Token | a |
Token ID | 257 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
fault | +0.042 |
Majesty | +0.039 |
endant | +0.037 |
jewelry | +0.035 |
collection | +0.035 |
Neg logit contributions | |
Run | -0.026 |
bel | -0.025 |
distance | -0.024 |
amed | -0.023 |
Mark | -0.023 |
Token | heart |
Token ID | 2612 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
fault | +0.027 |
jewelry | +0.023 |
endant | +0.023 |
Majesty | +0.023 |
ott | +0.021 |
Neg logit contributions | |
Run | -0.032 |
bel | -0.032 |
distance | -0.031 |
Uh | -0.030 |
he | -0.029 |
Token | - |
Token ID | 12 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
fault | +0.050 |
jewelry | +0.047 |
Majesty | +0.047 |
endant | +0.046 |
brace | +0.043 |
Neg logit contributions | |
bel | -0.037 |
he | -0.034 |
Run | -0.034 |
distance | -0.033 |
amed | -0.030 |
Token | shaped |
Token ID | 16760 |
Feature activation | +6.31e-02 |
Pos logit contributions | |
bel | +0.205 |
Run | +0.192 |
he | +0.179 |
Mark | +0.174 |
amed | +0.168 |
Neg logit contributions | |
fault | -0.281 |
Majesty | -0.262 |
jewelry | -0.260 |
collection | -0.255 |
endant | -0.250 |
Token | p |
Token ID | 279 |
Feature activation | +1.54e-01 |
Pos logit contributions | |
fault | +0.078 |
endant | +0.070 |
Majesty | +0.068 |
jewelry | +0.066 |
ott | +0.066 |
Neg logit contributions | |
Run | -0.143 |
bel | -0.142 |
distance | -0.140 |
he | -0.130 |
Uh | -0.130 |
Token | endant |
Token ID | 23048 |
Feature activation | +3.09e-01 |
Pos logit contributions | |
fault | +0.103 |
jewelry | +0.081 |
Majesty | +0.080 |
bean | +0.076 |
forgiveness | +0.075 |
Neg logit contributions | |
bel | -0.465 |
he | -0.412 |
amed | -0.409 |
aming | -0.399 |
Run | -0.396 |
Token | . |
Token ID | 13 |
Feature activation | -6.85e-04 |
Pos logit contributions | |
Majesty | +0.633 |
fault | +0.631 |
jewelry | +0.613 |
endant | +0.585 |
majesty | +0.552 |
Neg logit contributions | |
bel | -0.498 |
Run | -0.463 |
he | -0.458 |
distance | -0.418 |
amed | -0.405 |
Token |
|
Token ID | 198 |
Feature activation | -1.45e-02 |
Pos logit contributions | |
Run | +0.001 |
bel | +0.001 |
Quick | +0.001 |
Mark | +0.001 |
he | +0.001 |
Neg logit contributions | |
jewelry | -0.001 |
fault | -0.001 |
endant | -0.001 |
necklace | -0.001 |
Majesty | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.52e-02 |
Pos logit contributions | |
Run | +0.027 |
bel | +0.026 |
Uh | +0.024 |
distance | +0.024 |
Quick | +0.023 |
Neg logit contributions | |
jewelry | -0.026 |
endant | -0.024 |
fault | -0.024 |
necklace | -0.023 |
brace | -0.023 |
Token | L |
Token ID | 43 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
bel | +0.023 |
Run | +0.021 |
distance | +0.021 |
amer | +0.019 |
umbling | +0.018 |
Neg logit contributions | |
jewelry | -0.031 |
necklace | -0.028 |
sewing | -0.028 |
brace | -0.028 |
endant | -0.028 |
Token | ily |
Token ID | 813 |
Feature activation | +2.85e-02 |
Pos logit contributions | |
jewelry | +0.030 |
fault | +0.029 |
endant | +0.029 |
Majesty | +0.028 |
collection | +0.028 |
Neg logit contributions | |
Run | -0.024 |
he | -0.023 |
bel | -0.023 |
Quick | -0.022 |
Mark | -0.021 |
Token | who |
Token ID | 508 |
Feature activation | +8.15e-02 |
Pos logit contributions | |
Run | +0.067 |
bel | +0.066 |
distance | +0.062 |
he | +0.058 |
Mark | +0.057 |
Neg logit contributions | |
fault | -0.130 |
endant | -0.120 |
jewelry | -0.119 |
Majesty | -0.119 |
brace | -0.112 |
Token | was |
Token ID | 373 |
Feature activation | -5.63e-03 |
Pos logit contributions | |
jewelry | +0.164 |
fault | +0.156 |
endant | +0.154 |
necklace | +0.147 |
Majesty | +0.144 |
Neg logit contributions | |
Run | -0.140 |
bel | -0.134 |
Quick | -0.126 |
Uh | -0.123 |
distance | -0.123 |
Token | looking |
Token ID | 2045 |
Feature activation | +5.68e-02 |
Pos logit contributions | |
Run | +0.008 |
bel | +0.008 |
Quick | +0.007 |
distance | +0.007 |
he | +0.007 |
Neg logit contributions | |
jewelry | -0.013 |
endant | -0.012 |
fault | -0.012 |
necklace | -0.012 |
brace | -0.012 |
Token | for |
Token ID | 329 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
jewelry | +0.109 |
fault | +0.103 |
endant | +0.101 |
necklace | +0.097 |
Majesty | +0.096 |
Neg logit contributions | |
Run | -0.105 |
bel | -0.099 |
Quick | -0.095 |
Uh | -0.093 |
Mark | -0.092 |
Token | some |
Token ID | 617 |
Feature activation | +3.38e-02 |
Pos logit contributions | |
distance | +0.140 |
Run | +0.138 |
bel | +0.120 |
Quick | +0.114 |
Mark | +0.113 |
Neg logit contributions | |
fault | -0.247 |
endant | -0.238 |
Majesty | -0.225 |
Acc | -0.204 |
ccoli | -0.204 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
fault | +0.057 |
endant | +0.056 |
Majesty | +0.055 |
jewelry | +0.053 |
brace | +0.049 |
Neg logit contributions | |
Run | -0.066 |
bel | -0.063 |
distance | -0.061 |
Quick | -0.058 |
he | -0.057 |
Token | . |
Token ID | 13 |
Feature activation | +4.14e-03 |
Pos logit contributions | |
Majesty | +0.670 |
fault | +0.659 |
endant | +0.621 |
jewelry | +0.611 |
Acc | +0.604 |
Neg logit contributions | |
he | -0.457 |
bel | -0.446 |
distance | -0.405 |
Run | -0.398 |
amer | -0.371 |
Token | She |
Token ID | 1375 |
Feature activation | +6.04e-02 |
Pos logit contributions | |
fault | +0.009 |
jewelry | +0.008 |
endant | +0.008 |
necklace | +0.008 |
brace | +0.008 |
Neg logit contributions | |
Run | -0.007 |
bel | -0.006 |
Quick | -0.006 |
distance | -0.006 |
he | -0.006 |
Token | searched |
Token ID | 16499 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
fault | +0.137 |
jewelry | +0.131 |
endant | +0.128 |
Majesty | +0.122 |
necklace | +0.119 |
Neg logit contributions | |
Run | -0.090 |
bel | -0.089 |
Quick | -0.083 |
distance | -0.078 |
Uh | -0.077 |
Token | high |
Token ID | 1029 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
jewelry | +0.160 |
necklace | +0.142 |
fault | +0.140 |
endant | +0.137 |
collection | +0.137 |
Neg logit contributions | |
Run | -0.167 |
Quick | -0.154 |
Uh | -0.153 |
bel | -0.153 |
Mark | -0.148 |
Token | and |
Token ID | 290 |
Feature activation | -4.09e-03 |
Pos logit contributions | |
Run | +0.062 |
bel | +0.059 |
Quick | +0.055 |
distance | +0.054 |
he | +0.054 |
Neg logit contributions | |
jewelry | -0.081 |
fault | -0.079 |
endant | -0.078 |
necklace | -0.074 |
Majesty | -0.074 |
Token | where |
Token ID | 810 |
Feature activation | +6.27e-02 |
Pos logit contributions | |
bel | +0.193 |
Run | +0.191 |
distance | +0.175 |
he | +0.173 |
Mark | +0.169 |
Neg logit contributions | |
jewelry | -0.203 |
endant | -0.199 |
fault | -0.190 |
necklace | -0.183 |
Majesty | -0.180 |
Token | he |
Token ID | 339 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
jewelry | +0.122 |
endant | +0.119 |
fault | +0.118 |
Majesty | +0.111 |
necklace | +0.110 |
Neg logit contributions | |
Run | -0.109 |
bel | -0.107 |
distance | -0.097 |
Quick | -0.096 |
Mark | -0.095 |
Token | got |
Token ID | 1392 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Run | +0.022 |
bel | +0.022 |
Quick | +0.020 |
distance | +0.019 |
he | +0.019 |
Neg logit contributions | |
jewelry | -0.028 |
fault | -0.027 |
endant | -0.027 |
necklace | -0.025 |
Majesty | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -3.30e-02 |
Pos logit contributions | |
Run | +0.176 |
bel | +0.171 |
distance | +0.159 |
Quick | +0.155 |
he | +0.154 |
Neg logit contributions | |
jewelry | -0.198 |
fault | -0.191 |
endant | -0.190 |
Majesty | -0.181 |
necklace | -0.180 |
Token | little |
Token ID | 1310 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
Run | +0.065 |
bel | +0.063 |
distance | +0.060 |
Quick | +0.058 |
he | +0.058 |
Neg logit contributions | |
jewelry | -0.055 |
endant | -0.054 |
fault | -0.053 |
Majesty | -0.050 |
brace | -0.049 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.07e-01 |
Pos logit contributions | |
endant | +0.046 |
jewelry | +0.046 |
fault | +0.046 |
Majesty | +0.042 |
brace | +0.042 |
Neg logit contributions | |
Run | -0.048 |
bel | -0.046 |
distance | -0.045 |
Quick | -0.042 |
Uh | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -3.06e-03 |
Pos logit contributions | |
jewelry | +0.570 |
fault | +0.568 |
endant | +0.550 |
Majesty | +0.540 |
brace | +0.505 |
Neg logit contributions | |
bel | -0.541 |
Run | -0.533 |
he | -0.507 |
distance | -0.492 |
roads | -0.470 |
Token |
|
Token ID | 198 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
Run | +0.005 |
bel | +0.005 |
Quick | +0.005 |
Uh | +0.005 |
distance | +0.005 |
Neg logit contributions | |
jewelry | -0.006 |
fault | -0.006 |
endant | -0.006 |
Majesty | -0.005 |
necklace | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
bel | +0.041 |
Run | +0.039 |
Uh | +0.037 |
distance | +0.036 |
Quick | +0.034 |
Neg logit contributions | |
jewelry | -0.032 |
fault | -0.032 |
endant | -0.030 |
Majesty | -0.029 |
Ros | -0.028 |
Token | Joe |
Token ID | 19585 |
Feature activation | -9.63e-02 |
Pos logit contributions | |
bel | +0.031 |
Run | +0.026 |
distance | +0.026 |
he | +0.025 |
Uh | +0.024 |
Neg logit contributions | |
jewelry | -0.041 |
Majesty | -0.039 |
ott | -0.038 |
endant | -0.038 |
fault | -0.038 |
Token | smiled |
Token ID | 13541 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
Run | +0.173 |
bel | +0.166 |
Quick | +0.157 |
Uh | +0.152 |
distance | +0.152 |
Neg logit contributions | |
jewelry | -0.185 |
fault | -0.178 |
endant | -0.175 |
necklace | -0.166 |
Majesty | -0.164 |
Token | was |
Token ID | 373 |
Feature activation | +6.22e-04 |
Pos logit contributions | |
fault | +0.087 |
jewelry | +0.086 |
endant | +0.083 |
Majesty | +0.081 |
brace | +0.078 |
Neg logit contributions | |
Run | -0.053 |
bel | -0.052 |
Quick | -0.048 |
distance | -0.047 |
he | -0.047 |
Token | , |
Token ID | 11 |
Feature activation | +5.60e-02 |
Pos logit contributions | |
jewelry | +0.001 |
fault | +0.001 |
endant | +0.001 |
necklace | +0.001 |
Majesty | +0.001 |
Neg logit contributions | |
Run | -0.001 |
bel | -0.001 |
Quick | -0.001 |
Uh | -0.001 |
distance | -0.001 |
Token | lots |
Token ID | 6041 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
jewelry | +0.107 |
fault | +0.105 |
endant | +0.104 |
Majesty | +0.098 |
necklace | +0.096 |
Neg logit contributions | |
Run | -0.099 |
bel | -0.096 |
distance | -0.089 |
Quick | -0.088 |
Uh | -0.086 |
Token | of |
Token ID | 286 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
fault | +0.089 |
jewelry | +0.088 |
endant | +0.087 |
Majesty | +0.086 |
necklace | +0.083 |
Neg logit contributions | |
Run | -0.025 |
bel | -0.022 |
he | -0.020 |
distance | -0.019 |
Quick | -0.019 |
Token | beautiful |
Token ID | 4950 |
Feature activation | +8.95e-02 |
Pos logit contributions | |
Run | +0.202 |
bel | +0.193 |
Quick | +0.182 |
distance | +0.178 |
he | +0.178 |
Neg logit contributions | |
fault | -0.190 |
endant | -0.185 |
jewelry | -0.182 |
Majesty | -0.177 |
necklace | -0.169 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
fault | +0.182 |
endant | +0.167 |
Majesty | +0.158 |
jewelry | +0.156 |
necklace | +0.149 |
Neg logit contributions | |
Run | -0.149 |
bel | -0.149 |
distance | -0.134 |
Quick | -0.133 |
Uh | -0.126 |
Token | . |
Token ID | 13 |
Feature activation | +2.13e-03 |
Pos logit contributions | |
fault | +0.633 |
Majesty | +0.622 |
endant | +0.574 |
jewelry | +0.571 |
Acc | +0.549 |
Neg logit contributions | |
bel | -0.477 |
he | -0.467 |
Run | -0.449 |
distance | -0.413 |
amer | -0.393 |
Token | There |
Token ID | 1318 |
Feature activation | +5.27e-02 |
Pos logit contributions | |
jewelry | +0.004 |
fault | +0.003 |
endant | +0.003 |
necklace | +0.003 |
Majesty | +0.003 |
Neg logit contributions | |
Run | -0.004 |
bel | -0.004 |
Quick | -0.004 |
Uh | -0.004 |
distance | -0.004 |
Token | were |
Token ID | 547 |
Feature activation | -5.70e-02 |
Pos logit contributions | |
jewelry | +0.135 |
fault | +0.132 |
endant | +0.131 |
Majesty | +0.127 |
necklace | +0.123 |
Neg logit contributions | |
Run | -0.060 |
bel | -0.057 |
he | -0.052 |
Quick | -0.050 |
distance | -0.050 |
Token | neck |
Token ID | 7393 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
bel | +0.093 |
Run | +0.088 |
distance | +0.083 |
Quick | +0.080 |
Mark | +0.080 |
Neg logit contributions | |
fault | -0.112 |
endant | -0.107 |
jewelry | -0.105 |
Majesty | -0.102 |
necklace | -0.097 |
Token | l |
Token ID | 75 |
Feature activation | -9.27e-03 |
Pos logit contributions | |
fault | +0.081 |
jewelry | +0.075 |
Majesty | +0.073 |
endant | +0.072 |
necklace | +0.068 |
Neg logit contributions | |
bel | -0.060 |
Run | -0.055 |
he | -0.053 |
distance | -0.051 |
Mark | -0.049 |
Token | store |
Token ID | 3650 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
distance | +0.054 |
bel | +0.053 |
Run | +0.053 |
Quick | +0.050 |
he | +0.049 |
Neg logit contributions | |
fault | -0.051 |
endant | -0.050 |
jewelry | -0.048 |
brace | -0.048 |
Majesty | -0.047 |
Token | and |
Token ID | 290 |
Feature activation | -6.57e-03 |
Pos logit contributions | |
jewelry | +0.232 |
fault | +0.221 |
endant | +0.219 |
necklace | +0.206 |
Majesty | +0.205 |
Neg logit contributions | |
Run | -0.235 |
bel | -0.230 |
Quick | -0.214 |
distance | -0.211 |
he | -0.210 |
Token | gave |
Token ID | 2921 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
Run | +0.011 |
bel | +0.011 |
Quick | +0.010 |
distance | +0.010 |
he | +0.010 |
Neg logit contributions | |
jewelry | -0.013 |
fault | -0.013 |
endant | -0.013 |
Majesty | -0.012 |
necklace | -0.012 |
Token | him |
Token ID | 683 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
jewelry | +0.219 |
endant | +0.215 |
fault | +0.214 |
Majesty | +0.200 |
necklace | +0.200 |
Neg logit contributions | |
Run | -0.163 |
bel | -0.158 |
Quick | -0.144 |
he | -0.144 |
Mark | -0.142 |
Token | the |
Token ID | 262 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
Run | +0.045 |
bel | +0.044 |
Quick | +0.040 |
distance | +0.039 |
he | +0.039 |
Neg logit contributions | |
jewelry | -0.062 |
fault | -0.061 |
endant | -0.061 |
Majesty | -0.058 |
necklace | -0.057 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.05e-01 |
Pos logit contributions | |
Run | +0.052 |
bel | +0.051 |
distance | +0.049 |
Quick | +0.047 |
he | +0.046 |
Neg logit contributions | |
endant | -0.039 |
jewelry | -0.038 |
fault | -0.037 |
Majesty | -0.036 |
brace | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.80e-03 |
Pos logit contributions | |
fault | +0.637 |
Majesty | +0.630 |
endant | +0.577 |
jewelry | +0.575 |
Acc | +0.549 |
Neg logit contributions | |
bel | -0.488 |
he | -0.451 |
distance | -0.421 |
Run | -0.393 |
amer | -0.368 |
Token | The |
Token ID | 383 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
Run | +0.003 |
bel | +0.003 |
Quick | +0.003 |
distance | +0.003 |
Uh | +0.003 |
Neg logit contributions | |
jewelry | -0.004 |
fault | -0.004 |
endant | -0.004 |
necklace | -0.003 |
Majesty | -0.003 |
Token | little |
Token ID | 1310 |
Feature activation | +3.10e-02 |
Pos logit contributions | |
fault | +0.087 |
endant | +0.085 |
jewelry | +0.080 |
Majesty | +0.077 |
brace | +0.076 |
Neg logit contributions | |
bel | -0.063 |
Run | -0.062 |
distance | -0.060 |
Quick | -0.058 |
Uh | -0.056 |
Token | boy |
Token ID | 2933 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
fault | +0.081 |
endant | +0.081 |
jewelry | +0.081 |
Majesty | +0.077 |
necklace | +0.075 |
Neg logit contributions | |
Run | -0.033 |
bel | -0.030 |
Quick | -0.027 |
distance | -0.027 |
Uh | -0.026 |
Token | thanked |
Token ID | 26280 |
Feature activation | +6.89e-02 |
Pos logit contributions | |
bel | +0.199 |
Run | +0.197 |
Quick | +0.180 |
he | +0.176 |
Mark | +0.172 |
Neg logit contributions | |
fault | -0.271 |
jewelry | -0.269 |
endant | -0.258 |
Majesty | -0.247 |
necklace | -0.241 |
Token | necklace |
Token ID | 38987 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
fault | +0.151 |
endant | +0.134 |
Majesty | +0.124 |
jewelry | +0.122 |
brace | +0.115 |
Neg logit contributions | |
Run | -0.163 |
distance | -0.162 |
bel | -0.159 |
he | -0.152 |
Uh | -0.147 |
Token | with |
Token ID | 351 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
fault | +0.432 |
jewelry | +0.418 |
Majesty | +0.409 |
endant | +0.408 |
brace | +0.375 |
Neg logit contributions | |
bel | -0.293 |
Run | -0.292 |
he | -0.281 |
distance | -0.264 |
Mark | -0.252 |
Token | a |
Token ID | 257 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
fault | +0.036 |
Majesty | +0.033 |
endant | +0.032 |
jewelry | +0.031 |
collection | +0.030 |
Neg logit contributions | |
Run | -0.027 |
bel | -0.027 |
distance | -0.025 |
Mark | -0.024 |
he | -0.024 |
Token | cross |
Token ID | 3272 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
fault | +0.027 |
jewelry | +0.024 |
endant | +0.023 |
Majesty | +0.023 |
ott | +0.021 |
Neg logit contributions | |
Run | -0.031 |
bel | -0.031 |
distance | -0.031 |
Uh | -0.030 |
he | -0.029 |
Token | p |
Token ID | 279 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
Run | +0.343 |
bel | +0.328 |
Quick | +0.309 |
Uh | +0.299 |
distance | +0.297 |
Neg logit contributions | |
jewelry | -0.429 |
fault | -0.410 |
endant | -0.406 |
necklace | -0.389 |
Majesty | -0.379 |
Token | endant |
Token ID | 23048 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
fault | +0.098 |
jewelry | +0.097 |
Majesty | +0.073 |
necklace | +0.069 |
brace | +0.068 |
Neg logit contributions | |
bel | -0.451 |
Run | -0.434 |
he | -0.418 |
Quick | -0.409 |
distance | -0.407 |
Token | . |
Token ID | 13 |
Feature activation | -5.88e-03 |
Pos logit contributions | |
fault | +0.607 |
jewelry | +0.598 |
Majesty | +0.594 |
endant | +0.575 |
necklace | +0.529 |
Neg logit contributions | |
bel | -0.500 |
Run | -0.479 |
he | -0.459 |
distance | -0.433 |
Mark | -0.411 |
Token | He |
Token ID | 679 |
Feature activation | -4.76e-02 |
Pos logit contributions | |
Run | +0.006 |
bel | +0.006 |
Quick | +0.005 |
Mark | +0.005 |
he | +0.005 |
Neg logit contributions | |
jewelry | -0.015 |
fault | -0.015 |
endant | -0.015 |
necklace | -0.014 |
brace | -0.014 |
Token | recognized |
Token ID | 8018 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
Run | +0.074 |
bel | +0.073 |
Quick | +0.068 |
Uh | +0.063 |
Mark | +0.063 |
Neg logit contributions | |
fault | -0.109 |
jewelry | -0.099 |
endant | -0.094 |
Majesty | -0.092 |
necklace | -0.087 |
Token | it |
Token ID | 340 |
Feature activation | +3.52e-02 |
Pos logit contributions | |
fault | +0.247 |
endant | +0.233 |
Majesty | +0.222 |
jewelry | +0.219 |
collection | +0.209 |
Neg logit contributions | |
Run | -0.139 |
bel | -0.135 |
he | -0.126 |
Mark | -0.119 |
distance | -0.116 |
Token | as |
Token ID | 355 |
Feature activation | -1.85e-01 |
Pos logit contributions | |
fault | +0.081 |
endant | +0.080 |
Majesty | +0.078 |
jewelry | +0.077 |
majesty | +0.072 |
Neg logit contributions | |
bel | -0.045 |
Run | -0.042 |
distance | -0.042 |
he | -0.041 |
Mark | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | -6.48e-03 |
Pos logit contributions | |
jewelry | +0.072 |
fault | +0.071 |
endant | +0.071 |
Majesty | +0.069 |
brace | +0.066 |
Neg logit contributions | |
Run | -0.055 |
bel | -0.052 |
he | -0.051 |
Quick | -0.050 |
distance | -0.049 |
Token | difficult |
Token ID | 2408 |
Feature activation | +2.18e-02 |
Pos logit contributions | |
Run | +0.011 |
bel | +0.010 |
distance | +0.010 |
Uh | +0.010 |
Quick | +0.009 |
Neg logit contributions | |
endant | -0.012 |
jewelry | -0.012 |
Majesty | -0.012 |
fault | -0.012 |
necklace | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +3.00e-02 |
Pos logit contributions | |
endant | +0.047 |
jewelry | +0.043 |
necklace | +0.042 |
Majesty | +0.042 |
fault | +0.040 |
Neg logit contributions | |
Run | -0.033 |
bel | -0.033 |
ahead | -0.029 |
distance | -0.029 |
across | -0.029 |
Token | surrender |
Token ID | 16908 |
Feature activation | +1.69e-02 |
Pos logit contributions | |
endant | +0.068 |
fault | +0.067 |
jewelry | +0.063 |
Majesty | +0.061 |
cand | +0.058 |
Neg logit contributions | |
Run | -0.041 |
he | -0.041 |
distance | -0.041 |
bel | -0.038 |
Mark | -0.036 |
Token | her |
Token ID | 607 |
Feature activation | +7.11e-02 |
Pos logit contributions | |
jewelry | +0.031 |
endant | +0.030 |
fault | +0.029 |
Majesty | +0.029 |
necklace | +0.028 |
Neg logit contributions | |
Run | -0.030 |
bel | -0.030 |
distance | -0.028 |
he | -0.027 |
Quick | -0.027 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.03e-01 |
Pos logit contributions | |
fault | +0.076 |
Majesty | +0.075 |
endant | +0.063 |
ccoli | +0.062 |
majesty | +0.059 |
Neg logit contributions | |
bel | -0.170 |
distance | -0.141 |
across | -0.136 |
he | -0.132 |
ahead | -0.131 |
Token | , |
Token ID | 11 |
Feature activation | +5.36e-02 |
Pos logit contributions | |
endant | +0.533 |
jewelry | +0.520 |
Majesty | +0.519 |
fault | +0.519 |
brace | +0.462 |
Neg logit contributions | |
bel | -0.555 |
he | -0.505 |
distance | -0.493 |
Run | -0.484 |
roads | -0.455 |
Token | she |
Token ID | 673 |
Feature activation | +6.72e-02 |
Pos logit contributions | |
endant | +0.100 |
fault | +0.098 |
jewelry | +0.093 |
Majesty | +0.088 |
necklace | +0.086 |
Neg logit contributions | |
Run | -0.097 |
bel | -0.095 |
distance | -0.087 |
Quick | -0.085 |
Mark | -0.085 |
Token | now |
Token ID | 783 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
jewelry | +0.148 |
fault | +0.142 |
endant | +0.141 |
brace | +0.132 |
Majesty | +0.132 |
Neg logit contributions | |
Run | -0.096 |
bel | -0.095 |
he | -0.089 |
Quick | -0.087 |
distance | -0.085 |
Token | knew |
Token ID | 2993 |
Feature activation | +1.67e-02 |
Pos logit contributions | |
endant | +0.038 |
jewelry | +0.037 |
fault | +0.037 |
brace | +0.034 |
cand | +0.034 |
Neg logit contributions | |
distance | -0.021 |
Run | -0.020 |
bel | -0.020 |
he | -0.019 |
For | -0.018 |
Token | that |
Token ID | 326 |
Feature activation | +1.08e-01 |
Pos logit contributions | |
endant | +0.035 |
Majesty | +0.033 |
jewelry | +0.031 |
fault | +0.031 |
necklace | +0.030 |
Neg logit contributions | |
bel | -0.021 |
ahead | -0.021 |
Run | -0.020 |
across | -0.020 |
For | -0.020 |
Token | had |
Token ID | 550 |
Feature activation | +1.07e-02 |
Pos logit contributions | |
jewelry | +0.108 |
fault | +0.108 |
endant | +0.103 |
Majesty | +0.098 |
necklace | +0.097 |
Neg logit contributions | |
Run | -0.081 |
bel | -0.077 |
Quick | -0.073 |
Uh | -0.070 |
Mark | -0.069 |
Token | been |
Token ID | 587 |
Feature activation | +1.25e-02 |
Pos logit contributions | |
jewelry | +0.022 |
fault | +0.021 |
endant | +0.021 |
Majesty | +0.020 |
necklace | +0.020 |
Neg logit contributions | |
Run | -0.017 |
bel | -0.017 |
Quick | -0.016 |
distance | -0.016 |
Mark | -0.016 |
Token | living |
Token ID | 2877 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
jewelry | +0.023 |
fault | +0.023 |
endant | +0.023 |
necklace | +0.021 |
Majesty | +0.021 |
Neg logit contributions | |
Run | -0.023 |
bel | -0.022 |
Quick | -0.020 |
distance | -0.020 |
Mark | -0.020 |
Token | in |
Token ID | 287 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
jewelry | +0.329 |
endant | +0.319 |
fault | +0.319 |
necklace | +0.303 |
Majesty | +0.301 |
Neg logit contributions | |
Run | -0.201 |
bel | -0.195 |
Quick | -0.178 |
distance | -0.175 |
he | -0.172 |
Token | the |
Token ID | 262 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
Run | +0.046 |
bel | +0.043 |
distance | +0.041 |
Quick | +0.040 |
Mark | +0.040 |
Neg logit contributions | |
jewelry | -0.058 |
endant | -0.057 |
fault | -0.056 |
Majesty | -0.054 |
necklace | -0.053 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
Run | +0.052 |
bel | +0.050 |
distance | +0.049 |
Quick | +0.049 |
