Token | he |
Token ID | 339 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ados | +0.004 |
riages | +0.004 |
barrels | +0.004 |
pm | +0.004 |
drivers | +0.004 |
Neg logit contributions | |
kiss | -0.002 |
copy | -0.002 |
last | -0.002 |
dance | -0.002 |
blow | -0.002 |
Token | said |
Token ID | 531 |
Feature activation | +8.36e-01 |
Pos logit contributions | |
riages | +0.054 |
ados | +0.054 |
barrels | +0.052 |
pm | +0.051 |
drivers | +0.051 |
Neg logit contributions | |
kiss | -0.033 |
copy | -0.031 |
last | -0.028 |
massage | -0.028 |
dance | -0.028 |
Token | it |
Token ID | 340 |
Feature activation | +6.86e-01 |
Pos logit contributions | |
riages | +0.035 |
ados | +0.034 |
barrels | +0.034 |
pm | +0.033 |
drivers | +0.033 |
Neg logit contributions | |
kiss | -0.024 |
copy | -0.022 |
last | -0.020 |
dance | -0.020 |
massage | -0.020 |
Token | was |
Token ID | 373 |
Feature activation | +5.35e-01 |
Pos logit contributions | |
ados | +0.036 |
riages | +0.034 |
barrels | +0.034 |
pm | +0.033 |
drivers | +0.033 |
Neg logit contributions | |
kiss | -0.013 |
copy | -0.011 |
last | -0.010 |
dance | -0.010 |
blow | -0.010 |
Token | only |
Token ID | 691 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
ados | +0.025 |
riages | +0.025 |
pm | +0.024 |
barrels | +0.024 |
drivers | +0.024 |
Neg logit contributions | |
kiss | -0.012 |
copy | -0.011 |
last | -0.010 |
dance | -0.010 |
blow | -0.010 |
Token | $ |
Token ID | 720 |
Feature activation | +8.09e+00 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.006 |
last | +0.006 |
blow | +0.006 |
check | +0.006 |
Neg logit contributions | |
ados | -0.015 |
riages | -0.014 |
pm | -0.014 |
barrels | -0.014 |
drivers | -0.014 |
Token | 10 |
Token ID | 940 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
riages | +0.242 |
ados | +0.242 |
barrels | +0.232 |
bank | +0.230 |
drivers | +0.230 |
Neg logit contributions | |
kiss | -0.303 |
copy | -0.293 |
last | -0.277 |
check | -0.275 |
dab | -0.263 |
Token | . |
Token ID | 13 |
Feature activation | -5.27e-01 |
Pos logit contributions | |
ados | +0.071 |
riages | +0.068 |
barrels | +0.067 |
bank | +0.066 |
drivers | +0.065 |
Neg logit contributions | |
kiss | -0.036 |
last | -0.033 |
copy | -0.032 |
watch | -0.030 |
check | -0.029 |
Token | |
Token ID | 220 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
kiss | +0.017 |
copy | +0.016 |
last | +0.015 |
dance | +0.014 |
massage | +0.014 |
Neg logit contributions | |
ados | -0.020 |
riages | -0.020 |
barrels | -0.019 |
pm | -0.019 |
drivers | -0.019 |
Token |
|
Token ID | 198 |
Feature activation | -8.06e-01 |
Pos logit contributions | |
ados | +0.006 |
riages | +0.006 |
barrels | +0.006 |
pm | +0.006 |
drivers | +0.006 |
Neg logit contributions | |
kiss | -0.002 |
copy | -0.002 |
last | -0.002 |
dance | -0.002 |
blow | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -8.80e-01 |
Pos logit contributions | |
kiss | +0.023 |
copy | +0.023 |
blow | +0.021 |
dance | +0.021 |
last | +0.020 |
Neg logit contributions | |
ados | -0.030 |
ables | -0.029 |
These | -0.029 |
riages | -0.028 |
Those | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -6.80e-01 |
Pos logit contributions | |
kiss | +0.002 |
last | +0.002 |
copy | +0.002 |
happily | +0.001 |
massage | +0.001 |
Neg logit contributions | |
riages | -0.006 |
pm | -0.006 |
AW | -0.006 |
ados | -0.006 |
drivers | -0.006 |
Token | price |
Token ID | 2756 |
Feature activation | +1.68e+00 |
Pos logit contributions | |
kiss | +0.019 |
copy | +0.017 |
last | +0.017 |
massage | +0.016 |
check | +0.016 |
Neg logit contributions | |
ados | -0.029 |
riages | -0.029 |
pm | -0.028 |
ors | -0.028 |
barrels | -0.027 |
Token | - |
Token ID | 532 |
Feature activation | +3.43e-02 |
Pos logit contributions | |
ados | +0.077 |
riages | +0.075 |
pm | +0.074 |
AW | +0.074 |
ors | +0.073 |
Neg logit contributions | |
kiss | -0.039 |
last | -0.038 |
copy | -0.032 |
happily | -0.031 |
mom | -0.030 |
Token | it |
Token ID | 340 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
riages | +0.002 |
ados | +0.002 |
ors | +0.002 |
pm | +0.002 |
drivers | +0.002 |
Neg logit contributions | |
kiss | -0.001 |
copy | -0.001 |
last | -0.001 |
check | -0.001 |
massage | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
ados | +0.007 |
riages | +0.006 |
barrels | +0.006 |
pm | +0.006 |
drivers | +0.006 |
Neg logit contributions | |
kiss | -0.002 |
last | -0.002 |
copy | -0.002 |
blow | -0.002 |
dance | -0.002 |
Token | $ |
Token ID | 720 |
Feature activation | +7.55e+00 |
Pos logit contributions | |
ados | +0.018 |
riages | +0.017 |
pm | +0.017 |
ors | +0.017 |
barrels | +0.016 |
Neg logit contributions | |
kiss | -0.009 |
last | -0.008 |
copy | -0.008 |
check | -0.007 |
blow | -0.007 |
Token | 5 |
Token ID | 20 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
bank | +0.233 |
riages | +0.227 |
ados | +0.226 |
barrels | +0.222 |
pm | +0.217 |
Neg logit contributions | |
kiss | -0.271 |
copy | -0.267 |
last | -0.253 |
check | -0.249 |
folded | -0.237 |
Token | . |
Token ID | 13 |
Feature activation | -5.76e-01 |
Pos logit contributions | |
ados | +0.124 |
bank | +0.119 |
riages | +0.119 |
opes | +0.111 |
barrels | +0.111 |
Neg logit contributions | |
last | -0.078 |
copy | -0.068 |
kiss | -0.067 |
patiently | -0.065 |
folded | -0.064 |
Token | Sally |
Token ID | 25737 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
kiss | +0.016 |
Mom | +0.014 |
check | +0.013 |
Mum | +0.013 |
mum | +0.013 |
Neg logit contributions | |
riages | -0.024 |
drivers | -0.023 |
pm | -0.023 |
ors | -0.023 |
ados | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
ados | +0.055 |
riages | +0.054 |
barrels | +0.053 |
pm | +0.052 |
drivers | +0.052 |
Neg logit contributions | |
kiss | -0.020 |
copy | -0.017 |
last | -0.016 |
cl | -0.016 |
blow | -0.015 |
Token | so |
Token ID | 523 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
kiss | +0.016 |
copy | +0.015 |
last | +0.014 |
dance | +0.014 |
blow | +0.014 |
Neg logit contributions | |
ados | -0.018 |
riages | -0.017 |
barrels | -0.017 |
pm | -0.016 |
drivers | -0.016 |
Token | did |
Token ID | 750 |
Feature activation | +8.69e-01 |
Pos logit contributions | |
riages | +0.055 |
ados | +0.054 |
barrels | +0.053 |
pm | +0.052 |
drivers | +0.051 |
Neg logit contributions | |
kiss | -0.036 |
copy | -0.032 |
dance | -0.030 |
massage | -0.029 |
blow | -0.029 |
Token | not |
Token ID | 407 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ados | +0.035 |
riages | +0.034 |
barrels | +0.033 |
pm | +0.032 |
drivers | +0.032 |
Neg logit contributions | |
kiss | -0.026 |
copy | -0.025 |
last | -0.023 |
dance | -0.022 |
blow | -0.022 |
Token | want |
Token ID | 765 |
Feature activation | -5.24e-02 |
Pos logit contributions | |
ados | +0.073 |
riages | +0.072 |
barrels | +0.071 |
pm | +0.070 |
drivers | +0.069 |
Neg logit contributions | |
kiss | -0.033 |
copy | -0.030 |
last | -0.027 |
massage | -0.026 |
dance | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
kiss | +0.001 |
copy | +0.001 |
last | +0.001 |
dance | +0.001 |
blow | +0.001 |
Neg logit contributions | |
ados | -0.002 |
riages | -0.002 |
barrels | -0.002 |
pm | -0.002 |
drivers | -0.002 |
Token | be |
Token ID | 307 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
ados | +0.068 |
riages | +0.066 |
barrels | +0.065 |
pm | +0.065 |
drivers | +0.065 |
Neg logit contributions | |
kiss | -0.032 |
copy | -0.028 |
last | -0.026 |
dance | -0.025 |
check | -0.025 |
Token | dishonest |
Token ID | 30549 |
Feature activation | +7.11e+00 |
Pos logit contributions | |
comfort | +0.013 |
cl | +0.013 |
kiss | +0.013 |
copy | +0.013 |
happily | +0.013 |
Neg logit contributions | |
ados | -0.032 |
drivers | -0.031 |
cases | -0.031 |
ables | -0.031 |
These | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -2.28e-01 |
Pos logit contributions | |
riages | +0.406 |
ados | +0.394 |
Sale | +0.384 |
opes | +0.383 |
barrels | +0.383 |
Neg logit contributions | |
again | -0.092 |
last | -0.084 |
kiss | -0.084 |
happily | -0.082 |
copy | -0.069 |
Token | He |
Token ID | 679 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
kiss | +0.008 |
copy | +0.007 |
last | +0.007 |
check | +0.007 |
happily | +0.007 |
Neg logit contributions | |
riages | -0.008 |
ados | -0.008 |
pm | -0.007 |
barrels | -0.007 |
drivers | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
ados | +0.009 |
riages | +0.009 |
barrels | +0.009 |
pm | +0.009 |
drivers | +0.009 |
Neg logit contributions | |
kiss | -0.005 |
copy | -0.004 |
last | -0.004 |
dance | -0.004 |
blow | -0.004 |
Token | too |
Token ID | 1165 |
Feature activation | +9.89e-01 |
Pos logit contributions | |
kiss | +0.012 |
copy | +0.011 |
dance | +0.010 |
last | +0.010 |
blow | +0.010 |
Neg logit contributions | |
riages | -0.014 |
ados | -0.014 |
barrels | -0.014 |
pm | -0.014 |
drivers | -0.013 |
Token | honest |
Token ID | 5508 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
ados | +0.040 |
riages | +0.040 |
barrels | +0.038 |
pm | +0.038 |
drivers | +0.037 |
Neg logit contributions | |
kiss | -0.028 |
copy | -0.026 |
last | -0.024 |
dance | -0.024 |
blow | -0.023 |
Token | these |
Token ID | 777 |
Feature activation | +3.31e+00 |
Pos logit contributions | |
ados | +0.005 |
riages | +0.005 |
pm | +0.005 |
barrels | +0.005 |
drivers | +0.005 |
Neg logit contributions | |
kiss | -0.005 |
copy | -0.004 |
last | -0.004 |
dance | -0.004 |
blow | -0.004 |
Token | carrots |
Token ID | 34397 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
ados | +0.130 |
drivers | +0.128 |
riages | +0.128 |
pm | +0.127 |
bank | +0.123 |
Neg logit contributions | |
kiss | -0.102 |
last | -0.093 |
watch | -0.089 |
hug | -0.089 |
comfort | -0.089 |
Token | are |
Token ID | 389 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
ados | +0.024 |
riages | +0.023 |
pm | +0.023 |
drivers | +0.022 |
barrels | +0.022 |
Neg logit contributions | |
kiss | -0.018 |
last | -0.017 |
copy | -0.016 |
blow | -0.016 |
dance | -0.016 |
Token | not |
Token ID | 407 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
riages | +0.072 |
ados | +0.072 |
pm | +0.071 |
ors | +0.069 |
drivers | +0.069 |
Neg logit contributions | |
last | -0.036 |
kiss | -0.036 |
copy | -0.033 |
blow | -0.032 |
also | -0.031 |
Token | for |
Token ID | 329 |
Feature activation | +2.62e+00 |
Pos logit contributions | |
pm | +0.101 |
riages | +0.101 |
ados | +0.099 |
Those | +0.097 |
drivers | +0.096 |
Neg logit contributions | |
kiss | -0.054 |
last | -0.050 |
copy | -0.050 |
blow | -0.048 |
check | -0.046 |
Token | sale |
Token ID | 5466 |
Feature activation | +7.05e+00 |
Pos logit contributions | |
riages | +0.106 |
ables | +0.101 |
drivers | +0.100 |
cases | +0.099 |
ados | +0.099 |
Neg logit contributions | |
kiss | -0.075 |
last | -0.069 |
copy | -0.069 |
blow | -0.068 |
dance | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
ados | +0.392 |
riages | +0.367 |
AW | +0.360 |
bank | +0.357 |
barrels | +0.357 |
Neg logit contributions | |
last | -0.118 |
kiss | -0.091 |
copy | -0.072 |
next | -0.067 |
again | -0.065 |
Token | They |
Token ID | 1119 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
riages | +0.036 |
barrels | +0.035 |
drivers | +0.034 |
ados | +0.033 |
pm | +0.033 |
Neg logit contributions | |
kiss | -0.033 |
Mom | -0.029 |
rub | -0.027 |
massage | -0.027 |
check | -0.026 |
Token | are |
Token ID | 389 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ados | +0.124 |
drivers | +0.115 |
pm | +0.114 |
riages | +0.114 |
barrels | +0.110 |
Neg logit contributions | |
kiss | -0.073 |
last | -0.065 |
blow | -0.062 |
copy | -0.062 |
watch | -0.061 |
Token | for |
Token ID | 329 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
ados | +0.052 |
pm | +0.051 |
riages | +0.051 |
ors | +0.050 |
drivers | +0.050 |
Neg logit contributions | |
kiss | -0.026 |
last | -0.024 |
copy | -0.023 |
blow | -0.022 |
check | -0.022 |
Token | our |
Token ID | 674 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
riages | +0.094 |
ados | +0.092 |
ables | +0.088 |
drivers | +0.088 |
pm | +0.088 |
Neg logit contributions | |
kiss | -0.068 |
copy | -0.062 |
last | -0.061 |
blow | -0.060 |
check | -0.060 |
Token | blankets |
Token ID | 34794 |
Feature activation | +3.26e+00 |
Pos logit contributions | |
ados | +0.044 |
riages | +0.043 |
pm | +0.041 |
barrels | +0.041 |
drivers | +0.040 |
Neg logit contributions | |
kiss | -0.039 |
copy | -0.037 |
last | -0.034 |
dance | -0.034 |
blow | -0.034 |
Token | and |
Token ID | 290 |
Feature activation | -3.84e-01 |
Pos logit contributions | |
ados | +0.151 |
pm | +0.146 |
riages | +0.144 |
barrels | +0.142 |
assing | +0.138 |
Neg logit contributions | |
kiss | -0.081 |
last | -0.072 |
copy | -0.070 |
blow | -0.066 |
dance | -0.066 |
Token | chairs |
Token ID | 18791 |
Feature activation | +4.46e+00 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.012 |
last | +0.011 |
dance | +0.011 |
blow | +0.011 |
Neg logit contributions | |
ados | -0.014 |
riages | -0.014 |
pm | -0.013 |
barrels | -0.013 |
drivers | -0.013 |
Token | to |
Token ID | 284 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
ados | +0.204 |
riages | +0.193 |
pm | +0.193 |
barrels | +0.191 |
ors | +0.182 |
Neg logit contributions | |
kiss | -0.113 |
last | -0.100 |
copy | -0.099 |
blow | -0.094 |
dance | -0.094 |
Token | make |
Token ID | 787 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
riages | +0.051 |
ados | +0.051 |
barrels | +0.050 |
pm | +0.049 |
drivers | +0.048 |
Neg logit contributions | |
kiss | -0.022 |
copy | -0.020 |
last | -0.018 |
massage | -0.017 |
dance | -0.017 |
Token | cages |
Token ID | 37933 |
Feature activation | +7.03e+00 |
Pos logit contributions | |
kiss | +0.017 |
copy | +0.016 |
dance | +0.015 |
last | +0.015 |
blow | +0.014 |
Neg logit contributions | |
ados | -0.024 |
riages | -0.023 |
barrels | -0.022 |
drivers | -0.022 |
pm | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | +3.36e-01 |
Pos logit contributions | |
ados | +0.370 |
riages | +0.345 |
drivers | +0.342 |
pm | +0.340 |
AW | +0.339 |
Neg logit contributions | |
kiss | -0.151 |
dance | -0.128 |
copy | -0.124 |
last | -0.121 |
blow | -0.110 |
Token | fences |
Token ID | 35771 |
Feature activation | +4.38e+00 |
Pos logit contributions | |
ados | +0.013 |
riages | +0.012 |
pm | +0.012 |
ors | +0.012 |
drivers | +0.012 |
Neg logit contributions | |
kiss | -0.011 |
copy | -0.010 |
dance | -0.010 |
last | -0.010 |
blow | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
ados | +0.234 |
pm | +0.215 |
barrels | +0.210 |
riages | +0.209 |
pounds | +0.208 |
Neg logit contributions | |
kiss | -0.089 |
last | -0.074 |
dance | -0.073 |
blow | -0.070 |
watch | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | +3.97e-02 |
Pos logit contributions | |
ados | +0.037 |
riages | +0.035 |
barrels | +0.034 |
pm | +0.033 |
drivers | +0.033 |
Neg logit contributions | |
kiss | -0.028 |
copy | -0.025 |
last | -0.023 |
dance | -0.023 |
check | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
ados | +0.002 |
riages | +0.002 |
pm | +0.002 |
barrels | +0.002 |
drivers | +0.002 |
Neg logit contributions | |
kiss | -0.001 |
copy | -0.001 |
dance | -0.001 |
massage | -0.001 |
last | -0.001 |
Token | doing |
Token ID | 1804 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
ados | +0.045 |
riages | +0.044 |
barrels | +0.042 |
pm | +0.041 |
drivers | +0.040 |
Neg logit contributions | |
kiss | -0.059 |
copy | -0.056 |
last | -0.053 |
dance | -0.052 |
massage | -0.052 |
Token | here |
Token ID | 994 |
Feature activation | +2.51e+00 |
Pos logit contributions | |
riages | +0.111 |
ados | +0.110 |
barrels | +0.109 |
pm | +0.108 |
drivers | +0.107 |
Neg logit contributions | |
kiss | -0.053 |
copy | -0.050 |
dance | -0.047 |
massage | -0.045 |
last | -0.045 |
Token | ? |
Token ID | 30 |
Feature activation | +4.19e-01 |
Pos logit contributions | |
ados | +0.111 |
riages | +0.109 |
barrels | +0.106 |
pm | +0.105 |
drivers | +0.104 |
Neg logit contributions | |
kiss | -0.063 |
copy | -0.058 |
last | -0.053 |
dance | -0.052 |
blow | -0.051 |
Token | You |
Token ID | 921 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
ados | +0.014 |
barrels | +0.014 |
riages | +0.014 |
banks | +0.014 |
pounds | +0.014 |
Neg logit contributions | |
kiss | -0.014 |
check | -0.013 |
watch | -0.012 |
blow | -0.012 |
Mom | -0.012 |
Token | are |
Token ID | 389 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
ados | +0.084 |
pm | +0.078 |
bank | +0.077 |
drivers | +0.075 |
barrels | +0.075 |
Neg logit contributions | |
kiss | -0.053 |
blow | -0.048 |
copy | -0.048 |
last | -0.048 |
watch | -0.046 |
Token | tresp |
Token ID | 29723 |
Feature activation | +7.03e+00 |
Pos logit contributions | |
ados | +0.062 |
bank | +0.061 |
pm | +0.060 |
ables | +0.060 |
drivers | +0.060 |
Neg logit contributions | |
last | -0.032 |
blow | -0.032 |
cl | -0.032 |
kiss | -0.031 |
copy | -0.031 |
Token | assing |
Token ID | 19696 |
Feature activation | +5.78e+00 |
Pos logit contributions | |
ados | +0.130 |
barrels | +0.098 |
riages | +0.095 |
cases | +0.093 |
pm | +0.090 |
Neg logit contributions | |
copy | -0.342 |
kiss | -0.338 |
check | -0.334 |
cl | -0.332 |
last | -0.328 |
Token | ! |
Token ID | 0 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
ados | +0.290 |
riages | +0.280 |
bank | +0.277 |
drivers | +0.272 |
pm | +0.271 |
Neg logit contributions | |
last | -0.096 |
kiss | -0.089 |
cl | -0.086 |
copy | -0.079 |
slowly | -0.072 |
Token | You |
Token ID | 921 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
riages | +0.056 |
ados | +0.055 |
barrels | +0.055 |
banks | +0.055 |
nuts | +0.055 |
Neg logit contributions | |
kiss | -0.059 |
Mom | -0.055 |
check | -0.053 |
cl | -0.053 |
watch | -0.052 |
Token | are |
Token ID | 389 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
ados | +0.091 |
pm | +0.084 |
bank | +0.084 |
drivers | +0.080 |
barrels | +0.079 |
Neg logit contributions | |
kiss | -0.054 |
blow | -0.052 |
last | -0.051 |
cl | -0.050 |
watch | -0.049 |
Token | thieves |
Token ID | 26655 |
Feature activation | +6.04e+00 |
Pos logit contributions | |
ados | +0.060 |
pm | +0.059 |
bank | +0.059 |
ables | +0.058 |
cases | +0.057 |
Neg logit contributions | |
last | -0.032 |
cl | -0.032 |
blow | -0.031 |
check | -0.030 |
copy | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +8.29e-01 |
Pos logit contributions | |
riages | +0.048 |
ados | +0.047 |
barrels | +0.047 |
pm | +0.046 |
assing | +0.046 |
Neg logit contributions | |
kiss | -0.026 |
copy | -0.023 |
dance | -0.022 |
cl | -0.021 |
smile | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | +3.30e+00 |
Pos logit contributions | |
ados | +0.032 |
riages | +0.031 |
barrels | +0.030 |
pm | +0.030 |
drivers | +0.029 |
Neg logit contributions | |
kiss | -0.026 |
copy | -0.024 |
last | -0.022 |
dance | -0.022 |
blow | -0.022 |
Token | They |
Token ID | 2990 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ados | +0.183 |
riages | +0.175 |
barrels | +0.173 |
ors | +0.171 |
pm | +0.169 |
Neg logit contributions | |
kiss | -0.046 |
copy | -0.044 |
blow | -0.036 |
dance | -0.036 |
last | -0.035 |
Token | are |
Token ID | 389 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ados | +0.074 |
drivers | +0.072 |
pm | +0.070 |
riages | +0.068 |
keeping | +0.067 |
Neg logit contributions | |
kiss | -0.039 |
watch | -0.038 |
copy | -0.038 |
blow | -0.038 |
play | -0.038 |
Token | av |
Token ID | 1196 |
Feature activation | +5.54e+00 |
Pos logit contributions | |
ables | +0.062 |
oc | +0.061 |
ors | +0.061 |
situations | +0.061 |
bank | +0.059 |
Neg logit contributions | |
mom | -0.024 |
also | -0.023 |
tw | -0.023 |
Mom | -0.023 |
cl | -0.023 |
Token | oc |
Token ID | 420 |
Feature activation | +6.98e+00 |
Pos logit contributions | |
ados | +0.088 |
riages | +0.088 |
pm | +0.087 |
cases | +0.086 |
bank | +0.084 |
Neg logit contributions | |
copy | -0.267 |
kiss | -0.266 |
last | -0.244 |
cl | -0.243 |
blow | -0.243 |
Token | ados |
Token ID | 22484 |
Feature activation | +6.46e+00 |
Pos logit contributions | |
ditch | +0.081 |
ables | +0.073 |
riages | +0.073 |
ours | +0.073 |
banks | +0.072 |
Neg logit contributions | |
kiss | -0.347 |
cl | -0.327 |
last | -0.325 |
copy | -0.324 |
hug | -0.313 |
Token | . |
Token ID | 13 |
Feature activation | +7.97e-01 |
Pos logit contributions | |
ados | +0.319 |
pm | +0.312 |
AW | +0.310 |
roc | +0.309 |
riages | +0.307 |
Neg logit contributions | |
last | -0.120 |
kiss | -0.117 |
dance | -0.105 |
blow | -0.103 |
cl | -0.103 |
Token | My |
Token ID | 2011 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
riages | +0.029 |
ors | +0.029 |
AW | +0.029 |
ados | +0.028 |
tresp | +0.028 |
Neg logit contributions | |
kiss | -0.024 |
Mom | -0.023 |
watch | -0.021 |
copy | -0.021 |
check | -0.021 |
Token | mom |
Token ID | 1995 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
kiss | +0.032 |
copy | +0.029 |
last | +0.026 |
dance | +0.025 |
check | +0.025 |
Neg logit contributions | |
riages | -0.069 |
ados | -0.067 |
pm | -0.067 |
drivers | -0.065 |
barrels | -0.064 |
Token | makes |
Token ID | 1838 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
kiss | +0.039 |
copy | +0.035 |
last | +0.033 |
blow | +0.032 |
check | +0.031 |
Neg logit contributions | |
ados | -0.078 |
riages | -0.075 |
barrels | -0.074 |
pm | -0.073 |
drivers | -0.073 |
Token | that |
Token ID | 326 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
riages | +0.067 |
barrels | +0.064 |
pm | +0.063 |
ables | +0.062 |
ados | +0.062 |
Neg logit contributions | |
kiss | -0.033 |
mom | -0.026 |
comfort | -0.025 |
last | -0.025 |
massage | -0.025 |
Token | her |
Token ID | 607 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
ados | +0.003 |
riages | +0.003 |
barrels | +0.003 |
pm | +0.003 |
drivers | +0.003 |
Neg logit contributions | |
kiss | -0.002 |
copy | -0.002 |
last | -0.001 |
dance | -0.001 |
blow | -0.001 |
Token | injury |
Token ID | 5095 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
copy | +0.009 |
kiss | +0.009 |
dance | +0.008 |
blow | +0.008 |
last | +0.008 |
Neg logit contributions | |
ados | -0.015 |
riages | -0.014 |
drivers | -0.014 |
barrels | -0.014 |
ables | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | +5.00e-01 |
Pos logit contributions | |
kiss | +0.015 |
copy | +0.014 |
last | +0.013 |
dance | +0.012 |
blow | +0.012 |
Neg logit contributions | |
ados | -0.032 |
riages | -0.031 |
barrels | -0.030 |
pm | -0.030 |
drivers | -0.030 |
Token | too |
Token ID | 1165 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ados | +0.020 |
riages | +0.020 |
barrels | +0.019 |
pm | +0.019 |
drivers | +0.019 |
Neg logit contributions | |
kiss | -0.015 |
copy | -0.014 |
dance | -0.012 |
last | -0.012 |
blow | -0.012 |
Token | severe |
Token ID | 6049 |
Feature activation | +6.94e+00 |
Pos logit contributions | |
ados | +0.056 |
riages | +0.055 |
