Token | She |
Token ID | 3347 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
� | +0.009 |
wrench | +0.009 |
swollen | +0.009 |
bruises | +0.009 |
hammer | +0.008 |
Neg logit contributions | |
Julia | -0.012 |
Mae | -0.011 |
Amber | -0.011 |
dolphins | -0.011 |
La | -0.011 |
Token | helped |
Token ID | 4193 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
hammer | +0.045 |
bruised | +0.045 |
wrench | +0.045 |
� | +0.045 |
bruises | +0.044 |
Neg logit contributions | |
Julia | -0.058 |
La | -0.055 |
Mae | -0.053 |
Amber | -0.052 |
mate | -0.051 |
Token | Tom |
Token ID | 4186 |
Feature activation | -4.71e-02 |
Pos logit contributions | |
bruised | +0.262 |
� | +0.260 |
wrench | +0.256 |
swollen | +0.255 |
bruises | +0.249 |
Neg logit contributions | |
Julia | -0.241 |
Mae | -0.216 |
La | -0.213 |
Amber | -0.211 |
Jeff | -0.201 |
Token | use |
Token ID | 779 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Julia | +0.077 |
Mae | +0.071 |
Amber | +0.068 |
La | +0.068 |
Lou | +0.066 |
Neg logit contributions | |
swollen | -0.080 |
� | -0.080 |
bruised | -0.079 |
bruises | -0.077 |
wrench | -0.076 |
Token | the |
Token ID | 262 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
� | +0.235 |
leaning | +0.226 |
pling | +0.226 |
uc | +0.221 |
bruised | +0.220 |
Neg logit contributions | |
Julia | -0.133 |
Mae | -0.114 |
Jeff | -0.107 |
Amber | -0.107 |
Lou | -0.104 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.58e-01 |
Pos logit contributions | |
� | +0.078 |
leaning | +0.073 |
pling | +0.073 |
bruised | +0.070 |
uc | +0.069 |
Neg logit contributions | |
Julia | -0.100 |
Amber | -0.092 |
Mae | -0.092 |
dolphins | -0.092 |
dives | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | +4.91e-03 |
Pos logit contributions | |
� | +0.871 |
swollen | +0.868 |
bruised | +0.861 |
bruises | +0.846 |
wrench | +0.816 |
Neg logit contributions | |
Julia | -0.661 |
Mae | -0.576 |
Amber | -0.554 |
La | -0.551 |
Jeff | -0.550 |
Token | adjust |
Token ID | 4532 |
Feature activation | +8.64e-02 |
Pos logit contributions | |
bruised | +0.009 |
swollen | +0.009 |
pling | +0.009 |
� | +0.008 |
uc | +0.008 |
Neg logit contributions | |
dive | -0.006 |
Julia | -0.006 |
swim | -0.006 |
perform | -0.006 |
dance | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +3.78e-02 |
Pos logit contributions | |
� | +0.186 |
swollen | +0.172 |
leaning | +0.172 |
bruises | +0.170 |
bruised | +0.170 |
Neg logit contributions | |
Julia | -0.111 |
Zac | -0.092 |
Mae | -0.091 |
Amber | -0.086 |
Eric | -0.085 |
Token | nut |
Token ID | 6701 |
Feature activation | +1.38e-01 |
Pos logit contributions | |
� | +0.054 |
leaning | +0.049 |
uc | +0.048 |
pling | +0.047 |
swollen | +0.046 |
Neg logit contributions | |
Julia | -0.064 |
mate | -0.063 |
dolphins | -0.062 |
aliens | -0.060 |
La | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | +6.48e-03 |
Pos logit contributions | |
� | +0.241 |
bruised | +0.238 |
swollen | +0.233 |
bruises | +0.232 |
wrench | +0.232 |
Neg logit contributions | |
Julia | -0.221 |
Mae | -0.200 |
La | -0.195 |
Amber | -0.195 |
Zac | -0.189 |
Token | Can |
Token ID | 1680 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
bruised | +0.029 |
swollen | +0.028 |
wrench | +0.028 |
� | +0.028 |
bruises | +0.028 |
Neg logit contributions | |
Julia | -0.029 |
La | -0.026 |
Mae | -0.026 |
Amber | -0.025 |
Zac | -0.025 |
Token | you |
Token ID | 345 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
bruised | +0.061 |
wrench | +0.061 |
� | +0.060 |
swollen | +0.060 |
bruises | +0.059 |
Neg logit contributions | |
Julia | -0.039 |
La | -0.034 |
Mae | -0.034 |
Amber | -0.033 |
Zac | -0.032 |
Token | hand |
Token ID | 1021 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
Julia | +0.048 |
La | +0.044 |
Mae | +0.044 |
Amber | +0.043 |
Lou | +0.042 |
Neg logit contributions | |
bruised | -0.031 |
swollen | -0.030 |
bruises | -0.029 |
wrench | -0.029 |
� | -0.028 |
Token | me |
Token ID | 502 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
� | +0.234 |
bruised | +0.232 |
wrench | +0.228 |
swollen | +0.227 |
bruises | +0.224 |
Neg logit contributions | |
Julia | -0.188 |
La | -0.171 |
Mae | -0.168 |
Amber | -0.165 |
Zac | -0.157 |
Token | the |
Token ID | 262 |
Feature activation | +6.59e-02 |
Pos logit contributions | |
Julia | +0.169 |
La | +0.153 |
Mae | +0.151 |
Amber | +0.149 |
Zac | +0.141 |
Neg logit contributions | |
swollen | -0.213 |
bruised | -0.212 |
bruises | -0.210 |
� | -0.209 |
wrench | -0.207 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.52e-01 |
Pos logit contributions | |
� | +0.089 |
leaning | +0.084 |
uc | +0.080 |
ised | +0.078 |
swollen | +0.077 |
Neg logit contributions | |
princess | -0.117 |
mate | -0.115 |
knight | -0.114 |
actor | -0.113 |
pirate | -0.112 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
bruised | +0.850 |
� | +0.847 |
wrench | +0.839 |
swollen | +0.833 |
bruises | +0.827 |
Neg logit contributions | |
Julia | -0.655 |
La | -0.593 |
Mae | -0.581 |
Amber | -0.570 |
Zac | -0.544 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
Julia | +0.208 |
La | +0.195 |
Mae | +0.190 |
Amber | +0.186 |
Zac | +0.181 |
Neg logit contributions | |
� | -0.167 |
bruised | -0.166 |
wrench | -0.164 |
hammer | -0.160 |
swollen | -0.160 |
Token | picked |
Token ID | 6497 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Julia | +0.045 |
La | +0.042 |
Mae | +0.040 |
Amber | +0.040 |
Zac | +0.038 |
Neg logit contributions | |
� | -0.043 |
bruised | -0.042 |
wrench | -0.042 |
hammer | -0.042 |
swollen | -0.042 |
Token | up |
Token ID | 510 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
� | +0.379 |
bruised | +0.375 |
bruises | +0.375 |
swollen | +0.374 |
wrench | +0.359 |
Neg logit contributions | |
Julia | -0.183 |
Mae | -0.157 |
Amber | -0.155 |
La | -0.148 |
Zac | -0.146 |
Token | the |
Token ID | 262 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
Julia | +0.150 |
Amber | +0.131 |
Alicia | +0.123 |
Mae | +0.122 |
La | +0.121 |
Neg logit contributions | |
bruised | -0.227 |
wrench | -0.223 |
� | -0.221 |
supervision | -0.221 |
bruises | -0.220 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
bruised | +0.850 |
� | +0.847 |
wrench | +0.839 |
swollen | +0.833 |
bruises | +0.827 |
Neg logit contributions | |
Julia | -0.655 |
La | -0.593 |
Mae | -0.581 |
Amber | -0.570 |
Zac | -0.544 |
Token | Lily |
Token ID | 20037 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
Julia | +0.208 |
La | +0.195 |
Mae | +0.190 |
Amber | +0.186 |
Zac | +0.181 |
Neg logit contributions | |
� | -0.167 |
bruised | -0.166 |
wrench | -0.164 |
hammer | -0.160 |
swollen | -0.160 |
Token | picked |
Token ID | 6497 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Julia | +0.045 |
La | +0.042 |
Mae | +0.040 |
Amber | +0.040 |
Zac | +0.038 |
Neg logit contributions | |
� | -0.043 |
bruised | -0.042 |
wrench | -0.042 |
hammer | -0.042 |
swollen | -0.042 |
Token | up |
Token ID | 510 |
Feature activation | -1.08e-01 |
Pos logit contributions | |
� | +0.379 |
bruised | +0.375 |
bruises | +0.375 |
swollen | +0.374 |
wrench | +0.359 |
Neg logit contributions | |
Julia | -0.183 |
Mae | -0.157 |
Amber | -0.155 |
La | -0.148 |
Zac | -0.146 |
Token | the |
Token ID | 262 |
Feature activation | +6.24e-02 |
Pos logit contributions | |
Julia | +0.150 |
Amber | +0.131 |
Alicia | +0.123 |
Mae | +0.122 |
La | +0.121 |
Neg logit contributions | |
bruised | -0.227 |
wrench | -0.223 |
� | -0.221 |
supervision | -0.221 |
bruises | -0.220 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.50e-01 |
Pos logit contributions | |
� | +0.083 |
leaning | +0.077 |
uc | +0.071 |
ised | +0.071 |
anked | +0.069 |
Neg logit contributions | |
knight | -0.111 |
pirate | -0.106 |
princess | -0.104 |
mate | -0.101 |
king | -0.098 |
Token | and |
Token ID | 290 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
bruises | +0.901 |
� | +0.892 |
swollen | +0.891 |
bruised | +0.883 |
mercury | +0.850 |
Neg logit contributions | |
Julia | -0.606 |
Mae | -0.531 |
Amber | -0.512 |
Zac | -0.494 |
Jeff | -0.490 |
Token | gave |
Token ID | 2921 |
Feature activation | +7.44e-02 |
Pos logit contributions | |
� | +0.043 |
bruised | +0.042 |
swollen | +0.040 |
bruises | +0.040 |
wrench | +0.040 |
Neg logit contributions | |
Julia | -0.038 |
Mae | -0.034 |
Amber | -0.033 |
La | -0.033 |
Zac | -0.032 |
Token | it |
Token ID | 340 |
Feature activation | +1.94e-02 |
Pos logit contributions | |
� | +0.154 |
bruised | +0.146 |
swollen | +0.145 |
wrench | +0.144 |
bruises | +0.142 |
Neg logit contributions | |
Julia | -0.110 |
Mae | -0.096 |
Amber | -0.093 |
La | -0.090 |
Jeff | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
� | +0.034 |
bruised | +0.034 |
wrench | +0.033 |
swollen | +0.033 |
bruises | +0.032 |
Neg logit contributions | |
Julia | -0.031 |
La | -0.028 |
Mae | -0.028 |
Amber | -0.027 |
Lou | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +5.96e-02 |
Pos logit contributions | |
bruised | +0.023 |
� | +0.023 |
wrench | +0.022 |
swollen | +0.022 |
bruises | +0.022 |
Neg logit contributions | |
Julia | -0.022 |
Mae | -0.020 |
La | -0.020 |
Amber | -0.020 |
April | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | +3.83e-03 |
Pos logit contributions | |
wrench | +0.009 |
hammer | +0.009 |
bruised | +0.009 |
swollen | +0.009 |
� | +0.009 |
Neg logit contributions | |
Julia | -0.013 |
La | -0.013 |
Mae | -0.012 |
Amber | -0.012 |
April | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | +4.50e-02 |
Pos logit contributions | |
bruised | +0.006 |
bruises | +0.006 |
� | +0.006 |
swollen | +0.006 |
wrench | +0.006 |
Neg logit contributions | |
Julia | -0.006 |
Mae | -0.006 |
La | -0.006 |
Amber | -0.006 |
April | -0.006 |
Token | This |
Token ID | 1212 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
� | +0.055 |
bruised | +0.054 |
wrench | +0.054 |
bruises | +0.052 |
swollen | +0.052 |
Neg logit contributions | |
Julia | -0.095 |
La | -0.088 |
Mae | -0.087 |
Amber | -0.087 |
April | -0.084 |
Token | is |
Token ID | 318 |
Feature activation | -4.35e-02 |
Pos logit contributions | |
wrench | +0.224 |
hammer | +0.224 |
� | +0.220 |
bruised | +0.217 |
bruises | +0.209 |
Neg logit contributions | |
Julia | -0.229 |
La | -0.217 |
Amber | -0.206 |
Mae | -0.206 |
April | -0.197 |
Token | a |
Token ID | 257 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
Julia | +0.066 |
Mae | +0.059 |
La | +0.058 |
Amber | +0.058 |
Jeff | +0.057 |
Neg logit contributions | |
� | -0.082 |
bruised | -0.079 |
swollen | -0.077 |
wrench | -0.077 |
bruises | -0.076 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
Julia | +0.101 |
La | +0.093 |
Mae | +0.091 |
Amber | +0.090 |
Zac | +0.087 |
Neg logit contributions | |
bruised | -0.096 |
� | -0.095 |
wrench | -0.095 |
hammer | -0.093 |
swollen | -0.093 |
Token | . |
Token ID | 13 |
Feature activation | +1.98e-02 |
Pos logit contributions | |
� | +0.792 |
bruised | +0.787 |
swollen | +0.785 |
bruises | +0.773 |
wrench | +0.759 |
Neg logit contributions | |
Julia | -0.691 |
Mae | -0.620 |
La | -0.612 |
Amber | -0.600 |
Zac | -0.590 |
Token | I |
Token ID | 314 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
swollen | +0.035 |
bruised | +0.035 |
� | +0.035 |
wrench | +0.034 |
bruises | +0.033 |
Neg logit contributions | |
Julia | -0.031 |
La | -0.028 |
Mae | -0.028 |
Amber | -0.027 |
Zac | -0.027 |
Token | use |
Token ID | 779 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
Julia | +0.205 |
mate | +0.195 |
dives | +0.192 |
actor | +0.192 |
Mae | +0.186 |
Neg logit contributions | |
swollen | -0.250 |
bruised | -0.242 |
bruises | -0.235 |
ados | -0.234 |
supervision | -0.229 |
Token | it |
Token ID | 340 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
� | +0.254 |
bruised | +0.239 |
swollen | +0.238 |
cut | +0.229 |
pling | +0.227 |
Neg logit contributions | |
Julia | -0.110 |
mate | -0.096 |
dolphins | -0.091 |
dives | -0.090 |
dances | -0.088 |
Token | to |
Token ID | 284 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
� | +0.036 |
bruised | +0.036 |
wrench | +0.035 |
swollen | +0.035 |
bruises | +0.034 |
Neg logit contributions | |
Julia | -0.036 |
La | -0.033 |
Mae | -0.033 |
Amber | -0.032 |
Zac | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +5.75e-03 |
Pos logit contributions | |
bruised | +0.176 |
� | +0.175 |
wrench | +0.174 |
swollen | +0.171 |
hammer | +0.171 |
Neg logit contributions | |
Julia | -0.178 |
La | -0.165 |
Mae | -0.161 |
Amber | -0.157 |
Zac | -0.152 |
Token | you |
Token ID | 345 |
Feature activation | -2.68e-02 |
Pos logit contributions | |
� | +0.010 |
bruised | +0.010 |
wrench | +0.010 |
swollen | +0.010 |
bruises | +0.010 |
Neg logit contributions | |
Julia | -0.009 |
La | -0.008 |
Mae | -0.008 |
Amber | -0.008 |
Zac | -0.008 |
Token | can |
Token ID | 460 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
Julia | +0.044 |
mate | +0.043 |
dives | +0.041 |
La | +0.040 |
actor | +0.040 |
Neg logit contributions | |
swollen | -0.043 |
bruised | -0.041 |
� | -0.040 |
bruises | -0.040 |
pling | -0.039 |
Token | hold |
Token ID | 1745 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
bruised | +0.259 |
swollen | +0.251 |
� | +0.251 |
wrench | +0.251 |
bruises | +0.248 |
Neg logit contributions | |
Julia | -0.331 |
La | -0.306 |
Mae | -0.302 |
Amber | -0.297 |
Zac | -0.289 |
Token | this |
Token ID | 428 |
Feature activation | +4.86e-02 |
Pos logit contributions | |
� | +0.194 |
swollen | +0.187 |
bruised | +0.183 |
scrape | +0.178 |
cut | +0.174 |
Neg logit contributions | |
Julia | -0.098 |
Amber | -0.085 |
Mae | -0.083 |
dolphins | -0.079 |
April | -0.076 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.49e-01 |
Pos logit contributions | |
� | +0.077 |
swollen | +0.069 |
uc | +0.068 |
bruises | +0.068 |
bruised | +0.065 |
Neg logit contributions | |
mate | -0.080 |
Julia | -0.078 |
princess | -0.074 |
royal | -0.074 |
aliens | -0.073 |
Token | and |
Token ID | 290 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
� | +0.817 |
bruised | +0.803 |
swollen | +0.802 |
bruises | +0.789 |
wrench | +0.783 |
Neg logit contributions | |
Julia | -0.676 |
Mae | -0.608 |
La | -0.603 |
Amber | -0.593 |
Zac | -0.572 |
Token | this |
Token ID | 428 |
Feature activation | +4.66e-02 |
Pos logit contributions | |
� | +0.035 |
swollen | +0.033 |
bruised | +0.032 |
pling | +0.032 |
uc | +0.032 |
Neg logit contributions | |
Julia | -0.024 |
mate | -0.023 |
dives | -0.023 |
La | -0.022 |
astronauts | -0.022 |
Token | hammer |
Token ID | 15554 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
� | +0.069 |
swollen | +0.069 |
bruised | +0.066 |
uc | +0.065 |
bruises | +0.065 |
Neg logit contributions | |
mate | -0.075 |
Julia | -0.073 |
dives | -0.069 |
pirate | -0.069 |
actor | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
� | +0.776 |
bruised | +0.759 |
swollen | +0.749 |
wrench | +0.746 |
bruises | +0.736 |
Neg logit contributions | |
Julia | -0.678 |
La | -0.613 |
Mae | -0.605 |
Amber | -0.588 |
Zac | -0.576 |
Token | And |
Token ID | 843 |
Feature activation | +9.28e-03 |
Pos logit contributions | |
swollen | +0.017 |
bruised | +0.017 |
� | +0.017 |
wrench | +0.017 |
hammer | +0.017 |
Neg logit contributions | |
Julia | -0.017 |
La | -0.016 |
Mae | -0.016 |
Amber | -0.015 |
April | -0.015 |
Token | angry |
Token ID | 7954 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
Julia | +0.019 |
La | +0.017 |
Mae | +0.017 |
Amber | +0.017 |
Zac | +0.017 |
Neg logit contributions | |
bruised | -0.014 |
swollen | -0.014 |
hammer | -0.013 |
bruises | -0.013 |
wrench | -0.013 |
Token | when |
Token ID | 618 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
bruised | +0.126 |
hammer | +0.118 |
� | +0.117 |
bruises | +0.114 |
wrench | +0.113 |
Neg logit contributions | |
Julia | -0.147 |
La | -0.146 |
Mae | -0.140 |
Amber | -0.138 |
April | -0.138 |
Token | she |
Token ID | 673 |
Feature activation | -3.48e-02 |
Pos logit contributions | |
bruised | +0.025 |
� | +0.024 |
swollen | +0.024 |
hammer | +0.024 |
bruises | +0.024 |
Neg logit contributions | |
Julia | -0.030 |
La | -0.029 |
Mae | -0.027 |
Zac | -0.027 |
mate | -0.027 |
Token | used |
Token ID | 973 |
Feature activation | +8.09e-02 |
Pos logit contributions | |
Julia | +0.061 |
Mae | +0.056 |
La | +0.056 |
Amber | +0.055 |
Zac | +0.054 |
Neg logit contributions | |
bruised | -0.056 |
� | -0.055 |
hammer | -0.053 |
bruises | -0.053 |
swollen | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
� | +0.150 |
wrench | +0.146 |
bruised | +0.143 |
pling | +0.142 |
swollen | +0.139 |
Neg logit contributions | |
Julia | -0.127 |
Mae | -0.110 |
Amber | -0.109 |
La | -0.109 |
Jeff | -0.106 |
Token | hammer |
Token ID | 15554 |
Feature activation | +4.47e-01 |
Pos logit contributions | |
Julia | +0.115 |
La | +0.105 |
Mae | +0.103 |
Amber | +0.103 |
Lou | +0.101 |
Neg logit contributions | |
� | -0.102 |
bruised | -0.099 |
leaning | -0.098 |
bruises | -0.097 |
swollen | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
� | +0.812 |
bruised | +0.801 |
swollen | +0.797 |
wrench | +0.792 |
bruises | +0.790 |
Neg logit contributions | |
Julia | -0.682 |
La | -0.607 |
Mae | -0.599 |
Amber | -0.586 |
Lou | -0.562 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
� | +0.016 |
bruised | +0.016 |
wrench | +0.016 |
swollen | +0.016 |
hammer | +0.016 |
Neg logit contributions | |
Julia | -0.018 |
La | -0.016 |
Mae | -0.016 |
Amber | -0.016 |
Zac | -0.015 |
Token | John |
Token ID | 7554 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
� | +0.071 |
bruised | +0.063 |
hammer | +0.062 |
wrench | +0.062 |
pling | +0.062 |
Neg logit contributions | |
Julia | -0.095 |
La | -0.089 |
Mae | -0.087 |
actor | -0.086 |
Amber | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | +1.82e-02 |
Pos logit contributions | |
Julia | +0.048 |
La | +0.044 |
Mae | +0.043 |
Amber | +0.042 |
Zac | +0.041 |
Neg logit contributions | |
bruised | -0.043 |
� | -0.043 |
wrench | -0.042 |
hammer | -0.042 |
swollen | -0.041 |
Token | his |
Token ID | 465 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
bruised | +0.024 |
hammer | +0.024 |
wrench | +0.023 |
pling | +0.023 |
rag | +0.022 |
Neg logit contributions | |
Julia | -0.035 |
La | -0.034 |
Mae | -0.032 |
ae | -0.032 |
pir | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +2.21e-02 |
Pos logit contributions | |
bruised | +0.097 |
� | +0.097 |
wrench | +0.095 |
swollen | +0.095 |
bruises | +0.094 |
Neg logit contributions | |
Julia | -0.098 |
La | -0.089 |
Mae | -0.088 |
Amber | -0.086 |
Zac | -0.083 |
Token | She |
Token ID | 1375 |
Feature activation | +7.43e-03 |
Pos logit contributions | |
� | +0.034 |
bruised | +0.034 |
wrench | +0.034 |
swollen | +0.033 |
hammer | +0.033 |
Neg logit contributions | |
Julia | -0.039 |
La | -0.037 |
Mae | -0.036 |
Amber | -0.035 |
Zac | -0.034 |
Token | is |
Token ID | 318 |
Feature activation | -3.99e-02 |
Pos logit contributions | |
bruised | +0.013 |
bruises | +0.013 |
wrench | +0.012 |
swollen | +0.012 |
hammer | +0.012 |
Neg logit contributions | |
Julia | -0.012 |
La | -0.011 |
Mae | -0.011 |
Amber | -0.011 |
Lou | -0.010 |
Token | using |
Token ID | 1262 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Julia | +0.062 |
La | +0.057 |
Mae | +0.055 |
Amber | +0.053 |
Jeff | +0.051 |
Neg logit contributions | |
� | -0.075 |
wrench | -0.070 |
pling | -0.068 |
leaning | -0.067 |
bruised | -0.066 |
Token | a |
Token ID | 257 |
Feature activation | -6.54e-02 |
Pos logit contributions | |
� | +0.230 |
pling | +0.218 |
leaning | +0.206 |
uc | +0.205 |
cut | +0.203 |
Neg logit contributions | |
Julia | -0.138 |
Mae | -0.117 |
dolphins | -0.112 |
Amber | -0.111 |
dances | -0.109 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.46e-01 |
Pos logit contributions | |
Julia | +0.132 |
La | +0.122 |
Mae | +0.122 |
dives | +0.121 |
Amber | +0.119 |
Neg logit contributions | |
� | -0.081 |
bruised | -0.079 |
bruises | -0.077 |
leaning | -0.076 |
pling | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
� | +0.933 |
swollen | +0.907 |
bruised | +0.885 |
bruises | +0.872 |
leaning | +0.850 |
Neg logit contributions | |
Julia | -0.566 |
Mae | -0.482 |
La | -0.471 |
Zac | -0.460 |
Jeff | -0.457 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-03 |
Pos logit contributions | |
bruised | +0.035 |
� | +0.035 |
wrench | +0.035 |
swollen | +0.034 |
hammer | +0.034 |
Neg logit contributions | |
Julia | -0.037 |
La | -0.035 |
Mae | -0.034 |
Amber | -0.033 |
Zac | -0.032 |
Token |
|
Token ID | 198 |
Feature activation | -1.47e-03 |
Pos logit contributions | |
Julia | +0.004 |
La | +0.003 |
Mae | +0.003 |
Amber | +0.003 |
Zac | +0.003 |
Neg logit contributions | |
bruised | -0.004 |
� | -0.004 |
wrench | -0.004 |
swollen | -0.004 |
hammer | -0.004 |
Token | " |
Token ID | 1 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
Julia | +0.002 |
La | +0.002 |
Mae | +0.002 |
Amber | +0.002 |
Zac | +0.002 |
Neg logit contributions | |
bruised | -0.003 |
� | -0.003 |
wrench | -0.003 |
hammer | -0.002 |
swollen | -0.002 |
Token | Mom |
Token ID | 29252 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
bruised | +0.027 |
� | +0.026 |
wrench | +0.025 |
hammer | +0.025 |
pling | +0.025 |
Neg logit contributions | |
La | -0.028 |
Julia | -0.027 |
Mae | -0.025 |
Amber | -0.024 |
Zac | -0.023 |
Token | man |
Token ID | 582 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
� | +0.130 |
bruised | +0.129 |
wrench | +0.128 |
swollen | +0.125 |
bruises | +0.124 |
Neg logit contributions | |
Julia | -0.115 |
La | -0.105 |
Mae | -0.103 |
Amber | -0.101 |
Zac | -0.096 |
Token | started |
Token ID | 2067 |
Feature activation | -6.58e-03 |
Pos logit contributions | |
bruised | +0.015 |
hammer | +0.015 |
bruises | +0.014 |
wrench | +0.014 |
� | +0.014 |
Neg logit contributions | |
Julia | -0.019 |
La | -0.019 |
Mae | -0.018 |
Amber | -0.018 |
April | -0.017 |
Token | to |
Token ID | 284 |
Feature activation | +1.56e-02 |
Pos logit contributions | |
Julia | +0.012 |
La | +0.011 |
Mae | +0.011 |
Amber | +0.010 |
Zac | +0.010 |
Neg logit contributions | |
bruised | -0.010 |
wrench | -0.010 |
� | -0.010 |
hammer | -0.010 |
swollen | -0.010 |
Token | use |
Token ID | 779 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
bruised | +0.025 |
hammer | +0.025 |
wrench | +0.025 |
swollen | +0.025 |
� | +0.024 |
Neg logit contributions | |
Julia | -0.027 |
La | -0.025 |
Mae | -0.024 |
Amber | -0.024 |
Zac | -0.023 |
Token | his |
Token ID | 465 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
� | +0.241 |
pling | +0.226 |
leaning | +0.226 |
uc | +0.221 |
bruised | +0.220 |
Neg logit contributions | |
Julia | -0.163 |
Jeff | -0.139 |
Lou | -0.138 |
Mae | -0.138 |
Amber | -0.134 |
Token | hammer |
Token ID | 15554 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
Julia | +0.074 |
La | +0.069 |
Mae | +0.068 |
Amber | +0.067 |
Lou | +0.065 |
Neg logit contributions | |
� | -0.060 |
bruised | -0.059 |
wrench | -0.057 |
pling | -0.057 |
