Token | . |
Token ID | 13 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
ow | +0.004 |
feelings | +0.003 |
Christmas | +0.003 |
happen | +0.003 |
stories | +0.003 |
Neg logit contributions | |
Whoever | -0.008 |
ursion | -0.008 |
ounded | -0.008 |
ooters | -0.008 |
reaching | -0.008 |
Token | I |
Token ID | 314 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
ounded | +0.028 |
ursion | +0.026 |
ooters | +0.026 |
sc | +0.025 |
Whoever | +0.025 |
Neg logit contributions | |
ow | -0.037 |
Christmas | -0.031 |
feelings | -0.029 |
stories | -0.028 |
soup | -0.028 |
Token | am |
Token ID | 716 |
Feature activation | -2.32e-02 |
Pos logit contributions | |
Whoever | +0.098 |
ursion | +0.094 |
ounded | +0.094 |
ooters | +0.091 |
sc | +0.090 |
Neg logit contributions | |
ow | -0.101 |
Christmas | -0.080 |
feelings | -0.080 |
happen | -0.080 |
stories | -0.077 |
Token | sorry |
Token ID | 7926 |
Feature activation | -5.10e-03 |
Pos logit contributions | |
ow | +0.036 |
Christmas | +0.029 |
all | +0.027 |
soup | +0.026 |
happen | +0.026 |
Neg logit contributions | |
ounded | -0.046 |
sc | -0.046 |
Whoever | -0.046 |
ursion | -0.046 |
ooters | -0.045 |
Token | I |
Token ID | 314 |
Feature activation | +8.46e-02 |
Pos logit contributions | |
ow | +0.009 |
feelings | +0.007 |
Christmas | +0.007 |
happen | +0.007 |
stories | +0.006 |
Neg logit contributions | |
Whoever | -0.009 |
ursion | -0.009 |
ounded | -0.009 |
sc | -0.008 |
ooters | -0.008 |
Token | called |
Token ID | 1444 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
Whoever | +0.166 |
ounded | +0.161 |
ursion | +0.161 |
sc | +0.158 |
ooters | +0.157 |
Neg logit contributions | |
ow | -0.123 |
Christmas | -0.092 |
feelings | -0.092 |
happen | -0.089 |
stories | -0.088 |
Token | you |
Token ID | 345 |
Feature activation | +5.52e-02 |
Pos logit contributions | |
Whoever | +0.366 |
ursion | +0.360 |
ounded | +0.358 |
sc | +0.346 |
ooters | +0.344 |
Neg logit contributions | |
ow | -0.245 |
Christmas | -0.182 |
feelings | -0.178 |
happen | -0.177 |
stories | -0.174 |
Token | lazy |
Token ID | 16931 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Whoever | +0.103 |
ounded | +0.100 |
ursion | +0.099 |
ooters | +0.097 |
sc | +0.097 |
Neg logit contributions | |
ow | -0.086 |
feelings | -0.065 |
Christmas | -0.064 |
happen | -0.064 |
stories | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
Whoever | +0.032 |
ursion | +0.031 |
ounded | +0.030 |
ooters | +0.030 |
sc | +0.030 |
Neg logit contributions | |
ow | -0.023 |
Christmas | -0.017 |
happen | -0.017 |
feelings | -0.017 |
stories | -0.016 |
Token | You |
Token ID | 921 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
ounded | +0.045 |
ursion | +0.042 |
Whoever | +0.042 |
ooters | +0.041 |
sc | +0.041 |
Neg logit contributions | |
ow | -0.061 |
Christmas | -0.050 |
feelings | -0.048 |
soup | -0.045 |
stories | -0.045 |
Token | are |
Token ID | 389 |
Feature activation | -1.49e-02 |
Pos logit contributions | |
Whoever | +0.081 |
ounded | +0.078 |
ursion | +0.077 |
ooters | +0.075 |
sc | +0.075 |
Neg logit contributions | |
ow | -0.060 |
Christmas | -0.045 |
lf | -0.045 |
feelings | -0.045 |
happen | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | +3.37e-02 |
Pos logit contributions | |
Whoever | +0.106 |
ursion | +0.102 |
ounded | +0.101 |
ooters | +0.099 |
sc | +0.099 |
Neg logit contributions | |
ow | -0.104 |
happen | -0.079 |
Christmas | -0.079 |
feelings | -0.079 |
lf | -0.077 |
Token | You |
Token ID | 921 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
ounded | +0.051 |
ursion | +0.047 |
ooters | +0.046 |
sc | +0.044 |
Whoever | +0.044 |
Neg logit contributions | |
ow | -0.068 |
Christmas | -0.058 |
feelings | -0.054 |
stories | -0.053 |
each | -0.053 |
Token | can |
Token ID | 460 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
Whoever | +0.083 |
ounded | +0.080 |
ursion | +0.078 |
sc | +0.076 |
ooters | +0.074 |
Neg logit contributions | |
ow | -0.069 |
lf | -0.058 |
ating | -0.055 |
urg | -0.054 |
Christmas | -0.053 |
Token | play |
Token ID | 711 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
Whoever | +0.023 |
ursion | +0.021 |
ooters | +0.020 |
reaching | +0.020 |
ounded | +0.020 |
Neg logit contributions | |
ow | -0.018 |
feelings | -0.014 |
stories | -0.013 |
ling | -0.013 |
happen | -0.013 |
Token | hop |
Token ID | 1725 |
Feature activation | -7.67e-03 |
Pos logit contributions | |
ow | +0.056 |
ling | +0.043 |
urg | +0.041 |
Christmas | +0.041 |
feelings | +0.040 |
Neg logit contributions | |
Whoever | -0.072 |
sc | -0.071 |
Ready | -0.066 |
ounded | -0.065 |
Ready | -0.064 |
Token | sc |
Token ID | 1416 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
ow | +0.013 |
feelings | +0.010 |
Christmas | +0.010 |
happen | +0.010 |
stories | +0.010 |
Neg logit contributions | |
Whoever | -0.013 |
ursion | -0.013 |
ounded | -0.013 |
sc | -0.012 |
ooters | -0.012 |
Token | ot |
Token ID | 313 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
Whoever | +0.327 |
sc | +0.316 |
ooters | +0.313 |
ursion | +0.307 |
ounded | +0.306 |
Neg logit contributions | |
ow | -0.285 |
Christmas | -0.213 |
happen | -0.212 |
feelings | -0.204 |
lf | -0.202 |
Token | ch |
Token ID | 354 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
ow | +0.059 |
happen | +0.050 |
Christmas | +0.050 |
feelings | +0.050 |
stories | +0.049 |
Neg logit contributions | |
Whoever | -0.028 |
ounded | -0.027 |
ursion | -0.027 |
ooters | -0.025 |
reaching | -0.025 |
Token | again |
Token ID | 757 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
Whoever | +0.075 |
sc | +0.074 |
ursion | +0.070 |
ooters | +0.070 |
ounded | +0.069 |
Neg logit contributions | |
ow | -0.073 |
Christmas | -0.059 |
feelings | -0.057 |
happen | -0.056 |
stories | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +2.35e-02 |
Pos logit contributions | |
ow | +0.033 |
feelings | +0.025 |
happen | +0.025 |
Christmas | +0.025 |
each | +0.024 |
Neg logit contributions | |
Whoever | -0.043 |
ounded | -0.042 |
ursion | -0.042 |
ooters | -0.040 |
reaching | -0.040 |
Token | But |
Token ID | 887 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
ounded | +0.041 |
ursion | +0.036 |
ooters | +0.033 |
hole | +0.033 |
saw | +0.033 |
Neg logit contributions | |
ow | -0.042 |
Christmas | -0.036 |
songs | -0.034 |
each | -0.034 |
jokes | -0.034 |
Token | hurt |
Token ID | 5938 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Whoever | +0.024 |
sc | +0.023 |
ooters | +0.022 |
reaching | +0.022 |
wanna | +0.022 |
Neg logit contributions | |
ating | -0.017 |
ow | -0.016 |
lf | -0.016 |
stories | -0.016 |
ities | -0.015 |
Token | you |
Token ID | 345 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
ow | +0.037 |
Christmas | +0.028 |
feelings | +0.028 |
happen | +0.028 |
stories | +0.027 |
Neg logit contributions | |
Whoever | -0.038 |
ursion | -0.037 |
ounded | -0.037 |
ooters | -0.036 |
sc | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | +2.87e-03 |
Pos logit contributions | |
Whoever | +0.046 |
ursion | +0.045 |
ounded | +0.045 |
ooters | +0.043 |
sc | +0.043 |
Neg logit contributions | |
ow | -0.036 |
feelings | -0.027 |
Christmas | -0.027 |
happen | -0.027 |
stories | -0.026 |
Token | They |
Token ID | 1119 |
Feature activation | +4.69e-03 |
Pos logit contributions | |
ounded | +0.005 |
ursion | +0.005 |
ooters | +0.004 |
reaching | +0.004 |
sc | +0.004 |
Neg logit contributions | |
ow | -0.005 |
feelings | -0.004 |
each | -0.004 |
Christmas | -0.004 |
fur | -0.004 |
Token | are |
Token ID | 389 |
Feature activation | -1.40e-02 |
Pos logit contributions | |
Whoever | +0.008 |
ounded | +0.008 |
ursion | +0.008 |
sc | +0.008 |
ooters | +0.008 |
Neg logit contributions | |
ow | -0.007 |
lf | -0.006 |
Christmas | -0.006 |
feelings | -0.006 |
stories | -0.006 |
Token | called |
Token ID | 1444 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
ow | +0.019 |
Christmas | +0.014 |
happen | +0.014 |
feelings | +0.014 |
stories | +0.013 |
Neg logit contributions | |
Whoever | -0.029 |
ounded | -0.028 |
ursion | -0.028 |
ooters | -0.027 |
sc | -0.027 |
Token | sp |
Token ID | 599 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Whoever | +0.342 |
ounded | +0.325 |
ooters | +0.324 |
ursion | +0.324 |
sc | +0.322 |
Neg logit contributions | |
ow | -0.238 |
happen | -0.177 |
feelings | -0.175 |
Christmas | -0.173 |
lf | -0.172 |
Token | ines |
Token ID | 1127 |
Feature activation | +4.14e-02 |
Pos logit contributions | |
ow | +0.020 |
Christmas | +0.017 |
feelings | +0.017 |
stories | +0.016 |
happen | +0.016 |
Neg logit contributions | |
ooters | -0.020 |
Whoever | -0.020 |
ursion | -0.019 |
sc | -0.019 |
ounded | -0.019 |
Token | ." |
Token ID | 526 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
Whoever | +0.078 |
sc | +0.075 |
ooters | +0.074 |
ounded | +0.073 |
ursion | +0.073 |
Neg logit contributions | |
ow | -0.059 |
feelings | -0.044 |
ad | -0.043 |
happen | -0.042 |
ity | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +6.05e-03 |
Pos logit contributions | |
ounded | +0.048 |
ursion | +0.048 |
ooters | +0.046 |
Whoever | +0.045 |
sc | +0.045 |
Neg logit contributions | |
ow | -0.033 |
each | -0.025 |
Christmas | -0.025 |
stories | -0.025 |
feelings | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -5.90e-03 |
Pos logit contributions | |
ounded | +0.012 |
ursion | +0.012 |
Whoever | +0.011 |
ooters | +0.011 |
sc | +0.011 |
Neg logit contributions | |
ow | -0.009 |
each | -0.007 |
Christmas | -0.007 |
happen | -0.007 |
stories | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +2.91e-02 |
Pos logit contributions | |
ow | +0.003 |
Christmas | +0.002 |
happen | +0.002 |
lf | +0.002 |
feelings | +0.002 |
Neg logit contributions | |
Whoever | -0.004 |
ursion | -0.004 |
sc | -0.004 |
ounded | -0.004 |
reaching | -0.004 |
Token | I |
Token ID | 314 |
Feature activation | +6.23e-02 |
Pos logit contributions | |
ounded | +0.041 |
Whoever | +0.039 |
ooters | +0.038 |
ursion | +0.038 |
sc | +0.037 |
Neg logit contributions | |
ow | -0.060 |
Christmas | -0.050 |
feelings | -0.048 |
happen | -0.046 |
glow | -0.046 |
Token | 'm |
Token ID | 1101 |
Feature activation | +3.95e-03 |
Pos logit contributions | |
Whoever | +0.107 |
sc | +0.105 |
ounded | +0.104 |
ursion | +0.102 |
Ready | +0.102 |
Neg logit contributions | |
ow | -0.109 |
lf | -0.085 |
Christmas | -0.084 |
feelings | -0.079 |
urg | -0.079 |
Token | sorry |
Token ID | 7926 |
Feature activation | -7.79e-03 |
Pos logit contributions | |
sc | +0.009 |
ursion | +0.008 |
ounded | +0.008 |
ooters | +0.008 |
Ready | +0.008 |
Neg logit contributions | |
ow | -0.006 |
all | -0.005 |
Christmas | -0.004 |
happen | -0.004 |
soup | -0.004 |
Token | I |
Token ID | 314 |
Feature activation | +8.02e-02 |
Pos logit contributions | |
ow | +0.012 |
Christmas | +0.009 |
feelings | +0.009 |
happen | +0.009 |
stories | +0.009 |
Neg logit contributions | |
Whoever | -0.015 |
ursion | -0.014 |
ounded | -0.014 |
ooters | -0.014 |
sc | -0.014 |
Token | called |
Token ID | 1444 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
Whoever | +0.154 |
ounded | +0.149 |
sc | +0.148 |
ursion | +0.148 |
Ready | +0.143 |
Neg logit contributions | |
ow | -0.120 |
Christmas | -0.094 |
feelings | -0.091 |
stories | -0.089 |
lf | -0.089 |
Token | you |
Token ID | 345 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
Whoever | +0.312 |
ursion | +0.303 |
ounded | +0.302 |
ooters | +0.291 |
sc | +0.291 |
Neg logit contributions | |
ow | -0.282 |
feelings | -0.217 |
Christmas | -0.217 |
happen | -0.216 |
stories | -0.210 |
Token | mean |
Token ID | 1612 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
Whoever | +0.104 |
ounded | +0.100 |
ursion | +0.100 |
sc | +0.098 |
ooters | +0.098 |
Neg logit contributions | |
ow | -0.073 |
Christmas | -0.054 |
feelings | -0.053 |
happen | -0.053 |
lf | -0.051 |
Token | . |
Token ID | 13 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
ow | +0.015 |
Christmas | +0.011 |
happen | +0.011 |
feelings | +0.011 |
stories | +0.011 |
Neg logit contributions | |
Whoever | -0.020 |
ursion | -0.019 |
ounded | -0.019 |
ooters | -0.019 |
sc | -0.019 |
Token | You |
Token ID | 921 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
ounded | +0.041 |
ursion | +0.038 |
ooters | +0.037 |
Whoever | +0.037 |
sc | +0.036 |
Neg logit contributions | |
ow | -0.058 |
feelings | -0.047 |
Christmas | -0.047 |
stories | -0.045 |
jokes | -0.045 |
Token | are |
Token ID | 389 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
Whoever | +0.081 |
ounded | +0.077 |
ursion | +0.075 |
sc | +0.074 |
ooters | +0.073 |
Neg logit contributions | |
ow | -0.063 |
lf | -0.050 |
Christmas | -0.050 |
ating | -0.048 |
urg | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
ow | +0.025 |
feelings | +0.019 |
Christmas | +0.019 |
happen | +0.018 |
stories | +0.018 |
Neg logit contributions | |
Whoever | -0.034 |
ounded | -0.032 |
sc | -0.032 |
ursion | -0.032 |
ooters | -0.032 |
Token | fire |
Token ID | 2046 |
Feature activation | +4.83e-03 |
Pos logit contributions | |
Whoever | +0.046 |
ursion | +0.045 |
ounded | +0.044 |
ooters | +0.043 |
sc | +0.043 |
Neg logit contributions | |
ow | -0.028 |
feelings | -0.020 |
Christmas | -0.020 |
happen | -0.019 |
stories | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
Whoever | +0.010 |
ursion | +0.009 |
ounded | +0.009 |
sc | +0.009 |
ooters | +0.009 |
Neg logit contributions | |
ow | -0.007 |
feelings | -0.005 |
Christmas | -0.005 |
happen | -0.005 |
stories | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | +8.44e-03 |
Pos logit contributions | |
ounded | +0.042 |
ursion | +0.040 |
sc | +0.036 |
reaching | +0.036 |
shadow | +0.036 |
Neg logit contributions | |
ow | -0.041 |
Christmas | -0.034 |
feelings | -0.032 |
each | -0.032 |
ad | -0.030 |
Token | are |
Token ID | 389 |
Feature activation | -1.30e-02 |
Pos logit contributions | |
Whoever | +0.016 |
ursion | +0.015 |
ounded | +0.015 |
ooters | +0.015 |
sc | +0.015 |
Neg logit contributions | |
ow | -0.013 |
feelings | -0.009 |
Christmas | -0.009 |
lf | -0.009 |
happen | -0.009 |
Token | called |
Token ID | 1444 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
ow | +0.017 |
feelings | +0.013 |
Christmas | +0.013 |
happen | +0.012 |
stories | +0.012 |
Neg logit contributions | |
Whoever | -0.028 |
ursion | -0.027 |
ounded | -0.027 |
sc | -0.026 |
ooters | -0.026 |
Token | firefighters |
Token ID | 22611 |
Feature activation | +7.19e-02 |
Pos logit contributions | |
Whoever | +0.375 |
ursion | +0.370 |
ounded | +0.366 |
ooters | +0.356 |
sc | +0.355 |
Neg logit contributions | |
ow | -0.216 |
Christmas | -0.154 |
feelings | -0.152 |
happen | -0.151 |
stories | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
Whoever | +0.168 |
ursion | +0.165 |
ounded | +0.164 |
ooters | +0.159 |
sc | +0.159 |
Neg logit contributions | |
ow | -0.078 |
Christmas | -0.053 |
feelings | -0.053 |
happen | -0.052 |
stories | -0.049 |
Token | They |
Token ID | 1119 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
ounded | +0.059 |
ursion | +0.058 |
reaching | +0.053 |
shadow | +0.053 |
sc | +0.052 |
Neg logit contributions | |
ow | -0.049 |
Christmas | -0.040 |
each | -0.037 |
feelings | -0.036 |
stories | -0.034 |
Token | have |
Token ID | 423 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
Whoever | +0.029 |
ursion | +0.028 |
ounded | +0.028 |
ooters | +0.027 |
sc | +0.027 |
Neg logit contributions | |
ow | -0.022 |
feelings | -0.017 |
Christmas | -0.016 |
happen | -0.016 |
stories | -0.016 |
Token | water |
Token ID | 1660 |
Feature activation | +4.17e-02 |
Pos logit contributions | |
ow | +0.020 |
happen | +0.015 |
Christmas | +0.014 |
lf | +0.014 |
feelings | +0.014 |
Neg logit contributions | |
Whoever | -0.026 |
ursion | -0.025 |
ooters | -0.025 |
ounded | -0.025 |
sc | -0.024 |
Token | do |
Token ID | 466 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
Whoever | +0.008 |
ursion | +0.008 |
ounded | +0.008 |
ooters | +0.008 |
sc | +0.008 |
Neg logit contributions | |
ow | -0.007 |
feelings | -0.005 |
Christmas | -0.005 |
happen | -0.005 |
stories | -0.005 |
Token | not |
Token ID | 407 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
ow | +0.075 |
Christmas | +0.049 |
feelings | +0.046 |
soup | +0.045 |
ity | +0.045 |
Neg logit contributions | |
Whoever | -0.132 |
ounded | -0.126 |
ooters | -0.123 |
ursion | -0.122 |
reaching | -0.121 |
Token | like |
Token ID | 588 |
Feature activation | -2.59e-03 |
Pos logit contributions | |
ow | +0.047 |
feelings | +0.039 |
happen | +0.039 |
Christmas | +0.038 |
stories | +0.037 |
Neg logit contributions | |
Whoever | -0.037 |
ursion | -0.036 |
ounded | -0.036 |
sc | -0.034 |
ooters | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -7.27e-03 |
Pos logit contributions | |
ow | +0.004 |
ity | +0.003 |
Christmas | +0.003 |
ating | +0.003 |
lf | +0.003 |
Neg logit contributions | |
Whoever | -0.005 |
ooters | -0.005 |
ounded | -0.005 |
ursion | -0.005 |
reaching | -0.005 |
Token | be |
Token ID | 307 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
ow | +0.011 |
iflower | +0.008 |
Christmas | +0.008 |
ling | +0.008 |
unic | +0.008 |
Neg logit contributions | |
Whoever | -0.013 |
ooters | -0.013 |
ursion | -0.012 |
wanna | -0.012 |
reaching | -0.012 |
Token | called |
Token ID | 1444 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
ow | +0.080 |
feelings | +0.066 |
each | +0.060 |
happen | +0.059 |
stories | +0.059 |
Neg logit contributions | |
Whoever | -0.108 |
ursion | -0.106 |
ounded | -0.103 |
sc | -0.100 |
Ready | -0.100 |
Token | babies |
Token ID | 11903 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
Whoever | +0.352 |
ursion | +0.344 |
ounded | +0.342 |
sc | +0.329 |
ooters | +0.328 |
Neg logit contributions | |
ow | -0.238 |
feelings | -0.178 |
happen | -0.175 |
Christmas | -0.174 |
stories | -0.171 |
Token | . |
Token ID | 13 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
Whoever | +0.078 |
ursion | +0.075 |
ounded | +0.075 |
ooters | +0.074 |
sc | +0.074 |
Neg logit contributions | |
ow | -0.041 |
happen | -0.028 |
Christmas | -0.028 |
feelings | -0.028 |
stories | -0.026 |
Token | They |
Token ID | 1119 |
Feature activation | +8.98e-03 |
Pos logit contributions | |
ursion | +0.037 |
ounded | +0.036 |
ooters | +0.034 |
Whoever | +0.034 |
reaching | +0.034 |
Neg logit contributions | |
ow | -0.036 |
feelings | -0.028 |
Christmas | -0.028 |
stories | -0.027 |
each | -0.027 |
Token | are |
Token ID | 389 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
Whoever | +0.017 |
ounded | +0.016 |
ursion | +0.016 |
ooters | +0.016 |
sc | +0.016 |
Neg logit contributions | |
ow | -0.014 |
Christmas | -0.011 |
feelings | -0.011 |
happen | -0.010 |
lf | -0.010 |
Token | not |
Token ID | 407 |
Feature activation | +3.12e-03 |
Pos logit contributions | |
ow | +0.020 |
feelings | +0.015 |
Christmas | +0.015 |
happen | +0.015 |
stories | +0.014 |
Neg logit contributions | |
Whoever | -0.025 |
ursion | -0.025 |
ounded | -0.024 |
ooters | -0.024 |
sc | -0.024 |
Token | says |
Token ID | 1139 |
Feature activation | +5.77e-02 |
Pos logit contributions | |
ow | +0.053 |
happen | +0.044 |
feelings | +0.044 |
Christmas | +0.044 |
stories | +0.043 |
Neg logit contributions | |
Whoever | -0.030 |
ursion | -0.029 |
ounded | -0.029 |
sc | -0.028 |
ooters | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | +3.06e-03 |
Pos logit contributions | |
Whoever | +0.119 |
ounded | +0.114 |
ursion | +0.113 |
sc | +0.111 |
ooters | +0.111 |
Neg logit contributions | |
ow | -0.077 |
feelings | -0.057 |
Christmas | -0.055 |
happen | -0.055 |
lf | -0.053 |
Token | " |
Token ID | 366 |
Feature activation | +3.74e-02 |
Pos logit contributions | |
Whoever | +0.004 |
ounded | +0.004 |
ursion | +0.004 |
sc | +0.003 |
Ready | +0.003 |
Neg logit contributions | |
ow | -0.006 |
feelings | -0.006 |
happen | -0.006 |
ity | -0.005 |
Christmas | -0.005 |
Token | They |
Token ID | 2990 |
Feature activation | -2.07e-03 |
Pos logit contributions | |
ounded | +0.072 |
Whoever | +0.071 |
ursion | +0.070 |
ooters | +0.067 |
sc | +0.067 |
Neg logit contributions | |
ow | -0.056 |
feelings | -0.042 |
Christmas | -0.041 |
happen | -0.041 |
each | -0.040 |
Token | are |
Token ID | 389 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
ow | +0.003 |
lf | +0.002 |
Christmas | +0.002 |
feelings | +0.002 |
urg | +0.002 |
Neg logit contributions | |
Whoever | -0.004 |
ursion | -0.004 |
ounded | -0.004 |
sc | -0.004 |
ooters | -0.004 |
Token | called |
Token ID | 1444 |
Feature activation | +1.71e-01 |
Pos logit contributions | |
ow | +0.019 |
Christmas | +0.013 |
feelings | +0.013 |
happen | +0.013 |
stories | +0.013 |
Neg logit contributions | |
Whoever | -0.032 |
ursion | -0.032 |
ounded | -0.032 |
sc | -0.031 |
ooters | -0.031 |
Token | seasons |
Token ID | 7028 |
Feature activation | +6.80e-02 |
Pos logit contributions | |
Whoever | +0.346 |
ursion | +0.339 |
ounded | +0.337 |
ooters | +0.325 |
sc | +0.324 |
Neg logit contributions | |
ow | -0.243 |
Christmas | -0.181 |
feelings | -0.179 |
happen | -0.178 |
stories | -0.174 |
Token | . |
Token ID | 13 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
Whoever | +0.139 |
ooters | +0.134 |
sc | +0.134 |
ounded | +0.132 |
ursion | +0.132 |
Neg logit contributions | |
ow | -0.092 |
feelings | -0.066 |
lf | -0.064 |
happen | -0.063 |
Christmas | -0.063 |
Token | And |
Token ID | 843 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
ounded | +0.053 |
ursion | +0.050 |
shadow | +0.047 |
hole | +0.046 |
sc | +0.046 |
Neg logit contributions | |
ow | -0.049 |
Christmas | -0.040 |
each | -0.037 |
soup | -0.037 |
feelings | -0.036 |
Token | they |
Token ID | 484 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
Whoever | +0.047 |
ursion | +0.046 |
ounded | +0.046 |
ooters | +0.044 |
sc | +0.044 |
Neg logit contributions | |
ow | -0.045 |
Christmas | -0.035 |
feelings | -0.035 |
happen | -0.035 |
stories | -0.034 |
Token | happen |
Token ID | 1645 |
Feature activation | -5.05e-02 |
Pos logit contributions | |
Whoever | +0.060 |
ounded | +0.057 |
ursion | +0.056 |
sc | +0.056 |
ooters | +0.055 |
Neg logit contributions | |
ow | -0.052 |
feelings | -0.040 |
lf | -0.039 |
Christmas | -0.039 |
ity | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
ow | +0.002 |
lf | +0.002 |
each | +0.002 |
Christmas | +0.001 |
all | +0.001 |
Neg logit contributions | |
ounded | -0.004 |
ursion | -0.004 |
sc | -0.004 |
hole | -0.004 |
Whoever | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | -4.99e-03 |
Pos logit contributions | |
ow | +0.017 |
lf | +0.014 |
Christmas | +0.014 |
each | +0.014 |
feelings | +0.014 |
Neg logit contributions | |
ounded | -0.025 |
ursion | -0.024 |
sc | -0.024 |
Whoever | -0.024 |
ooters | -0.024 |
Token | Those |
Token ID | 9627 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
ow | +0.007 |
songs | +0.005 |
each | +0.005 |
Christmas | +0.005 |
end | +0.005 |
Neg logit contributions | |
ounded | -0.010 |
ooters | -0.010 |
ursion | -0.010 |
hole | -0.010 |
reaching | -0.009 |
Token | cars |
Token ID | 5006 |
Feature activation | +2.99e-02 |
Pos logit contributions | |
ow | +0.021 |
feelings | +0.014 |
happen | +0.014 |
Christmas | +0.014 |
stories | +0.013 |
Neg logit contributions | |
Whoever | -0.046 |
ursion | -0.045 |
ounded | -0.045 |
sc | -0.044 |
reaching | -0.044 |
Token | are |
Token ID | 389 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
Whoever | +0.065 |
ounded | +0.063 |
ursion | +0.063 |
ooters | +0.062 |
sc | +0.062 |
Neg logit contributions | |
ow | -0.038 |
feelings | -0.026 |
Christmas | -0.026 |
lf | -0.024 |
stories | -0.024 |
Token | called |