Uh | +0.048 |
Neg logit contributions | |
jewelry | -0.051 |
endant | -0.050 |
fault | -0.049 |
brace | -0.046 |
necklace | -0.046 |
Token | box |
Token ID | 3091 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
jewelry | +0.600 |
fault | +0.596 |
endant | +0.587 |
Majesty | +0.559 |
brace | +0.539 |
Neg logit contributions | |
bel | -0.486 |
Run | -0.486 |
he | -0.482 |
distance | -0.466 |
roads | -0.433 |
Token | and |
Token ID | 290 |
Feature activation | -9.53e-03 |
Pos logit contributions | |
jewelry | +0.206 |
fault | +0.200 |
endant | +0.198 |
Majesty | +0.190 |
brace | +0.185 |
Neg logit contributions | |
bel | -0.164 |
Run | -0.163 |
he | -0.156 |
distance | -0.154 |
Quick | -0.142 |
Token | had |
Token ID | 550 |
Feature activation | +1.10e-02 |
Pos logit contributions | |
bel | +0.015 |
Run | +0.015 |
distance | +0.014 |
Quick | +0.014 |
Mark | +0.013 |
Neg logit contributions | |
fault | -0.019 |
jewelry | -0.019 |
endant | -0.019 |
Majesty | -0.017 |
necklace | -0.017 |
Token | gotten |
Token ID | 7891 |
Feature activation | -3.31e-02 |
Pos logit contributions | |
fault | +0.023 |
endant | +0.023 |
jewelry | +0.022 |
Majesty | +0.022 |
collection | +0.020 |
Neg logit contributions | |
bel | -0.015 |
Run | -0.015 |
distance | -0.015 |
Mark | -0.014 |
Quick | -0.014 |
Token | in |
Token ID | 287 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
bel | +0.052 |
Run | +0.050 |
distance | +0.046 |
Quick | +0.043 |
Uh | +0.042 |
Neg logit contributions | |
fault | -0.069 |
Majesty | -0.067 |
endant | -0.067 |
jewelry | -0.066 |
necklace | -0.062 |
Token | old |
Token ID | 1468 |
Feature activation | +4.68e-02 |
Pos logit contributions | |
endant | +0.088 |
jewelry | +0.086 |
fault | +0.086 |
brace | +0.079 |
Majesty | +0.078 |
Neg logit contributions | |
Run | -0.071 |
distance | -0.065 |
bel | -0.064 |
he | -0.061 |
Quick | -0.060 |
Token | owl |
Token ID | 39610 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
fault | +0.107 |
endant | +0.107 |
jewelry | +0.104 |
Majesty | +0.097 |
brace | +0.096 |
Neg logit contributions | |
Run | -0.067 |
bel | -0.061 |
distance | -0.058 |
Quick | -0.056 |
Mark | -0.054 |
Token | , |
Token ID | 11 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
fault | +0.239 |
jewelry | +0.223 |
endant | +0.220 |
Majesty | +0.218 |
brace | +0.200 |
Neg logit contributions | |
Run | -0.196 |
bel | -0.181 |
distance | -0.168 |
Mark | -0.166 |
roads | -0.165 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.09e-02 |
Pos logit contributions | |
jewelry | +0.118 |
fault | +0.114 |
endant | +0.112 |
necklace | +0.106 |
Majesty | +0.105 |
Neg logit contributions | |
Run | -0.099 |
bel | -0.095 |
Quick | -0.089 |
distance | -0.086 |
Uh | -0.086 |
Token | the |
Token ID | 262 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
Run | +0.124 |
bel | +0.119 |
Quick | +0.107 |
Mark | +0.104 |
he | +0.104 |
Neg logit contributions | |
fault | -0.176 |
endant | -0.175 |
jewelry | -0.166 |
Majesty | -0.164 |
necklace | -0.161 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
distance | +0.042 |
Run | +0.041 |
bel | +0.040 |
Quick | +0.038 |
Uh | +0.037 |
Neg logit contributions | |
endant | -0.038 |
fault | -0.037 |
jewelry | -0.036 |
Majesty | -0.035 |
brace | -0.033 |
Token | and |
Token ID | 290 |
Feature activation | -2.22e-03 |
Pos logit contributions | |
fault | +0.677 |
Majesty | +0.653 |
jewelry | +0.641 |
endant | +0.640 |
majesty | +0.590 |
Neg logit contributions | |
Run | -0.427 |
bel | -0.424 |
he | -0.421 |
distance | -0.395 |
roads | -0.371 |
Token | told |
Token ID | 1297 |
Feature activation | +8.29e-02 |
Pos logit contributions | |
Run | +0.004 |
bel | +0.004 |
Quick | +0.003 |
distance | +0.003 |
Mark | +0.003 |
Neg logit contributions | |
fault | -0.005 |
jewelry | -0.004 |
endant | -0.004 |
Majesty | -0.004 |
necklace | -0.004 |
Token | Benny |
Token ID | 44275 |
Feature activation | +1.92e-01 |
Pos logit contributions | |
endant | +0.189 |
fault | +0.171 |
jewelry | +0.162 |
Majesty | +0.158 |
necklace | +0.156 |
Neg logit contributions | |
Run | -0.129 |
bel | -0.122 |
Mark | -0.113 |
Quick | -0.109 |
he | -0.107 |
Token | it |
Token ID | 340 |
Feature activation | +3.80e-02 |
Pos logit contributions | |
endant | +0.392 |
jewelry | +0.379 |
Majesty | +0.367 |
fault | +0.355 |
brace | +0.350 |
Neg logit contributions | |
bel | -0.297 |
Run | -0.278 |
distance | -0.268 |
he | -0.267 |
Mark | -0.259 |
Token | might |
Token ID | 1244 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
endant | +0.098 |
Majesty | +0.096 |
jewelry | +0.094 |
fault | +0.093 |
majesty | +0.089 |
Neg logit contributions | |
Run | -0.040 |
bel | -0.036 |
Mark | -0.034 |
distance | -0.031 |
he | -0.030 |
Token | had |
Token ID | 550 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
fault | +0.137 |
jewelry | +0.133 |
endant | +0.129 |
Majesty | +0.123 |
necklace | +0.121 |
Neg logit contributions | |
Run | -0.084 |
bel | -0.080 |
Quick | -0.076 |
Mark | -0.071 |
he | -0.070 |
Token | a |
Token ID | 257 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
endant | +0.024 |
fault | +0.023 |
jewelry | +0.023 |
necklace | +0.022 |
Majesty | +0.022 |
Neg logit contributions | |
Run | -0.017 |
bel | -0.017 |
Quick | -0.016 |
distance | -0.016 |
Mark | -0.015 |
Token | box |
Token ID | 3091 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
endant | +0.028 |
fault | +0.027 |
jewelry | +0.026 |
Majesty | +0.025 |
brace | +0.025 |
Neg logit contributions | |
bel | -0.023 |
distance | -0.023 |
Run | -0.021 |
Uh | -0.021 |
Quick | -0.021 |
Token | of |
Token ID | 286 |
Feature activation | -1.10e-01 |
Pos logit contributions | |
fault | +0.229 |
jewelry | +0.229 |
endant | +0.220 |
Majesty | +0.217 |
necklace | +0.209 |
Neg logit contributions | |
bel | -0.138 |
Run | -0.133 |
he | -0.123 |
distance | -0.120 |
Quick | -0.115 |
Token | special |
Token ID | 2041 |
Feature activation | +1.16e-01 |
Pos logit contributions | |
Run | +0.208 |
bel | +0.207 |
distance | +0.192 |
he | +0.189 |
Quick | +0.186 |
Neg logit contributions | |
fault | -0.185 |
Majesty | -0.179 |
endant | -0.173 |
jewelry | -0.149 |
collection | -0.146 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.02e-01 |
Pos logit contributions | |
fault | +0.198 |
endant | +0.184 |
Majesty | +0.182 |
ott | +0.164 |
jewelry | +0.163 |
Neg logit contributions | |
bel | -0.221 |
Run | -0.208 |
distance | -0.198 |
Quick | -0.186 |
roads | -0.185 |
Token | that |
Token ID | 326 |
Feature activation | +1.05e-01 |
Pos logit contributions | |
fault | +0.672 |
Majesty | +0.654 |
endant | +0.601 |
jewelry | +0.585 |
Acc | +0.580 |
Neg logit contributions | |
bel | -0.445 |
he | -0.443 |
Run | -0.383 |
distance | -0.375 |
amer | -0.356 |
Token | she |
Token ID | 673 |
Feature activation | +7.07e-02 |
Pos logit contributions | |
fault | +0.187 |
endant | +0.180 |
Majesty | +0.175 |
jewelry | +0.173 |
necklace | +0.155 |
Neg logit contributions | |
bel | -0.200 |
Run | -0.190 |
distance | -0.187 |
he | -0.186 |
Mark | -0.178 |
Token | kept |
Token ID | 4030 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
fault | +0.158 |
jewelry | +0.149 |
endant | +0.146 |
Majesty | +0.140 |
necklace | +0.136 |
Neg logit contributions | |
bel | -0.110 |
Run | -0.102 |
he | -0.095 |
Quick | -0.095 |
Mark | -0.090 |
Token | safe |
Token ID | 3338 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
bel | +0.030 |
distance | +0.025 |
across | +0.024 |
Run | +0.024 |
he | +0.024 |
Neg logit contributions | |
fault | -0.049 |
endant | -0.047 |
Majesty | -0.047 |
majesty | -0.043 |
jewelry | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -5.64e-03 |
Pos logit contributions | |
endant | +0.073 |
jewelry | +0.072 |
fault | +0.072 |
Majesty | +0.069 |
necklace | +0.067 |
Neg logit contributions | |
bel | -0.053 |
Run | -0.050 |
distance | -0.049 |
he | -0.047 |
roads | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -1.15e-03 |
Pos logit contributions | |
jewelry | +0.071 |
fault | +0.067 |
endant | +0.066 |
necklace | +0.064 |
brace | +0.062 |
Neg logit contributions | |
Run | -0.064 |
bel | -0.061 |
Quick | -0.058 |
Uh | -0.057 |
Mark | -0.056 |
Token | poured |
Token ID | 19036 |
Feature activation | +5.22e-03 |
Pos logit contributions | |
Run | +0.002 |
bel | +0.002 |
Quick | +0.002 |
distance | +0.002 |
Uh | +0.001 |
Neg logit contributions | |
jewelry | -0.003 |
fault | -0.002 |
endant | -0.002 |
Majesty | -0.002 |
necklace | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | +3.98e-02 |
Pos logit contributions | |
jewelry | +0.010 |
fault | +0.009 |
endant | +0.009 |
necklace | +0.009 |
Majesty | +0.009 |
Neg logit contributions | |
Run | -0.010 |
bel | -0.009 |
Quick | -0.009 |
Uh | -0.009 |
distance | -0.009 |
Token | on |
Token ID | 319 |
Feature activation | -2.93e-02 |
Pos logit contributions | |
jewelry | +0.097 |
fault | +0.095 |
endant | +0.094 |
Majesty | +0.091 |
necklace | +0.089 |
Neg logit contributions | |
Run | -0.050 |
bel | -0.049 |
Quick | -0.043 |
he | -0.042 |
distance | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -2.51e-02 |
Pos logit contributions | |
Run | +0.042 |
bel | +0.040 |
he | +0.038 |
distance | +0.037 |
Quick | +0.037 |
Neg logit contributions | |
jewelry | -0.065 |
endant | -0.065 |
fault | -0.064 |
Majesty | -0.061 |
necklace | -0.060 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
Run | +0.046 |
bel | +0.044 |
distance | +0.042 |
Quick | +0.042 |
he | +0.041 |
Neg logit contributions | |
jewelry | -0.046 |
fault | -0.045 |
endant | -0.045 |
Majesty | -0.041 |
brace | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | +1.86e-03 |
Pos logit contributions | |
fault | +0.567 |
jewelry | +0.560 |
endant | +0.544 |
Majesty | +0.540 |
brace | +0.487 |
Neg logit contributions | |
bel | -0.521 |
Run | -0.507 |
he | -0.499 |
distance | -0.487 |
roads | -0.455 |
Token | As |
Token ID | 1081 |
Feature activation | -5.78e-02 |
Pos logit contributions | |
jewelry | +0.004 |
endant | +0.004 |
fault | +0.004 |
necklace | +0.004 |
brace | +0.004 |
Neg logit contributions | |
Run | -0.003 |
bel | -0.002 |
Quick | -0.002 |
he | -0.002 |
Mark | -0.002 |
Token | she |
Token ID | 673 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
Run | +0.121 |
bel | +0.118 |
distance | +0.112 |
Quick | +0.110 |
Mark | +0.110 |
Neg logit contributions | |
endant | -0.092 |
fault | -0.092 |
jewelry | -0.091 |
Majesty | -0.083 |
necklace | -0.080 |
Token | poured |
Token ID | 19036 |
Feature activation | +5.42e-03 |
Pos logit contributions | |
fault | +0.150 |
jewelry | +0.149 |
endant | +0.141 |
Majesty | +0.137 |
necklace | +0.134 |
Neg logit contributions | |
Run | -0.115 |
bel | -0.111 |
Quick | -0.104 |
Uh | -0.099 |
he | -0.099 |
Token | , |
Token ID | 11 |
Feature activation | +5.82e-02 |
Pos logit contributions | |
jewelry | +0.011 |
fault | +0.011 |
endant | +0.011 |
Majesty | +0.010 |
necklace | +0.010 |
Neg logit contributions | |
Run | -0.009 |
bel | -0.009 |
distance | -0.008 |
he | -0.008 |
Quick | -0.008 |
Token | safe |
Token ID | 3338 |
Feature activation | +3.48e-02 |
Pos logit contributions | |
bel | +0.030 |
distance | +0.025 |
across | +0.024 |
Run | +0.024 |
he | +0.024 |
Neg logit contributions | |
fault | -0.049 |
endant | -0.047 |
Majesty | -0.047 |
majesty | -0.043 |
jewelry | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -5.64e-03 |
Pos logit contributions | |
endant | +0.073 |
jewelry | +0.072 |
fault | +0.072 |
Majesty | +0.069 |
necklace | +0.067 |
Neg logit contributions | |
bel | -0.053 |
Run | -0.050 |
distance | -0.049 |
he | -0.047 |
roads | -0.043 |
Token | secret |
Token ID | 3200 |
Feature activation | +1.26e-01 |
Pos logit contributions | |
bel | +0.010 |
distance | +0.008 |
Run | +0.008 |
he | +0.008 |
roads | +0.008 |
Neg logit contributions | |
fault | -0.012 |
endant | -0.011 |
jewelry | -0.011 |
Majesty | -0.011 |
bean | -0.010 |
Token | . |
Token ID | 13 |
Feature activation | +3.03e-03 |
Pos logit contributions | |
fault | +0.264 |
endant | +0.255 |
Majesty | +0.252 |
jewelry | +0.245 |
brace | +0.224 |
Neg logit contributions | |
bel | -0.197 |
distance | -0.167 |
he | -0.161 |
Run | -0.156 |
roads | -0.153 |
Token | The |
Token ID | 383 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
jewelry | +0.007 |
fault | +0.007 |
endant | +0.007 |
necklace | +0.007 |
brace | +0.007 |
Neg logit contributions | |
Run | -0.004 |
he | -0.004 |
Quick | -0.004 |
Mark | -0.003 |
bel | -0.003 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
fault | +0.090 |
endant | +0.083 |
jewelry | +0.077 |
Majesty | +0.077 |
brace | +0.076 |
Neg logit contributions | |
Run | -0.059 |
bel | -0.058 |
distance | -0.057 |
Quick | -0.055 |
he | -0.054 |
Token | was |
Token ID | 373 |
Feature activation | -1.57e-03 |
Pos logit contributions | |
fault | +0.667 |
jewelry | +0.626 |
Majesty | +0.615 |
endant | +0.611 |
brace | +0.561 |
Neg logit contributions | |
bel | -0.462 |
Run | -0.427 |
he | -0.422 |
distance | -0.391 |
Quick | -0.387 |
Token | so |
Token ID | 523 |
Feature activation | -1.17e-02 |
Pos logit contributions | |
bel | +0.003 |
Run | +0.002 |
distance | +0.002 |
he | +0.002 |
Uh | +0.002 |
Neg logit contributions | |
jewelry | -0.003 |
fault | -0.003 |
endant | -0.003 |
Majesty | -0.003 |
necklace | -0.003 |
Token | nice |
Token ID | 3621 |
Feature activation | +4.19e-02 |
Pos logit contributions | |
bel | +0.021 |
Run | +0.019 |
he | +0.019 |
distance | +0.017 |
amed | +0.017 |
Neg logit contributions | |
fault | -0.024 |
jewelry | -0.022 |
Majesty | -0.021 |
endant | -0.021 |
brace | -0.020 |
Token | that |
Token ID | 326 |
Feature activation | +1.06e-01 |
Pos logit contributions | |
endant | +0.092 |
Majesty | +0.091 |
jewelry | +0.091 |
fault | +0.088 |
brace | +0.085 |
Neg logit contributions | |
bel | -0.055 |
Run | -0.054 |
he | -0.052 |
distance | -0.052 |
ahead | -0.049 |
Token | the |
Token ID | 262 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
endant | +0.200 |
fault | +0.196 |
jewelry | +0.191 |
Majesty | +0.181 |
necklace | +0.171 |
Neg logit contributions | |
Run | -0.187 |
bel | -0.186 |
distance | -0.180 |
he | -0.176 |
Mark | -0.173 |
Token | man |
Token ID | 582 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
bel | +0.003 |
Run | +0.003 |
he | +0.003 |
distance | +0.003 |
Quick | +0.002 |
Neg logit contributions | |
jewelry | -0.009 |
endant | -0.009 |
fault | -0.009 |
Majesty | -0.008 |
brace | -0.008 |
Token | for |
Token ID | 329 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
bel | +0.194 |
Run | +0.194 |
he | +0.182 |
distance | +0.181 |
Mark | +0.167 |
Neg logit contributions | |
jewelry | -0.302 |
endant | -0.301 |
fault | -0.297 |
Majesty | -0.281 |
necklace | -0.273 |
Token | giving |
Token ID | 3501 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
Run | +0.186 |
bel | +0.179 |
Quick | +0.168 |
distance | +0.164 |
Mark | +0.163 |
Neg logit contributions | |
jewelry | -0.197 |
fault | -0.191 |
endant | -0.189 |
necklace | -0.177 |
Majesty | -0.175 |
Token | him |
Token ID | 683 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
endant | +0.169 |
jewelry | +0.168 |
fault | +0.167 |
Majesty | +0.157 |
necklace | +0.156 |
Neg logit contributions | |
Run | -0.109 |
bel | -0.104 |
distance | -0.096 |
he | -0.095 |
Quick | -0.094 |
Token | the |
Token ID | 262 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
bel | +0.042 |
Run | +0.041 |
distance | +0.039 |
he | +0.037 |
Quick | +0.035 |
Neg logit contributions | |
fault | -0.069 |
endant | -0.068 |
Majesty | -0.066 |
jewelry | -0.066 |
brace | -0.062 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
distance | +0.063 |
bel | +0.060 |
he | +0.055 |
middle | +0.053 |
across | +0.053 |
Neg logit contributions | |
fault | -0.026 |
Majesty | -0.024 |
ccoli | -0.024 |
ience | -0.023 |
oppy | -0.022 |
Token | and |
Token ID | 290 |
Feature activation | -8.22e-03 |
Pos logit contributions | |
fault | +0.611 |
Majesty | +0.610 |
endant | +0.583 |
jewelry | +0.574 |
brace | +0.531 |
Neg logit contributions | |
bel | -0.475 |
he | -0.449 |
distance | -0.429 |
Run | -0.396 |
amer | -0.370 |
Token | said |
Token ID | 531 |
Feature activation | +2.76e-02 |
Pos logit contributions | |
Run | +0.013 |
bel | +0.013 |
distance | +0.011 |
Quick | +0.011 |
Mark | +0.011 |
Neg logit contributions | |
fault | -0.017 |
endant | -0.017 |
jewelry | -0.017 |
Majesty | -0.016 |
necklace | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +5.42e-02 |
Pos logit contributions | |
endant | +0.057 |
jewelry | +0.055 |
fault | +0.053 |
necklace | +0.049 |
brace | +0.047 |
Neg logit contributions | |
bel | -0.047 |
Run | -0.045 |
he | -0.040 |
Mark | -0.040 |
Quick | -0.038 |
Token | " |
Token ID | 366 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
jewelry | +0.108 |
fault | +0.104 |
endant | +0.103 |
necklace | +0.097 |
Majesty | +0.095 |
Neg logit contributions | |
Run | -0.094 |
bel | -0.091 |
Quick | -0.085 |
distance | -0.082 |
Uh | -0.082 |
Token | I |
Token ID | 40 |
Feature activation | +4.71e-02 |
Pos logit contributions | |
Run | +0.088 |
bel | +0.086 |
Quick | +0.073 |
distance | +0.073 |
roads | +0.071 |
Neg logit contributions | |
jewelry | -0.197 |
endant | -0.185 |
fault | -0.183 |
necklace | -0.181 |
brace | -0.179 |
Token | surprised |
Token ID | 6655 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
Run | +0.021 |
bel | +0.019 |
Quick | +0.018 |
distance | +0.018 |
Uh | +0.018 |
Neg logit contributions | |
jewelry | -0.028 |
fault | -0.027 |
endant | -0.027 |
necklace | -0.025 |
Majesty | -0.025 |
Token | that |
Token ID | 326 |
Feature activation | +1.07e-01 |
Pos logit contributions | |
Run | +0.148 |
bel | +0.139 |
distance | +0.130 |
he | +0.129 |
Quick | +0.129 |
Neg logit contributions | |
jewelry | -0.147 |
fault | -0.143 |
endant | -0.143 |
Majesty | -0.136 |
necklace | -0.134 |
Token | she |
Token ID | 673 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
endant | +0.179 |
fault | +0.176 |
jewelry | +0.176 |
Majesty | +0.160 |
necklace | +0.156 |
Neg logit contributions | |
Run | -0.219 |
bel | -0.210 |
distance | -0.200 |
Mark | -0.197 |
he | -0.197 |
Token | dropped |
Token ID | 5710 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
jewelry | +0.134 |
endant | +0.126 |
fault | +0.124 |
necklace | +0.122 |
brace | +0.118 |
Neg logit contributions | |
Run | -0.121 |
bel | -0.117 |
Quick | -0.110 |
distance | -0.108 |
he | -0.108 |
Token | the |
Token ID | 262 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
Run | +0.034 |
bel | +0.031 |
distance | +0.030 |
Quick | +0.029 |
he | +0.029 |
Neg logit contributions | |
jewelry | -0.046 |
endant | -0.046 |
fault | -0.045 |
Majesty | -0.044 |
necklace | -0.043 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
bel | +0.068 |
distance | +0.067 |
Run | +0.066 |
he | +0.063 |
Uh | +0.062 |
Neg logit contributions | |
fault | -0.038 |
Majesty | -0.034 |
endant | -0.030 |
Acc | -0.028 |
ellery | -0.027 |
Token | box |
Token ID | 3091 |
Feature activation | +1.01e-01 |
Pos logit contributions | |
fault | +0.646 |
Majesty | +0.634 |
jewelry | +0.632 |
endant | +0.622 |
majesty | +0.566 |
Neg logit contributions | |
Run | -0.436 |
bel | -0.430 |
he | -0.425 |
distance | -0.411 |
roads | -0.365 |
Token | ! |
Token ID | 0 |
Feature activation | -4.44e-02 |
Pos logit contributions | |
jewelry | +0.211 |
fault | +0.205 |
Majesty | +0.201 |
endant | +0.201 |
necklace | +0.187 |
Neg logit contributions | |
Run | -0.158 |
bel | -0.157 |
he | -0.148 |
distance | -0.147 |
Quick | -0.136 |
Token |
|
Token ID | 198 |
Feature activation | -2.14e-02 |
Pos logit contributions | |
Run | +0.076 |
bel | +0.071 |
Quick | +0.068 |
Uh | +0.066 |
Mark | +0.066 |
Neg logit contributions | |
jewelry | -0.091 |
fault | -0.087 |
endant | -0.086 |
necklace | -0.083 |
Majesty | -0.081 |
Token |
|
Token ID | 198 |
Feature activation | -2.18e-02 |
Pos logit contributions | |
bel | +0.041 |
Run | +0.040 |
Uh | +0.037 |
distance | +0.036 |
amer | +0.036 |
Neg logit contributions | |
jewelry | -0.036 |
endant | -0.033 |
fault | -0.033 |
Ros | -0.031 |
sewing | -0.031 |
Token | When |
Token ID | 2215 |
Feature activation | -6.37e-02 |
Pos logit contributions | |
bel | +0.030 |
Run | +0.027 |
distance | +0.026 |
amer | +0.026 |
mer | +0.024 |
Neg logit contributions | |
jewelry | -0.046 |
sewing | -0.042 |
brace | -0.042 |
makeup | -0.041 |
ott | -0.041 |
Token | She |
Token ID | 1375 |
Feature activation | +5.80e-02 |
Pos logit contributions | |
jewelry | +0.002 |
fault | +0.002 |
endant | +0.002 |
necklace | +0.002 |
Majesty | +0.002 |
Neg logit contributions | |
Run | -0.002 |
bel | -0.001 |
Quick | -0.001 |
Mark | -0.001 |
Uh | -0.001 |
Token | found |
Token ID | 1043 |
Feature activation | +1.64e-02 |
Pos logit contributions | |
jewelry | +0.125 |
fault | +0.122 |
endant | +0.118 |
necklace | +0.113 |
Majesty | +0.112 |
Neg logit contributions | |
Run | -0.092 |
bel | -0.088 |
Quick | -0.083 |
Uh | -0.079 |
distance | -0.079 |
Token | a |
Token ID | 257 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
fault | +0.035 |
jewelry | +0.035 |
endant | +0.035 |
Majesty | +0.033 |
collection | +0.032 |
Neg logit contributions | |
Run | -0.025 |
bel | -0.024 |
Quick | -0.022 |
distance | -0.021 |
Mark | -0.021 |
Token | piece |
Token ID | 3704 |
Feature activation | +5.47e-03 |
Pos logit contributions | |
fault | +0.032 |
jewelry | +0.031 |
endant | +0.030 |
Majesty | +0.029 |
brace | +0.027 |
Neg logit contributions | |
Run | -0.033 |
bel | -0.033 |
distance | -0.032 |
Uh | -0.031 |
Quick | -0.031 |
Token | of |
Token ID | 286 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
jewelry | +0.014 |
fault | +0.014 |
endant | +0.014 |
necklace | +0.013 |
Majesty | +0.013 |
Neg logit contributions | |
Run | -0.006 |
bel | -0.006 |
Quick | -0.005 |
Uh | -0.005 |
distance | -0.005 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
Run | +0.221 |
distance | +0.203 |
bel | +0.201 |
Uh | +0.192 |
Quick | +0.190 |
Neg logit contributions | |
fault | -0.172 |
Majesty | -0.161 |
endant | -0.154 |
jewelry | -0.138 |
collection | -0.135 |
Token | ! |
Token ID | 0 |
Feature activation | -4.47e-02 |
Pos logit contributions | |
fault | +0.680 |
Majesty | +0.659 |
jewelry | +0.655 |
endant | +0.643 |
majesty | +0.595 |
Neg logit contributions | |
Run | -0.408 |
bel | -0.393 |
he | -0.391 |
distance | -0.364 |
roads | -0.340 |
Token | It |
Token ID | 632 |
Feature activation | +3.53e-02 |
Pos logit contributions | |
Run | +0.080 |
bel | +0.075 |
Quick | +0.072 |
he | +0.070 |
Uh | +0.070 |
Neg logit contributions | |
jewelry | -0.088 |
fault | -0.085 |
endant | -0.083 |
necklace | -0.080 |
Majesty | -0.079 |
Token | was |
Token ID | 373 |
Feature activation | -2.10e-03 |
Pos logit contributions | |
fault | +0.080 |
jewelry | +0.076 |
endant | +0.074 |
Majesty | +0.074 |
necklace | +0.067 |
Neg logit contributions | |
Run | -0.052 |
bel | -0.052 |
Quick | -0.047 |
Uh | -0.046 |
Mark | -0.046 |
Token | dull |
Token ID | 19222 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
bel | +0.003 |
Run | +0.003 |
distance | +0.003 |
Uh | +0.003 |
amed | +0.003 |
Neg logit contributions | |
fault | -0.004 |
jewelry | -0.004 |
Majesty | -0.004 |
endant | -0.004 |
necklace | -0.004 |
Token | and |
Token ID | 290 |
Feature activation | -7.92e-03 |
Pos logit contributions | |
Majesty | +0.089 |
endant | +0.087 |
fault | +0.086 |
jewelry | +0.083 |
majesty | +0.081 |
Neg logit contributions | |
bel | -0.047 |
Run | -0.045 |
distance | -0.043 |
he | -0.043 |
ahead | -0.038 |
Token | € |
Token ID | 26391 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
Run | +0.014 |
