Token | doll |
Token ID | 3654 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
ophone | +0.021 |
Jupiter | +0.021 |
belongs | +0.019 |
dark | +0.019 |
existed | +0.019 |
Neg logit contributions | |
distress | -0.020 |
No | -0.020 |
Son | -0.019 |
Dad | -0.019 |
Be | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | +1.15e-01 |
Pos logit contributions | |
ophone | +0.097 |
Jupiter | +0.096 |
belongs | +0.092 |
unic | +0.086 |
dark | +0.084 |
Neg logit contributions | |
Son | -0.151 |
distress | -0.148 |
Dad | -0.147 |
No | -0.146 |
Be | -0.143 |
Token | They |
Token ID | 1119 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
Son | +0.007 |
distress | +0.007 |
Dad | +0.007 |
No | +0.007 |
iver | +0.006 |
Neg logit contributions | |
Jupiter | -0.005 |
ophone | -0.004 |
belongs | -0.004 |
unic | -0.004 |
dark | -0.004 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
Son | +0.058 |
Dad | +0.051 |
distress | +0.050 |
iver | +0.048 |
Marcus | +0.048 |
Neg logit contributions | |
belongs | -0.068 |
ophone | -0.066 |
Jupiter | -0.063 |
belonged | -0.062 |
unic | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +1.96e-01 |
Pos logit contributions | |
Son | +0.301 |
iver | +0.293 |
Marcus | +0.285 |
distress | +0.284 |
Col | +0.279 |
Neg logit contributions | |
Jupiter | -0.165 |
belongs | -0.159 |
ophone | -0.156 |
lights | -0.151 |
they | -0.144 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.54e+00 |
Pos logit contributions | |
Son | +0.008 |
iver | +0.007 |
distress | +0.007 |
Col | +0.007 |
Dad | +0.007 |
Neg logit contributions | |
belongs | -0.012 |
ophone | -0.012 |
they | -0.011 |
Jupiter | -0.011 |
we | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
distress | +0.606 |
Son | +0.606 |
Be | +0.577 |
iver | +0.574 |
Baby | +0.550 |
Neg logit contributions | |
ophone | -0.226 |
belongs | -0.220 |
they | -0.200 |
lights | -0.200 |
ls | -0.200 |
Token | The |
Token ID | 383 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
No | +0.006 |
Dad | +0.006 |
Son | +0.006 |
distress | +0.006 |
Be | +0.005 |
Neg logit contributions | |
ophone | -0.005 |
Jupiter | -0.004 |
belongs | -0.004 |
unic | -0.004 |
endant | -0.004 |
Token | doll |
Token ID | 3654 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
ophone | +0.076 |
Jupiter | +0.076 |
belongs | +0.069 |
existed | +0.069 |
dark | +0.069 |
Neg logit contributions | |
No | -0.072 |
distress | -0.071 |
Be | -0.069 |
Dad | -0.068 |
Son | -0.067 |
Token | is |
Token ID | 318 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
ophone | +0.106 |
Jupiter | +0.104 |
belongs | +0.104 |
unic | +0.095 |
dark | +0.094 |
Neg logit contributions | |
Son | -0.144 |
distress | -0.140 |
Dad | -0.140 |
No | -0.136 |
iver | -0.134 |
Token | in |
Token ID | 287 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
ophone | +0.021 |
Jupiter | +0.021 |
belongs | +0.020 |
unic | +0.019 |
endant | +0.019 |
Neg logit contributions | |
Son | -0.020 |
distress | -0.019 |
Dad | -0.019 |
No | -0.019 |
iver | -0.018 |
Token |
|
Token ID | 198 |
Feature activation | -9.11e-02 |
Pos logit contributions | |
ophone | +0.004 |
Jupiter | +0.004 |
belongs | +0.004 |
dark | +0.004 |
unic | +0.004 |
Neg logit contributions | |
distress | -0.005 |
Son | -0.005 |
No | -0.005 |
Dad | -0.004 |
iver | -0.004 |
Token | They |
Token ID | 2990 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
they | +0.004 |
ophone | +0.004 |
Jupiter | +0.004 |
we | +0.004 |
dark | +0.004 |
Neg logit contributions | |
distress | -0.004 |
worry | -0.004 |
scrape | -0.004 |
irk | -0.004 |
iver | -0.004 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
Son | +0.132 |
distress | +0.130 |
No | +0.127 |
Dad | +0.127 |
iver | +0.125 |
Neg logit contributions | |
ophone | -0.132 |
Jupiter | -0.129 |
belongs | -0.125 |
unic | -0.120 |
dark | -0.117 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
Son | +0.500 |
distress | +0.497 |
iver | +0.477 |
Stan | +0.453 |
Marcus | +0.452 |
Neg logit contributions | |
ophone | -0.334 |
belongs | -0.301 |
they | -0.293 |
happens | -0.292 |
lights | -0.285 |
Token | and |
Token ID | 290 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
Son | +0.282 |
iver | +0.277 |
distress | +0.269 |
Marcus | +0.263 |
Dad | +0.262 |
Neg logit contributions | |
ophone | -0.164 |
belongs | -0.161 |
Jupiter | -0.160 |
existed | -0.143 |
they | -0.143 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
Son | +0.011 |
distress | +0.011 |
iver | +0.011 |
Dad | +0.010 |
Col | +0.010 |
Neg logit contributions | |
ophone | -0.018 |
Jupiter | -0.018 |
belongs | -0.017 |
belonged | -0.017 |
unic | -0.017 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
distress | +0.454 |
Son | +0.452 |
iver | +0.431 |
Be | +0.417 |
Brother | +0.403 |
Neg logit contributions | |
ophone | -0.379 |
belongs | -0.357 |
existed | -0.351 |
lights | -0.338 |
Jupiter | -0.338 |
Token | . |
Token ID | 13 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
Son | +0.284 |
iver | +0.277 |
distress | +0.276 |
Dad | +0.268 |
Be | +0.266 |
Neg logit contributions | |
ophone | -0.161 |
Jupiter | -0.157 |
belongs | -0.156 |
existed | -0.147 |
lights | -0.138 |
Token | The |
Token ID | 383 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
No | +0.004 |
Dad | +0.004 |
distress | +0.004 |
Son | +0.004 |
Be | +0.004 |
Neg logit contributions | |
ophone | -0.003 |
belongs | -0.003 |
Jupiter | -0.003 |
unic | -0.003 |
endant | -0.003 |
Token | train |
Token ID | 4512 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
ophone | +0.060 |
Jupiter | +0.059 |
existed | +0.057 |
belongs | +0.056 |
dark | +0.055 |
Neg logit contributions | |
Dad | -0.078 |
No | -0.078 |
distress | -0.078 |
Son | -0.076 |
Be | -0.076 |
Token | broke |
Token ID | 6265 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
Jupiter | +0.236 |
ophone | +0.228 |
endant | +0.223 |
unic | +0.217 |
dark | +0.217 |
Neg logit contributions | |
No | -0.186 |
distress | -0.179 |
tug | -0.175 |
Be | -0.174 |
peel | -0.174 |
Token | he |
Token ID | 339 |
Feature activation | -5.04e-01 |
Pos logit contributions | |
Son | +0.001 |
distress | +0.001 |
Dad | +0.001 |
No | +0.001 |
iver | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
belongs | -0.001 |
unic | -0.001 |
dark | -0.001 |
Token | started |
Token ID | 2067 |
Feature activation | -3.45e-01 |
Pos logit contributions | |
ophone | +0.026 |
Jupiter | +0.026 |
belongs | +0.025 |
unic | +0.024 |
dark | +0.023 |
Neg logit contributions | |
Son | -0.024 |
distress | -0.023 |
Dad | -0.023 |
No | -0.023 |
Be | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
ophone | +0.010 |
Jupiter | +0.010 |
unic | +0.009 |
belongs | +0.009 |
belonged | +0.008 |
Neg logit contributions | |
Dad | -0.024 |
Son | -0.024 |
iver | -0.023 |
No | -0.023 |
Col | -0.022 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
Dad | +0.038 |
Son | +0.037 |
Col | +0.033 |
No | +0.033 |
Marcus | +0.033 |
Neg logit contributions | |
Jupiter | -0.045 |
belongs | -0.045 |
ophone | -0.045 |
unic | -0.044 |
belonged | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +3.98e-01 |
Pos logit contributions | |
Son | +0.267 |
iver | +0.265 |
distress | +0.261 |
Dad | +0.251 |
Be | +0.247 |
Neg logit contributions | |
Jupiter | -0.201 |
ophone | -0.193 |
belongs | -0.179 |
dark | -0.172 |
existed | -0.170 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
No | +0.014 |
distress | +0.014 |
Son | +0.013 |
tug | +0.013 |
peel | +0.013 |
Neg logit contributions | |
Jupiter | -0.026 |
ophone | -0.026 |
belongs | -0.025 |
unic | -0.025 |
endant | -0.025 |
Token | at |
Token ID | 379 |
Feature activation | -7.14e-01 |
Pos logit contributions | |
distress | +0.431 |
Son | +0.417 |
iver | +0.415 |
Be | +0.399 |
Dad | +0.387 |
Neg logit contributions | |
ophone | -0.401 |
existed | -0.379 |
belongs | -0.378 |
Jupiter | -0.373 |
dark | -0.359 |
Token | his |
Token ID | 465 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
ophone | +0.038 |
Jupiter | +0.037 |
belongs | +0.036 |
endant | +0.035 |
unic | +0.035 |
Neg logit contributions | |
distress | -0.033 |
Son | -0.033 |
No | -0.032 |
Dad | -0.032 |
Be | -0.031 |
Token | toe |
Token ID | 21189 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
Son | +0.088 |
Dad | +0.082 |
distress | +0.078 |
Be | +0.078 |
No | +0.077 |
Neg logit contributions | |
ophone | -0.144 |
Jupiter | -0.140 |
belongs | -0.140 |
unic | -0.136 |
existed | -0.134 |
Token | until |
Token ID | 1566 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Son | +0.146 |
distress | +0.143 |
Dad | +0.143 |
Be | +0.140 |
No | +0.139 |
Neg logit contributions | |
ophone | -0.081 |
Jupiter | -0.075 |
belongs | -0.075 |
existed | -0.069 |
lights | -0.068 |
Token | it |
Token ID | 340 |
Feature activation | -5.92e-01 |
Pos logit contributions | |
ophone | +0.065 |
Jupiter | +0.062 |
belongs | +0.061 |
endant | +0.058 |
unic | +0.057 |
Neg logit contributions | |
No | -0.066 |
Dad | -0.066 |
Son | -0.065 |
distress | -0.065 |
Be | -0.062 |
Token | back |
Token ID | 736 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
ophone | +0.036 |
Jupiter | +0.036 |
belongs | +0.034 |
unic | +0.034 |
endant | +0.034 |
Neg logit contributions | |
No | -0.033 |
distress | -0.032 |
Son | -0.032 |
Dad | -0.031 |
peel | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | +4.78e-02 |
Pos logit contributions | |
distress | +0.058 |
No | +0.058 |
peel | +0.057 |
Son | +0.056 |
Dad | +0.056 |
Neg logit contributions | |
ophone | -0.064 |
Jupiter | -0.063 |
unic | -0.062 |
belongs | -0.060 |
dark | -0.059 |
Token |
|
Token ID | 198 |
Feature activation | -9.32e-02 |
Pos logit contributions | |
Son | +0.002 |
No | +0.002 |
Dad | +0.002 |
distress | +0.002 |
Be | +0.002 |
Neg logit contributions | |
ophone | -0.002 |
Jupiter | -0.002 |
belongs | -0.002 |
unic | -0.002 |
endant | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -9.11e-02 |
Pos logit contributions | |
ophone | +0.004 |
Jupiter | +0.004 |
belongs | +0.004 |
dark | +0.004 |
unic | +0.004 |
Neg logit contributions | |
distress | -0.005 |
Son | -0.005 |
No | -0.005 |
Dad | -0.004 |
iver | -0.004 |
Token | They |
Token ID | 2990 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
they | +0.004 |
ophone | +0.004 |
Jupiter | +0.004 |
we | +0.004 |
dark | +0.004 |
Neg logit contributions | |
distress | -0.004 |
worry | -0.004 |
scrape | -0.004 |
irk | -0.004 |
iver | -0.004 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
Son | +0.132 |
distress | +0.130 |
No | +0.127 |
Dad | +0.127 |
iver | +0.125 |
Neg logit contributions | |
ophone | -0.132 |
Jupiter | -0.129 |
belongs | -0.125 |
unic | -0.120 |
dark | -0.117 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
Son | +0.500 |
distress | +0.497 |
iver | +0.477 |
Stan | +0.453 |
Marcus | +0.452 |
Neg logit contributions | |
ophone | -0.334 |
belongs | -0.301 |
they | -0.293 |
happens | -0.292 |
lights | -0.285 |
Token | and |
Token ID | 290 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
Son | +0.282 |
iver | +0.277 |
distress | +0.269 |
Marcus | +0.263 |
Dad | +0.262 |
Neg logit contributions | |
ophone | -0.164 |
belongs | -0.161 |
Jupiter | -0.160 |
existed | -0.143 |
they | -0.143 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.52e+00 |
Pos logit contributions | |
Son | +0.011 |
distress | +0.011 |
iver | +0.011 |
Dad | +0.010 |
Col | +0.010 |
Neg logit contributions | |
ophone | -0.018 |
Jupiter | -0.018 |
belongs | -0.017 |
belonged | -0.017 |
unic | -0.017 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.55e+00 |
Pos logit contributions | |
distress | +0.454 |
Son | +0.452 |
iver | +0.431 |
Be | +0.417 |
Brother | +0.403 |
Neg logit contributions | |
ophone | -0.379 |
belongs | -0.357 |
existed | -0.351 |
lights | -0.338 |
Jupiter | -0.338 |
Token | . |
Token ID | 13 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
Son | +0.284 |
iver | +0.277 |
distress | +0.276 |
Dad | +0.268 |
Be | +0.266 |
Neg logit contributions | |
ophone | -0.161 |
Jupiter | -0.157 |
belongs | -0.156 |
existed | -0.147 |
lights | -0.138 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
Son | +0.098 |
iver | +0.091 |
Be | +0.091 |
No | +0.090 |
ouch | +0.088 |
Neg logit contributions | |
belongs | -0.130 |
ophone | -0.129 |
Jupiter | -0.125 |
endant | -0.117 |
belonged | -0.117 |
Token | . |
Token ID | 13 |
Feature activation | -4.05e-02 |
Pos logit contributions | |
Son | +0.075 |
distress | +0.074 |
Dad | +0.072 |
No | +0.072 |
Be | +0.070 |
Neg logit contributions | |
ophone | -0.069 |
Jupiter | -0.068 |
belongs | -0.066 |
unic | -0.062 |
dark | -0.061 |
Token | They |
Token ID | 1119 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
ophone | +0.002 |
Jupiter | +0.002 |
belongs | +0.002 |
unic | +0.002 |
dark | +0.002 |
Neg logit contributions | |
Son | -0.002 |
distress | -0.002 |
No | -0.002 |
Dad | -0.002 |
iver | -0.002 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
Son | +0.052 |
Dad | +0.047 |
distress | +0.046 |
iver | +0.045 |
Be | +0.044 |
Neg logit contributions | |
belongs | -0.071 |
ophone | -0.070 |
Jupiter | -0.069 |
belonged | -0.065 |
unic | -0.064 |
Token | and |
Token ID | 290 |
Feature activation | +1.52e-01 |
Pos logit contributions | |
iver | +0.304 |
Son | +0.299 |
distress | +0.290 |
Marcus | +0.277 |
Col | +0.271 |
Neg logit contributions | |
Jupiter | -0.166 |
ophone | -0.157 |
belongs | -0.154 |
they | -0.151 |
lights | -0.147 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.51e+00 |
Pos logit contributions | |
Son | +0.005 |
iver | +0.005 |
distress | +0.004 |
Dad | +0.004 |
Col | +0.004 |
Neg logit contributions | |
belongs | -0.010 |
ophone | -0.010 |
Jupiter | -0.010 |
they | -0.009 |
unic | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
distress | +0.562 |
Son | +0.555 |
iver | +0.540 |
Be | +0.534 |
Baby | +0.507 |
Neg logit contributions | |
ophone | -0.277 |
belongs | -0.264 |
Jupiter | -0.253 |
existed | -0.247 |
lights | -0.241 |
Token | The |
Token ID | 383 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
Jupiter | +0.001 |
ophone | +0.001 |
belongs | +0.001 |
unic | +0.001 |
dark | +0.001 |
Neg logit contributions | |
Son | -0.002 |
distress | -0.002 |
No | -0.002 |
Dad | -0.002 |
iver | -0.002 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
belongs | +0.063 |
ophone | +0.056 |
belonged | +0.055 |
floats | +0.052 |
lights | +0.052 |
Neg logit contributions | |
iver | -0.077 |
Son | -0.076 |
Baby | -0.073 |
distress | -0.072 |
Er | -0.070 |
Token | flies |
Token ID | 17607 |
Feature activation | -3.65e+00 |
Pos logit contributions | |
No | +0.083 |
distress | +0.082 |
Son | +0.080 |
peel | +0.080 |
Be | +0.080 |
Neg logit contributions | |
ophone | -0.060 |
Jupiter | -0.060 |
unic | -0.054 |
belongs | -0.054 |
endant | -0.053 |
Token | out |
Token ID | 503 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
ophone | +0.199 |
Jupiter | +0.197 |
unic | +0.188 |
belongs | +0.186 |
endant | +0.185 |
Neg logit contributions | |
No | -0.161 |
distress | -0.158 |
Son | -0.156 |
Be | -0.156 |
Dad | -0.153 |
Token | card |
Token ID | 2657 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
belongs | +0.017 |
ophone | +0.017 |
Jupiter | +0.017 |
endant | +0.015 |
unic | +0.015 |
Neg logit contributions | |
Son | -0.020 |
Dad | -0.019 |
iver | -0.019 |
Marcus | -0.019 |
No | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -1.13e-01 |
Pos logit contributions | |
distress | +0.033 |
No | +0.033 |
Son | +0.033 |
Dad | +0.032 |
peel | +0.032 |
Neg logit contributions | |
Jupiter | -0.041 |
ophone | -0.041 |
unic | -0.039 |
belongs | -0.039 |
dark | -0.039 |
Token | They |
Token ID | 1119 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
ophone | +0.006 |
Jupiter | +0.006 |
unic | +0.006 |
belongs | +0.006 |
endant | +0.006 |
Neg logit contributions | |
No | -0.005 |
Dad | -0.005 |
Son | -0.005 |
Be | -0.005 |
distress | -0.005 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
Son | +0.051 |
Dad | +0.047 |
distress | +0.047 |
No | +0.045 |
Be | +0.045 |
Neg logit contributions | |
belongs | -0.077 |
ophone | -0.076 |
Jupiter | -0.075 |
belonged | -0.070 |
unic | -0.070 |
Token | and |
Token ID | 290 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
Son | +0.263 |
iver | +0.255 |
distress | +0.255 |
Marcus | +0.247 |
Dad | +0.242 |
Neg logit contributions | |
Jupiter | -0.205 |
ophone | -0.200 |
belongs | -0.197 |
existed | -0.182 |
dark | -0.180 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.51e+00 |
Pos logit contributions | |
Son | +0.005 |
Dad | +0.004 |
iver | +0.004 |
distress | +0.004 |
Col | +0.004 |
Neg logit contributions | |
belongs | -0.009 |
ophone | -0.009 |
Jupiter | -0.009 |
they | -0.009 |
we | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
distress | +0.571 |
Son | +0.567 |
iver | +0.547 |
Be | +0.544 |
Brother | +0.516 |
Neg logit contributions | |
ophone | -0.264 |
belongs | -0.258 |
existed | -0.238 |
Jupiter | -0.232 |
ls | -0.227 |
Token | The |
Token ID | 383 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ophone | +0.006 |
belongs | +0.005 |
Jupiter | +0.005 |
unic | +0.005 |
endant | +0.005 |
Neg logit contributions | |
No | -0.005 |
Dad | -0.005 |
Son | -0.005 |
Be | -0.005 |
distress | -0.005 |
Token | card |
Token ID | 2657 |
Feature activation | +7.66e-01 |
Pos logit contributions | |
ophone | +0.077 |
Jupiter | +0.077 |
belongs | +0.075 |
unic | +0.072 |
dark | +0.070 |
Neg logit contributions | |
Son | -0.064 |
distress | -0.062 |
Dad | -0.061 |
No | -0.061 |
iver | -0.059 |
Token | r |
Token ID | 374 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
peel | +0.042 |
No | +0.041 |
distress | +0.041 |
tug | +0.039 |
Be | +0.038 |
Neg logit contributions | |
Jupiter | -0.035 |
ophone | -0.035 |
unic | -0.031 |
dark | -0.031 |
existed | -0.030 |
Token | ips |
Token ID | 2419 |
Feature activation | +3.34e+00 |
Pos logit contributions | |
ophone | +0.088 |
lights | +0.083 |
Jupiter | +0.078 |
endant | +0.078 |
ways | +0.075 |
Neg logit contributions | |
distress | -0.111 |
Son | -0.109 |
No | -0.108 |
worry | -0.107 |
pack | -0.106 |
Token |
|
Token ID | 198 |
Feature activation | -6.70e-03 |
Pos logit contributions | |
iver | +0.001 |
Son | +0.001 |
distress | +0.001 |
Marcus | +0.001 |
Col | +0.001 |
Neg logit contributions | |
Jupiter | -0.000 |
ophone | -0.000 |
belongs | -0.000 |
lights | -0.000 |
dark | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
ophone | +0.000 |
Jupiter | +0.000 |
belongs | +0.000 |
unic | +0.000 |
dark | +0.000 |
Neg logit contributions | |
Son | -0.000 |
distress | -0.000 |
No | -0.000 |
Dad | -0.000 |
Be | -0.000 |
Token | They |
Token ID | 2990 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
ophone | +0.001 |
belongs | +0.001 |
Jupiter | +0.001 |
dark | +0.001 |
lights | +0.001 |
Neg logit contributions | |
distress | -0.001 |
Son | -0.001 |
iver | -0.001 |
sausage | -0.001 |
reel | -0.001 |
Token | pulled |
Token ID | 5954 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
Son | +0.143 |
distress | +0.143 |
iver | +0.141 |
Dad | +0.134 |
No | +0.133 |
Neg logit contributions | |
ophone | -0.129 |
belongs | -0.121 |
Jupiter | -0.121 |
belonged | -0.117 |
unic | -0.113 |
Token | and |
Token ID | 290 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Son | +0.176 |
iver | +0.170 |
Col | +0.167 |
Marcus | +0.166 |
distress | +0.164 |
Neg logit contributions | |
Jupiter | -0.090 |
belongs | -0.088 |
ophone | -0.085 |
lights | -0.079 |
they | -0.077 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.51e+00 |
Pos logit contributions | |
iver | +0.010 |
distress | +0.009 |
Son | +0.009 |
Dad | +0.009 |
Col | +0.008 |
Neg logit contributions | |
ophone | -0.013 |
belongs | -0.012 |
Jupiter | -0.012 |
belonged | -0.012 |
dark | -0.012 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
distress | +0.462 |
Son | +0.456 |
iver | +0.439 |
Be | +0.421 |
Brother | +0.408 |
Neg logit contributions | |
ophone | -0.378 |
belongs | -0.357 |
existed | -0.343 |
lights | -0.341 |
Jupiter | -0.331 |
Token | on |
Token ID | 319 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
Son | +0.273 |
iver | +0.270 |
distress | +0.267 |
Dad | +0.258 |
Be | +0.257 |
Neg logit contributions | |
ophone | -0.171 |
belongs | -0.168 |
Jupiter | -0.164 |
existed | -0.155 |
lights | -0.148 |
Token | the |
Token ID | 262 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
Jupiter | +0.070 |
ophone | +0.066 |
belongs | +0.066 |
unic | +0.061 |
lights | +0.061 |
Neg logit contributions | |
Son | -0.094 |
iver | -0.092 |
distress | -0.090 |
Dad | -0.089 |
Be | -0.088 |
Token | hat |
Token ID | 6877 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
ophone | +0.031 |
Jupiter | +0.030 |
belongs | +0.029 |
unic | +0.028 |
endant | +0.028 |
Neg logit contributions | |
Son | -0.027 |
distress | -0.027 |
No | -0.026 |
Dad | -0.025 |
iver | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | +4.61e-03 |
Pos logit contributions | |
ophone | +0.078 |
Jupiter | +0.078 |
belongs | +0.074 |
unic | +0.071 |
dark | +0.070 |
Neg logit contributions | |
Son | -0.102 |
distress | -0.101 |
No | -0.100 |
Dad | -0.100 |
Be | -0.097 |
Token | . |
Token ID | 13 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
Son | +0.091 |
distress | +0.088 |
Dad | +0.087 |
Marcus | +0.085 |
No | +0.084 |
Neg logit contributions | |
ophone | -0.056 |
Jupiter | -0.056 |
belongs | -0.054 |
dark | -0.049 |
existed | -0.049 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.86e+00 |
Pos logit contributions | |
distress | +0.012 |
Son | +0.012 |
iver | +0.011 |
No | +0.011 |
Dad | +0.011 |
Neg logit contributions | |
Jupiter | -0.011 |
ophone | -0.011 |
belongs | -0.010 |
dark | -0.010 |
unic | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
Son | +0.180 |
Dad | +0.179 |
distress | +0.173 |
iver | +0.171 |
No | +0.170 |
Neg logit contributions | |
belongs | -0.107 |
ophone | -0.106 |
Jupiter | -0.102 |
unic | -0.092 |
dark | -0.089 |
Token | the |
Token ID | 262 |
Feature activation | -2.89e-01 |
Pos logit contributions | |
distress | +0.017 |
Son | +0.017 |
No | +0.017 |
Dad | +0.017 |
Be | +0.017 |
Neg logit contributions | |
ophone | -0.019 |
Jupiter | -0.019 |
belongs | -0.018 |
endant | -0.017 |
dark | -0.017 |
Token | cat |
Token ID | 3797 |
Feature activation | -6.73e-01 |
Pos logit contributions | |
ophone | +0.013 |
Jupiter | +0.013 |
belongs | +0.013 |
unic | +0.013 |
spaceship | +0.012 |
Neg logit contributions | |
Son | -0.015 |
iver | -0.015 |
Col | -0.014 |
Dad | -0.014 |
Marcus | -0.014 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.51e+00 |
Pos logit contributions | |
ophone | +0.041 |
Jupiter | +0.040 |
belongs | +0.040 |
unic | +0.038 |
dark | +0.037 |
Neg logit contributions | |
Son | -0.027 |
Dad | -0.026 |
distress | -0.026 |
No | -0.025 |
Marcus | -0.024 |
Token | and |
Token ID | 290 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
distress | +0.599 |
Son | +0.578 |
iver | +0.559 |
Stan | +0.528 |
Col | +0.521 |
Neg logit contributions | |
ls | -0.251 |
ophone | -0.236 |
they | -0.230 |
happens | -0.224 |