Token ID | 1444 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ow | +0.047 |
happen | +0.036 |
Christmas | +0.036 |
stories | +0.035 |
feelings | +0.035 |
Neg logit contributions | |
ursion | -0.070 |
ounded | -0.068 |
Whoever | -0.067 |
reaching | -0.065 |
sc | -0.064 |
Token | traffic |
Token ID | 4979 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
ursion | +0.336 |
ounded | +0.316 |
Whoever | +0.314 |
reaching | +0.313 |
Ready | +0.295 |
Neg logit contributions | |
ow | -0.253 |
stories | -0.207 |
Christmas | -0.204 |
each | -0.198 |
happen | -0.198 |
Token | ," |
Token ID | 553 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
Whoever | +0.083 |
ursion | +0.082 |
ounded | +0.081 |
reaching | +0.078 |
ooters | +0.077 |
Neg logit contributions | |
ow | -0.059 |
happen | -0.046 |
feelings | -0.045 |
Christmas | -0.044 |
stories | -0.043 |
Token | Mom |
Token ID | 11254 |
Feature activation | -1.84e-02 |
Pos logit contributions | |
Whoever | +0.032 |
ursion | +0.031 |
ounded | +0.031 |
ooters | +0.031 |
sc | +0.030 |
Neg logit contributions | |
ow | -0.020 |
feelings | -0.014 |
happen | -0.014 |
Christmas | -0.014 |
stories | -0.014 |
Token | my |
Token ID | 1820 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ow | +0.024 |
ad | +0.021 |
lf | +0.021 |
all | +0.020 |
each | +0.020 |
Neg logit contributions | |
Ready | -0.036 |
sc | -0.036 |
Whoever | -0.036 |
ursion | -0.035 |
ounded | -0.035 |
Token | said |
Token ID | 531 |
Feature activation | +5.85e-02 |
Pos logit contributions | |
ow | +0.022 |
lf | +0.018 |
ad | +0.018 |
happen | +0.018 |
each | +0.017 |
Neg logit contributions | |
ursion | -0.025 |
ooters | -0.024 |
Whoever | -0.024 |
sc | -0.024 |
Ready | -0.024 |
Token | hide |
Token ID | 7808 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
ow | +0.014 |
stories | +0.011 |
feelings | +0.011 |
Christmas | +0.011 |
happen | +0.010 |
Neg logit contributions | |
ursion | -0.021 |
ounded | -0.021 |
Whoever | -0.020 |
ooters | -0.020 |
reaching | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | -9.15e-03 |
Pos logit contributions | |
Whoever | +0.086 |
ounded | +0.083 |
sc | +0.082 |
ursion | +0.081 |
ooters | +0.080 |
Neg logit contributions | |
ow | -0.052 |
happen | -0.036 |
feelings | -0.035 |
Christmas | -0.035 |
stories | -0.033 |
Token | seek |
Token ID | 5380 |
Feature activation | -8.32e-03 |
Pos logit contributions | |
ow | +0.011 |
stories | +0.009 |
happen | +0.008 |
Christmas | +0.008 |
feelings | +0.008 |
Neg logit contributions | |
ursion | -0.020 |
Whoever | -0.020 |
ounded | -0.020 |
ooters | -0.019 |
reaching | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -8.08e-03 |
Pos logit contributions | |
ow | +0.011 |
happen | +0.008 |
feelings | +0.008 |
Christmas | +0.008 |
stories | +0.007 |
Neg logit contributions | |
Whoever | -0.018 |
ounded | -0.017 |
sc | -0.017 |
ursion | -0.017 |
ooters | -0.016 |
Token | hop |
Token ID | 1725 |
Feature activation | -1.03e-03 |
Pos logit contributions | |
ow | +0.009 |
stories | +0.007 |
Christmas | +0.007 |
all | +0.007 |
feelings | +0.007 |
Neg logit contributions | |
ursion | -0.019 |
ounded | -0.018 |
Whoever | -0.018 |
ooters | -0.018 |
reaching | -0.017 |
Token | sc |
Token ID | 1416 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ow | +0.001 |
happen | +0.001 |
Christmas | +0.001 |
all | +0.001 |
stories | +0.001 |
Neg logit contributions | |
Whoever | -0.002 |
ursion | -0.002 |
ounded | -0.002 |
ooters | -0.002 |
reaching | -0.002 |
Token | ot |
Token ID | 313 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
Whoever | +0.331 |
ounded | +0.317 |
ursion | +0.315 |
ooters | +0.313 |
sc | +0.312 |
Neg logit contributions | |
ow | -0.258 |
happen | -0.191 |
Christmas | -0.189 |
feelings | -0.187 |
stories | -0.179 |
Token | ch |
Token ID | 354 |
Feature activation | +5.21e-02 |
Pos logit contributions | |
ow | +0.051 |
Christmas | +0.042 |
stories | +0.042 |
happen | +0.041 |
feelings | +0.041 |
Neg logit contributions | |
Whoever | -0.034 |
ounded | -0.034 |
ursion | -0.033 |
ooters | -0.031 |
reaching | -0.031 |
Token | and |
Token ID | 290 |
Feature activation | -1.74e-03 |
Pos logit contributions | |
Whoever | +0.105 |
sc | +0.102 |
ooters | +0.100 |
ursion | +0.099 |
ounded | +0.099 |
Neg logit contributions | |
ow | -0.073 |
Christmas | -0.054 |
feelings | -0.054 |
happen | -0.052 |
stories | -0.050 |
Token | much |
Token ID | 881 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
ow | +0.002 |
happen | +0.002 |
Christmas | +0.002 |
stories | +0.002 |
feelings | +0.002 |
Neg logit contributions | |
Whoever | -0.004 |
ursion | -0.003 |
ounded | -0.003 |
ooters | -0.003 |
reaching | -0.003 |
Token | more |
Token ID | 517 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
Whoever | +0.040 |
ursion | +0.038 |
ounded | +0.038 |
ooters | +0.037 |
sc | +0.037 |
Neg logit contributions | |
ow | -0.028 |
happen | -0.021 |
feelings | -0.020 |
Christmas | -0.020 |
stories | -0.020 |
Token | does |
Token ID | 857 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
ow | +0.018 |
feelings | +0.014 |
stories | +0.014 |
happen | +0.014 |
jokes | +0.014 |
Neg logit contributions | |
ursion | -0.014 |
Whoever | -0.013 |
ounded | -0.013 |
ooters | -0.013 |
shadow | -0.013 |
Token | not |
Token ID | 407 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
ow | +0.013 |
happen | +0.008 |
feelings | +0.008 |
Christmas | +0.008 |
stories | +0.007 |
Neg logit contributions | |
Whoever | -0.036 |
ursion | -0.036 |
ounded | -0.036 |
ooters | -0.035 |
sc | -0.035 |
Token | like |
Token ID | 588 |
Feature activation | -3.90e-03 |
Pos logit contributions | |
ow | +0.048 |
happen | +0.038 |
feelings | +0.038 |
Christmas | +0.038 |
stories | +0.037 |
Neg logit contributions | |
Whoever | -0.043 |
ursion | -0.041 |
ounded | -0.041 |
sc | -0.039 |
ooters | -0.039 |
Token | to |
Token ID | 284 |
Feature activation | -7.17e-03 |
Pos logit contributions | |
ow | +0.006 |
Christmas | +0.004 |
happen | +0.004 |
ating | +0.004 |
feelings | +0.004 |
Neg logit contributions | |
Whoever | -0.008 |
ooters | -0.007 |
ounded | -0.007 |
ursion | -0.007 |
reaching | -0.007 |
Token | be |
Token ID | 307 |
Feature activation | -5.46e-02 |
Pos logit contributions | |
ow | +0.011 |
Christmas | +0.008 |
ating | +0.008 |
iflower | +0.008 |
kin | +0.008 |
Neg logit contributions | |
Whoever | -0.013 |
ooters | -0.013 |
reaching | -0.012 |
ursion | -0.012 |
ounded | -0.012 |
Token | called |
Token ID | 1444 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
ow | +0.082 |
feelings | +0.069 |
jokes | +0.062 |
each | +0.061 |
stories | +0.061 |
Neg logit contributions | |
ursion | -0.111 |
Whoever | -0.110 |
ounded | -0.106 |
Ready | -0.103 |
sc | -0.103 |
Token | mean |
Token ID | 1612 |
Feature activation | -1.36e-03 |
Pos logit contributions | |
Whoever | +0.357 |
ursion | +0.351 |
ounded | +0.347 |
sc | +0.334 |
ooters | +0.333 |
Neg logit contributions | |
ow | -0.228 |
feelings | -0.169 |
happen | -0.164 |
Christmas | -0.163 |
stories | -0.162 |
Token | . |
Token ID | 13 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
ow | +0.002 |
feelings | +0.001 |
happen | +0.001 |
Christmas | +0.001 |
stories | +0.001 |
Neg logit contributions | |
Whoever | -0.003 |
ursion | -0.003 |
ounded | -0.003 |
sc | -0.003 |
ooters | -0.003 |
Token | He |
Token ID | 679 |
Feature activation | +6.94e-04 |
Pos logit contributions | |
ounded | +0.046 |
ursion | +0.045 |
ooters | +0.043 |
sc | +0.043 |
Whoever | +0.043 |
Neg logit contributions | |
ow | -0.038 |
feelings | -0.029 |
stories | -0.028 |
jokes | -0.027 |
Christmas | -0.027 |
Token | is |
Token ID | 318 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
Whoever | +0.001 |
ursion | +0.001 |
ounded | +0.001 |
ooters | +0.001 |
sc | +0.001 |
Neg logit contributions | |
ow | -0.001 |
feelings | -0.001 |
happen | -0.001 |
stories | -0.001 |
Christmas | -0.001 |
Token | mean |
Token ID | 1612 |
Feature activation | -1.29e-02 |
Pos logit contributions | |
ursion | +0.035 |
Whoever | +0.035 |
ounded | +0.034 |
sc | +0.033 |
ooters | +0.033 |
Neg logit contributions | |
ow | -0.020 |
feelings | -0.015 |
stories | -0.014 |
happen | -0.014 |
jokes | -0.014 |
Token | She |
Token ID | 1375 |
Feature activation | +5.23e-03 |
Pos logit contributions | |
ounded | +0.023 |
sc | +0.023 |
ursion | +0.023 |
shadow | +0.023 |
Whoever | +0.022 |
Neg logit contributions | |
ow | -0.020 |
feelings | -0.015 |
each | -0.015 |
all | -0.014 |
stories | -0.014 |
Token | did |
Token ID | 750 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Whoever | +0.010 |
ounded | +0.010 |
ursion | +0.010 |
sc | +0.009 |
ooters | +0.009 |
Neg logit contributions | |
ow | -0.008 |
Christmas | -0.006 |
happen | -0.006 |
feelings | -0.006 |
stories | -0.006 |
Token | not |
Token ID | 407 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
ow | +0.049 |
Christmas | +0.036 |
urg | +0.035 |
ating | +0.034 |
ity | +0.034 |
Neg logit contributions | |
Whoever | -0.073 |
ounded | -0.066 |
reaching | -0.066 |
shadow | -0.065 |
Ready | -0.065 |
Token | like |
Token ID | 588 |
Feature activation | -5.34e-03 |
Pos logit contributions | |
ow | +0.061 |
Christmas | +0.049 |
feelings | +0.049 |
happen | +0.048 |
stories | +0.047 |
Neg logit contributions | |
Whoever | -0.046 |
ursion | -0.043 |
ounded | -0.043 |
ooters | -0.042 |
sc | -0.042 |
Token | being |
Token ID | 852 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
ow | +0.009 |
Christmas | +0.007 |
happen | +0.007 |
feelings | +0.007 |
ating | +0.007 |
Neg logit contributions | |
Whoever | -0.009 |
ooters | -0.009 |
ursion | -0.009 |
ounded | -0.009 |
reaching | -0.009 |
Token | called |
Token ID | 1444 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ow | +0.066 |
feelings | +0.052 |
happen | +0.049 |
Christmas | +0.049 |
stories | +0.049 |
Neg logit contributions | |
Whoever | -0.082 |
ursion | -0.081 |
ounded | -0.080 |
sc | -0.077 |
ooters | -0.077 |
Token | silly |
Token ID | 14397 |
Feature activation | +1.54e-02 |
Pos logit contributions | |
Whoever | +0.319 |
ooters | +0.306 |
ounded | +0.306 |
ursion | +0.305 |
sc | +0.302 |
Neg logit contributions | |
ow | -0.252 |
Christmas | -0.189 |
happen | -0.188 |
feelings | -0.185 |
lf | -0.180 |
Token | . |
Token ID | 13 |
Feature activation | +2.07e-02 |
Pos logit contributions | |
Whoever | +0.032 |
ooters | +0.031 |
sc | +0.031 |
ounded | +0.031 |
ursion | +0.030 |
Neg logit contributions | |
ow | -0.020 |
Christmas | -0.015 |
lf | -0.014 |
happen | -0.014 |
ating | -0.013 |
Token | She |
Token ID | 1375 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
ounded | +0.039 |
sc | +0.038 |
ursion | +0.038 |
Whoever | +0.037 |
shadow | +0.037 |
Neg logit contributions | |
ow | -0.035 |
feelings | -0.026 |
each | -0.024 |
stories | -0.024 |
Christmas | -0.024 |
Token | was |
Token ID | 373 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
Whoever | +0.022 |
ounded | +0.021 |
ursion | +0.021 |
sc | +0.020 |
ooters | +0.020 |
Neg logit contributions | |
ow | -0.017 |
Christmas | -0.013 |
happen | -0.013 |
feelings | -0.013 |
stories | -0.013 |
Token | ashamed |
Token ID | 22461 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Whoever | +0.048 |
ursion | +0.048 |
ounded | +0.048 |
sc | +0.047 |
ooters | +0.046 |
Neg logit contributions | |
ow | -0.031 |
feelings | -0.023 |
Christmas | -0.022 |
happen | -0.021 |
stories | -0.021 |
Token | It |
Token ID | 632 |
Feature activation | -8.24e-03 |
Pos logit contributions | |
ounded | +0.053 |
ursion | +0.050 |
shadow | +0.047 |
reaching | +0.046 |
hole | +0.045 |
Neg logit contributions | |
ow | -0.049 |
Christmas | -0.041 |
ad | -0.041 |
stories | -0.039 |
unic | -0.038 |
Token | has |
Token ID | 468 |
Feature activation | -1.03e-02 |
Pos logit contributions | |
ow | +0.010 |
feelings | +0.007 |
Christmas | +0.007 |
happen | +0.007 |
stories | +0.007 |
Neg logit contributions | |
Whoever | -0.018 |
ursion | -0.017 |
ounded | -0.017 |
ooters | -0.017 |
sc | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
ow | +0.012 |
feelings | +0.009 |
happen | +0.009 |
Christmas | +0.009 |
stories | +0.008 |
Neg logit contributions | |
Whoever | -0.023 |
ursion | -0.022 |
ounded | -0.022 |
ooters | -0.022 |
sc | -0.022 |
Token | long |
Token ID | 890 |
Feature activation | -4.05e-03 |
Pos logit contributions | |
ursion | +0.025 |
Whoever | +0.025 |
ounded | +0.025 |
sc | +0.024 |
ooters | +0.023 |
Neg logit contributions | |
ow | -0.015 |
Christmas | -0.010 |
stories | -0.010 |
feelings | -0.010 |
happen | -0.009 |
Token | nose |
Token ID | 9686 |
Feature activation | -3.54e-03 |
Pos logit contributions | |
ow | +0.005 |
feelings | +0.004 |
happen | +0.004 |
Christmas | +0.004 |
stories | +0.004 |
Neg logit contributions | |
Whoever | -0.009 |
ursion | -0.008 |
ounded | -0.008 |
ooters | -0.008 |
sc | -0.008 |
Token | called |
Token ID | 1444 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ow | +0.004 |
happen | +0.003 |
feelings | +0.003 |
Christmas | +0.003 |
stories | +0.003 |
Neg logit contributions | |
Whoever | -0.008 |
ursion | -0.007 |
ooters | -0.007 |
sc | -0.007 |
ounded | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | +3.29e-02 |
Pos logit contributions | |
ursion | +0.337 |
ounded | +0.334 |
Whoever | +0.334 |
reaching | +0.316 |
ooters | +0.314 |
Neg logit contributions | |
ow | -0.245 |
Christmas | -0.187 |
stories | -0.179 |
feelings | -0.178 |
happen | -0.172 |
Token | trunk |
Token ID | 21427 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
Whoever | +0.072 |
ounded | +0.072 |
ursion | +0.072 |
sc | +0.068 |
ooters | +0.068 |
Neg logit contributions | |
ow | -0.044 |
Christmas | -0.031 |
stories | -0.030 |
happen | -0.029 |
feelings | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
Whoever | +0.106 |
ounded | +0.105 |
ursion | +0.104 |
ooters | +0.101 |
Ready | +0.101 |
Neg logit contributions | |
ow | -0.043 |
Christmas | -0.028 |
happen | -0.027 |
feelings | -0.026 |
stories | -0.025 |
Token | It |
Token ID | 632 |
Feature activation | -5.41e-03 |
Pos logit contributions | |
ounded | +0.053 |
ursion | +0.050 |
reaching | +0.045 |
shadow | +0.044 |
hole | +0.043 |
Neg logit contributions | |
ow | -0.047 |
Christmas | -0.039 |
stories | -0.037 |
unic | -0.036 |
each | -0.036 |
Token | can |
Token ID | 460 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
ow | +0.007 |
happen | +0.005 |
Christmas | +0.005 |
feelings | +0.005 |
stories | +0.005 |
Neg logit contributions | |
Whoever | -0.011 |
ursion | -0.011 |
ounded | -0.011 |
ooters | -0.011 |
sc | -0.011 |
Token | We |
Token ID | 775 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
ounded | +0.032 |
ursion | +0.030 |
reaching | +0.027 |
shadow | +0.026 |
hole | +0.026 |
Neg logit contributions | |
ow | -0.029 |
Christmas | -0.025 |
unic | -0.023 |
stories | -0.023 |
each | -0.022 |
Token | live |
Token ID | 2107 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
Whoever | +0.096 |
ursion | +0.090 |
ounded | +0.089 |
sc | +0.087 |
ooters | +0.086 |
Neg logit contributions | |
ow | -0.096 |
feelings | -0.075 |
Christmas | -0.073 |
lf | -0.073 |
happen | -0.072 |
Token | on |
Token ID | 319 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
Whoever | +0.022 |
sc | +0.022 |
ounded | +0.022 |
ursion | +0.021 |
ooters | +0.021 |
Neg logit contributions | |
ow | -0.017 |
happen | -0.012 |
feelings | -0.012 |
lf | -0.012 |
Christmas | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +4.03e-03 |
Pos logit contributions | |
ow | +0.013 |
happen | +0.009 |
ating | +0.009 |
ity | +0.009 |
feelings | +0.009 |
Neg logit contributions | |
Whoever | -0.022 |
sc | -0.021 |
ooters | -0.020 |
ounded | -0.020 |
ursion | -0.020 |
Token | planet |
Token ID | 5440 |
Feature activation | +6.79e-03 |
Pos logit contributions | |
Whoever | +0.009 |
ursion | +0.009 |
ounded | +0.009 |
ooters | +0.008 |
reaching | +0.008 |
Neg logit contributions | |
ow | -0.005 |
Christmas | -0.004 |
happen | -0.004 |
stories | -0.004 |
feelings | -0.004 |
Token | called |
Token ID | 1444 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
Whoever | +0.015 |
ursion | +0.015 |
ounded | +0.015 |
ooters | +0.014 |
reaching | +0.014 |
Neg logit contributions | |
ow | -0.008 |
Christmas | -0.006 |
feelings | -0.006 |
happen | -0.006 |
stories | -0.005 |
Token | Earth |
Token ID | 3668 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
ursion | +0.350 |
Whoever | +0.349 |
ounded | +0.342 |
reaching | +0.333 |
ooters | +0.327 |
Neg logit contributions | |
ow | -0.221 |
Christmas | -0.169 |
stories | -0.168 |
feelings | -0.165 |
happen | -0.162 |
Token | . |
Token ID | 13 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Whoever | +0.116 |
ursion | +0.113 |
ounded | +0.112 |
ooters | +0.109 |
sc | +0.109 |
Neg logit contributions | |
ow | -0.070 |
feelings | -0.050 |
Christmas | -0.050 |
happen | -0.049 |
stories | -0.047 |
Token | It |
Token ID | 632 |
Feature activation | -5.56e-03 |
Pos logit contributions | |
ounded | +0.027 |
ursion | +0.026 |
reaching | +0.024 |
shadow | +0.023 |
ooters | +0.023 |
Neg logit contributions | |
ow | -0.028 |
Christmas | -0.024 |
unic | -0.023 |
songs | -0.023 |
stories | -0.023 |
Token | is |
Token ID | 318 |
Feature activation | +9.40e-03 |
Pos logit contributions | |
ow | +0.008 |
Christmas | +0.006 |
feelings | +0.006 |
happen | +0.006 |
lf | +0.006 |
Neg logit contributions | |
Whoever | -0.011 |
ounded | -0.011 |
sc | -0.010 |
ursion | -0.010 |
ooters | -0.010 |
Token | blue |
Token ID | 4171 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
Whoever | +0.021 |
ursion | +0.021 |
ounded | +0.020 |
ooters | +0.020 |
sc | +0.020 |
Neg logit contributions | |
ow | -0.012 |
Christmas | -0.008 |
feelings | -0.008 |
happen | -0.008 |
stories | -0.008 |
Token | would |
Token ID | 561 |
Feature activation | +6.91e-03 |
Pos logit contributions | |
ow | +0.046 |
happen | +0.034 |
feelings | +0.033 |
Christmas | +0.032 |
stories | +0.031 |
Neg logit contributions | |
Whoever | -0.062 |
ursion | -0.059 |
sc | -0.058 |
ounded | -0.058 |
ooters | -0.057 |
Token | play |
Token ID | 711 |
Feature activation | -4.73e-02 |
Pos logit contributions | |
Whoever | +0.014 |
ursion | +0.013 |
ounded | +0.013 |
ooters | +0.013 |
sc | +0.013 |
Neg logit contributions | |
ow | -0.010 |
feelings | -0.007 |
Christmas | -0.007 |
happen | -0.007 |
stories | -0.007 |
Token | tag |
Token ID | 7621 |
Feature activation | +3.07e-02 |
Pos logit contributions | |
ow | +0.075 |
feelings | +0.054 |
ling | +0.053 |
urg | +0.052 |
Christmas | +0.052 |
Neg logit contributions | |
Whoever | -0.094 |
sc | -0.093 |
Ready | -0.087 |
ounded | -0.081 |
Ready | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
sc | +0.062 |
Whoever | +0.061 |
ounded | +0.056 |
ooters | +0.055 |
Ready | +0.054 |
Neg logit contributions | |
ow | -0.048 |
feelings | -0.034 |
ling | -0.033 |
Christmas | -0.033 |
happen | -0.033 |
Token | hop |
Token ID | 1725 |
Feature activation | -4.01e-03 |
Pos logit contributions | |
ow | +0.017 |
stories | +0.013 |
feelings | +0.013 |
happen | +0.013 |
Christmas | +0.013 |
Neg logit contributions | |
ursion | -0.024 |
ounded | -0.023 |
Whoever | -0.023 |
ooters | -0.021 |
reaching | -0.021 |
Token | sc |
Token ID | 1416 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
ow | +0.007 |
Christmas | +0.005 |
happen | +0.005 |
stories | +0.005 |
feelings | +0.005 |
Neg logit contributions | |
Whoever | -0.007 |
ursion | -0.007 |
ounded | -0.007 |
ooters | -0.007 |
reaching | -0.006 |
Token | ot |
Token ID | 313 |
Feature activation | -2.61e-02 |
Pos logit contributions | |
Whoever | +0.316 |
ooters | +0.301 |
sc | +0.299 |
ounded | +0.297 |
ursion | +0.296 |
Neg logit contributions | |
ow | -0.265 |
happen | -0.197 |
Christmas | -0.195 |
feelings | -0.190 |
ling | -0.183 |
Token | ch |
Token ID | 354 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
ow | +0.062 |
Christmas | +0.052 |
feelings | +0.052 |
happen | +0.052 |
stories | +0.051 |
Neg logit contributions | |
Whoever | -0.027 |
ursion | -0.026 |
ounded | -0.026 |
ooters | -0.024 |
sc | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -4.04e-03 |
Pos logit contributions | |
sc | +0.086 |
Whoever | +0.084 |
ooters | +0.080 |
ursion | +0.074 |
ounded | +0.073 |
Neg logit contributions | |
ow | -0.069 |
Christmas | -0.056 |
feelings | -0.056 |
crystal | -0.055 |
unic | -0.055 |
Token | and |
Token ID | 290 |
Feature activation | -2.49e-03 |
Pos logit contributions | |
ow | +0.004 |
Christmas | +0.003 |
happen | +0.003 |
feelings | +0.003 |
stories | +0.003 |
Neg logit contributions | |
ursion | -0.009 |
Whoever | -0.009 |
ounded | -0.009 |
ooters | -0.009 |
reaching | -0.009 |
Token | many |
Token ID | 867 |
Feature activation | +1.73e-02 |
Pos logit contributions | |
ow | +0.003 |
happen | +0.003 |
stories | +0.002 |
Christmas | +0.002 |
feelings | +0.002 |
Neg logit contributions | |
ursion | -0.005 |
ounded | -0.005 |
Whoever | -0.005 |
reaching | -0.005 |
ooters | -0.005 |
Token | a |
Token ID | 257 |
Feature activation | +3.72e-03 |
Pos logit contributions | |
ow | +0.035 |
happen | +0.025 |
feelings | +0.025 |
lf | +0.025 |
Christmas | +0.024 |
Neg logit contributions | |
Whoever | -0.056 |
ursion | -0.053 |
ounded | -0.053 |
ooters | -0.052 |
sc | -0.052 |
Token | salad |
Token ID | 20698 |
Feature activation | -4.95e-02 |
Pos logit contributions | |
Whoever | +0.008 |
ursion | +0.008 |
ounded | +0.008 |
ooters | +0.008 |
sc | +0.008 |
Neg logit contributions | |
ow | -0.004 |
Christmas | -0.003 |
feelings | -0.003 |
happen | -0.003 |
stories | -0.003 |
Token | . |
Token ID | 13 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
ow | +0.076 |
feelings | +0.057 |
happen | +0.056 |
Christmas | +0.055 |
stories | +0.054 |
Neg logit contributions | |
Whoever | -0.091 |
ursion | -0.089 |
ounded | -0.088 |
sc | -0.087 |
ooters | -0.086 |
Token | It |
Token ID | 632 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
ounded | +0.040 |
ooters | +0.038 |
ursion | +0.038 |
reaching | +0.036 |
Whoever | +0.035 |
Neg logit contributions | |
ow | -0.058 |
Christmas | -0.049 |
soup | -0.047 |
feelings | -0.047 |
happen | -0.046 |
Token | is |
Token ID | 318 |
Feature activation | +7.87e-03 |
Pos logit contributions | |
ow | +0.018 |
feelings | +0.015 |
lf | +0.015 |
Christmas | +0.014 |
stories | +0.014 |
Neg logit contributions | |
Whoever | -0.019 |
sc | -0.019 |
ounded | -0.018 |
ursion | -0.017 |
Ready | -0.017 |
Token | called |
Token ID | 1444 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Whoever | +0.015 |
sc | +0.014 |
ounded | +0.013 |
ursion | +0.013 |
Ready | +0.013 |
Neg logit contributions | |
ow | -0.011 |
feelings | -0.009 |
ating | -0.009 |
lf | -0.008 |
ad | -0.008 |
Token | g |
Token ID | 308 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
Whoever | +0.351 |
sc | +0.332 |
ounded | +0.330 |
ursion | +0.329 |
ooters | +0.324 |
Neg logit contributions | |
ow | -0.209 |
feelings | -0.153 |
happen | -0.152 |
lf | -0.147 |
ating | -0.144 |
Token | aj |
Token ID | 1228 |
Feature activation | +7.87e-03 |
Pos logit contributions | |
ow | +0.060 |
feelings | +0.047 |
happen | +0.047 |
Christmas | +0.045 |
stories | +0.044 |
Neg logit contributions | |
Whoever | -0.086 |
ursion | -0.082 |
ooters | -0.081 |
ounded | -0.081 |
sc | -0.079 |
Token | ar |
Token ID | 283 |
Feature activation | +2.89e-03 |
Pos logit contributions | |
Whoever | +0.012 |
sc | +0.012 |
ooters | +0.011 |
ounded | +0.011 |
ursion | +0.011 |
Neg logit contributions | |
ow | -0.014 |
happen | -0.012 |
feelings | -0.011 |
ling | -0.011 |
unic | -0.011 |
Token | ka |
Token ID | 38387 |