bel | +0.013 |
Quick | +0.012 |
he | +0.012 |
distance | +0.012 |
Neg logit contributions | |
jewelry | -0.020 |
fault | -0.019 |
endant | -0.019 |
necklace | -0.018 |
Majesty | -0.018 |
Token | ™ |
Token ID | 8151 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
Run | +0.080 |
bel | +0.074 |
Quick | +0.072 |
Uh | +0.072 |
distance | +0.069 |
Neg logit contributions | |
jewelry | -0.078 |
fault | -0.077 |
endant | -0.075 |
Majesty | -0.070 |
necklace | -0.070 |
Token | t |
Token ID | 83 |
Feature activation | -7.26e-02 |
Pos logit contributions | |
Run | +0.027 |
bel | +0.025 |
Quick | +0.024 |
Uh | +0.024 |
distance | +0.021 |
Neg logit contributions | |
fault | -0.039 |
jewelry | -0.038 |
endant | -0.037 |
necklace | -0.034 |
collection | -0.034 |
Token | find |
Token ID | 1064 |
Feature activation | +5.98e-02 |
Pos logit contributions | |
Run | +0.118 |
bel | +0.113 |
Quick | +0.106 |
distance | +0.105 |
he | +0.104 |
Neg logit contributions | |
jewelry | -0.153 |
fault | -0.147 |
endant | -0.146 |
necklace | -0.138 |
Majesty | -0.137 |
Token | the |
Token ID | 262 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
fault | +0.116 |
endant | +0.113 |
Majesty | +0.110 |
jewelry | +0.104 |
necklace | +0.100 |
Neg logit contributions | |
Run | -0.096 |
distance | -0.095 |
bel | -0.093 |
Mark | -0.087 |
roads | -0.086 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
distance | +0.070 |
bel | +0.063 |
he | +0.059 |
across | +0.058 |
Uh | +0.058 |
Neg logit contributions | |
fault | -0.032 |
ccoli | -0.029 |
Acc | -0.029 |
Majesty | -0.028 |
endant | -0.028 |
Token | . |
Token ID | 13 |
Feature activation | -1.87e-03 |
Pos logit contributions | |
Majesty | +0.625 |
fault | +0.622 |
endant | +0.593 |
jewelry | +0.570 |
Acc | +0.566 |
Neg logit contributions | |
he | -0.454 |
bel | -0.443 |
distance | -0.426 |
Run | -0.415 |
roads | -0.391 |
Token |
|
Token ID | 198 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
Run | +0.003 |
bel | +0.003 |
Quick | +0.003 |
Uh | +0.003 |
distance | +0.003 |
Neg logit contributions | |
jewelry | -0.004 |
fault | -0.004 |
endant | -0.004 |
necklace | -0.004 |
brace | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
Run | +0.036 |
bel | +0.036 |
Uh | +0.032 |
distance | +0.032 |
Quick | +0.032 |
Neg logit contributions | |
jewelry | -0.037 |
fault | -0.035 |
endant | -0.035 |
necklace | -0.033 |
brace | -0.032 |
Token | Finally |
Token ID | 11158 |
Feature activation | -4.54e-02 |
Pos logit contributions | |
bel | +0.031 |
Run | +0.030 |
amer | +0.027 |
distance | +0.027 |
Uh | +0.025 |
Neg logit contributions | |
jewelry | -0.045 |
endant | -0.041 |
necklace | -0.040 |
brace | -0.039 |
sewing | -0.039 |
Token | , |
Token ID | 11 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
Run | +0.075 |
bel | +0.065 |
distance | +0.063 |
Quick | +0.062 |
he | +0.060 |
Neg logit contributions | |
fault | -0.101 |
endant | -0.096 |
jewelry | -0.087 |
Majesty | -0.086 |
brace | -0.082 |
Token | never |
Token ID | 1239 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
jewelry | +0.126 |
fault | +0.124 |
endant | +0.121 |
Majesty | +0.119 |
brace | +0.109 |
Neg logit contributions | |
Run | -0.086 |
bel | -0.079 |
Quick | -0.079 |
he | -0.073 |
distance | -0.071 |
Token | picked |
Token ID | 6497 |
Feature activation | +1.21e-01 |
Pos logit contributions | |
jewelry | +0.184 |
fault | +0.178 |
endant | +0.176 |
Majesty | +0.172 |
necklace | +0.166 |
Neg logit contributions | |
Run | -0.125 |
bel | -0.115 |
Quick | -0.107 |
distance | -0.107 |
Uh | -0.104 |
Token | her |
Token ID | 607 |
Feature activation | +7.05e-02 |
Pos logit contributions | |
endant | +0.278 |
Majesty | +0.258 |
fault | +0.250 |
jewelry | +0.248 |
collection | +0.248 |
Neg logit contributions | |
Run | -0.149 |
he | -0.126 |
distance | -0.124 |
across | -0.123 |
bel | -0.123 |
Token | mom |
Token ID | 1995 |
Feature activation | -3.85e-02 |
Pos logit contributions | |
endant | +0.155 |
Majesty | +0.150 |
fault | +0.149 |
ott | +0.144 |
majesty | +0.141 |
Neg logit contributions | |
bel | -0.086 |
distance | -0.076 |
Run | -0.072 |
below | -0.065 |
he | -0.065 |
Token | 's |
Token ID | 338 |
Feature activation | +1.49e-01 |
Pos logit contributions | |
Run | +0.079 |
bel | +0.078 |
he | +0.074 |
distance | +0.073 |
Quick | +0.072 |
Neg logit contributions | |
jewelry | -0.062 |
fault | -0.060 |
endant | -0.059 |
Majesty | -0.056 |
necklace | -0.054 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
endant | +0.244 |
fault | +0.225 |
Majesty | +0.223 |
jewelry | +0.214 |
Ros | +0.210 |
Neg logit contributions | |
bel | -0.288 |
distance | -0.282 |
Run | -0.269 |
Quick | -0.243 |
roads | -0.241 |
Token | again |
Token ID | 757 |
Feature activation | +1.18e-01 |
Pos logit contributions | |
Majesty | +0.510 |
fault | +0.503 |
endant | +0.490 |
jewelry | +0.473 |
majesty | +0.454 |
Neg logit contributions | |
he | -0.576 |
bel | -0.555 |
distance | -0.522 |
Run | -0.500 |
roads | -0.474 |
Token | without |
Token ID | 1231 |
Feature activation | +3.69e-02 |
Pos logit contributions | |
jewelry | +0.233 |
endant | +0.233 |
fault | +0.230 |
Majesty | +0.216 |
necklace | +0.211 |
Neg logit contributions | |
Run | -0.201 |
bel | -0.196 |
distance | -0.188 |
he | -0.180 |
Quick | -0.177 |
Token | asking |
Token ID | 4737 |
Feature activation | +8.35e-03 |
Pos logit contributions | |
endant | +0.077 |
fault | +0.070 |
Majesty | +0.069 |
collection | +0.068 |
jewelry | +0.066 |
Neg logit contributions | |
Run | -0.052 |
distance | -0.047 |
crashing | -0.045 |
aming | -0.044 |
Help | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | -7.55e-04 |
Pos logit contributions | |
endant | +0.018 |
fault | +0.017 |
Majesty | +0.017 |
jewelry | +0.017 |
necklace | +0.016 |
Neg logit contributions | |
Run | -0.012 |
bel | -0.012 |
distance | -0.011 |
he | -0.011 |
Mark | -0.011 |
Token | She |
Token ID | 1375 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
Run | +0.001 |
Mark | +0.001 |
Quick | +0.001 |
distance | +0.001 |
bel | +0.001 |
Neg logit contributions | |
jewelry | -0.002 |
necklace | -0.002 |
brace | -0.002 |
endant | -0.002 |
collection | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.59e-02 |
Pos logit contributions | |
jewelry | +0.004 |
fault | +0.004 |
endant | +0.004 |
necklace | +0.004 |
Majesty | +0.004 |
Neg logit contributions | |
Run | -0.003 |
bel | -0.002 |
Quick | -0.002 |
he | -0.002 |
Mark | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | -1.62e-02 |
Pos logit contributions | |
Run | +0.029 |
bel | +0.028 |
Uh | +0.026 |
distance | +0.026 |
Quick | +0.025 |
Neg logit contributions | |
jewelry | -0.029 |
fault | -0.027 |
endant | -0.027 |
necklace | -0.026 |
brace | -0.025 |
Token | She |
Token ID | 3347 |
Feature activation | +5.03e-03 |
Pos logit contributions | |
bel | +0.021 |
Run | +0.021 |
distance | +0.019 |
Uh | +0.018 |
amer | +0.017 |
Neg logit contributions | |
jewelry | -0.036 |
brace | -0.033 |
necklace | -0.033 |
endant | -0.033 |
bracelet | -0.032 |
Token | put |
Token ID | 1234 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
fault | +0.011 |
jewelry | +0.011 |
endant | +0.011 |
Majesty | +0.010 |
necklace | +0.010 |
Neg logit contributions | |
Run | -0.008 |
bel | -0.007 |
Quick | -0.007 |
distance | -0.006 |
Uh | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -2.80e-02 |
Pos logit contributions | |
jewelry | +0.058 |
fault | +0.058 |
endant | +0.056 |
Majesty | +0.054 |
necklace | +0.053 |
Neg logit contributions | |
Run | -0.041 |
bel | -0.040 |
Quick | -0.037 |
he | -0.036 |
distance | -0.036 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
bel | +0.074 |
distance | +0.073 |
he | +0.070 |
Uh | +0.067 |
amed | +0.067 |
Neg logit contributions | |
fault | -0.031 |
Majesty | -0.022 |
ccoli | -0.019 |
endant | -0.019 |
onna | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -2.81e-02 |
Pos logit contributions | |
fault | +0.723 |
jewelry | +0.722 |
endant | +0.701 |
Majesty | +0.680 |
necklace | +0.659 |
Neg logit contributions | |
bel | -0.368 |
Run | -0.367 |
he | -0.320 |
Quick | -0.319 |
distance | -0.319 |
Token | a |
Token ID | 257 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
Run | +0.049 |
bel | +0.047 |
distance | +0.046 |
he | +0.045 |
Quick | +0.044 |
Neg logit contributions | |
endant | -0.054 |
fault | -0.053 |
jewelry | -0.053 |
Majesty | -0.051 |
necklace | -0.048 |
Token | safe |
Token ID | 3338 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
fault | +0.033 |
endant | +0.031 |
jewelry | +0.031 |
Majesty | +0.029 |
brace | +0.028 |
Neg logit contributions | |
Run | -0.032 |
bel | -0.032 |
distance | -0.032 |
Quick | -0.030 |
he | -0.030 |
Token | place |
Token ID | 1295 |
Feature activation | -2.90e-02 |
Pos logit contributions | |
fault | +0.075 |
endant | +0.074 |
jewelry | +0.074 |
Majesty | +0.070 |
brace | +0.068 |
Neg logit contributions | |
distance | -0.040 |
bel | -0.039 |
Run | -0.037 |
he | -0.034 |
Quick | -0.032 |
Token | and |
Token ID | 290 |
Feature activation | -4.65e-03 |
Pos logit contributions | |
bel | +0.051 |
Run | +0.050 |
distance | +0.047 |
he | +0.046 |
Quick | +0.044 |
Neg logit contributions | |
jewelry | -0.055 |
fault | -0.053 |
endant | -0.053 |
Majesty | -0.051 |
necklace | -0.049 |
Token | not |
Token ID | 407 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
jewelry | +0.025 |
endant | +0.025 |
fault | +0.024 |
Majesty | +0.023 |
necklace | +0.022 |
Neg logit contributions | |
Run | -0.019 |
bel | -0.019 |
Mark | -0.017 |
Quick | -0.017 |
distance | -0.017 |
Token | played |
Token ID | 2826 |
Feature activation | +8.61e-02 |
Pos logit contributions | |
jewelry | +0.024 |
fault | +0.023 |
endant | +0.023 |
Majesty | +0.023 |
necklace | +0.022 |
Neg logit contributions | |
Run | -0.018 |
bel | -0.017 |
distance | -0.016 |
Quick | -0.015 |
Mark | -0.015 |
Token | with |
Token ID | 351 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
jewelry | +0.182 |
fault | +0.173 |
endant | +0.166 |
necklace | +0.165 |
brace | +0.160 |
Neg logit contributions | |
Run | -0.143 |
bel | -0.134 |
Quick | -0.130 |
Uh | -0.129 |
distance | -0.125 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.36e-02 |
Pos logit contributions | |
endant | +0.041 |
jewelry | +0.041 |
fault | +0.040 |
Majesty | +0.039 |
necklace | +0.038 |
Neg logit contributions | |
Run | -0.027 |
bel | -0.025 |
Mark | -0.023 |
distance | -0.023 |
Quick | -0.023 |
Token | 's |
Token ID | 338 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
Run | +0.088 |
bel | +0.085 |
Quick | +0.082 |
Uh | +0.079 |
Mark | +0.077 |
Neg logit contributions | |
jewelry | -0.076 |
fault | -0.072 |
endant | -0.069 |
necklace | -0.067 |
brace | -0.065 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
endant | +0.239 |
jewelry | +0.234 |
fault | +0.231 |
Majesty | +0.221 |
brace | +0.206 |
Neg logit contributions | |
bel | -0.291 |
Run | -0.287 |
distance | -0.270 |
Quick | -0.260 |
he | -0.256 |
Token | . |
Token ID | 13 |
Feature activation | -1.65e-03 |
Pos logit contributions | |
endant | +0.563 |
jewelry | +0.559 |
Majesty | +0.556 |
fault | +0.553 |
brace | +0.497 |
Neg logit contributions | |
bel | -0.504 |
he | -0.496 |
Run | -0.467 |
distance | -0.462 |
amer | -0.424 |
Token | They |
Token ID | 1119 |
Feature activation | -5.55e-02 |
Pos logit contributions | |
Run | +0.002 |
bel | +0.002 |
Quick | +0.002 |
Uh | +0.002 |
Mark | +0.002 |
Neg logit contributions | |
jewelry | -0.004 |
fault | -0.004 |
endant | -0.004 |
necklace | -0.004 |
brace | -0.004 |
Token | wished |
Token ID | 16555 |
Feature activation | +1.30e-01 |
Pos logit contributions | |
Run | +0.107 |
bel | +0.101 |
Quick | +0.097 |
distance | +0.096 |
Uh | +0.094 |
Neg logit contributions | |
jewelry | -0.100 |
fault | -0.098 |
endant | -0.095 |
Majesty | -0.091 |
necklace | -0.088 |
Token | they |
Token ID | 484 |
Feature activation | -3.93e-02 |
Pos logit contributions | |
endant | +0.304 |
jewelry | +0.300 |
Majesty | +0.293 |
fault | +0.288 |
necklace | +0.276 |
Neg logit contributions | |
bel | -0.173 |
Run | -0.172 |
Mark | -0.152 |
distance | -0.145 |
Quick | -0.140 |
Token | had |
Token ID | 550 |
Feature activation | +1.49e-02 |
Pos logit contributions | |
Run | +0.067 |
bel | +0.064 |
Quick | +0.061 |
Uh | +0.059 |
distance | +0.058 |
Neg logit contributions | |
jewelry | -0.080 |
fault | -0.076 |
endant | -0.075 |
necklace | -0.072 |
brace | -0.070 |
Token | are |
Token ID | 389 |
Feature activation | -7.90e-03 |
Pos logit contributions | |
Run | +0.101 |
bel | +0.097 |
Quick | +0.091 |
Uh | +0.089 |
distance | +0.088 |
Neg logit contributions | |
jewelry | -0.112 |
fault | -0.107 |
endant | -0.106 |
necklace | -0.101 |
Majesty | -0.099 |
Token | happy |
Token ID | 3772 |
Feature activation | -2.13e-02 |
Pos logit contributions | |
Run | +0.015 |
bel | +0.014 |
Quick | +0.014 |
Uh | +0.013 |
he | +0.013 |
Neg logit contributions | |
jewelry | -0.015 |
fault | -0.014 |
endant | -0.014 |
necklace | -0.013 |
Majesty | -0.013 |
Token | with |
Token ID | 351 |
Feature activation | +1.48e-02 |
Pos logit contributions | |
Run | +0.035 |
bel | +0.034 |
distance | +0.030 |
Quick | +0.029 |
Mark | +0.029 |
Neg logit contributions | |
jewelry | -0.043 |
endant | -0.043 |
fault | -0.043 |
Majesty | -0.040 |
necklace | -0.039 |
Token | their |
Token ID | 511 |
Feature activation | -8.04e-02 |
Pos logit contributions | |
endant | +0.037 |
jewelry | +0.037 |
fault | +0.037 |
Majesty | +0.036 |
necklace | +0.035 |
Neg logit contributions | |
Run | -0.017 |
bel | -0.015 |
Mark | -0.014 |
distance | -0.014 |
Quick | -0.014 |
Token | wire |
Token ID | 6503 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
Run | +0.153 |
bel | +0.152 |
distance | +0.141 |
Quick | +0.140 |
he | +0.135 |
Neg logit contributions | |
jewelry | -0.134 |
endant | -0.131 |
fault | -0.129 |
Majesty | -0.125 |
brace | -0.119 |
Token | jewelry |
Token ID | 22634 |
Feature activation | +2.98e-01 |
Pos logit contributions | |
jewelry | +0.119 |
endant | +0.116 |
fault | +0.114 |
Majesty | +0.111 |
ott | +0.104 |
Neg logit contributions | |
bel | -0.113 |
Run | -0.109 |
distance | -0.101 |
Mark | -0.096 |
amer | -0.095 |
Token | . |
Token ID | 13 |
Feature activation | -9.75e-03 |
Pos logit contributions | |
Majesty | +0.581 |
fault | +0.580 |
jewelry | +0.577 |
endant | +0.564 |
Acc | +0.515 |
Neg logit contributions | |
bel | -0.474 |
he | -0.444 |
Run | -0.437 |
distance | -0.423 |
amer | -0.400 |
Token | They |
Token ID | 1119 |
Feature activation | -5.74e-02 |
Pos logit contributions | |
Run | +0.014 |
bel | +0.013 |
Quick | +0.012 |
Uh | +0.012 |
distance | +0.012 |
Neg logit contributions | |
jewelry | -0.023 |
fault | -0.022 |
endant | -0.021 |
necklace | -0.021 |
Majesty | -0.020 |
Token | put |
Token ID | 1234 |
Feature activation | +2.69e-02 |
Pos logit contributions | |
Run | +0.095 |
bel | +0.091 |
he | +0.087 |
distance | +0.084 |
Quick | +0.083 |
Neg logit contributions | |
jewelry | -0.116 |
fault | -0.114 |
endant | -0.111 |
Majesty | -0.105 |
necklace | -0.103 |
Token | them |
Token ID | 606 |
Feature activation | -7.10e-02 |
Pos logit contributions | |
jewelry | +0.048 |
fault | +0.047 |
endant | +0.047 |
Majesty | +0.045 |
necklace | +0.042 |
Neg logit contributions | |
Run | -0.050 |
bel | -0.047 |
distance | -0.045 |
he | -0.044 |
Quick | -0.044 |
Token | on |
Token ID | 319 |
Feature activation | -3.42e-02 |
Pos logit contributions | |
Run | +0.095 |
bel | +0.092 |
Quick | +0.083 |
distance | +0.081 |
he | +0.080 |
Neg logit contributions | |
jewelry | -0.167 |
fault | -0.163 |
endant | -0.161 |
Majesty | -0.154 |
necklace | -0.153 |
Token | John |
Token ID | 1757 |
Feature activation | -5.59e-02 |
Pos logit contributions | |
Run | +0.008 |
bel | +0.008 |
Quick | +0.008 |
Uh | +0.007 |
distance | +0.007 |
Neg logit contributions | |
jewelry | -0.009 |
fault | -0.009 |
endant | -0.009 |
Majesty | -0.008 |
necklace | -0.008 |
Token | began |
Token ID | 2540 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
Run | +0.092 |
bel | +0.089 |
Quick | +0.083 |
distance | +0.081 |
he | +0.081 |
Neg logit contributions | |
jewelry | -0.116 |
fault | -0.114 |
endant | -0.110 |
necklace | -0.103 |
Majesty | -0.102 |
Token | to |
Token ID | 284 |
Feature activation | +2.64e-02 |
Pos logit contributions | |
bel | +0.186 |
Run | +0.174 |
distance | +0.167 |
Quick | +0.158 |
Mark | +0.155 |
Neg logit contributions | |
endant | -0.222 |
jewelry | -0.213 |
fault | -0.210 |
necklace | -0.197 |
Majesty | -0.196 |
Token | make |
Token ID | 787 |
Feature activation | -2.35e-02 |
Pos logit contributions | |
fault | +0.055 |
endant | +0.055 |
jewelry | +0.054 |
Majesty | +0.049 |
necklace | +0.048 |
Neg logit contributions | |
Run | -0.042 |
bel | -0.041 |
distance | -0.039 |
he | -0.037 |
Quick | -0.037 |
Token | - |
Token ID | 12 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Run | +0.039 |
bel | +0.038 |
Quick | +0.035 |
Mark | +0.035 |
distance | +0.035 |
Neg logit contributions | |
jewelry | -0.047 |
endant | -0.046 |
fault | -0.045 |
necklace | -0.042 |
Majesty | -0.041 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
Run | +0.143 |
Uh | +0.114 |
Quick | +0.107 |
distance | +0.102 |
he | +0.099 |
Neg logit contributions | |
jewelry | -0.356 |
necklace | -0.331 |
endant | -0.326 |
brace | -0.321 |
Majesty | -0.318 |
Token | ieve |
Token ID | 12311 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
Run | +0.721 |
bel | +0.673 |
Quick | +0.665 |
Uh | +0.661 |
distance | +0.658 |
Neg logit contributions | |
jewelry | -0.505 |
fault | -0.452 |
endant | -0.443 |
necklace | -0.437 |
brace | -0.435 |
Token | he |
Token ID | 339 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
jewelry | +0.140 |
fault | +0.134 |
endant | +0.132 |
necklace | +0.126 |
Majesty | +0.124 |
Neg logit contributions | |
Run | -0.128 |
bel | -0.124 |
Quick | -0.116 |
distance | -0.113 |
Uh | -0.112 |
Token | was |
Token ID | 373 |
Feature activation | -5.52e-03 |
Pos logit contributions | |
Run | +0.019 |
bel | +0.019 |
Quick | +0.018 |
distance | +0.017 |
Uh | +0.017 |
Neg logit contributions | |
jewelry | -0.021 |
fault | -0.021 |
endant | -0.020 |
necklace | -0.019 |
Majesty | -0.019 |
Token | a |
Token ID | 257 |
Feature activation | +1.05e-02 |
Pos logit contributions | |
Run | +0.009 |
bel | +0.009 |
Quick | +0.009 |
distance | +0.008 |
Uh | +0.008 |
Neg logit contributions | |
jewelry | -0.011 |
fault | -0.011 |
endant | -0.010 |
necklace | -0.010 |
Majesty | -0.010 |
Token | hero |
Token ID | 4293 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
endant | +0.020 |
jewelry | +0.020 |
fault | +0.019 |
brace | +0.017 |
cand | +0.017 |
Neg logit contributions | |
bel | -0.018 |
distance | -0.017 |
Run | -0.017 |
Quick | -0.016 |
Ar | -0.016 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
Run | +0.014 |
bel | +0.013 |
Quick | +0.012 |
he | +0.012 |
distance | +0.012 |
Neg logit contributions | |
jewelry | -0.018 |
fault | -0.018 |
endant | -0.017 |
Majesty | -0.016 |
necklace | -0.016 |
Token | to |
Token ID | 284 |
Feature activation | +2.65e-02 |
Pos logit contributions | |
bel | +0.049 |
Run | +0.046 |
Quick | +0.042 |
Mark | +0.042 |
distance | +0.042 |
Neg logit contributions | |
endant | -0.057 |
fault | -0.054 |
jewelry | -0.053 |
Majesty | -0.052 |
necklace | -0.051 |
Token | play |
Token ID | 711 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
fault | +0.055 |
endant | +0.055 |
jewelry | +0.053 |
Majesty | +0.049 |
necklace | +0.048 |
Neg logit contributions | |
Run | -0.042 |
bel | -0.042 |
he | -0.039 |
distance | -0.037 |
Quick | -0.037 |
Token | make |
Token ID | 787 |
Feature activation | -2.47e-02 |
Pos logit contributions | |
jewelry | +0.187 |
fault | +0.183 |
endant | +0.181 |
necklace | +0.171 |
Majesty | +0.171 |
Neg logit contributions | |
Run | -0.120 |
bel | -0.115 |
Quick | -0.105 |
he | -0.102 |
distance | -0.101 |
Token | - |
Token ID | 12 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Run | +0.043 |
bel | +0.043 |
Quick | +0.039 |
he | +0.039 |
Mark | +0.038 |
Neg logit contributions | |
endant | -0.048 |
fault | -0.046 |
jewelry | -0.046 |
Majesty | -0.042 |
necklace | -0.042 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
Run | +0.137 |
Uh | +0.107 |
Quick | +0.101 |
distance | +0.100 |
roads | +0.094 |
Neg logit contributions | |
jewelry | -0.369 |
necklace | -0.339 |
endant | -0.337 |
brace | -0.330 |
fault | -0.329 |
Token | ieve |
Token ID | 12311 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
Run | +0.704 |
distance | +0.647 |
Uh | +0.644 |
Quick | +0.642 |
bel | +0.639 |
Neg logit contributions | |
jewelry | -0.532 |
fault | -0.469 |
endant | -0.463 |
brace | -0.461 |
necklace | -0.460 |
Token | . |
Token ID | 13 |
Feature activation | -2.36e-03 |
Pos logit contributions | |
jewelry | +0.140 |
fault | +0.133 |
endant | +0.131 |
necklace | +0.125 |
Majesty | +0.123 |
Neg logit contributions | |
Run | -0.134 |
bel | -0.128 |
Quick | -0.121 |
Uh | -0.118 |
distance | -0.118 |
Token |
|
Token ID | 198 |
Feature activation | -2.48e-02 |
Pos logit contributions | |
he | +0.003 |
distance | +0.003 |
Run | +0.003 |
For | +0.003 |
When | +0.003 |
Neg logit contributions | |
endant | -0.005 |
jewelry | -0.005 |
fault | -0.005 |
necklace | -0.005 |
Majesty | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | -2.52e-02 |
Pos logit contributions | |
Run | +0.043 |
bel | +0.041 |
Quick | +0.038 |
Uh | +0.038 |
distance | +0.036 |
Neg logit contributions | |
jewelry | -0.049 |
fault | -0.047 |
endant | -0.046 |
necklace | -0.044 |
brace | -0.043 |
Token | One |
Token ID | 3198 |
Feature activation | +2.63e-02 |
Pos logit contributions | |
Run | +0.037 |
bel | +0.036 |
Quick | +0.033 |
Uh | +0.032 |
distance | +0.032 |
Neg logit contributions | |
jewelry | -0.057 |
fault | -0.054 |
endant | -0.053 |
necklace | -0.052 |
brace | -0.051 |
Token | and |
Token ID | 290 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
Majesty | +0.061 |
jewelry | +0.060 |
endant | +0.059 |
fault | +0.059 |
brace | +0.055 |
Neg logit contributions | |
bel | -0.045 |
he | -0.044 |
Run | -0.043 |
distance | -0.041 |
roads | -0.038 |
Token | they |
Token ID | 484 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
bel | +0.021 |
Run | +0.019 |
Mark | +0.018 |
distance | +0.018 |
Quick | +0.017 |
Neg logit contributions | |
endant | -0.024 |
fault | -0.023 |
jewelry | -0.022 |
Majesty | -0.021 |
necklace | -0.020 |
Token | played |
Token ID | 2826 |
Feature activation | +8.31e-02 |
Pos logit contributions | |
Run | +0.068 |
bel | +0.066 |
he | +0.061 |
Quick | +0.059 |
distance | +0.059 |
Neg logit contributions | |
jewelry | -0.095 |
fault | -0.094 |
endant | -0.091 |
Majesty | -0.087 |
necklace | -0.086 |
Token | make |
Token ID | 787 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
jewelry | +0.163 |
fault | +0.149 |
necklace | +0.147 |
brace | +0.143 |
endant | +0.143 |
Neg logit contributions | |
Run | -0.152 |
Uh | -0.141 |
Quick | -0.140 |
bel | -0.139 |
distance | -0.135 |
Token | - |
Token ID | 12 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
Run | +0.046 |
bel | +0.045 |
Quick | +0.042 |
he | +0.041 |
distance | +0.041 |
Neg logit contributions | |
jewelry | -0.050 |
fault | -0.048 |
endant | -0.047 |
necklace | -0.045 |
Majesty | -0.044 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.22e-01 |
Pos logit contributions | |
Run | +0.086 |
Quick | +0.050 |
Uh | +0.049 |
distance | +0.041 |
roads | +0.036 |
Neg logit contributions | |
jewelry | -0.427 |
endant | -0.403 |
necklace | -0.396 |
fault | -0.391 |
brace | -0.388 |
Token | ieve |
