Token | a |
Token ID | 64 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
ths | +0.078 |
agg | +0.076 |
eking | +0.074 |
aman | +0.070 |
Kevin | +0.066 |
Neg logit contributions | |
. | -0.108 |
." | -0.103 |
ashes | -0.098 |
He | -0.096 |
rider | -0.094 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
agg | +0.190 |
eking | +0.189 |
ths | +0.188 |
ering | +0.179 |
Jess | +0.178 |
Neg logit contributions | |
ashes | -0.023 |
rider | -0.016 |
." | -0.014 |
Let | -0.012 |
He | -0.011 |
Token | |
Token ID | 220 |
Feature activation | +1.82e-01 |
Pos logit contributions | |
. | +0.174 |
." | +0.161 |
ashes | +0.156 |
rider | +0.143 |
He | +0.138 |
Neg logit contributions | |
ths | -0.314 |
agg | -0.307 |
eking | -0.297 |
aman | -0.294 |
Kevin | -0.284 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ths | +0.355 |
agg | +0.352 |
eking | +0.338 |
aman | +0.335 |
Kevin | +0.322 |
Neg logit contributions | |
. | -0.244 |
." | -0.229 |
ashes | -0.227 |
rider | -0.216 |
He | -0.206 |
Token | " |
Token ID | 1 |
Feature activation | +2.86e-02 |
Pos logit contributions | |
ths | +0.324 |
agg | +0.313 |
eking | +0.311 |
aman | +0.302 |
prisingly | +0.286 |
Neg logit contributions | |
. | -0.188 |
." | -0.172 |
He | -0.156 |
ashes | -0.153 |
They | -0.146 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.50e-01 |
Pos logit contributions | |
ths | +0.061 |
eking | +0.060 |
agg | +0.059 |
aman | +0.057 |
ering | +0.055 |
Neg logit contributions | |
. | -0.038 |
." | -0.035 |
He | -0.034 |
ashes | -0.032 |
Let | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
ths | +0.417 |
agg | +0.396 |
aman | +0.391 |
eking | +0.377 |
Kevin | +0.364 |
Neg logit contributions | |
. | -0.475 |
." | -0.446 |
He | -0.389 |
rider | -0.384 |
Let | -0.382 |
Token | what |
Token ID | 644 |
Feature activation | -2.91e-02 |
Pos logit contributions | |
ths | +0.175 |
agg | +0.170 |
eking | +0.166 |
aman | +0.162 |
Kevin | +0.155 |
Neg logit contributions | |
. | -0.124 |
." | -0.115 |
ashes | -0.110 |
He | -0.105 |
rider | -0.101 |
Token | are |
Token ID | 389 |
Feature activation | +1.02e-01 |
Pos logit contributions | |
ashes | +0.034 |
. | +0.034 |
." | +0.032 |
towards | +0.031 |
He | +0.030 |
Neg logit contributions | |
agg | -0.059 |
ths | -0.058 |
eking | -0.057 |
Kevin | -0.055 |
Jess | -0.054 |
Token | you |
Token ID | 345 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
ths | +0.160 |
agg | +0.155 |
eking | +0.149 |
aman | +0.147 |
Kevin | +0.138 |
Neg logit contributions | |
. | -0.194 |
." | -0.184 |
ashes | -0.169 |
He | -0.168 |
They | -0.163 |
Token | ?" |
Token ID | 1701 |
Feature activation | -2.05e-03 |
Pos logit contributions | |
ths | +0.015 |
agg | +0.014 |
Jess | +0.013 |
aman | +0.013 |
prisingly | +0.013 |
Neg logit contributions | |
. | -0.016 |
." | -0.015 |
like | -0.013 |
name | -0.013 |
play | -0.013 |
Token | on |
Token ID | 319 |
Feature activation | -5.57e-02 |
Pos logit contributions | |
rider | +0.000 |
." | +0.000 |
ashes | +0.000 |
He | +0.000 |
Let | +0.000 |
Neg logit contributions | |
ths | -0.001 |
eking | -0.001 |
agg | -0.001 |
aman | -0.001 |
ering | -0.001 |
Token | her |
Token ID | 607 |
Feature activation | -3.53e-02 |
Pos logit contributions | |
. | +0.080 |
." | +0.074 |
ashes | +0.067 |
He | +0.067 |
They | +0.064 |
Neg logit contributions | |
ths | -0.114 |
agg | -0.110 |
eking | -0.109 |
aman | -0.105 |
Kevin | -0.100 |
Token | costume |
Token ID | 16569 |
Feature activation | -5.03e-02 |
Pos logit contributions | |
ashes | +0.048 |
He | +0.045 |
They | +0.045 |
Let | +0.045 |
rider | +0.044 |
Neg logit contributions | |
agg | -0.065 |
ths | -0.065 |
eking | -0.065 |
aman | -0.059 |
Kevin | -0.058 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ashes | +0.023 |
." | +0.020 |
. | +0.020 |
Let | +0.018 |
rider | +0.018 |
Neg logit contributions | |
ths | -0.138 |
eking | -0.136 |
aman | -0.135 |
agg | -0.135 |
ering | -0.127 |
Token | " |
Token ID | 366 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
. | +0.137 |
." | +0.122 |
He | +0.109 |
ashes | +0.105 |
They | +0.101 |
Neg logit contributions | |
ths | -0.346 |
agg | -0.337 |
eking | -0.336 |
aman | -0.323 |
prisingly | -0.315 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
ths | +0.130 |
eking | +0.128 |
agg | +0.128 |
aman | +0.121 |
ering | +0.120 |
Neg logit contributions | |
. | -0.092 |
." | -0.087 |
ashes | -0.082 |
He | -0.080 |
Let | -0.080 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.20e-02 |
Pos logit contributions | |
ths | +0.460 |
aman | +0.448 |
agg | +0.446 |
eking | +0.422 |
Kevin | +0.421 |
Neg logit contributions | |
. | -0.426 |
." | -0.395 |
ashes | -0.324 |
Let | -0.323 |
play | -0.322 |
Token | , |
Token ID | 11 |
Feature activation | +9.24e-02 |
Pos logit contributions | |
ths | +0.169 |
agg | +0.164 |
eking | +0.160 |
aman | +0.155 |
prisingly | +0.148 |
Neg logit contributions | |
. | -0.144 |
." | -0.135 |
ashes | -0.131 |
He | -0.127 |
They | -0.122 |
Token | Mom |
Token ID | 11254 |
Feature activation | +4.34e-02 |
Pos logit contributions | |
ths | +0.191 |
agg | +0.185 |
eking | +0.181 |
aman | +0.177 |
Kevin | +0.169 |
Neg logit contributions | |
. | -0.128 |
." | -0.118 |
ashes | -0.110 |
He | -0.108 |
rider | -0.102 |
Token | my |
Token ID | 1820 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
ths | +0.075 |
eking | +0.071 |
aman | +0.068 |
agg | +0.066 |
ering | +0.063 |
Neg logit contributions | |
ashes | -0.073 |
five | -0.067 |
powers | -0.067 |
He | -0.065 |
They | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -1.53e-01 |
Pos logit contributions | |
ths | +0.053 |
eking | +0.052 |
aman | +0.049 |
ering | +0.049 |
agg | +0.048 |
Neg logit contributions | |
ashes | -0.031 |
five | -0.026 |
They | -0.025 |
He | -0.025 |
powers | -0.025 |
Token | 's |
Token ID | 338 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
ths | +0.052 |
agg | +0.051 |
eking | +0.050 |
aman | +0.049 |
prisingly | +0.047 |
Neg logit contributions | |
. | -0.029 |
." | -0.026 |
He | -0.023 |
ashes | -0.023 |
They | -0.022 |
Token | daddy |
Token ID | 35822 |
Feature activation | +2.34e-02 |
Pos logit contributions | |
ths | +0.039 |
agg | +0.037 |
eking | +0.036 |
aman | +0.034 |
Kevin | +0.034 |
Neg logit contributions | |
. | -0.043 |
." | -0.041 |
ashes | -0.041 |
He | -0.039 |
Let | -0.038 |
Token | said |
Token ID | 531 |
Feature activation | +7.58e-02 |
Pos logit contributions | |
ths | +0.044 |
agg | +0.044 |
eking | +0.042 |
aman | +0.041 |
prisingly | +0.039 |
Neg logit contributions | |
. | -0.038 |
." | -0.034 |
He | -0.032 |
ashes | -0.030 |
They | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +8.59e-02 |
Pos logit contributions | |
ths | +0.180 |
agg | +0.175 |
eking | +0.174 |
aman | +0.171 |
prisingly | +0.164 |
Neg logit contributions | |
. | -0.093 |
." | -0.079 |
He | -0.072 |
They | -0.065 |
towards | -0.064 |
Token | " |
Token ID | 366 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
ths | +0.186 |
agg | +0.180 |
eking | +0.177 |
aman | +0.173 |
Kevin | +0.166 |
Neg logit contributions | |
. | -0.108 |
." | -0.100 |
ashes | -0.094 |
He | -0.088 |
rider | -0.086 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.49e-01 |
Pos logit contributions | |
agg | +0.132 |
eking | +0.129 |
ths | +0.128 |
aman | +0.118 |
Kevin | +0.114 |
Neg logit contributions | |
. | -0.096 |
." | -0.089 |
He | -0.089 |
Let | -0.087 |
They | -0.084 |
Token | - |
Token ID | 12 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
ths | +0.458 |
agg | +0.452 |
aman | +0.438 |
eking | +0.427 |
Kevin | +0.413 |
Neg logit contributions | |
. | -0.420 |
." | -0.382 |
He | -0.330 |
ashes | -0.326 |
rider | -0.325 |
Token | oh |
Token ID | 1219 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
ths | +0.031 |
agg | +0.030 |
eking | +0.029 |
aman | +0.028 |
Jess | +0.025 |
Neg logit contributions | |
. | -0.053 |
." | -0.050 |
ashes | -0.048 |
He | -0.047 |
Let | -0.046 |
Token | , |
Token ID | 11 |
Feature activation | +8.46e-02 |
Pos logit contributions | |
ths | +0.306 |
eking | +0.298 |
agg | +0.288 |
aman | +0.285 |
prisingly | +0.271 |
Neg logit contributions | |
ashes | -0.231 |
. | -0.220 |
." | -0.216 |
He | -0.212 |
They | -0.211 |
Token | there |
Token ID | 612 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
ths | +0.181 |
agg | +0.177 |
eking | +0.174 |
aman | +0.169 |
Kevin | +0.163 |
Neg logit contributions | |
. | -0.112 |
." | -0.103 |
He | -0.093 |
ashes | -0.093 |
They | -0.087 |
Token | 's |
Token ID | 338 |
Feature activation | +2.09e-02 |
Pos logit contributions | |
ashes | +0.020 |
. | +0.019 |
five | +0.019 |
rider | +0.018 |
He | +0.018 |
Neg logit contributions | |
ths | -0.036 |
aman | -0.035 |
agg | -0.034 |
eking | -0.034 |
ering | -0.032 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
. | +0.000 |
." | +0.000 |
rider | +0.000 |
ashes | +0.000 |
Let | +0.000 |
Neg logit contributions | |
ths | -0.000 |
eking | -0.000 |
agg | -0.000 |
impressed | -0.000 |
unfortunately | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
powers | +0.009 |
ashes | +0.009 |
rider | +0.008 |
pedal | +0.008 |
five | +0.008 |
Neg logit contributions | |
ths | -0.029 |
eking | -0.028 |
unfortunately | -0.027 |
ering | -0.027 |
Jane | -0.027 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
. | +0.131 |
." | +0.117 |
ashes | +0.107 |
rider | +0.097 |
He | +0.096 |
Neg logit contributions | |
ths | -0.348 |
agg | -0.341 |
eking | -0.332 |
aman | -0.328 |
Kevin | -0.318 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
ths | +0.369 |
eking | +0.355 |
agg | +0.345 |
aman | +0.341 |
prisingly | +0.322 |
Neg logit contributions | |
. | -0.162 |
." | -0.140 |
He | -0.125 |
They | -0.121 |
Let | -0.103 |
Token | " |
Token ID | 1 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
ths | +0.249 |
agg | +0.240 |
eking | +0.234 |
aman | +0.227 |
Kevin | +0.216 |
Neg logit contributions | |
. | -0.244 |
." | -0.230 |
ashes | -0.218 |
He | -0.212 |
rider | -0.206 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
eking | +0.045 |
agg | +0.044 |
ths | +0.043 |
ering | +0.041 |
aman | +0.041 |
Neg logit contributions | |
. | -0.034 |
." | -0.032 |
Let | -0.030 |
They | -0.030 |
ashes | -0.029 |
Token | - |
Token ID | 12 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
ths | +0.550 |
agg | +0.541 |
aman | +0.529 |
eking | +0.518 |
Kevin | +0.508 |
Neg logit contributions | |
. | -0.334 |
." | -0.301 |
ashes | -0.247 |
He | -0.246 |
Let | -0.238 |
Token | oh |
Token ID | 1219 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
ths | +0.027 |
agg | +0.026 |
eking | +0.025 |
aman | +0.023 |
Kevin | +0.021 |
Neg logit contributions | |
. | -0.061 |
." | -0.058 |
ashes | -0.056 |
He | -0.055 |
rider | -0.054 |
Token | !" |
Token ID | 2474 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
ths | +0.358 |
eking | +0.349 |
agg | +0.346 |
aman | +0.334 |
prisingly | +0.321 |
Neg logit contributions | |
. | -0.184 |
ashes | -0.181 |
He | -0.177 |
." | -0.172 |
They | -0.169 |
Token | Tom |
Token ID | 4186 |
Feature activation | +3.31e-02 |
Pos logit contributions | |
. | +0.036 |
." | +0.032 |
He | +0.029 |
ashes | +0.028 |
They | +0.027 |
Neg logit contributions | |
ths | -0.075 |
agg | -0.072 |
eking | -0.072 |
aman | -0.070 |
prisingly | -0.067 |
Token | said |
Token ID | 531 |
Feature activation | +6.54e-02 |
Pos logit contributions | |
ths | +0.069 |
agg | +0.067 |
eking | +0.067 |
aman | +0.064 |
Kevin | +0.061 |
Neg logit contributions | |
. | -0.039 |
." | -0.038 |
ashes | -0.037 |
rider | -0.034 |
five | -0.033 |
Token | open |
Token ID | 1280 |
Feature activation | -5.60e-02 |
Pos logit contributions | |
. | +0.053 |
." | +0.049 |
ashes | +0.045 |
He | +0.045 |
rider | +0.043 |
Neg logit contributions | |
ths | -0.068 |
agg | -0.066 |
eking | -0.065 |
aman | -0.063 |
prisingly | -0.060 |
Token | freezer |
Token ID | 30967 |
Feature activation | +3.99e-02 |
Pos logit contributions | |
ashes | +0.089 |
." | +0.083 |
. | +0.080 |
powers | +0.079 |
Let | +0.079 |
Neg logit contributions | |
aman | -0.095 |
ths | -0.094 |
agg | -0.092 |
eking | -0.083 |
Jess | -0.079 |
Token | and |
Token ID | 290 |
Feature activation | +5.20e-02 |
Pos logit contributions | |
ths | +0.098 |
eking | +0.093 |
agg | +0.092 |
aman | +0.092 |
ering | +0.087 |
Neg logit contributions | |
ashes | -0.035 |
. | -0.035 |
." | -0.035 |
powers | -0.029 |
rider | -0.029 |
Token | said |
Token ID | 531 |
Feature activation | +7.76e-02 |
Pos logit contributions | |
ths | +0.100 |
agg | +0.097 |
eking | +0.095 |
aman | +0.092 |
Kevin | +0.088 |
Neg logit contributions | |
. | -0.079 |
." | -0.074 |
ashes | -0.070 |
He | -0.068 |
rider | -0.065 |
Token | " |
Token ID | 366 |
Feature activation | +6.33e-02 |
Pos logit contributions | |
ths | +0.170 |
agg | +0.164 |
eking | +0.164 |
aman | +0.160 |
prisingly | +0.152 |
Neg logit contributions | |
. | -0.102 |
." | -0.091 |
He | -0.084 |
ashes | -0.081 |
They | -0.078 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
ths | +0.114 |
agg | +0.111 |
eking | +0.111 |
aman | +0.100 |
ering | +0.098 |
Neg logit contributions | |
. | -0.104 |
." | -0.097 |
He | -0.092 |
Let | -0.090 |
towards | -0.088 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.19e-02 |
Pos logit contributions | |
ths | +0.475 |
agg | +0.459 |
eking | +0.449 |
aman | +0.438 |
Kevin | +0.416 |
Neg logit contributions | |
. | -0.385 |
." | -0.359 |
ashes | -0.333 |
He | -0.327 |
rider | -0.313 |
Token | !" |
Token ID | 2474 |
Feature activation | -5.29e-02 |
Pos logit contributions | |
ths | +0.165 |
agg | +0.160 |
eking | +0.156 |
aman | +0.150 |
prisingly | +0.145 |
Neg logit contributions | |
ashes | -0.140 |
. | -0.137 |
." | -0.129 |
He | -0.126 |
rider | -0.125 |
Token | He |
Token ID | 679 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
. | +0.064 |
." | +0.058 |
ashes | +0.053 |
He | +0.052 |
rider | +0.049 |
Neg logit contributions | |
ths | -0.119 |
agg | -0.116 |
eking | -0.114 |
aman | -0.111 |
Kevin | -0.106 |
Token | quickly |
Token ID | 2952 |
Feature activation | -7.20e-02 |
Pos logit contributions | |
. | +0.031 |
." | +0.027 |
He | +0.026 |
bar | +0.024 |
They | +0.024 |
Neg logit contributions | |
ths | -0.044 |
eking | -0.043 |
agg | -0.043 |
aman | -0.042 |
Kevin | -0.040 |
Token | shut |
Token ID | 4423 |
Feature activation | -5.37e-02 |
Pos logit contributions | |
. | +0.105 |
." | +0.103 |
ashes | +0.099 |
rider | +0.093 |
He | +0.092 |
Neg logit contributions | |
ths | -0.139 |
agg | -0.135 |
eking | -0.131 |
aman | -0.127 |
Kevin | -0.120 |
Token | what |
Token ID | 644 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
ths | +0.079 |
agg | +0.077 |
eking | +0.074 |
aman | +0.073 |
Kevin | +0.069 |
Neg logit contributions | |
. | -0.082 |
." | -0.077 |
He | -0.070 |
ashes | -0.070 |
Let | -0.067 |
Token | to |
Token ID | 284 |
Feature activation | +3.23e-04 |
Pos logit contributions | |
. | +0.030 |
." | +0.027 |
ashes | +0.024 |
He | +0.023 |
rider | +0.022 |
Neg logit contributions | |
ths | -0.075 |
agg | -0.073 |
eking | -0.072 |
aman | -0.070 |
Kevin | -0.068 |
Token | say |
Token ID | 910 |
Feature activation | +5.57e-02 |
Pos logit contributions | |
agg | +0.001 |
ths | +0.001 |
eking | +0.001 |
aman | +0.001 |
Kevin | +0.001 |
Neg logit contributions | |
ashes | -0.000 |
. | -0.000 |
." | -0.000 |
He | -0.000 |
Let | -0.000 |
Token | . |
Token ID | 13 |
Feature activation | -1.39e-01 |
Pos logit contributions | |
eking | +0.149 |
ths | +0.148 |
agg | +0.142 |
aman | +0.138 |
ering | +0.138 |
Neg logit contributions | |
ashes | -0.034 |
ny | -0.025 |
powers | -0.025 |
rider | -0.024 |
He | -0.021 |
Token | " |
Token ID | 366 |
Feature activation | +6.79e-02 |
Pos logit contributions | |
. | +0.163 |
." | +0.148 |
ashes | +0.141 |
rider | +0.131 |
He | +0.128 |
Neg logit contributions | |
ths | -0.313 |
agg | -0.306 |
eking | -0.295 |
aman | -0.291 |
Jess | -0.281 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
eking | +0.140 |
ths | +0.138 |
agg | +0.137 |
aman | +0.134 |
ering | +0.129 |
Neg logit contributions | |
. | -0.089 |
." | -0.083 |
Let | -0.081 |
He | -0.080 |
rider | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | +8.85e-02 |
Pos logit contributions | |
ths | +0.487 |
agg | +0.470 |
aman | +0.463 |
eking | +0.453 |
Kevin | +0.436 |
Neg logit contributions | |
. | -0.387 |
." | -0.359 |
He | -0.309 |
ashes | -0.308 |
Let | -0.301 |
Token | ok |
Token ID | 12876 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
ths | +0.184 |
agg | +0.178 |
eking | +0.175 |
aman | +0.170 |
Jess | +0.163 |
Neg logit contributions | |
. | -0.118 |
." | -0.109 |
ashes | -0.105 |
He | -0.100 |
rider | -0.095 |
Token | !" |
Token ID | 2474 |
Feature activation | -5.44e-02 |
Pos logit contributions | |
ths | +0.197 |
agg | +0.185 |
eking | +0.179 |
aman | +0.178 |
Jess | +0.171 |
Neg logit contributions | |
ashes | -0.135 |
five | -0.123 |
. | -0.118 |
He | -0.116 |
powers | -0.115 |
Token | it |
Token ID | 340 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
. | +0.089 |
." | +0.085 |
ashes | +0.083 |
rider | +0.079 |
He | +0.077 |
Neg logit contributions | |
ths | -0.094 |
agg | -0.092 |
eking | -0.088 |
aman | -0.085 |
Jess | -0.082 |
Token | said |
Token ID | 531 |
Feature activation | +7.43e-02 |
Pos logit contributions | |
ths | +0.043 |
agg | +0.041 |
eking | +0.040 |
aman | +0.039 |
prisingly | +0.037 |
Neg logit contributions | |
. | -0.035 |
." | -0.033 |
ashes | -0.031 |
He | -0.030 |
rider | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
Let | +0.002 |
." | +0.002 |
five | +0.002 |
eer | +0.002 |
ashes | +0.002 |
Neg logit contributions | |
ths | -0.008 |
eking | -0.008 |
aman | -0.007 |
agg | -0.007 |
ering | -0.007 |
Token | Emma |
Token ID | 18966 |
Feature activation | +2.49e-02 |
Pos logit contributions | |
. | +0.278 |
ashes | +0.265 |
." | +0.264 |
rider | +0.252 |
He | +0.243 |
Neg logit contributions | |
ths | -0.207 |
agg | -0.198 |
aman | -0.187 |
eking | -0.186 |
Jess | -0.175 |
Token | said |
Token ID | 531 |
Feature activation | +7.68e-02 |
Pos logit contributions | |
ths | +0.052 |
agg | +0.050 |
eking | +0.050 |
aman | +0.048 |
prisingly | +0.046 |
Neg logit contributions | |
. | -0.034 |
." | -0.031 |
ashes | -0.029 |
He | -0.028 |
They | -0.027 |
Token | , |
Token ID | 11 |
Feature activation | +8.68e-02 |
Pos logit contributions | |
ths | +0.181 |
agg | +0.175 |
eking | +0.174 |
aman | +0.171 |
Kevin | +0.164 |
Neg logit contributions | |
. | -0.092 |
." | -0.080 |
He | -0.071 |
ashes | -0.067 |
towards | -0.066 |
Token | " |
Token ID | 366 |
Feature activation | +6.67e-02 |
Pos logit contributions | |
ths | +0.211 |
agg | +0.205 |
eking | +0.202 |
aman | +0.197 |
prisingly | +0.192 |
Neg logit contributions | |
. | -0.099 |
." | -0.086 |
He | -0.079 |
They | -0.073 |
Let | -0.071 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
eking | +0.124 |
ths | +0.123 |
agg | +0.119 |
ering | +0.113 |
aman | +0.107 |
Neg logit contributions | |
. | -0.105 |
." | -0.098 |
Let | -0.095 |
He | -0.091 |
They | -0.090 |
Token | - |
Token ID | 12 |
Feature activation | +2.26e-02 |
Pos logit contributions | |
ths | +0.456 |
agg | +0.442 |
eking | +0.427 |
aman | +0.425 |
Kevin | +0.406 |
Neg logit contributions | |
. | -0.412 |
." | -0.382 |
ashes | -0.341 |
He | -0.337 |
rider | -0.329 |
Token | oh |
Token ID | 1219 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
ths | +0.031 |
agg | +0.030 |
eking | +0.029 |
aman | +0.028 |
Kevin | +0.026 |
Neg logit contributions | |
. | -0.047 |
." | -0.044 |
ashes | -0.042 |
He | -0.042 |
rider | -0.040 |
Token | , |
Token ID | 11 |
Feature activation | +8.57e-02 |
Pos logit contributions | |
ths | +0.302 |
eking | +0.294 |
agg | +0.292 |
aman | +0.288 |
prisingly | +0.270 |
Neg logit contributions | |
ashes | -0.243 |
He | -0.224 |
. | -0.223 |
They | -0.223 |
." | -0.219 |
Token | now |
Token ID | 783 |
Feature activation | +2.95e-02 |
Pos logit contributions | |
ths | +0.167 |
agg | +0.161 |
eking | +0.158 |
aman | +0.154 |
Kevin | +0.146 |
Neg logit contributions | |
. | -0.127 |
." | -0.118 |
ashes | -0.114 |
He | -0.110 |
rider | -0.105 |
Token | what |
Token ID | 644 |
Feature activation | -3.16e-02 |
Pos logit contributions | |
ths | +0.049 |
aman | +0.042 |
eking | +0.041 |
agg | +0.041 |
unfortunately | +0.040 |
Neg logit contributions | |
ashes | -0.054 |
five | -0.048 |
v | -0.048 |
eer | -0.047 |
ny | -0.047 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ths | +0.356 |
eking | +0.346 |
agg | +0.342 |
aman | +0.329 |
prisingly | +0.314 |
Neg logit contributions | |
. | -0.170 |
." | -0.150 |
He | -0.135 |
They | -0.126 |
Let | -0.120 |
Token | â |
Token ID | 22940 |
Feature activation | +6.88e-02 |
Pos logit contributions | |
ths | +0.166 |
agg | +0.155 |
eking | +0.149 |
aman | +0.143 |
Kevin | +0.131 |
Neg logit contributions | |
. | -0.337 |
." | -0.323 |
ashes | -0.312 |
He | -0.304 |
rider | -0.297 |
Token | € |
Token ID | 26391 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
ths | +0.111 |
eking | +0.108 |
agg | +0.106 |
aman | +0.104 |
prisingly | +0.097 |
Neg logit contributions | |
. | -0.119 |
." | -0.114 |
ashes | -0.111 |
He | -0.106 |
powers | -0.103 |
Token | � |
Token ID | 129 |
Feature activation | +3.88e-02 |
Pos logit contributions | |
ths | +0.155 |
agg | +0.151 |
eking | +0.148 |
aman | +0.144 |
prisingly | +0.138 |
Neg logit contributions | |
. | -0.115 |
." | -0.106 |
He | -0.097 |
ashes | -0.096 |
They | -0.093 |
Token | � |
Token ID | 241 |
Feature activation | -1.15e-02 |
Pos logit contributions | |
ths | +0.065 |
eking | +0.063 |
agg | +0.061 |
aman | +0.059 |
Kevin | +0.055 |
Neg logit contributions | |
. | -0.066 |
." | -0.063 |
ashes | -0.061 |
Let | -0.059 |
He | -0.059 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
. | +0.019 |
." | +0.019 |
ashes | +0.019 |
rider | +0.017 |
bars | +0.017 |
Neg logit contributions | |
agg | -0.017 |
Kevin | -0.017 |
Jess | -0.017 |
ths | -0.017 |
aman | -0.017 |
Token | - |
Token ID | 12 |
Feature activation | +2.59e-02 |
Pos logit contributions | |
ths | +0.542 |
agg | +0.524 |
aman | +0.515 |
eking | +0.504 |
Kevin | +0.489 |
Neg logit contributions | |
. | -0.336 |
." | -0.299 |
He | -0.255 |
ashes | -0.250 |
towards | -0.245 |
Token | oh |
Token ID | 1219 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
