0.021 |
tickets | +0.019 |
worker | +0.019 |
money | +0.018 |
Neg logit contributions | |
stub | -0.046 |
Chick | -0.042 |
streamed | -0.042 |
Grac | -0.042 |
burn | -0.041 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.93e-01 |
Pos logit contributions | |
stub | +0.207 |
burn | +0.202 |
Chick | +0.197 |
light | +0.196 |
berry | +0.194 |
Neg logit contributions | |
Tom | -0.108 |
booth | -0.101 |
tickets | -0.090 |
Tomas | -0.090 |
employee | -0.087 |
Token | S |
Token ID | 50 |
Feature activation | -2.95e+00 |
Pos logit contributions | |
stub | +0.061 |
Chick | +0.060 |
burn | +0.059 |
light | +0.058 |
berry | +0.058 |
Neg logit contributions | |
booth | -0.036 |
Tom | -0.036 |
cheating | -0.034 |
Tomas | -0.033 |
employee | -0.033 |
Token | ara |
Token ID | 3301 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
berry | +0.137 |
burn | +0.130 |
stub | +0.127 |
Chick | +0.126 |
light | +0.124 |
Neg logit contributions | |
cheating | -0.155 |
employee | -0.152 |
worker | -0.148 |
workers | -0.148 |
site | -0.146 |
Token | they |
Token ID | 484 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
booth | +0.030 |
site | +0.028 |
employee | +0.028 |
cheating | +0.028 |
worker | +0.027 |
Neg logit contributions | |
Chick | -0.038 |
stub | -0.037 |
Grac | -0.037 |
light | -0.036 |
burn | -0.035 |
Token | didn |
Token ID | 1422 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
stub | +0.077 |
Chick | +0.075 |
burn | +0.074 |
berry | +0.073 |
light | +0.073 |
Neg logit contributions | |
Tom | -0.062 |
booth | -0.062 |
cheating | -0.059 |
employee | -0.058 |
Tomas | -0.058 |
Token | 't |
Token ID | 470 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
stub | +0.010 |
Chick | +0.010 |
burn | +0.009 |
berry | +0.009 |
light | +0.009 |
Neg logit contributions | |
booth | -0.011 |
Tom | -0.011 |
cheating | -0.011 |
employee | -0.010 |
Tomas | -0.010 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
stub | +0.059 |
Chick | +0.058 |
burn | +0.057 |
berry | +0.057 |
light | +0.057 |
Neg logit contributions | |
Tom | -0.044 |
cheating | -0.042 |
booth | -0.042 |
money | -0.040 |
employee | -0.039 |
Token | e |
Token ID | 68 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
stub | +0.171 |
Grac | +0.165 |
Chick | +0.165 |
burn | +0.159 |
skirt | +0.158 |
Neg logit contributions | |
Tom | -0.278 |
cheating | -0.265 |
money | -0.256 |
booth | -0.255 |
her | -0.254 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.71e+00 |
Pos logit contributions | |
stub | +0.079 |
Chick | +0.072 |
Grac | +0.072 |
streamed | +0.068 |
light | +0.068 |
Neg logit contributions | |
Tom | -0.176 |
booth | -0.168 |
worker | -0.166 |
tickets | -0.165 |
cheating | -0.164 |
Token | anymore |
Token ID | 7471 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
stub | +0.533 |
Chick | +0.525 |
berry | +0.509 |
burn | +0.508 |
Grac | +0.505 |
Neg logit contributions | |
Tom | -0.330 |
booth | -0.326 |
cheating | -0.315 |
worker | -0.301 |
employee | -0.299 |
Token | . |
Token ID | 13 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
booth | +0.028 |
Tom | +0.027 |
cheating | +0.026 |
employee | +0.025 |
Tomas | +0.025 |
Neg logit contributions | |
stub | -0.049 |
Chick | -0.048 |
burn | -0.047 |
berry | -0.047 |
light | -0.046 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -8.19e-01 |
Pos logit contributions | |
booth | +0.011 |
employee | +0.010 |
cheating | +0.010 |
worker | +0.010 |
tickets | +0.010 |
Neg logit contributions | |
Chick | -0.015 |
stub | -0.015 |
berry | -0.014 |
Grac | -0.014 |
burn | -0.014 |
Token | Tom |
Token ID | 13787 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
stub | +0.049 |
Chick | +0.048 |
burn | +0.047 |
berry | +0.047 |
light | +0.046 |
Neg logit contributions | |
booth | -0.032 |
Tom | -0.030 |
cheating | -0.030 |
employee | -0.029 |
Tomas | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +4.76e-01 |
Pos logit contributions | |
Tom | +0.150 |
booth | +0.143 |
cheating | +0.136 |
Tomas | +0.133 |
employee | +0.132 |
Neg logit contributions | |
stub | -0.235 |
Chick | -0.231 |
burn | -0.229 |
Grac | -0.224 |
berry | -0.223 |
Token | uffy |
Token ID | 15352 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
stub | +0.119 |
Grac | +0.114 |
burn | +0.113 |
lull | +0.111 |
streamed | +0.110 |
Neg logit contributions | |
Tom | -0.178 |
her | -0.169 |
cheating | -0.167 |
worker | -0.163 |
booth | -0.162 |
Token | started |
Token ID | 2067 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
burn | +0.147 |
stub | +0.147 |
Chick | +0.145 |
berry | +0.141 |
Grac | +0.141 |
Neg logit contributions | |
Tom | -0.123 |
booth | -0.122 |
cheating | -0.118 |
worker | -0.115 |
Tomas | -0.115 |
Token | to |
Token ID | 284 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
Tom | +0.055 |
money | +0.050 |
cheating | +0.050 |
worker | +0.049 |
booth | +0.049 |
Neg logit contributions | |
light | -0.119 |
Chick | -0.118 |
burn | -0.118 |
maid | -0.118 |
skirt | -0.116 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
light | +0.040 |
stub | +0.039 |
Chick | +0.039 |
burn | +0.039 |
berry | +0.038 |
Neg logit contributions | |
Tom | -0.040 |
cheating | -0.039 |
booth | -0.038 |
money | -0.037 |
Tomas | -0.037 |
Token | e |
Token ID | 68 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
stub | +0.165 |
burn | +0.156 |
Chick | +0.156 |
Grac | +0.155 |
maid | +0.154 |
Neg logit contributions | |
Tom | -0.272 |
cheating | -0.259 |
money | -0.251 |
Tomas | -0.248 |
booth | -0.246 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.71e+00 |
Pos logit contributions | |
stub | +0.082 |
Grac | +0.077 |
days | +0.077 |
light | +0.073 |
skirt | +0.073 |
Neg logit contributions | |
Tom | -0.177 |
worker | -0.165 |
her | -0.164 |
money | -0.162 |
workers | -0.161 |
Token | and |
Token ID | 290 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
stub | +0.556 |
Chick | +0.548 |
burn | +0.536 |
berry | +0.533 |
light | +0.529 |
Neg logit contributions | |
booth | -0.301 |
Tom | -0.284 |
cheating | -0.281 |
employee | -0.276 |
tickets | -0.275 |
Token | cough |
Token ID | 22094 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
Tom | +0.020 |
cheating | +0.016 |
money | +0.016 |
gas | +0.015 |
workers | +0.015 |
Neg logit contributions | |
burn | -0.027 |
ini | -0.027 |
berry | -0.026 |
maid | -0.026 |
days | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
stub | +0.275 |
berry | +0.269 |
days | +0.265 |
light | +0.265 |
maid | +0.264 |
Neg logit contributions | |
Tom | -0.137 |
her | -0.115 |
workers | -0.112 |
worker | -0.112 |
booth | -0.112 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | +3.83e-02 |
Pos logit contributions | |
Tom | +0.007 |
booth | +0.006 |
Tomas | +0.006 |
Officer | +0.006 |
money | +0.006 |
Neg logit contributions | |
burn | -0.007 |
maid | -0.007 |
stub | -0.007 |
light | -0.007 |
skirt | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
Tom | +0.001 |
money | +0.001 |
booth | +0.001 |
cheating | +0.001 |
Tomas | +0.001 |
Neg logit contributions | |
burn | -0.003 |
stub | -0.003 |
days | -0.003 |
berry | -0.003 |
lull | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
Tom | +0.053 |
cheating | +0.047 |
money | +0.046 |
numbers | +0.044 |
worker | +0.043 |
Neg logit contributions | |
light | -0.129 |
maid | -0.129 |
Chick | -0.127 |
berry | -0.127 |
oppy | -0.126 |
Token | cough |
Token ID | 22094 |
Feature activation | -4.15e+00 |
Pos logit contributions | |
light | +0.044 |
Chick | +0.043 |
berry | +0.043 |
stub | +0.042 |
ifying | +0.042 |
Neg logit contributions | |
Tom | -0.042 |
cheating | -0.041 |
money | -0.038 |
booth | -0.037 |
worker | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +3.28e-01 |
Pos logit contributions | |
stub | +0.316 |
Chick | +0.305 |
berry | +0.302 |
Grac | +0.302 |
burn | +0.300 |
Neg logit contributions | |
Tom | -0.099 |
booth | -0.093 |
cheating | -0.083 |
worker | -0.081 |
Tomas | -0.080 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.40e+00 |
Pos logit contributions | |
Tom | +0.016 |
money | +0.014 |
cheating | +0.014 |
gas | +0.014 |
signs | +0.014 |
Neg logit contributions | |
ini | -0.019 |
berry | -0.018 |
maid | -0.018 |
burn | -0.018 |
days | -0.017 |
Token | e |
Token ID | 68 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
stub | +0.157 |
Grac | +0.150 |
skirt | +0.148 |
maid | +0.147 |
burn | +0.141 |
Neg logit contributions | |
Tom | -0.289 |
cheating | -0.260 |
money | -0.260 |
Tom | -0.258 |
her | -0.256 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.71e+00 |
Pos logit contributions | |
stub | +0.093 |
days | +0.091 |
Grac | +0.091 |
skirt | +0.082 |
berry | +0.081 |
Neg logit contributions | |
Tom | -0.171 |
worker | -0.151 |
her | -0.150 |
money | -0.149 |
tickets | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
stub | +0.571 |
Chick | +0.556 |
burn | +0.548 |
berry | +0.548 |
Grac | +0.543 |
Neg logit contributions | |
Tom | -0.294 |
booth | -0.286 |
cheating | -0.273 |
worker | -0.264 |
employee | -0.262 |
Token |
|
Token ID | 198 |
Feature activation | -3.99e-01 |
Pos logit contributions | |
Tom | +0.005 |
booth | +0.004 |
Tomas | +0.004 |
cheating | +0.004 |
Officer | +0.004 |
Neg logit contributions | |
burn | -0.010 |
light | -0.010 |
stub | -0.010 |
berry | -0.010 |
maid | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
Chick | +0.020 |
stub | +0.020 |
Grac | +0.019 |
berry | +0.019 |
oppy | +0.018 |
Neg logit contributions | |
booth | -0.019 |
cheating | -0.019 |
employee | -0.019 |
tickets | -0.018 |
worker | -0.018 |
Token | " |
Token ID | 1 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
stub | +0.019 |
Chick | +0.018 |
oppy | +0.018 |
light | +0.017 |
berry | +0.017 |
Neg logit contributions | |
booth | -0.018 |
tickets | -0.018 |
inconvenience | -0.017 |
employee | -0.017 |
cheating | -0.017 |
Token | Help |
Token ID | 22087 |
Feature activation | +2.53e+00 |
Pos logit contributions | |
Tom | +0.032 |
Tom | +0.030 |
cheating | +0.027 |
Ready | +0.026 |
her | +0.026 |
Neg logit contributions | |
maid | -0.043 |
Grac | -0.043 |
oppy | -0.043 |
days | -0.042 |
light | -0.041 |
Token | dust |
Token ID | 8977 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