swollen | -0.056 |
Token | to |
Token ID | 284 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
� | +0.939 |
swollen | +0.873 |
bruised | +0.873 |
bruises | +0.855 |
leaning | +0.855 |
Neg logit contributions | |
Julia | -0.567 |
Lou | -0.480 |
La | -0.469 |
Jeff | -0.461 |
Mae | -0.460 |
Token | fix |
Token ID | 4259 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
� | +0.022 |
pling | +0.021 |
bruised | +0.021 |
leaning | +0.021 |
swollen | +0.021 |
Neg logit contributions | |
Julia | -0.020 |
Lou | -0.018 |
dive | -0.017 |
Mae | -0.017 |
La | -0.017 |
Token | his |
Token ID | 465 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
� | +0.514 |
bruised | +0.478 |
swollen | +0.474 |
leaning | +0.467 |
bruises | +0.466 |
Neg logit contributions | |
Julia | -0.388 |
Mae | -0.333 |
Jeff | -0.329 |
Zac | -0.324 |
Lou | -0.320 |
Token | house |
Token ID | 2156 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
Julia | +0.074 |
ia | +0.071 |
mate | +0.071 |
royal | +0.070 |
pirate | +0.070 |
Neg logit contributions | |
� | -0.062 |
leaning | -0.056 |
Here | -0.055 |
pling | -0.055 |
bruised | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
� | +0.093 |
swollen | +0.091 |
leaning | +0.089 |
bruises | +0.089 |
bruised | +0.088 |
Neg logit contributions | |
Julia | -0.053 |
Mae | -0.044 |
Jeff | -0.043 |
Lou | -0.042 |
Zac | -0.041 |
Token | bottom |
Token ID | 4220 |
Feature activation | -5.41e-02 |
Pos logit contributions | |
bruised | +0.094 |
� | +0.093 |
swollen | +0.092 |
bruises | +0.091 |
wrench | +0.091 |
Neg logit contributions | |
Julia | -0.072 |
La | -0.065 |
Mae | -0.065 |
Amber | -0.063 |
Zac | -0.062 |
Token | of |
Token ID | 286 |
Feature activation | -5.17e-02 |
Pos logit contributions | |
Julia | +0.095 |
La | +0.092 |
Amber | +0.087 |
Mae | +0.086 |
Zac | +0.085 |
Neg logit contributions | |
� | -0.084 |
hammer | -0.084 |
wrench | -0.083 |
bruised | -0.083 |
bruises | -0.080 |
Token | the |
Token ID | 262 |
Feature activation | +4.92e-02 |
Pos logit contributions | |
Julia | +0.088 |
Mae | +0.079 |
Amber | +0.078 |
La | +0.076 |
Jeff | +0.075 |
Neg logit contributions | |
bruised | -0.089 |
� | -0.088 |
wrench | -0.087 |
pling | -0.086 |
swollen | -0.085 |
Token | hill |
Token ID | 12788 |
Feature activation | +7.78e-02 |
Pos logit contributions | |
bruised | +0.085 |
swollen | +0.083 |
uc | +0.082 |
bruises | +0.081 |
� | +0.081 |
Neg logit contributions | |
Julia | -0.081 |
Mae | -0.073 |
La | -0.073 |
Zac | -0.071 |
Jeff | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | +4.24e-03 |
Pos logit contributions | |
bruised | +0.129 |
� | +0.128 |
wrench | +0.128 |
swollen | +0.126 |
bruises | +0.124 |
Neg logit contributions | |
Julia | -0.131 |
La | -0.119 |
Mae | -0.118 |
Amber | -0.116 |
Zac | -0.112 |
Token | bruised |
Token ID | 44379 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
� | +0.007 |
wrench | +0.007 |
swollen | +0.007 |
bruised | +0.007 |
pling | +0.007 |
Neg logit contributions | |
Julia | -0.007 |
Mae | -0.007 |
Amber | -0.006 |
La | -0.006 |
Zac | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
� | +0.805 |
wrench | +0.785 |
bruised | +0.782 |
swollen | +0.778 |
hammer | +0.771 |
Neg logit contributions | |
Julia | -0.693 |
Mae | -0.625 |
La | -0.610 |
Amber | -0.609 |
Zac | -0.598 |
Token | bleeding |
Token ID | 16832 |
Feature activation | +3.20e-01 |
Pos logit contributions | |
� | +0.027 |
wrench | +0.024 |
leaning | +0.024 |
pling | +0.024 |
issors | +0.024 |
Neg logit contributions | |
Julia | -0.030 |
Mae | -0.028 |
La | -0.027 |
Amber | -0.027 |
Jeff | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.06e-02 |
Pos logit contributions | |
bruised | +0.510 |
� | +0.500 |
wrench | +0.498 |
swollen | +0.490 |
hammer | +0.486 |
Neg logit contributions | |
Julia | -0.570 |
La | -0.528 |
Mae | -0.516 |
Amber | -0.510 |
Zac | -0.488 |
Token | He |
Token ID | 679 |
Feature activation | -1.92e-02 |
Pos logit contributions | |
� | +0.017 |
bruised | +0.016 |
wrench | +0.016 |
bruises | +0.016 |
swollen | +0.015 |
Neg logit contributions | |
Julia | -0.018 |
La | -0.017 |
actor | -0.016 |
Mae | -0.016 |
oper | -0.016 |
Token | was |
Token ID | 373 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
Julia | +0.033 |
La | +0.030 |
Mae | +0.030 |
Amber | +0.029 |
Zac | +0.028 |
Neg logit contributions | |
bruised | -0.032 |
wrench | -0.031 |
� | -0.031 |
swollen | -0.031 |
bruises | -0.031 |
Token | you |
Token ID | 345 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
� | +0.046 |
bruised | +0.045 |
wrench | +0.045 |
swollen | +0.044 |
hammer | +0.044 |
Neg logit contributions | |
Julia | -0.035 |
La | -0.032 |
Mae | -0.031 |
Amber | -0.030 |
Zac | -0.029 |
Token | help |
Token ID | 1037 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
Julia | +0.058 |
La | +0.054 |
Lou | +0.052 |
mate | +0.052 |
Mae | +0.052 |
Neg logit contributions | |
bruised | -0.042 |
swollen | -0.041 |
bruises | -0.041 |
� | -0.040 |
mercury | -0.040 |
Token | me |
Token ID | 502 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
hammer | +0.480 |
wrench | +0.477 |
bruised | +0.475 |
� | +0.473 |
swollen | +0.466 |
Neg logit contributions | |
Julia | -0.342 |
La | -0.310 |
Mae | -0.301 |
Amber | -0.294 |
Zac | -0.290 |
Token | find |
Token ID | 1064 |
Feature activation | -7.00e-02 |
Pos logit contributions | |
Julia | +0.198 |
La | +0.183 |
Mae | +0.177 |
Amber | +0.175 |
Lou | +0.170 |
Neg logit contributions | |
bruised | -0.208 |
� | -0.208 |
swollen | -0.206 |
bruises | -0.205 |
wrench | -0.200 |
Token | the |
Token ID | 262 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
Julia | +0.110 |
La | +0.101 |
Mae | +0.099 |
Amber | +0.097 |
Zac | +0.093 |
Neg logit contributions | |
� | -0.124 |
bruised | -0.124 |
wrench | -0.122 |
swollen | -0.121 |
hammer | -0.119 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
� | +0.105 |
bruised | +0.102 |
wrench | +0.101 |
bruises | +0.100 |
swollen | +0.100 |
Neg logit contributions | |
Julia | -0.087 |
La | -0.081 |
Mae | -0.078 |
Amber | -0.077 |
April | -0.076 |
Token | ?". |
Token ID | 43634 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
bruised | +0.861 |
� | +0.858 |
swollen | +0.848 |
wrench | +0.845 |
bruises | +0.842 |
Neg logit contributions | |
Julia | -0.620 |
La | -0.556 |
Mae | -0.548 |
Amber | -0.539 |
Zac | -0.511 |
Token |
|
Token ID | 198 |
Feature activation | -2.85e-03 |
Pos logit contributions | |
Julia | +0.179 |
La | +0.153 |
Amber | +0.151 |
Mae | +0.150 |
Lou | +0.150 |
Neg logit contributions | |
bruised | -0.213 |
swollen | -0.210 |
wrench | -0.203 |
bruises | -0.201 |
mercury | -0.200 |
Token |
|
Token ID | 198 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
Julia | +0.005 |
La | +0.004 |
Mae | +0.004 |
Amber | +0.004 |
Zac | +0.004 |
Neg logit contributions | |
� | -0.005 |
bruised | -0.005 |
wrench | -0.005 |
swollen | -0.005 |
hammer | -0.005 |
Token | Sarah |
Token ID | 29284 |
Feature activation | -9.62e-03 |
Pos logit contributions | |
Julia | +0.004 |
La | +0.004 |
Mae | +0.004 |
Amber | +0.004 |
Lou | +0.004 |
Neg logit contributions | |
bruised | -0.004 |
� | -0.004 |
wrench | -0.004 |
swollen | -0.004 |
bruises | -0.004 |
Token | happily |
Token ID | 18177 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
Julia | +0.016 |
La | +0.014 |
Mae | +0.014 |
Amber | +0.014 |
Zac | +0.013 |
Neg logit contributions | |
� | -0.016 |
bruised | -0.016 |
wrench | -0.016 |
swollen | -0.016 |
hammer | -0.016 |
Token | them |
Token ID | 606 |
Feature activation | -7.39e-02 |
Pos logit contributions | |
bruised | +0.113 |
wrench | +0.113 |
hammer | +0.112 |
oil | +0.110 |
bruises | +0.109 |
Neg logit contributions | |
Julia | -0.113 |
La | -0.110 |
Mae | -0.105 |
Amber | -0.102 |
ae | -0.101 |
Token | a |
Token ID | 257 |
Feature activation | -6.62e-02 |
Pos logit contributions | |
Julia | +0.110 |
La | +0.099 |
Mae | +0.098 |
Amber | +0.097 |
dolphins | +0.092 |
Neg logit contributions | |
� | -0.141 |
bruised | -0.138 |
swollen | -0.136 |
wrench | -0.135 |
hammer | -0.130 |
Token | jack |
Token ID | 14509 |
Feature activation | -7.74e-03 |
Pos logit contributions | |
Julia | +0.093 |
La | +0.088 |
Mae | +0.083 |
dives | +0.083 |
Amber | +0.083 |
Neg logit contributions | |
� | -0.120 |
leaning | -0.116 |
bruised | -0.115 |
supervision | -0.113 |
bruises | -0.111 |
Token | and |
Token ID | 290 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
Julia | +0.012 |
La | +0.011 |
Mae | +0.011 |
Amber | +0.010 |
mate | +0.010 |
Neg logit contributions | |
� | -0.014 |
bruised | -0.014 |
swollen | -0.014 |
wrench | -0.013 |
bruises | -0.013 |
Token | a |
Token ID | 257 |
Feature activation | -6.59e-02 |
Pos logit contributions | |
� | +0.026 |
bruised | +0.025 |
leaning | +0.024 |
mercury | +0.023 |
swollen | +0.023 |
Neg logit contributions | |
Julia | -0.022 |
La | -0.020 |
Amber | -0.020 |
Mae | -0.020 |
April | -0.019 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
Julia | +0.095 |
mate | +0.092 |
La | +0.091 |
royal | +0.090 |
princess | +0.089 |
Neg logit contributions | |
leaning | -0.113 |
� | -0.108 |
ised | -0.106 |
issors | -0.104 |
mons | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
� | +0.894 |
swollen | +0.877 |
bruised | +0.863 |
leaning | +0.845 |
bruises | +0.838 |
Neg logit contributions | |
Julia | -0.563 |
Zac | -0.488 |
Mae | -0.483 |
mate | -0.481 |
Jeff | -0.476 |
Token | They |
Token ID | 1119 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
bruised | +0.019 |
� | +0.019 |
wrench | +0.019 |
swollen | +0.019 |
hammer | +0.018 |
Neg logit contributions | |
Julia | -0.017 |
La | -0.016 |
Mae | -0.015 |
Amber | -0.015 |
Zac | -0.015 |
Token | helped |
Token ID | 4193 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
Julia | +0.036 |
Mae | +0.032 |
La | +0.031 |
mate | +0.031 |
Ro | +0.031 |
Neg logit contributions | |
� | -0.046 |
bruised | -0.046 |
swollen | -0.045 |
supervision | -0.044 |
mercury | -0.043 |
Token | him |
Token ID | 683 |
Feature activation | +4.22e-02 |
Pos logit contributions | |
bruised | +0.263 |
� | +0.262 |
wrench | +0.261 |
swollen | +0.256 |
hammer | +0.255 |
Neg logit contributions | |
Julia | -0.219 |
La | -0.200 |
Mae | -0.195 |
Amber | -0.192 |
Zac | -0.183 |
Token | lift |
Token ID | 10303 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
swollen | +0.067 |
� | +0.064 |
bruised | +0.064 |
leaning | +0.062 |
bruises | +0.061 |
Neg logit contributions | |
Julia | -0.067 |
dives | -0.061 |
dive | -0.060 |
La | -0.059 |
dance | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
� | +0.234 |
leaning | +0.217 |
mons | +0.210 |
pling | +0.209 |
uc | +0.208 |
Neg logit contributions | |
Julia | -0.140 |
Mae | -0.117 |
Amber | -0.114 |
April | -0.112 |
Jeff | -0.109 |
Token | screw |
Token ID | 9580 |
Feature activation | +2.97e-01 |
Pos logit contributions | |
Julia | +0.108 |
mate | +0.103 |
Mae | +0.100 |
royal | +0.100 |
pir | +0.099 |
Neg logit contributions | |
leaning | -0.101 |
bruises | -0.101 |
� | -0.100 |
mons | -0.098 |
bruised | -0.097 |
Token | driver |
Token ID | 26230 |
Feature activation | +2.74e-01 |
Pos logit contributions | |
� | +0.458 |
swollen | +0.441 |
leaning | +0.412 |
bruised | +0.405 |
bruises | +0.400 |
Neg logit contributions | |
Julia | -0.532 |
mate | -0.482 |
Mae | -0.468 |
Zac | -0.463 |
Lou | -0.459 |
Token | , |
Token ID | 11 |
Feature activation | +1.51e-03 |
Pos logit contributions | |
� | +0.665 |
swollen | +0.633 |
leaning | +0.620 |
uc | +0.611 |
bruises | +0.611 |
Neg logit contributions | |
Julia | -0.237 |
dragons | -0.193 |
Mae | -0.192 |
April | -0.182 |
mate | -0.180 |
Token | a |
Token ID | 257 |
Feature activation | -6.51e-02 |
Pos logit contributions | |
� | +0.003 |
uc | +0.003 |
pling | +0.003 |
leaning | +0.003 |
wrench | +0.003 |
Neg logit contributions | |
Julia | -0.002 |
Mae | -0.002 |
La | -0.002 |
Amber | -0.002 |
Lou | -0.002 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
Julia | +0.130 |
Mae | +0.120 |
mate | +0.118 |
La | +0.118 |
pir | +0.118 |
Neg logit contributions | |
� | -0.083 |
leaning | -0.083 |
mons | -0.076 |
bruises | -0.075 |
bruised | -0.074 |
Token | , |
Token ID | 11 |
Feature activation | +1.39e-03 |
Pos logit contributions | |
� | +1.031 |
swollen | +1.011 |
leaning | +0.992 |
bruises | +0.980 |
moderation | +0.973 |
Neg logit contributions | |
Julia | -0.384 |
Ro | -0.318 |
Zac | -0.316 |
mate | -0.309 |
Jeff | -0.298 |
Token | and |
Token ID | 290 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
� | +0.003 |
leaning | +0.003 |
uc | +0.003 |
pling | +0.003 |
swollen | +0.003 |
Neg logit contributions | |
Julia | -0.001 |
Mae | -0.001 |
La | -0.001 |
April | -0.001 |
Lou | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | -6.50e-02 |
Pos logit contributions | |
� | +0.029 |
bruised | +0.028 |
wrench | +0.027 |
swollen | +0.027 |
pling | +0.027 |
Neg logit contributions | |
Julia | -0.022 |
Mae | -0.020 |
La | -0.020 |
Amber | -0.019 |
Lou | -0.018 |
Token | flashlight |
Token ID | 38371 |
Feature activation | +4.76e-02 |
Pos logit contributions | |
Julia | +0.138 |
mate | +0.129 |
pirate | +0.128 |
princess | +0.127 |
pir | +0.126 |
Neg logit contributions | |
leaning | -0.075 |
� | -0.074 |
mons | -0.067 |
issors | -0.063 |
uc | -0.063 |
Token | . |
Token ID | 13 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
� | +0.088 |
swollen | +0.086 |
bruised | +0.085 |
bruises | +0.084 |
pling | +0.083 |
Neg logit contributions | |
Julia | -0.069 |
Mae | -0.060 |
La | -0.060 |
Zac | -0.058 |
Jeff | -0.058 |
Token | it |
Token ID | 340 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
� | +0.018 |
bruised | +0.018 |
swollen | +0.018 |
wrench | +0.017 |
bruises | +0.017 |
Neg logit contributions | |
Julia | -0.016 |
Mae | -0.015 |
La | -0.015 |
Amber | -0.014 |
Jeff | -0.014 |
Token | was |
Token ID | 373 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
bruised | +0.035 |
wrench | +0.035 |
� | +0.035 |
hammer | +0.034 |
swollen | +0.034 |
Neg logit contributions | |
Julia | -0.031 |
La | -0.028 |
Mae | -0.028 |
Amber | -0.027 |
Zac | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -5.36e-02 |
Pos logit contributions | |
Julia | +0.050 |
Mae | +0.044 |
Amber | +0.042 |
mate | +0.042 |
La | +0.042 |
Neg logit contributions | |
� | -0.089 |
leaning | -0.083 |
bruises | -0.082 |
bruised | -0.082 |
pling | -0.081 |
Token | very |
Token ID | 845 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
Julia | +0.088 |
La | +0.081 |
actor | +0.080 |
Mae | +0.080 |
mate | +0.078 |
Neg logit contributions | |
� | -0.086 |
bruised | -0.084 |
bruises | -0.084 |
pling | -0.081 |
swollen | -0.081 |
Token | special |
Token ID | 2041 |
Feature activation | -1.18e-02 |
Pos logit contributions | |
Julia | +0.046 |
La | +0.041 |
Mae | +0.041 |
Amber | +0.040 |
actor | +0.039 |
Neg logit contributions | |
� | -0.046 |
bruised | -0.046 |
bruises | -0.044 |
swollen | -0.044 |
wrench | -0.044 |
Token | hammer |
Token ID | 15554 |
Feature activation | +4.45e-01 |
Pos logit contributions | |
Julia | +0.024 |
actor | +0.023 |
mate | +0.022 |
pirate | +0.022 |
Lou | +0.022 |
Neg logit contributions | |
� | -0.013 |
swollen | -0.013 |
bruised | -0.012 |
bruises | -0.012 |
arl | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
� | +0.846 |
swollen | +0.818 |
bruises | +0.816 |
bruised | +0.812 |
leaning | +0.781 |
Neg logit contributions | |
Julia | -0.613 |
Mae | -0.539 |
La | -0.523 |
actor | -0.519 |
Jeff | -0.516 |
Token | Thomas |
Token ID | 5658 |
Feature activation | -1.11e-01 |
Pos logit contributions | |
� | +0.023 |
wrench | +0.022 |
bruised | +0.022 |
hammer | +0.021 |
swollen | +0.021 |
Neg logit contributions | |
Julia | -0.027 |
La | -0.026 |
Mae | -0.025 |
Amber | -0.024 |
dolphins | -0.024 |
Token | noticed |
Token ID | 6810 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
Julia | +0.184 |
La | +0.171 |
Mae | +0.165 |
Amber | +0.164 |
Zac | +0.157 |
Neg logit contributions | |
� | -0.184 |
bruised | -0.184 |
wrench | -0.183 |
hammer | -0.180 |
swollen | -0.179 |
Token | that |
Token ID | 326 |
Feature activation | +1.42e-01 |
Pos logit contributions | |
Julia | +0.120 |
La | +0.107 |
Mae | +0.106 |
Amber | +0.105 |
dolphins | +0.102 |
Neg logit contributions | |
� | -0.123 |
bruised | -0.121 |
swollen | -0.117 |
wrench | -0.117 |
bruises | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
� | +0.278 |
bruised | +0.268 |
swollen | +0.263 |
bruises | +0.262 |
wrench | +0.259 |
Neg logit contributions | |
Julia | -0.209 |
Mae | -0.185 |
La | -0.182 |
Amber | -0.180 |
Zac | -0.173 |
Token | hit |
Token ID | 2277 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
� | +0.096 |
bruised | +0.092 |
wrench | +0.090 |
swollen | +0.090 |
bruises | +0.087 |
Neg logit contributions | |
Julia | -0.132 |
La | -0.123 |
Mae | -0.121 |
Amber | -0.117 |
Lou | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
� | +0.160 |
wrench | +0.152 |
bruised | +0.150 |
swollen | +0.149 |
pling | +0.148 |
Neg logit contributions | |
Julia | -0.136 |
Mae | -0.120 |
La | -0.120 |
Amber | -0.116 |
Jeff | -0.116 |
Token | bird |
Token ID | 6512 |
Feature activation | -4.00e-02 |
Pos logit contributions | |
� | +0.096 |
leaning | +0.091 |
bruised | +0.087 |
swollen | +0.087 |
uc | +0.087 |
Neg logit contributions | |
Julia | -0.084 |
dolphins | -0.077 |
princess | -0.077 |
La | -0.076 |
Mae | -0.076 |
Token | with |
Token ID | 351 |
Feature activation | -2.75e-02 |
Pos logit contributions | |
Julia | +0.061 |
La | +0.055 |
Mae | +0.054 |
Amber | +0.053 |
Lou | +0.051 |
Neg logit contributions | |
� | -0.073 |
bruised | -0.072 |
wrench | -0.072 |
swollen | -0.070 |
bruises | -0.070 |
Token | his |
Token ID | 465 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
Julia | +0.026 |
Lou | +0.023 |
Jeff | +0.022 |
Mae | +0.022 |
Leo | +0.021 |
Neg logit contributions | |
� | -0.067 |
leaning | -0.064 |
pling | -0.062 |
uc | -0.062 |
mons | -0.061 |
Token | hammer |
Token ID | 15554 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
mate | +0.083 |
Julia | +0.081 |
ia | +0.081 |
Lou | +0.080 |
royal | +0.080 |
Neg logit contributions | |
leaning | -0.052 |
Here | -0.046 |
� | -0.045 |
checkout | -0.042 |
pling | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
� | +0.921 |
leaning | +0.888 |
bruises | +0.865 |
wrench | +0.859 |
bruised | +0.855 |
Neg logit contributions | |
Julia | -0.522 |
La | -0.457 |
Mae | -0.441 |
Lou | -0.435 |
Jeff | -0.428 |
Token | The |
Token ID | 383 |
Feature activation | +8.90e-02 |
Pos logit contributions | |
bruised | +0.026 |
swollen | +0.026 |
wrench | +0.026 |
bruises | +0.025 |
pling | +0.025 |
Neg logit contributions | |
Julia | -0.028 |
Mae | -0.026 |
La | -0.025 |
Amber | -0.025 |
Jeff | -0.025 |
Token | bird |
Token ID | 6512 |
Feature activation | -4.06e-02 |
Pos logit contributions | |
� | +0.184 |
bruised | +0.170 |
leaning | +0.170 |
wrench | +0.169 |
issors | +0.169 |
Neg logit contributions | |
Julia | -0.113 |
La | -0.100 |
dolphins | -0.098 |
princess | -0.097 |
Mae | -0.097 |
Token | could |
Token ID | 714 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
Julia | +0.060 |
La | +0.055 |
Mae | +0.054 |
Lou | +0.052 |
Amber | +0.051 |
Neg logit contributions | |
� | -0.076 |
wrench | -0.074 |
bruised | -0.072 |
bruises | -0.070 |
swollen | -0.070 |
Token | not |
Token ID | 407 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
bruised | +0.026 |
� | +0.026 |
wrench | +0.026 |
pling | +0.025 |
swollen | +0.025 |
Neg logit contributions | |
Julia | -0.027 |
La | -0.025 |
Mae | -0.024 |
Amber | -0.024 |
Jeff | -0.023 |
Token | home |
Token ID | 1363 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
Julia | +0.140 |
La | +0.124 |
Mae | +0.123 |
Amber | +0.123 |
Zac | +0.118 |
Neg logit contributions | |
� | -0.147 |
bruised | -0.145 |
wrench | -0.145 |
swollen | -0.144 |
bruises | -0.141 |
Token | , |
Token ID | 11 |
Feature activation | +2.25e-03 |
Pos logit contributions | |
� | +0.023 |
bruised | +0.023 |
wrench | +0.023 |
swollen | +0.022 |
hammer | +0.022 |
Neg logit contributions | |
Julia | -0.023 |
Mae | -0.020 |
La | -0.020 |
Amber | -0.020 |
Zac | -0.019 |
Token | but |
Token ID | 475 |
Feature activation | +7.69e-02 |
Pos logit contributions | |
� | +0.004 |
bruised | +0.004 |
wrench | +0.004 |
swollen | +0.004 |
bruises | +0.004 |
Neg logit contributions | |
Julia | -0.004 |
La | -0.003 |
Mae | -0.003 |
Amber | -0.003 |
Zac | -0.003 |
Token | he |
Token ID | 339 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
bruised | +0.130 |
wrench | +0.128 |
� | +0.128 |
swollen | +0.126 |
hammer | +0.125 |
Neg logit contributions | |
Julia | -0.126 |
La | -0.119 |
Mae | -0.115 |
Amber | -0.112 |
April | -0.107 |
Token | was |
Token ID | 373 |
Feature activation | -4.52e-02 |
Pos logit contributions | |
Julia | +0.112 |
La | +0.109 |
Mae | +0.103 |
Amber | +0.102 |
April | +0.098 |
Neg logit contributions | |
bruised | -0.092 |
� | -0.089 |
swollen | -0.088 |
hammer | -0.088 |
wrench | -0.088 |
Token | bruised |
Token ID | 44379 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
Julia | +0.085 |
La | +0.080 |
Mae | +0.077 |
Amber | +0.076 |
Lou | +0.074 |
Neg logit contributions | |
bruised | -0.069 |
swollen | -0.066 |
� | -0.065 |
wrench | -0.065 |
bruises | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
� | +0.807 |
wrench | +0.783 |
hammer | +0.774 |
bruises | +0.772 |
swollen | +0.768 |
Neg logit contributions | |
Julia | -0.677 |
Mae | -0.604 |
La | -0.595 |
Amber | -0.593 |
Zac | -0.570 |
Token | battered |
Token ID | 33613 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
� | +0.027 |
leaning | +0.026 |
issors | +0.025 |
wrench | +0.025 |
mons | +0.025 |
Neg logit contributions | |
Julia | -0.024 |
Mae | -0.022 |
Jeff | -0.021 |
Lou | -0.021 |
Amber | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
� | +0.467 |
wrench | +0.450 |
swollen | +0.448 |
bruised | +0.443 |
hammer | +0.441 |
Neg logit contributions | |
Julia | -0.413 |
Mae | -0.365 |
Amber | -0.360 |
La | -0.360 |
Jeff | -0.347 |
Token | His |
Token ID | 2399 |
Feature activation | -2.57e-02 |
Pos logit contributions | |
bruised | +0.025 |
swollen | +0.024 |
wrench | +0.024 |
� | +0.024 |
bruises | +0.024 |
Neg logit contributions | |
Julia | -0.023 |
La | -0.020 |
Mae | -0.020 |
Amber | -0.020 |
Zac | -0.019 |
Token | mummy |
Token ID | 38567 |
Feature activation | -1.93e-02 |
Pos logit contributions | |
Julia | +0.043 |
La | +0.039 |
Mae | +0.038 |
Amber | +0.038 |
dolphins | +0.037 |
Neg logit contributions | |
� | -0.042 |
bruised | -0.042 |
wrench | -0.041 |
pling | -0.040 |
swollen | -0.040 |
Token | and |
Token ID | 290 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
Julia | +0.067 |
La | +0.062 |
Mae | +0.060 |
Amber | +0.060 |
Zac | +0.057 |
Neg logit contributions | |
bruised | -0.059 |
� | -0.059 |
wrench | -0.059 |
swollen | -0.058 |
hammer | -0.057 |
Token | kept |