barrels | +0.053 |
pm | +0.052 |
drivers | +0.052 |
Neg logit contributions | |
kiss | -0.048 |
copy | -0.046 |
dance | -0.042 |
last | -0.041 |
blow | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +6.11e-01 |
Pos logit contributions | |
riages | +0.352 |
ados | +0.351 |
barrels | +0.343 |
pm | +0.338 |
drivers | +0.328 |
Neg logit contributions | |
kiss | -0.139 |
last | -0.122 |
copy | -0.113 |
blow | -0.107 |
dance | -0.105 |
Token | she |
Token ID | 673 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
ados | +0.024 |
riages | +0.023 |
barrels | +0.022 |
pm | +0.022 |
drivers | +0.022 |
Neg logit contributions | |
kiss | -0.018 |
copy | -0.017 |
dance | -0.016 |
last | -0.016 |
massage | -0.015 |
Token | would |
Token ID | 561 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
ados | +0.093 |
riages | +0.092 |
drivers | +0.088 |
barrels | +0.087 |
pm | +0.086 |
Neg logit contributions | |
kiss | -0.082 |
copy | -0.079 |
dance | -0.072 |
last | -0.070 |
massage | -0.070 |
Token | never |
Token ID | 1239 |
Feature activation | +7.67e-01 |
Pos logit contributions | |
kiss | +0.042 |
copy | +0.040 |
last | +0.035 |
dance | +0.035 |
massage | +0.034 |
Neg logit contributions | |
ados | -0.076 |
riages | -0.076 |
barrels | -0.073 |
pm | -0.072 |
drivers | -0.072 |
Token | be |
Token ID | 307 |
Feature activation | -2.37e-01 |
Pos logit contributions | |
riages | +0.029 |
bank | +0.028 |
assing | +0.027 |
drivers | +0.027 |
tresp | +0.027 |
Neg logit contributions | |
patiently | -0.016 |
copy | -0.016 |
dancer | -0.016 |
mum | -0.015 |
kiss | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ados | +0.088 |
riages | +0.084 |
drivers | +0.084 |
barrels | +0.084 |
pm | +0.082 |
Neg logit contributions | |
kiss | -0.034 |
last | -0.031 |
copy | -0.028 |
dance | -0.028 |
play | -0.026 |
Token | He |
Token ID | 679 |
Feature activation | +7.48e-01 |
Pos logit contributions | |
barrels | +0.045 |
riages | +0.045 |
ados | +0.045 |
pm | +0.043 |
drivers | +0.043 |
Neg logit contributions | |
kiss | -0.043 |
Mom | -0.036 |
massage | -0.035 |
copy | -0.035 |
check | -0.034 |
Token | is |
Token ID | 318 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
ados | +0.038 |
riages | +0.034 |
keeping | +0.034 |
barrels | +0.034 |
structures | +0.034 |
Neg logit contributions | |
kiss | -0.017 |
last | -0.016 |
rub | -0.015 |
check | -0.015 |
watch | -0.014 |
Token | not |
Token ID | 407 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
ados | +0.013 |
riages | +0.012 |
pm | +0.012 |
drivers | +0.012 |
ors | +0.012 |
Neg logit contributions | |
kiss | -0.009 |
copy | -0.008 |
last | -0.008 |
check | -0.008 |
blow | -0.008 |
Token | for |
Token ID | 329 |
Feature activation | +2.45e+00 |
Pos logit contributions | |
ados | +0.094 |
riages | +0.092 |
pm | +0.091 |
ors | +0.089 |
drivers | +0.089 |
Neg logit contributions | |
kiss | -0.046 |
copy | -0.042 |
check | -0.039 |
last | -0.039 |
blow | -0.038 |
Token | sale |
Token ID | 5466 |
Feature activation | +6.91e+00 |
Pos logit contributions | |
riages | +0.100 |
ables | +0.098 |
ados | +0.097 |
drivers | +0.094 |
bank | +0.094 |
Neg logit contributions | |
kiss | -0.068 |
last | -0.063 |
copy | -0.062 |
dance | -0.061 |
check | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
ados | +0.379 |
riages | +0.359 |
bank | +0.357 |
AW | +0.354 |
barrels | +0.350 |
Neg logit contributions | |
last | -0.118 |
kiss | -0.100 |
copy | -0.083 |
next | -0.074 |
watch | -0.071 |
Token | He |
Token ID | 679 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
ados | +0.051 |
riages | +0.050 |
barrels | +0.049 |
bank | +0.048 |
pm | +0.047 |
Neg logit contributions | |
kiss | -0.044 |
Mom | -0.040 |
massage | -0.036 |
check | -0.036 |
mom | -0.036 |
Token | is |
Token ID | 318 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
ados | +0.036 |
riages | +0.032 |
barrels | +0.032 |
keeping | +0.032 |
drivers | +0.032 |
Neg logit contributions | |
kiss | -0.016 |
last | -0.015 |
rub | -0.014 |
watch | -0.014 |
check | -0.014 |
Token | my |
Token ID | 616 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ados | +0.018 |
riages | +0.018 |
pm | +0.017 |
drivers | +0.017 |
ors | +0.017 |
Neg logit contributions | |
kiss | -0.012 |
copy | -0.011 |
last | -0.010 |
check | -0.010 |
blow | -0.010 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.56e-01 |
Pos logit contributions | |
ados | +0.078 |
riages | +0.077 |
pm | +0.075 |
barrels | +0.074 |
drivers | +0.073 |
Neg logit contributions | |
kiss | -0.040 |
copy | -0.036 |
last | -0.034 |
dance | -0.032 |
massage | -0.032 |
Token | the |
Token ID | 262 |
Feature activation | +8.56e-01 |
Pos logit contributions | |
riages | +0.028 |
ados | +0.027 |
ables | +0.026 |
barrels | +0.026 |
bank | +0.026 |
Neg logit contributions | |
kiss | -0.017 |
copy | -0.014 |
last | -0.014 |
dance | -0.014 |
blow | -0.013 |
Token | cars |
Token ID | 5006 |
Feature activation | +5.75e-01 |
Pos logit contributions | |
ados | +0.035 |
riages | +0.034 |
barrels | +0.033 |
pm | +0.033 |
drivers | +0.032 |
Neg logit contributions | |
kiss | -0.025 |
copy | -0.023 |
last | -0.022 |
dance | -0.021 |
blow | -0.021 |
Token | are |
Token ID | 389 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ados | +0.026 |
pm | +0.024 |
riages | +0.024 |
barrels | +0.023 |
bank | +0.023 |
Neg logit contributions | |
kiss | -0.016 |
last | -0.016 |
copy | -0.014 |
blow | -0.014 |
watch | -0.014 |
Token | not |
Token ID | 407 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
ados | +0.053 |
riages | +0.050 |
pm | +0.049 |
ors | +0.049 |
bank | +0.048 |
Neg logit contributions | |
kiss | -0.030 |
copy | -0.028 |
last | -0.027 |
blow | -0.026 |
check | -0.025 |
Token | for |
Token ID | 329 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
ados | +0.091 |
pm | +0.087 |
riages | +0.084 |
Those | +0.084 |
These | +0.084 |
Neg logit contributions | |
kiss | -0.043 |
copy | -0.041 |
last | -0.038 |
blow | -0.038 |
check | -0.037 |
Token | sale |
Token ID | 5466 |
Feature activation | +6.90e+00 |
Pos logit contributions | |
ados | +0.112 |
riages | +0.110 |
barrels | +0.106 |
pm | +0.104 |
ables | +0.104 |
Neg logit contributions | |
kiss | -0.077 |
copy | -0.072 |
last | -0.069 |
dance | -0.068 |
blow | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
ados | +0.384 |
AW | +0.355 |
riages | +0.350 |
barrels | +0.349 |
bank | +0.349 |
Neg logit contributions | |
last | -0.119 |
kiss | -0.100 |
copy | -0.085 |
watch | -0.074 |
turn | -0.074 |
Token | They |
Token ID | 1119 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
riages | +0.040 |
ados | +0.039 |
barrels | +0.039 |
pm | +0.038 |
drivers | +0.037 |
Neg logit contributions | |
kiss | -0.025 |
copy | -0.022 |
last | -0.020 |
check | -0.020 |
dance | -0.020 |
Token | are |
Token ID | 389 |
Feature activation | +9.53e-01 |
Pos logit contributions | |
ados | +0.122 |
riages | +0.119 |
barrels | +0.115 |
pm | +0.115 |
drivers | +0.114 |
Neg logit contributions | |
kiss | -0.049 |
copy | -0.044 |
last | -0.039 |
dance | -0.038 |
blow | -0.037 |
Token | only |
Token ID | 691 |
Feature activation | -6.47e-01 |
Pos logit contributions | |
ados | +0.039 |
riages | +0.039 |
barrels | +0.038 |
pm | +0.037 |
drivers | +0.036 |
Neg logit contributions | |
kiss | -0.026 |
copy | -0.025 |
last | -0.022 |
dance | -0.022 |
massage | -0.022 |
Token | for |
Token ID | 329 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
kiss | +0.012 |
copy | +0.011 |
last | +0.010 |
check | +0.010 |
cl | +0.010 |
Neg logit contributions | |
ados | -0.034 |
riages | -0.032 |
ors | -0.032 |
bank | -0.031 |
pm | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +7.23e-01 |
Pos logit contributions | |
ados | +0.010 |
riages | +0.010 |
barrels | +0.009 |
drivers | +0.009 |
pm | +0.009 |
Neg logit contributions | |
kiss | -0.007 |
last | -0.006 |
copy | -0.006 |
dance | -0.006 |
watch | -0.006 |
Token | they |
Token ID | 484 |
Feature activation | +2.54e+00 |
Pos logit contributions | |
ados | +0.029 |
riages | +0.029 |
drivers | +0.028 |
barrels | +0.027 |
ors | +0.027 |
Neg logit contributions | |
kiss | -0.022 |
copy | -0.020 |
dance | -0.019 |
last | -0.019 |
massage | -0.019 |
Token | pretend |
Token ID | 16614 |
Feature activation | -1.32e+00 |
Pos logit contributions | |
ados | +0.124 |
riages | +0.119 |
drivers | +0.115 |
pm | +0.115 |
barrels | +0.114 |
Neg logit contributions | |
kiss | -0.059 |
copy | -0.053 |
dance | -0.049 |
last | -0.048 |
blow | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | +5.90e-01 |
Pos logit contributions | |
kiss | +0.023 |
dance | +0.021 |
last | +0.019 |
watch | +0.019 |
copy | +0.017 |
Neg logit contributions | |
ados | -0.073 |
riages | -0.070 |
drivers | -0.069 |
ors | -0.067 |
assing | -0.067 |
Token | be |
Token ID | 307 |
Feature activation | -8.75e-01 |
Pos logit contributions | |
riages | +0.026 |
barrels | +0.026 |
pm | +0.026 |
ados | +0.026 |
drivers | +0.025 |
Neg logit contributions | |
kiss | -0.013 |
copy | -0.012 |
last | -0.010 |
massage | -0.010 |
scrub | -0.010 |
Token | lawyers |
Token ID | 9326 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
kiss | +0.021 |
copy | +0.020 |
dance | +0.019 |
cl | +0.018 |
watch | +0.018 |
Neg logit contributions | |
ados | -0.041 |
ors | -0.038 |
riages | -0.038 |
pm | -0.038 |
drivers | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
ados | +0.358 |
barrels | +0.331 |
AW | +0.331 |
riages | +0.331 |
drivers | +0.323 |
Neg logit contributions | |
kiss | -0.134 |
last | -0.117 |
dance | -0.116 |
copy | -0.115 |
watch | -0.110 |
Token | They |
Token ID | 1119 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
ados | +0.039 |
riages | +0.038 |
barrels | +0.036 |
pm | +0.036 |
drivers | +0.036 |
Neg logit contributions | |
kiss | -0.035 |
copy | -0.033 |
last | -0.031 |
dance | -0.030 |
massage | -0.030 |
Token | have |
Token ID | 423 |
Feature activation | -4.64e-01 |
Pos logit contributions | |
ados | +0.127 |
riages | +0.119 |
drivers | +0.116 |
pm | +0.116 |
barrels | +0.116 |
Neg logit contributions | |
kiss | -0.051 |
copy | -0.046 |
last | -0.041 |
dance | -0.041 |
blow | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -8.28e-01 |
Pos logit contributions | |
kiss | +0.009 |
copy | +0.008 |
last | +0.008 |
watch | +0.008 |
dance | +0.008 |
Neg logit contributions | |
ados | -0.024 |
ors | -0.023 |
riages | -0.023 |
pm | -0.022 |
bank | -0.022 |
Token | big |
Token ID | 1263 |
Feature activation | +8.36e-02 |
Pos logit contributions | |
kiss | +0.022 |
copy | +0.021 |
last | +0.019 |
blow | +0.019 |
dance | +0.018 |
Neg logit contributions | |
ados | -0.034 |
riages | -0.033 |
barrels | -0.033 |
drivers | -0.032 |
pm | -0.032 |
Token | Mom |
Token ID | 11254 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
riages | +0.002 |
ados | +0.002 |
barrels | +0.002 |
drivers | +0.002 |
pm | +0.002 |
Neg logit contributions | |
kiss | -0.002 |
copy | -0.002 |
last | -0.002 |
dance | -0.002 |
massage | -0.002 |
Token | says |
Token ID | 1139 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
kiss | +0.057 |
copy | +0.052 |
last | +0.048 |
dance | +0.046 |
blow | +0.046 |
Neg logit contributions | |
ados | -0.106 |
riages | -0.104 |
barrels | -0.100 |
pm | -0.100 |
drivers | -0.099 |
Token | they |
Token ID | 484 |
Feature activation | +3.27e+00 |
Pos logit contributions | |
riages | +0.028 |
ables | +0.027 |
ors | +0.026 |
bank | +0.026 |
keeping | +0.026 |
Neg logit contributions | |
kiss | -0.012 |
mom | -0.010 |
rose | -0.010 |
cl | -0.010 |
dance | -0.009 |
Token | are |
Token ID | 389 |
Feature activation | +9.95e-01 |
Pos logit contributions | |
ados | +0.163 |
riages | +0.157 |
pm | +0.153 |
bank | +0.152 |
drivers | +0.151 |
Neg logit contributions | |
kiss | -0.070 |
copy | -0.062 |
last | -0.061 |
check | -0.061 |
watch | -0.059 |
Token | av |
Token ID | 1196 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
ors | +0.050 |
riages | +0.048 |
drivers | +0.048 |
ados | +0.048 |
bank | +0.047 |
Neg logit contributions | |
kiss | -0.019 |
last | -0.018 |
check | -0.018 |
mom | -0.017 |
cl | -0.017 |
Token | oc |
Token ID | 420 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
ados | +0.081 |
riages | +0.078 |
pm | +0.073 |
barrels | +0.071 |
drivers | +0.070 |
Neg logit contributions | |
kiss | -0.259 |
copy | -0.252 |
last | -0.239 |
dance | -0.232 |
check | -0.232 |
Token | ados |
Token ID | 22484 |
Feature activation | +5.72e+00 |
Pos logit contributions | |
ditch | +0.076 |
riages | +0.076 |
ables | +0.072 |
pm | +0.071 |
banks | +0.068 |
Neg logit contributions | |
kiss | -0.340 |
last | -0.323 |
patiently | -0.321 |
cl | -0.318 |
happily | -0.317 |
Token | . |
Token ID | 13 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ados | +0.295 |
AW | +0.288 |
pm | +0.286 |
riages | +0.284 |
barrels | +0.272 |
Neg logit contributions | |
kiss | -0.106 |
last | -0.104 |
dance | -0.089 |
copy | -0.080 |
watch | -0.079 |
Token | Sara |
Token ID | 24799 |
Feature activation | -8.98e-01 |
Pos logit contributions | |
riages | +0.005 |
ados | +0.005 |
barrels | +0.005 |
drivers | +0.005 |
pm | +0.004 |
Neg logit contributions | |
kiss | -0.002 |
copy | -0.002 |
massage | -0.002 |
last | -0.002 |
dance | -0.002 |
Token | wants |
Token ID | 3382 |
Feature activation | -1.99e+00 |
Pos logit contributions | |
kiss | +0.028 |
copy | +0.026 |
last | +0.024 |
massage | +0.023 |
dance | +0.023 |
Neg logit contributions | |
ados | -0.034 |
riages | -0.033 |
barrels | -0.033 |
pm | -0.032 |
drivers | -0.031 |
Token | to |
Token ID | 284 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
kiss | +0.031 |
last | +0.028 |
copy | +0.025 |
blow | +0.022 |
check | +0.022 |
Neg logit contributions | |
riages | -0.111 |
pm | -0.106 |
drivers | -0.105 |
ors | -0.104 |
AW | -0.103 |
Token | 's |
Token ID | 338 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
kiss | +0.095 |
last | +0.089 |
blow | +0.081 |
watch | +0.078 |
also | +0.078 |
Neg logit contributions | |
ados | -0.217 |
pm | -0.203 |
barrels | -0.202 |
riages | -0.201 |
drivers | -0.193 |
Token | yard |
Token ID | 12699 |
Feature activation | +1.62e+00 |
Pos logit contributions | |
kiss | +0.038 |
copy | +0.032 |
last | +0.032 |
watch | +0.031 |
check | +0.031 |
Neg logit contributions | |
ados | -0.073 |
riages | -0.072 |
pm | -0.071 |
drivers | -0.069 |
ors | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
ados | +0.093 |
pm | +0.088 |
riages | +0.086 |
AW | +0.085 |
assing | +0.085 |
Neg logit contributions | |
last | -0.023 |
kiss | -0.021 |
watch | -0.015 |
play | -0.015 |
also | -0.015 |
Token | not |
Token ID | 407 |
Feature activation | +2.34e+00 |
Pos logit contributions | |
ados | +0.012 |
pm | +0.011 |
riages | +0.011 |
ors | +0.011 |
drivers | +0.011 |
Neg logit contributions | |
last | -0.006 |
kiss | -0.006 |
check | -0.005 |
copy | -0.005 |
scrub | -0.005 |
Token | for |
Token ID | 329 |
Feature activation | +2.74e+00 |
Pos logit contributions | |
ados | +0.113 |
pm | +0.110 |
riages | +0.109 |
Those | +0.108 |
ors | +0.108 |
Neg logit contributions | |
kiss | -0.049 |
last | -0.048 |
check | -0.046 |
copy | -0.045 |
blow | -0.045 |
Token | sale |
Token ID | 5466 |
Feature activation | +6.82e+00 |
Pos logit contributions | |
ables | +0.118 |
riages | +0.116 |
ados | +0.114 |
AW | +0.112 |
pm | +0.112 |
Neg logit contributions | |
last | -0.064 |
kiss | -0.062 |
check | -0.060 |
cl | -0.060 |
watch | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
ados | +0.354 |
AW | +0.328 |
riages | +0.325 |
barrels | +0.324 |
bank | +0.320 |
Neg logit contributions | |
last | -0.135 |
kiss | -0.113 |
copy | -0.104 |
check | -0.093 |
watch | -0.093 |
Token | Tim |
Token ID | 5045 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
Mom | +0.004 |
kiss | +0.003 |
mom | +0.003 |
happily | +0.003 |
Mum | +0.003 |
Neg logit contributions | |
pounds | -0.006 |
pm | -0.006 |
Street | -0.006 |
barrels | -0.006 |
dollars | -0.006 |
Token | my |
Token ID | 1820 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.007 |
dance | +0.006 |
blow | +0.006 |
massage | +0.006 |
Neg logit contributions | |
ados | -0.018 |
riages | -0.018 |
barrels | -0.017 |
pm | -0.017 |
drivers | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
kiss | +0.086 |
copy | +0.078 |
last | +0.071 |
blow | +0.068 |
dance | +0.067 |
Neg logit contributions | |
ados | -0.184 |
riages | -0.179 |
barrels | -0.175 |
pm | -0.173 |
drivers | -0.171 |
Token | very |
Token ID | 845 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
kiss | +0.023 |
copy | +0.021 |
last | +0.019 |
dance | +0.019 |
blow | +0.019 |
Neg logit contributions | |
ados | -0.030 |
riages | -0.029 |
barrels | -0.028 |
pm | -0.028 |
drivers | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +2.11e+00 |
Pos logit contributions | |
ados | +0.093 |
riages | +0.088 |
barrels | +0.087 |
pm | +0.086 |
drivers | +0.084 |
Neg logit contributions | |
kiss | -0.052 |
last | -0.046 |
copy | -0.045 |
dance | -0.043 |
watch | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
ados | +0.101 |
riages | +0.099 |
barrels | +0.097 |
pm | +0.096 |
drivers | +0.095 |
Neg logit contributions | |
kiss | -0.045 |
copy | -0.042 |
last | -0.036 |
dance | -0.036 |
blow | -0.035 |
Token | phone |
Token ID | 3072 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
ados | +0.001 |
riages | +0.001 |
barrels | +0.001 |
drivers | +0.001 |
pm | +0.001 |
Neg logit contributions | |
kiss | -0.000 |
copy | -0.000 |
last | -0.000 |
blow | -0.000 |
dance | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +9.47e-01 |
Pos logit contributions | |
ados | +0.158 |
barrels | +0.148 |
pm | +0.148 |
pounds | +0.144 |
riages | +0.141 |
Neg logit contributions | |
kiss | -0.088 |
watch | -0.075 |
copy | -0.073 |
dance | -0.073 |
last | -0.073 |
Token | some |
Token ID | 617 |
Feature activation | +3.15e-01 |
Pos logit contributions | |
ados | +0.047 |
riages | +0.046 |
pm | +0.044 |
barrels | +0.044 |
drivers | +0.044 |
Neg logit contributions | |
kiss | -0.019 |
copy | -0.017 |
last | -0.015 |
dance | -0.015 |
blow | -0.015 |
Token | files |
Token ID | 3696 |
Feature activation | +6.82e+00 |
Pos logit contributions | |
ados | +0.011 |
riages | +0.010 |
pm | +0.010 |
barrels | +0.009 |
ors | +0.009 |
Neg logit contributions | |
kiss | -0.012 |
copy | -0.011 |
last | -0.010 |
dance | -0.010 |
watch | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +9.81e-01 |
Pos logit contributions | |
ados | +0.358 |
barrels | +0.333 |
riages | +0.333 |
pm | +0.333 |
bank | +0.329 |
Neg logit contributions | |
kiss | -0.127 |
last | -0.114 |
copy | -0.105 |
cl | -0.102 |
dance | -0.101 |
Token |
|
Token ID | 198 |
Feature activation | +5.28e-01 |
Pos logit contributions | |
ados | +0.041 |
riages | +0.040 |
barrels | +0.039 |
drivers | +0.038 |
pm | +0.038 |
Neg logit contributions | |
kiss | -0.028 |
copy | -0.025 |
last | -0.023 |
dance | -0.023 |
massage | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +4.03e-01 |
Pos logit contributions | |
pm | +0.020 |
These | +0.020 |
riages | +0.019 |
drivers | +0.019 |
barrels | +0.019 |
Neg logit contributions | |
kiss | -0.015 |
copy | -0.015 |
dance | -0.013 |
blow | -0.013 |
massage | -0.013 |
Token | " |
Token ID | 1 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
ados | +0.018 |
riages | +0.017 |
barrels | +0.016 |
drivers | +0.016 |
ors | +0.016 |
Neg logit contributions | |
kiss | -0.011 |
copy | -0.011 |
blow | -0.010 |
last | -0.010 |
massage | -0.010 |
Token | Hello |
Token ID | 15496 |
Feature activation | +1.59e-01 |
Pos logit contributions | |
ados | +0.102 |
pm | +0.101 |
barrels | +0.100 |
riages | +0.099 |
These | +0.099 |
Neg logit contributions | |
kiss | -0.085 |
copy | -0.076 |
last | -0.075 |
patiently | -0.072 |
smile | -0.072 |
Token | The |
Token ID | 464 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
ados | +0.127 |
riages | +0.125 |
pm | +0.121 |
barrels | +0.120 |
drivers | +0.119 |
Neg logit contributions | |
kiss | -0.071 |
copy | -0.064 |
last | -0.059 |
dance | -0.058 |
blow | -0.056 |
Token | le |
Token ID | 443 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
kiss | +0.001 |
check | +0.001 |
last | +0.001 |
watch | +0.001 |
copy | +0.001 |
Neg logit contributions | |
ados | -0.001 |
ors | -0.001 |
cases | -0.001 |
riages | -0.001 |
ables | -0.001 |
Token | opard |
Token ID | 15478 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
ados | +0.003 |
pm | +0.003 |
riages | +0.003 |
barrels | +0.003 |
cases | +0.003 |
Neg logit contributions | |
kiss | -0.004 |
copy | -0.004 |
watch | -0.004 |
last | -0.004 |
check | -0.004 |
Token | toy |
Token ID | 13373 |
Feature activation | -8.83e-01 |
Pos logit contributions | |
kiss | +0.002 |
last | +0.002 |
watch | +0.002 |
rub | +0.002 |
check | +0.002 |
Neg logit contributions | |
ados | -0.005 |
riages | -0.005 |
drivers | -0.005 |
barrels | -0.005 |
keeping | -0.005 |
Token | costs |
Token ID | 3484 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
last | +0.018 |
watch | +0.017 |
kiss | +0.017 |
cl | +0.016 |
rose | +0.016 |
Neg logit contributions | |
ados | -0.040 |
pm | -0.039 |
keeping | -0.038 |
kilograms | -0.037 |
structures | -0.037 |
Token | $ |
Token ID | 720 |
Feature activation | +6.82e+00 |
Pos logit contributions | |
ados | +0.125 |
riages | +0.122 |
bank | +0.118 |
barrels | +0.115 |
pm | +0.115 |
Neg logit contributions | |
last | -0.064 |
kiss | -0.063 |
copy | -0.061 |
check | -0.058 |
folded | -0.054 |
Token | 10 |
Token ID | 940 |
Feature activation | +6.03e-01 |
Pos logit contributions | |
ados | +0.194 |
riages | +0.192 |
bank | +0.189 |
barrels | +0.185 |
drivers | +0.179 |
Neg logit contributions | |
kiss | -0.258 |
copy | -0.256 |
last | -0.245 |
check | -0.239 |
str | -0.226 |
Token | ." |
Token ID | 526 |
Feature activation | -3.28e-01 |
Pos logit contributions | |
ados | +0.022 |
riages | +0.021 |
bank | +0.021 |
barrels | +0.020 |
pm | +0.020 |
Neg logit contributions | |
last | -0.018 |
kiss | -0.018 |
copy | -0.018 |
check | -0.017 |
watch | -0.017 |
Token | |
Token ID | 220 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
kiss | +0.008 |
Mom | +0.007 |
play | +0.007 |
cl | +0.007 |
watch | +0.007 |
Neg logit contributions | |
pounds | -0.015 |
tresp | -0.015 |
pm | -0.014 |
riages | -0.014 |
ados | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -6.30e-01 |
Pos logit contributions | |
ados | +0.023 |
riages | +0.022 |
barrels | +0.022 |
pm | +0.022 |
drivers | +0.021 |
Neg logit contributions | |
kiss | -0.008 |
copy | -0.007 |
last | -0.007 |
dance | -0.006 |
blow | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | -7.77e-01 |
Pos logit contributions | |
kiss | +0.014 |
copy | +0.013 |
dance | +0.012 |
massage | +0.012 |
last | +0.012 |
Neg logit contributions | |
ados | -0.029 |
riages | -0.029 |
barrels | -0.028 |
pm | -0.027 |