stub | +0.036 |
Chick | +0.034 |
berry | +0.034 |
burn | +0.034 |
light | +0.033 |
Neg logit contributions | |
booth | -0.037 |
Tom | -0.036 |
cheating | -0.035 |
employee | -0.035 |
worker | -0.035 |
Token | made |
Token ID | 925 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
oppy | +0.074 |
burn | +0.072 |
Grac | +0.072 |
Chick | +0.072 |
berry | +0.072 |
Neg logit contributions | |
workers | -0.039 |
Tom | -0.039 |
worker | -0.037 |
man | -0.036 |
site | -0.036 |
Token | him |
Token ID | 683 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
stub | +0.165 |
Chick | +0.162 |
burn | +0.159 |
berry | +0.158 |
light | +0.157 |
Neg logit contributions | |
booth | -0.091 |
Tom | -0.088 |
cheating | -0.086 |
employee | -0.084 |
Tomas | -0.083 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.45e+00 |
Pos logit contributions | |
Chick | +0.010 |
berry | +0.010 |
stub | +0.010 |
Grac | +0.010 |
burn | +0.009 |
Neg logit contributions | |
Tom | -0.006 |
money | -0.006 |
cheating | -0.006 |
tickets | -0.006 |
booth | -0.006 |
Token | e |
Token ID | 68 |
Feature activation | -2.53e+00 |
Pos logit contributions | |
stub | +0.162 |
Chick | +0.152 |
burn | +0.151 |
Grac | +0.150 |
berry | +0.150 |
Neg logit contributions | |
Tom | -0.277 |
cheating | -0.267 |
booth | -0.264 |
her | -0.259 |
Tomas | -0.257 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.71e+00 |
Pos logit contributions | |
stub | +0.087 |
Grac | +0.087 |
berry | +0.085 |
Chick | +0.081 |
oppy | +0.079 |
Neg logit contributions | |
Tom | -0.158 |
her | -0.153 |
numbers | -0.150 |
signs | -0.148 |
money | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
stub | +0.537 |
Chick | +0.530 |
burn | +0.520 |
light | +0.514 |
berry | +0.512 |
Neg logit contributions | |
booth | -0.318 |
Tom | -0.298 |
cheating | -0.296 |
employee | -0.295 |
tickets | -0.292 |
Token | Tim |
Token ID | 5045 |
Feature activation | +1.82e+00 |
Pos logit contributions | |
booth | +0.010 |
cheating | +0.009 |
employee | +0.009 |
tickets | +0.009 |
worker | +0.009 |
Neg logit contributions | |
Chick | -0.011 |
stub | -0.011 |
berry | -0.011 |
Grac | -0.011 |
burn | -0.010 |
Token | 's |
Token ID | 338 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
Tom | +0.076 |
booth | +0.074 |
cheating | +0.072 |
Tomas | +0.069 |
worker | +0.069 |
Neg logit contributions | |
stub | -0.101 |
Chick | -0.100 |
burn | -0.099 |
berry | -0.098 |
Grac | -0.097 |
Token | face |
Token ID | 1986 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
Grac | +0.025 |
Chick | +0.025 |
berry | +0.025 |
stub | +0.024 |
streamed | +0.024 |
Neg logit contributions | |
Tom | -0.020 |
money | -0.019 |
booth | -0.019 |
worker | -0.019 |
tickets | -0.019 |
Token | turned |
Token ID | 2900 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
ubb | +0.087 |
oppy | +0.085 |
berry | +0.085 |
rys | +0.083 |
Grac | +0.083 |
Neg logit contributions | |
her | -0.042 |
Tom | -0.039 |
numbers | -0.038 |
fun | -0.036 |
harder | -0.036 |
Token | her |
Token ID | 607 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
stub | +0.082 |
burn | +0.079 |
Chick | +0.079 |
berry | +0.078 |
light | +0.076 |
Neg logit contributions | |
Tom | -0.041 |
booth | -0.039 |
cheating | -0.039 |
money | -0.037 |
tickets | -0.037 |
Token | cough |
Token ID | 22094 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
stub | +0.089 |
Chick | +0.089 |
streamed | +0.087 |
berry | +0.087 |
burn | +0.086 |
Neg logit contributions | |
Tom | -0.078 |
money | -0.076 |
booth | -0.074 |
cheating | -0.073 |
tickets | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | +5.35e-01 |
Pos logit contributions | |
stub | +0.301 |
Chick | +0.289 |
berry | +0.288 |
Grac | +0.285 |
oppy | +0.284 |
Neg logit contributions | |
Tom | -0.112 |
her | -0.101 |
booth | -0.099 |
cheating | -0.097 |
money | -0.096 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
Tom | +0.022 |
cheating | +0.020 |
money | +0.019 |
numbers | +0.018 |
old | +0.018 |
Neg logit contributions | |
berry | -0.033 |
burn | -0.033 |
stub | -0.031 |
maid | -0.031 |
days | -0.031 |
Token | e |
Token ID | 68 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
stub | +0.157 |
maid | +0.151 |
skirt | +0.151 |
Grac | +0.149 |
berry | +0.144 |
Neg logit contributions | |
Tom | -0.279 |
cheating | -0.263 |
her | -0.263 |
Tom | -0.253 |
money | -0.251 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.70e+00 |
Pos logit contributions | |
stub | +0.089 |
Grac | +0.088 |
berry | +0.085 |
days | +0.081 |
rys | +0.081 |
Neg logit contributions | |
Tom | -0.147 |
her | -0.145 |
money | -0.136 |
wrong | -0.134 |
numbers | -0.134 |
Token | . |
Token ID | 13 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
stub | +0.591 |
berry | +0.576 |
Chick | +0.573 |
oppy | +0.566 |
Grac | +0.565 |
Neg logit contributions | |
Tom | -0.283 |
cheating | -0.253 |
booth | -0.249 |
workers | -0.243 |
her | -0.240 |
Token | She |
Token ID | 1375 |
Feature activation | -5.90e-01 |
Pos logit contributions | |
Tom | +0.017 |
booth | +0.017 |
Tomas | +0.016 |
cheating | +0.016 |
employee | +0.016 |
Neg logit contributions | |
burn | -0.020 |
stub | -0.020 |
light | -0.020 |
Chick | -0.019 |
berry | -0.019 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.03e+00 |
Pos logit contributions | |
burn | +0.032 |
stub | +0.032 |
Chick | +0.032 |
berry | +0.031 |
light | +0.031 |
Neg logit contributions | |
Tom | -0.025 |
booth | -0.025 |
cheating | -0.024 |
employee | -0.023 |
Tomas | -0.023 |
Token | sorry |
Token ID | 7926 |
Feature activation | -2.40e+00 |
Pos logit contributions | |
burn | +0.209 |
berry | +0.205 |
light | +0.203 |
beam | +0.199 |
maid | +0.196 |
Neg logit contributions | |
Tom | -0.098 |
money | -0.093 |
old | -0.090 |
cheating | -0.088 |
her | -0.086 |
Token | for |
Token ID | 329 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
stub | +0.202 |
burn | +0.196 |
berry | +0.196 |
Chick | +0.193 |
light | +0.191 |
Neg logit contributions | |
Tom | -0.040 |
cheating | -0.034 |
booth | -0.034 |
money | -0.030 |
worker | -0.028 |
Token | made |
Token ID | 925 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
Tom | +0.016 |
cheating | +0.013 |
booth | +0.013 |
numbers | +0.013 |
money | +0.013 |
Neg logit contributions | |
berry | -0.023 |
days | -0.022 |
oppy | -0.022 |
burn | -0.022 |
stub | -0.022 |
Token | a |
Token ID | 257 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
stub | +0.188 |
burn | +0.182 |
berry | +0.182 |
Chick | +0.180 |
light | +0.178 |
Neg logit contributions | |
Tom | -0.078 |
booth | -0.074 |
cheating | -0.070 |
tickets | -0.069 |
money | -0.068 |
Token | loud |
Token ID | 7812 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
booth | +0.014 |
Tom | +0.013 |
cheating | +0.013 |
worker | +0.013 |
money | +0.013 |
Neg logit contributions | |
stub | -0.030 |
burn | -0.029 |
berry | -0.029 |
Chick | -0.029 |
light | -0.029 |
Token | sne |
Token ID | 10505 |
Feature activation | -4.38e+00 |
Pos logit contributions | |
stub | +0.001 |
Chick | +0.001 |
burn | +0.001 |
berry | +0.001 |
oppy | +0.001 |
Neg logit contributions | |
Tom | -0.000 |
cheating | -0.000 |
booth | -0.000 |
worker | -0.000 |
site | -0.000 |
Token | e |
Token ID | 68 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
Grac | +0.187 |
stub | +0.185 |
Chick | +0.179 |
maid | +0.178 |
skirt | +0.177 |
Neg logit contributions | |
Tom | -0.242 |
her | -0.232 |
cheating | -0.227 |
Tom | -0.225 |
inconvenience | -0.223 |
Token | ze |
Token ID | 2736 |
Feature activation | -8.69e+00 |
Pos logit contributions | |
Grac | +0.098 |
oppy | +0.095 |
stub | +0.093 |
berry | +0.091 |
Chick | +0.089 |
Neg logit contributions | |
Tom | -0.153 |
sign | -0.148 |
signs | -0.148 |
worker | -0.147 |
her | -0.147 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
stub | +0.632 |
oppy | +0.625 |
Grac | +0.619 |
Chick | +0.616 |
berry | +0.608 |
Neg logit contributions | |
Tom | -0.239 |
booth | -0.219 |
cheating | -0.216 |
worker | -0.206 |
workers | -0.199 |
Token | " |
Token ID | 366 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
booth | +0.007 |
Tom | +0.007 |
cheating | +0.006 |
Tomas | +0.006 |
employee | +0.006 |
Neg logit contributions | |
stub | -0.009 |
burn | -0.009 |
Chick | -0.009 |
berry | -0.009 |
light | -0.009 |
Token | A |
Token ID | 32 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
Tom | +0.045 |
Tom | +0.043 |
booth | +0.043 |
her | +0.043 |
cheating | +0.043 |
Neg logit contributions | |
maid | -0.067 |
berry | -0.067 |
stub | -0.066 |
Chick | -0.065 |
Grac | -0.065 |
Token | ch |
Token ID | 354 |
Feature activation | -2.98e-01 |
Pos logit contributions | |
booth | +0.076 |
Tom | +0.076 |
worker | +0.075 |
cheating | +0.074 |
inconvenience | +0.072 |
Neg logit contributions | |
Grac | -0.116 |
light | -0.116 |
stub | -0.115 |
Chick | -0.114 |
burn | -0.112 |
Token | oo |
Token ID | 2238 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
numb | +0.019 |
Grac | +0.018 |
Chick | +0.018 |
light | +0.018 |
stub | +0.018 |
Neg logit contributions | |
Tom | -0.011 |
worker | -0.010 |
inconvenience | -0.010 |
her | -0.008 |
site | -0.008 |
Token | place |
Token ID | 1295 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
cheating | +0.111 |
Tomas | +0.109 |
Tom | +0.109 |
nine | +0.103 |
booth | +0.103 |
Neg logit contributions | |
stub | -0.158 |
berry | -0.155 |
burn | -0.152 |
skirt | -0.151 |
Chick | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
light | +0.002 |
stub | +0.002 |
days | +0.002 |
skirt | +0.002 |
berry | +0.002 |
Neg logit contributions | |
Tomas | -0.002 |
cheating | -0.002 |
nine | -0.002 |
2 | -0.002 |
Officer | -0.002 |
Token | It |
Token ID | 632 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
booth | +0.011 |
Tom | +0.011 |
cheating | +0.011 |
employee | +0.010 |
Tomas | +0.010 |
Neg logit contributions | |
stub | -0.012 |
burn | -0.011 |
Chick | -0.011 |
light | -0.011 |
berry | -0.011 |
Token | is |
Token ID | 318 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
booth | +0.014 |
employee | +0.014 |
cheating | +0.013 |
Tomas | +0.013 |
Tom | +0.013 |
Neg logit contributions | |
stub | -0.025 |
Chick | -0.024 |