Token ID | 4030 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
bruised | +0.039 |
wrench | +0.038 |
hammer | +0.038 |
swollen | +0.038 |
bruises | +0.038 |
Neg logit contributions | |
Julia | -0.041 |
La | -0.039 |
Mae | -0.037 |
Amber | -0.036 |
April | -0.035 |
Token | playing |
Token ID | 2712 |
Feature activation | +4.82e-03 |
Pos logit contributions | |
� | +0.191 |
wrench | +0.179 |
bruised | +0.178 |
leaning | +0.177 |
pling | +0.176 |
Neg logit contributions | |
Julia | -0.154 |
La | -0.138 |
Mae | -0.135 |
Amber | -0.133 |
Jeff | -0.132 |
Token | with |
Token ID | 351 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
� | +0.009 |
bruised | +0.009 |
swollen | +0.008 |
wrench | +0.008 |
bruises | +0.008 |
Neg logit contributions | |
Julia | -0.007 |
La | -0.007 |
Mae | -0.007 |
Amber | -0.006 |
Jeff | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
Julia | +0.049 |
La | +0.046 |
Mae | +0.044 |
Amber | +0.044 |
Zac | +0.042 |
Neg logit contributions | |
bruised | -0.044 |
� | -0.044 |
wrench | -0.044 |
swollen | -0.043 |
hammer | -0.043 |
Token | hammer |
Token ID | 15554 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
� | +0.103 |
bruised | +0.095 |
pling | +0.094 |
swollen | +0.093 |
wrench | +0.093 |
Neg logit contributions | |
Julia | -0.091 |
dolphins | -0.083 |
La | -0.083 |
actor | -0.082 |
Mae | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-02 |
Pos logit contributions | |
� | +0.784 |
bruised | +0.765 |
swollen | +0.754 |
wrench | +0.754 |
bruises | +0.751 |
Neg logit contributions | |
Julia | -0.699 |
La | -0.624 |
Mae | -0.623 |
Amber | -0.607 |
Zac | -0.592 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
� | +0.026 |
bruised | +0.025 |
wrench | +0.025 |
hammer | +0.024 |
bruises | +0.024 |
Neg logit contributions | |
Julia | -0.033 |
La | -0.032 |
Mae | -0.030 |
Amber | -0.030 |
dolphins | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +7.87e-03 |
Pos logit contributions | |
La | +0.096 |
Julia | +0.095 |
Mae | +0.088 |
actor | +0.087 |
Amber | +0.087 |
Neg logit contributions | |
wrench | -0.065 |
hammer | -0.065 |
bruised | -0.063 |
bruises | -0.060 |
nails | -0.060 |
Token | he |
Token ID | 339 |
Feature activation | -5.84e-02 |
Pos logit contributions | |
wrench | +0.013 |
bruised | +0.013 |
hammer | +0.012 |
� | +0.012 |
bruises | +0.012 |
Neg logit contributions | |
Julia | -0.013 |
La | -0.013 |
Mae | -0.012 |
Amber | -0.012 |
actor | -0.011 |
Token | hit |
Token ID | 2277 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
Julia | +0.111 |
Mae | +0.104 |
La | +0.103 |
Amber | +0.102 |
oper | +0.101 |
Neg logit contributions | |
hammer | -0.086 |
swollen | -0.084 |
bruised | -0.084 |
wrench | -0.083 |
bruises | -0.082 |
Token | , |
Token ID | 11 |
Feature activation | +7.89e-03 |
Pos logit contributions | |
Julia | +0.151 |
La | +0.146 |
Mae | +0.138 |
Amber | +0.136 |
Zac | +0.131 |
Neg logit contributions | |
bruised | -0.126 |
wrench | -0.125 |
hammer | -0.123 |
� | -0.122 |
swollen | -0.120 |
Token | " |
Token ID | 366 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
bruised | +0.011 |
wrench | +0.011 |
� | +0.011 |
hammer | +0.011 |
swollen | +0.011 |
Neg logit contributions | |
Julia | -0.015 |
La | -0.014 |
Mae | -0.014 |
Amber | -0.014 |
Zac | -0.013 |
Token | I |
Token ID | 40 |
Feature activation | +6.05e-02 |
Pos logit contributions | |
� | +0.074 |
bruised | +0.074 |
wrench | +0.073 |
swollen | +0.071 |
hammer | +0.071 |
Neg logit contributions | |
Julia | -0.094 |
La | -0.088 |
Mae | -0.086 |
Amber | -0.084 |
Zac | -0.081 |
Token | miss |
Token ID | 2051 |
Feature activation | -4.92e-03 |
Pos logit contributions | |
� | +0.090 |
bruised | +0.089 |
wrench | +0.089 |
swollen | +0.088 |
bruises | +0.087 |
Neg logit contributions | |
Julia | -0.112 |
La | -0.104 |
Mae | -0.102 |
Amber | -0.100 |
Zac | -0.097 |
Token | my |
Token ID | 616 |
Feature activation | -7.85e-02 |
Pos logit contributions | |
Julia | +0.007 |
La | +0.006 |
Mae | +0.006 |
Amber | +0.006 |
Jeff | +0.006 |
Neg logit contributions | |
� | -0.009 |
bruised | -0.009 |
wrench | -0.009 |
swollen | -0.009 |
bruises | -0.009 |
Token | hammer |
Token ID | 15554 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
princess | +0.157 |
mate | +0.149 |
pirate | +0.147 |
king | +0.146 |
royal | +0.145 |
Neg logit contributions | |
� | -0.086 |
leaning | -0.081 |
arl | -0.077 |
swollen | -0.076 |
bruised | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
bruised | +0.736 |
� | +0.728 |
wrench | +0.726 |
swollen | +0.712 |
hammer | +0.711 |
Neg logit contributions | |
Julia | -0.754 |
La | -0.696 |
Mae | -0.681 |
Amber | -0.671 |
Zac | -0.643 |
Token | Can |
Token ID | 1680 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
bruised | +0.035 |
� | +0.034 |
wrench | +0.034 |
hammer | +0.034 |
swollen | +0.033 |
Neg logit contributions | |
Julia | -0.033 |
La | -0.030 |
Mae | -0.029 |
Amber | -0.029 |
Zac | -0.028 |
Token | you |
Token ID | 345 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
bruised | +0.063 |
� | +0.063 |
wrench | +0.063 |
swollen | +0.062 |
hammer | +0.062 |
Neg logit contributions | |
Julia | -0.042 |
La | -0.038 |
Mae | -0.037 |
Amber | -0.036 |
Zac | -0.034 |
Token | help |
Token ID | 1037 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
Julia | +0.047 |
La | +0.045 |
Mae | +0.043 |
Amber | +0.043 |
actor | +0.042 |
Neg logit contributions | |
bruised | -0.031 |
� | -0.030 |
swollen | -0.030 |
bruises | -0.029 |
wrench | -0.029 |
Token | me |
Token ID | 502 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
hammer | +0.474 |
wrench | +0.469 |
bruised | +0.465 |
swollen | +0.455 |
bruises | +0.451 |
Neg logit contributions | |
Julia | -0.384 |
La | -0.349 |
Mae | -0.340 |
Amber | -0.335 |
Zac | -0.325 |
Token | car |
Token ID | 1097 |
Feature activation | +6.43e-03 |
Pos logit contributions | |
� | +0.085 |
uc | +0.081 |
pling | +0.080 |
swollen | +0.079 |
bruised | +0.077 |
Neg logit contributions | |
Julia | -0.064 |
La | -0.060 |
dolphins | -0.058 |
pirate | -0.056 |
actor | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | +8.41e-03 |
Pos logit contributions | |
bruised | +0.010 |
wrench | +0.010 |
� | +0.010 |
hammer | +0.009 |
bruises | +0.009 |
Neg logit contributions | |
Julia | -0.012 |
La | -0.011 |
Mae | -0.011 |
Amber | -0.011 |
Zac | -0.010 |
Token | He |
Token ID | 679 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
bruised | +0.016 |
swollen | +0.015 |
wrench | +0.015 |
� | +0.015 |
pling | +0.015 |
Neg logit contributions | |
Julia | -0.013 |
Mae | -0.011 |
La | -0.011 |
Amber | -0.011 |
Jeff | -0.011 |
Token | had |
Token ID | 550 |
Feature activation | +4.29e-03 |
Pos logit contributions | |
Julia | +0.045 |
La | +0.042 |
Mae | +0.041 |
Amber | +0.041 |
Lou | +0.039 |
Neg logit contributions | |
bruised | -0.040 |
wrench | -0.038 |
bruises | -0.038 |
swollen | -0.038 |
hammer | -0.038 |
Token | a |
Token ID | 257 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
bruised | +0.007 |
wrench | +0.007 |
bruises | +0.007 |
swollen | +0.007 |
hammer | +0.007 |
Neg logit contributions | |
Julia | -0.007 |
La | -0.007 |
Mae | -0.006 |
Amber | -0.006 |
Zac | -0.006 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
Julia | +0.106 |
La | +0.101 |
mate | +0.096 |
Mae | +0.093 |
Amber | +0.093 |
Neg logit contributions | |
� | -0.114 |
pling | -0.111 |
leaning | -0.109 |
bruised | -0.109 |
uc | -0.108 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
� | +0.860 |
swollen | +0.820 |
bruised | +0.804 |
leaning | +0.798 |
bruises | +0.797 |
Neg logit contributions | |
Julia | -0.599 |
La | -0.538 |
Zac | -0.529 |
Ro | -0.518 |
Jeff | -0.513 |
Token | a |
Token ID | 257 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
� | +0.024 |
bruised | +0.024 |
wrench | +0.023 |
swollen | +0.023 |
hammer | +0.023 |
Neg logit contributions | |
Julia | -0.018 |
La | -0.017 |
Mae | -0.016 |
Amber | -0.016 |
Zac | -0.015 |
Token | jack |
Token ID | 14509 |
Feature activation | -8.62e-03 |
Pos logit contributions | |
Julia | +0.120 |
mate | +0.120 |
La | +0.117 |
pirate | +0.117 |
princess | +0.111 |
Neg logit contributions | |
� | -0.095 |
leaning | -0.094 |
pling | -0.088 |
mons | -0.087 |
uc | -0.086 |
Token | . |
Token ID | 13 |
Feature activation | +9.20e-03 |
Pos logit contributions | |
Julia | +0.015 |
La | +0.014 |
Mae | +0.013 |
Amber | +0.013 |
Zac | +0.013 |
Neg logit contributions | |
bruised | -0.014 |
wrench | -0.014 |
� | -0.014 |
hammer | -0.014 |
swollen | -0.014 |
Token | He |
Token ID | 679 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
swollen | +0.019 |
bruised | +0.019 |
bruises | +0.018 |
wrench | +0.018 |
pling | +0.018 |
Neg logit contributions | |
Julia | -0.013 |
Mae | -0.011 |
Jeff | -0.011 |
Amber | -0.010 |
Lou | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +7.88e-04 |
Pos logit contributions | |
bruised | +0.243 |
swollen | +0.238 |
wrench | +0.237 |
� | +0.237 |
bruises | +0.233 |
Neg logit contributions | |
Julia | -0.219 |
Mae | -0.192 |
La | -0.191 |
Amber | -0.190 |
Lou | -0.183 |
Token | she |
Token ID | 673 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
bruised | +0.002 |
swollen | +0.002 |
wrench | +0.002 |
mercury | +0.001 |
� | +0.001 |
Neg logit contributions | |
Julia | -0.001 |
Amber | -0.001 |
Mae | -0.001 |
Lou | -0.001 |
La | -0.001 |
Token | found |
Token ID | 1043 |
Feature activation | -4.77e-02 |
Pos logit contributions | |
Julia | +0.078 |
La | +0.075 |
Mae | +0.072 |
Amber | +0.071 |
Zac | +0.069 |
Neg logit contributions | |
hammer | -0.069 |
wrench | -0.068 |
� | -0.068 |
bruised | -0.068 |
swollen | -0.066 |
Token | a |
Token ID | 257 |
Feature activation | -6.70e-02 |
Pos logit contributions | |
Julia | +0.078 |
Mae | +0.069 |
Amber | +0.068 |
La | +0.067 |
Lou | +0.066 |
Neg logit contributions | |
� | -0.086 |
bruised | -0.084 |
wrench | -0.084 |
swollen | -0.082 |
bruises | -0.081 |
Token | shiny |
Token ID | 22441 |
Feature activation | +6.23e-02 |
Pos logit contributions | |
Julia | +0.112 |
La | +0.103 |
Mae | +0.101 |
Amber | +0.099 |
April | +0.097 |
Neg logit contributions | |
� | -0.110 |
bruised | -0.107 |
bruises | -0.106 |
leaning | -0.103 |
swollen | -0.101 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
� | +0.101 |
bruises | +0.100 |
supervision | +0.098 |
bruised | +0.095 |
uc | +0.094 |
Neg logit contributions | |
Julia | -0.101 |
princess | -0.095 |
actor | -0.093 |
mate | -0.093 |
pirate | -0.092 |
Token | hidden |
Token ID | 7104 |
Feature activation | -1.71e-02 |
Pos logit contributions | |
� | +0.808 |
swollen | +0.803 |
bruises | +0.798 |
bruised | +0.795 |
leaning | +0.771 |
Neg logit contributions | |
Julia | -0.663 |
Mae | -0.584 |
Amber | -0.566 |
Jeff | -0.561 |
La | -0.558 |
Token | among |
Token ID | 1871 |
Feature activation | -6.57e-02 |
Pos logit contributions | |
Julia | +0.025 |
pirate | +0.022 |
mate | +0.022 |
actor | +0.022 |
princess | +0.021 |
Neg logit contributions | |
� | -0.035 |
swollen | -0.031 |
uc | -0.030 |
bruised | -0.029 |
leaning | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
Julia | +0.116 |
La | +0.104 |
Mae | +0.103 |
Amber | +0.103 |
Jeff | +0.099 |
Neg logit contributions | |
� | -0.109 |
bruised | -0.108 |
wrench | -0.107 |
swollen | -0.103 |
hammer | -0.103 |
Token | plants |
Token ID | 6134 |
Feature activation | -2.29e-02 |
Pos logit contributions | |
� | +0.098 |
bruised | +0.096 |
wrench | +0.095 |
swollen | +0.093 |
bruises | +0.092 |
Neg logit contributions | |
Julia | -0.089 |
La | -0.082 |
Mae | -0.080 |
Amber | -0.079 |
April | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | +7.27e-03 |
Pos logit contributions | |
Julia | +0.035 |
Mae | +0.031 |
Amber | +0.031 |
Jeff | +0.030 |
La | +0.029 |
Neg logit contributions | |
bruised | -0.041 |
� | -0.041 |
wrench | -0.041 |
swollen | -0.041 |
bruises | -0.040 |
Token |
|
Token ID | 198 |
Feature activation | -2.31e-03 |
Pos logit contributions | |
bruised | +0.032 |
� | +0.032 |
wrench | +0.031 |
swollen | +0.031 |
hammer | +0.031 |
Neg logit contributions | |
Julia | -0.031 |
La | -0.029 |
Mae | -0.028 |
Amber | -0.028 |
Zac | -0.026 |
Token |
|
Token ID | 198 |
Feature activation | -2.12e-03 |
Pos logit contributions | |
Julia | +0.003 |
La | +0.003 |
Mae | +0.003 |
Amber | +0.003 |
Lou | +0.003 |
Neg logit contributions | |
bruised | -0.004 |
wrench | -0.004 |
� | -0.004 |
hammer | -0.004 |
swollen | -0.004 |
Token | She |
Token ID | 3347 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
Julia | +0.003 |
La | +0.003 |
Amber | +0.003 |
Mae | +0.003 |
ia | +0.003 |
Neg logit contributions | |
� | -0.004 |
bruised | -0.004 |
wrench | -0.004 |
hammer | -0.004 |
pling | -0.004 |
Token | twists |
Token ID | 30953 |
Feature activation | +3.23e-02 |
Pos logit contributions | |
bruised | +0.032 |
wrench | +0.032 |
bruises | +0.032 |
� | +0.031 |
swollen | +0.031 |
Neg logit contributions | |
Julia | -0.039 |
La | -0.036 |
Mae | -0.036 |
Amber | -0.035 |
Lou | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | +5.01e-02 |
Pos logit contributions | |
� | +0.072 |
pling | +0.068 |
uc | +0.068 |
wrench | +0.066 |
supervision | +0.065 |
Neg logit contributions | |
Julia | -0.041 |
La | -0.035 |
Amber | -0.033 |
Jeff | -0.033 |
Mae | -0.033 |
Token | wrench |
Token ID | 34561 |
Feature activation | +4.44e-01 |
Pos logit contributions | |
bruised | +0.096 |
� | +0.096 |
swollen | +0.094 |
bruises | +0.092 |
supervision | +0.089 |
Neg logit contributions | |
dives | -0.068 |
Julia | -0.067 |
dolphins | -0.065 |
dances | -0.064 |
La | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
� | +0.815 |
bruised | +0.800 |
swollen | +0.792 |
bruises | +0.786 |
wrench | +0.785 |
Neg logit contributions | |
Julia | -0.686 |
La | -0.615 |
Mae | -0.605 |
Amber | -0.587 |
Zac | -0.575 |
Token | She |
Token ID | 1375 |
Feature activation | +5.40e-03 |
Pos logit contributions | |
bruised | +0.032 |
wrench | +0.032 |
pling | +0.032 |
swollen | +0.032 |
hammer | +0.031 |
Neg logit contributions | |
Julia | -0.029 |
La | -0.025 |
Mae | -0.025 |
Amber | -0.025 |
April | -0.023 |
Token | turns |
Token ID | 4962 |
Feature activation | +1.96e-02 |
Pos logit contributions | |
bruised | +0.009 |
bruises | +0.009 |
wrench | +0.009 |
swollen | +0.008 |
hammer | +0.008 |
Neg logit contributions | |
Julia | -0.009 |
Mae | -0.009 |
Amber | -0.009 |
La | -0.009 |
Jeff | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | +4.83e-02 |
Pos logit contributions | |
� | +0.036 |
bruised | +0.035 |
swollen | +0.035 |
wrench | +0.034 |
pling | +0.034 |
Neg logit contributions | |
Julia | -0.031 |
La | -0.028 |
Mae | -0.027 |
Amber | -0.026 |
Zac | -0.026 |
Token | tap |
Token ID | 9814 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
� | +0.089 |
bruised | +0.086 |
swollen | +0.085 |
pling | +0.084 |
bruises | +0.081 |
Neg logit contributions | |
Julia | -0.069 |
dolphins | -0.068 |
dives | -0.064 |
diver | -0.064 |
dances | -0.063 |
Token |
|
Token ID | 198 |
Feature activation | -5.34e-03 |
Pos logit contributions | |
actor | +0.012 |
mate | +0.012 |
diver | +0.012 |
Mae | +0.012 |
Julia | +0.012 |
Neg logit contributions | |
� | -0.005 |
hammer | -0.004 |
rag | -0.004 |
bruises | -0.004 |
pling | -0.004 |
Token | From |
Token ID | 4863 |
Feature activation | +4.18e-03 |
Pos logit contributions | |
mate | +0.011 |
dolphins | +0.010 |
dives | +0.010 |
Dragon | +0.010 |
pirate | +0.010 |
Neg logit contributions | |
leaning | -0.005 |
issors | -0.005 |
Here | -0.004 |
ised | -0.004 |
� | -0.004 |
Token | then |
Token ID | 788 |
Feature activation | -1.95e-03 |
Pos logit contributions | |
� | +0.006 |
bruised | +0.006 |
wrench | +0.006 |
swollen | +0.005 |
bruises | +0.005 |
Neg logit contributions | |
Julia | -0.008 |
La | -0.008 |
Mae | -0.008 |
Amber | -0.008 |
Zac | -0.007 |
Token | on |
Token ID | 319 |
Feature activation | +5.67e-02 |
Pos logit contributions | |
Julia | +0.003 |
La | +0.002 |
Mae | +0.002 |
Ro | +0.002 |
dances | +0.002 |
Neg logit contributions | |
bruised | -0.004 |
swollen | -0.004 |
supervision | -0.004 |
bruises | -0.004 |
wrench | -0.004 |
Token | , |
Token ID | 11 |
Feature activation | -4.15e-03 |
Pos logit contributions | |
� | +0.070 |
hammer | +0.066 |
bruised | +0.065 |
swollen | +0.064 |
wrench | +0.064 |
Neg logit contributions | |
La | -0.111 |
Julia | -0.107 |
actor | -0.107 |
mate | -0.104 |
dolphins | -0.103 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
mate | +0.005 |
camel | +0.005 |
awoke | +0.005 |
land | +0.005 |
ull | +0.005 |
Neg logit contributions | |
� | -0.006 |
hammer | -0.006 |
file | -0.006 |
don | -0.006 |
free | -0.006 |
Token | would |
Token ID | 561 |
Feature activation | -2.73e-02 |
Pos logit contributions | |
Julia | +0.859 |
La | +0.780 |
Mae | +0.769 |
Amber | +0.763 |
Zac | +0.760 |
Neg logit contributions | |
bruised | -0.734 |
hammer | -0.733 |
� | -0.727 |
wrench | -0.721 |
swollen | -0.707 |
Token | take |
Token ID | 1011 |
Feature activation | +2.71e-02 |
Pos logit contributions | |
Julia | +0.044 |
La | +0.041 |
Mae | +0.040 |
Amber | +0.039 |
Zac | +0.039 |
Neg logit contributions | |
hammer | -0.046 |
� | -0.046 |
wrench | -0.046 |
bruised | -0.046 |
swollen | -0.044 |
Token | her |
Token ID | 607 |
Feature activation | -9.22e-03 |
Pos logit contributions | |
hammer | +0.036 |
bruised | +0.036 |
wrench | +0.036 |
� | +0.035 |
bruises | +0.035 |
Neg logit contributions | |
Julia | -0.051 |
Mae | -0.048 |
La | -0.047 |
Zac | -0.046 |
Amber | -0.046 |
Token | orange |
Token ID | 10912 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
Julia | +0.015 |
April | +0.014 |
royal | +0.013 |
Lou | +0.013 |
La | +0.013 |
Neg logit contributions | |
� | -0.016 |
bruised | -0.015 |
bruises | -0.015 |
swollen | -0.015 |
mercury | -0.015 |
Token | telephone |
Token ID | 11426 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
bruised | +0.020 |
bruises | +0.019 |
swollen | +0.019 |
mercury | +0.019 |
supervision | +0.019 |
Neg logit contributions | |
Julia | -0.019 |
Lou | -0.017 |
April | -0.017 |
La | -0.017 |
Boo | -0.017 |
Token | telephone |
Token ID | 11426 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
Julia | +0.032 |
Lou | +0.027 |
La | +0.027 |
April | +0.027 |
Mae | +0.027 |
Neg logit contributions | |
� | -0.038 |
bruised | -0.037 |
bruises | -0.036 |
swollen | -0.036 |
wrench | -0.035 |
Token | as |
Token ID | 355 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
bruised | +0.196 |
wrench | +0.193 |
hammer | +0.191 |
� | +0.191 |
swollen | +0.187 |
Neg logit contributions | |
Julia | -0.238 |
La | -0.226 |
Mae | -0.220 |
Amber | -0.216 |
Zac | -0.210 |
Token | a |
Token ID | 257 |
Feature activation | -6.85e-02 |
Pos logit contributions | |
hammer | +0.033 |
bruised | +0.032 |
sugar | +0.031 |
nails | +0.030 |
wrench | +0.030 |
Neg logit contributions | |
La | -0.048 |
Julia | -0.047 |
Mae | -0.046 |
Zac | -0.045 |
dives | -0.044 |
Token | gift |
Token ID | 6979 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Julia | +0.112 |
La | +0.102 |
Mae | +0.100 |
Amber | +0.099 |
April | +0.095 |
Neg logit contributions | |
� | -0.117 |
bruised | -0.116 |
wrench | -0.113 |
swollen | -0.113 |
bruises | -0.112 |
Token | . |
Token ID | 13 |
Feature activation | +5.13e-03 |
Pos logit contributions | |
bruised | +0.148 |
wrench | +0.147 |
hammer | +0.146 |
� | +0.143 |
rag | +0.139 |
Neg logit contributions | |
Julia | -0.206 |
La | -0.203 |
Mae | -0.194 |
Amber | -0.190 |
Zac | -0.186 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
� | +0.003 |
leaning | +0.003 |
Here | +0.003 |
| +0.002 |
hammer | +0.002 |
Neg logit contributions | |
actor | -0.012 |
diver | -0.011 |
mate | -0.011 |
dolphins | -0.011 |
royal | -0.011 |
Token | gigg |
Token ID | 30442 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
Julia | +0.885 |
La | +0.865 |
Mae | +0.822 |
Zac | +0.816 |
actor | +0.812 |
Neg logit contributions | |
hammer | -0.666 |
swollen | -0.629 |
bruised | -0.626 |
� | -0.620 |
bruises | -0.611 |
Token | led |
Token ID | 992 |
Feature activation | +2.47e-02 |
Pos logit contributions | |
Julia | +0.209 |
La | +0.194 |
Mae | +0.191 |
Amber | +0.187 |
Zac | +0.181 |
Neg logit contributions | |
bruised | -0.173 |
� | -0.172 |
wrench | -0.170 |
swollen | -0.168 |
hammer | -0.166 |
Token | with |
Token ID | 351 |
Feature activation | -3.82e-02 |
Pos logit contributions | |
bruised | +0.028 |
hammer | +0.026 |
swollen | +0.026 |
wrench | +0.026 |
� | +0.025 |
Neg logit contributions | |
Julia | -0.052 |
La | -0.052 |
Mae | -0.049 |
Zac | -0.049 |
ae | -0.048 |
Token | joy |
Token ID | 8716 |
Feature activation | -2.50e-02 |
Pos logit contributions | |
Julia | +0.062 |
La | +0.057 |
Mae | +0.056 |
Amber | +0.055 |
Zac | +0.052 |
Neg logit contributions | |
bruised | -0.066 |
� | -0.066 |
wrench | -0.066 |
swollen | -0.064 |
hammer | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
Julia | +0.040 |
Mae | +0.036 |
Amber | +0.036 |
La | +0.036 |
April | +0.034 |
Neg logit contributions | |
� | -0.044 |
wrench | -0.044 |
bruised | -0.044 |
bruises | -0.043 |
swollen | -0.043 |
Token | s |
Token ID | 82 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
Julia | +0.216 |
La | +0.199 |
Mae | +0.196 |
Amber | +0.193 |
Zac | +0.186 |
Neg logit contributions | |
bruised | -0.179 |
� | -0.178 |
wrench | -0.177 |
swollen | -0.174 |
bruises | -0.173 |
Token | that |
Token ID | 326 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
Julia | +0.050 |
La | +0.047 |
Mae | +0.046 |
Amber | +0.045 |
April | +0.044 |
Neg logit contributions | |
bruised | -0.033 |
� | -0.033 |
wrench | -0.033 |
swollen | -0.032 |
bruises | -0.032 |
Token | ? |
Token ID | 30 |
Feature activation | +2.21e-03 |
Pos logit contributions | |
hammer | +0.222 |
bruised | +0.221 |
wrench | +0.220 |
swollen | +0.214 |
� | +0.212 |
Neg logit contributions | |
Julia | -0.231 |
La | -0.215 |
Mae | -0.209 |
Amber | -0.205 |
Zac | -0.197 |
Token | â |
Token ID | 22940 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
� | +0.003 |
bruised | +0.003 |
wrench | +0.003 |
hammer | +0.003 |
swollen | +0.003 |
Neg logit contributions | |
Julia | -0.004 |
La | -0.004 |
Mae | -0.004 |
Amber | -0.004 |
dolphins | -0.004 |
Token | € |
Token ID | 26391 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
bruised | +0.072 |
wrench | +0.071 |
swollen | +0.070 |
bruises | +0.070 |
� | +0.070 |
Neg logit contributions | |
Julia | -0.039 |
La | -0.035 |
Mae | -0.033 |
Amber | -0.033 |
Lou | -0.030 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
Julia | +0.251 |
La | +0.231 |
Mae | +0.229 |
Amber | +0.225 |
Zac | +0.221 |
Neg logit contributions | |
� | -0.221 |
bruised | -0.211 |
wrench | -0.209 |
hammer | -0.206 |
swollen | -0.205 |
Token | asked |
Token ID | 1965 |
Feature activation | -3.17e-02 |
Pos logit contributions | |
Julia | +0.899 |
La | +0.843 |
Mae | +0.823 |
Amber | +0.815 |
Zac | +0.794 |
Neg logit contributions | |
swollen | -0.688 |