Feature activation | +8.81e-02 |
Pos logit contributions | |
Whoever | +0.005 |
ursion | +0.005 |
ounded | +0.005 |
ooters | +0.004 |
sc | +0.004 |
Neg logit contributions | |
ow | -0.005 |
feelings | -0.004 |
Christmas | -0.004 |
happen | -0.004 |
stories | -0.004 |
Token | hal |
Token ID | 10284 |
Feature activation | +5.51e-02 |
Pos logit contributions | |
Whoever | +0.162 |
ursion | +0.156 |
ounded | +0.155 |
ooters | +0.152 |
sc | +0.151 |
Neg logit contributions | |
ow | -0.140 |
happen | -0.107 |
feelings | -0.107 |
Christmas | -0.106 |
stories | -0.102 |
Token | did |
Token ID | 750 |
Feature activation | -3.26e-02 |
Pos logit contributions | |
ow | +0.018 |
feelings | +0.014 |
Christmas | +0.014 |
happen | +0.014 |
stories | +0.014 |
Neg logit contributions | |
Whoever | -0.015 |
ursion | -0.014 |
ounded | -0.014 |
ooters | -0.014 |
sc | -0.014 |
Token | not |
Token ID | 407 |
Feature activation | -2.88e-02 |
Pos logit contributions | |
ow | +0.048 |
Christmas | +0.036 |
ity | +0.035 |
ating | +0.035 |
unic | +0.034 |
Neg logit contributions | |
Whoever | -0.066 |
reaching | -0.059 |
shadow | -0.059 |
ounded | -0.059 |
Ready | -0.058 |
Token | like |
Token ID | 588 |
Feature activation | -6.59e-03 |
Pos logit contributions | |
ow | +0.053 |
happen | +0.044 |
feelings | +0.043 |
stories | +0.042 |
jokes | +0.041 |
Neg logit contributions | |
ursion | -0.048 |
Whoever | -0.047 |
ounded | -0.046 |
sc | -0.044 |
reaching | -0.042 |
Token | to |
Token ID | 284 |
Feature activation | -8.01e-03 |
Pos logit contributions | |
ow | +0.009 |
Christmas | +0.007 |
happen | +0.007 |
feelings | +0.007 |
ating | +0.006 |
Neg logit contributions | |
Whoever | -0.013 |
ooters | -0.012 |
ounded | -0.012 |
ursion | -0.012 |
sc | -0.012 |
Token | be |
Token ID | 307 |
Feature activation | -5.86e-02 |
Pos logit contributions | |
ow | +0.013 |
ating | +0.011 |
iflower | +0.010 |
ling | +0.010 |
Christmas | +0.010 |
Neg logit contributions | |
Whoever | -0.013 |
wanna | -0.012 |
ooters | -0.012 |
reaching | -0.012 |
ursion | -0.011 |
Token | called |
Token ID | 1444 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
ow | +0.094 |
feelings | +0.074 |
Christmas | +0.070 |
stories | +0.070 |
happen | +0.070 |
Neg logit contributions | |
Whoever | -0.111 |
ursion | -0.110 |
ounded | -0.108 |
sc | -0.103 |
ooters | -0.102 |
Token | a |
Token ID | 257 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
Whoever | +0.334 |
ursion | +0.323 |
ounded | +0.322 |
ooters | +0.315 |
sc | +0.313 |
Neg logit contributions | |
ow | -0.232 |
feelings | -0.170 |
Christmas | -0.170 |
happen | -0.169 |
stories | -0.161 |
Token | silly |
Token ID | 14397 |
Feature activation | -1.23e-03 |
Pos logit contributions | |
ursion | +0.081 |
ounded | +0.079 |
Whoever | +0.079 |
ooters | +0.077 |
hole | +0.076 |
Neg logit contributions | |
ow | -0.044 |
happen | -0.030 |
Christmas | -0.030 |
stories | -0.029 |
feelings | -0.029 |
Token | monkey |
Token ID | 21657 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
ow | +0.002 |
feelings | +0.001 |
Christmas | +0.001 |
happen | +0.001 |
stories | +0.001 |
Neg logit contributions | |
Whoever | -0.002 |
ursion | -0.002 |
ounded | -0.002 |
ooters | -0.002 |
sc | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +4.36e-03 |
Pos logit contributions | |
Whoever | +0.129 |
ursion | +0.128 |
ounded | +0.128 |
ooters | +0.123 |
reaching | +0.121 |
Neg logit contributions | |
ow | -0.054 |
feelings | -0.035 |
Christmas | -0.034 |
happen | -0.033 |
stories | -0.033 |
Token | He |
Token ID | 679 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
ursion | +0.009 |
ounded | +0.009 |
hole | +0.009 |
sc | +0.009 |
reaching | +0.009 |
Neg logit contributions | |
ow | -0.006 |
his | -0.004 |
all | -0.004 |
feelings | -0.004 |
stories | -0.004 |
Token | worse |
Token ID | 4785 |
Feature activation | +1.23e-02 |
Pos logit contributions | |
Whoever | +0.052 |
ursion | +0.050 |
ounded | +0.050 |
ooters | +0.049 |
sc | +0.048 |
Neg logit contributions | |
ow | -0.039 |
feelings | -0.029 |
happen | -0.029 |
Christmas | -0.029 |
stories | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Whoever | +0.024 |
ooters | +0.023 |
ounded | +0.023 |
ursion | +0.022 |
hole | +0.022 |
Neg logit contributions | |
ow | -0.017 |
Christmas | -0.013 |
stories | -0.012 |
lf | -0.012 |
feelings | -0.012 |
Token | They |
Token ID | 1119 |
Feature activation | +7.51e-03 |
Pos logit contributions | |
ursion | +0.023 |
ounded | +0.023 |
Whoever | +0.022 |
ooters | +0.022 |
sc | +0.021 |
Neg logit contributions | |
ow | -0.027 |
feelings | -0.022 |
Christmas | -0.021 |
each | -0.020 |
happen | -0.020 |
Token | cried |
Token ID | 16896 |
Feature activation | +5.08e-02 |
Pos logit contributions | |
Whoever | +0.013 |
ursion | +0.012 |
ounded | +0.012 |
sc | +0.012 |
reaching | +0.012 |
Neg logit contributions | |
ow | -0.012 |
happen | -0.010 |
Christmas | -0.010 |
feelings | -0.010 |
ating | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
Whoever | +0.087 |
sc | +0.085 |
ounded | +0.081 |
ooters | +0.078 |
reaching | +0.078 |
Neg logit contributions | |
ow | -0.081 |
lf | -0.074 |
happen | -0.068 |
ating | -0.067 |
ad | -0.064 |
Token | called |
Token ID | 1444 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Whoever | +0.030 |
ursion | +0.029 |
ounded | +0.029 |
sc | +0.028 |
ooters | +0.028 |
Neg logit contributions | |
ow | -0.036 |
Christmas | -0.029 |
happen | -0.029 |
feelings | -0.029 |
stories | -0.028 |
Token | for |
Token ID | 329 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
Whoever | +0.289 |
sc | +0.277 |
ounded | +0.276 |
ooters | +0.271 |
ursion | +0.267 |
Neg logit contributions | |
ow | -0.267 |
lf | -0.212 |
ating | -0.205 |
ity | -0.204 |
ad | -0.196 |
Token | help |
Token ID | 1037 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
ity | +0.018 |
ating | +0.018 |
ow | +0.017 |
iflower | +0.016 |
ling | +0.015 |
Neg logit contributions | |
Whoever | -0.019 |
First | -0.018 |
Ready | -0.017 |
ooters | -0.017 |
reaching | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
ow | +0.019 |
Christmas | +0.013 |
feelings | +0.013 |
happen | +0.013 |
lf | +0.013 |
Neg logit contributions | |
Whoever | -0.031 |
sc | -0.030 |
ounded | -0.030 |
ursion | -0.029 |
ooters | -0.029 |
Token | But |
Token ID | 887 |
Feature activation | +4.05e-02 |
Pos logit contributions | |
ursion | +0.022 |
ounded | +0.022 |
Whoever | +0.022 |
ooters | +0.021 |
sc | +0.021 |
Neg logit contributions | |
ow | -0.026 |
feelings | -0.021 |
Christmas | -0.021 |
each | -0.020 |
happen | -0.020 |
Token | no |
Token ID | 645 |
Feature activation | +1.01e-03 |
Pos logit contributions | |
Whoever | +0.063 |
ounded | +0.057 |
ursion | +0.055 |
Ready | +0.055 |
sc | +0.054 |
Neg logit contributions | |
ow | -0.074 |
ity | -0.062 |
feelings | -0.062 |
happen | -0.060 |
ad | -0.059 |
Token | on |
Token ID | 319 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
Whoever | +0.001 |
ursion | +0.001 |
ounded | +0.001 |
reaching | +0.001 |
ooters | +0.001 |
Neg logit contributions | |
ow | -0.001 |
happen | -0.000 |
Christmas | -0.000 |
feelings | -0.000 |
stories | -0.000 |
Token | it |
Token ID | 340 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
ow | +0.025 |
ity | +0.019 |
feelings | +0.019 |
ling | +0.019 |
urg | +0.018 |
Neg logit contributions | |
Whoever | -0.029 |
sc | -0.025 |
First | -0.024 |
ooters | -0.024 |
Ready | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -8.67e-03 |
Pos logit contributions | |
ow | +0.037 |
Christmas | +0.025 |
feelings | +0.025 |
happen | +0.024 |
stories | +0.023 |
Neg logit contributions | |
Whoever | -0.070 |
ursion | -0.067 |
ounded | -0.067 |
sc | -0.067 |
ooters | -0.067 |
Token | They |
Token ID | 1119 |
Feature activation | +1.75e-03 |
Pos logit contributions | |
ow | +0.010 |
each | +0.010 |
all | +0.009 |
every | +0.009 |
end | +0.008 |
Neg logit contributions | |
ounded | -0.019 |
shadow | -0.019 |
ursion | -0.018 |
reaching | -0.018 |
hole | -0.018 |
Token | were |
Token ID | 547 |
Feature activation | -7.21e-03 |
Pos logit contributions | |
Whoever | +0.003 |
ursion | +0.003 |
ounded | +0.003 |
ooters | +0.003 |
sc | +0.003 |
Neg logit contributions | |
ow | -0.003 |
happen | -0.002 |
feelings | -0.002 |
Christmas | -0.002 |
stories | -0.002 |
Token | called |
Token ID | 1444 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
ow | +0.010 |
Christmas | +0.007 |
feelings | +0.007 |
stories | +0.007 |
happen | +0.007 |
Neg logit contributions | |
Whoever | -0.015 |
ursion | -0.015 |
ounded | -0.015 |
sc | -0.014 |
ooters | -0.014 |
Token | or |
Token ID | 393 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
Whoever | +0.345 |
ounded | +0.331 |
ursion | +0.330 |
ooters | +0.326 |
sc | +0.322 |
Neg logit contributions | |
ow | -0.220 |
feelings | -0.158 |
happen | -0.157 |
Christmas | -0.155 |
stories | -0.147 |
Token | nam |
Token ID | 7402 |
Feature activation | +6.25e-02 |
Pos logit contributions | |
Whoever | +0.049 |
ooters | +0.043 |
ounded | +0.043 |
reaching | +0.036 |
Ready | +0.036 |
Neg logit contributions | |
ow | -0.137 |
ating | -0.123 |
urg | -0.117 |
happen | -0.116 |
ad | -0.115 |
Token | ents |
Token ID | 658 |
Feature activation | +5.58e-02 |
Pos logit contributions | |
Whoever | +0.057 |
ursion | +0.057 |
Ready | +0.056 |
sc | +0.055 |
ounded | +0.055 |
Neg logit contributions | |
ow | -0.151 |
Christmas | -0.131 |
feelings | -0.130 |
each | -0.129 |
ity | -0.127 |
Token | and |
Token ID | 290 |
Feature activation | +2.92e-02 |
Pos logit contributions | |
Whoever | +0.105 |
ounded | +0.101 |
sc | +0.101 |
ooters | +0.101 |
ursion | +0.097 |
Neg logit contributions | |
ow | -0.085 |
feelings | -0.060 |
lf | -0.058 |
Christmas | -0.058 |
happen | -0.057 |
Token | they |
Token ID | 484 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
Whoever | +0.058 |
ursion | +0.055 |
ooters | +0.054 |
ounded | +0.054 |
sc | +0.054 |
Neg logit contributions | |
ow | -0.042 |
happen | -0.031 |
feelings | -0.030 |
Christmas | -0.030 |
ating | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | -3.49e-03 |
Pos logit contributions | |
Whoever | +0.062 |
ursion | +0.060 |
ounded | +0.060 |
ooters | +0.059 |
sc | +0.058 |
Neg logit contributions | |
ow | -0.032 |
feelings | -0.022 |
Christmas | -0.022 |
happen | -0.022 |
stories | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
ow | +0.007 |
feelings | +0.006 |
happen | +0.006 |
Christmas | +0.005 |
stories | +0.005 |
Neg logit contributions | |
Whoever | -0.005 |
ursion | -0.005 |
ounded | -0.005 |
sc | -0.005 |
ooters | -0.005 |
Token | The |
Token ID | 464 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
ounded | +0.065 |
hole | +0.063 |
shadow | +0.063 |
ursion | +0.063 |
ooters | +0.062 |
Neg logit contributions | |
ow | -0.058 |
each | -0.052 |
songs | -0.052 |
feelings | -0.048 |
happen | -0.047 |
Token | game |
Token ID | 983 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
Whoever | +0.041 |
ursion | +0.040 |
ounded | +0.040 |
sc | +0.039 |
ooters | +0.038 |
Neg logit contributions | |
ow | -0.032 |
Christmas | -0.024 |
happen | -0.023 |
feelings | -0.023 |
stories | -0.022 |
Token | is |
Token ID | 318 |
Feature activation | +7.09e-03 |
Pos logit contributions | |
Whoever | +0.066 |
ounded | +0.065 |
ursion | +0.064 |
ooters | +0.063 |
reaching | +0.063 |
Neg logit contributions | |
ow | -0.027 |
happen | -0.018 |
feelings | -0.018 |
Christmas | -0.017 |
stories | -0.017 |
Token | called |
Token ID | 1444 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
ursion | +0.017 |
ounded | +0.016 |
Whoever | +0.016 |
reaching | +0.016 |
sc | +0.015 |
Neg logit contributions | |
ow | -0.008 |
stories | -0.006 |
happen | -0.006 |
feelings | -0.006 |
Christmas | -0.006 |
Token | hide |
Token ID | 7808 |
Feature activation | +5.40e-02 |
Pos logit contributions | |
ursion | +0.336 |
ounded | +0.329 |
Whoever | +0.327 |
reaching | +0.320 |
ooters | +0.309 |
Neg logit contributions | |
ow | -0.219 |
stories | -0.175 |
Christmas | -0.170 |
happen | -0.165 |
feelings | -0.165 |
Token | and |
Token ID | 290 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
Whoever | +0.114 |
sc | +0.110 |
ounded | +0.110 |
ursion | +0.109 |
ooters | +0.108 |
Neg logit contributions | |
ow | -0.072 |
happen | -0.051 |
Christmas | -0.049 |
feelings | -0.048 |
lf | -0.046 |
Token | seek |
Token ID | 5380 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Whoever | +0.035 |
ursion | +0.034 |
ounded | +0.034 |
ooters | +0.033 |
sc | +0.032 |
Neg logit contributions | |
ow | -0.024 |
feelings | -0.018 |
happen | -0.018 |
Christmas | -0.018 |
stories | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
Whoever | +0.028 |
ursion | +0.027 |
sc | +0.026 |
ounded | +0.026 |
ooters | +0.026 |
Neg logit contributions | |
ow | -0.018 |
feelings | -0.013 |
Christmas | -0.013 |
happen | -0.013 |
stories | -0.012 |
Token | You |
Token ID | 921 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
ounded | +0.038 |
ursion | +0.036 |
shadow | +0.036 |
reaching | +0.035 |
hole | +0.034 |
Neg logit contributions | |
ow | -0.032 |
each | -0.029 |
songs | -0.029 |
Christmas | -0.028 |
stories | -0.028 |
Token | She |
Token ID | 1375 |
Feature activation | +7.68e-03 |
Pos logit contributions | |
ounded | +0.042 |
shadow | +0.039 |
sc | +0.039 |
ooters | +0.039 |
ursion | +0.039 |
Neg logit contributions | |
ow | -0.030 |
each | -0.020 |
feelings | -0.019 |
pur | -0.019 |
jokes | -0.019 |
Token | says |
Token ID | 1139 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
Whoever | +0.010 |
ounded | +0.010 |
ursion | +0.010 |
ooters | +0.009 |
sc | +0.009 |
Neg logit contributions | |
ow | -0.017 |
happen | -0.014 |
stories | -0.014 |
feelings | -0.014 |
Christmas | -0.014 |
Token | the |
Token ID | 262 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
Whoever | +0.101 |
ursion | +0.097 |
ounded | +0.097 |
ooters | +0.094 |
sc | +0.093 |
Neg logit contributions | |
ow | -0.108 |
feelings | -0.085 |
Christmas | -0.085 |
happen | -0.085 |
stories | -0.082 |
Token | flower |
Token ID | 15061 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
Whoever | +0.043 |
ursion | +0.042 |
ounded | +0.042 |
sc | +0.041 |
Ready | +0.041 |
Neg logit contributions | |
ow | -0.023 |
happen | -0.016 |
Christmas | -0.016 |
feelings | -0.016 |
lf | -0.015 |
Token | is |
Token ID | 318 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
Whoever | +0.047 |
sc | +0.045 |
ursion | +0.045 |
ooters | +0.045 |
ounded | +0.044 |
Neg logit contributions | |
ow | -0.034 |
lf | -0.025 |
Christmas | -0.024 |
happen | -0.023 |
urg | -0.023 |
Token | called |
Token ID | 1444 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
Whoever | +0.030 |
ounded | +0.029 |
ursion | +0.029 |
ooters | +0.028 |
sc | +0.028 |
Neg logit contributions | |
ow | -0.021 |
feelings | -0.015 |
Christmas | -0.015 |
happen | -0.015 |
lf | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
Whoever | +0.340 |
ursion | +0.332 |
ounded | +0.330 |
ooters | +0.322 |
sc | +0.320 |
Neg logit contributions | |
ow | -0.224 |
Christmas | -0.163 |
feelings | -0.162 |
happen | -0.162 |
stories | -0.155 |
Token | tiger |
Token ID | 26241 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
Whoever | +0.068 |
ounded | +0.067 |
ursion | +0.067 |
ooters | +0.065 |
reaching | +0.064 |
Neg logit contributions | |
ow | -0.036 |
Christmas | -0.026 |
stories | -0.025 |
happen | -0.025 |
feelings | -0.024 |
Token | l |
Token ID | 300 |
Feature activation | -2.14e-03 |
Pos logit contributions | |
Whoever | +0.108 |
ounded | +0.108 |
ursion | +0.107 |
ooters | +0.104 |
reaching | +0.103 |
Neg logit contributions | |
ow | -0.042 |
Christmas | -0.028 |
feelings | -0.027 |
happen | -0.027 |
stories | -0.026 |
Token | ily |
Token ID | 813 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
ow | +0.003 |
soup | +0.002 |
ad | +0.002 |
kin | +0.002 |
Christmas | +0.002 |
Neg logit contributions | |
Whoever | -0.005 |
ounded | -0.005 |
shadow | -0.004 |
ursion | -0.004 |
Ready | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
Whoever | +0.113 |
ounded | +0.112 |
ursion | +0.111 |
ooters | +0.107 |
reaching | +0.107 |
Neg logit contributions | |
ow | -0.044 |
Christmas | -0.030 |
feelings | -0.030 |
happen | -0.028 |
stories | -0.028 |
Token | ." |
Token ID | 526 |
Feature activation | +1.60e-02 |
Pos logit contributions | |
Whoever | +0.040 |
ursion | +0.039 |
ounded | +0.039 |
ooters | +0.038 |
sc | +0.037 |
Neg logit contributions | |
ow | -0.030 |
feelings | -0.022 |
Christmas | -0.022 |
happen | -0.022 |
stories | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
ounded | +0.032 |
ooters | +0.030 |
sc | +0.030 |
ursion | +0.030 |
reaching | +0.029 |
Neg logit contributions | |
ow | -0.021 |
Christmas | -0.018 |
each | -0.018 |
all | -0.017 |
every | -0.017 |
Token | |
Token ID | 220 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
ow | +0.018 |
Christmas | +0.013 |
stories | +0.012 |
happen | +0.012 |
feelings | +0.012 |
Neg logit contributions | |
ounded | -0.034 |
Whoever | -0.034 |
ursion | -0.033 |
ooters | -0.032 |
sc | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
sc | +0.076 |
ounded | +0.074 |
Whoever | +0.070 |
ursion | +0.070 |
ooters | +0.069 |
Neg logit contributions | |
ow | -0.039 |
each | -0.034 |
Christmas | -0.032 |
all | -0.032 |
his | -0.031 |
Token | This |
Token ID | 1212 |
Feature activation | -7.80e-02 |
Pos logit contributions | |
ow | +0.030 |
Christmas | +0.023 |
lf | +0.022 |
feelings | +0.021 |
stories | +0.021 |
Neg logit contributions | |
ounded | -0.059 |
Whoever | -0.058 |
ursion | -0.058 |
ooters | -0.057 |
sc | -0.056 |
Token | story |
Token ID | 1621 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ow | +0.108 |
urg | +0.080 |
feelings | +0.075 |
happen | +0.074 |
ts | +0.073 |
Neg logit contributions | |
Whoever | -0.157 |
ursion | -0.153 |
sc | -0.143 |
bet | -0.142 |
Ready | -0.142 |
Token | teaches |
Token ID | 17324 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
ow | +0.265 |
feelings | +0.200 |
urg | +0.200 |
ities | +0.194 |
injuries | +0.194 |
Neg logit contributions | |
Whoever | -0.208 |
sc | -0.191 |
ursion | -0.190 |
ounded | -0.187 |
Ready | -0.174 |
Token | us |
Token ID | 514 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
ow | +0.046 |
kin | +0.033 |
happen | +0.032 |
lf | +0.032 |
feelings | +0.031 |
Neg logit contributions | |
Whoever | -0.062 |
ursion | -0.058 |
ooters | -0.056 |
sc | -0.056 |
ounded | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
ow | +0.029 |
Christmas | +0.019 |
lf | +0.018 |
happen | +0.018 |
feelings | +0.017 |
Neg logit contributions | |
Whoever | -0.055 |
ounded | -0.054 |
ursion | -0.053 |
ooters | -0.052 |
sc | -0.052 |
Token | valuable |
Token ID | 8119 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
ow | +0.020 |
Christmas | +0.016 |
feelings | +0.016 |
happen | +0.016 |
stories | +0.015 |
Neg logit contributions | |
Whoever | -0.029 |
ursion | -0.029 |
ooters | -0.028 |
ounded | -0.028 |
reaching | -0.028 |
Token | lesson |
Token ID | 11483 |
Feature activation | -5.15e-02 |
Pos logit contributions | |
ow | +0.079 |
feelings | +0.054 |
Christmas | +0.054 |
happen | +0.054 |
stories | +0.051 |
Neg logit contributions | |
Whoever | -0.137 |
ursion | -0.134 |
ounded | -0.133 |
sc | -0.130 |
ooters | -0.130 |
Token | beautiful |
Token ID | 4950 |
Feature activation | -6.07e-02 |
Pos logit contributions | |
ow | +0.053 |
happen | +0.039 |
feelings | +0.038 |
Christmas | +0.037 |
lf | +0.036 |
Neg logit contributions | |
Whoever | -0.074 |
ounded | -0.072 |
ursion | -0.072 |
sc | -0.071 |
ooters | -0.070 |
Token | melody |
Token ID | 32859 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
ow | +0.085 |
happen | +0.061 |
feelings | +0.060 |
Christmas | +0.057 |
urg | +0.057 |
Neg logit contributions | |
Whoever | -0.112 |
ursion | -0.112 |
sc | -0.110 |
ounded | -0.110 |
ooters | -0.108 |
Token | , |
Token ID | 11 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
ow | +0.024 |
Christmas | +0.017 |
feelings | +0.017 |
happen | +0.016 |
urg | +0.016 |
Neg logit contributions | |
Whoever | -0.034 |
sc | -0.034 |
ursion | -0.034 |
ounded | -0.033 |
ooters | -0.032 |
Token | which |
Token ID | 543 |
Feature activation | +1.63e-02 |
Pos logit contributions | |
ow | +0.014 |
feelings | +0.010 |
stories | +0.010 |
Christmas | +0.010 |
happen | +0.010 |
Neg logit contributions | |
ursion | -0.021 |
Whoever | -0.021 |
ounded | -0.021 |
ooters | -0.021 |
sc | -0.020 |
Token | still |
Token ID | 991 |
Feature activation | +1.84e-03 |
Pos logit contributions | |
Whoever | +0.033 |
ursion | +0.032 |
sc | +0.032 |
ounded | +0.031 |
Ready | +0.031 |
Neg logit contributions | |
ow | -0.021 |
feelings | -0.015 |
Christmas | -0.015 |
urg | -0.015 |
ity | -0.014 |
Token | ling |
Token ID | 18459 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Whoever | +0.004 |
ounded | +0.004 |
ursion | +0.004 |
sc | +0.004 |
ooters | +0.003 |
Neg logit contributions | |
ow | -0.002 |
feelings | -0.002 |
Christmas | -0.002 |
lf | -0.002 |
happen | -0.002 |
Token | ers |
Token ID | 364 |
Feature activation | +3.60e-02 |
Pos logit contributions | |
ow | +0.178 |
Christmas | +0.127 |
feelings | +0.126 |
happen | +0.122 |
stories | +0.120 |
Neg logit contributions | |
Whoever | -0.274 |
ounded | -0.267 |
ursion | -0.267 |
sc | -0.259 |
ooters | -0.258 |
Token | around |
Token ID | 1088 |
Feature activation | -2.99e-03 |
Pos logit contributions | |
Whoever | +0.080 |
ursion | +0.078 |
ounded | +0.078 |
ooters | +0.076 |
sc | +0.076 |
Neg logit contributions | |
ow | -0.043 |
feelings | -0.029 |
Christmas | -0.029 |
happen | -0.029 |
stories | -0.028 |
Token | today |
Token ID | 1909 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
ow | +0.004 |
feelings | +0.002 |
Christmas | +0.002 |
urg | +0.002 |
happen | +0.002 |
Neg logit contributions | |
Whoever | -0.006 |
sc | -0.006 |
ursion | -0.006 |
ounded | -0.006 |
Ready | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
ow | +0.015 |
feelings | +0.010 |
Christmas | +0.010 |
happen | +0.010 |
stories | +0.009 |
Neg logit contributions | |
Whoever | -0.028 |
ursion | -0.027 |
ounded | -0.027 |
sc | -0.027 |
ooters | -0.027 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
ow | +0.012 |
Christmas | +0.011 |
lived | +0.011 |
every | +0.010 |
her | +0.010 |
Neg logit contributions | |
ounded | -0.028 |
ooters | -0.025 |
hole | -0.024 |
sc | -0.024 |
ursion | -0.024 |
Token | happened |
Token ID | 3022 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
ow | +0.058 |
lf | +0.041 |
Christmas | +0.040 |
feelings | +0.040 |
ity | +0.040 |
Neg logit contributions | |
Whoever | -0.080 |
ounded | -0.077 |
ursion | -0.077 |
ooters | -0.075 |
sc | -0.074 |
Token | and |
Token ID | 290 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
ow | +0.034 |
feelings | +0.023 |
happen | +0.023 |
Christmas | +0.023 |
stories | +0.022 |
Neg logit contributions | |
Whoever | -0.064 |
ursion | -0.062 |
ounded | -0.062 |
ooters | -0.060 |
sc | -0.060 |
Token | Teddy |
Token ID | 29345 |
Feature activation | -5.33e-02 |
Pos logit contributions | |
ow | +0.045 |
happen | +0.035 |
Christmas | +0.035 |
feelings | +0.035 |
each | +0.034 |
Neg logit contributions | |
Whoever | -0.056 |
ursion | -0.056 |
ooters | -0.055 |
sc | -0.054 |
ounded | -0.054 |
Token | shared |
Token ID | 4888 |
Feature activation | -9.08e-02 |
Pos logit contributions | |
ow | +0.101 |
Christmas | +0.083 |
happen | +0.080 |
feelings | +0.080 |
stories | +0.078 |
Neg logit contributions | |
Whoever | -0.081 |
ounded | -0.077 |