belongs | -0.224 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
Son | +0.014 |
Dad | +0.014 |
distress | +0.013 |
iver | +0.013 |
Col | +0.012 |
Neg logit contributions | |
Jupiter | -0.023 |
belongs | -0.023 |
ophone | -0.023 |
unic | -0.022 |
existed | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
Son | +0.295 |
iver | +0.291 |
distress | +0.287 |
Marcus | +0.273 |
Dad | +0.272 |
Neg logit contributions | |
Jupiter | -0.172 |
ophone | -0.156 |
belongs | -0.154 |
existed | -0.145 |
unic | -0.145 |
Token | ball |
Token ID | 2613 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
belongs | +0.013 |
ophone | +0.012 |
belonged | +0.011 |
spaceship | +0.010 |
Jupiter | +0.010 |
Neg logit contributions | |
Son | -0.018 |
Col | -0.017 |
Marcus | -0.017 |
iver | -0.017 |
Todd | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +2.15e-01 |
Pos logit contributions | |
Son | +0.090 |
distress | +0.087 |
Dad | +0.086 |
Marcus | +0.084 |
Col | +0.083 |
Neg logit contributions | |
ophone | -0.057 |
Jupiter | -0.056 |
belongs | -0.055 |
existed | -0.049 |
dark | -0.049 |
Token | of |
Token ID | 286 |
Feature activation | +6.08e-02 |
Pos logit contributions | |
No | +0.060 |
Son | +0.060 |
Dad | +0.059 |
distress | +0.058 |
peel | +0.057 |
Neg logit contributions | |
Jupiter | -0.077 |
ophone | -0.075 |
belongs | -0.071 |
unic | -0.071 |
endant | -0.069 |
Token | the |
Token ID | 262 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
Son | +0.003 |
iver | +0.003 |
Dad | +0.003 |
No | +0.003 |
distress | +0.003 |
Neg logit contributions | |
Jupiter | -0.003 |
ophone | -0.003 |
belongs | -0.003 |
unic | -0.002 |
existed | -0.002 |
Token | shirt |
Token ID | 10147 |
Feature activation | +2.49e+00 |
Pos logit contributions | |
ophone | +0.019 |
belongs | +0.019 |
Jupiter | +0.018 |
unic | +0.018 |
endant | +0.017 |
Neg logit contributions | |
Son | -0.016 |
distress | -0.015 |
iver | -0.015 |
Col | -0.014 |
Dad | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.22e-02 |
Pos logit contributions | |
Son | +0.145 |
distress | +0.143 |
No | +0.141 |
Dad | +0.141 |
Be | +0.138 |
Neg logit contributions | |
ophone | -0.102 |
Jupiter | -0.100 |
belongs | -0.096 |
unic | -0.090 |
dark | -0.088 |
Token | He |
Token ID | 679 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
ophone | +0.001 |
Jupiter | +0.001 |
belongs | +0.001 |
unic | +0.001 |
dark | +0.001 |
Neg logit contributions | |
Son | -0.001 |
No | -0.001 |
distress | -0.001 |
Dad | -0.001 |
Be | -0.001 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
Son | +0.109 |
distress | +0.108 |
No | +0.106 |
Dad | +0.105 |
Be | +0.102 |
Neg logit contributions | |
ophone | -0.126 |
Jupiter | -0.126 |
belongs | -0.120 |
unic | -0.116 |
dark | -0.115 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
distress | +0.485 |
Son | +0.465 |
iver | +0.458 |
Be | +0.425 |
pack | +0.416 |
Neg logit contributions | |
ophone | -0.370 |
belongs | -0.343 |
existed | -0.326 |
dark | -0.317 |
Jupiter | -0.313 |
Token | and |
Token ID | 290 |
Feature activation | +2.12e-01 |
Pos logit contributions | |
iver | +0.269 |
Son | +0.267 |
distress | +0.266 |
Dad | +0.252 |
Be | +0.247 |
Neg logit contributions | |
ophone | -0.172 |
belongs | -0.163 |
Jupiter | -0.160 |
existed | -0.156 |
dark | -0.148 |
Token | shook |
Token ID | 14682 |
Feature activation | +4.40e-01 |
Pos logit contributions | |
iver | +0.010 |
Son | +0.010 |
distress | +0.010 |
Dad | +0.010 |
Col | +0.009 |
Neg logit contributions | |
ophone | -0.010 |
belongs | -0.010 |
Jupiter | -0.010 |
belonged | -0.009 |
unic | -0.009 |
Token | his |
Token ID | 465 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
Son | +0.019 |
distress | +0.019 |
No | +0.018 |
Dad | +0.018 |
Be | +0.018 |
Neg logit contributions | |
ophone | -0.025 |
Jupiter | -0.024 |
belongs | -0.024 |
unic | -0.023 |
dark | -0.022 |
Token | head |
Token ID | 1182 |
Feature activation | +3.81e+00 |
Pos logit contributions | |
Son | +0.115 |
Be | +0.106 |
distress | +0.100 |
Dad | +0.099 |
Col | +0.099 |
Neg logit contributions | |
ophone | -0.109 |
belongs | -0.105 |
Jupiter | -0.095 |
endant | -0.095 |
favorite | -0.094 |
Token | doll |
Token ID | 3654 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
Jupiter | +0.024 |
ophone | +0.024 |
belongs | +0.022 |
unic | +0.022 |
dark | +0.022 |
Neg logit contributions | |
No | -0.023 |
distress | -0.023 |
Be | -0.022 |
Son | -0.022 |
Dad | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
ophone | +0.113 |
Jupiter | +0.110 |
belongs | +0.109 |
dark | +0.098 |
endant | +0.097 |
Neg logit contributions | |
Son | -0.149 |
distress | -0.144 |
Dad | -0.143 |
No | -0.142 |
Be | -0.142 |
Token | They |
Token ID | 1119 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
No | +0.004 |
Dad | +0.004 |
Son | +0.004 |
distress | +0.004 |
Be | +0.004 |
Neg logit contributions | |
ophone | -0.004 |
Jupiter | -0.004 |
unic | -0.004 |
belongs | -0.004 |
existed | -0.003 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
Son | +0.056 |
Dad | +0.051 |
Be | +0.048 |
distress | +0.048 |
iver | +0.048 |
Neg logit contributions | |
belongs | -0.066 |
ophone | -0.062 |
Jupiter | -0.061 |
belonged | -0.059 |
endant | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
Son | +0.296 |
iver | +0.290 |
distress | +0.283 |
Marcus | +0.280 |
Col | +0.271 |
Neg logit contributions | |
Jupiter | -0.172 |
belongs | -0.163 |
ophone | -0.160 |
dark | -0.144 |
existed | -0.144 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
Son | +0.003 |
iver | +0.003 |
distress | +0.003 |
Dad | +0.003 |
Col | +0.003 |
Neg logit contributions | |
belongs | -0.006 |
ophone | -0.006 |
Jupiter | -0.006 |
they | -0.006 |
we | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | +7.49e-02 |
Pos logit contributions | |
distress | +0.575 |
Son | +0.575 |
iver | +0.557 |
Be | +0.555 |
Brother | +0.526 |
Neg logit contributions | |
ophone | -0.259 |
belongs | -0.255 |
existed | -0.239 |
Jupiter | -0.239 |
ls | -0.224 |
Token | They |
Token ID | 1119 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
No | +0.004 |
Dad | +0.004 |
distress | +0.004 |
Be | +0.004 |
Son | +0.004 |
Neg logit contributions | |
ophone | -0.004 |
belongs | -0.003 |
Jupiter | -0.003 |
unic | -0.003 |
endant | -0.003 |
Token | start |
Token ID | 923 |
Feature activation | +3.61e+00 |
Pos logit contributions | |
Son | +0.061 |
Dad | +0.055 |
distress | +0.052 |
Baby | +0.052 |
Be | +0.052 |
Neg logit contributions | |
belongs | -0.058 |
ophone | -0.056 |
Jupiter | -0.054 |
belonged | -0.053 |
endant | -0.051 |
Token | to |
Token ID | 284 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
Son | +0.238 |
distress | +0.231 |
Dad | +0.231 |
iver | +0.229 |
No | +0.226 |
Neg logit contributions | |
ophone | -0.121 |
Jupiter | -0.119 |
belongs | -0.113 |
unic | -0.104 |
dark | -0.102 |
Token | fight |
Token ID | 1907 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
Son | +0.042 |
umber | +0.040 |
P | +0.040 |
ascal | +0.038 |
Dad | +0.038 |
Neg logit contributions | |
where | -0.024 |
Ellie | -0.023 |
Lily | -0.023 |
Jupiter | -0.023 |
they | -0.023 |
Token | They |
Token ID | 1119 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
Dad | +0.001 |
Son | +0.001 |
No | +0.001 |
distress | +0.001 |
Be | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
belongs | -0.001 |
unic | -0.001 |
endant | -0.001 |
Token | pulled |
Token ID | 5954 |
Feature activation | +2.63e+00 |
Pos logit contributions | |
Son | +0.070 |
iver | +0.068 |
distress | +0.068 |
No | +0.067 |
Dad | +0.066 |
Neg logit contributions | |
ophone | -0.046 |
belongs | -0.045 |
Jupiter | -0.044 |
belonged | -0.043 |
existed | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | +2.30e-01 |
Pos logit contributions | |
Son | +0.156 |
iver | +0.151 |
distress | +0.150 |
Marcus | +0.146 |
Col | +0.144 |
Neg logit contributions | |
Jupiter | -0.110 |
belongs | -0.101 |
ophone | -0.100 |
existed | -0.094 |
lights | -0.092 |
Token | pushed |
Token ID | 7121 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
Son | +0.010 |
distress | +0.010 |
iver | +0.010 |
Dad | +0.009 |
No | +0.009 |
Neg logit contributions | |
ophone | -0.013 |
belongs | -0.012 |
Jupiter | -0.012 |
unic | -0.012 |
dark | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
Son | +0.128 |
distress | +0.124 |
iver | +0.122 |
Be | +0.120 |
Marcus | +0.119 |
Neg logit contributions | |
Jupiter | -0.070 |
ophone | -0.064 |
belongs | -0.063 |
existed | -0.061 |
lights | -0.061 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
No | +0.008 |
Son | +0.008 |
distress | +0.008 |
peel | +0.008 |
Dad | +0.008 |
Neg logit contributions | |
Jupiter | -0.014 |
ophone | -0.014 |
belongs | -0.014 |
unic | -0.013 |
dark | -0.013 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.57e+00 |
Pos logit contributions | |
distress | +0.550 |
Son | +0.539 |
iver | +0.524 |
Be | +0.523 |
No | +0.499 |
Neg logit contributions | |
ophone | -0.280 |
existed | -0.274 |
belongs | -0.266 |
Jupiter | -0.258 |
lights | -0.256 |
Token | , |
Token ID | 11 |
Feature activation | -2.57e-01 |
Pos logit contributions | |
Son | +0.297 |
distress | +0.291 |
iver | +0.287 |
Dad | +0.285 |
Be | +0.285 |
Neg logit contributions | |
ophone | -0.149 |
Jupiter | -0.149 |
belongs | -0.146 |
existed | -0.138 |
lights | -0.130 |
Token | but |
Token ID | 475 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
Jupiter | +0.012 |
ophone | +0.012 |
belongs | +0.012 |
unic | +0.011 |
dark | +0.011 |
Neg logit contributions | |
Son | -0.013 |
distress | -0.013 |
Dad | -0.013 |
No | -0.013 |
iver | -0.012 |
Token | they |
Token ID | 484 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
ophone | +0.002 |
Jupiter | +0.002 |
dark | +0.002 |
belongs | +0.002 |
endant | +0.002 |
Neg logit contributions | |
No | -0.002 |
Dad | -0.002 |
Be | -0.002 |
distress | -0.002 |
Son | -0.002 |
Token | could |
Token ID | 714 |
Feature activation | -4.32e+00 |
Pos logit contributions | |
ophone | +0.041 |
Jupiter | +0.036 |
belonged | +0.036 |
belongs | +0.036 |
existed | +0.035 |
Neg logit contributions | |
Son | -0.097 |
distress | -0.096 |
iver | -0.095 |
No | -0.094 |
Dad | -0.092 |
Token | her |
Token ID | 607 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
ophone | +0.031 |
Jupiter | +0.029 |
belongs | +0.029 |
endant | +0.027 |
unic | +0.027 |
Neg logit contributions | |
distress | -0.036 |
No | -0.036 |
Dad | -0.036 |
Son | -0.036 |
peel | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
ophone | +0.043 |
Jupiter | +0.043 |
dark | +0.041 |
belongs | +0.040 |
unic | +0.040 |
Neg logit contributions | |
distress | -0.033 |
Son | -0.031 |
peel | -0.031 |
No | -0.030 |
Dad | -0.030 |
Token | They |
Token ID | 1119 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
Son | +0.005 |
distress | +0.005 |
Dad | +0.005 |
No | +0.005 |
iver | +0.005 |
Neg logit contributions | |
ophone | -0.004 |
Jupiter | -0.004 |
belongs | -0.003 |
dark | -0.003 |
unic | -0.003 |
Token | pulled |
Token ID | 5954 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
iver | +0.074 |
Son | +0.072 |
distress | +0.072 |
Col | +0.069 |
No | +0.068 |
Neg logit contributions | |
ophone | -0.054 |
belonged | -0.053 |
belongs | -0.051 |
Jupiter | -0.051 |
couldn | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
Son | +0.163 |
iver | +0.155 |
Col | +0.154 |
distress | +0.153 |
Dad | +0.153 |
Neg logit contributions | |
Jupiter | -0.104 |
belongs | -0.098 |
ophone | -0.097 |
lights | -0.089 |
they | -0.088 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
iver | +0.013 |
distress | +0.012 |
Son | +0.012 |
Dad | +0.012 |
Col | +0.012 |
Neg logit contributions | |
ophone | -0.017 |
Jupiter | -0.017 |
belongs | -0.016 |
belonged | -0.016 |
unic | -0.016 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.53e+00 |
Pos logit contributions | |
distress | +0.445 |
Son | +0.440 |
iver | +0.419 |
Be | +0.407 |
Dad | +0.396 |
Neg logit contributions | |
ophone | -0.391 |
belongs | -0.367 |
Jupiter | -0.361 |
existed | -0.360 |
lights | -0.354 |
Token | , |
Token ID | 11 |
Feature activation | -1.48e-01 |
Pos logit contributions | |
Son | +0.271 |
distress | +0.264 |
iver | +0.261 |
Dad | +0.260 |
Col | +0.254 |
Neg logit contributions | |
ophone | -0.170 |
Jupiter | -0.167 |
belongs | -0.164 |
existed | -0.157 |
lights | -0.151 |
Token | and |
Token ID | 290 |
Feature activation | +3.06e-01 |
Pos logit contributions | |
Jupiter | +0.006 |
ophone | +0.006 |
belongs | +0.006 |
unic | +0.006 |
existed | +0.006 |
Neg logit contributions | |
Son | -0.008 |
Dad | -0.008 |
distress | -0.008 |
No | -0.008 |
iver | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
Son | +0.016 |
distress | +0.016 |
No | +0.015 |
Dad | +0.015 |
Be | +0.015 |
Neg logit contributions | |
ophone | -0.015 |
Jupiter | -0.014 |
belongs | -0.014 |
unic | -0.013 |
dark | -0.013 |
Token | hose |
Token ID | 31489 |
Feature activation | +7.84e-02 |
Pos logit contributions | |
Jupiter | +0.015 |
belongs | +0.015 |
ophone | +0.014 |
lights | +0.014 |
unic | +0.013 |
Neg logit contributions | |
Son | -0.018 |
Col | -0.017 |
Dad | -0.017 |
Marcus | -0.017 |
iver | -0.017 |
Token | book |
Token ID | 1492 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
Jupiter | +0.018 |
ophone | +0.018 |
belongs | +0.017 |
unic | +0.017 |
dark | +0.015 |
Neg logit contributions | |
Son | -0.023 |
Dad | -0.022 |
distress | -0.022 |
iver | -0.021 |
No | -0.021 |
Token | . |
Token ID | 13 |
Feature activation | -7.83e-02 |
Pos logit contributions | |
ophone | +0.066 |
Jupiter | +0.066 |
unic | +0.063 |
belongs | +0.061 |
dark | +0.061 |
Neg logit contributions | |
No | -0.071 |
distress | -0.071 |
peel | -0.070 |
Son | -0.069 |
Dad | -0.069 |
Token | They |
Token ID | 1119 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
ophone | +0.003 |
Jupiter | +0.003 |
belongs | +0.003 |
unic | +0.003 |
dark | +0.003 |
Neg logit contributions | |
Son | -0.004 |
distress | -0.004 |
No | -0.004 |
Dad | -0.004 |
Be | -0.004 |
Token | pulled |
Token ID | 5954 |
Feature activation | +2.66e+00 |
Pos logit contributions | |
Son | +0.062 |
distress | +0.061 |
iver | +0.060 |
Dad | +0.059 |
No | +0.058 |
Neg logit contributions | |
ophone | -0.056 |
Jupiter | -0.054 |
belongs | -0.054 |
belonged | -0.051 |
unic | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
Son | +0.161 |
iver | +0.157 |
distress | +0.151 |
Marcus | +0.151 |
Col | +0.149 |
Neg logit contributions | |
Jupiter | -0.107 |
belongs | -0.103 |
ophone | -0.100 |
existed | -0.091 |
lights | -0.090 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.50e+00 |
Pos logit contributions | |
Son | +0.007 |
iver | +0.007 |
distress | +0.007 |
Dad | +0.007 |
Marcus | +0.006 |
Neg logit contributions | |
ophone | -0.009 |
belongs | -0.009 |
Jupiter | -0.009 |
belonged | -0.008 |
dark | -0.008 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
distress | +0.493 |
Son | +0.486 |
iver | +0.475 |
Be | +0.456 |
Brother | +0.441 |
Neg logit contributions | |
ophone | -0.342 |
belongs | -0.327 |
existed | -0.321 |
Jupiter | -0.314 |
lights | -0.303 |
Token | , |
Token ID | 11 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
Son | +0.264 |
iver | +0.259 |
distress | +0.257 |
Dad | +0.251 |
Be | +0.247 |
Neg logit contributions | |
ophone | -0.178 |
belongs | -0.173 |
Jupiter | -0.172 |
existed | -0.163 |
unic | -0.152 |
Token | and |
Token ID | 290 |
Feature activation | +1.77e-01 |
Pos logit contributions | |
Jupiter | +0.015 |
ophone | +0.015 |
belongs | +0.014 |
unic | +0.013 |
existed | +0.013 |
Neg logit contributions | |
Son | -0.016 |
distress | -0.015 |
Dad | -0.015 |
No | -0.015 |
iver | -0.015 |
Token | the |
Token ID | 262 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
No | +0.008 |
distress | +0.008 |
Son | +0.008 |
Dad | +0.008 |
Be | +0.008 |
Neg logit contributions | |
ophone | -0.010 |
Jupiter | -0.009 |
belongs | -0.009 |
unic | -0.009 |
dark | -0.009 |
Token | book |
Token ID | 1492 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ophone | +0.016 |
Jupiter | +0.016 |
belongs | +0.015 |
unic | +0.014 |
dark | +0.014 |
Neg logit contributions | |
Son | -0.016 |
distress | -0.016 |
No | -0.016 |
Dad | -0.016 |
Be | -0.015 |
Token | mine |
Token ID | 6164 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
distress | +0.094 |
No | +0.092 |
peel | +0.092 |
tug | +0.091 |
pack | +0.091 |
Neg logit contributions | |
ophone | -0.084 |
Jupiter | -0.082 |
dark | -0.079 |
lights | -0.077 |
unic | -0.077 |
Token | !" |
Token ID | 2474 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
Jupiter | +0.048 |
ophone | +0.047 |
unic | +0.044 |
dark | +0.043 |
belongs | +0.043 |
Neg logit contributions | |
No | -0.055 |
distress | -0.055 |
Son | -0.054 |
Dad | -0.054 |
iver | -0.052 |
Token | Mia |
Token ID | 32189 |
Feature activation | -9.32e-02 |
Pos logit contributions | |
Son | +0.039 |
distress | +0.039 |
Dad | +0.038 |
No | +0.038 |
iver | +0.037 |
Neg logit contributions | |
ophone | -0.023 |
Jupiter | -0.023 |
belongs | -0.021 |
unic | -0.020 |
dark | -0.020 |
Token | yelled |
Token ID | 22760 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ophone | +0.004 |
belongs | +0.003 |
Jupiter | +0.003 |
belonged | +0.003 |
existed | +0.003 |
Neg logit contributions | |
Son | -0.006 |
distress | -0.005 |
Dad | -0.005 |
Be | -0.005 |
iver | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
Son | +0.013 |
distress | +0.013 |
No | +0.013 |
Dad | +0.012 |
Be | +0.012 |
Neg logit contributions | |
ophone | -0.010 |
Jupiter | -0.009 |
belongs | -0.009 |
unic | -0.009 |
dark | -0.008 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
Jupiter | +0.005 |
ophone | +0.005 |
belongs | +0.005 |
unic | +0.004 |
belonged | +0.004 |
Neg logit contributions | |
Son | -0.006 |
Dad | -0.006 |
distress | -0.006 |
No | -0.006 |
Col | -0.006 |
Token | ging |
Token ID | 2667 |
Feature activation | +4.87e+00 |
Pos logit contributions | |
distress | +0.462 |
Son | +0.444 |
Be | +0.435 |
iver | +0.435 |
pack | +0.407 |
Neg logit contributions | |
ophone | -0.354 |
existed | -0.338 |
they | -0.332 |
belongs | -0.322 |
ls | -0.314 |
Token | on |
Token ID | 319 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
iver | +0.334 |
Son | +0.329 |
distress | +0.318 |
Marcus | +0.308 |
Brother | +0.307 |
Neg logit contributions | |
belongs | -0.133 |
ophone | -0.130 |
Jupiter | -0.125 |
existed | -0.123 |
they | -0.122 |
Token | the |
Token ID | 262 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
Jupiter | +0.067 |
ophone | +0.064 |
belongs | +0.062 |
unic | +0.058 |
existed | +0.057 |
Neg logit contributions | |
Son | -0.097 |
Dad | -0.093 |
distress | -0.093 |
iver | -0.093 |
No | -0.092 |
Token | other |
Token ID | 584 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ophone | +0.015 |
Jupiter | +0.015 |
belongs | +0.014 |
unic | +0.013 |
dark | +0.013 |
Neg logit contributions | |
iver | -0.017 |
Son | -0.017 |
Dad | -0.016 |
No | -0.016 |
distress | -0.016 |
Token | side |
Token ID | 1735 |
Feature activation | +3.36e+00 |
Pos logit contributions | |
ophone | +0.070 |
Jupiter | +0.069 |
belongs | +0.068 |
dark | +0.064 |
unic | +0.064 |
Neg logit contributions | |
Son | -0.056 |
distress | -0.055 |
No | -0.054 |
iver | -0.054 |
Be | -0.054 |
Token |
|
Token ID | 198 |
Feature activation | -7.16e-02 |
Pos logit contributions | |
Son | +0.022 |
distress | +0.022 |
No | +0.021 |
Dad | +0.021 |
Be | +0.021 |
Neg logit contributions | |
ophone | -0.023 |
Jupiter | -0.023 |
belongs | -0.022 |
unic | -0.021 |
dark | -0.021 |
Token |
|
Token ID | 198 |
Feature activation | -7.19e-02 |
Pos logit contributions | |
Jupiter | +0.003 |
ophone | +0.003 |
belongs | +0.003 |
unic | +0.003 |
dark | +0.003 |
Neg logit contributions | |
distress | -0.004 |
Son | -0.004 |
No | -0.004 |
Dad | -0.004 |
iver | -0.003 |
Token | They |
Token ID | 2990 |
Feature activation | +2.71e+00 |
Pos logit contributions | |
belongs | +0.003 |
dark | +0.003 |
ophone | +0.003 |
Jupiter | +0.003 |
they | +0.003 |
Neg logit contributions | |
distress | -0.003 |
reel | -0.003 |
Son | -0.003 |
worry | -0.003 |
umb | -0.003 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
Son | +0.121 |
distress | +0.109 |
Dad | +0.108 |
Be | +0.104 |
iver | +0.103 |
Neg logit contributions | |
ophone | -0.154 |
Jupiter | -0.151 |
belongs | -0.149 |
unic | -0.141 |
existed | -0.140 |
Token | and |
Token ID | 290 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
Son | +0.282 |
iver | +0.266 |
Marcus | +0.262 |
distress | +0.260 |
Col | +0.257 |
Neg logit contributions | |
Jupiter | -0.187 |
lights | -0.177 |
belongs | -0.176 |
ophone | -0.174 |
they | -0.164 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
Son | +0.008 |
iver | +0.007 |
distress | +0.007 |
Col | +0.007 |
Dad | +0.007 |
Neg logit contributions | |
belongs | -0.014 |
ophone | -0.014 |
they | -0.014 |
Jupiter | -0.014 |
we | -0.013 |
Token | on |
Token ID | 319 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
Son | +0.545 |
distress | +0.541 |
iver | +0.519 |
Be | +0.505 |
Baby | +0.488 |
Neg logit contributions | |
ophone | -0.290 |
belongs | -0.279 |
lights | -0.265 |
Jupiter | -0.261 |
existed | -0.261 |
Token | the |
Token ID | 262 |
Feature activation | -4.57e-01 |
Pos logit contributions | |
Jupiter | +0.077 |
ophone | +0.076 |
belongs | +0.074 |
unic | +0.070 |
dark | +0.068 |
Neg logit contributions | |
Son | -0.084 |
distress | -0.082 |
Dad | -0.081 |
No | -0.080 |
iver | -0.079 |
Token | v |
Token ID | 410 |
Feature activation | -4.47e+00 |
Pos logit contributions | |
ophone | +0.014 |
Jupiter | +0.014 |
belongs | +0.012 |
existed | +0.012 |
dark | +0.011 |
Neg logit contributions | |
distress | -0.031 |
No | -0.031 |
Son | -0.031 |
Be | -0.030 |
Dad | -0.030 |
Token | ase |
Token ID | 589 |
Feature activation | -7.08e-01 |
Pos logit contributions | |
ophone | +0.200 |
Jupiter | +0.187 |
belongs | +0.184 |
dark | +0.176 |
endant | +0.173 |
Neg logit contributions | |
distress | -0.251 |
Son | -0.250 |
No | -0.239 |
Dad | -0.237 |
Be | -0.234 |
Token | . |
Token ID | 13 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
ophone | +0.028 |
Jupiter | +0.028 |
belongs | +0.026 |
unic | +0.025 |
dark | +0.024 |
Neg logit contributions | |
Son | -0.042 |
distress | -0.042 |
Dad | -0.041 |
No | -0.041 |
Be | -0.040 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
No | +0.001 |
Son | +0.001 |
distress | +0.001 |
Dad | +0.001 |
Be | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
belongs | -0.001 |
unic | -0.001 |
endant | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e-01 |
Pos logit contributions | |
ophone | +0.068 |
belongs | +0.066 |
Jupiter | +0.066 |
unic | +0.061 |
belonged | +0.060 |
Neg logit contributions | |
Son | -0.074 |
distress | -0.073 |
Dad | -0.072 |
No | -0.071 |
iver | -0.070 |
Token | Ben |
Token ID | 3932 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
iver | +0.011 |
distress | +0.011 |
Son | +0.011 |
Dad | +0.011 |
No | +0.011 |
Neg logit contributions | |
ophone | -0.011 |
Jupiter | -0.011 |