ths | +0.029 |
agg | +0.027 |
eking | +0.026 |
aman | +0.025 |
Kevin | +0.023 |
Neg logit contributions | |
. | -0.060 |
." | -0.058 |
ashes | -0.056 |
He | -0.055 |
They | -0.053 |
Token | , |
Token ID | 11 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
eking | +0.361 |
ths | +0.355 |
agg | +0.341 |
aman | +0.328 |
ering | +0.325 |
Neg logit contributions | |
ashes | -0.188 |
They | -0.178 |
He | -0.176 |
. | -0.172 |
." | -0.167 |
Token | â |
Token ID | 22940 |
Feature activation | +6.85e-02 |
Pos logit contributions | |
ths | +0.147 |
agg | +0.139 |
eking | +0.136 |
aman | +0.133 |
Jess | +0.124 |
Neg logit contributions | |
. | -0.158 |
." | -0.150 |
ashes | -0.149 |
He | -0.140 |
rider | -0.138 |
Token | € |
Token ID | 26391 |
Feature activation | +7.85e-02 |
Pos logit contributions | |
ths | +0.095 |
eking | +0.092 |
agg | +0.089 |
aman | +0.089 |
prisingly | +0.079 |
Neg logit contributions | |
. | -0.132 |
ashes | -0.129 |
." | -0.128 |
powers | -0.122 |
He | -0.121 |
Token | to |
Token ID | 284 |
Feature activation | -2.25e-03 |
Pos logit contributions | |
ths | +0.181 |
agg | +0.176 |
eking | +0.175 |
aman | +0.171 |
Kevin | +0.164 |
Neg logit contributions | |
. | -0.092 |
." | -0.082 |
He | -0.073 |
ashes | -0.072 |
They | -0.068 |
Token | Tim |
Token ID | 5045 |
Feature activation | +5.34e-02 |
Pos logit contributions | |
. | +0.003 |
." | +0.003 |
He | +0.002 |
play | +0.002 |
They | +0.002 |
Neg logit contributions | |
ths | -0.005 |
agg | -0.005 |
eking | -0.005 |
prisingly | -0.005 |
Kevin | -0.005 |
Token | my |
Token ID | 1820 |
Feature activation | +2.22e-02 |
Pos logit contributions | |
ths | +0.100 |
agg | +0.098 |
eking | +0.095 |
aman | +0.092 |
Jess | +0.088 |
Neg logit contributions | |
. | -0.086 |
." | -0.079 |
ashes | -0.073 |
He | -0.073 |
rider | -0.070 |
Token | , |
Token ID | 11 |
Feature activation | +9.16e-02 |
Pos logit contributions | |
agg | +0.053 |
ths | +0.052 |
eking | +0.051 |
aman | +0.049 |
Jess | +0.048 |
Neg logit contributions | |
. | -0.027 |
." | -0.023 |
He | -0.021 |
They | -0.019 |
ashes | -0.019 |
Token | " |
Token ID | 366 |
Feature activation | +6.35e-02 |
Pos logit contributions | |
ths | +0.220 |
agg | +0.217 |
eking | +0.212 |
aman | +0.205 |
prisingly | +0.198 |
Neg logit contributions | |
. | -0.106 |
." | -0.093 |
He | -0.085 |
They | -0.078 |
Let | -0.078 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
agg | +0.138 |
ths | +0.136 |
eking | +0.134 |
aman | +0.125 |
Kevin | +0.122 |
Neg logit contributions | |
. | -0.083 |
." | -0.077 |
He | -0.072 |
Let | -0.072 |
They | -0.068 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.28e-02 |
Pos logit contributions | |
ths | +0.379 |
agg | +0.375 |
aman | +0.360 |
eking | +0.350 |
Kevin | +0.339 |
Neg logit contributions | |
. | -0.497 |
." | -0.461 |
He | -0.408 |
ashes | -0.405 |
play | -0.401 |
Token | , |
Token ID | 11 |
Feature activation | +8.93e-02 |
Pos logit contributions | |
ths | +0.167 |
agg | +0.160 |
eking | +0.158 |
aman | +0.152 |
prisingly | +0.146 |
Neg logit contributions | |
. | -0.147 |
." | -0.139 |
ashes | -0.137 |
He | -0.131 |
rider | -0.128 |
Token | the |
Token ID | 262 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
ths | +0.188 |
agg | +0.183 |
eking | +0.179 |
aman | +0.174 |
Kevin | +0.167 |
Neg logit contributions | |
. | -0.122 |
." | -0.112 |
ashes | -0.102 |
He | -0.101 |
Let | -0.096 |
Token | roof |
Token ID | 9753 |
Feature activation | -3.81e-02 |
Pos logit contributions | |
. | +0.045 |
." | +0.042 |
ashes | +0.040 |
He | +0.039 |
rider | +0.037 |
Neg logit contributions | |
ths | -0.077 |
agg | -0.074 |
eking | -0.072 |
aman | -0.071 |
Jess | -0.068 |
Token | has |
Token ID | 468 |
Feature activation | -3.04e-02 |
Pos logit contributions | |
. | +0.038 |
ashes | +0.038 |
rider | +0.036 |
He | +0.036 |
." | +0.036 |
Neg logit contributions | |
ths | -0.086 |
eking | -0.083 |
agg | -0.079 |
aman | -0.078 |
ering | -0.076 |
Token | worried |
Token ID | 7960 |
Feature activation | +1.51e-02 |
Pos logit contributions | |
ths | +0.118 |
agg | +0.116 |
eking | +0.113 |
aman | +0.106 |
Kevin | +0.102 |
Neg logit contributions | |
. | -0.120 |
." | -0.114 |
ashes | -0.109 |
rider | -0.107 |
He | -0.107 |
Token | . |
Token ID | 13 |
Feature activation | -1.35e-01 |
Pos logit contributions | |
ths | +0.042 |
eking | +0.040 |
agg | +0.040 |
ering | +0.039 |
aman | +0.039 |
Neg logit contributions | |
ashes | -0.008 |
rider | -0.006 |
five | -0.006 |
powers | -0.006 |
pedal | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
. | +0.140 |
." | +0.125 |
He | +0.116 |
ashes | +0.109 |
They | +0.108 |
Neg logit contributions | |
ths | -0.331 |
eking | -0.322 |
agg | -0.320 |
aman | -0.309 |
prisingly | -0.299 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
eking | +0.401 |
ths | +0.399 |
aman | +0.378 |
agg | +0.367 |
ering | +0.361 |
Neg logit contributions | |
. | -0.144 |
." | -0.122 |
He | -0.117 |
They | -0.110 |
Let | -0.087 |
Token | " |
Token ID | 1 |
Feature activation | +3.01e-02 |
Pos logit contributions | |
eking | +0.273 |
ths | +0.267 |
aman | +0.258 |
agg | +0.255 |
ering | +0.239 |
Neg logit contributions | |
. | -0.239 |
." | -0.222 |
He | -0.219 |
They | -0.212 |
Let | -0.200 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.48e-01 |
Pos logit contributions | |
eking | +0.061 |
agg | +0.060 |
ths | +0.059 |
ering | +0.058 |
aman | +0.057 |
Neg logit contributions | |
. | -0.039 |
Let | -0.036 |
." | -0.036 |
He | -0.034 |
They | -0.033 |
Token | - |
Token ID | 12 |
Feature activation | +2.53e-02 |
Pos logit contributions | |
ths | +0.465 |
agg | +0.452 |
aman | +0.440 |
eking | +0.437 |
Kevin | +0.417 |
Neg logit contributions | |
. | -0.403 |
." | -0.374 |
He | -0.328 |
ashes | -0.324 |
Let | -0.318 |
Token | oh |
Token ID | 1219 |
Feature activation | +1.68e-01 |
Pos logit contributions | |
ths | +0.029 |
agg | +0.029 |
eking | +0.027 |
aman | +0.026 |
Kevin | +0.023 |
Neg logit contributions | |
. | -0.058 |
." | -0.055 |
ashes | -0.052 |
He | -0.052 |
Let | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | +9.02e-02 |
Pos logit contributions | |
ths | +0.375 |
agg | +0.367 |
eking | +0.364 |
aman | +0.351 |
prisingly | +0.337 |
Neg logit contributions | |
. | -0.182 |
ashes | -0.171 |
." | -0.168 |
He | -0.164 |
They | -0.157 |
Token | this |
Token ID | 428 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
ths | +0.169 |
agg | +0.163 |
eking | +0.160 |
aman | +0.155 |
Kevin | +0.147 |
Neg logit contributions | |
. | -0.143 |
." | -0.134 |
ashes | -0.124 |
He | -0.123 |
They | -0.117 |
Token | fire |
Token ID | 2046 |
Feature activation | +9.39e-03 |
Pos logit contributions | |
ths | +0.066 |
agg | +0.064 |
eking | +0.062 |
aman | +0.061 |
Kevin | +0.058 |
Neg logit contributions | |
. | -0.042 |
." | -0.039 |
ashes | -0.037 |
He | -0.036 |
rider | -0.034 |
Token | in |
Token ID | 287 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
eking | +0.044 |
ths | +0.042 |
ering | +0.039 |
nowhere | +0.037 |
agg | +0.037 |
Neg logit contributions | |
powers | -0.024 |
five | -0.024 |
ashes | -0.023 |
rider | -0.022 |
." | -0.020 |
Token | the |
Token ID | 262 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
ashes | +0.079 |
. | +0.079 |
." | +0.075 |
rider | +0.071 |
Let | +0.070 |
Neg logit contributions | |
ths | -0.109 |
agg | -0.104 |
aman | -0.102 |
eking | -0.101 |
Jess | -0.099 |
Token | distance |
Token ID | 5253 |
Feature activation | +2.12e-02 |
Pos logit contributions | |
. | +0.055 |
." | +0.052 |
ashes | +0.052 |
He | +0.050 |
rider | +0.049 |
Neg logit contributions | |
ths | -0.061 |
agg | -0.060 |
aman | -0.058 |
eking | -0.058 |
Jess | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
eking | +0.052 |
aman | +0.050 |
ering | +0.049 |
ths | +0.048 |
Jane | +0.047 |
Neg logit contributions | |
powers | -0.017 |
ashes | -0.016 |
rider | -0.013 |
five | -0.012 |
player | -0.012 |
Token | " |
Token ID | 366 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
. | +0.174 |
ashes | +0.170 |
." | +0.157 |
rider | +0.145 |
five | +0.141 |
Neg logit contributions | |
ths | -0.295 |
agg | -0.279 |
aman | -0.270 |
eking | -0.270 |
Jess | -0.262 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
agg | +0.125 |
ths | +0.125 |
eking | +0.123 |
aman | +0.115 |
ering | +0.111 |
Neg logit contributions | |
. | -0.102 |
." | -0.096 |
He | -0.091 |
rider | -0.088 |
Let | -0.088 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.59e-02 |
Pos logit contributions | |
ths | +0.560 |
aman | +0.543 |
agg | +0.537 |
Kevin | +0.525 |
eking | +0.509 |
Neg logit contributions | |
. | -0.342 |
." | -0.303 |
towards | -0.232 |
play | -0.231 |
like | -0.228 |
Token | , |
Token ID | 11 |
Feature activation | +8.76e-02 |
Pos logit contributions | |
ths | +0.201 |
agg | +0.193 |
eking | +0.191 |
aman | +0.185 |
prisingly | +0.180 |
Neg logit contributions | |
. | -0.120 |
ashes | -0.116 |
." | -0.111 |
He | -0.106 |
They | -0.102 |
Token | that |
Token ID | 326 |
Feature activation | -1.74e-02 |
Pos logit contributions | |
ths | +0.196 |
agg | +0.195 |
eking | +0.191 |
aman | +0.184 |
Kevin | +0.179 |
Neg logit contributions | |
. | -0.109 |
." | -0.100 |
He | -0.089 |
rider | -0.084 |
ashes | -0.082 |
Token | cloud |
Token ID | 6279 |
Feature activation | +1.86e-02 |
Pos logit contributions | |
. | +0.021 |
." | +0.020 |
ashes | +0.017 |
He | +0.017 |
rider | +0.016 |
Neg logit contributions | |
ths | -0.039 |
agg | -0.038 |
eking | -0.038 |
aman | -0.037 |
Kevin | -0.035 |
Token | looks |
Token ID | 3073 |
Feature activation | +5.39e-03 |
Pos logit contributions | |
ths | +0.042 |
eking | +0.041 |
agg | +0.040 |
aman | +0.039 |
ering | +0.038 |
Neg logit contributions | |
. | -0.019 |
ashes | -0.018 |
He | -0.017 |
." | -0.017 |
rider | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
rider | +0.056 |
five | +0.051 |
powers | +0.046 |
ashes | +0.046 |
likes | +0.043 |
Neg logit contributions | |
ths | -0.106 |
eking | -0.101 |
aman | -0.098 |
agg | -0.098 |
unfortunately | -0.095 |
Token | |
Token ID | 220 |
Feature activation | +1.81e-01 |
Pos logit contributions | |
. | +0.169 |
." | +0.152 |
He | +0.141 |
They | +0.136 |
ashes | +0.128 |
Neg logit contributions | |
ths | -0.341 |
eking | -0.332 |
agg | -0.331 |
aman | -0.314 |
prisingly | -0.309 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
ths | +0.471 |
eking | +0.446 |
agg | +0.441 |
aman | +0.428 |
Kevin | +0.410 |
Neg logit contributions | |
. | -0.205 |
." | -0.181 |
He | -0.161 |
They | -0.155 |
ashes | -0.135 |
Token |
|
Token ID | 198 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
ths | +0.393 |
eking | +0.378 |
agg | +0.368 |
aman | +0.353 |
Kevin | +0.337 |
Neg logit contributions | |
. | -0.158 |
." | -0.137 |
He | -0.119 |
They | -0.116 |
Let | -0.101 |
Token | " |
Token ID | 1 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
ths | +0.240 |
agg | +0.230 |
eking | +0.224 |
aman | +0.219 |
prisingly | +0.207 |
Neg logit contributions | |
. | -0.252 |
." | -0.238 |
ashes | -0.230 |
He | -0.222 |
rider | -0.219 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
agg | +0.051 |
eking | +0.051 |
ths | +0.048 |
ering | +0.047 |
aman | +0.044 |
Neg logit contributions | |
. | -0.031 |
Let | -0.028 |
." | -0.028 |
player | -0.028 |
play | -0.028 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.55e-02 |
Pos logit contributions | |
ths | +0.459 |
agg | +0.453 |
aman | +0.446 |
eking | +0.429 |
Kevin | +0.427 |
Neg logit contributions | |
. | -0.424 |
." | -0.391 |
play | -0.323 |
Let | -0.321 |
ashes | -0.320 |
Token | , |
Token ID | 11 |
Feature activation | +8.61e-02 |
Pos logit contributions | |
ths | +0.175 |
agg | +0.168 |
eking | +0.166 |
aman | +0.159 |
prisingly | +0.155 |
Neg logit contributions | |
. | -0.141 |
ashes | -0.139 |
." | -0.132 |
He | -0.130 |
They | -0.125 |
Token | we |
Token ID | 356 |
Feature activation | -3.49e-02 |
Pos logit contributions | |
ths | +0.195 |
agg | +0.191 |
eking | +0.188 |
aman | +0.183 |
Kevin | +0.178 |
Neg logit contributions | |
. | -0.105 |
." | -0.096 |
He | -0.084 |
ashes | -0.082 |
Let | -0.080 |
Token | have |
Token ID | 423 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
. | +0.038 |
." | +0.034 |
ashes | +0.028 |
play | +0.027 |
They | +0.026 |
Neg logit contributions | |
ths | -0.087 |
agg | -0.086 |
aman | -0.082 |
eking | -0.082 |
Jess | -0.079 |
Token | a |
Token ID | 257 |
Feature activation | -2.92e-02 |
Pos logit contributions | |
. | +0.011 |
." | +0.010 |
ashes | +0.007 |
He | +0.006 |
They | +0.006 |
Neg logit contributions | |
ths | -0.047 |
agg | -0.047 |
eking | -0.046 |
aman | -0.045 |
Kevin | -0.044 |
Token | angry |
Token ID | 7954 |
Feature activation | +1.57e-02 |
Pos logit contributions | |
ths | +0.114 |
agg | +0.108 |
aman | +0.105 |
eking | +0.101 |
Kevin | +0.098 |
Neg logit contributions | |
. | -0.138 |
." | -0.129 |
ashes | -0.120 |
He | -0.117 |
towards | -0.113 |
Token | . |
Token ID | 13 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
ths | +0.038 |
eking | +0.037 |
ering | +0.035 |
unfortunately | +0.034 |
prisingly | +0.034 |
Neg logit contributions | |
pedal | -0.011 |
ashes | -0.011 |
powers | -0.011 |
rider | -0.010 |
tire | -0.010 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
. | +0.084 |
." | +0.064 |
They | +0.063 |
He | +0.062 |
Let | +0.042 |
Neg logit contributions | |
ths | -0.424 |
eking | -0.420 |
agg | -0.416 |
aman | -0.396 |
prisingly | -0.393 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
ths | +0.388 |
eking | +0.377 |
agg | +0.365 |
aman | +0.362 |
ering | +0.343 |
Neg logit contributions | |
. | -0.156 |
." | -0.132 |
He | -0.118 |
They | -0.116 |
play | -0.095 |
Token | " |
Token ID | 1 |
Feature activation | +2.41e-02 |
Pos logit contributions | |
agg | +0.279 |
ths | +0.276 |
eking | +0.274 |
aman | +0.267 |
Kevin | +0.242 |
Neg logit contributions | |
. | -0.231 |
." | -0.209 |
He | -0.194 |
They | -0.193 |
Let | -0.185 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
agg | +0.053 |
eking | +0.051 |
ths | +0.048 |
ering | +0.047 |
aman | +0.047 |
Neg logit contributions | |
. | -0.032 |
Let | -0.029 |
." | -0.028 |
dog | -0.028 |
They | -0.028 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.61e-02 |
Pos logit contributions | |
ths | +0.517 |
agg | +0.514 |
aman | +0.505 |
Kevin | +0.485 |
eking | +0.482 |
Neg logit contributions | |
. | -0.373 |
." | -0.338 |
He | -0.271 |
Let | -0.264 |
play | -0.263 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
ths | +0.211 |
agg | +0.205 |
eking | +0.200 |
aman | +0.198 |
Kevin | +0.191 |
Neg logit contributions | |
. | -0.136 |
." | -0.125 |
He | -0.105 |
Let | -0.102 |
ashes | -0.100 |
Token | We |
Token ID | 775 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
. | +0.176 |
rider | +0.155 |
ashes | +0.153 |
." | +0.150 |
pedal | +0.142 |
Neg logit contributions | |
ths | -0.336 |
agg | -0.325 |
aman | -0.310 |
eking | -0.310 |
Jess | -0.303 |
Token | 're |
Token ID | 821 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
. | +0.031 |
." | +0.028 |
ashes | +0.026 |
He | +0.025 |
They | +0.024 |
Neg logit contributions | |
ths | -0.042 |
agg | -0.041 |
eking | -0.040 |
aman | -0.039 |
Kevin | -0.037 |
Token | in |
Token ID | 287 |
Feature activation | -5.39e-02 |
Pos logit contributions | |
ths | +0.212 |
agg | +0.211 |
aman | +0.203 |
Kevin | +0.193 |
eking | +0.189 |
Neg logit contributions | |
. | -0.224 |
." | -0.210 |
like | -0.178 |
ashes | -0.177 |
He | -0.177 |
Token | in |
Token ID | 287 |
Feature activation | -5.54e-02 |
Pos logit contributions | |
ths | +0.094 |
agg | +0.091 |
eking | +0.089 |
aman | +0.087 |
Kevin | +0.082 |
Neg logit contributions | |
. | -0.075 |
." | -0.070 |
ashes | -0.066 |
He | -0.064 |
rider | -0.062 |
Token | their |
Token ID | 511 |
Feature activation | +1.70e-02 |
Pos logit contributions | |
. | +0.075 |
." | +0.066 |
He | +0.060 |
ashes | +0.058 |
They | +0.057 |
Neg logit contributions | |
ths | -0.122 |
agg | -0.119 |
eking | -0.117 |
aman | -0.113 |
prisingly | -0.109 |
Token | tracks |
Token ID | 8339 |
Feature activation | -1.06e-01 |
Pos logit contributions | |
ths | +0.034 |
agg | +0.033 |
eking | +0.031 |
aman | +0.031 |
Kevin | +0.030 |
Neg logit contributions | |
. | -0.024 |
." | -0.023 |
ashes | -0.022 |
rider | -0.021 |
He | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
ashes | +0.070 |
powers | +0.062 |
." | +0.061 |
rider | +0.055 |
. | +0.052 |
Neg logit contributions | |
ths | -0.278 |
aman | -0.272 |
eking | -0.271 |
agg | -0.264 |
ering | -0.255 |
Token | " |
Token ID | 366 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
. | +0.138 |
." | +0.118 |
They | +0.114 |
He | +0.114 |
Let | +0.098 |
Neg logit contributions | |
ths | -0.370 |
eking | -0.368 |
agg | -0.355 |
aman | -0.342 |
prisingly | -0.342 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
eking | +0.147 |
ths | +0.145 |
agg | +0.144 |
ering | +0.140 |
aman | +0.133 |
Neg logit contributions | |
. | -0.079 |
." | -0.071 |
Let | -0.071 |
They | -0.070 |
He | -0.068 |
Token | , |
Token ID | 11 |
Feature activation | +8.73e-02 |
Pos logit contributions | |
ths | +0.419 |
agg | +0.412 |
aman | +0.410 |
eking | +0.383 |
Kevin | +0.380 |
Neg logit contributions | |
. | -0.458 |
." | -0.422 |
Let | -0.357 |
play | -0.355 |
He | -0.351 |
Token | nothing |
Token ID | 2147 |
Feature activation | +1.61e-02 |
Pos logit contributions | |
ths | +0.166 |
agg | +0.161 |
eking | +0.158 |
aman | +0.153 |
Kevin | +0.146 |
Neg logit contributions | |
. | -0.135 |
." | -0.126 |
ashes | -0.117 |
He | -0.115 |
rider | -0.110 |
Token | ," |
Token ID | 553 |
Feature activation | +1.87e-02 |
Pos logit contributions | |
eking | +0.026 |
ths | +0.026 |
agg | +0.025 |
ering | +0.024 |
prisingly | +0.023 |
Neg logit contributions | |
He | -0.024 |
ashes | -0.024 |
They | -0.024 |
. | -0.024 |
rider | -0.023 |
Token | they |
Token ID | 484 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
eking | +0.039 |
ths | +0.038 |
agg | +0.037 |
aman | +0.035 |
iter | +0.035 |
Neg logit contributions | |
. | -0.030 |
He | -0.026 |
." | -0.026 |
They | -0.025 |
Let | -0.022 |
Token | said |
Token ID | 531 |
Feature activation | +7.26e-02 |
Pos logit contributions | |
. | +0.046 |
." | +0.040 |
He | +0.038 |
They | +0.036 |
ny | +0.033 |
Neg logit contributions | |
ths | -0.064 |
agg | -0.061 |
eking | -0.059 |
aman | -0.058 |
prisingly | -0.057 |
Token | The |
Token ID | 464 |
Feature activation | +1.32e-02 |
Pos logit contributions | |
ths | +0.201 |
agg | +0.192 |
eking | +0.180 |
Jess | +0.177 |
Kevin | +0.175 |
Neg logit contributions | |
. | -0.284 |
ashes | -0.271 |
." | -0.270 |
rider | -0.247 |
eer | -0.244 |
Token | fox |
Token ID | 21831 |
Feature activation | -3.38e-03 |
Pos logit contributions | |
ths | +0.030 |
agg | +0.029 |
eking | +0.029 |
aman | +0.028 |
Kevin | +0.027 |
Neg logit contributions | |
. | -0.016 |
." | -0.014 |
ashes | -0.013 |
He | -0.013 |
rider | -0.012 |
Token | said |
Token ID | 531 |
Feature activation | +7.64e-02 |
Pos logit contributions | |
. | +0.005 |
." | +0.004 |
He | +0.004 |
ny | +0.004 |
play | +0.004 |
Neg logit contributions | |
ths | -0.007 |
eking | -0.007 |
agg | -0.007 |
aman | -0.007 |
Kevin | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +8.54e-02 |
Pos logit contributions | |
ths | +0.169 |
agg | +0.164 |
eking | +0.161 |
aman | +0.157 |
Kevin | +0.150 |
Neg logit contributions | |
. | -0.095 |
." | -0.087 |
ashes | -0.081 |
He | -0.078 |
rider | -0.074 |
Token | " |
Token ID | 366 |
Feature activation | +6.49e-02 |
Pos logit contributions | |
ths | +0.191 |
agg | +0.186 |
eking | +0.183 |
aman | +0.179 |
prisingly | +0.171 |
Neg logit contributions | |
. | -0.106 |
." | -0.096 |
He | -0.087 |
ashes | -0.085 |
They | -0.081 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
ths | +0.132 |
agg | +0.131 |
eking | +0.130 |
aman | +0.121 |
ering | +0.117 |
Neg logit contributions | |
. | -0.091 |
." | -0.084 |
He | -0.082 |
Let | -0.079 |
ny | -0.078 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.67e-02 |
Pos logit contributions | |
ths | +0.393 |
agg | +0.379 |
eking | +0.365 |
aman | +0.362 |
Kevin | +0.339 |
Neg logit contributions | |
. | -0.472 |
." | -0.442 |
He | -0.405 |
ashes | -0.404 |
rider | -0.393 |
Token | , |
Token ID | 11 |
Feature activation | +8.45e-02 |
Pos logit contributions | |
ths | +0.163 |
agg | +0.159 |
eking | +0.157 |
prisingly | +0.146 |
aman | +0.146 |
Neg logit contributions | |
ashes | -0.158 |
. | -0.141 |
They | -0.136 |
." | -0.135 |
rider | -0.135 |
Token | I |
Token ID | 314 |
Feature activation | -7.14e-02 |
Pos logit contributions | |
ths | +0.181 |
agg | +0.176 |
eking | +0.173 |
aman | +0.169 |
prisingly | +0.162 |
Neg logit contributions | |
. | -0.114 |
." | -0.105 |
He | -0.094 |
ashes | -0.090 |
Let | -0.089 |
Token | shouldn |
Token ID | 6584 |
Feature activation | +1.02e-02 |
Pos logit contributions | |
. | +0.090 |
." | +0.082 |
ashes | +0.076 |
He | +0.074 |
rider | +0.070 |
Neg logit contributions | |
ths | -0.158 |
agg | -0.153 |
eking | -0.150 |
aman | -0.147 |
Kevin | -0.140 |
Token | 't |
Token ID | 470 |
Feature activation | +9.81e-02 |
Pos logit contributions | |
ths | +0.013 |
agg | +0.013 |
eking | +0.012 |
aman | +0.012 |
prisingly | +0.011 |
Neg logit contributions | |
. | -0.023 |
." | -0.022 |
He | -0.021 |
Let | -0.020 |
They | -0.020 |
Token | her |
Token ID | 607 |
Feature activation | -3.57e-02 |
Pos logit contributions | |
ths | +0.088 |
agg | +0.086 |
eking | +0.084 |
aman | +0.083 |
Jess | +0.080 |
Neg logit contributions | |
. | -0.052 |
." | -0.046 |
He | -0.042 |
ashes | -0.041 |
They | -0.039 |
Token | . |
Token ID | 13 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
ashes | +0.043 |
five | +0.042 |
v | +0.041 |
round | +0.039 |
They | +0.039 |
Neg logit contributions | |
ths | -0.066 |
agg | -0.063 |
nowhere | -0.062 |
eking | -0.062 |
aman | -0.061 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
. | +0.155 |
." | +0.140 |
ashes | +0.132 |
rider | +0.119 |
He | +0.116 |
Neg logit contributions | |
ths | -0.340 |
agg | -0.334 |
eking | -0.325 |
aman | -0.321 |
Kevin | -0.311 |
Token |
|
Token ID | 198 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
ths | +0.361 |
eking | +0.347 |
agg | +0.341 |
aman | +0.333 |
prisingly | +0.319 |