drivers | -0.027 |
Token | grandma |
Token ID | 49890 |
Feature activation | -1.09e+00 |
Pos logit contributions | |
kiss | +0.000 |
copy | +0.000 |
blow | +0.000 |
last | +0.000 |
massage | +0.000 |
Neg logit contributions | |
riages | -0.000 |
ados | -0.000 |
barrels | -0.000 |
pm | -0.000 |
drivers | -0.000 |
Token | 's |
Token ID | 338 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
kiss | +0.029 |
copy | +0.025 |
last | +0.025 |
dance | +0.024 |
blow | +0.023 |
Neg logit contributions | |
ados | -0.049 |
riages | -0.046 |
barrels | -0.046 |
pm | -0.046 |
drivers | -0.045 |
Token | house |
Token ID | 2156 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
kiss | +0.027 |
copy | +0.024 |
last | +0.022 |
dance | +0.022 |
check | +0.021 |
Neg logit contributions | |
ados | -0.048 |
riages | -0.047 |
pm | -0.045 |
barrels | -0.045 |
drivers | -0.044 |
Token | for |
Token ID | 329 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
ados | +0.120 |
pm | +0.114 |
riages | +0.114 |
barrels | +0.112 |
drivers | +0.110 |
Neg logit contributions | |
kiss | -0.047 |
last | -0.042 |
copy | -0.039 |
dance | -0.036 |
check | -0.035 |
Token | safe |
Token ID | 3338 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
ados | +0.099 |
riages | +0.097 |
pm | +0.095 |
ables | +0.094 |
barrels | +0.094 |
Neg logit contributions | |
kiss | -0.046 |
copy | -0.040 |
last | -0.040 |
check | -0.039 |
dance | -0.037 |
Token | keeping |
Token ID | 19934 |
Feature activation | +6.80e+00 |
Pos logit contributions | |
ados | +0.077 |
pm | +0.077 |
AW | +0.076 |
riages | +0.076 |
These | +0.075 |
Neg logit contributions | |
kiss | -0.054 |
last | -0.049 |
dance | -0.048 |
check | -0.048 |
scrub | -0.047 |
Token | . |
Token ID | 13 |
Feature activation | -5.41e-01 |
Pos logit contributions | |
ados | +0.348 |
barrels | +0.339 |
pm | +0.333 |
riages | +0.330 |
crates | +0.314 |
Neg logit contributions | |
kiss | -0.132 |
last | -0.107 |
watch | -0.102 |
copy | -0.102 |
cl | -0.101 |
Token | |
Token ID | 220 |
Feature activation | +3.85e-01 |
Pos logit contributions | |
kiss | +0.014 |
Mom | +0.013 |
check | +0.012 |
mom | +0.012 |
play | +0.011 |
Neg logit contributions | |
ados | -0.024 |
pounds | -0.024 |
pm | -0.024 |
keys | -0.023 |
tresp | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
ados | +0.020 |
riages | +0.019 |
barrels | +0.019 |
pm | +0.019 |
drivers | +0.018 |
Neg logit contributions | |
kiss | -0.007 |
copy | -0.006 |
last | -0.005 |
blow | -0.005 |
check | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -7.74e-01 |
Pos logit contributions | |
kiss | +0.015 |
copy | +0.014 |
last | +0.012 |
dance | +0.012 |
blow | +0.012 |
Neg logit contributions | |
ados | -0.034 |
riages | -0.033 |
barrels | -0.032 |
pm | -0.032 |
drivers | -0.032 |
Token | When |
Token ID | 2215 |
Feature activation | -5.12e-01 |
Pos logit contributions | |
kiss | +0.020 |
copy | +0.019 |
scrub | +0.017 |
blow | +0.016 |
comfort | +0.016 |
Neg logit contributions | |
ados | -0.036 |
pm | -0.033 |
riages | -0.033 |
barrels | -0.033 |
ors | -0.032 |
Token | wrong |
Token ID | 2642 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
ados | +0.095 |
riages | +0.092 |
barrels | +0.089 |
pm | +0.088 |
drivers | +0.087 |
Neg logit contributions | |
kiss | -0.093 |
copy | -0.088 |
last | -0.082 |
dance | -0.081 |
blow | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
riages | +0.151 |
ados | +0.151 |
drivers | +0.144 |
barrels | +0.142 |
pm | +0.139 |
Neg logit contributions | |
kiss | -0.046 |
copy | -0.040 |
last | -0.039 |
watch | -0.036 |
dance | -0.036 |
Token | You |
Token ID | 921 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
riages | +0.031 |
barrels | +0.030 |
drivers | +0.029 |
cases | +0.029 |
nuts | +0.028 |
Neg logit contributions | |
kiss | -0.036 |
copy | -0.033 |
check | -0.033 |
Mom | -0.032 |
watch | -0.032 |
Token | are |
Token ID | 389 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
ados | +0.072 |
riages | +0.066 |
pm | +0.066 |
bank | +0.065 |
barrels | +0.065 |
Neg logit contributions | |
kiss | -0.048 |
copy | -0.045 |
last | -0.043 |
blow | -0.042 |
dance | -0.041 |
Token | being |
Token ID | 852 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ados | +0.051 |
riages | +0.050 |
pm | +0.049 |
ors | +0.049 |
drivers | +0.048 |
Neg logit contributions | |
kiss | -0.030 |
copy | -0.029 |
last | -0.027 |
cl | -0.026 |
blow | -0.026 |
Token | disrespectful |
Token ID | 41987 |
Feature activation | +6.78e+00 |
Pos logit contributions | |
ados | +0.043 |
riages | +0.041 |
pm | +0.040 |
barrels | +0.040 |
ors | +0.040 |
Neg logit contributions | |
kiss | -0.029 |
copy | -0.029 |
last | -0.026 |
cl | -0.026 |
dance | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +1.36e+00 |
Pos logit contributions | |
ados | +0.357 |
riages | +0.349 |
barrels | +0.339 |
pm | +0.335 |
drivers | +0.330 |
Neg logit contributions | |
kiss | -0.118 |
last | -0.106 |
copy | -0.100 |
cl | -0.089 |
watch | -0.085 |
Token | You |
Token ID | 921 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
riages | +0.043 |
barrels | +0.041 |
cases | +0.041 |
drivers | +0.040 |
nuts | +0.040 |
Neg logit contributions | |
kiss | -0.050 |
Mom | -0.046 |
copy | -0.046 |
check | -0.045 |
watch | -0.044 |
Token | do |
Token ID | 466 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ados | +0.085 |
pm | +0.079 |
riages | +0.078 |
bank | +0.078 |
barrels | +0.077 |
Neg logit contributions | |
kiss | -0.052 |
copy | -0.048 |
last | -0.047 |
blow | -0.046 |
watch | -0.044 |
Token | not |
Token ID | 407 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
copy | +0.029 |
kiss | +0.028 |
last | +0.028 |
watch | +0.026 |
blow | +0.026 |
Neg logit contributions | |
ados | -0.080 |
riages | -0.075 |
barrels | -0.073 |
drivers | -0.072 |
cases | -0.072 |
Token | know |
Token ID | 760 |
Feature activation | -6.76e-01 |
Pos logit contributions | |
ados | +0.052 |
riages | +0.050 |
barrels | +0.049 |
pm | +0.049 |
drivers | +0.048 |
Neg logit contributions | |
kiss | -0.024 |
copy | -0.022 |
last | -0.019 |
dance | -0.019 |
blow | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +7.62e-01 |
Pos logit contributions | |
ados | +0.021 |
riages | +0.021 |
barrels | +0.021 |
drivers | +0.020 |
pm | +0.020 |
Neg logit contributions | |
kiss | -0.012 |
copy | -0.010 |
last | -0.010 |
dance | -0.010 |
blow | -0.009 |
Token | You |
Token ID | 921 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
ados | +0.019 |
riages | +0.018 |
These | +0.018 |
pm | +0.017 |
Those | +0.017 |
Neg logit contributions | |
copy | -0.032 |
kiss | -0.032 |
last | -0.030 |
dance | -0.029 |
blow | -0.029 |
Token | are |
Token ID | 389 |
Feature activation | +1.12e+00 |
Pos logit contributions | |
ados | +0.065 |
riages | +0.062 |
pm | +0.061 |
barrels | +0.061 |
drivers | +0.059 |
Neg logit contributions | |
kiss | -0.056 |
copy | -0.051 |
last | -0.048 |
dance | -0.048 |
blow | -0.047 |
Token | not |
Token ID | 407 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
ados | +0.049 |
riages | +0.048 |
pm | +0.047 |
barrels | +0.046 |
drivers | +0.046 |
Neg logit contributions | |
kiss | -0.030 |
copy | -0.028 |
last | -0.026 |
dance | -0.025 |
blow | -0.025 |
Token | for |
Token ID | 329 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
ados | +0.087 |
riages | +0.086 |
pm | +0.084 |
barrels | +0.083 |
drivers | +0.082 |
Neg logit contributions | |
kiss | -0.051 |
copy | -0.047 |
last | -0.043 |
check | -0.043 |
dance | -0.043 |
Token | sale |
Token ID | 5466 |
Feature activation | +6.77e+00 |
Pos logit contributions | |
ados | +0.093 |
riages | +0.092 |
barrels | +0.089 |
pm | +0.087 |
drivers | +0.087 |
Neg logit contributions | |
kiss | -0.070 |
copy | -0.065 |
last | -0.061 |
dance | -0.060 |
blow | -0.059 |
Token | . |
Token ID | 13 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ados | +0.351 |
riages | +0.338 |
barrels | +0.336 |
bank | +0.331 |
AW | +0.327 |
Neg logit contributions | |
last | -0.124 |
kiss | -0.124 |
copy | -0.103 |
dance | -0.089 |
cl | -0.088 |
Token | You |
Token ID | 921 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
ados | +0.034 |
riages | +0.033 |
barrels | +0.032 |
pm | +0.031 |
drivers | +0.031 |
Neg logit contributions | |
kiss | -0.052 |
copy | -0.049 |
last | -0.047 |
dance | -0.046 |
blow | -0.046 |
Token | are |
Token ID | 389 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
ados | +0.075 |
barrels | +0.070 |
riages | +0.070 |
pm | +0.070 |
drivers | +0.069 |
Neg logit contributions | |
kiss | -0.057 |
copy | -0.052 |
last | -0.050 |
blow | -0.049 |
dance | -0.049 |
Token | for |
Token ID | 329 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
ados | +0.055 |
riages | +0.054 |
pm | +0.053 |
drivers | +0.053 |
bank | +0.052 |
Neg logit contributions | |
kiss | -0.032 |
copy | -0.030 |
last | -0.028 |
dance | -0.027 |
check | -0.027 |
Token | me |
Token ID | 502 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
ados | +0.094 |
riages | +0.092 |
barrels | +0.089 |
pm | +0.088 |
drivers | +0.087 |
Neg logit contributions | |
kiss | -0.074 |
copy | -0.069 |
last | -0.063 |
dance | -0.062 |
blow | -0.062 |
Token | model |
Token ID | 2746 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
kiss | +0.034 |
copy | +0.031 |
last | +0.029 |
dance | +0.028 |
check | +0.028 |
Neg logit contributions | |
ados | -0.062 |
riages | -0.061 |
pm | -0.058 |
barrels | -0.058 |
drivers | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
ados | +0.005 |
pm | +0.005 |
riages | +0.005 |
barrels | +0.005 |
assing | +0.005 |
Neg logit contributions | |
kiss | -0.003 |
last | -0.003 |
copy | -0.003 |
dance | -0.003 |
watch | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | +8.90e-02 |
Pos logit contributions | |
kiss | +0.004 |
copy | +0.004 |
last | +0.004 |
dance | +0.004 |
blow | +0.004 |
Neg logit contributions | |
ados | -0.006 |
riages | -0.006 |
barrels | -0.006 |
pm | -0.006 |
drivers | -0.006 |
Token | was |
Token ID | 373 |
Feature activation | +4.17e-01 |
Pos logit contributions | |
ados | +0.004 |
barrels | +0.004 |
riages | +0.004 |
pm | +0.004 |
drivers | +0.004 |
Neg logit contributions | |
kiss | -0.002 |
copy | -0.002 |
last | -0.002 |
dance | -0.002 |
blow | -0.002 |
Token | on |
Token ID | 319 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
ados | +0.019 |
riages | +0.018 |
pm | +0.018 |
drivers | +0.018 |
barrels | +0.017 |
Neg logit contributions | |
kiss | -0.010 |
copy | -0.010 |
last | -0.009 |
blow | -0.009 |
dance | -0.009 |
Token | sale |
Token ID | 5466 |
Feature activation | +6.75e+00 |
Pos logit contributions | |
riages | +0.129 |
ados | +0.126 |
drivers | +0.125 |
ables | +0.122 |
pm | +0.122 |
Neg logit contributions | |
kiss | -0.047 |
last | -0.044 |
copy | -0.044 |
dance | -0.040 |
blow | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | +1.98e+00 |
Pos logit contributions | |
ados | +0.335 |
AW | +0.307 |
riages | +0.306 |
drivers | +0.305 |
bank | +0.303 |
Neg logit contributions | |
last | -0.141 |
kiss | -0.129 |
copy | -0.120 |
cl | -0.117 |
next | -0.115 |
Token | so |
Token ID | 523 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
ados | +0.089 |
riages | +0.087 |
drivers | +0.085 |
pm | +0.084 |
barrels | +0.084 |
Neg logit contributions | |
kiss | -0.052 |
copy | -0.046 |
dance | -0.043 |
watch | -0.042 |
blow | -0.042 |
Token | I |
Token ID | 314 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ados | +0.064 |
riages | +0.062 |
barrels | +0.060 |
drivers | +0.060 |
pm | +0.059 |
Neg logit contributions | |
kiss | -0.051 |
copy | -0.047 |
dance | -0.045 |
last | -0.045 |
blow | -0.043 |
Token | decided |
Token ID | 3066 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
kiss | +0.032 |
copy | +0.030 |
last | +0.027 |
massage | +0.027 |
dance | +0.027 |
Neg logit contributions | |
riages | -0.051 |
ados | -0.051 |
barrels | -0.050 |
pm | -0.049 |
drivers | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | +4.99e-02 |
Pos logit contributions | |
ados | +0.063 |
riages | +0.061 |
barrels | +0.060 |
pm | +0.060 |
drivers | +0.060 |
Neg logit contributions | |
kiss | -0.021 |
copy | -0.018 |
last | -0.017 |
check | -0.016 |
watch | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | +2.90e-01 |
Pos logit contributions | |
kiss | +0.012 |
copy | +0.012 |
dance | +0.010 |
massage | +0.010 |
last | +0.010 |
Neg logit contributions | |
ados | -0.017 |
riages | -0.017 |
barrels | -0.016 |
pm | -0.016 |
drivers | -0.016 |
Token | scared |
Token ID | 12008 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
ados | +0.009 |
tresp | +0.009 |
riages | +0.008 |
barrels | +0.008 |
ards | +0.008 |
Neg logit contributions | |
kiss | -0.011 |
copy | -0.010 |
dance | -0.009 |
dancer | -0.009 |
last | -0.009 |
Token | she |
Token ID | 673 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
ados | +0.141 |
riages | +0.139 |
barrels | +0.136 |
pm | +0.135 |
drivers | +0.133 |
Neg logit contributions | |
kiss | -0.076 |
copy | -0.068 |
last | -0.066 |
dance | -0.063 |
blow | -0.061 |
Token | would |
Token ID | 561 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
ados | +0.097 |
riages | +0.096 |
barrels | +0.092 |
pm | +0.091 |
drivers | +0.090 |
Neg logit contributions | |
kiss | -0.073 |
copy | -0.069 |
last | -0.063 |
dance | -0.062 |
blow | -0.061 |
Token | be |
Token ID | 307 |
Feature activation | -5.40e-01 |
Pos logit contributions | |
kiss | +0.046 |
copy | +0.042 |
last | +0.039 |
dance | +0.038 |
blow | +0.038 |
Neg logit contributions | |
ados | -0.074 |
riages | -0.073 |
barrels | -0.070 |
pm | -0.070 |
drivers | -0.069 |
Token | punished |
Token ID | 16851 |
Feature activation | +6.74e+00 |
Pos logit contributions | |
kiss | +0.018 |
copy | +0.017 |
dance | +0.016 |
last | +0.016 |
blow | +0.016 |
Neg logit contributions | |
ados | -0.019 |
riages | -0.018 |
barrels | -0.018 |
pm | -0.017 |
drivers | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | -2.12e-01 |
Pos logit contributions | |
ados | +0.322 |
riages | +0.318 |
barrels | +0.313 |
pm | +0.310 |
drivers | +0.300 |
Neg logit contributions | |
kiss | -0.147 |
copy | -0.131 |
last | -0.129 |
dance | -0.117 |
check | -0.114 |
Token | Her |
Token ID | 2332 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.006 |
last | +0.006 |
blow | +0.006 |
dance | +0.006 |
Neg logit contributions | |
ados | -0.008 |
riages | -0.008 |
barrels | -0.008 |
pm | -0.008 |
drivers | -0.007 |
Token | mom |
Token ID | 1995 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
copy | +0.031 |
kiss | +0.027 |
blow | +0.026 |
admire | +0.023 |
dance | +0.023 |
Neg logit contributions | |
ados | -0.059 |
barrels | -0.055 |
files | -0.055 |
peanuts | -0.054 |
pistols | -0.054 |
Token | my |
Token ID | 1820 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
copy | +0.058 |
kiss | +0.056 |
massage | +0.054 |
dance | +0.053 |
dancer | +0.053 |
Neg logit contributions | |
riages | -0.083 |
drivers | -0.082 |
ados | -0.080 |
barrels | -0.079 |
ards | -0.078 |
Token | took |
Token ID | 1718 |
Feature activation | -6.67e-01 |
Pos logit contributions | |
copy | +0.082 |
kiss | +0.081 |
dance | +0.073 |
massage | +0.070 |
admire | +0.068 |
Neg logit contributions | |
riages | -0.132 |
ados | -0.132 |
drivers | -0.130 |
barrels | -0.127 |
pm | -0.125 |
Token | She |
Token ID | 1375 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
ados | +0.007 |
riages | +0.007 |
barrels | +0.007 |
pm | +0.007 |
drivers | +0.007 |
Neg logit contributions | |
kiss | -0.006 |
copy | -0.005 |
dance | -0.005 |
last | -0.005 |
check | -0.005 |
Token | put |
Token ID | 1234 |
Feature activation | -2.83e+00 |
Pos logit contributions | |
barrels | +0.040 |
drivers | +0.040 |
cases | +0.040 |
riages | +0.039 |
ados | +0.038 |
Neg logit contributions | |
copy | -0.032 |
kiss | -0.029 |
last | -0.028 |
dancer | -0.028 |
dance | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
copy | +0.087 |
kiss | +0.087 |
dance | +0.080 |
watch | +0.075 |
dancer | +0.074 |
Neg logit contributions | |
ados | -0.108 |
drivers | -0.096 |
riages | -0.095 |
barrels | -0.093 |
ors | -0.089 |
Token | Band |
Token ID | 10243 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
copy | +0.003 |
kiss | +0.003 |
admire | +0.003 |
dance | +0.003 |
patiently | +0.003 |
Neg logit contributions | |
ados | -0.004 |
riages | -0.004 |
drivers | -0.004 |
dollars | -0.004 |
barrels | -0.004 |
Token | - |
Token ID | 12 |
Feature activation | -8.44e-01 |
Pos logit contributions | |
kiss | +0.062 |
copy | +0.059 |
last | +0.053 |
dance | +0.053 |
blow | +0.051 |
Neg logit contributions | |
ados | -0.080 |
riages | -0.079 |
drivers | -0.076 |
barrels | -0.075 |
pm | -0.074 |
Token | Aid |
Token ID | 44245 |
Feature activation | -7.38e+00 |
Pos logit contributions | |
copy | +0.025 |
kiss | +0.025 |
dance | +0.025 |
admire | +0.023 |
cl | +0.023 |
Neg logit contributions | |
ados | -0.030 |
riages | -0.028 |
drivers | -0.028 |
ages | -0.027 |
cigarettes | -0.027 |
Token | on |
Token ID | 319 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
kiss | +0.210 |
copy | +0.207 |
dance | +0.185 |
admire | +0.180 |
check | +0.176 |
Neg logit contributions | |
ados | -0.292 |
riages | -0.279 |
drivers | -0.266 |
ors | -0.263 |
barrels | -0.260 |
Token | Lily |
Token ID | 20037 |
Feature activation | -8.85e-01 |
Pos logit contributions | |
ados | +0.059 |
riages | +0.057 |
barrels | +0.056 |
pm | +0.055 |
drivers | +0.055 |
Neg logit contributions | |
kiss | -0.023 |
copy | -0.021 |
last | -0.018 |
dance | -0.017 |
blow | -0.017 |
Token | 's |
Token ID | 338 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
kiss | +0.033 |
copy | +0.032 |
last | +0.030 |
dance | +0.029 |
blow | +0.029 |
Neg logit contributions | |
ados | -0.027 |
riages | -0.027 |
barrels | -0.025 |
pm | -0.025 |
drivers | -0.025 |
Token | knee |
Token ID | 10329 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
kiss | +0.038 |
copy | +0.038 |
blow | +0.034 |
dance | +0.033 |
last | +0.032 |
Neg logit contributions | |
ados | -0.065 |
riages | -0.061 |
barrels | -0.061 |
drivers | -0.060 |
ables | -0.057 |
Token | and |
Token ID | 290 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
kiss | +0.052 |
copy | +0.049 |
last | +0.042 |
dance | +0.042 |
blow | +0.041 |
Neg logit contributions | |
ados | -0.103 |
riages | -0.102 |
barrels | -0.098 |
drivers | -0.097 |
pm | -0.096 |
Token | of |
Token ID | 286 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
ados | +0.090 |
barrels | +0.088 |
riages | +0.087 |
pm | +0.086 |
drivers | +0.086 |
Neg logit contributions | |
kiss | -0.028 |
last | -0.024 |
copy | -0.023 |
dance | -0.021 |
cl | -0.020 |
Token | adventures |
Token ID | 17545 |
Feature activation | -4.34e-01 |
Pos logit contributions | |
ados | +0.007 |
riages | +0.006 |
drivers | +0.006 |
barrels | +0.006 |
pm | +0.006 |
Neg logit contributions | |
kiss | -0.005 |
copy | -0.005 |
dance | -0.005 |
last | -0.005 |
massage | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -9.98e-01 |
Pos logit contributions | |
last | +0.008 |
kiss | +0.007 |
dance | +0.007 |
cl | +0.007 |
copy | +0.007 |
Neg logit contributions | |
ados | -0.022 |
pm | -0.021 |
AW | -0.021 |
barrels | -0.020 |
Sale | -0.020 |
Token | Every |
Token ID | 3887 |
Feature activation | -4.96e+00 |
Pos logit contributions | |
kiss | +0.026 |
copy | +0.024 |
last | +0.022 |
dance | +0.022 |
massage | +0.021 |
Neg logit contributions | |
ados | -0.043 |
riages | -0.043 |
barrels | -0.042 |
pm | -0.041 |
drivers | -0.040 |
Token | day |
Token ID | 1110 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
kiss | +0.106 |
copy | +0.101 |
blow | +0.082 |
check | +0.080 |
scrub | +0.080 |
Neg logit contributions | |
pm | -0.221 |
riages | -0.219 |
barrels | -0.212 |
ados | -0.210 |
bank | -0.209 |
Token | Will |
Token ID | 2561 |
Feature activation | -7.09e+00 |
Pos logit contributions | |
kiss | +0.053 |
copy | +0.049 |
last | +0.047 |
dance | +0.046 |
cl | +0.045 |
Neg logit contributions | |
ados | -0.087 |
barrels | -0.083 |
riages | -0.081 |
pm | -0.079 |
assing | -0.078 |
Token | was |
Token ID | 373 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
kiss | +0.154 |
copy | +0.142 |
dance | +0.129 |
last | +0.128 |
blow | +0.126 |
Neg logit contributions | |
ados | -0.344 |
riages | -0.329 |
barrels | -0.326 |
drivers | -0.320 |
pm | -0.317 |
Token | sure |
Token ID | 1654 |
Feature activation | -2.09e+00 |
Pos logit contributions | |
kiss | +0.042 |
copy | +0.038 |
massage | +0.034 |
last | +0.033 |
blow | +0.033 |
Neg logit contributions | |
riages | -0.060 |
ados | -0.056 |
barrels | -0.056 |
pm | -0.056 |
drivers | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
kiss | +0.048 |
copy | +0.045 |
last | +0.043 |
dance | +0.041 |
check | +0.040 |
Neg logit contributions | |
ados | -0.098 |
riages | -0.095 |
barrels | -0.091 |
pm | -0.091 |
drivers | -0.090 |
Token | use |
Token ID | 779 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
kiss | +0.000 |
copy | +0.000 |
dab | +0.000 |
last | +0.000 |
massage | +0.000 |
Neg logit contributions | |
riages | -0.001 |
barrels | -0.001 |
ados | -0.001 |
pm | -0.001 |
bank | -0.001 |
Token | his |
Token ID | 465 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
kiss | +0.041 |
copy | +0.038 |
last | +0.035 |
dance | +0.034 |
massage | +0.033 |
Neg logit contributions | |
ados | -0.078 |
riages | -0.075 |
pm | -0.073 |
barrels | -0.072 |
drivers | -0.072 |
Token | and |
Token ID | 290 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
kiss | +0.028 |
copy | +0.026 |
last | +0.024 |
dance | +0.023 |
blow | +0.022 |
Neg logit contributions | |
ados | -0.053 |
riages | -0.052 |
pm | -0.051 |
drivers | -0.051 |
barrels | -0.051 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
kiss | +0.067 |
copy | +0.061 |
last | +0.055 |
dance | +0.054 |
blow | +0.053 |
Neg logit contributions | |
ados | -0.123 |
riages | -0.121 |
barrels | -0.117 |
pm | -0.116 |
drivers | -0.115 |
Token | to |
Token ID | 284 |
Feature activation | +1.97e-01 |
Pos logit contributions | |
kiss | +0.040 |
copy | +0.037 |
last | +0.032 |
blow | +0.032 |
dance | +0.032 |
Neg logit contributions | |
ados | -0.080 |
riages | -0.078 |
barrels | -0.077 |
drivers | -0.075 |
pm | -0.075 |
Token | clean |
Token ID | 3424 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
ados | +0.009 |
riages | +0.009 |
barrels | +0.009 |
bank | +0.008 |
pm | +0.008 |
Neg logit contributions | |
kiss | -0.004 |
copy | -0.004 |
last | -0.003 |
massage | -0.003 |
dancer | -0.003 |
Token | the |
Token ID | 262 |
Feature activation | -8.03e-01 |
Pos logit contributions | |
riages | +0.018 |
pm | +0.018 |
ados | +0.017 |
barrels | +0.017 |
drivers | +0.017 |
Neg logit contributions | |
kiss | -0.006 |
copy | -0.006 |
last | -0.005 |
scrub | -0.005 |
dance | -0.005 |
Token | mark |