belongs | -0.010 |
unic | -0.010 |
existed | -0.010 |
Token | started |
Token ID | 2067 |
Feature activation | -5.49e-01 |
Pos logit contributions | |
Son | +0.154 |
distress | +0.149 |
Dad | +0.148 |
iver | +0.148 |
No | +0.146 |
Neg logit contributions | |
ophone | -0.127 |
belongs | -0.123 |
Jupiter | -0.118 |
belonged | -0.118 |
existed | -0.111 |
Token | to |
Token ID | 284 |
Feature activation | +7.05e-01 |
Pos logit contributions | |
Jupiter | +0.013 |
ophone | +0.013 |
belongs | +0.012 |
unic | +0.011 |
belonged | +0.010 |
Neg logit contributions | |
Son | -0.041 |
iver | -0.040 |
Dad | -0.040 |
Marcus | -0.039 |
Brother | -0.038 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.49e+00 |
Pos logit contributions | |
Son | +0.042 |
Dad | +0.042 |
umber | +0.038 |
ascal | +0.038 |
Col | +0.038 |
Neg logit contributions | |
belongs | -0.028 |
Ellie | -0.026 |
Jupiter | -0.025 |
they | -0.023 |
belonged | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +2.56e-01 |
Pos logit contributions | |
Son | +0.585 |
distress | +0.584 |
iver | +0.566 |
Be | +0.552 |
Baby | +0.541 |
Neg logit contributions | |
belongs | -0.227 |
ophone | -0.227 |
existed | -0.221 |
they | -0.212 |
ls | -0.208 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.74e+00 |
Pos logit contributions | |
Son | +0.008 |
Dad | +0.007 |
iver | +0.007 |
distress | +0.007 |
No | +0.007 |
Neg logit contributions | |
ophone | -0.018 |
Jupiter | -0.017 |
belongs | -0.017 |
unic | -0.016 |
existed | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | +6.91e-03 |
Pos logit contributions | |
Son | +0.329 |
iver | +0.325 |
distress | +0.320 |
Marcus | +0.304 |
Be | +0.304 |
Neg logit contributions | |
Jupiter | -0.132 |
belongs | -0.124 |
ophone | -0.121 |
lights | -0.118 |
existed | -0.117 |
Token | The |
Token ID | 383 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
Son | +0.000 |
distress | +0.000 |
No | +0.000 |
Dad | +0.000 |
iver | +0.000 |
Neg logit contributions | |
Jupiter | -0.000 |
ophone | -0.000 |
belongs | -0.000 |
unic | -0.000 |
dark | -0.000 |
Token | blue |
Token ID | 4171 |
Feature activation | -2.70e+00 |
Pos logit contributions | |
Jupiter | +0.058 |
ophone | +0.058 |
belongs | +0.057 |
unic | +0.054 |
dark | +0.052 |
Neg logit contributions | |
Son | -0.083 |
distress | -0.079 |
iver | -0.079 |
Dad | -0.078 |
Marcus | -0.077 |
Token | both |
Token ID | 1111 |
Feature activation | +3.25e+00 |
Pos logit contributions | |
Son | +0.062 |
iver | +0.061 |
distress | +0.060 |
Dad | +0.059 |
No | +0.058 |
Neg logit contributions | |
ophone | -0.052 |
Jupiter | -0.050 |
belongs | -0.050 |
belonged | -0.049 |
unic | -0.046 |
Token | pulled |
Token ID | 5954 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
Son | +0.145 |
distress | +0.141 |
Dad | +0.139 |
iver | +0.136 |
No | +0.136 |
Neg logit contributions | |
ophone | -0.177 |
belongs | -0.171 |
Jupiter | -0.171 |
existed | -0.160 |
unic | -0.160 |
Token | and |
Token ID | 290 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
Son | +0.166 |
iver | +0.162 |
distress | +0.156 |
Col | +0.155 |
Marcus | +0.155 |
Neg logit contributions | |
Jupiter | -0.099 |
belongs | -0.095 |
ophone | -0.092 |
they | -0.087 |
existed | -0.084 |
Token | pushed |
Token ID | 7121 |
Feature activation | +2.02e+00 |
Pos logit contributions | |
Son | +0.008 |
iver | +0.008 |
distress | +0.008 |
Dad | +0.007 |
No | +0.007 |
Neg logit contributions | |
ophone | -0.010 |
belongs | -0.009 |
Jupiter | -0.009 |
belonged | -0.009 |
unic | -0.009 |
Token | and |
Token ID | 290 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
Son | +0.132 |
distress | +0.126 |
iver | +0.125 |
Marcus | +0.123 |
Col | +0.122 |
Neg logit contributions | |
Jupiter | -0.066 |
belongs | -0.062 |
ophone | -0.061 |
they | -0.059 |
existed | -0.058 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.48e+00 |
Pos logit contributions | |
No | +0.006 |
Son | +0.006 |
distress | +0.006 |
Be | +0.006 |
Dad | +0.006 |
Neg logit contributions | |
Jupiter | -0.011 |
ophone | -0.010 |
belongs | -0.010 |
unic | -0.010 |
dark | -0.010 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.49e+00 |
Pos logit contributions | |
distress | +0.503 |
Son | +0.498 |
iver | +0.477 |
Be | +0.470 |
Brother | +0.454 |
Neg logit contributions | |
ophone | -0.329 |
belongs | -0.314 |
existed | -0.311 |
Jupiter | -0.299 |
lights | -0.298 |
Token | on |
Token ID | 319 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
Son | +0.296 |
iver | +0.288 |
Dad | +0.285 |
distress | +0.285 |
Be | +0.279 |
Neg logit contributions | |
ophone | -0.143 |
belongs | -0.142 |
Jupiter | -0.138 |
existed | -0.131 |
lights | -0.122 |
Token | the |
Token ID | 262 |
Feature activation | -3.75e-01 |
Pos logit contributions | |
Jupiter | +0.075 |
ophone | +0.072 |
belongs | +0.070 |
unic | +0.067 |
lights | +0.065 |
Neg logit contributions | |
Son | -0.092 |
distress | -0.089 |
Dad | -0.089 |
iver | -0.089 |
No | -0.087 |
Token | tray |
Token ID | 26473 |
Feature activation | +2.17e+00 |
Pos logit contributions | |
ophone | +0.022 |
Jupiter | +0.022 |
belongs | +0.021 |
unic | +0.020 |
dark | +0.020 |
Neg logit contributions | |
Son | -0.016 |
distress | -0.015 |
Dad | -0.015 |
No | -0.015 |
iver | -0.014 |
Token | . |
Token ID | 13 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
Son | +0.122 |
distress | +0.122 |
No | +0.120 |
Dad | +0.119 |
Be | +0.116 |
Neg logit contributions | |
ophone | -0.091 |
Jupiter | -0.090 |
belongs | -0.085 |
unic | -0.083 |
dark | -0.081 |
Token | fight |
Token ID | 1907 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
Son | +0.030 |
Col | +0.030 |
Dad | +0.029 |
umber | +0.028 |
aleb | +0.027 |
Neg logit contributions | |
Jupiter | -0.028 |
belongs | -0.028 |
unic | -0.026 |
Ellie | -0.026 |
ophone | -0.025 |
Token | . |
Token ID | 13 |
Feature activation | -8.14e-02 |
Pos logit contributions | |
Son | +0.217 |
distress | +0.215 |
No | +0.212 |
Dad | +0.212 |
Be | +0.206 |
Neg logit contributions | |
ophone | -0.150 |
Jupiter | -0.148 |
belongs | -0.141 |
unic | -0.134 |
dark | -0.131 |
Token | They |
Token ID | 1119 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ophone | +0.004 |
Jupiter | +0.004 |
belongs | +0.004 |
unic | +0.004 |
endant | +0.004 |
Neg logit contributions | |
Son | -0.004 |
No | -0.004 |
Dad | -0.004 |
distress | -0.004 |
Be | -0.004 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.73e+00 |
Pos logit contributions | |
Son | +0.052 |
Dad | +0.049 |
distress | +0.045 |
iver | +0.044 |
Col | +0.044 |
Neg logit contributions | |
ophone | -0.076 |
belongs | -0.075 |
Jupiter | -0.074 |
belonged | -0.070 |
unic | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +9.97e-02 |
Pos logit contributions | |
Son | +0.291 |
iver | +0.288 |
distress | +0.276 |
Marcus | +0.276 |
Col | +0.270 |
Neg logit contributions | |
Jupiter | -0.174 |
belongs | -0.168 |
ophone | -0.164 |
lights | -0.164 |
they | -0.155 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.47e+00 |
Pos logit contributions | |
Son | +0.003 |
Col | +0.003 |
iver | +0.003 |
distress | +0.003 |
Dad | +0.003 |
Neg logit contributions | |
belongs | -0.007 |
ophone | -0.007 |
Jupiter | -0.007 |
they | -0.006 |
unic | -0.006 |
Token | on |
Token ID | 319 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
distress | +0.567 |
Son | +0.562 |
iver | +0.539 |
Be | +0.534 |
Col | +0.506 |
Neg logit contributions | |
ophone | -0.257 |
belongs | -0.250 |
ls | -0.241 |
lights | -0.240 |
they | -0.238 |
Token | the |
Token ID | 262 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
Jupiter | +0.066 |
belongs | +0.062 |
ophone | +0.062 |
unic | +0.057 |
lights | +0.056 |
Neg logit contributions | |
Son | -0.099 |
iver | -0.098 |
Dad | -0.095 |
distress | -0.093 |
Be | -0.092 |
Token | er |
Token ID | 1931 |
Feature activation | -4.82e-01 |
Pos logit contributions | |
ophone | +0.019 |
Jupiter | +0.019 |
belongs | +0.019 |
unic | +0.018 |
dark | +0.017 |
Neg logit contributions | |
Son | -0.021 |
Dad | -0.020 |
distress | -0.020 |
iver | -0.019 |
No | -0.019 |
Token | aser |
Token ID | 6005 |
Feature activation | -8.36e-02 |
Pos logit contributions | |
ophone | +0.023 |
Jupiter | +0.022 |
belongs | +0.021 |
unic | +0.020 |
dark | +0.020 |
Neg logit contributions | |
Son | -0.025 |
distress | -0.025 |
Dad | -0.025 |
No | -0.025 |
Be | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
ophone | +0.003 |
Jupiter | +0.003 |
belongs | +0.003 |
unic | +0.003 |
dark | +0.003 |
Neg logit contributions | |
Son | -0.005 |
distress | -0.005 |
No | -0.005 |
Dad | -0.005 |
Be | -0.005 |
Token | They |
Token ID | 1119 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
No | +0.008 |
Dad | +0.008 |
Son | +0.008 |
distress | +0.008 |
Be | +0.008 |
Neg logit contributions | |
ophone | -0.010 |
Jupiter | -0.009 |
belongs | -0.009 |
unic | -0.009 |
endant | -0.009 |
Token | start |
Token ID | 923 |
Feature activation | +3.63e+00 |
Pos logit contributions | |
Son | +0.059 |
Dad | +0.054 |
distress | +0.051 |
iver | +0.051 |
Col | +0.050 |
Neg logit contributions | |
belongs | -0.069 |
ophone | -0.068 |
Jupiter | -0.067 |
belonged | -0.063 |
unic | -0.062 |
Token | to |
Token ID | 284 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
Son | +0.246 |
Dad | +0.237 |
distress | +0.234 |
iver | +0.233 |
No | +0.230 |
Neg logit contributions | |
ophone | -0.120 |
Jupiter | -0.117 |
belongs | -0.110 |
unic | -0.101 |
dark | -0.099 |
Token | pull |
Token ID | 2834 |
Feature activation | +4.68e+00 |
Pos logit contributions | |
Son | +0.039 |
Dad | +0.038 |
umber | +0.037 |
P | +0.036 |
Col | +0.035 |
Neg logit contributions | |
Jupiter | -0.025 |
Zig | -0.024 |
Ellie | -0.023 |
belongs | -0.023 |
ophone | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
Son | +0.288 |
iver | +0.281 |
distress | +0.275 |
Marcus | +0.265 |
Col | +0.263 |
Neg logit contributions | |
Jupiter | -0.177 |
belongs | -0.169 |
ophone | -0.168 |
existed | -0.155 |
lights | -0.153 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.47e+00 |
Pos logit contributions | |
Son | +0.007 |
Dad | +0.006 |
Col | +0.006 |
iver | +0.006 |
distress | +0.006 |
Neg logit contributions | |
belongs | -0.015 |
ophone | -0.015 |
Jupiter | -0.015 |
existed | -0.014 |
belonged | -0.014 |
Token | on |
Token ID | 319 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
distress | +0.562 |
Son | +0.557 |
Be | +0.535 |
iver | +0.534 |
Baby | +0.503 |
Neg logit contributions | |
ophone | -0.267 |
belongs | -0.261 |
existed | -0.251 |
Jupiter | -0.239 |
lights | -0.239 |
Token | the |
Token ID | 262 |
Feature activation | -4.11e-01 |
Pos logit contributions | |
Jupiter | +0.074 |
ophone | +0.074 |
belongs | +0.071 |
unic | +0.067 |
existed | +0.066 |
Neg logit contributions | |
Son | -0.089 |
distress | -0.086 |
Dad | -0.085 |
No | -0.084 |
iver | -0.084 |
Token | doll |
Token ID | 3654 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
Jupiter | +0.021 |
ophone | +0.020 |
belongs | +0.019 |
unic | +0.019 |
existed | +0.018 |
Neg logit contributions | |
distress | -0.020 |
No | -0.020 |
Son | -0.020 |
Dad | -0.019 |
Be | -0.019 |
Token | , |
Token ID | 11 |
Feature activation | -2.52e-01 |
Pos logit contributions | |
ophone | +0.106 |
Jupiter | +0.105 |
belongs | +0.100 |
unic | +0.095 |
dark | +0.093 |
Neg logit contributions | |
Son | -0.143 |
distress | -0.141 |
No | -0.139 |
Dad | -0.139 |
Be | -0.135 |
Token | making |
Token ID | 1642 |
Feature activation | +6.83e-01 |
Pos logit contributions | |
ophone | +0.012 |
Jupiter | +0.012 |
belongs | +0.011 |
unic | +0.011 |
existed | +0.010 |
Neg logit contributions | |
Son | -0.013 |
Dad | -0.013 |
distress | -0.013 |
No | -0.012 |
iver | -0.012 |
Token | hands |
Token ID | 2832 |
Feature activation | +5.89e+00 |
Pos logit contributions | |
distress | +0.038 |
Son | +0.034 |
Dad | +0.032 |
peel | +0.032 |
No | +0.032 |
Neg logit contributions | |
ophone | -0.144 |
Jupiter | -0.142 |
dark | -0.138 |
belongs | -0.137 |
unic | -0.135 |
Token | . |
Token ID | 13 |
Feature activation | +2.07e-01 |
Pos logit contributions | |
Son | +0.288 |
distress | +0.284 |
Dad | +0.278 |
No | +0.277 |
iver | +0.266 |
Neg logit contributions | |
ophone | -0.292 |
Jupiter | -0.291 |
belongs | -0.279 |
unic | -0.272 |
dark | -0.267 |
Token | They |
Token ID | 1119 |
Feature activation | +1.35e+00 |
Pos logit contributions | |
Son | +0.011 |
No | +0.011 |
distress | +0.011 |
Dad | +0.011 |
Be | +0.010 |
Neg logit contributions | |
ophone | -0.010 |
Jupiter | -0.009 |
belongs | -0.009 |
unic | -0.009 |
endant | -0.009 |
Token | pulled |
Token ID | 5954 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
Son | +0.067 |
distress | +0.066 |
No | +0.065 |
Dad | +0.065 |
Be | +0.063 |
Neg logit contributions | |
ophone | -0.066 |
Jupiter | -0.066 |
belongs | -0.063 |
unic | -0.061 |
dark | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | +3.69e-01 |
Pos logit contributions | |
iver | +0.154 |
Son | +0.153 |
distress | +0.148 |
Dad | +0.147 |
Marcus | +0.144 |
Neg logit contributions | |
Jupiter | -0.111 |
ophone | -0.102 |
belongs | -0.102 |
Zig | -0.094 |
existed | -0.094 |
Token | tug |
Token ID | 27762 |
Feature activation | +8.46e+00 |
Pos logit contributions | |
Son | +0.014 |
iver | +0.014 |
distress | +0.014 |
Dad | +0.014 |
No | +0.013 |
Neg logit contributions | |
ophone | -0.021 |
Jupiter | -0.021 |
belongs | -0.021 |
unic | -0.020 |
belonged | -0.020 |
Token | ged |
Token ID | 2004 |
Feature activation | +4.48e+00 |
Pos logit contributions | |
Son | +0.419 |
distress | +0.418 |
iver | +0.412 |
Be | +0.400 |
Dad | +0.388 |
Neg logit contributions | |
ophone | -0.394 |
existed | -0.381 |
Jupiter | -0.379 |
belongs | -0.375 |
lights | -0.360 |
Token | and |
Token ID | 290 |
Feature activation | +3.71e-01 |
Pos logit contributions | |
Son | +0.255 |
distress | +0.251 |
iver | +0.249 |
Dad | +0.248 |
Be | +0.242 |
Neg logit contributions | |
ophone | -0.182 |
Jupiter | -0.181 |
belongs | -0.175 |
existed | -0.165 |
unic | -0.163 |
Token | ripped |
Token ID | 19551 |
Feature activation | +2.22e-01 |
Pos logit contributions | |
Son | +0.014 |
distress | +0.013 |
Dad | +0.013 |
No | +0.013 |
iver | +0.012 |
Neg logit contributions | |
ophone | -0.023 |
Jupiter | -0.023 |
belongs | -0.022 |
unic | -0.022 |
existed | -0.021 |
Token | Bob |
Token ID | 5811 |
Feature activation | -4.33e-01 |
Pos logit contributions | |
Son | +0.012 |
distress | +0.012 |
iver | +0.012 |
Be | +0.012 |
Dad | +0.012 |
Neg logit contributions | |
Jupiter | -0.009 |
ophone | -0.009 |
belongs | -0.009 |
lights | -0.009 |
unic | -0.008 |
Token | o |
Token ID | 78 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
ophone | +0.021 |
Jupiter | +0.020 |
belongs | +0.020 |
dark | +0.018 |
existed | +0.018 |
Neg logit contributions | |
Son | -0.022 |
distress | -0.022 |
No | -0.022 |
Dad | -0.022 |
Be | -0.021 |
Token | especially |
Token ID | 2592 |
Feature activation | -3.67e+00 |
Pos logit contributions | |
Jupiter | +0.028 |
ophone | +0.028 |
belongs | +0.026 |
lights | +0.026 |
dark | +0.025 |
Neg logit contributions | |
Son | -0.025 |
Dad | -0.025 |
distress | -0.024 |
No | -0.024 |
Be | -0.024 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
Jupiter | +0.183 |
ophone | +0.177 |
belongs | +0.167 |
endant | +0.162 |
lights | +0.162 |
Neg logit contributions | |
Son | -0.187 |
distress | -0.186 |
No | -0.185 |
Be | -0.183 |
Dad | -0.179 |
Token | rain |
Token ID | 6290 |
Feature activation | -4.27e+00 |
Pos logit contributions | |
ophone | +0.100 |
belongs | +0.094 |
dark | +0.090 |
endant | +0.088 |
Jupiter | +0.087 |
Neg logit contributions | |
distress | -0.079 |
Son | -0.077 |
No | -0.076 |
Be | -0.075 |
Dad | -0.074 |
Token | bows |
Token ID | 25435 |
Feature activation | -5.21e+00 |
Pos logit contributions | |
ophone | +0.167 |
Jupiter | +0.165 |
belongs | +0.157 |
dark | +0.151 |
endant | +0.147 |
Neg logit contributions | |
No | -0.247 |
distress | -0.247 |
Son | -0.245 |
iver | -0.244 |
Be | -0.238 |
Token | and |
Token ID | 290 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
ophone | +0.312 |
Jupiter | +0.309 |
belongs | +0.299 |
dark | +0.285 |
endant | +0.285 |
Neg logit contributions | |
distress | -0.194 |
Son | -0.192 |
No | -0.190 |
iver | -0.184 |
Dad | -0.182 |
Token | unic |
Token ID | 28000 |
Feature activation | -1.03e+01 |
Pos logit contributions | |
Be | +0.001 |
No | +0.001 |
distress | +0.001 |
worry | +0.001 |
pack | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
belongs | -0.001 |
Jupiter | -0.001 |
endant | -0.001 |
dark | -0.001 |
Token | orns |
Token ID | 19942 |
Feature activation | -2.49e+00 |
Pos logit contributions | |
ophone | +0.628 |
belongs | +0.614 |
Jupiter | +0.611 |
existed | +0.566 |
dark | +0.556 |
Neg logit contributions | |
Son | -0.386 |
Dad | -0.375 |
iver | -0.371 |
distress | -0.369 |
Be | -0.357 |
Token | . |
Token ID | 13 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
Jupiter | +0.133 |
ophone | +0.133 |
belongs | +0.127 |
dark | +0.125 |
endant | +0.120 |
Neg logit contributions | |
Son | -0.105 |
distress | -0.105 |
peel | -0.100 |
iver | -0.099 |
No | -0.099 |
Token |
|
Token ID | 198 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
ophone | +0.002 |
belongs | +0.002 |
lights | +0.002 |
endant | +0.002 |
Jupiter | +0.002 |
Neg logit contributions | |
Son | -0.002 |
Dad | -0.002 |
No | -0.001 |
Be | -0.001 |
Brother | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
Son | +0.001 |
distress | +0.001 |
No | +0.001 |
Dad | +0.001 |
Be | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
belongs | -0.001 |
unic | -0.001 |
dark | -0.001 |
Token | One |
Token ID | 3198 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
distress | +0.001 |
Son | +0.001 |
reel | +0.001 |
peel | +0.001 |
iver | +0.001 |
Neg logit contributions | |
Jupiter | -0.002 |
ophone | -0.002 |
dark | -0.001 |
unic | -0.001 |
belongs | -0.001 |
Token | and |
Token ID | 290 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
ophone | +0.003 |
belongs | +0.003 |
Jupiter | +0.003 |
dark | +0.002 |
endant | +0.002 |
Neg logit contributions | |
distress | -0.002 |
Son | -0.002 |
No | -0.002 |
Dad | -0.002 |
Be | -0.002 |
Token | had |
Token ID | 550 |
Feature activation | -1.22e+00 |
Pos logit contributions | |
pack | +0.010 |
worry | +0.010 |
M | +0.010 |
panic | +0.010 |
tug | +0.010 |
Neg logit contributions | |
endant | -0.015 |
ophone | -0.015 |
belongs | -0.015 |
Jupiter | -0.015 |
lights | -0.015 |
Token | sweet |
Token ID | 6029 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
ophone | +0.080 |
Jupiter | +0.076 |
belongs | +0.076 |
lights | +0.071 |
dark | +0.071 |
Neg logit contributions | |
distress | -0.044 |
Be | -0.044 |
No | -0.043 |
Son | -0.042 |
M | -0.041 |
Token | dreams |
Token ID | 10625 |
Feature activation | -4.35e+00 |
Pos logit contributions | |
distress | +0.116 |
No | +0.111 |
Son | +0.110 |
iver | +0.110 |
Dad | +0.109 |
Neg logit contributions | |
ophone | -0.088 |
Jupiter | -0.084 |
belongs | -0.079 |
existed | -0.073 |
endant | -0.072 |
Token | about |
Token ID | 546 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
belongs | +0.229 |
Jupiter | +0.228 |
ophone | +0.225 |
endant | +0.214 |
dark | +0.208 |
Neg logit contributions | |
iver | -0.197 |
he | -0.194 |
pack | -0.187 |
No | -0.184 |
distress | -0.182 |
Token | unic |
Token ID | 28000 |
Feature activation | -1.02e+01 |
Pos logit contributions | |
ophone | +0.197 |
belongs | +0.180 |
dark | +0.173 |
endant | +0.172 |
Jupiter | +0.160 |
Neg logit contributions | |
distress | -0.107 |
worry | -0.089 |
Son | -0.086 |
Be | -0.086 |
sausage | -0.085 |
Token | orns |
Token ID | 19942 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ophone | +0.632 |
belongs | +0.623 |
Jupiter | +0.620 |
existed | +0.572 |
Venus | +0.564 |
Neg logit contributions | |
iver | -0.374 |
Son | -0.368 |
distress | -0.359 |
Dad | -0.349 |
he | -0.347 |
Token | and |
Token ID | 290 |
Feature activation | +2.23e-01 |
Pos logit contributions | |
Jupiter | +0.132 |
ophone | +0.129 |
belongs | +0.121 |
endant | +0.121 |
dark | +0.120 |
Neg logit contributions | |
he | -0.089 |
peel | -0.086 |
iver | -0.085 |
Son | -0.085 |
distress | -0.084 |
Token | rain |
Token ID | 6290 |
Feature activation | -4.29e+00 |
Pos logit contributions | |
distress | +0.010 |
worry | +0.010 |
Be | +0.010 |
pack | +0.009 |
panic | +0.009 |
Neg logit contributions | |
ophone | -0.013 |
belongs | -0.012 |
dark | -0.011 |
endant | -0.011 |
Jupiter | -0.011 |
Token | bows |
Token ID | 25435 |
Feature activation | -5.08e+00 |
Pos logit contributions | |
Jupiter | +0.174 |
ophone | +0.171 |
belongs | +0.169 |
endant | +0.156 |
dark | +0.152 |
Neg logit contributions | |
iver | -0.244 |
No | -0.237 |
distress | -0.235 |
ouch | -0.234 |
he | -0.231 |
Token | . |
Token ID | 13 |
Feature activation | +2.20e-01 |
Pos logit contributions | |
ophone | +0.269 |
Jupiter | +0.269 |
belongs | +0.258 |
endant | +0.245 |
dark | +0.241 |
Neg logit contributions | |
distress | -0.222 |
No | -0.220 |
Son | -0.218 |
iver | -0.213 |
peel | -0.213 |
Token | dreaming |
Token ID | 28368 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
ophone | +0.002 |
belongs | +0.002 |
Jupiter | +0.002 |
dark | +0.002 |
endant | +0.002 |
Neg logit contributions | |
distress | -0.002 |
Son | -0.002 |
Be | -0.002 |
pack | -0.002 |
No | -0.002 |
Token | of |
Token ID | 286 |
Feature activation | +2.26e-01 |
Pos logit contributions | |
ophone | +0.160 |
Jupiter | +0.157 |
belongs | +0.155 |
endant | +0.143 |
Venus | +0.139 |
Neg logit contributions | |
No | -0.069 |
distress | -0.068 |
he | -0.064 |
Be | -0.062 |
peel | -0.061 |
Token | rain |
Token ID | 6290 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
distress | +0.009 |
No | +0.008 |
Son | +0.008 |
Be | +0.008 |
peel | +0.008 |
Neg logit contributions | |
ophone | -0.015 |
belongs | -0.013 |
Jupiter | -0.013 |
dark | -0.013 |
endant | -0.012 |
Token | bows |
Token ID | 25435 |
Feature activation | -4.96e+00 |
Pos logit contributions | |
Jupiter | +0.190 |
ophone | +0.184 |
belongs | +0.178 |
endant | +0.175 |
dark | +0.167 |
Neg logit contributions | |
iver | -0.224 |
No | -0.218 |
distress | -0.215 |
ouch | -0.211 |
Be | -0.210 |
Token | and |
Token ID | 290 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
Jupiter | +0.302 |
ophone | +0.301 |
belongs | +0.286 |
endant | +0.279 |
dark | +0.273 |
Neg logit contributions | |
No | -0.176 |
distress | -0.175 |
Son | -0.172 |
peel | -0.172 |
he | -0.171 |
Token | unic |
Token ID | 28000 |
Feature activation | -1.01e+01 |
Pos logit contributions | |
Be | +0.007 |
distress | +0.006 |
worry | +0.006 |
No | +0.006 |
panic | +0.006 |
Neg logit contributions | |
ophone | -0.010 |
belongs | -0.009 |
Jupiter | -0.008 |
endant | -0.008 |
existed | -0.008 |
Token | orns |
Token ID | 19942 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
ophone | +0.623 |
Jupiter | +0.610 |
belongs | +0.602 |
existed | +0.561 |
Venus | +0.551 |
Neg logit contributions | |
iver | -0.377 |
Son | -0.368 |
distress | -0.366 |