berry | -0.024 |
light | -0.023 |
burn | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
inconvenience | +0.031 |
booth | +0.031 |
2 | +0.031 |
Tomas | +0.030 |
employee | +0.030 |
Neg logit contributions | |
Chick | -0.044 |
stub | -0.043 |
skirt | -0.042 |
numb | -0.041 |
burn | -0.041 |
Token | museum |
Token ID | 13257 |
Feature activation | +5.79e+00 |
Pos logit contributions | |
Tom | +0.019 |
booth | +0.019 |
Tomas | +0.019 |
cheating | +0.019 |
employee | +0.019 |
Neg logit contributions | |
stub | -0.025 |
Chick | -0.024 |
burn | -0.024 |
berry | -0.023 |
skirt | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
inconvenience | +0.308 |
Tomas | +0.306 |
icult | +0.297 |
cheating | +0.293 |
40 | +0.290 |
Neg logit contributions | |
light | -0.235 |
berry | -0.215 |
skirt | -0.214 |
ouse | -0.198 |
days | -0.198 |
Token | There |
Token ID | 1318 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
Tom | +0.010 |
booth | +0.010 |
cheating | +0.010 |
employee | +0.009 |
Tomas | +0.009 |
Neg logit contributions | |
stub | -0.013 |
burn | -0.013 |
Chick | -0.013 |
light | -0.012 |
berry | -0.012 |
Token | are |
Token ID | 389 |
Feature activation | -3.72e-02 |
Pos logit contributions | |
booth | +0.090 |
cheating | +0.088 |
employee | +0.087 |
Tomas | +0.086 |
Tom | +0.084 |
Neg logit contributions | |
stub | -0.099 |
Chick | -0.098 |
berry | -0.096 |
light | -0.094 |
burn | -0.093 |
Token | many |
Token ID | 867 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
stub | +0.002 |
Chick | +0.002 |
light | +0.002 |
berry | +0.002 |
Grac | +0.002 |
Neg logit contributions | |
Tomas | -0.002 |
employee | -0.001 |
cheating | -0.001 |
booth | -0.001 |
inconvenience | -0.001 |
Token | pictures |
Token ID | 5986 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
Tomas | +0.042 |
employee | +0.039 |
booth | +0.039 |
her | +0.039 |
inconvenience | +0.039 |
Neg logit contributions | |
berry | -0.045 |
Grac | -0.042 |
skirt | -0.041 |
light | -0.041 |
stub | -0.041 |
Token | couldn |
Token ID | 3521 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
burn | +0.125 |
stub | +0.124 |
Chick | +0.124 |
berry | +0.120 |
light | +0.119 |
Neg logit contributions | |
Tom | -0.075 |
booth | -0.075 |
cheating | -0.073 |
worker | -0.069 |
Tomas | -0.068 |
Token | â |
Token ID | 22940 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
light | +0.039 |
stub | +0.039 |
skirt | +0.039 |
maid | +0.039 |
burn | +0.038 |
Neg logit contributions | |
Tom | -0.030 |
booth | -0.028 |
Tomas | -0.027 |
money | -0.027 |
cheating | -0.027 |
Token | € |
Token ID | 26391 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
stub | +0.021 |
burn | +0.021 |
berry | +0.020 |
streamed | +0.020 |
Chick | +0.020 |
Neg logit contributions | |
booth | -0.017 |
tickets | -0.017 |
cheating | -0.016 |
Tom | -0.016 |
clues | -0.016 |
Token | ™ |
Token ID | 8151 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
Tom | +0.200 |
booth | +0.195 |
Tomas | +0.185 |
cheating | +0.178 |
tickets | +0.177 |
Neg logit contributions | |
Chick | -0.198 |
stub | -0.196 |
burn | -0.193 |
Grac | -0.192 |
berry | -0.191 |
Token | t |
Token ID | 83 |
Feature activation | -4.08e-01 |
Pos logit contributions | |
Chick | +0.195 |
stub | +0.189 |
skirt | +0.186 |
light | +0.186 |
burn | +0.184 |
Neg logit contributions | |
Tom | -0.185 |
booth | -0.182 |
Tomas | -0.169 |
worker | -0.167 |
employee | -0.166 |
Token | believe |
Token ID | 1975 |
Feature activation | +6.00e+00 |
Pos logit contributions | |
stub | +0.025 |
burn | +0.025 |
Chick | +0.025 |
light | +0.025 |
berry | +0.025 |
Neg logit contributions | |
money | -0.014 |
Tom | -0.014 |
employee | -0.013 |
cheating | -0.013 |
worker | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
Tom | +0.174 |
cheating | +0.144 |
booth | +0.143 |
employee | +0.135 |
Tomas | +0.128 |
Neg logit contributions | |
burn | -0.441 |
stub | -0.436 |
light | -0.419 |
berry | -0.419 |
Chick | -0.415 |
Token | . |
Token ID | 13 |
Feature activation | +7.59e-01 |
Pos logit contributions | |
booth | +0.016 |
Tom | +0.016 |
cheating | +0.015 |
employee | +0.015 |
Tomas | +0.015 |
Neg logit contributions | |
stub | -0.026 |
burn | -0.025 |
Chick | -0.025 |
berry | -0.024 |
light | -0.024 |
Token | He |
Token ID | 679 |
Feature activation | +5.05e-01 |
Pos logit contributions | |
booth | +0.034 |
Tom | +0.034 |
cheating | +0.033 |
employee | +0.032 |
Tomas | +0.032 |
Neg logit contributions | |
stub | -0.040 |
Chick | -0.039 |
burn | -0.039 |
berry | -0.038 |
light | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
booth | +0.021 |
Tom | +0.021 |
cheating | +0.020 |
employee | +0.020 |
Tomas | +0.020 |
Neg logit contributions | |
burn | -0.028 |
stub | -0.028 |
Chick | -0.028 |
berry | -0.027 |
light | -0.027 |
Token | so |
Token ID | 523 |
Feature activation | -3.59e-01 |
Pos logit contributions | |
stub | +0.007 |
Chick | +0.007 |
burn | +0.007 |
berry | +0.007 |
light | +0.007 |
Neg logit contributions | |
booth | -0.004 |
Tom | -0.004 |
cheating | -0.004 |
employee | -0.004 |
Tomas | -0.004 |
Token | elephants |
Token ID | 25368 |
Feature activation | +2.16e+00 |
Pos logit contributions | |
stub | +0.032 |
Chick | +0.032 |
berry | +0.032 |
burn | +0.031 |
light | +0.031 |
Neg logit contributions | |
booth | -0.017 |
employee | -0.017 |
cheating | -0.016 |
Tom | -0.016 |
Tomas | -0.016 |
Token | , |
Token ID | 11 |
Feature activation | -5.01e-01 |
Pos logit contributions | |
employee | +0.105 |
cheating | +0.103 |
inconvenience | +0.103 |
booth | +0.101 |
Tomas | +0.100 |
Neg logit contributions | |
berry | -0.112 |
Chick | -0.105 |
light | -0.098 |
burn | -0.096 |
berries | -0.095 |
Token | monkeys |
Token ID | 26697 |
Feature activation | +2.56e+00 |
Pos logit contributions | |
Chick | +0.033 |
berry | +0.032 |
stub | +0.032 |
burn | +0.032 |
light | +0.031 |
Neg logit contributions | |
booth | -0.016 |
employee | -0.016 |
cheating | -0.015 |
Tomas | -0.015 |
Tom | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
employee | +0.117 |
booth | +0.116 |
cheating | +0.115 |
inconvenience | +0.114 |
Tomas | +0.111 |
Neg logit contributions | |
berry | -0.136 |
Chick | -0.134 |
burn | -0.125 |
light | -0.124 |
stub | -0.123 |
Token | gir |
Token ID | 37370 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
booth | +0.013 |
Tom | +0.012 |
cheating | +0.012 |
employee | +0.012 |
tickets | +0.012 |
Neg logit contributions | |
stub | -0.023 |
Chick | -0.022 |
burn | -0.022 |
berry | -0.022 |
light | -0.021 |
Token | aff |
Token ID | 2001 |
Feature activation | +6.94e+00 |
Pos logit contributions | |
Chick | +0.030 |
berry | +0.029 |
burn | +0.028 |
stub | +0.028 |
light | +0.027 |
Neg logit contributions | |
booth | -0.021 |
cheating | -0.020 |
employee | -0.020 |
Tomas | -0.019 |
site | -0.019 |
Token | es |
Token ID | 274 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
Tom | +0.443 |
booth | +0.433 |
cheating | +0.424 |
tickets | +0.420 |
worker | +0.415 |
Neg logit contributions | |
stub | -0.250 |
burn | -0.224 |
Chick | -0.223 |
light | -0.217 |
skirt | -0.216 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Chick | +0.175 |
stub | +0.175 |
berry | +0.173 |
burn | +0.170 |
light | +0.168 |
Neg logit contributions | |
booth | -0.123 |
cheating | -0.119 |
Tom | -0.117 |
employee | -0.117 |
Tomas | -0.115 |
Token | They |
Token ID | 1119 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
booth | +0.006 |
cheating | +0.006 |
employee | +0.006 |
worker | +0.006 |
site | +0.006 |
Neg logit contributions | |
Chick | -0.007 |
stub | -0.007 |
berry | -0.007 |
Grac | -0.007 |
burn | -0.007 |
Token | pretend |
Token ID | 16614 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
booth | +0.000 |
Tom | +0.000 |
cheating | +0.000 |
employee | +0.000 |
Tomas | +0.000 |
Neg logit contributions | |
stub | -0.000 |
Chick | -0.000 |
burn | -0.000 |
berry | -0.000 |
light | -0.000 |
Token | that |
Token ID | 326 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
stub | +0.211 |
burn | +0.206 |
Chick | +0.203 |
berry | +0.202 |
light | +0.200 |
Neg logit contributions | |
Tom | -0.110 |
booth | -0.108 |
tickets | -0.103 |
cheating | -0.100 |
workers | -0.099 |
Token | saw |
Token ID | 2497 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
stub | +0.059 |
Chick | +0.058 |
berry | +0.056 |
Grac | +0.055 |
light | +0.055 |
Neg logit contributions | |
booth | -0.052 |
cheating | -0.052 |
Tom | -0.050 |
employee | -0.050 |
Tomas | -0.050 |
Token | lots |
Token ID | 6041 |
Feature activation | -2.93e+00 |
Pos logit contributions | |
cheating | +0.012 |
inconvenience | +0.012 |
employee | +0.012 |
booth | +0.012 |
nine | +0.012 |
Neg logit contributions | |
Chick | -0.017 |
Grac | -0.016 |
stub | -0.016 |
skirt | -0.016 |
berry | -0.016 |
Token | of |
Token ID | 286 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
berry | +0.082 |
Grac | +0.069 |
Chick | +0.067 |
skirt | +0.065 |
streamed | +0.064 |
Neg logit contributions | |
Tomas | -0.208 |
cheating | -0.205 |
booth | -0.204 |
40 | -0.204 |
inconvenience | -0.201 |
Token | colourful |
Token ID | 44799 |
Feature activation | +7.20e-01 |
Pos logit contributions | |
stub | +0.097 |
Chick | +0.097 |
berry | +0.096 |
Grac | +0.095 |
skirt | +0.094 |
Neg logit contributions | |
booth | -0.067 |
Tomas | -0.064 |
Tom | -0.062 |
cheating | -0.062 |
employee | -0.061 |
Token | cereal |
Token ID | 33158 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
booth | +0.030 |
cheating | +0.029 |
Tomas | +0.029 |
Tom | +0.028 |
employee | +0.028 |
Neg logit contributions | |
berry | -0.041 |
Chick | -0.040 |
stub | -0.040 |
Grac | -0.039 |
light | -0.039 |
Token | boxes |
Token ID | 10559 |
Feature activation | +5.39e+00 |
Pos logit contributions | |
Officer | +0.014 |
nine | +0.014 |
inconvenience | +0.014 |
Tomas | +0.014 |
Ready | +0.013 |
Neg logit contributions | |
berry | -0.013 |
light | -0.013 |
skirt | -0.012 |
Chick | -0.012 |
berries | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
cheating | +0.258 |
Tomas | +0.252 |
nine | +0.249 |
Officer | +0.247 |
inconvenience | +0.243 |
Neg logit contributions | |
berry | -0.265 |
light | -0.261 |
streaming | -0.250 |
stub | -0.248 |