Neg logit contributions | |
. | -0.169 |
." | -0.148 |
He | -0.136 |
They | -0.126 |
ashes | -0.114 |
Token | " |
Token ID | 1 |
Feature activation | +2.42e-02 |
Pos logit contributions | |
ths | +0.256 |
agg | +0.246 |
eking | +0.241 |
aman | +0.234 |
Kevin | +0.221 |
Neg logit contributions | |
. | -0.242 |
." | -0.227 |
ashes | -0.215 |
He | -0.210 |
rider | -0.202 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
ths | +0.047 |
agg | +0.046 |
eking | +0.046 |
aman | +0.044 |
ering | +0.041 |
Neg logit contributions | |
. | -0.037 |
." | -0.034 |
He | -0.033 |
ashes | -0.032 |
They | -0.031 |
Token | - |
Token ID | 12 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
ths | +0.546 |
agg | +0.532 |
aman | +0.524 |
eking | +0.514 |
Kevin | +0.499 |
Neg logit contributions | |
. | -0.329 |
." | -0.297 |
He | -0.248 |
ashes | -0.247 |
rider | -0.236 |
Token | oh |
Token ID | 1219 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
ths | +0.025 |
agg | +0.024 |
eking | +0.023 |
aman | +0.021 |
Kevin | +0.020 |
Neg logit contributions | |
. | -0.057 |
." | -0.054 |
ashes | -0.052 |
He | -0.051 |
rider | -0.050 |
Token | , |
Token ID | 11 |
Feature activation | +8.66e-02 |
Pos logit contributions | |
ths | +0.348 |
eking | +0.339 |
agg | +0.334 |
aman | +0.323 |
prisingly | +0.310 |
Neg logit contributions | |
ashes | -0.192 |
. | -0.188 |
He | -0.181 |
." | -0.179 |
They | -0.178 |
Token | Lily |
Token ID | 20037 |
Feature activation | +3.78e-03 |
Pos logit contributions | |
ths | +0.198 |
agg | +0.195 |
eking | +0.190 |
aman | +0.185 |
Kevin | +0.180 |
Neg logit contributions | |
. | -0.106 |
." | -0.096 |
He | -0.085 |
ashes | -0.082 |
Let | -0.079 |
Token | ," |
Token ID | 553 |
Feature activation | +3.24e-02 |
Pos logit contributions | |
ths | +0.007 |
eking | +0.006 |
aman | +0.006 |
agg | +0.006 |
ering | +0.006 |
Neg logit contributions | |
ashes | -0.006 |
powers | -0.005 |
five | -0.005 |
towards | -0.005 |
He | -0.005 |
Token | everywhere |
Token ID | 8347 |
Feature activation | -1.76e-03 |
Pos logit contributions | |
. | +0.110 |
." | +0.107 |
ashes | +0.101 |
rider | +0.101 |
He | +0.100 |
Neg logit contributions | |
ths | -0.168 |
eking | -0.162 |
agg | -0.151 |
aman | -0.150 |
prisingly | -0.145 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ashes | +0.001 |
rider | +0.001 |
five | +0.000 |
pedal | +0.000 |
eer | +0.000 |
Neg logit contributions | |
ths | -0.005 |
eking | -0.005 |
agg | -0.005 |
ering | -0.005 |
Jess | -0.005 |
Token |
|
Token ID | 198 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
. | +0.153 |
." | +0.138 |
He | +0.131 |
They | +0.123 |
ashes | +0.113 |
Neg logit contributions | |
ths | -0.362 |
eking | -0.351 |
agg | -0.350 |
aman | -0.337 |
prisingly | -0.323 |
Token |
|
Token ID | 198 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
ths | +0.393 |
eking | +0.382 |
aman | +0.368 |
agg | +0.367 |
ering | +0.350 |
Neg logit contributions | |
. | -0.148 |
." | -0.127 |
He | -0.121 |
They | -0.113 |
Let | -0.091 |
Token | " |
Token ID | 1 |
Feature activation | +2.56e-02 |
Pos logit contributions | |
agg | +0.286 |
eking | +0.281 |
ths | +0.281 |
aman | +0.270 |
Kevin | +0.252 |
Neg logit contributions | |
. | -0.218 |
." | -0.198 |
He | -0.191 |
They | -0.183 |
Let | -0.177 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
agg | +0.049 |
eking | +0.048 |
ths | +0.048 |
ering | +0.046 |
aman | +0.045 |
Neg logit contributions | |
. | -0.039 |
." | -0.037 |
Let | -0.035 |
He | -0.035 |
ashes | -0.034 |
Token | - |
Token ID | 12 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
ths | +0.558 |
agg | +0.548 |
aman | +0.539 |
eking | +0.527 |
Kevin | +0.517 |
Neg logit contributions | |
. | -0.318 |
." | -0.287 |
ashes | -0.235 |
He | -0.234 |
Let | -0.224 |
Token | oh |
Token ID | 1219 |
Feature activation | +1.66e-01 |
Pos logit contributions | |
ths | +0.028 |
agg | +0.027 |
eking | +0.026 |
aman | +0.025 |
Kevin | +0.022 |
Neg logit contributions | |
. | -0.062 |
." | -0.059 |
ashes | -0.057 |
He | -0.056 |
rider | -0.055 |
Token | !" |
Token ID | 2474 |
Feature activation | -2.96e-02 |
Pos logit contributions | |
ths | +0.383 |
agg | +0.371 |
eking | +0.371 |
aman | +0.358 |
prisingly | +0.345 |
Neg logit contributions | |
. | -0.169 |
ashes | -0.158 |
." | -0.155 |
He | -0.150 |
They | -0.144 |
Token | Tim |
Token ID | 5045 |
Feature activation | +3.94e-02 |
Pos logit contributions | |
. | +0.036 |
." | +0.031 |
He | +0.030 |
They | +0.028 |
ashes | +0.026 |
Neg logit contributions | |
ths | -0.069 |
agg | -0.068 |
eking | -0.067 |
aman | -0.065 |
Kevin | -0.063 |
Token | said |
Token ID | 531 |
Feature activation | +6.61e-02 |
Pos logit contributions | |
ths | +0.069 |
eking | +0.065 |
agg | +0.065 |
aman | +0.062 |
prisingly | +0.059 |
Neg logit contributions | |
. | -0.063 |
." | -0.060 |
ashes | -0.060 |
He | -0.056 |
rider | -0.056 |
Token | Lily |
Token ID | 20037 |
Feature activation | +5.56e-03 |
Pos logit contributions | |
. | +0.006 |
ashes | +0.005 |
." | +0.005 |
rider | +0.005 |
powers | +0.005 |
Neg logit contributions | |
ths | -0.008 |
agg | -0.008 |
eking | -0.008 |
aman | -0.007 |
Kevin | -0.007 |
Token | . |
Token ID | 13 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
eking | +0.015 |
ths | +0.015 |
ering | +0.014 |
aman | +0.014 |
agg | +0.014 |
Neg logit contributions | |
ashes | -0.003 |
rider | -0.002 |
powers | -0.002 |
." | -0.002 |
five | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
. | +0.170 |
." | +0.149 |
He | +0.142 |
They | +0.131 |
ashes | +0.130 |
Neg logit contributions | |
ths | -0.346 |
agg | -0.337 |
eking | -0.336 |
aman | -0.323 |
prisingly | -0.312 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
ths | +0.330 |
eking | +0.317 |
agg | +0.317 |
aman | +0.306 |
prisingly | +0.292 |
Neg logit contributions | |
. | -0.179 |
." | -0.161 |
He | -0.146 |
ashes | -0.138 |
They | -0.137 |
Token | " |
Token ID | 1 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
ths | +0.254 |
agg | +0.245 |
eking | +0.239 |
aman | +0.232 |
Kevin | +0.220 |
Neg logit contributions | |
. | -0.243 |
." | -0.228 |
ashes | -0.215 |
He | -0.211 |
rider | -0.203 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
eking | +0.051 |
agg | +0.050 |
aman | +0.050 |
ering | +0.048 |
ths | +0.047 |
Neg logit contributions | |
. | -0.031 |
They | -0.029 |
." | -0.028 |
ashes | -0.028 |
He | -0.027 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.65e-02 |
Pos logit contributions | |
ths | +0.524 |
agg | +0.518 |
aman | +0.508 |
eking | +0.490 |
Kevin | +0.486 |
Neg logit contributions | |
. | -0.359 |
." | -0.324 |
He | -0.267 |
ashes | -0.260 |
Let | -0.257 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
ths | +0.199 |
agg | +0.193 |
eking | +0.189 |
aman | +0.185 |
Kevin | +0.177 |
Neg logit contributions | |
. | -0.139 |
." | -0.128 |
ashes | -0.115 |
He | -0.114 |
Let | -0.109 |
Token | We |
Token ID | 775 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
. | +0.171 |
ashes | +0.152 |
rider | +0.151 |
." | +0.143 |
pedal | +0.139 |
Neg logit contributions | |
ths | -0.338 |
agg | -0.326 |
aman | -0.310 |
eking | -0.309 |
Jess | -0.308 |
Token | 're |
Token ID | 821 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
. | +0.033 |
." | +0.031 |
ashes | +0.028 |
He | +0.028 |
They | +0.027 |
Neg logit contributions | |
ths | -0.043 |
agg | -0.041 |
eking | -0.040 |
aman | -0.039 |
Kevin | -0.037 |
Token | in |
Token ID | 287 |
Feature activation | -5.43e-02 |
Pos logit contributions | |
ths | +0.215 |
agg | +0.212 |
aman | +0.202 |
eking | +0.195 |
Kevin | +0.193 |
Neg logit contributions | |
. | -0.215 |
." | -0.202 |
ashes | -0.174 |
He | -0.173 |
like | -0.170 |
Token | scared |
Token ID | 12008 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
ths | +0.129 |
aman | +0.120 |
eking | +0.117 |
agg | +0.116 |
prisingly | +0.109 |
Neg logit contributions | |
. | -0.153 |
." | -0.141 |
He | -0.133 |
ashes | -0.130 |
They | -0.127 |
Token | , |
Token ID | 11 |
Feature activation | +9.13e-02 |
Pos logit contributions | |
ashes | +0.015 |
powers | +0.013 |
rider | +0.013 |
." | +0.013 |
five | +0.013 |
Neg logit contributions | |
ths | -0.030 |
aman | -0.029 |
eking | -0.029 |
agg | -0.029 |
unfortunately | -0.027 |
Token | she |
Token ID | 673 |
Feature activation | -4.65e-02 |
Pos logit contributions | |
ths | +0.192 |
agg | +0.190 |
eking | +0.182 |
aman | +0.175 |
prisingly | +0.174 |
Neg logit contributions | |
. | -0.137 |
." | -0.119 |
He | -0.114 |
They | -0.106 |
towards | -0.104 |
Token | said |
Token ID | 531 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
. | +0.077 |
." | +0.073 |
ashes | +0.070 |
He | +0.067 |
rider | +0.065 |
Neg logit contributions | |
ths | -0.082 |
agg | -0.079 |
eking | -0.078 |
aman | -0.075 |
Kevin | -0.071 |
Token | " |
Token ID | 366 |
Feature activation | +6.62e-02 |
Pos logit contributions | |
ths | +0.185 |
agg | +0.179 |
eking | +0.177 |
aman | +0.173 |
prisingly | +0.167 |
Neg logit contributions | |
. | -0.093 |
." | -0.080 |
He | -0.070 |
ashes | -0.065 |
They | -0.065 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
agg | +0.136 |
ths | +0.135 |
eking | +0.132 |
aman | +0.122 |
prisingly | +0.121 |
Neg logit contributions | |
. | -0.092 |
." | -0.084 |
He | -0.077 |
rider | -0.076 |
Let | -0.075 |
Token | - |
Token ID | 12 |
Feature activation | +2.33e-02 |
Pos logit contributions | |
ths | +0.443 |
agg | +0.424 |
eking | +0.423 |
aman | +0.397 |
prisingly | +0.386 |
Neg logit contributions | |
. | -0.400 |
." | -0.376 |
ashes | -0.372 |
He | -0.361 |
They | -0.348 |
Token | oh |
Token ID | 1219 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
ths | +0.029 |
agg | +0.027 |
eking | +0.026 |
aman | +0.025 |
Kevin | +0.023 |
Neg logit contributions | |
. | -0.052 |
." | -0.049 |
ashes | -0.048 |
He | -0.047 |
They | -0.046 |
Token | ", |
Token ID | 1600 |
Feature activation | +1.80e-03 |
Pos logit contributions | |
ths | +0.302 |
eking | +0.294 |
agg | +0.288 |
aman | +0.282 |
tha | +0.271 |
Neg logit contributions | |
ashes | -0.236 |
They | -0.215 |
He | -0.211 |
powers | -0.209 |
. | -0.204 |
Token | and |
Token ID | 290 |
Feature activation | +4.81e-02 |
Pos logit contributions | |
ths | +0.004 |
agg | +0.004 |
eking | +0.004 |
aman | +0.004 |
prisingly | +0.004 |
Neg logit contributions | |
. | -0.002 |
." | -0.002 |
He | -0.002 |
They | -0.001 |
ashes | -0.001 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.22e-01 |
Pos logit contributions | |
ths | +0.087 |
agg | +0.085 |
eking | +0.083 |
aman | +0.081 |
Kevin | +0.078 |
Neg logit contributions | |
. | -0.080 |
." | -0.074 |
He | -0.069 |
ashes | -0.066 |
They | -0.065 |
Token | in |
Token ID | 287 |
Feature activation | -5.68e-02 |
Pos logit contributions | |
ths | +0.063 |
eking | +0.062 |
agg | +0.060 |
aman | +0.057 |
prisingly | +0.057 |
Neg logit contributions | |
ashes | -0.029 |
rider | -0.028 |
. | -0.028 |
." | -0.027 |
five | -0.026 |
Token | and |
Token ID | 290 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
. | +0.042 |
ashes | +0.042 |
." | +0.041 |
rider | +0.037 |
He | +0.034 |
Neg logit contributions | |
ths | -0.142 |
agg | -0.138 |
aman | -0.137 |
eking | -0.136 |
Kevin | -0.131 |
Token | said |
Token ID | 531 |
Feature activation | +7.77e-02 |
Pos logit contributions | |
agg | +0.106 |
ths | +0.103 |
eking | +0.100 |
aman | +0.099 |
prisingly | +0.094 |
Neg logit contributions | |
. | -0.065 |
." | -0.054 |
play | -0.048 |
He | -0.048 |
towards | -0.045 |
Token | , |
Token ID | 11 |
Feature activation | +8.82e-02 |
Pos logit contributions | |
ths | +0.191 |
eking | +0.187 |
agg | +0.187 |
aman | +0.183 |
prisingly | +0.177 |
Neg logit contributions | |
. | -0.092 |
." | -0.077 |
He | -0.070 |
They | -0.064 |
Let | -0.059 |
Token | " |
Token ID | 366 |
Feature activation | +6.48e-02 |
Pos logit contributions | |
agg | +0.228 |
ths | +0.226 |
eking | +0.219 |
aman | +0.215 |
prisingly | +0.211 |
Neg logit contributions | |
. | -0.095 |
." | -0.077 |
He | -0.073 |
They | -0.068 |
Let | -0.062 |
Token | Uh |
Token ID | 34653 |
Feature activation | +2.47e-01 |
Pos logit contributions | |
agg | +0.146 |
eking | +0.137 |
ths | +0.136 |
iter | +0.124 |
ering | +0.123 |
Neg logit contributions | |
. | -0.085 |
He | -0.081 |
Let | -0.078 |
." | -0.076 |
They | -0.075 |
Token | oh |
Token ID | 11752 |
Feature activation | +9.62e-02 |
Pos logit contributions | |
ths | +0.403 |
agg | +0.402 |
aman | +0.384 |
eking | +0.377 |
Kevin | +0.366 |
Neg logit contributions | |
. | -0.468 |
." | -0.432 |
He | -0.380 |
play | -0.374 |
rider | -0.374 |
Token | , |
Token ID | 11 |
Feature activation | +8.71e-02 |
Pos logit contributions | |
ths | +0.180 |
agg | +0.171 |
eking | +0.170 |
aman | +0.163 |
prisingly | +0.157 |
Neg logit contributions | |
. | -0.141 |
ashes | -0.139 |
." | -0.134 |
He | -0.126 |
rider | -0.125 |
Token | you |
Token ID | 345 |
Feature activation | +5.75e-03 |
Pos logit contributions | |
agg | +0.208 |
ths | +0.204 |
eking | +0.201 |
aman | +0.195 |
Kevin | +0.194 |
Neg logit contributions | |
. | -0.099 |
." | -0.090 |
He | -0.080 |
They | -0.075 |
Let | -0.075 |
Token | boys |
Token ID | 6510 |
Feature activation | -5.21e-02 |
Pos logit contributions | |
ths | +0.012 |
agg | +0.011 |
eking | +0.011 |
aman | +0.011 |
Kevin | +0.010 |
Neg logit contributions | |
. | -0.008 |
ashes | -0.007 |
." | -0.007 |
He | -0.006 |
towards | -0.006 |
Token | made |
Token ID | 925 |
Feature activation | -2.86e-02 |
Pos logit contributions | |
ashes | +0.049 |
He | +0.049 |
powers | +0.048 |
rider | +0.045 |
five | +0.045 |
Neg logit contributions | |
eking | -0.114 |
ths | -0.113 |
aman | -0.109 |
ering | -0.105 |
done | -0.102 |
Token | " |
Token ID | 366 |
Feature activation | +5.84e-02 |
Pos logit contributions | |
ashes | +0.071 |
powers | +0.061 |
eer | +0.056 |
ny | +0.055 |
five | +0.054 |
Neg logit contributions | |
ths | -0.083 |
unfortunately | -0.077 |
eking | -0.077 |
aman | -0.075 |
lf | -0.073 |
Token | news |
Token ID | 10827 |
Feature activation | +6.63e-02 |
Pos logit contributions | |
ths | +0.108 |
agg | +0.104 |
eking | +0.102 |
aman | +0.099 |
prisingly | +0.094 |
Neg logit contributions | |
. | -0.092 |
." | -0.086 |
ashes | -0.081 |
He | -0.079 |
rider | -0.076 |
Token | " |
Token ID | 1 |
Feature activation | +1.47e-02 |
Pos logit contributions | |
ths | +0.109 |
agg | +0.104 |
eking | +0.101 |
aman | +0.097 |
Kevin | +0.093 |
Neg logit contributions | |
. | -0.113 |
ashes | -0.108 |
." | -0.106 |
He | -0.103 |
rider | -0.100 |
Token | on |
Token ID | 319 |
Feature activation | -6.15e-02 |
Pos logit contributions | |
ths | +0.033 |
agg | +0.031 |
eking | +0.031 |
aman | +0.030 |
Kevin | +0.029 |
Neg logit contributions | |
. | -0.016 |
." | -0.015 |
ashes | -0.015 |
Let | -0.013 |
five | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | +1.85e-02 |
Pos logit contributions | |
. | +0.086 |
." | +0.081 |
ashes | +0.079 |
He | +0.073 |
rider | +0.071 |
Neg logit contributions | |
ths | -0.123 |
agg | -0.118 |
eking | -0.116 |
aman | -0.114 |
Kevin | -0.108 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ths | +0.045 |
unfortunately | +0.043 |
eking | +0.043 |
ering | +0.041 |
nowhere | +0.041 |
Neg logit contributions | |
ashes | -0.014 |
five | -0.013 |
powers | -0.011 |
eer | -0.011 |
likes | -0.010 |
Token | Tim |
Token ID | 5045 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ashes | +0.223 |
rider | +0.223 |
. | +0.217 |
." | +0.208 |
ny | +0.202 |
Neg logit contributions | |
agg | -0.315 |
ths | -0.312 |
aman | -0.289 |
Jane | -0.279 |
eking | -0.278 |
Token | was |
Token ID | 373 |
Feature activation | +5.10e-02 |
Pos logit contributions | |
ths | +0.076 |
agg | +0.073 |
eking | +0.073 |
aman | +0.071 |
Kevin | +0.067 |
Neg logit contributions | |
. | -0.048 |
." | -0.045 |
ashes | -0.044 |
He | -0.041 |
rider | -0.040 |
Token | very |
Token ID | 845 |
Feature activation | +6.50e-02 |
Pos logit contributions | |
ths | +0.100 |
agg | +0.097 |
aman | +0.095 |
eking | +0.092 |
Kevin | +0.088 |
Neg logit contributions | |
. | -0.082 |
." | -0.075 |
He | -0.070 |
ashes | -0.067 |
They | -0.065 |
Token | curious |
Token ID | 11040 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
ths | +0.102 |
agg | +0.098 |
eking | +0.096 |
aman | +0.092 |
Kevin | +0.087 |
Neg logit contributions | |
. | -0.123 |
." | -0.116 |
ashes | -0.110 |
He | -0.108 |
rider | -0.105 |
Token | about |
Token ID | 546 |
Feature activation | +4.04e-02 |
Pos logit contributions | |
ashes | +0.015 |
five | +0.012 |
powers | +0.011 |
atters | +0.011 |
rider | +0.010 |
Neg logit contributions | |
ths | -0.030 |
ering | -0.028 |
Jane | -0.028 |
agg | -0.028 |
nowhere | -0.028 |
Token | " |
Token ID | 366 |
Feature activation | +5.91e-02 |
Pos logit contributions | |
ths | +0.184 |
agg | +0.180 |
eking | +0.176 |
aman | +0.173 |
prisingly | +0.166 |
Neg logit contributions | |
. | -0.095 |
." | -0.085 |
He | -0.077 |
ashes | -0.074 |
They | -0.072 |
Token | I |
Token ID | 40 |
Feature activation | -8.29e-02 |
Pos logit contributions | |
ths | +0.074 |
Fore | +0.069 |
aman | +0.069 |
Jess | +0.069 |
Kevin | +0.068 |
Neg logit contributions | |
. | -0.120 |
." | -0.112 |
ashes | -0.109 |
powers | -0.102 |
towards | -0.100 |
Token | love |
Token ID | 1842 |
Feature activation | +2.96e-02 |
Pos logit contributions | |
. | +0.093 |
." | +0.087 |
ashes | +0.086 |
He | +0.082 |
Let | +0.079 |
Neg logit contributions | |
agg | -0.175 |
ths | -0.170 |
prisingly | -0.167 |
eking | -0.166 |
Kevin | -0.163 |
Token | this |
Token ID | 428 |
Feature activation | +2.90e-02 |
Pos logit contributions | |
ths | +0.041 |
eking | +0.037 |
agg | +0.033 |
ering | +0.031 |
aman | +0.031 |
Neg logit contributions | |
ashes | -0.058 |
ny | -0.054 |
. | -0.053 |
He | -0.052 |
five | -0.052 |
Token | shirt |
Token ID | 10147 |
Feature activation | -4.53e-02 |
Pos logit contributions | |
pron | +0.047 |
mber | +0.047 |
agg | +0.046 |
aman | +0.046 |
eking | +0.046 |
Neg logit contributions | |
ashes | -0.040 |
He | -0.038 |
five | -0.037 |
They | -0.036 |
powers | -0.035 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ashes | +0.056 |
five | +0.049 |
Let | +0.046 |
pedal | +0.046 |
powers | +0.046 |
Neg logit contributions | |
ths | -0.080 |
eking | -0.075 |
exactly | -0.074 |
done | -0.074 |
agg | -0.074 |
Token | It |
Token ID | 632 |
Feature activation | +3.41e-02 |
Pos logit contributions | |
ashes | +0.255 |
. | +0.245 |
powers | +0.206 |
towards | +0.202 |
bars | +0.198 |
Neg logit contributions | |
mmm | -0.285 |
omm | -0.264 |
unfortunately | -0.263 |
hh | -0.256 |
prisingly | -0.256 |
Token | is |
Token ID | 318 |
Feature activation | +5.75e-02 |
Pos logit contributions | |
ths | +0.062 |
eking | +0.061 |
agg | +0.058 |
aman | +0.055 |
prisingly | +0.055 |
Neg logit contributions | |
ashes | -0.045 |
. | -0.042 |
five | -0.042 |
towards | -0.040 |
." | -0.038 |
Token | so |
Token ID | 523 |
Feature activation | +7.21e-02 |
Pos logit contributions | |
ths | +0.098 |
eking | +0.095 |
agg | +0.094 |
aman | +0.090 |
prisingly | +0.086 |
Neg logit contributions | |
. | -0.091 |
ashes | -0.089 |
." | -0.086 |
five | -0.083 |
He | -0.081 |
Token | nice |
Token ID | 3621 |
Feature activation | +4.02e-04 |
Pos logit contributions | |
agg | +0.124 |
ths | +0.119 |
eking | +0.119 |
Kevin | +0.109 |
impressed | +0.109 |
Neg logit contributions | |
. | -0.107 |
ashes | -0.106 |
." | -0.100 |
five | -0.099 |
rider | -0.098 |
Token | !" |
Token ID | 2474 |
Feature activation | -6.48e-02 |
Pos logit contributions | |
mber | +0.001 |
ths | +0.001 |
agg | +0.001 |
ll | +0.001 |
prisingly | +0.001 |
Neg logit contributions | |
pedal | -0.000 |
pedals | -0.000 |
tire | -0.000 |
ashes | -0.000 |
five | -0.000 |
Token | and |
Token ID | 290 |
Feature activation | +4.43e-02 |
Pos logit contributions | |
ashes | +0.062 |
powers | +0.058 |
five | +0.057 |
Let | +0.057 |
They | +0.053 |
Neg logit contributions | |
eking | -0.077 |
ths | -0.074 |
ering | -0.074 |
aman | -0.072 |
setup | -0.068 |
Token | wide |
Token ID | 3094 |
Feature activation | -6.94e-03 |
Pos logit contributions | |
ths | +0.076 |
eking | +0.070 |
agg | +0.070 |
prisingly | +0.065 |
impressed | +0.064 |
Neg logit contributions | |
. | -0.074 |
ashes | -0.072 |
." | -0.071 |
ny | -0.065 |
He | -0.065 |
Token | for |
Token ID | 329 |
Feature activation | -6.43e-03 |
Pos logit contributions | |
ashes | +0.007 |
five | +0.007 |
round | +0.005 |
iffs | +0.005 |
ny | +0.005 |
Neg logit contributions | |
ths | -0.016 |
eking | -0.015 |
exactly | -0.014 |
impressed | -0.014 |
ering | -0.014 |
Token | his |
Token ID | 465 |
Feature activation | -1.67e-02 |
Pos logit contributions | |
. | +0.010 |
." | +0.009 |
ashes | +0.008 |
He | +0.008 |
rider | +0.008 |
Neg logit contributions | |
ths | -0.013 |
agg | -0.012 |
eking | -0.012 |
aman | -0.012 |
Kevin | -0.011 |
Token | amazing |
Token ID | 4998 |
Feature activation | +4.80e-02 |
Pos logit contributions | |
ashes | +0.025 |
. | +0.024 |
." | +0.024 |
ny | +0.023 |
He | +0.023 |
Neg logit contributions | |
ths | -0.032 |
agg | -0.028 |
eking | -0.027 |
aman | -0.027 |
Kevin | -0.026 |
Token | feats |
Token ID | 35664 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
ths | +0.098 |
agg | +0.088 |
aman | +0.085 |
eking | +0.082 |
Kevin | +0.078 |
Neg logit contributions | |
ashes | -0.065 |
They | -0.060 |
." | -0.058 |
ny | -0.058 |
He | -0.057 |
Token | ! |
Token ID | 0 |
Feature activation | -1.30e-01 |
Pos logit contributions | |
ashes | +0.122 |
ny | +0.083 |
bone | +0.081 |
fur | +0.074 |
powers | +0.070 |
Neg logit contributions | |
ths | -0.435 |
eking | -0.426 |
agg | -0.413 |
prisingly | -0.401 |
ering | -0.398 |
Token | He |
Token ID | 679 |
Feature activation | -2.15e-02 |
Pos logit contributions | |
ashes | +0.176 |
. | +0.156 |
rider | +0.148 |
." | +0.146 |
towards | +0.144 |
Neg logit contributions | |
aman | -0.246 |
agg | -0.242 |
ths | -0.234 |
Jess | -0.232 |
Jane | -0.228 |
Token | could |
Token ID | 714 |
Feature activation | -1.85e-02 |
Pos logit contributions | |
. | +0.028 |
." | +0.027 |
ashes | +0.027 |
He | +0.024 |
Let | +0.023 |
Neg logit contributions | |
ths | -0.045 |
agg | -0.043 |
eking | -0.041 |
aman | -0.040 |
Kevin | -0.039 |
Token | jump |
Token ID | 4391 |
Feature activation | -5.66e-02 |