Token ID | 1317 |
Feature activation | -6.90e+00 |
Pos logit contributions | |
kiss | +0.025 |
copy | +0.024 |
last | +0.022 |
dance | +0.022 |
blow | +0.021 |
Neg logit contributions | |
ados | -0.030 |
riages | -0.030 |
barrels | -0.029 |
pm | -0.028 |
drivers | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
kiss | +0.167 |
copy | +0.154 |
last | +0.144 |
dance | +0.137 |
blow | +0.135 |
Neg logit contributions | |
ados | -0.311 |
riages | -0.304 |
barrels | -0.298 |
pm | -0.296 |
drivers | -0.289 |
Token | but |
Token ID | 475 |
Feature activation | -3.44e-01 |
Pos logit contributions | |
ados | +0.008 |
riages | +0.008 |
pm | +0.008 |
barrels | +0.008 |
drivers | +0.008 |
Neg logit contributions | |
kiss | -0.007 |
copy | -0.007 |
scrub | -0.006 |
blow | -0.006 |
last | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
kiss | +0.010 |
copy | +0.009 |
last | +0.008 |
dance | +0.008 |
blow | +0.008 |
Neg logit contributions | |
ados | -0.014 |
riages | -0.014 |
barrels | -0.014 |
pm | -0.014 |
drivers | -0.013 |
Token | didn |
Token ID | 1422 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
kiss | +0.001 |
copy | +0.001 |
check | +0.001 |
last | +0.001 |
dance | +0.001 |
Neg logit contributions | |
ados | -0.002 |
riages | -0.002 |
barrels | -0.002 |
drivers | -0.002 |
pm | -0.001 |
Token | 't |
Token ID | 470 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ados | +0.060 |
riages | +0.059 |
barrels | +0.058 |
pm | +0.056 |
drivers | +0.054 |
Neg logit contributions | |
kiss | -0.060 |
copy | -0.056 |
last | -0.055 |
blow | -0.053 |
dance | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
kiss | +0.082 |
copy | +0.071 |
dance | +0.069 |
last | +0.068 |
check | +0.066 |
Neg logit contributions | |
ados | -0.163 |
riages | -0.160 |
pm | -0.158 |
barrels | -0.156 |
drivers | -0.156 |
Token | Whenever |
Token ID | 21326 |
Feature activation | -5.32e+00 |
Pos logit contributions | |
kiss | +0.054 |
copy | +0.048 |
last | +0.045 |
dance | +0.044 |
massage | +0.043 |
Neg logit contributions | |
ados | -0.088 |
riages | -0.087 |
barrels | -0.085 |
drivers | -0.084 |
pm | -0.083 |
Token | Sarah |
Token ID | 10490 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
kiss | +0.102 |
dance | +0.089 |
copy | +0.084 |
last | +0.083 |
massage | +0.080 |
Neg logit contributions | |
riages | -0.275 |
drivers | -0.274 |
ados | -0.271 |
barrels | -0.268 |
ors | -0.258 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
kiss | +0.029 |
copy | +0.027 |
last | +0.024 |
blow | +0.023 |
dance | +0.023 |
Neg logit contributions | |
riages | -0.049 |
ados | -0.049 |
barrels | -0.047 |
pm | -0.047 |
ors | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
kiss | +0.038 |
copy | +0.035 |
last | +0.032 |
dance | +0.031 |
blow | +0.030 |
Neg logit contributions | |
ados | -0.065 |
riages | -0.064 |
barrels | -0.062 |
pm | -0.061 |
drivers | -0.061 |
Token | mark |
Token ID | 1317 |
Feature activation | -6.88e+00 |
Pos logit contributions | |
kiss | +0.023 |
copy | +0.021 |
blow | +0.019 |
last | +0.019 |
dance | +0.019 |
Neg logit contributions | |
ados | -0.035 |
riages | -0.034 |
barrels | -0.033 |
pm | -0.032 |
drivers | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | -4.17e-01 |
Pos logit contributions | |
kiss | +0.186 |
copy | +0.172 |
last | +0.157 |
dance | +0.154 |
blow | +0.153 |
Neg logit contributions | |
ados | -0.289 |
riages | -0.283 |
barrels | -0.275 |
pm | -0.272 |
drivers | -0.269 |
Token | she |
Token ID | 673 |
Feature activation | +8.42e-01 |
Pos logit contributions | |
kiss | +0.009 |
copy | +0.009 |
last | +0.008 |
dance | +0.008 |
blow | +0.007 |
Neg logit contributions | |
ados | -0.019 |
riages | -0.019 |
pm | -0.019 |
barrels | -0.018 |
drivers | -0.018 |
Token | remembered |
Token ID | 12086 |
Feature activation | -5.61e-01 |
Pos logit contributions | |
riages | +0.034 |
ados | +0.032 |
barrels | +0.032 |
pm | +0.031 |
bank | +0.031 |
Neg logit contributions | |
kiss | -0.022 |
copy | -0.020 |
massage | -0.019 |
last | -0.019 |
blow | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -1.96e-01 |
Pos logit contributions | |
kiss | +0.014 |
copy | +0.012 |
last | +0.011 |
dance | +0.011 |
blow | +0.011 |
Neg logit contributions | |
ados | -0.025 |
riages | -0.025 |
barrels | -0.024 |
pm | -0.024 |
drivers | -0.024 |
Token | moment |
Token ID | 2589 |
Feature activation | -3.02e+00 |
Pos logit contributions | |
kiss | +0.005 |
copy | +0.005 |
blow | +0.004 |
dance | +0.004 |
scrub | +0.004 |
Neg logit contributions | |
barrels | -0.008 |
ados | -0.007 |
riages | -0.007 |
pm | -0.007 |
files | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
kiss | +0.037 |
mom | +0.031 |
last | +0.031 |
Mom | +0.031 |
mum | +0.030 |
Neg logit contributions | |
drivers | -0.074 |
assing | -0.072 |
ados | -0.072 |
riages | -0.071 |
pm | -0.071 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
kiss | +0.038 |
copy | +0.033 |
dance | +0.032 |
last | +0.031 |
watch | +0.030 |
Neg logit contributions | |
ados | -0.066 |
drivers | -0.065 |
riages | -0.065 |
pm | -0.064 |
ors | -0.063 |
Token | my |
Token ID | 1820 |
Feature activation | -5.06e+00 |
Pos logit contributions | |
kiss | +0.035 |
copy | +0.033 |
massage | +0.028 |
dance | +0.027 |
last | +0.027 |
Neg logit contributions | |
riages | -0.081 |
ados | -0.081 |
barrels | -0.079 |
pm | -0.077 |
drivers | -0.076 |
Token | showed |
Token ID | 3751 |
Feature activation | -6.52e-01 |
Pos logit contributions | |
kiss | +0.108 |
copy | +0.098 |
last | +0.087 |
dance | +0.085 |
blow | +0.084 |
Neg logit contributions | |
ados | -0.244 |
riages | -0.239 |
barrels | -0.233 |
pm | -0.231 |
drivers | -0.228 |
Token | Billy |
Token ID | 15890 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
kiss | +0.019 |
copy | +0.018 |
dance | +0.016 |
blow | +0.016 |
last | +0.016 |
Neg logit contributions | |
ados | -0.027 |
riages | -0.025 |
barrels | -0.025 |
drivers | -0.024 |
ors | -0.024 |
Token | how |
Token ID | 703 |
Feature activation | -6.68e+00 |
Pos logit contributions | |
kiss | +0.040 |
copy | +0.035 |
last | +0.034 |
dance | +0.031 |
massage | +0.030 |
Neg logit contributions | |
riages | -0.106 |
ados | -0.104 |
pm | -0.103 |
drivers | -0.102 |
barrels | -0.099 |
Token | to |
Token ID | 284 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
kiss | +0.109 |
copy | +0.095 |
last | +0.087 |
dance | +0.083 |
check | +0.081 |
Neg logit contributions | |
ados | -0.357 |
riages | -0.349 |
pm | -0.345 |
drivers | -0.340 |
barrels | -0.337 |
Token | play |
Token ID | 711 |
Feature activation | -4.65e+00 |
Pos logit contributions | |
kiss | +0.033 |
copy | +0.029 |
last | +0.026 |
massage | +0.025 |
scrub | +0.024 |
Neg logit contributions | |
riages | -0.086 |
ados | -0.086 |
barrels | -0.084 |
pm | -0.081 |
drivers | -0.081 |
Token | with |
Token ID | 351 |
Feature activation | -6.74e-01 |
Pos logit contributions | |
kiss | +0.100 |
copy | +0.096 |
dance | +0.092 |
cl | +0.086 |
scrub | +0.082 |
Neg logit contributions | |
ados | -0.223 |
riages | -0.218 |
pm | -0.215 |
barrels | -0.209 |
drivers | -0.208 |
Token | the |
Token ID | 262 |
Feature activation | -2.55e-03 |
Pos logit contributions | |
kiss | +0.015 |
copy | +0.014 |
dance | +0.013 |
last | +0.013 |
comfort | +0.013 |
Neg logit contributions | |
pm | -0.032 |
ados | -0.030 |
riages | -0.030 |
ables | -0.030 |
bank | -0.029 |
Token | cars |
Token ID | 5006 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
kiss | +0.000 |
copy | +0.000 |
last | +0.000 |
dance | +0.000 |
blow | +0.000 |
Neg logit contributions | |
ados | -0.000 |
riages | -0.000 |
pm | -0.000 |
ors | -0.000 |
drivers | -0.000 |
Token | tray |
Token ID | 26473 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
kiss | +0.061 |
copy | +0.056 |
last | +0.055 |
dance | +0.053 |
blow | +0.052 |
Neg logit contributions | |
ados | -0.076 |
riages | -0.074 |
pm | -0.071 |
drivers | -0.071 |
ors | -0.070 |
Token | ." |
Token ID | 526 |
Feature activation | -1.83e+00 |
Pos logit contributions | |
ados | +0.071 |
riages | +0.070 |
drivers | +0.068 |
pm | +0.068 |
barrels | +0.068 |
Neg logit contributions | |
kiss | -0.030 |
last | -0.026 |
copy | -0.024 |
dance | -0.023 |
cl | -0.023 |
Token | She |
Token ID | 1375 |
Feature activation | +6.93e-01 |
Pos logit contributions | |
kiss | +0.038 |
copy | +0.031 |
dance | +0.027 |
cl | +0.027 |
last | +0.027 |
Neg logit contributions | |
ados | -0.092 |
riages | -0.091 |
barrels | -0.089 |
drivers | -0.088 |
pm | -0.087 |
Token | showed |
Token ID | 3751 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
ados | +0.031 |
riages | +0.031 |
barrels | +0.030 |
pm | +0.030 |
drivers | +0.029 |
Neg logit contributions | |
kiss | -0.016 |
copy | -0.015 |
last | -0.014 |
dance | -0.013 |
massage | -0.013 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
kiss | +0.003 |
copy | +0.003 |
dance | +0.003 |
blow | +0.002 |
last | +0.002 |
Neg logit contributions | |
ados | -0.005 |
riages | -0.005 |
barrels | -0.005 |
ors | -0.005 |
drivers | -0.005 |
Token | how |
Token ID | 703 |
Feature activation | -6.68e+00 |
Pos logit contributions | |
kiss | +0.059 |
copy | +0.054 |
last | +0.048 |
dance | +0.047 |
blow | +0.047 |
Neg logit contributions | |
ados | -0.114 |
riages | -0.112 |
barrels | -0.109 |
pm | -0.108 |
drivers | -0.107 |
Token | to |
Token ID | 284 |
Feature activation | -1.38e+00 |
Pos logit contributions | |
kiss | +0.162 |
copy | +0.151 |
last | +0.132 |
dance | +0.131 |
blow | +0.131 |
Neg logit contributions | |
ados | -0.300 |
riages | -0.291 |
barrels | -0.285 |
pm | -0.280 |
drivers | -0.277 |
Token | roll |
Token ID | 4836 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
kiss | +0.023 |
copy | +0.021 |
last | +0.019 |
dancer | +0.017 |
massage | +0.017 |
Neg logit contributions | |
riages | -0.067 |
ados | -0.066 |
barrels | -0.066 |
pm | -0.064 |
drivers | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
kiss | +0.074 |
copy | +0.067 |
last | +0.062 |
dance | +0.059 |
blow | +0.058 |
Neg logit contributions | |
ados | -0.153 |
riages | -0.152 |
pm | -0.148 |
barrels | -0.147 |
drivers | -0.144 |
Token | mix |
Token ID | 5022 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
kiss | +0.029 |
copy | +0.026 |
last | +0.026 |
dance | +0.024 |
blow | +0.024 |
Neg logit contributions | |
riages | -0.046 |
ados | -0.045 |
pm | -0.044 |
bank | -0.044 |
drivers | -0.043 |
Token | with |
Token ID | 351 |
Feature activation | +4.34e-01 |
Pos logit contributions | |
kiss | +0.062 |
copy | +0.058 |
last | +0.053 |
dance | +0.051 |
blow | +0.050 |
Neg logit contributions | |
ados | -0.106 |
riages | -0.105 |
barrels | -0.102 |
pm | -0.101 |
drivers | -0.100 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
kiss | +0.010 |
copy | +0.009 |
last | +0.008 |
blow | +0.008 |
dance | +0.008 |
Neg logit contributions | |
ados | -0.013 |
riages | -0.013 |
barrels | -0.013 |
pm | -0.013 |
drivers | -0.012 |
Token | Tim |
Token ID | 5045 |
Feature activation | -6.58e-01 |
Pos logit contributions | |
kiss | +0.052 |
copy | +0.043 |
cl | +0.042 |
comfort | +0.041 |
check | +0.041 |
Neg logit contributions | |
pm | -0.077 |
riages | -0.077 |
ados | -0.076 |
barrels | -0.076 |
pounds | -0.075 |
Token | my |
Token ID | 1820 |
Feature activation | -4.39e+00 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.012 |
dance | +0.010 |
last | +0.010 |
massage | +0.010 |
Neg logit contributions | |
ados | -0.031 |
riages | -0.031 |
barrels | -0.030 |
pm | -0.030 |
drivers | -0.029 |
Token | showed |
Token ID | 3751 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
kiss | +0.104 |
copy | +0.095 |
last | +0.085 |
dance | +0.084 |
blow | +0.083 |
Neg logit contributions | |
ados | -0.201 |
riages | -0.197 |
barrels | -0.191 |
pm | -0.190 |
drivers | -0.188 |
Token | Billy |
Token ID | 15890 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.007 |
last | +0.007 |
dance | +0.006 |
blow | +0.006 |
Neg logit contributions | |
ados | -0.007 |
riages | -0.007 |
barrels | -0.007 |
pm | -0.007 |
drivers | -0.007 |
Token | how |
Token ID | 703 |
Feature activation | -6.67e+00 |
Pos logit contributions | |
kiss | +0.046 |
copy | +0.041 |
last | +0.038 |
dance | +0.036 |
massage | +0.035 |
Neg logit contributions | |
riages | -0.098 |
ados | -0.097 |
pm | -0.095 |
drivers | -0.094 |
barrels | -0.094 |
Token | to |
Token ID | 284 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
kiss | +0.125 |
copy | +0.109 |
last | +0.101 |
dance | +0.096 |
blow | +0.093 |
Neg logit contributions | |
ados | -0.339 |
riages | -0.335 |
pm | -0.331 |
barrels | -0.326 |
drivers | -0.322 |
Token | hold |
Token ID | 1745 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
kiss | +0.033 |
copy | +0.030 |
last | +0.026 |
massage | +0.025 |
dance | +0.024 |
Neg logit contributions | |
ados | -0.085 |
riages | -0.084 |
barrels | -0.082 |
drivers | -0.080 |
pm | -0.080 |
Token | the |
Token ID | 262 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
kiss | +0.102 |
copy | +0.091 |
last | +0.086 |
blow | +0.081 |
massage | +0.080 |
Neg logit contributions | |
riages | -0.190 |
ados | -0.189 |
pm | -0.189 |
barrels | -0.182 |
drivers | -0.178 |
Token | arrow |
Token ID | 15452 |
Feature activation | -3.10e+00 |
Pos logit contributions | |
ados | +0.009 |
riages | +0.008 |
pm | +0.008 |
barrels | +0.008 |
drivers | +0.008 |
Neg logit contributions | |
kiss | -0.007 |
copy | -0.006 |
last | -0.006 |
dance | -0.006 |
massage | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
kiss | +0.065 |
copy | +0.058 |
last | +0.056 |
dance | +0.051 |
blow | +0.050 |
Neg logit contributions | |
ados | -0.150 |
pm | -0.150 |
riages | -0.148 |
barrels | -0.146 |
drivers | -0.143 |
Token | Then |
Token ID | 6423 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
copy | +0.014 |
kiss | +0.013 |
comfort | +0.011 |
last | +0.011 |
scrub | +0.011 |
Neg logit contributions | |
ados | -0.032 |
riages | -0.030 |
pm | -0.030 |
barrels | -0.029 |
drivers | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
kiss | +0.065 |
copy | +0.061 |
last | +0.059 |
dance | +0.057 |
watch | +0.055 |
Neg logit contributions | |
ados | -0.110 |
pm | -0.108 |
riages | -0.107 |
drivers | -0.106 |
ors | -0.105 |
Token | Jack |
Token ID | 3619 |
Feature activation | -9.37e-01 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.012 |
last | +0.011 |
dance | +0.010 |
blow | +0.010 |
Neg logit contributions | |
ados | -0.020 |
riages | -0.019 |
barrels | -0.019 |
pm | -0.018 |
drivers | -0.018 |
Token | showed |
Token ID | 3751 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
kiss | +0.025 |
copy | +0.023 |
last | +0.021 |
dance | +0.020 |
massage | +0.020 |
Neg logit contributions | |
ados | -0.040 |
riages | -0.039 |
barrels | -0.038 |
pm | -0.038 |
drivers | -0.037 |
Token | Sam |
Token ID | 3409 |
Feature activation | -1.75e+00 |
Pos logit contributions | |
kiss | +0.021 |
copy | +0.020 |
massage | +0.019 |
blow | +0.019 |
dance | +0.019 |
Neg logit contributions | |
ados | -0.021 |
riages | -0.020 |
barrels | -0.019 |
ors | -0.018 |
drivers | -0.018 |
Token | how |
Token ID | 703 |
Feature activation | -6.63e+00 |
Pos logit contributions | |
kiss | +0.043 |
copy | +0.040 |
last | +0.036 |
dance | +0.035 |
blow | +0.035 |
Neg logit contributions | |
ados | -0.078 |
riages | -0.077 |
barrels | -0.074 |
pm | -0.074 |
drivers | -0.073 |
Token | to |
Token ID | 284 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
kiss | +0.159 |
copy | +0.149 |
dance | +0.129 |
last | +0.128 |
blow | +0.128 |
Neg logit contributions | |
ados | -0.297 |
riages | -0.292 |
barrels | -0.285 |
drivers | -0.278 |
pm | -0.278 |
Token | look |
Token ID | 804 |
Feature activation | -2.71e+00 |
Pos logit contributions | |
dancer | +0.035 |
kiss | +0.034 |
happier | +0.033 |
Sweet | +0.032 |
night | +0.031 |
Neg logit contributions | |
riages | -0.066 |
assing | -0.059 |
barrels | -0.059 |
ors | -0.059 |
drivers | -0.058 |
Token | carefully |
Token ID | 7773 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
kiss | +0.063 |
copy | +0.058 |
last | +0.053 |
dance | +0.050 |
check | +0.049 |
Neg logit contributions | |
ados | -0.126 |
riages | -0.123 |
pm | -0.121 |
barrels | -0.119 |
drivers | -0.119 |
Token | before |
Token ID | 878 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
kiss | +0.046 |
copy | +0.042 |
last | +0.038 |
dance | +0.037 |
blow | +0.037 |
Neg logit contributions | |
ados | -0.082 |
riages | -0.081 |
barrels | -0.079 |
pm | -0.077 |
drivers | -0.077 |
Token | doing |
Token ID | 1804 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
kiss | +0.029 |
copy | +0.027 |
dance | +0.026 |
last | +0.026 |
blow | +0.025 |
Neg logit contributions | |
ados | -0.042 |
riages | -0.041 |
pm | -0.040 |
barrels | -0.040 |
ables | -0.039 |
Token | She |
Token ID | 1375 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
kiss | +0.062 |
Mom | +0.052 |
copy | +0.052 |
dance | +0.051 |
cl | +0.050 |
Neg logit contributions | |
ados | -0.095 |
riages | -0.093 |
barrels | -0.093 |
drivers | -0.091 |
pm | -0.089 |
Token | showed |
Token ID | 3751 |
Feature activation | -6.89e-01 |
Pos logit contributions | |
kiss | +0.008 |
copy | +0.008 |
last | +0.007 |
dance | +0.007 |
massage | +0.006 |
Neg logit contributions | |
ados | -0.016 |
riages | -0.016 |
barrels | -0.015 |
pm | -0.015 |
drivers | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
kiss | +0.018 |
copy | +0.016 |
last | +0.015 |
dance | +0.014 |
blow | +0.014 |
Neg logit contributions | |
ados | -0.030 |
riages | -0.029 |
barrels | -0.028 |
pm | -0.028 |
drivers | -0.028 |
Token | little |
Token ID | 1310 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
kiss | +0.065 |
copy | +0.059 |
last | +0.055 |
dance | +0.054 |
massage | +0.053 |
Neg logit contributions | |
ados | -0.090 |
riages | -0.089 |
pm | -0.085 |
drivers | -0.084 |
barrels | -0.083 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.80e+00 |
Pos logit contributions | |
kiss | +0.108 |
copy | +0.093 |
last | +0.085 |
dance | +0.085 |
massage | +0.082 |
Neg logit contributions | |
riages | -0.245 |
ados | -0.242 |
pm | -0.235 |
drivers | -0.232 |
barrels | -0.231 |
Token | how |
Token ID | 703 |
Feature activation | -6.62e+00 |
Pos logit contributions | |
kiss | +0.046 |
copy | +0.037 |
dance | +0.034 |
last | +0.033 |
comfort | +0.032 |
Neg logit contributions | |
riages | -0.153 |
drivers | -0.151 |
ados | -0.150 |
pm | -0.147 |
AW | -0.146 |
Token | to |
Token ID | 284 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
kiss | +0.128 |
copy | +0.111 |
last | +0.102 |
dance | +0.099 |
check | +0.095 |
Neg logit contributions | |
ados | -0.335 |
riages | -0.332 |
pm | -0.322 |
barrels | -0.321 |
drivers | -0.320 |
Token | put |
Token ID | 1234 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
kiss | +0.023 |
copy | +0.021 |
last | +0.019 |
massage | +0.018 |
dance | +0.017 |
Neg logit contributions | |
riages | -0.066 |
ados | -0.066 |
barrels | -0.064 |
drivers | -0.063 |
pm | -0.062 |
Token | the |
Token ID | 262 |
Feature activation | -7.18e-01 |
Pos logit contributions | |
kiss | +0.077 |
copy | +0.068 |
last | +0.064 |
dance | +0.060 |
check | +0.059 |
Neg logit contributions | |
riages | -0.175 |
ados | -0.174 |
pm | -0.173 |
barrels | -0.167 |
drivers | -0.167 |
Token | tiny |
Token ID | 7009 |
Feature activation | -3.58e+00 |
Pos logit contributions | |
kiss | +0.021 |
copy | +0.019 |
last | +0.018 |
dance | +0.018 |
massage | +0.017 |
Neg logit contributions | |
ados | -0.029 |
riages | -0.029 |
pm | -0.028 |
drivers | -0.028 |
barrels | -0.028 |
Token | socks |
Token ID | 24359 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
kiss | +0.100 |
copy | +0.089 |
dance | +0.084 |
last | +0.083 |
massage | +0.081 |
Neg logit contributions | |
riages | -0.157 |
ados | -0.149 |
pm | -0.149 |
drivers | -0.148 |
bank | -0.143 |
Token | , |
Token ID | 11 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
kiss | +0.087 |
last | +0.085 |
dance | +0.079 |
night | +0.079 |
copy | +0.075 |
Neg logit contributions | |
ados | -0.181 |
riages | -0.180 |
drivers | -0.176 |
barrels | -0.174 |
ors | -0.172 |
Token | the |
Token ID | 262 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
kiss | +0.041 |
copy | +0.038 |
last | +0.034 |
massage | +0.033 |
blow | +0.033 |
Neg logit contributions | |
ados | -0.062 |
riages | -0.059 |
barrels | -0.058 |
pm | -0.057 |
drivers | -0.055 |
Token | dad |
Token ID | 9955 |
Feature activation | -3.09e+00 |
Pos logit contributions | |
kiss | +0.005 |
copy | +0.005 |
idget | +0.005 |
stroke | +0.005 |
scrub | +0.005 |
Neg logit contributions | |
files | -0.006 |
sacks | -0.005 |
barrels | -0.005 |
wallets | -0.005 |
ados | -0.005 |
Token | showed |
Token ID | 3751 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
kiss | +0.073 |
copy | +0.068 |
massage | +0.061 |
last | +0.060 |
dance | +0.059 |
Neg logit contributions | |
riages | -0.136 |
ados | -0.135 |
barrels | -0.131 |
drivers | -0.129 |
pm | -0.128 |
Token | her |
Token ID | 607 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
kiss | +0.009 |
copy | +0.008 |
last | +0.007 |
dance | +0.007 |
massage | +0.007 |
Neg logit contributions | |
riages | -0.024 |
ados | -0.024 |
barrels | -0.023 |
pm | -0.023 |
drivers | -0.023 |
Token | how |
Token ID | 703 |
Feature activation | -6.61e+00 |
Pos logit contributions | |
kiss | +0.049 |
copy | +0.046 |
last | +0.041 |
dance | +0.040 |
blow | +0.040 |
Neg logit contributions | |
ados | -0.081 |
riages | -0.079 |
barrels | -0.077 |
pm | -0.076 |
drivers | -0.075 |
Token | to |
Token ID | 284 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
kiss | +0.162 |
copy | +0.144 |
last | +0.129 |
massage | +0.127 |
blow | +0.127 |
Neg logit contributions | |
ados | -0.298 |
riages | -0.287 |
barrels | -0.285 |
pm | -0.278 |
drivers | -0.274 |
Token | make |
Token ID | 787 |
Feature activation | -2.55e+00 |
Pos logit contributions | |
kiss | +0.026 |
Sweet | +0.023 |
massage | +0.023 |
dancer | +0.023 |
listener | +0.022 |
Neg logit contributions | |
riages | -0.051 |
barrels | -0.049 |
ados | -0.049 |
banks | -0.048 |
drivers | -0.047 |
Token | a |
Token ID | 257 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
kiss | +0.066 |
copy | +0.060 |
last | +0.055 |
dance | +0.053 |
massage | +0.053 |
Neg logit contributions | |
ados | -0.111 |
riages | -0.108 |
barrels | -0.106 |
pm | -0.105 |
drivers | -0.103 |
Token | pretty |
Token ID | 2495 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
kiss | +0.037 |
copy | +0.033 |
last | +0.033 |
massage | +0.032 |
check | +0.032 |
Neg logit contributions | |
ados | -0.053 |
riages | -0.052 |
barrels | -0.052 |
files | -0.050 |
pm | -0.049 |
Token | paper |
Token ID | 3348 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
kiss | +0.054 |
copy | +0.049 |
last | +0.048 |
check | +0.048 |