bruised | -0.688 |
hammer | -0.681 |
� | -0.678 |
wrench | -0.663 |
Token | her |
Token ID | 607 |
Feature activation | -9.28e-03 |
Pos logit contributions | |
Julia | +0.056 |
La | +0.052 |
Mae | +0.051 |
Amber | +0.050 |
Zac | +0.048 |
Neg logit contributions | |
bruised | -0.050 |
wrench | -0.049 |
� | -0.049 |
swollen | -0.048 |
hammer | -0.048 |
Token | cat |
Token ID | 3797 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
Julia | +0.017 |
La | +0.017 |
Mae | +0.016 |
Amber | +0.016 |
Zac | +0.015 |
Neg logit contributions | |
hammer | -0.014 |
wrench | -0.013 |
bruised | -0.013 |
pling | -0.013 |
swollen | -0.012 |
Token | curiously |
Token ID | 43211 |
Feature activation | -6.39e-02 |
Pos logit contributions | |
hammer | +0.075 |
bruised | +0.074 |
wrench | +0.071 |
swollen | +0.067 |
pling | +0.066 |
Neg logit contributions | |
La | -0.108 |
Julia | -0.107 |
Mae | -0.103 |
actor | -0.099 |
Amber | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | +7.23e-03 |
Pos logit contributions | |
Julia | +0.118 |
La | +0.113 |
Mae | +0.108 |
Amber | +0.106 |
Zac | +0.104 |
Neg logit contributions | |
bruised | -0.094 |
wrench | -0.091 |
hammer | -0.090 |
� | -0.089 |
swollen | -0.088 |
Token | waved |
Token ID | 26834 |
Feature activation | -1.97e-01 |
Pos logit contributions | |
bruised | +0.017 |
bruises | +0.017 |
swollen | +0.016 |
rag | +0.015 |
scrape | +0.015 |
Neg logit contributions | |
La | -0.022 |
Julia | -0.022 |
mate | -0.021 |
actor | -0.021 |
adies | -0.021 |
Token | as |
Token ID | 355 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
Julia | +0.318 |
La | +0.301 |
Mae | +0.292 |
Amber | +0.279 |
actor | +0.278 |
Neg logit contributions | |
bruised | -0.336 |
swollen | -0.320 |
wrench | -0.320 |
� | -0.314 |
bruises | -0.311 |
Token | they |
Token ID | 484 |
Feature activation | -9.07e-02 |
Pos logit contributions | |
bruised | +0.036 |
hammer | +0.035 |
bruises | +0.034 |
wrench | +0.034 |
� | +0.033 |
Neg logit contributions | |
La | -0.044 |
mate | -0.044 |
Julia | -0.042 |
actor | -0.041 |
echo | -0.041 |
Token | left |
Token ID | 1364 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
Julia | +0.143 |
La | +0.131 |
Mae | +0.128 |
Amber | +0.126 |
Zac | +0.120 |
Neg logit contributions | |
bruised | -0.161 |
� | -0.160 |
wrench | -0.159 |
swollen | -0.156 |
hammer | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | +9.79e-03 |
Pos logit contributions | |
bruised | +0.054 |
wrench | +0.053 |
� | +0.053 |
swollen | +0.053 |
hammer | +0.052 |
Neg logit contributions | |
Julia | -0.054 |
Mae | -0.049 |
La | -0.048 |
Amber | -0.048 |
Zac | -0.046 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
� | +0.014 |
bruises | +0.011 |
bruised | +0.011 |
hammer | +0.010 |
wrench | +0.010 |
Neg logit contributions | |
mate | -0.018 |
dolphins | -0.018 |
actor | -0.018 |
diver | -0.017 |
aliens | -0.017 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.19e-02 |
Pos logit contributions | |
Julia | +0.887 |
La | +0.838 |
mate | +0.823 |
Mae | +0.815 |
adies | +0.815 |
Neg logit contributions | |
bruised | -0.625 |
swollen | -0.621 |
� | -0.610 |
bruises | -0.604 |
pling | -0.586 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.34e-02 |
Pos logit contributions | |
Julia | +0.061 |
La | +0.056 |
Mae | +0.056 |
Lou | +0.055 |
Jeff | +0.053 |
Neg logit contributions | |
bruised | -0.048 |
swollen | -0.045 |
bruises | -0.043 |
� | -0.042 |
free | -0.040 |
Token | to |
Token ID | 284 |
Feature activation | +1.81e-03 |
Pos logit contributions | |
Julia | +0.143 |
La | +0.139 |
Mae | +0.133 |
Amber | +0.130 |
Zac | +0.128 |
Neg logit contributions | |
bruised | -0.103 |
� | -0.100 |
bruises | -0.097 |
swollen | -0.097 |
wrench | -0.097 |
Token | have |
Token ID | 423 |
Feature activation | +8.67e-04 |
Pos logit contributions | |
� | +0.003 |
hammer | +0.003 |
bruised | +0.002 |
wrench | +0.002 |
swollen | +0.002 |
Neg logit contributions | |
Julia | -0.003 |
Mae | -0.003 |
Amber | -0.003 |
dolphins | -0.003 |
Zac | -0.003 |
Token | borrowed |
Token ID | 22546 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
bruises | +0.001 |
� | +0.001 |
bruised | +0.001 |
swollen | +0.001 |
pling | +0.001 |
Neg logit contributions | |
Julia | -0.002 |
Zac | -0.002 |
Mae | -0.002 |
actor | -0.002 |
La | -0.002 |
Token | little |
Token ID | 1310 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
Julia | +0.130 |
Mae | +0.118 |
La | +0.118 |
Amber | +0.115 |
Zac | +0.115 |
Neg logit contributions | |
hammer | -0.095 |
wrench | -0.094 |
bruised | -0.093 |
� | -0.092 |
swollen | -0.091 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.97e-03 |
Pos logit contributions | |
bruised | +0.030 |
� | +0.030 |
wrench | +0.030 |
swollen | +0.029 |
hammer | +0.029 |
Neg logit contributions | |
Julia | -0.032 |
La | -0.030 |
Mae | -0.029 |
Amber | -0.029 |
Zac | -0.028 |
Token | called |
Token ID | 1444 |
Feature activation | -6.02e-02 |
Pos logit contributions | |
Julia | +0.009 |
La | +0.008 |
Mae | +0.008 |
Amber | +0.008 |
Zac | +0.007 |
Neg logit contributions | |
bruised | -0.008 |
� | -0.008 |
wrench | -0.008 |
swollen | -0.008 |
hammer | -0.008 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
La | +0.097 |
actor | +0.095 |
Julia | +0.093 |
dolphins | +0.092 |
Zac | +0.090 |
Neg logit contributions | |
� | -0.096 |
bruised | -0.092 |
hammer | -0.091 |
rag | -0.088 |
bruises | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +4.34e-03 |
Pos logit contributions | |
Julia | +1.061 |
Zac | +1.026 |
La | +1.000 |
Jeff | +0.991 |
dives | +0.990 |
Neg logit contributions | |
hammer | -0.460 |
bruised | -0.453 |
rag | -0.448 |
pling | -0.435 |
therm | -0.423 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
� | +0.003 |
leaning | +0.003 |
issors | +0.003 |
file | +0.003 |
hammer | +0.003 |
Neg logit contributions | |
dolphins | -0.009 |
diver | -0.009 |
mate | -0.009 |
actor | -0.009 |
orse | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
Julia | +0.858 |
La | +0.789 |
Mae | +0.776 |
Amber | +0.766 |
Zac | +0.757 |
Neg logit contributions | |
bruised | -0.740 |
� | -0.736 |
hammer | -0.729 |
swollen | -0.719 |
wrench | -0.717 |
Token | a |
Token ID | 257 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
Julia | +0.077 |
La | +0.071 |
Mae | +0.069 |
Amber | +0.068 |
Zac | +0.066 |
Neg logit contributions | |
bruised | -0.082 |
� | -0.080 |
wrench | -0.079 |
hammer | -0.079 |
swollen | -0.078 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
Julia | +0.115 |
La | +0.105 |
Mae | +0.103 |
Amber | +0.102 |
Zac | +0.099 |
Neg logit contributions | |
bruised | -0.109 |
� | -0.109 |
wrench | -0.108 |
hammer | -0.106 |
swollen | -0.106 |
Token | child |
Token ID | 1200 |
Feature activation | +2.02e-02 |
Pos logit contributions | |
Julia | +0.140 |
La | +0.135 |
Zac | +0.130 |
Mae | +0.127 |
Jeff | +0.125 |
Neg logit contributions | |
hammer | -0.095 |
bruised | -0.094 |
rag | -0.094 |
wrench | -0.093 |
� | -0.093 |
Token | and |
Token ID | 290 |
Feature activation | +1.26e-02 |
Pos logit contributions | |
bruised | +0.032 |
� | +0.032 |
wrench | +0.032 |
swollen | +0.031 |
hammer | +0.031 |
Neg logit contributions | |
Julia | -0.035 |
La | -0.033 |
Mae | -0.032 |
Amber | -0.031 |
Zac | -0.030 |
Token | outside |
Token ID | 2354 |
Feature activation | -6.18e-02 |
Pos logit contributions | |
Julia | +0.046 |
Mae | +0.043 |
La | +0.041 |
Amber | +0.041 |
Zac | +0.041 |
Neg logit contributions | |
� | -0.036 |
hammer | -0.035 |
wrench | -0.034 |
bruised | -0.033 |
rag | -0.033 |
Token | . |
Token ID | 13 |
Feature activation | +4.03e-03 |
Pos logit contributions | |
Julia | +0.122 |
La | +0.116 |
Amber | +0.112 |
Mae | +0.112 |
Zac | +0.110 |
Neg logit contributions | |
� | -0.085 |
bruised | -0.083 |
hammer | -0.082 |
wrench | -0.079 |
swollen | -0.077 |
Token | One |
Token ID | 1881 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
� | +0.004 |
leaning | +0.003 |
rag | +0.003 |
hammer | +0.003 |
issors | +0.003 |
Neg logit contributions | |
diver | -0.009 |
dolphins | -0.009 |
actor | -0.008 |
divers | -0.008 |
mate | -0.008 |
Token | day |
Token ID | 1110 |
Feature activation | +7.45e-03 |
Pos logit contributions | |
Julia | +0.065 |
La | +0.060 |
Amber | +0.059 |
Mae | +0.059 |
Zac | +0.058 |
Neg logit contributions | |
bruised | -0.043 |
hammer | -0.043 |
rag | -0.042 |
� | -0.042 |
wrench | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -2.24e-03 |
Pos logit contributions | |
� | +0.005 |
hammer | +0.005 |
uc | +0.005 |
wrench | +0.004 |
mons | +0.004 |
Neg logit contributions | |
La | -0.017 |
nes | -0.017 |
actor | -0.017 |
aliens | -0.017 |
echo | -0.017 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
mate | +0.004 |
rina | +0.004 |
La | +0.004 |
aliens | +0.004 |
awoke | +0.004 |
Neg logit contributions | |
hammer | -0.002 |
� | -0.002 |
lipstick | -0.002 |
mons | -0.002 |
don | -0.002 |
Token | found |
Token ID | 1043 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
Julia | +0.943 |
Zac | +0.890 |
La | +0.877 |
Jeff | +0.857 |
Mae | +0.841 |
Neg logit contributions | |
hammer | -0.660 |
rag | -0.638 |
bruised | -0.621 |
oil | -0.618 |
pling | -0.613 |
Token | an |
Token ID | 281 |
Feature activation | -1.30e-03 |
Pos logit contributions | |
Julia | +0.074 |
La | +0.066 |
Mae | +0.065 |
Amber | +0.064 |
Lou | +0.061 |
Neg logit contributions | |
� | -0.101 |
bruised | -0.100 |
wrench | -0.100 |
swollen | -0.098 |
hammer | -0.097 |
Token | orange |
Token ID | 10912 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
Julia | +0.002 |
La | +0.002 |
Mae | +0.002 |
Amber | +0.002 |
Zac | +0.002 |
Neg logit contributions | |
bruised | -0.002 |
� | -0.002 |
wrench | -0.002 |
swollen | -0.002 |
hammer | -0.002 |
Token | telephone |
Token ID | 11426 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
bruised | +0.022 |
supervision | +0.022 |
bruises | +0.021 |
swollen | +0.021 |
pling | +0.020 |
Neg logit contributions | |
Julia | -0.023 |
Mae | -0.021 |
Amber | -0.021 |
Jeff | -0.021 |
La | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | +7.37e-03 |
Pos logit contributions | |
swollen | +0.230 |
bruised | +0.229 |
� | +0.227 |
wrench | +0.226 |
bruises | +0.225 |
Neg logit contributions | |
Julia | -0.201 |
La | -0.181 |
Mae | -0.177 |
Amber | -0.173 |
April | -0.171 |
Token | lived |
Token ID | 5615 |
Feature activation | -4.21e-02 |
Pos logit contributions | |
� | +0.070 |
bruised | +0.069 |
hammer | +0.068 |
wrench | +0.067 |
swollen | +0.066 |
Neg logit contributions | |
Julia | -0.069 |
La | -0.066 |
Amber | -0.063 |
Mae | -0.063 |
April | -0.060 |
Token | a |
Token ID | 257 |
Feature activation | -6.66e-02 |
Pos logit contributions | |
Julia | +0.069 |
La | +0.062 |
Mae | +0.062 |
Amber | +0.060 |
Zac | +0.059 |
Neg logit contributions | |
� | -0.072 |
bruised | -0.072 |
hammer | -0.070 |
bruises | -0.070 |
wrench | -0.070 |
Token | little |
Token ID | 1310 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
Julia | +0.130 |
Mae | +0.118 |
La | +0.118 |
Amber | +0.115 |
Zac | +0.115 |
Neg logit contributions | |
hammer | -0.095 |
wrench | -0.094 |
bruised | -0.093 |
� | -0.092 |
swollen | -0.091 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.97e-03 |
Pos logit contributions | |
bruised | +0.030 |
� | +0.030 |
wrench | +0.030 |
swollen | +0.029 |
hammer | +0.029 |
Neg logit contributions | |
Julia | -0.032 |
La | -0.030 |
Mae | -0.029 |
Amber | -0.029 |
Zac | -0.028 |
Token | called |
Token ID | 1444 |
Feature activation | -6.02e-02 |
Pos logit contributions | |
Julia | +0.009 |
La | +0.008 |
Mae | +0.008 |
Amber | +0.008 |
Zac | +0.007 |
Neg logit contributions | |
bruised | -0.008 |
� | -0.008 |
wrench | -0.008 |
swollen | -0.008 |
hammer | -0.008 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
La | +0.097 |
actor | +0.095 |
Julia | +0.093 |
dolphins | +0.092 |
Zac | +0.090 |
Neg logit contributions | |
� | -0.096 |
bruised | -0.092 |
hammer | -0.091 |
rag | -0.088 |
bruises | -0.088 |
Token | . |
Token ID | 13 |
Feature activation | +4.34e-03 |
Pos logit contributions | |
Julia | +1.061 |
Zac | +1.026 |
La | +1.000 |
Jeff | +0.991 |
dives | +0.990 |
Neg logit contributions | |
hammer | -0.460 |
bruised | -0.453 |
rag | -0.448 |
pling | -0.435 |
therm | -0.423 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.79e-01 |
Pos logit contributions | |
� | +0.003 |
leaning | +0.003 |
issors | +0.003 |
file | +0.003 |
hammer | +0.003 |
Neg logit contributions | |
dolphins | -0.009 |
diver | -0.009 |
mate | -0.009 |
actor | -0.009 |
orse | -0.009 |
Token | was |
Token ID | 373 |
Feature activation | -4.69e-02 |
Pos logit contributions | |
Julia | +0.858 |
La | +0.789 |
Mae | +0.776 |
Amber | +0.766 |
Zac | +0.757 |
Neg logit contributions | |
bruised | -0.740 |
� | -0.736 |
hammer | -0.729 |
swollen | -0.719 |
wrench | -0.717 |
Token | a |
Token ID | 257 |
Feature activation | -6.68e-02 |
Pos logit contributions | |
Julia | +0.077 |
La | +0.071 |
Mae | +0.069 |
Amber | +0.068 |
Zac | +0.066 |
Neg logit contributions | |
bruised | -0.082 |
� | -0.080 |
wrench | -0.079 |
hammer | -0.079 |
swollen | -0.078 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
Julia | +0.115 |
La | +0.105 |
Mae | +0.103 |
Amber | +0.102 |
Zac | +0.099 |
Neg logit contributions | |
bruised | -0.109 |
� | -0.109 |
wrench | -0.108 |
hammer | -0.106 |
swollen | -0.106 |
Token | was |
Token ID | 373 |
Feature activation | -4.92e-02 |
Pos logit contributions | |
� | +0.065 |
bruised | +0.065 |
wrench | +0.063 |
hammer | +0.063 |
swollen | +0.063 |
Neg logit contributions | |
Julia | -0.061 |
La | -0.058 |
Mae | -0.055 |
Amber | -0.055 |
April | -0.052 |
Token | a |
Token ID | 257 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
Julia | +0.090 |
La | +0.082 |
Mae | +0.081 |
Amber | +0.079 |
Zac | +0.079 |
Neg logit contributions | |
� | -0.074 |
bruised | -0.073 |
bruises | -0.073 |
hammer | -0.072 |
rag | -0.071 |
Token | little |
Token ID | 1310 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
Julia | +0.128 |
La | +0.117 |
Mae | +0.116 |
Amber | +0.114 |
Zac | +0.111 |
Neg logit contributions | |
� | -0.109 |
bruised | -0.108 |
wrench | -0.108 |
hammer | -0.106 |
swollen | -0.105 |
Token | girl |
Token ID | 2576 |
Feature activation | -8.30e-03 |
Pos logit contributions | |
� | +0.025 |
bruised | +0.025 |
bruises | +0.025 |
wrench | +0.025 |
swollen | +0.025 |
Neg logit contributions | |
Julia | -0.023 |
La | -0.021 |
Mae | -0.021 |
Amber | -0.020 |
dolphins | -0.020 |
Token | named |
Token ID | 3706 |
Feature activation | +5.93e-03 |
Pos logit contributions | |
Julia | +0.012 |
La | +0.011 |
Mae | +0.011 |
Amber | +0.010 |
April | +0.010 |
Neg logit contributions | |
wrench | -0.016 |
� | -0.016 |
bruised | -0.015 |
swollen | -0.015 |
pling | -0.015 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
wrench | +0.011 |
bruised | +0.011 |
� | +0.011 |
swollen | +0.011 |
bruises | +0.011 |
Neg logit contributions | |
Julia | -0.009 |
La | -0.008 |
Mae | -0.008 |
Amber | -0.008 |
April | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | +4.76e-03 |
Pos logit contributions | |
Julia | +0.931 |
La | +0.857 |
Mae | +0.856 |
Zac | +0.849 |
Amber | +0.840 |
Neg logit contributions | |
bruised | -0.661 |
hammer | -0.639 |
� | -0.637 |
swollen | -0.627 |
wrench | -0.619 |
Token | She |
Token ID | 1375 |
Feature activation | +9.26e-03 |
Pos logit contributions | |
� | +0.004 |
leaning | +0.003 |
hammer | +0.003 |
issors | +0.003 |
file | +0.003 |
Neg logit contributions | |
actor | -0.010 |
mate | -0.010 |
dolphins | -0.010 |
oper | -0.010 |
diver | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -4.84e-02 |
Pos logit contributions | |
bruised | +0.012 |
� | +0.012 |
hammer | +0.012 |
swollen | +0.011 |
bruises | +0.011 |
Neg logit contributions | |
Julia | -0.019 |
Mae | -0.018 |
Amber | -0.018 |
Zac | -0.017 |
La | -0.017 |
Token | only |
Token ID | 691 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
Julia | +0.078 |
La | +0.071 |
Mae | +0.070 |
Amber | +0.068 |
Zac | +0.066 |
Neg logit contributions | |
bruised | -0.085 |
� | -0.084 |
wrench | -0.084 |
swollen | -0.082 |
hammer | -0.082 |
Token | three |
Token ID | 1115 |
Feature activation | -9.59e-02 |
Pos logit contributions | |
bruised | +0.091 |
wrench | +0.088 |
hammer | +0.088 |
� | +0.087 |
bruises | +0.086 |
Neg logit contributions | |
Julia | -0.077 |
La | -0.072 |
Mae | -0.069 |
Amber | -0.068 |
Zac | -0.066 |
Token | ne |
Token ID | 497 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
Julia | +0.067 |
La | +0.061 |
Mae | +0.060 |
Amber | +0.058 |
Zac | +0.056 |
Neg logit contributions | |
bruised | -0.080 |
� | -0.079 |
wrench | -0.079 |
swollen | -0.078 |
hammer | -0.077 |
Token | ared |
Token ID | 1144 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
Julia | +0.124 |
La | +0.115 |
Mae | +0.111 |
Amber | +0.111 |
Zac | +0.108 |
Neg logit contributions | |
bruised | -0.093 |
wrench | -0.092 |
� | -0.092 |
bruises | -0.091 |
hammer | -0.089 |
Token | the |
Token ID | 262 |
Feature activation | +5.04e-02 |
Pos logit contributions | |
bruised | +0.240 |
hammer | +0.231 |
swollen | +0.225 |
wrench | +0.224 |
bruises | +0.222 |
Neg logit contributions | |
La | -0.303 |
Julia | -0.303 |
actor | -0.290 |
Mae | -0.283 |
Amber | -0.279 |
Token | top |
Token ID | 1353 |
Feature activation | +5.27e-03 |
Pos logit contributions | |
� | +0.092 |
uc | +0.088 |
leaning | +0.087 |
bruised | +0.086 |
pling | +0.086 |
Neg logit contributions | |
Julia | -0.074 |
dolphins | -0.066 |
La | -0.066 |
Lou | -0.066 |
Jeff | -0.065 |
Token | , |
Token ID | 11 |
Feature activation | -1.32e-03 |
Pos logit contributions | |
� | +0.008 |
wrench | +0.008 |
hammer | +0.008 |
bruised | +0.008 |
swollen | +0.007 |
Neg logit contributions | |
Julia | -0.010 |
La | -0.009 |
Mae | -0.009 |
Amber | -0.009 |
actor | -0.009 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
mate | +0.003 |
nes | +0.003 |
actor | +0.003 |
ozo | +0.003 |
Birthday | +0.003 |
Neg logit contributions | |
hammer | -0.001 |
sugar | -0.001 |
nails | -0.001 |
knives | -0.001 |
bruises | -0.001 |
Token | stopped |
Token ID | 5025 |
Feature activation | -7.32e-02 |
Pos logit contributions | |
Julia | +0.911 |
La | +0.826 |
Mae | +0.824 |
Amber | +0.824 |
Zac | +0.812 |
Neg logit contributions | |
hammer | -0.671 |
bruised | -0.646 |
wrench | -0.644 |
swollen | -0.640 |
� | -0.637 |
Token | in |
Token ID | 287 |
Feature activation | +6.70e-03 |
Pos logit contributions | |
Julia | +0.117 |
La | +0.107 |
Mae | +0.103 |
Amber | +0.101 |
Lou | +0.099 |
Neg logit contributions | |
� | -0.130 |
bruised | -0.128 |
wrench | -0.127 |
swollen | -0.124 |
bruises | -0.123 |
Token | her |
Token ID | 607 |
Feature activation | -7.99e-03 |
Pos logit contributions | |
wrench | +0.014 |
uc | +0.014 |
pling | +0.014 |
issors | +0.014 |
mons | +0.014 |
Neg logit contributions | |
Julia | -0.010 |
Mae | -0.009 |
Amber | -0.008 |
La | -0.008 |
April | -0.008 |
Token | tracks |
Token ID | 8339 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
Julia | +0.016 |
La | +0.014 |
Mae | +0.014 |
Amber | +0.014 |
Zac | +0.014 |
Neg logit contributions | |
bruised | -0.011 |
� | -0.011 |
wrench | -0.011 |
swollen | -0.011 |
pling | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +2.51e-03 |
Pos logit contributions | |
Julia | +0.196 |
La | +0.182 |
Mae | +0.178 |
Amber | +0.175 |
Zac | +0.169 |
Neg logit contributions | |
bruised | -0.171 |
� | -0.171 |
wrench | -0.169 |
hammer | -0.166 |
swollen | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | +6.90e-03 |
Pos logit contributions | |
Julia | +0.066 |
La | +0.061 |
Mae | +0.060 |
Amber | +0.059 |
Zac | +0.057 |
Neg logit contributions | |
bruised | -0.059 |
� | -0.058 |
wrench | -0.058 |
hammer | -0.057 |
swollen | -0.057 |
Token |
|
Token ID | 198 |
Feature activation | -3.22e-03 |
Pos logit contributions | |
� | +0.006 |
Here | +0.004 |
leaning | +0.004 |
hammer | +0.004 |
issors | +0.004 |
Neg logit contributions | |
actor | -0.015 |
mate | -0.015 |
dolphins | -0.015 |
diver | -0.015 |
aliens | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -3.03e-03 |
Pos logit contributions | |
actor | +0.008 |
mate | +0.007 |
diver | +0.007 |
dolphins | +0.007 |
echo | +0.007 |
Neg logit contributions | |
� | -0.002 |
hammer | -0.002 |
Here | -0.001 |
leaning | -0.001 |
itt | -0.001 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
Dragon | +0.006 |
mate | +0.006 |
aliens | +0.005 |
actor | +0.005 |
diver | +0.005 |
Neg logit contributions | |
leaning | -0.003 |
ised | -0.002 |
Here | -0.002 |
issors | -0.002 |
If | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +1.53e-04 |
Pos logit contributions | |
actor | +0.159 |
La | +0.154 |
nes | +0.152 |
echo | +0.144 |
April | +0.143 |
Neg logit contributions | |
hammer | -0.032 |
nails | -0.028 |
bruised | -0.026 |
sugar | -0.026 |
knives | -0.025 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
hammer | +0.000 |
nails | +0.000 |
sugar | +0.000 |
mons | +0.000 |
bruises | +0.000 |
Neg logit contributions | |
mate | -0.000 |
actor | -0.000 |
nes | -0.000 |
La | -0.000 |
anna | -0.000 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Julia | +1.008 |
Zac | +0.957 |
Jeff | +0.941 |
oper | +0.919 |
actor | +0.915 |
Neg logit contributions | |
hammer | -0.589 |
pling | -0.545 |
bruised | -0.533 |
wrench | -0.525 |
swollen | -0.518 |
Token | a |
Token ID | 257 |
Feature activation | -6.60e-02 |
Pos logit contributions | |
Zac | +0.249 |
ozo | +0.247 |
diver | +0.246 |
Julia | +0.243 |
Mae | +0.241 |
Neg logit contributions | |
bruises | -0.128 |
nails | -0.125 |
bruised | -0.119 |
hammer | -0.119 |
oil | -0.118 |
Token | mysterious |
Token ID | 11428 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
Julia | +0.106 |
Mae | +0.096 |
La | +0.096 |
Amber | +0.095 |
Zac | +0.091 |
Neg logit contributions | |
swollen | -0.115 |
bruised | -0.115 |
wrench | -0.114 |
hammer | -0.114 |
� | -0.111 |
Token | voice |
Token ID | 3809 |
Feature activation | -1.00e-01 |
Pos logit contributions | |
hammer | +0.048 |
bruised | +0.047 |
wrench | +0.047 |
swollen | +0.046 |
� | +0.046 |
Neg logit contributions | |
Julia | -0.045 |
La | -0.041 |
Mae | -0.041 |
Amber | -0.040 |
Zac | -0.038 |
Token | . |
Token ID | 13 |
Feature activation | +7.03e-03 |
Pos logit contributions | |
Julia | +0.205 |
Mae | +0.195 |
La | +0.191 |
Amber | +0.190 |
Zac | +0.188 |
Neg logit contributions | |
hammer | -0.128 |
bruised | -0.126 |
therm | -0.120 |
wrench | -0.119 |
cut | -0.116 |
Token | strong |
Token ID | 1913 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
bruised | +0.019 |
swollen | +0.019 |
bruises | +0.019 |
hammer | +0.019 |
wrench | +0.018 |
Neg logit contributions | |
Julia | -0.023 |
La | -0.022 |
Mae | -0.021 |
Amber | -0.021 |
Zac | -0.021 |
Token | ! |
Token ID | 0 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
� | +0.085 |
wrench | +0.084 |
bruised | +0.084 |
swollen | +0.083 |