burn | -0.247 |
Token |
|
Token ID | 198 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
cheating | +0.030 |
inconvenience | +0.028 |
employee | +0.027 |
booth | +0.027 |
tickets | +0.027 |
Neg logit contributions | |
Grac | -0.024 |
Chick | -0.023 |
berry | -0.023 |
stub | -0.023 |
M | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
cheating | +0.010 |
bills | +0.009 |
inconvenience | +0.009 |
booth | +0.009 |
employee | +0.009 |
Neg logit contributions | |
stub | -0.007 |
Chick | -0.007 |
Grac | -0.007 |
berry | -0.006 |
bowl | -0.006 |
Token | J |
Token ID | 41 |
Feature activation | -3.47e-01 |
Pos logit contributions | |
tickets | +0.008 |
cheating | +0.008 |
booth | +0.008 |
money | +0.008 |
inconvenience | +0.008 |
Neg logit contributions | |
stub | -0.008 |
Chick | -0.008 |
Grac | -0.008 |
berry | -0.008 |
streamed | -0.008 |
Token | ill |
Token ID | 359 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
stub | +0.013 |
berry | +0.012 |
Chick | +0.012 |
burn | +0.012 |
light | +0.012 |
Neg logit contributions | |
booth | -0.021 |
cheating | -0.021 |
Tom | -0.020 |
employee | -0.020 |
Tomas | -0.020 |
Token | fun |
Token ID | 1257 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
booth | +0.083 |
cheating | +0.076 |
inconvenience | +0.076 |
employee | +0.076 |
Tomas | +0.075 |
Neg logit contributions | |
Chick | -0.119 |
Grac | -0.115 |
stub | -0.114 |
berry | -0.114 |
skirt | -0.113 |
Token | and |
Token ID | 290 |
Feature activation | +8.19e-01 |
Pos logit contributions | |
booth | +0.108 |
Tom | +0.103 |
cheating | +0.100 |
employee | +0.098 |
Tomas | +0.097 |
Neg logit contributions | |
stub | -0.217 |
Chick | -0.214 |
burn | -0.209 |
berry | -0.209 |
light | -0.207 |
Token | different |
Token ID | 1180 |
Feature activation | +3.04e+00 |
Pos logit contributions | |
booth | +0.031 |
Tom | +0.031 |
cheating | +0.030 |
employee | +0.029 |
money | +0.029 |
Neg logit contributions | |
stub | -0.049 |
Chick | -0.048 |
berry | -0.048 |
burn | -0.048 |
light | -0.046 |
Token | ways |
Token ID | 2842 |
Feature activation | -6.97e-01 |
Pos logit contributions | |
booth | +0.112 |
Tom | +0.106 |
cheating | +0.103 |
Tomas | +0.103 |
employee | +0.102 |
Neg logit contributions | |
stub | -0.186 |
Chick | -0.184 |
berry | -0.181 |
light | -0.179 |
burn | -0.179 |
Token | to |
Token ID | 284 |
Feature activation | -4.67e-01 |
Pos logit contributions | |
stub | +0.042 |
Chick | +0.042 |
burn | +0.041 |
berry | +0.041 |
light | +0.040 |
Neg logit contributions | |
booth | -0.026 |
Tom | -0.026 |
cheating | -0.025 |
tickets | -0.024 |
employee | -0.024 |
Token | work |
Token ID | 670 |
Feature activation | +5.36e+00 |
Pos logit contributions | |
stub | +0.026 |
Chick | +0.025 |
burn | +0.025 |
berry | +0.025 |
light | +0.024 |
Neg logit contributions | |
booth | -0.020 |
Tom | -0.020 |
cheating | -0.019 |
employee | -0.019 |
Tomas | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +5.22e-01 |
Pos logit contributions | |
booth | +0.245 |
employee | +0.228 |
Tomas | +0.227 |
Tom | +0.223 |
tickets | +0.222 |
Neg logit contributions | |
Chick | -0.276 |
Grac | -0.272 |
stub | -0.270 |
light | -0.267 |
burn | -0.264 |
Token | One |
Token ID | 1881 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
Tom | +0.022 |
booth | +0.021 |
Tomas | +0.019 |
employee | +0.019 |
tickets | +0.018 |
Neg logit contributions | |
stub | -0.031 |
burn | -0.030 |
numb | -0.029 |
Chick | -0.029 |
light | -0.029 |
Token | day |
Token ID | 1110 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
booth | +0.055 |
Tom | +0.054 |
cheating | +0.052 |
employee | +0.051 |
worker | +0.051 |
Neg logit contributions | |
stub | -0.081 |
Chick | -0.079 |
burn | -0.078 |
berry | -0.077 |
Grac | -0.076 |
Token | , |
Token ID | 11 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
stub | +0.055 |
burn | +0.054 |
numb | +0.053 |
light | +0.053 |
Chick | +0.053 |
Neg logit contributions | |
Tom | -0.028 |
booth | -0.024 |
money | -0.023 |
cheating | -0.023 |
Tomas | -0.022 |
Token | she |
Token ID | 673 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
stub | +0.015 |
burn | +0.015 |
lull | +0.015 |
light | +0.015 |
ivering | +0.015 |
Neg logit contributions | |
Tom | -0.012 |
money | -0.010 |
booth | -0.010 |
worker | -0.010 |
Tomas | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +4.64e-01 |
Pos logit contributions | |
stub | +0.046 |
light | +0.045 |
streamed | +0.044 |
berry | +0.044 |
skirt | +0.043 |
Neg logit contributions | |
booth | -0.041 |
Tomas | -0.039 |
employee | -0.039 |
tickets | -0.039 |
Tom | -0.039 |
Token | He |
Token ID | 679 |
Feature activation | +3.87e-01 |
Pos logit contributions | |
cheating | +0.023 |
booth | +0.023 |
employee | +0.022 |
tickets | +0.022 |
worker | +0.022 |
Neg logit contributions | |
stub | -0.023 |
Chick | -0.022 |
berry | -0.022 |
Grac | -0.021 |
light | -0.021 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.13e-01 |
Pos logit contributions | |
booth | +0.018 |
Tom | +0.017 |
cheating | +0.017 |
employee | +0.017 |
Tomas | +0.017 |
Neg logit contributions | |
stub | -0.020 |
Chick | -0.020 |
burn | -0.019 |
berry | -0.019 |
light | -0.019 |
Token | pepper |
Token ID | 13385 |
Feature activation | +1.71e+00 |
Pos logit contributions | |
booth | +0.005 |
Tom | +0.005 |
cheating | +0.005 |
employee | +0.005 |
tickets | +0.005 |
Neg logit contributions | |
stub | -0.016 |
Chick | -0.015 |
burn | -0.015 |
berry | -0.015 |
light | -0.015 |
Token | sprinkled |
Token ID | 49692 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
booth | +0.070 |
Tom | +0.067 |
cheating | +0.066 |
employee | +0.065 |
Tomas | +0.064 |
Neg logit contributions | |
stub | -0.098 |
Chick | -0.097 |
berry | -0.095 |
burn | -0.095 |
light | -0.094 |
Token | in |
Token ID | 287 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
stub | +0.117 |
burn | +0.113 |
Chick | +0.113 |
light | +0.110 |
berry | +0.110 |
Neg logit contributions | |
Tom | -0.065 |
booth | -0.065 |
cheating | -0.061 |
money | -0.061 |
tickets | -0.061 |
Token | the |
Token ID | 262 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
booth | +0.043 |
employee | +0.041 |
cheating | +0.040 |
Tomas | +0.039 |
Tom | +0.039 |
Neg logit contributions | |
Chick | -0.078 |
stub | -0.077 |
berry | -0.077 |
Grac | -0.075 |
light | -0.074 |
Token | grass |
Token ID | 8701 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
booth | +0.026 |
cheating | +0.026 |
Tom | +0.025 |
Tomas | +0.025 |
employee | +0.025 |
Neg logit contributions | |
stub | -0.031 |
Chick | -0.031 |
berry | -0.030 |
light | -0.030 |
burn | -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +6.35e-01 |
Pos logit contributions | |
booth | +0.106 |
Tom | +0.103 |
cheating | +0.098 |
employee | +0.095 |
worker | +0.095 |
Neg logit contributions | |
stub | -0.198 |
Chick | -0.194 |
burn | -0.190 |
berry | -0.187 |
Grac | -0.186 |
Token | decided |
Token ID | 3066 |
Feature activation | -8.54e-01 |
Pos logit contributions | |
booth | +0.027 |
Tom | +0.027 |
cheating | +0.026 |
employee | +0.025 |
tickets | +0.025 |
Neg logit contributions | |
stub | -0.035 |
burn | -0.034 |
Chick | -0.034 |
berry | -0.033 |
light | -0.033 |
Token | to |
Token ID | 284 |
Feature activation | -6.88e-01 |
Pos logit contributions | |
stub | +0.043 |
Chick | +0.040 |
Grac | +0.039 |
burn | +0.038 |
berry | +0.038 |
Neg logit contributions | |
booth | -0.041 |
employee | -0.040 |
cheating | -0.039 |
Tomas | -0.039 |
Tom | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | +4.32e-01 |
Pos logit contributions | |
Tom | +0.058 |
booth | +0.057 |
worker | +0.054 |
cheating | +0.053 |
site | +0.051 |
Neg logit contributions | |
stub | -0.106 |
burn | -0.104 |
Chick | -0.103 |
Grac | -0.103 |
numb | -0.101 |
Token | He |
Token ID | 679 |
Feature activation | +2.64e-01 |
Pos logit contributions | |
Tom | +0.019 |
booth | +0.019 |
cheating | +0.018 |
Tomas | +0.018 |
employee | +0.018 |
Neg logit contributions | |
stub | -0.024 |
burn | -0.023 |
Chick | -0.023 |
light | -0.022 |
berry | -0.022 |
Token | lived |
Token ID | 5615 |
Feature activation | -8.86e-01 |
Pos logit contributions | |
booth | +0.014 |
Tom | +0.014 |
cheating | +0.014 |
employee | +0.013 |
tickets | +0.013 |
Neg logit contributions | |
stub | -0.012 |
Chick | -0.012 |
berry | -0.011 |
burn | -0.011 |
light | -0.011 |
Token | in |
Token ID | 287 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
stub | +0.047 |
Chick | +0.045 |
burn | +0.043 |
berry | +0.043 |
Grac | +0.042 |
Neg logit contributions | |
booth | -0.041 |
Tom | -0.039 |
cheating | -0.039 |
employee | -0.039 |
tickets | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | +6.10e-01 |
Pos logit contributions | |
booth | +0.039 |
cheating | +0.036 |
employee | +0.036 |
worker | +0.034 |
inconvenience | +0.034 |
Neg logit contributions | |
Chick | -0.075 |
stub | -0.073 |
light | -0.071 |
burn | -0.070 |
berry | -0.069 |
Token | jungle |
Token ID | 20712 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
booth | +0.025 |
cheating | +0.025 |
employee | +0.025 |
Tomas | +0.024 |
Tom | +0.024 |
Neg logit contributions | |
stub | -0.034 |
Chick | -0.034 |
berry | -0.033 |
burn | -0.033 |
light | -0.032 |
Token | with |
Token ID | 351 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
booth | +0.111 |
employee | +0.111 |
Tomas | +0.110 |
cheating | +0.110 |
tickets | +0.109 |
Neg logit contributions | |
stub | -0.106 |
light | -0.103 |
berry | -0.100 |
Chick | -0.098 |
bowl | -0.094 |
Token | his |
Token ID | 465 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
stub | +0.098 |
burn | +0.097 |
ivering | +0.095 |
berry | +0.095 |
light | +0.094 |
Neg logit contributions | |
Tom | -0.056 |
money | -0.047 |
numbers | -0.047 |
tickets | -0.046 |
old | -0.042 |
Token | friends |
Token ID | 2460 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
Chick | +0.060 |
stub | +0.059 |
berry | +0.057 |
burn | +0.056 |
light | +0.055 |
Neg logit contributions | |
booth | -0.074 |
cheating | -0.072 |
employee | -0.071 |
Tom | -0.069 |
Tomas | -0.069 |
Token | . |
Token ID | 13 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
stub | +0.093 |
Chick | +0.091 |