ursion | -0.076 |
sc | -0.072 |
reaching | -0.071 |
Token | his |
Token ID | 465 |
Feature activation | -6.29e-02 |
Pos logit contributions | |
ow | +0.120 |
happen | +0.078 |
Christmas | +0.076 |
urg | +0.075 |
lf | +0.075 |
Neg logit contributions | |
Whoever | -0.193 |
sc | -0.187 |
Ready | -0.174 |
ooters | -0.174 |
ounded | -0.174 |
Token | story |
Token ID | 1621 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
ow | +0.081 |
ating | +0.056 |
happen | +0.055 |
Christmas | +0.055 |
urg | +0.053 |
Neg logit contributions | |
Whoever | -0.126 |
ooters | -0.124 |
sc | -0.124 |
ounded | -0.124 |
ursion | -0.122 |
Token | of |
Token ID | 286 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
ow | +0.177 |
Christmas | +0.116 |
feelings | +0.112 |
happen | +0.110 |
unic | +0.106 |
Neg logit contributions | |
Whoever | -0.282 |
ounded | -0.272 |
ooters | -0.268 |
sc | -0.268 |
ursion | -0.265 |
Token | how |
Token ID | 703 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
ow | +0.042 |
urg | +0.027 |
ating | +0.026 |
Christmas | +0.026 |
happen | +0.025 |
Neg logit contributions | |
Whoever | -0.073 |
ooters | -0.068 |
sc | -0.067 |
Ready | -0.067 |
Ready | -0.065 |
Token | time |
Token ID | 640 |
Feature activation | +1.76e-02 |
Pos logit contributions | |
ow | +0.057 |
feelings | +0.045 |
Christmas | +0.044 |
happen | +0.044 |
stories | +0.043 |
Neg logit contributions | |
Whoever | -0.065 |
ursion | -0.062 |
ounded | -0.062 |
sc | -0.061 |
ooters | -0.060 |
Token | and |
Token ID | 290 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
Whoever | +0.030 |
ursion | +0.028 |
ounded | +0.028 |
sc | +0.028 |
ooters | +0.028 |
Neg logit contributions | |
ow | -0.030 |
feelings | -0.023 |
Christmas | -0.023 |
happen | -0.023 |
stories | -0.022 |
Token | patience |
Token ID | 16336 |
Feature activation | -7.49e-03 |
Pos logit contributions | |
ow | +0.044 |
Christmas | +0.032 |
happen | +0.032 |
feelings | +0.032 |
ity | +0.030 |
Neg logit contributions | |
Whoever | -0.052 |
ounded | -0.050 |
ursion | -0.050 |
sc | -0.048 |
Ready | -0.048 |
Token | But |
Token ID | 887 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
ow | +0.021 |
all | +0.017 |
each | +0.017 |
his | +0.017 |
himself | +0.016 |
Neg logit contributions | |
ooters | -0.047 |
ounded | -0.047 |
sc | -0.047 |
ursion | -0.045 |
hole | -0.043 |
Token | the |
Token ID | 262 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
Whoever | +0.049 |
ounded | +0.047 |
ursion | +0.047 |
sc | +0.045 |
reaching | +0.045 |
Neg logit contributions | |
ow | -0.034 |
feelings | -0.025 |
happen | -0.024 |
Christmas | -0.024 |
stories | -0.023 |
Token | harsh |
Token ID | 11859 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
Whoever | +0.026 |
ursion | +0.026 |
ounded | +0.025 |
Ready | +0.025 |
sc | +0.024 |
Neg logit contributions | |
ow | -0.019 |
happen | -0.015 |
Christmas | -0.015 |
ating | -0.014 |
feelings | -0.014 |
Token | morning |
Token ID | 3329 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
ow | +0.039 |
Christmas | +0.031 |
happen | +0.030 |
ils | +0.030 |
ts | +0.029 |
Neg logit contributions | |
Whoever | -0.047 |
ounded | -0.046 |
ursion | -0.045 |
ooters | -0.044 |
sc | -0.043 |
Token | still |
Token ID | 991 |
Feature activation | -5.56e-03 |
Pos logit contributions | |
Whoever | +0.052 |
ursion | +0.050 |
ounded | +0.050 |
sc | +0.049 |
Ready | +0.048 |
Neg logit contributions | |
ow | -0.032 |
feelings | -0.024 |
happen | -0.023 |
Christmas | -0.022 |
stories | -0.022 |
Token | ling |
Token ID | 18459 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
ow | +0.008 |
feelings | +0.006 |
Christmas | +0.006 |
lf | +0.006 |
stories | +0.006 |
Neg logit contributions | |
Whoever | -0.010 |
ounded | -0.010 |
ursion | -0.010 |
reaching | -0.010 |
sc | -0.010 |
Token | ered |
Token ID | 1068 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
ow | +0.178 |
Christmas | +0.128 |
feelings | +0.128 |
happen | +0.123 |
stories | +0.121 |
Neg logit contributions | |
Whoever | -0.269 |
ounded | -0.262 |
ursion | -0.262 |
sc | -0.255 |
ooters | -0.253 |
Token | in |
Token ID | 287 |
Feature activation | +2.08e-03 |
Pos logit contributions | |
Whoever | +0.047 |
ursion | +0.046 |
ounded | +0.046 |
ooters | +0.044 |
sc | +0.044 |
Neg logit contributions | |
ow | -0.035 |
feelings | -0.026 |
Christmas | -0.026 |
happen | -0.026 |
stories | -0.025 |
Token | his |
Token ID | 465 |
Feature activation | -3.88e-02 |
Pos logit contributions | |
Whoever | +0.004 |
ursion | +0.004 |
reaching | +0.004 |
Ready | +0.004 |
sc | +0.004 |
Neg logit contributions | |
ow | -0.003 |
feelings | -0.002 |
lf | -0.002 |
happen | -0.002 |
ity | -0.002 |
Token | heart |
Token ID | 2612 |
Feature activation | -9.28e-02 |
Pos logit contributions | |
ow | +0.054 |
happen | +0.042 |
lf | +0.042 |
Christmas | +0.039 |
ating | +0.039 |
Neg logit contributions | |
Whoever | -0.074 |
ounded | -0.074 |
ursion | -0.073 |
ooters | -0.072 |
reaching | -0.071 |
Token | . |
Token ID | 13 |
Feature activation | -2.31e-02 |
Pos logit contributions | |
ow | +0.118 |
feelings | +0.080 |
Christmas | +0.080 |
happen | +0.080 |
stories | +0.075 |
Neg logit contributions | |
Whoever | -0.200 |
ounded | -0.194 |
ursion | -0.194 |
ooters | -0.189 |
sc | -0.189 |
Token | met |
Token ID | 1138 |
Feature activation | +3.02e-02 |
Pos logit contributions | |
Whoever | +0.045 |
ursion | +0.044 |
ounded | +0.043 |
sc | +0.042 |
ooters | +0.041 |
Neg logit contributions | |
ow | -0.043 |
feelings | -0.034 |
happen | -0.033 |
Christmas | -0.032 |
stories | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
Whoever | +0.066 |
ounded | +0.063 |
ursion | +0.063 |
sc | +0.063 |
ooters | +0.062 |
Neg logit contributions | |
ow | -0.038 |
happen | -0.026 |
Christmas | -0.025 |
feelings | -0.025 |
lf | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
ow | +0.019 |
all | +0.018 |
each | +0.018 |
every | +0.017 |
his | +0.017 |
Neg logit contributions | |
ounded | -0.044 |
hole | -0.040 |
shadow | -0.040 |
ursion | -0.039 |
sc | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
ow | +0.023 |
ad | +0.017 |
Christmas | +0.017 |
each | +0.017 |
all | +0.017 |
Neg logit contributions | |
ounded | -0.046 |
Whoever | -0.044 |
ursion | -0.044 |
hole | -0.043 |
ooters | -0.043 |
Token | This |
Token ID | 1212 |
Feature activation | -7.86e-02 |
Pos logit contributions | |
ow | +0.038 |
lf | +0.030 |
Christmas | +0.030 |
stories | +0.030 |
end | +0.030 |
Neg logit contributions | |
ounded | -0.055 |
Whoever | -0.053 |
ursion | -0.052 |
ooters | -0.052 |
hole | -0.051 |
Token | story |
Token ID | 1621 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
ow | +0.105 |
urg | +0.076 |
feelings | +0.072 |
ts | +0.071 |
ble | +0.070 |
Neg logit contributions | |
Whoever | -0.162 |
ursion | -0.159 |
sc | -0.156 |
Ready | -0.151 |
ounded | -0.149 |
Token | teaches |
Token ID | 17324 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
ow | +0.223 |
feelings | +0.166 |
urg | +0.159 |
ities | +0.154 |
Christmas | +0.154 |
Neg logit contributions | |
Whoever | -0.234 |
sc | -0.225 |
ursion | -0.215 |
ounded | -0.215 |
Ready | -0.211 |
Token | us |
Token ID | 514 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
ow | +0.034 |
Christmas | +0.023 |
happen | +0.023 |
lf | +0.023 |
kin | +0.023 |
Neg logit contributions | |
Whoever | -0.052 |
ooters | -0.048 |
ursion | -0.048 |
sc | -0.048 |
ounded | -0.047 |
Token | that |
Token ID | 326 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
ow | +0.022 |
Christmas | +0.015 |
lf | +0.014 |
happen | +0.014 |
feelings | +0.014 |
Neg logit contributions | |
Whoever | -0.039 |
ounded | -0.038 |
ursion | -0.037 |
ooters | -0.037 |
sc | -0.037 |
Token | you |
Token ID | 345 |
Feature activation | -5.56e-03 |
Pos logit contributions | |
ow | +0.055 |
feelings | +0.041 |
Christmas | +0.041 |
happen | +0.041 |
stories | +0.039 |
Neg logit contributions | |
Whoever | -0.072 |
ursion | -0.070 |
ounded | -0.070 |
ooters | -0.068 |
sc | -0.068 |
Token | should |
Token ID | 815 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
ow | +0.011 |
Christmas | +0.008 |
lf | +0.008 |
feelings | +0.008 |
happen | +0.008 |
Neg logit contributions | |
Whoever | -0.009 |
ounded | -0.009 |
ursion | -0.008 |
sc | -0.008 |
ooters | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +3.71e-03 |
Pos logit contributions | |
ow | +0.150 |
feelings | +0.107 |
Christmas | +0.105 |
happen | +0.105 |
stories | +0.098 |
Neg logit contributions | |
Whoever | -0.222 |
ounded | -0.214 |
ursion | -0.213 |
ooters | -0.210 |
sc | -0.209 |
Token |
|
Token ID | 198 |
Feature activation | -5.97e-03 |
Pos logit contributions | |
ounded | +0.008 |
ursion | +0.008 |
sc | +0.008 |
ooters | +0.008 |
shadow | +0.008 |
Neg logit contributions | |
ow | -0.004 |
Christmas | -0.003 |
each | -0.003 |
her | -0.003 |
every | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
ow | +0.007 |
lf | +0.006 |
Christmas | +0.006 |
each | +0.005 |
ad | +0.005 |
Neg logit contributions | |
ounded | -0.013 |
ursion | -0.012 |
ooters | -0.012 |
Whoever | -0.012 |
sc | -0.012 |
Token | " |
Token ID | 1 |
Feature activation | -6.73e-03 |
Pos logit contributions | |
ow | +0.016 |
lf | +0.015 |
Christmas | +0.014 |
end | +0.013 |
each | +0.013 |
Neg logit contributions | |
ounded | -0.027 |
ooters | -0.026 |
ursion | -0.026 |
Whoever | -0.025 |
sc | -0.025 |
Token | Which |
Token ID | 13828 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
ow | +0.011 |
Christmas | +0.009 |
feelings | +0.008 |
stories | +0.008 |
happen | +0.008 |
Neg logit contributions | |
Whoever | -0.012 |
ounded | -0.012 |
ursion | -0.012 |
ooters | -0.012 |
reaching | -0.011 |
Token | story |
Token ID | 1621 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ow | +0.067 |
urg | +0.052 |
ating | +0.049 |
feelings | +0.048 |
ling | +0.047 |
Neg logit contributions | |
Whoever | -0.092 |
ursion | -0.089 |
sc | -0.089 |
Ready | -0.088 |
ooters | -0.086 |
Token | do |
Token ID | 466 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
ow | +0.167 |
feelings | +0.118 |
Christmas | +0.118 |
happen | +0.118 |
stories | +0.112 |
Neg logit contributions | |
Whoever | -0.277 |
ursion | -0.269 |
ounded | -0.268 |
ooters | -0.262 |
sc | -0.261 |
Token | you |
Token ID | 345 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
ow | +0.225 |
feelings | +0.181 |
Christmas | +0.180 |
happen | +0.178 |
stories | +0.173 |
Neg logit contributions | |
Whoever | -0.160 |
ursion | -0.148 |
ooters | -0.146 |
ounded | -0.145 |
sc | -0.142 |
Token | want |
Token ID | 765 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
ow | +0.057 |
feelings | +0.046 |
Christmas | +0.045 |
happen | +0.045 |
lf | +0.044 |
Neg logit contributions | |
Whoever | -0.038 |
ursion | -0.036 |
ounded | -0.035 |
ooters | -0.033 |
sc | -0.033 |
Token | , |
Token ID | 11 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
ow | +0.074 |
feelings | +0.057 |
Christmas | +0.055 |
happen | +0.055 |
ity | +0.055 |
Neg logit contributions | |
Whoever | -0.068 |
ursion | -0.065 |
sc | -0.064 |
ooters | -0.063 |
ounded | -0.063 |
Token | Anna |
Token ID | 11735 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
ow | +0.048 |
feelings | +0.032 |
Christmas | +0.032 |
happen | +0.031 |
stories | +0.030 |
Neg logit contributions | |
Whoever | -0.099 |
ursion | -0.098 |
ounded | -0.097 |
sc | -0.095 |
ooters | -0.095 |
Token |
|
Token ID | 198 |
Feature activation | -2.21e-02 |
Pos logit contributions | |
ow | +0.012 |
Christmas | +0.008 |
stories | +0.007 |
happen | +0.007 |
feelings | +0.007 |
Neg logit contributions | |
Whoever | -0.029 |
ounded | -0.029 |
ursion | -0.029 |
sc | -0.028 |
ooters | -0.028 |
Token | The |
Token ID | 464 |
Feature activation | -9.09e-03 |
Pos logit contributions | |
ow | +0.039 |
Christmas | +0.031 |
feelings | +0.031 |
lf | +0.031 |
happen | +0.031 |
Neg logit contributions | |
Whoever | -0.035 |
ounded | -0.035 |
ursion | -0.035 |
ooters | -0.034 |
sc | -0.033 |
Token | moral |
Token ID | 6573 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
ow | +0.009 |
happen | +0.005 |
feelings | +0.005 |
Christmas | +0.005 |
ating | +0.005 |
Neg logit contributions | |
Whoever | -0.022 |
ursion | -0.022 |
ounded | -0.021 |
Ready | -0.021 |
ooters | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
ow | +0.053 |
stories | +0.044 |
ad | +0.044 |
Christmas | +0.044 |
feelings | +0.042 |
Neg logit contributions | |
ounded | -0.078 |
Whoever | -0.077 |
ursion | -0.074 |
shadow | -0.073 |
ooters | -0.072 |
Token | his |
Token ID | 465 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
ow | +0.061 |
feelings | +0.049 |
ities | +0.047 |
ity | +0.045 |
ling | +0.043 |
Neg logit contributions | |
Whoever | -0.055 |
old | -0.050 |
sc | -0.050 |
ursion | -0.049 |
First | -0.048 |
Token | story |
Token ID | 1621 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ow | +0.057 |
ating | +0.040 |
feelings | +0.035 |
ts | +0.035 |
happen | +0.034 |
Neg logit contributions | |
Whoever | -0.130 |
ursion | -0.123 |
ounded | -0.123 |
ooters | -0.123 |
reaching | -0.118 |
Token | is |
Token ID | 318 |
Feature activation | -3.94e-02 |
Pos logit contributions | |
ow | +0.161 |
Christmas | +0.109 |
feelings | +0.105 |
happen | +0.102 |
soup | +0.100 |
Neg logit contributions | |
Whoever | -0.287 |
ounded | -0.279 |
ursion | -0.277 |
sc | -0.273 |
ooters | -0.272 |
Token | that |
Token ID | 326 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
ow | +0.035 |
happen | +0.022 |
Christmas | +0.022 |
feelings | +0.021 |
stories | +0.020 |
Neg logit contributions | |
Whoever | -0.099 |
ursion | -0.097 |
ounded | -0.097 |
sc | -0.097 |
ooters | -0.095 |
Token | brave |
Token ID | 14802 |
Feature activation | -9.54e-03 |
Pos logit contributions | |
ow | +0.059 |
feelings | +0.044 |
happen | +0.044 |
Christmas | +0.044 |
stories | +0.042 |
Neg logit contributions | |
Whoever | -0.077 |
ursion | -0.075 |
ounded | -0.075 |
ooters | -0.073 |
sc | -0.073 |
Token | doesn |
Token ID | 1595 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
ow | +0.014 |
feelings | +0.010 |
happen | +0.010 |
Christmas | +0.010 |
stories | +0.010 |
Neg logit contributions | |
Whoever | -0.019 |
ursion | -0.019 |
ounded | -0.019 |
ooters | -0.018 |
sc | -0.018 |
Token | 't |
Token ID | 470 |
Feature activation | -1.71e-03 |
Pos logit contributions | |
ow | +0.023 |
feelings | +0.020 |
stories | +0.019 |
Christmas | +0.019 |
soup | +0.018 |
Neg logit contributions | |
Whoever | -0.019 |
ursion | -0.017 |
ounded | -0.017 |
sc | -0.016 |
reaching | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ow | +0.024 |
feelings | +0.016 |
Christmas | +0.016 |
happen | +0.016 |
stories | +0.015 |
Neg logit contributions | |
Whoever | -0.043 |
ursion | -0.042 |
ounded | -0.042 |
ooters | -0.041 |
sc | -0.041 |
Token | |
Token ID | 220 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
ow | +0.017 |
Christmas | +0.015 |
ad | +0.015 |
each | +0.015 |
every | +0.014 |
Neg logit contributions | |
ounded | -0.049 |
sc | -0.043 |
hole | -0.041 |
ursion | -0.041 |
ooters | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
ounded | +0.108 |
sc | +0.105 |
ursion | +0.103 |
Whoever | +0.100 |
Ready | +0.097 |
Neg logit contributions | |
ow | -0.045 |
all | -0.036 |
each | -0.036 |
ad | -0.032 |
every | -0.031 |
Token |
|
Token ID | 198 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
ow | +0.028 |
ad | +0.022 |
Christmas | +0.021 |
stories | +0.021 |
lf | +0.021 |
Neg logit contributions | |
ounded | -0.053 |
ursion | -0.050 |
Whoever | -0.049 |
sc | -0.048 |
hole | -0.048 |
Token | This |
Token ID | 1212 |
Feature activation | -7.49e-02 |
Pos logit contributions | |
ow | +0.034 |
Christmas | +0.028 |
ad | +0.028 |
lf | +0.028 |
stories | +0.027 |
Neg logit contributions | |
ounded | -0.063 |
ursion | -0.059 |
Whoever | -0.059 |
hole | -0.058 |
ooters | -0.058 |
Token | story |
Token ID | 1621 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
ow | +0.102 |
urg | +0.073 |
feelings | +0.072 |
happen | +0.070 |
ts | +0.069 |
Neg logit contributions | |
Whoever | -0.149 |
ursion | -0.147 |
sc | -0.141 |
Ready | -0.139 |
ounded | -0.138 |
Token | teaches |
Token ID | 17324 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
ow | +0.243 |
feelings | +0.181 |
urg | +0.178 |
Christmas | +0.174 |
ity | +0.172 |
Neg logit contributions | |
Whoever | -0.212 |
sc | -0.198 |
ursion | -0.197 |
ounded | -0.194 |
Ready | -0.187 |
Token | us |
Token ID | 514 |
Feature activation | -2.00e-02 |
Pos logit contributions | |
ow | +0.042 |
happen | +0.031 |
lf | +0.030 |
Christmas | +0.030 |
kin | +0.030 |
Neg logit contributions | |
Whoever | -0.052 |
ursion | -0.048 |
ooters | -0.048 |
sc | -0.047 |
ounded | -0.046 |
Token | that |
Token ID | 326 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
ow | +0.024 |
Christmas | +0.016 |
lf | +0.015 |
happen | +0.015 |
feelings | +0.014 |
Neg logit contributions | |
Whoever | -0.045 |
ounded | -0.043 |
ursion | -0.043 |
ooters | -0.043 |
sc | -0.042 |
Token | when |
Token ID | 618 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
ow | +0.053 |
Christmas | +0.039 |
feelings | +0.039 |
happen | +0.038 |
stories | +0.037 |
Neg logit contributions | |
Whoever | -0.068 |
ursion | -0.065 |
ounded | -0.065 |
sc | -0.064 |
ooters | -0.063 |
Token | we |
Token ID | 356 |
Feature activation | +2.10e-02 |
Pos logit contributions | |
Whoever | +0.089 |
ursion | +0.085 |
ounded | +0.084 |
ooters | +0.082 |
sc | +0.081 |
Neg logit contributions | |
ow | -0.091 |
feelings | -0.071 |
Christmas | -0.070 |
happen | -0.070 |
stories | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
ow | +0.015 |
ad | +0.013 |
lf | +0.012 |
each | +0.011 |
story | +0.011 |
Neg logit contributions | |
ounded | -0.032 |
hole | -0.030 |
ursion | -0.030 |
Whoever | -0.029 |
ooters | -0.029 |
Token | The |
Token ID | 464 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
ow | +0.021 |
lf | +0.019 |
ad | +0.019 |
each | +0.017 |
end | +0.017 |
Neg logit contributions | |
ounded | -0.050 |
hole | -0.048 |
ursion | -0.046 |
ooters | -0.046 |
Whoever | -0.045 |
Token | moral |
Token ID | 6573 |
Feature activation | -5.13e-02 |
Pos logit contributions | |
ow | +0.018 |
Christmas | +0.013 |
feelings | +0.013 |
happen | +0.012 |
stories | +0.012 |
Neg logit contributions | |
Whoever | -0.033 |
ounded | -0.032 |
ursion | -0.032 |
sc | -0.031 |
ooters | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
ow | +0.068 |
stories | +0.060 |
ad | +0.058 |
Christmas | +0.056 |
feelings | +0.056 |
Neg logit contributions | |
ounded | -0.102 |
Whoever | -0.100 |
shadow | -0.098 |
ursion | -0.097 |
reaching | -0.093 |
Token | this |
Token ID | 428 |
Feature activation | -5.77e-02 |
Pos logit contributions | |
ow | +0.059 |
feelings | +0.044 |
ity | +0.040 |
ities | +0.040 |
creations | +0.039 |
Neg logit contributions | |
Whoever | -0.083 |
ursion | -0.076 |
sc | -0.076 |
shadow | -0.074 |
reaching | -0.073 |
Token | story |
Token ID | 1621 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ow | +0.062 |
feelings | +0.040 |
ts | +0.038 |
Christmas | +0.038 |
ities | +0.035 |
Neg logit contributions | |
Whoever | -0.134 |
ounded | -0.128 |
ursion | -0.128 |
ooters | -0.127 |
sc | -0.125 |
Token | is |
Token ID | 318 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
ow | +0.164 |
Christmas | +0.111 |
feelings | +0.108 |
happen | +0.103 |
soup | +0.100 |
Neg logit contributions | |
Whoever | -0.281 |
ursion | -0.270 |
ounded | -0.269 |
sc | -0.265 |
ooters | -0.265 |
Token | that |
Token ID | 326 |
Feature activation | -4.15e-02 |
Pos logit contributions | |
ow | +0.037 |
happen | +0.024 |
Christmas | +0.023 |
feelings | +0.023 |
stories | +0.022 |
Neg logit contributions | |
Whoever | -0.105 |
ursion | -0.103 |
ounded | -0.103 |
sc | -0.102 |
ooters | -0.100 |
Token | it |
Token ID | 340 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
ow | +0.063 |
happen | +0.050 |
feelings | +0.050 |
Christmas | +0.049 |
stories | +0.048 |
Neg logit contributions | |
Whoever | -0.078 |
ursion | -0.078 |
ounded | -0.077 |
ooters | -0.075 |
sc | -0.074 |
Token | â |
Token ID | 22940 |
Feature activation | +4.06e-03 |
Pos logit contributions | |
ow | +0.056 |
feelings | +0.043 |
happen | +0.043 |
Christmas | +0.043 |
stories | +0.042 |
Neg logit contributions | |
Whoever | -0.065 |
ursion | -0.062 |
ounded | -0.062 |
ooters | -0.061 |
sc | -0.060 |
Token | € |
Token ID | 26391 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
Whoever | +0.005 |
ounded | +0.005 |
ursion | +0.005 |
ooters | +0.005 |
sc | +0.005 |
Neg logit contributions | |
ow | -0.008 |
happen | -0.007 |
feelings | -0.007 |
Christmas | -0.007 |
stories | -0.007 |
Token | of |
Token ID | 286 |
Feature activation | +6.42e-03 |
Pos logit contributions | |
ow | +0.151 |
Christmas | +0.132 |
feelings | +0.132 |
ities | +0.130 |
urg | +0.123 |
Neg logit contributions | |
sc | -0.119 |
Whoever | -0.106 |
ounded | -0.102 |
ursion | -0.098 |
reaching | -0.091 |
Token | that |
Token ID | 326 |
Feature activation | -3.55e-02 |
Pos logit contributions | |
Whoever | +0.013 |
sc | +0.012 |
Ready | +0.012 |
ursion | +0.012 |
ooters | +0.012 |
Neg logit contributions | |
ow | -0.008 |
lf | -0.006 |
ating | -0.006 |
Christmas | -0.006 |
ity | -0.006 |
Token | crazy |
Token ID | 7165 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
ow | +0.059 |
Christmas | +0.045 |
feelings | +0.045 |
lf | +0.044 |
urg | +0.043 |
Neg logit contributions | |
Whoever | -0.060 |
ounded | -0.058 |
ursion | -0.058 |
sc | -0.057 |
Ready | -0.055 |
Token | tuna |
Token ID | 38883 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
ow | +0.020 |
Christmas | +0.014 |
feelings | +0.013 |
happen | +0.013 |
lf | +0.013 |
Neg logit contributions | |
Whoever | -0.031 |
sc | -0.031 |
ursion | -0.029 |
ounded | -0.029 |
ooters | -0.029 |
Token | still |
Token ID | 991 |
Feature activation | +4.33e-04 |
Pos logit contributions | |
ow | +0.039 |
Christmas | +0.030 |
feelings | +0.028 |
unic | +0.027 |
urg | +0.026 |
Neg logit contributions | |
sc | -0.056 |
Whoever | -0.056 |
ounded | -0.053 |
ursion | -0.053 |
reaching | -0.050 |
Token | ling |
Token ID | 18459 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
Whoever | +0.001 |
ounded | +0.001 |
sc | +0.001 |
ursion | +0.001 |
reaching | +0.001 |
Neg logit contributions | |
ow | -0.001 |
Christmas | -0.000 |
feelings | -0.000 |
lf | -0.000 |
urg | -0.000 |
Token | ered |
Token ID | 1068 |
Feature activation | +2.80e-02 |
Pos logit contributions | |
ow | +0.179 |
Christmas | +0.128 |
feelings | +0.126 |
happen | +0.121 |
stories | +0.120 |
Neg logit contributions | |
Whoever | -0.262 |
ounded | -0.257 |
ursion | -0.255 |
sc | -0.251 |
ooters | -0.246 |
Token | in |
Token ID | 287 |
Feature activation | +9.07e-03 |
Pos logit contributions | |
Whoever | +0.055 |
ounded | +0.054 |
ursion | +0.053 |
sc | +0.052 |
ooters | +0.052 |
Neg logit contributions | |
ow | -0.040 |
happen | -0.030 |
feelings | -0.030 |
Christmas | -0.029 |
lf | -0.028 |
Token | her |
Token ID | 607 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
Whoever | +0.019 |
sc | +0.018 |
Ready | +0.018 |
ursion | +0.018 |
wanna | +0.017 |
Neg logit contributions | |
ow | -0.011 |
happen | -0.008 |
feelings | -0.007 |
lf | -0.007 |
Christmas | -0.007 |
Token | mouth |
Token ID | 5422 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
ow | +0.042 |
happen | +0.032 |
lf | +0.032 |
Christmas | +0.030 |
ating | +0.029 |
Neg logit contributions | |
Whoever | -0.052 |
ounded | -0.050 |
ursion | -0.050 |
sc | -0.050 |