bruises | +0.083 |
Neg logit contributions | |
Julia | -0.081 |
La | -0.073 |
Mae | -0.072 |
Amber | -0.071 |
Zac | -0.069 |
Token |
|
Token ID | 198 |
Feature activation | -4.43e-03 |
Pos logit contributions | |
actor | +0.035 |
mate | +0.035 |
dolphins | +0.035 |
La | +0.033 |
diver | +0.033 |
Neg logit contributions | |
� | -0.023 |
bruises | -0.018 |
bruised | -0.018 |
hammer | -0.018 |
wrench | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -3.87e-03 |
Pos logit contributions | |
actor | +0.009 |
mate | +0.009 |
Julia | +0.009 |
dolphins | +0.009 |
diver | +0.009 |
Neg logit contributions | |
� | -0.005 |
bruises | -0.005 |
hammer | -0.004 |
swollen | -0.004 |
pling | -0.004 |
Token | Suddenly |
Token ID | 38582 |
Feature activation | -6.13e-02 |
Pos logit contributions | |
mate | +0.007 |
actor | +0.007 |
aliens | +0.007 |
pirate | +0.007 |
dolphins | +0.007 |
Neg logit contributions | |
� | -0.005 |
bruises | -0.004 |
leaning | -0.004 |
Here | -0.004 |
swollen | -0.004 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
actor | +0.137 |
La | +0.135 |
mate | +0.128 |
villains | +0.127 |
ozo | +0.127 |
Neg logit contributions | |
hammer | -0.043 |
nails | -0.042 |
sugar | -0.041 |
� | -0.040 |
bruises | -0.039 |
Token | heard |
Token ID | 2982 |
Feature activation | -1.23e-01 |
Pos logit contributions | |
Julia | +0.844 |
actor | +0.792 |
Zac | +0.786 |
La | +0.786 |
oper | +0.780 |
Neg logit contributions | |
hammer | -0.691 |
� | -0.681 |
swollen | -0.674 |
pling | -0.664 |
bruised | -0.662 |
Token | a |
Token ID | 257 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
diver | +0.236 |
dives | +0.234 |
ozo | +0.234 |
nes | +0.231 |
actor | +0.229 |
Neg logit contributions | |
bruises | -0.126 |
nails | -0.123 |
hammer | -0.120 |
sugar | -0.119 |
oil | -0.118 |
Token | sn |
Token ID | 3013 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
Julia | +0.108 |
La | +0.099 |
Mae | +0.097 |
Amber | +0.096 |
Zac | +0.091 |
Neg logit contributions | |
bruised | -0.127 |
wrench | -0.126 |
swollen | -0.126 |
� | -0.125 |
hammer | -0.125 |
Token | ort |
Token ID | 419 |
Feature activation | -3.20e-02 |
Pos logit contributions | |
Julia | +0.266 |
La | +0.242 |
Mae | +0.235 |
Amber | +0.234 |
ia | +0.226 |
Neg logit contributions | |
bruised | -0.285 |
swollen | -0.284 |
wrench | -0.283 |
bruises | -0.281 |
hammer | -0.274 |
Token | behind |
Token ID | 2157 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
Julia | +0.060 |
La | +0.058 |
Mae | +0.055 |
Amber | +0.054 |
Zac | +0.052 |
Neg logit contributions | |
bruised | -0.045 |
hammer | -0.045 |
wrench | -0.045 |
� | -0.044 |
swollen | -0.042 |
Token | lend |
Token ID | 22096 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Julia | +0.042 |
La | +0.041 |
Mae | +0.038 |
Amber | +0.038 |
Zac | +0.037 |
Neg logit contributions | |
� | -0.039 |
hammer | -0.039 |
wrench | -0.038 |
bruised | -0.038 |
bruises | -0.036 |
Token | him |
Token ID | 683 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
gum | +0.153 |
bruises | +0.146 |
hammer | +0.146 |
bruised | +0.141 |
handcuffs | +0.139 |
Neg logit contributions | |
Julia | -0.288 |
La | -0.285 |
actor | -0.278 |
diver | -0.276 |
ae | -0.274 |
Token | her |
Token ID | 607 |
Feature activation | -7.28e-03 |
Pos logit contributions | |
bruised | +0.068 |
wrench | +0.067 |
hammer | +0.066 |
� | +0.066 |
bruises | +0.065 |
Neg logit contributions | |
Julia | -0.072 |
La | -0.067 |
Mae | -0.065 |
Amber | -0.064 |
April | -0.062 |
Token | boat |
Token ID | 8848 |
Feature activation | -9.41e-02 |
Pos logit contributions | |
Julia | +0.013 |
La | +0.012 |
Mae | +0.011 |
Amber | +0.011 |
Zac | +0.011 |
Neg logit contributions | |
bruised | -0.012 |
� | -0.012 |
wrench | -0.012 |
swollen | -0.011 |
hammer | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-03 |
Pos logit contributions | |
Julia | +0.159 |
La | +0.146 |
Mae | +0.144 |
Amber | +0.142 |
Zac | +0.136 |
Neg logit contributions | |
bruised | -0.155 |
� | -0.154 |
wrench | -0.153 |
swollen | -0.151 |
bruises | -0.149 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
� | +0.006 |
hammer | +0.006 |
wrench | +0.006 |
rag | +0.005 |
issors | +0.005 |
Neg logit contributions | |
dolphins | -0.004 |
La | -0.003 |
diver | -0.003 |
royal | -0.003 |
mate | -0.003 |
Token | said |
Token ID | 531 |
Feature activation | +3.46e-03 |
Pos logit contributions | |
La | +0.919 |
Julia | +0.910 |
Mae | +0.838 |
Amber | +0.828 |
April | +0.817 |
Neg logit contributions | |
hammer | -0.617 |
� | -0.607 |
swollen | -0.596 |
bruises | -0.595 |
bruised | -0.593 |
Token | yes |
Token ID | 3763 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
� | +0.005 |
bruised | +0.005 |
wrench | +0.005 |
hammer | +0.005 |
swollen | +0.005 |
Neg logit contributions | |
Julia | -0.006 |
La | -0.006 |
Mae | -0.006 |
Amber | -0.005 |
actor | -0.005 |
Token | ; |
Token ID | 26 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
Julia | +0.086 |
La | +0.084 |
Mae | +0.080 |
Amber | +0.079 |
dolphins | +0.077 |
Neg logit contributions | |
� | -0.054 |
wrench | -0.054 |
bruised | -0.054 |
hammer | -0.054 |
pling | -0.051 |
Token | with |
Token ID | 351 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
wrench | +0.065 |
bruised | +0.064 |
hammer | +0.064 |
� | +0.064 |
swollen | +0.062 |
Neg logit contributions | |
Julia | -0.064 |
La | -0.060 |
Mae | -0.058 |
Amber | -0.057 |
actor | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | -6.56e-02 |
Pos logit contributions | |
Julia | +0.056 |
La | +0.050 |
Mae | +0.050 |
Amber | +0.049 |
Zac | +0.047 |
Neg logit contributions | |
� | -0.066 |
bruised | -0.066 |
wrench | -0.065 |
swollen | -0.064 |
hammer | -0.063 |
Token | Sam |
Token ID | 3409 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
hammer | +0.013 |
bruises | +0.013 |
supervision | +0.013 |
� | +0.013 |
bruised | +0.013 |
Neg logit contributions | |
La | -0.008 |
actor | -0.008 |
royal | -0.008 |
mate | -0.008 |
ae | -0.008 |
Token | waved |
Token ID | 26834 |
Feature activation | -1.94e-01 |
Pos logit contributions | |
Julia | +0.151 |
La | +0.139 |
Mae | +0.138 |
Amber | +0.136 |
Lou | +0.129 |
Neg logit contributions | |
bruised | -0.128 |
� | -0.125 |
hammer | -0.123 |
wrench | -0.123 |
bruises | -0.122 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Julia | +0.299 |
La | +0.290 |
Mae | +0.282 |
actor | +0.265 |
Zac | +0.263 |
Neg logit contributions | |
bruised | -0.335 |
wrench | -0.318 |
swollen | -0.316 |
bruises | -0.316 |
� | -0.313 |
Token | and |
Token ID | 290 |
Feature activation | +6.88e-03 |
Pos logit contributions | |
Julia | +0.265 |
La | +0.249 |
Mae | +0.242 |
Amber | +0.237 |
Zac | +0.231 |
Neg logit contributions | |
bruised | -0.218 |
� | -0.218 |
wrench | -0.213 |
hammer | -0.210 |
swollen | -0.209 |
Token | thanked |
Token ID | 26280 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
bruises | +0.010 |
bruised | +0.010 |
swollen | +0.009 |
hammer | +0.009 |
rag | +0.009 |
Neg logit contributions | |
La | -0.011 |
actor | -0.011 |
adies | -0.011 |
oper | -0.011 |
Julia | -0.011 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
� | +0.020 |
hammer | +0.019 |
bruises | +0.019 |
rag | +0.018 |
free | +0.018 |
Neg logit contributions | |
dives | -0.022 |
mate | -0.022 |
royal | -0.021 |
awoke | -0.021 |
diver | -0.021 |
Token | for |
Token ID | 329 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
La | +0.960 |
Julia | +0.952 |
Mae | +0.895 |
actor | +0.891 |
diver | +0.862 |
Neg logit contributions | |
bruised | -0.542 |
� | -0.509 |
pling | -0.506 |
wrench | -0.505 |
hammer | -0.504 |
Token | being |
Token ID | 852 |
Feature activation | +4.03e-03 |
Pos logit contributions | |
� | +0.087 |
wrench | +0.084 |
bruised | +0.082 |
pling | +0.082 |
hammer | +0.081 |
Neg logit contributions | |
Julia | -0.078 |
La | -0.068 |
Mae | -0.068 |
Jeff | -0.067 |
Amber | -0.066 |
Token | so |
Token ID | 523 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
bruised | +0.006 |
swollen | +0.005 |
bruises | +0.005 |
oil | +0.005 |
hammer | +0.005 |
Neg logit contributions | |
Julia | -0.007 |
La | -0.007 |
dances | -0.007 |
dives | -0.007 |
Mae | -0.007 |
Token | clever |
Token ID | 14169 |
Feature activation | +2.28e-02 |
Pos logit contributions | |
Julia | +0.120 |
La | +0.112 |
Mae | +0.111 |
Amber | +0.109 |
April | +0.104 |
Neg logit contributions | |
bruised | -0.103 |
swollen | -0.101 |
bruises | -0.100 |
� | -0.100 |
wrench | -0.099 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-03 |
Pos logit contributions | |
bruised | +0.029 |
� | +0.028 |
wrench | +0.028 |
swollen | +0.027 |
bruises | +0.027 |
Neg logit contributions | |
La | -0.044 |
Julia | -0.043 |
Mae | -0.041 |
Amber | -0.040 |
Zac | -0.040 |
Token | the |
Token ID | 262 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
mate | +0.109 |
actor | +0.108 |
La | +0.106 |
adies | +0.103 |
Zac | +0.103 |
Neg logit contributions | |
bruises | -0.088 |
� | -0.087 |
sugar | -0.086 |
hammer | -0.085 |
swollen | -0.085 |
Token | leaves |
Token ID | 5667 |
Feature activation | +1.35e-01 |
Pos logit contributions | |
bruised | +0.057 |
pling | +0.054 |
supervision | +0.053 |
swollen | +0.052 |
cut | +0.052 |
Neg logit contributions | |
Julia | -0.083 |
Lou | -0.077 |
ia | -0.077 |
Alicia | -0.075 |
Mae | -0.075 |
Token | began |
Token ID | 2540 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
bruised | +0.261 |
wrench | +0.260 |
supervision | +0.256 |
leaning | +0.255 |
bruises | +0.254 |
Neg logit contributions | |
Julia | -0.175 |
La | -0.152 |
Mae | -0.150 |
dives | -0.146 |
Amber | -0.145 |
Token | to |
Token ID | 284 |
Feature activation | +1.93e-03 |
Pos logit contributions | |
Julia | +0.057 |
Mae | +0.048 |
La | +0.048 |
Amber | +0.048 |
Lou | +0.047 |
Neg logit contributions | |
� | -0.071 |
leaning | -0.071 |
bruised | -0.071 |
wrench | -0.070 |
pling | -0.070 |
Token | move |
Token ID | 1445 |
Feature activation | -1.01e-01 |
Pos logit contributions | |
� | +0.004 |
bruises | +0.004 |
bruised | +0.004 |
wrench | +0.004 |
pling | +0.004 |
Neg logit contributions | |
Julia | -0.003 |
La | -0.002 |
Mae | -0.002 |
Amber | -0.002 |
Zac | -0.002 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
Julia | +0.172 |
La | +0.158 |
Mae | +0.155 |
Amber | +0.152 |
Zac | +0.146 |
Neg logit contributions | |
bruised | -0.165 |
� | -0.164 |
wrench | -0.163 |
swollen | -0.160 |
hammer | -0.159 |
Token | smiled |
Token ID | 13541 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
Julia | +0.684 |
La | +0.640 |
Mae | +0.605 |
Amber | +0.600 |
actor | +0.589 |
Neg logit contributions | |
bruised | -0.909 |
� | -0.890 |
hammer | -0.888 |
wrench | -0.881 |
swollen | -0.875 |
Token | . |
Token ID | 13 |
Feature activation | +9.06e-03 |
Pos logit contributions | |
Julia | +0.114 |
La | +0.109 |
Mae | +0.106 |
Amber | +0.102 |
Zac | +0.101 |
Neg logit contributions | |
bruised | -0.081 |
wrench | -0.078 |
� | -0.077 |
swollen | -0.077 |
hammer | -0.076 |
Token | She |
Token ID | 1375 |
Feature activation | +7.20e-03 |
Pos logit contributions | |
� | +0.012 |
hammer | +0.010 |
wrench | +0.009 |
bruised | +0.009 |
bruises | +0.009 |
Neg logit contributions | |
actor | -0.018 |
mate | -0.018 |
dolphins | -0.018 |
La | -0.017 |
diver | -0.017 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
swollen | +0.010 |
hammer | +0.010 |
bruised | +0.010 |
bruises | +0.010 |
� | +0.010 |
Neg logit contributions | |
Julia | -0.013 |
La | -0.013 |
Amber | -0.013 |
Mae | -0.012 |
adies | -0.012 |
Token | brave |
Token ID | 14802 |
Feature activation | -6.73e-02 |
Pos logit contributions | |
Julia | +0.048 |
La | +0.044 |
Mae | +0.043 |
Amber | +0.043 |
Zac | +0.040 |
Neg logit contributions | |
� | -0.056 |
wrench | -0.055 |
bruised | -0.055 |
hammer | -0.054 |
swollen | -0.054 |
Token | me |
Token ID | 502 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
diver | +0.013 |
royal | +0.013 |
actor | +0.012 |
pirate | +0.012 |
La | +0.012 |
Neg logit contributions | |
sugar | -0.008 |
supervision | -0.008 |
knives | -0.007 |
oil | -0.007 |
issors | -0.007 |
Token | about |
Token ID | 546 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
Julia | +0.246 |
La | +0.245 |
Mae | +0.229 |
Zac | +0.224 |
Amber | +0.221 |
Neg logit contributions | |
hammer | -0.163 |
bruised | -0.162 |
wrench | -0.160 |
bruises | -0.148 |
pling | -0.148 |
Token | the |
Token ID | 262 |
Feature activation | +5.56e-02 |
Pos logit contributions | |
Julia | +0.339 |
La | +0.319 |
Mae | +0.309 |
Amber | +0.301 |
Zac | +0.299 |
Neg logit contributions | |
bruised | -0.270 |
� | -0.263 |
wrench | -0.261 |
hammer | -0.261 |
swollen | -0.259 |
Token | crystal |
Token ID | 15121 |
Feature activation | +8.87e-02 |
Pos logit contributions | |
� | +0.097 |
wrench | +0.093 |
bruised | +0.090 |
bruises | +0.089 |
uc | +0.089 |
Neg logit contributions | |
Julia | -0.090 |
Amber | -0.082 |
Mae | -0.081 |
La | -0.080 |
actor | -0.079 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
hammer | +0.121 |
bruised | +0.120 |
� | +0.116 |
pling | +0.115 |
wrench | +0.113 |
Neg logit contributions | |
La | -0.177 |
Julia | -0.176 |
Zac | -0.159 |
Mae | -0.159 |
Amber | -0.155 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
mate | +0.157 |
actor | +0.143 |
diver | +0.137 |
dolphins | +0.136 |
La | +0.135 |
Neg logit contributions | |
� | -0.281 |
nails | -0.258 |
bruises | -0.258 |
powd | -0.256 |
hammer | -0.255 |
Token | went |
Token ID | 1816 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
Julia | +0.884 |
La | +0.870 |
Amber | +0.813 |
Mae | +0.812 |
actor | +0.807 |
Neg logit contributions | |
hammer | -0.624 |
� | -0.591 |
bruised | -0.589 |
uc | -0.588 |
swollen | -0.576 |
Token | closer |
Token ID | 5699 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Julia | +0.104 |
La | +0.100 |
Mae | +0.098 |
Amber | +0.094 |
ae | +0.093 |
Neg logit contributions | |
hammer | -0.086 |
wrench | -0.086 |
bruised | -0.085 |
� | -0.082 |
bruises | -0.079 |
Token | and |
Token ID | 290 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
La | +0.251 |
Julia | +0.251 |
actor | +0.237 |
Mae | +0.233 |
Amber | +0.231 |
Neg logit contributions | |
hammer | -0.153 |
bruised | -0.152 |
� | -0.151 |
wrench | -0.147 |
pling | -0.142 |
Token | saw |
Token ID | 2497 |
Feature activation | -6.08e-03 |
Pos logit contributions | |
bruised | +0.020 |
hammer | +0.019 |
pling | +0.019 |
wrench | +0.018 |
� | +0.018 |
Neg logit contributions | |
La | -0.026 |
Julia | -0.026 |
ia | -0.024 |
actor | -0.024 |
oper | -0.024 |
Token | that |
Token ID | 326 |
Feature activation | +1.39e-01 |
Pos logit contributions | |
Julia | +0.011 |
La | +0.011 |
Mae | +0.010 |
Amber | +0.010 |
Zac | +0.010 |
Neg logit contributions | |
bruised | -0.009 |
hammer | -0.009 |
swollen | -0.009 |
bruises | -0.008 |
wrench | -0.008 |
Token | couldn |
Token ID | 3521 |
Feature activation | +6.98e-02 |
Pos logit contributions | |
Julia | +0.092 |
La | +0.091 |
Mae | +0.087 |
oper | +0.086 |
Zac | +0.086 |
Neg logit contributions | |
� | -0.057 |
bruised | -0.056 |
hammer | -0.056 |
swollen | -0.055 |
pling | -0.055 |
Token | 't |
Token ID | 470 |
Feature activation | +2.93e-02 |
Pos logit contributions | |
� | +0.089 |
hammer | +0.066 |
swollen | +0.063 |
wrench | +0.063 |
therm | +0.061 |
Neg logit contributions | |
Julia | -0.162 |
mate | -0.155 |
actor | -0.154 |
dolphins | -0.153 |
La | -0.153 |
Token | see |
Token ID | 766 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
hammer | +0.046 |
� | +0.044 |
wrench | +0.042 |
bruised | +0.041 |
pling | +0.040 |
Neg logit contributions | |
Julia | -0.049 |
La | -0.048 |
Zac | -0.047 |
Amber | -0.044 |
dances | -0.044 |
Token | anyone |
Token ID | 2687 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
Julia | +0.209 |
La | +0.199 |
Mae | +0.190 |
Amber | +0.188 |
Zac | +0.185 |
Neg logit contributions | |
bruised | -0.213 |
swollen | -0.201 |
hammer | -0.200 |
bruises | -0.199 |
� | -0.197 |
Token | . |
Token ID | 13 |
Feature activation | +8.92e-03 |
Pos logit contributions | |
Julia | +0.326 |
La | +0.317 |
Mae | +0.307 |
Amber | +0.301 |
Zac | +0.296 |
Neg logit contributions | |
bruised | -0.239 |
hammer | -0.229 |
swollen | -0.219 |
wrench | -0.218 |
� | -0.216 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
� | +0.010 |
nails | +0.008 |
hammer | +0.008 |
bruises | +0.008 |
issors | +0.008 |
Neg logit contributions | |
actor | -0.018 |
dolphins | -0.017 |
diver | -0.017 |
mate | -0.017 |
La | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -4.39e-02 |
Pos logit contributions | |
Julia | +0.942 |
La | +0.906 |
Mae | +0.869 |
Zac | +0.863 |
Amber | +0.855 |
Neg logit contributions | |
hammer | -0.599 |
bruised | -0.579 |
� | -0.570 |
swollen | -0.570 |
pling | -0.563 |
Token | scared |
Token ID | 12008 |
Feature activation | +2.84e-04 |
Pos logit contributions | |
La | +0.093 |
Julia | +0.092 |
Mae | +0.087 |
actor | +0.086 |
Amber | +0.085 |
Neg logit contributions | |
bruised | -0.055 |
swollen | -0.049 |
struggling | -0.047 |
bruises | -0.047 |
hammer | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | +4.45e-04 |
Pos logit contributions | |
bruised | +0.000 |
wrench | +0.000 |
hammer | +0.000 |
swollen | +0.000 |
� | +0.000 |
Neg logit contributions | |
Julia | -0.001 |
La | -0.001 |
Mae | -0.001 |
Zac | -0.001 |
Amber | -0.001 |
Token | so |
Token ID | 523 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
bruised | +0.001 |
� | +0.001 |
wrench | +0.001 |
swollen | +0.001 |
hammer | +0.001 |
Neg logit contributions | |
Julia | -0.001 |
La | -0.001 |
Mae | -0.001 |
Amber | -0.001 |
Zac | -0.001 |
Token | she |
Token ID | 673 |
Feature activation | -4.55e-02 |
Pos logit contributions | |
mate | +0.129 |
actor | +0.125 |
La | +0.122 |
oper | +0.118 |
aliens | +0.118 |
Neg logit contributions | |
bruises | -0.053 |
bruised | -0.051 |
swollen | -0.049 |
hammer | -0.048 |
� | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | +8.15e-04 |
Pos logit contributions | |
� | +0.009 |
hammer | +0.008 |
wrench | +0.008 |
bruised | +0.007 |
uc | +0.007 |
Neg logit contributions | |
Julia | -0.018 |
La | -0.018 |
actor | -0.017 |
oper | -0.017 |
echo | -0.017 |
Token | a |
Token ID | 257 |
Feature activation | -6.38e-02 |
Pos logit contributions | |
� | +0.001 |
bruised | +0.001 |
wrench | +0.001 |
hammer | +0.001 |
swollen | +0.001 |
Neg logit contributions | |
Julia | -0.001 |
La | -0.001 |
Mae | -0.001 |
Amber | -0.001 |
actor | -0.001 |
Token | little |
Token ID | 1310 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
Julia | +0.109 |
La | +0.101 |
Mae | +0.099 |
Amber | +0.097 |
Zac | +0.093 |
Neg logit contributions | |
bruised | -0.104 |
� | -0.103 |
wrench | -0.102 |
swollen | -0.101 |
bruises | -0.100 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.25e-03 |
Pos logit contributions | |
� | +0.043 |
bruised | +0.043 |
bruises | +0.042 |
swollen | +0.042 |
wrench | +0.042 |
Neg logit contributions | |
Julia | -0.038 |
La | -0.035 |
Mae | -0.034 |
Amber | -0.034 |
dolphins | -0.033 |
Token | named |
Token ID | 3706 |
Feature activation | +9.90e-03 |
Pos logit contributions | |
Julia | +0.008 |
La | +0.007 |
Mae | +0.007 |
Amber | +0.007 |
Zac | +0.007 |
Neg logit contributions | |
bruised | -0.007 |
� | -0.006 |
hammer | -0.006 |
wrench | -0.006 |
bruises | -0.006 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
bruised | +0.017 |
� | +0.017 |
wrench | +0.016 |
hammer | +0.016 |
swollen | +0.016 |
Neg logit contributions | |
Julia | -0.016 |
La | -0.015 |
Mae | -0.015 |
Amber | -0.014 |
Zac | -0.014 |
Token | went |
Token ID | 1816 |
Feature activation | -5.76e-02 |
Pos logit contributions | |
Julia | +0.833 |
La | +0.761 |
Zac | +0.756 |
Mae | +0.751 |
Amber | +0.738 |
Neg logit contributions | |
bruised | -0.753 |
hammer | -0.749 |
� | -0.741 |
bruises | -0.721 |
pling | -0.718 |
Token | exploring |
Token ID | 13504 |
Feature activation | -5.71e-02 |
Pos logit contributions | |
Julia | +0.112 |
La | +0.103 |
Mae | +0.103 |
Amber | +0.100 |
Zac | +0.098 |
Neg logit contributions | |
hammer | -0.083 |
wrench | -0.083 |
bruised | -0.082 |
� | -0.080 |
bruises | -0.078 |
Token | in |
Token ID | 287 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
Julia | +0.091 |
La | +0.082 |
Mae | +0.081 |
Amber | +0.080 |
dolphins | +0.076 |
Neg logit contributions | |
bruised | -0.102 |
� | -0.102 |
wrench | -0.101 |
swollen | -0.100 |
bruises | -0.098 |
Token | the |
Token ID | 262 |
Feature activation | +5.03e-02 |
Pos logit contributions | |
� | +0.018 |
bruised | +0.018 |
wrench | +0.018 |
swollen | +0.017 |
hammer | +0.017 |
Neg logit contributions | |
Julia | -0.019 |
La | -0.017 |
Mae | -0.017 |
Amber | -0.017 |
April | -0.016 |
Token | woods |
Token ID | 16479 |
Feature activation | -7.06e-02 |
Pos logit contributions | |
rag | +0.069 |
hammer | +0.067 |
oil | +0.065 |
bruised | +0.064 |
therm | +0.063 |
Neg logit contributions | |
Julia | -0.101 |
Zac | -0.096 |
Mae | -0.095 |
Jeff | -0.091 |
ae | -0.091 |
Token | the |
Token ID | 262 |
Feature activation | +5.49e-02 |
Pos logit contributions | |
Julia | +0.283 |
La | +0.268 |
Mae | +0.263 |
Amber | +0.257 |
Zac | +0.250 |
Neg logit contributions | |
bruised | -0.182 |
hammer | -0.181 |
wrench | -0.180 |
bruises | -0.175 |
� | -0.174 |
Token | creek |
Token ID | 39459 |
Feature activation | -9.60e-02 |
Pos logit contributions | |
hammer | +0.089 |
wrench | +0.087 |
bruised | +0.086 |
� | +0.086 |
oil | +0.085 |
Neg logit contributions | |
Julia | -0.097 |
La | -0.088 |
Mae | -0.088 |
Amber | -0.085 |
Lou | -0.082 |
Token | . |
Token ID | 13 |
Feature activation | +3.04e-03 |
Pos logit contributions | |
Julia | +0.185 |
La | +0.180 |
Mae | +0.168 |
Amber | +0.166 |
April | +0.163 |
Neg logit contributions | |
� | -0.136 |
bruised | -0.135 |
hammer | -0.135 |
wrench | -0.133 |
bruises | -0.126 |
Token | He |
Token ID | 679 |
Feature activation | -2.16e-02 |
Pos logit contributions | |
wrench | +0.004 |
hammer | +0.004 |
nails | +0.003 |
rag | +0.003 |
screws | +0.003 |
Neg logit contributions | |
La | -0.006 |
dolphins | -0.006 |
Amber | -0.005 |
oper | -0.005 |
diver | -0.005 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
Julia | +0.045 |
La | +0.043 |
oper | +0.041 |
Amber | +0.041 |
Mae | +0.041 |
Neg logit contributions | |
hammer | -0.026 |
wrench | -0.025 |
rag | -0.025 |
oil | -0.025 |
pling | -0.024 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
La | +0.043 |
Amber | +0.040 |
diver | +0.039 |
royal | +0.039 |
ae | +0.038 |
Neg logit contributions | |
hammer | -0.047 |
wrench | -0.045 |
tools | -0.045 |
screws | -0.043 |
nails | -0.042 |
Token | if |
Token ID | 611 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
La | +0.996 |
Julia | +0.990 |
Mae | +0.905 |
actor | +0.899 |
Amber | +0.884 |
Neg logit contributions | |
hammer | -0.568 |
wrench | -0.553 |
bruised | -0.544 |
pling | -0.525 |
supervision | -0.507 |