Dad | -0.357 |
ouch | -0.347 |
Token | instead |
Token ID | 2427 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
Jupiter | +0.122 |
ophone | +0.118 |
belongs | +0.112 |
dark | +0.110 |
endant | +0.109 |
Neg logit contributions | |
Son | -0.093 |
he | -0.093 |
peel | -0.092 |
distress | -0.092 |
iver | -0.091 |
Token | of |
Token ID | 286 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
distress | +0.011 |
No | +0.011 |
M | +0.011 |
iver | +0.011 |
he | +0.010 |
Neg logit contributions | |
ophone | -0.023 |
Jupiter | -0.023 |
endant | -0.021 |
icol | -0.021 |
Venus | -0.020 |
Token | nightmares |
Token ID | 33301 |
Feature activation | -5.32e+00 |
Pos logit contributions | |
distress | +0.007 |
worry | +0.006 |
peel | +0.006 |
No | +0.006 |
Be | +0.006 |
Neg logit contributions | |
ophone | -0.009 |
belongs | -0.008 |
dark | -0.008 |
Jupiter | -0.008 |
endant | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
Jupiter | +0.298 |
ophone | +0.296 |
endant | +0.293 |
belongs | +0.281 |
dark | +0.268 |
Neg logit contributions | |
iver | -0.217 |
he | -0.216 |
distress | -0.202 |
No | -0.201 |
peel | -0.200 |
Token | full |
Token ID | 1336 |
Feature activation | +2.01e-01 |
Pos logit contributions | |
Jupiter | +0.157 |
ophone | +0.157 |
unic | +0.150 |
dark | +0.147 |
belongs | +0.146 |
Neg logit contributions | |
distress | -0.116 |
Son | -0.115 |
he | -0.114 |
Dad | -0.111 |
No | -0.111 |
Token | of |
Token ID | 286 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
distress | +0.006 |
Dad | +0.006 |
Son | +0.006 |
No | +0.006 |
peel | +0.005 |
Neg logit contributions | |
ophone | -0.015 |
Jupiter | -0.013 |
belongs | -0.013 |
endant | -0.013 |
dark | -0.012 |
Token | rain |
Token ID | 6290 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
distress | +0.001 |
Dad | +0.001 |
Be | +0.001 |
No | +0.001 |
pack | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
belongs | -0.001 |
dark | -0.001 |
lights | -0.001 |
Jupiter | -0.001 |
Token | bows |
Token ID | 25435 |
Feature activation | -4.87e+00 |
Pos logit contributions | |
ophone | +0.213 |
Jupiter | +0.210 |
belongs | +0.201 |
dark | +0.191 |
unic | +0.183 |
Neg logit contributions | |
iver | -0.194 |
distress | -0.190 |
pack | -0.187 |
peel | -0.187 |
he | -0.178 |
Token | and |
Token ID | 290 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
ophone | +0.281 |
Jupiter | +0.272 |
belongs | +0.263 |
dark | +0.252 |
endant | +0.249 |
Neg logit contributions | |
distress | -0.195 |
peel | -0.188 |
Son | -0.185 |
pack | -0.184 |
No | -0.180 |
Token | unic |
Token ID | 28000 |
Feature activation | -1.00e+01 |
Pos logit contributions | |
distress | +0.005 |
Be | +0.005 |
worry | +0.005 |
panic | +0.005 |
No | +0.005 |
Neg logit contributions | |
ophone | -0.007 |
belongs | -0.006 |
existed | -0.006 |
dark | -0.006 |
endant | -0.006 |
Token | orns |
Token ID | 19942 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
ophone | +0.668 |
belongs | +0.632 |
Jupiter | +0.632 |
dark | +0.587 |
existed | +0.587 |
Neg logit contributions | |
Son | -0.325 |
distress | -0.325 |
iver | -0.312 |
Dad | -0.310 |
Col | -0.304 |
Token | . |
Token ID | 13 |
Feature activation | -8.48e-03 |
Pos logit contributions | |
Jupiter | +0.116 |
ophone | +0.116 |
dark | +0.111 |
belongs | +0.108 |
lights | +0.104 |
Neg logit contributions | |
distress | -0.090 |
peel | -0.085 |
Son | -0.085 |
he | -0.083 |
iver | -0.082 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
belongs | +0.001 |
ophone | +0.001 |
existed | +0.001 |
unic | +0.000 |
endant | +0.000 |
Neg logit contributions | |
Dad | -0.000 |
Son | -0.000 |
M | -0.000 |
Brother | -0.000 |
Be | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
ophone | +0.009 |
belongs | +0.009 |
Jupiter | +0.008 |
existed | +0.008 |
endant | +0.008 |
Neg logit contributions | |
Dad | -0.006 |
Son | -0.006 |
No | -0.005 |
Be | -0.005 |
distress | -0.005 |
Token | But |
Token ID | 1537 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
ophone | +0.009 |
Jupiter | +0.009 |
belongs | +0.009 |
dark | +0.008 |
unic | +0.008 |
Neg logit contributions | |
distress | -0.005 |
Son | -0.005 |
iver | -0.005 |
No | -0.004 |
Dad | -0.004 |
Token | about |
Token ID | 546 |
Feature activation | -2.88e+00 |
Pos logit contributions | |
ophone | +0.079 |
Jupiter | +0.077 |
belongs | +0.074 |
endant | +0.073 |
unic | +0.067 |
Neg logit contributions | |
No | -0.134 |
iver | -0.133 |
Son | -0.130 |
distress | -0.130 |
peel | -0.129 |
Token | different |
Token ID | 1180 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ophone | +0.172 |
dark | +0.155 |
belongs | +0.155 |
lights | +0.152 |
endant | +0.149 |
Neg logit contributions | |
distress | -0.121 |
pack | -0.115 |
Be | -0.115 |
peel | -0.110 |
No | -0.108 |
Token | things |
Token ID | 1243 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
ophone | +0.084 |
resolved | +0.074 |
belongs | +0.070 |
endant | +0.069 |
existed | +0.068 |
Neg logit contributions | |
distress | -0.048 |
Dad | -0.045 |
No | -0.044 |
br | -0.044 |
pack | -0.044 |
Token | , |
Token ID | 11 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
he | +0.006 |
peel | +0.006 |
pack | +0.006 |
No | +0.006 |
distress | +0.005 |
Neg logit contributions | |
ophone | -0.014 |
majesty | -0.013 |
Jupiter | -0.013 |
icol | -0.013 |
dark | -0.012 |
Token | like |
Token ID | 588 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
No | +0.006 |
distress | +0.006 |
Son | +0.006 |
Dad | +0.005 |
Be | +0.005 |
Neg logit contributions | |
ophone | -0.007 |
Jupiter | -0.007 |
belongs | -0.007 |
dark | -0.006 |
endant | -0.006 |
Token | unic |
Token ID | 28000 |
Feature activation | -9.92e+00 |
Pos logit contributions | |
ophone | +0.091 |
belongs | +0.086 |
dark | +0.081 |
lights | +0.080 |
existed | +0.080 |
Neg logit contributions | |
Be | -0.068 |
pack | -0.066 |
distress | -0.066 |
No | -0.066 |
peel | -0.065 |
Token | orns |
Token ID | 19942 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ophone | +0.538 |
Jupiter | +0.523 |
belongs | +0.514 |
existed | +0.479 |
dark | +0.477 |
Neg logit contributions | |
Son | -0.443 |
distress | -0.437 |
Dad | -0.433 |
iver | -0.422 |
No | -0.421 |
Token | and |
Token ID | 290 |
Feature activation | +4.75e-01 |
Pos logit contributions | |
ophone | +0.116 |
Jupiter | +0.113 |
belongs | +0.106 |
dark | +0.106 |
lights | +0.104 |
Neg logit contributions | |
distress | -0.091 |
peel | -0.090 |
Son | -0.089 |
iver | -0.088 |
No | -0.086 |
Token | rain |
Token ID | 6290 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
distress | +0.023 |
worry | +0.022 |
pack | +0.022 |
Be | +0.022 |
peel | +0.022 |
Neg logit contributions | |
ophone | -0.026 |
belongs | -0.023 |
dark | -0.022 |
endant | -0.022 |
lights | -0.022 |
Token | bows |
Token ID | 25435 |
Feature activation | -4.93e+00 |
Pos logit contributions | |
ophone | +0.160 |
Jupiter | +0.154 |
belongs | +0.151 |
dark | +0.139 |
endant | +0.135 |
Neg logit contributions | |
iver | -0.241 |
peel | -0.237 |
distress | -0.236 |
No | -0.234 |
ouch | -0.232 |
Token | . |
Token ID | 13 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
ophone | +0.237 |
Jupiter | +0.233 |
belongs | +0.224 |
endant | +0.210 |
dark | +0.210 |
Neg logit contributions | |
distress | -0.243 |
Son | -0.240 |
No | -0.239 |
Dad | -0.236 |
peel | -0.234 |
Token | lots |
Token ID | 6041 |
Feature activation | -8.43e-01 |
Pos logit contributions | |
Dad | +0.041 |
No | +0.040 |
distress | +0.038 |
Be | +0.038 |
Son | +0.037 |
Neg logit contributions | |
ophone | -0.056 |
belongs | -0.055 |
Jupiter | -0.053 |
endant | -0.052 |
dark | -0.052 |
Token | of |
Token ID | 286 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ophone | +0.061 |
Jupiter | +0.059 |
belongs | +0.056 |
existed | +0.054 |
Venus | +0.052 |
Neg logit contributions | |
distress | -0.024 |
peel | -0.024 |
No | -0.023 |
Dad | -0.022 |
Son | -0.022 |
Token | silly |
Token ID | 14397 |
Feature activation | +2.42e-01 |
Pos logit contributions | |
No | +0.009 |
distress | +0.008 |
Dad | +0.008 |
Be | +0.008 |
pack | +0.008 |
Neg logit contributions | |
ophone | -0.009 |
Jupiter | -0.008 |
belongs | -0.008 |
existed | -0.008 |
endant | -0.008 |
Token | things |
Token ID | 1243 |
Feature activation | +1.34e-01 |
Pos logit contributions | |
distress | +0.011 |
No | +0.010 |
Dad | +0.010 |
peel | +0.010 |
pack | +0.010 |
Neg logit contributions | |
ophone | -0.013 |
Jupiter | -0.013 |
belongs | -0.012 |
endant | -0.012 |
existed | -0.012 |
Token | like |
Token ID | 588 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
distress | +0.005 |
Son | +0.005 |
pack | +0.005 |
peel | +0.005 |
No | +0.005 |
Neg logit contributions | |
Jupiter | -0.008 |
ophone | -0.007 |
dark | -0.007 |
unic | -0.007 |
endant | -0.007 |
Token | unic |
Token ID | 28000 |
Feature activation | -9.84e+00 |
Pos logit contributions | |
ophone | +0.083 |
belongs | +0.079 |
dark | +0.077 |
endant | +0.077 |
Jupiter | +0.076 |
Neg logit contributions | |
No | -0.071 |
distress | -0.071 |
Dad | -0.068 |
peel | -0.067 |
Son | -0.067 |
Token | orns |
Token ID | 19942 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
ophone | +0.590 |
Jupiter | +0.579 |
belongs | +0.568 |
existed | +0.536 |
endant | +0.525 |
Neg logit contributions | |
Son | -0.367 |
iver | -0.363 |
distress | -0.358 |
Dad | -0.354 |
Col | -0.339 |
Token | and |
Token ID | 290 |
Feature activation | +3.14e-01 |
Pos logit contributions | |
Jupiter | +0.103 |
ophone | +0.102 |
dark | +0.094 |
unic | +0.093 |
lights | +0.092 |
Neg logit contributions | |
distress | -0.083 |
peel | -0.080 |
Son | -0.080 |
iver | -0.078 |
pack | -0.077 |
Token | dragons |
Token ID | 20308 |
Feature activation | -3.57e+00 |
Pos logit contributions | |
distress | +0.014 |
worry | +0.014 |
No | +0.014 |
pack | +0.014 |
Be | +0.014 |
Neg logit contributions | |
ophone | -0.016 |
belongs | -0.015 |
endant | -0.015 |
dark | -0.015 |
Jupiter | -0.015 |
Token | ! |
Token ID | 0 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
ophone | +0.170 |
Jupiter | +0.169 |
belongs | +0.163 |
dark | +0.158 |
endant | +0.151 |
Neg logit contributions | |
distress | -0.181 |
Son | -0.175 |
No | -0.167 |
peel | -0.166 |
Dad | -0.165 |
Token | |
Token ID | 220 |
Feature activation | +9.85e-02 |
Pos logit contributions | |
Son | +0.051 |
Dad | +0.048 |
No | +0.044 |
distress | +0.042 |
toddler | +0.042 |
Neg logit contributions | |
ophone | -0.083 |
endant | -0.081 |
belongs | -0.078 |
unic | -0.078 |
existed | -0.074 |
Token | full |
Token ID | 1336 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ophone | +0.078 |
majesty | +0.072 |
Jupiter | +0.070 |
lights | +0.067 |
Majesty | +0.067 |
Neg logit contributions | |
distress | -0.054 |
he | -0.053 |
No | -0.051 |
Be | -0.050 |
pack | -0.049 |
Token | of |
Token ID | 286 |
Feature activation | +1.57e-01 |
Pos logit contributions | |
distress | +0.005 |
No | +0.005 |
Son | +0.004 |
Dad | +0.004 |
Be | +0.004 |
Neg logit contributions | |
ophone | -0.011 |
Jupiter | -0.010 |
belongs | -0.010 |
dark | -0.009 |
endant | -0.009 |
Token | fair |
Token ID | 3148 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
distress | +0.008 |
Be | +0.007 |
Son | +0.007 |
worry | +0.007 |
No | +0.007 |
Neg logit contributions | |
ophone | -0.009 |
belongs | -0.008 |
dark | -0.007 |
lights | -0.007 |
Jupiter | -0.007 |
Token | ies |
Token ID | 444 |
Feature activation | +6.87e-01 |
Pos logit contributions | |
distress | +0.005 |
pack | +0.004 |
Dad | +0.004 |
Son | +0.004 |
worry | +0.004 |
Neg logit contributions | |
ophone | -0.007 |
Jupiter | -0.007 |
dark | -0.006 |
belongs | -0.006 |
existed | -0.006 |
Token | and |
Token ID | 290 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
distress | +0.026 |
pack | +0.024 |
peel | +0.023 |
Son | +0.022 |
worry | +0.022 |
Neg logit contributions | |
ophone | -0.041 |
Jupiter | -0.039 |
dark | -0.039 |
belongs | -0.038 |
lights | -0.038 |
Token | unic |
Token ID | 28000 |
Feature activation | -9.75e+00 |
Pos logit contributions | |
worry | +0.013 |
distress | +0.012 |
pack | +0.012 |
panic | +0.012 |
Be | +0.011 |
Neg logit contributions | |
ophone | -0.019 |
dark | -0.015 |
belongs | -0.015 |
endant | -0.015 |
existed | -0.015 |
Token | orns |
Token ID | 19942 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
ophone | +0.585 |
Jupiter | +0.562 |
belongs | +0.549 |
dark | +0.521 |
existed | +0.514 |
Neg logit contributions | |
distress | -0.386 |
Son | -0.386 |
Dad | -0.368 |
iver | -0.366 |
Be | -0.355 |
Token | . |
Token ID | 13 |
Feature activation | +1.95e-01 |
Pos logit contributions | |
Jupiter | +0.106 |
ophone | +0.105 |
dark | +0.100 |
lights | +0.096 |
belongs | +0.094 |
Neg logit contributions | |
distress | -0.078 |
peel | -0.071 |
pack | -0.069 |
he | -0.068 |
Son | -0.067 |
Token | They |
Token ID | 1119 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
Be | +0.004 |
No | +0.004 |
toddler | +0.003 |
Dad | +0.003 |
distress | +0.003 |
Neg logit contributions | |
ophone | -0.016 |
aro | -0.014 |
existed | -0.014 |
lights | -0.014 |
belongs | -0.014 |
Token | welcomed |
Token ID | 15846 |
Feature activation | -3.97e+00 |
Pos logit contributions | |
distress | +0.053 |
Son | +0.052 |
Dad | +0.051 |
Be | +0.051 |
pack | +0.050 |
Neg logit contributions | |
ophone | -0.076 |
belongs | -0.074 |
lights | -0.072 |
dark | -0.071 |
Jupiter | -0.069 |
Token | her |
Token ID | 607 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
ophone | +0.255 |
belongs | +0.239 |
endant | +0.238 |
dark | +0.234 |
Jupiter | +0.230 |
Neg logit contributions | |
distress | -0.148 |
No | -0.143 |
Dad | -0.137 |
Son | -0.135 |
Be | -0.131 |
Token | She |
Token ID | 1375 |
Feature activation | -2.82e-01 |
Pos logit contributions | |
No | +0.009 |
Be | +0.008 |
distress | +0.008 |
peel | +0.008 |
Dad | +0.008 |
Neg logit contributions | |
ophone | -0.020 |
Jupiter | -0.019 |
lights | -0.019 |
unic | -0.018 |
endant | -0.018 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
lights | +0.017 |
ophone | +0.016 |
dark | +0.016 |
Jupiter | +0.015 |
belongs | +0.015 |
Neg logit contributions | |
tug | -0.011 |
Be | -0.011 |
pack | -0.011 |
scoop | -0.010 |
No | -0.010 |
Token | rain |
Token ID | 6290 |
Feature activation | -4.21e+00 |
Pos logit contributions | |
ophone | +0.075 |
belongs | +0.070 |
endant | +0.070 |
dark | +0.069 |
Jupiter | +0.067 |
Neg logit contributions | |
distress | -0.059 |
No | -0.057 |
Be | -0.054 |
Son | -0.054 |
Dad | -0.051 |
Token | bows |
Token ID | 25435 |
Feature activation | -4.65e+00 |
Pos logit contributions | |
ophone | +0.193 |
Jupiter | +0.186 |
belongs | +0.185 |
dark | +0.175 |
endant | +0.174 |
Neg logit contributions | |
distress | -0.217 |
No | -0.215 |
Be | -0.212 |
Son | -0.209 |
iver | -0.207 |
Token | and |
Token ID | 290 |
Feature activation | +2.95e-01 |
Pos logit contributions | |
ophone | +0.262 |
Jupiter | +0.256 |
belongs | +0.248 |
dark | +0.237 |
endant | +0.235 |
Neg logit contributions | |
distress | -0.195 |
Son | -0.192 |
No | -0.190 |
Dad | -0.185 |
Be | -0.183 |
Token | unic |
Token ID | 28000 |
Feature activation | -9.69e+00 |
Pos logit contributions | |
Be | +0.013 |
No | +0.013 |
distress | +0.013 |
worry | +0.012 |
panic | +0.012 |
Neg logit contributions | |
ophone | -0.017 |
belongs | -0.016 |
endant | -0.015 |
dark | -0.015 |
Jupiter | -0.014 |
Token | orns |
Token ID | 19942 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ophone | +0.490 |
Jupiter | +0.486 |
belongs | +0.468 |
unic | +0.447 |
dark | +0.442 |
Neg logit contributions | |
Son | -0.469 |
distress | -0.463 |
No | -0.456 |
Dad | -0.454 |
Be | -0.440 |
Token | and |
Token ID | 290 |
Feature activation | +2.89e-01 |
Pos logit contributions | |
ophone | +0.099 |
Jupiter | +0.098 |
belongs | +0.094 |
dark | +0.093 |
unic | +0.089 |
Neg logit contributions | |
distress | -0.082 |
Son | -0.079 |
peel | -0.078 |
No | -0.077 |
Be | -0.077 |
Token | lots |
Token ID | 6041 |
Feature activation | -9.22e-01 |
Pos logit contributions | |
No | +0.012 |
Be | +0.012 |
distress | +0.012 |
worry | +0.012 |
pack | +0.012 |
Neg logit contributions | |
ophone | -0.016 |
endant | -0.015 |
belongs | -0.015 |
dark | -0.015 |
Jupiter | -0.015 |
Token | of |
Token ID | 286 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
ophone | +0.076 |
Jupiter | +0.073 |
belongs | +0.070 |
endant | +0.066 |
Venus | +0.064 |
Neg logit contributions | |
peel | -0.018 |
No | -0.018 |
he | -0.017 |
Be | -0.016 |
distress | -0.015 |
Token | other |
Token ID | 584 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
distress | +0.006 |
Be | +0.006 |
No | +0.006 |
panic | +0.006 |
tug | +0.006 |
Neg logit contributions | |
ophone | -0.009 |
belongs | -0.008 |
Jupiter | -0.008 |
endant | -0.008 |
existed | -0.007 |
Token | is |
Token ID | 318 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
lights | +0.035 |
Jupiter | +0.035 |
ophone | +0.035 |
endant | +0.034 |
existed | +0.032 |
Neg logit contributions | |
No | -0.036 |
pack | -0.036 |
Be | -0.035 |
Son | -0.034 |
distress | -0.034 |
Token | an |
Token ID | 281 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
lights | +0.013 |
ophone | +0.013 |
endant | +0.012 |
ways | +0.012 |
belongs | +0.012 |
Neg logit contributions | |
pack | -0.007 |
sausage | -0.006 |
peel | -0.006 |
M | -0.006 |
half | -0.006 |
Token | incredible |
Token ID | 8082 |
Feature activation | -1.65e+00 |
Pos logit contributions | |
ophone | +0.042 |
Jupiter | +0.042 |
belongs | +0.040 |
unic | +0.038 |
dark | +0.037 |
Neg logit contributions | |
Son | -0.069 |
distress | -0.069 |
No | -0.068 |
Dad | -0.068 |
Be | -0.066 |
Token | magazine |
Token ID | 7093 |
Feature activation | +3.92e-01 |
Pos logit contributions | |
ophone | +0.091 |
Jupiter | +0.089 |
belongs | +0.086 |
dark | +0.083 |
endant | +0.083 |
Neg logit contributions | |
distress | -0.071 |
No | -0.071 |
Son | -0.070 |
Dad | -0.069 |
Be | -0.068 |
Token | about |
Token ID | 546 |
Feature activation | -2.94e+00 |
Pos logit contributions | |
peel | +0.016 |
No | +0.016 |
he | +0.016 |
pack | +0.016 |
customer | +0.015 |
Neg logit contributions | |
Jupiter | -0.023 |
lights | -0.021 |
ophone | -0.021 |
endant | -0.021 |
Venus | -0.020 |
Token | unic |
Token ID | 28000 |
Feature activation | -9.68e+00 |
Pos logit contributions | |
ophone | +0.171 |
lights | +0.169 |
dark | +0.157 |
endant | +0.156 |
belongs | +0.150 |
Neg logit contributions | |
distress | -0.111 |
pack | -0.108 |
Be | -0.108 |
peel | -0.105 |
sausage | -0.105 |
Token | orns |
Token ID | 19942 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
ophone | +0.523 |
Jupiter | +0.520 |
belongs | +0.505 |
existed | +0.472 |
unic | +0.472 |
Neg logit contributions | |
Son | -0.426 |
distress | -0.416 |
Dad | -0.414 |
iver | -0.410 |
No | -0.407 |
Token | !" |
Token ID | 2474 |
Feature activation | +8.83e-01 |
Pos logit contributions | |
Jupiter | +0.093 |
ophone | +0.089 |
lights | +0.087 |
endant | +0.084 |
dark | +0.083 |
Neg logit contributions | |
peel | -0.084 |
iver | -0.083 |
he | -0.081 |
pack | -0.080 |
Son | -0.080 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
M | +0.032 |
Dad | +0.032 |
Son | +0.031 |
Brother | +0.031 |
Be | +0.031 |
Neg logit contributions | |
Jupiter | -0.055 |
ophone | -0.054 |
belongs | -0.054 |
existed | -0.053 |
endant | -0.050 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Son | +0.135 |
distress | +0.134 |
No | +0.132 |
Dad | +0.131 |
Be | +0.127 |
Neg logit contributions | |
ophone | -0.142 |
Jupiter | -0.141 |
belongs | -0.135 |
unic | -0.130 |
dark | -0.128 |
Token | upon |
Token ID | 2402 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
Dad | +0.034 |
No | +0.032 |
Son | +0.032 |
distress | +0.032 |
he | +0.031 |
Neg logit contributions | |
ophone | -0.066 |
Jupiter | -0.065 |
unic | -0.062 |
belongs | -0.061 |
endant | -0.061 |
Token | on |
Token ID | 319 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
Jupiter | +0.343 |
belongs | +0.336 |
ophone | +0.335 |
lights | +0.319 |
dark | +0.315 |
Neg logit contributions | |
peel | -0.265 |
distress | -0.259 |
No | -0.247 |
pack | -0.245 |
Son | -0.245 |
Token | it |
Token ID | 340 |
Feature activation | -6.93e-01 |
Pos logit contributions | |
ophone | +0.081 |
Jupiter | +0.077 |
belongs | +0.076 |
endant | +0.073 |
dark | +0.071 |
Neg logit contributions | |
distress | -0.079 |
No | -0.079 |
Son | -0.078 |
Dad | -0.078 |
Be | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | +3.08e-03 |
Pos logit contributions | |
Jupiter | +0.039 |
ophone | +0.039 |
endant | +0.037 |
lights | +0.036 |
dark | +0.036 |
Neg logit contributions | |
peel | -0.029 |
distress | -0.027 |
No | -0.026 |
pack | -0.025 |
iver | -0.025 |
Token | Anna |
Token ID | 11735 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
Son | +0.000 |
Dad | +0.000 |
No | +0.000 |
distress | +0.000 |
Be | +0.000 |
Neg logit contributions | |
ophone | -0.000 |
Jupiter | -0.000 |
belongs | -0.000 |
unic | -0.000 |
endant | -0.000 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Son | +0.003 |
distress | +0.003 |
No | +0.003 |
Dad | +0.003 |
Be | +0.002 |
Neg logit contributions | |
ophone | -0.003 |
Jupiter | -0.003 |
belongs | -0.002 |
unic | -0.002 |
dark | -0.002 |
Token | unic |
Token ID | 28000 |
Feature activation | -9.66e+00 |
Pos logit contributions | |
ophone | +0.079 |
belongs | +0.075 |
dark | +0.070 |
Jupiter | +0.069 |
existed | +0.069 |
Neg logit contributions | |
distress | -0.054 |
peel | -0.052 |
No | -0.052 |
Dad | -0.051 |
Be | -0.051 |
Token | orns |
Token ID | 19942 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ophone | +0.408 |
Jupiter | +0.403 |
unic | +0.386 |
belongs | +0.385 |
dark | +0.374 |
Neg logit contributions | |
Son | -0.544 |
No | -0.539 |
distress | -0.536 |
Dad | -0.529 |
Be | -0.512 |
Token | . |
Token ID | 13 |
Feature activation | -4.57e-03 |
Pos logit contributions | |
ophone | +0.093 |
Jupiter | +0.092 |
belongs | +0.086 |
dark | +0.084 |
unic | +0.083 |
Neg logit contributions | |
distress | -0.090 |
Son | -0.088 |
No | -0.087 |
peel | -0.085 |
iver | -0.085 |
Token | They |
Token ID | 1119 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
ophone | +0.000 |
Jupiter | +0.000 |
belongs | +0.000 |
unic | +0.000 |
endant | +0.000 |
Neg logit contributions | |
Son | -0.000 |
Dad | -0.000 |
No | -0.000 |
distress | -0.000 |
Be | -0.000 |
Token | were |
Token ID | 547 |
Feature activation | +8.50e-01 |
Pos logit contributions | |
Son | +0.066 |
distress | +0.065 |
Dad | +0.064 |
No | +0.064 |
Be | +0.062 |
Neg logit contributions | |
ophone | -0.062 |
Jupiter | -0.062 |
belongs | -0.060 |
dark | -0.056 |
unic | -0.056 |
Token | pretty |
Token ID | 2495 |
Feature activation | -2.78e+00 |
Pos logit contributions | |
distress | +0.030 |
pack | +0.030 |
worry | +0.029 |
Dad | +0.028 |
Son | +0.028 |
Neg logit contributions | |
ophone | -0.052 |
belongs | -0.051 |
dark | -0.051 |
existed | -0.048 |
Jupiter | -0.047 |
Token | the |
Token ID | 262 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
ophone | +0.046 |
Jupiter | +0.044 |