Pos logit contributions | |
. | +0.024 |
." | +0.023 |
ashes | +0.022 |
Let | +0.021 |
He | +0.020 |
Neg logit contributions | |
ths | -0.038 |
agg | -0.037 |
eking | -0.036 |
aman | -0.034 |
Kevin | -0.034 |
Token | higher |
Token ID | 2440 |
Feature activation | -7.21e-03 |
Pos logit contributions | |
. | +0.066 |
." | +0.064 |
ashes | +0.063 |
rider | +0.058 |
Let | +0.058 |
Neg logit contributions | |
ths | -0.118 |
agg | -0.115 |
eking | -0.112 |
aman | -0.108 |
Jess | -0.105 |
Token | - |
Token ID | 12 |
Feature activation | +2.14e-02 |
Pos logit contributions | |
ths | +0.069 |
eking | +0.066 |
agg | +0.066 |
aman | +0.064 |
ering | +0.060 |
Neg logit contributions | |
. | -0.044 |
ashes | -0.043 |
He | -0.041 |
." | -0.040 |
Let | -0.038 |
Token | ch |
Token ID | 354 |
Feature activation | +2.36e-02 |
Pos logit contributions | |
ths | +0.042 |
eking | +0.040 |
agg | +0.039 |
aman | +0.038 |
prisingly | +0.037 |
Neg logit contributions | |
. | -0.030 |
." | -0.028 |
ashes | -0.027 |
He | -0.026 |
They | -0.025 |
Token | oo |
Token ID | 2238 |
Feature activation | +3.45e-02 |
Pos logit contributions | |
ths | +0.026 |
agg | +0.025 |
eking | +0.024 |
aman | +0.023 |
Kevin | +0.021 |
Neg logit contributions | |
. | -0.055 |
." | -0.053 |
ashes | -0.050 |
He | -0.050 |
rider | -0.048 |
Token | " |
Token ID | 1 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
ths | +0.073 |
eking | +0.070 |
agg | +0.069 |
aman | +0.067 |
ering | +0.064 |
Neg logit contributions | |
. | -0.040 |
ashes | -0.039 |
He | -0.037 |
." | -0.036 |
Let | -0.035 |
Token | sound |
Token ID | 2128 |
Feature activation | +1.70e-03 |
Pos logit contributions | |
ths | +0.042 |
agg | +0.040 |
eking | +0.039 |
aman | +0.038 |
prisingly | +0.038 |
Neg logit contributions | |
. | -0.020 |
ashes | -0.019 |
." | -0.018 |
five | -0.017 |
He | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -1.69e-01 |
Pos logit contributions | |
ths | +0.004 |
eking | +0.004 |
agg | +0.004 |
aman | +0.004 |
Kevin | +0.003 |
Neg logit contributions | |
. | -0.002 |
ashes | -0.002 |
." | -0.002 |
He | -0.001 |
Let | -0.001 |
Token | The |
Token ID | 383 |
Feature activation | -4.78e-03 |
Pos logit contributions | |
ashes | +0.161 |
. | +0.153 |
rider | +0.152 |
ny | +0.135 |
powers | +0.133 |
Neg logit contributions | |
agg | -0.380 |
ths | -0.370 |
aman | -0.356 |
Jess | -0.348 |
prisingly | -0.339 |
Token | train |
Token ID | 4512 |
Feature activation | -2.59e-02 |
Pos logit contributions | |
. | +0.005 |
." | +0.004 |
ashes | +0.004 |
He | +0.004 |
rider | +0.003 |
Neg logit contributions | |
ths | -0.012 |
agg | -0.012 |
eking | -0.011 |
aman | -0.011 |
Kevin | -0.011 |
Token | liked |
Token ID | 8288 |
Feature activation | -9.43e-03 |
Pos logit contributions | |
. | +0.037 |
." | +0.035 |
ashes | +0.034 |
He | +0.032 |
Let | +0.030 |
Neg logit contributions | |
ths | -0.051 |
agg | -0.048 |
eking | -0.048 |
aman | -0.045 |
prisingly | -0.044 |
Token | to |
Token ID | 284 |
Feature activation | -4.30e-03 |
Pos logit contributions | |
. | +0.009 |
." | +0.008 |
ashes | +0.006 |
He | +0.006 |
They | +0.006 |
Neg logit contributions | |
ths | -0.024 |
agg | -0.024 |
eking | -0.023 |
aman | -0.023 |
Kevin | -0.022 |
Token | go |
Token ID | 467 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
ashes | +0.004 |
. | +0.004 |
powers | +0.004 |
pads | +0.004 |
Let | +0.004 |
Neg logit contributions | |
eking | -0.009 |
ths | -0.009 |
agg | -0.009 |
aman | -0.008 |
prisingly | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +8.42e-02 |
Pos logit contributions | |
powers | +0.122 |
ashes | +0.106 |
He | +0.102 |
Let | +0.100 |
rider | +0.098 |
Neg logit contributions | |
ths | -0.234 |
agg | -0.228 |
eking | -0.225 |
aman | -0.224 |
prisingly | -0.216 |
Token | slide |
Token ID | 10649 |
Feature activation | -6.22e-02 |
Pos logit contributions | |
ths | +0.218 |
agg | +0.212 |
eking | +0.209 |
aman | +0.205 |
Kevin | +0.198 |
Neg logit contributions | |
. | -0.074 |
." | -0.065 |
ashes | -0.057 |
He | -0.055 |
rider | -0.050 |
Token | and |
Token ID | 290 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
powers | +0.061 |
ashes | +0.058 |
He | +0.055 |
. | +0.055 |
." | +0.055 |
Neg logit contributions | |
ths | -0.138 |
aman | -0.130 |
eking | -0.128 |
agg | -0.127 |
ering | -0.121 |
Token | see |
Token ID | 766 |
Feature activation | +1.11e-02 |
Pos logit contributions | |
ths | +0.095 |
agg | +0.092 |
eking | +0.091 |
Jess | +0.088 |
aman | +0.087 |
Neg logit contributions | |
. | -0.043 |
." | -0.041 |
He | -0.038 |
ashes | -0.037 |
powers | -0.036 |
Token | - |
Token ID | 12 |
Feature activation | +2.38e-02 |
Pos logit contributions | |
ths | +0.015 |
aman | +0.014 |
impressed | +0.014 |
eking | +0.014 |
agg | +0.014 |
Neg logit contributions | |
powers | -0.020 |
ny | -0.020 |
ashes | -0.019 |
ro | -0.019 |
five | -0.019 |
Token | saw |
Token ID | 43439 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
ths | +0.053 |
agg | +0.051 |
eking | +0.051 |
aman | +0.049 |
Kevin | +0.047 |
Neg logit contributions | |
. | -0.029 |
." | -0.026 |
ashes | -0.024 |
He | -0.024 |
rider | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
powers | +0.099 |
pedal | +0.091 |
ray | +0.080 |
He | +0.079 |
name | +0.077 |
Neg logit contributions | |
ths | -0.409 |
eking | -0.391 |
ering | -0.376 |
Fore | -0.366 |
agg | -0.362 |
Token | But |
Token ID | 887 |
Feature activation | +8.67e-02 |
Pos logit contributions | |
. | +0.149 |
." | +0.136 |
rider | +0.129 |
ashes | +0.122 |
powers | +0.115 |
Neg logit contributions | |
agg | -0.339 |
ths | -0.333 |
aman | -0.329 |
eking | -0.315 |
Jess | -0.308 |
Token | their |
Token ID | 511 |
Feature activation | +1.39e-02 |
Pos logit contributions | |
agg | +0.216 |
ths | +0.212 |
eking | +0.210 |
Kevin | +0.206 |
Jess | +0.201 |
Neg logit contributions | |
. | -0.101 |
." | -0.083 |
play | -0.073 |
like | -0.073 |
bar | -0.072 |
Token | favorite |
Token ID | 4004 |
Feature activation | +6.18e-02 |
Pos logit contributions | |
ths | +0.026 |
agg | +0.025 |
eking | +0.024 |
aman | +0.024 |
Kevin | +0.022 |
Neg logit contributions | |
. | -0.021 |
." | -0.020 |
He | -0.019 |
ashes | -0.019 |
Let | -0.018 |
Token | thing |
Token ID | 1517 |
Feature activation | +1.15e-02 |
Pos logit contributions | |
ths | +0.132 |
agg | +0.127 |
aman | +0.127 |
eking | +0.124 |
Jane | +0.117 |
Neg logit contributions | |
ashes | -0.064 |
. | -0.061 |
." | -0.060 |
He | -0.059 |
powers | -0.057 |
Token | ck |
Token ID | 694 |
Feature activation | -2.80e-03 |
Pos logit contributions | |
. | +0.002 |
." | +0.002 |
ashes | +0.002 |
He | +0.002 |
Let | +0.001 |
Neg logit contributions | |
ths | -0.003 |
agg | -0.003 |
eking | -0.003 |
aman | -0.003 |
prisingly | -0.003 |
Token | " |
Token ID | 1 |
Feature activation | +2.04e-02 |
Pos logit contributions | |
ashes | +0.003 |
He | +0.003 |
. | +0.003 |
rider | +0.003 |
Let | +0.003 |
Neg logit contributions | |
ths | -0.006 |
agg | -0.006 |
eking | -0.006 |
aman | -0.005 |
ering | -0.005 |
Token | all |
Token ID | 477 |
Feature activation | -7.38e-02 |
Pos logit contributions | |
ths | +0.045 |
agg | +0.042 |
eking | +0.042 |
prisingly | +0.040 |
aman | +0.040 |
Neg logit contributions | |
ashes | -0.020 |
. | -0.020 |
." | -0.018 |
He | -0.017 |
five | -0.017 |
Token | day |
Token ID | 1110 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
. | +0.103 |
ashes | +0.103 |
powers | +0.102 |
bone | +0.100 |
rider | +0.100 |
Neg logit contributions | |
ths | -0.125 |
agg | -0.121 |
eking | -0.119 |
aman | -0.116 |
prisingly | -0.114 |
Token | long |
Token ID | 890 |
Feature activation | -4.89e-02 |
Pos logit contributions | |
ths | +0.110 |
eking | +0.108 |
agg | +0.106 |
aman | +0.102 |
ering | +0.100 |
Neg logit contributions | |
He | -0.019 |
ashes | -0.019 |
. | -0.016 |
rider | -0.016 |
Let | -0.016 |
Token | . |
Token ID | 13 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
ashes | +0.027 |
He | +0.024 |
rider | +0.021 |
pedal | +0.020 |
ny | +0.020 |
Neg logit contributions | |
ths | -0.132 |
eking | -0.126 |
agg | -0.121 |
aman | -0.119 |
ering | -0.118 |
Token |
|
Token ID | 198 |
Feature activation | +1.32e-01 |
Pos logit contributions | |
. | +0.199 |
." | +0.179 |
ashes | +0.177 |
rider | +0.164 |
He | +0.159 |
Neg logit contributions | |
ths | -0.366 |
agg | -0.359 |
eking | -0.344 |
aman | -0.342 |
Jess | -0.331 |
Token |
|
Token ID | 198 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
ths | +0.353 |
eking | +0.344 |
aman | +0.326 |
agg | +0.324 |
ering | +0.307 |
Neg logit contributions | |
. | -0.147 |
." | -0.123 |
He | -0.107 |
They | -0.102 |
play | -0.087 |
Token | One |
Token ID | 3198 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
ths | +0.227 |
eking | +0.220 |
agg | +0.220 |
aman | +0.209 |
Kevin | +0.196 |
Neg logit contributions | |
. | -0.246 |
." | -0.227 |
He | -0.212 |
They | -0.209 |
ashes | -0.205 |
Token | day |
Token ID | 1110 |
Feature activation | +4.11e-02 |
Pos logit contributions | |
agg | +0.028 |
ths | +0.027 |
prisingly | +0.024 |
eking | +0.024 |
Jess | +0.023 |
Neg logit contributions | |
ashes | -0.033 |
. | -0.032 |
." | -0.031 |
He | -0.030 |
powers | -0.030 |
Token | , |
Token ID | 11 |
Feature activation | +8.22e-02 |
Pos logit contributions | |
ths | +0.081 |
agg | +0.078 |
eking | +0.075 |
aman | +0.073 |
prisingly | +0.071 |
Neg logit contributions | |
. | -0.057 |
ashes | -0.054 |
." | -0.054 |
He | -0.051 |
rider | -0.050 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.73e-01 |
Pos logit contributions | |
agg | +0.165 |
ths | +0.160 |
Jess | +0.152 |
aman | +0.150 |
eking | +0.149 |
Neg logit contributions | |
ashes | -0.163 |
rider | -0.158 |
. | -0.156 |
." | -0.153 |
five | -0.150 |
Token | there |
Token ID | 612 |
Feature activation | -2.17e-02 |
Pos logit contributions | |
ths | +0.287 |
eking | +0.269 |
agg | +0.242 |
aman | +0.236 |
prisingly | +0.232 |
Neg logit contributions | |
. | -0.388 |
." | -0.340 |
He | -0.317 |
They | -0.306 |
play | -0.291 |
Token | was |
Token ID | 373 |
Feature activation | +5.23e-02 |
Pos logit contributions | |
. | +0.026 |
." | +0.019 |
He | +0.018 |
play | +0.017 |
They | +0.016 |
Neg logit contributions | |
eking | -0.057 |
ths | -0.057 |
agg | -0.053 |
prisingly | -0.051 |
Kevin | -0.051 |
Token | a |
Token ID | 257 |
Feature activation | -3.45e-02 |
Pos logit contributions | |
ths | +0.118 |
agg | +0.116 |
eking | +0.112 |
aman | +0.111 |
Jess | +0.107 |
Neg logit contributions | |
. | -0.059 |
." | -0.055 |
ashes | -0.051 |
He | -0.048 |
rider | -0.047 |
Token | bunny |
Token ID | 44915 |
Feature activation | -7.96e-03 |
Pos logit contributions | |
. | +0.053 |
." | +0.051 |
ashes | +0.050 |
He | +0.048 |
towards | +0.047 |
Neg logit contributions | |
ths | -0.061 |
agg | -0.060 |
aman | -0.056 |
eking | -0.056 |
Kevin | -0.054 |
Token | . |
Token ID | 13 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
ashes | +0.015 |
five | +0.015 |
pedals | +0.014 |
powers | +0.014 |
pedal | +0.014 |
Neg logit contributions | |
agg | -0.010 |
eking | -0.010 |
ths | -0.010 |
sudden | -0.009 |
Jess | -0.009 |
Token | The |
Token ID | 383 |
Feature activation | -6.59e-03 |
Pos logit contributions | |
ashes | +0.164 |
. | +0.155 |
." | +0.147 |
bars | +0.138 |
rider | +0.138 |
Neg logit contributions | |
agg | -0.374 |
ths | -0.354 |
Jess | -0.349 |
aman | -0.347 |
Kevin | -0.339 |
Token | bunny |
Token ID | 44915 |
Feature activation | -7.41e-03 |
Pos logit contributions | |
. | +0.007 |
ashes | +0.006 |
." | +0.006 |
five | +0.006 |
Let | +0.005 |
Neg logit contributions | |
ths | -0.015 |
agg | -0.015 |
eking | -0.014 |
aman | -0.014 |
Jess | -0.014 |
Token | had |
Token ID | 550 |
Feature activation | -3.41e-02 |
Pos logit contributions | |
ashes | +0.010 |
. | +0.010 |
." | +0.010 |
five | +0.010 |
rider | +0.009 |
Neg logit contributions | |
agg | -0.013 |
ths | -0.013 |
eking | -0.013 |
Jess | -0.012 |
aman | -0.012 |
Token | two |
Token ID | 734 |
Feature activation | -3.44e-02 |
Pos logit contributions | |
. | +0.042 |
." | +0.039 |
ashes | +0.038 |
rider | +0.035 |
Let | +0.034 |
Neg logit contributions | |
agg | -0.071 |
ths | -0.071 |
eking | -0.070 |
aman | -0.065 |
Jess | -0.065 |
Token | big |
Token ID | 1263 |
Feature activation | -3.77e-02 |
Pos logit contributions | |
. | +0.051 |
." | +0.050 |
ashes | +0.048 |
Let | +0.046 |
five | +0.046 |
Neg logit contributions | |
ths | -0.063 |
agg | -0.062 |
eking | -0.061 |
aman | -0.056 |
sudden | -0.055 |
Token | were |
Token ID | 547 |
Feature activation | +1.00e-01 |
Pos logit contributions | |
. | +0.064 |
ashes | +0.061 |
." | +0.061 |
He | +0.058 |
Let | +0.056 |
Neg logit contributions | |
ths | -0.073 |
agg | -0.071 |
eking | -0.069 |
aman | -0.067 |
Kevin | -0.064 |
Token | applauded |
Token ID | 38911 |
Feature activation | +2.84e-02 |
Pos logit contributions | |
ths | +0.181 |
agg | +0.177 |
eking | +0.172 |
aman | +0.171 |
Kevin | +0.158 |
Neg logit contributions | |
. | -0.174 |
." | -0.158 |
He | -0.139 |
like | -0.136 |
towards | -0.135 |
Token | for |
Token ID | 329 |
Feature activation | -3.98e-03 |
Pos logit contributions | |
ths | +0.069 |
agg | +0.067 |
eking | +0.066 |
aman | +0.064 |
Kevin | +0.062 |
Neg logit contributions | |
. | -0.025 |
ashes | -0.025 |
." | -0.024 |
He | -0.021 |
rider | -0.021 |
Token | their |
Token ID | 511 |
Feature activation | +1.95e-02 |
Pos logit contributions | |
. | +0.006 |
." | +0.005 |
He | +0.005 |
They | +0.004 |
play | +0.004 |
Neg logit contributions | |
ths | -0.008 |
agg | -0.008 |
eking | -0.008 |
aman | -0.008 |
prisingly | -0.007 |
Token | brave |
Token ID | 14802 |
Feature activation | -1.57e-02 |
Pos logit contributions | |
ths | +0.034 |
agg | +0.033 |
eking | +0.033 |
aman | +0.032 |
prisingly | +0.030 |
Neg logit contributions | |
. | -0.034 |
." | -0.031 |
ashes | -0.029 |
He | -0.029 |
rider | -0.028 |
Token | feats |
Token ID | 35664 |
Feature activation | -1.68e-01 |
Pos logit contributions | |
ashes | +0.023 |
. | +0.022 |
." | +0.022 |
Let | +0.021 |
He | +0.021 |
Neg logit contributions | |
ths | -0.027 |
agg | -0.027 |
aman | -0.025 |
eking | -0.025 |
Kevin | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
ashes | +0.079 |
He | +0.046 |
rider | +0.046 |
. | +0.045 |
." | +0.044 |
Neg logit contributions | |
ths | -0.473 |
agg | -0.462 |
eking | -0.460 |
aman | -0.452 |
Kevin | -0.440 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
. | +0.212 |
." | +0.198 |
ashes | +0.185 |
He | +0.184 |
They | +0.177 |
Neg logit contributions | |
ths | -0.244 |
agg | -0.234 |
eking | -0.232 |
aman | -0.223 |
prisingly | -0.212 |
Token | Once |
Token ID | 7454 |
Feature activation | +1.74e-01 |
Pos logit contributions | |
agg | +0.148 |
ths | +0.141 |
Jess | +0.137 |
Kevin | +0.135 |
aman | +0.134 |
Neg logit contributions | |
ashes | -0.187 |
rider | -0.177 |
powers | -0.173 |
five | -0.171 |
." | -0.171 |
Token | upon |
Token ID | 2402 |
Feature activation | +8.44e-02 |
Pos logit contributions | |
ths | +0.296 |
eking | +0.279 |
agg | +0.251 |
aman | +0.244 |
prisingly | +0.242 |
Neg logit contributions | |
. | -0.386 |
." | -0.332 |
He | -0.321 |
They | -0.300 |
play | -0.288 |
Token | a |
Token ID | 257 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
ths | +0.184 |
agg | +0.184 |
eking | +0.179 |
aman | +0.173 |
Jess | +0.170 |
Neg logit contributions | |
. | -0.096 |
." | -0.090 |
ashes | -0.088 |
He | -0.079 |
rider | -0.079 |
Token | " |
Token ID | 366 |
Feature activation | +6.19e-02 |
Pos logit contributions | |
ths | +0.187 |
agg | +0.181 |
eking | +0.178 |
aman | +0.174 |
Kevin | +0.167 |
Neg logit contributions | |
. | -0.103 |
." | -0.094 |
ashes | -0.086 |
He | -0.084 |
rider | -0.079 |
Token | Be |
Token ID | 3856 |
Feature activation | +1.60e-01 |
Pos logit contributions | |
ths | +0.142 |
agg | +0.141 |
eking | +0.138 |
aman | +0.132 |
Kevin | +0.128 |
Neg logit contributions | |
. | -0.073 |
." | -0.066 |
ashes | -0.060 |
He | -0.060 |
Let | -0.057 |
Token | careful |
Token ID | 8161 |
Feature activation | +3.16e-02 |
Pos logit contributions | |
aman | +0.299 |
ths | +0.292 |
agg | +0.286 |
Jess | +0.263 |
Kevin | +0.261 |
Neg logit contributions | |
. | -0.286 |
." | -0.260 |
like | -0.229 |
play | -0.226 |
ashes | -0.217 |
Token | , |
Token ID | 11 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
ths | +0.062 |
eking | +0.059 |
agg | +0.058 |
aman | +0.054 |
ering | +0.054 |
Neg logit contributions | |
. | -0.042 |
ashes | -0.042 |
." | -0.039 |
He | -0.039 |
rider | -0.038 |
Token | Tim |
Token ID | 5045 |
Feature activation | +3.50e-02 |
Pos logit contributions | |
ths | +0.180 |
agg | +0.176 |
eking | +0.173 |
aman | +0.169 |
Kevin | +0.162 |
Neg logit contributions | |
. | -0.105 |
." | -0.096 |
ashes | -0.086 |
He | -0.086 |
They | -0.081 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
eking | +0.087 |
ths | +0.086 |
ering | +0.082 |
tha | +0.080 |
aman | +0.080 |
Neg logit contributions | |
ashes | -0.029 |
powers | -0.026 |
five | -0.025 |
pedals | -0.022 |
bars | -0.019 |
Token | This |
Token ID | 770 |
Feature activation | +2.70e-02 |
Pos logit contributions | |
. | +0.181 |
ashes | +0.158 |
." | +0.146 |
rider | +0.143 |
towards | +0.136 |
Neg logit contributions | |
ths | -0.377 |
agg | -0.363 |
eking | -0.346 |
aman | -0.345 |
prisingly | -0.344 |
Token | box |
Token ID | 3091 |
Feature activation | +3.70e-03 |
Pos logit contributions | |
ths | +0.051 |
agg | +0.049 |
eking | +0.047 |
prisingly | +0.046 |
aman | +0.046 |
Neg logit contributions | |
ashes | -0.037 |
. | -0.036 |
." | -0.034 |
rider | -0.033 |
He | -0.033 |
Token | may |
Token ID | 743 |
Feature activation | +3.92e-02 |
Pos logit contributions | |
ths | +0.008 |
eking | +0.008 |
agg | +0.008 |
aman | +0.007 |
prisingly | +0.007 |
Neg logit contributions | |
ashes | -0.004 |
. | -0.004 |
five | -0.004 |
rider | -0.004 |
." | -0.004 |
Token | have |
Token ID | 423 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
ths | +0.067 |
eking | +0.066 |
agg | +0.065 |
aman | +0.061 |
prisingly | +0.058 |
Neg logit contributions | |
. | -0.059 |
ashes | -0.058 |
five | -0.056 |
." | -0.055 |
rider | -0.055 |
Token | a |
Token ID | 257 |
Feature activation | -2.89e-02 |
Pos logit contributions | |
. | +0.028 |
ashes | +0.027 |
." | +0.027 |
rider | +0.026 |
He | +0.025 |
Neg logit contributions | |
ths | -0.038 |
eking | -0.037 |
agg | -0.037 |
aman | -0.035 |
prisingly | -0.034 |
Token | No |
Token ID | 2949 |
Feature activation | +8.74e-02 |
Pos logit contributions | |
agg | +0.123 |
ths | +0.123 |
eking | +0.121 |
aman | +0.113 |
Kevin | +0.110 |
Neg logit contributions | |
. | -0.087 |
." | -0.080 |
He | -0.076 |
ashes | -0.074 |
rider | -0.072 |
Token | , |
Token ID | 11 |
Feature activation | +8.18e-02 |
Pos logit contributions | |
ths | +0.163 |
aman | +0.153 |
eking | +0.150 |
agg | +0.149 |
prisingly | +0.137 |
Neg logit contributions | |
ashes | -0.129 |
. | -0.122 |
." | -0.116 |
He | -0.116 |
towards | -0.114 |
Token | it |
Token ID | 340 |
Feature activation | +1.89e-02 |
Pos logit contributions | |
ths | +0.179 |
agg | +0.174 |
eking | +0.171 |
aman | +0.167 |
Kevin | +0.160 |
Neg logit contributions | |
. | -0.104 |
." | -0.095 |
ashes | -0.087 |
He | -0.085 |
rider | -0.080 |
Token | is |
Token ID | 318 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
ths | +0.040 |
eking | +0.038 |
aman | +0.035 |
agg | +0.035 |
ering | +0.034 |
Neg logit contributions | |
ashes | -0.023 |
. | -0.022 |
He | -0.021 |
five | -0.021 |
towards | -0.021 |
Token | fragile |
Token ID | 21049 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
ths | +0.085 |
eking | +0.082 |
aman | +0.074 |
agg | +0.074 |
prisingly | +0.073 |
Neg logit contributions | |
ashes | -0.097 |
. | -0.093 |
five | -0.093 |
He | -0.090 |
Let | -0.087 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
five | +0.014 |
ashes | +0.013 |
powers | +0.011 |
pet | +0.011 |
stroke | +0.010 |
Neg logit contributions | |
ths | -0.019 |
ering | -0.019 |
eking | -0.018 |
done | -0.018 |
unfortunately | -0.018 |
Token | You |
Token ID | 921 |
Feature activation | +2.27e-02 |
Pos logit contributions | |
. | +0.214 |
ashes | +0.198 |
towards | +0.176 |
rider | +0.171 |
five | +0.170 |
Neg logit contributions | |
ths | -0.319 |
unfortunately | -0.300 |
omm | -0.300 |
mmm | -0.298 |
agg | -0.297 |
Token | might |
Token ID | 1244 |
Feature activation | +1.50e-02 |
Pos logit contributions | |
ths | +0.046 |
eking | +0.044 |
agg | +0.043 |
prisingly | +0.042 |
aman | +0.041 |
Neg logit contributions | |
. | -0.027 |
ashes | -0.026 |
He | -0.024 |
." | -0.024 |
five | -0.024 |
Token | break |
Token ID | 2270 |
Feature activation | -4.04e-02 |
Pos logit contributions | |
eking | +0.026 |
ths | +0.026 |
agg | +0.024 |
aman | +0.024 |
prisingly | +0.023 |
Neg logit contributions | |
. | -0.021 |
ashes | -0.020 |
He | -0.020 |
Let | -0.020 |
five | -0.019 |
Token | it |
Token ID | 340 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
ashes | +0.062 |
He | +0.057 |
. | +0.057 |
five | +0.056 |
rider | +0.056 |
Neg logit contributions | |
ths | -0.072 |
eking | -0.068 |
agg | -0.062 |
aman | -0.062 |
ering | -0.061 |
Token | ." |
Token ID | 526 |
Feature activation | -8.94e-02 |
Pos logit contributions | |
ths | +0.053 |
eking | +0.053 |
aman | +0.050 |
prisingly | +0.050 |
agg | +0.050 |
Neg logit contributions | |
ashes | -0.007 |
five | -0.006 |
rider | -0.005 |
He | -0.005 |
. | -0.005 |
Token | is |
Token ID | 318 |
Feature activation | +5.87e-02 |
Pos logit contributions | |
ths | +0.160 |
eking | +0.152 |
agg | +0.150 |
aman | +0.142 |
ering | +0.141 |
Neg logit contributions | |
ashes | -0.103 |
. | -0.095 |
five | -0.092 |
towards | -0.089 |
rider | -0.089 |
Token | a |
Token ID | 257 |
Feature activation | -3.01e-02 |
Pos logit contributions | |
ths | +0.099 |
agg | +0.095 |
eking | +0.092 |
aman | +0.090 |
Kevin | +0.085 |
Neg logit contributions | |
. | -0.107 |
." | -0.100 |
ashes | -0.093 |
He | -0.093 |
They | -0.089 |
Token | vest |
Token ID | 19750 |
Feature activation | -7.09e-02 |
Pos logit contributions | |
. | +0.034 |
ashes | +0.032 |
." | +0.031 |
They | +0.027 |
He | +0.027 |
Neg logit contributions | |
ths | -0.067 |
agg | -0.065 |
eking | -0.064 |
aman | -0.063 |
Kevin | -0.061 |
Token | , |
Token ID | 11 |
Feature activation | +8.33e-02 |