Token | she |
Token ID | 673 |
Feature activation | -4.29e-02 |
Pos logit contributions | |
hammer | +0.035 |
wrench | +0.032 |
knives | +0.032 |
tools | +0.030 |
bruises | +0.030 |
Neg logit contributions | |
La | -0.076 |
echo | -0.073 |
ack | -0.072 |
pir | -0.071 |
oper | -0.071 |
Token | would |
Token ID | 561 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
La | +0.089 |
Julia | +0.088 |
oper | +0.082 |
Amber | +0.081 |
Mae | +0.081 |
Neg logit contributions | |
hammer | -0.054 |
bruised | -0.052 |
� | -0.051 |
wrench | -0.050 |
pling | -0.050 |
Token | lend |
Token ID | 22096 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Julia | +0.042 |
La | +0.041 |
Mae | +0.038 |
Amber | +0.038 |
Zac | +0.037 |
Neg logit contributions | |
� | -0.039 |
hammer | -0.039 |
wrench | -0.038 |
bruised | -0.038 |
bruises | -0.036 |
Token | him |
Token ID | 683 |
Feature activation | +4.20e-02 |
Pos logit contributions | |
gum | +0.153 |
bruises | +0.146 |
hammer | +0.146 |
bruised | +0.141 |
handcuffs | +0.139 |
Neg logit contributions | |
Julia | -0.288 |
La | -0.285 |
actor | -0.278 |
diver | -0.276 |
ae | -0.274 |
Token | |
Token ID | 220 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
� | +0.011 |
issors | +0.009 |
leaning | +0.008 |
bruises | +0.008 |
hammer | +0.008 |
Neg logit contributions | |
actor | -0.020 |
La | -0.019 |
diver | -0.019 |
dolphins | -0.019 |
oper | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -2.51e-03 |
Pos logit contributions | |
� | +0.141 |
wrench | +0.126 |
hammer | +0.125 |
bruises | +0.122 |
bruised | +0.122 |
Neg logit contributions | |
actor | -0.197 |
La | -0.196 |
dolphins | -0.195 |
Julia | -0.194 |
oper | -0.187 |
Token |
|
Token ID | 198 |
Feature activation | -2.54e-03 |
Pos logit contributions | |
actor | +0.005 |
aliens | +0.005 |
diver | +0.005 |
mate | +0.005 |
dolphins | +0.005 |
Neg logit contributions | |
� | -0.002 |
pling | -0.002 |
leaning | -0.002 |
Here | -0.002 |
hammer | -0.002 |
Token | Together |
Token ID | 41631 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
aliens | +0.005 |
Dragon | +0.005 |
actor | +0.005 |
mate | +0.005 |
dolphins | +0.005 |
Neg logit contributions | |
leaning | -0.002 |
� | -0.002 |
issors | -0.002 |
Here | -0.002 |
ised | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | -1.41e-03 |
Pos logit contributions | |
� | +0.036 |
wrench | +0.036 |
hammer | +0.035 |
bruised | +0.034 |
bruises | +0.034 |
Neg logit contributions | |
Julia | -0.050 |
La | -0.050 |
Mae | -0.046 |
actor | -0.046 |
Amber | -0.046 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
mate | +0.002 |
La | +0.002 |
aliens | +0.002 |
villains | +0.002 |
actor | +0.002 |
Neg logit contributions | |
� | -0.002 |
uc | -0.002 |
mons | -0.002 |
ised | -0.002 |
hammer | -0.002 |
Token | and |
Token ID | 290 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
Julia | +0.865 |
La | +0.812 |
Mae | +0.780 |
Amber | +0.778 |
Zac | +0.756 |
Neg logit contributions | |
hammer | -0.736 |
wrench | -0.729 |
� | -0.721 |
bruised | -0.702 |
pling | -0.700 |
Token | her |
Token ID | 607 |
Feature activation | -9.85e-03 |
Pos logit contributions | |
hammer | +0.016 |
wrench | +0.016 |
bruised | +0.015 |
swollen | +0.015 |
bruises | +0.015 |
Neg logit contributions | |
Julia | -0.023 |
La | -0.023 |
Mae | -0.022 |
ae | -0.022 |
Zac | -0.022 |
Token | friends |
Token ID | 2460 |
Feature activation | -5.88e-02 |
Pos logit contributions | |
Julia | +0.018 |
La | +0.017 |
Mae | +0.017 |
Amber | +0.016 |
Zac | +0.016 |
Neg logit contributions | |
hammer | -0.015 |
� | -0.015 |
wrench | -0.015 |
bruised | -0.015 |
bruises | -0.014 |
Token | worked |
Token ID | 3111 |
Feature activation | +1.55e-01 |
Pos logit contributions | |
Julia | +0.092 |
Mae | +0.083 |
La | +0.082 |
Amber | +0.080 |
Jeff | +0.079 |
Neg logit contributions | |
bruised | -0.106 |
swollen | -0.105 |
� | -0.104 |
bruises | -0.102 |
wrench | -0.100 |
Token | hard |
Token ID | 1327 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
bruised | +0.296 |
swollen | +0.287 |
bruises | +0.281 |
� | +0.281 |
wrench | +0.278 |
Neg logit contributions | |
Julia | -0.238 |
La | -0.209 |
Mae | -0.207 |
Amber | -0.205 |
Jeff | -0.201 |
Token | to |
Token ID | 284 |
Feature activation | +4.70e-03 |
Pos logit contributions | |
Julia | +0.031 |
La | +0.029 |
Mae | +0.028 |
Amber | +0.028 |
Zac | +0.027 |
Neg logit contributions | |
bruised | -0.027 |
� | -0.027 |
wrench | -0.027 |
hammer | -0.026 |
bruises | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
wrench | +0.007 |
hammer | +0.007 |
bruised | +0.007 |
� | +0.007 |
swollen | +0.007 |
Neg logit contributions | |
Julia | -0.008 |
La | -0.008 |
Mae | -0.008 |
Amber | -0.007 |
Zac | -0.007 |
Token | town |
Token ID | 3240 |
Feature activation | +1.08e-02 |
Pos logit contributions | |
bruised | +0.073 |
� | +0.073 |
wrench | +0.073 |
hammer | +0.072 |
swollen | +0.071 |
Neg logit contributions | |
Julia | -0.070 |
La | -0.064 |
Mae | -0.063 |
Amber | -0.062 |
Zac | -0.059 |
Token | ," |
Token ID | 553 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
hammer | +0.015 |
bruised | +0.015 |
wrench | +0.015 |
� | +0.014 |
bruises | +0.014 |
Neg logit contributions | |
Julia | -0.021 |
La | -0.020 |
Mae | -0.019 |
Amber | -0.019 |
Zac | -0.018 |
Token | pleaded |
Token ID | 16407 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
bruises | +0.017 |
hammer | +0.016 |
swollen | +0.016 |
wrench | +0.016 |
� | +0.016 |
Neg logit contributions | |
La | -0.027 |
Julia | -0.026 |
actor | -0.025 |
Mae | -0.025 |
mate | -0.025 |
Token | Julia |
Token ID | 22300 |
Feature activation | -4.77e-01 |
Pos logit contributions | |
hammer | +0.181 |
wrench | +0.162 |
uc | +0.161 |
bruises | +0.157 |
nails | +0.157 |
Neg logit contributions | |
La | -0.223 |
awoke | -0.216 |
actor | -0.213 |
royal | -0.211 |
mate | -0.207 |
Token | . |
Token ID | 13 |
Feature activation | +8.78e-03 |
Pos logit contributions | |
La | +0.725 |
Julia | +0.716 |
dolphins | +0.637 |
Amber | +0.634 |
Mae | +0.627 |
Neg logit contributions | |
bruised | -0.851 |
hammer | -0.838 |
wrench | -0.827 |
bruises | -0.818 |
swollen | -0.807 |
Token | |
Token ID | 220 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
� | +0.011 |
hammer | +0.010 |
wrench | +0.010 |
bruised | +0.010 |
bruises | +0.010 |
Neg logit contributions | |
Julia | -0.018 |
La | -0.018 |
actor | -0.017 |
Mae | -0.017 |
dolphins | -0.017 |
Token |
|
Token ID | 198 |
Feature activation | -2.43e-03 |
Pos logit contributions | |
� | +0.141 |
wrench | +0.130 |
hammer | +0.129 |
bruised | +0.128 |
bruises | +0.128 |
Neg logit contributions | |
La | -0.196 |
Julia | -0.196 |
actor | -0.194 |
dolphins | -0.189 |
Mae | -0.185 |
Token |
|
Token ID | 198 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
actor | +0.005 |
dolphins | +0.005 |
diver | +0.005 |
Julia | +0.005 |
mate | +0.005 |
Neg logit contributions | |
� | -0.003 |
hammer | -0.002 |
leaning | -0.002 |
bruises | -0.002 |
issors | -0.002 |
Token | Her |
Token ID | 9360 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
dolphins | +0.004 |
Julia | +0.004 |
Zac | +0.004 |
actor | +0.004 |
Dragon | +0.004 |
Neg logit contributions | |
leaning | -0.002 |
� | -0.002 |
issors | -0.002 |
ised | -0.002 |
bruises | -0.002 |
Token | had |
Token ID | 550 |
Feature activation | +3.00e-03 |
Pos logit contributions | |
Julia | +0.113 |
Mae | +0.099 |
Jeff | +0.098 |
Lou | +0.098 |
mate | +0.095 |
Neg logit contributions | |
� | -0.188 |
wrench | -0.179 |
swollen | -0.178 |
bruised | -0.174 |
bruises | -0.172 |
Token | to |
Token ID | 284 |
Feature activation | +1.55e-03 |
Pos logit contributions | |
� | +0.006 |
wrench | +0.006 |
pling | +0.006 |
bruised | +0.006 |
swollen | +0.006 |
Neg logit contributions | |
Julia | -0.004 |
Mae | -0.003 |
La | -0.003 |
mate | -0.003 |
dolphins | -0.003 |
Token | be |
Token ID | 307 |
Feature activation | -8.33e-02 |
Pos logit contributions | |
bruised | +0.002 |
swollen | +0.002 |
� | +0.002 |
pling | +0.002 |
wrench | +0.002 |
Neg logit contributions | |
Julia | -0.003 |
La | -0.002 |
mate | -0.002 |
Mae | -0.002 |
Lou | -0.002 |
Token | brave |
Token ID | 14802 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
Julia | +0.135 |
La | +0.125 |
mate | +0.120 |
Mae | +0.119 |
Amber | +0.119 |
Neg logit contributions | |
� | -0.146 |
wrench | -0.132 |
pling | -0.131 |
bruised | -0.130 |
swollen | -0.129 |
Token | and |
Token ID | 290 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
Julia | +0.126 |
La | +0.119 |
Mae | +0.115 |
Amber | +0.113 |
Zac | +0.109 |
Neg logit contributions | |
bruised | -0.104 |
wrench | -0.101 |
� | -0.099 |
swollen | -0.098 |
bruises | -0.098 |
Token | help |
Token ID | 1037 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
� | +0.016 |
bruised | +0.015 |
wrench | +0.015 |
swollen | +0.015 |
pling | +0.014 |
Neg logit contributions | |
Julia | -0.022 |
La | -0.020 |
Mae | -0.020 |
Amber | -0.019 |
Lou | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +4.59e-02 |
Pos logit contributions | |
hammer | +0.333 |
bruised | +0.331 |
wrench | +0.331 |
bruises | +0.315 |
oil | +0.309 |
Neg logit contributions | |
Julia | -0.470 |
La | -0.463 |
Amber | -0.436 |
Mae | -0.434 |
actor | -0.421 |
Token | animal |
Token ID | 5044 |
Feature activation | -7.75e-02 |
Pos logit contributions | |
� | +0.088 |
bruised | +0.088 |
wrench | +0.088 |
swollen | +0.086 |
bruises | +0.085 |
Neg logit contributions | |
Julia | -0.066 |
La | -0.060 |
Mae | -0.058 |
Amber | -0.057 |
Lou | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +2.07e-03 |
Pos logit contributions | |
Julia | +0.130 |
La | +0.122 |
Mae | +0.118 |
Amber | +0.116 |
Zac | +0.111 |
Neg logit contributions | |
bruised | -0.130 |
� | -0.129 |
wrench | -0.128 |
hammer | -0.127 |
swollen | -0.126 |
Token | They |
Token ID | 1119 |
Feature activation | -2.74e-02 |
Pos logit contributions | |
� | +0.003 |
bruises | +0.002 |
bruised | +0.002 |
wrench | +0.002 |
oil | +0.002 |
Neg logit contributions | |
La | -0.004 |
actor | -0.004 |
Julia | -0.004 |
oper | -0.004 |
royal | -0.004 |
Token | used |
Token ID | 973 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
Julia | +0.038 |
Mae | +0.033 |
Lou | +0.033 |
mate | +0.033 |
Amber | +0.033 |
Neg logit contributions | |
� | -0.054 |
bruised | -0.052 |
wrench | -0.051 |
swollen | -0.051 |
supervision | -0.050 |
Token | went |
Token ID | 1816 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
wrench | +0.265 |
bruised | +0.264 |
� | +0.262 |
hammer | +0.262 |
bruises | +0.253 |
Neg logit contributions | |
Julia | -0.258 |
La | -0.240 |
Mae | -0.234 |
Amber | -0.231 |
Zac | -0.218 |
Token | to |
Token ID | 284 |
Feature activation | +1.43e-02 |
Pos logit contributions | |
Julia | +0.084 |
La | +0.077 |
Mae | +0.076 |
Amber | +0.075 |
Zac | +0.072 |
Neg logit contributions | |
bruised | -0.081 |
� | -0.080 |
wrench | -0.079 |
swollen | -0.078 |
bruises | -0.078 |
Token | a |
Token ID | 257 |
Feature activation | -5.52e-02 |
Pos logit contributions | |
hammer | +0.021 |
wrench | +0.021 |
� | +0.020 |
bruised | +0.020 |
bruises | +0.020 |
Neg logit contributions | |
Julia | -0.027 |
La | -0.026 |
Mae | -0.025 |
Amber | -0.025 |
Zac | -0.023 |
Token | store |
Token ID | 3650 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
Julia | +0.099 |
La | +0.090 |
Mae | +0.090 |
Amber | +0.088 |
Zac | +0.085 |
Neg logit contributions | |
wrench | -0.089 |
hammer | -0.088 |
� | -0.086 |
bruised | -0.086 |
swollen | -0.082 |
Token | to |
Token ID | 284 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
� | +0.276 |
bruised | +0.275 |
swollen | +0.274 |
wrench | +0.269 |
bruises | +0.265 |
Neg logit contributions | |
Julia | -0.238 |
Mae | -0.210 |
La | -0.209 |
Amber | -0.207 |
Jeff | -0.198 |
Token | buy |
Token ID | 2822 |
Feature activation | +2.58e-01 |
Pos logit contributions | |
bruised | +0.018 |
swollen | +0.018 |
� | +0.018 |
wrench | +0.017 |
bruises | +0.017 |
Neg logit contributions | |
Julia | -0.031 |
La | -0.029 |
Mae | -0.029 |
Amber | -0.029 |
actor | -0.028 |
Token | some |
Token ID | 617 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
� | +0.501 |
bruised | +0.494 |
swollen | +0.490 |
wrench | +0.474 |
bruises | +0.469 |
Neg logit contributions | |
Julia | -0.369 |
La | -0.327 |
Mae | -0.322 |
Amber | -0.318 |
dolphins | -0.314 |
Token | things |
Token ID | 1243 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
� | +0.050 |
swollen | +0.049 |
bruised | +0.048 |
pling | +0.045 |
bruises | +0.044 |
Neg logit contributions | |
Julia | -0.053 |
dolphins | -0.051 |
actor | -0.050 |
La | -0.049 |
oper | -0.049 |
Token | . |
Token ID | 13 |
Feature activation | +1.97e-02 |
Pos logit contributions | |
bruised | +0.290 |
� | +0.289 |
swollen | +0.284 |
wrench | +0.284 |
bruises | +0.280 |
Neg logit contributions | |
Julia | -0.271 |
La | -0.244 |
Mae | -0.242 |
Amber | -0.239 |
Zac | -0.230 |
Token | The |
Token ID | 383 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
� | +0.029 |
wrench | +0.029 |
bruised | +0.028 |
hammer | +0.028 |
bruises | +0.027 |
Neg logit contributions | |
Julia | -0.037 |
La | -0.035 |
Mae | -0.034 |
Amber | -0.033 |
Zac | -0.032 |
Token | store |
Token ID | 3650 |
Feature activation | +1.53e-01 |
Pos logit contributions | |
swollen | +0.175 |
bruised | +0.167 |
uc | +0.161 |
pling | +0.161 |
� | +0.160 |
Neg logit contributions | |
pirate | -0.106 |
mate | -0.105 |
princess | -0.102 |
actor | -0.102 |
theater | -0.101 |
Token | if |
Token ID | 611 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
Julia | +0.065 |
La | +0.060 |
Mae | +0.059 |
Amber | +0.058 |
Zac | +0.056 |
Neg logit contributions | |
bruised | -0.056 |
� | -0.055 |
wrench | -0.055 |
swollen | -0.054 |
hammer | -0.053 |
Token | he |
Token ID | 339 |
Feature activation | -6.10e-02 |
Pos logit contributions | |
bruised | +0.058 |
swollen | +0.057 |
wrench | +0.057 |
hammer | +0.056 |
bruises | +0.055 |
Neg logit contributions | |
Julia | -0.069 |
La | -0.067 |
Mae | -0.064 |
actor | -0.061 |
ae | -0.061 |
Token | could |
Token ID | 714 |
Feature activation | +9.76e-03 |
Pos logit contributions | |
Julia | +0.118 |
La | +0.113 |
Mae | +0.108 |
Amber | +0.107 |
April | +0.105 |
Neg logit contributions | |
bruised | -0.088 |
hammer | -0.085 |
� | -0.085 |
wrench | -0.084 |
bruises | -0.082 |
Token | borrow |
Token ID | 8804 |
Feature activation | +9.23e-02 |
Pos logit contributions | |
bruised | +0.017 |
wrench | +0.017 |
swollen | +0.017 |
� | +0.016 |
bruises | +0.016 |
Neg logit contributions | |
Julia | -0.015 |
La | -0.014 |
Mae | -0.014 |
Amber | -0.014 |
Zac | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | -6.86e-03 |
Pos logit contributions | |
� | +0.176 |
wrench | +0.170 |
bruised | +0.168 |
pling | +0.166 |
hammer | +0.165 |
Neg logit contributions | |
Julia | -0.137 |
La | -0.124 |
Amber | -0.122 |
Mae | -0.120 |
Zac | -0.118 |
Token | comb |
Token ID | 1974 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
Julia | +0.012 |
La | +0.011 |
Mae | +0.010 |
Amber | +0.010 |
Zac | +0.010 |
Neg logit contributions | |
� | -0.011 |
wrench | -0.011 |
bruised | -0.011 |
bruises | -0.011 |
swollen | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +3.65e-03 |
Pos logit contributions | |
� | +0.413 |
bruised | +0.410 |
wrench | +0.406 |
swollen | +0.402 |
bruises | +0.399 |
Neg logit contributions | |
Julia | -0.413 |
La | -0.375 |
Mae | -0.370 |
Amber | -0.366 |
Zac | -0.351 |
Token | comb |
Token ID | 1974 |
Feature activation | +2.46e-01 |
Pos logit contributions | |
� | +0.006 |
bruised | +0.006 |
wrench | +0.006 |
swollen | +0.006 |
pling | +0.006 |
Neg logit contributions | |
Julia | -0.006 |
La | -0.005 |
Mae | -0.005 |
Amber | -0.005 |
Zac | -0.005 |
Token | his |
Token ID | 465 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
� | +0.441 |
bruised | +0.432 |
wrench | +0.432 |
bruises | +0.424 |
swollen | +0.421 |
Neg logit contributions | |
Julia | -0.393 |
La | -0.355 |
Mae | -0.351 |
Amber | -0.345 |
Zac | -0.333 |
Token | fur |
Token ID | 9230 |
Feature activation | +4.48e-02 |
Pos logit contributions | |
Julia | +0.081 |
La | +0.076 |
Mae | +0.073 |
Amber | +0.072 |
April | +0.071 |
Neg logit contributions | |
� | -0.085 |
wrench | -0.082 |
bruised | -0.082 |
pling | -0.081 |
leaning | -0.081 |
Token | and |
Token ID | 290 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
� | +0.079 |
bruises | +0.079 |
bruised | +0.079 |
swollen | +0.079 |
wrench | +0.078 |
Neg logit contributions | |
Julia | -0.068 |
Mae | -0.061 |
Amber | -0.060 |
La | -0.060 |
Zac | -0.057 |
Token | cat |
Token ID | 3797 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
wrench | +0.212 |
hammer | +0.212 |
bruised | +0.207 |
� | +0.203 |
rag | +0.197 |
Neg logit contributions | |
Julia | -0.281 |
La | -0.265 |
Mae | -0.254 |
Amber | -0.251 |
Zac | -0.244 |
Token | named |
Token ID | 3706 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
bruised | +0.077 |
hammer | +0.076 |
wrench | +0.074 |
� | +0.074 |
swollen | +0.073 |
Neg logit contributions | |
Julia | -0.109 |
La | -0.102 |
Mae | -0.100 |
Amber | -0.099 |
Zac | -0.097 |
Token | Mitt |
Token ID | 16627 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
bruised | +0.015 |
� | +0.015 |
wrench | +0.014 |
swollen | +0.014 |
hammer | +0.014 |
Neg logit contributions | |
Julia | -0.021 |
La | -0.020 |
Mae | -0.019 |
Amber | -0.019 |
Zac | -0.019 |
Token | ens |
Token ID | 641 |
Feature activation | +2.53e-01 |
Pos logit contributions | |
bruised | +0.238 |
wrench | +0.237 |
swollen | +0.231 |
� | +0.230 |
bruises | +0.229 |
Neg logit contributions | |
Julia | -0.591 |
La | -0.572 |
Mae | -0.564 |
Amber | -0.557 |
Alicia | -0.545 |
Token | . |
Token ID | 13 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
bruised | +0.416 |
� | +0.416 |
wrench | +0.412 |
swollen | +0.405 |
hammer | +0.401 |
Neg logit contributions | |
Julia | -0.431 |
La | -0.396 |
Mae | -0.390 |
Amber | -0.383 |
Zac | -0.368 |
Token | Mitt |
Token ID | 16627 |
Feature activation | +2.54e-01 |
Pos logit contributions | |
cut | +0.020 |
isher | +0.020 |
supervision | +0.020 |
pling | +0.018 |
plug | +0.018 |
Neg logit contributions | |
Ro | -0.024 |
Julia | -0.023 |
Mae | -0.023 |
La | -0.023 |
Leo | -0.022 |
Token | ens |
Token ID | 641 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
wrench | +0.276 |
bruised | +0.262 |
hammer | +0.262 |
swollen | +0.254 |
bruises | +0.253 |
Neg logit contributions | |
Julia | -0.556 |
La | -0.541 |
Mae | -0.533 |
ia | -0.525 |
Amber | -0.521 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.01e-02 |
Pos logit contributions | |
bruised | +0.390 |
� | +0.383 |
hammer | +0.382 |
swollen | +0.381 |
wrench | +0.378 |
Neg logit contributions | |
Julia | -0.453 |
La | -0.415 |
Mae | -0.406 |
Amber | -0.404 |
Zac | -0.389 |
Token | to |
Token ID | 284 |
Feature activation | +4.27e-03 |
Pos logit contributions | |
� | +0.019 |
bruised | +0.018 |
wrench | +0.018 |
swollen | +0.018 |
pling | +0.018 |
Neg logit contributions | |
Julia | -0.015 |
La | -0.014 |
Mae | -0.014 |
Amber | -0.013 |
dolphins | -0.013 |
Token | play |
Token ID | 711 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
hammer | +0.007 |
bruised | +0.007 |
� | +0.007 |
wrench | +0.007 |
bruises | +0.007 |
Neg logit contributions | |
Julia | -0.007 |
La | -0.006 |
Amber | -0.006 |
Mae | -0.006 |
Zac | -0.006 |
Token | with |
Token ID | 351 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
Julia | +0.037 |
La | +0.035 |
Mae | +0.033 |
Amber | +0.033 |
Jeff | +0.032 |
Neg logit contributions | |
bruised | -0.047 |
� | -0.046 |
wrench | -0.046 |
swollen | -0.046 |
bruises | -0.045 |
Token | down |
Token ID | 866 |
Feature activation | +5.19e-02 |
Pos logit contributions | |
Julia | +0.156 |
La | +0.139 |
Amber | +0.137 |
Mae | +0.133 |
dolphins | +0.131 |
Neg logit contributions | |
� | -0.225 |
supervision | -0.220 |
wrench | -0.212 |
rag | -0.209 |
bruises | -0.209 |
Token | and |
Token ID | 290 |
Feature activation | +1.72e-02 |
Pos logit contributions | |
bruised | +0.095 |
� | +0.094 |
wrench | +0.094 |
swollen | +0.094 |
bruises | +0.092 |
Neg logit contributions | |
Julia | -0.078 |
Mae | -0.070 |
La | -0.069 |
Amber | -0.068 |
Zac | -0.064 |
Token | looked |
Token ID | 3114 |
Feature activation | -3.39e-02 |
Pos logit contributions | |
� | +0.031 |
wrench | +0.030 |
bruised | +0.030 |
hammer | +0.029 |
swollen | +0.029 |
Neg logit contributions | |
Julia | -0.027 |
La | -0.025 |
Mae | -0.024 |
Amber | -0.024 |
Zac | -0.023 |
Token | in |
Token ID | 287 |
Feature activation | +1.42e-02 |
Pos logit contributions | |
Julia | +0.054 |
La | +0.049 |
Mae | +0.049 |
Amber | +0.048 |
dolphins | +0.046 |
Neg logit contributions | |
� | -0.061 |
wrench | -0.058 |
bruised | -0.058 |
bruises | -0.058 |
swollen | -0.058 |
Token | his |
Token ID | 465 |
Feature activation | -4.70e-02 |
Pos logit contributions | |
� | +0.031 |
pling | +0.031 |
uc | +0.030 |
mons | +0.030 |
wrench | +0.030 |
Neg logit contributions | |
Julia | -0.020 |
Mae | -0.016 |
Amber | -0.016 |
Jeff | -0.016 |
La | -0.015 |
Token | sack |
Token ID | 23704 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
Julia | +0.083 |
Amber | +0.078 |
royal | +0.073 |
La | +0.073 |
April | +0.072 |
Neg logit contributions | |
pling | -0.071 |
� | -0.070 |
bruised | -0.069 |
swollen | -0.069 |
free | -0.068 |
Token | . |
Token ID | 13 |
Feature activation | +1.40e-02 |
Pos logit contributions | |
� | +0.557 |
bruised | +0.549 |
swollen | +0.548 |
bruises | +0.542 |
wrench | +0.540 |
Neg logit contributions | |
Julia | -0.422 |
Mae | -0.369 |
Amber | -0.365 |
La | -0.355 |
Jeff | -0.352 |
Token | The |
Token ID | 383 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
� | +0.024 |
bruised | +0.024 |
wrench | +0.024 |
hammer | +0.023 |
swollen | +0.023 |
Neg logit contributions | |
Julia | -0.023 |
La | -0.021 |
Mae | -0.021 |
Amber | -0.020 |
Zac | -0.019 |
Token | toy |
Token ID | 13373 |
Feature activation | +6.85e-02 |
Pos logit contributions | |
pling | +0.130 |
� | +0.130 |
bruised | +0.129 |
wrench | +0.126 |
uc | +0.124 |
Neg logit contributions | |
Julia | -0.135 |
Amber | -0.128 |
dolphins | -0.127 |
princess | -0.126 |
mate | -0.124 |
Token | marriage |
Token ID | 4845 |
Feature activation | -1.62e-01 |
Pos logit contributions | |
bruised | +0.097 |
� | +0.097 |
wrench | +0.096 |
swollen | +0.094 |
hammer | +0.093 |
Neg logit contributions | |
Julia | -0.132 |
La | -0.123 |
Mae | -0.121 |
Amber | -0.119 |
Zac | -0.115 |
Token | made |
Token ID | 925 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
Julia | +0.267 |
La | +0.252 |
Mae | +0.244 |
Amber | +0.239 |
Zac | +0.232 |
Neg logit contributions | |
bruised | -0.271 |
� | -0.269 |
wrench | -0.267 |
hammer | -0.264 |
swollen | -0.262 |
Token |
|
Token ID | 198 |
Feature activation | -4.86e-03 |
Pos logit contributions | |
Julia | +0.009 |
La | +0.008 |
Mae | +0.008 |