burn | +0.090 |
berry | +0.090 |
light | +0.088 |
Neg logit contributions | |
booth | -0.060 |
Tom | -0.057 |
cheating | -0.057 |
employee | -0.056 |
Tomas | -0.055 |
Token | One |
Token ID | 1881 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
booth | +0.018 |
cheating | +0.018 |
employee | +0.017 |
tickets | +0.017 |
Tomas | +0.017 |
Neg logit contributions | |
Chick | -0.018 |
stub | -0.018 |
berry | -0.018 |
Grac | -0.017 |
light | -0.017 |
Token | is |
Token ID | 318 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
booth | +0.005 |
Tom | +0.005 |
cheating | +0.005 |
employee | +0.005 |
Tomas | +0.005 |
Neg logit contributions | |
stub | -0.009 |
Chick | -0.008 |
burn | -0.008 |
berry | -0.008 |
light | -0.008 |
Token | so |
Token ID | 523 |
Feature activation | -6.61e-01 |
Pos logit contributions | |
booth | +0.033 |
Tom | +0.032 |
cheating | +0.031 |
employee | +0.030 |
Tomas | +0.030 |
Neg logit contributions | |
stub | -0.062 |
Chick | -0.061 |
burn | -0.060 |
berry | -0.060 |
light | -0.059 |
Token | adorable |
Token ID | 29012 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
stub | +0.042 |
burn | +0.041 |
Chick | +0.041 |
berry | +0.040 |
light | +0.040 |
Neg logit contributions | |
Tom | -0.024 |
booth | -0.023 |
cheating | -0.022 |
employee | -0.022 |
worker | -0.022 |
Token | ! |
Token ID | 0 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
Chick | +0.060 |
stub | +0.060 |
berry | +0.060 |
light | +0.059 |
burn | +0.059 |
Neg logit contributions | |
booth | -0.043 |
cheating | -0.041 |
Tom | -0.040 |
tickets | -0.040 |
employee | -0.040 |
Token | Zip |
Token ID | 38636 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
Chick | +0.001 |
stub | +0.001 |
berry | +0.001 |
Grac | +0.001 |
light | +0.001 |
Neg logit contributions | |
booth | -0.001 |
cheating | -0.001 |
employee | -0.001 |
worker | -0.001 |
tickets | -0.001 |
Token | is |
Token ID | 318 |
Feature activation | +9.59e-01 |
Pos logit contributions | |
booth | +0.121 |
Tom | +0.119 |
cheating | +0.114 |
employee | +0.111 |
tickets | +0.111 |
Neg logit contributions | |
stub | -0.197 |
Chick | -0.193 |
burn | -0.191 |
berry | -0.187 |
light | -0.186 |
Token | the |
Token ID | 262 |
Feature activation | +6.85e-01 |
Pos logit contributions | |
Tom | +0.030 |
booth | +0.028 |
cheating | +0.027 |
employee | +0.026 |
Tomas | +0.026 |
Neg logit contributions | |
stub | -0.066 |
burn | -0.065 |
berry | -0.064 |
Chick | -0.064 |
light | -0.063 |
Token | perfect |
Token ID | 2818 |
Feature activation | -1.84e-01 |
Pos logit contributions | |
booth | +0.029 |
Tom | +0.028 |
cheating | +0.028 |
site | +0.027 |
Tomas | +0.027 |
Neg logit contributions | |
stub | -0.038 |
burn | -0.037 |
Chick | -0.037 |
berry | -0.036 |
Grac | -0.036 |
Token | addition |
Token ID | 3090 |
Feature activation | -5.77e-01 |
Pos logit contributions | |
stub | +0.010 |
burn | +0.010 |
berry | +0.010 |
Chick | +0.010 |
skirt | +0.009 |
Neg logit contributions | |
booth | -0.008 |
Tom | -0.008 |
cheating | -0.008 |
site | -0.007 |
Tomas | -0.007 |
Token | to |
Token ID | 284 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
stub | +0.031 |
berry | +0.031 |
Chick | +0.031 |
burn | +0.030 |
light | +0.030 |
Neg logit contributions | |
booth | -0.024 |
cheating | -0.024 |
Tom | -0.023 |
Tomas | -0.023 |
employee | -0.023 |
Token | the |
Token ID | 262 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
stub | +0.037 |
Chick | +0.036 |
burn | +0.035 |
berry | +0.035 |
light | +0.035 |
Neg logit contributions | |
booth | -0.026 |
Tom | -0.025 |
cheating | -0.025 |
employee | -0.024 |
Tomas | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +4.09e-01 |
Pos logit contributions | |
booth | +0.063 |
Tom | +0.061 |
cheating | +0.059 |
employee | +0.058 |
tickets | +0.057 |
Neg logit contributions | |
stub | -0.097 |
Chick | -0.095 |
burn | -0.093 |
berry | -0.093 |
light | -0.092 |
Token | the |
Token ID | 262 |
Feature activation | +2.31e-01 |
Pos logit contributions | |
Tom | +0.014 |
booth | +0.014 |
cheating | +0.013 |
money | +0.013 |
workers | +0.013 |
Neg logit contributions | |
stub | -0.027 |
berry | -0.026 |
burn | -0.026 |
Chick | -0.026 |
light | -0.026 |
Token | firefighters |
Token ID | 22611 |
Feature activation | +3.43e+00 |
Pos logit contributions | |
booth | +0.010 |
Tom | +0.010 |
Tomas | +0.010 |
employee | +0.010 |
cheating | +0.010 |
Neg logit contributions | |
stub | -0.013 |
Chick | -0.012 |
burn | -0.012 |
berry | -0.012 |
light | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
booth | +0.130 |
Tom | +0.126 |
cheating | +0.123 |
employee | +0.119 |
tickets | +0.119 |
Neg logit contributions | |
stub | -0.207 |
Chick | -0.204 |
burn | -0.199 |
berry | -0.197 |
Grac | -0.194 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
Tom | +0.010 |
booth | +0.009 |
Tomas | +0.009 |
Officer | +0.008 |
cheating | +0.008 |
Neg logit contributions | |
stub | -0.016 |
burn | -0.016 |
light | -0.016 |
berry | -0.016 |
streamed | -0.015 |
Token | said |
Token ID | 531 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
stub | +0.011 |
Chick | +0.010 |
burn | +0.010 |
berry | +0.010 |
light | +0.010 |
Neg logit contributions | |
booth | -0.011 |
Tom | -0.010 |
cheating | -0.010 |
employee | -0.010 |
Tomas | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
Tom | +0.086 |
cheating | +0.070 |
money | +0.070 |
booth | +0.069 |
tickets | +0.066 |
Neg logit contributions | |
stub | -0.149 |
light | -0.149 |
burn | -0.146 |
ifying | -0.144 |
berry | -0.143 |
Token | " |
Token ID | 366 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
stub | +0.033 |
burn | +0.032 |
light | +0.032 |
Chick | +0.032 |
berry | +0.032 |
Neg logit contributions | |
Tom | -0.013 |
booth | -0.013 |
cheating | -0.011 |
worker | -0.011 |
Tomas | -0.011 |
Token | Wow |
Token ID | 22017 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
Tom | +0.035 |
booth | +0.035 |
Tomas | +0.033 |
employee | +0.033 |
cheating | +0.033 |
Neg logit contributions | |
stub | -0.054 |
Chick | -0.053 |
berry | -0.052 |
burn | -0.052 |
Grac | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | -4.05e-01 |
Pos logit contributions | |
Tom | +0.134 |
cheating | +0.118 |
booth | +0.116 |
money | +0.108 |
tickets | +0.108 |
Neg logit contributions | |
stub | -0.280 |
Chick | -0.269 |
burn | -0.266 |
light | -0.261 |
berry | -0.260 |
Token | that |
Token ID | 326 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
stub | +0.026 |
Chick | +0.025 |
burn | +0.025 |
berry | +0.025 |
light | +0.024 |
Neg logit contributions | |
booth | -0.014 |
Tom | -0.014 |
cheating | -0.014 |
employee | -0.013 |
Tomas | -0.013 |
Token | Many |
Token ID | 4650 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
booth | +0.013 |
cheating | +0.013 |
employee | +0.013 |
tickets | +0.013 |
worker | +0.012 |
Neg logit contributions | |
Chick | -0.013 |
stub | -0.013 |
berry | -0.013 |
light | -0.012 |
Grac | -0.012 |
Token | animals |
Token ID | 4695 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
booth | +0.079 |
employee | +0.071 |
Tomas | +0.071 |
cheating | +0.067 |
inconvenience | +0.066 |
Neg logit contributions | |
Chick | -0.110 |
berry | -0.110 |
stub | -0.109 |
ie | -0.105 |
streamed | -0.103 |
Token | lived |
Token ID | 5615 |
Feature activation | -7.94e-01 |
Pos logit contributions | |
booth | +0.110 |
Tom | +0.107 |
cheating | +0.106 |
employee | +0.104 |
Tomas | +0.104 |
Neg logit contributions | |
stub | -0.097 |
Chick | -0.094 |
burn | -0.092 |
berry | -0.091 |
light | -0.090 |
Token | near |
Token ID | 1474 |
Feature activation | +5.65e-01 |
Pos logit contributions | |
stub | +0.037 |
Chick | +0.036 |
berry | +0.035 |
Grac | +0.034 |
light | +0.034 |
Neg logit contributions | |
booth | -0.040 |
inconvenience | -0.038 |
bills | -0.038 |
cheating | -0.038 |
Tomas | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
employee | +0.026 |
booth | +0.026 |
cheating | +0.025 |
inconvenience | +0.025 |
nine | +0.024 |
Neg logit contributions | |
Chick | -0.030 |
berry | -0.027 |
light | -0.026 |
c | -0.026 |
Grac | -0.026 |
Token | path |
Token ID | 3108 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
cheating | +0.030 |
nine | +0.029 |
employee | +0.028 |
inconvenience | +0.028 |
Tomas | +0.028 |
Neg logit contributions | |
Chick | -0.025 |
berry | -0.022 |
Po | -0.022 |
bur | -0.021 |
Grac | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +2.85e-01 |
Pos logit contributions | |
cheating | +0.081 |
booth | +0.081 |
Tomas | +0.081 |
employee | +0.081 |
Officer | +0.076 |
Neg logit contributions | |
stub | -0.098 |
light | -0.094 |
berry | -0.094 |
Chick | -0.093 |
streamed | -0.090 |
Token | One |
Token ID | 1881 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
cheating | +0.016 |
employee | +0.016 |
booth | +0.015 |
worker | +0.015 |
tickets | +0.015 |
Neg logit contributions | |
Chick | -0.012 |
berry | -0.012 |
stub | -0.011 |
Grac | -0.011 |
light | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
booth | +0.059 |
cheating | +0.057 |
Tom | +0.057 |
employee | +0.056 |
Tomas | +0.055 |
Neg logit contributions | |
stub | -0.071 |
Chick | -0.070 |
burn | -0.068 |
berry | -0.068 |
light | -0.067 |
Token | , |
Token ID | 11 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
stub | +0.068 |
Chick | +0.066 |
berry | +0.065 |
streamed | +0.063 |
burn | +0.063 |
Neg logit contributions | |
booth | -0.057 |
employee | -0.054 |
cheating | -0.054 |
Tomas | -0.052 |
site | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | +4.87e-01 |
Pos logit contributions | |
Chick | +0.028 |
stub | +0.028 |
berry | +0.027 |
Grac | +0.027 |
light | +0.027 |
Neg logit contributions | |
booth | -0.014 |
cheating | -0.013 |
employee | -0.013 |
Tomas | -0.012 |
site | -0.012 |
Token | there |
Token ID | 612 |
Feature activation | -6.49e-01 |
Pos logit contributions | |
stub | +0.023 |
Chick | +0.023 |
burn | +0.023 |
berry | +0.022 |
light | +0.022 |
Neg logit contributions | |
Tom | -0.017 |
booth | -0.016 |
cheating | -0.016 |
Tomas | -0.015 |
worker | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -5.09e-01 |
Pos logit contributions | |
burn | +0.042 |
Chick | +0.041 |
stub | +0.041 |