Pos logit contributions | |
ashes | +0.096 |
pedals | +0.083 |
spots | +0.080 |
powers | +0.080 |
irms | +0.078 |
Neg logit contributions | |
ths | -0.124 |
lf | -0.117 |
unfortunately | -0.114 |
eking | -0.110 |
done | -0.109 |
Token | Tim |
Token ID | 5045 |
Feature activation | +3.95e-02 |
Pos logit contributions | |
ths | +0.196 |
agg | +0.193 |
eking | +0.189 |
aman | +0.181 |
Kevin | +0.180 |
Neg logit contributions | |
. | -0.100 |
." | -0.091 |
He | -0.082 |
They | -0.076 |
Let | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ma | +0.092 |
aman | +0.091 |
eking | +0.090 |
ths | +0.090 |
tha | +0.087 |
Neg logit contributions | |
ashes | -0.036 |
powers | -0.031 |
five | -0.029 |
towards | -0.024 |
pedals | -0.024 |
Token | You |
Token ID | 921 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
. | +0.185 |
ashes | +0.184 |
rider | +0.152 |
five | +0.149 |
." | +0.146 |
Neg logit contributions | |
ths | -0.346 |
agg | -0.332 |
unfortunately | -0.330 |
prisingly | -0.325 |
mmm | -0.325 |
Token | wear |
Token ID | 5806 |
Feature activation | -9.57e-02 |
Pos logit contributions | |
ths | +0.050 |
eking | +0.049 |
agg | +0.049 |
prisingly | +0.047 |
Kevin | +0.046 |
Neg logit contributions | |
ashes | -0.034 |
. | -0.032 |
towards | -0.029 |
five | -0.029 |
." | -0.029 |
Token | it |
Token ID | 340 |
Feature activation | +1.99e-02 |
Pos logit contributions | |
ashes | +0.163 |
. | +0.160 |
rider | +0.150 |
." | +0.150 |
ny | +0.149 |
Neg logit contributions | |
ths | -0.148 |
eking | -0.143 |
agg | -0.136 |
aman | -0.127 |
prisingly | -0.126 |
Token | on |
Token ID | 319 |
Feature activation | -5.81e-02 |
Pos logit contributions | |
ths | +0.044 |
eking | +0.043 |
agg | +0.042 |
aman | +0.040 |
prisingly | +0.040 |
Neg logit contributions | |
. | -0.020 |
ashes | -0.020 |
." | -0.018 |
He | -0.018 |
five | -0.017 |
Token | your |
Token ID | 534 |
Feature activation | -4.62e-03 |
Pos logit contributions | |
ashes | +0.104 |
. | +0.098 |
rider | +0.094 |
He | +0.092 |
Let | +0.092 |
Neg logit contributions | |
ths | -0.086 |
eking | -0.084 |
agg | -0.083 |
aman | -0.082 |
Kevin | -0.077 |
Token | seat |
Token ID | 5852 |
Feature activation | -3.87e-02 |
Pos logit contributions | |
. | +0.036 |
." | +0.036 |
ashes | +0.035 |
Let | +0.033 |
He | +0.033 |
Neg logit contributions | |
ths | -0.063 |
aman | -0.061 |
agg | -0.060 |
eking | -0.058 |
str | -0.054 |
Token | on |
Token ID | 319 |
Feature activation | -5.85e-02 |
Pos logit contributions | |
powers | +0.049 |
ashes | +0.048 |
He | +0.045 |
." | +0.044 |
five | +0.044 |
Neg logit contributions | |
eking | -0.079 |
ths | -0.078 |
ering | -0.071 |
aman | -0.070 |
unfortunately | -0.068 |
Token | the |
Token ID | 262 |
Feature activation | -3.79e-02 |
Pos logit contributions | |
. | +0.081 |
ashes | +0.081 |
." | +0.078 |
He | +0.073 |
Let | +0.073 |
Neg logit contributions | |
ths | -0.114 |
agg | -0.108 |
aman | -0.107 |
eking | -0.104 |
Jess | -0.100 |
Token | see |
Token ID | 766 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ashes | +0.057 |
They | +0.055 |
powers | +0.053 |
Let | +0.053 |
. | +0.053 |
Neg logit contributions | |
ths | -0.062 |
eking | -0.062 |
aman | -0.062 |
agg | -0.058 |
str | -0.056 |
Token | - |
Token ID | 12 |
Feature activation | +2.45e-02 |
Pos logit contributions | |
ths | +0.029 |
eking | +0.029 |
agg | +0.028 |
aman | +0.028 |
Jane | +0.028 |
Neg logit contributions | |
ashes | -0.012 |
powers | -0.010 |
rider | -0.009 |
ny | -0.009 |
eer | -0.009 |
Token | saw |
Token ID | 43439 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ths | +0.045 |
agg | +0.043 |
eking | +0.043 |
aman | +0.043 |
Jane | +0.042 |
Neg logit contributions | |
. | -0.035 |
ashes | -0.033 |
." | -0.033 |
He | -0.031 |
They | -0.031 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
powers | +0.068 |
ashes | +0.065 |
He | +0.049 |
rider | +0.047 |
pedal | +0.043 |
Neg logit contributions | |
ths | -0.453 |
eking | -0.448 |
aman | -0.430 |
ering | -0.427 |
unfortunately | -0.423 |
Token | She |
Token ID | 1375 |
Feature activation | -3.09e-02 |
Pos logit contributions | |
ashes | +0.194 |
rider | +0.189 |
. | +0.177 |
five | +0.177 |
." | +0.175 |
Neg logit contributions | |
ths | -0.262 |
agg | -0.262 |
aman | -0.259 |
omm | -0.244 |
mber | -0.234 |
Token | ran |
Token ID | 4966 |
Feature activation | -1.25e-01 |
Pos logit contributions | |
. | +0.036 |
." | +0.030 |
He | +0.029 |
bar | +0.026 |
They | +0.025 |
Neg logit contributions | |
ths | -0.072 |
agg | -0.071 |
eking | -0.071 |
aman | -0.070 |
prisingly | -0.068 |
Token | to |
Token ID | 284 |
Feature activation | -9.06e-04 |
Pos logit contributions | |
. | +0.111 |
." | +0.106 |
ashes | +0.095 |
eer | +0.087 |
Let | +0.086 |
Neg logit contributions | |
ths | -0.302 |
agg | -0.284 |
eking | -0.283 |
aman | -0.281 |
prisingly | -0.270 |
Token | it |
Token ID | 340 |
Feature activation | +2.08e-02 |
Pos logit contributions | |
. | +0.001 |
." | +0.001 |
ashes | +0.001 |
He | +0.001 |
rider | +0.001 |
Neg logit contributions | |
ths | -0.002 |
agg | -0.002 |
eking | -0.002 |
aman | -0.002 |
Kevin | -0.002 |
Token | " |
Token ID | 366 |
Feature activation | +5.95e-02 |
Pos logit contributions | |
. | +0.185 |
rider | +0.175 |
." | +0.172 |
ashes | +0.167 |
eer | +0.166 |
Neg logit contributions | |
agg | -0.307 |
aman | -0.301 |
ths | -0.298 |
Kevin | -0.276 |
Jess | -0.274 |
Token | Be |
Token ID | 3856 |
Feature activation | +1.61e-01 |
Pos logit contributions | |
ths | +0.105 |
agg | +0.101 |
eking | +0.099 |
aman | +0.095 |
Kevin | +0.090 |
Neg logit contributions | |
. | -0.101 |
." | -0.095 |
ashes | -0.090 |
He | -0.088 |
rider | -0.085 |
Token | careful |
Token ID | 8161 |
Feature activation | +3.19e-02 |
Pos logit contributions | |
aman | +0.304 |
ths | +0.296 |
agg | +0.282 |
Jess | +0.266 |
Kevin | +0.266 |
Neg logit contributions | |
. | -0.282 |
." | -0.253 |
like | -0.220 |
fur | -0.211 |
ashes | -0.206 |
Token | , |
Token ID | 11 |
Feature activation | +8.20e-02 |
Pos logit contributions | |
ths | +0.070 |
agg | +0.068 |
eking | +0.067 |
aman | +0.066 |
Kevin | +0.063 |
Neg logit contributions | |
. | -0.041 |
." | -0.037 |
ashes | -0.033 |
He | -0.033 |
rider | -0.031 |
Token | Sara |
Token ID | 24799 |
Feature activation | +3.79e-02 |
Pos logit contributions | |
ths | +0.191 |
eking | +0.187 |
agg | +0.185 |
prisingly | +0.176 |
Kevin | +0.176 |
Neg logit contributions | |
. | -0.099 |
." | -0.093 |
He | -0.076 |
They | -0.075 |
bar | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ths | +0.082 |
agg | +0.079 |
eking | +0.078 |
aman | +0.077 |
Jess | +0.076 |
Neg logit contributions | |
ashes | -0.043 |
five | -0.036 |
powers | -0.036 |
pedals | -0.034 |
pedal | -0.033 |
Token | The |
Token ID | 383 |
Feature activation | -1.53e-03 |
Pos logit contributions | |
. | +0.191 |
." | +0.163 |
ashes | +0.162 |
rider | +0.158 |
towards | +0.152 |
Neg logit contributions | |
ths | -0.371 |
agg | -0.363 |
aman | -0.347 |
eking | -0.346 |
Jess | -0.333 |
Token | onion |
Token ID | 21670 |
Feature activation | +5.20e-03 |
Pos logit contributions | |
. | +0.002 |
." | +0.002 |
He | +0.002 |
ashes | +0.002 |
They | +0.002 |
Neg logit contributions | |
ths | -0.003 |
agg | -0.003 |
eking | -0.003 |
aman | -0.003 |
prisingly | -0.003 |
Token | can |
Token ID | 460 |
Feature activation | +3.89e-02 |
Pos logit contributions | |
ths | +0.012 |
agg | +0.011 |
eking | +0.011 |
aman | +0.011 |
prisingly | +0.010 |
Neg logit contributions | |
. | -0.006 |
." | -0.005 |
ashes | -0.005 |
He | -0.005 |
rider | -0.005 |
Token | make |
Token ID | 787 |
Feature activation | -4.40e-02 |
Pos logit contributions | |
ths | +0.083 |
agg | +0.081 |
eking | +0.079 |
aman | +0.077 |
Kevin | +0.074 |
Neg logit contributions | |
. | -0.051 |
." | -0.047 |
ashes | -0.043 |
He | -0.042 |
rider | -0.040 |
Token | your |
Token ID | 534 |
Feature activation | -3.87e-03 |
Pos logit contributions | |
. | +0.067 |
." | +0.062 |
He | +0.054 |
ashes | +0.053 |
They | +0.052 |
Neg logit contributions | |
ths | -0.089 |
agg | -0.086 |
eking | -0.084 |
aman | -0.084 |
Kevin | -0.080 |
Token | , |
Token ID | 11 |
Feature activation | +8.49e-02 |
Pos logit contributions | |
powers | +0.146 |
ashes | +0.139 |
eer | +0.129 |
Let | +0.125 |
five | +0.123 |
Neg logit contributions | |
ths | -0.208 |
aman | -0.205 |
eking | -0.197 |
agg | -0.193 |
unfortunately | -0.188 |
Token | slide |
Token ID | 10649 |
Feature activation | -6.03e-02 |
Pos logit contributions | |
ths | +0.185 |
agg | +0.180 |
eking | +0.174 |
aman | +0.171 |
prisingly | +0.164 |
Neg logit contributions | |
. | -0.114 |
." | -0.104 |
ashes | -0.095 |
He | -0.094 |
They | -0.089 |
Token | and |
Token ID | 290 |
Feature activation | +4.41e-02 |
Pos logit contributions | |
ashes | +0.077 |
powers | +0.075 |
five | +0.071 |
eer | +0.070 |
." | +0.070 |
Neg logit contributions | |
ths | -0.120 |
aman | -0.115 |
eking | -0.111 |
agg | -0.107 |
ering | -0.105 |
Token | see |
Token ID | 766 |
Feature activation | +1.04e-02 |
Pos logit contributions | |
ths | +0.097 |
eking | +0.095 |
agg | +0.094 |
aman | +0.090 |
Jess | +0.089 |
Neg logit contributions | |
. | -0.045 |
." | -0.043 |
ashes | -0.041 |
He | -0.038 |
powers | -0.037 |
Token | - |
Token ID | 12 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
ths | +0.014 |
aman | +0.013 |
eking | +0.013 |
agg | +0.013 |
exactly | +0.012 |
Neg logit contributions | |
ashes | -0.021 |
ny | -0.020 |
powers | -0.019 |
five | -0.019 |
Let | -0.018 |
Token | saw |
Token ID | 43439 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ths | +0.051 |
agg | +0.050 |
eking | +0.049 |
aman | +0.048 |
Kevin | +0.046 |
Neg logit contributions | |
. | -0.030 |
." | -0.028 |
ashes | -0.026 |
He | -0.025 |
rider | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -1.46e-01 |
Pos logit contributions | |
powers | +0.113 |
ashes | +0.108 |
eer | +0.101 |
ny | +0.090 |
ray | +0.090 |
Neg logit contributions | |
ths | -0.401 |
eking | -0.383 |
aman | -0.368 |
ering | -0.367 |
exactly | -0.358 |
Token |
|
Token ID | 198 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
. | +0.160 |
." | +0.143 |
ashes | +0.130 |
He | +0.128 |
They | +0.119 |
Neg logit contributions | |
ths | -0.349 |
agg | -0.339 |
eking | -0.336 |
aman | -0.326 |
Kevin | -0.314 |
Token |
|
Token ID | 198 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
ths | +0.400 |
eking | +0.392 |
agg | +0.365 |
aman | +0.358 |
ering | +0.350 |
Neg logit contributions | |
. | -0.139 |
." | -0.112 |
He | -0.105 |
They | -0.100 |
play | -0.080 |
Token | But |
Token ID | 1537 |
Feature activation | +9.36e-02 |
Pos logit contributions | |
ths | +0.246 |
eking | +0.242 |
agg | +0.239 |
aman | +0.222 |
Kevin | +0.209 |
Neg logit contributions | |
. | -0.249 |
." | -0.227 |
He | -0.215 |
They | -0.211 |
play | -0.209 |
Token | when |
Token ID | 618 |
Feature activation | -2.77e-02 |
Pos logit contributions | |
agg | +0.223 |
eking | +0.215 |
Kevin | +0.211 |
ths | +0.211 |
Jess | +0.205 |
Neg logit contributions | |
. | -0.120 |
." | -0.093 |
play | -0.088 |
like | -0.086 |
He | -0.085 |
Token | down |
Token ID | 866 |
Feature activation | -1.21e-01 |
Pos logit contributions | |
ths | +0.119 |
eking | +0.110 |
aman | +0.107 |
agg | +0.106 |
tha | +0.104 |
Neg logit contributions | |
. | -0.037 |
." | -0.036 |
ashes | -0.034 |
powers | -0.032 |
eer | -0.027 |
Token | on |
Token ID | 319 |
Feature activation | -5.77e-02 |
Pos logit contributions | |
eer | +0.119 |
ashes | +0.117 |
powers | +0.112 |
rider | +0.110 |
pads | +0.110 |
Neg logit contributions | |
ths | -0.266 |
agg | -0.254 |
aman | -0.247 |
eking | -0.245 |
Jess | -0.237 |
Token | the |
Token ID | 262 |
Feature activation | -3.58e-02 |
Pos logit contributions | |
. | +0.080 |
ashes | +0.078 |
." | +0.078 |
He | +0.074 |
rider | +0.072 |
Neg logit contributions | |
ths | -0.114 |
agg | -0.107 |
aman | -0.105 |
eking | -0.104 |
Jess | -0.099 |
Token | see |
Token ID | 766 |
Feature activation | +1.21e-02 |
Pos logit contributions | |
ashes | +0.041 |
. | +0.039 |
Let | +0.039 |
." | +0.038 |
They | +0.038 |
Neg logit contributions | |
ths | -0.075 |
eking | -0.075 |
agg | -0.071 |
aman | -0.070 |
ering | -0.066 |
Token | - |
Token ID | 12 |
Feature activation | +2.43e-02 |
Pos logit contributions | |
ths | +0.024 |
agg | +0.023 |
aman | +0.022 |
eking | +0.022 |
Jess | +0.021 |
Neg logit contributions | |
ashes | -0.016 |
ny | -0.014 |
powers | -0.014 |
rider | -0.014 |
eer | -0.014 |
Token | saw |
Token ID | 43439 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ths | +0.048 |
agg | +0.047 |
eking | +0.046 |
aman | +0.045 |
Jess | +0.044 |
Neg logit contributions | |
. | -0.032 |
." | -0.030 |
ashes | -0.029 |
He | -0.027 |
They | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -1.40e-01 |
Pos logit contributions | |
ashes | +0.066 |
powers | +0.061 |
He | +0.052 |
." | +0.051 |
Let | +0.048 |
Neg logit contributions | |
ths | -0.465 |
eking | -0.444 |
aman | -0.434 |
agg | -0.430 |
ering | -0.427 |
Token | |
Token ID | 220 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
. | +0.272 |
." | +0.259 |
ashes | +0.250 |
rider | +0.249 |
He | +0.235 |
Neg logit contributions | |
ths | -0.197 |
agg | -0.192 |
aman | -0.181 |
eking | -0.179 |
Jess | -0.167 |
Token |
|
Token ID | 198 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ths | +0.417 |
agg | +0.401 |
eking | +0.398 |
aman | +0.386 |
Kevin | +0.370 |
Neg logit contributions | |
. | -0.228 |
." | -0.207 |
He | -0.186 |
ashes | -0.184 |
They | -0.178 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ths | +0.422 |
eking | +0.410 |
agg | +0.389 |
aman | +0.386 |
ering | +0.370 |
Neg logit contributions | |
. | -0.143 |
." | -0.126 |
They | -0.109 |
He | -0.104 |
Let | -0.081 |
Token | After |
Token ID | 3260 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
ths | +0.231 |
agg | +0.229 |
eking | +0.226 |
aman | +0.214 |
Kevin | +0.200 |
Neg logit contributions | |
. | -0.279 |
." | -0.264 |
They | -0.246 |
ashes | -0.242 |
He | -0.242 |
Token | Max |
Token ID | 11518 |
Feature activation | +2.37e-02 |
Pos logit contributions | |
ths | +0.113 |
agg | +0.108 |
aman | +0.104 |
eking | +0.104 |
Jess | +0.100 |
Neg logit contributions | |
. | -0.102 |
." | -0.095 |
ashes | -0.090 |
He | -0.084 |
towards | -0.083 |
Token | , |
Token ID | 11 |
Feature activation | +8.25e-02 |
Pos logit contributions | |
ths | +0.036 |
agg | +0.032 |
eking | +0.031 |
aman | +0.031 |
tha | +0.030 |
Neg logit contributions | |
ashes | -0.042 |
five | -0.038 |
powers | -0.037 |
bars | -0.037 |
pedals | -0.036 |
Token | I |
Token ID | 314 |
Feature activation | -7.45e-02 |
Pos logit contributions | |
ths | +0.176 |
agg | +0.171 |
eking | +0.168 |
aman | +0.163 |
Kevin | +0.157 |
Neg logit contributions | |
. | -0.110 |
." | -0.101 |
ashes | -0.093 |
He | -0.091 |
rider | -0.086 |
Token | am |
Token ID | 716 |
Feature activation | +7.56e-02 |
Pos logit contributions | |
. | +0.078 |
." | +0.072 |
He | +0.069 |
ashes | +0.068 |
towards | +0.065 |
Neg logit contributions | |
agg | -0.164 |
ths | -0.162 |
eking | -0.159 |
prisingly | -0.154 |
aman | -0.153 |
Token | cold |
Token ID | 4692 |
Feature activation | -1.66e-02 |
Pos logit contributions | |
ths | +0.143 |
agg | +0.136 |
aman | +0.132 |
eking | +0.131 |
Kevin | +0.124 |
Neg logit contributions | |
. | -0.129 |
." | -0.121 |
He | -0.108 |
ashes | -0.107 |
play | -0.103 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
five | +0.015 |
powers | +0.014 |
ashes | +0.013 |
pedals | +0.012 |
name | +0.012 |
Neg logit contributions | |
ths | -0.037 |
eking | -0.036 |
agg | -0.035 |
aman | -0.034 |
ering | -0.033 |
Token | Let |
Token ID | 3914 |
Feature activation | -4.68e-02 |
Pos logit contributions | |
. | +0.226 |
ashes | +0.222 |
bars | +0.203 |
towards | +0.186 |
v | +0.184 |
Neg logit contributions | |
unfortunately | -0.283 |
agg | -0.278 |
ths | -0.270 |
mmm | -0.265 |
hh | -0.263 |
Token | 's |
Token ID | 338 |
Feature activation | +1.88e-02 |
Pos logit contributions | |
. | +0.048 |
." | +0.043 |
He | +0.036 |
ashes | +0.036 |
They | +0.033 |
Neg logit contributions | |
ths | -0.116 |
agg | -0.114 |
eking | -0.112 |
aman | -0.109 |
Kevin | -0.106 |
Token | sit |
Token ID | 1650 |
Feature activation | -9.04e-03 |
Pos logit contributions | |
ths | +0.041 |
agg | +0.041 |
eking | +0.040 |
aman | +0.039 |
unfortunately | +0.037 |
Neg logit contributions | |
ashes | -0.018 |
powers | -0.017 |
five | -0.017 |
. | -0.017 |
towards | -0.016 |
Token | closer |
Token ID | 5699 |
Feature activation | +9.23e-03 |
Pos logit contributions | |
. | +0.009 |
." | +0.008 |
ashes | +0.008 |
He | +0.007 |
five | +0.007 |
Neg logit contributions | |
ths | -0.022 |
eking | -0.021 |
agg | -0.021 |
aman | -0.020 |
prisingly | -0.020 |
Token | to |
Token ID | 284 |
Feature activation | -1.58e-03 |
Pos logit contributions | |
ths | +0.022 |
agg | +0.021 |
eking | +0.021 |
aman | +0.020 |
prisingly | +0.020 |
Neg logit contributions | |
ashes | -0.007 |
. | -0.007 |
rider | -0.007 |
He | -0.007 |
pedal | -0.006 |
Token | " |
Token ID | 366 |
Feature activation | +5.86e-02 |
Pos logit contributions | |
ths | +0.183 |
agg | +0.178 |
eking | +0.175 |
aman | +0.171 |
Kevin | +0.164 |
Neg logit contributions | |
. | -0.103 |
." | -0.094 |
ashes | -0.086 |
He | -0.084 |
rider | -0.079 |
Token | My |
Token ID | 3666 |
Feature activation | +1.92e-02 |
Pos logit contributions | |
agg | +0.113 |
ths | +0.112 |
eking | +0.110 |
aman | +0.103 |
prisingly | +0.099 |
Neg logit contributions | |
. | -0.090 |
." | -0.084 |
ashes | -0.080 |
He | -0.078 |
They | -0.076 |
Token | block |
Token ID | 2512 |
Feature activation | -1.37e-02 |
Pos logit contributions | |
ths | +0.039 |
agg | +0.037 |
eking | +0.036 |
prisingly | +0.034 |
iter | +0.033 |
Neg logit contributions | |
ashes | -0.023 |
. | -0.022 |
ny | -0.021 |
rider | -0.020 |
towards | -0.020 |
Token | is |
Token ID | 318 |
Feature activation | +5.65e-02 |
Pos logit contributions | |
ashes | +0.017 |
. | +0.015 |
five | +0.015 |
towards | +0.015 |
." | +0.014 |
Neg logit contributions | |
ths | -0.028 |
eking | -0.025 |
agg | -0.024 |
prisingly | -0.024 |
ering | -0.024 |
Token | better |
Token ID | 1365 |
Feature activation | +3.86e-02 |
Pos logit contributions | |
ths | +0.106 |
agg | +0.101 |
eking | +0.100 |
aman | +0.096 |
prisingly | +0.092 |
Neg logit contributions | |
. | -0.088 |
." | -0.082 |
ashes | -0.079 |
He | -0.077 |
rider | -0.074 |
Token | . |
Token ID | 13 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ths | +0.089 |
eking | +0.080 |
ering | +0.078 |
prisingly | +0.077 |
agg | +0.077 |
Neg logit contributions | |
pedals | -0.036 |
He | -0.035 |
ashes | -0.034 |
bars | -0.033 |
five | -0.033 |
Token | It |
Token ID | 632 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
. | +0.221 |
ashes | +0.210 |
bars | +0.193 |
rider | +0.191 |
towards | +0.188 |
Neg logit contributions | |
ths | -0.302 |
mmm | -0.295 |
omm | -0.293 |
agg | -0.286 |
unfortunately | -0.283 |
Token | is |
Token ID | 318 |
Feature activation | +5.81e-02 |
Pos logit contributions | |
ths | +0.073 |
eking | +0.068 |
agg | +0.065 |
prisingly | +0.064 |
aman | +0.063 |
Neg logit contributions | |
. | -0.039 |
ashes | -0.038 |
five | -0.035 |
." | -0.034 |
towards | -0.034 |
Token | red |
Token ID | 2266 |
Feature activation | -1.10e-02 |
Pos logit contributions | |
ths | +0.106 |
agg | +0.102 |
eking | +0.101 |
aman | +0.097 |
prisingly | +0.092 |
Neg logit contributions | |
. | -0.093 |
." | -0.087 |
ashes | -0.084 |
He | -0.081 |
rider | -0.078 |
Token | like |
Token ID | 588 |
Feature activation | -4.64e-03 |
Pos logit contributions | |
five | +0.016 |
powers | +0.015 |
ashes | +0.015 |
pet | +0.014 |
rider | +0.013 |
Neg logit contributions | |
ths | -0.016 |
eking | -0.015 |
ering | -0.015 |
aman | -0.015 |
omm | -0.015 |
Token | fire |
Token ID | 2046 |
Feature activation | +1.31e-02 |
Pos logit contributions | |
. | +0.006 |
." | +0.006 |
ashes | +0.005 |
He | +0.005 |
rider | +0.005 |
Neg logit contributions | |
ths | -0.010 |
agg | -0.009 |
eking | -0.009 |
aman | -0.009 |
Kevin | -0.009 |
Token | t |
Token ID | 83 |
Feature activation | +1.11e-01 |
Pos logit contributions | |
ths | +0.005 |
agg | +0.004 |
eking | +0.004 |
aman | +0.004 |
Jess | +0.004 |
Neg logit contributions | |
. | -0.005 |
." | -0.005 |
He | -0.004 |
rider | -0.004 |
ashes | -0.004 |
Token | about |
Token ID | 546 |
Feature activation | +4.61e-02 |
Pos logit contributions | |
ths | +0.248 |
agg | +0.241 |
eking | +0.237 |
aman | +0.231 |
Kevin | +0.222 |
Neg logit contributions | |
. | -0.135 |
." | -0.123 |
ashes | -0.112 |
He | -0.110 |
rider | -0.103 |
Token | the |
Token ID | 262 |
Feature activation | -3.47e-02 |
Pos logit contributions | |
ths | +0.083 |
agg | +0.082 |
eking | +0.078 |
aman | +0.076 |
prisingly | +0.074 |
Neg logit contributions | |
. | -0.078 |
." | -0.072 |
He | -0.065 |
ashes | -0.064 |
They | -0.062 |
Token | see |
Token ID | 766 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
. | +0.046 |
." | +0.043 |
ashes | +0.041 |
He | +0.040 |
They | +0.039 |
Neg logit contributions | |
ths | -0.072 |
eking | -0.068 |
agg | -0.068 |
aman | -0.066 |
Kevin | -0.062 |
Token | - |
Token ID | 12 |
Feature activation | +2.39e-02 |
Pos logit contributions | |
ths | +0.027 |
eking | +0.025 |
agg | +0.025 |
aman | +0.025 |
Jess | +0.023 |
Neg logit contributions | |
ashes | -0.015 |
. | -0.014 |
." | -0.014 |
He | -0.013 |
five | -0.012 |
Token | saw |
Token ID | 43439 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ths | +0.048 |
agg | +0.046 |
eking | +0.045 |
aman | +0.044 |
Kevin | +0.042 |
Neg logit contributions | |
. | -0.035 |
." | -0.032 |
ashes | -0.030 |
He | -0.029 |
rider | -0.028 |
Token | and |
Token ID | 290 |
Feature activation | +4.58e-02 |
Pos logit contributions | |
ashes | +0.172 |
." | +0.166 |
. | +0.166 |
He | +0.153 |
powers | +0.151 |
Neg logit contributions | |
ths | -0.372 |
eking | -0.351 |
agg | -0.345 |
aman | -0.344 |
unfortunately | -0.327 |
Token | how |
Token ID | 703 |
Feature activation | -5.66e-02 |
Pos logit contributions | |
ths | +0.081 |
agg | +0.078 |
eking | +0.076 |
aman | +0.074 |
Kevin | +0.070 |
Neg logit contributions | |
. | -0.078 |
." | -0.073 |
ashes | -0.068 |
He | -0.067 |