Amber | +0.008 |
actor | +0.008 |
Neg logit contributions | |
� | -0.007 |
wrench | -0.007 |
bruised | -0.007 |
hammer | -0.007 |
swollen | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
Julia | +0.009 |
La | +0.008 |
Mae | +0.008 |
Amber | +0.008 |
Zac | +0.008 |
Neg logit contributions | |
bruised | -0.007 |
� | -0.007 |
wrench | -0.007 |
swollen | -0.007 |
hammer | -0.007 |
Token | rabbit |
Token ID | 22746 |
Feature activation | -7.60e-04 |
Pos logit contributions | |
� | +0.110 |
bruised | +0.110 |
wrench | +0.108 |
bruises | +0.107 |
pling | +0.103 |
Neg logit contributions | |
Julia | -0.082 |
La | -0.075 |
Amber | -0.075 |
Mae | -0.074 |
Lou | -0.073 |
Token | hopped |
Token ID | 45230 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
Julia | +0.001 |
La | +0.001 |
actor | +0.001 |
Mae | +0.001 |
Amber | +0.001 |
Neg logit contributions | |
swollen | -0.001 |
bruised | -0.001 |
� | -0.001 |
hammer | -0.001 |
bruises | -0.001 |
Token | away |
Token ID | 1497 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
Julia | +0.355 |
La | +0.337 |
Mae | +0.327 |
Amber | +0.317 |
actor | +0.316 |
Neg logit contributions | |
bruised | -0.224 |
� | -0.223 |
wrench | -0.217 |
hammer | -0.213 |
swollen | -0.212 |
Token | , |
Token ID | 11 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
bruised | +0.212 |
� | +0.204 |
swollen | +0.203 |
hammer | +0.197 |
wrench | +0.195 |
Neg logit contributions | |
Julia | -0.279 |
La | -0.278 |
Mae | -0.259 |
actor | -0.253 |
Amber | -0.253 |
Token | but |
Token ID | 475 |
Feature activation | +7.44e-02 |
Pos logit contributions | |
Julia | +0.004 |
La | +0.003 |
Mae | +0.003 |
Amber | +0.003 |
Zac | +0.003 |
Neg logit contributions | |
bruised | -0.003 |
� | -0.003 |
wrench | -0.003 |
swollen | -0.003 |
hammer | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | +9.96e-03 |
Pos logit contributions | |
bruised | +0.119 |
� | +0.115 |
swollen | +0.113 |
bruises | +0.111 |
hammer | +0.110 |
Neg logit contributions | |
Julia | -0.128 |
La | -0.128 |
Mae | -0.115 |
Amber | -0.114 |
actor | -0.114 |
Token | turned |
Token ID | 2900 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
bruised | +0.015 |
swollen | +0.015 |
� | +0.014 |
bruises | +0.014 |
hammer | +0.013 |
Neg logit contributions | |
Julia | -0.019 |
La | -0.018 |
Amber | -0.017 |
Mae | -0.017 |
actor | -0.017 |
Token | its |
Token ID | 663 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
Julia | +0.018 |
La | +0.016 |
Mae | +0.016 |
Amber | +0.016 |
Zac | +0.015 |
Neg logit contributions | |
� | -0.021 |
wrench | -0.021 |
bruised | -0.020 |
bruises | -0.020 |
hammer | -0.020 |
Token | head |
Token ID | 1182 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
Julia | +0.108 |
La | +0.100 |
Mae | +0.098 |
Amber | +0.096 |
Zac | +0.093 |
Neg logit contributions | |
bruised | -0.099 |
� | -0.097 |
wrench | -0.097 |
swollen | -0.096 |
hammer | -0.095 |
Token | shore |
Token ID | 15191 |
Feature activation | -1.78e-01 |
Pos logit contributions | |
bruised | +0.004 |
� | +0.004 |
bruises | +0.004 |
hammer | +0.004 |
wrench | +0.004 |
Neg logit contributions | |
Julia | -0.006 |
La | -0.006 |
Mae | -0.005 |
Amber | -0.005 |
April | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
Julia | +0.308 |
La | +0.292 |
Mae | +0.281 |
Amber | +0.276 |
April | +0.266 |
Neg logit contributions | |
bruised | -0.289 |
wrench | -0.283 |
� | -0.281 |
hammer | -0.277 |
swollen | -0.274 |
Token | he |
Token ID | 339 |
Feature activation | -6.21e-02 |
Pos logit contributions | |
bruised | +0.027 |
hammer | +0.026 |
wrench | +0.026 |
bruises | +0.026 |
swollen | +0.025 |
Neg logit contributions | |
Julia | -0.028 |
La | -0.027 |
Mae | -0.026 |
Amber | -0.026 |
Zac | -0.024 |
Token | drowned |
Token ID | 30859 |
Feature activation | +7.09e-02 |
Pos logit contributions | |
Julia | +0.126 |
La | +0.119 |
Mae | +0.117 |
Amber | +0.117 |
April | +0.111 |
Neg logit contributions | |
bruised | -0.087 |
hammer | -0.084 |
swollen | -0.083 |
wrench | -0.082 |
bruises | -0.081 |
Token | in |
Token ID | 287 |
Feature activation | +7.91e-03 |
Pos logit contributions | |
bruised | +0.108 |
wrench | +0.103 |
� | +0.101 |
hammer | +0.101 |
bruises | +0.100 |
Neg logit contributions | |
Julia | -0.135 |
La | -0.129 |
Mae | -0.124 |
Amber | -0.123 |
April | -0.118 |
Token | the |
Token ID | 262 |
Feature activation | +5.14e-02 |
Pos logit contributions | |
� | +0.015 |
pling | +0.015 |
wrench | +0.015 |
uc | +0.015 |
mons | +0.015 |
Neg logit contributions | |
Julia | -0.012 |
Mae | -0.010 |
Amber | -0.010 |
La | -0.010 |
dolphins | -0.010 |
Token | sea |
Token ID | 5417 |
Feature activation | -2.00e-01 |
Pos logit contributions | |
bruised | +0.108 |
� | +0.108 |
wrench | +0.107 |
swollen | +0.106 |
hammer | +0.105 |
Neg logit contributions | |
Julia | -0.065 |
La | -0.058 |
Mae | -0.056 |
Amber | -0.055 |
Zac | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Julia | +0.368 |
La | +0.348 |
Mae | +0.335 |
Amber | +0.331 |
Zac | +0.320 |
Neg logit contributions | |
bruised | -0.306 |
� | -0.301 |
wrench | -0.300 |
hammer | -0.295 |
swollen | -0.290 |
Token |
|
Token ID | 198 |
Feature activation | +5.20e-06 |
Pos logit contributions | |
wrench | +0.029 |
swollen | +0.029 |
bruised | +0.029 |
supervision | +0.029 |
pling | +0.028 |
Neg logit contributions | |
Julia | -0.019 |
Jeff | -0.016 |
Boo | -0.016 |
Lou | -0.015 |
Ro | -0.015 |
Token |
|
Token ID | 198 |
Feature activation | +2.80e-05 |
Token | The |
Token ID | 464 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
supervision | +0.000 |
bruised | +0.000 |
wrench | +0.000 |
� | +0.000 |
pling | +0.000 |
Neg logit contributions | |
La | -0.000 |
Julia | -0.000 |
ia | -0.000 |
idi | -0.000 |
Ro | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +2.06e-02 |
Pos logit contributions | |
Julia | +0.020 |
La | +0.019 |
Mae | +0.018 |
Amber | +0.018 |
Zac | +0.017 |
Neg logit contributions | |
bruised | -0.018 |
� | -0.018 |
wrench | -0.018 |
swollen | -0.017 |
hammer | -0.017 |
Token | faster |
Token ID | 5443 |
Feature activation | -2.29e-01 |
Pos logit contributions | |
bruised | +0.034 |
swollen | +0.033 |
bruises | +0.033 |
wrench | +0.033 |
hammer | +0.033 |
Neg logit contributions | |
Julia | -0.036 |
La | -0.033 |
Mae | -0.033 |
Amber | -0.032 |
April | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
Julia | +0.386 |
La | +0.354 |
Mae | +0.348 |
Amber | +0.343 |
Zac | +0.329 |
Neg logit contributions | |
� | -0.382 |
bruised | -0.381 |
wrench | -0.377 |
swollen | -0.371 |
hammer | -0.368 |
Token | Do |
Token ID | 2141 |
Feature activation | +4.83e-03 |
Pos logit contributions | |
� | +0.022 |
bruised | +0.022 |
wrench | +0.022 |
bruises | +0.021 |
hammer | +0.021 |
Neg logit contributions | |
Julia | -0.027 |
La | -0.025 |
Mae | -0.024 |
Amber | -0.024 |
dolphins | -0.024 |
Token | you |
Token ID | 345 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
bruises | +0.006 |
� | +0.006 |
hammer | +0.006 |
wrench | +0.006 |
swollen | +0.006 |
Neg logit contributions | |
Julia | -0.009 |
Mae | -0.009 |
La | -0.009 |
actor | -0.009 |
Amber | -0.008 |
Token | want |
Token ID | 765 |
Feature activation | -8.53e-03 |
Pos logit contributions | |
Julia | +0.053 |
La | +0.049 |
Mae | +0.048 |
Amber | +0.047 |
Zac | +0.046 |
Neg logit contributions | |
bruised | -0.046 |
wrench | -0.046 |
hammer | -0.045 |
� | -0.045 |
bruises | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
Julia | +0.017 |
La | +0.016 |
Mae | +0.016 |
Amber | +0.015 |
Zac | +0.015 |
Neg logit contributions | |
hammer | -0.012 |
bruises | -0.011 |
wrench | -0.011 |
oil | -0.011 |
bruised | -0.011 |
Token | buy |
Token ID | 2822 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
� | +0.022 |
bruised | +0.022 |
wrench | +0.022 |
hammer | +0.021 |
swollen | +0.021 |
Neg logit contributions | |
Julia | -0.024 |
La | -0.022 |
Mae | -0.022 |
Amber | -0.022 |
Zac | -0.021 |
Token | some |
Token ID | 617 |
Feature activation | +2.87e-02 |
Pos logit contributions | |
oil | +0.367 |
hammer | +0.367 |
wrench | +0.365 |
bruised | +0.362 |
bruises | +0.357 |
Neg logit contributions | |
Julia | -0.471 |
La | -0.447 |
Mae | -0.429 |
Amber | -0.412 |
April | -0.406 |
Token | ?" |
Token ID | 1701 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
bruised | +0.045 |
� | +0.045 |
wrench | +0.044 |
swollen | +0.044 |
bruises | +0.043 |
Neg logit contributions | |
Julia | -0.051 |
La | -0.047 |
Mae | -0.046 |
Amber | -0.045 |
Zac | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -1.71e-03 |
Pos logit contributions | |
Julia | +0.190 |
La | +0.171 |
Mae | +0.170 |
Amber | +0.167 |
Zac | +0.160 |
Neg logit contributions | |
bruised | -0.205 |
wrench | -0.201 |
� | -0.201 |
swollen | -0.200 |
bruises | -0.197 |
Token | want |
Token ID | 765 |
Feature activation | -8.15e-04 |
Pos logit contributions | |
Julia | +0.266 |
La | +0.247 |
Mae | +0.243 |
Amber | +0.240 |
Zac | +0.232 |
Neg logit contributions | |
bruised | -0.191 |
� | -0.190 |
wrench | -0.188 |
hammer | -0.184 |
swollen | -0.183 |
Token | to |
Token ID | 284 |
Feature activation | +1.90e-02 |
Pos logit contributions | |
Julia | +0.002 |
La | +0.002 |
Zac | +0.002 |
Amber | +0.002 |
ia | +0.002 |
Neg logit contributions | |
oil | -0.001 |
hammer | -0.001 |
bruises | -0.001 |
gum | -0.001 |
sugar | -0.001 |
Token | hug |
Token ID | 16225 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
hammer | +0.029 |
wrench | +0.025 |
scrape | +0.025 |
bruised | +0.025 |
rag | +0.025 |
Neg logit contributions | |
Julia | -0.035 |
Mae | -0.033 |
Zac | -0.032 |
Amber | -0.032 |
La | -0.031 |
Token | the |
Token ID | 262 |
Feature activation | +6.39e-02 |
Pos logit contributions | |
hammer | +0.123 |
gum | +0.122 |
bruised | +0.120 |
oil | +0.120 |
wrench | +0.116 |
Neg logit contributions | |
Julia | -0.210 |
dives | -0.199 |
diver | -0.199 |
ae | -0.194 |
La | -0.194 |
Token | bear |
Token ID | 6842 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
bruised | +0.120 |
� | +0.120 |
wrench | +0.119 |
swollen | +0.117 |
hammer | +0.116 |
Neg logit contributions | |
Julia | -0.093 |
La | -0.085 |
Mae | -0.083 |
Amber | -0.081 |
Zac | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
bruised | +0.039 |
hammer | +0.038 |
pling | +0.037 |
cut | +0.037 |
wrench | +0.037 |
Neg logit contributions | |
Julia | -0.062 |
La | -0.061 |
Amber | -0.055 |
Mae | -0.055 |
dives | -0.054 |
Token | And |
Token ID | 843 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
bruised | +0.027 |
wrench | +0.026 |
� | +0.026 |
hammer | +0.025 |
bruises | +0.025 |
Neg logit contributions | |
Julia | -0.036 |
dolphins | -0.034 |
La | -0.034 |
Amber | -0.034 |
actor | -0.033 |
Token | I |
Token ID | 314 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
bruised | +0.020 |
hammer | +0.019 |
wrench | +0.019 |
bruises | +0.019 |
� | +0.018 |
Neg logit contributions | |
La | -0.019 |
Julia | -0.019 |
Mae | -0.017 |
Amber | -0.017 |
actor | -0.017 |
Token | want |
Token ID | 765 |
Feature activation | -1.24e-03 |
Pos logit contributions | |
Julia | +0.270 |
La | +0.251 |
Mae | +0.247 |
Amber | +0.244 |
Zac | +0.235 |
Neg logit contributions | |
bruised | -0.190 |
� | -0.189 |
wrench | -0.187 |
swollen | -0.183 |
hammer | -0.182 |
Token | to |
Token ID | 284 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
Julia | +0.003 |
La | +0.003 |
Amber | +0.003 |
Zac | +0.003 |
Mae | +0.003 |
Neg logit contributions | |
oil | -0.001 |
hammer | -0.001 |
bruises | -0.001 |
gum | -0.001 |
bruised | -0.001 |
Token | hug |
Token ID | 16225 |
Feature activation | +1.09e-01 |
Pos logit contributions | |
hammer | +0.026 |
bruised | +0.024 |
scrape | +0.024 |
rag | +0.024 |
bruises | +0.023 |
Neg logit contributions | |
Julia | -0.032 |
Mae | -0.031 |
Zac | -0.030 |
Amber | -0.029 |
La | -0.029 |
Token | the |
Token ID | 262 |
Feature activation | +4.97e-02 |
Pos logit contributions | |
swollen | +0.039 |
hammer | +0.039 |
bruised | +0.039 |
� | +0.038 |
bruises | +0.037 |
Neg logit contributions | |
La | -0.057 |
mate | -0.057 |
Julia | -0.055 |
dives | -0.054 |
actor | -0.053 |
Token | clown |
Token ID | 25573 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
bruises | +0.079 |
bruised | +0.079 |
therm | +0.078 |
swollen | +0.077 |
rag | +0.076 |
Neg logit contributions | |
Julia | -0.071 |
La | -0.069 |
Mae | -0.069 |
Lou | -0.068 |
Amber | -0.067 |
Token | jugg |
Token ID | 24873 |
Feature activation | -2.55e-01 |
Pos logit contributions | |
Julia | +0.513 |
La | +0.491 |
Mae | +0.486 |
Amber | +0.476 |
ae | +0.453 |
Neg logit contributions | |
� | -0.419 |
hammer | -0.413 |
wrench | -0.402 |
bruised | -0.402 |
uc | -0.400 |
Token | led |
Token ID | 992 |
Feature activation | +3.22e-02 |
Pos logit contributions | |
Julia | +0.583 |
April | +0.554 |
La | +0.552 |
Mae | +0.551 |
throne | +0.547 |
Neg logit contributions | |
tant | -0.251 |
ised | -0.233 |
uc | -0.221 |
anked | -0.199 |
bruised | -0.193 |
Token | balls |
Token ID | 11333 |
Feature activation | +9.51e-02 |
Pos logit contributions | |
bruised | +0.043 |
hammer | +0.042 |
wrench | +0.041 |
swollen | +0.041 |
bruises | +0.041 |
Neg logit contributions | |
Julia | -0.058 |
La | -0.058 |
ae | -0.056 |
Mae | -0.056 |
Amber | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +1.88e-03 |
Pos logit contributions | |
hammer | +0.131 |
wrench | +0.129 |
� | +0.128 |
bruised | +0.127 |
swollen | +0.119 |
Neg logit contributions | |
Julia | -0.189 |
La | -0.182 |
Mae | -0.175 |
Amber | -0.171 |
April | -0.168 |
Token | spun |
Token ID | 26843 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
bruised | +0.003 |
wrench | +0.003 |
hammer | +0.003 |
swollen | +0.003 |
� | +0.003 |
Neg logit contributions | |
Julia | -0.003 |
La | -0.003 |
Mae | -0.003 |
Amber | -0.003 |
ae | -0.003 |
Token | plates |
Token ID | 13375 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
Julia | +0.263 |
ae | +0.262 |
Mae | +0.252 |
La | +0.246 |
ia | +0.243 |
Neg logit contributions | |
oil | -0.192 |
wrench | -0.191 |
hammer | -0.191 |
swollen | -0.191 |
tant | -0.189 |
Token | and |
Token ID | 290 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
hammer | +0.058 |
wrench | +0.058 |
� | +0.056 |
bruised | +0.056 |
pling | +0.052 |
Neg logit contributions | |
Julia | -0.085 |
La | -0.083 |
Mae | -0.077 |
Amber | -0.076 |
April | -0.075 |
Token | rode |
Token ID | 22075 |
Feature activation | -1.12e-02 |
Pos logit contributions | |
swollen | +0.026 |
hammer | +0.026 |
bruises | +0.026 |
bruised | +0.025 |
wrench | +0.025 |
Neg logit contributions | |
Julia | -0.029 |
ae | -0.026 |
La | -0.026 |
ia | -0.026 |
actor | -0.026 |
Token | a |
Token ID | 257 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
Julia | +0.021 |
ae | +0.020 |
La | +0.020 |
Mae | +0.020 |
actor | +0.020 |
Neg logit contributions | |
free | -0.014 |
hammer | -0.013 |
tant | -0.013 |
cut | -0.013 |
wrench | -0.013 |
Token | One |
Token ID | 3198 |
Feature activation | -4.81e-02 |
Pos logit contributions | |
� | +0.058 |
bruised | +0.051 |
pling | +0.050 |
wrench | +0.050 |
hammer | +0.050 |
Neg logit contributions | |
Julia | -0.088 |
La | -0.084 |
Mae | -0.082 |
actor | -0.081 |
Amber | -0.080 |
Token | day |
Token ID | 1110 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
Julia | +0.104 |
La | +0.102 |
Amber | +0.098 |
Mae | +0.096 |
aliens | +0.096 |
Neg logit contributions | |
rag | -0.049 |
hammer | -0.049 |
bruised | -0.049 |
� | -0.047 |
therm | -0.046 |
Token | a |
Token ID | 257 |
Feature activation | -6.46e-02 |
Pos logit contributions | |
� | +0.014 |
hammer | +0.012 |
wrench | +0.012 |
bruised | +0.012 |
swollen | +0.011 |
Neg logit contributions | |
La | -0.025 |
Julia | -0.025 |
actor | -0.025 |
aliens | -0.023 |
oper | -0.023 |
Token | pilot |
Token ID | 8022 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
Julia | +0.127 |
Mae | +0.116 |
La | +0.114 |
Amber | +0.113 |
Zac | +0.112 |
Neg logit contributions | |
� | -0.092 |
wrench | -0.089 |
hammer | -0.089 |
rag | -0.086 |
therm | -0.084 |
Token | flew |
Token ID | 13112 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Julia | +0.411 |
La | +0.380 |
Mae | +0.378 |
Amber | +0.371 |
ae | +0.355 |
Neg logit contributions | |
hammer | -0.329 |
bruised | -0.327 |
cut | -0.320 |
wrench | -0.318 |
pling | -0.316 |
Token | through |
Token ID | 832 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
Julia | +0.264 |
Mae | +0.248 |
La | +0.243 |
Amber | +0.234 |
ae | +0.233 |
Neg logit contributions | |
bruised | -0.176 |
hammer | -0.174 |
wrench | -0.171 |
swollen | -0.168 |
cut | -0.164 |
Token | the |
Token ID | 262 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
Julia | +0.089 |
La | +0.084 |
Mae | +0.083 |
Amber | +0.081 |
Zac | +0.079 |
Neg logit contributions | |
hammer | -0.058 |
bruised | -0.058 |
� | -0.057 |
wrench | -0.057 |
swollen | -0.057 |
Token | sky |
Token ID | 6766 |
Feature activation | -6.63e-02 |
Pos logit contributions | |
hammer | +0.079 |
oil | +0.077 |
rag | +0.076 |
therm | +0.075 |
bruised | +0.075 |
Neg logit contributions | |
Julia | -0.104 |
Mae | -0.098 |
Amber | -0.093 |
Zac | -0.092 |
ae | -0.091 |
Token | in |
Token ID | 287 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
Julia | +0.147 |
La | +0.147 |
Mae | +0.137 |
Amber | +0.137 |
April | +0.135 |
Neg logit contributions | |
hammer | -0.070 |
bruised | -0.067 |
� | -0.066 |
oil | -0.062 |
cut | -0.062 |
Token | a |
Token ID | 257 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
bruised | +0.015 |
hammer | +0.014 |
bruises | +0.014 |
wrench | +0.014 |
� | +0.014 |
Neg logit contributions | |
Julia | -0.022 |
La | -0.021 |
Mae | -0.020 |
Amber | -0.020 |
Zac | -0.020 |
Token | big |
Token ID | 1263 |
Feature activation | -4.90e-02 |
Pos logit contributions | |
Julia | +0.126 |
Mae | +0.119 |
Zac | +0.116 |
Amber | +0.114 |
Jeff | +0.111 |
Neg logit contributions | |
hammer | -0.091 |
swollen | -0.088 |
rag | -0.086 |
wrench | -0.086 |
therm | -0.084 |
Token | a |
Token ID | 257 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
Julia | +0.088 |
Mae | +0.081 |
La | +0.080 |
ae | +0.077 |
Amber | +0.076 |
Neg logit contributions | |
hammer | -0.059 |
oil | -0.059 |
bruised | -0.058 |
wrench | -0.057 |
rag | -0.055 |
Token | rocket |
Token ID | 10701 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
Julia | +0.132 |
Mae | +0.121 |
La | +0.115 |
Amber | +0.114 |
Alicia | +0.112 |
Neg logit contributions | |
hammer | -0.096 |
wrench | -0.094 |
rag | -0.092 |
� | -0.091 |
swollen | -0.091 |
Token | or |
Token ID | 393 |
Feature activation | -4.72e-02 |
Pos logit contributions | |
Julia | +0.327 |
La | +0.315 |
Mae | +0.299 |
Amber | +0.299 |
Jul | +0.287 |
Neg logit contributions | |
hammer | -0.235 |
wrench | -0.226 |
cut | -0.217 |
bruised | -0.211 |
� | -0.210 |
Token | a |
Token ID | 257 |
Feature activation | -6.48e-02 |
Pos logit contributions | |
Julia | +0.091 |
Mae | +0.085 |
La | +0.084 |
ae | +0.083 |
ia | +0.083 |
Neg logit contributions | |
hammer | -0.069 |
bruises | -0.066 |
wrench | -0.066 |
bruised | -0.065 |
nails | -0.064 |
Token | bird |
Token ID | 6512 |
Feature activation | -5.19e-02 |
Pos logit contributions | |
Julia | +0.115 |
Mae | +0.104 |
La | +0.102 |
Amber | +0.099 |
Alicia | +0.097 |
Neg logit contributions | |
hammer | -0.109 |
wrench | -0.108 |
bruised | -0.103 |
swollen | -0.102 |
rag | -0.101 |
Token | ." |
Token ID | 526 |
Feature activation | -9.06e-02 |
Pos logit contributions | |
Julia | +0.102 |
Mae | +0.092 |
Amber | +0.091 |
La | +0.090 |
ae | +0.087 |
Neg logit contributions | |
hammer | -0.076 |
bruised | -0.074 |
wrench | -0.072 |
cut | -0.071 |
oil | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | -1.04e-03 |
Pos logit contributions | |
La | +0.150 |
actor | +0.140 |
oper | +0.139 |
royal | +0.136 |
April | +0.134 |
Neg logit contributions | |
� | -0.150 |
hammer | -0.135 |
oil | -0.134 |
wrench | -0.134 |
bruises | -0.132 |
Token |
|
Token ID | 198 |
Feature activation | -6.79e-04 |
Pos logit contributions | |
actor | +0.002 |
Julia | +0.002 |
La | +0.002 |
Amber | +0.002 |
oper | +0.002 |
Neg logit contributions | |
� | -0.001 |
hammer | -0.001 |
oil | -0.001 |
wrench | -0.001 |
bruises | -0.001 |
Token | " |
Token ID | 1 |
Feature activation | +1.83e-02 |
Pos logit contributions | |
Julia | +0.001 |
April | +0.001 |
oper | +0.001 |
actor | +0.001 |
Amber | +0.001 |
Neg logit contributions | |
leaning | -0.001 |
wrench | -0.001 |
oil | -0.001 |
hammer | -0.001 |
� | -0.001 |
Token | OK |
Token ID | 11380 |
Feature activation | +1.62e-01 |
Pos logit contributions | |
bruised | +0.023 |
� | +0.023 |
wrench | +0.023 |
swollen | +0.022 |
hammer | +0.022 |
Neg logit contributions | |
Julia | -0.039 |
La | -0.037 |
Mae | -0.036 |
Amber | -0.035 |
Zac | -0.034 |
Token | !" |
Token ID | 2474 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
wrench | +0.227 |
hammer | +0.226 |
bruised | +0.224 |
� | +0.220 |
swollen | +0.212 |
Neg logit contributions | |
Julia | -0.305 |
La | -0.304 |
Mae | -0.281 |
Amber | -0.279 |
actor | -0.272 |
Token | and |
Token ID | 290 |
Feature activation | +1.34e-02 |
Pos logit contributions | |
Julia | +0.020 |
La | +0.019 |
Mae | +0.019 |
Amber | +0.018 |
Zac | +0.018 |
Neg logit contributions | |
bruised | -0.016 |
� | -0.016 |
wrench | -0.016 |
swollen | -0.015 |
hammer | -0.015 |
Token | from |
Token ID | 422 |
Feature activation | -5.14e-02 |
Pos logit contributions | |
bruised | +0.020 |
swollen | +0.020 |
bruises | +0.020 |
hammer | +0.019 |
� | +0.019 |
Neg logit contributions | |
Julia | -0.023 |
La | -0.023 |
adies | -0.021 |
Mae | -0.021 |
actor | -0.021 |
Token | then |
Token ID | 788 |
Feature activation | +7.66e-04 |
Pos logit contributions | |
Julia | +0.089 |
La | +0.085 |
Mae | +0.081 |
Amber | +0.080 |
Zac | +0.078 |
Neg logit contributions | |
bruised | -0.081 |
swollen | -0.078 |
� | -0.078 |
wrench | -0.078 |
bruises | -0.077 |
Token | on |
Token ID | 319 |
Feature activation | +5.78e-02 |
Pos logit contributions | |
bruised | +0.001 |
� | +0.001 |
wrench | +0.001 |
hammer | +0.001 |
swollen | +0.001 |
Neg logit contributions | |
Julia | -0.002 |
La | -0.001 |
Mae | -0.001 |
Amber | -0.001 |
Zac | -0.001 |
Token | she |
Token ID | 673 |
Feature activation | -4.64e-02 |
Pos logit contributions | |
� | +0.094 |
bruised | +0.093 |
wrench | +0.092 |
swollen | +0.091 |
hammer | +0.091 |
Neg logit contributions | |
Julia | -0.098 |
La | -0.093 |
Mae | -0.089 |
Amber | -0.088 |
Zac | -0.086 |
Token | was |
Token ID | 373 |
Feature activation | -4.60e-02 |
Pos logit contributions | |
Julia | +0.085 |
La | +0.080 |
Zac | +0.078 |
Mae | +0.078 |
mate | +0.077 |
Neg logit contributions | |
bruised | -0.066 |
� | -0.065 |
hammer | -0.064 |
swollen | -0.063 |
pling | -0.062 |
Token | never |
Token ID | 1239 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Julia | +0.086 |
La | +0.081 |
Zac | +0.078 |
dives | +0.077 |
Amber | +0.077 |