belongs | +0.043 |
unic | +0.041 |
endant | +0.041 |
Neg logit contributions | |
distress | -0.039 |
Son | -0.039 |
Dad | -0.038 |
No | -0.038 |
Be | -0.037 |
Token | box |
Token ID | 3091 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
ophone | +0.030 |
lights | +0.028 |
Jupiter | +0.027 |
belongs | +0.026 |
dark | +0.025 |
Neg logit contributions | |
peel | -0.021 |
Be | -0.021 |
distress | -0.020 |
No | -0.020 |
Dad | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
Son | +0.068 |
distress | +0.067 |
No | +0.066 |
Dad | +0.066 |
Be | +0.063 |
Neg logit contributions | |
ophone | -0.062 |
Jupiter | -0.062 |
belongs | -0.059 |
unic | -0.057 |
dark | -0.056 |
Token | One |
Token ID | 1881 |
Feature activation | -2.95e-01 |
Pos logit contributions | |
ophone | +0.003 |
belongs | +0.003 |
Jupiter | +0.003 |
existed | +0.003 |
unic | +0.003 |
Neg logit contributions | |
Dad | -0.002 |
No | -0.002 |
Be | -0.002 |
Son | -0.002 |
distress | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ophone | +0.017 |
Jupiter | +0.016 |
unic | +0.016 |
endant | +0.015 |
existed | +0.015 |
Neg logit contributions | |
No | -0.012 |
distress | -0.012 |
Be | -0.012 |
peel | -0.012 |
Dad | -0.012 |
Token | ind |
Token ID | 773 |
Feature activation | -9.61e+00 |
Pos logit contributions | |
ophone | +0.066 |
belongs | +0.065 |
Jupiter | +0.064 |
lights | +0.061 |
dark | +0.061 |
Neg logit contributions | |
No | -0.044 |
Dad | -0.043 |
distress | -0.043 |
Be | -0.043 |
Son | -0.043 |
Token | igo |
Token ID | 14031 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
ophone | +0.463 |
Jupiter | +0.458 |
belongs | +0.439 |
unic | +0.421 |
dark | +0.415 |
Neg logit contributions | |
Son | -0.487 |
distress | -0.481 |
No | -0.474 |
Dad | -0.472 |
Be | -0.458 |
Token | and |
Token ID | 290 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Jupiter | +0.129 |
ophone | +0.125 |
belongs | +0.121 |
lights | +0.120 |
unic | +0.118 |
Neg logit contributions | |
M | -0.079 |
peel | -0.078 |
No | -0.076 |
Be | -0.074 |
Dad | -0.073 |
Token | one |
Token ID | 530 |
Feature activation | +5.91e-01 |
Pos logit contributions | |
Be | +0.007 |
No | +0.007 |
M | +0.007 |
Dad | +0.007 |
peel | +0.006 |
Neg logit contributions | |
ophone | -0.008 |
belongs | -0.007 |
Jupiter | -0.007 |
lights | -0.007 |
existed | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
No | +0.026 |
Be | +0.025 |
peel | +0.025 |
M | +0.025 |
distress | +0.024 |
Neg logit contributions | |
ophone | -0.032 |
Jupiter | -0.032 |
unic | -0.030 |
belongs | -0.029 |
lights | -0.029 |
Token | violet |
Token ID | 45725 |
Feature activation | -2.69e+00 |
Pos logit contributions | |
ophone | +0.063 |
belongs | +0.062 |
Jupiter | +0.058 |
dark | +0.057 |
lights | +0.057 |
Neg logit contributions | |
M | -0.047 |
No | -0.047 |
Be | -0.046 |
Dad | -0.045 |
distress | -0.045 |
Token | legs |
Token ID | 7405 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
distress | +0.006 |
Be | +0.006 |
No | +0.006 |
worry | +0.005 |
panic | +0.005 |
Neg logit contributions | |
ophone | -0.010 |
belongs | -0.009 |
dark | -0.009 |
existed | -0.008 |
Jupiter | -0.008 |
Token | . |
Token ID | 13 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
distress | +0.074 |
Son | +0.073 |
No | +0.070 |
Dad | +0.070 |
iver | +0.070 |
Neg logit contributions | |
Jupiter | -0.073 |
ophone | -0.072 |
belongs | -0.070 |
dark | -0.066 |
unic | -0.066 |
Token | They |
Token ID | 1119 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Dad | +0.002 |
Son | +0.002 |
No | +0.002 |
distress | +0.002 |
Be | +0.002 |
Neg logit contributions | |
ophone | -0.003 |
belongs | -0.003 |
unic | -0.003 |
endant | -0.003 |
existed | -0.003 |
Token | were |
Token ID | 547 |
Feature activation | +7.40e-01 |
Pos logit contributions | |
Son | +0.056 |
distress | +0.055 |
Dad | +0.054 |
No | +0.054 |
Be | +0.053 |
Neg logit contributions | |
ophone | -0.064 |
Jupiter | -0.063 |
belongs | -0.062 |
dark | -0.060 |
lights | -0.059 |
Token | called |
Token ID | 1444 |
Feature activation | -5.97e+00 |
Pos logit contributions | |
distress | +0.032 |
pack | +0.029 |
Son | +0.029 |
Be | +0.029 |
No | +0.028 |
Neg logit contributions | |
ophone | -0.042 |
belongs | -0.041 |
dark | -0.040 |
endant | -0.039 |
lights | -0.038 |
Token | ost |
Token ID | 23619 |
Feature activation | -9.55e+00 |
Pos logit contributions | |
ophone | +0.302 |
belongs | +0.285 |
dark | +0.282 |
lights | +0.275 |
endant | +0.273 |
Neg logit contributions | |
distress | -0.291 |
No | -0.278 |
Son | -0.269 |
pack | -0.268 |
Be | -0.268 |
Token | ric |
Token ID | 1173 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
belongs | +0.496 |
ophone | +0.494 |
Jupiter | +0.490 |
dark | +0.477 |
unic | +0.474 |
Neg logit contributions | |
distress | -0.420 |
Son | -0.415 |
Dad | -0.390 |
iver | -0.388 |
Col | -0.378 |
Token | hes |
Token ID | 956 |
Feature activation | -5.94e-02 |
Pos logit contributions | |
Jupiter | +0.023 |
belongs | +0.023 |
ophone | +0.022 |
unic | +0.022 |
endant | +0.022 |
Neg logit contributions | |
Son | -0.018 |
Dad | -0.017 |
distress | -0.017 |
No | -0.017 |
Be | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +9.47e-02 |
Pos logit contributions | |
Jupiter | +0.003 |
ophone | +0.003 |
belongs | +0.003 |
endant | +0.003 |
unic | +0.003 |
Neg logit contributions | |
Son | -0.003 |
distress | -0.003 |
No | -0.003 |
Dad | -0.003 |
Be | -0.003 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
Dad | +0.005 |
Son | +0.005 |
No | +0.004 |
Be | +0.004 |
distress | +0.004 |
Neg logit contributions | |
ophone | -0.005 |
belongs | -0.005 |
unic | -0.005 |
endant | -0.005 |
Jupiter | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
Jupiter | +0.086 |
ophone | +0.084 |
lights | +0.082 |
dark | +0.080 |
belongs | +0.080 |
Neg logit contributions | |
Son | -0.059 |
distress | -0.058 |
Dad | -0.057 |
No | -0.055 |
Be | -0.055 |
Token | said |
Token ID | 531 |
Feature activation | -4.69e-01 |
Pos logit contributions | |
No | +0.006 |
Son | +0.006 |
Dad | +0.006 |
peel | +0.006 |
Be | +0.006 |
Neg logit contributions | |
ophone | -0.008 |
Jupiter | -0.008 |
belongs | -0.007 |
lights | -0.007 |
dark | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
ophone | +0.031 |
dark | +0.029 |
belongs | +0.029 |
Jupiter | +0.028 |
lights | +0.028 |
Neg logit contributions | |
No | -0.016 |
M | -0.014 |
Dad | -0.014 |
worry | -0.014 |
Be | -0.013 |
Token | " |
Token ID | 366 |
Feature activation | +4.45e-02 |
Pos logit contributions | |
ophone | +0.012 |
belongs | +0.011 |
Jupiter | +0.011 |
dark | +0.011 |
endant | +0.011 |
Neg logit contributions | |
Dad | -0.012 |
No | -0.012 |
Be | -0.012 |
Son | -0.012 |
distress | -0.012 |
Token | This |
Token ID | 1212 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
Son | +0.001 |
distress | +0.001 |
No | +0.001 |
Dad | +0.001 |
Be | +0.001 |
Neg logit contributions | |
ophone | -0.003 |
Jupiter | -0.003 |
belongs | -0.003 |
unic | -0.003 |
dark | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
Be | +0.074 |
No | +0.074 |
peel | +0.072 |
M | +0.069 |
pack | +0.066 |
Neg logit contributions | |
lights | -0.079 |
ophone | -0.079 |
Jupiter | -0.075 |
endant | -0.069 |
dark | -0.069 |
Token | ind |
Token ID | 773 |
Feature activation | -9.53e+00 |
Pos logit contributions | |
ophone | +0.028 |
lights | +0.027 |
belongs | +0.025 |
dark | +0.025 |
Pole | +0.025 |
Neg logit contributions | |
M | -0.015 |
No | -0.014 |
Be | -0.014 |
half | -0.014 |
Dad | -0.013 |
Token | igo |
Token ID | 14031 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
ophone | +0.482 |
Jupiter | +0.476 |
belongs | +0.462 |
unic | +0.437 |
existed | +0.435 |
Neg logit contributions | |
distress | -0.446 |
Son | -0.444 |
No | -0.438 |
Dad | -0.437 |
iver | -0.431 |
Token | , |
Token ID | 11 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
Jupiter | +0.110 |
ophone | +0.106 |
belongs | +0.104 |
lights | +0.103 |
existed | +0.101 |
Neg logit contributions | |
M | -0.087 |
peel | -0.087 |
No | -0.087 |
Dad | -0.085 |
Be | -0.084 |
Token | and |
Token ID | 290 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ophone | +0.011 |
belongs | +0.010 |
existed | +0.010 |
lights | +0.010 |
Jupiter | +0.010 |
Neg logit contributions | |
Be | -0.013 |
No | -0.013 |
Dad | -0.012 |
M | -0.012 |
peel | -0.012 |
Token | this |
Token ID | 428 |
Feature activation | -7.21e-01 |
Pos logit contributions | |
Be | +0.008 |
No | +0.007 |
Dad | +0.007 |
M | +0.007 |
peel | +0.007 |
Neg logit contributions | |
ophone | -0.009 |
belongs | -0.008 |
lights | -0.008 |
Jupiter | -0.008 |
existed | -0.008 |
Token | is |
Token ID | 318 |
Feature activation | -4.40e-01 |
Pos logit contributions | |
ophone | +0.035 |
Jupiter | +0.035 |
lights | +0.034 |
unic | +0.031 |
endant | +0.031 |
Neg logit contributions | |
Be | -0.034 |
No | -0.034 |
peel | -0.032 |
M | -0.032 |
Dad | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
ophone | +0.098 |
Jupiter | +0.096 |
belongs | +0.087 |
existed | +0.086 |
unic | +0.086 |
Neg logit contributions | |
No | -0.090 |
peel | -0.086 |
Be | -0.086 |
distress | -0.085 |
Dad | -0.084 |
Token | green |
Token ID | 4077 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
ophone | +0.011 |
belongs | +0.010 |
Jupiter | +0.010 |
existed | +0.010 |
dark | +0.010 |
Neg logit contributions | |
No | -0.010 |
Be | -0.010 |
peel | -0.009 |
Dad | -0.009 |
Son | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -2.18e-01 |
Pos logit contributions | |
ophone | +0.073 |
Jupiter | +0.072 |
belongs | +0.066 |
existed | +0.064 |
unic | +0.064 |
Neg logit contributions | |
No | -0.068 |
peel | -0.066 |
Be | -0.065 |
distress | -0.064 |
Dad | -0.063 |
Token | blue |
Token ID | 4171 |
Feature activation | -2.61e+00 |
Pos logit contributions | |
ophone | +0.012 |
belongs | +0.011 |
Jupiter | +0.011 |
existed | +0.011 |
dark | +0.011 |
Neg logit contributions | |
No | -0.010 |
Be | -0.010 |
peel | -0.009 |
Dad | -0.009 |
tug | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -2.05e-01 |
Pos logit contributions | |
ophone | +0.114 |
Jupiter | +0.112 |
belongs | +0.107 |
unic | +0.102 |
dark | +0.100 |
Neg logit contributions | |
Son | -0.143 |
distress | -0.142 |
No | -0.141 |
Dad | -0.140 |
Be | -0.137 |
Token | ind |
Token ID | 773 |
Feature activation | -9.40e+00 |
Pos logit contributions | |
ophone | +0.012 |
belongs | +0.011 |
Jupiter | +0.011 |
existed | +0.011 |
dark | +0.010 |
Neg logit contributions | |
No | -0.009 |
Be | -0.009 |
peel | -0.008 |
Dad | -0.008 |
pack | -0.008 |
Token | igo |
Token ID | 14031 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
ophone | +0.497 |
Jupiter | +0.481 |
belongs | +0.478 |
existed | +0.454 |
dark | +0.449 |
Neg logit contributions | |
distress | -0.415 |
Dad | -0.408 |
No | -0.408 |
Son | -0.407 |
iver | -0.404 |
Token | , |
Token ID | 11 |
Feature activation | -2.20e-01 |
Pos logit contributions | |
Jupiter | +0.095 |
ophone | +0.092 |
belongs | +0.088 |
unic | +0.085 |
existed | +0.084 |
Neg logit contributions | |
No | -0.092 |
peel | -0.092 |
Dad | -0.090 |
Be | -0.089 |
Son | -0.089 |
Token | violet |
Token ID | 45725 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
ophone | +0.011 |
belongs | +0.010 |
Jupiter | +0.010 |
existed | +0.010 |
lights | +0.009 |
Neg logit contributions | |
Be | -0.011 |
No | -0.011 |
peel | -0.011 |
Dad | -0.011 |
M | -0.010 |
Token | ." |
Token ID | 526 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
Jupiter | +0.131 |
ophone | +0.129 |
unic | +0.118 |
belongs | +0.117 |
dark | +0.116 |
Neg logit contributions | |
peel | -0.114 |
No | -0.113 |
M | -0.109 |
iver | -0.105 |
Be | -0.104 |
Token |
|
Token ID | 198 |
Feature activation | -7.35e-02 |
Pos logit contributions | |
ophone | +0.100 |
Jupiter | +0.096 |
belongs | +0.094 |
unic | +0.091 |
existed | +0.089 |
Neg logit contributions | |
Dad | -0.064 |
No | -0.063 |
Son | -0.061 |
Be | -0.060 |
distress | -0.059 |
Token | , |
Token ID | 11 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
ophone | +0.102 |
Jupiter | +0.097 |
belongs | +0.094 |
existed | +0.089 |
dark | +0.088 |
Neg logit contributions | |
distress | -0.090 |
No | -0.089 |
Dad | -0.089 |
Be | -0.088 |
Son | -0.087 |
Token | green |
Token ID | 4077 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ophone | +0.013 |
belongs | +0.012 |
existed | +0.012 |
Jupiter | +0.011 |
endant | +0.011 |
Neg logit contributions | |
Be | -0.010 |
No | -0.009 |
Dad | -0.009 |
peel | -0.009 |
M | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | -2.16e-01 |
Pos logit contributions | |
ophone | +0.078 |
Jupiter | +0.073 |
belongs | +0.071 |
existed | +0.067 |
dark | +0.067 |
Neg logit contributions | |
distress | -0.068 |
No | -0.067 |
Be | -0.067 |
Dad | -0.066 |
peel | -0.065 |
Token | blue |
Token ID | 4171 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ophone | +0.013 |
belongs | +0.012 |
existed | +0.011 |
Jupiter | +0.011 |
endant | +0.011 |
Neg logit contributions | |
Be | -0.009 |
No | -0.009 |
peel | -0.009 |
Dad | -0.009 |
M | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -2.22e-01 |
Pos logit contributions | |
ophone | +0.122 |
Jupiter | +0.120 |
belongs | +0.115 |
unic | +0.109 |
dark | +0.108 |
Neg logit contributions | |
Son | -0.138 |
distress | -0.137 |
No | -0.135 |
Dad | -0.135 |
Be | -0.132 |
Token | ind |
Token ID | 773 |
Feature activation | -9.39e+00 |
Pos logit contributions | |
ophone | +0.013 |
belongs | +0.012 |
Jupiter | +0.012 |
existed | +0.012 |
endant | +0.011 |
Neg logit contributions | |
Be | -0.009 |
No | -0.009 |
peel | -0.009 |
Dad | -0.009 |
M | -0.009 |
Token | igo |
Token ID | 14031 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
ophone | +0.482 |
Jupiter | +0.472 |
belongs | +0.461 |
unic | +0.436 |
dark | +0.434 |
Neg logit contributions | |
Son | -0.441 |
distress | -0.440 |
Dad | -0.431 |
No | -0.429 |
iver | -0.418 |
Token | and |
Token ID | 290 |
Feature activation | +2.27e-01 |
Pos logit contributions | |
ophone | +0.110 |
Jupiter | +0.109 |
belongs | +0.105 |
unic | +0.100 |
existed | +0.099 |
Neg logit contributions | |
No | -0.082 |
Dad | -0.082 |
Son | -0.082 |
distress | -0.081 |
peel | -0.081 |
Token | violet |
Token ID | 45725 |
Feature activation | -2.62e+00 |
Pos logit contributions | |
Be | +0.011 |
No | +0.010 |
Dad | +0.010 |
M | +0.010 |
peel | +0.010 |
Neg logit contributions | |
ophone | -0.013 |
belongs | -0.012 |
lights | -0.011 |
Jupiter | -0.011 |
existed | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +1.78e-01 |
Pos logit contributions | |
ophone | +0.153 |
Jupiter | +0.149 |
belongs | +0.138 |
dark | +0.134 |
unic | +0.134 |
Neg logit contributions | |
peel | -0.093 |
M | -0.091 |
No | -0.088 |
ouch | -0.084 |
distress | -0.084 |
Token | It |
Token ID | 632 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Dad | +0.007 |
Be | +0.007 |
No | +0.007 |
Ben | +0.007 |
M | +0.007 |
Neg logit contributions | |
ophone | -0.011 |
belongs | -0.011 |
existed | -0.010 |
unic | -0.010 |
Jupiter | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +6.47e-02 |
Pos logit contributions | |
ophone | +0.028 |
Jupiter | +0.027 |
belongs | +0.026 |
dark | +0.026 |
endant | +0.025 |
Neg logit contributions | |
distress | -0.017 |
No | -0.017 |
Son | -0.017 |
Dad | -0.016 |
peel | -0.015 |
Token | Do |
Token ID | 2141 |
Feature activation | -3.06e+00 |
Pos logit contributions | |
No | +0.003 |
Be | +0.003 |
Dad | +0.003 |
M | +0.003 |
Son | +0.003 |
Neg logit contributions | |
lights | -0.003 |
ophone | -0.003 |
endant | -0.003 |
ls | -0.003 |
brand | -0.003 |
Token | you |
Token ID | 345 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
ophone | +0.123 |
Jupiter | +0.120 |
belongs | +0.114 |
dark | +0.108 |
unic | +0.107 |
Neg logit contributions | |
No | -0.180 |
Son | -0.180 |
Dad | -0.178 |
distress | -0.177 |
Be | -0.174 |
Token | know |
Token ID | 760 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
ophone | +0.097 |
Jupiter | +0.097 |
belongs | +0.093 |
unic | +0.090 |
dark | +0.089 |
Neg logit contributions | |
Son | -0.060 |
distress | -0.060 |
No | -0.058 |
Dad | -0.058 |
Be | -0.056 |
Token | what |
Token ID | 644 |
Feature activation | -2.03e-01 |
Pos logit contributions | |
No | +0.100 |
Dad | +0.094 |
half | +0.094 |
What | +0.093 |
peel | +0.093 |
Neg logit contributions | |
ophone | -0.200 |
icol | -0.194 |
belongs | -0.192 |
lights | -0.190 |
dark | -0.189 |
Token | ind |
Token ID | 773 |
Feature activation | -9.37e+00 |
Pos logit contributions | |
icol | +0.012 |
lights | +0.012 |
ophone | +0.011 |
endant | +0.011 |
Jupiter | +0.011 |
Neg logit contributions | |
No | -0.009 |
Be | -0.008 |
Dad | -0.008 |
do | -0.008 |
M | -0.007 |
Token | igo |
Token ID | 14031 |
Feature activation | -1.94e+00 |
Pos logit contributions | |
ophone | +0.516 |
Jupiter | +0.501 |
belongs | +0.495 |
existed | +0.477 |
dark | +0.468 |
Neg logit contributions | |
iver | -0.378 |
distress | -0.375 |
No | -0.368 |
Dad | -0.365 |
Son | -0.362 |
Token | and |
Token ID | 290 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
Jupiter | +0.092 |
ophone | +0.087 |
unic | +0.084 |
lights | +0.084 |
endant | +0.082 |
Neg logit contributions | |
No | -0.094 |
peel | -0.092 |
Be | -0.091 |
Dad | -0.091 |
Son | -0.090 |
Token | violet |
Token ID | 45725 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
Be | +0.012 |
No | +0.012 |
Dad | +0.011 |
M | +0.011 |
peel | +0.010 |
Neg logit contributions | |
ophone | -0.013 |
lights | -0.013 |
belongs | -0.013 |
Jupiter | -0.012 |
existed | -0.012 |
Token | are |
Token ID | 389 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
Jupiter | +0.128 |
ophone | +0.123 |
unic | +0.120 |
icol | +0.118 |
Zig | +0.115 |
Neg logit contributions | |
No | -0.115 |
peel | -0.108 |
iver | -0.108 |
Be | -0.108 |
Dad | -0.107 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
Dad | +0.093 |
M | +0.089 |
No | +0.089 |
Be | +0.088 |
peel | +0.088 |
Neg logit contributions | |
ophone | -0.120 |
lights | -0.115 |
belongs | -0.113 |
Zig | -0.110 |
Jupiter | -0.110 |
Token | fence |
Token ID | 13990 |
Feature activation | -3.05e-01 |
Pos logit contributions | |
ophone | +0.028 |
Jupiter | +0.027 |
belongs | +0.026 |
lights | +0.026 |
existed | +0.026 |
Neg logit contributions | |
distress | -0.023 |
Son | -0.023 |
Dad | -0.023 |
peel | -0.023 |
No | -0.023 |
Token | and |
Token ID | 290 |
Feature activation | +7.06e-02 |
Pos logit contributions | |
ophone | +0.019 |
Jupiter | +0.019 |
unic | +0.018 |
belongs | +0.018 |
endant | +0.018 |
Neg logit contributions | |
distress | -0.011 |
No | -0.011 |
peel | -0.011 |
Son | -0.010 |
Dad | -0.010 |
Token | run |
Token ID | 1057 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
distress | +0.002 |
peel | +0.002 |
No | +0.002 |
Son | +0.002 |
Be | +0.002 |
Neg logit contributions | |
ophone | -0.005 |
Jupiter | -0.005 |
belongs | -0.005 |
unic | -0.005 |
dark | -0.005 |
Token | to |
Token ID | 284 |
Feature activation | +5.23e-01 |
Pos logit contributions | |
No | +0.072 |
peel | +0.067 |
tug | +0.065 |
Be | +0.065 |
Dad | +0.065 |
Neg logit contributions | |
ophone | -0.096 |
Jupiter | -0.092 |
endant | -0.092 |
unic | -0.091 |
belongs | -0.089 |
Token | the |
Token ID | 262 |
Feature activation | -5.53e-01 |
Pos logit contributions | |
Son | +0.028 |
distress | +0.027 |
Dad | +0.027 |
No | +0.026 |
iver | +0.026 |
Neg logit contributions | |
Jupiter | -0.025 |
ophone | -0.024 |
belongs | -0.023 |
unic | -0.022 |
Venus | -0.021 |
Token | ost |
Token ID | 23619 |
Feature activation | -9.36e+00 |
Pos logit contributions | |
ophone | +0.030 |
Jupiter | +0.029 |
belongs | +0.028 |
lights | +0.027 |
dark | +0.027 |
Neg logit contributions | |
distress | -0.025 |
Dad | -0.025 |
Son | -0.025 |
No | -0.025 |
Be | -0.024 |
Token | ric |
Token ID | 1173 |
Feature activation | -5.99e-02 |
Pos logit contributions | |
ophone | +0.488 |
Jupiter | +0.488 |
belongs | +0.480 |
unic | +0.467 |
dark | +0.456 |
Neg logit contributions | |
Son | -0.413 |
distress | -0.412 |
Dad | -0.398 |
iver | -0.398 |
No | -0.383 |
Token | hes |
Token ID | 956 |
Feature activation | +4.26e-02 |
Pos logit contributions | |
Jupiter | +0.003 |
ophone | +0.003 |
unic | +0.003 |
belongs | +0.003 |
endant | +0.003 |
Neg logit contributions | |
Son | -0.002 |
Dad | -0.002 |
No | -0.002 |
iver | -0.002 |
distress | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | -9.38e-02 |
Pos logit contributions | |
No | +0.002 |
Son | +0.002 |
Dad | +0.002 |
distress | +0.002 |
peel | +0.002 |
Neg logit contributions | |
ophone | -0.002 |
Jupiter | -0.002 |
belongs | -0.002 |
endant | -0.002 |
unic | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
ophone | +0.006 |
unic | +0.005 |
belongs | +0.005 |
Jupiter | +0.005 |
endant | +0.005 |
Neg logit contributions | |
No | -0.004 |
Dad | -0.004 |
Be | -0.004 |
Son | -0.004 |
distress | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
ophone | +0.007 |
Jupiter | +0.007 |
belongs | +0.007 |
unic | +0.007 |
endant | +0.007 |
Neg logit contributions | |
Son | -0.006 |
Dad | -0.006 |
distress | -0.006 |
No | -0.006 |
Be | -0.006 |
Token | zz |
Token ID | 3019 |
Feature activation | +6.50e-01 |
Pos logit contributions | |
ophone | +0.059 |
Jupiter | +0.058 |
belongs | +0.056 |
unic | +0.052 |
dark | +0.052 |
Neg logit contributions | |
distress | -0.085 |
Son | -0.084 |
No | -0.084 |
Dad | -0.083 |
Be | -0.081 |
Token | ie |
Token ID | 494 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
Son | +0.028 |
distress | +0.028 |
iver | +0.027 |
ouch | +0.027 |
No | +0.027 |
Neg logit contributions | |
Jupiter | -0.035 |
belongs | -0.034 |
ophone | -0.033 |
dark | -0.032 |
existed | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | +3.61e-01 |
Pos logit contributions | |
ophone | +0.008 |
Jupiter | +0.008 |
belongs | +0.008 |
lights | +0.008 |
dark | +0.008 |
Neg logit contributions | |
Son | -0.008 |
distress | -0.008 |
Dad | -0.008 |
M | -0.007 |
peel | -0.007 |
Token | Unlike |
Token ID | 12101 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
Son | +0.019 |
No | +0.019 |
Dad | +0.019 |
distress | +0.019 |
Be | +0.018 |
Neg logit contributions | |
ophone | -0.017 |
Jupiter | -0.016 |
belongs | -0.016 |
unic | -0.015 |
lights | -0.015 |
Token | other |
Token ID | 584 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Son | +0.004 |
No | +0.004 |
Be | +0.004 |
pack | +0.004 |
peel | +0.004 |
Neg logit contributions | |
ophone | -0.005 |
lights | -0.005 |
endant | -0.005 |
dark | -0.005 |
belongs | -0.005 |
Token | ost |
Token ID | 23619 |
Feature activation | -9.34e+00 |
Pos logit contributions | |
Jupiter | +0.058 |
ophone | +0.057 |
belongs | +0.056 |
unic | +0.053 |
dark | +0.051 |
Neg logit contributions | |
Son | -0.069 |
distress | -0.067 |
Dad | -0.066 |
No | -0.066 |
iver | -0.065 |
Token | ric |
Token ID | 1173 |
Feature activation | -2.41e-01 |
Pos logit contributions | |
ophone | +0.485 |
Jupiter | +0.483 |
belongs | +0.477 |
dark | +0.453 |
unic | +0.451 |
Neg logit contributions | |
Son | -0.421 |
distress | -0.420 |