Token ID | 12311 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
Run | +0.738 |
distance | +0.689 |
Uh | +0.678 |
Quick | +0.676 |
bel | +0.649 |
Neg logit contributions | |
jewelry | -0.502 |
brace | -0.439 |
necklace | -0.428 |
endant | -0.426 |
fault | -0.419 |
Token | . |
Token ID | 13 |
Feature activation | -4.93e-03 |
Pos logit contributions | |
jewelry | +0.134 |
fault | +0.128 |
endant | +0.126 |
necklace | +0.120 |
Majesty | +0.118 |
Neg logit contributions | |
Run | -0.122 |
bel | -0.118 |
Quick | -0.111 |
distance | -0.108 |
Uh | -0.108 |
Token | Finally |
Token ID | 9461 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
Run | +0.008 |
he | +0.008 |
Quick | +0.008 |
bel | +0.008 |
distance | +0.007 |
Neg logit contributions | |
jewelry | -0.010 |
endant | -0.009 |
necklace | -0.009 |
fault | -0.009 |
brace | -0.009 |
Token | , |
Token ID | 11 |
Feature activation | +5.17e-02 |
Pos logit contributions | |
Run | +0.055 |
bel | +0.050 |
he | +0.049 |
Quick | +0.049 |
distance | +0.048 |
Neg logit contributions | |
jewelry | -0.053 |
endant | -0.053 |
fault | -0.053 |
Majesty | -0.047 |
necklace | -0.046 |
Token | they |
Token ID | 484 |
Feature activation | -4.42e-02 |
Pos logit contributions | |
jewelry | +0.086 |
fault | +0.081 |
necklace | +0.078 |
endant | +0.077 |
collection | +0.075 |
Neg logit contributions | |
Run | -0.103 |
bel | -0.102 |
Quick | -0.098 |
Mark | -0.096 |
Uh | -0.095 |
Token | red |
Token ID | 2266 |
Feature activation | +3.49e-02 |
Pos logit contributions | |
bel | +0.052 |
distance | +0.047 |
Run | +0.047 |
Quick | +0.045 |
Ar | +0.044 |
Neg logit contributions | |
endant | -0.035 |
fault | -0.035 |
jewelry | -0.034 |
Majesty | -0.031 |
ott | -0.030 |
Token | cape |
Token ID | 36023 |
Feature activation | +7.05e-03 |
Pos logit contributions | |
fault | +0.057 |
endant | +0.055 |
jewelry | +0.051 |
Majesty | +0.050 |
ott | +0.050 |
Neg logit contributions | |
bel | -0.071 |
Run | -0.062 |
distance | -0.062 |
he | -0.059 |
Ar | -0.059 |
Token | and |
Token ID | 290 |
Feature activation | -8.51e-03 |
Pos logit contributions | |
jewelry | +0.015 |
fault | +0.014 |
endant | +0.014 |
necklace | +0.013 |
Majesty | +0.013 |
Neg logit contributions | |
Run | -0.011 |
bel | -0.011 |
Quick | -0.010 |
he | -0.010 |
distance | -0.010 |
Token | make |
Token ID | 787 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
bel | +0.016 |
Run | +0.015 |
distance | +0.014 |
Quick | +0.014 |
Mark | +0.014 |
Neg logit contributions | |
fault | -0.016 |
endant | -0.016 |
jewelry | -0.015 |
necklace | -0.014 |
Majesty | -0.014 |
Token | - |
Token ID | 12 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
bel | +0.044 |
Run | +0.044 |
Quick | +0.040 |
he | +0.039 |
Mark | +0.039 |
Neg logit contributions | |
jewelry | -0.045 |
endant | -0.044 |
fault | -0.044 |
necklace | -0.040 |
Majesty | -0.039 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.21e-01 |
Pos logit contributions | |
Run | +0.082 |
Uh | +0.051 |
Quick | +0.049 |
distance | +0.041 |
he | +0.038 |
Neg logit contributions | |
jewelry | -0.423 |
endant | -0.399 |
necklace | -0.396 |
fault | -0.393 |
brace | -0.386 |
Token | ieve |
Token ID | 12311 |
Feature activation | +7.15e-02 |
Pos logit contributions | |
Run | +0.681 |
bel | +0.632 |
Quick | +0.622 |
distance | +0.615 |
Uh | +0.614 |
Neg logit contributions | |
jewelry | -0.535 |
fault | -0.488 |
endant | -0.482 |
necklace | -0.468 |
brace | -0.462 |
Token | he |
Token ID | 339 |
Feature activation | -8.26e-03 |
Pos logit contributions | |
jewelry | +0.144 |
fault | +0.138 |
endant | +0.137 |
necklace | +0.130 |
Majesty | +0.128 |
Neg logit contributions | |
Run | -0.122 |
bel | -0.118 |
Quick | -0.110 |
distance | -0.107 |
Uh | -0.106 |
Token | had |
Token ID | 550 |
Feature activation | +8.30e-03 |
Pos logit contributions | |
bel | +0.014 |
Run | +0.013 |
he | +0.012 |
distance | +0.012 |
Quick | +0.012 |
Neg logit contributions | |
jewelry | -0.016 |
fault | -0.016 |
endant | -0.016 |
Majesty | -0.015 |
necklace | -0.015 |
Token | super |
Token ID | 2208 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
jewelry | +0.017 |
endant | +0.017 |
fault | +0.017 |
Majesty | +0.015 |
necklace | +0.015 |
Neg logit contributions | |
bel | -0.013 |
Run | -0.013 |
distance | -0.012 |
Mark | -0.012 |
he | -0.012 |
Token | powers |
Token ID | 30132 |
Feature activation | +1.14e-01 |
Pos logit contributions | |
bel | +0.032 |
Run | +0.031 |
he | +0.028 |
distance | +0.028 |
Quick | +0.028 |
Neg logit contributions | |
jewelry | -0.033 |
fault | -0.031 |
endant | -0.031 |
necklace | -0.029 |
Majesty | -0.029 |
Token | - |
Token ID | 12 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
fault | +0.350 |
jewelry | +0.347 |
endant | +0.339 |
Majesty | +0.321 |
necklace | +0.315 |
Neg logit contributions | |
bel | -0.277 |
Run | -0.274 |
he | -0.251 |
Quick | -0.245 |
distance | -0.240 |
Token | up |
Token ID | 929 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
Run | +0.292 |
bel | +0.281 |
Quick | +0.269 |
Uh | +0.264 |
distance | +0.263 |
Neg logit contributions | |
jewelry | -0.201 |
fault | -0.188 |
endant | -0.186 |
necklace | -0.175 |
Majesty | -0.169 |
Token | and |
Token ID | 290 |
Feature activation | -8.30e-03 |
Pos logit contributions | |
jewelry | +0.121 |
fault | +0.116 |
endant | +0.115 |
necklace | +0.109 |
Majesty | +0.108 |
Neg logit contributions | |
Run | -0.098 |
bel | -0.094 |
Quick | -0.088 |
Uh | -0.086 |
distance | -0.085 |
Token | make |
Token ID | 787 |
Feature activation | -2.01e-02 |
Pos logit contributions | |
bel | +0.015 |
Run | +0.014 |
distance | +0.013 |
Quick | +0.013 |
he | +0.012 |
Neg logit contributions | |
fault | -0.017 |
endant | -0.016 |
jewelry | -0.015 |
Majesty | -0.015 |
necklace | -0.013 |
Token | - |
Token ID | 12 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
bel | +0.039 |
Run | +0.036 |
he | +0.034 |
Quick | +0.034 |
amed | +0.033 |
Neg logit contributions | |
fault | -0.035 |
endant | -0.035 |
jewelry | -0.032 |
Majesty | -0.030 |
collection | -0.029 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
Run | +0.219 |
Quick | +0.189 |
Uh | +0.186 |
distance | +0.184 |
bel | +0.183 |
Neg logit contributions | |
jewelry | -0.286 |
endant | -0.267 |
fault | -0.261 |
necklace | -0.259 |
brace | -0.251 |
Token | ieve |
Token ID | 12311 |
Feature activation | +7.38e-02 |
Pos logit contributions | |
Run | +0.667 |
distance | +0.614 |
Uh | +0.602 |
Quick | +0.601 |
bel | +0.593 |
Neg logit contributions | |
jewelry | -0.565 |
brace | -0.499 |
endant | -0.493 |
necklace | -0.493 |
fault | -0.491 |
Token | . |
Token ID | 13 |
Feature activation | -4.62e-03 |
Pos logit contributions | |
jewelry | +0.148 |
fault | +0.144 |
endant | +0.142 |
necklace | +0.134 |
Majesty | +0.133 |
Neg logit contributions | |
Run | -0.125 |
bel | -0.123 |
Quick | -0.112 |
he | -0.110 |
distance | -0.110 |
Token | One |
Token ID | 1881 |
Feature activation | +6.44e-02 |
Pos logit contributions | |
he | +0.006 |
distance | +0.005 |
At | +0.005 |
When | +0.005 |
Mark | +0.005 |
Neg logit contributions | |
endant | -0.011 |
Majesty | -0.011 |
fault | -0.010 |
jewelry | -0.010 |
necklace | -0.010 |
Token | day |
Token ID | 1110 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
jewelry | +0.137 |
fault | +0.128 |
endant | +0.127 |
necklace | +0.126 |
collection | +0.119 |
Neg logit contributions | |
Run | -0.106 |
bel | -0.100 |
Quick | -0.093 |
Uh | -0.091 |
Mark | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
Run | +0.067 |
bel | +0.063 |
Quick | +0.060 |
he | +0.059 |
Uh | +0.058 |
Neg logit contributions | |
fault | -0.062 |
jewelry | -0.060 |
endant | -0.060 |
Majesty | -0.057 |
necklace | -0.055 |
Token | he |
Token ID | 339 |
Feature activation | -8.17e-03 |
Pos logit contributions | |
Run | +0.104 |
bel | +0.098 |
distance | +0.092 |
Quick | +0.092 |
Mark | +0.090 |
Neg logit contributions | |
jewelry | -0.117 |
endant | -0.115 |
fault | -0.114 |
necklace | -0.107 |
brace | -0.104 |
Token | was |
Token ID | 373 |
Feature activation | -5.23e-03 |
Pos logit contributions | |
Run | +0.014 |
bel | +0.013 |
Quick | +0.012 |
Uh | +0.012 |
Mark | +0.012 |
Neg logit contributions | |
jewelry | -0.017 |
endant | -0.016 |
fault | -0.016 |
necklace | -0.015 |
Majesty | -0.015 |
Token | playing |
Token ID | 2712 |
Feature activation | +3.28e-02 |
Pos logit contributions | |
Run | +0.009 |
bel | +0.009 |
Quick | +0.008 |
distance | +0.008 |
he | +0.008 |
Neg logit contributions | |
jewelry | -0.010 |
fault | -0.010 |
endant | -0.010 |
necklace | -0.009 |
Majesty | -0.009 |
Token | make |
Token ID | 787 |
Feature activation | -2.25e-02 |
Pos logit contributions | |
jewelry | +0.065 |
fault | +0.059 |
necklace | +0.058 |
brace | +0.056 |
endant | +0.056 |
Neg logit contributions | |
Run | -0.059 |
bel | -0.056 |
Quick | -0.055 |
Uh | -0.055 |
distance | -0.051 |
Token | - |
Token ID | 12 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Run | +0.040 |
bel | +0.039 |
Quick | +0.036 |
he | +0.035 |
distance | +0.035 |
Neg logit contributions | |
jewelry | -0.043 |
endant | -0.042 |
fault | -0.042 |
necklace | -0.039 |
Majesty | -0.039 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
Run | +0.108 |
Uh | +0.077 |
Quick | +0.076 |
distance | +0.069 |
roads | +0.063 |
Neg logit contributions | |
jewelry | -0.395 |
endant | -0.373 |
necklace | -0.366 |
fault | -0.363 |
brace | -0.358 |
Token | ieve |
Token ID | 12311 |
Feature activation | +7.42e-02 |
Pos logit contributions | |
Run | +0.742 |
distance | +0.686 |
Quick | +0.684 |
Uh | +0.679 |
bel | +0.670 |
Neg logit contributions | |
jewelry | -0.480 |
brace | -0.414 |
fault | -0.409 |
endant | -0.408 |
necklace | -0.407 |
Token | at |
Token ID | 379 |
Feature activation | -7.07e-02 |
Pos logit contributions | |
jewelry | +0.140 |
fault | +0.134 |
endant | +0.132 |
necklace | +0.125 |
Majesty | +0.123 |
Neg logit contributions | |
Run | -0.136 |
bel | -0.131 |
Quick | -0.124 |
Uh | -0.121 |
distance | -0.120 |
Token | home |
Token ID | 1363 |
Feature activation | -5.71e-03 |
Pos logit contributions | |
bel | +0.090 |
he | +0.087 |
distance | +0.087 |
Run | +0.086 |
Mark | +0.085 |
Neg logit contributions | |
endant | -0.160 |
jewelry | -0.158 |
fault | -0.151 |
Majesty | -0.149 |
necklace | -0.146 |
Token | , |
Token ID | 11 |
Feature activation | +5.31e-02 |
Pos logit contributions | |
Run | +0.011 |
bel | +0.011 |
Quick | +0.010 |
Uh | +0.010 |
distance | +0.010 |
Neg logit contributions | |
jewelry | -0.010 |
fault | -0.009 |
endant | -0.009 |
necklace | -0.009 |
Majesty | -0.009 |
Token | too |
Token ID | 1165 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
jewelry | +0.110 |
endant | +0.108 |
fault | +0.105 |
necklace | +0.099 |
Majesty | +0.098 |
Neg logit contributions | |
Run | -0.085 |
bel | -0.083 |
distance | -0.078 |
Quick | -0.075 |
Mark | -0.074 |
Token | a |
Token ID | 257 |
Feature activation | +1.14e-02 |
Pos logit contributions | |
Run | +0.052 |
bel | +0.051 |
distance | +0.048 |
he | +0.047 |
Quick | +0.046 |
Neg logit contributions | |
jewelry | -0.069 |
fault | -0.069 |
endant | -0.069 |
Majesty | -0.065 |
necklace | -0.063 |
Token | world |
Token ID | 995 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
fault | +0.023 |
endant | +0.022 |
jewelry | +0.021 |
Majesty | +0.020 |
brace | +0.019 |
Neg logit contributions | |
distance | -0.020 |
Run | -0.019 |
bel | -0.019 |
Uh | -0.018 |
Quick | -0.018 |
Token | of |
Token ID | 286 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Run | +0.033 |
bel | +0.031 |
Quick | +0.031 |
Uh | +0.030 |
Mark | +0.029 |
Neg logit contributions | |
jewelry | -0.030 |
endant | -0.027 |
fault | -0.027 |
necklace | -0.026 |
brace | -0.025 |
Token | make |
Token ID | 787 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
Run | +0.191 |
bel | +0.184 |
distance | +0.183 |
Mark | +0.173 |
he | +0.167 |
Neg logit contributions | |
fault | -0.215 |
endant | -0.212 |
jewelry | -0.196 |
Majesty | -0.193 |
necklace | -0.181 |
Token | - |
Token ID | 12 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
Run | +0.047 |
bel | +0.046 |
Quick | +0.042 |
he | +0.042 |
distance | +0.042 |
Neg logit contributions | |
jewelry | -0.049 |
fault | -0.047 |
endant | -0.047 |
necklace | -0.043 |
Majesty | -0.043 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.19e-01 |
Pos logit contributions | |
Run | +0.125 |
Quick | +0.096 |
Uh | +0.092 |
distance | +0.082 |
bel | +0.081 |
Neg logit contributions | |
jewelry | -0.376 |
endant | -0.354 |
fault | -0.347 |
necklace | -0.346 |
brace | -0.342 |
Token | ieve |
Token ID | 12311 |
Feature activation | +7.16e-02 |
Pos logit contributions | |
Run | +0.739 |
Quick | +0.681 |
Uh | +0.673 |
distance | +0.672 |
bel | +0.671 |
Neg logit contributions | |
jewelry | -0.486 |
endant | -0.421 |
brace | -0.419 |
fault | -0.417 |
necklace | -0.411 |
Token | . |
Token ID | 13 |
Feature activation | -4.37e-03 |
Pos logit contributions | |
jewelry | +0.133 |
fault | +0.128 |
endant | +0.127 |
necklace | +0.119 |
Majesty | +0.117 |
Neg logit contributions | |
Run | -0.131 |
bel | -0.129 |
Quick | -0.119 |
distance | -0.118 |
he | -0.117 |
Token |
|
Token ID | 198 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
Run | +0.005 |
distance | +0.005 |
he | +0.005 |
bel | +0.005 |
Mark | +0.005 |
Neg logit contributions | |
fault | -0.010 |
jewelry | -0.010 |
endant | -0.010 |
necklace | -0.010 |
brace | -0.009 |
Token |
|
Token ID | 198 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
Run | +0.044 |
bel | +0.042 |
Quick | +0.039 |
Uh | +0.039 |
distance | +0.038 |
Neg logit contributions | |
jewelry | -0.052 |
fault | -0.049 |
endant | -0.049 |
necklace | -0.047 |
brace | -0.046 |
Token | She |
Token ID | 3347 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
Run | +0.034 |
bel | +0.033 |
amer | +0.029 |
distance | +0.029 |
Uh | +0.028 |
Neg logit contributions | |
jewelry | -0.061 |
necklace | -0.055 |
brace | -0.055 |
endant | -0.055 |
bracelet | -0.054 |
Token | Finally |
Token ID | 9461 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
Run | +0.007 |
bel | +0.007 |
Mark | +0.007 |
Quick | +0.007 |
he | +0.007 |
Neg logit contributions | |
jewelry | -0.012 |
fault | -0.011 |
endant | -0.011 |
necklace | -0.011 |
brace | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +4.95e-02 |
Pos logit contributions | |
Run | +0.046 |
bel | +0.041 |
he | +0.040 |
distance | +0.040 |
Quick | +0.039 |
Neg logit contributions | |
fault | -0.050 |
endant | -0.050 |
jewelry | -0.047 |
Majesty | -0.043 |
necklace | -0.042 |
Token | her |
Token ID | 607 |
Feature activation | +6.51e-02 |
Pos logit contributions | |
jewelry | +0.083 |
fault | +0.080 |
endant | +0.078 |
necklace | +0.073 |
Majesty | +0.072 |
Neg logit contributions | |
Run | -0.102 |
bel | -0.099 |
Quick | -0.093 |
distance | -0.091 |
he | -0.091 |
Token | make |
Token ID | 787 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
jewelry | +0.103 |
endant | +0.100 |
fault | +0.099 |
brace | +0.089 |
necklace | +0.086 |
Neg logit contributions | |
bel | -0.131 |
Run | -0.129 |
Quick | -0.121 |
Mark | -0.117 |
distance | -0.116 |
Token | - |
Token ID | 12 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
bel | +0.034 |
Run | +0.033 |
he | +0.030 |
distance | +0.030 |
Mark | +0.030 |
Neg logit contributions | |
endant | -0.042 |
jewelry | -0.041 |
fault | -0.040 |
necklace | -0.037 |
Majesty | -0.036 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.18e-01 |
Pos logit contributions | |
Run | +0.169 |
distance | +0.137 |
Uh | +0.137 |
Quick | +0.134 |
roads | +0.127 |
Neg logit contributions | |
jewelry | -0.329 |
endant | -0.309 |
necklace | -0.305 |
fault | -0.302 |
brace | -0.295 |
Token | ieve |
Token ID | 12311 |
Feature activation | +7.52e-02 |
Pos logit contributions | |
Run | +0.733 |
distance | +0.677 |
Quick | +0.671 |
bel | +0.670 |
Uh | +0.666 |
Neg logit contributions | |
jewelry | -0.482 |
brace | -0.419 |
necklace | -0.415 |
endant | -0.413 |
fault | -0.411 |
Token | adventures |
Token ID | 17545 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
jewelry | +0.132 |
endant | +0.126 |
fault | +0.126 |
necklace | +0.116 |
Majesty | +0.115 |
Neg logit contributions | |
Run | -0.145 |
bel | -0.143 |
Quick | -0.132 |
distance | -0.131 |
he | -0.130 |
Token | were |
Token ID | 547 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
jewelry | +0.243 |
fault | +0.231 |
endant | +0.227 |
necklace | +0.218 |
brace | +0.211 |
Neg logit contributions | |
Run | -0.239 |
bel | -0.227 |
Quick | -0.218 |
Uh | -0.213 |
distance | -0.209 |
Token | recorded |
Token ID | 6264 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
Run | +0.106 |
bel | +0.104 |
Quick | +0.096 |
distance | +0.095 |
Mark | +0.093 |
Neg logit contributions | |
jewelry | -0.120 |
fault | -0.115 |
endant | -0.115 |
necklace | -0.108 |
Majesty | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | -7.79e-03 |
Pos logit contributions | |
jewelry | +0.031 |
fault | +0.029 |
endant | +0.029 |
necklace | +0.028 |
Majesty | +0.027 |
Neg logit contributions | |
Run | -0.030 |
bel | -0.028 |
Quick | -0.027 |
Uh | -0.026 |
distance | -0.026 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
bel | +0.013 |
Run | +0.013 |
Quick | +0.012 |
he | +0.011 |
Mark | +0.011 |
Neg logit contributions | |
fault | -0.023 |
jewelry | -0.022 |
endant | -0.021 |
necklace | -0.020 |
Majesty | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
endant | +0.061 |
fault | +0.058 |
Majesty | +0.055 |
necklace | +0.054 |
jewelry | +0.054 |
Neg logit contributions | |
bel | -0.049 |
Run | -0.045 |
distance | -0.042 |
Mark | -0.041 |
Quick | -0.040 |
Token | play |
Token ID | 711 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
fault | +0.073 |
endant | +0.071 |
jewelry | +0.065 |
Majesty | +0.062 |
penn | +0.060 |
Neg logit contributions | |
bel | -0.041 |
Run | -0.040 |
he | -0.037 |
distance | -0.036 |
Mark | -0.034 |
Token | make |
Token ID | 787 |
Feature activation | -1.94e-02 |
Pos logit contributions | |
fault | +0.195 |
jewelry | +0.194 |
endant | +0.192 |
Majesty | +0.181 |
necklace | +0.178 |
Neg logit contributions | |
Run | -0.122 |
bel | -0.119 |
he | -0.107 |
Quick | -0.107 |
Mark | -0.103 |
Token | - |
Token ID | 12 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
bel | +0.035 |
Run | +0.034 |
he | +0.031 |
Quick | +0.031 |
Mark | +0.030 |
Neg logit contributions | |
endant | -0.036 |
fault | -0.036 |
jewelry | -0.035 |
Majesty | -0.032 |
necklace | -0.032 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.17e-01 |
Pos logit contributions | |
Run | +0.194 |
distance | +0.164 |
Uh | +0.164 |
Quick | +0.161 |
roads | +0.152 |
Neg logit contributions | |
jewelry | -0.307 |
endant | -0.280 |
necklace | -0.276 |
brace | -0.274 |
fault | -0.273 |
Token | ieve |
Token ID | 12311 |
Feature activation | +8.12e-02 |
Pos logit contributions | |
Run | +0.672 |
distance | +0.619 |
bel | +0.616 |
Quick | +0.612 |
Uh | +0.609 |
Neg logit contributions | |
jewelry | -0.536 |
fault | -0.484 |
endant | -0.478 |
brace | -0.466 |
necklace | -0.463 |
Token | . |
Token ID | 13 |
Feature activation | -2.41e-03 |
Pos logit contributions | |
jewelry | +0.162 |
fault | +0.152 |
endant | +0.150 |
necklace | +0.145 |
brace | +0.142 |
Neg logit contributions | |
Run | -0.144 |
bel | -0.134 |
Quick | -0.130 |
Uh | -0.127 |
distance | -0.125 |
Token | Today |
Token ID | 6288 |
Feature activation | -5.35e-02 |
Pos logit contributions | |
he | +0.003 |
distance | +0.003 |
bel | +0.003 |
Run | +0.003 |
Mark | +0.003 |
Neg logit contributions | |
endant | -0.005 |
jewelry | -0.005 |
necklace | -0.005 |
brace | -0.005 |
fault | -0.005 |
Token | she |
Token ID | 673 |
Feature activation | +6.81e-02 |
Pos logit contributions | |
bel | +0.084 |
Run | +0.080 |
he | +0.080 |
distance | +0.077 |
roads | +0.073 |
Neg logit contributions | |
endant | -0.106 |
fault | -0.103 |
Majesty | -0.103 |
jewelry | -0.102 |
brace | -0.098 |
Token | was |
Token ID | 373 |
Feature activation | -2.61e-03 |
Pos logit contributions | |
jewelry | +0.154 |
fault | +0.154 |
endant | +0.150 |
Majesty | +0.141 |
necklace | +0.139 |
Neg logit contributions | |
Run | -0.094 |
bel | -0.093 |
Quick | -0.085 |
he | -0.085 |
Uh | -0.080 |
Token | was |
Token ID | 373 |
Feature activation | -7.83e-03 |
Pos logit contributions | |
Run | +0.208 |
bel | +0.203 |
Quick | +0.185 |
distance | +0.184 |
he | +0.183 |
Neg logit contributions | |
jewelry | -0.260 |
fault | -0.250 |
endant | -0.249 |
necklace | -0.231 |
Majesty | -0.229 |
Token | ready |
Token ID | 3492 |
Feature activation | -3.50e-02 |
Pos logit contributions | |
Run | +0.014 |
bel | +0.013 |
Quick | +0.013 |
Uh | +0.012 |
he | +0.012 |
Neg logit contributions | |
jewelry | -0.015 |
fault | -0.015 |
endant | -0.014 |
necklace | -0.014 |
Majesty | -0.013 |
Token | for |
Token ID | 329 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
Run | +0.064 |
bel | +0.063 |
distance | +0.058 |
he | +0.058 |
Quick | +0.058 |
Neg logit contributions | |
jewelry | -0.065 |
endant | -0.062 |
fault | -0.061 |
Majesty | -0.057 |
necklace | -0.057 |
Token | make |
Token ID | 787 |
Feature activation | -1.89e-02 |
Pos logit contributions | |
Run | +0.171 |
bel | +0.157 |
Mark | +0.154 |
distance | +0.151 |
he | +0.147 |
Neg logit contributions | |
endant | -0.230 |
jewelry | -0.229 |
fault | -0.224 |
necklace | -0.208 |
Majesty | -0.205 |
Token | - |
Token ID | 12 |
Feature activation | -1.33e-01 |
Pos logit contributions | |
Run | +0.034 |
bel | +0.033 |
Quick | +0.030 |
he | +0.030 |
Mark | +0.030 |
Neg logit contributions | |
jewelry | -0.036 |
endant | -0.035 |
fault | -0.034 |
necklace | -0.032 |
Majesty | -0.031 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.16e-01 |
Pos logit contributions | |
Run | +0.230 |
Uh | +0.205 |
Quick | +0.203 |
distance | +0.197 |
bel | +0.192 |
Neg logit contributions | |
jewelry | -0.268 |
fault | -0.249 |
endant | -0.244 |
necklace | -0.242 |
Majesty | -0.237 |
Token | ieve |
Token ID | 12311 |
Feature activation | +8.13e-02 |
Pos logit contributions | |
Run | +0.704 |
distance | +0.653 |
Uh | +0.650 |
Quick | +0.649 |
bel | +0.644 |
Neg logit contributions | |
jewelry | -0.493 |
fault | -0.451 |
brace | -0.432 |
endant | -0.428 |
Majesty | -0.426 |
Token | fun |
Token ID | 1257 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
jewelry | +0.146 |
fault | +0.139 |
endant | +0.135 |
necklace | +0.130 |
Majesty | +0.127 |
Neg logit contributions | |
Run | -0.158 |
bel | -0.150 |
Quick | -0.145 |
Uh | -0.142 |
distance | -0.139 |
Token | . |
Token ID | 13 |
Feature activation | -5.78e-03 |
Pos logit contributions | |
jewelry | +0.122 |
fault | +0.115 |
endant | +0.111 |
necklace | +0.109 |
brace | +0.105 |
Neg logit contributions | |
Run | -0.129 |
bel | -0.121 |
Quick | -0.119 |
Uh | -0.116 |
distance | -0.112 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.12e-01 |
Pos logit contributions | |
Run | +0.010 |
bel | +0.010 |
Quick | +0.009 |
Uh | +0.009 |
distance | +0.009 |
Neg logit contributions | |
fault | -0.011 |
jewelry | -0.011 |
endant | -0.010 |
Majesty | -0.010 |
necklace | -0.010 |
Token | Once |
Token ID | 7454 |
Feature activation | +6.61e-02 |
Pos logit contributions | |
bel | +0.176 |
Run | +0.170 |
Quick | +0.154 |
Uh | +0.151 |
distance | +0.143 |
Neg logit contributions | |
jewelry | -0.237 |
fault | -0.223 |
endant | -0.221 |
necklace | -0.213 |
Ros | -0.206 |
Token | He |
Token ID | 679 |
Feature activation | -5.11e-02 |
Pos logit contributions | |
Run | +0.005 |
bel | +0.005 |
Quick | +0.004 |
Uh | +0.004 |
distance | +0.004 |