numb | +0.040 |
light | +0.040 |
Neg logit contributions | |
Tom | -0.025 |
booth | -0.021 |
cheating | -0.020 |
Tomas | -0.020 |
site | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | +4.90e-01 |
Pos logit contributions | |
Chick | +0.034 |
stub | +0.034 |
berry | +0.033 |
burn | +0.032 |
Grac | +0.032 |
Neg logit contributions | |
booth | -0.017 |
cheating | -0.016 |
employee | -0.016 |
site | -0.015 |
worker | -0.015 |
Token | big |
Token ID | 1263 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
booth | +0.023 |
cheating | +0.022 |
Tom | +0.022 |
tickets | +0.022 |
Tomas | +0.022 |
Neg logit contributions | |
stub | -0.025 |
Chick | -0.025 |
berry | -0.024 |
burn | -0.024 |
Grac | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -4.72e-01 |
Pos logit contributions | |
stub | +0.076 |
berry | +0.075 |
Chick | +0.075 |
burn | +0.073 |
skirt | +0.071 |
Neg logit contributions | |
booth | -0.070 |
cheating | -0.067 |
Tomas | -0.067 |
employee | -0.067 |
Tom | -0.066 |
Token | blue |
Token ID | 4171 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
stub | +0.023 |
Chick | +0.023 |
Grac | +0.021 |
skirt | +0.021 |
berry | +0.021 |
Neg logit contributions | |
booth | -0.024 |
tickets | -0.023 |
Tomas | -0.023 |
cheating | -0.023 |
inconvenience | -0.023 |
Token | sea |
Token ID | 5417 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
stub | +0.113 |
Chick | +0.112 |
Grac | +0.109 |
streamed | +0.108 |
light | +0.106 |
Neg logit contributions | |
booth | -0.087 |
Tom | -0.082 |
worker | -0.080 |
tickets | -0.077 |
money | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
booth | +0.036 |
Tom | +0.035 |
worker | +0.034 |
site | +0.032 |
employee | +0.031 |
Neg logit contributions | |
stub | -0.076 |
burn | -0.076 |
Chick | -0.076 |
Grac | -0.074 |
streamed | -0.073 |
Token | In |
Token ID | 554 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
Tom | +0.015 |
booth | +0.015 |
cheating | +0.014 |
Tomas | +0.014 |
employee | +0.014 |
Neg logit contributions | |
stub | -0.020 |
Chick | -0.019 |
burn | -0.019 |
berry | -0.018 |
light | -0.018 |
Token | this |
Token ID | 428 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
Tom | +0.081 |
booth | +0.077 |
site | +0.067 |
money | +0.066 |
tickets | +0.066 |
Neg logit contributions | |
stub | -0.231 |
burn | -0.225 |
Chick | -0.223 |
light | -0.220 |
lull | -0.218 |
Token | sea |
Token ID | 5417 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
booth | +0.007 |
Tom | +0.007 |
cheating | +0.007 |
tickets | +0.007 |
site | +0.007 |
Neg logit contributions | |
stub | -0.008 |
Chick | -0.008 |
burn | -0.008 |
berry | -0.008 |
Grac | -0.008 |
Token | that |
Token ID | 326 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
stub | +0.022 |
Chick | +0.021 |
burn | +0.021 |
berry | +0.021 |
streamed | +0.020 |
Neg logit contributions | |
booth | -0.013 |
cheating | -0.012 |
tickets | -0.012 |
employee | -0.012 |
Tomas | -0.012 |
Token | day |
Token ID | 1110 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
stub | +0.027 |
Chick | +0.026 |
burn | +0.026 |
berry | +0.025 |
light | +0.025 |
Neg logit contributions | |
booth | -0.022 |
cheating | -0.022 |
Tom | -0.021 |
employee | -0.021 |
tickets | -0.021 |
Token | , |
Token ID | 11 |
Feature activation | -6.09e-01 |
Pos logit contributions | |
stub | +0.082 |
Chick | +0.081 |
light | +0.080 |
burn | +0.080 |
berry | +0.079 |
Neg logit contributions | |
Tom | -0.044 |
booth | -0.042 |
money | -0.039 |
worker | -0.039 |
cheating | -0.038 |
Token | after |
Token ID | 706 |
Feature activation | +2.92e-01 |
Pos logit contributions | |
burn | +0.043 |
light | +0.042 |
stub | +0.041 |
ubb | +0.041 |
berry | +0.041 |
Neg logit contributions | |
Tom | -0.022 |
money | -0.016 |
booth | -0.016 |
worker | -0.016 |
workers | -0.015 |
Token | taking |
Token ID | 2263 |
Feature activation | -5.00e-02 |
Pos logit contributions | |
booth | +0.013 |
employee | +0.012 |
worker | +0.011 |
site | +0.011 |
Tomas | +0.011 |
Neg logit contributions | |
Chick | -0.017 |
stub | -0.017 |
Grac | -0.016 |
burn | -0.016 |
berry | -0.015 |
Token | a |
Token ID | 257 |
Feature activation | +4.72e-01 |
Pos logit contributions | |
Chick | +0.003 |
stub | +0.003 |
burn | +0.003 |
Grac | +0.003 |
berry | +0.003 |
Neg logit contributions | |
booth | -0.002 |
cheating | -0.002 |
employee | -0.001 |
site | -0.001 |
worker | -0.001 |
Token | bath |
Token ID | 7837 |
Feature activation | -4.99e+00 |
Pos logit contributions | |
booth | +0.022 |
Tom | +0.022 |
cheating | +0.022 |
employee | +0.021 |
Tomas | +0.021 |
Neg logit contributions | |
stub | -0.024 |
Chick | -0.023 |
burn | -0.023 |
berry | -0.023 |
light | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
Chick | +0.361 |
stub | +0.360 |
ifying | +0.351 |
light | +0.348 |
maid | +0.346 |
Neg logit contributions | |
worker | -0.141 |
Tom | -0.140 |
employee | -0.122 |
booth | -0.121 |
workers | -0.120 |
Token | Lily |
Token ID | 20037 |
Feature activation | -5.52e-02 |
Pos logit contributions | |
burn | +0.040 |
stub | +0.039 |
light | +0.039 |
berry | +0.038 |
beam | +0.037 |
Neg logit contributions | |
Tom | -0.023 |
booth | -0.018 |
worker | -0.017 |
money | -0.017 |
cheating | -0.017 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
ubb | +0.003 |
Grac | +0.003 |
Chick | +0.003 |
berry | +0.003 |
burn | +0.003 |
Neg logit contributions | |
Tom | -0.002 |
booth | -0.002 |
cheating | -0.002 |
her | -0.002 |
Tomas | -0.002 |
Token | her |
Token ID | 607 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
stub | +0.001 |
Chick | +0.001 |
berry | +0.001 |
burn | +0.001 |
light | +0.001 |
Neg logit contributions | |
Tom | -0.000 |
booth | -0.000 |
cheating | -0.000 |
Tomas | -0.000 |
tickets | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | +2.73e-01 |
Pos logit contributions | |
stub | +0.084 |
Chick | +0.082 |
burn | +0.081 |
berry | +0.079 |
oppy | +0.079 |
Neg logit contributions | |
Tom | -0.036 |
cheating | -0.033 |
booth | -0.033 |
worker | -0.031 |
employee | -0.031 |
Token | From |
Token ID | 3574 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
Tom | +0.010 |
booth | +0.009 |
Tomas | +0.008 |
Officer | +0.008 |
her | +0.008 |
Neg logit contributions | |
burn | -0.018 |
stub | -0.017 |
numb | -0.017 |
light | -0.017 |
berry | -0.017 |
Token | then |
Token ID | 788 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
inconvenience | +0.057 |
booth | +0.056 |
cheating | +0.056 |
tickets | +0.055 |
site | +0.055 |
Neg logit contributions | |
stub | -0.082 |
Chick | -0.082 |
ie | -0.081 |
skirt | -0.080 |
burn | -0.080 |
Token | on |
Token ID | 319 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
Tom | +0.074 |
booth | +0.073 |
cheating | +0.071 |
Tomas | +0.068 |
money | +0.067 |
Neg logit contributions | |
stub | -0.118 |
Chick | -0.115 |
berry | -0.113 |
light | -0.113 |
maid | -0.112 |
Token | , |
Token ID | 11 |
Feature activation | -5.33e-01 |
Pos logit contributions | |
stub | +0.037 |
burn | +0.037 |
numb | +0.036 |
light | +0.036 |
Chick | +0.036 |
Neg logit contributions | |
Tom | -0.017 |
booth | -0.015 |
cheating | -0.015 |
money | -0.015 |
Tomas | -0.014 |
Token | she |
Token ID | 673 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
light | +0.036 |
maid | +0.036 |
stub | +0.036 |
burn | +0.035 |
numb | +0.035 |
Neg logit contributions | |
Tom | -0.021 |
money | -0.016 |
cheating | -0.015 |
worker | -0.015 |
workers | -0.014 |
Token | always |
Token ID | 1464 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
stub | +0.111 |
Chick | +0.110 |
berry | +0.108 |
burn | +0.108 |
numb | +0.107 |
Neg logit contributions | |
Tom | -0.064 |
booth | -0.062 |
cheating | -0.062 |
Tomas | -0.059 |
worker | -0.059 |
Token | made |
Token ID | 925 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
Tom | +0.007 |
cheating | +0.007 |
booth | +0.007 |
tickets | +0.006 |
money | +0.006 |
Neg logit contributions | |
stub | -0.008 |
Chick | -0.008 |
berry | -0.008 |
burn | -0.007 |
Grac | -0.007 |
Token | sure |
Token ID | 1654 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
stub | +0.182 |
Chick | +0.178 |
berry | +0.176 |
burn | +0.175 |
light | +0.175 |
Neg logit contributions | |
booth | -0.072 |
Tom | -0.071 |
cheating | -0.068 |
employee | -0.066 |
tickets | -0.065 |
Token | to |
Token ID | 284 |
Feature activation | -8.11e-01 |
Pos logit contributions | |
Tom | +0.036 |
booth | +0.035 |
money | +0.034 |
cheating | +0.034 |
tickets | +0.033 |
Neg logit contributions | |
Chick | -0.063 |
berry | -0.062 |
stub | -0.062 |
light | -0.062 |
burn | -0.061 |
Token | bring |
Token ID | 2222 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
stub | +0.045 |
Chick | +0.044 |
berry | +0.043 |
burn | +0.043 |
light | +0.042 |
Neg logit contributions | |
booth | -0.035 |
Tom | -0.034 |
cheating | -0.033 |
employee | -0.032 |
Tomas | -0.032 |
Token | no |
Token ID | 645 |
Feature activation | -4.73e+00 |
Pos logit contributions | |
stub | +0.023 |
Chick | +0.023 |
burn | +0.023 |
berry | +0.023 |
streamed | +0.022 |
Neg logit contributions | |
Tom | -0.012 |
booth | -0.012 |
cheating | -0.012 |
Tomas | -0.011 |
money | -0.011 |
Token | matter |
Token ID | 2300 |
Feature activation | +4.72e+00 |
Pos logit contributions | |
stub | +0.289 |
Chick | +0.282 |
burn | +0.278 |
berry | +0.276 |
light | +0.275 |
Neg logit contributions | |
booth | -0.179 |
Tom | -0.171 |
cheating | -0.166 |
employee | -0.165 |
Tomas | -0.163 |
Token | how |
Token ID | 703 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
booth | +0.150 |
Tom | +0.147 |
cheating | +0.143 |
employee | +0.136 |
Tomas | +0.136 |
Neg logit contributions | |
stub | -0.315 |
Chick | -0.307 |
burn | -0.304 |
berry | -0.301 |
light | -0.297 |
Token | scared |
Token ID | 12008 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
burn | +0.066 |
stub | +0.066 |
berry | +0.065 |
Chick | +0.065 |
light | +0.063 |
Neg logit contributions | |
Tom | -0.032 |
booth | -0.032 |
cheating | -0.031 |
money | -0.030 |
tickets | -0.029 |
Token | he |
Token ID | 339 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
stub | +0.166 |
burn | +0.166 |
Chick | +0.160 |
light | +0.158 |
berry | +0.157 |