Dad | -0.404 |
iver | -0.402 |
No | -0.387 |
Token | hes |
Token ID | 956 |
Feature activation | +3.23e-03 |
Pos logit contributions | |
Jupiter | +0.013 |
belongs | +0.013 |
unic | +0.013 |
ophone | +0.013 |
lights | +0.013 |
Neg logit contributions | |
Son | -0.010 |
Dad | -0.009 |
distress | -0.009 |
iver | -0.009 |
M | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +9.25e-02 |
Pos logit contributions | |
Son | +0.000 |
No | +0.000 |
distress | +0.000 |
Dad | +0.000 |
peel | +0.000 |
Neg logit contributions | |
Jupiter | -0.000 |
ophone | -0.000 |
endant | -0.000 |
unic | -0.000 |
dark | -0.000 |
Token | his |
Token ID | 465 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
No | +0.005 |
Son | +0.005 |
Be | +0.005 |
Dad | +0.005 |
distress | +0.004 |
Neg logit contributions | |
ophone | -0.005 |
belongs | -0.004 |
endant | -0.004 |
dark | -0.004 |
Jupiter | -0.004 |
Token | feathers |
Token ID | 27400 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
distress | +0.104 |
No | +0.097 |
Son | +0.096 |
Dad | +0.095 |
worry | +0.092 |
Neg logit contributions | |
ophone | -0.133 |
Jupiter | -0.127 |
dark | -0.123 |
lights | -0.123 |
belongs | -0.121 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
No | +0.009 |
Be | +0.009 |
Dad | +0.009 |
distress | +0.009 |
Son | +0.008 |
Neg logit contributions | |
ophone | -0.006 |
Jupiter | -0.005 |
belongs | -0.005 |
endant | -0.005 |
existed | -0.005 |
Token | cried |
Token ID | 16896 |
Feature activation | -5.39e-01 |
Pos logit contributions | |
ophone | +0.041 |
belongs | +0.039 |
they | +0.038 |
belonged | +0.037 |
could | +0.036 |
Neg logit contributions | |
Son | -0.101 |
Col | -0.101 |
aleb | -0.099 |
Dad | -0.099 |
distress | -0.098 |
Token | . |
Token ID | 13 |
Feature activation | -1.26e-01 |
Pos logit contributions | |
Jupiter | +0.014 |
ophone | +0.013 |
lights | +0.012 |
dark | +0.011 |
unic | +0.011 |
Neg logit contributions | |
Son | -0.039 |
Marcus | -0.038 |
distress | -0.038 |
No | -0.038 |
Col | -0.037 |
Token | " |
Token ID | 366 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
Jupiter | +0.005 |
ophone | +0.005 |
belongs | +0.004 |
lights | +0.004 |
unic | +0.004 |
Neg logit contributions | |
Son | -0.008 |
distress | -0.008 |
iver | -0.007 |
Dad | -0.007 |
No | -0.007 |
Token | The |
Token ID | 464 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
customer | +0.007 |
ippy | +0.007 |
aleb | +0.007 |
distress | +0.007 |
id | +0.007 |
Neg logit contributions | |
dark | -0.006 |
they | -0.006 |
ophone | -0.006 |
we | -0.006 |
belonged | -0.006 |
Token | ost |
Token ID | 23619 |
Feature activation | -9.34e+00 |
Pos logit contributions | |
ophone | +0.038 |
Jupiter | +0.038 |
belongs | +0.037 |
unic | +0.036 |
endant | +0.035 |
Neg logit contributions | |
Son | -0.031 |
distress | -0.030 |
Dad | -0.029 |
No | -0.029 |
iver | -0.029 |
Token | ric |
Token ID | 1173 |
Feature activation | -9.83e-02 |
Pos logit contributions | |
ophone | +0.443 |
Jupiter | +0.436 |
belongs | +0.419 |
unic | +0.399 |
dark | +0.393 |
Neg logit contributions | |
Son | -0.485 |
distress | -0.479 |
No | -0.475 |
Dad | -0.473 |
Be | -0.459 |
Token | hes |
Token ID | 956 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ophone | +0.005 |
Jupiter | +0.005 |
belongs | +0.004 |
dark | +0.004 |
unic | +0.004 |
Neg logit contributions | |
Son | -0.005 |
distress | -0.005 |
No | -0.005 |
Dad | -0.005 |
Be | -0.005 |
Token | are |
Token ID | 389 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
Son | +0.007 |
distress | +0.007 |
No | +0.007 |
Dad | +0.007 |
Be | +0.007 |
Neg logit contributions | |
ophone | -0.007 |
Jupiter | -0.007 |
belongs | -0.007 |
unic | -0.007 |
dark | -0.006 |
Token | mean |
Token ID | 1612 |
Feature activation | -5.81e-01 |
Pos logit contributions | |
distress | +0.090 |
No | +0.089 |
Son | +0.086 |
peel | +0.085 |
Be | +0.085 |
Neg logit contributions | |
ophone | -0.114 |
Jupiter | -0.113 |
belongs | -0.106 |
dark | -0.104 |
lights | -0.103 |
Token | ! |
Token ID | 0 |
Feature activation | +9.66e-01 |
Pos logit contributions | |
ophone | +0.023 |
Jupiter | +0.022 |
belongs | +0.022 |
unic | +0.020 |
dark | +0.020 |
Neg logit contributions | |
Son | -0.035 |
distress | -0.034 |
Dad | -0.034 |
No | -0.034 |
Be | -0.033 |
Token | normal |
Token ID | 3487 |
Feature activation | -4.20e-01 |
Pos logit contributions | |
ophone | +0.043 |
Jupiter | +0.042 |
belongs | +0.040 |
dark | +0.040 |
endant | +0.039 |
Neg logit contributions | |
distress | -0.035 |
No | -0.032 |
Be | -0.032 |
Son | -0.032 |
Dad | -0.032 |
Token | for |
Token ID | 329 |
Feature activation | +4.31e-01 |
Pos logit contributions | |
ophone | +0.024 |
Jupiter | +0.023 |
endant | +0.022 |
dark | +0.022 |
belongs | +0.021 |
Neg logit contributions | |
distress | -0.018 |
No | -0.017 |
peel | -0.016 |
pack | -0.016 |
worry | -0.016 |
Token | old |
Token ID | 1468 |
Feature activation | +1.16e-02 |
Pos logit contributions | |
distress | +0.019 |
Son | +0.018 |
Dad | +0.018 |
No | +0.018 |
peel | +0.017 |
Neg logit contributions | |
ophone | -0.025 |
Jupiter | -0.023 |
belongs | -0.023 |
endant | -0.022 |
dark | -0.022 |
Token | k |
Token ID | 479 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
distress | +0.001 |
No | +0.000 |
Son | +0.000 |
worry | +0.000 |
peel | +0.000 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
endant | -0.001 |
dark | -0.001 |
belongs | -0.001 |
Token | ang |
Token ID | 648 |
Feature activation | -2.03e+00 |
Pos logit contributions | |
ophone | +0.107 |
Jupiter | +0.105 |
belongs | +0.101 |
unic | +0.097 |
dark | +0.096 |
Neg logit contributions | |
Son | -0.106 |
distress | -0.105 |
Dad | -0.103 |
No | -0.103 |
Be | -0.099 |
Token | aro |
Token ID | 12022 |
Feature activation | -9.33e+00 |
Pos logit contributions | |
Jupiter | +0.105 |
unic | +0.100 |
endant | +0.098 |
ophone | +0.097 |
belongs | +0.095 |
Neg logit contributions | |
No | -0.097 |
Son | -0.096 |
distress | -0.094 |
Dad | -0.091 |
peel | -0.087 |
Token | os |
Token ID | 418 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
Jupiter | +0.472 |
unic | +0.439 |
belongs | +0.437 |
endant | +0.420 |
ophone | +0.417 |
Neg logit contributions | |
Son | -0.443 |
iver | -0.431 |
distress | -0.423 |
Dad | -0.421 |
No | -0.419 |
Token | ." |
Token ID | 526 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
Jupiter | +0.099 |
ophone | +0.095 |
endant | +0.094 |
unic | +0.091 |
belongs | +0.089 |
Neg logit contributions | |
Son | -0.088 |
No | -0.087 |
distress | -0.087 |
iver | -0.086 |
Be | -0.083 |
Token |
|
Token ID | 198 |
Feature activation | -4.67e-02 |
Pos logit contributions | |
ophone | +0.074 |
Jupiter | +0.072 |
belongs | +0.071 |
endant | +0.069 |
unic | +0.068 |
Neg logit contributions | |
Son | -0.060 |
Dad | -0.060 |
No | -0.059 |
distress | -0.056 |
Brother | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | -7.94e-02 |
Pos logit contributions | |
Jupiter | +0.002 |
ophone | +0.002 |
belongs | +0.002 |
unic | +0.002 |
dark | +0.002 |
Neg logit contributions | |
distress | -0.002 |
Son | -0.002 |
No | -0.002 |
Dad | -0.002 |
iver | -0.002 |
Token | " |
Token ID | 1 |
Feature activation | +2.52e+00 |
Pos logit contributions | |
Jupiter | +0.005 |
ophone | +0.005 |
belongs | +0.005 |
dark | +0.005 |
unic | +0.004 |
Neg logit contributions | |
distress | -0.003 |
iver | -0.002 |
Son | -0.002 |
peel | -0.002 |
reel | -0.002 |
Token | everyone |
Token ID | 2506 |
Feature activation | -9.94e-01 |
Pos logit contributions | |
ophone | +0.126 |
endant | +0.124 |
Jupiter | +0.121 |
belongs | +0.120 |
dark | +0.119 |
Neg logit contributions | |
No | -0.061 |
distress | -0.057 |
Son | -0.056 |
Dad | -0.055 |
peel | -0.052 |
Token | . |
Token ID | 13 |
Feature activation | +6.77e-02 |
Pos logit contributions | |
Jupiter | +0.055 |
ophone | +0.054 |
endant | +0.050 |
belongs | +0.049 |
dark | +0.049 |
Neg logit contributions | |
distress | -0.042 |
No | -0.042 |
Son | -0.042 |
pack | -0.040 |
Be | -0.039 |
Token |
|
Token ID | 198 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
Son | +0.003 |
No | +0.003 |
Dad | +0.003 |
distress | +0.003 |
Be | +0.003 |
Neg logit contributions | |
ophone | -0.004 |
Jupiter | -0.004 |
belongs | -0.004 |
unic | -0.004 |
endant | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | -6.19e-02 |
Pos logit contributions | |
Jupiter | +0.003 |
ophone | +0.003 |
belongs | +0.003 |
dark | +0.003 |
unic | +0.003 |
Neg logit contributions | |
distress | -0.003 |
Son | -0.003 |
No | -0.003 |
iver | -0.003 |
Dad | -0.003 |
Token | M |
Token ID | 44 |
Feature activation | +4.06e+00 |
Pos logit contributions | |
dark | +0.003 |
ophone | +0.003 |
fant | +0.003 |
they | +0.003 |
Jupiter | +0.003 |
Neg logit contributions | |
distress | -0.003 |
scrape | -0.002 |
sausage | -0.002 |
shave | -0.002 |
pack | -0.002 |
Token | olly |
Token ID | 5098 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
Son | +0.195 |
distress | +0.190 |
Be | +0.186 |
No | +0.185 |
Dad | +0.183 |
Neg logit contributions | |
ophone | -0.203 |
belongs | -0.202 |
Jupiter | -0.200 |
endant | -0.192 |
dark | -0.191 |
Token | ran |
Token ID | 4966 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
Son | +0.200 |
distress | +0.199 |
Dad | +0.194 |
No | +0.193 |
Be | +0.185 |
Neg logit contributions | |
Jupiter | -0.229 |
ophone | -0.228 |
belongs | -0.215 |
endant | -0.211 |
lights | -0.211 |
Token | inside |
Token ID | 2641 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
ophone | +0.032 |
Jupiter | +0.031 |
belongs | +0.031 |
unic | +0.030 |
endant | +0.029 |
Neg logit contributions | |
No | -0.036 |
distress | -0.035 |
Son | -0.035 |
Dad | -0.035 |
Be | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | +5.87e-01 |
Pos logit contributions | |
Jupiter | +0.089 |
ophone | +0.089 |
belongs | +0.085 |
unic | +0.082 |
dark | +0.080 |
Neg logit contributions | |
Son | -0.088 |
distress | -0.087 |
Dad | -0.085 |
No | -0.085 |
iver | -0.083 |
Token | show |
Token ID | 905 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
Son | +0.022 |
distress | +0.022 |
No | +0.021 |
Dad | +0.021 |
Be | +0.020 |
Neg logit contributions | |
ophone | -0.037 |
Jupiter | -0.036 |
belongs | -0.035 |
unic | -0.034 |
dark | -0.034 |
Token | her |
Token ID | 607 |
Feature activation | -8.14e-01 |
Pos logit contributions | |
ophone | +0.128 |
endant | +0.124 |
belongs | +0.124 |
Jupiter | +0.123 |
dark | +0.121 |
Neg logit contributions | |
distress | -0.054 |
No | -0.052 |
Son | -0.051 |
Dad | -0.049 |
peel | -0.045 |
Token | B |
Token ID | 33 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
distress | +0.004 |
peel | +0.003 |
Son | +0.003 |
Col | +0.003 |
iver | +0.003 |
Neg logit contributions | |
Jupiter | -0.005 |
ophone | -0.005 |
belonged | -0.005 |
belongs | -0.004 |
unic | -0.004 |
Token | ongo |
Token ID | 25162 |
Feature activation | +1.12e-02 |
Pos logit contributions | |
distress | +0.091 |
Son | +0.088 |
No | +0.087 |
Dad | +0.086 |
Be | +0.084 |
Neg logit contributions | |
ophone | -0.057 |
endant | -0.053 |
belongs | -0.053 |
Jupiter | -0.051 |
dark | -0.049 |
Token | had |
Token ID | 550 |
Feature activation | -8.23e-01 |
Pos logit contributions | |
Son | +0.001 |
distress | +0.001 |
No | +0.001 |
Dad | +0.001 |
Be | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
belongs | -0.001 |
unic | -0.000 |
dark | -0.000 |
Token | a |
Token ID | 257 |
Feature activation | -9.46e-03 |
Pos logit contributions | |
ophone | +0.039 |
Jupiter | +0.038 |
belongs | +0.036 |
unic | +0.035 |
dark | +0.035 |
Neg logit contributions | |
distress | -0.043 |
Son | -0.043 |
No | -0.043 |
Dad | -0.042 |
Be | -0.042 |
Token | brilliant |
Token ID | 10457 |
Feature activation | -7.54e-01 |
Pos logit contributions | |
Jupiter | +0.000 |
ophone | +0.000 |
belongs | +0.000 |
unic | +0.000 |
belonged | +0.000 |
Neg logit contributions | |
Son | -0.000 |
Dad | -0.000 |
iver | -0.000 |
distress | -0.000 |
No | -0.000 |
Token | idea |
Token ID | 2126 |
Feature activation | +5.04e+00 |
Pos logit contributions | |
Jupiter | +0.042 |
ophone | +0.041 |
belongs | +0.041 |
unic | +0.038 |
endant | +0.038 |
Neg logit contributions | |
Son | -0.034 |
Dad | -0.032 |
distress | -0.031 |
Brother | -0.031 |
Marcus | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
distress | +0.246 |
No | +0.243 |
Son | +0.239 |
Dad | +0.233 |
peel | +0.233 |
Neg logit contributions | |
Jupiter | -0.252 |
ophone | -0.251 |
unic | -0.234 |
dark | -0.232 |
endant | -0.232 |
Token | He |
Token ID | 679 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
Son | +0.010 |
distress | +0.010 |
No | +0.010 |
Dad | +0.010 |
Be | +0.009 |
Neg logit contributions | |
ophone | -0.015 |
Jupiter | -0.014 |
belongs | -0.014 |
unic | -0.013 |
dark | -0.013 |
Token | found |
Token ID | 1043 |
Feature activation | -8.26e-01 |
Pos logit contributions | |
Be | +0.094 |
Son | +0.093 |
peel | +0.092 |
tug | +0.091 |
distress | +0.090 |
Neg logit contributions | |
lights | -0.164 |
Jupiter | -0.159 |
ophone | -0.156 |
existed | -0.149 |
dark | -0.147 |
Token | a |
Token ID | 257 |
Feature activation | -2.84e-02 |
Pos logit contributions | |
ophone | +0.053 |
belongs | +0.051 |
Jupiter | +0.050 |
lights | +0.049 |
existed | +0.049 |
Neg logit contributions | |
Son | -0.030 |
distress | -0.030 |
No | -0.029 |
Be | -0.029 |
Dad | -0.027 |
Token | big |
Token ID | 1263 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
Jupiter | +0.001 |
ophone | +0.001 |
belongs | +0.001 |
unic | +0.001 |
endant | +0.001 |
Neg logit contributions | |
Son | -0.001 |
distress | -0.001 |
Dad | -0.001 |
No | -0.001 |
iver | -0.001 |
Token | remembered |
Token ID | 12086 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
Jupiter | +0.022 |
ophone | +0.022 |
belongs | +0.021 |
lights | +0.021 |
endant | +0.020 |
Neg logit contributions | |
Son | -0.018 |
distress | -0.018 |
No | -0.018 |
Be | -0.017 |
Dad | -0.017 |
Token | the |
Token ID | 262 |
Feature activation | -4.70e-01 |
Pos logit contributions | |
distress | +0.009 |
Son | +0.009 |
No | +0.009 |
Dad | +0.009 |
Be | +0.009 |
Neg logit contributions | |
ophone | -0.011 |
belongs | -0.010 |
Jupiter | -0.010 |
endant | -0.010 |
dark | -0.010 |
Token | exit |
Token ID | 8420 |
Feature activation | -9.90e-01 |
Pos logit contributions | |
ophone | +0.024 |
Jupiter | +0.024 |
belongs | +0.023 |
unic | +0.022 |
dark | +0.022 |
Neg logit contributions | |
distress | -0.023 |
Son | -0.022 |
No | -0.022 |
Dad | -0.022 |
Be | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +3.10e-01 |
Pos logit contributions | |
Jupiter | +0.055 |
ophone | +0.054 |
endant | +0.053 |
unic | +0.052 |
dark | +0.052 |
Neg logit contributions | |
distress | -0.040 |
No | -0.040 |
Son | -0.039 |
peel | -0.038 |
pack | -0.037 |
Token | made |
Token ID | 925 |
Feature activation | +7.28e-01 |
Pos logit contributions | |
distress | +0.013 |
Be | +0.013 |
Son | +0.013 |
No | +0.013 |
worry | +0.013 |
Neg logit contributions | |
ophone | -0.017 |
endant | -0.017 |
Jupiter | -0.016 |
lights | -0.016 |
dark | -0.016 |
Token | sure |
Token ID | 1654 |
Feature activation | +4.37e+00 |
Pos logit contributions | |
Son | +0.028 |
distress | +0.028 |
No | +0.027 |
Dad | +0.027 |
Be | +0.026 |
Neg logit contributions | |
ophone | -0.044 |
Jupiter | -0.043 |
belongs | -0.043 |
endant | -0.042 |
existed | -0.041 |
Token | to |
Token ID | 284 |
Feature activation | +5.84e-01 |
Pos logit contributions | |
No | +0.206 |
Son | +0.203 |
distress | +0.200 |
Dad | +0.190 |
Be | +0.188 |
Neg logit contributions | |
ophone | -0.238 |
Jupiter | -0.227 |
dark | -0.221 |
endant | -0.218 |
belongs | -0.214 |
Token | tell |
Token ID | 1560 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
Son | +0.024 |
distress | +0.024 |
No | +0.023 |
Dad | +0.023 |
Be | +0.023 |
Neg logit contributions | |
ophone | -0.033 |
Jupiter | -0.033 |
belongs | -0.032 |
unic | -0.031 |
dark | -0.030 |
Token | a |
Token ID | 257 |
Feature activation | -3.23e-02 |
Pos logit contributions | |
No | +0.012 |
Son | +0.012 |
distress | +0.012 |
Dad | +0.011 |
pack | +0.011 |
Neg logit contributions | |
ophone | -0.021 |
endant | -0.020 |
belongs | -0.020 |
dark | -0.019 |
Jupiter | -0.019 |
Token | joke |
Token ID | 9707 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
ophone | +0.001 |
Jupiter | +0.001 |
belongs | +0.001 |
endant | +0.001 |
dark | +0.001 |
Neg logit contributions | |
No | -0.002 |
Son | -0.002 |
distress | -0.002 |
Be | -0.002 |
Dad | -0.002 |
Token | before |
Token ID | 878 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
No | +0.107 |
Son | +0.103 |
peel | +0.099 |
Dad | +0.099 |
Be | +0.099 |
Neg logit contributions | |
Jupiter | -0.109 |
endant | -0.107 |
ophone | -0.103 |
belongs | -0.102 |
unic | -0.099 |
Token | neck |
Token ID | 7393 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
ophone | +0.037 |
Jupiter | +0.035 |
dark | +0.035 |
belongs | +0.035 |
happens | +0.034 |
Neg logit contributions | |
distress | -0.021 |
peel | -0.019 |
pack | -0.019 |
iver | -0.019 |
Dad | -0.018 |
Token | . |
Token ID | 13 |
Feature activation | -7.77e-03 |
Pos logit contributions | |
Jupiter | +0.008 |
ophone | +0.008 |
belongs | +0.008 |
dark | +0.008 |
unic | +0.008 |
Neg logit contributions | |
Son | -0.008 |
distress | -0.008 |
Dad | -0.007 |
iver | -0.007 |
No | -0.007 |
Token |
|
Token ID | 198 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
ophone | +0.000 |
belongs | +0.000 |
endant | +0.000 |
Jupiter | +0.000 |
lights | +0.000 |
Neg logit contributions | |
Dad | -0.000 |
Son | -0.000 |
No | -0.000 |
Be | -0.000 |
M | -0.000 |
Token |
|
Token ID | 198 |
Feature activation | -2.61e-03 |
Pos logit contributions | |
Son | +0.001 |
Dad | +0.001 |
No | +0.001 |
distress | +0.001 |
Be | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
belongs | -0.001 |
unic | -0.001 |
endant | -0.001 |
Token | The |
Token ID | 464 |
Feature activation | -7.10e-01 |
Pos logit contributions | |
ophone | +0.000 |
Jupiter | +0.000 |
belongs | +0.000 |
dark | +0.000 |
unic | +0.000 |
Neg logit contributions | |
distress | -0.000 |
Son | -0.000 |
iver | -0.000 |
peel | -0.000 |
No | -0.000 |
Token | little |
Token ID | 1310 |
Feature activation | +3.89e+00 |
Pos logit contributions | |
ophone | +0.042 |
Jupiter | +0.041 |
belongs | +0.040 |
dark | +0.038 |
existed | +0.038 |
Neg logit contributions | |
distress | -0.029 |
Son | -0.029 |
Dad | -0.028 |
No | -0.028 |
Be | -0.027 |
Token | duck |
Token ID | 22045 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
Son | +0.125 |
distress | +0.123 |
No | +0.119 |
Dad | +0.119 |
Be | +0.113 |
Neg logit contributions | |
ophone | -0.259 |
Jupiter | -0.258 |
belongs | -0.250 |
unic | -0.243 |
dark | -0.240 |
Token | wanted |
Token ID | 2227 |
Feature activation | -5.35e-01 |
Pos logit contributions | |
Son | +0.007 |
distress | +0.007 |
No | +0.007 |
Dad | +0.007 |
Be | +0.007 |
Neg logit contributions | |
ophone | -0.007 |
Jupiter | -0.007 |
belongs | -0.007 |
unic | -0.007 |
dark | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
ophone | +0.031 |
belongs | +0.029 |
lights | +0.029 |
endant | +0.029 |
Jupiter | +0.029 |
Neg logit contributions | |
No | -0.024 |
distress | -0.023 |
Be | -0.023 |
Son | -0.022 |
Dad | -0.022 |
Token | scarf |
Token ID | 36953 |
Feature activation | -2.64e+00 |
Pos logit contributions | |
ophone | +0.028 |
Jupiter | +0.028 |
belongs | +0.027 |
unic | +0.026 |
dark | +0.026 |
Neg logit contributions | |
Son | -0.025 |
distress | -0.025 |
No | -0.024 |
Dad | -0.024 |
Be | -0.023 |
Token | so |
Token ID | 523 |
Feature activation | +2.65e-01 |
Pos logit contributions | |
Jupiter | +0.132 |
ophone | +0.131 |
belongs | +0.125 |
unic | +0.125 |
dark | +0.125 |
Neg logit contributions | |
distress | -0.123 |
No | -0.123 |
Son | -0.121 |
iver | -0.118 |
worry | -0.117 |
Token | the |
Token ID | 262 |
Feature activation | -2.59e-01 |
Pos logit contributions | |
distress | +0.010 |
No | +0.009 |
peel | +0.009 |
pack | +0.009 |
Son | +0.009 |
Neg logit contributions | |
ophone | -0.014 |
dark | -0.013 |
Jupiter | -0.013 |
belongs | -0.012 |
endant | -0.012 |
Token | big |
Token ID | 1263 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
ophone | +0.014 |
Jupiter | +0.014 |
belongs | +0.014 |
unic | +0.013 |
dark | +0.013 |
Neg logit contributions | |
Son | -0.011 |
distress | -0.011 |
Dad | -0.011 |
No | -0.011 |
iver | -0.011 |
Token | net |
Token ID | 2010 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
Son | +0.043 |
No | +0.041 |
Dad | +0.041 |
distress | +0.041 |
Be | +0.040 |
Neg logit contributions | |
Jupiter | -0.060 |
ophone | -0.059 |
belongs | -0.058 |
unic | -0.056 |
endant | -0.054 |
Token | at |
Token ID | 379 |
Feature activation | -8.62e-01 |
Pos logit contributions | |
distress | +0.141 |
peel | +0.141 |
pack | +0.130 |
No | +0.128 |
Son | +0.128 |
Neg logit contributions | |
Jupiter | -0.181 |
ophone | -0.175 |
dark | -0.161 |
unic | -0.160 |
endant | -0.158 |
Token | the |
Token ID | 262 |
Feature activation | -2.35e-01 |
Pos logit contributions | |
ophone | +0.045 |
Jupiter | +0.044 |
belongs | +0.043 |
unic | +0.041 |
dark | +0.041 |
Neg logit contributions | |
Son | -0.040 |
distress | -0.040 |
Dad | -0.039 |
No | -0.039 |
Be | -0.037 |
Token | end |
Token ID | 886 |
Feature activation | +4.60e+00 |
Pos logit contributions | |
ophone | +0.013 |
Jupiter | +0.013 |
belongs | +0.013 |
unic | +0.012 |
dark | +0.012 |
Neg logit contributions | |
Son | -0.010 |
distress | -0.010 |
No | -0.010 |
Dad | -0.010 |
Be | -0.009 |
Token | of |
Token ID | 286 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
Dad | +0.169 |
Son | +0.159 |
distress | +0.154 |
No | +0.151 |
peel | +0.146 |
Neg logit contributions | |
ophone | -0.297 |
Jupiter | -0.283 |
unic | -0.276 |
endant | -0.268 |
belongs | -0.267 |
Token | the |
Token ID | 262 |
Feature activation | -2.45e-01 |
Pos logit contributions | |
Son | +0.006 |
iver | +0.006 |
No | +0.006 |
distress | +0.006 |
Dad | +0.006 |
Neg logit contributions | |
Jupiter | -0.005 |
ophone | -0.004 |
unic | -0.004 |
belongs | -0.004 |
Venus | -0.004 |
Token | first |
Token ID | 717 |
Feature activation | +2.57e+00 |
Pos logit contributions | |
Jupiter | +0.011 |
ophone | +0.011 |
belongs | +0.011 |
unic | +0.010 |
dark | +0.010 |
Neg logit contributions | |
Son | -0.013 |
distress | -0.013 |
No | -0.013 |
Dad | -0.013 |
iver | -0.013 |
Token | paragraph |
Token ID | 7322 |
Feature activation | -7.00e-01 |
Pos logit contributions | |
Son | +0.128 |
distress | +0.126 |
No | +0.124 |
Dad | +0.122 |
Be | +0.120 |
Neg logit contributions | |
ophone | -0.126 |
Jupiter | -0.125 |
belongs | -0.121 |
unic | -0.115 |
dark | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | +9.60e-02 |
Pos logit contributions | |
Jupiter | +0.033 |
ophone | +0.032 |
endant | +0.031 |
unic | +0.031 |
belongs | +0.030 |
Neg logit contributions | |
Son | -0.036 |
Dad | -0.035 |
distress | -0.035 |
No | -0.034 |
iver | -0.034 |
Token | . |
Token ID | 13 |
Feature activation | +2.93e-01 |
Pos logit contributions | |
Son | +0.010 |