Neg logit contributions | |
jewelry | -0.006 |
fault | -0.006 |
endant | -0.006 |
necklace | -0.006 |
Majesty | -0.006 |
Token | loved |
Token ID | 6151 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
Run | +0.076 |
bel | +0.073 |
Quick | +0.067 |
Mark | +0.067 |
he | +0.066 |
Neg logit contributions | |
jewelry | -0.112 |
fault | -0.110 |
endant | -0.108 |
necklace | -0.102 |
Majesty | -0.102 |
Token | playing |
Token ID | 2712 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
endant | +0.052 |
fault | +0.050 |
jewelry | +0.050 |
necklace | +0.047 |
Majesty | +0.047 |
Neg logit contributions | |
bel | -0.051 |
Run | -0.049 |
distance | -0.045 |
Quick | -0.044 |
Mark | -0.044 |
Token | make |
Token ID | 787 |
Feature activation | -2.03e-02 |
Pos logit contributions | |
jewelry | +0.072 |
necklace | +0.064 |
fault | +0.064 |
brace | +0.063 |
endant | +0.063 |
Neg logit contributions | |
Run | -0.059 |
bel | -0.055 |
Quick | -0.053 |
Uh | -0.053 |
distance | -0.052 |
Token | - |
Token ID | 12 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
Run | +0.034 |
bel | +0.033 |
Quick | +0.031 |
distance | +0.030 |
he | +0.030 |
Neg logit contributions | |
jewelry | -0.041 |
fault | -0.040 |
endant | -0.039 |
necklace | -0.037 |
Majesty | -0.037 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
Run | +0.128 |
Uh | +0.094 |
Quick | +0.093 |
distance | +0.090 |
roads | +0.078 |
Neg logit contributions | |
jewelry | -0.370 |
endant | -0.345 |
fault | -0.340 |
necklace | -0.338 |
brace | -0.334 |
Token | ieve |
Token ID | 12311 |
Feature activation | +8.08e-02 |
Pos logit contributions | |
Run | +0.687 |
distance | +0.629 |
Quick | +0.625 |
bel | +0.622 |
Uh | +0.620 |
Neg logit contributions | |
jewelry | -0.522 |
fault | -0.462 |
endant | -0.459 |
brace | -0.452 |
necklace | -0.448 |
Token | with |
Token ID | 351 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
jewelry | +0.157 |
fault | +0.147 |
endant | +0.143 |
necklace | +0.140 |
brace | +0.136 |
Neg logit contributions | |
Run | -0.148 |
bel | -0.136 |
Quick | -0.135 |
Uh | -0.133 |
distance | -0.129 |
Token | his |
Token ID | 465 |
Feature activation | -2.49e-02 |
Pos logit contributions | |
endant | +0.034 |
fault | +0.033 |
jewelry | +0.033 |
Majesty | +0.031 |
necklace | +0.031 |
Neg logit contributions | |
bel | -0.022 |
Run | -0.021 |
Mark | -0.020 |
distance | -0.019 |
he | -0.019 |
Token | friends |
Token ID | 2460 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
bel | +0.049 |
Run | +0.048 |
Quick | +0.045 |
distance | +0.045 |
he | +0.043 |
Neg logit contributions | |
fault | -0.040 |
endant | -0.040 |
jewelry | -0.040 |
brace | -0.036 |
Majesty | -0.036 |
Token | . |
Token ID | 13 |
Feature activation | -4.34e-03 |
Pos logit contributions | |
Run | +0.076 |
bel | +0.075 |
Quick | +0.069 |
he | +0.069 |
distance | +0.068 |
Neg logit contributions | |
jewelry | -0.083 |
fault | -0.080 |
endant | -0.080 |
necklace | -0.075 |
Majesty | -0.074 |
Token | " |
Token ID | 366 |
Feature activation | -7.79e-02 |
Pos logit contributions | |
jewelry | +0.110 |
fault | +0.107 |
endant | +0.105 |
necklace | +0.099 |
Majesty | +0.097 |
Neg logit contributions | |
Run | -0.101 |
bel | -0.096 |
Quick | -0.091 |
Uh | -0.087 |
distance | -0.087 |
Token | Please |
Token ID | 5492 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
bel | +0.099 |
Run | +0.092 |
ditch | +0.085 |
distance | +0.082 |
roads | +0.081 |
Neg logit contributions | |
jewelry | -0.172 |
endant | -0.159 |
necklace | -0.159 |
brace | -0.157 |
bracelet | -0.150 |
Token | keep |
Token ID | 1394 |
Feature activation | +6.20e-02 |
Pos logit contributions | |
Run | +0.086 |
bel | +0.083 |
Quick | +0.078 |
distance | +0.077 |
he | +0.077 |
Neg logit contributions | |
jewelry | -0.084 |
fault | -0.081 |
endant | -0.080 |
necklace | -0.074 |
brace | -0.074 |
Token | your |
Token ID | 534 |
Feature activation | +1.31e-01 |
Pos logit contributions | |
fault | +0.124 |
jewelry | +0.121 |
endant | +0.121 |
necklace | +0.111 |
Majesty | +0.110 |
Neg logit contributions | |
Run | -0.105 |
bel | -0.099 |
Quick | -0.092 |
distance | -0.092 |
roads | -0.090 |
Token | seat |
Token ID | 5852 |
Feature activation | -2.63e-02 |
Pos logit contributions | |
jewelry | +0.238 |
endant | +0.231 |
fault | +0.231 |
brace | +0.212 |
Majesty | +0.208 |
Neg logit contributions | |
bel | -0.241 |
Run | -0.232 |
distance | -0.222 |
Quick | -0.214 |
he | -0.206 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
Run | +0.047 |
bel | +0.044 |
Quick | +0.042 |
Uh | +0.041 |
distance | +0.040 |
Neg logit contributions | |
jewelry | -0.052 |
fault | -0.049 |
endant | -0.049 |
necklace | -0.047 |
Majesty | -0.045 |
Token | ts |
Token ID | 912 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
Run | +0.677 |
bel | +0.636 |
distance | +0.627 |
Quick | +0.625 |
Uh | +0.625 |
Neg logit contributions | |
jewelry | -0.512 |
fault | -0.458 |
endant | -0.448 |
brace | -0.447 |
necklace | -0.447 |
Token | on |
Token ID | 319 |
Feature activation | -3.12e-02 |
Pos logit contributions | |
Run | +0.223 |
bel | +0.214 |
Quick | +0.202 |
Uh | +0.196 |
distance | +0.195 |
Neg logit contributions | |
jewelry | -0.249 |
fault | -0.239 |
endant | -0.235 |
necklace | -0.224 |
Majesty | -0.220 |
Token | . |
Token ID | 13 |
Feature activation | -8.34e-04 |
Pos logit contributions | |
Run | +0.060 |
bel | +0.058 |
Quick | +0.055 |
distance | +0.054 |
he | +0.054 |
Neg logit contributions | |
jewelry | -0.056 |
fault | -0.053 |
endant | -0.053 |
necklace | -0.049 |
Majesty | -0.048 |
Token | We |
Token ID | 775 |
Feature activation | +2.61e-02 |
Pos logit contributions | |
Run | +0.002 |
bel | +0.002 |
Quick | +0.002 |
Uh | +0.001 |
Mark | +0.001 |
Neg logit contributions | |
jewelry | -0.001 |
fault | -0.001 |
endant | -0.001 |
necklace | -0.001 |
brace | -0.001 |
Token | will |
Token ID | 481 |
Feature activation | +4.98e-02 |
Pos logit contributions | |
jewelry | +0.047 |
fault | +0.045 |
endant | +0.044 |
necklace | +0.042 |
Majesty | +0.041 |
Neg logit contributions | |
Run | -0.050 |
bel | -0.049 |
Quick | -0.046 |
Uh | -0.045 |
distance | -0.045 |
Token | He |
Token ID | 679 |
Feature activation | -4.57e-02 |
Pos logit contributions | |
jewelry | +0.002 |
fault | +0.002 |
endant | +0.002 |
necklace | +0.002 |
Majesty | +0.002 |
Neg logit contributions | |
Run | -0.002 |
bel | -0.002 |
Quick | -0.001 |
Uh | -0.001 |
distance | -0.001 |
Token | would |
Token ID | 561 |
Feature activation | +7.88e-03 |
Pos logit contributions | |
Run | +0.080 |
bel | +0.077 |
Quick | +0.072 |
Uh | +0.070 |
distance | +0.070 |
Neg logit contributions | |
jewelry | -0.090 |
fault | -0.087 |
endant | -0.086 |
necklace | -0.081 |
Majesty | -0.080 |
Token | race |
Token ID | 3234 |
Feature activation | -5.62e-02 |
Pos logit contributions | |
jewelry | +0.016 |
fault | +0.015 |
endant | +0.015 |
necklace | +0.014 |
Majesty | +0.014 |
Neg logit contributions | |
Run | -0.013 |
bel | -0.013 |
Quick | -0.012 |
Uh | -0.012 |
distance | -0.012 |
Token | make |
Token ID | 787 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
Run | +0.114 |
Uh | +0.107 |
Quick | +0.107 |
bel | +0.099 |
he | +0.096 |
Neg logit contributions | |
jewelry | -0.101 |
fault | -0.093 |
necklace | -0.089 |
brace | -0.088 |
endant | -0.085 |
Token | - |
Token ID | 12 |
Feature activation | -1.29e-01 |
Pos logit contributions | |
Run | +0.023 |
bel | +0.022 |
Quick | +0.021 |
Uh | +0.020 |
distance | +0.020 |
Neg logit contributions | |
jewelry | -0.027 |
fault | -0.026 |
endant | -0.025 |
necklace | -0.024 |
Majesty | -0.024 |
Token | bel |
Token ID | 6667 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
Run | +0.174 |
Uh | +0.152 |
distance | +0.148 |
Quick | +0.141 |
he | +0.137 |
Neg logit contributions | |
jewelry | -0.305 |
endant | -0.280 |
necklace | -0.276 |
fault | -0.276 |
brace | -0.276 |
Token | ieve |
Token ID | 12311 |
Feature activation | +8.11e-02 |
Pos logit contributions | |
Run | +0.641 |
distance | +0.604 |
Uh | +0.588 |
Quick | +0.586 |
bel | +0.583 |
Neg logit contributions | |
jewelry | -0.551 |
fault | -0.491 |
brace | -0.489 |
endant | -0.486 |
necklace | -0.479 |
Token | cars |
Token ID | 5006 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
jewelry | +0.156 |
fault | +0.147 |
endant | +0.142 |
necklace | +0.141 |
brace | +0.137 |
Neg logit contributions | |
Run | -0.148 |
bel | -0.137 |
Quick | -0.135 |
Uh | -0.134 |
distance | -0.129 |
Token | around |
Token ID | 1088 |
Feature activation | -8.40e-02 |
Pos logit contributions | |
Run | +0.129 |
Quick | +0.122 |
Uh | +0.121 |
bel | +0.118 |
distance | +0.113 |
Neg logit contributions | |
jewelry | -0.107 |
fault | -0.099 |
necklace | -0.096 |
endant | -0.095 |
brace | -0.093 |
Token | the |
Token ID | 262 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
Run | +0.148 |
bel | +0.139 |
Quick | +0.136 |
Uh | +0.134 |
distance | +0.126 |
Neg logit contributions | |
jewelry | -0.168 |
fault | -0.157 |
necklace | -0.150 |
endant | -0.150 |
brace | -0.147 |
Token | track |
Token ID | 2610 |
Feature activation | -9.08e-02 |
Pos logit contributions | |
Run | +0.050 |
bel | +0.048 |
Quick | +0.045 |
Uh | +0.044 |
he | +0.044 |
Neg logit contributions | |
jewelry | -0.056 |
fault | -0.053 |
endant | -0.052 |
necklace | -0.051 |
Majesty | -0.049 |
Token | in |
Token ID | 287 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
Run | +0.559 |
bel | +0.539 |
Quick | +0.518 |
Uh | +0.505 |
Mark | +0.491 |
Neg logit contributions | |
jewelry | -0.616 |
fault | -0.578 |
endant | -0.574 |
necklace | -0.551 |
brace | -0.534 |
Token | yellow |
Token ID | 7872 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
bel | +0.074 |
Run | +0.070 |
Quick | +0.069 |
Uh | +0.068 |
amed | +0.067 |
Neg logit contributions | |
jewelry | -0.063 |
necklace | -0.055 |
fault | -0.054 |
brace | -0.053 |
sewing | -0.052 |
Token | suits |
Token ID | 14803 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
jewelry | +0.074 |
endant | +0.064 |
necklace | +0.064 |
fault | +0.063 |
brace | +0.062 |
Neg logit contributions | |
Run | -0.083 |
bel | -0.083 |
Quick | -0.079 |
Uh | -0.077 |
Mark | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -9.45e-03 |
Pos logit contributions | |
Run | +0.141 |
Quick | +0.138 |
bel | +0.137 |
Uh | +0.135 |
Mark | +0.127 |
Neg logit contributions | |
jewelry | -0.088 |
fault | -0.083 |
endant | -0.080 |
necklace | -0.077 |
brace | -0.072 |
Token | The |
Token ID | 383 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
bel | +0.017 |
Run | +0.017 |
he | +0.016 |
Mark | +0.016 |
Uh | +0.016 |
Neg logit contributions | |
jewelry | -0.017 |
fault | -0.016 |
endant | -0.016 |
necklace | -0.015 |
Majesty | -0.015 |
Token | men |
Token ID | 1450 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
jewelry | +0.055 |
necklace | +0.049 |
fault | +0.047 |
collection | +0.046 |
Majesty | +0.044 |
Neg logit contributions | |
Run | -0.067 |
he | -0.067 |
bel | -0.065 |
aming | -0.063 |
Uh | -0.062 |
Token | are |
Token ID | 389 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
Run | +0.565 |
bel | +0.555 |
Quick | +0.535 |
Uh | +0.517 |
Mark | +0.508 |
Neg logit contributions | |
jewelry | -0.599 |
endant | -0.557 |
fault | -0.549 |
necklace | -0.537 |
brace | -0.520 |
Token | firefighters |
Token ID | 22611 |
Feature activation | -1.61e-01 |
Pos logit contributions | |
Run | +0.032 |
Quick | +0.031 |
Uh | +0.029 |
he | +0.029 |
bel | +0.029 |
Neg logit contributions | |
jewelry | -0.027 |
necklace | -0.024 |
sewing | -0.023 |
brace | -0.022 |
fault | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -9.82e-03 |
Pos logit contributions | |
Run | +0.365 |
Quick | +0.352 |
bel | +0.349 |
Uh | +0.344 |
Mark | +0.325 |
Neg logit contributions | |
jewelry | -0.238 |
fault | -0.228 |
endant | -0.211 |
necklace | -0.208 |
collection | -0.196 |
Token | They |
Token ID | 1119 |
Feature activation | -6.92e-02 |
Pos logit contributions | |
bel | +0.020 |
Run | +0.018 |
Mark | +0.018 |
amed | +0.018 |
he | +0.018 |
Neg logit contributions | |
jewelry | -0.015 |
fault | -0.015 |
endant | -0.014 |
cand | -0.014 |
penn | -0.013 |
Token | are |
Token ID | 389 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
Run | +0.130 |
bel | +0.126 |
Quick | +0.121 |
Uh | +0.119 |
Mark | +0.115 |
Neg logit contributions | |
jewelry | -0.128 |
fault | -0.116 |
endant | -0.116 |
necklace | -0.114 |
Majesty | -0.111 |
Token | truck |
Token ID | 7779 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
bel | +0.072 |
Run | +0.071 |
aming | +0.067 |
amed | +0.066 |
he | +0.065 |
Neg logit contributions | |
jewelry | -0.053 |
necklace | -0.048 |
fault | -0.043 |
endant | -0.042 |
collection | -0.041 |
Token | . |
Token ID | 13 |
Feature activation | -8.87e-03 |
Pos logit contributions | |
Run | +0.211 |
bel | +0.201 |
Quick | +0.198 |
Uh | +0.194 |
Mark | +0.185 |
Neg logit contributions | |
jewelry | -0.170 |
fault | -0.159 |
endant | -0.158 |
necklace | -0.152 |
brace | -0.143 |
Token | They |
Token ID | 1119 |
Feature activation | -6.83e-02 |
Pos logit contributions | |
bel | +0.015 |
Run | +0.015 |
he | +0.014 |
Quick | +0.014 |
Mark | +0.014 |
Neg logit contributions | |
jewelry | -0.017 |
fault | -0.016 |
endant | -0.016 |
necklace | -0.015 |
Majesty | -0.015 |
Token | see |
Token ID | 766 |
Feature activation | -8.52e-02 |
Pos logit contributions | |
Run | +0.112 |
bel | +0.110 |
Quick | +0.103 |
Uh | +0.102 |
Mark | +0.098 |
Neg logit contributions | |
jewelry | -0.141 |
fault | -0.131 |
endant | -0.131 |
necklace | -0.128 |
Majesty | -0.125 |
Token | some |
Token ID | 617 |
Feature activation | +3.66e-02 |
Pos logit contributions | |
Run | +0.157 |
bel | +0.154 |
Quick | +0.146 |
Uh | +0.143 |
he | +0.142 |
Neg logit contributions | |
jewelry | -0.161 |
fault | -0.153 |
endant | -0.149 |
necklace | -0.144 |
Majesty | -0.139 |
Token | men |
Token ID | 1450 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
jewelry | +0.072 |
necklace | +0.063 |
endant | +0.062 |
fault | +0.062 |
brace | +0.061 |
Neg logit contributions | |
bel | -0.067 |
Run | -0.065 |
Quick | -0.063 |
Mark | -0.062 |
he | -0.061 |
Token | in |
Token ID | 287 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
Run | +0.559 |
bel | +0.539 |
Quick | +0.518 |
Uh | +0.505 |
Mark | +0.491 |
Neg logit contributions | |
jewelry | -0.616 |
fault | -0.578 |
endant | -0.574 |
necklace | -0.551 |
brace | -0.534 |
Token | yellow |
Token ID | 7872 |
Feature activation | +4.16e-02 |
Pos logit contributions | |
bel | +0.074 |
Run | +0.070 |
Quick | +0.069 |
Uh | +0.068 |
amed | +0.067 |
Neg logit contributions | |
jewelry | -0.063 |
necklace | -0.055 |
fault | -0.054 |
brace | -0.053 |
sewing | -0.052 |
Token | suits |
Token ID | 14803 |
Feature activation | -6.14e-02 |
Pos logit contributions | |
jewelry | +0.074 |
endant | +0.064 |
necklace | +0.064 |
fault | +0.063 |
brace | +0.062 |
Neg logit contributions | |
Run | -0.083 |
bel | -0.083 |
Quick | -0.079 |
Uh | -0.077 |
Mark | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -9.45e-03 |
Pos logit contributions | |
Run | +0.141 |
Quick | +0.138 |
bel | +0.137 |
Uh | +0.135 |
Mark | +0.127 |
Neg logit contributions | |
jewelry | -0.088 |
fault | -0.083 |
endant | -0.080 |
necklace | -0.077 |
brace | -0.072 |
Token | The |
Token ID | 383 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
bel | +0.017 |
Run | +0.017 |
he | +0.016 |
Mark | +0.016 |
Uh | +0.016 |
Neg logit contributions | |
jewelry | -0.017 |
fault | -0.016 |
endant | -0.016 |
necklace | -0.015 |
Majesty | -0.015 |
Token | , |
Token ID | 11 |
Feature activation | +5.25e-02 |
Pos logit contributions | |
Run | +0.247 |
bel | +0.236 |
he | +0.235 |
distance | +0.228 |
roads | +0.214 |
Neg logit contributions | |
jewelry | -0.283 |
endant | -0.282 |
fault | -0.274 |
Majesty | -0.272 |
necklace | -0.252 |
Token | but |
Token ID | 475 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
endant | +0.109 |
jewelry | +0.108 |
fault | +0.107 |
Majesty | +0.106 |
collection | +0.099 |
Neg logit contributions | |
Run | -0.082 |
bel | -0.078 |
distance | -0.072 |
Mark | -0.069 |
Quick | -0.067 |
Token | she |
Token ID | 673 |
Feature activation | +6.42e-02 |
Pos logit contributions | |
Run | +0.036 |
bel | +0.035 |
Mark | +0.033 |
he | +0.032 |
distance | +0.031 |
Neg logit contributions | |
Majesty | -0.048 |
endant | -0.046 |
fault | -0.046 |
jewelry | -0.043 |
collection | -0.042 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
jewelry | +0.139 |
fault | +0.135 |
endant | +0.133 |
Majesty | +0.126 |
necklace | +0.125 |
Neg logit contributions | |
Run | -0.098 |
bel | -0.095 |
Quick | -0.088 |
he | -0.087 |
Mark | -0.087 |
Token | charming |
Token ID | 23332 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
bel | +0.034 |
Run | +0.032 |
Mark | +0.030 |
distance | +0.030 |
Quick | +0.027 |
Neg logit contributions | |
endant | -0.054 |
fault | -0.050 |
Majesty | -0.049 |
jewelry | -0.049 |
necklace | -0.047 |
Token | men |
Token ID | 1450 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
fault | +0.111 |
endant | +0.110 |
jewelry | +0.109 |
Majesty | +0.101 |
necklace | +0.100 |
Neg logit contributions | |
Run | -0.103 |
bel | -0.101 |
distance | -0.092 |
Quick | -0.091 |
he | -0.089 |
Token | . |
Token ID | 13 |
Feature activation | -5.84e-03 |
Pos logit contributions | |
bel | +0.401 |
he | +0.398 |
distance | +0.375 |
Run | +0.363 |
Mark | +0.333 |
Neg logit contributions | |
endant | -0.702 |
Majesty | -0.682 |
fault | -0.677 |
jewelry | -0.663 |
majesty | -0.612 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.19e-01 |
Pos logit contributions | |
Run | +0.008 |
Mark | +0.007 |
he | +0.007 |
bel | +0.007 |
Quick | +0.007 |
Neg logit contributions | |
jewelry | -0.013 |
fault | -0.013 |
endant | -0.013 |
necklace | -0.013 |
collection | -0.013 |
Token | Anna |
Token ID | 31160 |
Feature activation | +9.85e-03 |
Pos logit contributions | |
Run | +0.161 |
bel | +0.161 |
Quick | +0.145 |
Uh | +0.140 |
distance | +0.135 |
Neg logit contributions | |
jewelry | -0.279 |
fault | -0.263 |
endant | -0.260 |
necklace | -0.253 |
brace | -0.246 |
Token | liked |
Token ID | 8288 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
fault | +0.023 |
jewelry | +0.022 |
endant | +0.022 |
Majesty | +0.021 |
necklace | +0.020 |
Neg logit contributions | |
Run | -0.013 |
bel | -0.012 |
he | -0.012 |
distance | -0.012 |
Quick | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
bel | +0.045 |
Run | +0.044 |
Quick | +0.039 |
Mark | +0.038 |
distance | +0.038 |
Neg logit contributions | |
endant | -0.052 |
fault | -0.051 |
Majesty | -0.050 |
jewelry | -0.049 |
necklace | -0.047 |
Token | pond |
Token ID | 16723 |
Feature activation | -3.96e-02 |
Pos logit contributions | |
Run | +0.068 |
bel | +0.066 |
Quick | +0.062 |
Uh | +0.060 |
Mark | +0.060 |
Neg logit contributions | |
jewelry | -0.069 |
fault | -0.065 |
endant | -0.064 |
necklace | -0.063 |
Majesty | -0.060 |
Token | . |
Token ID | 13 |
Feature activation | -1.84e-03 |
Pos logit contributions | |
Run | +0.084 |
bel | +0.081 |
Quick | +0.080 |
Uh | +0.078 |
Mark | +0.077 |
Neg logit contributions | |
jewelry | -0.062 |
fault | -0.060 |
endant | -0.058 |
necklace | -0.055 |
Majesty | -0.052 |
Token | Sara |
Token ID | 24799 |
Feature activation | -1.26e-02 |
Pos logit contributions | |
Run | +0.003 |
bel | +0.003 |
Quick | +0.003 |
Mark | +0.003 |
Uh | +0.003 |
Neg logit contributions | |
jewelry | -0.004 |
fault | -0.004 |
endant | -0.004 |
necklace | -0.004 |
brace | -0.003 |
Token | sees |
Token ID | 7224 |
Feature activation | -4.32e-02 |
Pos logit contributions | |
Run | +0.023 |
bel | +0.023 |
Quick | +0.021 |
Uh | +0.021 |
distance | +0.021 |
Neg logit contributions | |
jewelry | -0.024 |
fault | -0.023 |
endant | -0.022 |
necklace | -0.021 |
Majesty | -0.021 |
Token | some |
Token ID | 617 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
Run | +0.074 |
bel | +0.070 |
Quick | +0.065 |
he | +0.064 |
distance | +0.064 |
Neg logit contributions | |
jewelry | -0.086 |
endant | -0.083 |
fault | -0.083 |
necklace | -0.079 |
Majesty | -0.078 |
Token | men |
Token ID | 1450 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
jewelry | +0.066 |
fault | +0.062 |
endant | +0.062 |
necklace | +0.060 |
brace | +0.058 |
Neg logit contributions | |
Run | -0.057 |
bel | -0.056 |
Quick | -0.053 |
Mark | -0.051 |
Uh | -0.051 |
Token | in |
Token ID | 287 |
Feature activation | -3.38e-02 |
Pos logit contributions | |
Run | +0.530 |
bel | +0.512 |
Quick | +0.492 |
Uh | +0.477 |
Mark | +0.466 |
Neg logit contributions | |
jewelry | -0.641 |
fault | -0.603 |
endant | -0.596 |
necklace | -0.576 |
brace | -0.556 |
Token | yellow |
Token ID | 7872 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
bel | +0.065 |
Run | +0.060 |
Quick | +0.059 |
amed | +0.059 |
Uh | +0.057 |
Neg logit contributions | |
jewelry | -0.062 |
fault | -0.055 |
necklace | -0.053 |
brace | -0.052 |
makeup | -0.052 |
Token | hats |
Token ID | 23910 |
Feature activation | -6.65e-02 |
Pos logit contributions | |
jewelry | +0.085 |
fault | +0.077 |
endant | +0.076 |
necklace | +0.074 |
brace | +0.072 |
Neg logit contributions | |
Run | -0.093 |
bel | -0.091 |
Quick | -0.087 |
Mark | -0.086 |
he | -0.086 |
Token | and |
Token ID | 290 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
Run | +0.118 |
bel | +0.113 |
he | +0.106 |
Quick | +0.105 |
distance | +0.105 |
Neg logit contributions | |
jewelry | -0.128 |
fault | -0.122 |
endant | -0.121 |
necklace | -0.115 |
Majesty | -0.114 |
Token | coats |
Token ID | 30720 |
Feature activation | -9.72e-02 |
Pos logit contributions | |
Run | +0.023 |
bel | +0.022 |
Quick | +0.021 |
distance | +0.020 |
Uh | +0.020 |
Neg logit contributions | |
jewelry | -0.023 |
fault | -0.023 |
endant | -0.023 |
necklace | -0.021 |
Majesty | -0.021 |
Token | let |
Token ID | 1309 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
jewelry | +0.094 |
fault | +0.092 |
endant | +0.090 |
necklace | +0.082 |
Majesty | +0.082 |
Neg logit contributions | |
Run | -0.104 |
bel | -0.101 |
Quick | -0.095 |
distance | -0.093 |
Uh | -0.092 |
Token | 's |
Token ID | 338 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
jewelry | +0.084 |
fault | +0.081 |
endant | +0.080 |
necklace | +0.071 |
Majesty | +0.070 |
Neg logit contributions | |
Run | -0.157 |
bel | -0.151 |
Quick | -0.147 |
he | -0.144 |
distance | -0.143 |
Token | go |
Token ID | 467 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
jewelry | +0.325 |
fault | +0.316 |
endant | +0.313 |
necklace | +0.296 |
Majesty | +0.294 |
Neg logit contributions | |
Run | -0.211 |
bel | -0.202 |
Quick | -0.186 |
he | -0.182 |
distance | -0.181 |
Token | to |
Token ID | 284 |
Feature activation | +2.74e-02 |
Pos logit contributions | |
Run | +0.197 |
bel | +0.191 |
Quick | +0.180 |
distance | +0.176 |
he | +0.175 |
Neg logit contributions | |
jewelry | -0.187 |
fault | -0.178 |
endant | -0.175 |
necklace | -0.166 |
Majesty | -0.164 |
Token | the |
Token ID | 262 |
Feature activation | -3.07e-02 |
Pos logit contributions | |
endant | +0.069 |
fault | +0.068 |
jewelry | +0.068 |
Majesty | +0.064 |
necklace | +0.063 |
Neg logit contributions | |
Run | -0.030 |
bel | -0.028 |
distance | -0.027 |
he | -0.027 |
Quick | -0.025 |
Token | men |
Token ID | 1450 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
distance | +0.049 |
Run | +0.049 |
bel | +0.048 |
Quick | +0.046 |
he | +0.046 |
Neg logit contributions | |
fault | -0.062 |
endant | -0.060 |
jewelry | -0.057 |
Majesty | -0.055 |
brace | -0.053 |
Token | 's |
Token ID | 338 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