massage | +0.047 |
Neg logit contributions | |
ados | -0.076 |
riages | -0.073 |
barrels | -0.072 |
pm | -0.069 |
files | -0.068 |
Token | top |
Token ID | 1353 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
kiss | +0.109 |
comfort | +0.099 |
massage | +0.097 |
scrub | +0.096 |
copy | +0.096 |
Neg logit contributions | |
ados | -0.131 |
barrels | -0.122 |
structures | -0.115 |
riages | -0.112 |
pounds | -0.111 |
Token | . |
Token ID | 13 |
Feature activation | -4.60e-01 |
Pos logit contributions | |
kiss | +0.070 |
copy | +0.066 |
massage | +0.064 |
check | +0.061 |
comfort | +0.061 |
Neg logit contributions | |
riages | -0.071 |
ados | -0.070 |
barrels | -0.067 |
pm | -0.066 |
drivers | -0.063 |
Token | He |
Token ID | 679 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
kiss | +0.017 |
copy | +0.015 |
massage | +0.014 |
last | +0.014 |
dance | +0.014 |
Neg logit contributions | |
riages | -0.015 |
ados | -0.015 |
barrels | -0.015 |
pm | -0.014 |
drivers | -0.014 |
Token | shows |
Token ID | 2523 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
kiss | +0.004 |
copy | +0.003 |
last | +0.003 |
check | +0.003 |
dab | +0.003 |
Neg logit contributions | |
ados | -0.004 |
riages | -0.004 |
barrels | -0.004 |
drivers | -0.004 |
pm | -0.004 |
Token | her |
Token ID | 607 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
kiss | +0.004 |
copy | +0.003 |
dance | +0.003 |
last | +0.003 |
blow | +0.003 |
Neg logit contributions | |
ados | -0.009 |
riages | -0.008 |
barrels | -0.008 |
pm | -0.008 |
drivers | -0.008 |
Token | how |
Token ID | 703 |
Feature activation | -6.59e+00 |
Pos logit contributions | |
kiss | +0.057 |
copy | +0.053 |
check | +0.048 |
blow | +0.048 |
massage | +0.048 |
Neg logit contributions | |
ados | -0.082 |
riages | -0.078 |
barrels | -0.077 |
drivers | -0.074 |
pm | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
kiss | +0.164 |
copy | +0.152 |
blow | +0.132 |
last | +0.132 |
massage | +0.131 |
Neg logit contributions | |
ados | -0.289 |
riages | -0.281 |
barrels | -0.276 |
pm | -0.271 |
drivers | -0.266 |
Token | hold |
Token ID | 1745 |
Feature activation | -4.11e+00 |
Pos logit contributions | |
Sweet | +0.024 |
night | +0.024 |
reaction | +0.022 |
listener | +0.022 |
Aid | +0.021 |
Neg logit contributions | |
riages | -0.042 |
clubs | -0.040 |
ors | -0.039 |
assing | -0.039 |
hips | -0.039 |
Token | it |
Token ID | 340 |
Feature activation | -4.62e-01 |
Pos logit contributions | |
kiss | +0.110 |
copy | +0.102 |
dance | +0.092 |
massage | +0.091 |
blow | +0.091 |
Neg logit contributions | |
ados | -0.174 |
riages | -0.166 |
barrels | -0.164 |
drivers | -0.160 |
pm | -0.160 |
Token | and |
Token ID | 290 |
Feature activation | -7.25e-01 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.012 |
massage | +0.011 |
blow | +0.011 |
check | +0.011 |
Neg logit contributions | |
ados | -0.019 |
riages | -0.018 |
barrels | -0.018 |
drivers | -0.017 |
pm | -0.017 |
Token | twist |
Token ID | 14528 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.012 |
dab | +0.012 |
last | +0.012 |
listener | +0.011 |
Neg logit contributions | |
riages | -0.032 |
ados | -0.031 |
barrels | -0.030 |
drivers | -0.030 |
bullets | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
kiss | +0.065 |
copy | +0.060 |
last | +0.052 |
dance | +0.052 |
check | +0.051 |
Neg logit contributions | |
ados | -0.139 |
riages | -0.135 |
barrels | -0.132 |
drivers | -0.130 |
pm | -0.128 |
Token | Mom |
Token ID | 29252 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
kiss | +0.055 |
copy | +0.052 |
dance | +0.045 |
check | +0.044 |
massage | +0.044 |
Neg logit contributions | |
ados | -0.103 |
riages | -0.098 |
barrels | -0.096 |
drivers | -0.095 |
ors | -0.094 |
Token | my |
Token ID | 1820 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
copy | +0.070 |
kiss | +0.070 |
dance | +0.061 |
massage | +0.060 |
last | +0.058 |
Neg logit contributions | |
ados | -0.139 |
riages | -0.137 |
barrels | -0.135 |
drivers | -0.134 |
pm | -0.133 |
Token | showed |
Token ID | 3751 |
Feature activation | -3.33e-01 |
Pos logit contributions | |
kiss | +0.077 |
copy | +0.072 |
last | +0.063 |
dance | +0.062 |
massage | +0.061 |
Neg logit contributions | |
ados | -0.162 |
riages | -0.159 |
barrels | -0.155 |
pm | -0.154 |
drivers | -0.153 |
Token | Sammy |
Token ID | 44067 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
kiss | +0.009 |
copy | +0.008 |
last | +0.008 |
massage | +0.007 |
dance | +0.007 |
Neg logit contributions | |
riages | -0.014 |
pm | -0.014 |
ados | -0.014 |
barrels | -0.014 |
drivers | -0.013 |
Token | how |
Token ID | 703 |
Feature activation | -6.58e+00 |
Pos logit contributions | |
kiss | +0.048 |
copy | +0.044 |
last | +0.040 |
dance | +0.038 |
blow | +0.038 |
Neg logit contributions | |
ados | -0.105 |
riages | -0.105 |
pm | -0.101 |
barrels | -0.101 |
drivers | -0.099 |
Token | to |
Token ID | 284 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
kiss | +0.118 |
copy | +0.105 |
last | +0.096 |
dance | +0.092 |
check | +0.088 |
Neg logit contributions | |
ados | -0.339 |
riages | -0.335 |
pm | -0.328 |
barrels | -0.323 |
drivers | -0.320 |
Token | use |
Token ID | 779 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
kiss | +0.033 |
copy | +0.029 |
last | +0.026 |
dance | +0.024 |
massage | +0.024 |
Neg logit contributions | |
ados | -0.091 |
riages | -0.090 |
barrels | -0.088 |
pm | -0.086 |
drivers | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
kiss | +0.056 |
copy | +0.052 |
last | +0.051 |
dance | +0.047 |
scrub | +0.047 |
Neg logit contributions | |
pm | -0.127 |
riages | -0.125 |
ados | -0.125 |
ables | -0.119 |
keys | -0.117 |
Token | oven |
Token ID | 14361 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
kiss | +0.033 |
copy | +0.031 |
last | +0.030 |
dance | +0.029 |
massage | +0.028 |
Neg logit contributions | |
ados | -0.040 |
riages | -0.040 |
pm | -0.039 |
ors | -0.037 |
drivers | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | -7.09e-01 |
Pos logit contributions | |
ados | +0.071 |
riages | +0.070 |
pm | +0.069 |
assing | +0.068 |
barrels | +0.067 |
Neg logit contributions | |
kiss | -0.026 |
last | -0.023 |
copy | -0.022 |
blow | -0.019 |
watch | -0.019 |
Token | needle |
Token ID | 17598 |
Feature activation | +7.77e-01 |
Pos logit contributions | |
kiss | +0.018 |
copy | +0.017 |
last | +0.016 |
blow | +0.016 |
admire | +0.015 |
Neg logit contributions | |
ados | -0.030 |
riages | -0.029 |
barrels | -0.029 |
drivers | -0.029 |
files | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -5.71e-01 |
Pos logit contributions | |
ados | +0.036 |
riages | +0.035 |
barrels | +0.035 |
pm | +0.035 |
drivers | +0.033 |
Neg logit contributions | |
kiss | -0.019 |
copy | -0.016 |
last | -0.016 |
blow | -0.015 |
dance | -0.015 |
Token | She |
Token ID | 1375 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
kiss | +0.020 |
copy | +0.019 |
last | +0.018 |
dance | +0.017 |
blow | +0.017 |
Neg logit contributions | |
ados | -0.020 |
riages | -0.019 |
pm | -0.018 |
barrels | -0.018 |
drivers | -0.018 |
Token | shows |
Token ID | 2523 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
ados | +0.017 |
barrels | +0.016 |
riages | +0.016 |
drivers | +0.016 |
cases | +0.015 |
Neg logit contributions | |
kiss | -0.013 |
copy | -0.013 |
last | -0.012 |
check | -0.012 |
dancer | -0.012 |
Token | her |
Token ID | 607 |
Feature activation | -2.48e+00 |
Pos logit contributions | |
kiss | +0.004 |
copy | +0.004 |
dance | +0.003 |
blow | +0.003 |
scrub | +0.003 |
Neg logit contributions | |
ados | -0.005 |
riages | -0.004 |
barrels | -0.004 |
pm | -0.004 |
ors | -0.004 |
Token | how |
Token ID | 703 |
Feature activation | -6.56e+00 |
Pos logit contributions | |
kiss | +0.055 |
copy | +0.053 |
blow | +0.046 |
check | +0.046 |
last | +0.044 |
Neg logit contributions | |
ados | -0.113 |
barrels | -0.105 |
riages | -0.104 |
drivers | -0.103 |
files | -0.102 |
Token | to |
Token ID | 284 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
kiss | +0.161 |
copy | +0.151 |
blow | +0.130 |
last | +0.130 |
dance | +0.129 |
Neg logit contributions | |
ados | -0.292 |
riages | -0.280 |
barrels | -0.276 |
pm | -0.270 |
drivers | -0.268 |
Token | make |
Token ID | 787 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
kiss | +0.025 |
listener | +0.024 |
dancer | +0.024 |
Sweet | +0.023 |
reaction | +0.022 |
Neg logit contributions | |
riages | -0.050 |
ados | -0.050 |
drivers | -0.050 |
banks | -0.048 |
barrels | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
kiss | +0.059 |
copy | +0.054 |
last | +0.049 |
dance | +0.048 |
blow | +0.048 |
Neg logit contributions | |
ados | -0.102 |
riages | -0.099 |
barrels | -0.097 |
pm | -0.096 |
drivers | -0.095 |
Token | stitch |
Token ID | 24695 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
kiss | +0.034 |
copy | +0.032 |
last | +0.032 |
blow | +0.031 |
check | +0.029 |
Neg logit contributions | |
ados | -0.046 |
riages | -0.045 |
barrels | -0.043 |
drivers | -0.042 |
ables | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
kiss | +0.065 |
copy | +0.060 |
last | +0.055 |
dance | +0.054 |
blow | +0.052 |
Neg logit contributions | |
ados | -0.118 |
riages | -0.117 |
barrels | -0.114 |
pm | -0.114 |
drivers | -0.110 |
Token | . |
Token ID | 13 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
ados | +0.030 |
riages | +0.028 |
barrels | +0.028 |
drivers | +0.028 |
pm | +0.028 |
Neg logit contributions | |
kiss | -0.015 |
copy | -0.013 |
last | -0.012 |
dance | -0.012 |
watch | -0.012 |
Token | And |
Token ID | 843 |
Feature activation | +8.14e-01 |
Pos logit contributions | |
kiss | +0.008 |
copy | +0.008 |
last | +0.007 |
dance | +0.007 |
massage | +0.007 |
Neg logit contributions | |
ados | -0.007 |
riages | -0.007 |
barrels | -0.007 |
pm | -0.006 |
drivers | -0.006 |
Token | that |
Token ID | 326 |
Feature activation | -9.67e-01 |
Pos logit contributions | |
ados | +0.033 |
riages | +0.032 |
barrels | +0.031 |
pm | +0.031 |
drivers | +0.031 |
Neg logit contributions | |
kiss | -0.023 |
copy | -0.022 |
last | -0.021 |
dance | -0.020 |
blow | -0.019 |
Token | 's |
Token ID | 338 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
kiss | +0.024 |
copy | +0.022 |
last | +0.022 |
dance | +0.022 |
blow | +0.020 |
Neg logit contributions | |
ados | -0.045 |
riages | -0.043 |
drivers | -0.042 |
barrels | -0.041 |
pm | -0.040 |
Token | how |
Token ID | 703 |
Feature activation | -5.22e+00 |
Pos logit contributions | |
kiss | +0.040 |
copy | +0.037 |
last | +0.034 |
dance | +0.033 |
blow | +0.033 |
Neg logit contributions | |
ados | -0.060 |
riages | -0.058 |
barrels | -0.057 |
pm | -0.056 |
drivers | -0.055 |
Token | Will |
Token ID | 2561 |
Feature activation | -6.56e+00 |
Pos logit contributions | |
kiss | +0.146 |
copy | +0.135 |
last | +0.132 |
dance | +0.127 |
check | +0.124 |
Neg logit contributions | |
ados | -0.215 |
riages | -0.210 |
pm | -0.202 |
drivers | -0.201 |
barrels | -0.200 |
Token | y |
Token ID | 88 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
kiss | +0.164 |
copy | +0.154 |
last | +0.147 |
dance | +0.146 |
blow | +0.142 |
Neg logit contributions | |
ados | -0.292 |
riages | -0.279 |
drivers | -0.276 |
barrels | -0.274 |
pm | -0.268 |
Token | found |
Token ID | 1043 |
Feature activation | -4.56e-01 |
Pos logit contributions | |
kiss | +0.036 |
copy | +0.033 |
last | +0.030 |
dance | +0.029 |
blow | +0.029 |
Neg logit contributions | |
ados | -0.073 |
riages | -0.071 |
barrels | -0.069 |
drivers | -0.068 |
pm | -0.068 |
Token | his |
Token ID | 465 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
kiss | +0.012 |
copy | +0.011 |
last | +0.010 |
dance | +0.010 |
check | +0.010 |
Neg logit contributions | |
ados | -0.020 |
riages | -0.019 |
pm | -0.019 |
barrels | -0.019 |
drivers | -0.019 |
Token | ancient |
Token ID | 6156 |
Feature activation | -1.69e+00 |
Pos logit contributions | |
kiss | +0.027 |
copy | +0.025 |
blow | +0.022 |
scrub | +0.022 |
massage | +0.021 |
Neg logit contributions | |
ados | -0.042 |
barrels | -0.042 |
riages | -0.042 |
pm | -0.040 |
drivers | -0.039 |
Token | nest |
Token ID | 16343 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
kiss | +0.055 |
copy | +0.052 |
last | +0.049 |
dance | +0.049 |
massage | +0.047 |
Neg logit contributions | |
ados | -0.063 |
riages | -0.061 |
pm | -0.058 |
drivers | -0.057 |
barrels | -0.057 |
Token | safely |
Token ID | 11512 |
Feature activation | -5.07e-01 |
Pos logit contributions | |
kiss | +0.094 |
copy | +0.083 |
last | +0.083 |
dance | +0.078 |
cl | +0.073 |
Neg logit contributions | |
ados | -0.206 |
pm | -0.200 |
riages | -0.199 |
barrels | -0.199 |
drivers | -0.196 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
kiss | +0.011 |
last | +0.010 |
copy | +0.009 |
dance | +0.009 |
check | +0.008 |
Neg logit contributions | |
ados | -0.025 |
barrels | -0.023 |
pm | -0.023 |
riages | -0.023 |
drivers | -0.023 |
Token | She |
Token ID | 1375 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
kiss | +0.041 |
copy | +0.035 |
dance | +0.033 |
last | +0.032 |
check | +0.032 |
Neg logit contributions | |
ados | -0.067 |
barrels | -0.065 |
riages | -0.065 |
drivers | -0.063 |
pm | -0.063 |
Token | showed |
Token ID | 3751 |
Feature activation | -5.08e-01 |
Pos logit contributions | |
kiss | +0.001 |
copy | +0.001 |
last | +0.001 |
dance | +0.001 |
blow | +0.001 |
Neg logit contributions | |
ados | -0.001 |
riages | -0.001 |
barrels | -0.001 |
pm | -0.001 |
drivers | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.012 |
blow | +0.011 |
last | +0.011 |
dance | +0.011 |
Neg logit contributions | |
ados | -0.022 |
riages | -0.021 |
ors | -0.021 |
barrels | -0.020 |
drivers | -0.020 |
Token | how |
Token ID | 703 |
Feature activation | -6.55e+00 |
Pos logit contributions | |
kiss | +0.040 |
copy | +0.038 |
blow | +0.032 |
last | +0.032 |
check | +0.031 |
Neg logit contributions | |
ados | -0.097 |
riages | -0.094 |
barrels | -0.093 |
drivers | -0.090 |
pm | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
kiss | +0.145 |
copy | +0.132 |
last | +0.118 |
dance | +0.115 |
blow | +0.113 |
Neg logit contributions | |
ados | -0.310 |
riages | -0.304 |
barrels | -0.296 |
pm | -0.294 |
drivers | -0.291 |
Token | practice |
Token ID | 3357 |
Feature activation | -3.43e+00 |
Pos logit contributions | |
kiss | +0.024 |
copy | +0.022 |
last | +0.019 |
dab | +0.019 |
dancer | +0.019 |
Neg logit contributions | |
riages | -0.056 |
ados | -0.053 |
barrels | -0.052 |
drivers | -0.051 |
ors | -0.051 |
Token | on |
Token ID | 319 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
kiss | +0.084 |
copy | +0.078 |
last | +0.071 |
dance | +0.070 |
blow | +0.068 |
Neg logit contributions | |
ados | -0.155 |
riages | -0.151 |
barrels | -0.148 |
pm | -0.147 |
drivers | -0.142 |
Token | her |
Token ID | 607 |
Feature activation | -1.78e+00 |
Pos logit contributions | |
ados | +0.002 |
riages | +0.002 |
pm | +0.002 |
barrels | +0.002 |
drivers | +0.002 |
Neg logit contributions | |
kiss | -0.001 |
copy | -0.001 |
last | -0.001 |
dance | -0.001 |
massage | -0.001 |
Token | bike |
Token ID | 7161 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
kiss | +0.055 |
copy | +0.051 |
last | +0.046 |
blow | +0.046 |
dance | +0.046 |
Neg logit contributions | |
ados | -0.068 |
riages | -0.068 |
barrels | -0.065 |
pm | -0.064 |
drivers | -0.063 |
Token | inch |
Token ID | 11111 |
Feature activation | -3.22e+00 |
Pos logit contributions | |
kiss | +0.127 |
copy | +0.117 |
last | +0.105 |
dance | +0.103 |
blow | +0.102 |
Neg logit contributions | |
ados | -0.224 |
riages | -0.219 |
barrels | -0.213 |
pm | -0.211 |
drivers | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
kiss | +0.090 |
copy | +0.085 |
last | +0.084 |
dance | +0.078 |
blow | +0.075 |
Neg logit contributions | |
ados | -0.132 |
riages | -0.129 |
barrels | -0.126 |
pm | -0.124 |
drivers | -0.123 |
Token | She |
Token ID | 1375 |
Feature activation | -2.61e-01 |
Pos logit contributions | |
kiss | +0.056 |
copy | +0.049 |
Mom | +0.047 |
cl | +0.046 |
dance | +0.046 |
Neg logit contributions | |
ados | -0.084 |
riages | -0.083 |
barrels | -0.083 |
drivers | -0.080 |
pm | -0.079 |
Token | showed |
Token ID | 3751 |
Feature activation | -5.55e-01 |
Pos logit contributions | |
kiss | +0.006 |
copy | +0.005 |
last | +0.005 |
dance | +0.005 |
blow | +0.005 |
Neg logit contributions | |
ados | -0.012 |
riages | -0.012 |
barrels | -0.011 |
pm | -0.011 |
drivers | -0.011 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
kiss | +0.014 |
copy | +0.013 |
last | +0.012 |
dance | +0.011 |
blow | +0.011 |
Neg logit contributions | |
ados | -0.025 |
riages | -0.024 |
barrels | -0.023 |
pm | -0.023 |
drivers | -0.023 |
Token | how |
Token ID | 703 |
Feature activation | -6.52e+00 |
Pos logit contributions | |
kiss | +0.052 |
copy | +0.046 |
last | +0.042 |
dance | +0.042 |
massage | +0.039 |
Neg logit contributions | |
riages | -0.130 |
ados | -0.128 |
drivers | -0.126 |
pm | -0.126 |
barrels | -0.125 |
Token | to |
Token ID | 284 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
kiss | +0.141 |
copy | +0.128 |
last | +0.114 |
dance | +0.111 |
blow | +0.109 |
Neg logit contributions | |
ados | -0.313 |
riages | -0.307 |
barrels | -0.299 |
pm | -0.297 |
drivers | -0.294 |
Token | use |
Token ID | 779 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
kiss | +0.027 |
copy | +0.024 |
last | +0.022 |
massage | +0.020 |
dancer | +0.020 |
Neg logit contributions | |
ados | -0.071 |
riages | -0.070 |
barrels | -0.069 |
pm | -0.066 |
bank | -0.066 |
Token | it |
Token ID | 340 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
kiss | +0.051 |
copy | +0.046 |
last | +0.043 |
dance | +0.041 |
check | +0.039 |
Neg logit contributions | |
riages | -0.114 |
ados | -0.113 |
pm | -0.110 |
barrels | -0.108 |
drivers | -0.106 |
Token | to |
Token ID | 284 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
kiss | +0.047 |
copy | +0.044 |
last | +0.043 |
dance | +0.039 |
check | +0.037 |
Neg logit contributions | |
ados | -0.109 |
pm | -0.108 |
riages | -0.108 |
barrels | -0.105 |
drivers | -0.102 |
Token | measure |
Token ID | 3953 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.006 |
last | +0.005 |
massage | +0.005 |
dance | +0.005 |
Neg logit contributions | |
ados | -0.018 |
riages | -0.018 |
barrels | -0.017 |
pm | -0.017 |
drivers | -0.017 |
Token | then |
Token ID | 788 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
kiss | +0.020 |
copy | +0.018 |
last | +0.016 |
dance | +0.016 |
massage | +0.016 |
Neg logit contributions | |
ados | -0.038 |
riages | -0.037 |
barrels | -0.036 |
pm | -0.036 |
drivers | -0.036 |
Token | finally |
Token ID | 3443 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
kiss | +0.085 |
copy | +0.078 |
dance | +0.070 |
last | +0.070 |
blow | +0.069 |
Neg logit contributions | |
ados | -0.159 |
riages | -0.156 |
barrels | -0.152 |
pm | -0.151 |
drivers | -0.149 |
Token | the |
Token ID | 262 |
Feature activation | -2.43e-01 |
Pos logit contributions | |
kiss | +0.014 |
copy | +0.013 |
last | +0.012 |
dance | +0.012 |
blow | +0.012 |
Neg logit contributions | |
ados | -0.026 |
riages | -0.025 |
barrels | -0.025 |
pm | -0.024 |
drivers | -0.024 |
Token | right |
Token ID | 826 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.007 |
dance | +0.006 |
last | +0.006 |
massage | +0.006 |
Neg logit contributions | |
ados | -0.010 |
riages | -0.009 |
barrels | -0.009 |
pm | -0.009 |
drivers | -0.009 |
Token | - |
Token ID | 12 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
ados | +0.010 |
riages | +0.010 |
barrels | +0.010 |
pm | +0.009 |
drivers | +0.009 |
Neg logit contributions | |
kiss | -0.007 |
copy | -0.006 |
massage | -0.006 |
dance | -0.005 |
blow | -0.005 |
Token | shaped |
Token ID | 16760 |
Feature activation | -6.51e+00 |
Pos logit contributions | |
kiss | +0.043 |
copy | +0.041 |
last | +0.038 |
blow | +0.036 |
dance | +0.036 |
Neg logit contributions | |
ados | -0.061 |
riages | -0.059 |
barrels | -0.057 |
pm | -0.057 |
drivers | -0.056 |
Token | hook |
Token ID | 8011 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
kiss | +0.199 |
copy | +0.186 |
last | +0.172 |
dance | +0.169 |
blow | +0.167 |
Neg logit contributions | |
ados | -0.253 |
riages | -0.248 |
barrels | -0.238 |
pm | -0.237 |
drivers | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | +9.74e-02 |
Pos logit contributions | |
kiss | +0.036 |
copy | +0.033 |
last | +0.032 |
dance | +0.030 |
blow | +0.030 |
Neg logit contributions | |
ados | -0.070 |
riages | -0.067 |
pm | -0.066 |
barrels | -0.065 |
drivers | -0.064 |
Token |
|
Token ID | 198 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
riages | +0.004 |
ados | +0.004 |
barrels | +0.004 |
pm | +0.004 |
drivers | +0.004 |
Neg logit contributions | |
kiss | -0.003 |
copy | -0.002 |
last | -0.002 |
check | -0.002 |
dance | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -3.55e-01 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.007 |
last | +0.006 |
dance | +0.006 |
blow | +0.006 |
Neg logit contributions | |
ados | -0.014 |
riages | -0.014 |
pm | -0.014 |
barrels | -0.014 |
drivers | -0.014 |
Token | Once |
Token ID | 7454 |
Feature activation | +4.28e-01 |
Pos logit contributions | |
kiss | +0.009 |
copy | +0.009 |
dance | +0.008 |
massage | +0.008 |
blow | +0.007 |
Neg logit contributions | |
ados | -0.016 |
riages | -0.015 |
barrels | -0.015 |
drivers | -0.015 |
pm | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | -8.95e-01 |
Pos logit contributions | |
kiss | +0.051 |
copy | +0.046 |
last | +0.043 |
dance | +0.042 |
blow | +0.041 |
Neg logit contributions | |
ados | -0.085 |
riages | -0.083 |
barrels | -0.082 |
pm | -0.081 |
drivers | -0.080 |
Token | she |
Token ID | 673 |
Feature activation | -9.52e-02 |
Pos logit contributions | |
kiss | +0.024 |
copy | +0.021 |
dance | +0.021 |
watch | +0.020 |
blow | +0.020 |
Neg logit contributions | |
ados | -0.039 |
riages | -0.039 |
pm | -0.038 |
drivers | -0.037 |
ors | -0.037 |
Token | showed |
Token ID | 3751 |
Feature activation | -7.81e-01 |
Pos logit contributions | |
kiss | +0.002 |
copy | +0.002 |
last | +0.002 |
dance | +0.002 |
massage | +0.002 |
Neg logit contributions | |
ados | -0.004 |
riages | -0.004 |
barrels | -0.004 |
pm | -0.004 |
drivers | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
kiss | +0.017 |
copy | +0.016 |
last | +0.014 |
dance | +0.014 |
blow | +0.014 |
Neg logit contributions | |
ados | -0.037 |
riages | -0.036 |
barrels | -0.035 |
pm | -0.035 |
drivers | -0.034 |
Token | girl |
Token ID | 2576 |
Feature activation | -2.84e+00 |
Pos logit contributions | |
kiss | +0.058 |
copy | +0.053 |
last | +0.050 |
dance | +0.049 |
massage | +0.048 |
Neg logit contributions | |
ados | -0.084 |
riages | -0.083 |
pm | -0.079 |
drivers | -0.078 |
barrels | -0.078 |
Token | how |