distress | +0.010 |
No | +0.009 |
Dad | +0.009 |
pack | +0.009 |
Neg logit contributions | |
Jupiter | -0.010 |
ophone | -0.010 |
belongs | -0.010 |
lights | -0.009 |
endant | -0.009 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
Son | +0.011 |
Dad | +0.010 |
Brother | +0.010 |
No | +0.009 |
M | +0.009 |
Neg logit contributions | |
endant | -0.019 |
ophone | -0.019 |
belongs | -0.019 |
unic | -0.018 |
existed | -0.018 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
Son | +0.137 |
distress | +0.133 |
Dad | +0.133 |
No | +0.131 |
Be | +0.128 |
Neg logit contributions | |
ophone | -0.182 |
Jupiter | -0.176 |
belongs | -0.173 |
endant | -0.168 |
unic | -0.167 |
Token | upon |
Token ID | 2402 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
Son | +0.043 |
Dad | +0.042 |
distress | +0.039 |
he | +0.038 |
No | +0.037 |
Neg logit contributions | |
ophone | -0.084 |
belongs | -0.083 |
Jupiter | -0.081 |
endant | -0.079 |
dark | -0.079 |
Token | a |
Token ID | 257 |
Feature activation | +3.36e-02 |
Pos logit contributions | |
Son | +0.101 |
distress | +0.096 |
Dad | +0.095 |
No | +0.094 |
Marcus | +0.093 |
Neg logit contributions | |
Jupiter | -0.088 |
ophone | -0.086 |
unic | -0.081 |
belongs | -0.080 |
dark | -0.079 |
Token | time |
Token ID | 640 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
Son | +0.002 |
distress | +0.002 |
No | +0.002 |
Dad | +0.002 |
Be | +0.002 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
belongs | -0.001 |
dark | -0.001 |
unic | -0.001 |
Token | , |
Token ID | 11 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
No | +0.045 |
distress | +0.043 |
Dad | +0.042 |
worry | +0.042 |
Son | +0.041 |
Neg logit contributions | |
ophone | -0.064 |
Jupiter | -0.060 |
dark | -0.059 |
belongs | -0.059 |
endant | -0.059 |
Token | there |
Token ID | 612 |
Feature activation | -1.18e+00 |
Pos logit contributions | |
Son | +0.007 |
distress | +0.007 |
No | +0.007 |
Dad | +0.007 |
Be | +0.007 |
Neg logit contributions | |
ophone | -0.007 |
Jupiter | -0.007 |
belongs | -0.007 |
unic | -0.007 |
endant | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
Jupiter | +0.041 |
ophone | +0.041 |
belongs | +0.040 |
unic | +0.037 |
existed | +0.036 |
Neg logit contributions | |
Son | -0.074 |
distress | -0.073 |
Dad | -0.072 |
No | -0.071 |
iver | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | +4.18e-02 |
Pos logit contributions | |
Jupiter | +0.047 |
ophone | +0.047 |
belongs | +0.044 |
unic | +0.044 |
dark | +0.043 |
Neg logit contributions | |
Son | -0.043 |
distress | -0.042 |
No | -0.041 |
Dad | -0.041 |
iver | -0.039 |
Token | nice |
Token ID | 3621 |
Feature activation | +4.79e-01 |
Pos logit contributions | |
distress | +0.002 |
Son | +0.002 |
No | +0.002 |
iver | +0.002 |
Dad | +0.002 |
Neg logit contributions | |
Jupiter | -0.002 |
ophone | -0.002 |
unic | -0.002 |
belongs | -0.002 |
dark | -0.002 |
Token | ." |
Token ID | 526 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
pack | +0.137 |
Dad | +0.129 |
peel | +0.126 |
No | +0.125 |
customer | +0.125 |
Neg logit contributions | |
Jupiter | -0.161 |
ophone | -0.150 |
unic | -0.145 |
belongs | -0.145 |
lights | -0.145 |
Token |
|
Token ID | 198 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ophone | +0.096 |
belongs | +0.092 |
Jupiter | +0.090 |
unic | +0.088 |
endant | +0.087 |
Neg logit contributions | |
Dad | -0.058 |
Son | -0.057 |
No | -0.055 |
distress | -0.053 |
Be | -0.052 |
Token |
|
Token ID | 198 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ophone | +0.009 |
Jupiter | +0.009 |
belongs | +0.009 |
unic | +0.008 |
endant | +0.008 |
Neg logit contributions | |
Son | -0.007 |
Dad | -0.007 |
distress | -0.007 |
No | -0.007 |
Be | -0.007 |
Token | Tom |
Token ID | 13787 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
Jupiter | +0.009 |
ophone | +0.009 |
belongs | +0.008 |
Venus | +0.008 |
unic | +0.008 |
Neg logit contributions | |
distress | -0.007 |
iver | -0.007 |
reel | -0.006 |
worry | -0.006 |
sausage | -0.006 |
Token | feels |
Token ID | 5300 |
Feature activation | +2.31e+00 |
Pos logit contributions | |
No | +0.069 |
Son | +0.069 |
Dad | +0.067 |
distress | +0.065 |
Be | +0.064 |
Neg logit contributions | |
belongs | -0.044 |
ophone | -0.043 |
Jupiter | -0.040 |
reminds | -0.036 |
belonged | -0.035 |
Token | sorry |
Token ID | 7926 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
distress | +0.111 |
Son | +0.108 |
No | +0.107 |
Dad | +0.106 |
Be | +0.101 |
Neg logit contributions | |
ophone | -0.126 |
Jupiter | -0.122 |
belongs | -0.114 |
endant | -0.114 |
unic | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | +4.72e-02 |
Pos logit contributions | |
ophone | +0.018 |
Jupiter | +0.018 |
belongs | +0.017 |
unic | +0.016 |
dark | +0.016 |
Neg logit contributions | |
Son | -0.031 |
distress | -0.031 |
No | -0.030 |
Dad | -0.030 |
Be | -0.029 |
Token | He |
Token ID | 679 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
Son | +0.002 |
distress | +0.002 |
No | +0.002 |
Dad | +0.002 |
Be | +0.002 |
Neg logit contributions | |
ophone | -0.003 |
Jupiter | -0.003 |
belongs | -0.003 |
unic | -0.003 |
dark | -0.002 |
Token | did |
Token ID | 750 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
distress | +0.115 |
peel | +0.113 |
Be | +0.111 |
No | +0.111 |
Son | +0.110 |
Neg logit contributions | |
Jupiter | -0.116 |
ophone | -0.116 |
lights | -0.114 |
existed | -0.104 |
unic | -0.104 |
Token | not |
Token ID | 407 |
Feature activation | -5.96e-01 |
Pos logit contributions | |
No | +0.056 |
distress | +0.055 |
peel | +0.053 |
Dad | +0.053 |
tug | +0.052 |
Neg logit contributions | |
ophone | -0.063 |
Jupiter | -0.061 |
unic | -0.060 |
endant | -0.059 |
lights | -0.059 |
Token | mean |
Token ID | 1612 |
Feature activation | -6.20e-01 |
Pos logit contributions | |
ophone | +0.033 |
Jupiter | +0.032 |
belongs | +0.031 |
existed | +0.030 |
dark | +0.030 |
Neg logit contributions | |
Son | -0.028 |
distress | -0.027 |
Marcus | -0.026 |
Dad | -0.025 |
No | -0.025 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
ophone | +0.024 |
Jupiter | +0.024 |
belongs | +0.023 |
unic | +0.022 |
dark | +0.021 |
Neg logit contributions | |
Son | -0.040 |
distress | -0.040 |
No | -0.039 |
Dad | -0.039 |
Be | -0.038 |
Token | Her |
Token ID | 2332 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
ophone | +0.108 |
Jupiter | +0.106 |
belongs | +0.102 |
unic | +0.101 |
endant | +0.100 |
Neg logit contributions | |
Dad | -0.104 |
No | -0.102 |
Son | -0.099 |
distress | -0.097 |
Be | -0.097 |
Token | mum |
Token ID | 25682 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
Jupiter | +0.076 |
ophone | +0.074 |
belongs | +0.073 |
unic | +0.070 |
endant | +0.068 |
Neg logit contributions | |
Son | -0.084 |
No | -0.082 |
distress | -0.081 |
Be | -0.080 |
iver | -0.080 |
Token | replied |
Token ID | 8712 |
Feature activation | +9.97e-01 |
Pos logit contributions | |
Son | +0.056 |
distress | +0.055 |
No | +0.055 |
Dad | +0.054 |
Be | +0.053 |
Neg logit contributions | |
ophone | -0.046 |
Jupiter | -0.046 |
belongs | -0.043 |
unic | -0.042 |
dark | -0.041 |
Token | , |
Token ID | 11 |
Feature activation | -4.24e-03 |
Pos logit contributions | |
Son | +0.049 |
Dad | +0.049 |
No | +0.049 |
distress | +0.048 |
Be | +0.046 |
Neg logit contributions | |
ophone | -0.050 |
Jupiter | -0.049 |
belongs | -0.047 |
endant | -0.045 |
unic | -0.045 |
Token | " |
Token ID | 366 |
Feature activation | +6.29e-02 |
Pos logit contributions | |
ophone | +0.000 |
Jupiter | +0.000 |
belongs | +0.000 |
endant | +0.000 |
unic | +0.000 |
Neg logit contributions | |
Son | -0.000 |
Dad | -0.000 |
No | -0.000 |
distress | -0.000 |
Be | -0.000 |
Token | We |
Token ID | 1135 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
distress | +0.002 |
Son | +0.002 |
Dad | +0.002 |
No | +0.002 |
iver | +0.002 |
Neg logit contributions | |
ophone | -0.004 |
Jupiter | -0.004 |
belongs | -0.004 |
unic | -0.004 |
dark | -0.004 |
Token | 're |
Token ID | 821 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
ophone | +0.061 |
Jupiter | +0.061 |
belongs | +0.058 |
unic | +0.056 |
dark | +0.056 |
Neg logit contributions | |
Son | -0.051 |
distress | -0.050 |
No | -0.049 |
Dad | -0.049 |
Be | -0.047 |
Token | going |
Token ID | 1016 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
distress | +0.136 |
Son | +0.135 |
pack | +0.135 |
Dad | +0.130 |
No | +0.128 |
Neg logit contributions | |
ophone | -0.159 |
endant | -0.151 |
Jupiter | -0.150 |
lights | -0.148 |
belongs | -0.146 |
Token | to |
Token ID | 284 |
Feature activation | +7.98e-01 |
Pos logit contributions | |
ophone | +0.100 |
Jupiter | +0.098 |
belongs | +0.097 |
endant | +0.096 |
unic | +0.094 |
Neg logit contributions | |
pack | -0.097 |
Son | -0.094 |
Dad | -0.093 |
No | -0.092 |
distress | -0.092 |
Token | surprise |
Token ID | 5975 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
Son | +0.036 |
distress | +0.036 |
No | +0.035 |
Dad | +0.035 |
iver | +0.034 |
Neg logit contributions | |
Jupiter | -0.043 |
ophone | -0.042 |
belongs | -0.040 |
unic | -0.040 |
dark | -0.038 |
Token | We |
Token ID | 775 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
ophone | +0.004 |
lights | +0.004 |
endant | +0.004 |
unic | +0.003 |
existed | +0.003 |
Neg logit contributions | |
Dad | -0.003 |
No | -0.003 |
Son | -0.003 |
Be | -0.003 |
distress | -0.003 |
Token | have |
Token ID | 423 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ophone | +0.086 |
Jupiter | +0.085 |
belongs | +0.083 |
unic | +0.078 |
dark | +0.077 |
Neg logit contributions | |
Son | -0.081 |
distress | -0.079 |
Dad | -0.078 |
No | -0.078 |
Be | -0.075 |
Token | to |
Token ID | 284 |
Feature activation | +6.07e-01 |
Pos logit contributions | |
distress | +0.066 |
No | +0.063 |
Dad | +0.063 |
Be | +0.061 |
Son | +0.061 |
Neg logit contributions | |
ophone | -0.091 |
Jupiter | -0.088 |
lights | -0.088 |
dark | -0.084 |
existed | -0.079 |
Token | respect |
Token ID | 2461 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
Son | +0.029 |
Be | +0.028 |
No | +0.028 |
distress | +0.027 |
Brother | +0.027 |
Neg logit contributions | |
Jupiter | -0.028 |
belongs | -0.028 |
unic | -0.026 |
ophone | -0.025 |
belonged | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -4.95e-01 |
Pos logit contributions | |
Dad | +0.078 |
Son | +0.075 |
No | +0.074 |
distress | +0.073 |
peel | +0.070 |
Neg logit contributions | |
ophone | -0.113 |
endant | -0.106 |
Jupiter | -0.105 |
dark | -0.104 |
lights | -0.103 |
Token | hero |
Token ID | 4293 |
Feature activation | +2.19e-01 |
Pos logit contributions | |
ophone | +0.027 |
Jupiter | +0.027 |
belongs | +0.026 |
dark | +0.025 |
unic | +0.025 |
Neg logit contributions | |
Son | -0.022 |
distress | -0.022 |
No | -0.022 |
Dad | -0.021 |
Be | -0.021 |
Token | and |
Token ID | 290 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
peel | +0.009 |
distress | +0.009 |
Son | +0.009 |
pack | +0.009 |
No | +0.008 |
Neg logit contributions | |
Jupiter | -0.012 |
endant | -0.012 |
ophone | -0.011 |
icol | -0.011 |
unic | -0.011 |
Token | leave |
Token ID | 2666 |
Feature activation | +4.55e-01 |
Pos logit contributions | |
distress | +0.004 |
No | +0.004 |
pack | +0.004 |
Dad | +0.004 |
peel | +0.004 |
Neg logit contributions | |
ophone | -0.008 |
Jupiter | -0.007 |
dark | -0.007 |
lights | -0.007 |
endant | -0.007 |
Token | the |
Token ID | 262 |
Feature activation | -4.50e-01 |
Pos logit contributions | |
Dad | +0.017 |
distress | +0.016 |
Son | +0.016 |
No | +0.015 |
half | +0.014 |
Neg logit contributions | |
ophone | -0.030 |
endant | -0.029 |
lights | -0.027 |
dark | -0.027 |
Jupiter | -0.026 |
Token | statue |
Token ID | 15207 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ophone | +0.028 |
Jupiter | +0.027 |
lights | +0.026 |
belongs | +0.026 |
dark | +0.025 |
Neg logit contributions | |
distress | -0.017 |
No | -0.017 |
Dad | -0.017 |
Son | -0.017 |
peel | -0.016 |
Token | here |
Token ID | 994 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
icol | +0.123 |
Jupiter | +0.122 |
unic | +0.117 |
Venus | +0.110 |
endant | +0.109 |
Neg logit contributions | |
peel | -0.080 |
he | -0.078 |
pack | -0.074 |
iver | -0.070 |
' | -0.070 |
Token | was |
Token ID | 373 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ophone | +0.267 |
Jupiter | +0.266 |
belongs | +0.251 |
endant | +0.248 |
dark | +0.246 |
Neg logit contributions | |
Son | -0.256 |
distress | -0.250 |
Dad | -0.246 |
No | -0.242 |
Be | -0.235 |
Token | very |
Token ID | 845 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
ophone | +0.057 |
belongs | +0.056 |
Jupiter | +0.055 |
endant | +0.053 |
lights | +0.053 |
Neg logit contributions | |
distress | -0.044 |
Son | -0.043 |
No | -0.042 |
Dad | -0.041 |
Be | -0.040 |
Token | excited |
Token ID | 6568 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
ophone | +0.097 |
Jupiter | +0.095 |
unic | +0.090 |
lights | +0.090 |
belongs | +0.089 |
Neg logit contributions | |
Dad | -0.061 |
No | -0.061 |
Son | -0.060 |
distress | -0.059 |
worry | -0.058 |
Token | and |
Token ID | 290 |
Feature activation | +3.55e-01 |
Pos logit contributions | |
No | +0.090 |
distress | +0.090 |
Son | +0.089 |
Dad | +0.087 |
Be | +0.083 |
Neg logit contributions | |
ophone | -0.111 |
Jupiter | -0.107 |
belongs | -0.102 |
dark | -0.099 |
unic | -0.099 |
Token | thanked |
Token ID | 26280 |
Feature activation | -2.17e+00 |
Pos logit contributions | |
No | +0.013 |
pack | +0.013 |
worry | +0.012 |
Dad | +0.012 |
scoop | +0.012 |
Neg logit contributions | |
ophone | -0.023 |
endant | -0.022 |
lights | -0.022 |
Jupiter | -0.022 |
dark | -0.021 |
Token | the |
Token ID | 262 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
ophone | +0.137 |
belongs | +0.129 |
endant | +0.124 |
Jupiter | +0.120 |
dark | +0.120 |
Neg logit contributions | |
distress | -0.085 |
Dad | -0.084 |
No | -0.084 |
Son | -0.083 |
Be | -0.079 |
Token | kind |
Token ID | 1611 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
ophone | +0.028 |
lights | +0.028 |
belongs | +0.025 |
endant | +0.025 |
existed | +0.024 |
Neg logit contributions | |
distress | -0.019 |
No | -0.019 |
Dad | -0.018 |
customer | -0.018 |
peel | -0.018 |
Token | witch |
Token ID | 16365 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
distress | +0.006 |
Dad | +0.006 |
Son | +0.006 |
No | +0.006 |
Col | +0.006 |
Neg logit contributions | |
ophone | -0.008 |
Jupiter | -0.008 |
belongs | -0.008 |
lights | -0.007 |
endant | -0.007 |
Token | for |
Token ID | 329 |
Feature activation | +4.37e-01 |
Pos logit contributions | |
endant | +0.185 |
ophone | +0.185 |
lights | +0.183 |
belongs | +0.182 |
Jupiter | +0.179 |
Neg logit contributions | |
distress | -0.148 |
pack | -0.147 |
peel | -0.144 |
Son | -0.143 |
' | -0.135 |
Token | her |
Token ID | 607 |
Feature activation | -7.68e-01 |
Pos logit contributions | |
Son | +0.020 |
distress | +0.020 |
No | +0.020 |
Dad | +0.020 |
Be | +0.019 |
Neg logit contributions | |
ophone | -0.023 |
Jupiter | -0.023 |
belongs | -0.022 |
unic | -0.021 |
dark | -0.021 |
Token | useful |
Token ID | 4465 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
ophone | +0.048 |
Jupiter | +0.046 |
belongs | +0.045 |
lights | +0.045 |
dark | +0.044 |
Neg logit contributions | |
distress | -0.030 |
worry | -0.029 |
pack | -0.028 |
No | -0.027 |
tug | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | +7.58e-01 |
Pos logit contributions | |
ophone | +0.008 |
belongs | +0.008 |
endant | +0.008 |
Jupiter | +0.007 |
lights | +0.007 |
Neg logit contributions | |
pack | -0.006 |
peel | -0.005 |
distress | -0.005 |
Be | -0.005 |
Son | -0.005 |
Token | sleep |
Token ID | 3993 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
distress | +0.032 |
pack | +0.031 |
Son | +0.029 |
peel | +0.029 |
iver | +0.029 |
Neg logit contributions | |
ophone | -0.045 |
endant | -0.042 |
belongs | -0.042 |
lights | -0.041 |
dark | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
ophone | +0.017 |
endant | +0.017 |
belongs | +0.016 |
Jupiter | +0.015 |
dark | +0.015 |
Neg logit contributions | |
pack | -0.012 |
distress | -0.011 |
iver | -0.011 |
Dad | -0.011 |
Son | -0.011 |
Token | he |
Token ID | 339 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
ophone | +0.002 |
endant | +0.002 |
belongs | +0.002 |
dark | +0.002 |
Jupiter | +0.002 |
Neg logit contributions | |
pack | -0.002 |
Son | -0.002 |
distress | -0.002 |
worry | -0.002 |
No | -0.002 |
Token | would |
Token ID | 561 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
Jupiter | +0.024 |
lights | +0.024 |
endant | +0.024 |
ophone | +0.024 |
dark | +0.022 |
Neg logit contributions | |
pack | -0.029 |
Son | -0.029 |
distress | -0.028 |
peel | -0.027 |
Be | -0.027 |
Token | dream |
Token ID | 4320 |
Feature activation | -2.12e+00 |
Pos logit contributions | |
Jupiter | +0.169 |
ophone | +0.167 |
belongs | +0.161 |
unic | +0.158 |
dark | +0.152 |
Neg logit contributions | |
Son | -0.144 |
distress | -0.141 |
No | -0.140 |
Dad | -0.139 |
Be | -0.134 |
Token | of |
Token ID | 286 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
ophone | +0.113 |
Jupiter | +0.112 |
endant | +0.110 |
belongs | +0.110 |
unic | +0.102 |
Neg logit contributions | |
iver | -0.093 |
No | -0.091 |
Son | -0.089 |
Be | -0.089 |
distress | -0.088 |
Token | the |
Token ID | 262 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
distress | +0.005 |
worry | +0.004 |
Son | +0.004 |
pack | +0.004 |
panic | +0.004 |
Neg logit contributions | |
ophone | -0.008 |
dark | -0.008 |
belongs | -0.007 |
endant | -0.007 |
lights | -0.007 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.25e-01 |
Pos logit contributions | |
ophone | +0.023 |
lights | +0.020 |
endant | +0.020 |
belongs | +0.019 |
dark | +0.019 |
Neg logit contributions | |
distress | -0.018 |
pack | -0.017 |
worry | -0.017 |
br | -0.017 |
Col | -0.017 |
Token | kingdom |
Token ID | 13239 |
Feature activation | -3.14e+00 |
Pos logit contributions | |
ophone | +0.012 |
Jupiter | +0.010 |
belongs | +0.010 |
endant | +0.010 |
dark | +0.010 |
Neg logit contributions | |
distress | -0.011 |
Dad | -0.011 |
pack | -0.011 |
No | -0.011 |
worry | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
endant | +0.167 |
Jupiter | +0.166 |
ophone | +0.163 |
belongs | +0.157 |
dark | +0.155 |
Neg logit contributions | |
distress | -0.122 |
pack | -0.121 |
Son | -0.120 |
he | -0.119 |
No | -0.115 |
Token | his |
Token ID | 465 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
Son | +0.014 |
distress | +0.014 |
No | +0.014 |
Dad | +0.014 |
iver | +0.013 |
Neg logit contributions | |
Jupiter | -0.020 |
ophone | -0.020 |
belongs | -0.019 |
unic | -0.019 |
dark | -0.018 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
distress | +0.079 |
Dad | +0.071 |
worry | +0.068 |
No | +0.068 |
Son | +0.068 |
Neg logit contributions | |
ophone | -0.137 |
belongs | -0.130 |
Jupiter | -0.128 |
dark | -0.124 |
existed | -0.123 |
Token | went |
Token ID | 1816 |
Feature activation | -2.42e-01 |
Pos logit contributions | |
ophone | +0.022 |
Jupiter | +0.022 |
belongs | +0.021 |
unic | +0.020 |
dark | +0.020 |
Neg logit contributions | |
Son | -0.027 |
distress | -0.026 |
No | -0.026 |
Dad | -0.026 |
Be | -0.025 |
Token | on |
Token ID | 319 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ophone | +0.012 |
Jupiter | +0.012 |
belongs | +0.012 |
unic | +0.011 |
endant | +0.011 |
Neg logit contributions | |
distress | -0.012 |
Son | -0.012 |
Dad | -0.012 |
No | -0.012 |
pack | -0.012 |
Token | many |
Token ID | 867 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
ophone | +0.082 |
Jupiter | +0.078 |
belongs | +0.078 |
endant | +0.074 |
dark | +0.074 |
Neg logit contributions | |
distress | -0.073 |
Son | -0.071 |
No | -0.071 |
Dad | -0.070 |
Be | -0.067 |
Token | more |
Token ID | 517 |
Feature activation | -3.07e+00 |
Pos logit contributions | |
distress | +0.008 |
Son | +0.008 |
No | +0.008 |
Dad | +0.008 |
Be | +0.008 |
Neg logit contributions | |
ophone | -0.007 |
Jupiter | -0.007 |
belongs | -0.006 |
endant | -0.006 |
dark | -0.006 |
Token | adventures |
Token ID | 17545 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ophone | +0.164 |
Jupiter | +0.149 |
endant | +0.147 |
belongs | +0.144 |
dark | +0.139 |
Neg logit contributions | |
distress | -0.153 |
No | -0.145 |
peel | -0.143 |
pack | -0.141 |
iver | -0.141 |
Token | together |
Token ID | 1978 |
Feature activation | -9.01e-01 |
Pos logit contributions | |
ophone | +0.112 |
endant | +0.105 |
Jupiter | +0.103 |
dark | +0.102 |
belongs | +0.101 |
Neg logit contributions | |
distress | -0.086 |
tug | -0.083 |
No | -0.083 |
pack | -0.082 |
he | -0.081 |
Token | ! |
Token ID | 0 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
ophone | +0.044 |
Jupiter | +0.042 |
belongs | +0.041 |
endant | +0.040 |
dark | +0.039 |
Neg logit contributions | |
distress | -0.046 |
Son | -0.045 |
No | -0.045 |
Dad | -0.045 |
peel | -0.043 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +3.17e+00 |
Pos logit contributions | |
Dad | +0.068 |
Son | +0.065 |
No | +0.064 |
distress | +0.061 |
Be | +0.061 |
Neg logit contributions | |
ophone | -0.082 |
endant | -0.079 |
belongs | -0.078 |
unic | -0.078 |
Jupiter | -0.078 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
Son | +0.146 |
distress | +0.144 |
Dad | +0.142 |
No | +0.141 |
Be | +0.138 |
Neg logit contributions | |
ophone | -0.167 |
Jupiter | -0.165 |
belongs | -0.159 |
unic | -0.153 |
endant | -0.151 |
Token | . |
Token ID | 13 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
ophone | +0.408 |
belongs | +0.401 |
Jupiter | +0.399 |
endant | +0.392 |
existed | +0.386 |
Neg logit contributions | |
pack | -0.284 |
No | -0.276 |
Dad | -0.266 |
customer | -0.260 |
peel | -0.258 |
Token | It |
Token ID | 632 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
Son | +0.019 |
distress | +0.018 |
Dad | +0.018 |
No | +0.018 |
iver | +0.018 |
Neg logit contributions | |
Jupiter | -0.012 |
ophone | -0.012 |
belongs | -0.011 |
unic | -0.011 |
dark | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
Be | +0.001 |
M | +0.000 |
No | +0.000 |
That | +0.000 |
he | +0.000 |
Neg logit contributions | |
endant | -0.001 |
majesty | -0.001 |
icol | -0.001 |
lan | -0.001 |
Majesty | -0.001 |
Token | full |
Token ID | 1336 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
ophone | +0.057 |
belongs | +0.055 |
endant | +0.053 |
existed | +0.052 |
dark | +0.051 |
Neg logit contributions | |
Be | -0.032 |
No | -0.032 |
pack | -0.031 |
distress | -0.031 |
Dad | -0.030 |
Token | of |
Token ID | 286 |
Feature activation | +3.30e-01 |
Pos logit contributions | |
Dad | +0.009 |
No | +0.009 |
Son | +0.009 |
Be | +0.008 |
pack | +0.008 |
Neg logit contributions | |
ophone | -0.022 |
Jupiter | -0.022 |
belongs | -0.021 |
endant | -0.021 |
existed | -0.020 |
Token | people |
Token ID | 661 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
Dad | +0.013 |
Be | +0.012 |
No | +0.012 |
distress | +0.012 |
pack | +0.012 |
Neg logit contributions | |
ophone | -0.022 |
belongs | -0.020 |
endant | -0.019 |
existed | -0.019 |