Run | +0.555 |
bel | +0.542 |
he | +0.503 |
distance | +0.498 |
Quick | +0.498 |
Neg logit contributions | |
jewelry | -0.604 |
fault | -0.584 |
endant | -0.574 |
Majesty | -0.541 |
necklace | -0.541 |
Token | store |
Token ID | 3650 |
Feature activation | +1.17e-01 |
Pos logit contributions | |
fault | +0.265 |
endant | +0.264 |
jewelry | +0.254 |
Majesty | +0.234 |
brace | +0.234 |
Neg logit contributions | |
bel | -0.256 |
Run | -0.242 |
distance | -0.238 |
roads | -0.233 |
Quick | -0.227 |
Token | ," |
Token ID | 553 |
Feature activation | +1.33e-02 |
Pos logit contributions | |
jewelry | +0.188 |
fault | +0.177 |
endant | +0.173 |
necklace | +0.165 |
brace | +0.158 |
Neg logit contributions | |
Run | -0.252 |
bel | -0.239 |
Quick | -0.234 |
Uh | -0.230 |
Mark | -0.223 |
Token | mom |
Token ID | 1995 |
Feature activation | -4.61e-02 |
Pos logit contributions | |
jewelry | +0.021 |
fault | +0.019 |
endant | +0.019 |
necklace | +0.018 |
Majesty | +0.017 |
Neg logit contributions | |
Run | -0.029 |
bel | -0.028 |
Quick | -0.027 |
distance | -0.026 |
Uh | -0.026 |
Token | said |
Token ID | 531 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
Run | +0.092 |
bel | +0.089 |
Quick | +0.085 |
distance | +0.084 |
Uh | +0.083 |
Neg logit contributions | |
jewelry | -0.080 |
endant | -0.074 |
fault | -0.074 |
necklace | -0.070 |
brace | -0.068 |
Token | They |
Token ID | 1119 |
Feature activation | -6.52e-02 |
Pos logit contributions | |
bel | +0.006 |
Run | +0.006 |
amer | +0.006 |
distance | +0.006 |
aming | +0.006 |
Neg logit contributions | |
jewelry | -0.005 |
fault | -0.005 |
endant | -0.005 |
Ros | -0.005 |
brace | -0.005 |
Token | also |
Token ID | 635 |
Feature activation | -3.83e-03 |
Pos logit contributions | |
Run | +0.113 |
bel | +0.107 |
roads | +0.101 |
Quick | +0.100 |
distance | +0.100 |
Neg logit contributions | |
jewelry | -0.127 |
endant | -0.117 |
necklace | -0.115 |
fault | -0.112 |
collection | -0.111 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
Run | +0.007 |
bel | +0.006 |
he | +0.006 |
Quick | +0.006 |
Mark | +0.006 |
Neg logit contributions | |
jewelry | -0.007 |
endant | -0.007 |
necklace | -0.007 |
collection | -0.006 |
fault | -0.006 |
Token | some |
Token ID | 617 |
Feature activation | +3.71e-02 |
Pos logit contributions | |
Run | +0.149 |
bel | +0.141 |
Quick | +0.135 |
Uh | +0.131 |
he | +0.130 |
Neg logit contributions | |
jewelry | -0.170 |
fault | -0.162 |
endant | -0.158 |
necklace | -0.152 |
Majesty | -0.149 |
Token | wealthy |
Token ID | 11574 |
Feature activation | +1.38e-02 |
Pos logit contributions | |
jewelry | +0.074 |
fault | +0.067 |
endant | +0.065 |
necklace | +0.065 |
brace | +0.062 |
Neg logit contributions | |
Run | -0.068 |
bel | -0.063 |
Quick | -0.062 |
Uh | -0.060 |
he | -0.060 |
Token | men |
Token ID | 1450 |
Feature activation | -3.14e-01 |
Pos logit contributions | |
jewelry | +0.031 |
fault | +0.029 |
endant | +0.028 |
necklace | +0.027 |
brace | +0.027 |
Neg logit contributions | |
Run | -0.022 |
bel | -0.020 |
Quick | -0.020 |
Uh | -0.019 |
he | -0.019 |
Token | in |
Token ID | 287 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
Run | +0.551 |
bel | +0.528 |
Quick | +0.503 |
Uh | +0.488 |
distance | +0.480 |
Neg logit contributions | |
jewelry | -0.621 |
fault | -0.589 |
endant | -0.582 |
necklace | -0.558 |
brace | -0.541 |
Token | suits |
Token ID | 14803 |
Feature activation | -6.08e-02 |
Pos logit contributions | |
Run | +0.065 |
bel | +0.062 |
Quick | +0.060 |
Uh | +0.059 |
he | +0.056 |
Neg logit contributions | |
jewelry | -0.067 |
fault | -0.059 |
necklace | -0.058 |
sewing | -0.056 |
endant | -0.056 |
Token | , |
Token ID | 11 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
Run | +0.126 |
Quick | +0.120 |
bel | +0.119 |
Uh | +0.117 |
Mark | +0.112 |
Neg logit contributions | |
jewelry | -0.102 |
fault | -0.096 |
endant | -0.093 |
necklace | -0.090 |
collection | -0.085 |
Token | talking |
Token ID | 3375 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
jewelry | +0.104 |
fault | +0.097 |
endant | +0.095 |
necklace | +0.092 |
brace | +0.090 |
Neg logit contributions | |
Run | -0.095 |
bel | -0.090 |
Quick | -0.088 |
he | -0.085 |
Uh | -0.085 |
Token | on |
Token ID | 319 |
Feature activation | -3.95e-02 |
Pos logit contributions | |
Run | +0.039 |
Quick | +0.037 |
Uh | +0.037 |
bel | +0.035 |
he | +0.034 |
Neg logit contributions | |
jewelry | -0.033 |
fault | -0.030 |
endant | -0.029 |
necklace | -0.028 |
brace | -0.027 |
Token | hats |
Token ID | 23910 |
Feature activation | -6.86e-02 |
Pos logit contributions | |
jewelry | +0.076 |
fault | +0.072 |
endant | +0.071 |
necklace | +0.067 |
Majesty | +0.066 |
Neg logit contributions | |
Run | -0.086 |
bel | -0.083 |
Quick | -0.078 |
Uh | -0.076 |
distance | -0.076 |
Token | and |
Token ID | 290 |
Feature activation | -1.28e-02 |
Pos logit contributions | |
Run | +0.117 |
bel | +0.113 |
he | +0.110 |
distance | +0.106 |
Mark | +0.101 |
Neg logit contributions | |
jewelry | -0.133 |
endant | -0.127 |
fault | -0.127 |
Majesty | -0.125 |
necklace | -0.119 |
Token | coats |
Token ID | 30720 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
Run | +0.023 |
bel | +0.022 |
Quick | +0.021 |
distance | +0.020 |
Mark | +0.020 |
Neg logit contributions | |
jewelry | -0.024 |
fault | -0.024 |
endant | -0.024 |
Majesty | -0.022 |
necklace | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -2.37e-03 |
Pos logit contributions | |
Run | +0.199 |
bel | +0.194 |
Quick | +0.179 |
distance | +0.179 |
he | +0.179 |
Neg logit contributions | |
jewelry | -0.181 |
fault | -0.173 |
endant | -0.171 |
Majesty | -0.161 |
necklace | -0.161 |
Token | The |
Token ID | 383 |
Feature activation | +3.72e-02 |
Pos logit contributions | |
bel | +0.005 |
Run | +0.004 |
aming | +0.004 |
amed | +0.004 |
he | +0.004 |
Neg logit contributions | |
jewelry | -0.004 |
fault | -0.004 |
endant | -0.004 |
Ros | -0.003 |
necklace | -0.003 |
Token | men |
Token ID | 1450 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
jewelry | +0.052 |
necklace | +0.049 |
fault | +0.045 |
collection | +0.043 |
fridge | +0.042 |
Neg logit contributions | |
bel | -0.082 |
Run | -0.082 |
he | -0.081 |
aming | -0.079 |
umbling | -0.077 |
Token | were |
Token ID | 547 |
Feature activation | -5.95e-02 |
Pos logit contributions | |
Run | +0.536 |
bel | +0.522 |
Quick | +0.517 |
Uh | +0.496 |
Mark | +0.495 |
Neg logit contributions | |
jewelry | -0.601 |
endant | -0.557 |
necklace | -0.548 |
brace | -0.517 |
cand | -0.511 |
Token | firefighters |
Token ID | 22611 |
Feature activation | -1.60e-01 |
Pos logit contributions | |
Run | +0.123 |
Quick | +0.121 |
bel | +0.115 |
Uh | +0.113 |
he | +0.110 |
Neg logit contributions | |
jewelry | -0.103 |
sewing | -0.090 |
necklace | -0.089 |
fault | -0.089 |
brace | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -3.11e-03 |
Pos logit contributions | |
Run | +0.341 |
Quick | +0.328 |
bel | +0.320 |
Uh | +0.320 |
Mark | +0.299 |
Neg logit contributions | |
jewelry | -0.256 |
fault | -0.243 |
necklace | -0.227 |
endant | -0.227 |
collection | -0.215 |
Token |
|
Token ID | 198 |
Feature activation | -1.91e-02 |
Pos logit contributions | |
bel | +0.007 |
amed | +0.006 |
Run | +0.006 |
aming | +0.006 |
Quick | +0.006 |
Neg logit contributions | |
fault | -0.004 |
jewelry | -0.004 |
endant | -0.004 |
Ros | -0.004 |
Teddy | -0.003 |
Token |
|
Token ID | 198 |
Feature activation | -1.96e-02 |
Pos logit contributions | |
Run | +0.044 |
bel | +0.044 |
Uh | +0.043 |
folk | +0.041 |
amer | +0.040 |
Neg logit contributions | |
jewelry | -0.020 |
fault | -0.020 |
endant | -0.019 |
Ros | -0.019 |
sewing | -0.018 |
Token | junk |
Token ID | 18556 |
Feature activation | +1.27e-01 |
Pos logit contributions | |
Run | +0.200 |
bel | +0.187 |
Quick | +0.181 |
Uh | +0.178 |
distance | +0.177 |
Neg logit contributions | |
fault | -0.205 |
endant | -0.192 |
Majesty | -0.190 |
jewelry | -0.178 |
collection | -0.169 |
Token | on |
Token ID | 319 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
jewelry | +0.303 |
fault | +0.294 |
endant | +0.290 |
necklace | +0.278 |
Majesty | +0.278 |
Neg logit contributions | |
Run | -0.167 |
bel | -0.158 |
Quick | -0.144 |
distance | -0.139 |
he | -0.139 |
Token | the |
Token ID | 262 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
Run | +0.051 |
bel | +0.048 |
Quick | +0.045 |
distance | +0.044 |
Uh | +0.043 |
Neg logit contributions | |
jewelry | -0.079 |
fault | -0.076 |
endant | -0.075 |
necklace | -0.072 |
Majesty | -0.071 |
Token | ground |
Token ID | 2323 |
Feature activation | -1.04e-02 |
Pos logit contributions | |
Run | +0.048 |
bel | +0.047 |
Quick | +0.044 |
distance | +0.043 |
he | +0.042 |
Neg logit contributions | |
jewelry | -0.061 |
fault | -0.060 |
endant | -0.059 |
Majesty | -0.055 |
necklace | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -3.88e-04 |
Pos logit contributions | |
Run | +0.022 |
bel | +0.021 |
Quick | +0.021 |
Uh | +0.020 |
Mark | +0.020 |
Neg logit contributions | |
jewelry | -0.016 |
fault | -0.016 |
endant | -0.015 |
necklace | -0.014 |
brace | -0.014 |
Token | Blue |
Token ID | 4518 |
Feature activation | +1.99e-01 |
Pos logit contributions | |
Run | +0.001 |
bel | +0.001 |
Quick | +0.001 |
Uh | +0.001 |
distance | +0.001 |
Neg logit contributions | |
jewelry | -0.001 |
fault | -0.001 |
endant | -0.001 |
necklace | -0.001 |
Majesty | -0.001 |
Token | flew |
Token ID | 13112 |
Feature activation | +3.17e-02 |
Pos logit contributions | |
fault | +0.420 |
jewelry | +0.417 |
endant | +0.395 |
Majesty | +0.379 |
necklace | +0.375 |
Neg logit contributions | |
Run | -0.327 |
bel | -0.308 |
Quick | -0.298 |
Uh | -0.281 |
distance | -0.277 |
Token | down |
Token ID | 866 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
jewelry | +0.056 |
fault | +0.052 |
necklace | +0.051 |
endant | +0.050 |
Ros | +0.050 |
Neg logit contributions | |
Run | -0.062 |
Quick | -0.059 |
Uh | -0.057 |
bel | -0.056 |
Mark | -0.055 |
Token | to |
Token ID | 284 |
Feature activation | +2.89e-02 |
Pos logit contributions | |
Run | +0.279 |
bel | +0.268 |
Quick | +0.262 |
Uh | +0.259 |
roads | +0.251 |
Neg logit contributions | |
jewelry | -0.170 |
fault | -0.150 |
necklace | -0.144 |
endant | -0.141 |
Ros | -0.139 |
Token | see |
Token ID | 766 |
Feature activation | -8.13e-02 |
Pos logit contributions | |
jewelry | +0.063 |
fault | +0.063 |
endant | +0.062 |
Majesty | +0.058 |
necklace | +0.057 |
Neg logit contributions | |
Run | -0.044 |
bel | -0.042 |
Quick | -0.039 |
distance | -0.039 |
he | -0.038 |
Token | what |
Token ID | 644 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
Run | +0.126 |
Mark | +0.113 |
bel | +0.110 |
What | +0.109 |
he | +0.104 |
Neg logit contributions | |
fault | -0.173 |
endant | -0.165 |
Majesty | -0.162 |
collection | -0.159 |
jewelry | -0.151 |
Token | and |
Token ID | 290 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
jewelry | +0.098 |
fault | +0.093 |
endant | +0.092 |
necklace | +0.088 |
brace | +0.086 |
Neg logit contributions | |
Run | -0.085 |
bel | -0.082 |
Quick | -0.077 |
Uh | -0.075 |
distance | -0.075 |
Token | pulls |
Token ID | 16194 |
Feature activation | -7.22e-03 |
Pos logit contributions | |
Run | +0.024 |
bel | +0.023 |
Quick | +0.021 |
distance | +0.021 |
Uh | +0.020 |
Neg logit contributions | |
jewelry | -0.029 |
fault | -0.028 |
endant | -0.028 |
Majesty | -0.026 |
necklace | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
Run | +0.013 |
bel | +0.012 |
Quick | +0.011 |
Uh | +0.011 |
distance | +0.011 |
Neg logit contributions | |
jewelry | -0.014 |
fault | -0.014 |
endant | -0.013 |
necklace | -0.013 |
brace | -0.013 |
Token | They |
Token ID | 1119 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
Run | +0.014 |
bel | +0.013 |
Quick | +0.012 |
distance | +0.012 |
Uh | +0.012 |
Neg logit contributions | |
jewelry | -0.024 |
endant | -0.023 |
fault | -0.023 |
necklace | -0.022 |
Majesty | -0.021 |
Token | make |
Token ID | 787 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
Run | +0.106 |
bel | +0.101 |
Quick | +0.097 |
Uh | +0.094 |
Mark | +0.092 |
Neg logit contributions | |
jewelry | -0.118 |
fault | -0.113 |
endant | -0.111 |
necklace | -0.108 |
brace | -0.106 |
Token | bows |
Token ID | 37469 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
Run | +0.024 |
bel | +0.023 |
Quick | +0.021 |
distance | +0.021 |
Uh | +0.021 |
Neg logit contributions | |
jewelry | -0.029 |
fault | -0.028 |
endant | -0.028 |
necklace | -0.027 |
Majesty | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | -1.53e-02 |
Pos logit contributions | |
jewelry | +0.314 |
fault | +0.298 |
endant | +0.291 |
necklace | +0.283 |
brace | +0.275 |
Neg logit contributions | |
Run | -0.344 |
bel | -0.323 |
Quick | -0.319 |
Uh | -0.311 |
distance | -0.302 |
Token | t |
Token ID | 256 |
Feature activation | +1.12e-01 |
Pos logit contributions | |
Run | +0.032 |
bel | +0.031 |
Quick | +0.029 |
distance | +0.029 |
Uh | +0.028 |
Neg logit contributions | |
fault | -0.024 |
jewelry | -0.024 |
endant | -0.024 |
Majesty | -0.022 |
necklace | -0.020 |
Token | ucks |
Token ID | 6238 |
Feature activation | -8.23e-02 |
Pos logit contributions | |
jewelry | +0.184 |
endant | +0.178 |
fault | +0.175 |
brace | +0.159 |
necklace | +0.156 |
Neg logit contributions | |
Run | -0.243 |
distance | -0.221 |
Quick | -0.218 |
Uh | -0.212 |
roads | -0.208 |
Token | . |
Token ID | 13 |
Feature activation | -1.08e-02 |
Pos logit contributions | |
Run | +0.166 |
Quick | +0.157 |
bel | +0.156 |
Uh | +0.153 |
roads | +0.142 |
Neg logit contributions | |
jewelry | -0.146 |
fault | -0.141 |
necklace | -0.129 |
endant | -0.127 |
brace | -0.126 |
Token | They |
Token ID | 1119 |
Feature activation | -6.16e-02 |
Pos logit contributions | |
Run | +0.015 |
bel | +0.014 |
Quick | +0.013 |
distance | +0.012 |
Mark | +0.012 |
Neg logit contributions | |
jewelry | -0.026 |
endant | -0.025 |
fault | -0.025 |
necklace | -0.024 |
Majesty | -0.024 |
Token | drew |
Token ID | 9859 |
Feature activation | +1.58e-01 |
Pos logit contributions | |
jewelry | +0.101 |
fault | +0.099 |
endant | +0.096 |
Majesty | +0.091 |
necklace | +0.091 |
Neg logit contributions | |
Run | -0.089 |
bel | -0.086 |
Quick | -0.081 |
Uh | -0.079 |
distance | -0.077 |
Token | a |
Token ID | 257 |
Feature activation | +1.66e-02 |
Pos logit contributions | |
jewelry | +0.306 |
endant | +0.282 |
fault | +0.282 |
necklace | +0.273 |
brace | +0.265 |
Neg logit contributions | |
Run | -0.294 |
bel | -0.279 |
Quick | -0.269 |
Uh | -0.262 |
distance | -0.259 |
Token | big |
Token ID | 1263 |
Feature activation | -4.75e-02 |
Pos logit contributions | |
jewelry | +0.024 |
fault | +0.023 |
endant | +0.022 |
necklace | +0.021 |
Majesty | +0.020 |
Neg logit contributions | |
Run | -0.038 |
bel | -0.037 |
Quick | -0.035 |
Uh | -0.034 |
distance | -0.034 |
Token | heart |
Token ID | 2612 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
Run | +0.099 |
bel | +0.095 |
distance | +0.092 |
Uh | +0.090 |
he | +0.089 |
Neg logit contributions | |
fault | -0.076 |
jewelry | -0.071 |
endant | -0.070 |
Majesty | -0.065 |
brace | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
jewelry | +0.036 |
endant | +0.034 |
fault | +0.034 |
necklace | +0.033 |
brace | +0.032 |
Neg logit contributions | |
Run | -0.030 |
bel | -0.028 |
Quick | -0.027 |
Uh | -0.026 |
distance | -0.026 |
Token | wrote |
Token ID | 2630 |
Feature activation | +1.63e-01 |
Pos logit contributions | |
Run | +0.019 |
bel | +0.019 |
Quick | +0.017 |
Uh | +0.016 |
Mark | +0.016 |
Neg logit contributions | |
fault | -0.020 |
jewelry | -0.020 |
endant | -0.019 |
Majesty | -0.018 |
necklace | -0.018 |
Token | " |
Token ID | 366 |
Feature activation | -7.93e-02 |
Pos logit contributions | |
jewelry | +0.313 |
fault | +0.307 |
endant | +0.301 |
necklace | +0.281 |
Majesty | +0.280 |
Neg logit contributions | |
Run | -0.287 |
bel | -0.283 |
Quick | -0.258 |
Mark | -0.255 |
he | -0.255 |
Token | L |
Token ID | 43 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
Run | +0.118 |
Quick | +0.105 |
bel | +0.104 |
distance | +0.100 |
Uh | +0.097 |
Neg logit contributions | |
jewelry | -0.179 |
endant | -0.167 |
necklace | -0.161 |
brace | -0.156 |
fault | -0.155 |
Token | ily |
Token ID | 813 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
jewelry | +0.026 |
fault | +0.025 |
Majesty | +0.025 |
endant | +0.025 |
collection | +0.024 |
Neg logit contributions | |
Run | -0.020 |
he | -0.019 |
bel | -0.019 |
Mark | -0.018 |
aked | -0.018 |
Token | loves |
Token ID | 10408 |
Feature activation | +3.05e-02 |
Pos logit contributions | |
jewelry | +0.051 |
fault | +0.050 |
endant | +0.049 |
necklace | +0.047 |
Majesty | +0.046 |
Neg logit contributions | |
Run | -0.037 |
bel | -0.036 |
Quick | -0.033 |
he | -0.032 |
distance | -0.032 |
Token | Ben |
Token ID | 3932 |
Feature activation | -1.10e-03 |
Pos logit contributions | |
fault | +0.054 |
endant | +0.054 |
jewelry | +0.051 |
Majesty | +0.047 |
necklace | +0.047 |
Neg logit contributions | |
Run | -0.056 |
bel | -0.055 |
Mark | -0.051 |
Quick | -0.051 |
distance | -0.050 |
Token | his |
Token ID | 465 |
Feature activation | -1.99e-02 |
Pos logit contributions | |
endant | +0.047 |
Majesty | +0.045 |
jewelry | +0.045 |
fault | +0.044 |
necklace | +0.042 |
Neg logit contributions | |
bel | -0.022 |
Run | -0.022 |
distance | -0.021 |
Mark | -0.020 |
roads | -0.019 |
Token | cr |
Token ID | 1067 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
bel | +0.043 |
Run | +0.039 |
distance | +0.039 |
Quick | +0.037 |
Mark | +0.037 |
Neg logit contributions | |
endant | -0.026 |
Majesty | -0.026 |
fault | -0.025 |
jewelry | -0.024 |
cand | -0.022 |
Token | ay |
Token ID | 323 |
Feature activation | -9.03e-02 |
Pos logit contributions | |
jewelry | +0.348 |
fault | +0.337 |
endant | +0.334 |
necklace | +0.322 |
Majesty | +0.316 |
Neg logit contributions | |
Run | -0.147 |
bel | -0.134 |
Quick | -0.123 |
distance | -0.117 |
Uh | -0.117 |
Token | ons |
Token ID | 684 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
bel | +0.172 |
Run | +0.167 |
Mark | +0.158 |
he | +0.158 |
Quick | +0.155 |
Neg logit contributions | |
jewelry | -0.154 |
fault | -0.152 |
endant | -0.147 |
Majesty | -0.147 |
necklace | -0.138 |
Token | to |
Token ID | 284 |
Feature activation | +3.13e-02 |
Pos logit contributions | |
jewelry | +0.184 |
endant | +0.178 |
Majesty | +0.175 |
fault | +0.173 |
necklace | +0.163 |
Neg logit contributions | |
bel | -0.150 |
Run | -0.142 |
he | -0.138 |
distance | -0.136 |
Mark | -0.127 |
Token | draw |
Token ID | 3197 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
fault | +0.060 |
endant | +0.058 |
jewelry | +0.057 |
Majesty | +0.054 |
necklace | +0.051 |
Neg logit contributions | |
bel | -0.055 |
Run | -0.053 |
he | -0.051 |
distance | -0.050 |
Mark | -0.049 |
Token | a |
Token ID | 257 |
Feature activation | +1.41e-02 |
Pos logit contributions | |
jewelry | +0.333 |
fault | +0.328 |
endant | +0.322 |
Majesty | +0.304 |
necklace | +0.301 |
Neg logit contributions | |
Run | -0.266 |
bel | -0.264 |
he | -0.238 |
Quick | -0.236 |
distance | -0.234 |
Token | picture |
Token ID | 4286 |
Feature activation | +4.74e-02 |
Pos logit contributions | |
fault | +0.022 |
jewelry | +0.022 |
endant | +0.021 |
Majesty | +0.019 |
brace | +0.019 |
Neg logit contributions | |
Run | -0.029 |
bel | -0.029 |
distance | -0.027 |
Quick | -0.027 |
Uh | -0.027 |
Token | of |
Token ID | 286 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
jewelry | +0.102 |
fault | +0.098 |
endant | +0.097 |
necklace | +0.092 |
Majesty | +0.092 |
Neg logit contributions | |
Run | -0.073 |
bel | -0.071 |
Quick | -0.065 |
he | -0.064 |
distance | -0.063 |
Token | the |
Token ID | 262 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
Run | +0.146 |
bel | +0.146 |
distance | +0.138 |
Mark | +0.136 |
he | +0.129 |
Neg logit contributions | |
endant | -0.227 |
fault | -0.224 |
jewelry | -0.219 |
Majesty | -0.210 |
necklace | -0.201 |
Token | rain |
Token ID | 6290 |
Feature activation | +3.61e-02 |
Pos logit contributions | |
Run | +0.058 |
bel | +0.057 |
distance | +0.055 |
Quick | +0.054 |
Uh | +0.053 |
Neg logit contributions | |
fault | -0.049 |
endant | -0.048 |
jewelry | -0.048 |
Majesty | -0.044 |
brace | -0.043 |
Token | Sally |
Token ID | 25737 |
Feature activation | +3.53e-03 |
Pos logit contributions | |
Run | +0.009 |
bel | +0.009 |
Quick | +0.008 |
Uh | +0.008 |
distance | +0.008 |
Neg logit contributions | |
jewelry | -0.013 |
fault | -0.012 |
endant | -0.012 |
necklace | -0.012 |
Majesty | -0.011 |
Token | said |
Token ID | 531 |
Feature activation | +2.29e-02 |
Pos logit contributions | |
fault | +0.007 |
jewelry | +0.007 |
endant | +0.007 |
Majesty | +0.007 |
necklace | +0.007 |
Neg logit contributions | |
Run | -0.006 |
bel | -0.005 |
Quick | -0.005 |
he | -0.005 |
Mark | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +5.46e-02 |
Pos logit contributions | |
endant | +0.053 |
jewelry | +0.050 |
fault | +0.049 |
Majesty | +0.046 |
necklace | +0.046 |
Neg logit contributions | |
Run | -0.034 |
bel | -0.030 |
Quick | -0.027 |
Bob | -0.025 |
No | -0.025 |
Token | " |
Token ID | 366 |
Feature activation | -8.25e-02 |
Pos logit contributions | |
jewelry | +0.111 |
fault | +0.109 |
endant | +0.107 |
necklace | +0.101 |
Majesty | +0.099 |
Neg logit contributions | |
Run | -0.093 |
bel | -0.089 |
Quick | -0.084 |
Uh | -0.080 |
Mark | -0.080 |
Token | M |
Token ID | 44 |
Feature activation | -6.09e-02 |
Pos logit contributions | |
bel | +0.118 |
Run | +0.102 |
roads | +0.098 |
he | +0.095 |
distance | +0.094 |
Neg logit contributions | |
jewelry | -0.179 |
brace | -0.166 |
fault | -0.163 |
necklace | -0.160 |
endant | -0.160 |
Token | mm |
Token ID | 3020 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
Run | +0.134 |
bel | +0.127 |
Quick | +0.124 |
Uh | +0.121 |
he | +0.120 |
Neg logit contributions | |
jewelry | -0.090 |
fault | -0.087 |
necklace | -0.081 |
endant | -0.080 |
brace | -0.079 |
Token | , |
Token ID | 11 |
Feature activation | +5.43e-02 |
Pos logit contributions | |
fault | +0.338 |
jewelry | +0.322 |
endant | +0.322 |
sewing | +0.291 |
Majesty | +0.289 |
Neg logit contributions | |
Run | -0.281 |
bel | -0.242 |
he | -0.237 |
Quick | -0.235 |
Uh | -0.222 |
Token | this |
Token ID | 428 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
fault | +0.102 |
endant | +0.096 |
Majesty | +0.095 |
jewelry | +0.094 |
oppy | +0.086 |
Neg logit contributions | |
Run | -0.096 |
bel | -0.088 |
he | -0.087 |
Uh | -0.084 |
Mark | -0.082 |
Token | cake |
Token ID | 12187 |
Feature activation | +8.40e-02 |
Pos logit contributions | |
jewelry | +0.300 |
endant | +0.295 |
fault | +0.287 |
Ros | +0.282 |
Majesty | +0.280 |
Neg logit contributions | |
he | -0.326 |
Run | -0.323 |
bel | -0.318 |
distance | -0.301 |
Uh | -0.300 |
Token | is |
Token ID | 318 |
Feature activation | +4.79e-02 |
Pos logit contributions | |
jewelry | +0.138 |
fault | +0.134 |
endant | +0.131 |
Majesty | +0.124 |
brace | +0.119 |
Neg logit contributions | |
bel | -0.164 |
Run | -0.164 |
he | -0.155 |
Uh | -0.149 |
Quick | -0.149 |
Token | y |
Token ID | 331 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
jewelry | +0.093 |
fault | +0.093 |
endant | +0.092 |
Majesty | +0.089 |
necklace | +0.084 |
Neg logit contributions | |
Run | -0.076 |
bel | -0.075 |
distance | -0.068 |
Uh | -0.068 |
he | -0.067 |
Token |
|
Token ID | 198 |
Feature activation | -2.46e-02 |
Pos logit contributions | |
Run | +0.042 |
bel | +0.041 |
Uh | +0.037 |
Quick | +0.037 |
distance | +0.037 |
Neg logit contributions | |
jewelry | -0.047 |