Token ID | 703 |
Feature activation | -6.50e+00 |
Pos logit contributions | |
kiss | +0.048 |
copy | +0.037 |
dance | +0.035 |
last | +0.035 |
comfort | +0.033 |
Neg logit contributions | |
drivers | -0.154 |
riages | -0.154 |
ados | -0.152 |
pm | -0.147 |
AW | -0.146 |
Token | to |
Token ID | 284 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
kiss | +0.123 |
copy | +0.106 |
last | +0.100 |
dance | +0.099 |
comfort | +0.094 |
Neg logit contributions | |
ados | -0.331 |
riages | -0.328 |
drivers | -0.319 |
pm | -0.319 |
barrels | -0.317 |
Token | cut |
Token ID | 2005 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
kiss | +0.029 |
copy | +0.026 |
last | +0.023 |
massage | +0.023 |
dancer | +0.021 |
Neg logit contributions | |
riages | -0.076 |
ados | -0.075 |
barrels | -0.073 |
drivers | -0.071 |
pm | -0.071 |
Token | it |
Token ID | 340 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
kiss | +0.058 |
copy | +0.053 |
last | +0.049 |
dance | +0.045 |
scrub | +0.044 |
Neg logit contributions | |
riages | -0.144 |
pm | -0.140 |
ados | -0.140 |
barrels | -0.136 |
drivers | -0.134 |
Token | properly |
Token ID | 6105 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
kiss | +0.033 |
copy | +0.031 |
last | +0.029 |
dance | +0.027 |
blow | +0.027 |
Neg logit contributions | |
ados | -0.069 |
riages | -0.068 |
pm | -0.068 |
barrels | -0.067 |
drivers | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ados | +0.050 |
riages | +0.049 |
pm | +0.049 |
barrels | +0.048 |
drivers | +0.046 |
Neg logit contributions | |
kiss | -0.021 |
copy | -0.020 |
last | -0.019 |
blow | -0.018 |
scrub | -0.018 |
Token | my |
Token ID | 1820 |
Feature activation | -4.66e+00 |
Pos logit contributions | |
kiss | +0.046 |
copy | +0.044 |
dance | +0.038 |
massage | +0.038 |
blow | +0.036 |
Neg logit contributions | |
riages | -0.097 |
ados | -0.096 |
barrels | -0.094 |
pm | -0.093 |
drivers | -0.092 |
Token | showed |
Token ID | 3751 |
Feature activation | -7.49e-01 |
Pos logit contributions | |
kiss | +0.103 |
copy | +0.094 |
last | +0.084 |
dance | +0.081 |
blow | +0.080 |
Neg logit contributions | |
ados | -0.220 |
riages | -0.216 |
barrels | -0.210 |
pm | -0.208 |
drivers | -0.206 |
Token | his |
Token ID | 465 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
kiss | +0.023 |
copy | +0.022 |
dance | +0.020 |
blow | +0.020 |
last | +0.020 |
Neg logit contributions | |
ados | -0.029 |
riages | -0.027 |
barrels | -0.027 |
ors | -0.026 |
drivers | -0.026 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.48e+00 |
Pos logit contributions | |
kiss | +0.028 |
copy | +0.026 |
blow | +0.022 |
last | +0.021 |
massage | +0.020 |
Neg logit contributions | |
ados | -0.090 |
barrels | -0.088 |
riages | -0.087 |
drivers | -0.084 |
pm | -0.083 |
Token | the |
Token ID | 262 |
Feature activation | -5.48e-01 |
Pos logit contributions | |
kiss | +0.089 |
copy | +0.075 |
last | +0.075 |
comfort | +0.065 |
check | +0.064 |
Neg logit contributions | |
riages | -0.227 |
ados | -0.225 |
drivers | -0.218 |
pm | -0.216 |
AW | -0.212 |
Token | mark |
Token ID | 1317 |
Feature activation | -6.48e+00 |
Pos logit contributions | |
kiss | +0.017 |
copy | +0.015 |
last | +0.014 |
dance | +0.014 |
massage | +0.014 |
Neg logit contributions | |
ados | -0.022 |
riages | -0.022 |
pm | -0.021 |
ors | -0.021 |
drivers | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -4.49e-01 |
Pos logit contributions | |
kiss | +0.163 |
copy | +0.150 |
last | +0.136 |
dance | +0.133 |
blow | +0.131 |
Neg logit contributions | |
ados | -0.286 |
riages | -0.281 |
barrels | -0.273 |
pm | -0.270 |
drivers | -0.267 |
Token | asked |
Token ID | 1965 |
Feature activation | -6.96e-01 |
Pos logit contributions | |
kiss | +0.011 |
copy | +0.010 |
last | +0.009 |
dance | +0.009 |
blow | +0.009 |
Neg logit contributions | |
ados | -0.020 |
riages | -0.019 |
barrels | -0.019 |
pm | -0.019 |
drivers | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +3.78e-01 |
Pos logit contributions | |
kiss | +0.017 |
copy | +0.016 |
last | +0.015 |
check | +0.014 |
dance | +0.014 |
Neg logit contributions | |
riages | -0.031 |
ados | -0.031 |
pm | -0.030 |
barrels | -0.030 |
drivers | -0.029 |
Token | " |
Token ID | 366 |
Feature activation | +2.70e+00 |
Pos logit contributions | |
ados | +0.012 |
riages | +0.011 |
barrels | +0.011 |
drivers | +0.011 |
pm | +0.010 |
Neg logit contributions | |
kiss | -0.014 |
copy | -0.013 |
last | -0.013 |
dance | -0.013 |
massage | -0.012 |
Token | What |
Token ID | 2061 |
Feature activation | -2.47e+00 |
Pos logit contributions | |
ados | +0.123 |
barrels | +0.119 |
riages | +0.118 |
pm | +0.113 |
drivers | +0.113 |
Neg logit contributions | |
kiss | -0.062 |
copy | -0.052 |
dance | -0.052 |
last | -0.050 |
massage | -0.048 |
Token | said |
Token ID | 531 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
kiss | +0.063 |
copy | +0.057 |
last | +0.052 |
dance | +0.051 |
blow | +0.050 |
Neg logit contributions | |
ados | -0.126 |
riages | -0.121 |
barrels | -0.118 |
drivers | -0.117 |
pm | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | -9.71e-01 |
Pos logit contributions | |
kiss | +0.006 |
copy | +0.005 |
dance | +0.005 |
last | +0.005 |
cl | +0.005 |
Neg logit contributions | |
ados | -0.010 |
riages | -0.010 |
barrels | -0.010 |
pm | -0.009 |
drivers | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
kiss | +0.025 |
copy | +0.021 |
cl | +0.019 |
dance | +0.019 |
check | +0.019 |
Neg logit contributions | |
ados | -0.043 |
riages | -0.043 |
barrels | -0.042 |
pm | -0.041 |
drivers | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
kiss | +0.031 |
copy | +0.028 |
last | +0.024 |
dance | +0.023 |
blow | +0.023 |
Neg logit contributions | |
ados | -0.084 |
riages | -0.082 |
barrels | -0.080 |
pm | -0.079 |
drivers | -0.078 |
Token | So |
Token ID | 2396 |
Feature activation | -1.33e+00 |
Pos logit contributions | |
kiss | +0.039 |
copy | +0.037 |
dance | +0.032 |
massage | +0.032 |
scrub | +0.031 |
Neg logit contributions | |
ados | -0.069 |
riages | -0.066 |
barrels | -0.064 |
drivers | -0.063 |
ors | -0.063 |
Token | Will |
Token ID | 2561 |
Feature activation | -6.47e+00 |
Pos logit contributions | |
kiss | +0.032 |
last | +0.028 |
mom | +0.028 |
Mom | +0.027 |
copy | +0.027 |
Neg logit contributions | |
drivers | -0.062 |
ados | -0.062 |
assing | -0.062 |
riages | -0.060 |
ors | -0.060 |
Token | got |
Token ID | 1392 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
kiss | +0.147 |
copy | +0.131 |
dance | +0.122 |
last | +0.122 |
blow | +0.120 |
Neg logit contributions | |
ados | -0.318 |
barrels | -0.302 |
riages | -0.300 |
drivers | -0.295 |
pm | -0.285 |
Token | into |
Token ID | 656 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
kiss | +0.025 |
copy | +0.022 |
last | +0.021 |
dance | +0.020 |
check | +0.019 |
Neg logit contributions | |
ados | -0.058 |
riages | -0.057 |
pm | -0.056 |
barrels | -0.055 |
drivers | -0.055 |
Token | the |
Token ID | 262 |
Feature activation | +4.07e-01 |
Pos logit contributions | |
ados | +0.020 |
riages | +0.020 |
ors | +0.019 |
drivers | +0.019 |
ables | +0.019 |
Neg logit contributions | |
kiss | -0.008 |
copy | -0.007 |
last | -0.006 |
dance | -0.006 |
cl | -0.006 |
Token | je |
Token ID | 11223 |
Feature activation | -3.49e-01 |
Pos logit contributions | |
ados | +0.016 |
riages | +0.016 |
barrels | +0.015 |
pm | +0.015 |
drivers | +0.015 |
Neg logit contributions | |
kiss | -0.012 |
copy | -0.011 |
last | -0.010 |
blow | -0.010 |
dance | -0.010 |
Token | ep |
Token ID | 538 |
Feature activation | -4.12e-01 |
Pos logit contributions | |
kiss | +0.010 |
copy | +0.009 |
last | +0.009 |
check | +0.009 |
comfort | +0.008 |
Neg logit contributions | |
ados | -0.014 |
riages | -0.013 |
barrels | -0.013 |
pm | -0.013 |
These | -0.013 |
Token | there |
Token ID | 612 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
ados | +0.036 |
riages | +0.035 |
barrels | +0.035 |
These | +0.034 |
pm | +0.033 |
Neg logit contributions | |
kiss | -0.068 |
copy | -0.063 |
massage | -0.062 |
dance | -0.058 |
blow | -0.057 |
Token | was |
Token ID | 373 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
riages | +0.079 |
pm | +0.078 |
barrels | +0.078 |
ados | +0.077 |
These | +0.077 |
Neg logit contributions | |
kiss | -0.075 |
copy | -0.071 |
massage | -0.063 |
blow | -0.062 |
check | -0.061 |
Token | a |
Token ID | 257 |
Feature activation | +2.06e-01 |
Pos logit contributions | |
kiss | +0.008 |
copy | +0.008 |
dance | +0.007 |
last | +0.007 |
check | +0.007 |
Neg logit contributions | |
ados | -0.012 |
barrels | -0.011 |
riages | -0.011 |
pm | -0.011 |
These | -0.010 |
Token | little |
Token ID | 1310 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
barrels | +0.007 |
riages | +0.007 |
ados | +0.007 |
files | +0.007 |
pm | +0.007 |
Neg logit contributions | |
kiss | -0.006 |
copy | -0.005 |
last | -0.005 |
scrub | -0.005 |
dab | -0.005 |
Token | squirrel |
Token ID | 33039 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
kiss | +0.012 |
copy | +0.011 |
last | +0.009 |
scrub | +0.009 |
blow | +0.009 |
Neg logit contributions | |
ados | -0.027 |
riages | -0.026 |
barrels | -0.026 |
pm | -0.025 |
drivers | -0.025 |
Token | named |
Token ID | 3706 |
Feature activation | +4.22e+00 |
Pos logit contributions | |
riages | +0.012 |
ados | +0.012 |
barrels | +0.011 |
pm | +0.011 |
ors | +0.011 |
Neg logit contributions | |
kiss | -0.013 |
copy | -0.012 |
massage | -0.011 |
last | -0.011 |
blow | -0.011 |
Token | Nut |
Token ID | 11959 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
pm | +0.147 |
riages | +0.147 |
ados | +0.144 |
drivers | +0.141 |
barrels | +0.139 |
Neg logit contributions | |
kiss | -0.140 |
last | -0.133 |
copy | -0.129 |
check | -0.127 |
dance | -0.126 |
Token | ty |
Token ID | 774 |
Feature activation | +3.59e-01 |
Pos logit contributions | |
ados | +0.101 |
riages | +0.098 |
barrels | +0.095 |
pm | +0.094 |
drivers | +0.093 |
Neg logit contributions | |
kiss | -0.094 |
copy | -0.089 |
last | -0.084 |
blow | -0.081 |
dance | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
ados | +0.019 |
riages | +0.018 |
pm | +0.018 |
drivers | +0.018 |
barrels | +0.018 |
Neg logit contributions | |
kiss | -0.006 |
last | -0.006 |
copy | -0.005 |
watch | -0.005 |
check | -0.005 |
Token | Nut |
Token ID | 11959 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
pm | +0.026 |
riages | +0.023 |
ados | +0.023 |
keys | +0.022 |
cases | +0.022 |
Neg logit contributions | |
kiss | -0.041 |
last | -0.040 |
check | -0.039 |
rub | -0.039 |
blow | -0.039 |
Token | ty |
Token ID | 774 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ados | +0.047 |
riages | +0.046 |
barrels | +0.044 |
pm | +0.044 |
drivers | +0.043 |
Neg logit contributions | |
kiss | -0.050 |
copy | -0.047 |
last | -0.044 |
dance | -0.044 |
blow | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
kiss | +0.021 |
copy | +0.019 |
last | +0.017 |
dance | +0.016 |
blow | +0.016 |
Neg logit contributions | |
ados | -0.051 |
riages | -0.050 |
barrels | -0.048 |
pm | -0.048 |
drivers | -0.048 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
kiss | +0.035 |
copy | +0.032 |
scrub | +0.028 |
last | +0.028 |
check | +0.028 |
Neg logit contributions | |
ados | -0.065 |
barrels | -0.062 |
riages | -0.061 |
pm | -0.061 |
drivers | -0.058 |
Token | are |
Token ID | 389 |
Feature activation | +6.21e-01 |
Pos logit contributions | |
kiss | +0.032 |
copy | +0.030 |
last | +0.028 |
dance | +0.027 |
blow | +0.027 |
Neg logit contributions | |
ados | -0.040 |
riages | -0.039 |
barrels | -0.037 |
pm | -0.037 |
drivers | -0.037 |
Token | sad |
Token ID | 6507 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
ados | +0.023 |
riages | +0.023 |
barrels | +0.022 |
pm | +0.021 |
drivers | +0.021 |
Neg logit contributions | |
kiss | -0.019 |
copy | -0.018 |
massage | -0.017 |
last | -0.016 |
blow | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -3.73e-01 |
Pos logit contributions | |
ados | +0.104 |
riages | +0.101 |
barrels | +0.099 |
drivers | +0.096 |
pm | +0.096 |
Neg logit contributions | |
kiss | -0.072 |
copy | -0.069 |
massage | -0.061 |
blow | -0.060 |
dance | -0.060 |
Token | mad |
Token ID | 8805 |
Feature activation | +3.96e+00 |
Pos logit contributions | |
kiss | +0.011 |
copy | +0.011 |
last | +0.010 |
massage | +0.009 |
dance | +0.009 |
Neg logit contributions | |
ados | -0.015 |
riages | -0.014 |
barrels | -0.014 |
pm | -0.014 |
drivers | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | +1.88e-01 |
Pos logit contributions | |
ados | +0.191 |
barrels | +0.189 |
riages | +0.189 |
pm | +0.187 |
drivers | +0.183 |
Neg logit contributions | |
kiss | -0.082 |
copy | -0.073 |
last | -0.069 |
cl | -0.066 |
dance | -0.065 |
Token | They |
Token ID | 1119 |
Feature activation | +2.18e+00 |
Pos logit contributions | |
ados | +0.007 |
riages | +0.007 |
barrels | +0.006 |
drivers | +0.006 |
ors | +0.006 |
Neg logit contributions | |
kiss | -0.007 |
copy | -0.006 |
Mom | -0.006 |
cl | -0.006 |
dance | -0.006 |
Token | do |
Token ID | 466 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
ados | +0.108 |
riages | +0.104 |
pm | +0.102 |
barrels | +0.101 |
drivers | +0.101 |
Neg logit contributions | |
kiss | -0.047 |
copy | -0.042 |
last | -0.038 |
dance | -0.037 |
blow | -0.037 |
Token | not |
Token ID | 407 |
Feature activation | +6.70e-01 |
Pos logit contributions | |
kiss | +0.069 |
copy | +0.065 |
massage | +0.061 |
dance | +0.059 |
last | +0.059 |
Neg logit contributions | |
ados | -0.062 |
riages | -0.062 |
pm | -0.060 |
barrels | -0.060 |
drivers | -0.057 |
Token | want |
Token ID | 765 |
Feature activation | -7.85e-01 |
Pos logit contributions | |
ados | +0.029 |
riages | +0.029 |
barrels | +0.028 |
pm | +0.028 |
drivers | +0.027 |
Neg logit contributions | |
kiss | -0.017 |
copy | -0.016 |
last | -0.014 |
massage | -0.014 |
dance | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -9.65e-01 |
Pos logit contributions | |
pm | +0.026 |
ados | +0.025 |
riages | +0.025 |
barrels | +0.025 |
drivers | +0.024 |
Neg logit contributions | |
kiss | -0.012 |
last | -0.011 |
copy | -0.011 |
dance | -0.010 |
blow | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -9.69e-01 |
Pos logit contributions | |
kiss | +0.027 |
copy | +0.023 |
check | +0.022 |
dance | +0.022 |
Mom | +0.022 |
Neg logit contributions | |
ados | -0.040 |
riages | -0.040 |
drivers | -0.038 |
barrels | -0.038 |
pm | -0.037 |
Token | L |
Token ID | 43 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
kiss | +0.025 |
copy | +0.024 |
blow | +0.021 |
dance | +0.021 |
massage | +0.020 |
Neg logit contributions | |
ados | -0.041 |
riages | -0.040 |
barrels | -0.039 |
pm | -0.039 |
drivers | -0.038 |
Token | ily |
Token ID | 813 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
ados | +0.044 |
pm | +0.042 |
barrels | +0.042 |
riages | +0.042 |
drivers | +0.041 |
Neg logit contributions | |
kiss | -0.029 |
copy | -0.027 |
massage | -0.024 |
dance | -0.024 |
blow | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | -5.28e-01 |
Pos logit contributions | |
ados | +0.011 |
riages | +0.011 |
barrels | +0.010 |
pm | +0.010 |
ors | +0.010 |
Neg logit contributions | |
kiss | -0.006 |
copy | -0.006 |
last | -0.005 |
massage | -0.005 |
dance | -0.005 |
Token | hungry |
Token ID | 14720 |
Feature activation | +4.10e+00 |
Pos logit contributions | |
kiss | +0.019 |
copy | +0.019 |
last | +0.016 |
massage | +0.016 |
dance | +0.015 |
Neg logit contributions | |
ados | -0.016 |
riages | -0.016 |
ards | -0.015 |
barrels | -0.015 |
opes | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ados | +0.184 |
riages | +0.179 |
barrels | +0.175 |
pm | +0.172 |
drivers | +0.170 |
Neg logit contributions | |
kiss | -0.099 |
copy | -0.093 |
last | -0.082 |
dance | -0.081 |
blow | -0.079 |
Token | She |
Token ID | 1375 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
riages | +0.063 |
ados | +0.063 |
barrels | +0.060 |
pm | +0.059 |
drivers | +0.059 |
Neg logit contributions | |
kiss | -0.044 |
copy | -0.039 |
dance | -0.037 |
check | -0.036 |
massage | -0.036 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
ados | +0.061 |
riages | +0.059 |
barrels | +0.057 |
drivers | +0.057 |
pm | +0.056 |
Neg logit contributions | |
kiss | -0.033 |
copy | -0.030 |
last | -0.027 |
dance | -0.027 |
blow | -0.027 |
Token | to |
Token ID | 284 |
Feature activation | +9.60e-01 |
Pos logit contributions | |
kiss | +0.036 |
copy | +0.033 |
last | +0.029 |
dance | +0.029 |
blow | +0.028 |
Neg logit contributions | |
ados | -0.070 |
riages | -0.068 |
barrels | -0.066 |
pm | -0.066 |
drivers | -0.065 |
Token | eat |
Token ID | 4483 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
ados | +0.046 |
riages | +0.045 |
barrels | +0.044 |
pm | +0.043 |
drivers | +0.042 |
Neg logit contributions | |
kiss | -0.020 |
copy | -0.019 |
last | -0.017 |
dance | -0.016 |
blow | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
riages | +0.165 |
ados | +0.162 |
drivers | +0.158 |
barrels | +0.157 |
pm | +0.155 |
Neg logit contributions | |
kiss | -0.070 |
copy | -0.064 |
last | -0.059 |
dance | -0.057 |
cl | -0.054 |
Token | We |
Token ID | 775 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
riages | +0.036 |
ados | +0.036 |
barrels | +0.036 |
drivers | +0.034 |
bank | +0.034 |
Neg logit contributions | |
kiss | -0.038 |
Mom | -0.035 |
copy | -0.033 |
mom | -0.032 |
watch | -0.032 |
Token | want |
Token ID | 765 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
ados | +0.048 |
riages | +0.045 |
barrels | +0.044 |
pm | +0.044 |
drivers | +0.044 |
Neg logit contributions | |
kiss | -0.026 |
copy | -0.024 |
last | -0.022 |
blow | -0.021 |
watch | -0.021 |
Token | to |
Token ID | 284 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
kiss | +0.029 |
copy | +0.025 |
last | +0.024 |
dance | +0.022 |
massage | +0.021 |
Neg logit contributions | |
riages | -0.075 |
ados | -0.075 |
pm | -0.073 |
drivers | -0.072 |
barrels | -0.072 |
Token | be |
Token ID | 307 |
Feature activation | -5.70e-01 |
Pos logit contributions | |
riages | +0.027 |
ados | +0.027 |
barrels | +0.026 |
pm | +0.026 |
bank | +0.025 |
Neg logit contributions | |
kiss | -0.011 |
copy | -0.010 |
scrub | -0.009 |
last | -0.009 |
patiently | -0.009 |
Token | naughty |
Token ID | 45128 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
kiss | +0.015 |
copy | +0.014 |
dance | +0.012 |
last | +0.012 |
blow | +0.012 |
Neg logit contributions | |
ados | -0.025 |
riages | -0.025 |
pm | -0.024 |
drivers | -0.024 |
barrels | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
riages | +0.243 |
ados | +0.241 |
barrels | +0.232 |
drivers | +0.230 |
pm | +0.228 |
Neg logit contributions | |
kiss | -0.115 |
copy | -0.103 |
last | -0.096 |
cl | -0.092 |
dance | -0.092 |
Token | We |
Token ID | 775 |
Feature activation | +9.63e-01 |
Pos logit contributions | |
ados | +0.032 |
riages | +0.031 |
barrels | +0.030 |
drivers | +0.030 |
bank | +0.029 |
Neg logit contributions | |
kiss | -0.039 |
copy | -0.036 |
watch | -0.034 |
blow | -0.033 |
massage | -0.033 |
Token | want |
Token ID | 765 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ados | +0.047 |
riages | +0.045 |
barrels | +0.044 |
pm | +0.044 |
drivers | +0.044 |
Neg logit contributions | |
kiss | -0.024 |
copy | -0.021 |
last | -0.019 |
blow | -0.019 |
watch | -0.019 |
Token | to |
Token ID | 284 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
kiss | +0.030 |
copy | +0.026 |
last | +0.025 |
dance | +0.023 |
massage | +0.022 |
Neg logit contributions | |
riages | -0.078 |
ados | -0.078 |
pm | -0.075 |
drivers | -0.075 |
barrels | -0.074 |
Token | have |
Token ID | 423 |
Feature activation | -4.89e-01 |
Pos logit contributions | |
riages | +0.025 |
ados | +0.025 |
barrels | +0.024 |
pm | +0.024 |
bank | +0.023 |
Neg logit contributions | |
kiss | -0.010 |
copy | -0.009 |
scrub | -0.008 |
last | -0.008 |
massage | -0.008 |
Token | He |
Token ID | 679 |
Feature activation | -9.87e-02 |
Pos logit contributions | |
kiss | +0.012 |
copy | +0.011 |
last | +0.011 |
cl | +0.010 |
check | +0.010 |
Neg logit contributions | |
ados | -0.016 |
riages | -0.015 |
barrels | -0.015 |
pm | -0.014 |
tresp | -0.014 |
Token | told |
Token ID | 1297 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
kiss | +0.003 |
copy | +0.002 |
massage | +0.002 |
dance | +0.002 |
last | +0.002 |
Neg logit contributions | |
riages | -0.004 |
ados | -0.004 |
barrels | -0.004 |
drivers | -0.004 |
pm | -0.004 |
Token | them |
Token ID | 606 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
ados | +0.086 |
riages | +0.083 |
pm | +0.079 |
ables | +0.079 |
barrels | +0.079 |
Neg logit contributions | |
kiss | -0.051 |
copy | -0.049 |
last | -0.046 |
dance | -0.044 |
blow | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | +2.00e-01 |
Pos logit contributions | |
AW | +0.031 |
ados | +0.031 |
assing | +0.031 |
pm | +0.030 |
riages | +0.030 |
Neg logit contributions | |
kiss | -0.008 |
last | -0.008 |
mom | -0.008 |
also | -0.007 |
watch | -0.007 |
Token | get |
Token ID | 651 |
Feature activation | -1.45e+00 |
Pos logit contributions | |
ados | +0.010 |
riages | +0.010 |
barrels | +0.010 |
pm | +0.010 |
drivers | +0.010 |
Neg logit contributions | |
kiss | -0.004 |
copy | -0.003 |
last | -0.003 |
dance | -0.003 |
blow | -0.003 |
Token | cups |
Token ID | 14180 |
Feature activation | +4.61e+00 |
Pos logit contributions | |
kiss | +0.033 |
copy | +0.030 |
last | +0.028 |
dance | +0.026 |
blow | +0.026 |
Neg logit contributions | |
ados | -0.068 |
riages | -0.067 |
pm | -0.065 |
barrels | -0.065 |
drivers | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | -6.32e-01 |
Pos logit contributions | |
pm | +0.239 |
ados | +0.237 |
riages | +0.232 |
AW | +0.223 |
assing | +0.222 |
Neg logit contributions | |
kiss | -0.088 |
last | -0.081 |
copy | -0.070 |
blow | -0.065 |
dance | -0.063 |
Token | buckets |
Token ID | 38674 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
kiss | +0.016 |
copy | +0.014 |
blow | +0.013 |
last | +0.013 |
scrub | +0.013 |
Neg logit contributions | |
ados | -0.029 |
riages | -0.029 |
pm | -0.028 |
ors | -0.028 |
ables | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
ados | +0.117 |
pm | +0.116 |
riages | +0.114 |
assing | +0.112 |
AW | +0.108 |
Neg logit contributions | |
kiss | -0.047 |
last | -0.043 |
copy | -0.039 |
scrub | -0.035 |
blow | -0.035 |
Token | start |
Token ID | 923 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
kiss | +0.021 |
copy | +0.018 |
last | +0.017 |
blow | +0.017 |
dance | +0.017 |
Neg logit contributions | |
ados | -0.040 |
riages | -0.039 |
ors | -0.038 |
pm | -0.037 |
drivers | -0.036 |
Token | filling |
Token ID | 12591 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
kiss | +0.022 |
copy | +0.020 |
scrub | +0.020 |
last | +0.020 |
dance | +0.019 |
Neg logit contributions | |