Neg logit contributions | |
Tom | -0.088 |
cheating | -0.078 |
money | -0.077 |
booth | -0.075 |
Tomas | -0.074 |
Token | was |
Token ID | 373 |
Feature activation | -4.42e-01 |
Pos logit contributions | |
stub | +0.081 |
Chick | +0.077 |
light | +0.077 |
berry | +0.076 |
streamed | +0.075 |
Neg logit contributions | |
booth | -0.078 |
employee | -0.074 |
Tom | -0.074 |
worker | -0.073 |
Tomas | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
stub | +0.025 |
Chick | +0.024 |
light | +0.024 |
berry | +0.023 |
burn | +0.023 |
Neg logit contributions | |
booth | -0.018 |
Tom | -0.017 |
employee | -0.017 |
cheating | -0.017 |
Tomas | -0.017 |
Token | By |
Token ID | 2750 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
booth | +0.016 |
Tom | +0.016 |
cheating | +0.015 |
Tomas | +0.015 |
employee | +0.015 |
Neg logit contributions | |
stub | -0.025 |
Chick | -0.024 |
burn | -0.024 |
berry | -0.024 |
light | -0.023 |
Token | doing |
Token ID | 1804 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
booth | +0.163 |
employee | +0.152 |
Tomas | +0.145 |
worker | +0.145 |
tickets | +0.143 |
Neg logit contributions | |
Chick | -0.223 |
stub | -0.221 |
Grac | -0.220 |
light | -0.216 |
berry | -0.213 |
Token | so |
Token ID | 523 |
Feature activation | -7.67e-01 |
Pos logit contributions | |
booth | +0.062 |
Tom | +0.061 |
cheating | +0.060 |
Tomas | +0.058 |
tickets | +0.058 |
Neg logit contributions | |
stub | -0.095 |
Chick | -0.092 |
burn | -0.091 |
berry | -0.091 |
light | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | -3.01e-01 |
Pos logit contributions | |
stub | +0.047 |
Chick | +0.046 |
burn | +0.045 |
berry | +0.045 |
light | +0.045 |
Neg logit contributions | |
booth | -0.028 |
Tom | -0.027 |
cheating | -0.027 |
employee | -0.026 |
Tomas | -0.026 |
Token | enthusiastic |
Token ID | 21848 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
Chick | +0.019 |
stub | +0.018 |
berry | +0.018 |
skirt | +0.018 |
ie | +0.018 |
Neg logit contributions | |
booth | -0.014 |
cheating | -0.013 |
Tomas | -0.013 |
Tom | -0.013 |
tickets | -0.013 |
Token | little |
Token ID | 1310 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
berry | +0.111 |
stub | +0.111 |
Chick | +0.108 |
burn | +0.106 |
light | +0.105 |
Neg logit contributions | |
booth | -0.095 |
cheating | -0.093 |
Tomas | -0.091 |
tickets | -0.090 |
site | -0.089 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
stub | +0.073 |
Chick | +0.072 |
burn | +0.071 |
berry | +0.070 |
light | +0.069 |
Neg logit contributions | |
booth | -0.040 |
Tom | -0.039 |
cheating | -0.038 |
employee | -0.037 |
Tomas | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
booth | +0.068 |
tickets | +0.065 |
cheating | +0.064 |
employee | +0.064 |
Tomas | +0.063 |
Neg logit contributions | |
stub | -0.084 |
Chick | -0.084 |
berry | -0.083 |
light | -0.083 |
burn | -0.079 |
Token | He |
Token ID | 679 |
Feature activation | +2.35e-01 |
Pos logit contributions | |
booth | +0.030 |
tickets | +0.028 |
cheating | +0.028 |
employee | +0.028 |
Tomas | +0.027 |
Neg logit contributions | |
stub | -0.029 |
berry | -0.029 |
Chick | -0.029 |
Grac | -0.028 |
light | -0.027 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.62e+00 |
Pos logit contributions | |
booth | +0.010 |
Tom | +0.010 |
cheating | +0.010 |
employee | +0.010 |
Tomas | +0.010 |
Neg logit contributions | |
stub | -0.013 |
Chick | -0.012 |
burn | -0.012 |
berry | -0.012 |
light | -0.012 |
Token | to |
Token ID | 284 |
Feature activation | -5.82e-01 |
Pos logit contributions | |
stub | +0.102 |
Chick | +0.099 |
light | +0.099 |
berry | +0.099 |
burn | +0.098 |
Neg logit contributions | |
Tom | -0.058 |
booth | -0.055 |
cheating | -0.054 |
tickets | -0.054 |
numbers | -0.053 |
Token | swing |
Token ID | 9628 |
Feature activation | -4.55e-01 |
Pos logit contributions | |
stub | +0.034 |
maid | +0.033 |
Chick | +0.033 |
burn | +0.033 |
ubb | +0.033 |
Neg logit contributions | |
Tom | -0.024 |
numbers | -0.023 |
money | -0.022 |
equipment | -0.022 |
booth | -0.022 |
Token | on |
Token ID | 319 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
stub | +0.027 |
Chick | +0.026 |
burn | +0.026 |
berry | +0.025 |
light | +0.025 |
Neg logit contributions | |
booth | -0.018 |
Tom | -0.018 |
cheating | -0.017 |
employee | -0.017 |
Tomas | -0.017 |
Token | his |
Token ID | 465 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
stub | +0.024 |
Chick | +0.023 |
burn | +0.023 |
berry | +0.023 |
light | +0.023 |
Neg logit contributions | |
booth | -0.013 |
Tom | -0.012 |
cheating | -0.012 |
employee | -0.012 |
tickets | -0.012 |
Token | swing |
Token ID | 9628 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
Chick | +0.077 |
stub | +0.076 |
Grac | +0.074 |
burn | +0.074 |
berry | +0.074 |
Neg logit contributions | |
booth | -0.047 |
Tom | -0.046 |
tickets | -0.043 |
money | -0.042 |
cheating | -0.042 |
Token | is |
Token ID | 318 |
Feature activation | +7.65e-01 |
Pos logit contributions | |
booth | +0.090 |
Tom | +0.086 |
cheating | +0.085 |
employee | +0.085 |
Tomas | +0.084 |
Neg logit contributions | |
stub | -0.160 |
Chick | -0.158 |
berry | -0.157 |
light | -0.156 |
burn | -0.156 |
Token | when |
Token ID | 618 |
Feature activation | -6.35e-01 |
Pos logit contributions | |
booth | +0.031 |
Tom | +0.029 |
employee | +0.029 |
cheating | +0.029 |
Tomas | +0.028 |
Neg logit contributions | |
stub | -0.044 |
Chick | -0.044 |
burn | -0.043 |
light | -0.042 |
berry | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
stub | +0.041 |
burn | +0.039 |
Chick | +0.039 |
berry | +0.039 |
Grac | +0.038 |
Neg logit contributions | |
Tom | -0.023 |
booth | -0.022 |
cheating | -0.021 |
tickets | -0.021 |
money | -0.020 |
Token | sun |
Token ID | 4252 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
booth | +0.020 |
Tom | +0.019 |
cheating | +0.019 |
Tomas | +0.019 |
employee | +0.019 |
Neg logit contributions | |
stub | -0.021 |
Chick | -0.020 |
burn | -0.020 |
berry | -0.020 |
light | -0.020 |
Token | says |
Token ID | 1139 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
Tom | +0.002 |
booth | +0.002 |
tickets | +0.002 |
cheating | +0.002 |
money | +0.002 |
Neg logit contributions | |
stub | -0.004 |
Chick | -0.004 |
Grac | -0.004 |
skirt | -0.004 |
burn | -0.004 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -5.28e+00 |
Pos logit contributions | |
Tom | +0.108 |
booth | +0.106 |
tickets | +0.101 |
numbers | +0.098 |
money | +0.095 |
Neg logit contributions | |
stub | -0.203 |
Chick | -0.193 |
burn | -0.193 |
oppy | -0.187 |
berry | -0.186 |
Token | to |
Token ID | 284 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
stub | +0.318 |
Chick | +0.313 |
burn | +0.307 |
berry | +0.303 |
light | +0.303 |
Neg logit contributions | |
booth | -0.200 |
cheating | -0.192 |
Tom | -0.192 |
employee | -0.186 |
Tomas | -0.186 |
Token | the |
Token ID | 262 |
Feature activation | +3.94e-01 |
Pos logit contributions | |
stub | +0.049 |
burn | +0.047 |
Chick | +0.047 |
berry | +0.046 |
light | +0.046 |
Neg logit contributions | |
Tom | -0.028 |
booth | -0.027 |
cheating | -0.025 |
tickets | -0.025 |
Tomas | -0.025 |
Token | day |
Token ID | 1110 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
booth | +0.018 |
Tomas | +0.018 |
cheating | +0.018 |
employee | +0.017 |
Tom | +0.017 |
Neg logit contributions | |
stub | -0.020 |
berry | -0.020 |
burn | -0.019 |
Chick | -0.019 |
skirt | -0.019 |
Token | and |
Token ID | 290 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
stub | +0.065 |
Chick | +0.064 |
light | +0.064 |
berry | +0.063 |
burn | +0.063 |
Neg logit contributions | |
booth | -0.046 |
cheating | -0.046 |
employee | -0.045 |
Tomas | -0.045 |
Tom | -0.044 |
Token | hello |
Token ID | 23748 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
Tom | +0.013 |
booth | +0.012 |
tickets | +0.012 |
money | +0.012 |
cheating | +0.012 |
Neg logit contributions | |
stub | -0.021 |
Chick | -0.020 |
berry | -0.020 |
burn | -0.020 |
Grac | -0.019 |
Token | She |
Token ID | 1375 |
Feature activation | -5.94e-01 |
Pos logit contributions | |
booth | +0.013 |
Tom | +0.013 |
cheating | +0.012 |
employee | +0.012 |
tickets | +0.012 |
Neg logit contributions | |
stub | -0.012 |
Chick | -0.011 |
burn | -0.011 |
berry | -0.011 |
light | -0.011 |
Token | went |
Token ID | 1816 |
Feature activation | -8.10e-01 |
Pos logit contributions | |
stub | +0.032 |
Chick | +0.032 |
burn | +0.031 |
berry | +0.031 |
light | +0.030 |
Neg logit contributions | |
booth | -0.026 |
Tom | -0.025 |
cheating | -0.025 |
employee | -0.024 |
tickets | -0.024 |
Token | back |
Token ID | 736 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
stub | +0.039 |
burn | +0.037 |
Chick | +0.037 |
berry | +0.037 |
Grac | +0.037 |
Neg logit contributions | |
cheating | -0.040 |
Tom | -0.038 |
inconvenience | -0.038 |
booth | -0.038 |
Tomas | -0.037 |
Token | to |
Token ID | 284 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
booth | +0.009 |
cheating | +0.009 |
employee | +0.009 |
Tom | +0.009 |
Tomas | +0.009 |
Neg logit contributions | |
stub | -0.011 |
Chick | -0.011 |
berry | -0.011 |
burn | -0.010 |
Grac | -0.010 |
Token | her |
Token ID | 607 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
stub | +0.046 |
Chick | +0.045 |
burn | +0.045 |
berry | +0.044 |
light | +0.044 |
Neg logit contributions | |
booth | -0.033 |
Tom | -0.033 |
cheating | -0.032 |
employee | -0.031 |
Tomas | -0.031 |
Token | c |
Token ID | 269 |
Feature activation | -5.27e+00 |
Pos logit contributions | |
stub | +0.094 |
Chick | +0.091 |
burn | +0.090 |
berry | +0.089 |
light | +0.088 |
Neg logit contributions | |
booth | -0.085 |
Tom | -0.082 |
cheating | -0.080 |
employee | -0.079 |
tickets | -0.079 |
Token | auldron |
Token ID | 45637 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
stub | +0.234 |
Chick | +0.226 |
Grac | +0.224 |
burn | +0.219 |
streamed | +0.217 |
Neg logit contributions | |
booth | -0.288 |
her | -0.277 |
site | -0.273 |
Tom | -0.273 |
worker | -0.272 |
Token | and |
Token ID | 290 |
Feature activation | +4.84e-01 |
Pos logit contributions | |
Chick | +0.156 |
stub | +0.156 |
burn | +0.154 |
streamed | +0.150 |
berry | +0.149 |
Neg logit contributions | |