fault | -0.046 |
endant | -0.045 |
necklace | -0.042 |
Majesty | -0.042 |
Token | The |
Token ID | 464 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
Run | +0.029 |
bel | +0.028 |
distance | +0.024 |
he | +0.023 |
roads | +0.022 |
Neg logit contributions | |
jewelry | -0.059 |
fault | -0.057 |
endant | -0.056 |
necklace | -0.055 |
Majesty | -0.055 |
Token | mother |
Token ID | 2802 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
Run | +0.046 |
bel | +0.044 |
Quick | +0.043 |
distance | +0.042 |
Uh | +0.042 |
Neg logit contributions | |
jewelry | -0.038 |
fault | -0.037 |
endant | -0.037 |
necklace | -0.033 |
Majesty | -0.033 |
Token | couldn |
Token ID | 3521 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
Run | +0.098 |
bel | +0.095 |
Quick | +0.088 |
distance | +0.086 |
Uh | +0.086 |
Neg logit contributions | |
jewelry | -0.108 |
fault | -0.102 |
endant | -0.102 |
necklace | -0.098 |
Majesty | -0.095 |
Token | 't |
Token ID | 470 |
Feature activation | +3.62e-02 |
Pos logit contributions | |
jewelry | +0.043 |
fault | +0.042 |
endant | +0.040 |
necklace | +0.039 |
brace | +0.038 |
Neg logit contributions | |
Run | -0.036 |
bel | -0.036 |
Quick | -0.033 |
distance | -0.032 |
Uh | -0.032 |
Token | take |
Token ID | 1011 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
jewelry | +0.071 |
fault | +0.068 |
endant | +0.067 |
necklace | +0.064 |
Majesty | +0.062 |
Neg logit contributions | |
Run | -0.064 |
bel | -0.062 |
Quick | -0.058 |
Uh | -0.057 |
distance | -0.056 |
Token | it |
Token ID | 340 |
Feature activation | +3.27e-02 |
Pos logit contributions | |
endant | +0.071 |
jewelry | +0.070 |
fault | +0.068 |
necklace | +0.065 |
Majesty | +0.065 |
Neg logit contributions | |
Run | -0.045 |
bel | -0.042 |
distance | -0.039 |
Mark | -0.039 |
Quick | -0.038 |
Token | anymore |
Token ID | 7471 |
Feature activation | +3.20e-02 |
Pos logit contributions | |
jewelry | +0.069 |
endant | +0.067 |
fault | +0.066 |
Majesty | +0.064 |
necklace | +0.063 |
Neg logit contributions | |
Run | -0.051 |
bel | -0.049 |
distance | -0.047 |
he | -0.046 |
Quick | -0.044 |
Token | . |
Token ID | 13 |
Feature activation | -4.46e-03 |
Pos logit contributions | |
endant | +0.064 |
jewelry | +0.064 |
fault | +0.061 |
Majesty | +0.060 |
necklace | +0.059 |
Neg logit contributions | |
Run | -0.052 |
bel | -0.051 |
distance | -0.048 |
he | -0.047 |
Mark | -0.046 |
Token | Ups |
Token ID | 35949 |
Feature activation | +2.66e-02 |
Pos logit contributions | |
Run | +0.007 |
bel | +0.007 |
Quick | +0.007 |
distance | +0.006 |
Mark | +0.006 |
Neg logit contributions | |
jewelry | -0.009 |
endant | -0.009 |
necklace | -0.009 |
fault | -0.009 |
brace | -0.009 |
Token | et |
Token ID | 316 |
Feature activation | -8.54e-02 |
Pos logit contributions | |
jewelry | +0.051 |
fault | +0.049 |
endant | +0.047 |
necklace | +0.045 |
brace | +0.044 |
Neg logit contributions | |
Run | -0.048 |
bel | -0.048 |
Quick | -0.044 |
Uh | -0.043 |
Mark | -0.042 |
Token | J |
Token ID | 41 |
Feature activation | -3.75e-02 |
Pos logit contributions | |
Run | +0.042 |
bel | +0.040 |
distance | +0.037 |
roads | +0.036 |
folk | +0.034 |
Neg logit contributions | |
jewelry | -0.056 |
necklace | -0.052 |
endant | -0.051 |
makeup | -0.051 |
brace | -0.051 |
Token | ill |
Token ID | 359 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
Run | +0.063 |
bel | +0.061 |
Quick | +0.057 |
he | +0.055 |
Uh | +0.055 |
Neg logit contributions | |
jewelry | -0.077 |
fault | -0.073 |
endant | -0.072 |
necklace | -0.070 |
Majesty | -0.069 |
Token | went |
Token ID | 1816 |
Feature activation | -6.25e-02 |
Pos logit contributions | |
Run | +0.127 |
bel | +0.122 |
Quick | +0.115 |
distance | +0.111 |
he | +0.111 |
Neg logit contributions | |
jewelry | -0.142 |
fault | -0.137 |
endant | -0.134 |
necklace | -0.127 |
Majesty | -0.126 |
Token | to |
Token ID | 284 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
Run | +0.119 |
bel | +0.112 |
Quick | +0.108 |
Uh | +0.107 |
Mark | +0.103 |
Neg logit contributions | |
jewelry | -0.117 |
fault | -0.109 |
endant | -0.107 |
necklace | -0.104 |
brace | -0.100 |
Token | find |
Token ID | 1064 |
Feature activation | +5.29e-02 |
Pos logit contributions | |
jewelry | +0.052 |
endant | +0.051 |
fault | +0.051 |
necklace | +0.048 |
Majesty | +0.047 |
Neg logit contributions | |
Run | -0.037 |
bel | -0.035 |
Quick | -0.033 |
distance | -0.032 |
he | -0.032 |
Token | another |
Token ID | 1194 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
jewelry | +0.104 |
fault | +0.101 |
endant | +0.100 |
Majesty | +0.094 |
necklace | +0.094 |
Neg logit contributions | |
Run | -0.092 |
bel | -0.089 |
Quick | -0.083 |
distance | -0.082 |
he | -0.081 |
Token | task |
Token ID | 4876 |
Feature activation | -1.01e-02 |
Pos logit contributions | |
jewelry | +0.053 |
fault | +0.050 |
endant | +0.050 |
necklace | +0.047 |
Majesty | +0.046 |
Neg logit contributions | |
Run | -0.058 |
bel | -0.056 |
Quick | -0.053 |
distance | -0.052 |
Uh | -0.052 |
Token | to |
Token ID | 284 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
Run | +0.019 |
bel | +0.019 |
Quick | +0.017 |
distance | +0.017 |
he | +0.017 |
Neg logit contributions | |
jewelry | -0.018 |
fault | -0.018 |
endant | -0.017 |
necklace | -0.016 |
Majesty | -0.016 |
Token | do |
Token ID | 466 |
Feature activation | -6.95e-02 |
Pos logit contributions | |
jewelry | +0.044 |
fault | +0.041 |
endant | +0.040 |
necklace | +0.039 |
Majesty | +0.038 |
Neg logit contributions | |
Run | -0.046 |
bel | -0.044 |
Quick | -0.042 |
Uh | -0.041 |
distance | -0.040 |
Token | . |
Token ID | 13 |
Feature activation | -6.50e-03 |
Pos logit contributions | |
bel | +0.118 |
Run | +0.116 |
he | +0.108 |
roads | +0.105 |
Mark | +0.105 |
Neg logit contributions | |
endant | -0.132 |
jewelry | -0.131 |
fault | -0.130 |
Majesty | -0.125 |
necklace | -0.119 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -9.54e-03 |
Pos logit contributions | |
Run | +0.009 |
bel | +0.009 |
Quick | +0.009 |
he | +0.008 |
Mark | +0.008 |
Neg logit contributions | |
jewelry | -0.014 |
endant | -0.014 |
fault | -0.014 |
necklace | -0.014 |
brace | -0.013 |
Token | his |
Token ID | 465 |
Feature activation | -2.33e-02 |
Pos logit contributions | |
jewelry | +0.034 |
fault | +0.033 |
endant | +0.033 |
Majesty | +0.031 |
necklace | +0.031 |
Neg logit contributions | |
Run | -0.027 |
bel | -0.026 |
Quick | -0.024 |
distance | -0.024 |
Mark | -0.023 |
Token | hammer |
Token ID | 15554 |
Feature activation | +6.93e-02 |
Pos logit contributions | |
bel | +0.046 |
Run | +0.045 |
Quick | +0.043 |
distance | +0.042 |
Mark | +0.041 |
Neg logit contributions | |
endant | -0.037 |
jewelry | -0.037 |
fault | -0.037 |
Majesty | -0.035 |
brace | -0.033 |
Token | when |
Token ID | 618 |
Feature activation | -6.44e-02 |
Pos logit contributions | |
jewelry | +0.131 |
fault | +0.125 |
endant | +0.124 |
necklace | +0.117 |
Majesty | +0.117 |
Neg logit contributions | |
Run | -0.125 |
bel | -0.123 |
Quick | -0.113 |
he | -0.112 |
distance | -0.112 |
Token | his |
Token ID | 465 |
Feature activation | -2.45e-02 |
Pos logit contributions | |
Run | +0.111 |
bel | +0.105 |
Quick | +0.099 |
Mark | +0.099 |
distance | +0.097 |
Neg logit contributions | |
jewelry | -0.127 |
endant | -0.126 |
fault | -0.125 |
Majesty | -0.116 |
necklace | -0.115 |
Token | friend |
Token ID | 1545 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
bel | +0.050 |
Run | +0.047 |
Quick | +0.046 |
distance | +0.044 |
Mark | +0.043 |
Neg logit contributions | |
endant | -0.038 |
jewelry | -0.038 |
fault | -0.035 |
brace | -0.034 |
Majesty | -0.033 |
Token | Billy |
Token ID | 15890 |
Feature activation | +5.24e-02 |
Pos logit contributions | |
Run | +0.033 |
bel | +0.031 |
Quick | +0.030 |
distance | +0.029 |
Mark | +0.029 |
Neg logit contributions | |
jewelry | -0.032 |
fault | -0.031 |
endant | -0.031 |
necklace | -0.028 |
Majesty | -0.028 |
Token | came |
Token ID | 1625 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
jewelry | +0.117 |
fault | +0.115 |
endant | +0.115 |
Majesty | +0.108 |
necklace | +0.107 |
Neg logit contributions | |
Run | -0.078 |
bel | -0.071 |
Quick | -0.067 |
distance | -0.066 |
he | -0.065 |
Token | over |
Token ID | 625 |
Feature activation | -6.12e-02 |
Pos logit contributions | |
Run | +0.084 |
bel | +0.079 |
Quick | +0.076 |
Uh | +0.076 |
he | +0.071 |
Neg logit contributions | |
jewelry | -0.106 |
fault | -0.104 |
endant | -0.095 |
necklace | -0.094 |
brace | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | -1.91e-03 |
Pos logit contributions | |
Run | +0.134 |
bel | +0.129 |
Quick | +0.126 |
Uh | +0.124 |
distance | +0.120 |
Neg logit contributions | |
jewelry | -0.095 |
fault | -0.089 |
endant | -0.083 |
necklace | -0.081 |
brace | -0.078 |
Token | Billy |
Token ID | 15890 |
Feature activation | +5.13e-02 |
Pos logit contributions | |
Run | +0.003 |
bel | +0.003 |
Quick | +0.003 |
Uh | +0.003 |
distance | +0.003 |
Neg logit contributions | |
jewelry | -0.004 |
fault | -0.004 |
endant | -0.003 |
necklace | -0.003 |
Majesty | -0.003 |
Token | wanted |
Token ID | 2227 |
Feature activation | -8.07e-03 |
Pos logit contributions | |
jewelry | +0.107 |
fault | +0.104 |
endant | +0.103 |
necklace | +0.096 |
Majesty | +0.096 |
Neg logit contributions | |
Run | -0.085 |
bel | -0.080 |
Quick | -0.077 |
Mark | -0.074 |
distance | -0.073 |
Token | . |
Token ID | 13 |
Feature activation | -8.11e-03 |
Pos logit contributions | |
Run | +0.063 |
bel | +0.061 |
distance | +0.056 |
Mark | +0.056 |
he | +0.055 |
Neg logit contributions | |
endant | -0.076 |
jewelry | -0.076 |
fault | -0.073 |
Majesty | -0.070 |
necklace | -0.068 |
Token |
|
Token ID | 198 |
Feature activation | -2.64e-02 |
Pos logit contributions | |
Run | +0.014 |
bel | +0.014 |
Quick | +0.013 |
Uh | +0.013 |
he | +0.013 |
Neg logit contributions | |
jewelry | -0.015 |
fault | -0.015 |
endant | -0.014 |
Ros | -0.014 |
necklace | -0.014 |
Token |
|
Token ID | 198 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
Run | +0.052 |
bel | +0.052 |
Uh | +0.049 |
Quick | +0.047 |
distance | +0.046 |
Neg logit contributions | |
jewelry | -0.042 |
fault | -0.041 |
endant | -0.040 |
Ros | -0.038 |
necklace | -0.037 |
Token | " |
Token ID | 1 |
Feature activation | -1.24e-01 |
Pos logit contributions | |
bel | +0.045 |
he | +0.044 |
folk | +0.042 |
Whe | +0.040 |
ained | +0.040 |
Neg logit contributions | |
jewelry | -0.043 |
Ros | -0.041 |
necklace | -0.039 |
Po | -0.038 |
bean | -0.038 |
Token | Sir |
Token ID | 22788 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
he | +0.146 |
bel | +0.142 |
umbling | +0.122 |
amed | +0.122 |
tim | +0.121 |
Neg logit contributions | |
jewelry | -0.288 |
necklace | -0.275 |
brace | -0.269 |
Ros | -0.254 |
endant | -0.253 |
Token | , |
Token ID | 11 |
Feature activation | +5.16e-02 |
Pos logit contributions | |
jewelry | +0.025 |
fault | +0.024 |
endant | +0.024 |
necklace | +0.022 |
brace | +0.022 |
Neg logit contributions | |
Run | -0.031 |
bel | -0.029 |
Quick | -0.028 |
Mark | -0.027 |
Uh | -0.027 |
Token | sir |
Token ID | 15967 |
Feature activation | -8.96e-02 |
Pos logit contributions | |
jewelry | +0.093 |
fault | +0.089 |
endant | +0.087 |
necklace | +0.083 |
Majesty | +0.081 |
Neg logit contributions | |
Run | -0.100 |
bel | -0.096 |
Quick | -0.091 |
Uh | -0.089 |
distance | -0.089 |
Token | , |
Token ID | 11 |
Feature activation | +5.15e-02 |
Pos logit contributions | |
Run | +0.193 |
bel | +0.188 |
Quick | +0.179 |
Uh | +0.175 |
distance | +0.174 |
Neg logit contributions | |
jewelry | -0.140 |
fault | -0.133 |
endant | -0.130 |
necklace | -0.123 |
Majesty | -0.120 |
Token | please |
Token ID | 3387 |
Feature activation | -6.93e-02 |
Pos logit contributions | |
jewelry | +0.087 |
fault | +0.082 |
endant | +0.080 |
necklace | +0.077 |
Majesty | +0.074 |
Neg logit contributions | |
Run | -0.105 |
bel | -0.103 |
Quick | -0.097 |
Uh | -0.095 |
distance | -0.095 |
Token | help |
Token ID | 1037 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
Run | +0.136 |
bel | +0.130 |
Quick | +0.122 |
distance | +0.120 |
Mark | +0.119 |
Neg logit contributions | |
jewelry | -0.124 |
fault | -0.119 |
endant | -0.118 |
necklace | -0.110 |
Majesty | -0.108 |
Token | us |
Token ID | 514 |
Feature activation | -5.53e-02 |
Pos logit contributions | |
jewelry | +0.086 |
necklace | +0.076 |
fault | +0.073 |
brace | +0.070 |
Ros | +0.069 |
Neg logit contributions | |
bel | -0.149 |
Run | -0.145 |
Uh | -0.143 |
Quick | -0.139 |
distance | -0.133 |
Token | swing |
Token ID | 9628 |
Feature activation | +8.77e-03 |
Pos logit contributions | |
jewelry | +0.061 |
fault | +0.057 |
necklace | +0.057 |
endant | +0.055 |
brace | +0.055 |
Neg logit contributions | |
Run | -0.051 |
bel | -0.051 |
Quick | -0.048 |
Uh | -0.047 |
Mark | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -4.83e-03 |
Pos logit contributions | |
jewelry | +0.014 |
fault | +0.014 |
necklace | +0.013 |
brace | +0.013 |
endant | +0.012 |
Neg logit contributions | |
Run | -0.018 |
Quick | -0.018 |
bel | -0.017 |
Uh | -0.017 |
aming | -0.016 |
Token | But |
Token ID | 887 |
Feature activation | +5.69e-02 |
Pos logit contributions | |
Run | +0.008 |
bel | +0.008 |
Quick | +0.008 |
distance | +0.007 |
Uh | +0.007 |
Neg logit contributions | |
jewelry | -0.009 |
fault | -0.009 |
endant | -0.009 |
necklace | -0.008 |
brace | -0.008 |
Token | the |
Token ID | 262 |
Feature activation | -2.67e-02 |
Pos logit contributions | |
jewelry | +0.102 |
fault | +0.098 |
endant | +0.093 |
necklace | +0.092 |
brace | +0.088 |
Neg logit contributions | |
Run | -0.108 |
bel | -0.108 |
Quick | -0.101 |
he | -0.098 |
Uh | -0.098 |
Token | swing |
Token ID | 9628 |
Feature activation | +8.28e-03 |
Pos logit contributions | |
bel | +0.050 |
Run | +0.050 |
Quick | +0.046 |
Mark | +0.046 |
he | +0.046 |
Neg logit contributions | |
jewelry | -0.049 |
fault | -0.047 |
necklace | -0.045 |
endant | -0.045 |
brace | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | +7.35e-04 |
Pos logit contributions | |
jewelry | +0.013 |
fault | +0.013 |
necklace | +0.013 |
endant | +0.012 |
brace | +0.012 |
Neg logit contributions | |
Run | -0.017 |
bel | -0.016 |
Quick | -0.016 |
Mark | -0.015 |
Uh | -0.015 |
Token | too |
Token ID | 1165 |
Feature activation | -7.02e-02 |
Pos logit contributions | |
jewelry | +0.001 |
fault | +0.001 |
endant | +0.001 |
necklace | +0.001 |
brace | +0.001 |
Neg logit contributions | |
Run | -0.001 |
bel | -0.001 |
Quick | -0.001 |
he | -0.001 |
Uh | -0.001 |
Token | cold |
Token ID | 4692 |
Feature activation | -5.69e-02 |
Pos logit contributions | |
bel | +0.117 |
Run | +0.114 |
roads | +0.110 |
Mark | +0.106 |
Quick | +0.106 |
Neg logit contributions | |
fault | -0.134 |
jewelry | -0.134 |
endant | -0.133 |
necklace | -0.125 |
brace | -0.120 |
Token | and |
Token ID | 290 |
Feature activation | -1.27e-02 |
Pos logit contributions | |
Run | +0.118 |
bel | +0.116 |
Quick | +0.112 |
Uh | +0.108 |
Mark | +0.106 |
Neg logit contributions | |
jewelry | -0.094 |
fault | -0.091 |
endant | -0.085 |
necklace | -0.082 |
brace | -0.079 |
Token | made |
Token ID | 925 |
Feature activation | -3.52e-02 |
Pos logit contributions | |
bel | +0.023 |
Run | +0.023 |
he | +0.022 |
Quick | +0.022 |
Mark | +0.022 |
Neg logit contributions | |
jewelry | -0.023 |
fault | -0.022 |
endant | -0.022 |
necklace | -0.021 |
brace | -0.021 |
Token | their |
Token ID | 511 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
Run | +0.065 |
bel | +0.063 |
Quick | +0.059 |
Uh | +0.058 |
distance | +0.057 |
Neg logit contributions | |
jewelry | -0.067 |
fault | -0.064 |
endant | -0.063 |
necklace | -0.060 |
brace | -0.058 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
Run | +0.041 |
Quick | +0.037 |
bel | +0.037 |
Uh | +0.036 |
distance | +0.036 |
Neg logit contributions | |
fault | -0.037 |
jewelry | -0.037 |
endant | -0.036 |
Majesty | -0.034 |
brace | -0.033 |
Token | looked |
Token ID | 3114 |
Feature activation | +7.04e-02 |
Pos logit contributions | |
fault | +0.087 |
jewelry | +0.083 |
endant | +0.078 |
Majesty | +0.077 |
collection | +0.072 |
Neg logit contributions | |
Run | -0.064 |
bel | -0.059 |
Quick | -0.056 |
Uh | -0.055 |
distance | -0.055 |
Token | at |
Token ID | 379 |
Feature activation | -7.12e-02 |
Pos logit contributions | |
jewelry | +0.169 |
fault | +0.169 |
endant | +0.164 |
Majesty | +0.160 |
necklace | +0.156 |
Neg logit contributions | |
Run | -0.090 |
bel | -0.080 |
distance | -0.074 |
he | -0.071 |
Quick | -0.071 |
Token | Anna |
Token ID | 11735 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
Run | +0.151 |
bel | +0.147 |
Quick | +0.139 |
Uh | +0.136 |
distance | +0.135 |
Neg logit contributions | |
jewelry | -0.116 |
fault | -0.110 |
endant | -0.107 |
necklace | -0.102 |
Majesty | -0.099 |
Token | and |
Token ID | 290 |
Feature activation | -1.33e-02 |
Pos logit contributions | |
fault | +0.048 |
jewelry | +0.047 |
endant | +0.047 |
Majesty | +0.047 |
majesty | +0.043 |
Neg logit contributions | |
Run | -0.029 |
he | -0.028 |
distance | -0.026 |
bel | -0.026 |
Mark | -0.024 |
Token | tilted |
Token ID | 37229 |
Feature activation | -2.07e-02 |
Pos logit contributions | |
Run | +0.021 |
bel | +0.018 |
Quick | +0.017 |
Uh | +0.017 |
distance | +0.016 |
Neg logit contributions | |
fault | -0.032 |
jewelry | -0.029 |
endant | -0.029 |
Majesty | -0.028 |
collection | -0.027 |
Token | its |
Token ID | 663 |
Feature activation | +5.41e-02 |
Pos logit contributions | |
Run | +0.038 |
bel | +0.037 |
distance | +0.034 |
Mark | +0.033 |
Quick | +0.033 |
Neg logit contributions | |
jewelry | -0.038 |
fault | -0.038 |
endant | -0.037 |
Majesty | -0.036 |
collection | -0.034 |
Token | head |
Token ID | 1182 |
Feature activation | -9.72e-02 |
Pos logit contributions | |
jewelry | +0.098 |
fault | +0.093 |
endant | +0.092 |
necklace | +0.086 |
Majesty | +0.086 |
Neg logit contributions | |
Run | -0.104 |
bel | -0.101 |
Quick | -0.094 |
distance | -0.093 |
Uh | -0.092 |
Token | . |
Token ID | 13 |
Feature activation | -7.02e-03 |
Pos logit contributions | |
Run | +0.167 |
bel | +0.161 |
he | +0.149 |
distance | +0.148 |
Quick | +0.141 |
Neg logit contributions | |
jewelry | -0.192 |
fault | -0.187 |
endant | -0.187 |
Majesty | -0.184 |
necklace | -0.170 |
Token | It |
Token ID | 632 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
Run | +0.011 |
bel | +0.010 |
he | +0.010 |
Quick | +0.010 |
Mark | +0.009 |
Neg logit contributions | |
jewelry | -0.015 |
fault | -0.015 |
endant | -0.015 |
collection | -0.014 |
Majesty | -0.014 |
Token | saw |
Token ID | 2497 |
Feature activation | -8.39e-02 |
Pos logit contributions | |
jewelry | +0.059 |
fault | +0.058 |
endant | +0.056 |
Majesty | +0.053 |
necklace | +0.053 |
Neg logit contributions | |
Run | -0.052 |
bel | -0.050 |
Quick | -0.047 |
Uh | -0.045 |
distance | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -2.97e-02 |
Pos logit contributions | |
Run | +0.064 |
bel | +0.063 |
Quick | +0.058 |
Uh | +0.057 |
he | +0.056 |
Neg logit contributions | |
jewelry | -0.079 |
fault | -0.076 |
endant | -0.074 |
necklace | -0.072 |
brace | -0.071 |
Token | ground |
Token ID | 2323 |
Feature activation | -1.60e-02 |
Pos logit contributions | |
Run | +0.056 |
bel | +0.054 |
Uh | +0.051 |
Quick | +0.051 |
he | +0.049 |
Neg logit contributions | |
jewelry | -0.057 |
fault | -0.053 |
necklace | -0.052 |
endant | -0.052 |
brace | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -5.29e-03 |
Pos logit contributions | |
Run | +0.033 |
bel | +0.032 |
Quick | +0.031 |
Uh | +0.030 |
Mark | +0.030 |
Neg logit contributions | |
jewelry | -0.026 |
fault | -0.025 |
endant | -0.024 |
necklace | -0.023 |
brace | -0.022 |
Token | L |
Token ID | 406 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
Run | +0.010 |
bel | +0.009 |
Quick | +0.009 |
distance | +0.008 |
Uh | +0.008 |
Neg logit contributions | |
jewelry | -0.010 |
fault | -0.010 |
endant | -0.010 |
necklace | -0.009 |
Majesty | -0.009 |
Token | ila |
Token ID | 10102 |
Feature activation | -1.02e-01 |
Pos logit contributions | |
Run | +0.026 |
bel | +0.026 |
Quick | +0.024 |
distance | +0.024 |
Uh | +0.024 |
Neg logit contributions | |
jewelry | -0.031 |
fault | -0.030 |
endant | -0.030 |
necklace | -0.028 |
Majesty | -0.027 |
Token | and |
Token ID | 290 |
Feature activation | -1.35e-02 |
Pos logit contributions | |
Run | +0.173 |
bel | +0.159 |
Quick | +0.153 |
he | +0.147 |
distance | +0.147 |
Neg logit contributions | |
jewelry | -0.212 |
fault | -0.200 |
endant | -0.200 |
necklace | -0.191 |
Majesty | -0.189 |
Token | Tom |
Token ID | 4186 |
Feature activation | -8.05e-02 |
Pos logit contributions | |
Run | +0.022 |
bel | +0.021 |
Quick | +0.019 |
Uh | +0.019 |
distance | +0.019 |
Neg logit contributions | |
jewelry | -0.029 |
fault | -0.028 |
endant | -0.028 |
necklace | -0.026 |
Majesty | -0.026 |
Token | stumble |
Token ID | 35174 |
Feature activation | -1.32e-02 |
Pos logit contributions | |
bel | +0.148 |
Run | +0.143 |
Quick | +0.139 |
Uh | +0.137 |
Mark | +0.130 |
Neg logit contributions | |
jewelry | -0.149 |
fault | -0.145 |
endant | -0.135 |
necklace | -0.135 |
brace | -0.134 |
Token | on |
Token ID | 319 |
Feature activation | -3.84e-02 |
Pos logit contributions | |
Run | +0.025 |
Quick | +0.024 |
Uh | +0.024 |
bel | +0.023 |
Mark | +0.021 |
Neg logit contributions | |
fault | -0.023 |
jewelry | -0.022 |
necklace | -0.020 |
brace | -0.020 |
endant | -0.020 |
Token | them |
Token ID | 606 |
Feature activation | -7.16e-02 |
Pos logit contributions | |
Run | +0.055 |
bel | +0.051 |
Quick | +0.047 |
distance | +0.047 |
Mark | +0.046 |
Neg logit contributions | |
jewelry | -0.088 |
endant | -0.085 |
fault | -0.085 |
Majesty | -0.081 |
necklace | -0.080 |
Token | . |
Token ID | 13 |
Feature activation | -5.18e-03 |
Pos logit contributions | |
Run | +0.133 |
bel | +0.129 |
he | +0.121 |
distance | +0.121 |
Quick | +0.118 |
Neg logit contributions | |
jewelry | -0.132 |
endant | -0.126 |
fault | -0.125 |
Majesty | -0.120 |
necklace | -0.117 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.09e-02 |
Pos logit contributions | |
fault | +0.106 |
jewelry | +0.103 |
endant | +0.098 |
Majesty | +0.095 |
necklace | +0.091 |
Neg logit contributions | |
Run | -0.073 |
bel | -0.070 |
Quick | -0.066 |
he | -0.064 |
Mark | -0.063 |
Token | to |
Token ID | 284 |
Feature activation | +2.78e-02 |
Pos logit contributions | |
endant | +0.061 |
fault | +0.058 |
jewelry | +0.056 |
Majesty | +0.055 |
necklace | +0.054 |
Neg logit contributions | |
bel | -0.052 |
Run | -0.048 |
Mark | -0.048 |
distance | -0.044 |
he | -0.044 |
Token | make |
Token ID | 787 |
Feature activation | -2.30e-02 |
Pos logit contributions | |
endant | +0.062 |
fault | +0.062 |
jewelry | +0.060 |
Majesty | +0.054 |
necklace | +0.054 |
Neg logit contributions | |
bel | -0.038 |
Run | -0.037 |
he | -0.037 |
Mark | -0.036 |
distance | -0.035 |
Token | poems |
Token ID | 31888 |
Feature activation | +1.09e-02 |
Pos logit contributions | |
bel | +0.040 |
Run | +0.039 |
he | +0.038 |
Mark | +0.036 |
roads | +0.035 |
Neg logit contributions | |
endant | -0.044 |
jewelry | -0.041 |
fault | -0.040 |
necklace | -0.037 |
Majesty | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -8.49e-03 |
Pos logit contributions | |
jewelry | +0.021 |
fault | +0.021 |
endant | +0.020 |
Majesty | +0.019 |
necklace | +0.019 |
Neg logit contributions | |
Run | -0.019 |
bel | -0.018 |
he | -0.017 |
distance | -0.017 |
Quick | -0.017 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
Run | +0.010 |
Mark | +0.009 |
bel | +0.009 |
Quick | +0.009 |
he | +0.009 |