ooters | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | -2.40e-02 |
Pos logit contributions | |
ow | +0.050 |
feelings | +0.035 |
Christmas | +0.035 |
happen | +0.035 |
stories | +0.033 |
Neg logit contributions | |
Whoever | -0.082 |
ursion | -0.079 |
ounded | -0.079 |
sc | -0.078 |
ooters | -0.077 |
Token | belly |
Token ID | 19921 |
Feature activation | -6.79e-02 |
Pos logit contributions | |
ow | +0.046 |
ating | +0.033 |
happen | +0.033 |
feelings | +0.032 |
Christmas | +0.032 |
Neg logit contributions | |
Whoever | -0.071 |
ursion | -0.071 |
sc | -0.069 |
ounded | -0.068 |
ooters | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -1.98e-02 |
Pos logit contributions | |
ow | +0.094 |
feelings | +0.070 |
happen | +0.068 |
Christmas | +0.067 |
ating | +0.064 |
Neg logit contributions | |
Whoever | -0.131 |
sc | -0.125 |
ursion | -0.123 |
ounded | -0.122 |
ooters | -0.121 |
Token |
|
Token ID | 198 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
ow | +0.023 |
Christmas | +0.019 |
each | +0.018 |
all | +0.018 |
his | +0.017 |
Neg logit contributions | |
ounded | -0.044 |
ooters | -0.042 |
sc | -0.042 |
shadow | -0.040 |
hole | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -3.03e-02 |
Pos logit contributions | |
ow | +0.030 |
Christmas | +0.023 |
lf | +0.022 |
each | +0.022 |
happen | +0.022 |
Neg logit contributions | |
ounded | -0.044 |
ursion | -0.042 |
Whoever | -0.042 |
ooters | -0.041 |
sc | -0.041 |
Token | This |
Token ID | 1212 |
Feature activation | -7.55e-02 |
Pos logit contributions | |
ow | +0.038 |
Christmas | +0.030 |
lf | +0.030 |
each | +0.028 |
feelings | +0.028 |
Neg logit contributions | |
ounded | -0.063 |
ooters | -0.061 |
ursion | -0.060 |
Whoever | -0.060 |
hole | -0.059 |
Token | story |
Token ID | 1621 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ow | +0.115 |
urg | +0.099 |
ts | +0.097 |
ble | +0.093 |
ating | +0.092 |
Neg logit contributions | |
Whoever | -0.127 |
ursion | -0.120 |
sc | -0.116 |
Ready | -0.113 |
bet | -0.113 |
Token | ended |
Token ID | 4444 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
ow | +0.234 |
feelings | +0.184 |
urg | +0.184 |
ities | +0.183 |
injuries | +0.177 |
Neg logit contributions | |
Whoever | -0.207 |
sc | -0.203 |
ursion | -0.184 |
ounded | -0.184 |
Ready | -0.173 |
Token | well |
Token ID | 880 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
ow | +0.086 |
happen | +0.060 |
unic | +0.058 |
Christmas | +0.056 |
feelings | +0.056 |
Neg logit contributions | |
Whoever | -0.105 |
ounded | -0.101 |
sc | -0.101 |
shadow | -0.098 |
reaching | -0.097 |
Token | and |
Token ID | 290 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
ow | +0.063 |
feelings | +0.044 |
happen | +0.044 |
Christmas | +0.044 |
stories | +0.042 |
Neg logit contributions | |
Whoever | -0.092 |
ounded | -0.089 |
ursion | -0.089 |
sc | -0.088 |
ooters | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | -2.84e-03 |
Pos logit contributions | |
ow | +0.038 |
feelings | +0.026 |
happen | +0.025 |
Christmas | +0.025 |
lf | +0.024 |
Neg logit contributions | |
Whoever | -0.062 |
ounded | -0.060 |
ursion | -0.059 |
sc | -0.058 |
Ready | -0.057 |
Token | family |
Token ID | 1641 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
ow | +0.004 |
ating | +0.003 |
happen | +0.003 |
urg | +0.003 |
ity | +0.003 |
Neg logit contributions | |
Whoever | -0.005 |
ursion | -0.005 |
ounded | -0.005 |
Ready | -0.005 |
ooters | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -2.19e-02 |
Pos logit contributions | |
Whoever | +0.134 |
ursion | +0.130 |
ounded | +0.130 |
sc | +0.128 |
ooters | +0.128 |
Neg logit contributions | |
ow | -0.073 |
happen | -0.048 |
feelings | -0.048 |
Christmas | -0.048 |
stories | -0.045 |
Token | |
Token ID | 220 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
every | +0.023 |
Christmas | +0.022 |
ow | +0.022 |
ad | +0.022 |
all | +0.022 |
Neg logit contributions | |
ounded | -0.046 |
sc | -0.042 |
ursion | -0.041 |
ooters | -0.040 |
shadow | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
ounded | +0.092 |
sc | +0.089 |
Whoever | +0.089 |
ursion | +0.088 |
ooters | +0.084 |
Neg logit contributions | |
ow | -0.053 |
all | -0.042 |
Christmas | -0.042 |
each | -0.041 |
ad | -0.041 |
Token |
|
Token ID | 198 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
ow | +0.030 |
ad | +0.025 |
Christmas | +0.024 |
each | +0.024 |
all | +0.023 |
Neg logit contributions | |
ounded | -0.049 |
ursion | -0.046 |
Whoever | -0.046 |
sc | -0.044 |
ooters | -0.044 |
Token | This |
Token ID | 1212 |
Feature activation | -7.33e-02 |
Pos logit contributions | |
ow | +0.040 |
Christmas | +0.032 |
ad | +0.032 |
lf | +0.031 |
happen | +0.030 |
Neg logit contributions | |
ounded | -0.059 |
Whoever | -0.057 |
ursion | -0.057 |
ooters | -0.056 |
sc | -0.055 |
Token | story |
Token ID | 1621 |
Feature activation | -1.28e-01 |
Pos logit contributions | |
ow | +0.092 |
urg | +0.062 |
feelings | +0.058 |
ble | +0.056 |
ts | +0.055 |
Neg logit contributions | |
Whoever | -0.157 |
ursion | -0.153 |
sc | -0.148 |
Ready | -0.146 |
ounded | -0.145 |
Token | has |
Token ID | 468 |
Feature activation | -5.62e-02 |
Pos logit contributions | |
ow | +0.225 |
feelings | +0.168 |
Christmas | +0.161 |
urg | +0.157 |
soup | +0.154 |
Neg logit contributions | |
Whoever | -0.225 |
sc | -0.212 |
ounded | -0.210 |
ursion | -0.208 |
Ready | -0.199 |
Token | a |
Token ID | 257 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
ow | +0.079 |
happen | +0.057 |
Christmas | +0.057 |
feelings | +0.057 |
lf | +0.054 |
Neg logit contributions | |
Whoever | -0.113 |
ursion | -0.109 |
ounded | -0.109 |
ooters | -0.107 |
sc | -0.106 |
Token | moral |
Token ID | 6573 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
ow | +0.021 |
stories | +0.016 |
Christmas | +0.016 |
feelings | +0.015 |
happen | +0.015 |
Neg logit contributions | |
Whoever | -0.039 |
ursion | -0.038 |
ounded | -0.037 |
ooters | -0.037 |
sc | -0.036 |
Token | value |
Token ID | 1988 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
ow | +0.062 |
stories | +0.053 |
feelings | +0.050 |
ad | +0.050 |
Christmas | +0.050 |
Neg logit contributions | |
Whoever | -0.082 |
ounded | -0.080 |
ursion | -0.078 |
shadow | -0.077 |
ooters | -0.075 |
Token | : |
Token ID | 25 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
ow | +0.101 |
Christmas | +0.072 |
feelings | +0.072 |
happen | +0.071 |
stories | +0.067 |
Neg logit contributions | |
Whoever | -0.152 |
ursion | -0.148 |
ounded | -0.147 |
ooters | -0.145 |
sc | -0.144 |
Token |
|
Token ID | 198 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
ow | +0.023 |
ad | +0.019 |
Christmas | +0.018 |
stories | +0.018 |
lf | +0.018 |
Neg logit contributions | |
ounded | -0.040 |
Whoever | -0.039 |
ursion | -0.039 |
hole | -0.037 |
sc | -0.036 |
Token | The |
Token ID | 464 |
Feature activation | -1.65e-02 |
Pos logit contributions | |
ow | +0.030 |
ad | +0.025 |
Christmas | +0.024 |
lf | +0.024 |
stories | +0.023 |
Neg logit contributions | |
ounded | -0.051 |
Whoever | -0.048 |
ursion | -0.048 |
hole | -0.047 |
ooters | -0.046 |
Token | moral |
Token ID | 6573 |
Feature activation | -5.27e-02 |
Pos logit contributions | |
ow | +0.017 |
feelings | +0.011 |
Christmas | +0.011 |
happen | +0.011 |
stories | +0.010 |
Neg logit contributions | |
Whoever | -0.040 |
ursion | -0.039 |
ounded | -0.039 |
sc | -0.038 |
ooters | -0.038 |
Token | of |
Token ID | 286 |
Feature activation | -4.49e-02 |
Pos logit contributions | |
ow | +0.077 |
stories | +0.064 |
Christmas | +0.062 |
ad | +0.061 |
feelings | +0.061 |
Neg logit contributions | |
ounded | -0.100 |
Whoever | -0.100 |
ursion | -0.095 |
shadow | -0.094 |
reaching | -0.092 |
Token | this |
Token ID | 428 |
Feature activation | -5.73e-02 |
Pos logit contributions | |
ow | +0.067 |
feelings | +0.051 |
ities | +0.051 |
ity | +0.048 |
ling | +0.047 |
Neg logit contributions | |
Whoever | -0.080 |
ursion | -0.073 |
shadow | -0.072 |
old | -0.072 |
sc | -0.071 |
Token | story |
Token ID | 1621 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ow | +0.066 |
ts | +0.048 |
feelings | +0.045 |
ities | +0.043 |
Christmas | +0.041 |
Neg logit contributions | |
Whoever | -0.126 |
ursion | -0.119 |
ounded | -0.117 |
ooters | -0.117 |
sc | -0.114 |
Token | is |
Token ID | 318 |
Feature activation | -4.14e-02 |
Pos logit contributions | |
ow | +0.170 |
Christmas | +0.115 |
feelings | +0.112 |
happen | +0.108 |
soup | +0.106 |
Neg logit contributions | |
Whoever | -0.274 |
ounded | -0.263 |
ursion | -0.263 |
sc | -0.258 |
ooters | -0.257 |
Token | that |
Token ID | 326 |
Feature activation | -4.24e-02 |
Pos logit contributions | |
ow | +0.048 |
feelings | +0.032 |
Christmas | +0.032 |
happen | +0.031 |
stories | +0.029 |
Neg logit contributions | |
Whoever | -0.095 |
ursion | -0.093 |
ounded | -0.092 |
ooters | -0.091 |
sc | -0.090 |
Token | it |
Token ID | 340 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ow | +0.067 |
Christmas | +0.051 |
feelings | +0.051 |
happen | +0.050 |
stories | +0.049 |
Neg logit contributions | |
Whoever | -0.079 |
ursion | -0.076 |
ounded | -0.076 |
ooters | -0.074 |
sc | -0.074 |
Token | 's |
Token ID | 338 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
ow | +0.061 |
Christmas | +0.048 |
feelings | +0.047 |
happen | +0.047 |
stories | +0.046 |
Neg logit contributions | |
Whoever | -0.059 |
ursion | -0.057 |
ounded | -0.057 |
sc | -0.055 |
ooters | -0.055 |
Token | sometimes |
Token ID | 3360 |
Feature activation | -8.63e-03 |
Pos logit contributions | |
ow | +0.038 |
happen | +0.028 |
Christmas | +0.027 |
feelings | +0.027 |
lf | +0.027 |
Neg logit contributions | |
Whoever | -0.047 |
ursion | -0.045 |
ounded | -0.045 |
sc | -0.044 |
ooters | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
Whoever | +0.059 |
sc | +0.055 |
ounded | +0.054 |
ooters | +0.054 |
ursion | +0.053 |
Neg logit contributions | |
ow | -0.046 |
happen | -0.031 |
Christmas | -0.031 |
unic | -0.030 |
feelings | -0.030 |
Token | |
Token ID | 220 |
Feature activation | +4.23e-02 |
Pos logit contributions | |
ow | +0.036 |
Christmas | +0.031 |
each | +0.031 |
all | +0.030 |
his | +0.029 |
Neg logit contributions | |
ounded | -0.047 |
sc | -0.046 |
ooters | -0.045 |
ursion | -0.044 |
Whoever | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
sc | +0.089 |
ounded | +0.088 |
ursion | +0.086 |
Whoever | +0.084 |
ooters | +0.083 |
Neg logit contributions | |
ow | -0.051 |
each | -0.039 |
all | -0.037 |
Christmas | -0.036 |
feelings | -0.035 |
Token |
|
Token ID | 198 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
ow | +0.045 |
Christmas | +0.036 |
happen | +0.036 |
feelings | +0.036 |
stories | +0.035 |
Neg logit contributions | |
Whoever | -0.039 |
ursion | -0.038 |
ounded | -0.038 |
ooters | -0.036 |
sc | -0.036 |
Token | This |
Token ID | 1212 |
Feature activation | -7.78e-02 |
Pos logit contributions | |
ow | +0.053 |
feelings | +0.040 |
happen | +0.040 |
Christmas | +0.039 |
stories | +0.038 |
Neg logit contributions | |
Whoever | -0.062 |
ursion | -0.060 |
ounded | -0.058 |
sc | -0.057 |
ooters | -0.057 |
Token | story |
Token ID | 1621 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ow | +0.121 |
urg | +0.094 |
ts | +0.093 |
feelings | +0.090 |
happen | +0.088 |
Neg logit contributions | |
Whoever | -0.142 |
ursion | -0.135 |
sc | -0.132 |
bet | -0.126 |
Ready | -0.126 |
Token | teaches |
Token ID | 17324 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
ow | +0.267 |
feelings | +0.206 |
urg | +0.203 |
ities | +0.199 |
injuries | +0.197 |
Neg logit contributions | |
Whoever | -0.190 |
sc | -0.177 |
ursion | -0.169 |
ounded | -0.162 |
Ready | -0.154 |
Token | us |
Token ID | 514 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
ow | +0.044 |
kin | +0.035 |
happen | +0.033 |
lf | +0.033 |
ling | +0.032 |
Neg logit contributions | |
Whoever | -0.044 |
ursion | -0.040 |
First | -0.039 |
old | -0.039 |
ooters | -0.038 |
Token | two |
Token ID | 734 |
Feature activation | +4.67e-02 |
Pos logit contributions | |
ow | +0.023 |
Christmas | +0.015 |
lf | +0.015 |
happen | +0.014 |
feelings | +0.013 |
Neg logit contributions | |
Whoever | -0.045 |
ounded | -0.044 |
ursion | -0.043 |
ooters | -0.043 |
sc | -0.042 |
Token | valuable |
Token ID | 8119 |
Feature activation | -4.58e-02 |
Pos logit contributions | |
Whoever | +0.090 |
ooters | +0.084 |
ounded | +0.084 |
sc | +0.084 |
ursion | +0.082 |
Neg logit contributions | |
ow | -0.071 |
Christmas | -0.051 |
lf | -0.051 |
happen | -0.050 |
ating | -0.048 |
Token | lessons |
Token ID | 11658 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
ow | +0.072 |
Christmas | +0.049 |
lf | +0.047 |
ling | +0.046 |
happen | +0.045 |
Neg logit contributions | |
Whoever | -0.089 |
ooters | -0.086 |
sc | -0.086 |
ounded | -0.085 |
ursion | -0.083 |
Token | her |
Token ID | 607 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
ow | +0.015 |
feelings | +0.010 |
happen | +0.009 |
Christmas | +0.009 |
lf | +0.009 |
Neg logit contributions | |
Whoever | -0.026 |
sc | -0.026 |
ursion | -0.025 |
ooters | -0.025 |
ounded | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -7.54e-03 |
Pos logit contributions | |
ow | +0.044 |
happen | +0.031 |
Christmas | +0.031 |
feelings | +0.031 |
stories | +0.029 |
Neg logit contributions | |
Whoever | -0.071 |
ursion | -0.069 |
ounded | -0.069 |
ooters | -0.068 |
sc | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
every | +0.007 |
ow | +0.007 |
all | +0.007 |
each | +0.007 |
Christmas | +0.007 |
Neg logit contributions | |
ounded | -0.017 |
shadow | -0.015 |
hole | -0.015 |
sc | -0.015 |
ursion | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
ow | +0.017 |
ad | +0.013 |
each | +0.013 |
Christmas | +0.013 |
all | +0.013 |
Neg logit contributions | |
ounded | -0.036 |
Whoever | -0.034 |
ursion | -0.034 |
sc | -0.033 |
hole | -0.033 |
Token | This |
Token ID | 1212 |
Feature activation | -7.47e-02 |
Pos logit contributions | |
ow | +0.023 |
end | +0.021 |
ad | +0.020 |
lf | +0.020 |
Christmas | +0.020 |
Neg logit contributions | |
ounded | -0.041 |
ooters | -0.038 |
hole | -0.038 |
ursion | -0.037 |
Whoever | -0.037 |
Token | story |
Token ID | 1621 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ow | +0.103 |
urg | +0.074 |
feelings | +0.073 |
ts | +0.069 |
Christmas | +0.068 |
Neg logit contributions | |
Whoever | -0.148 |
ursion | -0.144 |
sc | -0.142 |
Ready | -0.137 |
ounded | -0.136 |
Token | is |
Token ID | 318 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
ow | +0.185 |
feelings | +0.131 |
urg | +0.124 |
Christmas | +0.122 |
ities | +0.121 |
Neg logit contributions | |
Whoever | -0.262 |
sc | -0.252 |
ounded | -0.248 |
ursion | -0.240 |
Ready | -0.231 |
Token | a |
Token ID | 257 |
Feature activation | -6.98e-03 |
Pos logit contributions | |
ow | +0.044 |
feelings | +0.031 |
happen | +0.031 |
Christmas | +0.030 |
stories | +0.029 |
Neg logit contributions | |
Whoever | -0.072 |
ursion | -0.070 |
ounded | -0.070 |
sc | -0.068 |
ooters | -0.068 |
Token | reminder |
Token ID | 15438 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
ow | +0.009 |
Christmas | +0.007 |
feelings | +0.007 |
happen | +0.007 |
stories | +0.007 |
Neg logit contributions | |
Whoever | -0.015 |
ursion | -0.014 |
ounded | -0.014 |
ooters | -0.014 |
sc | -0.014 |
Token | that |
Token ID | 326 |
Feature activation | -4.80e-02 |
Pos logit contributions | |
ow | +0.107 |
feelings | +0.081 |
happen | +0.079 |
Christmas | +0.079 |
lf | +0.075 |
Neg logit contributions | |
Whoever | -0.093 |
ursion | -0.090 |
ounded | -0.089 |
sc | -0.088 |
ooters | -0.087 |
Token | we |
Token ID | 356 |
Feature activation | +2.49e-03 |
Pos logit contributions | |
ow | +0.076 |
feelings | +0.057 |
Christmas | +0.057 |
happen | +0.056 |
stories | +0.054 |
Neg logit contributions | |
Whoever | -0.090 |
ursion | -0.086 |
ounded | -0.086 |
sc | -0.084 |
ooters | -0.083 |
Token | ran |
Token ID | 4966 |
Feature activation | +4.21e-02 |
Pos logit contributions | |
Whoever | +0.078 |
ursion | +0.075 |
ounded | +0.075 |
ooters | +0.073 |
sc | +0.072 |
Neg logit contributions | |
ow | -0.079 |
feelings | -0.062 |
Christmas | -0.062 |
happen | -0.062 |
stories | -0.060 |
Token | home |
Token ID | 1363 |
Feature activation | -7.65e-03 |
Pos logit contributions | |
ursion | +0.091 |
Whoever | +0.086 |
ounded | +0.085 |
ooters | +0.084 |
hole | +0.083 |
Neg logit contributions | |
all | -0.047 |
each | -0.043 |
ow | -0.042 |
his | -0.039 |
every | -0.038 |
Token | and |
Token ID | 290 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
ow | +0.010 |
feelings | +0.007 |
Christmas | +0.007 |
happen | +0.007 |
stories | +0.006 |
Neg logit contributions | |
Whoever | -0.017 |
ursion | -0.016 |
ounded | -0.016 |
ooters | -0.016 |
sc | -0.016 |
Token | shared |
Token ID | 4888 |
Feature activation | -8.28e-02 |
Pos logit contributions | |
ow | +0.035 |
Christmas | +0.027 |
feelings | +0.027 |
happen | +0.027 |
stories | +0.026 |
Neg logit contributions | |
Whoever | -0.040 |
ursion | -0.038 |
ounded | -0.038 |
ooters | -0.037 |
sc | -0.037 |
Token | his |
Token ID | 465 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
ow | +0.111 |
happen | +0.076 |
ling | +0.073 |
feelings | +0.070 |
ating | +0.070 |
Neg logit contributions | |
Whoever | -0.173 |
sc | -0.157 |
ooters | -0.153 |
ounded | -0.153 |
shadow | -0.151 |
Token | story |
Token ID | 1621 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ow | +0.077 |
unic | +0.058 |
lf | +0.057 |
ating | +0.055 |
ad | +0.055 |
Neg logit contributions | |
ounded | -0.105 |
ooters | -0.102 |
sc | -0.102 |
Whoever | -0.102 |
wanna | -0.100 |
Token | with |
Token ID | 351 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
ow | +0.165 |
Christmas | +0.107 |
feelings | +0.107 |
happen | +0.104 |
unic | +0.096 |
Neg logit contributions | |
Whoever | -0.279 |
ounded | -0.269 |
sc | -0.264 |
ooters | -0.264 |
ursion | -0.263 |
Token | the |
Token ID | 262 |
Feature activation | -6.69e-03 |
Pos logit contributions | |
ow | +0.035 |
happen | +0.022 |
Christmas | +0.021 |
feelings | +0.021 |
ity | +0.020 |
Neg logit contributions | |
Whoever | -0.076 |
ounded | -0.072 |
ursion | -0.071 |
ooters | -0.070 |
reaching | -0.070 |
Token | other |
Token ID | 584 |
Feature activation | +6.63e-02 |
Pos logit contributions | |
ow | +0.010 |
Christmas | +0.008 |
happen | +0.008 |
lf | +0.007 |
feelings | +0.007 |
Neg logit contributions | |
Whoever | -0.012 |
ounded | -0.012 |
ursion | -0.012 |
ooters | -0.012 |
reaching | -0.012 |
Token | ants |
Token ID | 27842 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
Whoever | +0.141 |
ursion | +0.137 |
ounded | +0.136 |
ooters | +0.133 |
sc | +0.133 |
Neg logit contributions | |
ow | -0.089 |
feelings | -0.064 |
Christmas | -0.064 |
happen | -0.063 |
stories | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
ow | +0.012 |
feelings | +0.008 |
happen | +0.008 |
Christmas | +0.008 |
stories | +0.007 |
Neg logit contributions | |
Whoever | -0.030 |
ursion | -0.029 |
ounded | -0.029 |
Ready | -0.028 |
sc | -0.028 |
Token |
|
Token ID | 198 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
ow | +0.038 |
lf | +0.031 |
all | +0.030 |
stories | +0.030 |
ad | +0.029 |
Neg logit contributions | |
ounded | -0.052 |
ursion | -0.052 |
Whoever | -0.051 |
ooters | -0.051 |
hole | -0.050 |
Token | The |
Token ID | 464 |
Feature activation | -1.81e-02 |
Pos logit contributions | |
ow | +0.056 |
feelings | +0.044 |
lf | +0.044 |
Christmas | +0.044 |
happen | +0.044 |
Neg logit contributions | |
Whoever | -0.060 |
ounded | -0.059 |
ursion | -0.059 |
ooters | -0.058 |
sc | -0.056 |
Token | moral |
Token ID | 6573 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
ow | +0.020 |
feelings | +0.013 |
Christmas | +0.013 |
happen | +0.013 |
stories | +0.012 |
Neg logit contributions | |
Whoever | -0.042 |
ursion | -0.041 |
ounded | -0.041 |
ooters | -0.040 |
sc | -0.040 |
Token | of |
Token ID | 286 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
ow | +0.066 |
stories | +0.054 |
ad | +0.053 |
Christmas | +0.053 |
feelings | +0.052 |
Neg logit contributions | |
ounded | -0.092 |
Whoever | -0.090 |
shadow | -0.087 |
ursion | -0.086 |
ooters | -0.084 |
Token | this |
Token ID | 428 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
ow | +0.071 |
feelings | +0.058 |
ities | +0.056 |
ity | +0.055 |
creations | +0.053 |
Neg logit contributions | |
Whoever | -0.064 |
shadow | -0.058 |
old | -0.057 |
ursion | -0.057 |
Would | -0.056 |
Token | story |
Token ID | 1621 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ow | +0.070 |
ts | +0.059 |
feelings | +0.053 |
ities | +0.052 |
ating | +0.048 |
Neg logit contributions | |
Whoever | -0.112 |
ursion | -0.104 |
ounded | -0.104 |
ooters | -0.102 |
reaching | -0.100 |
Token | is |
Token ID | 318 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
ow | +0.179 |
Christmas | +0.124 |
feelings | +0.119 |
soup | +0.116 |
urg | +0.113 |
Neg logit contributions | |
Whoever | -0.264 |
ounded | -0.254 |
ursion | -0.250 |
sc | -0.249 |
ooters | -0.244 |
Token | that |
Token ID | 326 |
Feature activation | -4.11e-02 |
Pos logit contributions | |
ow | +0.046 |
feelings | +0.030 |
Christmas | +0.029 |
happen | +0.029 |
lf | +0.027 |
Neg logit contributions | |
Whoever | -0.096 |
ursion | -0.093 |
ounded | -0.092 |
ooters | -0.091 |
reaching | -0.089 |
Token | if |
Token ID | 611 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
ow | +0.065 |
Christmas | +0.049 |
feelings | +0.049 |
happen | +0.049 |
stories | +0.047 |
Neg logit contributions | |
Whoever | -0.077 |
ursion | -0.074 |
ounded | -0.073 |
ooters | -0.071 |
sc | -0.071 |
Token | you |
Token ID | 345 |
Feature activation | +6.65e-03 |
Pos logit contributions | |
Whoever | +0.068 |
ursion | +0.064 |
ounded | +0.063 |
ooters | +0.061 |
sc | +0.061 |
Neg logit contributions | |
ow | -0.090 |
Christmas | -0.071 |
feelings | -0.071 |
happen | -0.070 |
stories | -0.069 |
Token | feel |
Token ID | 1254 |
Feature activation | -6.28e-02 |
Pos logit contributions | |
Whoever | +0.011 |
ursion | +0.011 |
ounded | +0.010 |
ooters | +0.010 |
sc | +0.010 |
Neg logit contributions | |
ow | -0.012 |
lf | -0.010 |
Christmas | -0.009 |
feelings | -0.009 |
stories | -0.009 |
Token | safe |
Token ID | 3338 |
Feature activation | -2.09e-03 |
Pos logit contributions | |
ow | +0.085 |
feelings | +0.067 |
Christmas | +0.067 |
happen | +0.066 |
stories | +0.064 |
Neg logit contributions | |
Whoever | -0.078 |
ursion | -0.076 |
ounded | -0.076 |
sc | -0.073 |
ooters | -0.072 |
Token | . |
Token ID | 13 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
ow | +0.003 |
lf | +0.002 |
feelings | +0.002 |
Christmas | +0.002 |
stories | +0.002 |
Neg logit contributions | |
Whoever | -0.004 |
sc | -0.004 |
ounded | -0.004 |
ursion | -0.004 |
hole | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
ow | +0.023 |
all | +0.020 |
his | +0.020 |
each | +0.019 |
every | +0.019 |
Neg logit contributions | |
ounded | -0.040 |
ooters | -0.039 |
sc | -0.039 |
ursion | -0.038 |
hole | -0.038 |
Token |
|
Token ID | 198 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
ow | +0.026 |
Christmas | +0.019 |
happen | +0.019 |
lf | +0.019 |
stories | +0.018 |
Neg logit contributions | |
ounded | -0.042 |
Whoever | -0.042 |
ursion | -0.042 |
ooters | -0.040 |
sc | -0.040 |
Token | This |
Token ID | 1212 |
Feature activation | -7.73e-02 |
Pos logit contributions | |
ow | +0.041 |
Christmas | +0.031 |
feelings | +0.031 |
happen | +0.031 |
lf | +0.030 |
Neg logit contributions | |
Whoever | -0.056 |
ounded | -0.055 |
ursion | -0.055 |
ooters | -0.054 |
sc | -0.053 |
Token | story |