Neg logit contributions | |
bruised | -0.070 |
swollen | -0.065 |
bruises | -0.062 |
hammer | -0.060 |
� | -0.059 |
Token | sad |
Token ID | 6507 |
Feature activation | +9.53e-02 |
Pos logit contributions | |
Julia | +0.019 |
La | +0.018 |
Mae | +0.018 |
Zac | +0.018 |
Amber | +0.017 |
Neg logit contributions | |
bruised | -0.015 |
swollen | -0.014 |
bruises | -0.014 |
wrench | -0.014 |
hammer | -0.014 |
Token | again |
Token ID | 757 |
Feature activation | -1.52e-01 |
Pos logit contributions | |
bruised | +0.157 |
� | +0.145 |
swollen | +0.141 |
wrench | +0.141 |
hammer | +0.140 |
Neg logit contributions | |
Julia | -0.162 |
La | -0.160 |
Mae | -0.148 |
Zac | -0.147 |
April | -0.146 |
Token | . |
Token ID | 13 |
Feature activation | +7.25e-03 |
Pos logit contributions | |
Julia | +0.271 |
La | +0.252 |
Mae | +0.246 |
Amber | +0.242 |
Zac | +0.236 |
Neg logit contributions | |
bruised | -0.239 |
� | -0.238 |
wrench | -0.233 |
swollen | -0.231 |
bruises | -0.228 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.91e-02 |
Pos logit contributions | |
� | +0.011 |
bruised | +0.011 |
wrench | +0.010 |
swollen | +0.010 |
hammer | +0.010 |
Neg logit contributions | |
Julia | -0.013 |
La | -0.012 |
actor | -0.012 |
dolphins | -0.012 |
Amber | -0.012 |
Token | called |
Token ID | 1444 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
Julia | +0.010 |
La | +0.009 |
Mae | +0.009 |
Amber | +0.009 |
April | +0.008 |
Neg logit contributions | |
� | -0.011 |
wrench | -0.011 |
bruised | -0.011 |
swollen | -0.011 |
bruises | -0.011 |
Token | Lisa |
Token ID | 15378 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
La | +0.092 |
Julia | +0.089 |
actor | +0.087 |
dolphins | +0.086 |
dives | +0.085 |
Neg logit contributions | |
� | -0.106 |
bruised | -0.105 |
hammer | -0.102 |
bruises | -0.099 |
wrench | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | +8.39e-03 |
Pos logit contributions | |
Julia | +0.640 |
La | +0.591 |
Mae | +0.583 |
Amber | +0.577 |
Zac | +0.569 |
Neg logit contributions | |
bruised | -0.482 |
hammer | -0.463 |
swollen | -0.462 |
wrench | -0.462 |
� | -0.461 |
Token | Lisa |
Token ID | 15378 |
Feature activation | -3.36e-01 |
Pos logit contributions | |
� | +0.015 |
hammer | +0.014 |
bruises | +0.014 |
bruised | +0.014 |
issors | +0.014 |
Neg logit contributions | |
dolphins | -0.011 |
diver | -0.011 |
mate | -0.011 |
Julia | -0.010 |
Zac | -0.010 |
Token | had |
Token ID | 550 |
Feature activation | +4.08e-03 |
Pos logit contributions | |
Julia | +0.617 |
La | +0.563 |
Mae | +0.556 |
Amber | +0.553 |
Zac | +0.545 |
Neg logit contributions | |
bruised | -0.508 |
hammer | -0.500 |
� | -0.497 |
swollen | -0.494 |
wrench | -0.488 |
Token | a |
Token ID | 257 |
Feature activation | -6.78e-02 |
Pos logit contributions | |
bruises | +0.006 |
bruised | +0.006 |
hammer | +0.006 |
� | +0.006 |
rag | +0.006 |
Neg logit contributions | |
Julia | -0.008 |
Zac | -0.007 |
dives | -0.007 |
Mae | -0.007 |
actor | -0.007 |
Token | diary |
Token ID | 26339 |
Feature activation | +5.07e-02 |
Pos logit contributions | |
Julia | +0.110 |
La | +0.101 |
Mae | +0.099 |
Amber | +0.098 |
April | +0.093 |
Neg logit contributions | |
bruised | -0.116 |
� | -0.114 |
wrench | -0.113 |
bruises | -0.113 |
swollen | -0.111 |
Token | that |
Token ID | 326 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
bruised | +0.075 |
� | +0.075 |
wrench | +0.074 |
swollen | +0.073 |
hammer | +0.072 |
Neg logit contributions | |
Julia | -0.094 |
La | -0.087 |
Mae | -0.086 |
Amber | -0.085 |
Zac | -0.082 |
Token | she |
Token ID | 673 |
Feature activation | -4.43e-02 |
Pos logit contributions | |
bruised | +0.219 |
� | +0.215 |
hammer | +0.215 |
wrench | +0.211 |
swollen | +0.211 |
Neg logit contributions | |
Julia | -0.223 |
La | -0.213 |
Mae | -0.202 |
Amber | -0.200 |
Zac | -0.198 |
Token | kept |
Token ID | 4030 |
Feature activation | +8.91e-02 |
Pos logit contributions | |
Julia | +0.078 |
La | +0.072 |
Mae | +0.070 |
Amber | +0.069 |
Zac | +0.067 |
Neg logit contributions | |
bruised | -0.071 |
� | -0.070 |
wrench | -0.070 |
swollen | -0.069 |
hammer | -0.068 |
Token | all |
Token ID | 477 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
� | +0.182 |
wrench | +0.168 |
bruised | +0.164 |
bruises | +0.164 |
swollen | +0.162 |
Neg logit contributions | |
Julia | -0.129 |
April | -0.114 |
La | -0.113 |
Amber | -0.113 |
Mae | -0.111 |
Token | fish |
Token ID | 5916 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
� | +0.164 |
bruised | +0.163 |
swollen | +0.160 |
wrench | +0.160 |
bruises | +0.157 |
Neg logit contributions | |
Julia | -0.101 |
La | -0.090 |
Mae | -0.088 |
Amber | -0.088 |
dolphins | -0.087 |
Token | sw |
Token ID | 1509 |
Feature activation | -8.83e-02 |
Pos logit contributions | |
Julia | +0.224 |
La | +0.216 |
Mae | +0.204 |
Amber | +0.202 |
April | +0.196 |
Neg logit contributions | |
bruised | -0.210 |
hammer | -0.209 |
� | -0.205 |
swollen | -0.204 |
wrench | -0.203 |
Token | am |
Token ID | 321 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Julia | +0.123 |
La | +0.113 |
Mae | +0.111 |
Amber | +0.107 |
April | +0.103 |
Neg logit contributions | |
bruised | -0.174 |
wrench | -0.172 |
� | -0.171 |
bruises | -0.169 |
hammer | -0.168 |
Token | away |
Token ID | 1497 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
Julia | +0.266 |
La | +0.263 |
Mae | +0.252 |
Amber | +0.243 |
April | +0.241 |
Neg logit contributions | |
bruised | -0.165 |
wrench | -0.160 |
hammer | -0.160 |
swollen | -0.153 |
� | -0.151 |
Token | and |
Token ID | 290 |
Feature activation | +1.30e-02 |
Pos logit contributions | |
bruised | +0.212 |
� | +0.205 |
swollen | +0.200 |
hammer | +0.196 |
wrench | +0.193 |
Neg logit contributions | |
Julia | -0.282 |
La | -0.282 |
Mae | -0.262 |
Amber | -0.256 |
actor | -0.253 |
Token | Daisy |
Token ID | 40355 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
bruised | +0.023 |
swollen | +0.022 |
bruises | +0.022 |
� | +0.021 |
hammer | +0.021 |
Neg logit contributions | |
Julia | -0.021 |
La | -0.021 |
Mae | -0.019 |
actor | -0.019 |
Zac | -0.019 |
Token | felt |
Token ID | 2936 |
Feature activation | -3.02e-02 |
Pos logit contributions | |
Julia | +0.201 |
La | +0.194 |
Amber | +0.184 |
Mae | +0.182 |
adies | +0.179 |
Neg logit contributions | |
bruised | -0.149 |
� | -0.142 |
hammer | -0.141 |
swollen | -0.141 |
bruises | -0.139 |
Token | happy |
Token ID | 3772 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
Julia | +0.047 |
La | +0.043 |
Mae | +0.042 |
Amber | +0.042 |
Zac | +0.040 |
Neg logit contributions | |
� | -0.054 |
wrench | -0.053 |
bruised | -0.053 |
hammer | -0.052 |
swollen | -0.052 |
Token | and |
Token ID | 290 |
Feature activation | +1.22e-02 |
Pos logit contributions | |
Julia | +0.129 |
La | +0.127 |
Mae | +0.120 |
Amber | +0.117 |
Zac | +0.116 |
Neg logit contributions | |
bruised | -0.109 |
� | -0.106 |
bruises | -0.103 |
swollen | -0.103 |
wrench | -0.102 |
Token | proud |
Token ID | 6613 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
bruised | +0.018 |
swollen | +0.017 |
bruises | +0.017 |
� | +0.017 |
hammer | +0.016 |
Neg logit contributions | |
Julia | -0.023 |
La | -0.022 |
Mae | -0.021 |
Zac | -0.021 |
Amber | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | +3.18e-03 |
Pos logit contributions | |
bruised | +0.115 |
� | +0.112 |
bruises | +0.109 |
wrench | +0.109 |
hammer | +0.108 |
Neg logit contributions | |
Julia | -0.159 |
La | -0.158 |
Mae | -0.149 |
Amber | -0.146 |
Zac | -0.145 |
Token | decided |
Token ID | 3066 |
Feature activation | -7.55e-02 |
Pos logit contributions | |
Julia | +0.387 |
La | +0.367 |
Mae | +0.354 |
Zac | +0.351 |
Amber | +0.345 |
Neg logit contributions | |
hammer | -0.278 |
� | -0.276 |
bruised | -0.275 |
wrench | -0.266 |
bruises | -0.261 |
Token | to |
Token ID | 284 |
Feature activation | +4.58e-03 |
Pos logit contributions | |
Julia | +0.155 |
La | +0.149 |
Mae | +0.145 |
Amber | +0.140 |
Zac | +0.139 |
Neg logit contributions | |
hammer | -0.098 |
wrench | -0.097 |
bruised | -0.095 |
bruises | -0.092 |
swollen | -0.092 |
Token | build |
Token ID | 1382 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
hammer | +0.007 |
� | +0.007 |
wrench | +0.007 |
bruised | +0.007 |
rag | +0.007 |
Neg logit contributions | |
Julia | -0.008 |
La | -0.008 |
Mae | -0.007 |
Amber | -0.007 |
Zac | -0.007 |
Token | a |
Token ID | 257 |
Feature activation | -6.77e-02 |
Pos logit contributions | |
hammer | +0.085 |
wrench | +0.083 |
bruises | +0.083 |
bruised | +0.081 |
oil | +0.080 |
Neg logit contributions | |
Julia | -0.094 |
La | -0.093 |
Mae | -0.089 |
Amber | -0.087 |
April | -0.085 |
Token | sand |
Token ID | 6450 |
Feature activation | -7.96e-02 |
Pos logit contributions | |
Julia | +0.126 |
La | +0.120 |
Mae | +0.116 |
Amber | +0.114 |
Zac | +0.113 |
Neg logit contributions | |
hammer | -0.113 |
wrench | -0.109 |
rag | -0.102 |
oil | -0.101 |
bruised | -0.098 |
Token | castle |
Token ID | 16669 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
La | +0.126 |
Julia | +0.123 |
April | +0.118 |
Amber | +0.117 |
Mae | +0.116 |
Neg logit contributions | |
hammer | -0.153 |
wrench | -0.148 |
� | -0.137 |
nails | -0.136 |
oil | -0.135 |
Token | at |
Token ID | 379 |
Feature activation | -1.01e-03 |
Pos logit contributions | |
Julia | +0.503 |
La | +0.497 |
Mae | +0.473 |
Amber | +0.471 |
April | +0.457 |
Neg logit contributions | |
hammer | -0.340 |
wrench | -0.332 |
� | -0.322 |
bruised | -0.319 |
bruises | -0.305 |
Token | the |
Token ID | 262 |
Feature activation | +4.90e-02 |
Pos logit contributions | |
Julia | +0.002 |
La | +0.002 |
Mae | +0.002 |
Amber | +0.002 |
Zac | +0.002 |
Neg logit contributions | |
� | -0.001 |
hammer | -0.001 |
bruised | -0.001 |
wrench | -0.001 |
bruises | -0.001 |
Token | beach |
Token ID | 10481 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
hammer | +0.077 |
oil | +0.073 |
� | +0.071 |
file | +0.071 |
wrench | +0.070 |
Neg logit contributions | |
Julia | -0.088 |
La | -0.086 |
April | -0.080 |
Zac | -0.078 |
Lou | -0.078 |
Token | . |
Token ID | 13 |
Feature activation | +7.63e-03 |
Pos logit contributions | |
Julia | +0.424 |
La | +0.407 |
Mae | +0.386 |
Amber | +0.383 |
Zac | +0.374 |
Neg logit contributions | |
� | -0.323 |
bruised | -0.321 |
wrench | -0.320 |
hammer | -0.320 |
bruises | -0.306 |
Token | They |
Token ID | 1119 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
� | +0.012 |
wrench | +0.011 |
hammer | +0.011 |
bruised | +0.010 |
rag | +0.010 |
Neg logit contributions | |
La | -0.014 |
Julia | -0.013 |
Zac | -0.012 |
Mae | -0.012 |
oper | -0.012 |
Token | was |
Token ID | 373 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
bruised | +0.084 |
� | +0.084 |
wrench | +0.083 |
swollen | +0.082 |
bruises | +0.081 |
Neg logit contributions | |
Julia | -0.065 |
La | -0.059 |
Mae | -0.058 |
Amber | -0.056 |
Zac | -0.054 |
Token | a |
Token ID | 257 |
Feature activation | -6.49e-02 |
Pos logit contributions | |
Julia | +0.075 |
La | +0.069 |
Mae | +0.066 |
Amber | +0.065 |
April | +0.064 |
Neg logit contributions | |
� | -0.066 |
bruised | -0.065 |
bruises | -0.064 |
hammer | -0.064 |
rag | -0.063 |
Token | young |
Token ID | 1862 |
Feature activation | +1.44e-02 |
Pos logit contributions | |
Julia | +0.118 |
La | +0.109 |
Mae | +0.107 |
Amber | +0.106 |
Zac | +0.102 |
Neg logit contributions | |
bruised | -0.099 |
� | -0.099 |
wrench | -0.099 |
hammer | -0.097 |
swollen | -0.096 |
Token | bear |
Token ID | 6842 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
bruised | +0.023 |
wrench | +0.023 |
� | +0.023 |
hammer | +0.022 |
swollen | +0.022 |
Neg logit contributions | |
Julia | -0.025 |
La | -0.024 |
Mae | -0.023 |
Amber | -0.023 |
Zac | -0.022 |
Token | called |
Token ID | 1444 |
Feature activation | -5.65e-02 |
Pos logit contributions | |
bruised | +0.036 |
� | +0.036 |
wrench | +0.036 |
swollen | +0.035 |
hammer | +0.035 |
Neg logit contributions | |
Julia | -0.038 |
La | -0.035 |
Mae | -0.035 |
Amber | -0.034 |
Zac | -0.033 |
Token | Flash |
Token ID | 9973 |
Feature activation | -3.06e-01 |
Pos logit contributions | |
Julia | +0.105 |
La | +0.105 |
actor | +0.094 |
dances | +0.094 |
Mae | +0.094 |
Neg logit contributions | |
wrench | -0.079 |
hammer | -0.079 |
bruised | -0.079 |
� | -0.078 |
pling | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Julia | +0.614 |
La | +0.576 |
Amber | +0.554 |
Mae | +0.554 |
Alicia | +0.533 |
Neg logit contributions | |
bruised | -0.420 |
hammer | -0.391 |
wrench | -0.390 |
� | -0.387 |
swollen | -0.386 |
Token | He |
Token ID | 679 |
Feature activation | -1.78e-02 |
Pos logit contributions | |
bruised | +0.018 |
� | +0.018 |
hammer | +0.018 |
cut | +0.018 |
therm | +0.018 |
Neg logit contributions | |
Julia | -0.017 |
actor | -0.016 |
Amber | -0.016 |
La | -0.016 |
diver | -0.015 |
Token | had |
Token ID | 550 |
Feature activation | +8.71e-03 |
Pos logit contributions | |
Julia | +0.035 |
Mae | +0.031 |
Amber | +0.031 |
La | +0.031 |
Alicia | +0.029 |
Neg logit contributions | |
bruised | -0.026 |
hammer | -0.025 |
� | -0.025 |
swollen | -0.025 |
wrench | -0.025 |
Token | very |
Token ID | 845 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
bruised | +0.014 |
bruises | +0.014 |
hammer | +0.014 |
swollen | +0.014 |
wrench | +0.014 |
Neg logit contributions | |
Julia | -0.016 |
Mae | -0.014 |
Amber | -0.014 |
La | -0.014 |
Alicia | -0.013 |
Token | soft |
Token ID | 2705 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
Julia | +0.064 |
La | +0.059 |
Mae | +0.058 |
Amber | +0.057 |
Zac | +0.055 |
Neg logit contributions | |
bruised | -0.057 |
� | -0.057 |
wrench | -0.056 |
swollen | -0.056 |
hammer | -0.055 |
Token | castle |
Token ID | 16669 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
Julia | +0.099 |
La | +0.092 |
Mae | +0.090 |
Alicia | +0.088 |
Amber | +0.087 |
Neg logit contributions | |
bruised | -0.081 |
hammer | -0.080 |
wrench | -0.078 |
bruises | -0.077 |
� | -0.076 |
Token | so |
Token ID | 523 |
Feature activation | -6.11e-02 |
Pos logit contributions | |
Julia | +0.431 |
La | +0.397 |
Mae | +0.390 |
Amber | +0.384 |
Zac | +0.369 |
Neg logit contributions | |
bruised | -0.403 |
� | -0.401 |
wrench | -0.398 |
swollen | -0.391 |
hammer | -0.389 |
Token | much |
Token ID | 881 |
Feature activation | -6.80e-02 |
Pos logit contributions | |
Julia | +0.113 |
La | +0.108 |
Mae | +0.105 |
Amber | +0.102 |
Zac | +0.100 |
Neg logit contributions | |
swollen | -0.093 |
bruised | -0.092 |
hammer | -0.086 |
wrench | -0.085 |
� | -0.084 |
Token | . |
Token ID | 13 |
Feature activation | +9.12e-03 |
Pos logit contributions | |
Julia | +0.132 |
La | +0.125 |
Mae | +0.122 |
Amber | +0.119 |
Zac | +0.117 |
Neg logit contributions | |
bruised | -0.097 |
wrench | -0.095 |
� | -0.093 |
swollen | -0.093 |
hammer | -0.093 |
Token | The |
Token ID | 383 |
Feature activation | +7.81e-02 |
Pos logit contributions | |
� | +0.012 |
hammer | +0.011 |
wrench | +0.011 |
bruised | +0.010 |
swollen | +0.010 |
Neg logit contributions | |
La | -0.018 |
Julia | -0.018 |
dolphins | -0.017 |
Mae | -0.017 |
Zac | -0.017 |
Token | mer |
Token ID | 4017 |
Feature activation | -3.02e-01 |
Pos logit contributions | |
hammer | +0.140 |
iron | +0.138 |
cloth | +0.136 |
swollen | +0.136 |
jug | +0.134 |
Neg logit contributions | |
La | -0.090 |
Julia | -0.088 |
Zac | -0.088 |
ae | -0.084 |
Mae | -0.083 |
Token | maid |
Token ID | 23151 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
La | +0.442 |
Julia | +0.395 |
Mae | +0.393 |
Ro | +0.389 |
Alicia | +0.371 |
Neg logit contributions | |
uc | -0.551 |
tant | -0.512 |
isher | -0.498 |
bruised | -0.490 |
ised | -0.488 |
Token | said |
Token ID | 531 |
Feature activation | +3.46e-03 |
Pos logit contributions | |
Julia | +0.515 |
La | +0.499 |
Mae | +0.482 |
Amber | +0.468 |
Zac | +0.457 |
Neg logit contributions | |
swollen | -0.352 |
bruised | -0.344 |
hammer | -0.341 |
wrench | -0.329 |
� | -0.328 |
Token | , |
Token ID | 11 |
Feature activation | -4.63e-04 |
Pos logit contributions | |
� | +0.005 |
bruised | +0.005 |
wrench | +0.005 |
swollen | +0.005 |
bruises | +0.005 |
Neg logit contributions | |
Julia | -0.006 |
La | -0.006 |
Mae | -0.006 |
Amber | -0.006 |
Jeff | -0.005 |
Token | " |
Token ID | 366 |
Feature activation | +4.31e-02 |
Pos logit contributions | |
Julia | +0.001 |
Mae | +0.001 |
Amber | +0.001 |
La | +0.001 |
April | +0.001 |
Neg logit contributions | |
bruised | -0.001 |
bruises | -0.001 |
� | -0.001 |
swollen | -0.001 |
wrench | -0.001 |
Token | I |
Token ID | 40 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
� | +0.052 |
swollen | +0.051 |
bruised | +0.051 |
wrench | +0.050 |
hammer | +0.050 |
Neg logit contributions | |
Julia | -0.091 |
Mae | -0.084 |
La | -0.083 |
Amber | -0.083 |
Zac | -0.081 |
Token | liked |
Token ID | 8288 |
Feature activation | +1.20e-02 |
Pos logit contributions | |
Julia | +0.144 |
La | +0.132 |
Mae | +0.131 |
Amber | +0.129 |
Zac | +0.124 |
Neg logit contributions | |
bruised | -0.132 |
� | -0.131 |
wrench | -0.130 |
swollen | -0.128 |
hammer | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
bruised | +0.015 |
hammer | +0.015 |
wrench | +0.015 |
oil | +0.015 |
bruises | +0.015 |
Neg logit contributions | |
Julia | -0.024 |
La | -0.023 |
Amber | -0.022 |
Mae | -0.022 |
Zac | -0.021 |
Token | Tom |
Token ID | 4186 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
bruised | +0.025 |
wrench | +0.025 |
� | +0.025 |
swollen | +0.025 |
hammer | +0.024 |
Neg logit contributions | |
Julia | -0.022 |
La | -0.020 |
Mae | -0.019 |
Amber | -0.019 |
Zac | -0.018 |
Token | picked |
Token ID | 6497 |
Feature activation | +1.56e-01 |
Pos logit contributions | |
Julia | +0.096 |
La | +0.090 |
Amber | +0.089 |
oper | +0.087 |
Mae | +0.085 |
Neg logit contributions | |
bruised | -0.056 |
hammer | -0.054 |
bruises | -0.053 |
wrench | -0.052 |
swollen | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
oil | +0.193 |
gum | +0.185 |
hammer | +0.180 |
bruised | +0.179 |
sugar | +0.178 |
Neg logit contributions | |
La | -0.300 |
Julia | -0.284 |
nes | -0.277 |
mate | -0.270 |
ae | -0.268 |
Token | movie |
Token ID | 3807 |
Feature activation | -2.96e-01 |
Pos logit contributions | |
Julia | +0.112 |
Amber | +0.100 |
Mae | +0.100 |
La | +0.099 |
Lou | +0.096 |
Neg logit contributions | |
wrench | -0.097 |
hammer | -0.097 |
bruised | -0.094 |
therm | -0.093 |
swollen | -0.091 |
Token | about |
Token ID | 546 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
Julia | +0.567 |
La | +0.555 |
Amber | +0.527 |
Mae | +0.523 |
April | +0.507 |
Neg logit contributions | |
wrench | -0.406 |
bruised | -0.402 |
� | -0.401 |
hammer | -0.396 |
bruises | -0.385 |
Token | cars |
Token ID | 5006 |
Feature activation | +7.08e-03 |
Pos logit contributions | |
Julia | +0.328 |
La | +0.314 |
Amber | +0.303 |
Mae | +0.301 |
ae | +0.287 |
Neg logit contributions | |
bruised | -0.261 |
bruises | -0.259 |
hammer | -0.258 |
swollen | -0.258 |
oil | -0.254 |
Token | and |
Token ID | 290 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
bruised | +0.012 |
� | +0.012 |
wrench | +0.012 |
swollen | +0.012 |
bruises | +0.011 |
Neg logit contributions | |
Julia | -0.012 |
La | -0.011 |
Mae | -0.011 |
Amber | -0.010 |
Zac | -0.010 |
Token | trucks |
Token ID | 13960 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
bruises | +0.021 |
bruised | +0.020 |
swollen | +0.020 |
wrench | +0.020 |
hammer | +0.020 |
Neg logit contributions | |
Julia | -0.030 |
Amber | -0.028 |
La | -0.028 |
ae | -0.027 |
Mae | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | +1.36e-02 |
Pos logit contributions | |
bruised | +0.178 |
� | +0.177 |
wrench | +0.175 |
swollen | +0.172 |
hammer | +0.171 |
Neg logit contributions | |
Julia | -0.195 |
La | -0.180 |
Mae | -0.176 |
Amber | -0.174 |
Zac | -0.167 |
Token | of |
Token ID | 286 |
Feature activation | -5.51e-02 |
Pos logit contributions | |
hammer | +0.044 |
bruised | +0.043 |
rag | +0.041 |
wrench | +0.041 |
swollen | +0.040 |
Neg logit contributions | |
La | -0.078 |
Mae | -0.074 |
Julia | -0.074 |
Amber | -0.074 |
actor | -0.073 |
Token | the |
Token ID | 262 |
Feature activation | +5.64e-02 |
Pos logit contributions | |
Julia | +0.102 |
La | +0.094 |
Mae | +0.093 |
Amber | +0.091 |
Zac | +0.088 |
Neg logit contributions | |
bruised | -0.084 |
� | -0.084 |
wrench | -0.084 |
swollen | -0.081 |
hammer | -0.081 |
Token | day |
Token ID | 1110 |
Feature activation | +9.05e-03 |
Pos logit contributions | |
� | +0.082 |
wrench | +0.077 |
hammer | +0.077 |
bruised | +0.077 |
swollen | +0.076 |
Neg logit contributions | |
Julia | -0.114 |
Mae | -0.103 |
La | -0.103 |
Amber | -0.102 |
Lou | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +1.83e-03 |
Pos logit contributions | |
� | +0.010 |
sugar | +0.008 |
hammer | +0.008 |
bruised | +0.008 |
oil | +0.008 |
Neg logit contributions | |
La | -0.021 |
mate | -0.020 |
Julia | -0.019 |
Zac | -0.019 |
actor | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | +5.59e-02 |
Pos logit contributions | |
wrench | +0.003 |
bruised | +0.003 |
� | +0.003 |
bruises | +0.003 |
swollen | +0.003 |
Neg logit contributions | |
Julia | -0.003 |
Amber | -0.003 |
Mae | -0.003 |
La | -0.003 |
April | -0.003 |
Token | superhero |
Token ID | 17343 |
Feature activation | -2.56e-01 |
Pos logit contributions | |
bruised | +0.073 |
� | +0.072 |
wrench | +0.071 |
hammer | +0.071 |
swollen | +0.069 |
Neg logit contributions | |
Julia | -0.116 |
Mae | -0.107 |
La | -0.106 |
Amber | -0.103 |
Jeff | -0.103 |
Token | brought |
Token ID | 3181 |
Feature activation | +4.49e-02 |
Pos logit contributions | |
Julia | +0.488 |
La | +0.477 |
Mae | +0.447 |
Amber | +0.435 |
Zac | +0.433 |
Neg logit contributions | |
pling | -0.349 |
hammer | -0.349 |
bruised | -0.345 |
wrench | -0.344 |
� | -0.337 |
Token | the |
Token ID | 262 |
Feature activation | +5.53e-02 |
Pos logit contributions | |
bruised | +0.061 |
bruises | +0.058 |
oil | +0.058 |
groceries | +0.058 |
hammer | +0.057 |
Neg logit contributions | |
Julia | -0.088 |
La | -0.086 |
actor | -0.081 |
Zac | -0.080 |
Mae | -0.080 |
Token | little |
Token ID | 1310 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
bruised | +0.082 |
wrench | +0.081 |
hammer | +0.080 |
� | +0.080 |
bruises | +0.078 |
Neg logit contributions | |
Julia | -0.106 |
La | -0.099 |
Mae | -0.097 |
Amber | -0.095 |
Lou | -0.093 |
Token | girl |
Token ID | 2576 |
Feature activation | -4.14e-03 |
Pos logit contributions | |
bruised | +0.022 |
wrench | +0.021 |
hammer | +0.020 |
rag | +0.020 |
oil | +0.019 |
Neg logit contributions | |
Julia | -0.043 |
La | -0.042 |
Jeff | -0.041 |
Mae | -0.041 |
Lou | -0.040 |
Token | safely |
Token ID | 11512 |
Feature activation | -4.34e-02 |
Pos logit contributions | |
Julia | +0.007 |
La | +0.007 |
Mae | +0.006 |
Amber | +0.006 |
Zac | +0.006 |
Neg logit contributions | |
bruised | -0.007 |
hammer | -0.007 |
bruises | -0.007 |
� | -0.007 |
wrench | -0.007 |