booth | -0.043 |
Tom | -0.037 |
money | -0.036 |
worker | -0.034 |
site | -0.033 |
Token | started |
Token ID | 2067 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
Tom | +0.021 |
booth | +0.020 |
money | +0.019 |
worker | +0.019 |
cheating | +0.019 |
Neg logit contributions | |
burn | -0.027 |
stub | -0.027 |
berry | -0.026 |
Chick | -0.026 |
numb | -0.025 |
Token | mixing |
Token ID | 17090 |
Feature activation | +4.13e-01 |
Pos logit contributions | |
Tom | +0.051 |
money | +0.051 |
numbers | +0.048 |
worker | +0.046 |
booth | +0.046 |
Neg logit contributions | |
stub | -0.131 |
burn | -0.130 |
maid | -0.128 |
ifying | -0.127 |
streamed | -0.126 |
Token | the |
Token ID | 262 |
Feature activation | +4.00e-01 |
Pos logit contributions | |
booth | +0.016 |
cheating | +0.015 |
employee | +0.015 |
Tom | +0.015 |
Tomas | +0.015 |
Neg logit contributions | |
Chick | -0.024 |
stub | -0.024 |
berry | -0.024 |
Grac | -0.023 |
light | -0.023 |
Token | , |
Token ID | 11 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
booth | +0.061 |
Tom | +0.059 |
cheating | +0.057 |
employee | +0.056 |
Tomas | +0.055 |
Neg logit contributions | |
stub | -0.107 |
Chick | -0.105 |
burn | -0.104 |
berry | -0.103 |
light | -0.102 |
Token | when |
Token ID | 618 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
stub | +0.012 |
Chick | +0.011 |
light | +0.011 |
burn | +0.011 |
Grac | +0.011 |
Neg logit contributions | |
booth | -0.007 |
tickets | -0.007 |
cheating | -0.007 |
employee | -0.007 |
worker | -0.007 |
Token | he |
Token ID | 339 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
stub | +0.027 |
Chick | +0.027 |
burn | +0.026 |
berry | +0.026 |
light | +0.026 |
Neg logit contributions | |
booth | -0.020 |
Tom | -0.020 |
cheating | -0.019 |
Tomas | -0.018 |
employee | -0.018 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
Chick | +0.086 |
burn | +0.086 |
stub | +0.084 |
Grac | +0.082 |
berry | +0.082 |
Neg logit contributions | |
booth | -0.062 |
Tom | -0.062 |
cheating | -0.060 |
Tomas | -0.057 |
employee | -0.056 |
Token | a |
Token ID | 257 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
booth | +0.001 |
Tom | +0.001 |
cheating | +0.001 |
employee | +0.001 |
Tomas | +0.001 |
Neg logit contributions | |
stub | -0.002 |
Chick | -0.002 |
burn | -0.002 |
berry | -0.002 |
light | -0.002 |
Token | bee |
Token ID | 20697 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
Tom | +0.025 |
booth | +0.025 |
Tomas | +0.025 |
tickets | +0.024 |
employee | +0.024 |
Neg logit contributions | |
stub | -0.029 |
berry | -0.027 |
Chick | -0.027 |
skirt | -0.026 |
burn | -0.026 |
Token | come |
Token ID | 1282 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
Chick | +0.303 |
Grac | +0.301 |
stub | +0.299 |
burn | +0.297 |
maid | +0.288 |
Neg logit contributions | |
booth | -0.160 |
Tom | -0.157 |
cheating | -0.156 |
site | -0.142 |
worker | -0.142 |
Token | flying |
Token ID | 7348 |
Feature activation | -8.26e-02 |
Pos logit contributions | |
stub | +0.014 |
Chick | +0.013 |
streamed | +0.013 |
Grac | +0.013 |
lull | +0.013 |
Neg logit contributions | |
Tom | -0.006 |
booth | -0.005 |
cheating | -0.005 |
worker | -0.005 |
her | -0.005 |
Token | by |
Token ID | 416 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
stub | +0.006 |
Chick | +0.005 |
burn | +0.005 |
berry | +0.005 |
Grac | +0.005 |
Neg logit contributions | |
Tom | -0.003 |
booth | -0.003 |
cheating | -0.002 |
worker | -0.002 |
her | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
booth | +0.129 |
employee | +0.121 |
Tom | +0.120 |
cheating | +0.119 |
tickets | +0.118 |
Neg logit contributions | |
stub | -0.183 |
Chick | -0.181 |
berry | -0.177 |
light | -0.175 |
burn | -0.175 |
Token | He |
Token ID | 679 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
booth | +0.019 |
cheating | +0.019 |
tickets | +0.018 |
employee | +0.018 |
worker | +0.017 |
Neg logit contributions | |
Chick | -0.020 |
stub | -0.019 |
berry | -0.019 |
Grac | -0.018 |
light | -0.018 |
Token | ! |
Token ID | 0 |
Feature activation | -1.51e-02 |
Pos logit contributions | |
booth | +0.017 |
Tom | +0.016 |
cheating | +0.016 |
Tomas | +0.016 |
employee | +0.016 |
Neg logit contributions | |
stub | -0.024 |
Chick | -0.024 |
berry | -0.024 |
burn | -0.024 |
light | -0.024 |
Token | She |
Token ID | 1375 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
burn | +0.001 |
stub | +0.001 |
light | +0.001 |
maid | +0.001 |
lull | +0.001 |
Neg logit contributions | |
Tom | -0.001 |
booth | -0.001 |
worker | -0.001 |
employee | -0.001 |
Tomas | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
burn | +0.025 |
stub | +0.024 |
light | +0.023 |
berry | +0.023 |
numb | +0.023 |
Neg logit contributions | |
Tom | -0.016 |
booth | -0.015 |
cheating | -0.014 |
worker | -0.014 |
numbers | -0.014 |
Token | that |
Token ID | 326 |
Feature activation | -4.22e-01 |
Pos logit contributions | |
Tom | +0.003 |
booth | +0.002 |
money | +0.002 |
numbers | +0.002 |
tickets | +0.002 |
Neg logit contributions | |
stub | -0.008 |
burn | -0.008 |
light | -0.008 |
berry | -0.007 |
days | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | +4.94e-01 |
Pos logit contributions | |
stub | +0.032 |
burn | +0.031 |
streamed | +0.030 |
maid | +0.030 |
berry | +0.030 |
Neg logit contributions | |
Tom | -0.012 |
money | -0.010 |
cheating | -0.009 |
booth | -0.009 |
numbers | -0.009 |
Token | pot |
Token ID | 1787 |
Feature activation | -5.56e+00 |
Pos logit contributions | |
booth | +0.028 |
worker | +0.028 |
money | +0.028 |
workers | +0.028 |
site | +0.028 |
Neg logit contributions | |
stub | -0.020 |
burn | -0.020 |
Chick | -0.018 |
streamed | -0.018 |
berry | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -4.24e-01 |
Pos logit contributions | |
burn | +0.370 |
stub | +0.365 |
Chick | +0.365 |
Grac | +0.347 |
oppy | +0.346 |
Neg logit contributions | |
Tom | -0.182 |
worker | -0.170 |
her | -0.169 |
booth | -0.167 |
workers | -0.165 |
Token | really |
Token ID | 1107 |
Feature activation | +5.83e-01 |
Pos logit contributions | |
stub | +0.027 |
Chick | +0.026 |
burn | +0.026 |
berry | +0.026 |
light | +0.026 |
Neg logit contributions | |
Tom | -0.015 |
booth | -0.015 |
cheating | -0.014 |
worker | -0.014 |
money | -0.014 |
Token | ancient |
Token ID | 6156 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
Tom | +0.023 |
booth | +0.022 |
cheating | +0.022 |
money | +0.021 |
worker | +0.021 |
Neg logit contributions | |
stub | -0.035 |
burn | -0.034 |
light | -0.034 |
Chick | -0.034 |
berry | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
booth | +0.073 |
Tom | +0.070 |
cheating | +0.068 |
employee | +0.067 |
tickets | +0.066 |
Neg logit contributions | |
stub | -0.141 |
Chick | -0.138 |
burn | -0.137 |
berry | -0.135 |
light | -0.134 |
Token | She |
Token ID | 1375 |
Feature activation | -4.18e-01 |
Pos logit contributions | |
Tom | +0.014 |
booth | +0.013 |
Tomas | +0.012 |
cheating | +0.012 |
employee | +0.012 |
Neg logit contributions | |
burn | -0.015 |
stub | -0.015 |
light | -0.014 |
Chick | -0.014 |
berry | -0.014 |
Token | Next |
Token ID | 7406 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
booth | +0.023 |
Tom | +0.023 |
Tomas | +0.021 |
employee | +0.021 |
tickets | +0.021 |
Neg logit contributions | |
stub | -0.033 |
burn | -0.031 |
Chick | -0.031 |
berry | -0.031 |
light | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
booth | +0.049 |
Tom | +0.046 |
cheating | +0.046 |
employee | +0.046 |
Tomas | +0.045 |
Neg logit contributions | |
stub | -0.066 |
Chick | -0.065 |
berry | -0.064 |
light | -0.064 |
burn | -0.063 |
Token | she |
Token ID | 673 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
stub | +0.007 |
burn | +0.007 |
berry | +0.007 |
Chick | +0.007 |
streamed | +0.007 |
Neg logit contributions | |
Tom | -0.006 |
booth | -0.006 |
tickets | -0.006 |
employee | -0.006 |
Tomas | -0.006 |
Token | placed |
Token ID | 4624 |
Feature activation | +2.88e-01 |
Pos logit contributions | |
stub | +0.094 |
Chick | +0.092 |
burn | +0.090 |
berry | +0.090 |
light | +0.089 |
Neg logit contributions | |
booth | -0.077 |
Tom | -0.075 |
cheating | -0.074 |
employee | -0.073 |
Tomas | -0.072 |
Token | her |
Token ID | 607 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
booth | +0.011 |
cheating | +0.011 |
employee | +0.011 |
site | +0.010 |
Tomas | +0.010 |
Neg logit contributions | |
Chick | -0.017 |
stub | -0.016 |
Grac | -0.016 |
burn | -0.016 |
berry | -0.016 |
Token | knee |
Token ID | 10329 |
Feature activation | -4.13e+00 |
Pos logit contributions | |
Chick | +0.086 |
burn | +0.083 |
stub | +0.082 |
berry | +0.080 |
light | +0.079 |
Neg logit contributions | |
money | -0.077 |
booth | -0.076 |
tickets | -0.074 |
numbers | -0.074 |
Tom | -0.073 |
Token | on |
Token ID | 319 |
Feature activation | -4.76e-01 |
Pos logit contributions | |
Chick | +0.246 |
stub | +0.244 |
berry | +0.240 |
Grac | +0.236 |
burn | +0.231 |
Neg logit contributions | |
booth | -0.159 |
Tom | -0.158 |
cheating | -0.146 |
her | -0.146 |
money | -0.146 |
Token | the |
Token ID | 262 |
Feature activation | +6.98e-01 |
Pos logit contributions | |
stub | +0.028 |
Chick | +0.028 |
burn | +0.027 |
berry | +0.027 |
light | +0.027 |
Neg logit contributions | |
booth | -0.018 |
cheating | -0.017 |
Tom | -0.017 |
employee | -0.017 |
Tomas | -0.017 |
Token | highest |
Token ID | 4511 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
booth | +0.030 |
Tom | +0.029 |
cheating | +0.028 |
employee | +0.028 |
tickets | +0.027 |
Neg logit contributions | |
stub | -0.039 |
Chick | -0.038 |
burn | -0.038 |
berry | -0.037 |
light | -0.037 |
Token | point |
Token ID | 966 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
booth | +0.044 |
Tom | +0.043 |
money | +0.040 |
cheating | +0.040 |
worker | +0.040 |
Neg logit contributions | |
Chick | -0.061 |
stub | -0.061 |
burn | -0.060 |
Grac | -0.058 |
berry | -0.058 |
Token | she |
Token ID | 673 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
booth | +0.168 |
Tom | +0.168 |
worker | +0.158 |
tickets | +0.156 |
cheating | +0.155 |
Neg logit contributions | |
stub | -0.130 |
Chick | -0.130 |
burn | -0.121 |
Grac | -0.119 |
numb | -0.118 |