Neg logit contributions | |
jewelry | -0.020 |
endant | -0.020 |
necklace | -0.020 |
fault | -0.020 |
collection | -0.019 |
Token | L |
Token ID | 43 |
Feature activation | +1.17e-02 |
Pos logit contributions | |
bel | +0.176 |
Run | +0.174 |
Quick | +0.159 |
Uh | +0.156 |
distance | +0.147 |
Neg logit contributions | |
jewelry | -0.261 |
fault | -0.242 |
endant | -0.240 |
necklace | -0.234 |
sewing | -0.230 |
Token | ily |
Token ID | 813 |
Feature activation | +2.03e-02 |
Pos logit contributions | |
jewelry | +0.022 |
fault | +0.022 |
endant | +0.021 |
Majesty | +0.021 |
necklace | +0.020 |
Neg logit contributions | |
he | -0.020 |
Run | -0.019 |
bel | -0.019 |
Quick | -0.018 |
distance | -0.018 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
fault | +0.047 |
jewelry | +0.047 |
endant | +0.046 |
Majesty | +0.044 |
necklace | +0.043 |
Neg logit contributions | |
Run | -0.026 |
he | -0.024 |
bel | -0.024 |
Quick | -0.023 |
distance | -0.023 |
Token | to |
Token ID | 284 |
Feature activation | +2.72e-02 |
Pos logit contributions | |
bel | +0.012 |
distance | +0.011 |
Run | +0.011 |
across | +0.010 |
For | +0.010 |
Neg logit contributions | |
Majesty | -0.025 |
endant | -0.025 |
majesty | -0.023 |
fault | -0.023 |
ig | -0.022 |
Token | help |
Token ID | 1037 |
Feature activation | +6.44e-02 |
Pos logit contributions | |
jewelry | +0.053 |
fault | +0.053 |
endant | +0.052 |
Majesty | +0.048 |
necklace | +0.047 |
Neg logit contributions | |
Run | -0.046 |
bel | -0.045 |
he | -0.042 |
distance | -0.042 |
Quick | -0.042 |
Token | a |
Token ID | 257 |
Feature activation | +6.55e-03 |
Pos logit contributions | |
fault | +0.115 |
endant | +0.110 |
jewelry | +0.107 |
Majesty | +0.105 |
brace | +0.103 |
Neg logit contributions | |
Run | -0.074 |
Mark | -0.060 |
Quick | -0.060 |
bel | -0.058 |
Uh | -0.058 |
Token | knife |
Token ID | 9845 |
Feature activation | +2.68e-02 |
Pos logit contributions | |
fault | +0.010 |
jewelry | +0.009 |
endant | +0.009 |
brace | +0.008 |
Majesty | +0.008 |
Neg logit contributions | |
Run | -0.014 |
Uh | -0.014 |
bel | -0.013 |
Quick | -0.013 |
distance | -0.013 |
Token | !" |
Token ID | 2474 |
Feature activation | -4.27e-02 |
Pos logit contributions | |
jewelry | +0.049 |
fault | +0.047 |
endant | +0.046 |
necklace | +0.044 |
Majesty | +0.043 |
Neg logit contributions | |
Run | -0.051 |
bel | -0.049 |
Quick | -0.047 |
distance | -0.045 |
Uh | -0.045 |
Token | Tom |
Token ID | 4186 |
Feature activation | -8.63e-02 |
Pos logit contributions | |
bel | +0.095 |
amer | +0.093 |
Run | +0.093 |
distance | +0.091 |
Quick | +0.090 |
Neg logit contributions | |
jewelry | -0.061 |
brace | -0.052 |
makeup | -0.051 |
necklace | -0.051 |
endant | -0.051 |
Token | said |
Token ID | 531 |
Feature activation | +2.16e-02 |
Pos logit contributions | |
Run | +0.174 |
bel | +0.169 |
Quick | +0.160 |
distance | +0.158 |
Uh | +0.156 |
Neg logit contributions | |
jewelry | -0.147 |
endant | -0.136 |
fault | -0.135 |
necklace | -0.129 |
brace | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | -8.71e-03 |
Pos logit contributions | |
jewelry | +0.038 |
fault | +0.036 |
endant | +0.036 |
necklace | +0.033 |
Majesty | +0.033 |
Neg logit contributions | |
Run | -0.043 |
bel | -0.042 |
Quick | -0.039 |
distance | -0.038 |
he | -0.038 |
Token | " |
Token ID | 366 |
Feature activation | -8.51e-02 |
Pos logit contributions | |
Run | +0.018 |
amer | +0.018 |
bel | +0.018 |
aming | +0.017 |
hing | +0.017 |
Neg logit contributions | |
jewelry | -0.012 |
Ros | -0.011 |
endant | -0.010 |
fault | -0.010 |
brace | -0.010 |
Token | Let |
Token ID | 5756 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
Run | +0.116 |
bel | +0.115 |
roads | +0.109 |
distance | +0.107 |
folk | +0.100 |
Neg logit contributions | |
jewelry | -0.185 |
endant | -0.170 |
necklace | -0.168 |
brace | -0.166 |
crystal | -0.158 |
Token | 's |
Token ID | 338 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
jewelry | +0.019 |
fault | +0.017 |
endant | +0.016 |
necklace | +0.016 |
brace | +0.015 |
Neg logit contributions | |
bel | -0.049 |
Run | -0.049 |
Quick | -0.046 |
Uh | -0.045 |
distance | -0.045 |
Token | play |
Token ID | 711 |
Feature activation | +7.73e-02 |
Pos logit contributions | |
fault | +0.292 |
jewelry | +0.284 |
endant | +0.283 |
Majesty | +0.260 |
brace | +0.258 |
Neg logit contributions | |
Run | -0.235 |
bel | -0.222 |
he | -0.211 |
Quick | -0.209 |
distance | -0.203 |
Token | with |
Token ID | 351 |
Feature activation | +1.13e-02 |
Pos logit contributions | |
jewelry | +0.151 |
endant | +0.139 |
fault | +0.139 |
necklace | +0.137 |
Majesty | +0.130 |
Neg logit contributions | |
Run | -0.140 |
bel | -0.134 |
Quick | -0.128 |
distance | -0.126 |
Uh | -0.125 |
Token | his |
Token ID | 465 |
Feature activation | -2.43e-02 |
Pos logit contributions | |
Run | +0.056 |
bel | +0.054 |
Quick | +0.050 |
Mark | +0.049 |
he | +0.049 |
Neg logit contributions | |
jewelry | -0.067 |
endant | -0.065 |
fault | -0.064 |
necklace | -0.060 |
Majesty | -0.059 |
Token | rh |
Token ID | 9529 |
Feature activation | -9.14e-03 |
Pos logit contributions | |
Run | +0.043 |
bel | +0.042 |
distance | +0.039 |
Quick | +0.039 |
Uh | +0.038 |
Neg logit contributions | |
jewelry | -0.047 |
fault | -0.045 |
endant | -0.045 |
necklace | -0.041 |
Majesty | -0.041 |
Token | in |
Token ID | 259 |
Feature activation | -8.29e-03 |
Pos logit contributions | |
Run | +0.020 |
bel | +0.019 |
distance | +0.019 |
Quick | +0.018 |
Uh | +0.018 |
Neg logit contributions | |
jewelry | -0.014 |
necklace | -0.013 |
endant | -0.013 |
fault | -0.012 |
brace | -0.012 |
Token | oc |
Token ID | 420 |
Feature activation | -5.09e-02 |
Pos logit contributions | |
Run | +0.015 |
Uh | +0.014 |
Quick | +0.014 |
bel | +0.013 |
roads | +0.013 |
Neg logit contributions | |
jewelry | -0.016 |
necklace | -0.015 |
fault | -0.015 |
endant | -0.015 |
collection | -0.014 |
Token | eros |
Token ID | 27498 |
Feature activation | -5.82e-02 |
Pos logit contributions | |
bel | +0.067 |
Run | +0.065 |
amer | +0.065 |
Mark | +0.060 |
umbling | +0.058 |
Neg logit contributions | |
jewelry | -0.112 |
endant | -0.110 |
fault | -0.108 |
necklace | -0.105 |
makeup | -0.102 |
Token | charge |
Token ID | 3877 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
Run | +0.097 |
bel | +0.095 |
Quick | +0.086 |
distance | +0.086 |
he | +0.084 |
Neg logit contributions | |
jewelry | -0.117 |
fault | -0.114 |
endant | -0.113 |
Majesty | -0.105 |
necklace | -0.105 |
Token | at |
Token ID | 379 |
Feature activation | -7.21e-02 |
Pos logit contributions | |
Run | +0.193 |
bel | +0.185 |
Quick | +0.174 |
Uh | +0.169 |
distance | +0.168 |
Neg logit contributions | |
jewelry | -0.263 |
fault | -0.252 |
endant | -0.248 |
necklace | -0.238 |
brace | -0.233 |
Token | Lily |
Token ID | 20037 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
Run | +0.103 |
bel | +0.100 |
distance | +0.093 |
Mark | +0.091 |
he | +0.090 |
Neg logit contributions | |
jewelry | -0.160 |
endant | -0.158 |
fault | -0.151 |
necklace | -0.146 |
Majesty | -0.145 |
Token | 's |
Token ID | 338 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
jewelry | +0.170 |
endant | +0.169 |
fault | +0.164 |
Majesty | +0.158 |
necklace | +0.151 |
Neg logit contributions | |
Run | -0.175 |
bel | -0.174 |
he | -0.164 |
distance | -0.163 |
Mark | -0.158 |
Token | rabbit |
Token ID | 22746 |
Feature activation | +1.62e-02 |
Pos logit contributions | |
endant | +0.283 |
jewelry | +0.271 |
fault | +0.255 |
makeup | +0.244 |
brace | +0.238 |
Neg logit contributions | |
distance | -0.230 |
bel | -0.225 |
Run | -0.216 |
Quick | -0.203 |
Ar | -0.201 |
Token | . |
Token ID | 13 |
Feature activation | -4.40e-03 |
Pos logit contributions | |
endant | +0.032 |
jewelry | +0.031 |
fault | +0.031 |
Majesty | +0.030 |
brace | +0.028 |
Neg logit contributions | |
Run | -0.026 |
bel | -0.025 |
distance | -0.024 |
he | -0.023 |
Mark | -0.023 |
Token | with |
Token ID | 351 |
Feature activation | +1.58e-02 |
Pos logit contributions | |
jewelry | +0.038 |
fault | +0.036 |
endant | +0.036 |
necklace | +0.035 |
Majesty | +0.034 |
Neg logit contributions | |
Run | -0.026 |
bel | -0.025 |
Quick | -0.023 |
Uh | -0.023 |
Mark | -0.022 |
Token | N |
Token ID | 399 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
endant | +0.032 |
Majesty | +0.032 |
fault | +0.031 |
jewelry | +0.029 |
majesty | +0.028 |
Neg logit contributions | |
distance | -0.023 |
bel | -0.022 |
he | -0.022 |
Run | -0.022 |
Mark | -0.021 |
Token | ana |
Token ID | 2271 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
jewelry | +0.038 |
necklace | +0.035 |
endant | +0.034 |
fault | +0.034 |
brace | +0.034 |
Neg logit contributions | |
Run | -0.030 |
bel | -0.030 |
he | -0.028 |
Quick | -0.026 |
Mark | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -3.98e-03 |
Pos logit contributions | |
bel | +0.259 |
Run | +0.239 |
he | +0.237 |
distance | +0.219 |
Quick | +0.208 |
Neg logit contributions | |
endant | -0.354 |
jewelry | -0.351 |
fault | -0.344 |
Majesty | -0.328 |
brace | -0.318 |
Token | N |
Token ID | 399 |
Feature activation | +1.80e-02 |
Pos logit contributions | |
Run | +0.006 |
bel | +0.006 |
distance | +0.005 |
he | +0.005 |
Quick | +0.005 |
Neg logit contributions | |
jewelry | -0.009 |
fault | -0.008 |
endant | -0.008 |
necklace | -0.008 |
brace | -0.008 |
Token | ana |
Token ID | 2271 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
jewelry | +0.040 |
fault | +0.037 |
endant | +0.037 |
necklace | +0.037 |
Majesty | +0.036 |
Neg logit contributions | |
Run | -0.026 |
bel | -0.026 |
he | -0.024 |
Quick | -0.023 |
Mark | -0.023 |
Token | was |
Token ID | 373 |
Feature activation | -2.99e-03 |
Pos logit contributions | |
Run | +0.234 |
bel | +0.225 |
Quick | +0.210 |
he | +0.207 |
Mark | +0.189 |
Neg logit contributions | |
fault | -0.387 |
jewelry | -0.384 |
endant | -0.376 |
Majesty | -0.355 |
brace | -0.349 |
Token | nice |
Token ID | 3621 |
Feature activation | +3.87e-02 |
Pos logit contributions | |
Run | +0.005 |
bel | +0.005 |
distance | +0.004 |
Quick | +0.004 |
he | +0.004 |
Neg logit contributions | |
jewelry | -0.006 |
fault | -0.006 |
endant | -0.006 |
necklace | -0.006 |
Majesty | -0.005 |
Token | , |
Token ID | 11 |
Feature activation | +5.28e-02 |
Pos logit contributions | |
jewelry | +0.082 |
endant | +0.082 |
fault | +0.080 |
Majesty | +0.078 |
brace | +0.076 |
Neg logit contributions | |
bel | -0.057 |
Run | -0.056 |
he | -0.052 |
distance | -0.052 |
Quick | -0.048 |
Token | but |
Token ID | 475 |
Feature activation | -2.27e-02 |
Pos logit contributions | |
Majesty | +0.120 |
fault | +0.119 |
endant | +0.118 |
jewelry | +0.116 |
collection | +0.110 |
Neg logit contributions | |
bel | -0.070 |
Run | -0.070 |
distance | -0.067 |
he | -0.061 |
Mark | -0.059 |
Token | L |
Token ID | 406 |
Feature activation | -1.31e-02 |
Pos logit contributions | |
Run | +0.041 |
bel | +0.038 |
Uh | +0.037 |
distance | +0.036 |
he | +0.036 |
Neg logit contributions | |
fault | -0.044 |
jewelry | -0.042 |
endant | -0.041 |
Majesty | -0.041 |
necklace | -0.038 |
Token | The |
Token ID | 383 |
Feature activation | +3.40e-02 |
Pos logit contributions | |
Run | +0.001 |
bel | +0.001 |
Quick | +0.001 |
distance | +0.001 |
he | +0.001 |
Neg logit contributions | |
jewelry | -0.001 |
fault | -0.001 |
endant | -0.001 |
necklace | -0.001 |
brace | -0.001 |
Token | cat |
Token ID | 3797 |
Feature activation | -2.87e-02 |
Pos logit contributions | |
fault | +0.053 |
jewelry | +0.052 |
necklace | +0.050 |
collection | +0.047 |
sewing | +0.044 |
Neg logit contributions | |
bel | -0.074 |
Run | -0.067 |
amer | -0.066 |
aming | -0.066 |
amed | -0.065 |
Token | wanted |
Token ID | 2227 |
Feature activation | -1.02e-02 |
Pos logit contributions | |
bel | +0.051 |
Run | +0.051 |
Quick | +0.046 |
Uh | +0.046 |
distance | +0.046 |
Neg logit contributions | |
jewelry | -0.055 |
endant | -0.051 |
fault | -0.051 |
necklace | -0.050 |
brace | -0.048 |
Token | to |
Token ID | 284 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
Run | +0.017 |
Mark | +0.016 |
distance | +0.016 |
bel | +0.016 |
Quick | +0.015 |
Neg logit contributions | |
endant | -0.021 |
jewelry | -0.020 |
Majesty | -0.020 |
fault | -0.019 |
necklace | -0.018 |
Token | speed |
Token ID | 2866 |
Feature activation | -4.17e-02 |
Pos logit contributions | |
jewelry | +0.043 |
fault | +0.041 |
necklace | +0.040 |
collection | +0.038 |
brace | +0.038 |
Neg logit contributions | |
bel | -0.051 |
Run | -0.050 |
Uh | -0.050 |
Quick | -0.048 |
amed | -0.048 |
Token | across |
Token ID | 1973 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
Run | +0.074 |
Quick | +0.073 |
Uh | +0.073 |
bel | +0.069 |
aming | +0.064 |
Neg logit contributions | |
jewelry | -0.080 |
fault | -0.078 |
necklace | -0.071 |
endant | -0.070 |
Ros | -0.070 |
Token | the |
Token ID | 262 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
bel | +0.313 |
Run | +0.308 |
Uh | +0.289 |
Quick | +0.286 |
he | +0.280 |
Neg logit contributions | |
jewelry | -0.347 |
fault | -0.321 |
necklace | -0.311 |
endant | -0.309 |
brace | -0.301 |
Token | road |
Token ID | 2975 |
Feature activation | -1.34e-01 |
Pos logit contributions | |
bel | +0.061 |
Run | +0.053 |
amer | +0.048 |
amed | +0.046 |
Uh | +0.046 |
Neg logit contributions | |
fault | -0.060 |
jewelry | -0.057 |
necklace | -0.054 |
collection | -0.052 |
endant | -0.050 |
Token | . |
Token ID | 13 |
Feature activation | -8.63e-04 |
Pos logit contributions | |
Run | +0.282 |
bel | +0.269 |
Quick | +0.263 |
Uh | +0.257 |
Mark | +0.248 |
Neg logit contributions | |
jewelry | -0.218 |
fault | -0.211 |
endant | -0.201 |
necklace | -0.193 |
brace | -0.185 |
Token | Zero |
Token ID | 12169 |
Feature activation | +1.24e-01 |
Pos logit contributions | |
Run | +0.002 |
bel | +0.002 |
Quick | +0.002 |
distance | +0.002 |
Uh | +0.002 |
Neg logit contributions | |
jewelry | -0.001 |
fault | -0.001 |
endant | -0.001 |
necklace | -0.001 |
Majesty | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -2.40e-03 |
Pos logit contributions | |
jewelry | +0.243 |
fault | +0.232 |
endant | +0.229 |
necklace | +0.219 |
Majesty | +0.214 |
Neg logit contributions | |
Run | -0.217 |
bel | -0.210 |
Quick | -0.196 |
Uh | -0.191 |
distance | -0.191 |
Token | shake |
Token ID | 13279 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
jewelry | +0.044 |
fault | +0.040 |
necklace | +0.040 |
endant | +0.040 |
brace | +0.038 |
Neg logit contributions | |
Run | -0.048 |
bel | -0.047 |
Quick | -0.044 |
Mark | -0.043 |
Uh | -0.043 |
Token | and |
Token ID | 290 |
Feature activation | -9.69e-03 |
Pos logit contributions | |
Run | +0.155 |
bel | +0.151 |
Quick | +0.146 |
Uh | +0.144 |
Mark | +0.139 |
Neg logit contributions | |
jewelry | -0.121 |
fault | -0.111 |
endant | -0.106 |
necklace | -0.105 |
brace | -0.102 |
Token | there |
Token ID | 612 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
Run | +0.016 |
bel | +0.016 |
Quick | +0.015 |
he | +0.015 |
Uh | +0.014 |
Neg logit contributions | |
jewelry | -0.020 |
fault | -0.019 |
endant | -0.018 |
necklace | -0.018 |
Majesty | -0.017 |
Token | was |
Token ID | 373 |
Feature activation | -6.39e-03 |
Pos logit contributions | |
jewelry | +0.051 |
fault | +0.050 |
necklace | +0.047 |
endant | +0.046 |
collection | +0.045 |
Neg logit contributions | |
Run | -0.057 |
bel | -0.057 |
Quick | -0.054 |
Uh | -0.053 |
distance | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | +9.44e-03 |
Pos logit contributions | |
Run | +0.013 |
bel | +0.013 |
he | +0.012 |
Quick | +0.012 |
Uh | +0.011 |
Neg logit contributions | |
jewelry | -0.011 |
fault | -0.011 |
endant | -0.010 |
necklace | -0.010 |
brace | -0.009 |
Token | loud |
Token ID | 7812 |
Feature activation | -1.82e-01 |
Pos logit contributions | |
jewelry | +0.019 |
fault | +0.018 |
necklace | +0.018 |
endant | +0.018 |
Majesty | +0.017 |
Neg logit contributions | |
Run | -0.016 |
bel | -0.016 |
he | -0.014 |
Quick | -0.014 |
Mark | -0.014 |
Token | noise |
Token ID | 7838 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
Run | +0.268 |
bel | +0.263 |
Quick | +0.242 |
Uh | +0.231 |
distance | +0.231 |
Neg logit contributions | |
jewelry | -0.406 |
fault | -0.388 |
endant | -0.381 |
necklace | -0.372 |
Majesty | -0.363 |
Token | . |
Token ID | 13 |
Feature activation | -4.78e-03 |
Pos logit contributions | |
Quick | +0.294 |
bel | +0.285 |
Uh | +0.284 |
Run | +0.282 |
amed | +0.280 |
Neg logit contributions | |
jewelry | -0.124 |
necklace | -0.102 |
makeup | -0.094 |
sewing | -0.093 |
fault | -0.090 |
Token | It |
Token ID | 632 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
Run | +0.008 |
bel | +0.008 |
Quick | +0.008 |
Uh | +0.007 |
distance | +0.007 |
Neg logit contributions | |
jewelry | -0.009 |
fault | -0.009 |
endant | -0.009 |
necklace | -0.008 |
Majesty | -0.008 |
Token | was |
Token ID | 373 |
Feature activation | -5.83e-03 |
Pos logit contributions | |
jewelry | +0.033 |
necklace | +0.032 |
bl | +0.031 |
crystal | +0.030 |
endant | +0.030 |
Neg logit contributions | |
bel | -0.052 |
he | -0.050 |
Run | -0.050 |
amer | -0.048 |
folk | -0.048 |
Token | an |
Token ID | 281 |
Feature activation | +5.55e-02 |
Pos logit contributions | |
Run | +0.012 |
bel | +0.012 |
Quick | +0.012 |
he | +0.012 |
roads | +0.011 |
Neg logit contributions | |
fault | -0.009 |
jewelry | -0.009 |
sewing | -0.008 |
Ros | -0.008 |
necklace | -0.008 |
Token | ae |
Token ID | 3609 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
Run | +0.099 |
bel | +0.092 |
Quick | +0.090 |
Uh | +0.088 |
distance | +0.086 |
Neg logit contributions | |
jewelry | -0.088 |
endant | -0.083 |
fault | -0.083 |
necklace | -0.079 |
Majesty | -0.077 |
Token | with |
Token ID | 351 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
bel | +0.186 |
Run | +0.184 |
distance | +0.172 |
he | +0.169 |
roads | +0.157 |
Neg logit contributions | |
fault | -0.237 |
jewelry | -0.236 |
endant | -0.231 |
Majesty | -0.225 |
brace | -0.211 |
Token | lots |
Token ID | 6041 |
Feature activation | +3.18e-02 |
Pos logit contributions | |
fault | +0.038 |
endant | +0.037 |
jewelry | +0.037 |
Majesty | +0.036 |
necklace | +0.034 |
Neg logit contributions | |
Run | -0.030 |
bel | -0.029 |
distance | -0.027 |
Quick | -0.027 |
he | -0.027 |
Token | of |
Token ID | 286 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
fault | +0.093 |
jewelry | +0.093 |
Majesty | +0.090 |
endant | +0.090 |
necklace | +0.087 |
Neg logit contributions | |
Run | -0.023 |
bel | -0.021 |
he | -0.019 |
distance | -0.018 |
Quick | -0.016 |
Token | topp |
Token ID | 23126 |
Feature activation | +5.37e-02 |
Pos logit contributions | |
Run | +0.222 |
bel | +0.215 |
Quick | +0.204 |
Uh | +0.200 |
distance | +0.199 |
Neg logit contributions | |
jewelry | -0.176 |
fault | -0.167 |
endant | -0.164 |
necklace | -0.155 |
Majesty | -0.151 |
Token | ings |
Token ID | 654 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
jewelry | +0.121 |
fault | +0.116 |
endant | +0.115 |
necklace | +0.110 |
Majesty | +0.109 |
Neg logit contributions | |
Run | -0.079 |
bel | -0.076 |
Quick | -0.069 |
he | -0.067 |
Uh | -0.067 |
Token | . |
Token ID | 13 |
Feature activation | -2.94e-04 |
Pos logit contributions | |
Run | +0.314 |
bel | +0.301 |
he | +0.283 |
Quick | +0.278 |
distance | +0.276 |
Neg logit contributions | |
jewelry | -0.311 |
fault | -0.303 |
endant | -0.301 |
Majesty | -0.286 |
necklace | -0.281 |
Token |
|
Token ID | 198 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
Run | +0.000 |
bel | +0.000 |
Quick | +0.000 |
he | +0.000 |
Uh | +0.000 |
Neg logit contributions | |
jewelry | -0.001 |
fault | -0.001 |
endant | -0.001 |
necklace | -0.001 |
Majesty | -0.001 |
Token |
|
Token ID | 198 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
Run | +0.030 |
bel | +0.028 |
Uh | +0.026 |
distance | +0.026 |
Quick | +0.026 |
Neg logit contributions | |
jewelry | -0.025 |
fault | -0.024 |
endant | -0.023 |
Ros | -0.022 |
necklace | -0.022 |
Token | They |
Token ID | 2990 |
Feature activation | -1.36e-01 |
Pos logit contributions | |
Run | +0.022 |
bel | +0.021 |
distance | +0.020 |
amer | +0.019 |
roads | +0.019 |
Neg logit contributions | |
jewelry | -0.034 |
sewing | -0.031 |
necklace | -0.031 |
makeup | -0.031 |
brace | -0.031 |
Token | sat |
Token ID | 3332 |
Feature activation | -2.36e-02 |
Pos logit contributions | |
Run | +0.227 |
bel | +0.217 |
Quick | +0.203 |
Uh | +0.201 |
distance | +0.197 |
Neg logit contributions | |
jewelry | -0.283 |
endant | -0.262 |
necklace | -0.256 |
fault | -0.253 |
collection | -0.249 |
Token | held |
Token ID | 2714 |
Feature activation | +3.34e-02 |
Pos logit contributions | |
Run | +0.104 |
bel | +0.101 |
Quick | +0.094 |
Uh | +0.092 |
distance | +0.092 |
Neg logit contributions | |
jewelry | -0.111 |
fault | -0.105 |
endant | -0.104 |
necklace | -0.100 |
brace | -0.097 |
Token | the |
Token ID | 262 |
Feature activation | -2.70e-02 |
Pos logit contributions | |
jewelry | +0.060 |
fault | +0.057 |
endant | +0.054 |
necklace | +0.053 |
brace | +0.052 |
Neg logit contributions | |
Run | -0.064 |
bel | -0.064 |
Quick | -0.060 |
Uh | -0.059 |
distance | -0.059 |
Token | stick |
Token ID | 4859 |
Feature activation | -1.05e-02 |
Pos logit contributions | |
Run | +0.053 |
bel | +0.051 |
Quick | +0.048 |
Uh | +0.047 |
he | +0.047 |
Neg logit contributions | |
jewelry | -0.048 |
fault | -0.046 |
endant | -0.045 |
necklace | -0.043 |
brace | -0.041 |
Token | in |
Token ID | 287 |
Feature activation | -3.06e-02 |
Pos logit contributions | |
Run | +0.016 |
bel | +0.016 |
distance | +0.015 |
he | +0.014 |
Quick | +0.014 |
Neg logit contributions | |
jewelry | -0.022 |
fault | -0.022 |
endant | -0.021 |
Majesty | -0.021 |
necklace | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
Run | +0.051 |
bel | +0.050 |
Quick | +0.046 |
Uh | +0.045 |
he | +0.045 |
Neg logit contributions | |
jewelry | -0.063 |
fault | -0.061 |
endant | -0.058 |
necklace | -0.057 |
brace | -0.056 |
Token | middle |
Token ID | 3504 |
Feature activation | -1.64e-01 |
Pos logit contributions | |
Run | +0.054 |
bel | +0.053 |
Quick | +0.050 |
he | +0.049 |
Mark | +0.049 |
Neg logit contributions | |
jewelry | -0.049 |
fault | -0.046 |
endant | -0.045 |
necklace | -0.044 |
Majesty | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
Run | +0.241 |
bel | +0.226 |
Quick | +0.214 |
Uh | +0.207 |
Mark | +0.201 |
Neg logit contributions | |
jewelry | -0.374 |
fault | -0.359 |
endant | -0.355 |
necklace | -0.343 |
brace | -0.333 |
Token | swung |
Token ID | 28507 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
Run | +0.017 |
bel | +0.017 |
Quick | +0.016 |
Uh | +0.015 |
distance | +0.015 |
Neg logit contributions | |
jewelry | -0.022 |
fault | -0.022 |
endant | -0.021 |
necklace | -0.020 |
Majesty | -0.020 |
Token | it |
Token ID | 340 |
Feature activation | +3.85e-02 |
Pos logit contributions | |
jewelry | +0.070 |
fault | +0.068 |
necklace | +0.065 |
brace | +0.063 |
endant | +0.063 |
Neg logit contributions | |
Run | -0.066 |
Quick | -0.063 |
bel | -0.063 |
Uh | -0.062 |
he | -0.058 |
Token | at |
Token ID | 379 |
Feature activation | -7.04e-02 |
Pos logit contributions | |
jewelry | +0.090 |
fault | +0.087 |
endant | +0.087 |
Majesty | +0.085 |
necklace | +0.082 |
Neg logit contributions | |
Run | -0.052 |
bel | -0.051 |
distance | -0.045 |
he | -0.044 |
Quick | -0.044 |
Token | the |
Token ID | 262 |
Feature activation | -2.65e-02 |
Pos logit contributions | |
Run | +0.116 |
bel | +0.113 |
Quick | +0.105 |
Uh | +0.103 |
he | +0.098 |
Neg logit contributions | |
jewelry | -0.147 |
fault | -0.143 |
endant | -0.137 |
necklace | -0.134 |
brace | -0.130 |