Token ID | 1621 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
ow | +0.113 |
urg | +0.101 |
ts | +0.097 |
ating | +0.094 |
ities | +0.092 |
Neg logit contributions | |
Whoever | -0.129 |
ursion | -0.126 |
sc | -0.124 |
bet | -0.121 |
Ready | -0.118 |
Token | teaches |
Token ID | 17324 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
ow | +0.213 |
feelings | +0.158 |
ities | +0.154 |
urg | +0.152 |
ity | +0.145 |
Neg logit contributions | |
Whoever | -0.233 |
sc | -0.223 |
ounded | -0.213 |
ursion | -0.211 |
Ready | -0.199 |
Token | us |
Token ID | 514 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
ow | +0.040 |
kin | +0.033 |
lf | +0.031 |
happen | +0.031 |
Christmas | +0.030 |
Neg logit contributions | |
Whoever | -0.042 |
ooters | -0.038 |
ursion | -0.038 |
sc | -0.037 |
old | -0.036 |
Token | that |
Token ID | 326 |
Feature activation | -3.70e-02 |
Pos logit contributions | |
ow | +0.026 |
lf | +0.017 |
Christmas | +0.017 |
happen | +0.016 |
feelings | +0.016 |
Neg logit contributions | |
Whoever | -0.043 |
ounded | -0.042 |
ursion | -0.042 |
ooters | -0.041 |
sc | -0.041 |
Token | courage |
Token ID | 11917 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
ow | +0.055 |
feelings | +0.040 |
Christmas | +0.040 |
happen | +0.040 |
stories | +0.038 |
Neg logit contributions | |
Whoever | -0.072 |
ursion | -0.070 |
ounded | -0.069 |
ooters | -0.068 |
sc | -0.068 |
Token | can |
Token ID | 460 |
Feature activation | +2.47e-03 |
Pos logit contributions | |
Whoever | +0.093 |
ursion | +0.090 |
ounded | +0.090 |
ooters | +0.087 |
sc | +0.087 |
Neg logit contributions | |
ow | -0.072 |
feelings | -0.054 |
happen | -0.054 |
Christmas | -0.054 |
stories | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | -2.83e-02 |
Pos logit contributions | |
ow | +0.024 |
Christmas | +0.017 |
happen | +0.017 |
stories | +0.016 |
feelings | +0.016 |
Neg logit contributions | |
Whoever | -0.043 |
ounded | -0.043 |
ursion | -0.043 |
ooters | -0.041 |
sc | -0.041 |
Token | The |
Token ID | 464 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
ow | +0.039 |
Christmas | +0.029 |
happen | +0.029 |
feelings | +0.029 |
stories | +0.028 |
Neg logit contributions | |
Whoever | -0.057 |
ounded | -0.056 |
ursion | -0.056 |
ooters | -0.054 |
sc | -0.053 |
Token | fisherman |
Token ID | 48750 |
Feature activation | +1.70e-03 |
Pos logit contributions | |
ow | +0.018 |
happen | +0.013 |
feelings | +0.013 |
Christmas | +0.013 |
lf | +0.013 |
Neg logit contributions | |
Whoever | -0.022 |
ursion | -0.021 |
ounded | -0.021 |
ooters | -0.021 |
sc | -0.021 |
Token | shared |
Token ID | 4888 |
Feature activation | -9.30e-02 |
Pos logit contributions | |
Whoever | +0.003 |
ursion | +0.003 |
ounded | +0.003 |
ooters | +0.003 |
sc | +0.003 |
Neg logit contributions | |
ow | -0.002 |
feelings | -0.002 |
Christmas | -0.002 |
happen | -0.002 |
stories | -0.002 |
Token | his |
Token ID | 465 |
Feature activation | -5.48e-02 |
Pos logit contributions | |
ow | +0.128 |
happen | +0.089 |
ating | +0.083 |
ling | +0.082 |
feelings | +0.081 |
Neg logit contributions | |
Whoever | -0.195 |
sc | -0.183 |
ooters | -0.172 |
ounded | -0.172 |
Ready | -0.169 |
Token | story |
Token ID | 1621 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
ow | +0.077 |
happen | +0.052 |
lf | +0.052 |
unic | +0.051 |
feelings | +0.051 |
Neg logit contributions | |
Whoever | -0.107 |
ounded | -0.105 |
ooters | -0.104 |
sc | -0.103 |
ursion | -0.103 |
Token | with |
Token ID | 351 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
ow | +0.177 |
feelings | +0.115 |
happen | +0.112 |
Christmas | +0.112 |
unic | +0.105 |
Neg logit contributions | |
Whoever | -0.271 |
ounded | -0.259 |
sc | -0.256 |
ooters | -0.254 |
ursion | -0.252 |
Token | a |
Token ID | 257 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
ow | +0.036 |
happen | +0.027 |
lf | +0.025 |
ating | +0.024 |
ity | +0.024 |
Neg logit contributions | |
Whoever | -0.064 |
sc | -0.058 |
ursion | -0.057 |
ooters | -0.057 |
Ready | -0.057 |
Token | special |
Token ID | 2041 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
ow | +0.018 |
Christmas | +0.012 |
feelings | +0.012 |
happen | +0.012 |
stories | +0.012 |
Neg logit contributions | |
Whoever | -0.033 |
ursion | -0.032 |
ounded | -0.031 |
ooters | -0.031 |
sc | -0.031 |
Token | friend |
Token ID | 1545 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
ow | +0.047 |
happen | +0.033 |
feelings | +0.033 |
Christmas | +0.032 |
lf | +0.030 |
Neg logit contributions | |
Whoever | -0.071 |
ursion | -0.069 |
ounded | -0.069 |
sc | -0.068 |
ooters | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
ounded | +0.050 |
Whoever | +0.049 |
ursion | +0.049 |
Ready | +0.047 |
reaching | +0.047 |
Neg logit contributions | |
ow | -0.015 |
each | -0.010 |
Christmas | -0.009 |
feelings | -0.009 |
stories | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-02 |
Pos logit contributions | |
Whoever | +0.052 |
sc | +0.048 |
hole | +0.046 |
ounded | +0.045 |
ooters | +0.045 |
Neg logit contributions | |
ow | -0.043 |
urg | -0.030 |
feelings | -0.029 |
happen | -0.028 |
Christmas | -0.028 |
Token | |
Token ID | 220 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
ow | +0.021 |
Christmas | +0.019 |
each | +0.019 |
all | +0.018 |
every | +0.017 |
Neg logit contributions | |
ounded | -0.027 |
ooters | -0.025 |
shadow | -0.025 |
ursion | -0.025 |
reaching | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -1.83e-02 |
Pos logit contributions | |
ounded | +0.103 |
ursion | +0.100 |
sc | +0.099 |
Whoever | +0.098 |
reaching | +0.094 |
Neg logit contributions | |
ow | -0.059 |
each | -0.049 |
Christmas | -0.046 |
all | -0.046 |
ad | -0.045 |
Token |
|
Token ID | 198 |
Feature activation | -2.58e-02 |
Pos logit contributions | |
ow | +0.029 |
Christmas | +0.023 |
happen | +0.022 |
lf | +0.022 |
each | +0.022 |
Neg logit contributions | |
ounded | -0.033 |
Whoever | -0.032 |
ursion | -0.032 |
ooters | -0.031 |
sc | -0.031 |
Token | This |
Token ID | 1212 |
Feature activation | -7.53e-02 |
Pos logit contributions | |
ow | +0.039 |
Christmas | +0.031 |
happen | +0.030 |
feelings | +0.030 |
stories | +0.029 |
Neg logit contributions | |
Whoever | -0.048 |
ounded | -0.047 |
ursion | -0.047 |
ooters | -0.046 |
sc | -0.045 |
Token | story |
Token ID | 1621 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
ow | +0.113 |
urg | +0.088 |
ts | +0.082 |
ble | +0.081 |
feelings | +0.078 |
Neg logit contributions | |
Whoever | -0.143 |
ursion | -0.140 |
sc | -0.135 |
Ready | -0.131 |
ounded | -0.128 |
Token | teaches |
Token ID | 17324 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
ow | +0.219 |
feelings | +0.158 |
urg | +0.158 |
injuries | +0.156 |
ities | +0.153 |
Neg logit contributions | |
Whoever | -0.229 |
sc | -0.221 |
ounded | -0.213 |
ursion | -0.210 |
Ready | -0.197 |
Token | us |
Token ID | 514 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
ow | +0.042 |
lf | +0.031 |
happen | +0.030 |
kin | +0.030 |
glow | +0.028 |
Neg logit contributions | |
Whoever | -0.051 |
ursion | -0.046 |
sc | -0.046 |
ooters | -0.046 |
ounded | -0.045 |
Token | that |
Token ID | 326 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
ow | +0.024 |
lf | +0.015 |
Christmas | +0.015 |
happen | +0.015 |
feelings | +0.014 |
Neg logit contributions | |
Whoever | -0.043 |
ounded | -0.042 |
ursion | -0.042 |
ooters | -0.041 |
sc | -0.041 |
Token | helping |
Token ID | 5742 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
ow | +0.050 |
Christmas | +0.036 |
feelings | +0.036 |
happen | +0.036 |
stories | +0.034 |
Neg logit contributions | |
Whoever | -0.068 |
ursion | -0.065 |
ounded | -0.064 |
sc | -0.063 |
ooters | -0.063 |
Token | others |
Token ID | 1854 |
Feature activation | +1.19e-02 |
Pos logit contributions | |
ow | +0.052 |
Christmas | +0.036 |
happen | +0.036 |
feelings | +0.036 |
stories | +0.034 |
Neg logit contributions | |
Whoever | -0.085 |
ursion | -0.082 |
ounded | -0.082 |
ooters | -0.080 |
sc | -0.080 |
Token | there |
Token ID | 612 |
Feature activation | +4.62e-02 |
Pos logit contributions | |
Whoever | +0.046 |
ursion | +0.046 |
ounded | +0.045 |
sc | +0.045 |
ooters | +0.044 |
Neg logit contributions | |
ow | -0.020 |
Christmas | -0.013 |
happen | -0.012 |
feelings | -0.012 |
lf | -0.011 |
Token | was |
Token ID | 373 |
Feature activation | -3.33e-02 |
Pos logit contributions | |
Whoever | +0.089 |
ounded | +0.082 |
ursion | +0.081 |
Ready | +0.081 |
sc | +0.081 |
Neg logit contributions | |
ow | -0.061 |
feelings | -0.053 |
stories | -0.052 |
ities | -0.051 |
songs | -0.050 |
Token | a |
Token ID | 257 |
Feature activation | -6.27e-03 |
Pos logit contributions | |
ow | +0.034 |
happen | +0.023 |
feelings | +0.023 |
Christmas | +0.021 |
stories | +0.021 |
Neg logit contributions | |
Whoever | -0.080 |
ursion | -0.078 |
ounded | -0.076 |
sc | -0.075 |
ooters | -0.075 |
Token | little |
Token ID | 1310 |
Feature activation | +4.37e-02 |
Pos logit contributions | |
ow | +0.009 |
feelings | +0.007 |
happen | +0.007 |
Christmas | +0.007 |
stories | +0.006 |
Neg logit contributions | |
Whoever | -0.012 |
ursion | -0.012 |
ounded | -0.012 |
ooters | -0.012 |
sc | -0.012 |
Token | girl |
Token ID | 2576 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
Whoever | +0.075 |
ounded | +0.072 |
ursion | +0.072 |
ooters | +0.070 |
sc | +0.070 |
Neg logit contributions | |
ow | -0.078 |
Christmas | -0.061 |
feelings | -0.059 |
soup | -0.058 |
happen | -0.057 |
Token | called |
Token ID | 1444 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Whoever | +0.046 |
ursion | +0.044 |
ounded | +0.044 |
ooters | +0.042 |
sc | +0.042 |
Neg logit contributions | |
ow | -0.063 |
feelings | -0.051 |
Christmas | -0.051 |
happen | -0.051 |
stories | -0.049 |
Token | Emily |
Token ID | 17608 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
ursion | +0.214 |
ounded | +0.211 |
Whoever | +0.207 |
sc | +0.198 |
ooters | +0.197 |
Neg logit contributions | |
ow | -0.182 |
Christmas | -0.149 |
each | -0.132 |
happen | -0.132 |
stories | -0.131 |
Token | . |
Token ID | 13 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
ow | +0.026 |
urg | +0.021 |
happen | +0.021 |
ating | +0.021 |
boredom | +0.019 |
Neg logit contributions | |
Whoever | -0.025 |
hole | -0.024 |
sc | -0.024 |
pad | -0.023 |
reaching | -0.023 |
Token | She |
Token ID | 1375 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
ow | +0.071 |
happen | +0.061 |
Christmas | +0.060 |
feelings | +0.059 |
ating | +0.058 |
Neg logit contributions | |
Whoever | -0.036 |
ursion | -0.034 |
ounded | -0.033 |
sc | -0.032 |
reaching | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | -8.44e-03 |
Pos logit contributions | |
ow | +0.032 |
urg | +0.032 |
injuries | +0.031 |
ities | +0.031 |
forgiveness | +0.030 |
Neg logit contributions | |
sc | -0.025 |
Whoever | -0.025 |
Ready | -0.025 |
reaching | -0.024 |
ursion | -0.024 |
Token | very |
Token ID | 845 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
ow | +0.012 |
feelings | +0.009 |
Christmas | +0.009 |
happen | +0.009 |
stories | +0.008 |
Neg logit contributions | |
Whoever | -0.017 |
ursion | -0.017 |
ounded | -0.017 |
ooters | -0.016 |
sc | -0.016 |
Token | day |
Token ID | 1110 |
Feature activation | -3.80e-04 |
Pos logit contributions | |
ursion | +0.054 |
Whoever | +0.053 |
ooters | +0.050 |
ounded | +0.050 |
reaching | +0.049 |
Neg logit contributions | |
ow | -0.065 |
feelings | -0.051 |
happen | -0.051 |
urg | -0.050 |
ities | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
ow | +0.000 |
Christmas | +0.000 |
happen | +0.000 |
feelings | +0.000 |
stories | +0.000 |
Neg logit contributions | |
Whoever | -0.001 |
ursion | -0.001 |
ounded | -0.001 |
ooters | -0.001 |
sc | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
ow | +0.036 |
feelings | +0.023 |
Christmas | +0.023 |
happen | +0.023 |
stories | +0.022 |
Neg logit contributions | |
Whoever | -0.079 |
ursion | -0.076 |
ounded | -0.076 |
ooters | -0.075 |
sc | -0.074 |
Token | little |
Token ID | 1310 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
Whoever | +0.022 |
ursion | +0.021 |
ounded | +0.021 |
ooters | +0.020 |
sc | +0.020 |
Neg logit contributions | |
ow | -0.017 |
happen | -0.013 |
feelings | -0.013 |
Christmas | -0.012 |
stories | -0.012 |
Token | mouse |
Token ID | 10211 |
Feature activation | +5.93e-03 |
Pos logit contributions | |
Whoever | +0.100 |
ounded | +0.096 |
ursion | +0.095 |
ooters | +0.093 |
sc | +0.092 |
Neg logit contributions | |
ow | -0.098 |
Christmas | -0.078 |
feelings | -0.074 |
happen | -0.072 |
soup | -0.072 |
Token | named |
Token ID | 3706 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
ursion | +0.006 |
Whoever | +0.006 |
sc | +0.006 |
reaching | +0.006 |
ounded | +0.005 |
Neg logit contributions | |
ow | -0.013 |
feelings | -0.012 |
happen | -0.011 |
stories | -0.011 |
urg | -0.011 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
Whoever | +0.185 |
ursion | +0.178 |
ounded | +0.177 |
ooters | +0.172 |
sc | +0.171 |
Neg logit contributions | |
ow | -0.194 |
feelings | -0.153 |
happen | -0.152 |
Christmas | -0.152 |
stories | -0.147 |
Token | was |
Token ID | 373 |
Feature activation | +1.60e-03 |
Pos logit contributions | |
ursion | +0.032 |
Whoever | +0.031 |
reaching | +0.030 |
sc | +0.030 |
ounded | +0.030 |
Neg logit contributions | |
ow | -0.037 |
feelings | -0.031 |
stories | -0.031 |
jokes | -0.031 |
songs | -0.031 |
Token | hungry |
Token ID | 14720 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
Whoever | +0.003 |
ounded | +0.003 |
ursion | +0.003 |
ooters | +0.003 |
sc | +0.003 |
Neg logit contributions | |
ow | -0.002 |
Christmas | -0.002 |
all | -0.002 |
feelings | -0.002 |
happen | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
ow | +0.032 |
feelings | +0.024 |
ating | +0.024 |
happen | +0.023 |
unic | +0.023 |
Neg logit contributions | |
Whoever | -0.040 |
sc | -0.040 |
ursion | -0.039 |
ooters | -0.037 |
hole | -0.037 |
Token | He |
Token ID | 679 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
ow | +0.038 |
happen | +0.033 |
feelings | +0.033 |
ating | +0.033 |
ity | +0.033 |
Neg logit contributions | |
Whoever | -0.021 |
ooters | -0.018 |
ursion | -0.017 |
ounded | -0.017 |
reaching | -0.017 |
Token | there |
Token ID | 612 |
Feature activation | +5.38e-02 |
Pos logit contributions | |
ow | +0.059 |
Christmas | +0.046 |
happen | +0.044 |
feelings | +0.043 |
all | +0.043 |
Neg logit contributions | |
Whoever | -0.062 |
ursion | -0.062 |
ooters | -0.062 |
ounded | -0.061 |
sc | -0.059 |
Token | was |
Token ID | 373 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
Whoever | +0.089 |
sc | +0.079 |
Ready | +0.078 |
bet | +0.077 |
Would | +0.076 |
Neg logit contributions | |
ow | -0.083 |
ities | -0.075 |
feelings | -0.075 |
songs | -0.073 |
urg | -0.072 |
Token | a |
Token ID | 257 |
Feature activation | +1.20e-03 |
Pos logit contributions | |
ow | +0.029 |
happen | +0.021 |
feelings | +0.021 |
ating | +0.019 |
stories | +0.018 |
Neg logit contributions | |
Whoever | -0.060 |
ursion | -0.058 |
sc | -0.056 |
ounded | -0.056 |
Ready | -0.056 |
Token | little |
Token ID | 1310 |
Feature activation | +4.82e-02 |
Pos logit contributions | |
Whoever | +0.002 |
ursion | +0.002 |
ounded | +0.002 |
ooters | +0.002 |
sc | +0.002 |
Neg logit contributions | |
ow | -0.002 |
feelings | -0.001 |
happen | -0.001 |
Christmas | -0.001 |
stories | -0.001 |
Token | boy |
Token ID | 2933 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
Whoever | +0.081 |
ounded | +0.078 |
ursion | +0.077 |
ooters | +0.076 |
sc | +0.075 |
Neg logit contributions | |
ow | -0.087 |
Christmas | -0.069 |
feelings | -0.067 |
happen | -0.065 |
soup | -0.065 |
Token | called |
Token ID | 1444 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Whoever | +0.021 |
ursion | +0.019 |
ooters | +0.019 |
sc | +0.019 |
ounded | +0.018 |
Neg logit contributions | |
ow | -0.036 |
happen | -0.031 |
feelings | -0.030 |
stories | -0.030 |
Christmas | -0.029 |
Token | Peter |
Token ID | 5613 |
Feature activation | +4.24e-03 |
Pos logit contributions | |
ursion | +0.243 |
ounded | +0.241 |
Whoever | +0.230 |
sc | +0.227 |
reaching | +0.222 |
Neg logit contributions | |
ow | -0.171 |
Christmas | -0.136 |
happen | -0.124 |
soup | -0.121 |
each | -0.121 |
Token | . |
Token ID | 13 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
Whoever | +0.007 |
ooters | +0.007 |
ursion | +0.007 |
ounded | +0.007 |
sc | +0.007 |
Neg logit contributions | |
ow | -0.007 |
happen | -0.005 |
ating | -0.005 |
feelings | -0.005 |
lf | -0.005 |
Token | Peter |
Token ID | 5613 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
ow | +0.058 |
feelings | +0.048 |
happen | +0.047 |
kin | +0.045 |
Christmas | +0.045 |
Neg logit contributions | |
Whoever | -0.037 |
ursion | -0.034 |
ounded | -0.033 |
ooters | -0.031 |
Ready | -0.031 |
Token | was |
Token ID | 373 |
Feature activation | -8.48e-03 |
Pos logit contributions | |
ow | +0.021 |
urg | +0.016 |
happen | +0.016 |
feelings | +0.015 |
injuries | +0.015 |
Neg logit contributions | |
Whoever | -0.024 |
ursion | -0.023 |
ounded | -0.022 |
reaching | -0.022 |
sc | -0.022 |
Token | three |
Token ID | 1115 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ow | +0.012 |
happen | +0.009 |
feelings | +0.009 |
Christmas | +0.009 |
stories | +0.009 |
Neg logit contributions | |
Whoever | -0.017 |
ursion | -0.016 |
ounded | -0.016 |
ooters | -0.016 |
sc | -0.016 |
Token | party |
Token ID | 2151 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
Ready | +0.042 |
ursion | +0.042 |
Whoever | +0.041 |
sc | +0.041 |
ounded | +0.041 |
Neg logit contributions | |
ow | -0.028 |
ating | -0.022 |
urg | -0.021 |
ity | -0.020 |
lf | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +6.29e-03 |
Pos logit contributions | |
ow | +0.028 |
urg | +0.022 |
iflower | +0.020 |
unic | +0.020 |
Christmas | +0.020 |
Neg logit contributions | |
sc | -0.034 |
Whoever | -0.034 |
ounded | -0.033 |
ursion | -0.032 |
ooters | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -5.44e-03 |
Pos logit contributions | |
ounded | +0.012 |
ooters | +0.011 |
ursion | +0.011 |
sc | +0.011 |
shadow | +0.011 |
Neg logit contributions | |
ow | -0.010 |
feelings | -0.008 |
each | -0.008 |
Christmas | -0.007 |
stories | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | -1.46e-02 |
Pos logit contributions | |
ow | +0.008 |
Christmas | +0.006 |
happen | +0.006 |
each | +0.006 |
lf | +0.006 |
Neg logit contributions | |
ounded | -0.010 |
ursion | -0.010 |
Whoever | -0.010 |
sc | -0.010 |
ooters | -0.010 |
Token | She |
Token ID | 3347 |
Feature activation | -9.18e-03 |
Pos logit contributions | |
ow | +0.020 |
lf | +0.016 |
end | +0.016 |
Christmas | +0.015 |
feelings | +0.015 |
Neg logit contributions | |
ounded | -0.029 |
Whoever | -0.028 |
ooters | -0.028 |
ursion | -0.028 |
hole | -0.027 |
Token | called |
Token ID | 1444 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ow | +0.015 |
happen | +0.012 |
feelings | +0.012 |
Christmas | +0.012 |
ating | +0.012 |
Neg logit contributions | |
Whoever | -0.016 |
ounded | -0.015 |
ursion | -0.015 |
sc | -0.015 |
reaching | -0.014 |
Token | Ben |
Token ID | 3932 |
Feature activation | -9.47e-03 |
Pos logit contributions | |
Whoever | +0.301 |
sc | +0.282 |
ursion | +0.277 |
ounded | +0.276 |
ooters | +0.274 |
Neg logit contributions | |
ow | -0.202 |
feelings | -0.142 |
ating | -0.141 |
ity | -0.138 |
happen | -0.137 |
Token | on |
Token ID | 319 |
Feature activation | -2.75e-03 |
Pos logit contributions | |
ow | +0.015 |
happen | +0.011 |
feelings | +0.011 |
Christmas | +0.011 |
lf | +0.011 |
Neg logit contributions | |
Whoever | -0.018 |
ounded | -0.017 |
ursion | -0.017 |
sc | -0.017 |
ooters | -0.016 |
Token | the |
Token ID | 262 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
ow | +0.004 |
happen | +0.003 |
ating | +0.003 |
ity | +0.003 |
ling | +0.003 |
Neg logit contributions | |
Whoever | -0.005 |
sc | -0.005 |
ooters | -0.005 |
Ready | -0.004 |
Ready | -0.004 |
Token | phone |
Token ID | 3072 |
Feature activation | -3.56e-03 |
Pos logit contributions | |
Whoever | +0.018 |
sc | +0.018 |
ounded | +0.018 |
ooters | +0.018 |
ursion | +0.018 |
Neg logit contributions | |
ow | -0.014 |
ating | -0.011 |
Christmas | -0.011 |
ity | -0.011 |
ad | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +8.44e-03 |
Pos logit contributions | |
ow | +0.004 |
happen | +0.003 |
feelings | +0.003 |
Christmas | +0.003 |
stories | +0.003 |
Neg logit contributions | |
Whoever | -0.008 |
ursion | -0.008 |
ounded | -0.008 |
ooters | -0.008 |
sc | -0.008 |
Token | there |
Token ID | 612 |
Feature activation | +4.77e-02 |
Pos logit contributions | |
ow | +0.066 |
Christmas | +0.051 |
feelings | +0.050 |
happen | +0.049 |
each | +0.048 |
Neg logit contributions | |
Whoever | -0.072 |
ursion | -0.071 |
ooters | -0.070 |
ounded | -0.069 |
sc | -0.068 |
Token | was |
Token ID | 373 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
Whoever | +0.088 |
Ready | +0.080 |
ursion | +0.080 |
ounded | +0.080 |
sc | +0.080 |
Neg logit contributions | |
ow | -0.067 |
stories | -0.057 |
feelings | -0.056 |
ities | -0.055 |
jokes | -0.054 |
Token | a |
Token ID | 257 |
Feature activation | -6.22e-03 |
Pos logit contributions | |
ow | +0.031 |
happen | +0.021 |
feelings | +0.021 |
Christmas | +0.020 |
stories | +0.019 |
Neg logit contributions | |
Whoever | -0.074 |
ursion | -0.072 |
ounded | -0.071 |
sc | -0.070 |
ooters | -0.069 |
Token | good |
Token ID | 922 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
ow | +0.009 |
happen | +0.007 |
feelings | +0.007 |
Christmas | +0.007 |
stories | +0.007 |
Neg logit contributions | |
Whoever | -0.012 |
ursion | -0.012 |
ounded | -0.012 |
ooters | -0.011 |
sc | -0.011 |
Token | cat |
Token ID | 3797 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
ow | +0.066 |
happen | +0.050 |
feelings | +0.050 |
Christmas | +0.049 |
stories | +0.048 |
Neg logit contributions | |
Whoever | -0.083 |
ursion | -0.080 |
ounded | -0.079 |
ooters | -0.078 |
sc | -0.077 |
Token | named |
Token ID | 3706 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
ow | +0.056 |
happen | +0.049 |
feelings | +0.048 |
urg | +0.047 |
stories | +0.046 |
Neg logit contributions | |
Whoever | -0.026 |
sc | -0.024 |
ursion | -0.024 |
ooters | -0.024 |
reaching | -0.023 |
Token | Fl |
Token ID | 1610 |
Feature activation | +2.85e-03 |
Pos logit contributions | |
ursion | +0.203 |
ounded | +0.203 |
Whoever | +0.198 |
sc | +0.191 |
ooters | +0.190 |
Neg logit contributions | |
ow | -0.153 |
Christmas | -0.113 |
each | -0.106 |
feelings | -0.105 |
stories | -0.103 |
Token | uffy |
Token ID | 15352 |
Feature activation | -5.22e-03 |
Pos logit contributions | |
Whoever | +0.003 |
ursion | +0.003 |
urry | +0.003 |
ooters | +0.003 |
reaching | +0.003 |
Neg logit contributions | |
ow | -0.006 |
happen | -0.006 |
stories | -0.006 |
jokes | -0.005 |
lf | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
ow | +0.009 |
happen | +0.008 |
urg | +0.008 |
accidents | +0.008 |
ling | +0.008 |
Neg logit contributions | |
Whoever | -0.006 |
sc | -0.006 |
reaching | -0.006 |
ooters | -0.006 |
ruff | -0.006 |
Token | Fl |
Token ID | 1610 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
ow | +0.041 |
happen | +0.038 |
ating | +0.037 |
feelings | +0.037 |
lf | +0.036 |
Neg logit contributions | |
Whoever | -0.045 |
sc | -0.042 |
ooters | -0.040 |
ursion | -0.040 |
reaching | -0.039 |
Token | uffy |
Token ID | 15352 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
ow | +0.026 |
happen | +0.026 |
stories | +0.025 |
lf | +0.023 |
ad | +0.023 |