They | -0.064 |
Token | much |
Token ID | 881 |
Feature activation | -8.31e-03 |
Pos logit contributions | |
. | +0.100 |
." | +0.094 |
ashes | +0.088 |
He | +0.087 |
They | +0.083 |
Neg logit contributions | |
ths | -0.097 |
agg | -0.093 |
eking | -0.091 |
aman | -0.089 |
Kevin | -0.084 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
. | +0.014 |
." | +0.013 |
ashes | +0.013 |
He | +0.012 |
Let | +0.012 |
Neg logit contributions | |
ths | -0.015 |
agg | -0.014 |
eking | -0.013 |
aman | -0.013 |
Jess | -0.012 |
Token | it |
Token ID | 340 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
ths | +0.031 |
agg | +0.027 |
eking | +0.026 |
aman | +0.025 |
Jess | +0.024 |
Neg logit contributions | |
ashes | -0.044 |
ny | -0.042 |
. | -0.041 |
." | -0.041 |
He | -0.041 |
Token | Leo |
Token ID | 19632 |
Feature activation | +4.63e-02 |
Pos logit contributions | |
ths | +0.175 |
eking | +0.170 |
agg | +0.170 |
aman | +0.165 |
prisingly | +0.153 |
Neg logit contributions | |
. | -0.145 |
." | -0.134 |
He | -0.127 |
ashes | -0.122 |
They | -0.120 |
Token | continued |
Token ID | 3767 |
Feature activation | -9.97e-02 |
Pos logit contributions | |
ths | +0.098 |
agg | +0.095 |
eking | +0.094 |
aman | +0.091 |
Kevin | +0.087 |
Neg logit contributions | |
. | -0.062 |
." | -0.057 |
ashes | -0.052 |
He | -0.052 |
rider | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | +2.14e-03 |
Pos logit contributions | |
. | +0.085 |
." | +0.077 |
ashes | +0.069 |
He | +0.064 |
rider | +0.062 |
Neg logit contributions | |
ths | -0.256 |
agg | -0.247 |
eking | -0.244 |
aman | -0.237 |
Kevin | -0.231 |
Token | perform |
Token ID | 1620 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
ths | +0.005 |
agg | +0.005 |
eking | +0.005 |
aman | +0.005 |
Kevin | +0.004 |
Neg logit contributions | |
. | -0.002 |
." | -0.002 |
ashes | -0.002 |
He | -0.002 |
rider | -0.002 |
Token | amazing |
Token ID | 4998 |
Feature activation | +5.12e-02 |
Pos logit contributions | |
. | +0.057 |
." | +0.053 |
ashes | +0.050 |
He | +0.047 |
Let | +0.046 |
Neg logit contributions | |
ths | -0.118 |
agg | -0.115 |
eking | -0.113 |
aman | -0.109 |
Kevin | -0.106 |
Token | feats |
Token ID | 35664 |
Feature activation | -1.67e-01 |
Pos logit contributions | |
ths | +0.099 |
agg | +0.093 |
eking | +0.088 |
ering | +0.084 |
sudden | +0.081 |
Neg logit contributions | |
." | -0.070 |
ashes | -0.069 |
Let | -0.069 |
. | -0.069 |
He | -0.067 |
Token | for |
Token ID | 329 |
Feature activation | -4.43e-03 |
Pos logit contributions | |
ashes | +0.185 |
." | +0.151 |
Let | +0.148 |
ny | +0.147 |
They | +0.146 |
Neg logit contributions | |
ths | -0.360 |
agg | -0.349 |
eking | -0.344 |
aman | -0.329 |
prisingly | -0.328 |
Token | his |
Token ID | 465 |
Feature activation | -1.16e-02 |
Pos logit contributions | |
. | +0.006 |
." | +0.005 |
He | +0.004 |
play | +0.004 |
bar | +0.004 |
Neg logit contributions | |
aman | -0.010 |
ths | -0.010 |
agg | -0.010 |
eking | -0.010 |
str | -0.009 |
Token | friends |
Token ID | 2460 |
Feature activation | -7.71e-02 |
Pos logit contributions | |
. | +0.018 |
." | +0.017 |
ashes | +0.017 |
He | +0.016 |
They | +0.016 |
Neg logit contributions | |
ths | -0.021 |
agg | -0.021 |
eking | -0.020 |
aman | -0.019 |
Kevin | -0.018 |
Token | in |
Token ID | 287 |
Feature activation | -5.34e-02 |
Pos logit contributions | |
ashes | +0.068 |
powers | +0.053 |
atters | +0.053 |
Let | +0.053 |
five | +0.052 |
Neg logit contributions | |
ths | -0.180 |
eking | -0.170 |
Kevin | -0.169 |
agg | -0.167 |
Jess | -0.164 |
Token | the |
Token ID | 262 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
. | +0.070 |
." | +0.064 |
ashes | +0.059 |
He | +0.058 |
rider | +0.055 |
Neg logit contributions | |
ths | -0.115 |
agg | -0.111 |
eking | -0.109 |
aman | -0.107 |
Kevin | -0.102 |
Token | to |
Token ID | 284 |
Feature activation | -4.38e-03 |
Pos logit contributions | |
. | +0.069 |
ashes | +0.066 |
rider | +0.065 |
He | +0.063 |
." | +0.061 |
Neg logit contributions | |
ths | -0.202 |
eking | -0.192 |
agg | -0.189 |
prisingly | -0.181 |
aman | -0.180 |
Token | make |
Token ID | 787 |
Feature activation | -4.66e-02 |
Pos logit contributions | |
. | +0.005 |
." | +0.005 |
ashes | +0.005 |
He | +0.005 |
rider | +0.004 |
Neg logit contributions | |
ths | -0.010 |
agg | -0.009 |
eking | -0.009 |
aman | -0.009 |
Kevin | -0.009 |
Token | a |
Token ID | 257 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
. | +0.066 |
." | +0.061 |
ashes | +0.052 |
He | +0.051 |
They | +0.050 |
Neg logit contributions | |
ths | -0.098 |
agg | -0.094 |
eking | -0.094 |
aman | -0.091 |
Kevin | -0.087 |
Token | delicious |
Token ID | 12625 |
Feature activation | +4.06e-02 |
Pos logit contributions | |
. | +0.055 |
." | +0.051 |
ashes | +0.048 |
He | +0.046 |
rider | +0.045 |
Neg logit contributions | |
ths | -0.059 |
agg | -0.057 |
eking | -0.056 |
aman | -0.054 |
Kevin | -0.051 |
Token | cake |
Token ID | 12187 |
Feature activation | +2.81e-02 |
Pos logit contributions | |
ths | +0.068 |
agg | +0.065 |
eking | +0.064 |
aman | +0.062 |
Kevin | +0.058 |
Neg logit contributions | |
. | -0.072 |
." | -0.068 |
ashes | -0.064 |
He | -0.063 |
rider | -0.061 |
Token | . |
Token ID | 13 |
Feature activation | -1.66e-01 |
Pos logit contributions | |
ths | +0.067 |
agg | +0.064 |
aman | +0.062 |
eking | +0.062 |
prisingly | +0.060 |
Neg logit contributions | |
ashes | -0.025 |
He | -0.023 |
powers | -0.022 |
. | -0.022 |
pedal | -0.021 |
Token | I |
Token ID | 314 |
Feature activation | -7.37e-02 |
Pos logit contributions | |
. | +0.175 |
ashes | +0.152 |
." | +0.145 |
rider | +0.135 |
powers | +0.130 |
Neg logit contributions | |
ths | -0.386 |
agg | -0.375 |
aman | -0.364 |
prisingly | -0.355 |
Jess | -0.354 |
Token | can |
Token ID | 460 |
Feature activation | +3.73e-02 |
Pos logit contributions | |
. | +0.085 |
." | +0.077 |
ashes | +0.072 |
He | +0.070 |
rider | +0.067 |
Neg logit contributions | |
ths | -0.165 |
agg | -0.162 |
eking | -0.158 |
aman | -0.154 |
prisingly | -0.150 |
Token | 't |
Token ID | 470 |
Feature activation | +9.43e-02 |
Pos logit contributions | |
ths | +0.082 |
agg | +0.080 |
eking | +0.078 |
aman | +0.076 |
prisingly | +0.073 |
Neg logit contributions | |
. | -0.044 |
." | -0.039 |
ashes | -0.039 |
He | -0.037 |
rider | -0.037 |
Token | wait |
Token ID | 4043 |
Feature activation | +2.97e-02 |
Pos logit contributions | |
ths | +0.185 |
agg | +0.180 |
eking | +0.175 |
aman | +0.168 |
prisingly | +0.164 |
Neg logit contributions | |
. | -0.131 |
." | -0.122 |
ashes | -0.121 |
He | -0.119 |
rider | -0.115 |
Token | to |
Token ID | 284 |
Feature activation | -3.47e-03 |
Pos logit contributions | |
ths | +0.073 |
agg | +0.071 |
eking | +0.070 |
aman | +0.068 |
prisingly | +0.066 |
Neg logit contributions | |
. | -0.025 |
ashes | -0.023 |
rider | -0.022 |
He | -0.022 |
." | -0.022 |
Token | their |
Token ID | 511 |
Feature activation | +2.11e-02 |
Pos logit contributions | |
. | +0.027 |
." | +0.025 |
ashes | +0.024 |
He | +0.023 |
rider | +0.022 |
Neg logit contributions | |
ths | -0.045 |
agg | -0.044 |
eking | -0.043 |
aman | -0.042 |
Kevin | -0.040 |
Token | tasty |
Token ID | 25103 |
Feature activation | +3.46e-02 |
Pos logit contributions | |
ths | +0.042 |
agg | +0.040 |
eking | +0.040 |
aman | +0.039 |
Kevin | +0.037 |
Neg logit contributions | |
. | -0.032 |
." | -0.029 |
ashes | -0.027 |
He | -0.026 |
They | -0.025 |
Token | home |
Token ID | 1363 |
Feature activation | -1.22e-02 |
Pos logit contributions | |
ths | +0.062 |
agg | +0.062 |
eking | +0.058 |
aman | +0.054 |
ering | +0.053 |
Neg logit contributions | |
He | -0.048 |
. | -0.047 |
rider | -0.046 |
Let | -0.046 |
." | -0.045 |
Token | . |
Token ID | 13 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
ashes | +0.007 |
He | +0.006 |
five | +0.005 |
Let | +0.005 |
pedal | +0.005 |
Neg logit contributions | |
ths | -0.033 |
eking | -0.032 |
agg | -0.031 |
ering | -0.030 |
prisingly | -0.030 |
Token |
|
Token ID | 198 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
ashes | +0.179 |
rider | +0.166 |
pedal | +0.166 |
. | +0.163 |
towards | +0.161 |
Neg logit contributions | |
agg | -0.258 |
ths | -0.248 |
Jess | -0.236 |
aman | -0.236 |
Fore | -0.232 |
Token |
|
Token ID | 198 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ths | +0.305 |
eking | +0.289 |
aman | +0.281 |
agg | +0.277 |
prisingly | +0.253 |
Neg logit contributions | |
. | -0.251 |
." | -0.225 |
He | -0.214 |
They | -0.204 |
play | -0.200 |
Token | Then |
Token ID | 6423 |
Feature activation | +4.36e-02 |
Pos logit contributions | |
eking | +0.283 |
ths | +0.277 |
agg | +0.274 |
aman | +0.270 |
tha | +0.249 |
Neg logit contributions | |
. | -0.245 |
." | -0.226 |
He | -0.221 |
They | -0.209 |
play | -0.205 |
Token | one |
Token ID | 530 |
Feature activation | -1.79e-02 |
Pos logit contributions | |
ths | +0.097 |
agg | +0.095 |
eking | +0.094 |
aman | +0.092 |
Kevin | +0.087 |
Neg logit contributions | |
. | -0.056 |
." | -0.050 |
He | -0.045 |
ashes | -0.044 |
They | -0.042 |
Token | day |
Token ID | 1110 |
Feature activation | +4.30e-02 |
Pos logit contributions | |
. | +0.028 |
." | +0.027 |
ashes | +0.025 |
He | +0.024 |
They | +0.024 |
Neg logit contributions | |
ths | -0.033 |
agg | -0.032 |
eking | -0.031 |
aman | -0.030 |
prisingly | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +9.17e-02 |
Pos logit contributions | |
ths | +0.085 |
agg | +0.084 |
eking | +0.079 |
aman | +0.078 |
Jess | +0.076 |
Neg logit contributions | |
. | -0.059 |
." | -0.056 |
ashes | -0.056 |
rider | -0.051 |
He | -0.050 |
Token | some |
Token ID | 617 |
Feature activation | -2.08e-02 |
Pos logit contributions | |
ths | +0.197 |
agg | +0.191 |
eking | +0.186 |
aman | +0.182 |
prisingly | +0.175 |
Neg logit contributions | |
. | -0.117 |
." | -0.109 |
ashes | -0.104 |
He | -0.097 |
rider | -0.094 |
Token | using |
Token ID | 1262 |
Feature activation | -7.82e-02 |
Pos logit contributions | |
. | +0.043 |
." | +0.041 |
ashes | +0.039 |
He | +0.038 |
rider | +0.037 |
Neg logit contributions | |
ths | -0.044 |
agg | -0.042 |
eking | -0.041 |
aman | -0.040 |
Kevin | -0.038 |
Token | the |
Token ID | 262 |
Feature activation | -3.32e-02 |
Pos logit contributions | |
. | +0.110 |
." | +0.104 |
ashes | +0.102 |
He | +0.096 |
rider | +0.095 |
Neg logit contributions | |
ths | -0.153 |
agg | -0.146 |
eking | -0.145 |
aman | -0.141 |
prisingly | -0.134 |
Token | tools |
Token ID | 4899 |
Feature activation | -7.60e-02 |
Pos logit contributions | |
. | +0.043 |
." | +0.040 |
ashes | +0.037 |
He | +0.035 |
rider | +0.034 |
Neg logit contributions | |
ths | -0.072 |
agg | -0.069 |
eking | -0.069 |
aman | -0.066 |
Kevin | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
ashes | +0.044 |
rider | +0.038 |
." | +0.035 |
five | +0.034 |
Let | +0.033 |
Neg logit contributions | |
ths | -0.201 |
eking | -0.194 |
aman | -0.185 |
ering | -0.185 |
agg | -0.183 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
. | +0.202 |
He | +0.182 |
." | +0.181 |
They | +0.174 |
Let | +0.163 |
Neg logit contributions | |
eking | -0.293 |
ths | -0.288 |
agg | -0.287 |
aman | -0.269 |
prisingly | -0.262 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e-01 |
Pos logit contributions | |
eking | +0.409 |
ths | +0.398 |
agg | +0.380 |
aman | +0.373 |
ering | +0.365 |
Neg logit contributions | |
. | -0.139 |
He | -0.115 |
." | -0.110 |
They | -0.105 |
Let | -0.078 |
Token | Steve |
Token ID | 19206 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
eking | +0.208 |
agg | +0.204 |
ths | +0.185 |
aman | +0.183 |
ering | +0.169 |
Neg logit contributions | |
. | -0.307 |
He | -0.285 |
." | -0.279 |
They | -0.275 |
towards | -0.264 |
Token | worked |
Token ID | 3111 |
Feature activation | +2.48e-02 |
Pos logit contributions | |
ths | +0.065 |
agg | +0.061 |
eking | +0.058 |
aman | +0.057 |
prisingly | +0.055 |
Neg logit contributions | |
. | -0.050 |
ashes | -0.049 |
." | -0.049 |
He | -0.043 |
towards | -0.043 |
Token | on |
Token ID | 319 |
Feature activation | -5.64e-02 |
Pos logit contributions | |
agg | +0.052 |
ths | +0.050 |
aman | +0.049 |
eking | +0.049 |
Jess | +0.048 |
Neg logit contributions | |
. | -0.039 |
." | -0.033 |
He | -0.031 |
They | -0.028 |
towards | -0.028 |
Token | the |
Token ID | 262 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
. | +0.073 |
." | +0.064 |
He | +0.059 |
ashes | +0.056 |
They | +0.055 |
Neg logit contributions | |
ths | -0.125 |
agg | -0.123 |
eking | -0.120 |
aman | -0.117 |
Kevin | -0.113 |
Token | black |
Token ID | 2042 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
. | +0.044 |
." | +0.041 |
ashes | +0.039 |
He | +0.037 |
rider | +0.036 |
Neg logit contributions | |
ths | -0.071 |
eking | -0.068 |
agg | -0.068 |
aman | -0.065 |
Kevin | -0.063 |
Token | had |
Token ID | 550 |
Feature activation | -2.11e-02 |
Pos logit contributions | |
ashes | +0.008 |
. | +0.008 |
." | +0.008 |
five | +0.007 |
powers | +0.007 |
Neg logit contributions | |
ths | -0.014 |
eking | -0.013 |
agg | -0.013 |
aman | -0.013 |
Jess | -0.012 |
Token | ever |
Token ID | 1683 |
Feature activation | +5.54e-02 |
Pos logit contributions | |
. | +0.034 |
." | +0.032 |
ashes | +0.031 |
He | +0.030 |
rider | +0.029 |
Neg logit contributions | |
ths | -0.038 |
agg | -0.037 |
eking | -0.036 |
aman | -0.035 |
Kevin | -0.033 |
Token | seen |
Token ID | 1775 |
Feature activation | +1.55e-02 |
Pos logit contributions | |
ths | +0.101 |
agg | +0.097 |
eking | +0.093 |
aman | +0.092 |
prisingly | +0.087 |
Neg logit contributions | |
. | -0.094 |
." | -0.087 |
ashes | -0.080 |
He | -0.079 |
towards | -0.076 |
Token | . |
Token ID | 13 |
Feature activation | -1.41e-01 |
Pos logit contributions | |
ths | +0.036 |
eking | +0.036 |
agg | +0.035 |
aman | +0.034 |
ering | +0.033 |
Neg logit contributions | |
ashes | -0.015 |
They | -0.013 |
He | -0.012 |
. | -0.012 |
powers | -0.012 |
Token | |
Token ID | 220 |
Feature activation | +1.83e-01 |
Pos logit contributions | |
. | +0.208 |
ashes | +0.195 |
." | +0.194 |
rider | +0.183 |
five | +0.168 |
Neg logit contributions | |
ths | -0.261 |
agg | -0.259 |
aman | -0.247 |
eking | -0.239 |
Kevin | -0.239 |
Token |
|
Token ID | 198 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
ths | +0.383 |
agg | +0.374 |
eking | +0.365 |
aman | +0.357 |
Kevin | +0.343 |
Neg logit contributions | |
. | -0.240 |
." | -0.222 |
ashes | -0.208 |
He | -0.198 |
rider | -0.195 |
Token |
|
Token ID | 198 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ths | +0.396 |
eking | +0.383 |
agg | +0.366 |
aman | +0.364 |
ering | +0.344 |
Neg logit contributions | |
. | -0.153 |
." | -0.137 |
He | -0.127 |
They | -0.111 |
Let | -0.099 |
Token | He |
Token ID | 1544 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
ths | +0.234 |
agg | +0.225 |
eking | +0.220 |
aman | +0.212 |
Kevin | +0.199 |
Neg logit contributions | |
. | -0.277 |
." | -0.261 |
ashes | -0.247 |
He | -0.244 |
rider | -0.236 |
Token | began |
Token ID | 2540 |
Feature activation | -1.07e-01 |
Pos logit contributions | |
. | +0.020 |
." | +0.017 |
He | +0.017 |
play | +0.016 |
bar | +0.016 |
Neg logit contributions | |
ths | -0.031 |
eking | -0.030 |
agg | -0.030 |
aman | -0.030 |
prisingly | -0.028 |
Token | to |
Token ID | 284 |
Feature activation | +1.14e-03 |
Pos logit contributions | |
. | +0.079 |
." | +0.063 |
ashes | +0.053 |
He | +0.053 |
towards | +0.047 |
Neg logit contributions | |
ths | -0.296 |
agg | -0.289 |
eking | -0.285 |
aman | -0.280 |
prisingly | -0.271 |
Token | gra |
Token ID | 7933 |
Feature activation | -8.90e-03 |
Pos logit contributions | |
ths | +0.002 |
agg | +0.002 |
eking | +0.002 |
aman | +0.002 |
Kevin | +0.002 |
Neg logit contributions | |
. | -0.001 |
." | -0.001 |
ashes | -0.001 |
He | -0.001 |
rider | -0.001 |
Token | replied |
Token ID | 8712 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
eking | +0.066 |
agg | +0.064 |
ths | +0.063 |
aman | +0.060 |
iter | +0.059 |
Neg logit contributions | |
. | -0.050 |
." | -0.042 |
He | -0.042 |
They | -0.039 |
play | -0.038 |
Token | his |
Token ID | 465 |
Feature activation | -1.43e-02 |
Pos logit contributions | |
ths | +0.204 |
agg | +0.198 |
eking | +0.190 |
aman | +0.188 |
Jess | +0.180 |
Neg logit contributions | |
. | -0.139 |
ashes | -0.134 |
." | -0.133 |
He | -0.121 |
rider | -0.120 |
Token | mom |
Token ID | 1995 |
Feature activation | +1.18e-02 |
Pos logit contributions | |
. | +0.025 |
." | +0.024 |
ashes | +0.024 |
He | +0.022 |
Let | +0.022 |
Neg logit contributions | |
ths | -0.024 |
agg | -0.023 |
eking | -0.021 |
aman | -0.021 |
Kevin | -0.020 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
ths | +0.035 |
eking | +0.034 |
agg | +0.034 |
aman | +0.033 |
ering | +0.033 |
Neg logit contributions | |
ashes | -0.005 |
." | -0.003 |
. | -0.003 |
powers | -0.003 |
rider | -0.002 |
Token |
|
Token ID | 198 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
. | +0.167 |
." | +0.151 |
He | +0.144 |
They | +0.134 |
Let | +0.128 |
Neg logit contributions | |
ths | -0.324 |
eking | -0.319 |
agg | -0.314 |
aman | -0.298 |
prisingly | -0.291 |
Token |
|
Token ID | 198 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
ths | +0.419 |
eking | +0.417 |
agg | +0.388 |
aman | +0.380 |
ering | +0.373 |
Neg logit contributions | |
. | -0.142 |
." | -0.121 |
He | -0.110 |
They | -0.100 |
Let | -0.082 |
Token | The |
Token ID | 464 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
ths | +0.260 |
agg | +0.257 |
eking | +0.257 |
aman | +0.240 |
ering | +0.223 |
Neg logit contributions | |
. | -0.251 |
." | -0.236 |
He | -0.223 |
They | -0.212 |
Let | -0.211 |
Token | little |
Token ID | 1310 |
Feature activation | +3.70e-02 |
Pos logit contributions | |
agg | +0.032 |
ths | +0.032 |
eking | +0.031 |
aman | +0.031 |
prisingly | +0.030 |
Neg logit contributions | |
. | -0.019 |
." | -0.017 |
He | -0.015 |
rider | -0.014 |
ride | -0.014 |
Token | owl |
Token ID | 39610 |
Feature activation | +1.24e-02 |
Pos logit contributions | |
ths | +0.080 |
agg | +0.075 |
eking | +0.075 |
aman | +0.074 |
Jess | +0.071 |
Neg logit contributions | |
. | -0.046 |
ashes | -0.044 |
." | -0.043 |
He | -0.038 |
They | -0.038 |
Token | was |
Token ID | 373 |
Feature activation | +6.42e-02 |
Pos logit contributions | |
ths | +0.023 |
agg | +0.023 |
eking | +0.023 |
aman | +0.022 |
Kevin | +0.021 |
Neg logit contributions | |
. | -0.020 |
." | -0.018 |
He | -0.016 |
play | -0.016 |
rider | -0.015 |
Token | so |
Token ID | 523 |
Feature activation | +8.01e-02 |
Pos logit contributions | |
ths | +0.126 |
aman | +0.118 |
agg | +0.118 |
eking | +0.114 |
Kevin | +0.108 |
Neg logit contributions | |
. | -0.109 |
." | -0.094 |
like | -0.089 |
He | -0.086 |
pet | -0.083 |
Token | kind |
Token ID | 1611 |
Feature activation | -6.94e-03 |
Pos logit contributions | |
ths | +0.133 |
agg | +0.128 |
eking | +0.126 |
aman | +0.120 |
Kevin | +0.116 |
Neg logit contributions | |
. | -0.132 |
." | -0.125 |
ashes | -0.120 |
He | -0.117 |
rider | -0.113 |
Token | to |
Token ID | 284 |
Feature activation | -6.79e-04 |
Pos logit contributions | |
ashes | +0.006 |
pedal | +0.006 |
play | +0.005 |
rider | +0.005 |
bar | +0.005 |
Neg logit contributions | |
ths | -0.016 |
agg | -0.015 |
eking | -0.014 |
Jess | -0.014 |
prisingly | -0.014 |
Token | each |
Token ID | 1123 |
Feature activation | -5.45e-02 |
Pos logit contributions | |
. | +0.001 |
." | +0.001 |
ashes | +0.001 |
He | +0.001 |
rider | +0.001 |
Neg logit contributions | |
ths | -0.001 |
agg | -0.001 |
eking | -0.001 |
aman | -0.001 |
Kevin | -0.001 |
Token | other |
Token ID | 584 |
Feature activation | -6.41e-02 |
Pos logit contributions | |
. | +0.067 |
." | +0.062 |
ashes | +0.061 |
He | +0.059 |
rider | +0.055 |
Neg logit contributions | |
ths | -0.116 |
agg | -0.113 |
eking | -0.111 |
aman | -0.108 |
Jess | -0.105 |
Token | ." |
Token ID | 526 |
Feature activation | -7.15e-02 |
Pos logit contributions | |
ashes | +0.033 |
. | +0.033 |
He | +0.028 |
rider | +0.025 |
." | +0.025 |
Neg logit contributions | |
ths | -0.174 |
eking | -0.169 |
agg | -0.169 |
aman | -0.163 |
Kevin | -0.161 |
Token |
|
Token ID | 198 |
Feature activation | +1.51e-01 |
Pos logit contributions | |
. | +0.095 |
." | +0.087 |
ashes | +0.080 |
He | +0.079 |
rider | +0.074 |
Neg logit contributions | |
ths | -0.152 |
agg | -0.148 |
eking | -0.145 |
aman | -0.141 |
Kevin | -0.135 |
Token |
|
Token ID | 198 |
Feature activation | +1.50e-01 |
Pos logit contributions | |
ths | +0.371 |
eking | +0.365 |
agg | +0.352 |
aman | +0.348 |
ering | +0.328 |
Neg logit contributions | |
. | -0.173 |
." | -0.153 |
He | -0.139 |
They | -0.129 |
ashes | -0.121 |
Token | The |
Token ID | 464 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
eking | +0.268 |
ths | +0.266 |
agg | +0.263 |
aman | +0.251 |
ering | +0.232 |
Neg logit contributions | |
. | -0.252 |
." | -0.235 |
He | -0.224 |
They | -0.214 |
Let | -0.213 |
Token | two |
Token ID | 734 |
Feature activation | -2.62e-02 |
Pos logit contributions | |
ths | +0.033 |
agg | +0.031 |
eking | +0.030 |
aman | +0.029 |
prisingly | +0.028 |
Neg logit contributions | |
. | -0.028 |
." | -0.027 |
ashes | -0.025 |
He | -0.025 |
They | -0.024 |
Token | friends |
Token ID | 2460 |
Feature activation | -7.59e-02 |
Pos logit contributions | |
. | +0.035 |
." | +0.031 |
ashes | +0.029 |
He | +0.028 |
rider | +0.027 |
Neg logit contributions | |
ths | -0.056 |
agg | -0.055 |
eking | -0.054 |
aman | -0.053 |
Kevin | -0.051 |
Token | were |
Token ID | 547 |
Feature activation | +9.99e-02 |
Pos logit contributions | |
. | +0.125 |
." | +0.114 |
He | +0.106 |
ashes | +0.105 |
They | +0.102 |
Neg logit contributions | |
ths | -0.138 |
agg | -0.133 |
eking | -0.132 |
aman | -0.128 |
Kevin | -0.121 |
Token | liked |
Token ID | 8288 |
Feature activation | -1.11e-02 |
Pos logit contributions | |
. | +0.048 |
." | +0.046 |
ashes | +0.044 |
He | +0.042 |
rider | +0.040 |
Neg logit contributions | |
ths | -0.071 |
agg | -0.070 |
eking | -0.067 |
aman | -0.065 |
Kevin | -0.064 |
Token | to |
Token ID | 284 |
Feature activation | -5.74e-03 |
Pos logit contributions | |
. | +0.009 |
." | +0.009 |
ashes | +0.008 |
He | +0.008 |
rider | +0.008 |
Neg logit contributions | |
ths | -0.027 |
agg | -0.026 |
eking | -0.026 |
aman | -0.025 |
Kevin | -0.025 |
Token | slide |
Token ID | 10649 |
Feature activation | -6.27e-02 |