ados | -0.054 |
riages | -0.053 |
pm | -0.052 |
ors | -0.050 |
barrels | -0.050 |
Token | something |
Token ID | 1223 |
Feature activation | +3.23e-01 |
Pos logit contributions | |
kiss | +0.001 |
copy | +0.001 |
last | +0.001 |
dance | +0.001 |
blow | +0.001 |
Neg logit contributions | |
ados | -0.002 |
riages | -0.002 |
barrels | -0.002 |
pm | -0.002 |
drivers | -0.002 |
Token | amazing |
Token ID | 4998 |
Feature activation | +6.74e-03 |
Pos logit contributions | |
ados | +0.015 |
riages | +0.014 |
pm | +0.014 |
barrels | +0.014 |
drivers | +0.013 |
Neg logit contributions | |
kiss | -0.008 |
copy | -0.008 |
last | -0.007 |
dance | -0.007 |
blow | -0.007 |
Token | ! |
Token ID | 0 |
Feature activation | -5.14e-01 |
Pos logit contributions | |
ados | +0.000 |
riages | +0.000 |
barrels | +0.000 |
pm | +0.000 |
drivers | +0.000 |
Neg logit contributions | |
kiss | -0.000 |
copy | -0.000 |
last | -0.000 |
dance | -0.000 |
blow | -0.000 |
Token | It |
Token ID | 632 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
kiss | +0.016 |
copy | +0.014 |
check | +0.014 |
last | +0.013 |
massage | +0.013 |
Neg logit contributions | |
riages | -0.020 |
ados | -0.019 |
barrels | -0.019 |
pm | -0.018 |
drivers | -0.018 |
Token | was |
Token ID | 373 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
kiss | +0.010 |
copy | +0.009 |
last | +0.008 |
dance | +0.008 |
cl | +0.007 |
Neg logit contributions | |
ados | -0.031 |
riages | -0.030 |
pm | -0.029 |
barrels | -0.029 |
drivers | -0.029 |
Token | filled |
Token ID | 5901 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.006 |
last | +0.006 |
dance | +0.006 |
blow | +0.006 |
Neg logit contributions | |
ados | -0.019 |
riages | -0.019 |
pm | -0.018 |
ors | -0.018 |
barrels | -0.018 |
Token | with |
Token ID | 351 |
Feature activation | +1.01e+00 |
Pos logit contributions | |
ados | +0.077 |
riages | +0.077 |
drivers | +0.076 |
barrels | +0.075 |
pm | +0.074 |
Neg logit contributions | |
kiss | -0.034 |
copy | -0.032 |
last | -0.028 |
dance | -0.028 |
blow | -0.026 |
Token | ch |
Token ID | 442 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
ados | +0.045 |
riages | +0.045 |
ors | +0.045 |
pm | +0.044 |
drivers | +0.044 |
Neg logit contributions | |
kiss | -0.026 |
copy | -0.022 |
last | -0.022 |
dance | -0.021 |
cl | -0.021 |
Token | ocol |
Token ID | 4668 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.013 |
last | +0.012 |
dance | +0.012 |
check | +0.012 |
Neg logit contributions | |
riages | -0.008 |
bank | -0.008 |
ados | -0.008 |
pm | -0.008 |
drivers | -0.008 |
Token | ates |
Token ID | 689 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
riages | +0.088 |
ados | +0.087 |
pm | +0.083 |
barrels | +0.082 |
drivers | +0.078 |
Neg logit contributions | |
kiss | -0.149 |
copy | -0.143 |
last | -0.136 |
blow | -0.133 |
dance | -0.132 |
Token | and |
Token ID | 290 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
ados | +0.037 |
pm | +0.036 |
riages | +0.036 |
drivers | +0.035 |
barrels | +0.035 |
Neg logit contributions | |
kiss | -0.016 |
last | -0.014 |
dance | -0.014 |
copy | -0.014 |
blow | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | -5.91e-01 |
Pos logit contributions | |
kiss | +0.004 |
copy | +0.004 |
last | +0.003 |
blow | +0.003 |
dance | +0.003 |
Neg logit contributions | |
ados | -0.008 |
riages | -0.008 |
barrels | -0.007 |
pm | -0.007 |
drivers | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -7.34e-01 |
Pos logit contributions | |
copy | +0.017 |
kiss | +0.016 |
blow | +0.015 |
last | +0.015 |
dance | +0.014 |
Neg logit contributions | |
riages | -0.021 |
ados | -0.021 |
ables | -0.021 |
pm | -0.021 |
These | -0.021 |
Token | The |
Token ID | 464 |
Feature activation | -6.27e-01 |
Pos logit contributions | |
kiss | +0.020 |
copy | +0.019 |
last | +0.017 |
dance | +0.017 |
massage | +0.016 |
Neg logit contributions | |
ados | -0.032 |
riages | -0.031 |
barrels | -0.030 |
pm | -0.030 |
drivers | -0.030 |
Token | girl |
Token ID | 2576 |
Feature activation | -3.43e-01 |
Pos logit contributions | |
copy | +0.015 |
kiss | +0.015 |
scrub | +0.014 |
admire | +0.013 |
pat | +0.013 |
Neg logit contributions | |
ados | -0.024 |
pm | -0.024 |
files | -0.024 |
barrels | -0.024 |
sacks | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
kiss | +0.008 |
copy | +0.008 |
last | +0.007 |
massage | +0.007 |
check | +0.007 |
Neg logit contributions | |
ados | -0.014 |
riages | -0.014 |
pm | -0.014 |
barrels | -0.014 |
drivers | -0.013 |
Token | so |
Token ID | 523 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
kiss | +0.046 |
copy | +0.044 |
blow | +0.040 |
massage | +0.040 |
dance | +0.040 |
Neg logit contributions | |
riages | -0.035 |
ados | -0.034 |
pm | -0.033 |
barrels | -0.032 |
bank | -0.030 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.15e-01 |
Pos logit contributions | |
pm | +0.040 |
riages | +0.040 |
These | +0.039 |
ados | +0.037 |
ages | +0.037 |
Neg logit contributions | |
kiss | -0.034 |
copy | -0.033 |
massage | -0.032 |
scrub | -0.031 |
check | -0.030 |
Token | she |
Token ID | 673 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
kiss | +0.005 |
copy | +0.004 |
last | +0.004 |
dance | +0.004 |
cl | +0.004 |
Neg logit contributions | |
riages | -0.010 |
ados | -0.010 |
barrels | -0.010 |
drivers | -0.010 |
pm | -0.010 |
Token | could |
Token ID | 714 |
Feature activation | -3.39e+00 |
Pos logit contributions | |
riages | +0.046 |
ados | +0.045 |
barrels | +0.043 |
pm | +0.043 |
drivers | +0.042 |
Neg logit contributions | |
kiss | -0.028 |
copy | -0.026 |
last | -0.024 |
massage | -0.023 |
dance | -0.023 |
Token | play |
Token ID | 711 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
kiss | +0.071 |
copy | +0.068 |
scrub | +0.066 |
dab | +0.064 |
massage | +0.062 |
Neg logit contributions | |
riages | -0.144 |
ados | -0.140 |
barrels | -0.133 |
Sale | -0.131 |
pm | -0.129 |
Token | in |
Token ID | 287 |
Feature activation | -1.56e+00 |
Pos logit contributions | |
kiss | +0.079 |
copy | +0.074 |
dance | +0.068 |
last | +0.064 |
watch | +0.062 |
Neg logit contributions | |
riages | -0.174 |
ados | -0.173 |
drivers | -0.171 |
barrels | -0.168 |
pm | -0.167 |
Token | it |
Token ID | 340 |
Feature activation | +9.71e-01 |
Pos logit contributions | |
ados | +0.035 |
drivers | +0.034 |
barrels | +0.034 |
riages | +0.034 |
bank | +0.033 |
Neg logit contributions | |
copy | -0.015 |
kiss | -0.015 |
dance | -0.014 |
watch | -0.014 |
cl | -0.013 |
Token | too |
Token ID | 1165 |
Feature activation | +8.90e-01 |
Pos logit contributions | |
ados | +0.031 |
riages | +0.030 |
barrels | +0.028 |
pm | +0.028 |
drivers | +0.028 |
Neg logit contributions | |
kiss | -0.037 |
copy | -0.035 |
last | -0.032 |
dance | -0.032 |
blow | -0.032 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ados | +0.030 |
riages | +0.030 |
barrels | +0.028 |
pm | +0.028 |
drivers | +0.028 |
Neg logit contributions | |
kiss | -0.032 |
copy | -0.030 |
last | -0.028 |
dance | -0.028 |
blow | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +7.15e-01 |
Pos logit contributions | |
ados | +0.058 |
riages | +0.057 |
barrels | +0.055 |
pm | +0.055 |
drivers | +0.054 |
Neg logit contributions | |
kiss | -0.032 |
copy | -0.030 |
last | -0.027 |
dance | -0.027 |
blow | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +4.61e-01 |
Pos logit contributions | |
pm | +0.022 |
barrels | +0.021 |
ables | +0.021 |
oc | +0.021 |
riages | +0.021 |
Neg logit contributions | |
copy | -0.024 |
kiss | -0.023 |
massage | -0.022 |
blow | -0.022 |
groom | -0.022 |
Token | L |
Token ID | 43 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
riages | +0.018 |
ados | +0.018 |
barrels | +0.018 |
pm | +0.017 |
drivers | +0.017 |
Neg logit contributions | |
kiss | -0.013 |
copy | -0.012 |
dance | -0.011 |
massage | -0.011 |
last | -0.011 |
Token | ily |
Token ID | 813 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
ados | +0.032 |
pm | +0.030 |
barrels | +0.029 |
riages | +0.029 |
drivers | +0.029 |
Neg logit contributions | |
kiss | -0.028 |
copy | -0.026 |
massage | -0.024 |
dance | -0.024 |
blow | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
kiss | +0.021 |
copy | +0.020 |
last | +0.018 |
massage | +0.018 |
check | +0.018 |
Neg logit contributions | |
ados | -0.029 |
riages | -0.029 |
barrels | -0.028 |
pm | -0.028 |
drivers | -0.027 |
Token | Ben |
Token ID | 3932 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
kiss | +0.031 |
copy | +0.031 |
scrub | +0.028 |
dab | +0.028 |
folded | +0.027 |
Neg logit contributions | |
ados | -0.047 |
barrels | -0.042 |
guns | -0.040 |
riages | -0.039 |
pm | -0.039 |
Token | hug |
Token ID | 16225 |
Feature activation | -9.22e-01 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.007 |
massage | +0.006 |
last | +0.006 |
dab | +0.006 |
Neg logit contributions | |
riages | -0.016 |
ados | -0.016 |
barrels | -0.015 |
assing | -0.015 |
pm | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | +5.57e-01 |
Pos logit contributions | |
kiss | +0.024 |
copy | +0.023 |
last | +0.021 |
blow | +0.020 |
massage | +0.020 |
Neg logit contributions | |
ados | -0.039 |
barrels | -0.036 |
riages | -0.036 |
pm | -0.035 |
ors | -0.034 |
Token | He |
Token ID | 679 |
Feature activation | -2.23e-01 |
Pos logit contributions | |
kiss | +0.018 |
copy | +0.016 |
check | +0.015 |
massage | +0.015 |
dance | +0.015 |
Neg logit contributions | |
ados | -0.025 |
riages | -0.025 |
barrels | -0.024 |
drivers | -0.024 |
pm | -0.024 |
Token | felt |
Token ID | 2936 |
Feature activation | -4.47e-01 |
Pos logit contributions | |
kiss | +0.005 |
copy | +0.005 |
last | +0.004 |
massage | +0.004 |
blow | +0.004 |
Neg logit contributions | |
riages | -0.009 |
ados | -0.009 |
pm | -0.009 |
barrels | -0.009 |
bank | -0.009 |
Token | happy |
Token ID | 3772 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.012 |
last | +0.011 |
blow | +0.011 |
dance | +0.011 |
Neg logit contributions | |
ados | -0.017 |
riages | -0.017 |
barrels | -0.017 |
pm | -0.016 |
bank | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | -6.08e-01 |
Pos logit contributions | |
ados | +0.026 |
riages | +0.025 |
barrels | +0.024 |
pm | +0.024 |
drivers | +0.023 |
Neg logit contributions | |
kiss | -0.016 |
copy | -0.015 |
last | -0.013 |
blow | -0.013 |
massage | -0.013 |
Token | safe |
Token ID | 3338 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
kiss | +0.016 |
copy | +0.015 |
last | +0.013 |
massage | +0.013 |
blow | +0.012 |
Neg logit contributions | |
riages | -0.025 |
ados | -0.024 |
barrels | -0.024 |
bank | -0.023 |
pm | -0.023 |
Token | . |
Token ID | 13 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
ados | +0.084 |
riages | +0.082 |
barrels | +0.080 |
pm | +0.079 |
drivers | +0.079 |
Neg logit contributions | |
kiss | -0.047 |
copy | -0.043 |
last | -0.039 |
dance | -0.038 |
blow | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
kiss | +0.003 |
copy | +0.003 |
last | +0.003 |
check | +0.003 |
dance | +0.003 |
Neg logit contributions | |
ados | -0.005 |
riages | -0.005 |
barrels | -0.005 |
pm | -0.004 |
drivers | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
kiss | +0.017 |
copy | +0.016 |
blow | +0.014 |
dance | +0.014 |
last | +0.014 |
Neg logit contributions | |
riages | -0.027 |
pm | -0.027 |
ados | -0.027 |
barrels | -0.026 |
drivers | -0.026 |
Token | R |
Token ID | 49 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
kiss | +0.017 |
copy | +0.017 |
dance | +0.015 |
last | +0.015 |
massage | +0.014 |
Neg logit contributions | |
ados | -0.030 |
riages | -0.028 |
barrels | -0.027 |
drivers | -0.027 |
pm | -0.027 |
Token | oo |
Token ID | 2238 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
kiss | +0.066 |
copy | +0.062 |
last | +0.058 |
dance | +0.057 |
blow | +0.056 |
Neg logit contributions | |
ados | -0.076 |
riages | -0.074 |
barrels | -0.071 |
drivers | -0.070 |
pm | -0.070 |
Token | 's |
Token ID | 338 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
kiss | +0.030 |
copy | +0.026 |
last | +0.024 |
blow | +0.024 |
dance | +0.024 |
Neg logit contributions | |
ados | -0.064 |
riages | -0.062 |
barrels | -0.061 |
drivers | -0.060 |
pm | -0.059 |
Token | so |
Token ID | 523 |
Feature activation | +2.77e+00 |
Pos logit contributions | |
ados | +0.028 |
pm | +0.028 |
riages | +0.027 |
ors | +0.027 |
drivers | +0.027 |
Neg logit contributions | |
kiss | -0.014 |
copy | -0.013 |
last | -0.013 |
blow | -0.012 |
dance | -0.012 |
Token | pretty |
Token ID | 2495 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
ados | +0.130 |
riages | +0.127 |
barrels | +0.124 |
pm | +0.123 |
drivers | +0.121 |
Neg logit contributions | |
kiss | -0.064 |
copy | -0.058 |
last | -0.052 |
dance | -0.051 |
blow | -0.051 |
Token | !" |
Token ID | 2474 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
riages | +0.044 |
assing | +0.043 |
drivers | +0.043 |
pm | +0.043 |
barrels | +0.043 |
Neg logit contributions | |
kiss | -0.032 |
last | -0.029 |
cl | -0.029 |
copy | -0.028 |
dance | -0.028 |
Token | L |
Token ID | 406 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
ados | +0.041 |
riages | +0.040 |
ables | +0.039 |
pounds | +0.039 |
barrels | +0.039 |
Neg logit contributions | |
kiss | -0.014 |
cl | -0.011 |
copy | -0.011 |
Mom | -0.011 |
mom | -0.011 |
Token | ila |
Token ID | 10102 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
kiss | +0.002 |
copy | +0.001 |
massage | +0.001 |
last | +0.001 |
dance | +0.001 |
Neg logit contributions | |
ados | -0.003 |
riages | -0.003 |
pm | -0.003 |
barrels | -0.003 |
drivers | -0.003 |
Token | said |
Token ID | 531 |
Feature activation | +1.53e+00 |
Pos logit contributions | |
ados | +0.003 |
riages | +0.003 |
pm | +0.003 |
barrels | +0.003 |
drivers | +0.003 |
Neg logit contributions | |
kiss | -0.002 |
copy | -0.002 |
last | -0.002 |
massage | -0.002 |
dance | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +9.31e-01 |
Pos logit contributions | |
ados | +0.053 |
riages | +0.050 |
drivers | +0.049 |
ors | +0.048 |
pm | +0.046 |
Neg logit contributions | |
kiss | -0.051 |
copy | -0.049 |
massage | -0.046 |
last | -0.045 |
dance | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | +8.44e-02 |
Pos logit contributions | |
ados | +0.033 |
riages | +0.032 |
drivers | +0.031 |
barrels | +0.031 |
pm | +0.030 |
Neg logit contributions | |
kiss | -0.031 |
copy | -0.030 |
last | -0.028 |
dance | -0.027 |
massage | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | -7.21e-03 |
Pos logit contributions | |
riages | +0.003 |
drivers | +0.003 |
ados | +0.003 |
pm | +0.003 |
ors | +0.003 |
Neg logit contributions | |
kiss | -0.002 |
copy | -0.002 |
massage | -0.002 |
dance | -0.002 |
last | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
kiss | +0.000 |
copy | +0.000 |
dance | +0.000 |
blow | +0.000 |
last | +0.000 |
Neg logit contributions | |
ados | -0.000 |
riages | -0.000 |
barrels | -0.000 |
pm | -0.000 |
drivers | -0.000 |
Token | Maybe |
Token ID | 13300 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
ados | +0.081 |
riages | +0.081 |
These | +0.080 |
Those | +0.077 |
drivers | +0.077 |
Neg logit contributions | |
kiss | -0.080 |
last | -0.073 |
scrub | -0.070 |
patiently | -0.069 |
check | -0.069 |
Token | he |
Token ID | 339 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
kiss | +0.074 |
copy | +0.065 |
last | +0.063 |
dance | +0.063 |
massage | +0.059 |
Neg logit contributions | |
ados | -0.148 |
riages | -0.144 |
drivers | -0.141 |
pm | -0.139 |
barrels | -0.137 |
Token | could |
Token ID | 714 |
Feature activation | -4.19e+00 |
Pos logit contributions | |
kiss | +0.038 |
copy | +0.035 |
last | +0.032 |
dance | +0.031 |
blow | +0.031 |
Neg logit contributions | |
ados | -0.068 |
riages | -0.066 |
barrels | -0.064 |
pm | -0.064 |
drivers | -0.063 |
Token | not |
Token ID | 407 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
kiss | +0.090 |
copy | +0.081 |
last | +0.071 |
massage | +0.070 |
dab | +0.070 |
Neg logit contributions | |
riages | -0.188 |
ados | -0.186 |
barrels | -0.178 |
drivers | -0.176 |
pm | -0.176 |
Token | catch |
Token ID | 4929 |
Feature activation | +2.29e-01 |
Pos logit contributions | |
kiss | +0.007 |
copy | +0.006 |
last | +0.006 |
dance | +0.005 |
blow | +0.005 |
Neg logit contributions | |
ados | -0.017 |
riages | -0.017 |
barrels | -0.016 |
pm | -0.016 |
drivers | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
riages | +0.011 |
drivers | +0.011 |
pm | +0.011 |
cases | +0.010 |
ables | +0.010 |
Neg logit contributions | |
kiss | -0.004 |
copy | -0.004 |
dance | -0.004 |
last | -0.004 |
rub | -0.004 |
Token | The |
Token ID | 383 |
Feature activation | -6.00e-01 |
Pos logit contributions | |
kiss | +0.007 |
Mom | +0.006 |
copy | +0.006 |
check | +0.006 |
massage | +0.006 |
Neg logit contributions | |
ados | -0.011 |
riages | -0.011 |
barrels | -0.010 |
drivers | -0.010 |
pm | -0.010 |
Token | fle |
Token ID | 5104 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
kiss | +0.016 |
copy | +0.015 |
blow | +0.013 |
scrub | +0.013 |
last | +0.013 |
Neg logit contributions | |
ados | -0.025 |
barrels | -0.024 |
riages | -0.024 |
pm | -0.024 |
drivers | -0.023 |
Token | a |
Token ID | 64 |
Feature activation | -9.60e-01 |
Pos logit contributions | |
kiss | +0.020 |
copy | +0.019 |
last | +0.016 |
dance | +0.015 |
scrub | +0.015 |
Neg logit contributions | |
ados | -0.063 |
barrels | -0.061 |
riages | -0.061 |
pm | -0.059 |
drivers | -0.057 |
Token | was |
Token ID | 373 |
Feature activation | -3.12e-01 |
Pos logit contributions | |
kiss | +0.021 |
copy | +0.019 |
dance | +0.017 |
last | +0.017 |
blow | +0.017 |
Neg logit contributions | |
ados | -0.047 |
riages | -0.045 |
barrels | -0.045 |
drivers | -0.044 |
pm | -0.043 |
Token | very |
Token ID | 845 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
kiss | +0.009 |
copy | +0.008 |
dance | +0.008 |
massage | +0.007 |
last | +0.007 |
Neg logit contributions | |
riages | -0.012 |
ados | -0.012 |
barrels | -0.012 |
drivers | -0.011 |
pm | -0.011 |
Token | fast |
Token ID | 3049 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
riages | +0.116 |
ados | +0.116 |
barrels | +0.112 |
pm | +0.111 |
drivers | +0.107 |
Neg logit contributions | |
kiss | -0.079 |
copy | -0.073 |
last | -0.066 |
massage | -0.066 |
dance | -0.065 |
Token | making |
Token ID | 1642 |
Feature activation | +2.33e-01 |
Pos logit contributions | |
kiss | +0.050 |
copy | +0.046 |
last | +0.041 |
massage | +0.041 |
dance | +0.040 |
Neg logit contributions | |
ados | -0.082 |
riages | -0.080 |
barrels | -0.079 |
pm | -0.076 |
drivers | -0.076 |
Token | noise |
Token ID | 7838 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
ados | +0.011 |
riages | +0.011 |
pm | +0.011 |
drivers | +0.011 |
barrels | +0.011 |
Neg logit contributions | |
kiss | -0.005 |
copy | -0.004 |
last | -0.004 |
dance | -0.004 |
blow | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
ados | +0.088 |
riages | +0.085 |
barrels | +0.084 |
pm | +0.083 |
drivers | +0.082 |
Neg logit contributions | |
kiss | -0.041 |
copy | -0.037 |
last | -0.034 |
dance | -0.034 |
blow | -0.032 |
Token | |
Token ID | 220 |
Feature activation | +4.99e-01 |
Pos logit contributions | |
kiss | +0.006 |
Mom | +0.005 |
cl | +0.005 |
mom | +0.005 |
copy | +0.005 |
Neg logit contributions | |
ados | -0.011 |
riages | -0.010 |
barrels | -0.010 |
pounds | -0.010 |
dollars | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
ados | +0.027 |
riages | +0.026 |
barrels | +0.025 |
pm | +0.025 |
drivers | +0.025 |
Neg logit contributions | |
kiss | -0.008 |
copy | -0.007 |
last | -0.006 |
dance | -0.006 |
check | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -8.52e-01 |
Pos logit contributions | |
kiss | +0.013 |
copy | +0.011 |
last | +0.009 |
scrub | +0.009 |
dance | +0.009 |
Neg logit contributions | |
ados | -0.042 |
riages | -0.040 |
barrels | -0.039 |
pm | -0.038 |
ors | -0.038 |
Token | Sur |
Token ID | 14214 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
kiss | +0.020 |
copy | +0.019 |
scrub | +0.016 |
dance | +0.016 |
cl | +0.016 |
Neg logit contributions | |
ados | -0.042 |
riages | -0.039 |
barrels | -0.038 |
ors | -0.038 |
drivers | -0.037 |
Token | pr |
Token ID | 1050 |
Feature activation | -4.21e-01 |
Pos logit contributions | |
kiss | +0.018 |
copy | +0.017 |
massage | +0.016 |
blow | +0.016 |
dance | +0.016 |
Neg logit contributions | |
riages | -0.019 |
ados | -0.018 |
pm | -0.018 |
barrels | -0.018 |
drivers | -0.018 |
Token | ised |
Token ID | 1417 |
Feature activation | -2.11e+00 |
Pos logit contributions | |
kiss | +0.015 |
copy | +0.014 |
last | +0.014 |
dance | +0.013 |
cl | +0.013 |
Neg logit contributions | |
ados | -0.014 |
barrels | -0.014 |
riages | -0.013 |
drivers | -0.013 |
assing | -0.013 |
Token | , |
Token ID | 11 |
Feature activation | +3.43e-01 |
Pos logit contributions | |
kiss | +0.042 |
copy | +0.041 |
last | +0.041 |
cl | +0.039 |
dance | +0.038 |
Neg logit contributions | |
ados | -0.105 |
riages | -0.102 |
drivers | -0.101 |
cases | -0.097 |
barrels | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | +7.44e-02 |
Pos logit contributions | |
ados | +0.017 |
riages | +0.017 |
ors | +0.017 |
drivers | +0.016 |
pm | +0.016 |
Neg logit contributions | |
kiss | -0.007 |
copy | -0.006 |
cl | -0.006 |
dance | -0.006 |
last | -0.006 |
Token | he |
Token ID | 339 |
Feature activation | +5.66e-01 |
Pos logit contributions | |
ados | +0.041 |
riages | +0.040 |
barrels | +0.038 |
pm | +0.038 |
drivers | +0.037 |
Neg logit contributions | |
kiss | -0.051 |
copy | -0.048 |
last | -0.044 |
blow | -0.044 |
dance | -0.044 |
Token | went |
Token ID | 1816 |
Feature activation | -9.46e-01 |
Pos logit contributions | |
riages | +0.024 |
ados | +0.023 |
barrels | +0.023 |
pm | +0.022 |
drivers | +0.022 |
Neg logit contributions | |
kiss | -0.016 |
copy | -0.014 |
last | -0.013 |
massage | -0.012 |
dance | -0.012 |
Token | on |
Token ID | 319 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
kiss | +0.023 |
copy | +0.021 |
last | +0.018 |
blow | +0.018 |
massage | +0.018 |
Neg logit contributions | |
ados | -0.043 |
riages | -0.043 |
barrels | -0.041 |
pm | -0.040 |
drivers | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ados | +0.068 |
riages | +0.066 |
drivers | +0.063 |
ors | +0.063 |
pm | +0.062 |
Neg logit contributions | |
kiss | -0.031 |
last | -0.029 |
copy | -0.029 |
dance | -0.028 |
check | -0.027 |
Token | long |
Token ID | 890 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
kiss | +0.028 |
copy | +0.025 |
last | +0.023 |
blow | +0.022 |
dance | +0.022 |
Neg logit contributions | |
ados | -0.045 |
riages | -0.045 |
barrels | -0.043 |
pm | -0.043 |
drivers | -0.042 |
Token | walk |
Token ID | 2513 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
ados | +0.015 |
riages | +0.014 |
drivers | +0.014 |
pm | +0.013 |
barrels | +0.013 |