Neg logit contributions | |
Whoever | -0.012 |
urry | -0.012 |
ursion | -0.012 |
ends | -0.011 |
reaching | -0.011 |
Token | ! |
Token ID | 0 |
Feature activation | -2.06e-02 |
Pos logit contributions | |
Whoever | +0.057 |
sc | +0.051 |
ounded | +0.051 |
ursion | +0.051 |
ooters | +0.050 |
Neg logit contributions | |
ow | -0.043 |
feelings | -0.032 |
happen | -0.030 |
ity | -0.030 |
urg | -0.029 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
ow | +0.036 |
all | +0.029 |
Christmas | +0.029 |
each | +0.028 |
stories | +0.028 |
Neg logit contributions | |
ursion | -0.035 |
ounded | -0.034 |
Whoever | -0.034 |
ooters | -0.034 |
shadow | -0.032 |
Token | Once |
Token ID | 7454 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
ow | +0.089 |
feelings | +0.072 |
happen | +0.072 |
Christmas | +0.072 |
stories | +0.070 |
Neg logit contributions | |
Whoever | -0.070 |
ursion | -0.067 |
ounded | -0.066 |
sc | -0.064 |
ooters | -0.064 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
ow | +0.045 |
Christmas | +0.037 |
all | +0.037 |
happen | +0.037 |
end | +0.036 |
Neg logit contributions | |
Whoever | -0.032 |
sc | -0.031 |
Ready | -0.031 |
ursion | -0.030 |
ounded | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | +4.64e-03 |
Pos logit contributions | |
ow | +0.019 |
ating | +0.016 |
ling | +0.015 |
urg | +0.015 |
feelings | +0.014 |
Neg logit contributions | |
Whoever | -0.027 |
Ready | -0.024 |
sc | -0.024 |
reaching | -0.024 |
ursion | -0.024 |
Token | time |
Token ID | 640 |
Feature activation | +2.82e-02 |
Pos logit contributions | |
Whoever | +0.008 |
ursion | +0.008 |
ounded | +0.008 |
sc | +0.008 |
reaching | +0.008 |
Neg logit contributions | |
ow | -0.007 |
happen | -0.006 |
feelings | -0.006 |
Christmas | -0.006 |
stories | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
Whoever | +0.060 |
ounded | +0.058 |
ursion | +0.058 |
ooters | +0.057 |
sc | +0.056 |
Neg logit contributions | |
ow | -0.036 |
feelings | -0.026 |
happen | -0.026 |
Christmas | -0.025 |
stories | -0.025 |
Token | there |
Token ID | 612 |
Feature activation | +5.75e-02 |
Pos logit contributions | |
ow | +0.066 |
feelings | +0.053 |
Christmas | +0.053 |
happen | +0.053 |
stories | +0.051 |
Neg logit contributions | |
Whoever | -0.054 |
ursion | -0.052 |
ounded | -0.052 |
ooters | -0.050 |
sc | -0.050 |
Token | was |
Token ID | 373 |
Feature activation | -2.28e-02 |
Pos logit contributions | |
Whoever | +0.091 |
sc | +0.080 |
Ready | +0.078 |
Would | +0.078 |
bet | +0.077 |
Neg logit contributions | |
ow | -0.091 |
ities | -0.086 |
feelings | -0.084 |
injuries | -0.083 |
songs | -0.082 |
Token | a |
Token ID | 257 |
Feature activation | +4.35e-03 |
Pos logit contributions | |
ow | +0.027 |
happen | +0.020 |
feelings | +0.019 |
ating | +0.018 |
lf | +0.017 |
Neg logit contributions | |
Whoever | -0.051 |
ursion | -0.048 |
sc | -0.047 |
ounded | -0.047 |
Ready | -0.047 |
Token | little |
Token ID | 1310 |
Feature activation | +4.73e-02 |
Pos logit contributions | |
Whoever | +0.008 |
ursion | +0.008 |
ounded | +0.008 |
ooters | +0.008 |
sc | +0.008 |
Neg logit contributions | |
ow | -0.006 |
happen | -0.005 |
feelings | -0.005 |
Christmas | -0.005 |
stories | -0.005 |
Token | they |
Token ID | 484 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
ow | +0.037 |
Christmas | +0.030 |
stories | +0.029 |
all | +0.029 |
happen | +0.029 |
Neg logit contributions | |
ursion | -0.039 |
ounded | -0.038 |
Whoever | -0.037 |
ooters | -0.037 |
reaching | -0.036 |
Token | were |
Token ID | 547 |
Feature activation | -1.63e-02 |
Pos logit contributions | |
Whoever | +0.033 |
ursion | +0.032 |
ounded | +0.032 |
sc | +0.030 |
ooters | +0.030 |
Neg logit contributions | |
ow | -0.032 |
feelings | -0.026 |
happen | -0.025 |
stories | -0.024 |
Christmas | -0.024 |
Token | able |
Token ID | 1498 |
Feature activation | +8.10e-02 |
Pos logit contributions | |
ow | +0.030 |
feelings | +0.024 |
Christmas | +0.024 |
happen | +0.024 |
stories | +0.023 |
Neg logit contributions | |
Whoever | -0.026 |
ursion | -0.025 |
ounded | -0.025 |
ooters | -0.024 |
sc | -0.024 |
Token | to |
Token ID | 284 |
Feature activation | +7.86e-04 |
Pos logit contributions | |
Whoever | +0.150 |
ursion | +0.146 |
ounded | +0.144 |
sc | +0.141 |
ooters | +0.139 |
Neg logit contributions | |
ow | -0.129 |
feelings | -0.100 |
Christmas | -0.099 |
happen | -0.099 |
stories | -0.096 |
Token | move |
Token ID | 1445 |
Feature activation | +7.59e-03 |
Pos logit contributions | |
Whoever | +0.001 |
ounded | +0.001 |
ursion | +0.001 |
ooters | +0.001 |
reaching | +0.001 |
Neg logit contributions | |
ow | -0.001 |
feelings | -0.001 |
stories | -0.001 |
Christmas | -0.001 |
lf | -0.001 |
Token | the |
Token ID | 262 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
Whoever | +0.018 |
ursion | +0.017 |
ounded | +0.017 |
ooters | +0.017 |
sc | +0.017 |
Neg logit contributions | |
ow | -0.008 |
feelings | -0.005 |
Christmas | -0.005 |
happen | -0.005 |
stories | -0.005 |
Token | tree |
Token ID | 5509 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Whoever | +0.046 |
ounded | +0.044 |
ursion | +0.044 |
ooters | +0.043 |
sc | +0.043 |
Neg logit contributions | |
ow | -0.036 |
feelings | -0.028 |
Christmas | -0.028 |
happen | -0.028 |
stories | -0.027 |
Token | out |
Token ID | 503 |
Feature activation | +7.53e-02 |
Pos logit contributions | |
Whoever | +0.057 |
ursion | +0.056 |
Ready | +0.055 |
ounded | +0.055 |
ooters | +0.054 |
Neg logit contributions | |
ow | -0.025 |
all | -0.020 |
each | -0.019 |
happen | -0.019 |
Christmas | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | +1.74e-02 |
Pos logit contributions | |
Whoever | +0.140 |
ursion | +0.138 |
ounded | +0.137 |
Ready | +0.134 |
ooters | +0.133 |
Neg logit contributions | |
ow | -0.116 |
happen | -0.091 |
feelings | -0.088 |
each | -0.088 |
Christmas | -0.088 |
Token | the |
Token ID | 262 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
Whoever | +0.040 |
ursion | +0.038 |
ounded | +0.038 |
sc | +0.037 |
Ready | +0.037 |
Neg logit contributions | |
ow | -0.019 |
ity | -0.013 |
feelings | -0.013 |
happen | -0.013 |
ating | -0.012 |
Token | way |
Token ID | 835 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Whoever | +0.050 |
ursion | +0.048 |
ounded | +0.048 |
sc | +0.047 |
ooters | +0.047 |
Neg logit contributions | |
ow | -0.032 |
Christmas | -0.023 |
feelings | -0.022 |
happen | -0.022 |
stories | -0.021 |
Token | is |
Token ID | 318 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
ow | +0.068 |
Christmas | +0.050 |
feelings | +0.047 |
ating | +0.047 |
happen | +0.047 |
Neg logit contributions | |
Whoever | -0.095 |
ounded | -0.090 |
ursion | -0.089 |
ooters | -0.089 |
sc | -0.087 |
Token | the |
Token ID | 262 |
Feature activation | +2.50e-02 |
Pos logit contributions | |
ow | +0.017 |
Christmas | +0.012 |
happen | +0.012 |
each | +0.011 |
all | +0.011 |
Neg logit contributions | |
ursion | -0.031 |
ounded | -0.031 |
sc | -0.031 |
Whoever | -0.030 |
ooters | -0.029 |
Token | sky |
Token ID | 6766 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
Whoever | +0.054 |
ursion | +0.053 |
ounded | +0.053 |
sc | +0.051 |
ooters | +0.051 |
Neg logit contributions | |
ow | -0.033 |
feelings | -0.024 |
Christmas | -0.023 |
happen | -0.023 |
stories | -0.022 |
Token | changing |
Token ID | 5609 |
Feature activation | -7.75e-03 |
Pos logit contributions | |
Whoever | +0.042 |
ounded | +0.042 |
ursion | +0.040 |
shadow | +0.039 |
saw | +0.039 |
Neg logit contributions | |
ow | -0.035 |
feelings | -0.030 |
ils | -0.029 |
ities | -0.029 |
foods | -0.028 |
Token | , |
Token ID | 11 |
Feature activation | -6.19e-03 |
Pos logit contributions | |
ow | +0.010 |
Christmas | +0.007 |
feelings | +0.007 |
happen | +0.007 |
lf | +0.007 |
Neg logit contributions | |
Whoever | -0.016 |
ursion | -0.016 |
ounded | -0.016 |
ooters | -0.016 |
sc | -0.016 |
Token | L |
Token ID | 406 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
ow | +0.010 |
feelings | +0.008 |
Christmas | +0.008 |
happen | +0.008 |
stories | +0.008 |
Neg logit contributions | |
Whoever | -0.011 |
ursion | -0.011 |
ounded | -0.011 |
sc | -0.010 |
ooters | -0.010 |
Token | ila |
Token ID | 10102 |
Feature activation | +4.56e-02 |
Pos logit contributions | |
ounded | +0.065 |
Whoever | +0.064 |
Ready | +0.061 |
ursion | +0.061 |
sc | +0.061 |
Neg logit contributions | |
ow | -0.079 |
happen | -0.064 |
ad | -0.064 |
kin | -0.063 |
lf | -0.062 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.83e-02 |
Pos logit contributions | |
Whoever | +0.104 |
ounded | +0.102 |
sc | +0.100 |
ooters | +0.099 |
Ready | +0.099 |
Neg logit contributions | |
ow | -0.053 |
Christmas | -0.038 |
happen | -0.038 |
feelings | -0.037 |
lf | -0.036 |
Token | he |
Token ID | 339 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
ounded | +0.070 |
sc | +0.069 |
ursion | +0.068 |
ooters | +0.064 |
pad | +0.063 |
Neg logit contributions | |
ow | -0.029 |
his | -0.027 |
him | -0.023 |
himself | -0.022 |
the | -0.022 |
Token | asked |
Token ID | 1965 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
ursion | +0.039 |
ooters | +0.038 |
Whoever | +0.038 |
ounded | +0.038 |
sc | +0.038 |
Neg logit contributions | |
ow | -0.047 |
feelings | -0.037 |
lf | -0.037 |
each | -0.036 |
all | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -3.50e-03 |
Pos logit contributions | |
Whoever | +0.178 |
ursion | +0.173 |
ounded | +0.173 |
ooters | +0.169 |
sc | +0.169 |
Neg logit contributions | |
ow | -0.091 |
Christmas | -0.061 |
feelings | -0.061 |
happen | -0.060 |
stories | -0.057 |
Token | Sc |
Token ID | 3351 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
ow | +0.032 |
Christmas | +0.024 |
happen | +0.024 |
feelings | +0.024 |
lf | +0.023 |
Neg logit contributions | |
Whoever | -0.035 |
ursion | -0.035 |
ounded | -0.034 |
ooters | -0.033 |
sc | -0.033 |
Token | amp |
Token ID | 696 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
Ready | +0.028 |
sc | +0.027 |
Ready | +0.026 |
Whoever | +0.026 |
ready | +0.025 |
Neg logit contributions | |
ad | -0.033 |
ow | -0.030 |
ating | -0.028 |
ity | -0.028 |
each | -0.028 |
Token | i |
Token ID | 72 |
Feature activation | -1.64e-02 |
Pos logit contributions | |
ow | +0.023 |
feelings | +0.020 |
stories | +0.019 |
Christmas | +0.019 |
happen | +0.019 |
Neg logit contributions | |
Whoever | -0.026 |
sc | -0.025 |
ursion | -0.025 |
ounded | -0.024 |
reaching | -0.024 |
Token | hopped |
Token ID | 45230 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
ow | +0.024 |
happen | +0.018 |
Christmas | +0.018 |
feelings | +0.018 |
stories | +0.017 |
Neg logit contributions | |
Whoever | -0.032 |
ounded | -0.031 |
ursion | -0.031 |
ooters | -0.030 |
sc | -0.030 |
Token | up |
Token ID | 510 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
ooters | +0.023 |
ursion | +0.022 |
Whoever | +0.022 |
ounded | +0.022 |
hole | +0.021 |
Neg logit contributions | |
all | -0.013 |
ow | -0.013 |
every | -0.011 |
each | -0.011 |
end | -0.011 |
Token | the |
Token ID | 262 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
Whoever | +0.085 |
ounded | +0.081 |
ursion | +0.081 |
sc | +0.080 |
reaching | +0.078 |
Neg logit contributions | |
ow | -0.072 |
feelings | -0.055 |
happen | -0.053 |
Christmas | -0.052 |
ity | -0.051 |
Token | hill |
Token ID | 12788 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
Whoever | +0.045 |
ounded | +0.042 |
shadow | +0.042 |
ursion | +0.041 |
reaching | +0.040 |
Neg logit contributions | |
ow | -0.033 |
Christmas | -0.025 |
lf | -0.025 |
all | -0.024 |
happen | -0.024 |
Token | excited |
Token ID | 6568 |
Feature activation | +1.46e-03 |
Pos logit contributions | |
Whoever | +0.098 |
ursion | +0.096 |
Ready | +0.096 |
ooters | +0.095 |
ounded | +0.094 |
Neg logit contributions | |
ow | -0.029 |
all | -0.022 |
each | -0.021 |
every | -0.019 |
Christmas | -0.019 |
Token | ly |
Token ID | 306 |
Feature activation | +2.55e-02 |
Pos logit contributions | |
Whoever | +0.002 |
ounded | +0.002 |
ooters | +0.002 |
sc | +0.002 |
reaching | +0.002 |
Neg logit contributions | |
ow | -0.003 |
Christmas | -0.002 |
kin | -0.002 |
feelings | -0.002 |
happen | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
Whoever | +0.057 |
ursion | +0.056 |
ounded | +0.055 |
ooters | +0.054 |
Ready | +0.054 |
Neg logit contributions | |
ow | -0.030 |
happen | -0.021 |
Christmas | -0.021 |
feelings | -0.021 |
stories | -0.021 |
Token | When |
Token ID | 1649 |
Feature activation | +4.53e-02 |
Pos logit contributions | |
ow | +0.039 |
all | +0.032 |
fur | +0.029 |
each | +0.029 |
Christmas | +0.029 |
Neg logit contributions | |
ursion | -0.060 |
sc | -0.059 |
Whoever | -0.058 |
ounded | -0.056 |
hole | -0.056 |
Token | shuts |
Token ID | 44854 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
Whoever | +0.040 |
ursion | +0.038 |
ounded | +0.038 |
ooters | +0.037 |
sc | +0.037 |
Neg logit contributions | |
ow | -0.041 |
Christmas | -0.032 |
feelings | -0.032 |
happen | -0.031 |
lf | -0.031 |
Token | his |
Token ID | 465 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
Whoever | +0.072 |
sc | +0.068 |
ursion | +0.068 |
ounded | +0.067 |
ooters | +0.067 |
Neg logit contributions | |
ow | -0.031 |
happen | -0.020 |
feelings | -0.019 |
ity | -0.018 |
Christmas | -0.018 |
Token | door |
Token ID | 3420 |
Feature activation | +2.88e-02 |
Pos logit contributions | |
ow | +0.040 |
happen | +0.030 |
Christmas | +0.030 |
feelings | +0.030 |
stories | +0.029 |
Neg logit contributions | |
Whoever | -0.047 |
ounded | -0.045 |
ursion | -0.045 |
ooters | -0.044 |
sc | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | +2.27e-04 |
Pos logit contributions | |
Whoever | +0.067 |
ursion | +0.065 |
ounded | +0.065 |
ooters | +0.063 |
sc | +0.063 |
Neg logit contributions | |
ow | -0.032 |
feelings | -0.022 |
Christmas | -0.021 |
happen | -0.021 |
stories | -0.020 |
Token | plays |
Token ID | 5341 |
Feature activation | -3.10e-02 |
Pos logit contributions | |
Whoever | +0.000 |
ounded | +0.000 |
ursion | +0.000 |
sc | +0.000 |
ooters | +0.000 |
Neg logit contributions | |
ow | -0.000 |
happen | -0.000 |
feelings | -0.000 |
Christmas | -0.000 |
stories | -0.000 |
Token | quietly |
Token ID | 12703 |
Feature activation | +3.57e-02 |
Pos logit contributions | |
ow | +0.043 |
Christmas | +0.028 |
feelings | +0.028 |
happen | +0.028 |
ity | +0.027 |
Neg logit contributions | |
Whoever | -0.065 |
sc | -0.063 |
ounded | -0.062 |
ooters | -0.061 |
Ready | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | +5.21e-03 |
Pos logit contributions | |
Whoever | +0.077 |
ounded | +0.075 |
ursion | +0.075 |
ooters | +0.074 |
sc | +0.074 |
Neg logit contributions | |
ow | -0.045 |
feelings | -0.030 |
happen | -0.030 |
Christmas | -0.030 |
stories | -0.028 |
Token | She |
Token ID | 1375 |
Feature activation | -8.39e-04 |
Pos logit contributions | |
sc | +0.010 |
ounded | +0.010 |
Whoever | +0.010 |
ursion | +0.010 |
hole | +0.009 |
Neg logit contributions | |
ow | -0.009 |
feelings | -0.006 |
Christmas | -0.006 |
each | -0.006 |
happen | -0.006 |
Token | says |
Token ID | 1139 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
ow | +0.002 |
feelings | +0.001 |
happen | +0.001 |
Christmas | +0.001 |
stories | +0.001 |
Neg logit contributions | |
Whoever | -0.001 |
ursion | -0.001 |
ounded | -0.001 |
sc | -0.001 |
ooters | -0.001 |
Token | she |
Token ID | 673 |
Feature activation | +4.57e-02 |
Pos logit contributions | |
Whoever | +0.078 |
ursion | +0.072 |
ooters | +0.071 |
ounded | +0.068 |
reaching | +0.067 |
Neg logit contributions | |
ow | -0.111 |
ating | -0.093 |
feelings | -0.092 |
ity | -0.091 |
happen | -0.091 |
Token | will |
Token ID | 481 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
Whoever | +0.085 |
ursion | +0.082 |
ounded | +0.082 |
sc | +0.079 |
ooters | +0.079 |
Neg logit contributions | |
ow | -0.072 |
feelings | -0.056 |
Christmas | -0.055 |
happen | -0.054 |
stories | -0.053 |
Token | fish |
Token ID | 11084 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
ow | +0.006 |
happen | +0.004 |
feelings | +0.004 |
Christmas | +0.004 |
stories | +0.004 |
Neg logit contributions | |
Whoever | -0.017 |
ounded | -0.016 |
ursion | -0.016 |
Ready | -0.016 |
ooters | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -4.03e-03 |
Pos logit contributions | |
Whoever | +0.068 |
ursion | +0.066 |
ounded | +0.066 |
ooters | +0.065 |
Ready | +0.065 |
Neg logit contributions | |
ow | -0.026 |
happen | -0.017 |
feelings | -0.016 |
Christmas | -0.016 |
stories | -0.015 |
Token | " |
Token ID | 366 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
ow | +0.004 |
his | +0.003 |
each | +0.003 |
every | +0.003 |
was | +0.003 |
Neg logit contributions | |
hole | -0.009 |
ursion | -0.009 |
ounded | -0.009 |
ooters | -0.009 |
reaching | -0.008 |
Token | I |
Token ID | 40 |
Feature activation | +4.03e-02 |
Pos logit contributions | |
ooters | +0.047 |
ounded | +0.046 |
hole | +0.045 |
ursion | +0.045 |
reaching | +0.043 |
Neg logit contributions | |
ow | -0.028 |
each | -0.027 |
Christmas | -0.022 |
ad | -0.021 |
end | -0.021 |
Token | want |
Token ID | 765 |
Feature activation | -4.74e-03 |
Pos logit contributions | |
ooters | +0.069 |
ursion | +0.067 |
Whoever | +0.065 |
ounded | +0.064 |
shadow | +0.064 |
Neg logit contributions | |
ow | -0.063 |
happen | -0.052 |
Christmas | -0.052 |
each | -0.052 |
glow | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
ow | +0.007 |
Christmas | +0.005 |
happen | +0.005 |
feelings | +0.005 |
stories | +0.005 |
Neg logit contributions | |
Whoever | -0.009 |
ursion | -0.009 |
ounded | -0.009 |
ooters | -0.009 |
sc | -0.009 |
Token | touch |
Token ID | 3638 |
Feature activation | -1.97e-02 |
Pos logit contributions | |
ow | +0.031 |
feelings | +0.024 |
Christmas | +0.024 |
happen | +0.023 |
stories | +0.023 |
Neg logit contributions | |
Whoever | -0.034 |
ursion | -0.033 |
ounded | -0.032 |
ooters | -0.032 |
sc | -0.031 |
Token | it |
Token ID | 340 |
Feature activation | -6.62e-03 |
Pos logit contributions | |
ow | +0.026 |
Christmas | +0.018 |
happen | +0.018 |
feelings | +0.018 |
stories | +0.018 |
Neg logit contributions | |
Whoever | -0.042 |
ursion | -0.041 |
ounded | -0.041 |
ooters | -0.040 |
sc | -0.040 |
Token | too |
Token ID | 1165 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
ow | +0.009 |
happen | +0.007 |
Christmas | +0.007 |
feelings | +0.007 |
stories | +0.006 |
Neg logit contributions | |
Whoever | -0.014 |
ursion | -0.013 |
ounded | -0.013 |
ooters | -0.013 |
reaching | -0.013 |
Token | !" |
Token ID | 2474 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
Ready | +0.064 |
ounded | +0.061 |
ready | +0.061 |
ursion | +0.061 |
ooters | +0.060 |
Neg logit contributions | |
ow | -0.033 |
all | -0.028 |
each | -0.026 |
Christmas | -0.023 |
every | -0.023 |
Token |
|
Token ID | 198 |
Feature activation | +4.71e-03 |
Pos logit contributions | |
hole | +0.049 |
ounded | +0.048 |
ursion | +0.048 |
ooters | +0.046 |
reaching | +0.046 |
Neg logit contributions | |
his | -0.018 |
ow | -0.017 |
all | -0.015 |
the | -0.015 |
every | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
ounded | +0.116 |
Whoever | +0.112 |
ursion | +0.112 |
Ready | +0.111 |
shadow | +0.109 |
Neg logit contributions | |
ow | -0.052 |
each | -0.049 |
all | -0.046 |
end | -0.046 |
every | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Whoever | +0.041 |
ursion | +0.038 |
Ready | +0.038 |
sc | +0.038 |
ooters | +0.037 |
Neg logit contributions | |
ow | -0.021 |
ity | -0.016 |
feelings | -0.015 |
happen | -0.015 |
urg | -0.014 |
Token | park |
Token ID | 3952 |
Feature activation | +4.60e-02 |
Pos logit contributions | |
Whoever | +0.027 |
ounded | +0.026 |
ursion | +0.026 |
sc | +0.025 |
ooters | +0.025 |
Neg logit contributions | |
ow | -0.020 |
Christmas | -0.015 |
feelings | -0.015 |
stories | -0.015 |
happen | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | +6.90e-03 |
Pos logit contributions | |
Whoever | +0.110 |
ursion | +0.109 |
ounded | +0.107 |
ooters | +0.105 |
sc | +0.104 |
Neg logit contributions | |
ow | -0.046 |
feelings | -0.031 |
Christmas | -0.030 |
happen | -0.030 |
stories | -0.029 |
Token | They |
Token ID | 1119 |
Feature activation | +8.59e-05 |
Pos logit contributions | |
ounded | +0.015 |
shadow | +0.015 |
ursion | +0.015 |
hole | +0.014 |
ooters | +0.014 |
Neg logit contributions | |
ow | -0.008 |
each | -0.007 |
every | -0.006 |
feelings | -0.006 |
Christmas | -0.006 |
Token | had |
Token ID | 550 |
Feature activation | -1.90e-02 |
Pos logit contributions | |
Whoever | +0.000 |
ounded | +0.000 |
ursion | +0.000 |
sc | +0.000 |
ooters | +0.000 |
Neg logit contributions | |
ow | -0.000 |
happen | -0.000 |
feelings | -0.000 |
Christmas | -0.000 |
stories | -0.000 |
Token | fun |
Token ID | 1257 |
Feature activation | -3.44e-03 |
Pos logit contributions | |
ow | +0.030 |
lf | +0.023 |
happen | +0.023 |
ating | +0.023 |
Christmas | +0.023 |
Neg logit contributions | |
Whoever | -0.035 |
ooters | -0.033 |
ounded | -0.033 |
ursion | -0.032 |
sc | -0.032 |
Token | together |
Token ID | 1978 |
Feature activation | +3.08e-02 |
Pos logit contributions | |
ow | +0.005 |
Christmas | +0.003 |
happen | +0.003 |
urg | +0.003 |
feelings | +0.003 |
Neg logit contributions | |
Whoever | -0.007 |
ooters | -0.007 |
sc | -0.007 |
ursion | -0.007 |
ounded | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
Whoever | +0.063 |
sc | +0.060 |
ounded | +0.059 |
ursion | +0.059 |
ooters | +0.058 |
Neg logit contributions | |
ow | -0.044 |
feelings | -0.030 |
Christmas | -0.030 |
happen | -0.029 |
urg | -0.029 |
Token | They |
Token ID | 1119 |
Feature activation | +4.21e-03 |
Pos logit contributions | |
ursion | +0.028 |
ounded | +0.027 |
shadow | +0.026 |
ooters | +0.026 |
sc | +0.026 |
Neg logit contributions | |
ow | -0.021 |
each | -0.016 |
feelings | -0.016 |
Christmas | -0.015 |
stories | -0.015 |
Token | did |
Token ID | 750 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
Whoever | +0.008 |
ounded | +0.008 |
ursion | +0.008 |
sc | +0.007 |
Ready | +0.007 |
Neg logit contributions | |
ow | -0.007 |
happen | -0.005 |
Christmas | -0.005 |
soup | -0.005 |
urg | -0.005 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.15e-02 |
Pos logit contributions | |
ow | +0.052 |
Christmas | +0.039 |
stories | +0.037 |
all | +0.037 |
each | +0.037 |
Neg logit contributions | |
ounded | -0.107 |
ursion | -0.107 |
Whoever | -0.106 |
reaching | -0.102 |
hole | -0.102 |
Token | and |
Token ID | 290 |
Feature activation | -4.50e-03 |
Pos logit contributions | |
Whoever | +0.050 |
ursion | +0.048 |
ounded | +0.048 |
ooters | +0.047 |
sc | +0.047 |
Neg logit contributions | |
ow | -0.025 |
feelings | -0.016 |
Christmas | -0.016 |
happen | -0.016 |
stories | -0.015 |
Token | started |
Token ID | 2067 |
Feature activation | +6.37e-02 |
Pos logit contributions | |
ow | +0.006 |
Christmas | +0.004 |
feelings | +0.004 |
stories | +0.004 |
happen | +0.004 |
Neg logit contributions | |
Whoever | -0.010 |
ursion | -0.010 |
ounded | -0.010 |
sc | -0.009 |
ooters | -0.009 |
Token | driving |
Token ID | 5059 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
Whoever | +0.117 |
ursion | +0.114 |
ounded | +0.113 |
sc | +0.110 |
ooters | +0.109 |
Neg logit contributions | |
ow | -0.101 |
Christmas | -0.079 |
feelings | -0.078 |
happen | -0.077 |
stories | -0.076 |
Token | together |
Token ID | 1978 |
Feature activation | +2.46e-02 |
Pos logit contributions | |
ursion | +0.076 |
Whoever | +0.075 |
ounded | +0.073 |
reaching | +0.071 |
ooters | +0.070 |
Neg logit contributions | |
ow | -0.057 |
each | -0.047 |
Christmas | -0.047 |