Pos logit contributions | |
powers | +0.006 |
ray | +0.006 |
. | +0.006 |
." | +0.006 |
eer | +0.006 |
Neg logit contributions | |
agg | -0.011 |
eking | -0.011 |
ths | -0.011 |
Kevin | -0.010 |
Jane | -0.010 |
Token | , |
Token ID | 11 |
Feature activation | +7.70e-02 |
Pos logit contributions | |
powers | +0.075 |
five | +0.072 |
ashes | +0.071 |
Let | +0.071 |
eer | +0.070 |
Neg logit contributions | |
ths | -0.129 |
eking | -0.116 |
aman | -0.116 |
agg | -0.114 |
ering | -0.111 |
Token | swing |
Token ID | 9628 |
Feature activation | -1.18e-01 |
Pos logit contributions | |
ths | +0.184 |
eking | +0.182 |
agg | +0.181 |
aman | +0.172 |
Kevin | +0.170 |
Neg logit contributions | |
. | -0.067 |
." | -0.061 |
ashes | -0.055 |
He | -0.052 |
rider | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +7.71e-02 |
Pos logit contributions | |
powers | +0.123 |
ashes | +0.121 |
." | +0.118 |
rider | +0.116 |
Let | +0.115 |
Neg logit contributions | |
ths | -0.246 |
agg | -0.234 |
eking | -0.232 |
aman | -0.229 |
prisingly | -0.215 |
Token | and |
Token ID | 290 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
ths | +0.202 |
agg | +0.196 |
eking | +0.196 |
aman | +0.190 |
Kevin | +0.184 |
Neg logit contributions | |
. | -0.059 |
." | -0.052 |
ashes | -0.045 |
He | -0.043 |
rider | -0.040 |
Token | run |
Token ID | 1057 |
Feature activation | -1.14e-01 |
Pos logit contributions | |
ths | +0.085 |
agg | +0.082 |
eking | +0.082 |
Jess | +0.079 |
Kevin | +0.079 |
Neg logit contributions | |
. | -0.037 |
." | -0.035 |
ashes | -0.032 |
He | -0.031 |
rider | -0.029 |
Token | . |
Token ID | 13 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
ray | +0.119 |
Let | +0.118 |
eer | +0.115 |
fixes | +0.107 |
rider | +0.107 |
Neg logit contributions | |
h | -0.192 |
eking | -0.192 |
ering | -0.192 |
done | -0.191 |
ths | -0.191 |
Token | One |
Token ID | 1881 |
Feature activation | +1.81e-03 |
Pos logit contributions | |
. | +0.165 |
." | +0.155 |
rider | +0.149 |
ashes | +0.138 |
pedal | +0.131 |
Neg logit contributions | |
agg | -0.352 |
ths | -0.344 |
aman | -0.330 |
eking | -0.326 |
Kevin | -0.321 |
Token | day |
Token ID | 1110 |
Feature activation | +3.90e-02 |
Pos logit contributions | |
agg | +0.003 |
ths | +0.003 |
aman | +0.003 |
eking | +0.003 |
prisingly | +0.002 |
Neg logit contributions | |
." | -0.003 |
. | -0.003 |
ashes | -0.003 |
powers | -0.003 |
He | -0.003 |
Token | He |
Token ID | 1544 |
Feature activation | -1.48e-02 |
Pos logit contributions | |
agg | +0.245 |
eking | +0.241 |
ths | +0.239 |
aman | +0.224 |
ering | +0.207 |
Neg logit contributions | |
. | -0.280 |
." | -0.264 |
He | -0.258 |
They | -0.243 |
Let | -0.239 |
Token | planted |
Token ID | 18969 |
Feature activation | -1.34e-02 |
Pos logit contributions | |
. | +0.021 |
." | +0.019 |
He | +0.018 |
ashes | +0.017 |
They | +0.017 |
Neg logit contributions | |
ths | -0.030 |
eking | -0.029 |
agg | -0.029 |
aman | -0.029 |
Kevin | -0.027 |
Token | the |
Token ID | 262 |
Feature activation | -3.43e-02 |
Pos logit contributions | |
. | +0.015 |
." | +0.013 |
He | +0.012 |
like | +0.010 |
They | +0.010 |
Neg logit contributions | |
ths | -0.034 |
eking | -0.032 |
aman | -0.032 |
agg | -0.032 |
Kevin | -0.030 |
Token | flowers |
Token ID | 12734 |
Feature activation | -9.35e-02 |
Pos logit contributions | |
. | +0.047 |
." | +0.042 |
ashes | +0.038 |
He | +0.036 |
towards | +0.035 |
Neg logit contributions | |
ths | -0.075 |
agg | -0.071 |
eking | -0.071 |
aman | -0.069 |
prisingly | -0.067 |
Token | and |
Token ID | 290 |
Feature activation | +4.88e-02 |
Pos logit contributions | |
. | +0.088 |
." | +0.068 |
He | +0.056 |
towards | +0.054 |
ashes | +0.052 |
Neg logit contributions | |
ths | -0.249 |
agg | -0.242 |
aman | -0.240 |
eking | -0.238 |
Kevin | -0.231 |
Token | waited |
Token ID | 13488 |
Feature activation | +5.97e-02 |
Pos logit contributions | |
ths | +0.103 |
eking | +0.101 |
aman | +0.100 |
agg | +0.098 |
Kevin | +0.093 |
Neg logit contributions | |
. | -0.069 |
." | -0.060 |
play | -0.054 |
He | -0.054 |
They | -0.053 |
Token | for |
Token ID | 329 |
Feature activation | -4.67e-03 |
Pos logit contributions | |
ths | +0.124 |
agg | +0.122 |
eking | +0.120 |
aman | +0.115 |
Kevin | +0.111 |
Neg logit contributions | |
. | -0.076 |
." | -0.072 |
ashes | -0.069 |
rider | -0.066 |
He | -0.066 |
Token | them |
Token ID | 606 |
Feature activation | +3.06e-02 |
Pos logit contributions | |
. | +0.006 |
." | +0.005 |
play | +0.004 |
He | +0.004 |
fur | +0.004 |
Neg logit contributions | |
ths | -0.012 |
eking | -0.011 |
aman | -0.011 |
agg | -0.011 |
mber | -0.010 |
Token | to |
Token ID | 284 |
Feature activation | +6.60e-04 |
Pos logit contributions | |
ths | +0.074 |
agg | +0.072 |
eking | +0.071 |
prisingly | +0.067 |
Jess | +0.066 |
Neg logit contributions | |
rider | -0.026 |
ashes | -0.025 |
." | -0.025 |
. | -0.024 |
pedal | -0.023 |
Token | grow |
Token ID | 1663 |
Feature activation | +3.39e-02 |
Pos logit contributions | |
ths | +0.002 |
agg | +0.001 |
eking | +0.001 |
aman | +0.001 |
Kevin | +0.001 |
Neg logit contributions | |
. | -0.001 |
." | -0.001 |
play | -0.001 |
He | -0.001 |
like | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
agg | +0.093 |
ths | +0.093 |
eking | +0.091 |
aman | +0.088 |
Jess | +0.086 |
Neg logit contributions | |
. | -0.019 |
." | -0.016 |
rider | -0.016 |
ashes | -0.016 |
He | -0.014 |
Token | would |
Token ID | 561 |
Feature activation | -8.21e-03 |
Pos logit contributions | |
. | +0.043 |
ashes | +0.042 |
." | +0.041 |
five | +0.038 |
towards | +0.038 |
Neg logit contributions | |
ths | -0.057 |
agg | -0.055 |
eking | -0.053 |
aman | -0.051 |
Jess | -0.050 |
Token | sleep |
Token ID | 3993 |
Feature activation | -2.34e-02 |
Pos logit contributions | |
. | +0.011 |
." | +0.010 |
ashes | +0.010 |
He | +0.009 |
Let | +0.009 |
Neg logit contributions | |
ths | -0.017 |
agg | -0.016 |
eking | -0.016 |
aman | -0.016 |
Kevin | -0.015 |
Token | all |
Token ID | 477 |
Feature activation | -7.66e-02 |
Pos logit contributions | |
. | +0.023 |
." | +0.020 |
He | +0.016 |
ashes | +0.015 |
rider | +0.015 |
Neg logit contributions | |
ths | -0.060 |
agg | -0.060 |
aman | -0.058 |
eking | -0.057 |
Kevin | -0.056 |
Token | day |
Token ID | 1110 |
Feature activation | +4.27e-02 |
Pos logit contributions | |
. | +0.117 |
." | +0.109 |
ashes | +0.105 |
He | +0.101 |
rider | +0.098 |
Neg logit contributions | |
ths | -0.143 |
agg | -0.138 |
eking | -0.136 |
aman | -0.131 |
prisingly | -0.125 |
Token | and |
Token ID | 290 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
ths | +0.107 |
agg | +0.104 |
eking | +0.102 |
aman | +0.100 |
prisingly | +0.097 |
Neg logit contributions | |
. | -0.044 |
." | -0.038 |
ashes | -0.033 |
He | -0.032 |
They | -0.030 |
Token | never |
Token ID | 1239 |
Feature activation | +8.75e-02 |
Pos logit contributions | |
ths | +0.085 |
agg | +0.083 |
eking | +0.081 |
aman | +0.078 |
Kevin | +0.075 |
Neg logit contributions | |
. | -0.070 |
." | -0.065 |
ashes | -0.061 |
He | -0.060 |
rider | -0.057 |
Token | play |
Token ID | 711 |
Feature activation | -1.17e-01 |
Pos logit contributions | |
eking | +0.147 |
ths | +0.141 |
agg | +0.133 |
ering | +0.126 |
aman | +0.124 |
Neg logit contributions | |
ashes | -0.126 |
. | -0.123 |
." | -0.121 |
They | -0.119 |
He | -0.118 |
Token | with |
Token ID | 351 |
Feature activation | -1.36e-02 |
Pos logit contributions | |
ashes | +0.134 |
Let | +0.124 |
ny | +0.121 |
pads | +0.119 |
pedal | +0.118 |
Neg logit contributions | |
ths | -0.233 |
eking | -0.219 |
agg | -0.210 |
unfortunately | -0.208 |
prisingly | -0.207 |
Token | the |
Token ID | 262 |
Feature activation | -3.69e-02 |
Pos logit contributions | |
. | +0.020 |
ashes | +0.020 |
." | +0.019 |
rider | +0.018 |
Let | +0.018 |
Neg logit contributions | |
ths | -0.025 |
eking | -0.024 |
agg | -0.024 |
aman | -0.023 |
Jess | -0.021 |
Token | other |
Token ID | 584 |
Feature activation | -7.01e-02 |
Pos logit contributions | |
. | +0.047 |
ashes | +0.045 |
." | +0.044 |
five | +0.044 |
He | +0.042 |
Neg logit contributions | |
ths | -0.069 |
agg | -0.067 |
eking | -0.067 |
aman | -0.066 |
sudden | -0.064 |
Token | cats |
Token ID | 11875 |
Feature activation | -6.82e-02 |
Pos logit contributions | |
He | +0.079 |
five | +0.075 |
ashes | +0.074 |
. | +0.072 |
." | +0.071 |
Neg logit contributions | |
ths | -0.141 |
eking | -0.134 |
agg | -0.131 |
sudden | -0.129 |
aman | -0.128 |
Token | it |
Token ID | 340 |
Feature activation | +1.81e-02 |
Pos logit contributions | |
ths | +0.217 |
agg | +0.216 |
eking | +0.209 |
aman | +0.208 |
Jess | +0.200 |
Neg logit contributions | |
. | -0.119 |
." | -0.101 |
He | -0.097 |
towards | -0.087 |
They | -0.086 |
Token | is |
Token ID | 318 |
Feature activation | +6.03e-02 |
Pos logit contributions | |
ths | +0.041 |
eking | +0.039 |
agg | +0.038 |
aman | +0.036 |
prisingly | +0.036 |
Neg logit contributions | |
. | -0.020 |
." | -0.020 |
ashes | -0.018 |
five | -0.017 |
rider | -0.017 |
Token | too |
Token ID | 1165 |
Feature activation | +2.75e-02 |
Pos logit contributions | |
ths | +0.115 |
eking | +0.109 |
agg | +0.107 |
aman | +0.103 |
prisingly | +0.101 |
Neg logit contributions | |
. | -0.088 |
ashes | -0.086 |
." | -0.085 |
five | -0.079 |
He | -0.077 |
Token | hard |
Token ID | 1327 |
Feature activation | -1.86e-02 |
Pos logit contributions | |
ths | +0.050 |
agg | +0.048 |
eking | +0.047 |
aman | +0.046 |
prisingly | +0.044 |
Neg logit contributions | |
. | -0.042 |
." | -0.040 |
ashes | -0.039 |
He | -0.037 |
Let | -0.037 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ashes | +0.012 |
rider | +0.010 |
five | +0.009 |
pet | +0.009 |
player | +0.009 |
Neg logit contributions | |
eking | -0.046 |
ths | -0.046 |
ering | -0.045 |
aman | -0.043 |
done | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
. | +0.118 |
." | +0.097 |
He | +0.095 |
They | +0.087 |
ashes | +0.083 |
Neg logit contributions | |
ths | -0.394 |
eking | -0.384 |
agg | -0.378 |
prisingly | -0.362 |
aman | -0.362 |
Token |
|
Token ID | 198 |
Feature activation | +1.41e-01 |
Pos logit contributions | |
ths | +0.401 |
eking | +0.394 |
agg | +0.364 |
aman | +0.361 |
ering | +0.353 |
Neg logit contributions | |
. | -0.140 |
." | -0.112 |
He | -0.110 |
They | -0.102 |
Let | -0.070 |
Token | " |
Token ID | 1 |
Feature activation | +2.05e-02 |
Pos logit contributions | |
ths | +0.272 |
eking | +0.261 |
agg | +0.257 |
aman | +0.247 |
prisingly | +0.236 |
Neg logit contributions | |
. | -0.225 |
." | -0.205 |
He | -0.194 |
ashes | -0.190 |
They | -0.186 |
Token | No |
Token ID | 2949 |
Feature activation | +8.58e-02 |
Pos logit contributions | |
eking | +0.042 |
ths | +0.041 |
agg | +0.041 |
aman | +0.039 |
ering | +0.038 |
Neg logit contributions | |
. | -0.028 |
ashes | -0.026 |
." | -0.025 |
He | -0.025 |
bar | -0.025 |
Token | , |
Token ID | 11 |
Feature activation | +8.67e-02 |
Pos logit contributions | |
ths | +0.158 |
agg | +0.153 |
aman | +0.151 |
eking | +0.148 |
ering | +0.136 |
Neg logit contributions | |
ashes | -0.127 |
. | -0.120 |
." | -0.116 |
He | -0.116 |
rider | -0.115 |
Token | Sara |
Token ID | 24799 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
ths | +0.199 |
agg | +0.191 |
eking | +0.189 |
aman | +0.183 |
Kevin | +0.178 |
Neg logit contributions | |
. | -0.109 |
." | -0.100 |
He | -0.087 |
ashes | -0.083 |
They | -0.082 |
Token | my |
Token ID | 1820 |
Feature activation | +2.01e-02 |
Pos logit contributions | |
ths | +0.093 |
agg | +0.088 |
eking | +0.088 |
aman | +0.084 |
prisingly | +0.081 |
Neg logit contributions | |
. | -0.095 |
." | -0.090 |
ashes | -0.090 |
He | -0.084 |
They | -0.082 |
Token | said |
Token ID | 531 |
Feature activation | +7.33e-02 |
Pos logit contributions | |
ths | +0.037 |
agg | +0.036 |
eking | +0.036 |
aman | +0.035 |
Jess | +0.033 |
Neg logit contributions | |
. | -0.034 |
." | -0.031 |
He | -0.029 |
ashes | -0.027 |
They | -0.027 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ths | +0.179 |
agg | +0.178 |
pleased | +0.178 |
impressed | +0.178 |
eking | +0.176 |
Neg logit contributions | |
ashes | -0.052 |
bars | -0.047 |
eer | -0.043 |
spots | -0.043 |
pads | -0.042 |
Token | His |
Token ID | 2399 |
Feature activation | +3.21e-02 |
Pos logit contributions | |
rider | +0.180 |
ashes | +0.178 |
bars | +0.172 |
. | +0.171 |
." | +0.169 |
Neg logit contributions | |
ths | -0.258 |
agg | -0.255 |
Kevin | -0.233 |
Jane | -0.230 |
mber | -0.227 |
Token | friends |
Token ID | 2460 |
Feature activation | -7.92e-02 |
Pos logit contributions | |
ths | +0.058 |
agg | +0.054 |
eking | +0.053 |
aman | +0.050 |
Kevin | +0.049 |
Neg logit contributions | |
. | -0.051 |
." | -0.049 |
ashes | -0.046 |
Let | -0.044 |
He | -0.044 |
Token | were |
Token ID | 547 |
Feature activation | +9.34e-02 |
Pos logit contributions | |
. | +0.132 |
." | +0.119 |
He | +0.110 |
ashes | +0.108 |
They | +0.106 |
Neg logit contributions | |
ths | -0.148 |
agg | -0.144 |
eking | -0.140 |
aman | -0.137 |
Jess | -0.130 |
Token | amazed |
Token ID | 24562 |
Feature activation | +9.04e-02 |
Pos logit contributions | |
ths | +0.182 |
agg | +0.180 |
aman | +0.179 |
eking | +0.170 |
prisingly | +0.162 |
Neg logit contributions | |
. | -0.166 |
." | -0.144 |
He | -0.134 |
like | -0.133 |
They | -0.132 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ths | +0.239 |
eking | +0.227 |
agg | +0.226 |
aman | +0.222 |
Kevin | +0.215 |
Neg logit contributions | |
ashes | -0.061 |
." | -0.059 |
. | -0.057 |
five | -0.051 |
Let | -0.050 |
Token | " |
Token ID | 366 |
Feature activation | +6.57e-02 |
Pos logit contributions | |
. | +0.143 |
rider | +0.136 |
." | +0.133 |
ashes | +0.132 |
five | +0.119 |
Neg logit contributions | |
ths | -0.323 |
agg | -0.312 |
aman | -0.298 |
eking | -0.292 |
Kevin | -0.288 |
Token | Wow |
Token ID | 22017 |
Feature activation | +1.43e-01 |
Pos logit contributions | |
ths | +0.119 |
agg | +0.113 |
eking | +0.110 |
aman | +0.108 |
Jess | +0.104 |
Neg logit contributions | |
. | -0.108 |
." | -0.100 |
ashes | -0.095 |
He | -0.091 |
rider | -0.088 |
Token | , |
Token ID | 11 |
Feature activation | +8.37e-02 |
Pos logit contributions | |
ths | +0.295 |
eking | +0.278 |
agg | +0.273 |
aman | +0.262 |
ering | +0.256 |
Neg logit contributions | |
ashes | -0.183 |
. | -0.174 |
." | -0.165 |
rider | -0.164 |
He | -0.162 |
Token | see |
Token ID | 766 |
Feature activation | +1.53e-02 |
Pos logit contributions | |
ashes | +0.005 |
. | +0.005 |
." | +0.004 |
rider | +0.004 |
powers | +0.004 |
Neg logit contributions | |
agg | -0.007 |
ths | -0.007 |
Kevin | -0.006 |
eking | -0.006 |
Jess | -0.006 |
Token | the |
Token ID | 262 |
Feature activation | -3.25e-02 |
Pos logit contributions | |
ths | +0.024 |
aman | +0.022 |
agg | +0.022 |
eking | +0.022 |
Jess | +0.021 |
Neg logit contributions | |
ashes | -0.027 |
. | -0.024 |
powers | -0.024 |
ny | -0.024 |
." | -0.024 |
Token | big |
Token ID | 1263 |
Feature activation | -2.99e-02 |
Pos logit contributions | |
. | +0.028 |
." | +0.025 |
ashes | +0.021 |
He | +0.021 |
rider | +0.019 |
Neg logit contributions | |
ths | -0.085 |
agg | -0.083 |
eking | -0.082 |
aman | -0.080 |
Kevin | -0.077 |
Token | , |
Token ID | 11 |
Feature activation | +9.15e-02 |
Pos logit contributions | |
ashes | +0.031 |
. | +0.030 |
." | +0.028 |
Let | +0.026 |
powers | +0.026 |
Neg logit contributions | |
ths | -0.065 |
eking | -0.063 |
aman | -0.062 |
agg | -0.062 |
ering | -0.058 |
Token | gray |
Token ID | 12768 |
Feature activation | -2.38e-02 |
Pos logit contributions | |
ths | +0.188 |
agg | +0.182 |
eking | +0.181 |
aman | +0.175 |
Kevin | +0.169 |
Neg logit contributions | |
. | -0.121 |
." | -0.113 |
ashes | -0.108 |
He | -0.100 |
Let | -0.097 |
Token | animal |
Token ID | 5044 |
Feature activation | +2.60e-02 |
Pos logit contributions | |
ashes | +0.032 |
Let | +0.027 |
." | +0.026 |
powers | +0.026 |
. | +0.025 |
Neg logit contributions | |
ths | -0.044 |
aman | -0.044 |
agg | -0.042 |
eking | -0.041 |
sudden | -0.041 |
Token | with |
Token ID | 351 |
Feature activation | -1.21e-02 |
Pos logit contributions | |
ths | +0.062 |
eking | +0.061 |
ering | +0.058 |
agg | +0.057 |
impressed | +0.057 |
Neg logit contributions | |
ashes | -0.023 |
five | -0.021 |
." | -0.020 |
Let | -0.020 |
ny | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -3.28e-02 |
Pos logit contributions | |
. | +0.016 |
." | +0.015 |
ashes | +0.014 |
He | +0.013 |
Let | +0.013 |
Neg logit contributions | |
ths | -0.025 |
agg | -0.025 |
eking | -0.024 |
aman | -0.023 |
Kevin | -0.023 |
Token | point |
Token ID | 966 |
Feature activation | -2.20e-02 |
Pos logit contributions | |
. | +0.040 |
." | +0.037 |
ashes | +0.037 |
He | +0.033 |
Let | +0.033 |
Neg logit contributions | |
ths | -0.069 |
eking | -0.067 |
agg | -0.067 |
aman | -0.065 |
Kevin | -0.062 |
Token | y |
Token ID | 88 |
Feature activation | +3.47e-03 |
Pos logit contributions | |
ashes | +0.026 |
five | +0.025 |
powers | +0.025 |
rider | +0.024 |
." | +0.024 |
Neg logit contributions | |
eking | -0.045 |
ths | -0.043 |
ering | -0.042 |
agg | -0.042 |
aman | -0.041 |
Token | horn |
Token ID | 12718 |
Feature activation | -1.23e-02 |
Pos logit contributions | |
ths | +0.006 |
eking | +0.006 |
agg | +0.006 |
sudden | +0.006 |
aman | +0.006 |
Neg logit contributions | |
ashes | -0.004 |
Let | -0.004 |
." | -0.004 |
They | -0.003 |
. | -0.003 |
Token | got |
Token ID | 1392 |
Feature activation | -5.02e-03 |
Pos logit contributions | |
. | +0.071 |
." | +0.068 |
ashes | +0.066 |
He | +0.062 |
rider | +0.060 |
Neg logit contributions | |
ths | -0.098 |
agg | -0.094 |
eking | -0.088 |
prisingly | -0.085 |
aman | -0.084 |
Token | a |
Token ID | 257 |
Feature activation | -2.85e-02 |
Pos logit contributions | |
. | +0.006 |
." | +0.006 |
rider | +0.006 |
ashes | +0.006 |
He | +0.006 |
Neg logit contributions | |
ths | -0.011 |
agg | -0.010 |
eking | -0.010 |
prisingly | -0.009 |
aman | -0.009 |
Token | bucket |
Token ID | 19236 |
Feature activation | -2.54e-02 |
Pos logit contributions | |
. | +0.038 |
." | +0.035 |
ashes | +0.033 |
He | +0.032 |
rider | +0.030 |
Neg logit contributions | |
ths | -0.060 |
agg | -0.058 |
eking | -0.057 |
aman | -0.056 |
Kevin | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +4.51e-02 |
Pos logit contributions | |
ashes | +0.028 |
rider | +0.028 |
eer | +0.027 |
powers | +0.027 |
." | +0.026 |
Neg logit contributions | |
ths | -0.051 |
eking | -0.048 |
aman | -0.048 |
agg | -0.047 |
prisingly | -0.047 |
Token | filled |
Token ID | 5901 |
Feature activation | -1.76e-02 |
Pos logit contributions | |
ths | +0.083 |
agg | +0.081 |
prisingly | +0.073 |
sudden | +0.073 |
aman | +0.071 |
Neg logit contributions | |
ashes | -0.066 |
rider | -0.065 |
. | -0.064 |
." | -0.064 |
ny | -0.061 |
Token | it |
Token ID | 340 |
Feature activation | +2.24e-02 |
Pos logit contributions | |
. | +0.024 |
." | +0.021 |
They | +0.018 |
towards | +0.018 |
He | +0.018 |
Neg logit contributions | |
ths | -0.040 |
agg | -0.038 |
eking | -0.038 |
aman | -0.038 |
Kevin | -0.036 |
Token | with |
Token ID | 351 |
Feature activation | -1.41e-02 |
Pos logit contributions | |
ths | +0.044 |
eking | +0.044 |
agg | +0.041 |
ering | +0.040 |
impressed | +0.039 |
Neg logit contributions | |
rider | -0.029 |
eer | -0.028 |
ashes | -0.026 |
five | -0.026 |
zone | -0.025 |
Token | pe |
Token ID | 613 |
Feature activation | +1.10e-01 |
Pos logit contributions | |
rider | +0.017 |
." | +0.017 |
. | +0.016 |
ashes | +0.016 |
ny | +0.016 |
Neg logit contributions | |
ths | -0.028 |
agg | -0.028 |
eking | -0.027 |
aman | -0.027 |
sudden | -0.025 |
Token | b |
Token ID | 65 |
Feature activation | +9.73e-02 |
Pos logit contributions | |
ths | +0.146 |
aman | +0.142 |
agg | +0.137 |
Jess | +0.131 |
Kevin | +0.125 |
Neg logit contributions | |
. | -0.234 |
." | -0.217 |
ashes | -0.206 |
bars | -0.192 |
He | -0.192 |
Token | bles |
Token ID | 7689 |
Feature activation | -4.87e-02 |
Pos logit contributions | |
ths | +0.118 |
agg | +0.116 |
eking | +0.116 |
aman | +0.103 |
Kevin | +0.098 |
Neg logit contributions | |
. | -0.206 |
." | -0.200 |
ashes | -0.195 |
rider | -0.193 |
He | -0.191 |
Token | . |
Token ID | 13 |
Feature activation | -1.45e-01 |
Pos logit contributions | |
rider | +0.036 |
eer | +0.029 |
Let | +0.026 |
He | +0.026 |
ny | +0.025 |
Neg logit contributions | |
ths | -0.122 |
agg | -0.121 |
eking | -0.121 |
ering | -0.117 |
impressed | -0.113 |
Token | there |
Token ID | 612 |
Feature activation | -1.73e-02 |
Pos logit contributions | |
ths | +0.149 |
eking | +0.140 |
agg | +0.140 |
aman | +0.139 |
Kevin | +0.128 |
Neg logit contributions | |
. | -0.151 |
." | -0.136 |
He | -0.129 |
They | -0.125 |
Let | -0.121 |
Token | was |
Token ID | 373 |
Feature activation | +5.75e-02 |
Pos logit contributions | |
. | +0.020 |
." | +0.016 |
He | +0.016 |
They | +0.014 |
play | +0.014 |
Neg logit contributions | |
ths | -0.044 |
eking | -0.044 |
agg | -0.042 |
aman | -0.041 |
Kevin | -0.040 |
Token | a |
Token ID | 257 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
ths | +0.152 |
eking | +0.148 |
agg | +0.144 |
aman | +0.141 |
ering | +0.137 |
Neg logit contributions | |
. | -0.054 |
." | -0.046 |
He | -0.040 |
They | -0.038 |
ashes | -0.037 |
Token | battery |
Token ID | 6555 |
Feature activation | -1.55e-02 |
Pos logit contributions | |
. | +0.046 |
." | +0.043 |
ashes | +0.041 |
He | +0.040 |
Let | +0.038 |
Neg logit contributions | |
ths | -0.060 |
agg | -0.059 |
eking | -0.057 |
aman | -0.056 |
Kevin | -0.053 |
Token | . |
Token ID | 13 |
Feature activation | -1.51e-01 |
Pos logit contributions | |
ashes | +0.022 |
five | +0.020 |
irms | +0.019 |
powers | +0.019 |
pads | +0.017 |
Neg logit contributions | |
ths | -0.028 |
lf | -0.027 |
eking | -0.026 |
unfortunately | -0.026 |
nowhere | -0.025 |
Token | It |
Token ID | 632 |
Feature activation | +3.76e-02 |
Pos logit contributions | |
ashes | +0.196 |
. | +0.163 |
pads | +0.156 |
." | +0.146 |
powers | +0.145 |
Neg logit contributions | |
agg | -0.290 |
ths | -0.289 |
Jess | -0.279 |
mmm | -0.276 |
unfortunately | -0.276 |
Token | had |
Token ID | 550 |
Feature activation | -2.78e-02 |
Pos logit contributions | |
ths | +0.053 |
agg | +0.050 |
eking | +0.050 |
prisingly | +0.046 |
nowhere | +0.045 |
Neg logit contributions | |
ashes | -0.069 |
five | -0.064 |
. | -0.062 |