Neg logit contributions | |
kiss | -0.008 |
copy | -0.007 |
dance | -0.007 |
last | -0.007 |
blow | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
kiss | +0.034 |
last | +0.033 |
copy | +0.032 |
dance | +0.031 |
watch | +0.029 |
Neg logit contributions | |
ados | -0.094 |
drivers | -0.089 |
barrels | -0.086 |
pm | -0.086 |
riages | -0.086 |
Token | Soon |
Token ID | 15894 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
ados | +0.011 |
barrels | +0.011 |
riages | +0.011 |
pm | +0.011 |
bank | +0.010 |
Neg logit contributions | |
kiss | -0.008 |
copy | -0.008 |
dance | -0.007 |
blow | -0.007 |
massage | -0.007 |
Token | he |
Token ID | 339 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
last | +0.018 |
kiss | +0.018 |
copy | +0.017 |
watch | +0.017 |
dance | +0.016 |
Neg logit contributions | |
ados | -0.033 |
riages | -0.033 |
drivers | -0.032 |
assing | -0.032 |
ors | -0.032 |
Token | came |
Token ID | 1625 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
kiss | +0.004 |
copy | +0.004 |
massage | +0.003 |
blow | +0.003 |
last | +0.003 |
Neg logit contributions | |
riages | -0.006 |
barrels | -0.006 |
pm | -0.006 |
ables | -0.005 |
bank | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
riages | +0.040 |
barrels | +0.038 |
ados | +0.038 |
pm | +0.037 |
ables | +0.036 |
Neg logit contributions | |
kiss | -0.036 |
copy | -0.032 |
massage | -0.031 |
scrub | -0.031 |
blow | -0.029 |
Token | very |
Token ID | 845 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
ados | +0.005 |
riages | +0.005 |
pm | +0.005 |
ors | +0.005 |
drivers | +0.005 |
Neg logit contributions | |
kiss | -0.003 |
copy | -0.003 |
last | -0.002 |
blow | -0.002 |
dance | -0.002 |
Token | spicy |
Token ID | 26880 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
ados | +0.143 |
riages | +0.140 |
drivers | +0.137 |
pm | +0.133 |
ors | +0.133 |
Neg logit contributions | |
kiss | -0.081 |
copy | -0.076 |
last | -0.071 |
dance | -0.068 |
check | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
riages | +0.133 |
ados | +0.129 |
pm | +0.128 |
AW | +0.128 |
drivers | +0.127 |
Neg logit contributions | |
kiss | -0.059 |
last | -0.056 |
dance | -0.054 |
watch | -0.051 |
copy | -0.050 |
Token | He |
Token ID | 679 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Mom | +0.007 |
kiss | +0.007 |
rubbed | +0.006 |
rub | +0.006 |
check | +0.006 |
Neg logit contributions | |
riages | -0.013 |
tresp | -0.013 |
drivers | -0.013 |
ados | -0.013 |
ors | -0.013 |
Token | decided |
Token ID | 3066 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
ados | +0.007 |
riages | +0.007 |
barrels | +0.007 |
pm | +0.007 |
drivers | +0.007 |
Neg logit contributions | |
kiss | -0.004 |
copy | -0.004 |
last | -0.003 |
dance | -0.003 |
massage | -0.003 |
Token | to |
Token ID | 284 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
ados | +0.027 |
pm | +0.027 |
riages | +0.027 |
drivers | +0.026 |
barrels | +0.026 |
Neg logit contributions | |
kiss | -0.005 |
last | -0.005 |
check | -0.004 |
copy | -0.004 |
watch | -0.004 |
Token | rest |
Token ID | 1334 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
kiss | +0.006 |
copy | +0.005 |
last | +0.005 |
dance | +0.005 |
blow | +0.005 |
Neg logit contributions | |
ados | -0.017 |
riages | -0.016 |
barrels | -0.016 |
pm | -0.016 |
drivers | -0.016 |
Token | under |
Token ID | 739 |
Feature activation | +2.14e-01 |
Pos logit contributions | |
kiss | +0.072 |
copy | +0.066 |
last | +0.061 |
dance | +0.058 |
blow | +0.057 |
Neg logit contributions | |
ados | -0.150 |
riages | -0.148 |
pm | -0.143 |
barrels | -0.142 |
drivers | -0.141 |
Token | the |
Token ID | 262 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
riages | +0.012 |
drivers | +0.012 |
cases | +0.012 |
ors | +0.012 |
ables | +0.012 |
Neg logit contributions | |
last | -0.003 |
kiss | -0.002 |
watch | -0.002 |
another | -0.002 |
dance | -0.002 |
Token | tree |
Token ID | 5509 |
Feature activation | -4.81e-01 |
Pos logit contributions | |
riages | +0.007 |
drivers | +0.007 |
ados | +0.007 |
pm | +0.007 |
ors | +0.006 |
Neg logit contributions | |
kiss | -0.004 |
last | -0.004 |
copy | -0.003 |
dance | -0.003 |
watch | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | +5.44e-01 |
Pos logit contributions | |
kiss | +0.010 |
last | +0.009 |
copy | +0.009 |
dance | +0.008 |
watch | +0.008 |
Neg logit contributions | |
ados | -0.023 |
riages | -0.023 |
drivers | -0.022 |
pm | -0.022 |
barrels | -0.022 |
Token | It |
Token ID | 632 |
Feature activation | -2.54e-01 |
Pos logit contributions | |
kiss | +0.024 |
copy | +0.023 |
last | +0.021 |
dance | +0.021 |
blow | +0.021 |
Neg logit contributions | |
ados | -0.017 |
riages | -0.017 |
barrels | -0.016 |
pm | -0.015 |
drivers | -0.015 |
Token | was |
Token ID | 373 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
kiss | +0.006 |
copy | +0.005 |
last | +0.005 |
dance | +0.004 |
blow | +0.004 |
Neg logit contributions | |
ados | -0.012 |
riages | -0.012 |
barrels | -0.011 |
pm | -0.011 |
drivers | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
kiss | +0.004 |
copy | +0.004 |
last | +0.003 |
dance | +0.003 |
blow | +0.003 |
Neg logit contributions | |
ados | -0.007 |
riages | -0.007 |
barrels | -0.007 |
pm | -0.007 |
drivers | -0.007 |
Token | great |
Token ID | 1049 |
Feature activation | +7.50e-01 |
Pos logit contributions | |
riages | +0.014 |
barrels | +0.013 |
ados | +0.013 |
pm | +0.013 |
tresp | +0.013 |
Neg logit contributions | |
kiss | -0.009 |
copy | -0.009 |
scrub | -0.008 |
blow | -0.008 |
dab | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | -3.91e-01 |
Pos logit contributions | |
ados | +0.028 |
barrels | +0.028 |
riages | +0.028 |
pm | +0.027 |
bullets | +0.027 |
Neg logit contributions | |
kiss | -0.019 |
copy | -0.019 |
dab | -0.017 |
rubbed | -0.017 |
blow | -0.017 |
Token | and |
Token ID | 290 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
kiss | +0.010 |
copy | +0.009 |
last | +0.009 |
dance | +0.008 |
blow | +0.008 |
Neg logit contributions | |
ados | -0.017 |
riages | -0.017 |
barrels | -0.016 |
pm | -0.016 |
drivers | -0.016 |
Token | they |
Token ID | 484 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
kiss | +0.016 |
copy | +0.015 |
last | +0.013 |
dance | +0.013 |
blow | +0.013 |
Neg logit contributions | |
ados | -0.018 |
riages | -0.018 |
barrels | -0.017 |
pm | -0.017 |
drivers | -0.017 |
Token | would |
Token ID | 561 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
riages | +0.117 |
ados | +0.116 |
barrels | +0.110 |
pm | +0.108 |
drivers | +0.107 |
Neg logit contributions | |
kiss | -0.092 |
copy | -0.087 |
massage | -0.078 |
last | -0.078 |
check | -0.077 |
Token | always |
Token ID | 1464 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
kiss | +0.054 |
copy | +0.050 |
last | +0.043 |
dance | +0.042 |
blow | +0.041 |
Neg logit contributions | |
ados | -0.128 |
riages | -0.127 |
barrels | -0.123 |
pm | -0.122 |
drivers | -0.120 |
Token | remember |
Token ID | 3505 |
Feature activation | -8.48e-01 |
Pos logit contributions | |
kiss | +0.027 |
copy | +0.023 |
last | +0.020 |
massage | +0.020 |
dab | +0.020 |
Neg logit contributions | |
riages | -0.074 |
ados | -0.073 |
barrels | -0.072 |
pm | -0.071 |
drivers | -0.069 |
Token | it |
Token ID | 340 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
kiss | +0.019 |
copy | +0.018 |
last | +0.017 |
dance | +0.016 |
check | +0.016 |
Neg logit contributions | |
ados | -0.040 |
riages | -0.039 |
pm | -0.038 |
drivers | -0.038 |
barrels | -0.037 |
Token | ully |
Token ID | 2132 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
kiss | +0.061 |
copy | +0.057 |
last | +0.053 |
dance | +0.052 |
blow | +0.052 |
Neg logit contributions | |
ados | -0.067 |
riages | -0.064 |
barrels | -0.063 |
pm | -0.062 |
drivers | -0.061 |
Token | couldn |
Token ID | 3521 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
kiss | +0.049 |
copy | +0.043 |
last | +0.040 |
dance | +0.038 |
blow | +0.037 |
Neg logit contributions | |
ados | -0.119 |
riages | -0.116 |
barrels | -0.113 |
drivers | -0.112 |
pm | -0.112 |
Token | 't |
Token ID | 470 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
kiss | +0.073 |
last | +0.068 |
copy | +0.066 |
blow | +0.064 |
dance | +0.064 |
Neg logit contributions | |
ados | -0.086 |
riages | -0.085 |
barrels | -0.084 |
pm | -0.081 |
files | -0.079 |
Token | decide |
Token ID | 5409 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
kiss | +0.025 |
copy | +0.022 |
massage | +0.020 |
dab | +0.020 |
last | +0.019 |
Neg logit contributions | |
riages | -0.044 |
ados | -0.044 |
barrels | -0.042 |
pm | -0.041 |
bank | -0.041 |
Token | which |
Token ID | 543 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
kiss | +0.006 |
copy | +0.005 |
last | +0.005 |
check | +0.005 |
dance | +0.005 |
Neg logit contributions | |
ados | -0.006 |
riages | -0.006 |
pm | -0.006 |
barrels | -0.006 |
drivers | -0.006 |
Token | one |
Token ID | 530 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
kiss | +0.050 |
last | +0.048 |
copy | +0.042 |
check | +0.041 |
watch | +0.039 |
Neg logit contributions | |
riages | -0.132 |
drivers | -0.126 |
ados | -0.126 |
pm | -0.126 |
ors | -0.124 |
Token | to |
Token ID | 284 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
kiss | +0.118 |
copy | +0.109 |
last | +0.100 |
dance | +0.097 |
blow | +0.095 |
Neg logit contributions | |
ados | -0.201 |
riages | -0.198 |
barrels | -0.191 |
pm | -0.190 |
drivers | -0.188 |
Token | eat |
Token ID | 4483 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
kiss | +0.023 |
copy | +0.021 |
last | +0.018 |
dance | +0.018 |
blow | +0.018 |
Neg logit contributions | |
ados | -0.050 |
riages | -0.049 |
barrels | -0.048 |
pm | -0.048 |
drivers | -0.047 |
Token | first |
Token ID | 717 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
kiss | +0.032 |
last | +0.032 |
copy | +0.027 |
rose | +0.027 |
dance | +0.026 |
Neg logit contributions | |
riages | -0.081 |
drivers | -0.078 |
pm | -0.077 |
ados | -0.077 |
cases | -0.075 |
Token | , |
Token ID | 11 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
kiss | +0.024 |
last | +0.022 |
copy | +0.021 |
dance | +0.019 |
happily | +0.019 |
Neg logit contributions | |
riages | -0.054 |
ados | -0.052 |
drivers | -0.051 |
pm | -0.050 |
barrels | -0.050 |
Token | so |
Token ID | 523 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
ados | +0.007 |
pm | +0.007 |
drivers | +0.007 |
riages | +0.007 |
ors | +0.007 |
Neg logit contributions | |
kiss | -0.005 |
dance | -0.004 |
copy | -0.004 |
cl | -0.004 |
blow | -0.004 |
Token | uffy |
Token ID | 15352 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
kiss | +0.068 |
copy | +0.065 |
last | +0.060 |
blow | +0.059 |
dance | +0.059 |
Neg logit contributions | |
ados | -0.056 |
riages | -0.055 |
pm | -0.052 |
barrels | -0.052 |
drivers | -0.051 |
Token | were |
Token ID | 547 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
kiss | +0.008 |
copy | +0.007 |
last | +0.006 |
dance | +0.006 |
blow | +0.006 |
Neg logit contributions | |
ados | -0.017 |
riages | -0.016 |
barrels | -0.016 |
pm | -0.016 |
drivers | -0.016 |
Token | tired |
Token ID | 10032 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
ados | +0.033 |
riages | +0.033 |
barrels | +0.032 |
pm | +0.031 |
drivers | +0.031 |
Neg logit contributions | |
kiss | -0.019 |
copy | -0.018 |
massage | -0.015 |
dance | -0.015 |
last | -0.015 |
Token | . |
Token ID | 13 |
Feature activation | -6.23e-01 |
Pos logit contributions | |
ados | +0.080 |
riages | +0.079 |
barrels | +0.078 |
pm | +0.074 |
drivers | +0.074 |
Neg logit contributions | |
kiss | -0.070 |
copy | -0.069 |
massage | -0.060 |
dab | -0.060 |
blow | -0.059 |
Token | They |
Token ID | 1119 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
kiss | +0.019 |
Mom | +0.018 |
massage | +0.017 |
cl | +0.017 |
rub | +0.017 |
Neg logit contributions | |
barrels | -0.023 |
riages | -0.023 |
pm | -0.023 |
ados | -0.023 |
drivers | -0.023 |
Token | lay |
Token ID | 3830 |
Feature activation | -4.52e+00 |
Pos logit contributions | |
riages | +0.070 |
ados | +0.069 |
barrels | +0.068 |
pm | +0.066 |
drivers | +0.064 |
Neg logit contributions | |
kiss | -0.037 |
copy | -0.036 |
last | -0.031 |
massage | -0.030 |
dance | -0.030 |
Token | down |
Token ID | 866 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
kiss | +0.129 |
copy | +0.125 |
dance | +0.109 |
blow | +0.107 |
massage | +0.107 |
Neg logit contributions | |
ados | -0.189 |
riages | -0.180 |
barrels | -0.178 |
drivers | -0.167 |
pm | -0.167 |
Token | on |
Token ID | 319 |
Feature activation | +8.31e-01 |
Pos logit contributions | |
copy | +0.056 |
kiss | +0.055 |
massage | +0.050 |
dab | +0.048 |
dance | +0.047 |
Neg logit contributions | |
ados | -0.058 |
riages | -0.056 |
barrels | -0.056 |
guns | -0.051 |
ards | -0.051 |
Token | the |
Token ID | 262 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ados | +0.038 |
riages | +0.037 |
barrels | +0.036 |
pm | +0.036 |
drivers | +0.035 |
Neg logit contributions | |
kiss | -0.020 |
copy | -0.018 |
last | -0.016 |
dance | -0.016 |
blow | -0.016 |
Token | soft |
Token ID | 2705 |
Feature activation | -2.72e+00 |
Pos logit contributions | |
kiss | +0.005 |
copy | +0.005 |
check | +0.004 |
admire | +0.004 |
blow | +0.004 |
Neg logit contributions | |
ados | -0.005 |
riages | -0.005 |
barrels | -0.005 |
ors | -0.005 |
pm | -0.004 |
Token | grass |
Token ID | 8701 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
kiss | +0.080 |
copy | +0.072 |
last | +0.068 |
dance | +0.068 |
blow | +0.066 |
Neg logit contributions | |
ados | -0.110 |
riages | -0.108 |
pm | -0.105 |
barrels | -0.104 |
drivers | -0.104 |
Token | " |
Token ID | 366 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
ados | +0.047 |
riages | +0.046 |
barrels | +0.045 |
pm | +0.045 |
drivers | +0.043 |
Neg logit contributions | |
kiss | -0.035 |
copy | -0.031 |
blow | -0.030 |
dance | -0.029 |
last | -0.029 |
Token | Can |
Token ID | 6090 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
ados | +0.157 |
riages | +0.153 |
barrels | +0.149 |
pm | +0.148 |
drivers | +0.145 |
Neg logit contributions | |
kiss | -0.094 |
copy | -0.087 |
last | -0.080 |
dance | -0.078 |
blow | -0.077 |
Token | I |
Token ID | 314 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
kiss | +0.056 |
last | +0.049 |
copy | +0.049 |
dance | +0.048 |
watch | +0.048 |
Neg logit contributions | |
barrels | -0.076 |
riages | -0.076 |
ados | -0.076 |
pm | -0.073 |
drivers | -0.073 |
Token | put |
Token ID | 1234 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
dab | +0.029 |
patiently | +0.029 |
scrub | +0.027 |
reaction | +0.027 |
Acc | +0.026 |
Neg logit contributions | |
riages | -0.066 |
These | -0.062 |
ados | -0.061 |
clubs | -0.061 |
pistols | -0.059 |
Token | my |
Token ID | 616 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
kiss | +0.078 |
copy | +0.073 |
last | +0.068 |
dance | +0.067 |
blow | +0.067 |
Neg logit contributions | |
ados | -0.084 |
riages | -0.082 |
barrels | -0.079 |
pm | -0.078 |
drivers | -0.077 |
Token | favorite |
Token ID | 4004 |
Feature activation | -4.97e+00 |
Pos logit contributions | |
kiss | +0.009 |
copy | +0.008 |
scrub | +0.008 |
check | +0.008 |
dab | +0.008 |
Neg logit contributions | |
ados | -0.011 |
riages | -0.011 |
barrels | -0.010 |
pistols | -0.010 |
bullets | -0.010 |
Token | song |
Token ID | 3496 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
kiss | +0.145 |
copy | +0.134 |
check | +0.127 |
scrub | +0.125 |
last | +0.125 |
Neg logit contributions | |
ados | -0.193 |
riages | -0.184 |
barrels | -0.179 |
pm | -0.173 |
ors | -0.170 |
Token | on |
Token ID | 319 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
kiss | +0.006 |
copy | +0.005 |
blow | +0.005 |
dab | +0.005 |
scrub | +0.005 |
Neg logit contributions | |
ados | -0.008 |
riages | -0.008 |
barrels | -0.008 |
drivers | -0.008 |
ables | -0.008 |
Token | your |
Token ID | 534 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
riages | +0.055 |
ados | +0.054 |
drivers | +0.054 |
barrels | +0.053 |
pm | +0.053 |
Neg logit contributions | |
kiss | -0.038 |
copy | -0.036 |
dance | -0.033 |
massage | -0.033 |
last | -0.033 |
Token | guitar |
Token ID | 10047 |
Feature activation | +6.47e-01 |
Pos logit contributions | |
ados | +0.020 |
riages | +0.020 |
barrels | +0.019 |
pm | +0.018 |
ors | +0.018 |
Neg logit contributions | |
kiss | -0.016 |
copy | -0.015 |
check | -0.014 |
scrub | -0.014 |
admire | -0.014 |
Token | ?" |
Token ID | 1701 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
ados | +0.027 |
riages | +0.027 |
barrels | +0.026 |
pm | +0.026 |
drivers | +0.025 |
Neg logit contributions | |
kiss | -0.017 |
copy | -0.016 |
last | -0.015 |
dance | -0.014 |
blow | -0.014 |
Token | and |
Token ID | 290 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
ados | +0.163 |
pm | +0.163 |
riages | +0.157 |
Sale | +0.157 |
Those | +0.156 |
Neg logit contributions | |
last | -0.050 |
kiss | -0.048 |
copy | -0.039 |
folded | -0.038 |
happily | -0.038 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +5.64e-01 |
Pos logit contributions | |
kiss | +0.017 |
last | +0.016 |
copy | +0.015 |
dance | +0.015 |
rubbed | +0.015 |
Neg logit contributions | |
ados | -0.014 |
pm | -0.014 |
ors | -0.013 |
riages | -0.013 |
assing | -0.013 |
Token | jewels |
Token ID | 42701 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ados | +0.015 |
riages | +0.015 |
pm | +0.015 |
assing | +0.015 |
Those | +0.014 |
Neg logit contributions | |
kiss | -0.023 |
copy | -0.021 |
last | -0.021 |
dance | -0.021 |
cl | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -8.46e-01 |
Pos logit contributions | |
ados | +0.088 |
pm | +0.088 |
barrels | +0.086 |
riages | +0.086 |
Sale | +0.084 |
Neg logit contributions | |
kiss | -0.034 |
last | -0.032 |
copy | -0.029 |
dance | -0.026 |
happily | -0.026 |
Token | Tim |
Token ID | 5045 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
kiss | +0.020 |
Mom | +0.018 |
happily | +0.016 |
check | +0.016 |
mum | +0.016 |
Neg logit contributions | |
riages | -0.038 |
ados | -0.038 |
barrels | -0.037 |
drivers | -0.036 |
tresp | -0.036 |
Token | my |
Token ID | 1820 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
kiss | +0.024 |
copy | +0.022 |
massage | +0.019 |
blow | +0.019 |
dance | +0.018 |
Neg logit contributions | |
riages | -0.050 |
ados | -0.050 |
barrels | -0.049 |
pm | -0.048 |
drivers | -0.047 |
Token | was |
Token ID | 373 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
kiss | +0.096 |
copy | +0.088 |
last | +0.078 |
dance | +0.075 |
massage | +0.075 |
Neg logit contributions | |
ados | -0.204 |
riages | -0.201 |
barrels | -0.196 |
pm | -0.194 |
drivers | -0.192 |
Token | so |
Token ID | 523 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
kiss | +0.044 |
copy | +0.042 |
last | +0.038 |
blow | +0.038 |
massage | +0.038 |
Neg logit contributions | |
ados | -0.036 |
riages | -0.035 |
barrels | -0.033 |
pm | -0.033 |
drivers | -0.032 |
Token | happy |
Token ID | 3772 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
riages | +0.075 |
ados | +0.074 |
pm | +0.073 |
barrels | +0.072 |
drivers | +0.069 |
Neg logit contributions | |
kiss | -0.034 |
copy | -0.031 |
massage | -0.028 |
last | -0.027 |
blow | -0.026 |
Token | that |
Token ID | 326 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
riages | +0.010 |
ados | +0.010 |
pm | +0.010 |
drivers | +0.010 |
barrels | +0.010 |
Neg logit contributions | |
kiss | -0.004 |
copy | -0.004 |
dance | -0.004 |
last | -0.004 |
cl | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
kiss | +0.035 |
copy | +0.032 |
last | +0.031 |
dance | +0.031 |
massage | +0.029 |
Neg logit contributions | |
ados | -0.041 |
riages | -0.041 |
pm | -0.039 |
drivers | -0.039 |
barrels | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
ados | +0.040 |
riages | +0.039 |
barrels | +0.038 |
pm | +0.037 |
drivers | +0.037 |
Neg logit contributions | |
kiss | -0.024 |
copy | -0.022 |
last | -0.020 |
dance | -0.020 |
blow | -0.020 |
Token |
|
Token ID | 198 |
Feature activation | -1.92e-01 |
Pos logit contributions | |
kiss | +0.000 |
copy | +0.000 |
last | +0.000 |
dance | +0.000 |
blow | +0.000 |
Neg logit contributions | |
ados | -0.000 |
riages | -0.000 |
barrels | -0.000 |
pm | -0.000 |
drivers | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -2.27e-01 |
Pos logit contributions | |
kiss | +0.006 |
copy | +0.006 |
dance | +0.005 |
blow | +0.005 |
last | +0.005 |
Neg logit contributions | |
riages | -0.007 |
ables | -0.007 |
pm | -0.007 |
barrels | -0.007 |
drivers | -0.007 |
Token | G |
Token ID | 38 |
Feature activation | -4.19e-01 |
Pos logit contributions | |
kiss | +0.006 |
copy | +0.006 |
last | +0.006 |
blow | +0.005 |
dance | +0.005 |
Neg logit contributions | |
ados | -0.010 |
riages | -0.009 |
barrels | -0.009 |
drivers | -0.009 |
pm | -0.009 |
Token | abe |
Token ID | 11231 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
kiss | +0.015 |
copy | +0.015 |
last | +0.014 |
blow | +0.014 |
check | +0.014 |
Neg logit contributions | |
ados | -0.013 |
riages | -0.013 |
barrels | -0.012 |
pm | -0.012 |
drivers | -0.012 |
Token | then |
Token ID | 788 |
Feature activation | -4.34e+00 |
Pos logit contributions | |
kiss | +0.010 |
copy | +0.009 |
last | +0.008 |
dance | +0.008 |
massage | +0.008 |
Neg logit contributions | |
ados | -0.018 |
riages | -0.018 |
barrels | -0.018 |
pm | -0.018 |
drivers | -0.017 |
Token | got |
Token ID | 1392 |
Feature activation | -2.85e-01 |
Pos logit contributions | |
kiss | +0.102 |
copy | +0.099 |
last | +0.089 |
massage | +0.082 |
dance | +0.081 |
Neg logit contributions | |
riages | -0.188 |
ados | -0.187 |
barrels | -0.182 |
pm | -0.176 |
drivers | -0.174 |
Token | some |
Token ID | 617 |
Feature activation | -2.90e-01 |
Pos logit contributions | |
kiss | +0.008 |
copy | +0.007 |
dance | +0.006 |
blow | +0.006 |
massage | +0.006 |
Neg logit contributions | |
ados | -0.012 |
barrels | -0.012 |
riages | -0.011 |
files | -0.011 |
sacks | -0.011 |
Token | paint |
Token ID | 7521 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
kiss | +0.010 |
copy | +0.009 |
last | +0.008 |
dance | +0.008 |
blow | +0.008 |
Neg logit contributions | |
ados | -0.010 |
riages | -0.010 |
barrels | -0.010 |
pm | -0.009 |
drivers | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
kiss | +0.000 |
copy | +0.000 |
last | +0.000 |
dance | +0.000 |
blow | +0.000 |
Neg logit contributions | |
ados | -0.001 |
riages | -0.001 |
pm | -0.001 |
barrels | -0.001 |
drivers | -0.001 |
Token | painted |
Token ID | 13055 |
Feature activation | -7.64e-01 |
Pos logit contributions | |
kiss | +0.012 |
copy | +0.011 |
massage | +0.010 |
last | +0.010 |
check | +0.010 |
Neg logit contributions | |
ados | -0.018 |
barrels | -0.018 |
riages | -0.017 |
drivers | -0.017 |
files | -0.016 |