stories | -0.046 |
feelings | -0.046 |
Token | . |
Token ID | 13 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
ursion | +0.060 |
Whoever | +0.060 |
ounded | +0.060 |
Ready | +0.057 |
reaching | +0.057 |
Neg logit contributions | |
ow | -0.022 |
Christmas | -0.016 |
each | -0.016 |
all | -0.016 |
happen | -0.015 |
Token | They |
Token ID | 1119 |
Feature activation | -2.05e-02 |
Pos logit contributions | |
ow | +0.036 |
each | +0.034 |
all | +0.034 |
every | +0.032 |
stories | +0.031 |
Neg logit contributions | |
ursion | -0.068 |
shadow | -0.067 |
hole | -0.067 |
ounded | -0.066 |
one | -0.062 |
Token | had |
Token ID | 550 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
ow | +0.034 |
feelings | +0.027 |
happen | +0.027 |
Christmas | +0.026 |
stories | +0.025 |
Neg logit contributions | |
Whoever | -0.037 |
ursion | -0.035 |
ounded | -0.035 |
ooters | -0.034 |
sc | -0.034 |
Token | so |
Token ID | 523 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
ow | +0.032 |
Christmas | +0.024 |
feelings | +0.023 |
stories | +0.023 |
each | +0.022 |
Neg logit contributions | |
ursion | -0.059 |
Whoever | -0.058 |
ounded | -0.058 |
sc | -0.056 |
ooters | -0.055 |
Token | much |
Token ID | 881 |
Feature activation | +1.35e-03 |
Pos logit contributions | |
Whoever | +0.046 |
ursion | +0.045 |
ounded | +0.044 |
ooters | +0.043 |
sc | +0.043 |
Neg logit contributions | |
ow | -0.060 |
feelings | -0.049 |
Christmas | -0.049 |
happen | -0.048 |
stories | -0.047 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.73e-03 |
Pos logit contributions | |
Whoever | +0.003 |
ursion | +0.002 |
ounded | +0.002 |
sc | +0.002 |
ooters | +0.002 |
Neg logit contributions | |
ow | -0.002 |
happen | -0.002 |
Christmas | -0.002 |
stories | -0.002 |
feelings | -0.002 |
Token | try |
Token ID | 1949 |
Feature activation | +6.27e-03 |
Pos logit contributions | |
Whoever | +0.032 |
ursion | +0.032 |
bet | +0.031 |
wanna | +0.030 |
ounded | +0.030 |
Neg logit contributions | |
ow | -0.036 |
lf | -0.033 |
ity | -0.030 |
kin | -0.030 |
urg | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
Whoever | +0.011 |
ounded | +0.010 |
ooters | +0.010 |
sc | +0.010 |
ursion | +0.010 |
Neg logit contributions | |
ow | -0.011 |
Christmas | -0.008 |
lf | -0.008 |
feelings | -0.008 |
happen | -0.008 |
Token | help |
Token ID | 1037 |
Feature activation | -3.40e-02 |
Pos logit contributions | |
ow | +0.046 |
ling | +0.037 |
Christmas | +0.036 |
lf | +0.036 |
urg | +0.035 |
Neg logit contributions | |
Whoever | -0.040 |
wanna | -0.036 |
Ready | -0.036 |
bet | -0.036 |
shadow | -0.036 |
Token | you |
Token ID | 345 |
Feature activation | +5.18e-03 |
Pos logit contributions | |
ow | +0.048 |
Christmas | +0.034 |
lf | +0.034 |
ling | +0.034 |
feelings | +0.033 |
Neg logit contributions | |
Whoever | -0.067 |
sc | -0.063 |
ooters | -0.063 |
ounded | -0.062 |
ursion | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
Whoever | +0.010 |
ounded | +0.010 |
ursion | +0.009 |
sc | +0.009 |
ooters | +0.009 |
Neg logit contributions | |
ow | -0.008 |
lf | -0.006 |
Christmas | -0.006 |
ling | -0.005 |
unic | -0.005 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
ow | +0.048 |
feelings | +0.042 |
ad | +0.041 |
Christmas | +0.041 |
each | +0.041 |
Neg logit contributions | |
ounded | -0.054 |
sc | -0.050 |
ursion | -0.048 |
ooters | -0.048 |
reaching | -0.046 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.82e-02 |
Pos logit contributions | |
ow | +0.089 |
feelings | +0.072 |
happen | +0.071 |
Christmas | +0.071 |
stories | +0.069 |
Neg logit contributions | |
Whoever | -0.069 |
ursion | -0.066 |
ounded | -0.065 |
sc | -0.063 |
ooters | -0.063 |
Token | upon |
Token ID | 2402 |
Feature activation | -7.25e-03 |
Pos logit contributions | |
ow | +0.036 |
Christmas | +0.030 |
all | +0.030 |
happen | +0.030 |
each | +0.029 |
Neg logit contributions | |
Whoever | -0.026 |
sc | -0.025 |
Ready | -0.025 |
ursion | -0.025 |
ounded | -0.024 |
Token | a |
Token ID | 257 |
Feature activation | -1.01e-04 |
Pos logit contributions | |
ow | +0.010 |
ating | +0.009 |
urg | +0.008 |
ling | +0.008 |
happen | +0.008 |
Neg logit contributions | |
Whoever | -0.014 |
Ready | -0.013 |
reaching | -0.013 |
ursion | -0.012 |
sc | -0.012 |
Token | time |
Token ID | 640 |
Feature activation | +2.52e-02 |
Pos logit contributions | |
ow | +0.000 |
happen | +0.000 |
feelings | +0.000 |
Christmas | +0.000 |
ating | +0.000 |
Neg logit contributions | |
Whoever | -0.000 |
ursion | -0.000 |
ounded | -0.000 |
sc | -0.000 |
reaching | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
Whoever | +0.055 |
ounded | +0.053 |
ursion | +0.053 |
ooters | +0.052 |
sc | +0.052 |
Neg logit contributions | |
ow | -0.031 |
feelings | -0.022 |
Christmas | -0.022 |
happen | -0.022 |
stories | -0.021 |
Token | home |
Token ID | 1363 |
Feature activation | -3.46e-03 |
Pos logit contributions | |
Whoever | +0.023 |
ounded | +0.022 |
Ready | +0.022 |
sc | +0.021 |
ursion | +0.021 |
Neg logit contributions | |
ow | -0.015 |
Christmas | -0.012 |
all | -0.012 |
each | -0.012 |
happen | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
ow | +0.004 |
feelings | +0.003 |
Christmas | +0.003 |
happen | +0.003 |
stories | +0.003 |
Neg logit contributions | |
Whoever | -0.008 |
ounded | -0.007 |
ursion | -0.007 |
ooters | -0.007 |
sc | -0.007 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
ow | +0.023 |
each | +0.021 |
all | +0.021 |
every | +0.019 |
dinner | +0.018 |
Neg logit contributions | |
ooters | -0.039 |
hole | -0.037 |
ounded | -0.037 |
ursion | -0.037 |
sc | -0.036 |
Token | Once |
Token ID | 7454 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
ow | +0.078 |
Christmas | +0.062 |
happen | +0.060 |
feelings | +0.058 |
stories | +0.057 |
Neg logit contributions | |
ounded | -0.081 |
Whoever | -0.081 |
ursion | -0.078 |
ooters | -0.078 |
reaching | -0.075 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.44e-02 |
Pos logit contributions | |
ow | +0.035 |
all | +0.030 |
Christmas | +0.029 |
end | +0.029 |
each | +0.029 |
Neg logit contributions | |
Whoever | -0.028 |
sc | -0.028 |
Ready | -0.028 |
ursion | -0.026 |
ooters | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | +2.19e-04 |
Pos logit contributions | |
ow | +0.020 |
ating | +0.016 |
happen | +0.015 |
urg | +0.015 |
ling | +0.015 |
Neg logit contributions | |
Whoever | -0.028 |
ursion | -0.026 |
sc | -0.026 |
reaching | -0.026 |
Ready | -0.026 |
Token | time |
Token ID | 640 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
Whoever | +0.000 |
ursion | +0.000 |
ounded | +0.000 |
ooters | +0.000 |
sc | +0.000 |
Neg logit contributions | |
ow | -0.000 |
feelings | -0.000 |
happen | -0.000 |
Christmas | -0.000 |
stories | -0.000 |
Token | , |
Token ID | 11 |
Feature activation | -3.37e-02 |
Pos logit contributions | |
Whoever | +0.055 |
ursion | +0.055 |
ounded | +0.054 |
ooters | +0.053 |
sc | +0.053 |
Neg logit contributions | |
ow | -0.025 |
Christmas | -0.016 |
happen | -0.016 |
feelings | -0.015 |
stories | -0.014 |
Token | there |
Token ID | 612 |
Feature activation | +5.48e-02 |
Pos logit contributions | |
ow | +0.060 |
Christmas | +0.047 |
happen | +0.046 |
feelings | +0.046 |
stories | +0.044 |
Neg logit contributions | |
Whoever | -0.057 |
ursion | -0.055 |
ooters | -0.055 |
ounded | -0.054 |
sc | -0.053 |
Token | was |
Token ID | 373 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
Whoever | +0.093 |
sc | +0.082 |
Ready | +0.080 |
bet | +0.080 |
ounded | +0.079 |
Neg logit contributions | |
ow | -0.084 |
feelings | -0.077 |
ities | -0.077 |
songs | -0.074 |
stories | -0.073 |
Token | a |
Token ID | 257 |
Feature activation | +3.94e-03 |
Pos logit contributions | |
ow | +0.027 |
happen | +0.020 |
feelings | +0.019 |
ating | +0.018 |
lf | +0.017 |
Neg logit contributions | |
Whoever | -0.055 |
ursion | -0.053 |
sc | -0.052 |
ounded | -0.052 |
Ready | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
Whoever | +0.025 |
ursion | +0.024 |
ounded | +0.024 |
sc | +0.023 |
ooters | +0.023 |
Neg logit contributions | |
ow | -0.021 |
Christmas | -0.016 |
happen | -0.016 |
feelings | -0.016 |
stories | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | +1.71e-02 |
Pos logit contributions | |
ow | +0.035 |
feelings | +0.026 |
ity | +0.026 |
ling | +0.025 |
ating | +0.025 |
Neg logit contributions | |
Whoever | -0.056 |
ursion | -0.052 |
Ready | -0.052 |
ounded | -0.051 |
ooters | -0.050 |
Token | park |
Token ID | 3952 |
Feature activation | +5.02e-02 |
Pos logit contributions | |
Whoever | +0.025 |
Ready | +0.024 |
ursion | +0.024 |
saw | +0.023 |
ooters | +0.023 |
Neg logit contributions | |
ow | -0.032 |
feelings | -0.027 |
ity | -0.026 |
happen | -0.026 |
ling | -0.026 |
Token | with |
Token ID | 351 |
Feature activation | -2.12e-02 |
Pos logit contributions | |
Whoever | +0.118 |
ursion | +0.115 |
ounded | +0.114 |
ooters | +0.112 |
sc | +0.112 |
Neg logit contributions | |
ow | -0.055 |
feelings | -0.036 |
Christmas | -0.036 |
happen | -0.036 |
stories | -0.033 |
Token | his |
Token ID | 465 |
Feature activation | -3.66e-02 |
Pos logit contributions | |
ow | +0.027 |
ling | +0.024 |
ating | +0.024 |
ities | +0.024 |
ity | +0.023 |
Neg logit contributions | |
Whoever | -0.038 |
First | -0.038 |
ursion | -0.037 |
First | -0.037 |
ooters | -0.037 |
Token | mom |
Token ID | 1995 |
Feature activation | -6.75e-02 |
Pos logit contributions | |
ow | +0.048 |
happen | +0.040 |
ating | +0.038 |
urg | +0.037 |
ity | +0.037 |
Neg logit contributions | |
Whoever | -0.071 |
ursion | -0.070 |
ooters | -0.070 |
ounded | -0.069 |
reaching | -0.066 |
Token | . |
Token ID | 13 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
ow | +0.063 |
Christmas | +0.042 |
feelings | +0.039 |
happen | +0.038 |
each | +0.037 |
Neg logit contributions | |
Whoever | -0.168 |
ounded | -0.166 |
ursion | -0.164 |
ooters | -0.160 |
sc | -0.159 |
Token | They |
Token ID | 1119 |
Feature activation | -1.70e-02 |
Pos logit contributions | |
ow | +0.048 |
Christmas | +0.039 |
feelings | +0.038 |
happen | +0.038 |
stories | +0.037 |
Neg logit contributions | |
Whoever | -0.036 |
ursion | -0.035 |
ounded | -0.035 |
sc | -0.033 |
ooters | -0.033 |
Token | saw |
Token ID | 2497 |
Feature activation | +7.58e-03 |
Pos logit contributions | |
ow | +0.025 |
Christmas | +0.019 |
feelings | +0.019 |
happen | +0.018 |
stories | +0.018 |
Neg logit contributions | |
Whoever | -0.033 |
ursion | -0.032 |
ounded | -0.032 |
ooters | -0.031 |
sc | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
Whoever | +0.015 |
saw | +0.014 |
ursion | +0.014 |
Ready | +0.014 |
ooters | +0.014 |
Neg logit contributions | |
ow | -0.009 |
lf | -0.008 |
happen | -0.008 |
ating | -0.008 |
kin | -0.007 |
Token | big |
Token ID | 1263 |
Feature activation | -9.10e-03 |
Pos logit contributions | |
Whoever | +0.025 |
ursion | +0.024 |
ounded | +0.024 |
ooters | +0.023 |
sc | +0.023 |
Neg logit contributions | |
ow | -0.013 |
Christmas | -0.008 |
feelings | -0.008 |
happen | -0.008 |
stories | -0.008 |
Token | very |
Token ID | 845 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
ow | +0.005 |
Christmas | +0.004 |
feelings | +0.004 |
happen | +0.004 |
stories | +0.004 |
Neg logit contributions | |
Whoever | -0.006 |
ursion | -0.006 |
ounded | -0.006 |
sc | -0.006 |
ooters | -0.006 |
Token | sad |
Token ID | 6507 |
Feature activation | -8.19e-03 |
Pos logit contributions | |
Whoever | +0.029 |
sc | +0.028 |
Ready | +0.027 |
ursion | +0.027 |
hole | +0.027 |
Neg logit contributions | |
ow | -0.021 |
Christmas | -0.017 |
feelings | -0.016 |
happen | -0.016 |
all | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +3.98e-03 |
Pos logit contributions | |
ow | +0.011 |
happen | +0.008 |
Christmas | +0.008 |
feelings | +0.008 |
lf | +0.008 |
Neg logit contributions | |
Whoever | -0.017 |
ounded | -0.016 |
ursion | -0.016 |
ooters | -0.016 |
sc | -0.016 |
Token | she |
Token ID | 673 |
Feature activation | +2.73e-02 |
Pos logit contributions | |
Whoever | +0.007 |
ursion | +0.007 |
ounded | +0.007 |
ooters | +0.007 |
sc | +0.006 |
Neg logit contributions | |
ow | -0.007 |
Christmas | -0.005 |
happen | -0.005 |
feelings | -0.005 |
stories | -0.005 |
Token | we |
Token ID | 356 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
Whoever | +0.055 |
ursion | +0.054 |
ooters | +0.053 |
ounded | +0.053 |
sc | +0.052 |
Neg logit contributions | |
ow | -0.038 |
feelings | -0.028 |
Christmas | -0.028 |
happen | -0.027 |
stories | -0.027 |
Token | pt |
Token ID | 457 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
Whoever | +0.070 |
ursion | +0.067 |
ounded | +0.067 |
ooters | +0.066 |
sc | +0.066 |
Neg logit contributions | |
ow | -0.038 |
Christmas | -0.027 |
feelings | -0.027 |
happen | -0.026 |
stories | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -2.10e-02 |
Pos logit contributions | |
ow | +0.100 |
happen | +0.080 |
ad | +0.077 |
lf | +0.075 |
feelings | +0.074 |
Neg logit contributions | |
ounded | -0.125 |
Whoever | -0.123 |
ursion | -0.120 |
sc | -0.117 |
reaching | -0.115 |
Token | She |
Token ID | 1375 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
ow | +0.031 |
each | +0.024 |
soup | +0.023 |
Christmas | +0.023 |
all | +0.022 |
Neg logit contributions | |
ooters | -0.043 |
sc | -0.042 |
ounded | -0.041 |
ursion | -0.041 |
Whoever | -0.040 |
Token | was |
Token ID | 373 |
Feature activation | +3.00e-03 |
Pos logit contributions | |
ow | +0.021 |
Christmas | +0.016 |
feelings | +0.016 |
happen | +0.016 |
stories | +0.015 |
Neg logit contributions | |
Whoever | -0.028 |
ursion | -0.028 |
ounded | -0.027 |
ooters | -0.027 |
sc | -0.026 |
Token | so |
Token ID | 523 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
Whoever | +0.005 |
ursion | +0.005 |
ounded | +0.005 |
ooters | +0.005 |
sc | +0.005 |
Neg logit contributions | |
ow | -0.005 |
feelings | -0.004 |
happen | -0.004 |
Christmas | -0.004 |
stories | -0.004 |
Token | sad |
Token ID | 6507 |
Feature activation | -1.80e-04 |
Pos logit contributions | |
Whoever | +0.071 |
ursion | +0.067 |
ounded | +0.067 |
sc | +0.066 |
ooters | +0.066 |
Neg logit contributions | |
ow | -0.081 |
Christmas | -0.065 |
feelings | -0.064 |
happen | -0.064 |
stories | -0.062 |
Token | soldier |
Token ID | 10686 |
Feature activation | +1.93e-02 |
Pos logit contributions | |
ursion | +0.062 |
Whoever | +0.062 |
ounded | +0.062 |
ooters | +0.061 |
hole | +0.060 |
Neg logit contributions | |
ow | -0.032 |
Christmas | -0.023 |
feelings | -0.022 |
happen | -0.021 |
stories | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
Whoever | +0.042 |
ursion | +0.040 |
ounded | +0.040 |
sc | +0.039 |
ooters | +0.039 |
Neg logit contributions | |
ow | -0.024 |
happen | -0.017 |
Christmas | -0.017 |
feelings | -0.017 |
stories | -0.016 |
Token | He |
Token ID | 679 |
Feature activation | -4.88e-03 |
Pos logit contributions | |
ounded | +0.033 |
ursion | +0.033 |
hole | +0.031 |
reaching | +0.031 |
sc | +0.031 |
Neg logit contributions | |
ow | -0.016 |
his | -0.014 |
feelings | -0.013 |
himself | -0.013 |
each | -0.012 |
Token | wanted |
Token ID | 2227 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
ow | +0.008 |
feelings | +0.006 |
Christmas | +0.006 |
happen | +0.006 |
stories | +0.005 |
Neg logit contributions | |
Whoever | -0.009 |
ursion | -0.009 |
ounded | -0.009 |
ooters | -0.009 |
sc | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
Whoever | +0.029 |
ooters | +0.027 |
ursion | +0.027 |
ounded | +0.027 |
sc | +0.027 |
Neg logit contributions | |
ow | -0.030 |
happen | -0.023 |
Christmas | -0.023 |
feelings | -0.023 |
lf | -0.022 |
Token | be |
Token ID | 307 |
Feature activation | -6.40e-02 |
Pos logit contributions | |
ow | +0.032 |
Christmas | +0.024 |
feelings | +0.024 |
ating | +0.023 |
happen | +0.023 |
Neg logit contributions | |
Whoever | -0.041 |
ursion | -0.039 |
ooters | -0.039 |
ounded | -0.039 |
Ready | -0.038 |
Token | brave |
Token ID | 14802 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
ow | +0.066 |
feelings | +0.043 |
Christmas | +0.042 |
happen | +0.042 |
stories | +0.040 |
Neg logit contributions | |
Whoever | -0.155 |
ursion | -0.151 |
ounded | -0.151 |
sc | -0.147 |
ooters | -0.147 |
Token | and |
Token ID | 290 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
Whoever | +0.068 |
ursion | +0.067 |
ounded | +0.066 |
ooters | +0.064 |
reaching | +0.064 |
Neg logit contributions | |
ow | -0.027 |
feelings | -0.019 |
stories | -0.017 |
Christmas | -0.017 |
happen | -0.016 |
Token | strong |
Token ID | 1913 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
Whoever | +0.043 |
ursion | +0.042 |
ounded | +0.041 |
sc | +0.041 |
ooters | +0.041 |
Neg logit contributions | |
ow | -0.027 |
feelings | -0.021 |
happen | -0.021 |
stories | -0.020 |
Christmas | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
Whoever | +0.035 |
ursion | +0.034 |
ounded | +0.034 |
ooters | +0.033 |
sc | +0.033 |
Neg logit contributions | |
ow | -0.018 |
Christmas | -0.012 |
happen | -0.012 |
feelings | -0.012 |
stories | -0.011 |
Token | He |
Token ID | 679 |
Feature activation | -3.25e-03 |
Pos logit contributions | |
ounded | +0.039 |
ursion | +0.039 |
hole | +0.037 |
reaching | +0.036 |
ooters | +0.036 |
Neg logit contributions | |
his | -0.016 |
ow | -0.015 |
himself | -0.014 |
each | -0.013 |
every | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
Whoever | +0.083 |
ursion | +0.081 |
ounded | +0.080 |
ooters | +0.079 |
sc | +0.079 |
Neg logit contributions | |
ow | -0.045 |
feelings | -0.031 |
Christmas | -0.031 |
happen | -0.030 |
stories | -0.029 |
Token | But |
Token ID | 887 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
ow | +0.027 |
each | +0.023 |
all | +0.022 |
jokes | +0.022 |
stories | +0.022 |
Neg logit contributions | |
ursion | -0.030 |
ounded | -0.029 |
sc | -0.028 |
shadow | -0.028 |
ooters | -0.027 |
Token | he |
Token ID | 339 |
Feature activation | +5.92e-03 |
Pos logit contributions | |
Whoever | +0.057 |
ursion | +0.055 |
ounded | +0.055 |
ooters | +0.054 |
sc | +0.053 |
Neg logit contributions | |
ow | -0.041 |
Christmas | -0.031 |
feelings | -0.030 |
happen | -0.030 |
stories | -0.029 |
Token | couldn |
Token ID | 3521 |
Feature activation | +2.91e-03 |
Pos logit contributions | |
Whoever | +0.011 |
ounded | +0.011 |
ursion | +0.011 |
reaching | +0.010 |
sc | +0.010 |
Neg logit contributions | |
ow | -0.009 |
happen | -0.007 |
Christmas | -0.007 |
feelings | -0.007 |
stories | -0.007 |
Token | 't |
Token ID | 470 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
Whoever | +0.005 |
ursion | +0.005 |
ounded | +0.005 |
sc | +0.005 |
ooters | +0.005 |
Neg logit contributions | |
ow | -0.005 |
feelings | -0.004 |
Christmas | -0.004 |
happen | -0.004 |
stories | -0.004 |
Token | do |
Token ID | 466 |
Feature activation | -7.12e-02 |
Pos logit contributions | |
ow | +0.021 |
feelings | +0.017 |
lf | +0.016 |
stories | +0.016 |
Christmas | +0.016 |
Neg logit contributions | |
Whoever | -0.021 |
ounded | -0.019 |
ursion | -0.019 |
bet | -0.019 |
ooters | -0.019 |
Token | that |
Token ID | 326 |
Feature activation | -3.15e-02 |
Pos logit contributions | |
ow | +0.110 |
Christmas | +0.082 |
feelings | +0.081 |
lf | +0.079 |
urg | +0.077 |
Neg logit contributions | |
Whoever | -0.135 |
ounded | -0.123 |
ooters | -0.120 |
ursion | -0.120 |
reaching | -0.118 |
Token | . |
Token ID | 13 |
Feature activation | -1.47e-02 |
Pos logit contributions | |
ow | +0.034 |
happen | +0.023 |
feelings | +0.023 |
Christmas | +0.022 |
stories | +0.021 |
Neg logit contributions | |
Whoever | -0.074 |
ursion | -0.073 |
ounded | -0.073 |
ooters | -0.071 |
sc | -0.071 |
Token | That |
Token ID | 1320 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
ow | +0.022 |
each | +0.017 |
all | +0.017 |
feelings | +0.017 |
Christmas | +0.016 |
Neg logit contributions | |
ursion | -0.028 |
ounded | -0.028 |
sc | -0.028 |
ooters | -0.027 |
shadow | -0.026 |
Token | 's |
Token ID | 338 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
ow | +0.063 |
feelings | +0.046 |
Christmas | +0.046 |
happen | +0.046 |
stories | +0.044 |
Neg logit contributions | |
Whoever | -0.081 |
ursion | -0.078 |
ounded | -0.078 |
sc | -0.076 |
ooters | -0.075 |
Token | why |
Token ID | 1521 |
Feature activation | -4.56e-02 |
Pos logit contributions | |
ow | +0.026 |
Christmas | +0.020 |
happen | +0.019 |
lf | +0.019 |
feelings | +0.019 |
Neg logit contributions | |
Whoever | -0.033 |
ounded | -0.032 |
ursion | -0.031 |
shadow | -0.031 |
ooters | -0.031 |
Token | toys |
Token ID | 14958 |
Feature activation | -2.89e-03 |
Pos logit contributions | |
ow | +0.054 |
urg | +0.038 |
ad | +0.037 |
happen | +0.035 |
ling | +0.035 |
Neg logit contributions | |
ooters | -0.069 |
sc | -0.068 |
Whoever | -0.068 |
ounded | -0.067 |
ursion | -0.066 |
Token | and |
Token ID | 290 |
Feature activation | -2.42e-02 |
Pos logit contributions | |
ow | +0.005 |
urg | +0.003 |
ling | +0.003 |
unic | +0.003 |
Christmas | +0.003 |
Neg logit contributions | |
Whoever | -0.005 |
sc | -0.005 |
ounded | -0.005 |
ooters | -0.005 |
ursion | -0.005 |
Token | how |
Token ID | 703 |
Feature activation | -1.69e-02 |
Pos logit contributions | |
ow | +0.033 |
Christmas | +0.023 |
feelings | +0.023 |
happen | +0.023 |
stories | +0.022 |
Neg logit contributions | |
Whoever | -0.050 |
ursion | -0.048 |
ounded | -0.048 |
ooters | -0.047 |
sc | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
ow | +0.026 |
Christmas | +0.021 |
feelings | +0.021 |
happen | +0.021 |
stories | +0.020 |
Neg logit contributions | |
Whoever | -0.032 |
ursion | -0.030 |
ounded | -0.030 |
sc | -0.030 |
ooters | -0.029 |
Token | not |
Token ID | 407 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
ow | +0.048 |
ling | +0.034 |
feelings | +0.034 |
urg | +0.034 |
Christmas | +0.034 |
Neg logit contributions | |
Whoever | -0.052 |
ursion | -0.046 |
Ready | -0.046 |
ooters | -0.045 |
ounded | -0.044 |
Token | be |
Token ID | 307 |
Feature activation | -6.99e-02 |
Pos logit contributions | |
ow | +0.067 |
Christmas | +0.049 |
feelings | +0.048 |
lf | +0.047 |
stories | +0.047 |
Neg logit contributions | |
Whoever | -0.086 |
ursion | -0.081 |
ounded | -0.080 |
ooters | -0.080 |
sc | -0.077 |
Token | mean |
Token ID | 1612 |
Feature activation | -2.69e-02 |
Pos logit contributions | |
ow | +0.110 |
feelings | +0.091 |
Christmas | +0.086 |
happen | +0.085 |
stories | +0.084 |
Neg logit contributions | |
Whoever | -0.131 |
ursion | -0.127 |
ounded | -0.125 |
sc | -0.123 |
ooters | -0.122 |
Token | to |
Token ID | 284 |
Feature activation | -9.51e-04 |
Pos logit contributions | |
ow | +0.031 |
feelings | +0.023 |
happen | +0.022 |
Christmas | +0.022 |
stories | +0.021 |
Neg logit contributions | |
Whoever | -0.061 |
ursion | -0.060 |
ounded | -0.060 |
ooters | -0.058 |
sc | -0.057 |
Token | her |
Token ID | 607 |
Feature activation | +2.25e-03 |
Pos logit contributions | |
ow | +0.001 |
Christmas | +0.001 |
feelings | +0.001 |
stories | +0.001 |
happen | +0.001 |
Neg logit contributions | |
Whoever | -0.002 |
ooters | -0.002 |
ursion | -0.002 |
ounded | -0.002 |
reaching | -0.002 |
Token | friends |
Token ID | 2460 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
Whoever | +0.005 |
ounded | +0.004 |
ursion | +0.004 |
ooters | +0.004 |
sc | +0.004 |
Neg logit contributions | |
ow | -0.003 |
happen | -0.002 |
Christmas | -0.002 |
lf | -0.002 |
ad | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
Whoever | +0.046 |
ursion | +0.044 |
ounded | +0.044 |
sc | +0.043 |
ooters | +0.043 |
Neg logit contributions | |
ow | -0.026 |
Christmas | -0.017 |
feelings | -0.017 |
happen | -0.017 |
stories | -0.016 |