." | -0.062 |
rider | -0.059 |
Token | a |
Token ID | 257 |
Feature activation | -3.08e-02 |
Pos logit contributions | |
ashes | +0.040 |
. | +0.039 |
." | +0.036 |
rider | +0.035 |
five | +0.034 |
Neg logit contributions | |
agg | -0.048 |
ths | -0.048 |
eking | -0.048 |
sudden | -0.045 |
prisingly | -0.044 |
Token | very |
Token ID | 845 |
Feature activation | +6.65e-02 |
Pos logit contributions | |
ashes | +0.046 |
. | +0.043 |
." | +0.042 |
five | +0.042 |
Let | +0.039 |
Neg logit contributions | |
ths | -0.055 |
sudden | -0.053 |
agg | -0.051 |
eking | -0.050 |
aman | -0.050 |
Token | important |
Token ID | 1593 |
Feature activation | +2.25e-02 |
Pos logit contributions | |
agg | +0.111 |
ths | +0.105 |
eking | +0.104 |
sudden | +0.095 |
impressed | +0.094 |
Neg logit contributions | |
. | -0.106 |
Let | -0.103 |
." | -0.101 |
ashes | -0.101 |
five | -0.098 |
Token | job |
Token ID | 1693 |
Feature activation | -2.04e-02 |
Pos logit contributions | |
ths | +0.034 |
agg | +0.033 |
sudden | +0.033 |
prisingly | +0.032 |
aman | +0.032 |
Neg logit contributions | |
ashes | -0.035 |
pads | -0.031 |
Let | -0.029 |
." | -0.029 |
They | -0.029 |
Token | She |
Token ID | 1375 |
Feature activation | -3.36e-02 |
Pos logit contributions | |
ashes | +0.197 |
rider | +0.197 |
eer | +0.190 |
." | +0.173 |
. | +0.172 |
Neg logit contributions | |
agg | -0.247 |
ths | -0.235 |
omm | -0.233 |
Jane | -0.231 |
unfortunately | -0.230 |
Token | opened |
Token ID | 4721 |
Feature activation | -3.90e-02 |
Pos logit contributions | |
. | +0.045 |
." | +0.039 |
He | +0.037 |
They | +0.034 |
Let | +0.034 |
Neg logit contributions | |
ths | -0.073 |
agg | -0.071 |
eking | -0.071 |
aman | -0.070 |
Kevin | -0.066 |
Token | the |
Token ID | 262 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
. | +0.059 |
." | +0.053 |
He | +0.048 |
ashes | +0.046 |
towards | +0.046 |
Neg logit contributions | |
ths | -0.079 |
agg | -0.077 |
eking | -0.075 |
aman | -0.073 |
Kevin | -0.070 |
Token | door |
Token ID | 3420 |
Feature activation | -4.28e-02 |
Pos logit contributions | |
. | +0.056 |
ashes | +0.056 |
." | +0.055 |
rider | +0.053 |
He | +0.052 |
Neg logit contributions | |
ths | -0.061 |
eking | -0.059 |
agg | -0.058 |
aman | -0.057 |
prisingly | -0.053 |
Token | and |
Token ID | 290 |
Feature activation | +4.64e-02 |
Pos logit contributions | |
ashes | +0.042 |
eer | +0.038 |
." | +0.037 |
rider | +0.037 |
powers | +0.035 |
Neg logit contributions | |
ths | -0.100 |
agg | -0.094 |
eking | -0.094 |
aman | -0.094 |
prisingly | -0.090 |
Token | saw |
Token ID | 2497 |
Feature activation | +2.23e-02 |
Pos logit contributions | |
ths | +0.085 |
agg | +0.082 |
eking | +0.081 |
aman | +0.078 |
Kevin | +0.074 |
Neg logit contributions | |
. | -0.074 |
." | -0.070 |
ashes | -0.066 |
He | -0.064 |
rider | -0.062 |
Token | stairs |
Token ID | 16046 |
Feature activation | -9.17e-02 |
Pos logit contributions | |
ths | +0.047 |
agg | +0.045 |
eking | +0.044 |
aman | +0.043 |
Jess | +0.041 |
Neg logit contributions | |
. | -0.028 |
." | -0.027 |
ashes | -0.026 |
rider | -0.024 |
ny | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | -1.42e-01 |
Pos logit contributions | |
ashes | +0.065 |
powers | +0.058 |
." | +0.054 |
eer | +0.053 |
five | +0.053 |
Neg logit contributions | |
ths | -0.242 |
eking | -0.238 |
aman | -0.222 |
ering | -0.221 |
agg | -0.218 |
Token | She |
Token ID | 1375 |
Feature activation | -3.35e-02 |
Pos logit contributions | |
rider | +0.196 |
eer | +0.193 |
ashes | +0.182 |
ny | +0.174 |
. | +0.170 |
Neg logit contributions | |
agg | -0.252 |
ths | -0.251 |
aman | -0.235 |
Jane | -0.232 |
Jess | -0.232 |
Token | climbed |
Token ID | 19952 |
Feature activation | -3.00e-02 |
Pos logit contributions | |
. | +0.045 |
." | +0.039 |
He | +0.037 |
They | +0.034 |
Let | +0.034 |
Neg logit contributions | |
ths | -0.072 |
aman | -0.072 |
eking | -0.071 |
agg | -0.071 |
Kevin | -0.067 |
Token | the |
Token ID | 262 |
Feature activation | -3.54e-02 |
Pos logit contributions | |
ashes | +0.036 |
powers | +0.032 |
rider | +0.032 |
eer | +0.032 |
." | +0.031 |
Neg logit contributions | |
ths | -0.064 |
eking | -0.059 |
agg | -0.059 |
aman | -0.057 |
prisingly | -0.055 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
. | +0.179 |
." | +0.163 |
He | +0.149 |
ashes | +0.149 |
They | +0.142 |
Neg logit contributions | |
ths | -0.322 |
agg | -0.311 |
eking | -0.310 |
aman | -0.297 |
prisingly | -0.287 |
Token |
|
Token ID | 198 |
Feature activation | +1.46e-01 |
Pos logit contributions | |
ths | +0.351 |
eking | +0.342 |
agg | +0.330 |
aman | +0.319 |
prisingly | +0.310 |
Neg logit contributions | |
. | -0.171 |
." | -0.152 |
He | -0.138 |
They | -0.132 |
ashes | -0.130 |
Token | Tom |
Token ID | 13787 |
Feature activation | +6.55e-02 |
Pos logit contributions | |
ths | +0.249 |
eking | +0.248 |
agg | +0.244 |
aman | +0.225 |
ering | +0.215 |
Neg logit contributions | |
. | -0.252 |
." | -0.235 |
ashes | -0.222 |
He | -0.218 |
They | -0.214 |
Token | che |
Token ID | 1125 |
Feature activation | +3.68e-02 |
Pos logit contributions | |
ths | +0.134 |
agg | +0.129 |
eking | +0.126 |
aman | +0.123 |
Kevin | +0.118 |
Neg logit contributions | |
. | -0.090 |
." | -0.085 |
ashes | -0.080 |
He | -0.076 |
rider | -0.073 |
Token | wed |
Token ID | 19103 |
Feature activation | -3.05e-02 |
Pos logit contributions | |
ths | +0.083 |
agg | +0.078 |
aman | +0.078 |
eking | +0.077 |
Kevin | +0.076 |
Neg logit contributions | |
. | -0.043 |
." | -0.039 |
ashes | -0.038 |
ny | -0.035 |
bars | -0.033 |
Token | his |
Token ID | 465 |
Feature activation | -1.50e-02 |
Pos logit contributions | |
. | +0.034 |
." | +0.031 |
ashes | +0.029 |
He | +0.028 |
rider | +0.026 |
Neg logit contributions | |
ths | -0.071 |
agg | -0.069 |
eking | -0.068 |
aman | -0.066 |
Kevin | -0.063 |
Token | gum |
Token ID | 27142 |
Feature activation | -1.07e-02 |
Pos logit contributions | |
. | +0.022 |
." | +0.021 |
ashes | +0.020 |
They | +0.019 |
towards | +0.019 |
Neg logit contributions | |
ths | -0.028 |
eking | -0.027 |
agg | -0.026 |
aman | -0.025 |
ering | -0.024 |
Token | too |
Token ID | 1165 |
Feature activation | +2.98e-02 |
Pos logit contributions | |
. | +0.011 |
." | +0.010 |
ashes | +0.010 |
He | +0.009 |
rider | +0.009 |
Neg logit contributions | |
ths | -0.025 |
agg | -0.024 |
eking | -0.024 |
aman | -0.023 |
prisingly | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ths | +0.060 |
agg | +0.058 |
eking | +0.057 |
aman | +0.055 |
Kevin | +0.053 |
Neg logit contributions | |
. | -0.038 |
." | -0.037 |
ashes | -0.037 |
Let | -0.035 |
rider | -0.034 |
Token | He |
Token ID | 679 |
Feature activation | -1.88e-02 |
Pos logit contributions | |
. | +0.115 |
." | +0.102 |
rider | +0.095 |
ashes | +0.094 |
eer | +0.088 |
Neg logit contributions | |
ths | -0.358 |
agg | -0.355 |
aman | -0.343 |
Kevin | -0.331 |
eking | -0.327 |
Token | smiled |
Token ID | 13541 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
. | +0.030 |
." | +0.028 |
ashes | +0.027 |
He | +0.026 |
rider | +0.025 |
Neg logit contributions | |
ths | -0.034 |
agg | -0.033 |
eking | -0.032 |
aman | -0.031 |
Kevin | -0.030 |
Token | her |
Token ID | 607 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
. | +0.015 |
." | +0.014 |
ashes | +0.013 |
He | +0.013 |
They | +0.012 |
Neg logit contributions | |
ths | -0.019 |
agg | -0.018 |
eking | -0.018 |
aman | -0.017 |
prisingly | -0.017 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.17e-03 |
Pos logit contributions | |
ashes | +0.073 |
. | +0.071 |
." | +0.070 |
He | +0.068 |
ny | +0.067 |
Neg logit contributions | |
ths | -0.065 |
agg | -0.063 |
aman | -0.059 |
eking | -0.058 |
Kevin | -0.056 |
Token | what |
Token ID | 644 |
Feature activation | -3.60e-02 |
Pos logit contributions | |
ths | +0.005 |
agg | +0.005 |
eking | +0.004 |
aman | +0.004 |
prisingly | +0.004 |
Neg logit contributions | |
. | -0.003 |
." | -0.002 |
ashes | -0.002 |
He | -0.002 |
rider | -0.002 |
Token | was |
Token ID | 373 |
Feature activation | +5.79e-02 |
Pos logit contributions | |
. | +0.060 |
ashes | +0.060 |
." | +0.058 |
He | +0.054 |
towards | +0.053 |
Neg logit contributions | |
ths | -0.060 |
agg | -0.058 |
aman | -0.054 |
eking | -0.054 |
Jess | -0.053 |
Token | inside |
Token ID | 2641 |
Feature activation | -6.00e-02 |
Pos logit contributions | |
ths | +0.108 |
agg | +0.104 |
eking | +0.101 |
aman | +0.098 |
Kevin | +0.094 |
Neg logit contributions | |
. | -0.095 |
." | -0.088 |
ashes | -0.081 |
He | -0.080 |
rider | -0.077 |
Token | . |
Token ID | 13 |
Feature activation | -1.38e-01 |
Pos logit contributions | |
. | +0.056 |
ashes | +0.052 |
." | +0.052 |
He | +0.048 |
rider | +0.046 |
Neg logit contributions | |
ths | -0.144 |
agg | -0.141 |
eking | -0.139 |
aman | -0.137 |
Kevin | -0.131 |
Token | Her |
Token ID | 2332 |
Feature activation | +2.32e-02 |
Pos logit contributions | |
. | +0.171 |
." | +0.153 |
He | +0.144 |
They | +0.138 |
Let | +0.132 |
Neg logit contributions | |
ths | -0.319 |
eking | -0.312 |
agg | -0.306 |
aman | -0.290 |
prisingly | -0.286 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.12e-03 |
Pos logit contributions | |
ths | +0.040 |
agg | +0.039 |
eking | +0.038 |
aman | +0.037 |
Kevin | +0.035 |
Neg logit contributions | |
. | -0.039 |
." | -0.037 |
ashes | -0.036 |
He | -0.035 |
rider | -0.033 |
Token | said |
Token ID | 531 |
Feature activation | +7.57e-02 |
Pos logit contributions | |
agg | +0.011 |
eking | +0.011 |
ths | +0.011 |
Kevin | +0.011 |
Jess | +0.011 |
Neg logit contributions | |
. | -0.007 |
." | -0.006 |
play | -0.005 |
likes | -0.005 |
bar | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | +1.77e-02 |
Pos logit contributions | |
ths | +0.176 |
eking | +0.170 |
aman | +0.168 |
agg | +0.166 |
Kevin | +0.161 |
Neg logit contributions | |
. | -0.109 |
." | -0.089 |
He | -0.077 |
like | -0.075 |
They | -0.072 |
Token | was |
Token ID | 373 |
Feature activation | +6.02e-02 |
Pos logit contributions | |
ths | +0.034 |
agg | +0.033 |
eking | +0.032 |
aman | +0.031 |
prisingly | +0.030 |
Neg logit contributions | |
. | -0.026 |
." | -0.025 |
ashes | -0.024 |
He | -0.022 |
rider | -0.022 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ths | +0.232 |
eking | +0.225 |
agg | +0.223 |
aman | +0.217 |
ering | +0.213 |
Neg logit contributions | |
ashes | -0.053 |
." | -0.048 |
rider | -0.046 |
. | -0.046 |
powers | -0.045 |
Token | They |
Token ID | 1119 |
Feature activation | -2.71e-02 |
Pos logit contributions | |
eer | +0.163 |
rider | +0.160 |
raid | +0.150 |
ashes | +0.150 |
five | +0.148 |
Neg logit contributions | |
ths | -0.259 |
sudden | -0.252 |
Da | -0.248 |
July | -0.243 |
Finally | -0.241 |
Token | loved |
Token ID | 6151 |
Feature activation | +6.25e-03 |
Pos logit contributions | |
." | +0.044 |
ashes | +0.044 |
. | +0.043 |
Let | +0.041 |
powers | +0.040 |
Neg logit contributions | |
agg | -0.045 |
ths | -0.042 |
aman | -0.040 |
Kevin | -0.040 |
eking | -0.039 |
Token | the |
Token ID | 262 |
Feature activation | -3.21e-02 |
Pos logit contributions | |
ths | +0.014 |
agg | +0.014 |
eking | +0.014 |
aman | +0.013 |
Kevin | +0.013 |
Neg logit contributions | |
. | -0.007 |
." | -0.007 |
ashes | -0.006 |
He | -0.006 |
rider | -0.006 |
Token | snow |
Token ID | 6729 |
Feature activation | +1.52e-02 |
Pos logit contributions | |
. | +0.032 |
." | +0.029 |
ashes | +0.027 |
He | +0.025 |
rider | +0.024 |
Neg logit contributions | |
ths | -0.078 |
agg | -0.076 |
eking | -0.075 |
aman | -0.073 |
Kevin | -0.070 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ths | +0.028 |
agg | +0.026 |
eking | +0.026 |
aman | +0.024 |
ering | +0.024 |
Neg logit contributions | |
ashes | -0.021 |
." | -0.021 |
. | -0.020 |
powers | -0.020 |
five | -0.020 |
Token | They |
Token ID | 1119 |
Feature activation | -2.72e-02 |
Pos logit contributions | |
rider | +0.141 |
eer | +0.138 |
ny | +0.125 |
five | +0.123 |
ashes | +0.123 |
Neg logit contributions | |
agg | -0.290 |
unfortunately | -0.286 |
ths | -0.285 |
Finally | -0.280 |
sudden | -0.278 |
Token | put |
Token ID | 1234 |
Feature activation | -5.06e-02 |
Pos logit contributions | |
ashes | +0.042 |
." | +0.041 |
. | +0.040 |
Let | +0.039 |
towards | +0.039 |
Neg logit contributions | |
agg | -0.047 |
ths | -0.043 |
Kevin | -0.041 |
aman | -0.040 |
eking | -0.040 |
Token | on |
Token ID | 319 |
Feature activation | -5.80e-02 |
Pos logit contributions | |
. | +0.050 |
." | +0.045 |
ashes | +0.040 |
He | +0.039 |
rider | +0.036 |
Neg logit contributions | |
ths | -0.125 |
agg | -0.122 |
eking | -0.120 |
aman | -0.117 |
Kevin | -0.113 |
Token | their |
Token ID | 511 |
Feature activation | +1.75e-02 |
Pos logit contributions | |
. | +0.088 |
." | +0.084 |
ashes | +0.079 |
He | +0.077 |
rider | +0.075 |
Neg logit contributions | |
ths | -0.110 |
agg | -0.107 |
eking | -0.104 |
aman | -0.102 |
Kevin | -0.096 |
Token | coats |
Token ID | 30720 |
Feature activation | -6.04e-02 |
Pos logit contributions | |
ths | +0.039 |
agg | +0.038 |
eking | +0.037 |
aman | +0.036 |
Kevin | +0.035 |
Neg logit contributions | |
. | -0.020 |
." | -0.019 |
ashes | -0.018 |
He | -0.017 |
rider | -0.016 |
Token | he |
Token ID | 339 |
Feature activation | -3.34e-02 |
Pos logit contributions | |
ths | +0.043 |
agg | +0.042 |
eking | +0.042 |
aman | +0.041 |
prisingly | +0.039 |
Neg logit contributions | |
. | -0.026 |
." | -0.023 |
He | -0.021 |
ashes | -0.020 |
They | -0.020 |
Token | will |
Token ID | 481 |
Feature activation | +3.75e-02 |
Pos logit contributions | |
. | +0.043 |
towards | +0.043 |
He | +0.042 |
ashes | +0.042 |
." | +0.041 |
Neg logit contributions | |
ths | -0.060 |
eking | -0.057 |
agg | -0.055 |
Fore | -0.052 |
aman | -0.052 |
Token | come |
Token ID | 1282 |
Feature activation | +4.38e-02 |
Pos logit contributions | |
ths | +0.068 |
agg | +0.068 |
eking | +0.066 |
aman | +0.064 |
Kevin | +0.062 |
Neg logit contributions | |
. | -0.055 |
." | -0.051 |
ashes | -0.051 |
He | -0.049 |
rider | -0.048 |
Token | back |
Token ID | 736 |
Feature activation | -5.16e-02 |
Pos logit contributions | |
eking | +0.086 |
ths | +0.085 |
agg | +0.084 |
prisingly | +0.082 |
Jess | +0.079 |
Neg logit contributions | |
rider | -0.047 |
ashes | -0.046 |
bar | -0.045 |
powers | -0.045 |
five | -0.045 |
Token | and |
Token ID | 290 |
Feature activation | +4.39e-02 |
Pos logit contributions | |
ashes | +0.056 |
rider | +0.053 |
. | +0.051 |
He | +0.051 |
pedal | +0.051 |
Neg logit contributions | |
ths | -0.107 |
agg | -0.104 |
eking | -0.098 |
Jess | -0.098 |
aman | -0.097 |
Token | play |
Token ID | 711 |
Feature activation | -1.15e-01 |
Pos logit contributions | |
ths | +0.109 |
agg | +0.105 |
eking | +0.104 |
aman | +0.101 |
prisingly | +0.099 |
Neg logit contributions | |
. | -0.037 |
." | -0.033 |
ashes | -0.032 |
He | -0.029 |
rider | -0.028 |
Token | with |
Token ID | 351 |
Feature activation | -1.58e-02 |
Pos logit contributions | |
rider | +0.136 |
pedal | +0.129 |
He | +0.129 |
ashes | +0.127 |
powers | +0.126 |
Neg logit contributions | |
agg | -0.219 |
ths | -0.218 |
eking | -0.218 |
prisingly | -0.208 |
Jess | -0.201 |
Token | them |
Token ID | 606 |
Feature activation | +3.15e-02 |
Pos logit contributions | |
. | +0.025 |
." | +0.023 |
ashes | +0.021 |
He | +0.020 |
They | +0.020 |
Neg logit contributions | |
ths | -0.031 |
agg | -0.030 |
eking | -0.029 |
aman | -0.029 |
prisingly | -0.027 |
Token | another |
Token ID | 1194 |
Feature activation | -3.68e-02 |
Pos logit contributions | |
ths | +0.065 |
eking | +0.063 |
agg | +0.061 |
prisingly | +0.060 |
unfortunately | +0.058 |
Neg logit contributions | |
ashes | -0.034 |
powers | -0.030 |
He | -0.030 |
five | -0.029 |
Let | -0.029 |
Token | day |
Token ID | 1110 |
Feature activation | +4.15e-02 |
Pos logit contributions | |
He | +0.054 |
. | +0.053 |
ashes | +0.052 |
rider | +0.052 |
five | +0.051 |
Neg logit contributions | |
ths | -0.062 |
agg | -0.061 |
eking | -0.057 |
Jess | -0.056 |
aman | -0.056 |
Token | . |
Token ID | 13 |
Feature activation | -1.43e-01 |
Pos logit contributions | |
ths | +0.111 |
agg | +0.109 |
eking | +0.109 |
Jess | +0.105 |
Kevin | +0.104 |
Neg logit contributions | |
pedal | -0.017 |
rider | -0.016 |
ashes | -0.016 |
tire | -0.014 |
pedals | -0.014 |
Token | a |
Token ID | 257 |
Feature activation | -3.11e-02 |
Pos logit contributions | |
ths | +0.165 |
eking | +0.156 |
aman | +0.153 |
agg | +0.152 |
Kevin | +0.146 |
Neg logit contributions | |
. | -0.057 |
." | -0.046 |
He | -0.038 |
They | -0.038 |
ashes | -0.037 |
Token | little |
Token ID | 1310 |
Feature activation | +3.64e-02 |
Pos logit contributions | |
. | +0.049 |
." | +0.047 |
ashes | +0.046 |
He | +0.043 |
towards | +0.043 |
Neg logit contributions | |
ths | -0.055 |
agg | -0.054 |
eking | -0.050 |
aman | -0.050 |
Kevin | -0.049 |
Token | girl |
Token ID | 2576 |
Feature activation | +1.45e-02 |
Pos logit contributions | |
ths | +0.071 |
agg | +0.070 |
aman | +0.066 |
Jess | +0.065 |
mber | +0.065 |
Neg logit contributions | |
. | -0.045 |
." | -0.044 |
ashes | -0.043 |
Let | -0.040 |
He | -0.040 |
Token | named |
Token ID | 3706 |
Feature activation | -2.82e-02 |
Pos logit contributions | |
ths | +0.022 |
agg | +0.021 |
eking | +0.021 |
Jess | +0.020 |
aman | +0.020 |
Neg logit contributions | |
ashes | -0.026 |
He | -0.024 |
powers | -0.024 |
five | -0.024 |
pedals | -0.023 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +2.62e-02 |
Pos logit contributions | |
. | +0.030 |
." | +0.028 |
ashes | +0.023 |
They | +0.022 |
He | +0.022 |
Neg logit contributions | |
ths | -0.071 |
agg | -0.068 |
eking | -0.067 |
aman | -0.066 |
Kevin | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -1.54e-01 |
Pos logit contributions | |
Jess | +0.061 |
agg | +0.061 |
aman | +0.059 |
Fore | +0.058 |
Mill | +0.058 |
Neg logit contributions | |
ashes | -0.026 |
powers | -0.023 |
five | -0.021 |
puck | -0.019 |
pedals | -0.019 |
Token | Daisy |
Token ID | 40355 |
Feature activation | +2.51e-02 |
Pos logit contributions | |
. | +0.222 |
." | +0.204 |
He | +0.188 |
ashes | +0.187 |
They | +0.181 |
Neg logit contributions | |
ths | -0.319 |
eking | -0.305 |
agg | -0.304 |
aman | -0.293 |
prisingly | -0.280 |
Token | wanted |
Token ID | 2227 |
Feature activation | -4.63e-02 |
Pos logit contributions | |
agg | +0.043 |
ths | +0.043 |
eking | +0.040 |
Jess | +0.039 |
aman | +0.039 |
Neg logit contributions | |
ashes | -0.040 |
five | -0.036 |
." | -0.036 |
. | -0.035 |
towards | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -4.72e-03 |
Pos logit contributions | |
. | +0.032 |
." | +0.027 |
ashes | +0.022 |
He | +0.021 |
They | +0.018 |
Neg logit contributions | |
ths | -0.129 |
agg | -0.126 |
eking | -0.125 |
aman | -0.122 |
Kevin | -0.118 |
Token | learn |
Token ID | 2193 |
Feature activation | -4.20e-02 |
Pos logit contributions | |
ashes | +0.006 |
ny | +0.005 |
eer | +0.005 |
. | +0.005 |
rider | +0.005 |
Neg logit contributions | |
agg | -0.010 |
ths | -0.009 |
Kevin | -0.009 |
Jess | -0.009 |
aman | -0.008 |
Token | how |
Token ID | 703 |
Feature activation | -6.01e-02 |
Pos logit contributions | |
ashes | +0.068 |
ny | +0.064 |
He | +0.061 |
rider | +0.061 |
five | +0.060 |
Neg logit contributions | |
ths | -0.072 |
eking | -0.068 |
agg | -0.066 |
aman | -0.066 |
prisingly | -0.062 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
ths | +0.029 |
lf | +0.027 |
eking | +0.027 |
nowhere | +0.027 |
unfortunately | +0.026 |
Neg logit contributions | |
ashes | -0.010 |
five | -0.008 |
round | -0.007 |
pedal | -0.007 |
powers | -0.007 |
Token | He |
Token ID | 679 |
Feature activation | -2.44e-02 |
Pos logit contributions | |
rider | +0.113 |
ride | +0.111 |
tie | +0.111 |
five | +0.107 |
towards | +0.106 |
Neg logit contributions | |
agg | -0.311 |
omm | -0.295 |
ths | -0.287 |
Kevin | -0.285 |
unfortunately | -0.279 |
Token | wonders |
Token ID | 19294 |
Feature activation | +3.91e-03 |
Pos logit contributions | |
. | +0.030 |
." | +0.030 |
round | +0.028 |
ashes | +0.028 |
rider | +0.028 |
Neg logit contributions | |
ths | -0.046 |
eking | -0.046 |
agg | -0.044 |
eling | -0.042 |
End | -0.040 |
Token | where |
Token ID | 810 |
Feature activation | -6.34e-02 |
Pos logit contributions | |
eking | +0.006 |
ths | +0.006 |
agg | +0.006 |
ering | +0.005 |
aman | +0.005 |
Neg logit contributions | |
ashes | -0.007 |
powers | -0.006 |
." | -0.006 |
rider | -0.006 |
ny | -0.006 |
Token | they |
Token ID | 484 |
Feature activation | -3.67e-02 |
Pos logit contributions | |
ashes | +0.095 |
. | +0.094 |
." | +0.091 |
rider | +0.085 |
five | +0.084 |
Neg logit contributions | |
ths | -0.111 |
agg | -0.107 |
eking | -0.103 |
Kevin | -0.097 |
Jess | -0.096 |
Token | went |
Token ID | 1816 |
Feature activation | -8.59e-02 |
Pos logit contributions | |
. | +0.044 |
." | +0.043 |
ashes | +0.043 |
Let | +0.042 |
rider | +0.041 |
Neg logit contributions | |
eking | -0.073 |
ths | -0.072 |
agg | -0.069 |
aman | -0.066 |
Kevin | -0.064 |
Token | . |
Token ID | 13 |
Feature activation | -1.44e-01 |
Pos logit contributions | |
five | +0.058 |
powers | +0.056 |
rider | +0.055 |
ashes | +0.053 |
likes | +0.049 |
Neg logit contributions | |
eking | -0.209 |
ths | -0.203 |
aman | -0.197 |
ering | -0.195 |
agg | -0.192 |
Token | He |
Token ID | 679 |
Feature activation | -2.41e-02 |
Pos logit contributions | |
. | +0.089 |
." | +0.087 |
ashes | +0.087 |
rider | +0.082 |
five | +0.080 |
Neg logit contributions | |
agg | -0.356 |
ths | -0.349 |
Kevin | -0.331 |
sudden | -0.325 |
unfortunately | -0.324 |
Token | hopes |
Token ID | 7176 |
Feature activation | +8.66e-04 |
Pos logit contributions | |
. | +0.030 |
." | +0.029 |
ashes | +0.027 |
rider | +0.026 |
five | +0.025 |
Neg logit contributions | |
ths | -0.049 |
eking | -0.048 |
agg | -0.048 |
aman | -0.045 |
Kevin | -0.042 |
Token | they |
Token ID | 484 |
Feature activation | -3.63e-02 |
Pos logit contributions | |
ths | +0.002 |
agg | +0.002 |
eking | +0.002 |
aman | +0.002 |
Kevin | +0.002 |
Neg logit contributions | |
. | -0.001 |
." | -0.001 |
ashes | -0.001 |
He | -0.001 |
rider | -0.001 |
Token | are |
Token ID | 389 |
Feature activation | +1.04e-01 |
Pos logit contributions | |
ashes | +0.045 |
. | +0.045 |
." | +0.042 |
five | +0.042 |
He | +0.041 |
Neg logit contributions | |
ths | -0.069 |
eking | -0.068 |
agg | -0.063 |
prisingly | -0.062 |
aman | -0.061 |