dark | -0.018 |
Token | b |
Token ID | 275 |
Feature activation | +2.28e-01 |
Pos logit contributions | |
Jupiter | +0.133 |
ophone | +0.132 |
belongs | +0.131 |
endant | +0.131 |
lights | +0.131 |
Neg logit contributions | |
Son | -0.076 |
pack | -0.075 |
distress | -0.070 |
Dad | -0.069 |
iver | -0.068 |
Token | owing |
Token ID | 7855 |
Feature activation | +2.95e+00 |
Pos logit contributions | |
Son | +0.013 |
distress | +0.013 |
Dad | +0.013 |
No | +0.013 |
iver | +0.012 |
Neg logit contributions | |
ophone | -0.010 |
Jupiter | -0.009 |
belongs | -0.009 |
unic | -0.008 |
dark | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +8.13e-01 |
Pos logit contributions | |
pack | +0.097 |
No | +0.096 |
Dad | +0.096 |
Son | +0.095 |
distress | +0.095 |
Neg logit contributions | |
ophone | -0.181 |
endant | -0.179 |
belongs | -0.176 |
dark | -0.172 |
Jupiter | -0.170 |
Token | the |
Token ID | 262 |
Feature activation | -3.51e-01 |
Pos logit contributions | |
Dad | +0.031 |
Son | +0.031 |
distress | +0.031 |
No | +0.031 |
Be | +0.030 |
Neg logit contributions | |
ophone | -0.049 |
Jupiter | -0.047 |
belongs | -0.047 |
endant | -0.046 |
dark | -0.045 |
Token | stage |
Token ID | 3800 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
ophone | +0.022 |
endant | +0.022 |
belongs | +0.021 |
existed | +0.021 |
Jupiter | +0.020 |
Neg logit contributions | |
Dad | -0.013 |
No | -0.012 |
pack | -0.012 |
distress | -0.011 |
br | -0.011 |
Token | here |
Token ID | 994 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
distress | +0.153 |
iver | +0.147 |
Son | +0.144 |
No | +0.140 |
Be | +0.137 |
Neg logit contributions | |
Jupiter | -0.235 |
ophone | -0.228 |
belongs | -0.227 |
lights | -0.227 |
dark | -0.223 |
Token | to |
Token ID | 284 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
ophone | +0.078 |
Jupiter | +0.076 |
endant | +0.076 |
belongs | +0.075 |
lights | +0.074 |
Neg logit contributions | |
No | -0.055 |
pack | -0.054 |
distress | -0.052 |
worry | -0.052 |
Dad | -0.051 |
Token | catch |
Token ID | 4929 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
distress | +0.018 |
No | +0.018 |
Be | +0.017 |
Son | +0.017 |
iver | +0.017 |
Neg logit contributions | |
ophone | -0.039 |
Jupiter | -0.038 |
belongs | -0.037 |
lights | -0.037 |
dark | -0.036 |
Token | you |
Token ID | 345 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
No | +0.028 |
distress | +0.028 |
Son | +0.027 |
Dad | +0.026 |
Be | +0.026 |
Neg logit contributions | |
ophone | -0.025 |
Jupiter | -0.025 |
belongs | -0.023 |
endant | -0.023 |
lights | -0.023 |
Token | if |
Token ID | 611 |
Feature activation | -6.78e-01 |
Pos logit contributions | |
lights | +0.072 |
Jupiter | +0.072 |
endant | +0.072 |
ophone | +0.071 |
belongs | +0.068 |
Neg logit contributions | |
iver | -0.083 |
peel | -0.081 |
distress | -0.081 |
No | -0.080 |
pack | -0.079 |
Token | you |
Token ID | 345 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ophone | +0.042 |
Jupiter | +0.040 |
endant | +0.039 |
lights | +0.038 |
icol | +0.037 |
Neg logit contributions | |
No | -0.030 |
distress | -0.027 |
worry | -0.027 |
Be | -0.025 |
Dad | -0.025 |
Token | fall |
Token ID | 2121 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
Jupiter | +0.104 |
ophone | +0.102 |
lights | +0.098 |
unic | +0.097 |
belongs | +0.096 |
Neg logit contributions | |
distress | -0.058 |
Son | -0.057 |
No | -0.057 |
iver | -0.056 |
peel | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -2.98e-02 |
Pos logit contributions | |
Jupiter | +0.083 |
ophone | +0.082 |
endant | +0.081 |
belongs | +0.080 |
dark | +0.077 |
Neg logit contributions | |
No | -0.067 |
iver | -0.066 |
distress | -0.065 |
ouch | -0.065 |
Be | -0.064 |
Token | I |
Token ID | 314 |
Feature activation | -9.85e-01 |
Pos logit contributions | |
ophone | +0.001 |
lights | +0.001 |
belongs | +0.001 |
endant | +0.001 |
Jupiter | +0.001 |
Neg logit contributions | |
No | -0.002 |
Son | -0.002 |
Be | -0.002 |
Dad | -0.002 |
Brother | -0.002 |
Token | promise |
Token ID | 6991 |
Feature activation | +3.21e+00 |
Pos logit contributions | |
Jupiter | +0.052 |
ophone | +0.051 |
lights | +0.050 |
unic | +0.049 |
dark | +0.047 |
Neg logit contributions | |
distress | -0.045 |
No | -0.045 |
Son | -0.045 |
Be | -0.044 |
iver | -0.043 |
Token | ." |
Token ID | 526 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
No | +0.156 |
distress | +0.141 |
pack | +0.139 |
worry | +0.138 |
Dad | +0.138 |
Neg logit contributions | |
ophone | -0.169 |
Jupiter | -0.169 |
lights | -0.167 |
dark | -0.166 |
endant | -0.161 |
Token | a |
Token ID | 257 |
Feature activation | -1.65e-01 |
Pos logit contributions | |
No | +0.149 |
distress | +0.145 |
Son | +0.140 |
Be | +0.138 |
pack | +0.135 |
Neg logit contributions | |
ophone | -0.204 |
Jupiter | -0.194 |
endant | -0.192 |
belongs | -0.192 |
lights | -0.191 |
Token | nap |
Token ID | 25422 |
Feature activation | +4.70e+00 |
Pos logit contributions | |
Jupiter | +0.012 |
ophone | +0.012 |
belongs | +0.012 |
unic | +0.011 |
lights | +0.011 |
Neg logit contributions | |
No | -0.004 |
distress | -0.004 |
Son | -0.004 |
Be | -0.004 |
Dad | -0.004 |
Token | Teddy |
Token ID | 29345 |
Feature activation | -4.25e-02 |
Pos logit contributions | |
Son | +0.303 |
distress | +0.301 |
Be | +0.297 |
No | +0.295 |
Dad | +0.295 |
Neg logit contributions | |
ophone | -0.157 |
Jupiter | -0.156 |
belongs | -0.148 |
lights | -0.137 |
unic | -0.134 |
Token | ? |
Token ID | 30 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
ophone | +0.002 |
Jupiter | +0.001 |
belongs | +0.001 |
unic | +0.001 |
dark | +0.001 |
Neg logit contributions | |
Son | -0.003 |
No | -0.003 |
Dad | -0.003 |
distress | -0.003 |
Be | -0.003 |
Token | You |
Token ID | 921 |
Feature activation | -2.60e-01 |
Pos logit contributions | |
Jupiter | +0.021 |
ophone | +0.021 |
belongs | +0.021 |
unic | +0.020 |
endant | +0.020 |
Neg logit contributions | |
Son | -0.029 |
No | -0.029 |
distress | -0.028 |
Dad | -0.028 |
Be | -0.028 |
Token | 'll |
Token ID | 1183 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ophone | +0.012 |
Jupiter | +0.012 |
belongs | +0.011 |
unic | +0.011 |
lights | +0.011 |
Neg logit contributions | |
distress | -0.013 |
Son | -0.013 |
No | -0.013 |
Dad | -0.013 |
iver | -0.012 |
Token | feel |
Token ID | 1254 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
Jupiter | +0.067 |
ophone | +0.066 |
unic | +0.064 |
belongs | +0.064 |
existed | +0.061 |
Neg logit contributions | |
Son | -0.048 |
Dad | -0.046 |
distress | -0.046 |
No | -0.045 |
Be | -0.043 |
Token | better |
Token ID | 1365 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
distress | +0.182 |
Son | +0.175 |
No | +0.170 |
iver | +0.168 |
worry | +0.164 |
Neg logit contributions | |
ophone | -0.227 |
Jupiter | -0.212 |
belongs | -0.205 |
endant | -0.202 |
lights | -0.202 |
Token | if |
Token ID | 611 |
Feature activation | -6.10e-01 |
Pos logit contributions | |
Jupiter | +0.073 |
ophone | +0.073 |
belongs | +0.069 |
unic | +0.065 |
existed | +0.064 |
Neg logit contributions | |
Son | -0.114 |
distress | -0.112 |
No | -0.111 |
Dad | -0.111 |
Be | -0.110 |
Token | you |
Token ID | 345 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ophone | +0.032 |
Jupiter | +0.030 |
endant | +0.030 |
dark | +0.027 |
lights | +0.027 |
Neg logit contributions | |
No | -0.032 |
Son | -0.031 |
distress | -0.030 |
worry | -0.030 |
Dad | -0.029 |
Token | do |
Token ID | 466 |
Feature activation | +6.38e+00 |
Pos logit contributions | |
ophone | +0.090 |
Jupiter | +0.089 |
belongs | +0.087 |
unic | +0.084 |
dark | +0.083 |
Neg logit contributions | |
Son | -0.056 |
distress | -0.055 |
No | -0.054 |
Dad | -0.054 |
Be | -0.052 |
Token | there |
Token ID | 612 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
distress | +0.007 |
No | +0.007 |
Son | +0.007 |
Dad | +0.007 |
Be | +0.007 |
Neg logit contributions | |
ophone | -0.008 |
Jupiter | -0.008 |
belongs | -0.007 |
unic | -0.007 |
endant | -0.007 |
Token | was |
Token ID | 373 |
Feature activation | -8.24e-01 |
Pos logit contributions | |
ophone | +0.051 |
Jupiter | +0.051 |
belongs | +0.049 |
unic | +0.046 |
dark | +0.045 |
Neg logit contributions | |
Son | -0.082 |
distress | -0.081 |
Dad | -0.080 |
No | -0.080 |
iver | -0.078 |
Token | a |
Token ID | 257 |
Feature activation | -9.45e-02 |
Pos logit contributions | |
Jupiter | +0.042 |
ophone | +0.042 |
belongs | +0.040 |
unic | +0.039 |
dark | +0.039 |
Neg logit contributions | |
Son | -0.039 |
distress | -0.038 |
No | -0.037 |
Dad | -0.037 |
iver | -0.036 |
Token | little |
Token ID | 1310 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
Jupiter | +0.005 |
ophone | +0.005 |
belongs | +0.005 |
unic | +0.005 |
dark | +0.005 |
Neg logit contributions | |
distress | -0.004 |
Son | -0.004 |
No | -0.004 |
iver | -0.004 |
Dad | -0.004 |
Token | boy |
Token ID | 2933 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
distress | +0.163 |
Son | +0.161 |
iver | +0.157 |
No | +0.152 |
Dad | +0.151 |
Neg logit contributions | |
Jupiter | -0.240 |
ophone | -0.234 |
belongs | -0.230 |
unic | -0.229 |
dark | -0.220 |
Token | called |
Token ID | 1444 |
Feature activation | -5.86e+00 |
Pos logit contributions | |
Son | +0.283 |
distress | +0.281 |
Dad | +0.279 |
No | +0.279 |
Be | +0.271 |
Neg logit contributions | |
ophone | -0.096 |
Jupiter | -0.095 |
belongs | -0.088 |
unic | -0.079 |
endant | -0.078 |
Token | Tom |
Token ID | 4186 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
ophone | +0.278 |
belongs | +0.270 |
Jupiter | +0.263 |
endant | +0.259 |
lights | +0.255 |
Neg logit contributions | |
Son | -0.303 |
Dad | -0.298 |
No | -0.294 |
distress | -0.292 |
peel | -0.283 |
Token | . |
Token ID | 13 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
Son | +0.050 |
distress | +0.049 |
No | +0.049 |
Dad | +0.048 |
Be | +0.048 |
Neg logit contributions | |
ophone | -0.028 |
Jupiter | -0.027 |
belongs | -0.026 |
unic | -0.024 |
dark | -0.024 |
Token | Tom |
Token ID | 4186 |
Feature activation | +7.73e-01 |
Pos logit contributions | |
Son | +0.025 |
distress | +0.024 |
iver | +0.023 |
Dad | +0.022 |
No | +0.022 |
Neg logit contributions | |
ophone | -0.017 |
Jupiter | -0.017 |
belongs | -0.017 |
dark | -0.015 |
existed | -0.015 |
Token | loved |
Token ID | 6151 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
Dad | +0.043 |
Son | +0.042 |
No | +0.041 |
distress | +0.041 |
Be | +0.040 |
Neg logit contributions | |
Jupiter | -0.034 |
ophone | -0.033 |
lights | -0.032 |
unic | -0.030 |
endant | -0.030 |
Token | to |
Token ID | 284 |
Feature activation | +8.12e-01 |
Pos logit contributions | |
Jupiter | +0.061 |
ophone | +0.059 |
unic | +0.057 |
belongs | +0.056 |
dark | +0.053 |
Neg logit contributions | |
Son | -0.107 |
distress | -0.106 |
No | -0.104 |
iver | -0.104 |
Dad | -0.104 |
Token | there |
Token ID | 612 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
Son | +0.001 |
distress | +0.001 |
No | +0.001 |
Dad | +0.001 |
Be | +0.001 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.001 |
belongs | -0.001 |
unic | -0.001 |
dark | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
ophone | +0.049 |
Jupiter | +0.049 |
belongs | +0.047 |
unic | +0.044 |
dark | +0.043 |
Neg logit contributions | |
Son | -0.078 |
distress | -0.077 |
Dad | -0.076 |
No | -0.076 |
iver | -0.074 |
Token | a |
Token ID | 257 |
Feature activation | -4.01e-02 |
Pos logit contributions | |
Jupiter | +0.047 |
ophone | +0.046 |
unic | +0.043 |
dark | +0.043 |
belongs | +0.043 |
Neg logit contributions | |
Son | -0.046 |
distress | -0.045 |
No | -0.044 |
Dad | -0.043 |
iver | -0.043 |
Token | little |
Token ID | 1310 |
Feature activation | +3.99e+00 |
Pos logit contributions | |
Jupiter | +0.002 |
ophone | +0.002 |
unic | +0.002 |
belongs | +0.002 |
dark | +0.002 |
Neg logit contributions | |
distress | -0.002 |
Son | -0.002 |
iver | -0.002 |
No | -0.002 |
Dad | -0.002 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
distress | +0.163 |
Son | +0.158 |
iver | +0.155 |
No | +0.150 |
Dad | +0.148 |
Neg logit contributions | |
Jupiter | -0.231 |
ophone | -0.224 |
unic | -0.221 |
belongs | -0.221 |
dark | -0.212 |
Token | named |
Token ID | 3706 |
Feature activation | -7.64e+00 |
Pos logit contributions | |
Dad | +0.112 |
M | +0.108 |
Son | +0.108 |
No | +0.108 |
distress | +0.106 |
Neg logit contributions | |
ophone | -0.072 |
Jupiter | -0.070 |
belongs | -0.069 |
lights | -0.068 |
existed | -0.067 |
Token | Lily |
Token ID | 20037 |
Feature activation | -1.29e+00 |
Pos logit contributions | |
ophone | +0.407 |
lights | +0.393 |
Jupiter | +0.378 |
belongs | +0.368 |
endant | +0.368 |
Neg logit contributions | |
M | -0.372 |
Dad | -0.356 |
No | -0.353 |
peel | -0.347 |
Be | -0.340 |
Token | . |
Token ID | 13 |
Feature activation | +3.01e-01 |
Pos logit contributions | |
ophone | +0.054 |
Jupiter | +0.053 |
belongs | +0.051 |
unic | +0.048 |
dark | +0.047 |
Neg logit contributions | |
Son | -0.074 |
distress | -0.073 |
Dad | -0.072 |
No | -0.072 |
Be | -0.070 |
Token | She |
Token ID | 1375 |
Feature activation | -2.08e-01 |
Pos logit contributions | |
No | +0.013 |
Dad | +0.013 |
distress | +0.013 |
Be | +0.013 |
Son | +0.013 |
Neg logit contributions | |
ophone | -0.017 |
Jupiter | -0.016 |
belongs | -0.016 |
endant | -0.015 |
unic | -0.015 |
Token | had |
Token ID | 550 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
Jupiter | +0.008 |
ophone | +0.008 |
lights | +0.008 |
endant | +0.008 |
belongs | +0.008 |
Neg logit contributions | |
Dad | -0.012 |
No | -0.012 |
Be | -0.012 |
Son | -0.012 |
M | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
ophone | +0.055 |
Jupiter | +0.054 |
belongs | +0.053 |
unic | +0.050 |
dark | +0.050 |
Neg logit contributions | |
Son | -0.042 |
distress | -0.042 |
No | -0.041 |
Dad | -0.041 |
Be | -0.039 |
Token | morning |
Token ID | 3329 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
ophone | +0.015 |
Jupiter | +0.015 |
belongs | +0.014 |
unic | +0.014 |
dark | +0.013 |
Neg logit contributions | |
Son | -0.011 |
distress | -0.011 |
No | -0.011 |
Dad | -0.010 |
Be | -0.010 |
Token | and |
Token ID | 290 |
Feature activation | +4.12e-01 |
Pos logit contributions | |
ophone | +0.065 |
belongs | +0.064 |
Jupiter | +0.062 |
unic | +0.060 |
endant | +0.059 |
Neg logit contributions | |
Dad | -0.047 |
pack | -0.047 |
iver | -0.047 |
peel | -0.046 |
Son | -0.045 |
Token | found |
Token ID | 1043 |
Feature activation | -9.34e-01 |
Pos logit contributions | |
Be | +0.014 |
Ben | +0.013 |
Dad | +0.012 |
pack | +0.012 |
M | +0.012 |
Neg logit contributions | |
lights | -0.030 |
endant | -0.028 |
belongs | -0.027 |
unic | -0.026 |
ophone | -0.026 |
Token | that |
Token ID | 326 |
Feature activation | +5.19e-01 |
Pos logit contributions | |
ophone | +0.058 |
Jupiter | +0.056 |
belongs | +0.055 |
unic | +0.053 |
lights | +0.053 |
Neg logit contributions | |
Son | -0.036 |
distress | -0.036 |
No | -0.035 |
Dad | -0.035 |
Be | -0.034 |
Token | the |
Token ID | 262 |
Feature activation | -2.49e-01 |
Pos logit contributions | |
No | +0.023 |
Be | +0.021 |
distress | +0.021 |
Dad | +0.021 |
worry | +0.020 |
Neg logit contributions | |
ophone | -0.032 |
endant | -0.030 |
dark | -0.029 |
belongs | -0.028 |
Jupiter | -0.028 |
Token | fog |
Token ID | 19558 |
Feature activation | -5.80e+00 |
Pos logit contributions | |
ophone | +0.012 |
Jupiter | +0.012 |
belongs | +0.011 |
unic | +0.011 |
dark | +0.011 |
Neg logit contributions | |
Son | -0.013 |
distress | -0.013 |
No | -0.013 |
Dad | -0.013 |
Be | -0.012 |
Token | had |
Token ID | 550 |
Feature activation | -9.60e-01 |
Pos logit contributions | |
Jupiter | +0.389 |
unic | +0.370 |
ophone | +0.363 |
Saturn | +0.358 |
Venus | +0.357 |
Neg logit contributions | |
distress | -0.185 |
No | -0.162 |
pack | -0.160 |
peel | -0.154 |
worry | -0.148 |
Token | covered |
Token ID | 5017 |
Feature activation | -1.51e+00 |
Pos logit contributions | |
ophone | +0.055 |
Jupiter | +0.051 |
belongs | +0.050 |
lights | +0.048 |
dark | +0.047 |
Neg logit contributions | |
Be | -0.044 |
distress | -0.043 |
No | -0.042 |
peel | -0.039 |
Dad | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
ophone | +0.078 |
Jupiter | +0.077 |
belongs | +0.074 |
unic | +0.071 |
endant | +0.071 |
Neg logit contributions | |
Son | -0.070 |
distress | -0.070 |
No | -0.070 |
Dad | -0.069 |
Be | -0.067 |
Token | whole |
Token ID | 2187 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
ophone | +0.010 |
Jupiter | +0.010 |
belongs | +0.009 |
endant | +0.009 |
dark | +0.009 |
Neg logit contributions | |
No | -0.008 |
distress | -0.008 |
Dad | -0.008 |
Be | -0.008 |
iver | -0.008 |
Token | park |
Token ID | 3952 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
Son | +0.186 |
distress | +0.183 |
No | +0.179 |
Dad | +0.178 |
Be | +0.173 |
Neg logit contributions | |
Jupiter | -0.192 |
ophone | -0.192 |
belongs | -0.185 |
unic | -0.178 |
dark | -0.175 |
Token | there |
Token ID | 612 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
Son | +0.005 |
distress | +0.005 |
No | +0.005 |
Dad | +0.004 |
Be | +0.004 |
Neg logit contributions | |
ophone | -0.005 |
Jupiter | -0.004 |
belongs | -0.004 |
unic | -0.004 |
endant | -0.004 |
Token | was |
Token ID | 373 |
Feature activation | -8.84e-01 |
Pos logit contributions | |
Jupiter | +0.045 |
ophone | +0.044 |
belongs | +0.044 |
unic | +0.040 |
dark | +0.039 |
Neg logit contributions | |
Son | -0.080 |
distress | -0.078 |
Dad | -0.077 |
No | -0.077 |
iver | -0.076 |
Token | a |
Token ID | 257 |
Feature activation | -6.17e-02 |
Pos logit contributions | |
Jupiter | +0.045 |
ophone | +0.044 |
unic | +0.042 |
belongs | +0.042 |
dark | +0.041 |
Neg logit contributions | |
Son | -0.043 |
distress | -0.042 |
No | -0.040 |
Dad | -0.040 |
iver | -0.039 |
Token | little |
Token ID | 1310 |
Feature activation | +4.03e+00 |
Pos logit contributions | |
Jupiter | +0.003 |
ophone | +0.003 |
unic | +0.003 |
belongs | +0.003 |
dark | +0.003 |
Neg logit contributions | |
distress | -0.003 |
Son | -0.003 |
No | -0.003 |
iver | -0.002 |
Dad | -0.002 |
Token | fish |
Token ID | 5916 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
distress | +0.166 |
Son | +0.162 |
iver | +0.159 |
No | +0.154 |
Marcus | +0.152 |
Neg logit contributions | |
Jupiter | -0.230 |
ophone | -0.223 |
unic | -0.221 |
belongs | -0.219 |
dark | -0.211 |
Token | named |
Token ID | 3706 |
Feature activation | -7.64e+00 |
Pos logit contributions | |
Son | +0.104 |
distress | +0.103 |
No | +0.102 |
Dad | +0.102 |
Be | +0.099 |
Neg logit contributions | |
ophone | -0.040 |
Jupiter | -0.039 |
belongs | -0.036 |
unic | -0.033 |
dark | -0.032 |
Token | Nem |
Token ID | 22547 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
ophone | +0.364 |
belongs | +0.359 |
lights | +0.351 |
endant | +0.348 |
existed | +0.342 |
Neg logit contributions | |
Son | -0.390 |
Dad | -0.382 |
sausage | -0.366 |
reel | -0.359 |
pack | -0.355 |
Token | o |
Token ID | 78 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
ophone | +0.072 |
Jupiter | +0.068 |
belongs | +0.066 |
dark | +0.063 |
unic | +0.062 |
Neg logit contributions | |
distress | -0.077 |
No | -0.077 |
Son | -0.075 |
Be | -0.074 |
Dad | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
Son | +0.093 |
distress | +0.091 |
Dad | +0.091 |
No | +0.089 |
iver | +0.087 |
Neg logit contributions | |
ophone | -0.070 |
Jupiter | -0.070 |
belongs | -0.068 |
endant | -0.063 |
unic | -0.063 |
Token | Nem |
Token ID | 22547 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
distress | +0.031 |
Son | +0.031 |
No | +0.031 |
Dad | +0.030 |
iver | +0.030 |
Neg logit contributions | |
Jupiter | -0.013 |
ophone | -0.013 |
belongs | -0.012 |
unic | -0.011 |
dark | -0.011 |
Token | o |
Token ID | 78 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
ophone | +0.070 |
Jupiter | +0.066 |
belongs | +0.065 |
dark | +0.062 |
endant | +0.059 |
Neg logit contributions | |
distress | -0.079 |
No | -0.078 |
Son | -0.076 |
Dad | -0.075 |
Be | -0.074 |
Token | there |
Token ID | 612 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
No | +0.000 |
distress | +0.000 |
Son | +0.000 |
Dad | +0.000 |
Be | +0.000 |
Neg logit contributions | |
ophone | -0.001 |
Jupiter | -0.000 |
belongs | -0.000 |
unic | -0.000 |
endant | -0.000 |
Token | was |
Token ID | 373 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
Jupiter | +0.050 |
ophone | +0.050 |
belongs | +0.049 |
unic | +0.045 |
dark | +0.044 |
Neg logit contributions | |
Son | -0.087 |
distress | -0.085 |
Dad | -0.084 |
No | -0.084 |
iver | -0.083 |
Token | a |
Token ID | 257 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
Jupiter | +0.052 |
ophone | +0.051 |
belongs | +0.048 |
unic | +0.048 |
dark | +0.047 |
Neg logit contributions | |
Son | -0.046 |
distress | -0.045 |
No | -0.044 |
Dad | -0.043 |
iver | -0.042 |
Token | little |
Token ID | 1310 |
Feature activation | +3.92e+00 |
Pos logit contributions | |
Jupiter | +0.007 |
ophone | +0.007 |
belongs | +0.007 |
unic | +0.007 |
dark | +0.006 |
Neg logit contributions | |
distress | -0.005 |
Son | -0.005 |
No | -0.005 |
iver | -0.005 |
Dad | -0.005 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
distress | +0.158 |
Son | +0.155 |
iver | +0.152 |
No | +0.147 |
Dad | +0.145 |
Neg logit contributions | |
Jupiter | -0.228 |
ophone | -0.220 |
unic | -0.217 |
belongs | -0.217 |
dark | -0.208 |
Token | named |
Token ID | 3706 |
Feature activation | -7.68e+00 |
Pos logit contributions | |
Dad | +0.110 |
M | +0.108 |
No | +0.106 |
Son | +0.104 |
distress | +0.104 |
Neg logit contributions | |
ophone | -0.078 |
Jupiter | -0.075 |
belongs | -0.074 |
lights | -0.073 |
existed | -0.072 |
Token | Amy |
Token ID | 14235 |
Feature activation | -5.99e-01 |
Pos logit contributions | |
ophone | +0.420 |
lights | +0.410 |
Jupiter | +0.389 |
belongs | +0.386 |
endant | +0.378 |
Neg logit contributions | |
M | -0.372 |
No | -0.354 |
Dad | -0.348 |
Baby | -0.338 |
Be | -0.337 |
Token | . |
Token ID | 13 |
Feature activation | +3.12e-01 |
Pos logit contributions | |
ophone | +0.030 |
Jupiter | +0.030 |
lights | +0.028 |
belongs | +0.027 |
endant | +0.027 |
Neg logit contributions | |
Dad | -0.030 |
distress | -0.029 |
Son | -0.029 |
No | -0.029 |
M | -0.028 |
Token | One |
Token ID | 1881 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
No | +0.016 |
distress | +0.016 |
Dad | +0.016 |
Be | +0.015 |
Son | +0.015 |
Neg logit contributions | |
ophone | -0.015 |
Jupiter | -0.015 |
belongs | -0.014 |
endant | -0.014 |
unic | -0.014 |
Token | day |
Token ID | 1110 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
ophone | +0.004 |
Jupiter | +0.004 |
belongs | +0.004 |
dark | +0.004 |
unic | +0.003 |
Neg logit contributions | |
Son | -0.007 |
distress | -0.007 |
No | -0.007 |
Dad | -0.007 |
Be | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +2.40e-02 |
Pos logit contributions | |
Jupiter | +0.007 |
ophone | +0.007 |
unic | +0.007 |
endant | +0.007 |
belongs | +0.007 |
Neg logit contributions | |
No | -0.006 |
Dad | -0.006 |
distress | -0.006 |
Be | -0.006 |
M | -0.006 |